key: cord-0041928-3okyyo4k
authors: nan
title: Poster Session Abstracts
date: 2010-12-15
journal: Pediatr Pulmonol
DOI: 10.1002/ppul.21339
sha: 93e41ab96746bff2cc0d899137cae2054645e6c6
doc_id: 41928
cord_uid: 3okyyo4k

nan

The high resolution, experimental 3D structures of complete ABC exporters, published since 2006, have been used to generate models of the 3D structure of the CFTR protein in different conformations. These models are useful to understand the molecular basis of the CFTR function, as well as to evaluate the impact of mutations on the CFTR 3D structure and function. The Sav1866 structure in an outward-facing conformation was first used to model the open form of the CFTR channel (1) , whereas a MsbA structure in an inward-facing conformation (called "closed apo") was afterwards considered to construct a plausible 3D model of the closed form of the channel (2) . Despite the large reorganization of the membrane-spanning domains and movements of the nucleotide-binding domains, the coupling interfaces linking these domains are relatively well conserved, suggesting that these act as pivots around which the CFTR channel dynamics occur.

We have further considered our previous models of the CFTR channel, based on the Sav1866 and MsbA 3D structures, as well as new ones constructed on the basis of the recent P-gp 3D structure, in order to highlight new structural features, which may account for specific functional features. The models especially support the hypothesis that CFTR may consist of a "broken" ABC transporter, having an "atrophied" gate at the cytoplasmic side (3) . According to our models, this gate would be located at the level of the bundle formed by the four intracellular loops (ICLs). Moreover, a careful analysis of the 3D structure models revealed several potential ligand binding sites at the interface between the domains (NBDs heterodimer interface, but also MSDs:NBDs), suggesting that these could be privileged targets for therapeutic strategies. Biosynthetic misprocessing of the CFTR, a plasma membrane chloride channel, is the hallmark of the CF cellular phenotype. Deletion of Phe508 (∆F508) in the NBD1, the most common CF mutation, compromises the membrane spanning domains (MSD1-2) and the NBD2 conformation, while it has marginal effect on the isolated NBD1 biochemical, biophysical and structural characteristics. In the ∆F508-NBD1 crystal structure, the conformational difference was limited to the F508 containing loop that could be attributed to solubilizing and revertant mutations incorporated in the domain. In contrast, more recent data and molecular modeling suggest that ∆F508 reduces the conformational stability of the NBD1 (1) (2) (3) (4) (5) . In the context of CFTR the ∆F508 manifests in a temperature-sensitive folding and stability defect of the channel, consistent with the notion that the F508 loop is involved in stabilization of the NBD1 as well as the NBD1-CL4 (MSD2) and the NBD1-CL1/2 (MSD1) interface, confirmed by Cys-crosslinking. Whether the conformational defect of multiple CFTR domains could be attributed to the ∆F508-NBD1 localized topological perturbation and/or its thermodynamic defect (1) (2) (3) (4) in concert with cooperative domain misfolding (6) remains incompletely understood. To assess the contribution of the ∆F508-NBD1 folding defect to CFTR misprocessing, the conformational stability of isolated NBD1 variants were established and correlated with the cell surface expression and biosynthetic processing efficiency of the CD4-NBD1 chimera and CFTR, respectively. The CD4-NBD1 chimera was implemented to probe the NBD1 recognition by the cellular quality control mechanism in comparison with CFTR, where extensive domain-domain interaction may compound the downstream effect of the mutation. Thermal and chemical unfolding in concert with global hydrogen-deuterium exchange revealed that the ∆F508-NBD1 was energetically destabilized, rendering the domain susceptible to proteolytic quality control both in vitro and in vivo, probed by a panel of biochemical and biophysical assays. While second site mutations improved the wt-, ∆F508-NBD1 and the wt-CFTR processing, they had modest effect on the ∆F508-CFTR. Ongoing experiments suggest that this is likely due to persistent domain misassembly. Thus restoring the ∆F508-NBD1 stability and domain-assembly defect are both required for efficient therapy in CF.

Supported by CFFT, CCFF and NSERC. 1. Richardson, JM et al. Ped. Pulm. Supp. 2007 : 30, 203. 2. Brouillette, C et al. Ped. Pulm. Suppl. 2008 : 31, 227. 3. Wang, C et al. Ped. Pulm. Suppl. 2008 : 31, 231. 4. Mulvihill, CM et al. Ped. Pulm. Supp. 2008 (5) Wieczorek, G et al. J Cystic Fibrosis 2008: 7, 295; (6) Du, K et al. Mol Biol Cell, 2009 : 20, 1903 .

The cystic fibrosis transmembrane conductance regulator (CFTR) is an epithelial chloride channel and the only member of the ABC transporter family that forms a channel. Mutations in CFTR (the major mutation is the deltaF508) cause cystic fibrosis, the most common genetic disease in the Caucasian population. CFTR consists of two membrane-spanning domains (MSDs), two nucleotide binding domains (NBDs) and an ~200-residue intrinsically disordered regulatory region (R region). Channel activity is regulated by phosphorylation (by PKA, PKC and AMPK), and ATP binding and hydrolysis, which promote the dimerization of the NBDs and channel opening. Most of the PKA consensus sites (9/10) are located in the R region, highlighting the critical role of this disordered segment in channel regulation.

In order to probe R region structure and its binding properties, we used mostly NMR based techniques, including NMR-derived chemical shifts, R2 values, residual dipolar couplings, paramagnetic relaxation enhancements, and hydrodynamic radius, as well as small angle X-ray scattering (SAXS), on non-phosphorylated and phosphorylated R region. These data were input into ENSEMBLE, a computational approach developed in the Forman-Kay group to describe disordered protein states as a set of conformations that in aggregate agree with the experimental data. To investigate the interaction between the R region and its partners we used HSQC and HNCO experiments, as well as tryptophan fluorescence.

To probe the R region interactions with other protein domains for which there is evidence of regulatory interaction, including NBD1, NBD2, the disordered C-terminal portion of CFTR and the STAS domain of the SLC26A3 chloride-bicarbonate exchanger which co-regulates CFTR. We recorded NMR spectra in the free and the bound forms to obtain residue specific information about the R region interaction within the molecule.

Examination of the regulatory region interactions revealed that the same segments in the R region are involved in binding to different partners and the binding sites are coincident with transient helical elements.

(K95C, V345C, M348C) were rapidly modified by MTSES both in activated and non-activated channels, those more deeply into the pore (Q98C, L102C, T338C, S341C, I344C) could only be modified after the channels had been activated by protein kinase A and ATP. The pattern of internal and external MTS modification that we observe allows us to divide the pore into distinct "regions" that we refer to as the outer vestibule, narrow region and inner vestibule. The outer vestibule (accessible to external but not internal MTS reagents) contains R104 (TM1) and R334 and K335 (TM6). The inner vestibule (accessible to internal but not external MTS reagents) contains K95, Q98 and L102 (TM1) as well as I344, V345, M348, A349, R352 and Q353 (TM6). The narrow region (accessible to both internal and external MTS reagents) contains the TM6 residues T338 and S341, however, no TM1 side chains were identified as being in this central region. In fact, TM1 appears to change from internal (L102) to external (R104) over a very short distance, suggesting that these two TMs are arranged differently around the CFTR channel pore. Prior to channel activation, internally applied MTS reagents are restricted to the inner vestibule, whereas following channel activation they are able to penetrate more deeply into the pore, reaching the narrow pore region. We therefore propose that channel activation involves removal of a restriction that prevents access from the cytoplasm to the pore narrow region, and suggest that the mechanism of activation involves opening of a channel "gate" in the pore inner vestibule.

Supported by CIHR and CCFF. Cystic fibrosis transmembrane conductance regulator (CFTR) is a unique ATP-binding cassette (ABC) transporter protein in that it functions as a chloride channel. Binding and hydrolysis of ATP in the nucleotide binding domains lead to gating conformational changes in the transmembrane domains (TMDs). To study the functional role of the cytoplasmic side of the sixth transmembrane segment (TM6) of TMDs, we mutated all 18 endogenous cysteines, introduced one cysteine at each position of into this cysless CFTR, and examined their reactivity to intracellularly applied thiol-specific methanethiosulfonate (MTS) reagents. The effects of differently charged reagents indicate that the cytoplasmic part (341-354) of TM6 assumes a secondary structure of an α-helix. Interestingly, modification by positively charged reagents of cysteine residues at one face (344, 348 and 352) of the helix can affect gating. In the cases of I344C and M348C, after modification, gating behavior is profoundly altered whereas the singlechannel amplitude is only slightly changed. The open time is prolonged and the closed time is shortened, suggesting that depositions of positive charges at these positions stabilize the open state but destabilize the closed state. For R352C, which exhibits reduced single-channel amplitude, modifications by two positively charged MTS reagents of different size completely restore the single-channel amplitude but have distinct effects on both the open time and the closed time, indicating that the arginine residue not only regulates anion permeation but is also involved in gating motion. These results are consistent with the idea that a helix rotation of TM6, which has been proposed to be part of the molecular motions during transport cycles in other ABC transporters, is associated with gating of the CFTR pore.

Phosphorylation modulates the helical propensity which affects the R region interactions; competition between binding of the same elements among the various binding partners enables this regulatory region to act as a unique dynamic integrator.

GATING DEFECTS IN HUMAN-MOUSE CFTR ∆F508 CHIMERAS Dong, Q. 1 ; Ostedgaard, L.S. 1 ; Zhang, Y. 1 ; Welsh, M.J. 1, 2 1. Internal Medicine, University of Iowa, Iowa City, IA, USA; 2. Howard Hughes Medical Institute, Iowa City, IA, USA CFTR ∆F508 is the most common mutation in cystic fibrosis. It has at least three defects: it disrupts biosynthetic processing, it reduces the channel opening rate, and it decreases the protein lifetime on the cell membrane. Mouse CFTR ∆F508 exhibits a similar gating defect as human CFTR ∆F508, however, its processing defect is less severe than hCFTR∆F508. We found that the C-terminal portion of mouse nucleotide binding domain 1 (the Regulatory Extension (RE) of mNBD1) was critical for the species-specific difference in CFTR ∆F508 processing. However, the RE region did not affect the ∆F508-induced gating defect. Models of CFTR structure suggest interactions between NBD1 and the intracellular loops (ICLs) of membrane spanning domain 2 (MSD2). We found that substituting mMSD2 into hCFTR (hmMSD2) did not rescue the CFTR ∆F508 processing defect, but it prevented the ∆F508-induced gating defect. To better understand which regions of MSD2 are important for the CFTR∆F508 gating defects, we designed a series of mutations in the ICLs of MSD2. We found that like the hmMSD2 chimera, mutating residues from the human to the mouse sequence at either ICL3 or ICL4 region prevented the ∆F508-induced gating defect. This was a surprise because both mCFTR ∆F508 and hCFTR ∆F508 exhibit similar gating defects. An explanation for these apparently paradoxical findings came when we compared chimeras with a human backbone but mouse ICLs and/or mouse NBDs. We found that a hmNBD1 chimera with a mICL4 mutation and a hmNBD2 chimera with a mICL3 mutation retained the ∆F508-induced gating defect (prolonged closed time). However hmNBD1 chimera with a mICL3 mutation and hmNBD2 chimera with a mICL4 mutation prevented the ∆F508-induced gating defect. These results suggest that CFTR channel gating involves interactons between ICL4 and NBD1 and between ICL3 and NBD2. By pointing to specific regions of CFTR that alter gating of CFTR∆F508, the data may suggest novel strategies for correcting the defects caused by this mutation.

Supported by the NIH and the Cystic Fibrosis Foundation. Opening of the CFTR chloride channel is dependent on both phosphorylation and on ATP binding and hydrolysis. However, the structural changes in the channel pore that are triggered by these cytoplasmic factors to cause channel opening are not known. To gain information on the conformational changes in the pore associated with channel opening, we have used patch clamp recording to investigate the accessibility of cysteine reactive reagents to cysteines introduced at different sites in the pore. Individual cysteines were engineered into the first (TM1) and sixth (TM6) transmembrane helices of a cysteine-less variant of CFTR using site-directed mutagenesis. Cysteine reactive methanethiosulfonate (MTS) reagents were then applied to the extracellular or intracellular side of the membrane. When applied to the cytoplasmic side of open CFTR channels, the MTS reagents [2-sulfonatoethyl] methanethiosulfonate (MTSES) and [2-(trimethylammonium) ethyl] methanethiosulfonate (MTSET) were able to modify irreversibly cysteines substituted for (from outside to inside) L102, P99, Q98 and K95 in TM1, and F337, T338, S341, I344, V345, M348, A349, R352 and Q353 in TM6, leading to increases or decreases in macroscopic current amplitude. External MTSET was able to modify R104C but not L102C, Q98C or K95C (in TM1), and T338C and S341C but not I344C, V345C or M348C (in TM6), confirming that MTS reagents are not permeant in CFTR. When used to pretreat excised membrane patches, the pattern of modification by cytoplasmic MTSES was dependent on the activation state of the channels. Thus, while cysteines located relatively superficially into the pore from the inside The most common mutation in CF patients, F508del, disrupts CFTR folding in the endoplasmic reticulum and targets the newly synthesized pro-tein for degradation via the ubiquitin proteasome pathway. Phe 508 is located in the first nucleotide binding domain (NBD1) of CFTR where it facilitates the coupling of ATP hydrolysis to transmembrane helix movement during channel gating. Importantly, F508del increases local dynamics of the domain near residues 507-511, but does not significantly alter native NBD1 structure. It has therefore been proposed that the mutation primarily disrupts the pathway of NBD1 folding, although how this occurs remains unknown. NBD1 exhibits a particularly complex topography consisting of three canonical subdomains: an antiparallel ABC beta-subdomain, an alpha-helical subdomain, and an F1-like ATPase beta-sheet core, as well as an unstructured regulatory insertion (RI; residues 405-436) and C-terminal helical extension (RE; residues 656-673). To understand how nascent NBD1 acquires its native structure, we developed a novel method using fluorescence resonance energy transfer (FRET) to characterize the cotranslational compaction of nascent polypeptides as they are synthesized and emerge from the ribosome exit tunnel. This strategy is based on the principle that the distance between incorporated donor and acceptor dyes should decrease, and hence the FRET should increase, as the polypeptide transitions from an extended to a folded conformation.

To characterize cotranslational NBD1 folding, a donor fluorophore (CFP) was fused to the N-terminus (residue 389) and a small acceptor dye (7-nitrobenz-2-oxa-1,3-diazole) was incorporated at engineered stop (UAG) codons using a synthetic suppressor tRNA. Truncated mRNAs were then synthesized in vitro to generate nascent NBD1 polypeptides that retain a covalent peptidyl-tRNA bond and mimic kinetically trapped translation intermediates. FRET efficiencies obtained for multiple surface-exposed residues indicated that ribosome bound NBD1 (truncated at residue 744) acquires a native-like fold. Translation intermediates truncated between residues 460 and 744 further identified an N-terminal subdomain (residues 389-500) that folds autonomously as residues 490-510 emerge from the ribosome exit tunnel. This region contains the ABC beta-subdomain, the RI, and key elements that stabilize ATP binding, including the Walker A motif (Lys464), Trp401, and other residues that contact the ribose and adenine ring. Remarkably, real-time folding kinetics of ribosome released nascent polypeptides demonstrated that that ATP serves as a critical ligand that facilitates rapid, de novo folding of this subdomain. Thus, we predict that NBD1 folding in cells is a vectorial, multi-state process initiated by rapid cotranslational compaction of an N-terminal ATP-binding sub-domain. Because Phe508 is still in the ribosome tunnel during these early folding events, we predict that the F508del mutation affects later stages of folding that involve the ABC alpha-sub-domain and/or formation of the F1-type ATPase betasheet core.

Supported by NIH and CFFT.

pathway, and rescuing the [∆F508]CFTR trafficking defect. The impact of this work is that C4-ceramide, which occurs in cells in low abundance, may be, or contribute to the development of, novel correctors for dystrafficked mutant CFTR in CF epithelial cells.

Swiatecka-Urban, A.; Cihil, K.; Swiatecka-Urban, A. Pediatric Nephrology, Children's Hospital of Pittsburgh, Pittsburgh, PA, USA

In our previous study we have shown that the ubiquitin ligase, c-Cbl facilitates endocytosis and lysosomal degradation of wild type (WT)-CFTR in human airway epithelial cells. To focus on the role of c-Cbl in cystic fibrosis, we examined whether c-Cbl regulates trafficking of ∆F508-CFTR in human airway epithelial cells. Further, we investigated whether protein adaptors and phosphorylation regulate the c-Cbl-CFTR interaction. Studies were conducted in the primary differentiated human bronchial epithelial cells obtained from the University of Pittsburgh Human Airway Cell Core (HBE cells; homozygous WT-CFTR or ∆F508-CFTR) and in polarized human airway epithelial cells (CFBE41o-) stably expressing WT-CFTR or ∆F508-CFTR. Decreased temperature (27 ο C) was used to increase the trafficking and expression of ∆F508-CFTR in the plasma membrane. Endogenous c-Cbl co-immunoprecipitated with ∆F508-CFTR in HBE and CFBE41o-cells and silencing the c-Cbl expression in CFBE41o-cells increased the plasma membrane abundance of ∆F508-CFTR. These data suggest that c-Cbl regulates the endocytosis and post-endocytic trafficking of ∆508-CFTR in human airway epithelial cells. The role of c-Cbl in endocytic trafficking depends on spatially and temporally regulated interactions with protein adaptors. CIN85 (c-Cbl interacting protein of 85 kDa) recruits the c-Cbl-target protein complex to the clathrin-coated vesicles in nonepithelial cells. We examined the role of CIN85 in human airway epithelial cells. Endogenous CIN85 formed a complex with c-Cbl and CFTR and silencing the CIN85 expression increased CFTR abundance in the plasma membrane in CFBE41o-cells expressing WT-CFTR or ∆F508-CFTR. These data indicate that CIN85 facilitates the c-Cbl mediated CFTR endocytosis. Phosphorylation plays a critical role in protein-protein interactions. We investigated whether phosphorylation regulates the interaction between c-Cbl and CFTR. c-Cbl is known to be regulated by phosphorylation on its serine-rich 30 amino acid region (residue 615-644) and on tyrosine residues (674-774) in its distal C-terminus. Treatment of HBE and CFBE41o-cells with pervanadate (Tyr-phosphatase inhibitor) and calyculin A (Ser/Thrphosphatase inhibitor) increased the amount and mobility of c-Cbl in an additive manner indicating that c-Cbl is regulated by Ser/Thr and Tyr phosphorylation in human airway epithelial cells. Endogenous c-Cbl was not found in complex with CFTR immunoprecipitated with the phospho-(Ser/Thr) PKA substrate antibody in the presence of pervanadate and calyculin A. These data suggest that c-Cbl does not interact with the PKA activated, (Ser/Thr)-phosphorylated CFTR. Overall, calyculin A together with pervanadate augmented the c-Cbl-CFTR interaction and the Tyr-774 phosphorylated c-Cbl was found in complex with CFTR in co-immunoprecipitation experiments. Taken together, our data indicate that in human airway epithelial cells the PKA activated CFTR is not targeted by c-Cbl and that the complex formation between c-Cbl and CFTR is augmented by phosphorylation of the Ser/Thr and Tyr residues on c-Cbl and/or its adaptor(s).

(Supported by NIH R01HL090767 & 3R01HL090767-02S1 to ASU.) research group established independent protocols to investigate effect(s) of gene knock-down (via Ad-shRNA) on ∆F508-CFTR maturation. Protocols included 1) A study of CF bronchial epithelial cells expressing the halide sensitive variant of eYFP and measurement of ∆F508-CFTR activity at the cell surface. 2) Short circuit current in primary human bronchial epithelial cells (∆F508/∆F508), 3) Appearance of rescued ∆F508 CFTR at the plasma membrane in CF bronchial epithelial cells monitored biochemically, 4) Effects on CFTR-dependent release of inflammatory markers (chemokines and cytokines) from IB3 cells (∆F508/W1282X), and 5) Short circuit current and Western blotting in CFBE cells transduced with lentivirus encoding ∆F508-CFTR. Preliminary results indicate significant activity of BioFocus shRNAs in several of the independent protocols and laboratories, particularly against gene products such as AHSA1 and 2 (activators of HSP 90) and HDAC7A (a member of the histone deacetylase family). A profile of shRNAs found to improve ∆F508 processing, including spectrum of activity data and gene-network annotation of the relevant pathways, will be presented. Identifying the most robust molecular targets for ∆F508 CFTR correction (from among hundreds of candidates in the CFTR "interactome") has been limited by the complexity of the relevant cellular pathways. The studies described here provide a means by which chaperones and other contributors to CFTR misprocessing can be evaluated, prioritized, and better understood in the future for development of new therapeutic approaches and delineating genetic modifiers that contribute to variation in severity of CF onset and disease progression. The project represents a collaboration among members of the CFTR Folding Consortium. Supported by the CFF and NIH.

to attach on the ER membrane. The ER fraction of Hdj-2 is farnesylated and the farnesylated Hdj-2 can be abolished by adding FTI-277. We further investigated the role of the molecular chaperone in CFTR degradation and maturation. We found that Hsp70 and its cochaperone Hdj-2 interacted significantly with ∆F508 CFTR and wt-CFTR in airway epithelial cells. Hsp70 interacted with the immature form of CFTR while Hdj-2 associated with both immature form and ubiquitylated CFTR. Immunoprecipitation and GST-fusion protein pull-down experiments revealed that Hdj-2 interacted with ubiquitin. In steady-state, over-expression of Hdj-2 increased both immature and mature forms of wt-CFTR, but it decreased immature form of ∆F508CFTR. Pulse-chase studies showed that co-expression of Hdj-2 promoted ∆F508 CFTR degradation, reducing its half-life from 85 to 50 min. In contrast, Hdj-2 expression did not significantly affect wt-CFTR biogenesis. These data are consistent with our finding that Hdj-2 shows a selective physical interaction with ∆F508 CFTR, which translates into a functional discrimination between the mutant and its wt counterpart. A role for Hsp70 in Hdj-2 mediated ∆F508CFTR degradation was tested in which Hsp70 knockdown increased ∆F508CFTR expression 3-4 fold. Hjd-2 regulates ∆F508CFTR maturation and knockdown of endogenous Hdj-2 promoted ∆F508CFTR and its cAMP-dependent anion conductance in airway epithelial cells. In contrast, we were unable to detect maturation of ∆F508 CFTR in Hsp70 knockdown experiments, indicating that the maturation of ∆F508 CFTR mediated by reduced Hdj-2 expression may be independent of Hsp70 function. Taken together, our data demonstrate that Hdj-2 is a molecular sensor that can detect differences in the folding related to the ∆F508CFTR maturation. These data also highlight a novel pathway for Hdj-2 mediated, ubiquitin-dependent degradation of ∆F508 CFTR. Reduction of Hdj-2 expression or its association with ∆F508 CFTR promotes maturation of this mutant. Farnesyl transferase inhibitor drugs, such as tipifarnib in clinical trials for cancer therapy, will represent a potential therapeutic approach for cystic fibrosis.

[Supported by CF Foundation SUN08G0 and NIH grant HL096800.]

The cystic fibrosis transmembrane conductance regulator (CFTR) was the first mammalian membrane protein implicated as a substrate for endoplasmic reticulum associated degradation (ERAD), and it has served as a model for the folding or disposal of polytopic membrane proteins. Steps in ERAD usually include the recognition and removal of misfolded proteins from the ER, followed by their ubiquitylation and proteasome-dependent degradation. Due to its complex folding and domain assembly requirements, much of WT CFTR and ~100% of the common folding mutant, ∆F508 CFTR, are degraded in most systems.

Small heat shock proteins (sHsps) bind destabilized proteins during cell stress and disease, but their physiological functions are less clear. Hsp27 is expressed in airway epithelial cells, and it selectively interacted with the common CFTR mutant, ∆F508, and targeted it for proteasomal degradation. This action of Hsp27 appears to result from its ability to interact physically with the SUMO E2 conjugating enzyme, Ubc9, and like Hsp27, Ubc9 selectively promoted ∆F508 CFTR proteolysis. Similarly, the knockdown of Hsp27 or the SUMO E1 enzyme, SAE1/2, increased CFTR expression 2-3 fold. Hsp27 expression promoted the sumoylation of ∆F508 CFTR in vivo, and disabling the SUMO pathway via SUMO E1 knockdown reduced the ability of Hsp27 to degrade mutant CFTR.

To begin to evaluate the properties of CFTR that lead to its sumoylation/degradation, we determined the modification of CFTR NBD1 by SUMO-1 in vitro. The reaction mixture included the purified components: E1, E2, SUMO-1, WT or ∆F508 NBD1 and ATP. NBD1 sumoylation was detected either by the molecular mass shift observed upon blotting with an NBD1 antibody or from blots performed using anti-SUMO-1. SUMO modification of NBD1 increased as a function of time over a 0-120 min assay period, with significantly greater modification observed for ∆F508 vs. WT NBD1. Reactions run at different ATP concentrations demonstrated the expected ATP dependence of NBD1 modification, and showed also that WT NBD1 sumoylation increased at low ATP whereas the ∆F508 NBD was less posed to act in a dominant negative manner, its mechanism of inhibiting the Hsc70-CHIP degradation pathway remains poorly understood.

To address this question, we used a cell-free system that reconstitutes CFTR ubiquitination, dislocation, and proteasome-mediated degradation. This system lacks transcriptional activity and therefore allows manipulation of ER-associated degradation components without activating cellular compensatory mechanisms. Exogenous addition of wild-type CHIP had minimal effect on CFTR stability, whereas addition of P269A CHIP strongly inhibited CFTR ubiquitination, and blocked degradation in a dose-dependent manner. In the presence of P269A CHIP, addition of recombinant wild-type CHIP and/or Hsc70 partially restored CFTR degradation. The apparent concentration of P269A CHIP required for 50% inhibition of CFTR degradation was ~2.5 µM, which was much higher than the predicted endogenous CHIP level and comparable to endogenous Hsc70 concentration (~2 µM). Thus, it is unlikely that P269A CHIP simply competes with wild-type CHIP for Hsc70 binding. Surprisingly, pull down experiments revealed that P269A CHIP associated much more efficiently with CFTR than wild-type CHIP, suggesting that the mutant may stabilize chaperone (presumably Hsc70) binding to substrate. To better understand how P269A CHIP inhibits CFTR degradation, cytosolic CHIP binding proteins were isolated using Histagged affinity purification. Preliminary studies have isolated several proteins that selectively bind to P269A CHIP; identification by mass spectrometry analysis is in progress.

(Supported by CFFT, NIH, and the Manpei Suzuki Diabetes Foundation.) The cystic fibrosis transmembrane conductance regulator (CFTR) operates as a macromolecular complex at the apical membranes of epithelial cells. Complex formation is facilitated by interaction with the scaffolding protein EBP50/NHERF1 via a C-terminal PDZ binding motif in CFTR. To elucidate the relationship between complex formation and CFTR localization in vivo, we have been investigating the consequences of S1455X mutation that truncates the C-terminus of CFTR but does not cause cystic fibrosis. It has been previously demonstrated that localization of CFTR is not fully dependent on the PDZ binding motif (Swiatecka-Urban et al., 2002; Ostedgaard et al., 2003) and that the C-terminus upstream of the PDZ binding motif contains sequences capable of directing apical localization (Milewski et al., 2001 (Milewski et al., , 2005 . Using protein overlay experiments, we discovered that a peptide encompassing the last 111 residues of the CFTR Cterminus enhanced EBP50 multimerization. Deletion analysis and alanine scanning mutagenesis mapped the region conferring enhanced multimerization to an internal sequence in the C-terminus (1417EENKVR1422). Full-length wild-type CFTR stably expressed in polarized Madin-Darby Canine Kidney (MDCK) cells enhanced interaction of HA-and c-myc tagged EBP50 and facilitated complex formation with actin. Full-length CFTR with the EENKVR region substituted with alanines was processed to mature, fully glycosylated protein but was unable to promote EBP50 multimerization. Furthermore, confocal microscopy and cell surface biotinylation demonstrated that CFTR with the EENKVR substitution was not localized to apical membranes of polarized MDCK cells. On the other hand, CFTR bearing the S1455X mutation could enhance EBP50 multimerization and localize to apical membranes, thereby providing an explanation for the moderate phenotype associated with this mutation. Together, these results suggest that CFTR and EBP50 have a dynamic interaction mediated by an internal C-terminal sequence that promotes multiprotein complex formation and apical membrane localization.

References sensitive to ATP in the range 0.1-2 mM. Thus, the native NBD1 conformation, as exemplified by the WT domain at high ATP, was a poor substrate for SUMO modification relative to ∆F508 NBD1. These observations, together with the eight degree lower melting temperature of the ∆F508 NBD1, suggest that the native NBD1 conformation is not modified by SUMO, whereas non-native conformational intermediates are preferentially sumoylated. These findings link sHsp-mediated mutant CFTR sumoylation to protein degradation, and they raise the possibility that folding intermediates formed during CFTR biogenesis are stabilized by SUMO addition until the native conformation is obtained.

[ The Hsp40 cochaperones, in concert with heat shock proteins of 70 kDa (Hsp70 and constitutive Hsc70), play a central role in determining the fate of newly synthesized CFTR in the endoplasmic reticulum. Hsp40s stimulate ATP hydrolysis by Hsp/c70 via their J domains, and are thereby believed to recruit substrates and/or stabilize substrate binding to Hsp/c70. Hdj-2 is a membrane-localized Hsp40 that is farnesylated at its C-terminus. Together with Hsc70, Hdj-2 binds CFTR NBD1 and is proposed to facilitate early steps in CFTR folding and maturation. Consistent with this, overexpression of Hdj-2 in cells increases the steady-state levels of both mature and immature CFTR. However, Hdj-2 also cooperates with Hsc70 and CHIP E3 ubiquitin ligase to mediate CFTR ubiquitination in vitro. The mechanism by which Hdj-2 carries out these opposing roles therefore has important implications for CFTR biogenesis and trafficking.

Here we investigate the effect of manipulating Hdj-2 levels on the fate of CFTR in vitro and in cells. When added to a reconstituted degradation system, addition of wild-type Hdj-2 or a mutant that lacks the C-terminal three amino acids required for farnesylation (∆QTS), both markedly inhibited CFTR degradation with an apparent IC 50 of ~6 µM and ~4 µM, respectively. Interestingly, inhibitory concentrations of Hdj-2 were approximately 50 fold higher than endogenous cytosolic Hdj-2 levels (0.1 µM) and 2 to 3 times higher than Hsc70 (~2 µM). Contrary to the current view that Hsp40 primarily stimulates Hsp/c70 binding to substrate, we found that addition of ∆QTS Hdj-2 at concentrations that block degradation, strongly inhibited CFTR-Hsc70 co-immunoprecipitation. Similarly, overexpression of wildtype or ∆QTS Hdj-2 in HEK293 cells increased both band B and C of wildtype CFTR and band B of ∆F508 CFTR. Therefore, excess Hdj-2 prevents CFTR degradation regardless of its farnesylation status. Based on these results, we predict that Hdj-2 stabilizes newly synthesized CFTR by competition with cytosolic proteins. (Supported by CFFT, NIH, and the Manpei Suzuki Diabetes Foundation.) Despite significant knowledge of the mechanisms and the molecular factors involved in CFTR biogenesis, trafficking and function, some aspects of these processes are not yet fully understood. Protein kinases and phosphatases have long been known to regulate CFTR function. However, their role in CFTR biogenesis and trafficking remains less explored. Spleen Tyrosine Kinase (SYK), a non-receptor tyrosine kinase with a role in signal transduction and signalling, has a consensus phosphorylation site consisting of a tyrosine followed by two negative residues. In CFTR sequence, this consensus appears only once at residue 512 (i.e., very close to residue F508, deleted in most CF patients).

Our aim here was to characterize the relevance of SYK in CFTR biogenesis and trafficking.

Firstly, we used RT-PCR analysis to show that SYK is endogenously expressed in epithelial respiratory cells (Calu-3 and virally transduced wt-or F508del-CFBE) suggesting a possible physiological role for this kinase. Coimmunoprecipitation of CFTR followed by Western blot (WB) with SYK antibody (and vice-versa) shows that SYK interacts in vivo with CFTR. Phosphorylation assays, using immunoprecipitated SYK and purified CFTR wt-NBD1 or Y512F-NBD1, revealed that in vitro SYK not only undergoes auto-phosphorylation but it also phosphorylates wt-NBD1, but not Y512F-NBD1. Furthermore, the observed phosphorylation is abolished when a dead kinase mutant of SYK is used in the same assay.

To analyse the biochemical and functional consequences of mutating tyrosine residue 512, target for SYK phosphorylation, we produced cell lines stably expressing CFTR mutants (both wt-and F508del-CFTR), where Y512 was substituted by either: 1) the neutral residue alanine (A); 2) phenylalanine (F), also neutral, but mimicking the bulky side group of tyrosine; or 3) acidic residues aspartic acid (D) and glutamic acid (E). WB analyses and pulse-chase experiments followed by CFTR immunoprecipitation using these cells lines show that Y512A/D, but not Y512F (on wt-CFTR) have decreased steady-state levels and low efficiency of processing. Analysis of CFTR function by the iodide efflux assay, revealed that cells expressing Y512A-CFTR have ~15% less IBMX/Fsk-stimulated iodide efflux than wt-CFTR cells, whereas Y512D-CFTR expressing cells show ~60% reduction.

Altogether, our data suggest that phosphorylation of wt-CFTR by SYK at Y512 exerts a positive effect on CFTR stability and processing.

Work supported by EU grant FP6-LSH-2005-037365 (TargetScreen2) and pluriannual funding of BioFig (FCT, Portugal) . Simão Luz is recipient of FCT PhD fellowship SFRH/BD/47445/2008 grant.

Watson, M.; Gilmore, R.; O'Neal, W.K.; Gentzsch, M.; Tarran, R. CF Center, UNC-Chapel Hill, Chapel Hill, NC, USA CFTR is a chloride channel that plays an important role in regulating airway fluid homeostasis. The adenosine 2B receptor (A2BR) and the β 2 adrenergic receptor (β 2 AR) activate CFTR via G protein linked stimulation of the cAMP/PKA second messenger system. The present study examines interaction of these two receptors with CFTR at the plasma membrane. In polarized human bronchial epithelia cells (HBECs) we found cytochalasin D (CD) inhibits the ability of the A2BR, but not the β 2 AR, to stimulate CFTR and thereby regulate airway surface liquid (ASL). To further characterize interactions of these membrane proteins we used wild type BHK cells and BHK cells transfected with CFTR (BHKcftr) and either A2BR or β 2 AR constructs fused with CFP or YFP. In BHKcftr cells A2BR stimulation increased cAMP levels significantly more than in wild type BHK cells (148±10% vs. 24±8% respectively). β 2 AR stimulated cAMP production was insensitive to the presence of CFTR. We confirmed these data in HBECs.

Upon CD treatment, cAMP stimulation in BHK cells was unchanged but the increased cAMP production in BHKcftr cells was abolished. The effect of CD on cAMP accumulation was reversed in BHKcftr cells expressing β 2 AR, where cAMP accumulation was greater in CD treated cells. We used FRET to determine spatial relationships between A2BR, β 2 AR, and CFTR. There was an adenosine (ADO) dose dependent increase (4-16%) in FRET between yfpA2BR and cfpCFTR (EC 50 = 1.7 µM). FRET between β 2 AR and CFTR (~4% basal FRET) was insensitive to isoproterenol (ISO) stimulation. ADO stimulated increases in yfpA2BR-cfpCFTR FRET was abolished with CD. Post synaptic density, Disc large, and ZO-1 protein (PDZ) motifs have been previously implicated in organization and location of membrane proteins. A2BR and β 2 AR have PDZ motifs that differ from one another, A2BR having a type 2 motif (GVGL) and β 2 AR having a type 1 motif (DSLL), each located on the c-terminus. To investigate the possibility that PDZ binding was involved in the specific interactions of A2BR and β 2 AR with CFTR we made A2BR and β 2 AR constructs where the c-terminal PDZ binding motifs were switched (i.e. A2BRpdz1 was switched from type 2 Ç type 1 and the reverse for the β 2 ARpdz2). In yfpA2BRpdz1 cells the FRET response was now insensitive to ADO stimulation. Conversely, by switching PDZ motifs, the yfpβ 2 ARpdz2 FRET response was now increased by ISO stimulation. cAMP measurements performed using the PDZ motif mutants reflected the results we saw in the FRET assays and β 2 AR became CFTR "sensitive." Immunofluorescence in HBECs indicates that proteins involved in PDZ membrane binding such as actin, ezrin, and NHERF2 were increased in normal compared to CF HBECs suggesting that CFTR organizes its protein environment to facilitate preferential insertion of proteins with type 2 PDZ binding motifs (such as A2BR). In summary, we demonstrate that A2BR and β 2 AR interact with CFTR in a similar, yet different manner. The differences appear to be, at least in part, due to the c-terminal PDZ motifs on these proteins. We are currently examining the role CFTR may play as an organizational hub in the plasma membrane for signal compartmentalization of its effectors.

This work supported by NIH P50HL084934, R01HL74158-3, P30DK34987 and the CFF.

The epithelial sodium channel (ENaC) and cystic fibrosis transmembrane conductance regulator (CFTR) are required for ion absorption and secretion at the apical membrane of epithelia in airways and other tissues. In airway epithelia inhibition of ENaC by CFTR prevents excessive sodium absorption. The absence of functional CFTR in cystic fibrosis results in hyperactive ENaC channels and consequently dehydration of the airway surface. In the past several years it has become obvious that limited proteolysis of ENaC's extracellular domains regulates its open probability. In addition, several studies have indicated that CFTR physically associates with ENaC (Kunzelmann et al. 1997 FEBS Lett. 400: 341-4; Ji et al. 2000 J Biol Chem. 275: 27947-56; Berdiev et al. 2007 J Biol Chem. 282: 36481-8) . In this study we have investigated the interaction of ENaC and CFTR in Xenopus oocytes and airway epithelial cultures and assessed its impact on ENaC regulation by proteolysis. We demonstrate for the first time that CFTR markedly impedes proteolytic processing of ENaC's extracellular domains. Further, co-immunoprecipitation experiments revealed that CFTR interacts mainly with α and γ subunits of ENaC. Proteolytic cleavage of both α and γ ENaC by a channel activating protease was drastically diminished in the presence of CFTR. Defined proteolysis of these subunits is central to the regulation of ENaC and thus inhibition of proteolytic cleavage by CFTR provides a means to regulate ENaC activity. While proteolytic processing of ENaC was drastically inhibited by wild-type CFTR, it was not significantly altered by ∆F508 CFTR, which is the most common mutant protein in cystic fibrosis patients. To verify a role of CFTR in ENaC proteolysis in human airway epithelia, we analyzed the proteolytic state of ENaC in normal and cystic fibrosis airway epithelial cultures that lack functional CFTR. In primary cystic fibrosis airway epithelial cells that were homozygous for ∆F508 CFTR the amount of proteolytically cleaved α ENaC was significantly increased when compared to normal cultures. These observations suggest

The severity of pulmonary disease in patients with cystic fibrosis (CF) varies substantially even among patients with identical genotypes. Genetic modifiers are an important source of this phenotypic variation. Transforming growth factor β1 (TGFβ1) gene polymorphisms are known to be modifiers of pulmonary function in CF. Our long-term objective is to elucidate the mechanisms by which TGFβ1 acts as a gene modifier through its effects on CFTR in human airway epithelial cells. Although TGFβ1 has been shown to decrease CFTR expression in nasal polyps of non-CF individuals and colonic epithelial cells, its effect on CFTR has not been studied in human airway epithelial cells. Recent work has demonstrated that the multifunctional adaptor protein Disabled-2 (Dab2) is integral to TGFβ1 signaling, although it remains unknown whether Dab2 plays a similar role in TGFβ1 signaling in human airway epithelial cells. Our previously published work demonstrated that Dab2 and CFTR co-immunoprecipitated in human airway epithelial cells. Thus, the goals of the study were: (1) to determine the effects of TGFβ1 on CFTR in human airway epithelial cells; and (2) to examine whether Dab2 mediates signal transduction from the TGFβ receptor I (TGFβ-RI) to the downstream transcription factor, Smad2. Studies were conducted in primary differentiated human bronchial epithelial cells obtained from the University of Pittsburgh Human Airway Cell Core (HBE cells; homozygous WT-CFTR) and in polarized human airway epithelial cells (CFBE41o-) stably expressing WT-CFTR or ∆F508-CFTR. Our studies demonstrate that TGFβ1 treatment induced transcriptional responses, including Smad2 phosphorylation in the HBE and CFBE41o-cells. Treatment with TGFβ1 significantly decreased CFTR expression in cell lysates and apical plasma membrane in a concentration and time-dependent man-that CFTR protects ENaC from proteolytic processing by proteases in airway epithelia. The lack of this control mechanism in cystic fibrosis may account for hyperactive sodium channels that have been implicated in dehydration of the airway surface liquid and subsequent inefficient mucociliary clearance.

Supported by the NIH [5R01HL080561, 5P01HL034322]. Sumoylation is the covalent and reversible post-translational conjugation of the small ubiquitin-like modifier (SUMO) protein to lysine residues of protein. The enzymatic machinery mediating sumoylation is similar to that of the ubiquitin system, including an E1 activating enzyme, an E2 conjugating enzyme, E3 ligases and desumoylating enzymes (Sentrin/SUMOspecific proteases, or SENPs). Sumoylation has been shown to play a role in transcription, cell division, DNA repair, and, more recently, in protein quality control and ion channel function.

We have reported earlier that over-expression of the SUMO-conjugating enzyme Ubc9 decreased the steady-state level of CFTR, whereas the overexpression of the desumoylating enzyme SENP1 increased it; therefore we have initiated a systematic analysis of the effect of sumoylation on CFTR biogenesis.

Increasing the intracellular level of SUMO1, SUMO2, and SUMO3 accelerated the intracellular degradation of ∆F508 CFTR in HEK293 cells. Over-expression of SUMO1 or SUMO3 reduced the half-life of ∆F508 CFTR by half, and the over-expression of SUMO3 had a smaller effect. The general increase in sumoylation observed during hypoxia presumably has a similar effect, making sumoylation physiologically relevant in CF patients.

Next, we performed RT-PCR reactions using total RNA isolated from human bronchial epithelial (HBE) cells, in order to ascertain which SENPs are expressed in a cell type physiologically relevant to CF. With the exception of SENP3, messenger RNA of each SENP could be detected in HBE cells from four different non-CF patients. We detected mRNA encoding each of the six SENPs in three frequently used cell lines, Calu-3 cells, HEK293 cells, and HeLa cells.

Co-expression of each of the six SENPs with ∆F508 CFTR in HEK293 cells increased the level of ∆F508 CFTR to varying degrees. On the other hand, SENPs had varying effects on the steady-state level of WT CFTR: SENP1, SENP2, and SENP6 increased the level of WT CFTR, whereas the others had no effect.

Comparing the effect of each SENP on ∆F508 CFTR with that on WT CFTR revealed that three of them increased the level of ∆F508 CFTR whereas they had no effect on WT CFTR, indicating that they might be able to distinguish between the two CFTR species.

Most of the research on SENPs focuses on nuclear events and extranuclear localization has been recognized for only a few of them. In order to establish the possibility of a physical interaction between CFTR and the SENPs we over-expressed each SENP in HEK293 cells and performed a crude fractionation of the cells to nuclear, cytosolic, and membrane fractions. The bulk of each SENP was detected in the nuclear or the non-nuclear membrane fraction, and only in the case of SENP5 was some protein detected in the cytosolic fraction. This data clearly indicate that the SENPs can be found in (on) non-nuclear membranes; therefore they may interact with CFTR directly.

In summary, these data point to desumoylating enzymes having therapeutic potential in CF and warrant further study of the interactions of SENPs with ∆F508 CFTR.

[Supported by NIH: DK068196, DK072506; CFF: ner, and treatment with TGFβ-RI inhibitors prior to stimulation with TGFβ1 prevented Smad2 phosphorylation. Endogenous Dab2 co-immunoprecipitated with members of the TGFβ1 signaling cascade Smad2 and 3, and with CFTR. Moreover, Dab2 silencing reduced Smad2 phosphorylation at baseline and after TGFβ1 treatment. These data indicate that Dab2 mediates TGFβ1 signaling in human airway epithelial cells. Our prior data have shown that silencing Dab2 increased CFTR abundance in cell lysates and in the plasma membrane and increased the CFTR mediated chloride secretion. Taken together, our data demonstrate that in human airway epithelial cells, including primary differentiated human bronchial epithelial cells, TGFβ1 downregulates CFTR expression by a mechanism that involves Dab2 mediated signal transduction to Smad2 and other downstream elements in the TGF β1 signaling cascade. We anticipate further studies will increase our understanding of the role of TGFβ1 in CF and its role as an important genetic modifier, having relevance for further clinical implications in individuals with CF.

(Supported by CFF Clinical Fellowship SNODGR08B0 to SMS, NIH R01DK075309 to DRM, NIH R01HL090767 and 3R01HL090767-02S1 to ASU.) establish the role of membrane and lipid-raft CFTR-dependent ceramide signaling in pathogenesis of CF lung disease.

Methods: Age matched C57BL/6 (WT) and Cftrtm1Unc-Tg(FABPCFTR) (Cftr-/-) mice were treated intra-tracheally with Pseudomonas aeruginosa lipopolysaccharide (Pa-LPS, 20 µg/mouse), fumonisin (FB1, 50 µg/mouse) and/or Amitriptyline (AMT, 50 µg/mouse). ELISA and immunoblotting were used to quantify the changes in cytokine, neutrophil activity and protein levels. For the in vitro studies, CFBE41o-and WT-CFTR CFBE41o-cells were treated with Pa-LPS (10 ng/ml), TNFα (10 ng/ml), methyl-β-cyclodextrin (CD, 5 mM) and/or FB1 (50 µM). The inflammatory markers (NFκB and NIMP-R14), CFTR, apoptotic marker (Fas/CD95), and raft proteins were localized and quantified using immunoflurorescence staining and flow cytometry, respectively. The changes in raft proteins expression was verified by immunoblotting after raft-enrichment.

Results: Cftr-/-mice lungs and inflammatory cells have hyper-inflammatory phenotype compared to the WT mice as seen by significantly (p<0.05) elevated pro-inflammatory cytokines. We also found significant increase in basal and Pa-LPS induced inflammatory (neutrophil/NFκB) and apoptotic (Fas/CD95) signaling in Cftr-/-mice lungs compared to the WTs. We show for the first time that the lungs and CD4+ splenocytes of Cftr-/mice have a constitutive increase in forkhead-box-P3 (FoxP3) expression compared to the WT indicative of counteractive protective mechanism. Moreover, CFBE41o-cells have inherently high levels of ceramide that is downregulated in the presence of stable WT-CFTR expression. Our data shows that membrane/raft CFTR expression and ceramide inhibition depletes expression of tight junction protein, . In the presence of WT-CFTR, inhibition of de novo ceramide synthesis by FB1 inhibits TNFα induced NFκB reporter activity (p<0.05). Similarly, Pa-LPS induced NFκB activation and recruitment of macrophages and neutrophils in the lungs of WT mice was controlled by FB1 treatment while its inhibitory effect was significantly lower in Cftr-/-mice indicating that WT-CFTR may be depleting NFkB activity (p<0.05) by negative regulation of TNF-R1 or sphingomyelin. In contrast, inhibition of membrane acid-sphingomyelinase (ASMase) by AMT showed an enhanced protective effect in controlling the Pa-LPS induced lung disease in Cftr-/-mice (p<0.05) as compared to the WT further confirming that CFTR negatively regulates lipid-raft ceramide levels.

Conclusion: We demonstrate that CFTR regulates tight junction formation, membrane ceramide accumulation, and apoptotic and inflammatory signaling in cystic fibrosis. Moreover, our data demonstrates the efficacy of membrane ceramide inhibitor, AMT in controlling Pa-LPS induced CF lung disease. Support: NIH (R025-CR07, R03HL096931), CFF & FAMRI (NV). Contact: nvij1@jhmi.edu

The cystic fibrosis transmembrane conductance regulator protein (CFTR) is a large, multispanning membrane glycoprotein. Malfunction or loss of CFTR at the plasma membrane disrupts chloride transport across epithelia, resulting in the lethal genetic disease cystic fibrosis (CF). CFTR contains two transmembrane spanning domains (TMDs), two nucleotide binding domains (NBDs), and one regulatory region (R) that are translated in the order: TMD1-NBD1-R-TMD2-NBD2. Each domain folds cotranslationally, and can then associate with previously folded domains to form the final, functional CFTR structure. The most common CF mutation, ∆F508, and other CF mutations, disrupt multiple steps in this folding pathway. Translation of individual domains provides insight into cotranslational formation of CFTR folding intermediates. We have used in vitro translation to probe folding and association of TMD1 and NBD1 using both wheat germ and rabbit reticulocyte lysate systems. System specific differences in domain translation were observed that highlight the relationship between TMD1 and NBD1 translation. Specific effects of CF-causing mutants, particularly ∆F508, are being examined in these systems.

Supported by the NIH and the CF Foundation.

mutants of NDPK-A immunoprecipitated from HEK-293T cell lysates. We found that wild-type and S122A mutant NDPK-A showed significant autophosphorylation in the absence of AMPK, which was enhanced by AMPK. However, the H118F mutant showed very little 32 P labeling with or without AMPK present. These results suggest that AMPK-dependent phosphorylation of NDPK-A requires NDPK-A catalytic function and occurs at a site distinct from Ser-122. To test the potential requirement of NDPK-A in the regulation of CFTR by AMPK, either the H118F mutant or wild-type NDPK-A were co-expressed with CFTR in Xenopus oocytes, and two-electrode voltage clamp (TEV) studies were performed. Oocytes were injected with either K-gluconate (control) or ZMP (AMPK activator) 2-4 h prior to recording, and CFTR conductance was measured after stimulation with 1 µM forskolin and 100 µM IBMX. AMPK activation significantly inhibited CFTR conductance in the presence of wild-type NDPK-A, but this inhibition was blocked with expression of the NDPK-A H118F mutant. This result suggests that functional NDPK-A is required for the regulation of CFTR by AMPK. To test whether AMPK binding to CFTR is functionally important for the regulation, we generated a purified peptide corresponding to the AMPK-interacting region in the CFTR tail (CFTR-1420 (CFTR- -1457 . This peptide, when injected into CFTR-and AMPK-α1-expressing Xenopus oocytes or HEK-293T cells, inhibited the binding of AMPK to CFTR as detected by co-immunoprecipitation and GST pull-down assays. Electrophysiological experiments to test whether this blocking peptide interferes with the ability of AMPK to regulate CFTR are underway. In conclusion, our studies highlight the importance of NDPK-A in the AMPK-dependent regulation of CFTR and support the substrate channeling hypothesis. Further elucidation of the mechanisms of NDPK-A and AMPK interregulation and their role in the regulation of CFTR could suggest novel therapeutic targets for CF. (Supported by NIH, CFF, and Wellcome Trust.)

In human sweat duct, CFTR can be activated by cAMP as well as cytosolic glutamate by phosphorylation dependent and phosphorylation independent mechanisms, respectively (Reddy and Quinton, 2003) . We have previously shown that both luminal and cytosolic monovalent cation composition (MCC) changes significantly during transepithelial salt absorption, which seems to alter phosphorylation activation of CFTR (Reddy and Quinton, 2006) . The objective of the present investigation is to further characterize the role of luminal and cytosolic MCC on cAMP and glutamate dependent regulation of CFTR in the native human sweat duct. After α-toxin permeabilization of the basolateral membrane as previously described (Reddy and Quinton, 2003) , we measured the effect of changing Na + and K + concentrations in the cytosol and in the lumen on the apical CFTR-Clconductance (CFTR-gCl). We found that in the cytosol, substituting K + with Na + completely inhibited cAMP-activated CFTR-gCl regardless of the luminal MCC. In contrast, the effect of substituting cytosolic Na + for K + on glutamate activation of CFTR-gCl is a function of luminal MCC. That is, glutamate-activated CFTR-gCl is stimulated by the presence of luminal Na + and inhibited by its absence (K + substitution). Preliminary results suggested that this inhibition of glutamate-activated CFTR-gCl was due to lowered luminal Na + and not due to increased [K + ] since luminal Na + substitution with Li + or NMDG had similar inhibitory effects.

We conclude that a) changes in MCC in the lumen and cytosol differentially control cAMP and glutamate regulation of CFTR, and b) changes in cytosolic [K + ] appear to control phosphorylation activation while changes in luminal Na + appear to control glutamate activation of CFTR activity.

The cystic fibrosis (CF) phenotype is species dependent, exemplified by CF gallbladder morphology (severe in pigs, but absent in mice). As underlying mechanism, species could differ in their functional activity of alternative chloride channels, such as calcium induced chloride channels. We tested the hypothesis that the more severe CF gallbladder phenotype in pigs compared with mice corresponds with a lower capacity of calcium induced chloride transport in the former.

Methods: In Ussing chambers, we measured trans-epithelial anion currents (representing Cl -+ HCO 3 -) in gallbladder samples from CF pigs and mice (both homozygous ∆F508 and CFTR knockout) and in their respective controls. CFTR dependent anion secretion was assessed by means of the cAMP agonist forskolin. Calcium dependent anion secretion was quantified using the muscarinic agonist carbachol. Values are expressed as median (range).

Results: Forskolin induced anion secretion was quantitatively similar in pigs and mice; both in wild type resp.) and in the two CF conditions (∆F508: delta Isc 38 (18-52) versus 17 (8-74) resp.) (knockout: delta Isc 11 (10-12) versus 22 resp.) (µAmp/cm 2 ). In contrast, carbachol induced anion secretion was considerably lower in pigs compared to mice, both in wild type and in the two CF conditions (figure) .

Conclusion: The absence of a morphological gallbladder phenotype in mice, compared with pigs, corresponds with a prominent capacity for calcium induced anion secretion. Our data support the concept that stimulation of calcium induced anion secretion could be a therapeutic strategy to prevent or mitigate the CF phenotype. CFTR bearing F508del, the most common CF-causing mutation, is retained intracellularly at the endoplasmic reticulum (ER) due to ineffective folding and sent to premature degradation through the ubiquitin-proteasome pathway (UPP). Accordingly, most mutant protein fails to reach its proper location at the plasma membrane. Several approaches have been used in the recent years to rescue the misfolding and trafficking defects of the F508del-CFTR mutant and to understand the underlying mechanisms. These strategies either improve folding or circumvent the ER quality control (ERQC) that targets F508del-CFTR to UPP, or both. They include: 1) low temperature incubation (26°C); 2) presence of genetic revertants (second-site mutations that either remove ERQC retention signals or stabilize its folded conformation); and 3) small molecules which either interact directly with F508del-CFTR to favour correct folding and/or indirectly promote its membrane trafficking. Understanding how F508del-CFTR is rescued to the cell surface by distinct agents enables us to mimic such effects by small molecules. Moreover, identifying distinct rescuing mechanisms will possibly allow usage of different therapeutic molecules for enhanced results.

Our aim is to assess the synergistic/additive effects of F508del-CFTR correctors with those of several genetic revertants of F508del-CFTR and of low temperature incubation to learn more about their mechanism of action (MoA). As an example, we tested here the VRT-325 small molecule.

To this end, the following F508del-CFTR revertants (stably expressed in BHK cells) were tested alone: V510D; I539T; G550E; R553M; R553Q; R555K; R1070W; or in combination: 4RK; G550E/R553Q; R553M/R555K; R553Q/R555K; 4RK/G550E; F494N/Q637R; and F429S/F494N/Q637R. Western Blot was performed after 26°C incubation (48h) and/or treatment with VRT-325 (6.7 µM; 24h). The percentage of band C vs. the total of CFTR expressed for each variant was calculated after densitometry and data normalized to wt-CFTR. Our results show that: i) I539T and R555K are the most effective single genetic revertants; ii) VRT-325 is less efficient than low temperature in rescuing F508del-CFTR; iii) low temperature additivity is evident for almost all revertants tested, being most evident for G550E and R1070W; iv) VRT-325 exhibits highest additiv- The cystic fibrosis transmembrane conductance regulator (CFTR) is a cAMP/PKA-regulated chloride channel whose phosphorylation controls both channel gating and its trafficking at the plasma membrane. The 14-3-3 proteins are a family of conserved regulatory molecules that generally interact with phosphopeptide consensus sequences (e.g., the canonical motifs, RSXpSXP and RXXXpSXP) within their protein targets, although binding at varied related sequences is observed. Several potential non-canonical 14-3-3 binding motifs appear in the CFTR sequence, particularly in the R domain. Therefore, we are evaluating the hypothesis that 14-3-3 proteins participate in the regulation of CFTR through association with its phosphorylated binding motifs. First, we found that the 14-3-3 β, γ and ε isoforms are expressed in Calu-3 cells and HEK293 cells using RT-PCR and immunoblot analysis. Physical interactions between CFTR and the β, γ and ε isoforms of 14-3-3 were identified using co-IP experiments. Using in vitro pairwise binding studies, we found that 14-3-3β interacted significantly with the CFTR N-terminus and R-domain. The dependence of 14-3-3 binding on CFTR phosphorylation was examined by treating HEK293 cells for 15 min with forskolin (10 µM), followed by 14-3-3β/CFTR co-IP; the results showed that forskolin increased the interaction between 14-3-3β and CFTR. In co-IP experiments, we used extracts from HEK293 cells transiently transfected with CFTR. 14-3-3β was found to associate with both the mature (band C) and immature (band B) forms of CFTR, and this interaction was significantly increased by prior forskolin treatment. Over-expression of 14-3-3β in HEK293 cells increased the expression of CFTR (bands B and C). We employed 14-3-3β targeted siRNAs to knock down 14-3-3β in HEK293 cells transfected with or without CFTR. Decreased 14-3-3β reduced the expression of both immature and mature CFTR (bands B and C) by ~90% and ~65%, respectively. We also found that altered 14-3-3 protein expression produces parallel changes in CFTR channel function in mammalian cells, as detected by SPQ fluorescence dequenching when bath solution is Misfolding and degradation of CFTR is the cause of disease in patients with the most prevalent CFTR mutation, deletion of phenylalanine 508 (F508del), which is located in the first nucleotide-binding domain of human CFTR (hNBD1). While (F508del) CFTR cellular folding is impaired, the structure and chemical unfolding transition of isolated (F508del)hNBD1 is not. We therefore studied the in vitro thermal unfolding of hNBD1 constructs ±F508del to shed light on the defective folding mechanism and thermal instability of (F508del)CFTR. Differential scanning calorimetry (DSC) and circular dichroism were used to study 7 sequence-matched hNBD1 pairs (with and without the F508del mutation), some pairs of which also contained suppressor mutations. The results show that regardless of the sequence, the F508del uniformly lowers the unfolding transition temperature (Tm) by 6-7 o C. A thermal unfolding mechanism derived from nonlinear least squares fitting of comprehensive DSC data sets indicates that hNBD1 unfolds as an isolated monomer. All data are consistent with a simple three-state thermal unfolding mechanism for hNBD1±F508del: N(±MgATP) Ä I T (±MgATP) Ç A T --> (A T ) n . The partially unfolded intermediate, I T , is in equilibrium with native hNBD1 and further unfolding leads to the formation of species A T , via a kinetically-controlled, irreversible transition. A T is a partially folded, aggregation-prone monomer. Its formation represents the rate-determining step leading to the aggregated product, (A T ) n . Fitted parameters indicate that F508del thermodynamically destabilizes the native state, N, i.e., ∆∆G o = 1.1 kcal/mol. Furthermore, the F508del mutation accelerates the formation of A T and aggregation of hNBD1. In summary, the studies reported here represent the first thermodynamic description of hNBD1 thermal unfolding. Temperature rescue of the defective in vivo processing of (F508del)CFTR is likely due, in part, to the inhibition of this unfolding pathway of the hNBD1 domain. It is possible that "solubilizing" mutations act by inhibiting formation of A T , rather than altering the physical properties of the folded domain; i.e., aggregation is initiated from the partially unfolded A T state, not the folded state, and solubilizing mutations act by inhibiting the formation of A T . Further, the inclusion of high MgATP concentration or osmolytes like glycerol and other so-called "stabilizing" additives lead to improved solubility by inhibiting the formation of A T .

Financial support from the CFF and the technical assistance of Jianli An is gratefully acknowledged.

Fisher, J.T. 1, 2 ; Liu, X. 1,2 ; Luo, M. 1 ; Lukacs, G. 3 ; Engelhardt, J.F. 1,2 1. Anatomy and Cell Biology, University of Iowa, Iowa City, IA, USA; 2. Center for Gene Therapy, University of Iowa, Iowa City, IA, USA; 3. Physiology, McGill University, Montreal, QC, Canada Essential to development of effective therapies for cystic fibrosis (CF) are animal models that recapitulate the human disease phenotype. To this end, ferret and pig CF models have been developed and their phenotypes are now being characterized. Differences in CFTR biology have been reported to exist across species. For example, murine and porcine ∆F508-CFTR proteins are more effectively processed to the apical membrane of airway epithelia than human ∆F508-CFTR. Understanding such species-specific differences is imperative to evaluation of CF animal models, developing and testing therapeutic agents, and elucidating mechanisms of CFTR processing, function, and regulation. In the present study we have compared protein changed from iodide to nitrate buffer. Based on these findings, we conclude that CFTR interacts physically with the 14-3-3β, and perhaps other isoforms and it associates with both mature and immature CFTR in vivo. Their interaction was increased significantly by cAMP/PKA stimulation and 14-3-3β directly interacted with CFTR's N-terminus and R-domain. We also propose that 14-3-3 protein interactions with CFTR regulate its biogenesis, providing what may be a late checkpoint in CFTR quality control.

[Supported by NIH DK054814 and DK072506 and the CF Foundation.] processing and functional differences between human and ferret wild type and ∆F508 CFTR. To this end, wild type and ∆F508 CFTR proteins, with and without an extracellular HA-tag in ECL-4, were used to study the processing, maturation, membrane stability, and function of CFTR across species. Heterologous expression of wild type CFTR in human cell lines (HT1080 and HEK293) demonstrate elevated total fCFTR protein levels relative to hCFTR by Western blot and in vitro phosphorylation assays. By 35 S-methionine metabolic pulse chase, we demonstrate that significant differences exist in the membrane stability of the fully processed band C forms of fCFTR and hCFTR (fCFTR is several fold more stable than hCFTR). No significant differences were observed in the initial amount of incorporated label, band B stability, and rate of band B to C conversion between the CFTR orthologs. The observation of increased fCFTR membrane stability relative to hCFTR has also been characterized by surface labeling HA immunofluorescence (IF) in non-permeabilized BHK21 cells. Expressing the ∆F508-CFTR orthologs in HT1080 and HEK293 cells results in some band C formation in the ferret protein with little or no processing of the human mutant, as determined by in vitro phosphorylation assays and metabolic pulse chase experiments. This data is similar to that reported in the literature of pig and mouse ∆F508-CFTR proteins suggesting that the ∆F508 band B processing block may be specific to human CFTR. Despite some processing of the ∆F508-ferret band B to C, surface HA IF staining of nonpermeabilized BHK21 cells was not observed relative to its wild type counterpart. Furthermore, ∆F508-fCFTR produced no detectable Clcurrents when expressed in polarized human airway epithelia. Comparative electrophysiologic and immunolocalization studies in ferret CFTR null tracheal xenografts transduced with human or ferret CFTR (wild type and ∆F508) are ongoing to determine if the species of the cell influences ∆F508 processing. Understanding the molecular mechanisms responsible for increased membrane stability of fCFTR and the species differences in ∆F508 processing will enhance our understanding of the CFTR biosynthetic pathway, lead to additional therapeutic targets, and aid in the interpretation of the phenotypes seen in the newly generated pig and ferret CF animal models. Cystic fibrosis (CF) is one of the most common lethal genetic diseases worldwide that is a consequence of failure of the proteostasis network (PN) to promote proper protein folding. CF is caused by mutations in the apical chloride channel CFTR, leading to impairment of function and disease. The majority of patients carry at least one allele of the ∆F508 mutation. ∆F508 CFTR is characterized by a folding/trafficking defect in which it fails to exit the endoplasmic reticulum (ER) and is subsequently targeted for proteasomal degradation. We have previously shown that wild type (WT) and ∆F508 CFTR interact with distinct cellular partners during CFTR biogenesis, and that the ∆F508 is found in complex with Hsp70/Hsp90 and its co-chaperones, suggesting that it accumulates in on-pathway folding intermediates in the ER. Molecular chaperones and co-chaperones found in the PN control cell homeostasis either promoting folding or degradation of misfolded proteins. However the regulatory mechanisms in the PN that direct proteins towards folding or degradation and the precise role of the co-chaperones involved in such processes are not well understood. Herein we have examined the role of co-chaperones in ∆F508 CFTR stability and trafficking, using human bronchial epithelial cells expressing either WT or ∆F508 CFTR. We now demonstrate that modulation of the PN can significantly promote ∆F508 CFTR stability/trafficking and a concomitant increase in channel activity, suggesting that co-chaperones can redirect ∆F508 CFTR biogenesis to a more functional state. Our results emphasize the importance of the PN in CF therapeutics through protein folding and trafficking pathways essential for its correction and proper function at the cell membrane.

Zegarra-Moran, O.L.; Melani, R.; Galietta, L.J. Lab. Genetica Molecolare, Istituto Giannina Gaslini, Genoa, Italy Cystic fibrosis transmembrane conductance regulator (CFTR) is a cAMP-activated Clchannel whose gating is controlled by the binding and hydrolysis of ATP at two nucleotide binding domains (NBD1 and NBD2). CFTR activity is pharmacologically modulated by potentiators, small molecules having dual effect. At lower concentration they increase the activity of CFTR, whereas at higher concentrations they inhibit it. The binding site for potentiators is hypothesized to be located in the interface of the dimer formed by the two nucleotide binding domains, interacting mostly with NBD1 and to a lesser extent with NBD2. From previous studies, the binding site seems to involve protonable residues, cysteines and histidines. We have reasoned that if the binding of potentiators to the NBDs relies on electrostatic interaction with any of these amino acids, then modifications of cytosolic pH would alter the binding affinity. To test this hypothesis, we have analyzed the effect of intracellular pH (pHi) on apical membrane currents of Fisher Rat Thyroid cells stably expressing CFTR. Currents were activated by using the membrane permeable cAMP analogue CPTcAMP. To study the role of pHi on potentiators binding we have employed genistein, a classical CFTR potentiator. We found that the activity of CFTR was sensitive to shifts in pHi, with the current significantly lower at pH 6. Also the concentration of CPTcAMP necessary to obtain half of the maximal current diminished at pHi 6. Addition of increasing concentration of genistein to partially phosphorylated channels showed that at pHi 8 the apparent dissociation constant from the activating site was shifted towards higher values relative to pH 7.3, with no effect on this parameter at pH 6. The affinity of genistein for the inhibitory site was instead less sensitive to pHi. These results indicate that cysteines, but not histidines, are involved in the activating binding site. Accordingly, we mutated to alanine two cysteine residues: Cys491 in NBD1, predicted by molecular docking to make contacts with CFTR potentiators, and Cys1344, located in NBD2 outside the putative binding site. We found that the mutation C491A, but not C1344A, abolishes the effect of alkaline pH on genistein binding. Our results confirm and extend the modulatory effect of pHi on CFTR activity, and demonstrate that binding of potentiators to CFTR depends on electrostatic interactions with amino acid residues of NBD1 including Cys491.

ent channel gating without altering protein processing. To address our aim, we studied CFTR Clchannels in inside-out membrane patches excised from BHK cells expressing wild-type (wt), G551D-, G550E-G551D-and 4RK-G551D-CFTR. The pipette (external) solution contained 10 mM Cland the bath (internal) solution contained 147 mM Cl -, 1 mM ATP and 75 nM PKA at 37°C; voltage was -50 mV. When compared with wt-CFTR, the gating behaviour of G551D-CFTR is characterised by infrequent channel openings separated by very prolonged channel closures (wt, open probability (P o ) = 0.47 ± 0.04, mean burst duration (MBD) = 149 ± 14 ms, interburst interval (IBI) = 166 ± 26 ms, means ± SEM, n = 7; G551D, P o = 0.007 ± 0.001, MBD = 39 ± 7 ms, IBI = 5,452 ± 599 ms, n = 8). Although, neither G550E nor 4RK rescued the gating defects of G551D-CFTR, 4RK caused a small, but significant increase in P o (G550E-G551D-, P o = 0.010 ± 0.001, n = 13; 4RK-G551D-, P o = 0.027 ± 0.005, n = 14; p < 0.05, Student's t-test). However, both revertants reduced dramatically the duration of G551D channel openings. Therefore, G550E-and 4RK-G551D-CFTR channels were characterised by frequent, very flickery channel openings (G550E-G551D-, MBD = 27 ± 4 ms, IBI = 2,772 ± 301 ms, n = 13; 4RK-G551D-, MBD = 11 ± 1 ms, IBI = 535 ± 113 ms, n = 14; p < 0.05). Because G551D-CFTR channel gating is ATP-independent, unlike that of wt-CFTR, we investigated whether revertant mutations restore ATP-dependence to G551D-CFTR. G550E had modest effects on the ATP-dependence of G551D-CFTR. By contrast, the ATP-dependence of 4RK-G551D-CFTR was similar to that of wt-CFTR (wt, dissociation constant (K d ) = 210 ± 47 µM, n ≥ 5; 4RK-G551D, K d = 209 ± 108 µM, n ≥ 6). In summary, the revertants G550E and 4RK altered the pattern of CFTR channel gating without rescuing channel activity. However, 4RK, but not G550E, restored the ATP-dependence of G551D-CFTR, suggesting that G550E and 4RK act by different mechanisms. Structural models of the nucleotide-binding domains predict that changes to local electrostatic potentials may partially explain the observed differences in channel gating.

Supported by CF Trust (UK) and FCT (Portugal). JL and ZX were supported by the Overseas Research Studentship (ORS) award and the University of Bristol Scholarship.

CFTR is a unique ABC family chloride channel that controls transepithelial salt and fluid movement across apical cell membranes. Greater than 1700 mutations in CFTR cause CF, which is characterized by chronic lung infection and early death as a result of reduced secretion of both Cland fluids in the airway which causes viscous and acidic airway surface liquid that can be easily colonized by bacteria that are poorly cleared by airway cilia. Mutations can affect both the amount of mature protein expressed on the cell surface and the chloride channel activity of the CFTR protein that does reach the plasma membrane.

Small molecule therapies are being developed to restore CFTR activity in vivo, however our understanding of the direct interaction between the protein and these therapeutics remains limited. The development of a variety of in vitro activity assays will allow us to study the interaction of small molecule therapies with the protein, with the goal of better understanding the mechanism of action of the protein and improving the prognosis for this disease.

CF research is currently limited by the availability of methods for the preparation of highly functional, purified CFTR. We have recently developed a new method for the purification of microgram quantities of functional CFTR protein. The purification and reconstitution procedure is rapid, taking less than one day, and yields highly functional protein as assessed by radioactive ATPase assays. The ATPase specific activity was >4 nmol/µg protein released per hour with a Km of 0.64 mM, comparable to CFTR isolated in more time-consuming purification procedures. The protein also possesses single channel conductance and PKA and MgATP-dependent gating typical of the CFTR protein. Using this purification method, we are developing new assays for the assessment of the molecular mechanisms underlying CFTR activity, the defects as a result of disease-causing mutations and the basis for small molecule activity. Our novel ELISA-based assay allows detection and quantitation of sub-nanomolar concentrations of CFTR in a sample. We are developing a sensitive fluorescence-based non-radioactive ATPase activity assay suitable for medium throughput of purified CFTR samples, to study their interaction with small molecule potentiators. We are also developing an iodide flux assay to measure the regulated chloride channel activity of purified CFTR protein reconstituted into PC liposomes. We can measure significant CFTR-dependent efflux over a period of several minutes at protein:lipid ratios of greater than 1:2000 to 1:200 (w/w), with as little as 1 µg of CFTR protein per assay. We believe these assays will have great utility in the elucidation of the direct interaction of CFTR with potential small molecule therapeutics and their effects on the channel and enzymatic activities of the protein.

This work is supported by the Canadian Cystic Fibrosis Foundation (CCFF) and the Canadian Institutes of Health Research (CIHR). P.D.W.E. was supported by a fellowship from the CCFF. phosphorylation in CF IB3-1 cells (p-values of 0.004 to 0.02). These phosphatases are: Cyclin-dependent kinase inhibitor 3, [CDKN3], Dual specificity protein phosphatase 4, [DUSP4], Tyrosine phosphatase zeta polypeptide 2 [HTPZP2] and Receptor-type tyrosine-protein phosphatase F, [PTPRF] .

Interpretation and Conclusion: We interpret these data to indicate that the CF mutation has profound effects on the dynamics of the phospho-proteome signaling, and conclude that the candidate phosphatases may contribute to the proinflammatory defect caused by mutant CFTR.

described. In addition, CFTR has been implicated in the transport of sphingosine-1 phosphate (S1P), a vascular barrier-enhancing sphingolipid (Boujaoude et al, 2001) , and in the organization of tight junctions between epithelial cells (LeSimple et al, 2010) .

Objective: We hypothesized that an impairment of CFTR function would decrease endothelial cell barrier function by alteration in cytoskeletal organization of actin and adherens junctional proteins, causing intercellular gaps that could then be restored by S1P supplementation.

Methods: The barrier of primary rat or human pulmonary endothelial cells or sheep bronchial artery endothelial cell monolayers was measured using Electric Cell Substrate Impedance Sensing (ECIS) in the presence of CFTR inhibitors GlyH-101 (1-10 µM) and or their respective vehicle controls. Cell morphology, actin cytoskeleton, and junctional proteins were studied in rat and human endothelial cells monolayers by fluorescence microscopy using Texas Red phalloidin and immunocytochemistry.

Results: There was a dose-dependent increase in pulmonary and bronchial artery endothelial monolayer permeability in response to treatment with specific CFTR inhibitors GlyH-101 or CFTRinh-172, which decreased the TER by up to 40% and 50%, respectively (p<0.05) reaching the maximum rate of decline at 4 hours after CFTR inhibition. Treatment with CFTR inhibitors was associated with both inter-cellular gaps and actin stress fiber formation compared to vehicle-treated cells. The addition of S1P (1 µM) significantly attenuated the decrease in TER in response to CFTR inhibitors, while decreasing actin stress fiber formation.

Conclusions: CFTR may function in endothelium to maintain its barrier properties, including adherence junction stability. The mechanism by which this chloride and ATP binding cassette-transporter channel regulates vascular permeability may involve sphingolipid transport regulation and availability, since the barrier dysfunction induced by CFTR inhibitors could be restored by S1P treatment.

Funding: R01 HL 077328.

The most common cystic fibrosis (CF)-causing mutation in the cystic fibrosis transmembrane conductance regulator (CFTR) is deletion of Phe508 (F508del) in the first of two nucleotide binding domains (NBDs). The F508del mutation leads to defects in cellular processing and folding, as well as channel gating. As a potential therapy for CF, small-molecule compounds have been identified that correct the processing/folding defects (correctors) and/or channel gating defects (potentiators) of F508del. Compounds having dual corrector/potentiator activity have been assumed to interact directly with CFTR, possibly at the NBD1, the site of the F508del mutation. Our goal is to determine how these small molecule modulators act. We report NMR studies of the interactions of a representative dual corrector/potentiator with human F508del NBD1 and have mapped the compound-affected residues onto the structure of NBD1. In order to obtain appropriate solubility and yields required for NMR, our constructs lack the regulatory insert (∆RI) and regulatory extension (∆RE) normally contained within NBD1. The compounds interact with a hydrophobic patch comprising residues in β-strands S3, S9 and S10, located within the ATP-binding core near the C-terminus of the molecule. Binding of the compounds disrupts an interaction between the C-terminal helices (H8,9) of NBD1 and this hydrophobic patch leading to a shift in a pre-existing conformational equilibrium from helix to coil. Disruption of the interaction between Helix 8, 9 and the NBD1 core by a mutation within β-strand S9, H620Q, shown to increase the open probability of the channel (Wei et al. FEBS Letters 439(1998) :121), partially mimics the effect of the compounds by promoting the coil state. A comparison of Helix 8 and 9 between F508del and wild-type show that the wild-type favours the coil conformation, suggesting that perturbations that shift the helices toward the coil state coincide with conditions that promote channel opening. Addition of compound to constructs in which Helix 9 has been deleted indicate a broader area of interaction within the hydrophobic patch. We hypothesize that the compounds compete with the The non-hydrolytic N-terminal nucleotide binding domain (NBD1) of CFTR contains a segment of ~ 30 amino acids in length near its N-terminus not present in NBD2 or NBDs of other ABC proteins. Structural studies (Lewis et al, EMBO J 23, 282-93, 2004; Kanelis et al, EMBO J 29, 263-77, 2010) indicate that this insertion (RI) is not highly ordered and strongly affects the thermal stability of the NBD1 folded state (Atwell et al., Protein Eng Des Sel. 23, 375-384, 2010) . Although RI contains a site phosphorylated by protein kinase A, its removal left the channel activity largely unaltered (Csanady et al, J Gen Physiol 125, 171-86, 2005 ).

Here we tested the possible role of RI in the thermal instability of rescued ∆F508 CFTR, which is reflected in its short life-time in cells (Sharma et al., J.Biol Chem. 276, 8942-8950, 2001 ) and functional run down (Hegedus et al, Biochim Biophys Acta 1758 , 565-72, 2006 . We found that deletion of all, but not portions, of RI restored ∆F508 CFTR maturation and traffic to the cell surface where it had a life-time similar to the wild-type and was fully functional. RI deletion restored the ability of ∆F508 CFTR to occlude ATP at NBD1. Most striking was the observation that robust and stable channel gating was maintained by ∆RI/∆F508 CFTR for hours at physiological temperature where channel activity of temperature and/or corrector rescued ∆F508 CFTR was lost in minutes. Thus we assume that the inclusion of RI in NBD1, while normally contributing to some aspect of CFTR structural dynamics also promotes thermodynamic instability, an effect that is amplified by the presence of the ∆F508 mutation. One may speculate that it might be possible to mimic the stabilizing effect of RI removal by its restriction to one particular conformation.

(Supported by the NIDDK of NIH and CFF.) Current model systems to bridge preclinical studies of CFTR modulators to clinical trials are limited to studies in cell lines, polarized primary CF airway epithelia, and transgenic CF mice. Each of these model systems has well described advantages and limitations, including variable behavior of mutant CFTR in heterologous expression systems, scarcity of primary human airway cells from CF patients with specific genotypes, and symptoms in CF mice that do not fully recapitulate disease manifestations in CF patients. Examination of CFTR outcome measures in human tissue (including rectal tissue) may complement the current model systems for several reasons, including high CFTR expression, ease of tissue acquisition, minimal organ damage, and several endpoints to monitor modulator effects on CFTR ex vivo (such as intestinal current measurements -ICM, Western blot -WB, and real time RT-PCR). The goal of the current study is to optimize rectal tissue survival and CFTR detection, characterizing CFTRdependent Clconductance, CFTR protein and RNA levels over time. After obtaining informed consent, biopsied rectal tissue from non-CF and CF subjects (Olympus "endojaw") was placed immediately in iced RPMI-1640 media (+25 mM HCO 3 -, pH 7.0), dissected to remove the muscularis mucosa, and mounted in recirculating Ussing chambers (Physiologic Instruments, San Diego, CA). Tissue was bubbled with 95%O 2 :5%CO 2 in RPMI-1640 + HCO 3 - at 37 o C. Isc was measured under voltage clamp conditions (Physiologic Instruments, San Diego, CA), and tissues were exposed to a standardized series of reagents to isolate and quantify CFTR activity [10 µM indomethacin x 30 min, 10 µM amiloride (mucosal), 10 µM forskolin/100 C-terminal helices (H8, 9) for the core of NBD1 resulting in a destabilization that displaces the RE/R region from the ATP-binding core and exposing the NBD dimerization interface, thereby promoting NBD dimerization and channel opening, accounting for the potentiating effect of the compound. Data for related compounds suggest that the primary corrector effect is not via a CFTR-dependent mechanism, but the compound may contribute to the corrector effect in one of two possible ways, one by binding to the core of NBD1 and shifting the equilibrium to a folded state, the other by promoting NBD dimerization that could stabilize the folding of full-length CFTR.

This work was funded by the CF Foundation.

µM IBMX (both compartments), 100 µM carbachol (basolateral), and 100 µM bumetanide (basolateral)]. Tissues were then rinsed, replaced in chambers (+100 µg/ml penicillin, 20 µg/ml ciprofloxacin, and 100 µg/ml gentamicin), and maintained at 37 o C with gas until restudied at 6 and 18-24 hrs later. At the end of the studies, tissues were lysed in RIPA buffer for Western blot (mouse anti-human CFTR 570 and 596 Abs), or placed in RNA later for subsequent real time RT-PCR. A total of 168 biopsies from 40 subjects were studied across all three study sites. 86.3% of biopsies provided interpretable data (145/168 biopsies -39/40 subjects) at t=0. For normal biopsies with repeated measures (n= 11-19 samples), 100%, 89.5%, and 91.7% of biopsies had measurable CFTR conductance at 4, 18, and 24 hrs, respectively (including cAMP and/or CCh-stimulated Isc that was sensitive to bumetanide blockade). While biopsies studied in Ringer's and RPMI buffer demonstrated similar cAMP and CCh stimulated Isc at t = 0, use of Ringer's was associated with reduction in CCh-stimulated Isc of ~50% over 6 hrs.

The results indicate that ex vivo monitoring of CFTR maturation and activity is feasible in rectal biopsy tissue from human subjects, and that this model system may serve an important role in the preclinical evaluation of CFTR-active compounds. Supported by CFFT and KPRI. The two nucleotide binding domains (NBDs) and the R region of the cystic fibrosis transmembrane conductance regulator (CFTR) are responsible for regulating the chloride pore formed by the membrane spanning domains of CFTR. The most common cystic fibrosis causing mutation, deletion of F508 (F508del), is a mutation in the first NBD (NBD1). F508del has deleterious effects on the folding and stability of CFTR and also severely reduces channel function. Unexpectedly, F508del was shown to have a minimal effect on the crystal structures of NBD1. Rather, F508del is thought to disrupt NBD1 surfaces required for interdomain interactions and/or the thermodynamics of NBD1 folding and interactions. We used NMR to measure fast time scale (ns-ps) motions for WT and F508del mutant NBD1 at residue level resolution. These studies identify flexible regions in WT NBD1 and uncover F508del induced differences in fast time scale motions. These changes are not restricted to the vicinity of the mutation, but propagate through the NBD1 structure. Broadening of NMR resonances and NMR dispersion measurements indicate that NBD1 is highly dynamic and that it samples one or more excited state conformations on an intermediate time scale (µs -ms) . Deletion of F508 appears to enhance sampling of higher energy conformations as indicated by increased resonance broadening and chemical shift dispersion. We hypothesize that these higher energy NBD1 conformations lead to unproductive folding intermediates during CFTR processing. Further we posit that the dynamic nature of NBD1 ensures that phosphorylation status and interactions with NBD2, ICLs and nucleotide have effects that propagate through NBD1 connecting these events to enable proper regulation of the channel. Subtle changes in the thermodynamics of these interactions are caused by deletion of F508 and may provide a mechanism for disruption of channel gating.

Dalton, J. 1 ; Villà-Freixa, J. 1 ; Mutlu, S. 1 ; Lalinde, W. 2 ; Hernandez, M. 2 ; Tortajada, J. 2 1. Computational Biochemistry and Biophysics Laboratory, Universitat Pompeu Fabra, Barcelona, Spain; 2. Asociación Madrileña Contra la Fibrosis Quística, O2 Health Link, Inc, Madrid, Spain The standard approach for homology modeling CFTR is to use the experimentally determined 3D structure of prokaryotic Sav1866 as a template. However this only reveals the structure of CFTR in an outward (or open) conformation. More recently, an attempt has been made to model CFTR in an inward (or closed) conformation by using the low-resolution 3D structure of prokaryotic MsbA as a template, although this requires significant template readjustment to provide a fully closed state (Mornon et al., 2009) . Here, we model the full length closed conformation of CFTR on the higher resolution eukaryotic 3D structure of P-glycoprotein, as well as the 3D structures of CFTR nucleotide binding domains: NBD1 and NBD2. This results in the NBDs adopting a fully separated state and the membranespanning domains (MSDs) significantly rearranged in relation to the open conformation. However, despite the large conformational shift involved, the connections between the NBDs and MSDs of CFTR remain principally unaltered, although flexibility in critical side-chains is apparent. In addition, one half of the regulatory domain of CFTR has been modeled upon a prokaryotic ABC-transporter TOBE domain, which despite very low sequence similarity, shows similar secondary structure to the regulatory region of CFTR, predicted by NMR (Baker et al., 2007) . The second half of the regulatory region has been modeled ab initio, to permit biologically feasible movement between different conformations, whilst also taking into account relevant amphipathic helical elements.

The open and closed conformations of CFTR, modeled within the context of the cellular epithelial membrane, have been integrated into a novel 3D-graphical platform that is part of the ActivA project: an online virtual world for visualizing all biological aspects of cystic fibrosis.

References We have established a simple, non-radioactive assay for monitoring enzymatic activity of phosphodiesterases relevant to CFTR activation, including PDE1 and PDE4. This assay tracks cAMP metabolism via malachite green and can be used to quantify effects of various small-molecules on PDEs, including identification of novel pharmacologic inhibitors. We investigated CFTR potentiators from the CFFT modulator library identified previously by high-throughput compound library screening. Our findings demonstrate that none of the available compounds tested to date (including VRT-532, UCCF-152, and UCCF-029) inhibit either PDE1 or PDE4, two phosphodiesterases potentially relevant to CFTR regulation within the local plasma membrane environment. These data are congruent with earlier findings that neither VRT-532, UCCF-152, or UCCF-029 increase cellular cAMP. In particular, UCCF-152 was previously demonstrated to activate CFTR, and phosphorylate isolated regulatory domain (R-D), but not augment PP2A activity (a phosphatase important for CFTR modulation through local cAMP regulation). We then validated the CFTR R-D phosphorylation assay for two CFTR activating mechanisms involving cAMP. In this assay, an HA-tagged regulatory domain of CFTR (amino acids 635-836) was expressed in HEK293 cells and phosphorylation monitored by a gel shift of 2-3 kDa. β2-selective adrenergic agonists such as salmeterol and salbuterol bind to the β2-adrenergic receptor and thereby augment adenylyl cyclase. These compounds, as well as selective PDE4 inhibitors such as rolipram (also a CFTR activator), were found to enhance R-D phosphorylation. For example, rolipram stimulated CFTR phosphorylation with an IC50 of approximately 5 nM, strongly suggesting that effects of this compound are mediated by selective inhibition of PDE4, rather than blocking other members of the PDE superfamily. Immunoblotting of epithelia used in the assay demonstrated two known PDE4D isoforms, as well as several other PDE4s that await further characterization. The results are concordant with high level PDE4 expression noted previously by our laboratory (and others) in airway epithelial cells, and help clarify PDE4D subtypes responsible for effects on CFTR gating. In summary, these experiments provide a means by which novel CFTR potentiators can be efficiently tested for their effects on PDEs, R-D phosphorylation, and CFTR function. The experiments also implicate PDE4D as a specific phosphodesterase class that enhances CFTR function in airway and other epithelia.

involves an abrupt compaction of a small ATP-binding subdomain comprising residues 378-495, 3) ATP acts as a ligand to facilitate NBD1 folding. Based on these results, we have focused efforts on identifying how the F508del mutation affects co-translational NBD1 folding. Because Phe508 is located in the alpha-helical subdomain (residue 501-564) and has not yet emerged from the ribosome when the N-terminal subdomain has folded, FRET was measured in full length NBD1 (truncated at residue 744) containing the acceptor dye at residues 487 and 567. Preliminary data indicate that F508del NBD1 domain folds less efficiently than wild-type as evidenced by a lower FRET in ribosome bound nascent chains. This suggests that F508del affects primarily alpha-helical subdomain folding and/or formation of the F1-type ATPase beta-sheet core. Further studies are underway to confirm these findings and compare both the folding efficiency and the kinetics of wild-type and F508del-subdomains. It is anticipated that these results will define the impact of F508del on co-translational NBD1 folding and provide a means to assess small molecule correction of this basic defect.

Supported by the NIH and CFFT. Gating of the CFTR chloride channel is regulated by ATP-binding and hydrolysis by the cytoplasmic nucleotide binding domains (NBD1 and NBD2). Nucleotide binding and hydrolysis are mediated by multiple canonical sequences, including the Walker A and B sequences and the ABC-transporter signature motif (LSGGQ). Mutations within these sequences, which putatively disrupt ATP-binding and hydrolysis, and within the putative NBD1-NBD2 dimer interface have been associated with CF. Structures of the human and mouse NBD1 and NBD2 proteins provide structural insight on the isolated domains. However, detailed functional and structural details of the putative CFTR NBD heterodimer have not been fully elucidated in vitro. Confounding our understanding of the NBD dimerization events that regulate channel gating, the canonical ATP-associated sequences are not strictly conserved in CFTR. Changes in the canonical ATP sequences are associated with the NBD1 ATP-binding site (Walker B, catalytic histidine, and LSGGQ). Further characterization of the functional and structural states associated with CFTR NBD dimerization is critical to our understanding of the NBD reaction cycle and its coupling to channel gating. To characterize the interactions between the CFTR NBDs and the effects of these noncanonical NBD sequences, we have purified and characterized NBDs from two highly related ABC-transporters: CFTR (ABCC7) and ABCC6. Native NBD1 and NBD2 proteins from murine CFTR have been expressed and purified for functional analyses. Consistent with prior studies, the isolated, purified NBDs have negligible ATPase activity alone. Combining the NBD1 and NBD2 proteins in vitro fails to stimulate significant changes in ATPase activity and biochemical studies fail to identify significant in vitro dimerization. Functional characterization of the individual ABCC6 NBDs (NBD1 or NBD2) demonstrates that both NBDs are capable of ATP hydrolysis, likely mediated via homodimerization. Significant differences in NBD1 and NBD2 ATPase rates are observed and may be attributed to conservative substitutions within the Walker B motif. Further, co-incubation of NBD1 and NBD2 in varying stoichiometries stimulates ATPase activity, consistent with NBD1-NBD2 heterodimerization. To better understand how the non-canonical ATP-associated sequences impact NBD activity, chimeric proteins were produced; the non-canonical sequences in CFTR were introduced into ABCC6 NBD proteins and the canonical sequences were introduced into the CFTR NBD proteins. Changes in protein expression and activity were observed in both systems. The introduction of the non-canonical CFTR sequences into ABCC6 significantly reduced the ATPase activity of NBD1 independent of NBD2. Further, ATPase stimulation by co-incubation of NBD1 and NBD2 was significantly reduced in the presence of the noncanonical (CFTR) sequences. The loss of ATPase activity in NBD1, and stimulation by co-incubation of NBD1 and NBD2, are consistent with current models of ATP binding and hydrolysis in CFTR. Further characterization of these NBD-NBD interactions is critical to understanding CFTR channel regulation and its disruption by CF-causing mutations.

Post-translational modifications (PTMs) play a crucial role during biogenesis of many transmembrane proteins, including both ion channels and ABC transporters. Previously, studies evaluating PTMs in CFTR have been limited by difficulty obtaining sufficient amounts of purified protein for analysis. We recently developed a recombinant over expression system that yields large amounts of highly purified CFTR using a lentivirus-based system in mammalian HEK293 cells. The vector, designated L2231/TRE-CFTR.TEV.His-IRES-GFP, was packaged and pseudotyped with the VSV-G envelope glycoprotein and used to transduce a recombinant cell line (293HEK.m2) constitutively expressing the M2 reverse transcriptional transactivator (rtTA). Using GFP expressed from a bicistronic mRNA, clonal cell cultures were selected and tested for inducible CFTR expression, multiplication kinetics, and stability. Expression was induced in 10 liter bioreactor cultures near peak density of ~3x10 6 cells/ml, generating CFTR protein yields >1 pg/cell. Using this technique followed by nickel affinity chromatography, we obtained purified protein samples that represent both properly folded and unfolded configurations of CFTR and examined sites of PTM by liquid chromatography mass spectrometry. Purified, full-length CFTR was subjected to enzymatic digestion using a variety of proteases singly or in combination (trypsin, chymotrypsin, Lys-C, Glu-C, Asp-N, among others). Resulting peptides were passed into an ionspray interface of a 4000 Q TRAP LC/MS/MS system (Applied Biosystems/MDS Sciex, Concorde, Ontario, Canada), and resulting data processed with an in-house MASCOT search engine. Our proteomic analysis identified numerous and specific CFTR modifications, including novel sites for CFTR phosphorylation and potential residues for ubiquitination, methylation, and palmitoylation, a modification never before reported for CFTR. Side chain attachments such as these are likely to play a role in channel maturation and activity, and may also account mechanistically for certain CFTR mutations associated with disease. Identification of novel PTMs could also lead to new therapeutic targets for cystic fibrosis.

Supported by CFF and NIH.

Kim, S.; Khushoo, A.; Skach, W.R. OHSU, Portland, OR, USA CFTR folding begins during synthesis as the nascent chain emerges from the ribosome. Analyzing co-translational folding therefore provides valuable insight into CFTR biogenesis and represents a novel method to decipher the effects of Phe508 deletion (F508del) on the CFTR folding pathway. To understand this process, our laboratory developed methodologies for using fluorescence resonance energy transfer (FRET) to examine cotranslational NBD1 folding. This strategy utilized a donor fluorophore (CFP) attached as in-frame N-terminal fusion to CFTR NBD1 and a small acceptor dye NBD (7-nitrobenz-2-oxa-1,3-diazole) incorporated at engineered amber (UAG) stop codons using a modified synthetic suppressor tRNA (ε-NBD-lysine-tRNA amb ). When translated in vitro, truncated mRNA templates generate a uniform cohort of nascent chains that mimic folding intermediates kinetically trapped at a specific stage of synthesis. FRET analyses of fluorescently labeled ribosome nascent chain complexes with the acceptor probe located at CFTR residue 450 revealed that: 1) NBD1 achieves a native-like conformation on the ribosome, 2) NBD1 folding

Randak, C.O. 1 ; Welsh, M.J. 2, 3 1. Pediatrics, University of Iowa, Iowa City, IA, USA; 2. Internal Medicine, University of Iowa, Iowa City, IA, USA; 3. Howard Hughes Medical Institute, Iowa City, IA, USA Previous work has shown that CFTR-NBDs possess the capacity for two different kinds of enzymatic activity coupled to channel gating, ATPase activity (ATP Ç ADP + P i ) and adenylate kinase activity (ATP + AMP Ä ADP + ADP). Adenylate kinases have distinct binding-sites for ATP and AMP. Pyrophosphate (PP i ) markedly increases the opening rate and open time of CFTR if ATP is present. However, how PP i interacts with CFTR remains uncertain. We hypothesized that PP i influences the interaction of nucleotides with CFTR. We tested this hypothesis by using patch-clamp electrophysiology with excised inside-out membrane patches and by studying the interactions of 8-azido-photoreactive nucleotide analogues with membrane-bound CFTR.

Three lines of evidence indicated that PP i did not directly interact with an ATP-site: first, we found that PP i reduced the ATP concentration generating half-maximal current, indicating that PP i did not compete with ATP. Second, PP i did not reduce radioactive labeling of CFTR with [α 32 P]8-N 3 -ATP. Third, PP i increased ATP-dependent CFTR current even when ATPbinding to either one of the two ATP-sites was abolished (A462F and S1248F mutations): PP i increased the ATP-dependent opening rate in both mutants; however, channel open time was increased only in CFTR A462F. These results indicate that PP i influenced gating by interacting with a site different from the ATP-sites and that its effect on open time required ATP to interact with the ATP-site of NBD2. We then investigated how the interaction of AMP with CFTR was influenced by PP i and vice versa. When we labeled CFTR with [α 32 P]8-N 3 -AMP in the presence of ATP, we found that PP i increased labeling. In addition, when we studied the influence of different nucleotides on PP i -induced CFTR current stimulation we found that AMP reduced the on-rate of PP i in a concentration-dependent manner. Neither ATP nor GMP, which does not bind to the AMP binding-site, had this effect. These findings indicate that AMP influenced the interaction of PP i with CFTR. In summary the data suggest that PP i , ATP and AMP bind to non-identical sites but mutually influence their interactions with CFTR. Because PP i produces a large increase in chloride current through CFTR, our data suggest that the mechanisms of its interaction with CFTR may be of interest for developing new therapeutics for cystic fibrosis and other diseases involving ATP-binding cassette (ABC) transporters.

Supported by NIH/NHLBI, CFF, Francis Family Foundation, and HHMI. The majority of cystic fibrosis is caused by CFTR trafficking defects at the endoplasmic reticulum (ER) or plasma membrane. Disease causing mutations in the amino-terminal tail (N-tail) of CFTR, from amino acids 1-78, have been implicated in altered CFTR biogenesis and reduced cell surface stability. Lysine residues are the site of several post-translational modifications, including ubiquitination, sumoylation, neddylation, and acetylation. We hypothesized that one or more lysines in the N-tail are subject to post-translational modifications that modulate the intracellular trafficking or cell surface stability of CFTR. We created a series of CFTR lysine mutants in which each lysine in the N-tail was mutated to arginine (K8R, K14R, K26R, K52R, K64/K65R, K68R) and an additional mutant with all lysines mutated (N-tail-lysineless) . Productive trafficking of the CFTR mutants was monitored by pulse-chase analysis, and cell surface stability was examined by biotinylation and cycloheximide chase. We observed an increased rate of disappearance of the N-tail-lysineless CFTR mutant from the plasma membrane, supporting a role for N-tail lysine residues in the cell surface stability of CFTR. Furthermore, we found no difference in the degree of ubiquitination between wild type and N-tail-lysineless CFTR, indicating that N-tail posttranslational modifications other than ubiquitination influence CFTR trafficking. These studies highlight the importance of the N-tail in CFTR fate determination and suggest novel mechanisms that regulate CFTR trafficking. Grove, D.; Fan, C.; Ren, H.; Cyr, D. Cell and Developmental Biology, University of North Carolina, Chapel Hill, NC, USA Protein quality control networks impact the folding efficiency of cellular proteins and the fate of mutant disease proteins. However, mechanisms for the partitioning of non-native proteins between folding and degradation pathways remain unclear. The fatal disease cystic fibrosis is caused by misfolding and premature degradation of CFTR ∆F508. To date two distinct endoplasmic reticulum quality control (ER QC) complexes have been identified which monitor the folding progression of CFTR. The membrane-associated RMA1 E3/Derlin-1 complex appears to act co-translationally to recognize folding defects in CFTR that involve the misassembly of its aminoterminal regions (1, 2) . Whereas, the CHIP E3/Hsp70 complex may act post-translationally to recognize misfolded regions of CFTR that include NBD2 (1) . Yet, it is unclear how these complexes recognize CFTR as misfolded as well as what other proteins are necessary for this recognition.

Herein, we characterized the human Hsp40 DNAJB12 (JB12) and defined its role in the life span of CFTR and CFTR∆ F508. JB12 spans the ER with a single transmembrane domain and utilizes a cytosolic J-domain to interact with Hsp70. Surprisingly, we found that JB12 forms complexes with the RMA1 E3 machinery, but a functional interaction between JB12 and the CHIP E3/Hsp70 complex was not detected. The RMA1 QC complex monitors the folding status of N-terminal regions of CFTR and we observed that JB12 interacts within this area of CFTR. Modest elevation of JB12 levels increases Hsp70 association with Derlin-1 and consequently CFTR degradation is accelerated. Conversely, depletion of endogenous JB12 by siRNA results in a dramatic 3-fold increase in the folding efficiency of CFTR and permits some CFTR ∆F508 to fold and escape the ER. Collectively, it appears that the Hsp40 JB12 recruits Hsp70 to function as a component of the RMA1 E3 ligase complex to select misfolded CFTR for proteasomal degradation. Interestingly, attenuation of JB12 along with introduction of the V510D suppressor mutation restored CFTR ∆F508 folding efficiency to 50% of wild-type. Thus, down regulation of JB12 activity combined with use of a small molecule chemical corrector that mimics the action of the V510D suppressor mutation has the potential for use as a combination therapy to treat CF.

Supported by the Cystic Fibrosis Foundation. 1. Younger, J.M., Chen, L., Ren, H.Y., et al. (2006) Cell 126, 571-582. 2. Rosser, M.F., Grove, D.E., Chen, L., and Cyr, D.M. (2008) Mol Biol Cell 19, 4570-4579. sive reactive oxygen species (ROS) production by phagocytes. Our studies suggest that at a basal level, without infection, there is an increased ROS production, thereby suggesting that there could be an abnormal regulation of ROS production in CF. Reduced glutathione (GSH) is a major antioxidant and has been shown to be decreased not only in the airway lining fluid but also in mitochondria. In addition, deficits in mitochondrial GSH were reported elsewhere to lead to respiratory chain dysfunction.

Aim: The aim was to evaluate i) the possible relationship between CFTR, mitochondrial redox state and respiratory chain activity in CF models; ii) if mitochondrial dysfunction could be involved in the increased ROS production and the altered inflammatory response in CF models; and iii) the efficacy of antioxidants in correcting mitochondrial and cellular defects.

Methods: Mitochondria from murine colonic epithelia (cftr -/vs. cftr +/+ mice) and from bronchial epithelial cells (IB3-1 CF vs. CFTR corrected C38 cells) were isolated using differential centrifugation. The redox state was assessed by GSH concentrations measured by HPLC coupled with electrochemical detection. Redox-dependent modulation of aconitase activity was measured by spectrophotometry and used as an index of mitochondrial oxidative stress. Complex I (CI) enzymatic activity was monitored using spectrophotometric methods and CI expression was analysed by blue native PAGE. IL-8 production was quantified by ELISA. GSH monoethyl ester (GSH-MEE), a drug able to selectively restore mitochondrial GSH and Nacetyl-cysteine (NAC), a drug that increases de novo GSH synthesis were used as GSH modulators.

Results: We observed a 30% decrease in mitochondrial GSH in CF vs. corrected cells whereas cytosolic GSH was unchanged. This was associated with a decrease in aconitase activity. CI activity was decreased by 50% in mitochondria from CF cells and mice and this was not due to a decrease in CI expression. In CF vs. corrected cells basal and TNFα-induced IL-8 production were increased 2-and 1.4-fold, respectively. GSH-MEE restored mitochondrial GSH, aconitase activity and CI activity. Basal and TNFαinduced IL-8 levels were markedly reduced by GSH-MEE treatment. NAC did not restore mitochondrial GSH in CF cells but increased cytosolic GSH. TNFα-induced IL-8 production was decreased by NAC but 2-fold less than with GSH-MEE.

Conclusion: We suggest that reversible redox alterations in the mitochondrial compartment and CI inhibition could be linked. A decrease in mitochondrial GSH could be implicated in the exaggerated basal or stimulated cytokine production in CF cells. We propose that antioxidants targeting mitochondria could be useful in restoring CI activity and controlling the redox balance in CF cells and as such could be considered as potential therapeutic agents in CF.

This work was supported by the association "Vaincre la Mucoviscidose." scription of IL-8, in bronchial cells (Dechecchi, 2008) . The involvement of the metabolism of SLs on the biological effect of miglustat was tested through the comparative analysis of the inhibitor amitriptyline. In agreement with the anti-inflammatory effect demonstrated in vivo in CF mice (Teichgraber, 2008 , Becker, 2009 , we observed that amitriptyline significantly inhibited the expression of IL-8 mRNA by about 50 % in IB3-1 and CuFi-1 cells (Dechecchi, NACFC 2009 ). To gain insights into how miglustat and amitriptyline modulate the inflammatory response to P.aeruginosa, RNA macroarrays by TaqMan Low Density Array (TLDA) custom designed of 93 genes, were conducted. Of the 93 genes analysed 53 were detectable in IB3-1 cells. P.aeruginosa (4 hours) up-modulated the expression of 16 genes, mainly chemokines (IL-8, Gro-α/β/γ, GCP, IP-10), pro-inflammatory cytokines TNFα, and the transcripts of ICAM-1, NFKB1, TLR-2, HBD-4 genes. Both miglustat and amitriptyline decreased the expression of almost all the up-regulated genes, indicating that these drugs are effective in reducing neutrophil chemotaxis, likely by decreasing ceramide expression dependent on P.aeruginosa, as we previously observed. The effect of miglustat was also studied in murine models of lung infection and inflammation in C57BL/6 mice challenged with intranasal instillation of LPS or with intratracheal inoculum of P. aeruginosa strain PAO1. Miglustat (100mg/Kg at 72, 48, 24 hr before LPS pro-inflammatory challenge or 400 mg/Kg 1 hr before PAO1 infection) significantly reduced the amount of neutrophils recruited in BALF and MPO activity in both LPS stimulated and PAO1 infected mice. Collectively these results demonstrate a down-modulation of neutrophils chemotactic signalling by targeting SLs metabolism.

Supported by: Italian FFC (#12/2008) with the contribution of "Delegazione FFC di Vicenza con le scuole vicentine."

Stick, S.M. 1, 2 ; Sutanto, E. 3, 2 ; Kicic, A. 2, 3 ; Foo, C. 2 Background: Neutrophilic airway inflammation begins early in life and is associated with structural lung disease including gas trapping and bronchiectasis in the absence of overt bacterial infection. One possible explanation is an exaggerated epithelial pro-inflammatory response to infections with human rhinovirus (HRV), a ubiquitous pathogen in the first years of life.

Methods: We used primary lower airway cell cultures obtained from children with cystic fibrosis (CF) and healthy controls (H). We measured IL-8 release in response to non-viral pro-inflammatory cytokines and to infection with HRV.

Results: CF and H cells responded similarly to non-viral stimulation. However, CF compared to H cells had a significant 7-fold increase in IL-8 at 24 hrs post-infection with HRV (p<0.01). Increased IL-8 was associated with decreased viability of CF (r2-0.38) compared to H cells. Apoptotic (7fold:4-fold) and interferon-β responses (3-fold:0) to HRV were greater in H than CF cells and viral replication lower suggesting decreased viability due to necrosis in CF cells and enhanced viral release.

Conclusions: Dysregulated innate immune responses to HRV in CF epithelial cells results in exaggerated IL-8 release and enhanced viral replication. These responses serve to amplify the IL-8 response to HRV and could act as a potent trigger for neutrophilic inflammation in early life. Agents that enhance anti-viral innate immunity could be useful to slow the onset of structural lung disease. There is compelling evidence that TMEM16A (ANO1) functions as a calcium-activated chloride channel (CaCC) . Recent studies suggest that TMEM16A is an important CaCC in airway epithelium. Here, we used pharmacological tools to investigate the contribution of TMEM16A to CaCC current in human airway and intestinal epithelial cells. A highthroughput screen of ~100,000 compounds revealed 6 chemical classes of TMEM16A inhibitors that fully blocked TMEM16A chloride current with IC50 < 10 µM. The inhibitors did not affect cytoplasmic calcium signaling, suggesting that they target TMEM16A directly. There were two distinct groups of inhibitors. Group A inhibitors fully inhibited CaCC current in multiple types of human epithelia, including primary bronchial cell cultures and T84 intestinal cells. Group B inhibitors poorly inhibited CaCC current in the same cell types, with maximum inhibition in the range 10-40 %. However, both group A and B inhibitors completely blocked CaCC current in salivary gland cells. Group A and B inhibitors strongly inhibited another TMEM16 isoform (TMEM16B) that has been associated with calcium-activated chloride channel activity. Together with knockdown and overexpression studies, these data show that TMEM16A carries most CaCC current in salivary gland epithelium, but is a minor contributor to CaCC current in airway and intestinal epithelium. The small-molecule modulators identified here permit pharmacological dissection of TMEM16A/CaCC function, and some are potential development candidates for drug therapy of hypertension, pain, diarrhea and excessive mucus production. This work was supported by the CFF and NIH. Objective: With the ability to inhibit protein expression, antisense oligonucleotides (AON) offer new perspectives for the treatment of cystic fibrosis (CF). This disease is characterised by a defective chloride secretion mediated by the cystic fibrosis transmembrane conductance regulator (CFTR) and excessively increased Na + -absorption via the amiloride-sensitive epithelial Na + channel (ENaC). It was shown in a mouse model, that Na + hyperabsorption by ENaC is sufficient to evoke typical symptoms of cystic fibrosis-like lung disease that is found in CF patients [1] . Therefore, by downregulation of the ENaC mediated Na + hyperabsorption in CF, AON could circumvent one of the basic problems of this severe disease. AON are short synthetic DNA molecules (15-18 bases) which are complementary to specific mRNA sequences and created in order to prevent the functional expression of the target protein. Thus, the objective of this investigation was to find effective AON against human ENaC with respect to improving therapy for CF patients. For that reason, we selected specific AON that efficiently suppress Na + hyperabsorption by inhibiting the expression of the α-ENaC subunit.

Methods: For this study we used primary cultured human nasal epithelial cells derived from CF and non-CF tissue, which were transfected with the specific AON. To analyze the inhibitory AON effect on the functional level, we used confluent monolayers of primary cultured cells for the measurement of the amiloride-sensitive ENaC current in Ussing chamber analyses. To prove the effect of AON treatment on the protein level, we carried out Western blot analyses with a specific anti α-ENaC antibody. Furthermore, we performed fluorescence optical analyses to visualize the cellular AON uptake and distribution.

Results: Using these methods we showed that AON repress effectively the amiloride-sensitive Na + absorption mediated by ENaC in CF and non-CF tissues. With the functional Ussing chamber studies we showed that the AON transfection inhibits the ENaC current by about 75% in CF and by about 66% in non-CF cells. Further, we demonstrated that the AON are able to sustain this functional ENaC inhibition over a time period of more than 72h. We verified this downregulation of the ENaC protein expression in Western blot experiments and found a 46% reduction of the α-ENaC protein in AON transfected cells [2] .

Conclusion: Our data are a first step towards a preclinical study using the AON to reduce Na + hyperabsorption in CF epithelia. From these data we conclude that ENaC specific AON could evoke a long-lasting repression of the excessive Na + absorption in CF, thereby improving life quality of the patients.

References: 1. Mall M, Grubb BR, Harkema JR, O'Neal WK and Boucher RC. Increased airway epithelial Na + absorption produces cystic fibrosis-like lung disease in mice. Nat Med 10: 487-493, 2004. 2. Sobczak K, Segal A, Bangel-Ruland N, Semmler J, Van Driessche W, Lindemann H, Heermann R and Weber WM. Specific inhibition of epithelial Na + channels by antisense oligonucleotides for the treatment of Na + hyperabsorption in cystic fibrosis. J Gene Med 11: 813-823, 2009. Bangel-Ruland, N.; Sobczak, K.; Leier, G.; Leciejewski, B.; Weber, W. Institute of Animal Physiology, University of Muenster, Muenster, Germany Cystic fibrosis (CF) is the most frequent genetic disease in the Caucasian population. CF is caused by a defective gene coding for the cystic fibrosis transmembrane conductance regulator (CFTR), a cAMP-and ATPdependent Clchannel, which plays also a central regulatory role in epithelial cells. The most common mutation is ∆F508 that impairs CFTR processing within the cell and thus prevents functional expression at the apical cell surface.

iodide-free buffer, fluorescence was measured in response to serial additions of iodide and ATP. TMEM16A inhibitors were identified by reduced YFP fluorescence quenching in response to ATP. Screening of drug and natural product collections identified multiple TMEM16A inhibitors. However, with the exception of tannic acid, the remaining TMEM16A inhibitors showed partial or weak inhibition of native CaCC chloride current in T84 cells. Tannic acid was further studied because of its high inhibition potency and efficacy, and because tannic acid and related gallotannins are present in a number of widely consumed food products. IC50 of tannic acid inhibition of TMEM16A was 6 µM. Tannic acid inhibited CaCCs in multiple cell types, including primary bronchial cell cultures and T84 cells, and did not inhibit CFTR chloride conductance. Structure-activity analysis indicated the requirement of gallic or digallic acid substituents on a macromolecular scaffold (gallotannins), as are present in green tea and red wine. Other polyphenolic components of teas and wines, including epicatechin, catechin, and malvidin-3-glucoside, poorly inhibited CaCCs. Remarkably, a 1000-fold dilution of red wine and 100-fold dilution of green tea inhibited CaCCs by > 50% in T84 cells. Tannic acid, red wine and green tea inhibited arterial smooth muscle contraction and intestinal chloride secretion. Gallotannins are thus potent CaCC inhibitors whose biological activity provides a potential molecular basis for the cardio-protective and antidiarrheal benefits of red wine and green tea.

This work was supported by the CFF and NIH. Introduction: Airway epithelial injury and remodelling is a critical component of the lung pathology in cystic fibrosis (CF) patients. Although chronic infection and inflammation has been clearly associated with this progressive epithelial damage, the effect of inflammatory products on the ability of airway epithelia to repair after injury is still unclear. Thus, our objective was to study the regulation of repair mechanisms by tumor necrosis factor alpha (TNF), a major component of lung inflammation in CF, using a model of mechanical wounding of non-CF (NuLi) and CF (CuFi) bronchial monolayers.

Results: As we previously demonstrated (AJP295 (5): L866-80,2008) , we first verified that NuLi monolayers repaired faster than CuFi. We then observed that TNF stimulated NuLi and CuFi wound healing. Surprisingly, this stimulation was higher after chronic (24-48 h) exposure to TNF before injury (up to 2.6 fold). Single cell analysis in time-lapse experiments revealed that TNF accelerated bronchial cell migration during repair. Accordingly, inhibition with GM6001 of metalloproteinases (MMP), which are key components of cell migration machinery, prevented TNF-induced stimulation of wound repair.

Because TNF exposure and MMP activation could be responsible for epidermal growth factor receptor (EGFR) transactivation, through the release of the EGFR ligands controlling bronchial repair, we then decided to examine the role of EGFR pathway in the regulation of wound repair by TNF. We showed that EGF titration (with an EGF antibody) or EGFR tyrosine-kinase inhibition (with tyrphostin inhibitors) prevented the TNF-stimulation. Moreover, an EGFR activation (i.e. increased phospho-EGFR levels) was observed after TNF exposure. All together, these results indicated a role of EGFR transactivation in the regulation of repair by TNF.

Finally, we recently reported a relationship between EGF response and K + channel function, which is critical for bronchial repair. We now show that exposure to TNF stimulated KvLQT1 and K ATP K + currents in NuLi and CuFi polarized cell monolayers (140% of stimulation in NuLi and 175% in CuFi). In addition, KvLQT1, K ATP and KCa3.1 channel inhibition (with clofilium, glibenclamide and TRAM-34) abolished the TNF-induced stimulation of bronchial repair, suggesting that TNF-stimulation could be dependent on functional K + channels.

In summary, TNF could up-regulate the repair processes of non-CF and CF epithelial cell monolayers after injury through EGFR pathway and K + channel stimulation. These results show that TNF could play a complex role in CF airways by promoting both epithelial injury and repair.

Supported by the Canadian Cystic Fibrosis Foundation.

The aim of the present study is to investigate the functional expression and insertion of wild type (wt) CFTR in the apical plasma membrane of human CF airway epithelia after wtCFTR-mRNA transfection. Using mRNA would have several advantages compared with a DNA plasmid based approach, e.g. the nuclear envelope, one of the major obstacles to non-viral transfection, could be circumvented and the risk of insertional mutagenesis avoided. We propose to develop a strategy for the delivery of CFTR-mRNA directly to epithelial cells.

The human bronchial epithelial cell line CFBE41ostably expressing ∆F508-CFTR is well suited for CF research: It forms polarized cell monolayers with a sufficient transepithelial resistance when grown on permeable filters that allows functional measurements of transepithelial transport processes. The cells can be mounted in modified Ussing chambers for electrophysiological measurements to determine transepithelial short-circuit current, conductance and capacitance.

In orientating experiments we determined the appropriate length of the CFTR-mRNA PolyA-tail for an effective and long lasting expression using the Xenopus laevis oocyte expression system. We show that a 120 PolyA-tail produces maximal increases in CFTR current, conductance and capacitance following cAMP stimulation.

Furthermore, we carried out Ussing chamber experiments with CFBE41ocells transfected with this wtCFTR-mRNA. In contrast to the non-transfected control cells, where no CFTR Cl --conductance could be detected, the transfected cells showed an increased CFTR current, conductance and capacitance after CFTR activation by cAMP (100 µM). Interestingly, CFTR currents in wtCFTR-mRNA transfected cells were approximately identical with cells expressing CFTR (16HBE14o -). The increase in capacitance shows that a major part of this activation is caused by exocytotic insertion of preformed functional CFTR molecules into the apical membrane. Furthermore, the CFTR activation is sensitive to the Clchannel blocker NPPB, which decreased CFTR current significantly.

Additionally, we determined the abundance of total CFTR protein in transfected CFBE41ocells by Western blot analyses and additionally detected CFTR molecules at the apical surface of these cells with immunofluorescence approaches. From these data we conclude that the CFTR-mRNA transfection was successful in these cells. Since there is no risk of insertional mutagenesis and/or vector-induced immunogenicity, in vivo manipulation of cells with mRNA is considered to be a valuable and safe alternative and could be a novel tool to restore impaired CFTR function in CF. CFTR function is required for efficient airway epithelial cell migration and wound repair. Blocking channel activity or silencing expression by RNA interference produces a significant delay in wound healing by impeding lamellipodia protrusion and reducing the rate of cell migration into the wound. The objective of this study was to investigate mechanisms responsible for Cland HCO 3 dependence of migration mediated by putative functional interactions between CFTR and anion transporters in human bronchial epithelial (HBE) cells and Calu-3 cells. Cell migration and wound closure was continuously measured by time-lapse video imaging and cellsubstrate impedance sensing. HBE cells or Calu-3 cells were grown to confluence on impedance sensing chamber arrays, each containing a circular 250 µm diameter electrode. Wound repair was monitored by applying alternating current (1 µA, 15 kHz) and measuring increases in impedance as cells migrated onto the electrode surface. As wound closure proceeded and more cells covered the electrode, impedance increased until reaching a plateau when wound closure was complete. Replacement of extracellular Clwith methanesulfonate produced a significant (3 fold) delay in the time required for complete wound closure in HBE cells, similar to the effects of the CFTR blocker, CFTR inh -172 (20 µM) and the anion exchange inhibitor, DIDS (30 µM). However, no effect of CFTR inh -172 was detected in HBE cells in the absence of Cland no effect of DIDS occurred after CFTR inhibition. In contrast, Clreplacement did not significantly affect Calu-3 cell wound closure. Lowering extracellular ] from 26 to 5 mM (with continuous 5% CO 2 gassing) did not significantly affect intracellular pH as determined by single cell ratio imaging using the pH sensitive dye, BCECF. However, 5 mM extracellular HCO 3 produced a significant (2 fold) delay in the time required for complete wound closure of Calu-3 cells. Interestingly when CFTR inh -172 was added to Calu-3 cells bathed in 5 mM HCO 3 media, wound closure was completely inhibited. In contrast, 5 mM HCO 3 media produced a minor delay in HBE cell restitution. Analysis of anion transporter expression revealed that HBE and Calu-3 cells express relatively high levels of SLC26A6 and SLC4A2 anion exchange mRNAs whereas the electrogenic Na + -HCO 3 cotransporters (SLC4A4 and SLC4A5) exhibited significantly (64 and 8 fold respectively) higher levels of expression in Calu-3 compared to HBE cells. These findings indicate that HBE and Calu-3 cells exhibit distinct anion dependencies with respect to cell migration and wound closure and that CFTR interacts with different anion transporters in these airway epithelia to facilitate wound repair. In HBE cells the data suggest that CFTR collaborates with SLC26A6 and/or SLC4A2 anion exchangers to sustain HCO 3 efflux, providing a pathway for Clrecycling across the membrane. However, in Calu-3 cells wound closure is HCO 3 dependent but Clindependent, suggesting that CFTR functions in parallel with electrogenic Na + -HCO 3 - A method for measuring liquid absorption in the airways would be useful for screening new corrective therapies for cystic fibrosis (CF). We hypothesize that changes in liquid absorption rate in the airways can be measured in vivo using imaging methods after a radiolabeled surrogate (diethylene triamine pentaacetic acid, DTPA) is delivered by inhalation. The objective of these studies was to characterize the absorption rate of Technetium 99m labeled DTPA (Tc-DTPA) in CF and non-CF human bronchial epithelial cell cultures (HBEs) in order to determine whether DTPA absorption is driven by liquid absorption.

Methods: DTPA absorption rates were compared in CF and non-CF HBEs. Tc-DTPA in 10 µL PBS was added to the apical surface of 4 non-CF and 2 CF cell lines (genotypes DF508/DF508 -homozygous severe mutations and DF508/D1152H-heterozygous severe/mild mutations with mild pulmonary phenotype) and DTPA absorption rate was measured over 24 hours. The effect of osmotic gradients on DTPA absorption was determined by repeating this experiment with different concentrations of mannitol on the apical and basolateral surfaces (2 CF and 2 non-CF lines included). Finally, measurements of DTPA and liquid absorption were made simultaneously after the addition of 50 µL PBS containing Tc-DTPA and Texas red dextran (TRD) to the apical surface of 3 non-CF and 2 CF cell lines. Changes in the concentration of the non-absorbable TRD were used to determine liquid absorption rate.

Results: CF HBEs absorbed DTPA at a significantly higher rate than non-CF HBEs (68.6 ± 19.8 vs 40.9 ± 9.7 %/24hr; p<0.001) and DF508/DF508 cells absorbed DTPA more quickly than DF508/D1152H cells (80.9 ± 7.4 vs. 56.2 ± 20.9 %/24 hr; p<0.001). DTPA absorption from the apical surface was accelerated by basolateral mannitol and slowed (but not stopped) by apical mannitol, and DTPA absorption rate varied linearly with mannitol gradient (R-squared 0.76-0.95). There was a linear relationship between DTPA and liquid absorption rates based on comparisons made using the dye based method (R-squared 0.22-0.85; overall 0.43).

Conclusions: CF HBEs, which have been previously shown to hyperabsorb liquid, also hyper-absorb DTPA. DTPA absorption, like liquid absorption, can be driven by osmotic gradients. And finally, liquid and DTPA absorption rates are proportional as assessed using a dye based technique. Previously published studies have demonstrated that DTPA absorption is increased in vivo in the CF airway. These studies provide evidence that this increased absorption rate is a consequence of the liquid hyperabsorption associated with the basic defect of CF.

Funding NIH K25 HL081533. Airway submucosal gland serous acini are major sites of CFTR expression in the lung and likely generate most of the fluid secreted by these glands. Agonists that signal through cAMP activate CFTR-dependent secretion from intact glands. To identify mechanisms of cAMP-evoked fluid secretion and the role of CFTR in this process, we isolated serous acinar cells from submucosal glands dissected from human nasal turbinates. Human serous cells were studied by DIC imaging of cell volume, to track agonist-induced changes in cell solute content reflective of changes in the secretory state of the cell, and simultaneous quantitative fluorescence microscopy to measure intracellular concentrations of ions involved in driving fluid secretion ([Cl -] i ) and regulating it ([Ca 2+ ] i ). Stimulation with the cAMP-elevating agonist vasoactive intestinal polypeptide (VIP; 1µM) elicited serous cell shrinkage (16±1%) that was indicative of the activation of Cl -/fluid secretion. VIP-evoked shrinkage was blocked by the CFTR inhibitor CFTR inh 172 but not by the Ca 2+ -activated Clchannel (CaCC) inhibitor niflumic acid (NFA). VIP-activated shrinkage was temporally correlated with a small increase in [Ca 2+ ] i (from 46±3nM at rest to 129±6nM peak). Both shrinkage and Ca 2+ responses were downstream of cAMP/PKA, as they were mimicked by the adenylyl cyclase-activating compound forskolin and inhibited by the PKA inhibitor H89. Similar small [Ca 2+ ] i elevations (130±5nM) observed during stimulation with a low concentration (100nM) of the cholinergic agonist carbachol (CCh; an IP 3 -generating cAMP-independent agonist) failed to activate shrinkage, suggesting these Ca 2+ signals are insufficient to activate secretion. Nonetheless, [Ca 2+ ] i elevation was required for VIP/cAMP-evoked secretion, as abolishing the Ca 2+ signal inhibited shrinkage. However, Ca 2+ was not required for CFTR activation; in SPQ-loaded serous cells, a NO 3 substitution protocol revealed that VIP activated a PKA-dependent, Ca 2+ -independent plasma membrane Clpermeability that was inhibited by CFTR inh 172. These data suggest that CFTR is the secretory Clchannel during cAMP-evoked Clsecretion, while [Ca 2+ ] i elevation is required for activation of K + channels mediating counter-ion current. We also observed important synergistic interactions between cAMP-and cholinergic stimulation in human serous cells. The success of airway gene transfer that can correct the bioelectrical defect in cystic fibrosis (CF) mice has not been studied in the same animals over long time periods. We have examined the sustainability of lentiviral (LV) CFTR gene transfer success via repeated nasal potential difference (PD) measures in individual CF mice over their lifetimes.

Methods: The nasal airway of anaesthetized CF tm1unc mice was instilled with either PBS (control) or 0.3% lysophosphatidylcholine (LPC) 1 hour prior to delivery of a LV-CFTR gene. A third group received LPC followed by an empty (MT) LV vector control. Nasal PD measurements were designed for assessment at 1 week and 1, 3, 6, 9, 12, 15, 18 and 21 months after treatment in each mouse. Krebs saline solutions delivered via the single-lumen recording cannula (basal, basal+amiloride (10 -4 M), then low chloride+amiloride (10 -4 M)) were perfused at 1µl/min. Statistical analysis was by repeated-measures ANOVA. Serum was collected at all time points for immunological analysis.

Results: Correction of the chloride channel response (∆PD: average: 32%; Range: 12%-54% towards normal) continued in mice receiving LPC and LV-CFTR for 15 months, but was lost at 18 months (p<0.05, RM ANOVA). In the two control groups, the mean ∆PD after PBS pre-treatment or LV-MT treatment at any time points was no different to untreated CF mice (n. s., ANOVA). There was no correction of sodium transport apparent at any time points (n.s., RM ANOVA). After completion of the 21 month time-point serum from all time points will be examined for the presence of antibodies to the LV vector.

Conclusion: In this ongoing study sustained correction of airway CFTR function persisted for at least 15 months using this LPC/LV-CFTR delivery method. Results from this third data set in our group examining LV-CFTR gene transfer further extends the evidence of functional CFTR transduction longevity. These findings support the feasibility of a long-lasting singledose gene transfer therapy.

Lee, R.J.; Foskett, J.K. Physiology, University of Pennsylvania School of Medicine, Philadelphia, PA, USA Serous acinar cells of submucosal glands are major sites of fluid secretion and CFTR expression in the lung. When stimulated with intracellular [cAMP] ([cAMP] i )-elevating agonists (VIP or forskolin), human and porcine serous cells exhibit cell shrinkage responses due to cellular KCl solute loss, reflective of the activation of fluid secretion. cAMP-evoked secretion requires Clefflux through CFTR and K + efflux through channels activated by PKA-dependent Ca 2+ signals. In contrast, murine nasal gland serous cells stimulated with 3µM VIP or 5µM forskolin exhibited neither shrinkage (secretion) nor Ca 2+ responses. Optical techniques were used to examine murine serous cells in more detail to determine if [cAMP] i -elevating agonists activate CFTR and whether or not CFTR plays a role in murine serous cell secretion. In Wt murine serous cells loaded with the Cl --sensitive fluorophore SPQ, NO 3 substitution revealed that stimulation with 1µM VIP or 3µM forskolin increased plasma membrane Clpermeability by >10-fold. This was blocked by the PKA-inhibitor H89 or the CFTR-inhibitor CFTR inh 172. Resting Clpermeability was identical in cells from Wt and cftr tm1Unc-/knockout (cftr -/-) mice, but, in contrast to Wt cells, cAMP stimulation did not increase Clpermeability in cftr -/cells. We hypothesized that, despite robust CFTR activation, murine serous cells secrete poorly during cAMP stimulation because absence of cAMP-evoked Ca 2+ signals results in responses, as VIP activated CFTR and CCh activated Ca 2+ -senstive K + channels. Stronger cAMP stimulation (1µM VIP, 3µM forskolin, or 100µM IBMX) markedly potentiated the 100nM CCh-evoked peak [Ca 2+ ] i responses (~350 nM), activating CaCC-dependent CFTR-independent secretion. Our data identify molecular mechanisms of human serous cell secretion and suggest that therapeutic targeting of crosstalk between cAMP and Ca 2+ signaling could allow restoration of secretion in CF serous cells. Goblet cell metaplasia and mucin hypersecretion leading to mucus accumulation and plaque formation are features of CF lung disease. Thus, identifying therapies to efficiently reduce mucus plaque production in the lung by controlling the secretion of mucins would be highly beneficial for CF patients. However, there is an incomplete understanding of the complex molecular machinery involved in mucin granule exocytosis. For example, the identity of the SNARE proteins that form the core of the exocytotic machine remain unidentified in airway goblet cells. We discovered that the vesicle SNARE VAMP-8 is expressed in lung tissues and airway epithelial cells. Immunofluorescence studies indicated that VAMP-8 was localized mainly in goblet cells of human lung tissues. Moreover, VAMP-8 was associated with mucin granules isolated from airway goblet cells, and redistributed into the cytoplasm upon agonist-stimulation of airway epithelial cell cultures. Importantly, silencing VAMP-8 mRNA expression by siRNA oligonucleotides drastically decreased basal and agonist-stimulated mucin secretion in airway epithelial cells. Thus, our data suggest that VAMP-8 is a major regulator of airway mucin granule exocytosis. Reduction of VAMP-8 activity / expression may provide a novel therapeutic target to ameliorate airway mucus obstruction in CF lung disease.

Supported by CFF KREDA10I0 and NIH P01-HL034322. Mucus obstruction of airways is one of the most vicious agents of morbidity and mortality in the hereditary disease cystic fibrosis (CF). Over the past three decades, great emphasis has been placed on abnormal transport of Cland Na + ions in CF, but no clear link between these defects and mucus obstruction has been established in all affected tissues. However, CFTR mutations also disrupt the transport of another biologically vital ion, HCO 3 -. The role of HCO 3 - in CF pathogenesis has received little consideration compared to other electrolytes affected in this disease. We examined the HCO 3 conductance of the porcine native small airway tissue ex vivo with a micro-Ussing chamber that we developed to monitor electrolyte transport functions in bronchioles. By replacing 150 mM Clin the apical solution with 150 mM HCO 3 or 150 mM gluconate in the presence of amiloride and forskolin+IBMX to block ENaC and activate CFTR, we determined that the permeability to HCO 3 is about 1/5th as that of Clfrom changes in transepithelial potential (V t ) and conductance (G t ). Since 150 mM HCO 3 may not be physiological in the airway, we reduced the concentration of both HCO 3 and Clto 25 mM and repeated the substitutions. These maneuvers also resulted in changes in V t and G t that were qualitatively similar to those above. These results show that the native porcine small airway exhibits a significant HCO 3 conductance, although notably less than that of Clwhich is consistent with previous findings for CFTR. The imminent question that remains is whether secretory agonists can effect changes in this conductance and whether anion conductance inhibitors will effectively block it. Answers to these questions may help define the role of HCO 3 in abnormal mucus clearance in CF affected organs in general. Defective expression or function of the cystic fibrosis transmembrane conductance regulator (CFTR) gene in airway epithelial cells results in abnormal regulation of Na + and Clmovement across epithelial cell surface leading to disruption of airway salt economy. The subsequent mucus plugging of the bronchial airways in CF may result in impaired ventilation/perfusion relationships and subsequent localized hypoxia/anoxia. We compared the ability of cell lines with defective CFTR and those with corrected CFTR to secrete certain cytokines (such as IL-8) under anoxic conditions. Two cell lines; IB3-1 (with F508del and W1282X mutations in the CFTR gene) and C38 (with corrected CFTR gene) were cultured in the presence or absence of oxygen in LHC-8 medium supplemented with 5% fetal bovine serum at 37°C for 1 to 3 days. IL-8 concentration in the supernatant was determined by ELISA (Biosource). Cellular viability was assessed by two methods: LDH release (Promega) indicating cell membrane integrity and ATP activity in mitochondria (MTS, Promega) indicating cellular metabolic activity.

IL-8 levels were significantly increased in CF cells in the presence of anoxia during the 2nd and 3rd days (P <0.001) (Figure 1 ). LDH and MTS measurements revealed that the CF cells had greater cell membrane permeability than corrected cells, but there was no difference in viability under conditions of anoxia. In addition, anoxia did not affect their metabolic or mitochondrial activity.

Anoxia appears to affect CF epithelial cells uniquely with an increased production of IL-8 with neutrophilic chemotactic activity. The viability of both types of epithelial cells was not altered in the presence of anoxia.

The ability of the CF cell line to respond with increased secretion of inflammatory cytokines may contribute to the inflammatory response seen in CF bronchial airways.

Funding of this study was provided by Long Beach Memorial Foundation. the human airway mucus clearance apparatus. In this report we measured mucociliary clearance time (MCT), cilia height as measured by scanning electron microscopy (SEM), surface liquid depth (SLD) by high speed video and intensity color mapping of the ciliated surface of epithelial cells and surface liquid, mucus viscosity (LogG *) using micro-rheometry and on ciliary beat frequency using high speed video analysis during normal conditions of hydration, after fluid-depletion using Cland Na + channel blockers to block Clsecretion and Na+ absorption in the palate, and after the instillation of NaCl 0.9% saline on the palate following fluid depletion. Normal saline is hypertonic to the frog palate at a concentration of 0.9%, Frog Ringers (made from normal Ringers solution and distilled water, 2:1) is nebulized continuously on the palate to maintain it in a 100% humidified and temperature controlled (20-22 °C) chamber. Channel blockers were made up in frog Ringers.

Results: Data from these studies is shown in the Table. The results may be summarized as follows: a) Compared to normal MCT, under FD conditions MCT is dramatically slowed. Treatment with NaCl results in MCT returning to a near normal. b) Cilia height measured perpendicular to cell surface is less under FD conditions and recovered to near normal by rehydration. c) Surface-liquid depth varied depending on the state of palate hydration and wave action. It was deepest under normal conditions, shallowest under FD conditions and near normal after rehydration. d) Mucus viscosity as indicated by Log G* indicated an increase in mucus viscosity after FD, improving to less than normal conditions after rehydration.

This study indicates that a reduction in cilia height due to a decrease in surface liquid and increased viscosity of mucus contribute to a slowing of MCT. Fluid rehydration which restores SLD and normalizes mucus viscostiy contributes to a recovery of MCT. Further studies on known agents of similar function need to be done to confirm this mechanism of action and other compounds need to be investigated for their beneficial action in restoring MCT in CF. Abdullah, L.H.; Davis, C. CF Center, University of North Carolina, Chapel Hill, NC, USA Our understanding of the regulation of mucin synthesis in the airways is essentially limited to an appreciation of mucin gene (MUC5AC, MUC5B) expression upon exposure of animal or culture models to inflammatory stimuli. Little is known about synthesis under control (i.e., "healthy") conditions, about the relationships between mucin gene transcription and translation, or about potential feedback mechanisms between mucin synthesis and secretory granule stores. We explored these relationships in human bronchial epithelial (HBE) cell cultures, using qRT-PCR to measure MUC5AC and MUC5B mRNA, and semi-quantitative Western blotting for the non-glycosylated (immature) mucin proteins, and the mature mucins. Goblet cells mount a full discharge of mucin secretory granules during a prolonged exposure to purinergic agonist [Kemp, et al., 2004. Am J Respir Cell Mol Biol. 31:446-55] , a feature we used to deplete mucin stores fully at the beginning of an experiment. HBE cultures were exposed for 45 min to 100 µM ATPγS, washed, then extracted as a function of time into 8 M guanidine hydrochloride or RLT buffer (Qiagen) to solubilize mucins and mRNAs, respectively. Mature MUC5B, measured using antibodies to the Cys-rich domains in the tandem repeat, remained at their post secretion lows for 2-4 hrs, then increased such that the stores were approximately half restored at 12 hrs and fully so at 24 hrs. Immature MUC5B was measured with an antibody to a peptide sequence in the glycosylated repeat domain; because positive signals are obtained exclusively from mucins in the ER, the antibody offers a measure of mucin peptide synthesis. This signal was elevated over baseline at 2 hrs, it peaked between 8-16 hrs, and then it declined over the next 24+ hrs. MUC5B mRNA, interestingly, was elevated less than 2-fold over baseline at a single time point, 4 hrs post-secretion; by 8 hrs the levels had

Harris, W.T. 1 ; Grenett, H.E. 2 ; Macewan, M. 1 ; Clancy, J.P. 1 1. Pediatric Pulmonology, UAB, Birmingham, AL, USA; 2. Cardiovascular Medicine, UAB, Birmingham, AL, USA Introduction: TGF-β1 is an established modifier of CF lung disease severity, yet mechanisms through which TGF-β1 influences CF lung disease progression remain elusive. In non-CF lung disease, dysregulated TGF-β1 activity results in pathologic airway remodeling inducing a progressive decline in lung function. We have previously demonstrated increased TGF-β1 secretion and altered fibroblast phenotype in CF model systems. In this study, we focus on dysregulated TGF-β1 activation in the CF context, first by utilizing CF-relevant inflammatory cytokines and then developing a CF bronchial epithelial-fibroblast cell co-culture system.

Hypothesis: TGF-β1 activation is dysregulated in cystic fibrosis as one mechanism through which TGF-β1 associated airway remodeling modifies CF lung disease progression.

Methods: Bronchial epithelial cells (CFBE41o-) stably transduced with F508del CFTR (∆F508 CFBE) or WT CFTR (WT CFBE) were grown in cell culture. Lung fibroblasts were collected from Cftr -/or WT mice (C57BL6/J background) and grown in primary cell culture. CF-relevant cytokine stimulation included exogenous TNF-α (2-10 ng/ml), IL-1β (2 ng/ml) and TGF-β1 (2 ng/ml). Bronchial cell polarity was induced by growing CFBE cells on permeable Costar 12-well filter inserts. In epithelialfibroblast co-culture systems, fibroblasts were plated at the bottom of each well with bronchial epithelial cells growing on the permeable filter inserts. Total and active TGF-β1 were measured in culture media utilizing the welldescribed mink lung epithelial cell bioassay. T-tests and ANOVA compared continuous variables with significance assumed at p<.05.

Results: TNF-α (10 ng/ml) stimulation doubled TGF-β1 activation in ∆F508 CFBE cell culture (No stim: 150±13.9 pg/ml, TNF-α stim: 309.4±43.5 pg/ml; p<.05). Similarly, TNF-α (2 ng/ml) and IL-1β (2 ng/ml) stimulation in combination with exogenous TGF-β1 (2 ng/ml) potently increased TGF-β1 activity in murine lung fibroblast culture (No stim: 39.91±1.78 pg/ml, TNF-α/IL-1β stim alone: 85.57±4.06 pg/ml, exogenous TGF-β1 alone: 117.8±6.67 pg/ml, TNF-α/IL-1β stim with exogenous TGF-β1: 441.3±59.82 pg/ml: p<.001). In polarized ∆F508 CFBE cell culture, basolateral (BL) secretion of TGF-β1 far exceeded apical secretion (BL: 8859±149 pg/ml, Apical: 1108±306 pg/ml; p<.001). In the bronchial epithelial:fibroblast co-culture system, culture of Cftr -/lung fibroblasts with ∆F508 CFBE resulted in a significantly increased fraction of active TGF-β1 compared to fibroblasts co-cultured with WT CFBE cells (∆F508 CFBE: 13.9±1.0% active TGF-β1, WT CFBE: 6.5±0.2% active TGF-β1; p<.001).

Conclusions: These preliminary in vitro cell culture model systems suggest that TGF-β1 activity is increased in CF by at least two CF-specific mechanisms: 1) stimulation with cytokines elevated in the CF airway and 2) aberrant CF epithelial:fibroblast interaction. As increased TGF-β1 activity is pathogenic in multiple non-CF lung diseases, these findings offer insight into the mechanisms through which TGF-β1 may modify CF lung disease severity and importantly introduce CF-specific model systems through which potential therapeutic interventions can be trialed to inhibit TGF-β1 activity in CF. The use of osmotically active solutions such as hypertonic saline for rehydration of fluid-depleted (FD) airways has been employed clinically in cystic fibrosis (CF) patients with a resultant improvement in lung function. Further studies are required to explain the mechanism by which osmotically active solutions exert a beneficial clinical effect in CF. The frog palate model of airway mucociliary clearance was used in this study for its versatility and resilience to be studied by multiple methodologies over a sustained period of time during a single experiment, including the ease of measurement mucus clearance time, easy access and large ciliated flat surface for high speed video, abundance of mucus and for its anatomical similarity to returned to baseline. That is, MUC5B transcription peaked as translation was increasing, and mucin peptide synthesis then continued to increase, but at basal levels of mRNA. To test whether elevated MUC5B transcription was essential for increased translation, we treated HBE cultures, post-secretion, with actinomycin D and found that mucin peptide synthesis at 8 hrs was unchanged from control. As expected, inhibiting translation with cycloheximide abolished mucin peptide synthesis. Hence, restoration of mucin stores following a full, agonist-induced discharge does not appear to depend on the synthesis of new mRNA. Rather, the regulation of mucin synthesis is posttranscriptional, with translation being modulated on stable, basal levels of mRNA. In control experiments, treating HBE cells with a supernatant of mucopurulent material harvested from lungs of CF patients removed at transplant caused massive, multi-fold changes in both mucin transcription and translation. Consequently, there appears to be at least two levels of control exercised over goblet cell mucin synthesis, one that operates independently of elevations in transcription levels under "healthy" conditions, and one triggered during inflammation that is highly dependent on elevated gene transcription.

These studies were supported by grants from the NIH (HL063756) and the North American CF Foundation.

Kernan, N.G. 1 ; Cullinan, P. 2 ; Alton, E.W. 1 ; Bilton, D. 3 ; Griesenbach, U. 1 1. Department of Gene Therapy, Imperial College London, London, United Kingdom; 2. Department of Occupational and Environmental Medicine, Imperial College London, London, United Kingdom; 3. Department of Respiratory Medicine, Royal Brompton Hospital, London, United Kingdom Several studies using a variety of in vitro models indicate that sex hormones such as oestrogen can alter ion transport across epithelial cells by either directly affecting CFTR or altering the activity of alternative chloride channels; such effects may in part explain the gender-difference in disease severity observed in some studies. However, published data are inconsistent with several studies postulating beneficial and others detrimental effects of oestrogen on CF ion transport abnormalities and, therefore, disease severity. A large proportion of women with CF regularly use oral contraceptives (OC), but the effect of OC use on disease severity has not been systematically studied. Here, we assessed the effects of OC use on clinical outcomes in a retrospective study of patient data from the Dendrite Database at the Royal Brompton Hospital, London, UK. The data include annual follow-up information from 681 women born between 1937 to 1992 of whom 42% have taken OC albeit for varying periods of time. Data regarding OC use is currently available from 1981 to 2010. We first performed an inter-patient analysis comparing average yearly changes in %FEV 1 and body mass index (BMI) and total days of intravenous (IV) antibiotic use over a five year period between matched cohorts of OC users (n=57), (median age at start of study period in years: 23 (16-45), median %FEV 1 at start of study period: 56. 2 (20.4-111.1) ), and OC non-users (n=57) (median age at start of study period in years: 22 (17-44), median %FEV 1 at start of study period: 48.4 (12.8 -119.6)). We found no differences between the two groups (median change in %FEV 1 : users: -1.87 (-11.5 to 10.4) , non-users: -1.03 (-11.8 to 17.9) ; median change in BMI: users: 0.051 (-1.1 to 1.6), non-users: -0.065 (-1.5 to 3.3) ; median total days on antibiotics: users: 49 (0 to 308), nonusers: 42 (0 to 378)). We next performed an intra-patient analysis of the same outcomes over a three year period on and a three year period off OC in the same patient (n=23 -27), but again did not detect any differences in any of the clinical outcomes studied. In conclusion, OC use in CF females did not affect %FEV 1 , BMI or intravenous antibiotic usage in this retrospective study; our findings suggest that there is no evidence of a clinically significant effect on CF outcomes. Paisley, D.; Czarnecki, S.; Coote, K.; Danahay, H. NIBR, Novartis, Horsham, United Kingdom The clinical methodology for evaluating nasal transepithelial potential difference (NPD) has been successfully adapted in a number of studies for use in the mouse. In addition to its use in evaluating the ion transport phenotype of both wild-type and CFTR knockout animals, this model has proven valuable in evaluating the ability of compounds to restore normal ion transport in CFTR knockout mice (e.g., Grubb et al., Am J Physiol. 266: C1478-83 (1994 ). Furthermore, UTP-mediated changes in NPD have also been demonstrated, most likely as a result of activation of calcium activated chloride channels (CaCC) (Ghosal et al., Eur Respir J. 15: 146-150 (2000) ).

The aim of our studies was to adapt the NPD model to enable transepithelial potential difference (TPD) to be measured in the murine trachea. Further, using pharmacological tools, we were able to demonstrate the presence of both CFTR and CaCC activity in this tissue. Development of this model may eventually enable the measurement of compound effects on ion channel function following administration by the inhaled route.

We adapted the NPD methodology to enable us to measure TPD and changes resulting from the perfusion of different solutions onto the tracheal epithelium. In our initial studies, baseline PD measured during perfusion with PBS containing 1mM amiloride resulted in a mean TPD measurement (± s.e.mean) of -2.03 ± 0.4 mV (n=20, across three separate studies). Switching from PBS to perfusion with Clfree PBS resulted in a ∆TPD (± s.e.mean) of -1.74 ± 0.7 mV (n=15, across two separate studies). This hyperpolarisation was completely reversed by subsequent perfusion with 20 µM CFTRinh-172 (∆TPD = +3.07 ± 0.7 mV (n=15, across the same two studies quoted for ∆TPD Cl-free PBS)), suggesting that the Clfree induced change is CFTR-mediated. Furthermore, that perfusion with CFTRinh-172 reversed the TPD beyond the PBS + amiloride baseline suggests that Clsecretion through CFTR may be a component of this basal PD.

We were particularly interested in investigating the presence of CaCC activity in the murine trachea in vivo. Changes in TPD induced acutely by perfusion with the purinergic agonist UTP (100 µM, mean peak ∆TPD = -6.5 ± 1.3 mV (n=8)) and also by the muscarinic agonist carbachol (1 mM, mean peak ∆TPD = -5.6 ± 0.9 mV (n=7)) were quantified. To demonstrate that these effects were not ENaC or CFTR-mediated, responses to these agents were evaluated in the presence of 1 mM amiloride and 20 µM CFTRinh-172 respectively. The change in TPD induced by carbachol was found to be partially inhibited by subsequent perfusion with the CaCC blocker DIDS (200 µM, 77% inhibition of sustained carbachol response), providing additional evidence that the carbachol response is, at least in part, mediated by CaCC.

Together, these data suggest that we can measure changes in murine TPD associated with secretion of Clthrough CaCC. To our knowledge this is the first report detailing measurement of TPD in the mouse. This model may prove useful in measuring changes in TPD induced by compounds which act by activating/potentiating CaCC. Furthermore, this model may be suitable for measuring the efficacy of compounds following administration via the inhaled route.

Sheridan, J. 1 ; Watson, M. 2 ; Tarran, R. 1, 2 1. Cell and Molecular Physiology, University of North Carolina, Chapel Hill, NC, USA; 2. Cystic Fibrosis Center, University of North Carolina, Chapel Hill, NC, USA Changes in intracellular Ca 2+ are fundamentally important for innate lung defense. For example, Ca 2+ regulates chloride secretion, ciliary beating, and mucin release. Recently, stromal interaction molecule 1 (STIM1) has been identified as a Ca 2+ sensor which is located in the endoplasmic reticulum membrane (ER). Upon ER Ca 2+ depletion, which occurs following UTP stimulation, STIM1 aggregates and moves via microtubules to the plasma membrane (PM)-ER junction to activate Ca 2+ entry. This paradigm has been well established in cell lines, such as HEK293 cells. However, since little is tion of XBP-1 during HBE inflammation expands the Golgi and mitochondrial compartments. Because XBP-1 activation is increased in freshly isolated and inflamed CF HBE, these responses are likely relevant for CF airways disease. We are currently addressing the functional consequences of XBP-1induced Golgi and mitochondrial expansion on HBE inflammatory responses, e.g., production of oxygen radicals. These studies may lead to new therapies for CF patients by targeting the XBP-1 pathway. CFF funded. Pseudomonas aeruginosa, bacteria which commonly infect lungs of cystic fibrosis (CF) patients, secrete N-(3-oxo-dodecanoyl)-S-homoserine lactone (3O-C12) to regulate bacterial gene expression. 3O-C12 also affects functions of eukaryotic cells. Transepithelial electrophysiology in Ussing chambers showed that 10-100 µM 3O-C12 increased Clsecretion (I Cl ) in CFTR-expressing airway epithelial cell lines, but not in CF cells. Bright field observations of "bubble" formation in oil-covered pig tracheas showed that 3O-C12 also increased fluid secretion by submucosal glands in non-CF pig tracheas. Fura-2 imaging showed that 3O-C12 increased cytosolic [Ca 2+ ], Ca cyto , similarly in Ca 2+ -containing and Ca 2+ -free solutions, indicating that Ca 2+ was being released from the endoplasmic reticulum (ER). FRET imaging of ER-targeted cameleon D1-transfected cells confirmed that 3O-C12 reduced [Ca 2+ ] in the ER (Ca er ). The Ca-ATPase inhibitor thapsigargin (tg) also increased I Cl and Ca cyto and decreased Ca er ; 3O-C12 had no effect in cells treated with tg. Patch clamp measurements on isolated nuclei from DT40 cells showed that 3O-C12 also activated IP 3 R1. TIRF imaging showed that 3O-C12 and tg both caused STIM1, a Ca-dependent ER protein that regulates plasma membrane Ca 2+ and cAMP signaling, to migrate to the plasma membrane, indicating STIM1 had become activated. FRET imaging of Epac H30-transfected cells showed that 3O-C12 increased cytosolic [cAMP] (cAMP cyto ). 3O-C12-stimulated I Cl was inhibited by a cAMP antagonist and increased by a phosphodiesterase inhibitor. The high Kd Ca 2+ chelator TPEN, which lowers Ca er without increasing Ca cyto in other cells, increased cAMP cyto and I Cl similar to 3O-C12. It is concluded that 3O-C12 stimulates CFTR-dependent Cland fluid secretion in airway epithelial cells by directly activating IP 3 R, lowering Ca er and activating STIM1 and storeoperated cAMP production. In CF airways, 3O-C12 (and other Ca-raising agonists) will alter Ca and cAMP signaling but will not increase Cland fluid secretion. An implication of these studies is that PKA-dependent activation of CFTR in airway epithelia may occur in response to agonists (e.g., VIP) that activate cAMP directly and also in response to Ca cyto -raising agonists (e.g., ATP and 3O-C12) that activate cAMP indirectly through the store-operated cAMP mechanism. Support: NIH, CFF and CFRI.

known about STIM1 translocation in native tissues, we tested the hypothesis that STIM1 movement is an important step in Ca 2+ influx in polarized human bronchial epithelial cells (HBECs). STIM1 knock down had no effect on ER-Ca 2+ release while Ca 2+ influx was abolished, demonstrating that STIM1 is required for Ca 2+ influx in HBECs. Next, we infected HBECs with a retrovirus containing STIM1-YFP, a construct that has previously been shown to regulate Ca 2+ entry in HEK293 cells, and cultured them on Transwell permeable supports for 4 weeks to induce polarization. Through qPCR we found that STIM1 expression levels in STIM1-YFP infected HBECs increased 3-fold, over mock infected cultures, which coincided with an increase in the Fura-2 emission ratio following UTP stimulation. These data imply that STIM1-YFP plays a role in airway Ca 2+ metabolism and that higher expression levels result in a larger influx of Ca 2+ . Using XZ and XY scanning confocal microscopy, we found that STIM1-YFP was distributed throughout the cell under basal conditions. Upon stimulation with 100 µM UTP, we observed no change in localization of STIM1-YFP. Inflammation of human bronchial epithelia (HBE) triggers endoplasmic reticulum (ER) stress-induced expansion of ER/calcium stores, a process responsible for amplification of calcium-dependent inflammatory responses and mediated by activation of X-box binding protein-1 (XBP-1). To date, it is not known whether XBP-1 also expands the Golgi and mitochondrial compartments during HBE inflammation. An expanded Golgi might contribute to the increased production of inflammatory factors and mucins by facilitating their glycosylation. Together with expanded ER/calcium stores, mitochondrial expansion might be relevant to oxidative stress responses of CF airways, since mitochondria uptake calcium released from ER stores, which stimulates mitochondrial production of oxygen radicals in HBE. We tested the hypothesis that HBE inflammation-induced XBP-1 activation promotes Golgi and mitochondrial expansion. We first evaluated whether mucosal exposure of primary cultures of HBE to supernatant from mucopurulent material (SMM) from CF airways, a condition that activates XBP-1, up-regulated the mRNA expression of Golgi and mitochondrial markers. SMM increased the mRNA expression of several markers of these compartments by 1.4-4.3 fold at 6 hr, whereas at 24 hr their mRNA expression was maintained or decreased (n=4). To test whether these responses were coupled to XBP-1-induced Golgi and mitochondrial expansion, 16HBE14ocell lines stably expressing control vector, the active form of XBP-1 (spliced XBP-1) or a dominant negative XBP-1 (DN-XBP-1) were studied following 48 hr PBS or SMM exposure. The Golgi and mitochondria were stained with antibodies to p230 trans-Golgi and cytochrome oxidase and their expression studied by confocal immunofluorescence microscopy. Data are expressed as mean fluorescence intensity + SEM, n=6-13 culture sections from 2 experiments. In PBS-exposed cultures, p230 fluorescence was slightly increased by expression of spliced XBP-1 and not affected by the DN-XBP-1 ( Table 1) . Expression of spliced XBP-1 or DN-XBP-1 potentiated or blunted SMM-up-regulated p230 fluorescence (Table 1) . Cytochrome oxidase fluorescence was similar in PBS-exposed cultures expressing spliced XBP-1 vs. control cultures, but decreased in DN-XBP-1-expressing cultures (Table 1) . SMM up-regulated cytochrome oxidase fluorescence in control cultures, and this response was increased or blunted in cultures expressing spliced XBP-1 or DN-XBP-1 ( Airway surface dehydration and mucus accumulation are found in cystic fibrosis (CF); however, the complex relationship between defective ion transport and changes in mucus rheology is not well understood. It is thought that the airway surface is covered by an airway surface liquid (ASL), comprising a mucus layer (ML) and a periciliary fluid layer (PCL), overlying motile and sensory cilia at the apical membrane of epithelial cells. In CF, increased viscosity of the ML has been postulated to be an important factor in the impaired clearance of pathogenic bacteria and reduced function of innate immune mediators. The PCL is dehydrated in CF, but its viscosity is thought to be low. In order to study the determinants of ASL microviscosity in the ML and PCL separately, we measured the diffusion of fluorescently labeled macromolecules as a function of distance from the epithelial cell surface. Studies were performed by laser scanning confocal microscopy in primary cultures of human CF and non-CF bronchial epithelial cells that were grown for 21 days at an air-liquid interface. Experiments were made under steady-state conditions in unperturbed cell cultures, without prior washing, but covered by perfluorocarbon for use of a high-numerical aperture immersion objective. Diffusion of FITC-dextrans of different molecular sizes (10-2000 kDa) was studied by fluorescence recovery after photobleaching, in which time-resolved confocal imaging was performed at different depths in the fluorescently stained PCL (5 µm above the epithelial cell surface, confirmed by the presence of cilia in xy-scans) and ML (10 µm above the epithelial cell surface, no cilia observed). Relative diffusion of FITC-dextran in the ASL was size-dependent and slower in CF vs. non-CF cell cultures. In the PCL the relative diffusion coefficients (compared to FITC-dextran diffusion in saline) were D/Do = 0.54 (CF) vs. ~1 (non-CF) for 10 kDa FITC-dextran, and D/Do = 0.43 (CF) vs. 0.68 (non-CF) for 2000 kDa FITC-dextran. In the ML of CF cells, diffusion was further reduced, with D/Do = 0.4 for 10 kDa FITC-dextran and D/Do = 0.35 for 2000 kDa FITC-dextran. However, in non-CF cells diffusion in the ML was similar to that in the PCL. Studies of effects of CFTR and ENaC modulators are underway, as are measurements in ex vivo fragments of intact trachea. These first data on ASL microviscosity in human airway cultures provide evidence for CF-related increased viscosity of both the PCL and ML, and indicate solute size-dependent diffusion in the ASL of CF airways. This work was supported by the CFF, NIH and CFRI. Defects in the cystic fibrosis transmembrane conductance regulator (CFTR), including the ∆F508 mutation that impairs CFTR trafficking, are associated with increased activity of the epithelial sodium channel (ENaC). This is hypothesized to result in airway surface liquid depletion, decreased mucociliary clearance, and increased bacterial colonization of the airway seen in CF patients. Restoration of proper airway surface liquid in the CF airway will likely require correction of both CFTR function and decreased ENaC activity. Evidence suggests that CFTR and ENaC trafficking involve similar mechanisms (Wagner, Ott et al. 2001; Lang, Henke et al. 2003; Dieter, Palmada et al. 2004; Sato, Chapline et al. 2007 ), leading us to examine mechanisms of CFTR trafficking control in understanding ENaC trafficking regulation. In particular, we are interested in the influence of ∆F508-CFTR correctors on ENaC trafficking and functional expression. We hypothesize that ENaC trafficking can be regulated by the molecular chaperone proteins Hsp70 and Hsc70, which have altered expression in response to the prototype corrector, 4-phenylbutyrate, in a manner similar to their proposed regulation of ∆F508-CFTR (Rubenstein and Zeitlin 2000; Choo-Kang and Zeitlin 2001).

We performed Ussing chamber assays to examine the effect of Hsc70 and Hsp70 on amiloride-sensitive short-circuit current (Isc), as a readout of ENaC functional expression. Consistent with our previous work in Xenopus oocytes (Goldfarb, Kashlan et al. 2006) , our preliminary data in MDCK epithelial cells show an inhibition of ENaC functional and apical surface expression when Hsc70 is overexpressed by ~20%. This appears to result from more rapid removal of functional ENaC from the plasma membrane. In contrast, 1.8-fold overexpression of Hsp70, a chaperone protein closely related (85% identical) to Hsc70, has the opposite effect and enhances the functional activity and surface expression of ENaC. Hsp70 does not alter the rate of ENaC's retrieval from the apical membrane, suggesting that it acts to stimulate ENaC delivery to the apical membrane. Our data support the hypothesis that Hsc70 and Hsp70 have opposing effects on ENaC trafficking and functional expression in epithelial cells via actions at distinct sites along ENaC intracellular itinerary. This research is supported by grants from R01 DK-58046 to RCR and T32 DK-07748 to RAC. Large conductance, Ca +2 activated and voltage-dependent K + (BK) channels control a variety of physiological processes in different tissues. BK is encoded by a single gene (KCNMA1), but alternative splicing, regulatory beta-subunits, and post translational modifications contribute to diverse activity and function of BK. Ion transport regulation in lung epithelium contributes to the adequate water balance needed to preserve mucociliary transport. Although other K + channels are recognized as modulators of ion transport in response to nucleotides, the involvement of BK in this process has not been described. Since BK channels are sensitive to intracellular Ca +2 , we hypothesized that these channels may be involved in ion transport changes in response to ATP stimulation in normal human bronchial epithelial cells (NBE), a process at least partially mediated by Ca +2 . BK channels expression and function were studied by electrophysiology and confocal microscopy in NBE cells. K + currents were elicited upon depolarization using patch-clamp in cell-attached and inside-out configurations. A typical BK single conductance (290 pS) was detected in trypsinized NBE cells at intracellular Ca +2 concentration of 100 µM. Fifty percent of the macroscopic current was blocked by paxilline (40 nM) added to the apical side of NBE cells. Confocal microscopy showed immuno-reactive apical BK channels co-localizing with acetylated tubulin in the base of cilia. In Ussing chamber experiments conducted in the presence of a basolateral/apical K + gradient, basolaterally permeabilized cells respond to 10 µM apical ATP with a change in short circuit current (Isc) of 41.97 ± 5.165 µA (n=6). Apical paxilline blocked the Isc change in a dose-dependent manner (r = 0.9431, logIC50=1.496 ± 0.0648) with 31.37 nM paxilline causing half maximal inhibition. Small hairpin RNA (shRNA) knock down of KCNMA1 was accomplished by infection of NBE cells with lentiviruses (pLKO.1) encoding puromycin resistance and shRNA targeting KCNMA1 (NM_001014797 and NM_002247, TRC clone Id TRCN0000000212) or non targeting shRNA (SHC002). Infected NBE cells were selected with puromycin and grown until differentiation (~20 days). Relative mRNA expression was measured by qPCR with GAPDH as the gene of reference and was normalized to the average mRNA level in untreated cells. Knock down of KCNMA1 mRNA expression to 39 ± 8.1% of control caused a decrease of the ATP-induced Isc change (19.0 ± 1.482, n=4); in contrast, no significant reduction was recorded in cells infected with non-targeting lentiviruses (49.5 ± 8.383, n=4) or untreated (56.1 ± 7.130, n=5) controls. Finally, after four days without washing of the apical surface, knocked down cultures were dryer than control and non-target cells. We conclude that BK channels are present in the apical membrane of NBE cells and contribute to ion transport in response to ATP, by transporting K + to the apical compartment. Therefore, BK channels play an important role in airway water regulation. Decrease of KCNMA1 activity might be especially relevant for cystic fibrosis where water balance is already dysfunctional due to defective CFTR.

Supported by NHLBI.

localization of βarr2-GFP which is co-localized with cholesterol as shown by filipin stain. These data demonstrate that overexpression of βarr2 in normal epithelial cells results in CF-like intracellular cholesterol accumulation. In order to elucidate the full effect of βarr2 on CF-like cholesterol accumulation, 9HTEo-cells and two different CF model cells, pCEPR and IB3, were transiently transfected with a GFP-tagged βarr2 shRNA construct to effectively knock down βarr2 expression. Both the pCEPR and IB3 cells transfected with βarr2-GFP shRNA show an alleviation of the intracellular cholesterol accumulation which is similar to that seen in wildtype cells. To further analyze the role of βarr2 in cholesterol processing, total cellular cholesterol was measured. βarr2-overexpressing cells show nearly 2-fold increase in total cholesterol compared to their respective controls (p < 0.05). Since βarr2 has been shown to bind to the Na+/K+ ATPase (NKA) and facilitate endocytosis of the pump, and also that knockdown of the sodium pump is implicated in the development of perinuclear cholesterol accumulation, we studied the effect of βarr2 overexpression on the NKA. βarr2-overexpressing cells show a 6-fold increase in NKA compared to controls (p < 0.05) and an increase in its internalization as determined by immunostaining. These data suggest an involvement of NKA in βarr2-mediated cholesterol regulation; this relationship is currently being explored. We previously have shown that two signaling proteins, NOS2 and RhoA, exhibit CFrelated expression differences that are downstream consequences of cholesterol changes in CF. To directly test the impact of βarr2 on CF phenotypes, Cftr/βarr2 double knockout mice were developed. Nasal epithelium was excised from wt mice (Cftr +/+; βarr2 +/+), CF mice (Cftr -/-; βarr2 +/+), and DKO mice (Cftr -/-; βarr2 -/-),and analyzed for NOS2 and RhoA expression. The DKO mice showed nearly 80% recovery of NOS2 and 50% decrease in RhoA compared to CF mice suggesting a significant role of βarr2 in regulating the secondary manifestations of lost CFTR function; studies are on-going. In conclusion, these data implicate a regulatory role of βarr2 on cholesterol processing and the development of the cholesterol phenotype in CF. Cystic fibrosis related diabetes (CFRD) is an emerging CF complication, diagnosed in 20 to 40% of CF adults, and associated with both decreased lung function as well as a 6-fold increase in morbidity and mortality. We then hypothesized that CFRD-associated hyperglycemia contributes to epithelial cell dysfunction. In particular, we postulated that hyperglycemia could worsen the airway ion transport dysregulation, leading to abnormal airway surface liquid volume and mucociliary clearance. To test this assumption, we then investigated the impact of hyperglycemia on ion transport through differentiated bronchial epithelial cell monolayers.

CFBE bronchial cells, expressing wild-type CFTR (wt-CFBE) were cultured at air-liquid interface and then mounted in Ussing chamber for shortcircuit current (Isc) measurements. After current stabilisation, forskolin (Fk, 10 µM) and IBMX (100 µM) application induced a rapid increase in Isc (from 2 µA/cm 2 to 6.2 µA/cm 2 , before and after cAMP stimulation, respectively). Although the larger part of this Fk-induced Isc (I Fk ) was sensitive to CFTR Inh172 (20 µM), we noted that there was a CFTR Inh172 -independent residual Isc (30%). This portion of I Fk was inhibited by clofilium (100 µM), a blocker of the cAMP-activated KvLQT1 K + channel, which appears crucial for airway Clsecretion.

The I Fk was then compared through monolayers cultured for 24 h before Ussing experiments in culture medium containing physiological (5 mM glucose + 20 mM mannitol, control condition) or high glucose concentrations (25 mM). Interestingly, we observed that treatment with 25mM glucose reduced by ≈50% the I Fk (4.2 vs 2.3 µA/cm2, in mannitol and glucose treated monolayers, respectively). To determine if this decrease could be due to a direct down-regulation of CFTR channel activity, the basolateral membrane of wt-CFBE monolayers was permeabilized with amphotericin B (7.5 µM) and a basolateral-to-apical Clgradient was established. In these conditions, high glucose failed to inhibit the I Fk . Furthermore, CFTR expres- We have previously demonstrated that cAMP response element binding protein (CREB) exhibits increased phosphorylation and activation in cystic fibrosis (CF) cells and tissues. This finding was consistent with our hypothesis that increased cAMP signaling in CF cells leads to a number of CFrelated phenotypes including intracellular cholesterol accumulation. We demonstrated that inhibition of cAMP signaling with Rp-cAMPS in CF cells results in the reversal of the cholesterol phenotype supporting this general hypothesis. However, when examining the effect of Rp-cAMPS treatment on CREB regulation, it was observed that PKA inhibition had no impact in CF samples. We believe that the regulation of CREB is key to a number of cell regulatory cascades that contribute to cellular events that characterize CF including inflammation and redox regulation. Therefore, the goal of this study is to determine how the CREB pathway is regulated in the absence of CFTR function. The hypothesis of the study is that increased expression of the protein beta-arrestin-2 (barr2) leads to CREB activation in CF cells through a PKA-independent mechanism. Arrestins are multifunctional proteins that influence several cell regulatory pathways. The function initially identified was as regulators of G-protein coupled receptors (GPCR) in concert with G-protein coupled receptor kinases (GRKs). Arrestins and GRKs act as silencers of a variety of GPCRs such as β-adrenergic receptors (β-AR). To begin testing this hypothesis, barr2-GFP expressing 9/HTEo-cells (barr2-GFP) were examined for pCREB content compared to GFP expressing control cells. Expression of pCREB normalized to total CREB (tCREB) content in barr2-GFP cells was 2.3 +/-0.3 fold increased compared to controls (n=9, p < 0.05), similar to what we observed in multiple CF models. Based on these data, a candidate pathway by which barr2 could regulate CREB activation would be the Src/ERK cascade. Both Src and ERK exhibit increased activation in barr2-GFP cells compared to GFP cell controls as determined by Western blot. The ability of Src inhibitor (PP2) and two MEK/ERK inhibitors (PD98059 and UO126) to inhibit cAMP response element (CRE)-luciferase activation was examined in barr2-GFP and CF model cells. In both cell models, Src and ERK inhibition significantly diminished CRE activation. To test directly the impact of barr2 on CREB activation in CF, Cftr/barr2 double knock-out mice were developed and analyzed for pCREB expression. Nasal epithelium was excised from wt mice (Cftr +/+;barr2 +/+), CF mice (Cftr -/-; barr2 +/+), and DKO mice (Cftr -/-;barr2 -/-). Ratios of pCREB/tCREB are reduced 75% in DKO mice compared to CF mice (n=4, p < 0.001). These data demonstrate directly that barr2 expression in CF is impacting the CREB signaling pathway. Other evidence suggests that predicted CREB signaling consequences are also reverted towards wt levels, although this work is ongoing. It is concluded from this study that CREB regulation in CF is mediated directly through barr2 expression via the Src/ERK pathway.

A consequence of lost CFTR function is specific alterations in cholesterol homeostasis. These alterations include intracellular cholesterol accumulation, increased membrane cholesterol content, and increased cholesterol synthesis. How these aberrations relate directly to CFTR function is not known. Previously we demonstrated that CF cells exhibit at least a 2fold increase in beta-arrestin-2 (βarr2) expression compared to non-CF controls. The hypothesis of this study is that a cellular feedback response to lost CFTR function leads to increased βarr2 expression and that βarr2 mediates cholesterol phenotypes associated with CF. To examine this, 9HTEo-cells were stably transfected with the control and βarr2-overexpressing pEGFP-N1 constructs, which express GFP or βarr2-tagged GFP (βarr2-GFP), respectively. Upon imaging, the GFP cells show a mostly nuclear localization of GFP, whereas the βarr2-overexpressing cells show a perinuclear sion, measured by western blotting, was not significantly different in either mannitol or glucose-treated wt-CFBE cells. We then postulated that the decrease in I Fk previously observed in intact monolayers could be due to an indirect effect through basolateral K + channels. The apical membrane of wt-CFBE was then permeabilized and an apical-to-basolateral K + gradient was established. Interestingly, we observed that the clofilium-sensitive, KvLQT1 basolateral K + current, was lower in glucose than mannitol pre-treated monolayers.

In summary, hyperglycemia could indirectly affect Clsecretion through the modulation of K + currents in bronchial monolayers. The effect of hyperglycemia and glucose fluctuations on ion and fluid transport through airway epithelia should be further studied to better understand the lung function deterioration associated with CFRD.

O. Bardou is supported by a studentship from FESP, University of Montreal, the project is funded by the Canadian Cystic Fibrosis Foundation. 1 1. Cystic Fibrosis Center, University of North Carolina, Chapel Hill, NC, USA; 2. Pharmacology, University of North Carolina, Chapel Hill, NC, USA; 3. Cell Biology, Emory, Atlanta, GA, USA Ano1, also known as TMEM16A, has recently been identified as the Ca 2+ activated Clchannel that is the apical membrane conduit for Clsecretion in CF airway epithelia. Since little is known about this protein, we used fluorescent resonance energy transfer (FRET) to test the hypothesis that Ano1 exists as a multimer in the plasma membrane. We C-terminally labeled Ano1 with either eGFP (FRET donor) or mCherry (FRET acceptor) and these constructs both displayed identical I-V relations and sensitivity to Ca 2+ as untagged Ano1 when expressed in HEK cells, suggesting that appending a fluorescent protein to the C-terminus did not affect Ano1 function. We then measured the % FRET by the acceptor photobleaching method using a Leica SP5 confocal microscope. When coexpressed in either primary human bronchial or HEK293 cells, Ano1-eGFP and Ano1-mCherry were clearly visible in the plasma membrane and underwent 8% FRET (n= 22 and 56 respectively). In contrast, Ano1-mCherry did not FRET with eGFP alone (n=19) and underwent a small, but significant amount of FRET with either A2BR-eGFP or with CFTR-eGFP of ~1.5% (both n=20; p<0.05). This FRET was significantly less than the 8% FRET seen between Ano1 constructs. Biochemical analysis revealed that Ano1-mCherry coimmunoprecipitated with Ano1-eGFP but did not co-immunoprecipitate with either eGFP alone or with A2BR-eGFP, suggesting that the Ano1-Ano1 FRET represents a specific physical interaction. The Ano1-Ano1 FRET was also ~8% when measured in intracellular compartments (n=36) and was not altered by addition of thapsigargin (n=27), which elevates intracellular Ca 2+ , suggesting that Ano1 association occurs early in the biosynthetic process and remains constant during channel activation. Furthermore, application of cytochalasin D, which disrupts the actin cytoskeleton, also failed to alter Ano1-Ano1 FRET (n=35). Since the Ano1 FRET was specific and constant, we used FRET to gain more detailed information about Ano1 stoichiometry. To this end, we varied the ratio of donor/acceptor cDNA expressed, whilst keeping the total amount of cDNA constant at 2 µg/ml and measured the resultant changes in FRET. There was a ratio-dependent increase in FRET which asymptoted at 8%. When both donor and acceptor molecules are present at equal concentrations, there is a finite probability that (i) all donors, (ii) all acceptors, or (iii) a mix of the two will combine. Thus, a mathematical model that describes the probability that an eGFP (donor) subunit is contained in a multimer with none, one, two or three mCherry (acceptors) can be compared to the actual FRET data. Based on this probability model, the shape of the curve is influenced by the number of subunits in the complex and can be used to determine the stoichiometry. Using this approach, we found that our data was best fit by the trimer probability model. In conclusion, we have demonstrated that Ano1 undergoes FRET in the plasma membrane of airway epithelial and HEK cells and most likely exists as a homotrimer. Whether or not these subunits interact and bind with other proteins remains to be determined.

Funded by the NIH. The thin layer of liquid at the surface of airway epithelium, the airway surface liquid (ASL), is important in normal airway physiology and in the pathophysiology of cystic fibrosis. The ASL is thought to consist of a mucus layer overlying a periciliary fluid layer. At present, the best method to measure ASL depth involves scanning confocal microscopy after staining with an aqueous-phase fluorescent dye. We designed and validated a simple, noninvasive imaging method to measure ASL depth by reflectance imaging of an epithelial mucosa in which the surface is illuminated at a 45-60 degree angle by an elongated 13-µm wide rectangular beam produced by a 670 nm diode micro-focus laser. The principle of the surface laser reflectance microscopy (SLRM) method is that air-liquid, liquid-liquid and liquid-cell interfaces produce distinct specular or diffuse reflections that can be imaged to give a micron-resolution replica of the mucosal surface. SLRM was validated using fluid layers of specified thicknesses, and in cell cultures by agreement with measurements using z-scanning confocal microscopy. The SLRM method was applied to measure ASL thickness in native, unperturbed cell cultures and in ex vivo fragments of pig trachea. ASL thickness was initially measured in well-differentiated bronchial cell cultures after washing, addition of a small volume of fluid, and waiting 6-12 h to reach a steadystate ASL thickness. SLRM showed an ASL thickness of 8.8 ± 0.2 µm, which agreed with confocal microscopy and many prior measurements of similarly prepared cultures. In contrast, SLRM in native, unperturbed cell cultures gave a wide range of ASL thicknesses (7-20 µm, average 11 ± 3 µm), which is likely due to a heterogeneous mucus layer. ASL measurements were also done in ex vivo fragments of pig trachea, giving an average ASL depth of 10 ± 3 µm. SLRM was adapted to measure transepithelial fluid transport from the dynamics of fluid layer thickness. Human bronchial epithelial cell cultures absorbed fluid over 4-6 hrs, which was inhibited by amiloride. SLRM measurements in pig trachea showed rapid apical fluid secretion (0.15 µL/min/cm 2 ) after stimulation by forskolin, IBMX and carbachol. The SLRM method was also adapted to measure transepithelial osmotic water permeability. Using mannitol (300 mM) in the basolateral perfusate, the transepithelial osmotic water permeability coefficient (P f ) in bronchial epithelial cells was P f = 48 ± 7 µm/s, which was not significantly different when the osmotic gradient was reversed (apical mannitol), P f = 42 ± 8 µm/s. Compared with confocal microscopy, ASL thickness measurement by SLRM does not require dye staining or costly instrumentation, and can potentially be adapted for in vivo measurements using fiberoptics.

Supported by the CFF and the NIH.

Airway epithelial injury and remodeling is responsible for the progressive and irreversible deterioration of lung function in CF patients. Thus, a better comprehension of epithelial repair mechanisms is crucial to develop new therapeutic strategies to stabilize CF patients. Our previous work (Trinh et al. AJP 295(5) : L866-80, 2008) , using a model of mechanical injury, showed that repair mechanisms of bronchial monolayers were highly dependent on EGF response and KvLQT1, K ATP and KCa3.1 K + channel function. We also demonstrated that bronchial CF cells (CuFi-1) exhibited a delay in repair after injury when compared to normal bronchial cells . This delay was associated with a lower EGF responsiveness and an altered K + channel function.

We then hypothesized that the CF bronchial epithelia could exhibit a defect in repair mechanisms and we further investigated the contribution of CFTR channels in that defect. Our first objective was then to compare the repair after mechanical injury of human bronchial cells isolated from lung specimens obtained from CF patients and healthy donors during lung trans-beat in HBE cells. Application of a small nebulized dose of HS resulted in near-complete ciliostasis by minutes into the treatment course, as measured by both the cilia beat frequency and the % of cells involved in cilia beat. Subsequent studies suggest that slower nebulization rates may not result in the same degree of ciliostasis. With the advent of more rapid delivery devices in use for delivery of HS, understanding the impact of HS delivery at higher rates is necessary to minimize the surprisingly negative effect of HS on CBF.

The Epithelial Na + Channel (ENaC) is a key protein involved in the regulation of airways function. In the cystic fibrosis (CF) airway, hyperactivity of ENaC is hypothesized to cause impaired mucociliary clearance and portend bacterial colonization of the airway. In our group, the studies of correction of ∆F508-CFTR trafficking by 4-phenylbutyrate (4PBA) led us to identify a novel 29 kDa molecular chaperone, ERp29, that was present in the lumen of the endoplasmic reticulum (ER) and that had in CF epithelial cells increased abundance after treatment with 4-Phenylbutyrate. We tested the hypothesis that ERp29 would regulate intracellular trafficking of ∆F508-CFTR in CF epithelial cells and wt-CFTR in human colon adenocarcinoma T84 cells. ERp29 is homologous to thioredoxin/PDI proteins, which are enzymes that catalyze the formation and rearrangement of disulfide bridges. However, ERp29 has a single cysteine residue and lacks the cysteine sequence motif (C-X-X-C) that is characteristic of thioredoxins. In order to test how this motif influenced ERp29 function, our group constructed a plantation. Our results revealed that CF bronchial monolayers repaired slower (55%) than non-CF. We then decided to define the role of CFTR channel in that phenomenon, by testing the impact of CFTR down-regulation in non-CF cells, and conversely, by exploring a possible beneficial effect of CFTR correction in CF cells. Therefore, we showed that CFTR silencing in non-CF bronchial cells, with CFTR-siRNAs, partially inhibited wound healing after injury. In addition, we made an interesting observation: wound repair in IB3 CF bronchial cells corrected with wt-CFTR (S9 cells) was enhanced by 127 % when compared to non-corrected IB3 (CF) cells. Similarly, wound healing in CFBE cells over-expressing wt-CFTR (wt-CFBE) was enhanced by 148 % when compared to CFBE cells over-expressing ∆F508-CFTR (∆F508-CFBE). Furthermore, we noted that CFTR correction was accompanied with an increase in K + currents in wt-CFBE monolayers. More precisely, short-circuit current measurements were performed in Ussing chamber on apical membrane permeabilized CFBE monolayers, in the presence of an apical-to-basolateral K + gradient. Our results revealed that the basolateral clofilium-(KvLQT1) and glibenclamide-sensitive (K ATP ) currents as well as the 1-EBIO-activated (KCa3.1) current were 3-7 fold higher in wt-CFBE than in ∆F508-CFBE monolayers.

Altogether our results suggested that functional CFTR and K + channels play a critical role in the epithelial repair capacity, which is crucial for effective airway regeneration in CF patients. NTN Trinh is supported by a CCFF studentship and the project was funded by a CCFF operating grant.

Adequate mucociliary clearance (MCC), a crucial factor in the maintenance of normal airway health, is dependent on mucus rheology, airway hydration, and cilia beat frequency (CBF). Control of cilia beat frequency (CBF) is mediated through various intracellular messengers as well as pharmacologic agents such as beta-agonists. Equally important, however, are pharmacotherapeutics that may limit CBF. Data from several otolaryngology studies indicate that application of topical hypertonic saline (HS) to the nasopharynx results in significant inhibition of CBF. These studies were done with very large concentrations of salt rapidly deposited on the airway surface. Little is known about the effects of hypertonic solutions on cilia beating in human airway epithelial cells at deposition rates consistent with inhalation nebulizers (i.e., in the range of nanoliters/min/cm 2 ). In this study, we sought to evaluate the effect of nebulized HS on CBF in airway epithelial cells.

Well-differentiated human bronchial epithelial cells (non-CF) under airliquid interface were imaged using phase contrast video microscopy to record CBF and the relative density of cilia beating in the culture (% area). Baseline CBF was recorded for 2 min and then 7% HS was applied to the surface of the cells using a novel nebulizer delivery system designed to deliver a clinically relevant nanoliter volume of fluid (~100 nl/min). Both CBF and % area were recorded every 30 sec for the duration of the 10 min nebulization as well as during the recovery period (minimum of 20 min) after nebulization.

Initial CBF averaged 5.02 ± 1.16 beats per minute (bpm) and % area involved averaged 44.6 ± 8.3%. At an average of 12.4 minutes (range 2-24.5 min) after the start of 7% HS nebulization, CBF had decreased to 2.7 ± 0.32 bpm. More importantly, the % area decreased to 4.0 ± 2.9%, suggesting that most cilia had completely stopped beating. By an average of 18.3 min after the nebulizer treatment had ended (range 10-33 min), CBF had recovered to a mean frequency of 4.22 ± 0.51 bpm but % area remained low at 6.4 ± 3.2%. Additional preliminary studies indicate that CBF is a dose-response phenomenon; when the nebulization rate is tripled to ~300nl/min, CBF and % area both decrease rapidly.

Hypertonic saline is a clinically proven therapy that can improve MCC in CF patients, purportedly by drawing water onto the airway surface. However, topical hypertonic saline has been shown to decrease CBF in human nasopharynx. We sought to determine the effect of nebulized HS on cilia mutant where this strictly conserved cysteine was mutated (C157S). Transfection of ERp29 C157S significantly decreased CFTR whole cell and surface expression in T-84 cells. These data suggest that C157 plays a critical role in ERp29's regulation of CFTR trafficking. Based on the similar trafficking pathway between CFTR and ENaC, we hypothesized that ERp 29 would promote ENaC functional expression, and that ERp29 C157S would inhibit ENaC trafficking. In Xenopus laevis oocytes, co-injection of ERp29 RNA with αβγ-ENaC RNA caused a ~2-fold increase in amiloride-sensitive whole oocyte current compared with oocytes injected with αβγ-ENaC RNA alone, suggesting that ERp29 can regulate ENaC functional expression. We then completed a selection of ENaC-overexpressing MDCK cell lines with tetracycline-inducible expression of wt or C157S ERp29 as an epithelial cell based model. Both wt and C157S ERp29 expression increased with increasing doxycycline concentrations (0, 0.005, 0.01 and 0.05 µg/ml). We established conditions by which these cells form high resistance polarized monolayers when grown on permeable supports that will allow assessment of ENaC function in Ussing chambers. Preliminary data suggest that full length (uncleaved subunit) γ-ENaC whole cell expression increased with increased ERp29 C157S expression (0.05 µg/ml). Interestingly, this increased C157S ERp29 expression was associated with decreased amiloride sensitive current in Ussing chamber experiments. These data support the hypothesis that ERp29 can regulate ENaC functional expression by regulating both trafficking or subunit proteolysis, and that the C157 may be a critical residue in this regulation.

This study was supported by grants from R01 DK58046 and R01 DK73185 to RCR. Cystic fibrosis (CF) is characterized by acute pulmonary exacerbations (APE) with reduced FEV1. The CF nasal airway exhibits a similar ion transport defect as the lung, and colonization, infection, and inflammation within the nasal passages are common among CF patients. In addition, chronic rhinosinusitis (CRS) has been postulated to contribute to CF pathogenesis, including its role as an early marker of descending pulmonary infection. Nasal lavage fluid (NLF) is a minimally invasive means to collect upper airway samples, and IL-8 has been suggested as a potential biomarker of CF pathogenesis in some but not all studies. The Luminex platform provides a ready means for rapid screening of the cytokine profile of biologic fluids, providing a new means to assess NLF. We collected NLF from CF patients at the onset and resolution of CF APE and compared the Luminex Bio-Rad 27 human 27-plex profile to stable CF outpatients and normal controls. We identified elevated levels of IP-10 in subjects hospitalized for APE (2582 pg/mL [95% CL of mean: 818, 8165], N=13) which significantly decreased (647 pg/mL [357, 1174] P<0.05, N=13) following antimicrobial therapy (mean duration 16 ± 0.6 days). Stable CF outpatients exhibited intermediately elevated levels (680 pg/mL [281, 1644] , N=13)) that were not significantly different than normal controls (342 pg/mL [110, 1061]; P=0.3, N=10) but less than CF inpatients upon admission (P=0.056). All comparisons were log-tranformed based on distribution of IP-10 levels in biologic samples. Resolution of elevated IP-10 was associated with concomitant improvement in FEV1% (46 ± 6 at admission vs 55 ± 8 upon discharge, P<0.05). In contrast to IP-10, no increase in other cytokine levels were observed in any of the CF groups, which included TNF, IL-8, and IL-1, among 27 cytokines evaluated. To add confidence to the clinical relevance of our findings, we tested IP-10 levels in the BALF of CF patients. Elevated IP-10 was observed in CF BALF (2673 pg/mL [1306, 5458] , N=10) which was significantly greater than healthy post-lung transplant patients (8.4 pg/mL [0.03, 2172] , N=5, P<0.001). Similarly elevated IP-10 levels were observed in the BALF of disease controls which included recurrent pneumonia, cough, and tracheomalacia, suggesting IP-10 elevation may not be unique to CF APE (267 pg/mL [36, 1968] , N=20, P=0.05 post-lung transplant controls). IP-10 levels in the basolateral compartment of fully-differentiated nasal airway epithelial monolayers derived from CF mice exposed to PAO-1 (bacteria-free prep from 20 hour log-phase growth) or LPS (100 nm) apically for 24 hours were significantly elevated compared with vehicle controls (1159 ± 147, N=13, P<0.001 for PAO-1 and 1373 ± 191 N=11, P<0.001 for LPS vs. 305 ± 68 for vehicle control specimen (N=13)), suggesting IP-10 levels in NLF/BALF may be due to stimulation of epithelia with bacterial pathogens. We conclude that IP-10 is elevated in the nasal lavage of CF patients with APE and responds to antimicrobial therapy. Further exploration of IP-10 as a potential biomarker and the pathogenic role of this TH-2 cytokine is warranted.

Funded by the NIH and the CFF. The functional expression of the epithelial sodium channel, ENaC, is elevated in cystic fibrosis (CF) airway epithelia, but the mechanism by which this occurs is not clear. We tested the hypothesis that correction of ∆F508-CFTR (cystic fibrosis transmembrane conductance regulator) would alter the functional expression of endogenously expressed human ENaC in the CFBE41o-model of CF epithelia. Functional expression of ENaC, as defined by amiloride-inhibited short circuit current (I sc ) in Ussing chamber experiments was absent under control conditions, but present in CFBE41oparental and ∆F508-CFTR-overexpressing cells after treatment with 1 µM dexamethasone (Dex) for 24 h. Application of trypsin to the apical surface of the epithelia to activate uncleaved, "near silent" ENaC caused an additional increase in amiloride-sensitive I sc in the Dex-treated cells, and was without effect in the control cells, suggesting that Dex caused increased expression of ENaC at the cell surface. We tested the influence of two strategies to improve ∆F508-CFTR trafficking, incubation at reduced temperature (25 o C) or with 1 mM sodium 4-phenylbutyrate (4PBA), on this endogenous ENaC functional expression. These methods of ∆F508 correction resulted in miminal CFTR-mediated I sc 's in the parental cells, and in modest CFTRmediated I sc 's in the ∆F508 overexpressing cells. Amiloride-sensitive I sc remained absent from Dex-treated CFBE41o-parental or ∆F508-overexpressing cells that had been incubated at reduced temperature, and these reduced temperature-incubated cells also failed to have an increase in amiloride-sensitive I sc with trypsin. In contrast, incubation with 4PBA was without significant effect on ENaC functional expression in Dex-treated CFBE41o-parental cells, but synergistically increased Dex-induced ENaC functional expression in ∆F508-CFTR overexpressing cells. This synergistic increase was not associated with change in the rate at which functional ENaC was retrieved from the apical surface, or in the fraction of total ENaCmediated current that could be stimulated by trypsin. These data suggest that 4PBA and overexpressed ∆F508 synergistically stimulated ENaC delivery to the apical membrane, and further suggest that the mechanisms by which reduced temperature and 4PBA correct ∆F508 trafficking have distinct effects on ENaC trafficking. These are also consistent with the hypothesis that lack of regulation of ENaC trafficking by CFTR is one potential mechanism by which ENaC becomes hyperactive in CF airway epithelia.

Supported by R01 DK58046, R01 DK73185 and the CFF. CF airways disease is characterized by inflammation and mucus overproduction. Airway inflammation induces endoplasmic reticulum stress mediated by activation of Inositol Requiring Enzyme 1 (IRE1α or β). IRE1α is functionally relevant for airway epithelial Th1 cytokine secretion.

Our findings now suggest that IRE1β regulates mucin production during airway inflammation. IRE1β is expressed in Clara and mucous cells, but not in ciliated or alveolar type II cells; moreover, airway inflammation-increased mucin production is blunted in IRE1β -/mice (Pediatric Pulmonol. Suppl. secretion and mucin secretion. We conclude that, in cultures of human tracheobronchial gland cells, mucin secretion is independent of Clsecretion and unaltered in CF, that the ratio of Clsecretion to mucus secretion varies markedly depending on mediator, and that secretions induced by stimulation of β-adrenergic receptors will be abnormally concentrated in CF.

A MODEL OF WATER AND ION TRANSPORT IN AIRWAY EPITHELIA: CF VS. NORMAL CELLS Garcia, G.J. 1,2 ; Boucher, R.C. 1 ; Elston, T.C. 2 1. CF Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 2. Pharmacology, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Experimental evidence suggests that the absence of functional CFTR channels leads to airway surface liquid (ASL) dehydration, impaired mucociliary clearance and persistent lung infections in cystic fibrosis (CF) patients. The complexity of the regulatory system that maintains ASL homeostasis presents a major challenge for understanding the pathophysiology of CF and therefore hinders the development of effective therapies. To overcome this challenge, mathematical models can be developed that quantitatively describe lung epithelia electrophysiology and the signaling pathways that regulate ASL hydration. Although models of epithelial transport have been reported, a full characterization of human airway epithelia electrophysiology is still missing. In particular, the distinct bioelectric properties of CF vs. normal epithelia have not yet been incorporated into a predictive model. We developed a mathematical model of airway epithelial cell electrophysiology that has three compartments (ASL, cytoplasm, and plasma) and incorporates the three main ions contributing to cell osmolarity (Na + , K + , and Cl − ), ion transport due to electrochemical gradients and water transport due to differences in osmotic pressure. Microelectrode measurements performed in primary cell cultures of human nasal epithelium were used to fit model parameters. The effects of amiloride, bumetanide and ouabain on the steady-state bioelectric properties were investigated. The model successfully reproduced the bioelectric properties of both CF and normal airway epithelia, including ion concentrations, membrane potentials, transepithelial resistance, short-circuit current and cell volume. Importantly, the model parameters that best fit the experimental data were consistent with known differences between normal and CF epithelia, namely CF cells have (a) lower apical Cl − permeability, (b) higher apical Na + permeability and (c) higher expression of the Na-K pump than normal cells. Finally, the model provided predictions that can be tested experimentally, such as the prediction that CF cells have higher basolateral K + permeability than normal cells. To achieve a comprehensive understanding of ASL height regulation, the next step in this research project will be to incorporate ion channel (ENaC, CFTR, CaCC and CaKC) regulation by extracellular nucleotides (ATP, Adenosine) via membrane receptors (P2Y 2 and A 2b ) using our previously published model of extracellular nucleotide regulation [Zuo et al. (2008) J. Biol. Chem. 283, 26805] .

Illek, B. 1 ; Meiss, L. 1 ; Modlin, S. 1 ; Cohen, F. 1 ; Gruenert, D.C. 2,3 ; Fischer, H. 1 1. Nutrition and Metabolism, Children Hospital and Research Center Oakland, Oakland, CA, USA; University of California, San Francisco, San Francisco, CA, USA; 3. Department of Laboratory Medicine, University of California San Francisco, San Francisco, CA, USA Brief hyperthermic treatments (BHT; 41-43 o C for 30-60 min) of the airways have been applied in clinical studies for the common cold and are known to increase the expression of cytoprotective heat shock proteins. The inducible Hsp72 (iHsp72) is one of the most highly inducible heat shock proteins and has been shown to restore normal fluid transport in response to cAMP-dependent stimulation in an animal model (Pittet et al., J. Physiol. 538, 583-597, 2002) . The goal of this study was to use CFTR-corrected CF bronchial epithelial cell monolayers (wtCFTR-CFBE41o-) to establish a heat shock treatment protocol for i) examination of the effect that BHT has on CFTR Cltransport and ii) evaluation of its putative protective role by 32, 259, 2009 ). This study was conducted to further evaluate the role of IRE1β in the regulation of mucin production. Utilizing Celsius, a Gene Expression Tool (http://genome.ucla.edu/projects/UGET), we found that IRE1β, but not IRE1α, exhibits a strong association with genes involved in mucous cell function and mucus production. A strong correlation of IRE1β gene expression was found with a) SAM pointed domain containing ets transcription factor (SPDEF), which is required for pulmonary goblet cell differentiation and regulation of genes associated with mucus production; b) anterior gradient homolog 2 (AGR2), a protein disulfide isomerase essential for production of intestinal mucus; and c) CLCA1/3 (GOB-5), a member of the calcium-activated Clchannel family associated with mucin gene regulation and goblet cell hyperplasia. Other genes highly correlated with IRE1β gene expression included several enzymes predicted to be involved in mucin glycosylation. To further address the role of IRE1β in mucin production, we used Calu-3 cells, an airway epithelial cell line that expresses mucous cells and secretes gel-forming mucins. In Calu-3 cultures, IRE1β mRNA was upregulated by interleukin-13 (IL-13), a response coupled to mucous cell metaplasia. Baseline and IL-13-induced IRE1β mRNA levels were knocked-down by ~60% in Calu-3 cultures stably expressing pSIREN IRE1β shRNA, an effect coupled with inhibition of IL-13-induced mucus cell metaplasia. In addition, while IL-13 up-regulated MUC5AC, MUC5B, SPDEF and AGR2 mRNA levels in control cultures, this effect was blunted in cultures expressing the IRE1β shRNA (Table 1 ; data correspond to mean + SEM of fold changes over control cultures at t=0; n=2-3). Finally, overexpression of IRE1β in Calu-3 cells increased the basal expression of MUC5AC. We conclude that 1) IRE1β is a key regulator of airway mucin production and 2) this IRE1β function results from, at least in part, IRE1βdependent regulation of genes associated with mucin production. We are currently investigating IRE1β-regulated airway mucin production in models relevant to CF. These studies may lead to the use of IRE1β inhibitors as therapeutic agents for CF airways disease. We investigated how cystic fibrosis (CF) affects the relationship between Cland mucin secretion in cultures of human tracheobronchial gland mucous (HTGM) and CF tracheobronchial gland mucous (CFTGM) cells. Biochemical studies showed that the cultures secreted typical airway mucins, and molecular studies showed that the cells expressed genes encoding MUC1, MUC4, MUC5B, MUC15, MUC16 and MUC20. An enzymelinked immunosorbant assay (ELISA) was used to measure mucin secretion, and short circuit current was used to estimate Clsecretion. Effects of cumulative doses of methacholine (MCH), phenylephrine (PHE), isoproterenol (ISO) and ATP on mucin and Clsecretion were studied. Baseline mucin secretion was not significantly altered in CF, nor were the increases in mucin secretion induced by mediators. In both CF and non-CF cells, mediators (at 10 -5 M) increased mucin secretion in the rank order ATP > PHE = ISO > MCH. They increased Clsecretion in the sequence ATP > MCH ≈ ISO > PHE. Responses of CF-HTMG to MCH, ATP and PHE were not significantly different from those of HTMG. By contrast, the response of CF-HTMG cells to ISO was significantly reduced to ~20% that of HTGM cells. Across mediators, there was no correlation between the maximal increases in Cl -mitigating the loss of CFTR function by the Pseudomonas aeruginosa virulence factor pyocyanin (Schwarzer et al., Free Radic. Biol. Med., 45: 1653 -62, 2008 . Western blot analysis showed that BHT caused iHsp72 protein expression to increase in a time-dependent fashion by a factor of 1.2-fold (2 hours), 1.7-fold (4 hours), 3.1-fold (6 hours), and 4.1-fold (24 hours). Transepithelial, CFTR-mediated chloride transport (I Cl ) of confluent, heatshocked epithelia was further investigated in Ussing chambers. BHT caused a gradual hyper-activation of forskolin-stimulated I Cl that progressively increased from 57±6 µA/cm 2 (n=12) to 153±9 µA/cm 2 (n=3) after 4 hours, and to 225±33 µA/cm 2 (n=3) after 6 hours. The largest stimulatory response to forskolin was obtained at 24 hours after recovery from BHT exposure with an average I Cl of 269±40 µA/cm 2 (n=4). The BHT-induced hyperactivation of the forskolin-stimulated I Cl was completely blocked by CFTRinh-172 (20 µM) applied apically. BHT effects were absent in non-complemented CFBE41o-monolayers homozygous for ∆F508 CFTR indicating that BHT specifically activated the CFTR-mediated Clconductance. The apical application of the virulence factor pyocyanin (100 µM) caused a gradual decline of forskolin-stimulated I Cl up to 63% in wtCFTR-CFBE41o-monolayers that were not exposed to BHT. In contrast, prior exposure to BHT prevented the loss of forskolin-stimulated I Cl by pyocyanin. Over the course of 2-3 hours, pyocyanin caused I Cl to drop to 25±3 µA/cm 2 (n=9) in untreated controls, whereas I Cl was stimulated to 181±35 µA/cm 2 (n=7) in BHT-treated monolayers. These data suggest that activation of the heat shock response in CFTR-corrected CF bronchial epithelial cells hyper-stimulated CFTR activity that protected against the loss of cAMP-dependent CFTR Cltransport in presence of the cytotoxic virulence factor pyocyanin.

Supported by CFF, CFRI, PACFI, Elizabeth Nash Foundation. Mucociliary clearance (MCC) depends critically on the volume and rheological properties of a thin airway surface liquid layer, ciliary beating and balanced mucin secretion. All components of MCC are under control of purinergic signaling. Its first step involves nucleotide release by surface epithelium, however, the stimuli and pathway(s) contributing to nucleotide release remain incompletely understood. We have previously demonstrated that ATP release from alveolar A549 cells is a mechano-sensitive, Ca 2+dependent process that is strongly stimulated by low volume of liquid covering cells. Such a stimulus involves cell mechanical deformation by tension forces at the air-liquid interface (ALI) leading to elevation of intracellular Ca 2+ and exocytotic ATP release. In this study we examined whether airway epithelial cells are also susceptible to such mechanical stimulation and, if different manipulations that expose cells to tension forces of ALI will produce ATP release. Human bronchial epithelial 16HBE14ocell monolayers grown on glass coverslips were mounted in a flow-through chamber and exposed to ALI by brief introduction of an air bubble into the perfusion chamber, or by tilting an open culture dish (6-well plate) with cells covered by small volume of media. Intracellular Ca 2+ was determined by Fura-2 fluorescence imaging, ATP content in the extracellular medium was measured by luciferase assay. Regardless of experimental approach, tension forces acting on 16HBE14ocells at the ALI consistently induced significant intracellular Ca 2+ elevation and ATP release. The latter was not affected by removal of extracellular Ca 2+ but was reduced by ~70% in cells loaded with BAPTA-AM. ATP release was completely abolished in N-methylmalemidetreated cells, suggesting involvement of a Ca 2+ -dependent exocytosis. Ethidium bromide staining did not disclose any cell damage in these experiments. Confocal fluorescence microscopy study showed reversible cell deformation (flattening) in the monolayer part that was in direct contact with the air bubble and confirmed that cell integrity was entirely preserved.

Conclusions: Airway epithelial cells are susceptible to mechanical deformation by tension forces at the ALI and respond by intracellular Ca 2+ elevation and Ca 2+ -dependent ATP release. We propose that a similar mechanism may operate in vivo in the airways, where tension forces acting directly on locally exposed epithelial cells may provide a mechanism to control mucociliary clearance via mechano-sensitive Ca 2+ -dependent ATP release and purinergic modulation of fluid and mucin secretion. This mechanism may be defective in cystic fibrosis due to excessive mucus layer covering the airways.

( Mucin proteins are present in abundance in the luminal airway space, and they are considered highly relevant to cystic fibrosis (CF) lung disease pathogenesis. Several mucin species can be found in airway secretions in both normal and CF lungs by proteomics analyses, including MUC5AC, MUC5B, and MUC16. Of these mucins, MUC5AC and MUC5B are generally considered the most important gel-forming mucins that contribute to the rheological properties of the mucus layer in both normal and CF lungs. MUC16 is the most abundant sulfated mucin, and, as such, may be altered in CF lungs, where mucin sulfation is known to be increased. The specific roles of each of these mucins in normal lung and the relative importance of the mucins in the development of disease are not fully understood. In Scnn1b-Tg mice, airway-targeted overexpression of the β subunit of the epithelial sodium channel (βENaC protein, Scnn1b gene) recapitulates many of the pathological features of human CF pulmonary disease, including airway surface dehydration, development of mucus plugs, impaired mucus clearance, early bacterial burden, and chronic inflammation. We hypothesized that the phenotype in Scnn1b-Tg mice would be significantly altered in the absence of mucin proteins, and that these phenotypic differences would provide clues about specific mucin functions during disease pathogenesis. We have evaluated these hypotheses by crossing Muc16, Muc5AC, and Muc5B knock-out mice with the Scnn1b-Tg model. Several parameters of the phenotype have been analyzed in the resulting mucin deficient/Scnn1b-Tg, including survival, bronchial mucus plugging, mucus cells abundance, airspace enlargement, airway inflammation, and lymphoid hyperplasia. No measurable differences in lung phenotype have been yet observed when Muc16-or Muc5AC-deficient/Scnn1b-Tg mice are compared to Scnn1b-Tg controls. However, Scnn1b-Tg mice lacking Muc5B contain significantly less bronchial mucus plugs at 5 weeks. Preliminary data indicated that absence of Muc5B does not significantly modify the inflammatory infiltrate in Scnn1b-Tg mice, but correlated with an increase in peribronchial lymphoid hyperplasia, which is a general indicator of ongoing adaptive immune responses. The results of these studies suggest that Muc5B is a particularly important mucin for plug development, and that it may play a role in modulating long-term chronic inflammation in this model. Presently, we are evaluating a number of other phenotypes, including early post-natal mucus plugging and the presence of bacteria in early stage disease. The insights from these mouse studies have obvious implications for human CF and corroborate human studies to control mucin production as a therapeutic target. 3 -) , and is severely impaired in the cystic fibrosis ∆F508 murine reproductive tract [1] . Although evidence suggests HCO 3 secretion is reduced in CF epithelia, these measurements have never been made in the native female reproductive tract. The present study demon-that CFTRinh172 modifies the formation of various cyclopentenone-protein complexes. Their identification by mass spectrometry, currently ongoing, will give important information about the molecular pathways affected by CFTR inhibition, and on the role of CFTR in the cell response to inflammatory stimuli. 1 1. Clinical Microbiology, Rigshospitalet, Copenhagen, Denmark; 2. Copenhagen CF Centre, Rigshospitalet, Copenhagen, Denmark Background: Chronic Pseudomonas aeruginosa lung infection is the most severe complication in CF patients. The infection is characterized by hypoxic endobronchial mucus with mucoid biofilms surrounded by numerous polymorphonuclear leukocytes (PMNs). We have recently identified the reduction of O 2 for production of O 2 by PMNs as the major mechanism for the O 2 depletion in sputum from infected CF patients. In this study, we hypothesized that the PMNs also consume O 2 for production of nitric oxide (NO) by the inducible NO synthases (iNOS) in infected endobronchial mucus.

Methods and Results: NO production by PMNs was examined in fresh expectorated sputum from chronically infected CF patients. The occurrence of discrete zones with NO in the sputum was demonstrated by microsensors and suggests that NO does not spread far from the source of production before being degraded. Fluorescence microscopy of sputum stained with the NO-indicator, DAF-FM, indicated PMNs a source of NO production. Furthermore, flow cytometry demonstrated the presence of iNOS in the majority of the host cells. To verify NO production by the iNOS of the PMNs, staining with the NO-indicator, DAF-FM, was reduced during inhibition of the iNOS with the arginine analog L-NMMA (p<0.002, n=8). Accordingly, O 2 consumption was also reduced during inhibition of the iNOS with L-NMMA (p<0.002, n=4). In addition, production of NO by PMNs was further indicated by the positive correlation observed between the concentration of PMNs and stable endproducts of NO generation; NO 3 and NO 2 -(p<0.04, n=11) as estimated with the Griess Reagent System. Conclusion: The present study demonstrates that sputum from chronically infected CF patients contains PMNs that consume O 2 for NO production. Therefore, in addition to depleting O 2 for the production of reactive oxygen species, the PMNs in the endobronchial mucus may also consume O 2 for the generation of reactive nitrogen species during chronic P. aeruginosa lung infection in CF. Mucociliary clearance (MCC) can be assessed in the clinic using gamma-scintigraphy. An aerosol of 99m Tc-particulate is inhaled and residual lung radioactivity is monitored over time. A potential therapeutic approach to the treatment of multiple respiratory diseases is to enhance MCC. To assist in the development of such novel "mucokinetics" it is desirable to establish a translational small animal model of MCC.

Our initial studies were designed to assess basal MCC in animals after short-acting anaesthesia. Guinea-pigs were anaesthetised with isoflurane/N 2 O and exposed to aerosolized 99m Tc-Sulfur colloid. Residual right lung radioactivity was monitored over time using an X-SPECT gamma-camera (GammaMedica-Ideas, USA), corrected for decay and expressed as mean percentage clearance of initial activity ± s.e.mean. Immediately following radio-aerosol exposure, guinea-pigs were restrained and allowed to regain consciousness while placed in the gamma-camera. Animals remained restrained throughout all acquisitions, however, they were returned to cages between the 10 minute and 90 minute acquisitions. Basal clearance with these animals was measured at 22.4 ± 1.9% (n=6) at 90 minutes. This study was repeated to assess inter-study variability and similar clearance was observed with 25.9 ± 3.1% (n=6) cleared at 90 minutes. strates that HCO 3 secretion is significantly reduced in CF ∆F508 murine uterine epithelium compared to wild type (WT).

Methods: Transepithelial short-circuit current (I sc ) measurements and HCO 3 secretion were examined in WT and CF ∆F508 murine uterine tissue mounted in Ussing chambers with an exposed area of 0.031 cm 2 . HCO 3 secretion was measured using the pH-stat technique. The luminal side of the epithelium was bathed with unbuffered HCO 3 free Ringer's solution bubbled with 100% O 2 . The basolateral side was bathed with buffered Ringer's solution containing 25 mM HCO 3 and bubbled with 95% O 2 / 5% CO 2 . Results: Basal I sc values were similar in the CF ∆F508 and wt uterine epithelium (67.5 ± 10 vs. 60.6 ± 6.5 mA.cm -2 respectively; p = 0.6), whereas basal HCO 3 secretion was significantly lower in the CF ∆F508 uterine epithelium compared to basal WT values (0.45 ± 0.18 vs. 1.28 ± 0.3 µmol.cm -2 .h -1 respectively; p < 0.05). The agonists PGE 2 (1 µM) and carbachol (10 µM) increased uterine HCO 3 secretion and I sc in WT mice with a maximal effect seen when added concurrently: ∆ HCO 3 -= 2.00 ± 0.3 µmol.cm -2 .h -1 and ∆ I sc = 27.5 ± 5.8 mA.cm -2 . Both stimulated HCO 3 secretion and I sc were significantly reduced in CF ∆F508 mice: ∆ HCO 3 -= 0.85 ± 0.3 µmol.cm -2 .h -1 (p < 0.05), and ∆ I sc = -16.6 ± 3.9 mA.cm -2 (p < 0.001).

Conclusion: PGE 2 and carbachol stimulate mucus release in the WT female reproductive tract, which is highly dependent on the presence of HCO 3 -, whereas mucus release is impaired in CF [1] . Herein, under similar conditions these agonists concurrently stimulated HCO 3 secretion. Both basal and stimulated HCO 3 secretion were significantly reduced in the CF ∆F508 uterine epithelium, providing the first direct measurements from native tissue of HCO 3 secretion in the CF female reproductive tract. The present findings provide further support of our hypothesis that normal mucus formation requires HCO 3 - [2] , demonstrating that where HCO 3 secretion is impaired, mucus release is also impaired. Inflammation is a physiological process that firstly suppresses both the initial injurious stimulus and its effects (onset and active phase), and secondly restores the normal tissue function (resolution). When cells cannot correctly respond to the aggression, the physiological phenomenon does not succeed and evolves toward a chronic inflammatory state leading to tissue destruction. This is the case in cystic fibrosis (CF), where both onset and resolution of inflammation are altered. While enhanced arachidonic acid release from membrane phospholipids is an important feature of the disease, a large number of eicosanoid mediators derived from this fatty acid may be either pro-or anti-inflammatory. These include prostaglandin D2 (PGD2), which enhances pulmonary recruitment of macrophages and neutrophils, and its intracellular dehydration product 15-deoxy-12,14D-PGJ2 which alters the function of many reactive proteins via its binding on their S-S bridges. The present study was designed to determine in human Calu-3 cells whether PGD2 and 15deoxy-12,14D-PGJ2 production ( measured by enzymatic immuno-assays) is affected by inhibition of CFTR channel activity with CFTRinh172.

CFTRinh172 treatment of cells for 48 hours decreased in a dosedependent manner the cell content in PGD2, as well as 15-deoxy-12,14D-PGJ2, with a maximal effect (about -50%) at 10 mM. The same effect was obtained by incubation with the pro-inflammatory cytokine IL-1b (250 pg/ml). Surprisingly, pre-treatment of cells for 48 hours with CFTRinh172 reversed the effect of IL-1b on both PGD2 and 15-deoxy-12,14D-PGJ2 cell content, which was increased by 2-fold. When the experiments were performed in the presence of cyclooxygenase-1 (COX-1) and -2 (COX-2) inhibitors SC560 and NS398, the results revealed that PGD2 production is due to COX-1 activation, which is inhibited by CFTRinh172 and negatively modulated by COX-2. Co-immunoprecipitation experiments demonstrated a physical link between CFTR and COX-1 that disappears under CFTRinh172 treatment. Moreover, by incubating cells under control or stimulated conditions with 15-deoxy-12,14D-PGJ2-biotinamide, we found To test if we could pharmacologically manipulate guinea-pig MCC, animals were exposed to an aerosol of either 5% dextrose (vehicle) or 100 µM benzamil, an ENaC blocker, for 60 minutes. Previously, the same exposure regimen of benzamil has been shown to significantly attenuate guinea-pig tracheal potential difference (Coote et al. Br J Pharmacol, 2008 , 155: 1025 . When tested in the guinea-pig MCC model, a 60-minute aerosol exposure to benzamil increased MCC compared to control (38.9 ± 4.3% and 18.1 ± 0.7% n=6, respectively). To assess if this model was sensitive to other "mucokinetics," we exposed guinea-pigs to a 10-minute aerosol of the endogenous purinergic agonist ATP (100 mM) immediately prior to the anaesthesia/radio-aerosol protocol, which induced an increase in MCC compared to vehicle (33.5 ± 4.0 % and 18.1 ± 0.3% respectively n=6). As these agents elicit their effects by different mechanisms we wanted to test if coadministration further enhanced guinea-pig MCC. Guinea-pigs were exposed to 50 minutes of 100 µM benzamil immediately followed by a 10minute exposure to 100 µM benzamil + 100 mM ATP immediately prior to the radio-aerosol protocol. There was no additive effect of the co-administration (34.9 ± 0.2% n=6).

We have measured basal MCC in guinea-pigs and furthermore pharmacologically manipulated guinea-pig MCC using two "mucokinetics." There was no observed additive effect when the two agents were co-administered, possibly due to MCC being fully stimulated by either agent at these concentrations. Dose-response studies and the utilisation of sub-maximal doses may give a greater insight into the potential utility of co-administered "mucokinetics" on guinea-pig MCC.

Buijs-Offerman, R. 1 ; Aarbiou, J. 1 ; Ooms, M. 2 ; Dik, W. 3 ; Wilke, M. 4 ; Scholte, B.J. 1 1. Cell biology, Erasmus MC, Rotterdam, Netherlands; 2. Reproduction and development, Erasmus MC, Rotterdam, Netherlands; 3. Immunology, Erasmus MC, Rotterdam, Netherlands; 4. Clinical genetics, Erasmus MC, Rotterdam, Netherlands Cystic fibrosis (CF) lung pathology is the main cause of morbidity and mortality in CF. It involves chronic bacterial infections with frequent exacerbations, intense inflammation, tissue injury and remodeling, which eventually results in fatal loss of lung function. If we can determine which molecular mechanisms are involved in the CF lung pathophysiology, we can use these as potential targets to develop new CF therapies.

We have used our mouse model of the most common CFTR mutation F508del (Cftr tm1eur FVB) to establish differential responses to airway injury and repair, which is one of the hallmarks of CF lung disease. Normal and mutant male mice (N=7, 9) were injected intra-peritoneally with naphthalene, which causes selective ablation of Clara cells followed by epithelial repair. We monitored the proteomic (1) and transcriptional responses to injury after 2 and 7 days by using Lab-on a Fluidigm™ microfluidics QPCR array, and we verified these by Fluorescent In Situ Hybridization (FISH), immunohistochemistry (IHC), and sequential ELISA.

We observed moderate chronic lung inflammation in untreated, pathogen free mutant mice compared to normal littermates. After injury, both mutant and normal mice showed a transient induction of IGF1 and EGFR-pathway related growth hormones including Amphiregulin (Areg), Epiregulin (EREG) and Heparin-binding EGF-like growth factor (HB-EGF). We demonstrate that Amphiregulin (in situ hybridization) as well as IGF1 and its co-receptor IGFBP5 (IHC) are produced by the repairing epithelium. We also observed transient mRNA induction of pro-inflammatory cytokines IL6 and CCL2/MCP-1 in both mutant and normal mice. This was accompanied by a substantial increase of extracellular matrix associated mRNA synthesis, pro-collagens (Col1a1, Col1a2 and Col3a1), Elastin (Eln), and TIMP1, indicative of a fibrotic response. In male CF mutant mice we observe a two-fold higher expression of IL6, Amphiregulin, and elastin compared to normal littermates, seven days after injury (N=9; p<0.01).

Our studies demonstrate in male F508del mouse lungs, that CFTR dysfunction results in a sustained inflammatory and fibrotic response to airway injury. Furthermore, we identify the converging EGFR and IL6 related paracrine and autocrine pathways as potential therapeutic targets in the CF lung pathology. Expression of TMEM16A protein is associated with the activity of calcium-activated chloride channels (CaCCs). There are multiple TMEM16A isoforms generated by alternative splicing and usage of an alternative promoter. Using patch-clamp recordings on heterologous expression systems (FRT and HEK-293 cells), we have investigated the properties of three different isoforms: TMEM16A(abc), TMEM16(ac), and TMEM16A(0). The (abc) and (ac) isoforms, generated by inclusion/skipping of exon 6b, are coexpressed in many cell types including primary bronchial epithelial cells and in the pancreatic CFPAC-1 cell line. Inclusion of this exon, coding for a segment of 22 amino acid residues (segment b) localized before the first transmembrane domain, decreases the calcium affinity of TMEM16Adependent chloride currents by nearly four-fold. Conversely, the expression of TMEM16A(0), an isoform devoid of all alternative segments including a region of 116 amino acid residues at the N-terminus, generates chloride currents activated by cytosolic calcium but lacking voltage-dependent activation at positive membrane potentials. We also studied endogenous CaCC activity in various cell lines such as CFPAC-1 cells. In these cells, chloride currents have characteristics consistent with the co-expression of multiple TMEM16A isoforms. Cells expressing TMEM16A show also an inhibition of chloride currents at high (> 1 µM) cytosolic free calcium concentrations. This mechanism may limit the extent of chloride secretion in epithelial cells following maximal stimulation.

In conclusion, alternative splicing of TMEM16A is an important regulatory mechanism controlling the activity of CaCCs. Elucidation of molecular mechanisms underlying TMEM16A function and regulation may help to design strategies aiming at maximizing chloride secretion in CF cells. Introduction: Cystic fibrosis (CF) is the most common inherited illness in Ireland that impacts directly on an individual's longevity. The deltaF508 CF phenotype affects secretory epithelia in different tissues, in particular the lung, the site of chronic infection in CF. The surface mucus is a key component of the innate immune system of the respiratory tract. Proper hydration of the airway surface layer (ASL) is prerequisite for efficient mucociliary clearance of inhaled particles and pathogens. There are significant differences in the progression of CF in male and female patients termed the « CF gender gap ». Lung function among female patients deteriorates 26% more rapidly than in male patients and on average, male CF patients survive 9 years longer than females. Moreover, measurement of nasal trans-epithelial potential difference in female CF subjects has shown an effect of menstrual cycle stage and circulating oestrogen on ion transport properties of the epithelium.

Methods: Confocal fluorescence microscopy was used to test the effects of 17β-Estradiol (E2) on ASL height in a normal human bronchial epithelial cell line (NuLi-1) and a deltaF508 CF cell line (CuFi-1), derived way disease and mucus hypersecetion in CF. Further, our studies suggest that several therapeutic agents affecting lipid metabolism as well as CFTR correctors can be tested in this paradigm. Current studies show a reduction of mucin lung content in Cftrtm1eur mutant mice treated with the cPLA2 inhibitor ATK (N=6; p<0.02) as reported previously in CFTR KO mice (3) The rate of mucociliary transport (MCT) in the airways is strongly influenced by the depth of airway surface liquid (ASL). Secretion of this liquid, from both the surface epithelium and the submucosal glands, is driven by a combination of active transepithelial Cland HCO 3 secretion. However, the relative contributions of Cland HCO 3 secretion to the ASL depth and MCT are unclear. This study was designed to assess the relative importance of Cland HCO 3 secretion to MCT in ex vivo porcine tracheas. Pigs (10-20 Kg) were euthanized, and the tracheas were excised and mounted in an environmental chamber designed for measurement of MCT. To test the effects of inhibiting HCO 3 secretion on MCT, one group of tissues was exposed to HCO 3 --free Krebs solution for the duration of the experiment while a parallel group was exposed to HCO 3 --replete (25 mEq/L) Krebs solution. Following three consecutive 20 min measurements of MCT, 100 µM acetylcholine (ACh) was added to stimulate submucosal gland mucous liquid secretion, and MCT was measured again for three additional 20 min periods. Prior to ACh addition, the mean MCT was higher in HCO 3 --depleted tissues (4.20±0.88 mm/min) than in tissues bathed in HCO 3 --replete Krebs (2.34±0.33 mm/min), but this difference did not reach statistical significance. Adding ACh significantly increased MCT in both groups, but the ACh-stimulated MCT was significantly lower in tissues exposed to HCO 3 -free Krebs (11.00±1.52 mm/min) than in tissues exposed to HCO 3 --replete Krebs (17.00±1.96 mm/min). A second series of experiments examined the effects on MCT of inhibiting active Clsecretion with 100 µM bumetanide, an NKCC inhibitor. In these experiments, which were all conducted in HCO 3 --replete Krebs, one group of tracheas was pretreated with bumetanide and the other group served as untreated controls. Before adding ACh, MCT in bumetanide-pretreated tracheas (0.97±0.18 mm/min) was significantly lower than in the control tracheas (3.82±1.06 mm/min). Adding ACh significantly increased MCT in both groups, but there was no significant difference between the bumetanide-pretreated (21.35±1.73 mm/min) and control groups (19.54±3.40 mm/min). These results indicate that HCO 3 secretion significantly enhances the ACh-stimulated MCT, whereas the loss of Clsecretion significantly impairs spontaneous MCT in the absence of ACh.

(Supported by CFFT grant Ballar07XX0). A prominent feature of cystic fibrosis (CF) airway disease is the production of an obstructive airway mucus with an unusually high (≥10%) solids content. However, the conditions within the airways that favor the produc-from a homozygous CFTR F508del/F508del individual. To investigate signal transduction via membrane and nuclear oestrogen receptors, we tested the effects of free E2 and a nuclear-impeded Estrogen Dendrimer Conjugate (EDC -EDC concentrations are expressed in terms of E2 equivalent) on ASL height.

Results: In control conditions, ASL height is significantly higher in normal cells than in CF cells (6.82 ± 0.53 µm versus 5.70 ± 0.21 µm, n=13, p=0.016, t test) . Moreover, ASL height showed a significant decrease in both normal (25% decrease in average after E2 treatment, n=5, p<0.05, ANOVA) and CF (20% decrease in average after E2 treatment, n=5, p<0.05, ANOVA) cell lines after 30 min treatment using concentrations of 17β-Estradiol between 0.1 and 10 nM. Finally, experiments using equivalent concentrations of EDC showed a significant reduction of ASL in both cell lines (4.72 ± 0.25 in NuLi-1 and 4.86 ± 0.42 µm in CuFi-1 after 1nM E2 equivalent EDC, n=5, p<0,05, ANOVA) whereas the empty dendrimer had no significant effect on ASL height.

Conclusion: These results demonstrate an antisecretory effect of oestrogen in both normal and CF bronchial epithelia which results in a reduction of ASL height. These rapid responses to oestrogen are initiated by membrane-initiated signalling events rather than via the classical nuclear receptor signal transduction pathway.

Acknowledgements: This work was supported by a NBIP Ireland Career Enhancement and Mobility Fellowship to V.S-C. co-funded by EU FP7 Marie Curie Actions and the Irish Higher Education Authority of Ireland PRTLI4. Support from the National Institutes of Health (R37 DK015556, to J.A.K.) is acknowledged.

Buijs-Offerman, R. 1 ; Aarbiou, J. 1 ; Ooms, M. 3 ; Tiddens, H. 2 ; Wojewodka, G. 5 ; DeSanctis, J. 6 ; Radzioch, D. 5 ; Touqui, L. 4 ; Scholte, B.J. 1 1. Cell biology, Erasmus MC, Rotterdam, Netherlands; 2. Sophia Children Hospital, Erasmus MC, Rotterdam, Netherlands; 3. reproduction and Development, Erasmus MC, Rotterdam, Netherlands; 4. Institut Pasteur, Paris, France; 5. McGill University, Montreal, QC, Canada; 6. Central university of Venezuela, Caracas, Venezuela Distal airway disease is observed in a majority of CF infants younger than five years as indicated by air trapping during expiration in CT scan analysis (1, 2) . This is accompanied by progressive airway remodelling, chronic inflammation and infections, frequent exacerbations, and eventually fatal loss of lung function. We hypothesize that early air trapping reflects distal mucus hyper-secretion caused by pro-inflammatory signalling, due to CFTR deficiency. Early intervention could potentially reduce the progression of CF lung disease. This requires an in depth study of the mechanisms involved in the process.

Mucins in distal airways are not secreted from sub-mucosal glands but from specialised (goblet cells) and other secretory (Clara) cells that are prominent in distal human airways. In mouse models, Clara cells can perform a phenotypic switch towards mucus secretion in models of induced inflammation. Recent data from our laboratory show that mucin content (ELISA) and cells expressing Goblet cell markers (Cacl3, Alcian blue) are enhanced in the airways of unchallenged and LPS treated and Cftrtm1eur F508del CFTR mutant mice kept under pathogen free conditions (N=9; p< 0.02), similar to data recently reported by us in CF knockout mice (3) . Importantly, this correlates with inflammation and distinct abnormalities in the metabolism of bioactive lipids in F508del CF mutant mice, which are known to be involved in the regulation of airway mucins. First, we observed a significant increase of phospholipase A2 (cPLA2) activity in the lung (N=6; p<0.002).

We previously reported a reduced expression of lung carbonyl reductase(Cbr2)and its product PGF2a in Cftrtm1eur mutant mice (4) . In lipid extracts from F508del mutant mouse lungs, arachidonic acid/docosahexanoic acid (AA/DHA) ratio was increased six fold (N=10, p<0.001) and the ratio of long-chain (C24) versus short chain (C14) ceramide subspecies was reduced threefold, with a reduction of total ceramides (N=10 p<0.001). We have concluded that the Cftrtm1eur F508del mouse model, which shows >80% survival to adulthood on normal chow, is a useful model of distal air-tion of mucus with high solids content are poorly understood. In the present study, we examined the relationship between the rate of airway liquid secretion (J v ) and mucous solids in ex vivo bronchi from porcine and non-CF human lungs. Bronchi were isolated as intact tubes and cannulated to collect the liquid secreted into the airway lumen. In pig bronchi, variable rates of J v were induced with different secretogogues (acetylcholine, substance P, and vasoactive intestinal peptide) and anion secretion inhibitors. The nonvolatile solids content of secreted liquid was determined from its dry weight. At high rates of liquid secretion (14-16 µl/cm 2 /h), the percent solids were about 2%. However, a curvilinear trend was observed for increased percent solids at lower liquid secretion rates with a maximum of 19.9% solids seen at 0.5 µl/cm 2 /h. When the secretion rates of solids mass (J s ) and water mass (J w ) were determined, both of these parameters were linearly related to the J v . The y-intercept of the J s /J v relationship was positive relative to the J w /J v , which accounts for the high percent solids at the low J v rates. A predictive model incorporating these two linear relationships fit the experimental data very well. In experiments with human bronchi, variable J v rates were achieved with acetylcholine and forskolin as secretogogues. The human bronchi (obtained from explanted and donor lungs) mirrored the behavior of the pig bronchi with high percent solids measured in airway liquid at low J v rates. J s and J w in human tissues were also linearly related to the J v , and the J s /J v relationship also exhibited a positive y-intercept. The predictive model incorporating the human bronchial J s /J v and J w /J v relationships closely approximates the model for the pig airways. These data indicate that secretion of mucus with high percent solids content is the normal consequence of sustained, low rates of mucus liquid secretion in both human and porcine airways.

Supported by NIH HL063302. Background: Cystic fibrosis (CF) is characterized by recessive mutation in the CFTR gene, resulting in a massive pro-inflammatory phenotype in the lung. The problem is that the mechanism by which mutant CFTR causes inflammation is completely unknown. Recently, microRNAs (miR-NAs) have been proven to be key post-transcriptional regulators of gene expression by directing their target mRNAs toward degradation and/or translational repression. The mis-regulation of specific miRNAs has been demonstrated in a variety of diseases in humans including cancer, heart disease and diabetes. Since each miRNA governs the expression of multiple genes, the most effective way to overcome the effects of a mis-regulated miRNA is to modulate its expression in diseased cells. Thus, miRNAs have emerged as important therapeutic targets in the frontier of biomedical research. Recent studies indicate that even a single miRNA can induce a therapeutic response in an animal model of disease. Our goal is to investigate the mechanism by which aberrant expression of miRNAs in CF neutrophils leads to the pro-inflammatory disease phenotype and develop novel miRNA-based anti-inflammatory therapeutics for CF.

Hypothesis: Our hypothesis is that microRNA expression profile in CF neutrophils will reveal novel biomarkers and help develop miRNA-based therapeutics for CF.

Methods: miRNA expression profiling of neutrophils isolated from blood of CF patients, homozygous for ∆F508, was carried out using the TaqMan miRNA low density arrays comprising 571 different miRNAs (ABI). Functional pathway analysis of miRNAs was carried out using Ingenuity Pathway Analysis (IPA). The mRNA levels corresponding to miRNA expressions were analyzed simultaneously using ILLUMINA bead chip HT-12 arrays.

Results: miRNA profiling of neutrophils from CF patients as well as controls revealed a group of miRNAs which exhibit altered expression levels. Functional pathway analyses identified signaling pathways, including TNFα and IL-1β signaling, relevant to the CF disease status. The corresponding mRNA expression further corroborates the miRNA expression data.

Our findings indicate that aberrant expression of miRNAs in CF neutrophils serve as novel biomarkers. Insight into the mechanisms associated with miRNA-mediated regulation of inflammation in CF will lead to novel miRNA-based therapy for CF. The TMEM16A protein (also known as anoctamin-1) appears as an important component of calcium-activated chloride channels (CaCCs). When expressed in heterologous expression systems, TMEM16A causes the appearance of calcium-activated chloride currents characterized by slow activation at positive membrane potentials (Caputo et al., Science 322:590-594, 2008) , a feature of classical CaCCs in different cell types. TMEM16A belongs to a protein family that includes nine other members, from TMEM16B to TMEM16K. Using electrophysiological and biochemical techniques, we have evaluated the expression and function of TMEM16A-K proteins in human epithelial cells. TMEM16A, TMEM16F, and TMEM16K are the anoctamins with the highest expression in primary bronchial epithelial cells, the bronchial cell line CFBE41o-, and pancreatic CFPAC-1 cells. However, only TMEM16A was clearly associated with a plasma membrane chloride channel function in such cells. By cell transfection with siRNA, we found that silencing of TMEM16A inhibited calciumdependent anion transport, whereas silencing of the other anoctamins was ineffective. We also transfected different cell types with plasmids coding for most TMEM16 proteins. Expression of TMEM16A, and to a lower extent TMEM16B, evoked calcium-dependent chloride transport. In contrast, the expression of TMEM16C, 16E, 16F, 16G, 16H, 16J, and 16K was not associated with the appearance of calcium-or cAMP-activated chloride channels, as measured by YFP fluorescence assay, patch-clamp, or short-circuit current experiments. Our experiments indicate that TMEM16C-K proteins are not involved in the function of CaCCs. They may constitute a different type of plasma membrane ion channel, activated by a stimulus other than calcium. Alternatively, they may have a non-channel function or be localized in intracellular compartments.

Supported by Italian Foundation for Cystic Fibrosis. Calcium-activated chloride channels (CaCCs) have long been thought to play a role in fluid transport in airway epithelia. The molecular identity of these channels has remained unknown until recently when TMEM16A / Anoctamin-1 was shown to be essential for tracheal secretion in mouse. These channels are gated both by calcium and by voltage, but the molecular mechanisms remain enigmatic. Here we have performed a detailed biophysical and structure-function analysis of the voltage-and calcium-dependent gating of Ano1 expressed in HEK-293 cells. We show that Ano1 is essentially a ligand-gated channel whose voltage dependence is conferred not by a voltage sensor composed of charged amino acids but by permeant ions stabilizing channel open state. Gating of Ano1 is controlled by at least two calcium binding sites. The high-affinity site, which exhibits low cooperativity and is weakly coupled to voltage, is abolished by neutralizing 4 glutamic acids (444-EEEE-447) in the first intracellular loop. The low affinity site exhibits high cooperativity, is intimately coupled to voltage-dependent gating, and is disrupted by a naturally-occurring splice variant that deletes 4 amino acids (448-EAVK-451). Our data support a model in which Ano1 is primarily gated by calcium with allosteric coupling of the calcium binding sites to a permeant anion binding site in the channel whose occupancy is voltage-dependent. This mechanism provides a unified explanation of CaCC channel gating by voltage and ligand that has long been enigmatic. the glandular stem/progenitor cell niche and what impact this might have on regeneration of the airways in the setting of disease. We have begun to address this question by comparing tracheal injury responses between CFTR-/-and CFTR+/-mice, and evaluating the localization and proliferative capacity of the slow-cycling nucleotide label-retaining cells (LRCs) thought to represent stem/progenitor cell populations in the mouse airway. Although mice have relatively few glands in their proximal trachea, others have shown that CFTR plays a similar role in glandular secretions of mice and humans. Our studies showed that LRCs accumulated in SMGs of the proximal trachea in CFTR+/-mice, whereas LRCs of CFTR-/-mice predominantly localized to the tracheal surface airway epithelium (SAE). Analysis of airway progenitor cell proliferative capacity using a colony forming efficiency (CFE) assay, demonstrated significant differences in the CFE index of proximal (SMG containing) and distal (SMG absent) portions of the trachea between CFTR-/-and CFTR+/-mice. The CFE index in the proximal glandular regions of CFTR+/-tracheas was 5.8-fold higher than distal segments, consistent with the highest density of LRC stem/progenitors in the SMGs of the proximal trachea. In contrast, the CFE and LRC indexes were both evenly distributed across proximal and distal tracheal segments of CFTR-/-mice. To evaluate what properties of CF glands might change and influence the glandular niche, we evaluated expression of a number of neuropeptides known to be produced by neuroendocrine cells associated with stem cell niches in the bronchioles. Significantly enhanced (10 fold) expression of the mitogenic neuropeptide calcitonin gene-related peptide (CGRP) was observed in CFTR-/-tracheal SMGs. CGRP is known to activate CFTR through G-protein signaling and thus its altered expression may be a compensatory response to enhance glandular secretions in CF. Interestingly, glandular expression of CGRP was also induced (6.4 fold) by airway injury in CFTR+/-mice, while no further enhancement in CGRP was seen in CFTR-/-mice following injury. Elevations in glandular CGRP were also seen in CF human and CFTR-/-ferret tracheobronchial tissues. Adenovirusmediated expression of CGRP in mouse trachea also led to a significant increase in epithelial cell proliferation. Collectively, our findings suggest that CFTR defects in SMGs may invoke an injury response that influences maintenance of the stem/progenitor niche in this region through the induction of CGRP, a discovery that may have important implications for injury/repair mechanisms in the CF airway. Surface expression of TLR4 was followed over 24h after stimulation with PA LPS (clinical isolate, 100ng/ml) in CFTE & HTE cells. Basally, CFTE showed increased TLR4 expression (p<0.05 vs. HTE). LPS exposure decreased TLR4 in both cells (8h). TLR4 remained significantly below basal levels (24h, p<0.05) in HTE, while in CFTE cells it returned to basal level, which was associated with a significant increase of IL-8 release (24h). Following stimulation with SA PGN (10µg/ml), a similar pattern was observed for TLR2, with a significantly higher basal TLR2 expression in CFTE, which remained significantly elevated at 24h (p<0.05). These data suggest a role for CFTR to downregulate TLR expression thus controlling the inflammatory response produced. To confirm this we repeatedly stimulated cells (24h LPS, wash, 24h LPS). After the second stimulation HTE cells showed a significant decrease in IL-8 (48h vs. 24h (p<0.05)), but CFTE cells still released similar amounts. This suggests that reduced TLR4 expression in HTE may provide a protective mechanism against repeated stimulation, whereas CFTE cells are continuously stimulated.

While Hyperabsortion of Na + through the epithelial Na + channel (ENaC) contributes to cystic fibrosis (CF) lung disease and thus emphasizes the need to further understand the regulation of this channel. The short, palate lung and nasal epithelial clone (SPLUNC1) is expressed in the submucosal glands and surface epithelia of the conducting airways and was recently identified as secreted protein, which negatively regulates ENaC. SPLUNC2 and long PLUNC1 (LPLUNC1) are related PLUNC family members that are also expressed in the airways and we tested whether these proteins could likewise inhibit ENaC by co-injecting Xenopus laevis oocytes with 0.3 ng of cRNA for each α, β, γ-ENaC subunit and either 1.5 ng of SPLUNC1, SPLUNC2, LPLUNC1, or water as a control, followed by voltage clamping 24 hours later. The resultant amiloride-sensitve current was recorded and normalized to ENaC control groups. Compared to controls, SPLUNC1 inhibited ENaC current while SPLUNC2 and LPLUNC1 had no effect [control 1 ± 0.10 (n=18), LPLUNC1 0.82 ± 0.17 (n=17), SPLUNC2 0.74 ± 0.09 (n=22), and SPLUNC1 0.30 ± 0.06 (n=22, p<0.05)]. Since SPLUNC1 is a small, secreted protein (256 amino acids), identifying its active site may enable us to design a therapeutic peptide that inhibits ENaC. To identify this region, we engineered truncation mutants that retained the N-terminal, 19 amino acid signal sequence but lacked 30, 60, or 85% of the full-length protein, deleted from the C-terminus. These truncants were co-expressed with ENaC in oocytes for voltage clamp measurements. Secretion of each truncant was confirmed by Western blot. SPLUNC1's inhibitory effects on ENaC were retained in truncants with deletions up to 85% [control 1 ± 0.04, SPLUNC1 0.52 ± 0.03, SPLUNC1 del30% 0.389± 0.09, SPLUNC1 del60% 0.45 ± 0.06, SPLUNC1 del85% 0.49 ± 0.06 (n=15-18 per group, p<0.05)]. Since SPLUNC1 del85% retained full function, we analyzed its peptide sequence. Residues 22 to 39 were ~40% homologous to the ENaC inhibitory peptides, α26 and γ43 that are released from the extracellular loop of ENaC following proteolytic cleavage. To test whether this region of SPLUNC1 was responsible for the inhibition of ENaC, we engineered additional SPLUNC1 truncants that bisected or omitted this putative active site. These mutants did not inhibit ENaC (control 1.00 ±0.06, SPLUNC1 del88.75% 1.01 ± 0.13, SPLUNC1 del92.5% 1.03 ± 0.06; all n=12), suggesting that the active site of SPLUNC1 lies in its first 40 amino acids and, specifically, appears to be in the region with high homology to ENaC inhibitory peptides α26 and γ43. However, despite its strong simiarity to these inhibitory peptides, SPLUNC1's mechanism of action may differ. Using surface biotinylation and Western blotting, we found that SPLUNC1 facilitated a reduction in the number (N) of ENaC subunits at the plasma membrane, which may contribute to the inhibition of macroscopic ENaC current. In conclusion, our data suggest that SPLUNC1 inhibits ENaC by acting through amino acids 22-39, which are homologous to the previously described ENaC inhibitory fragments. Identifying the active site of SPLUNC1 may provide a new avenue for therapeutic regulation of hyperactive ENaC in CF airways.

Funded by the NIH.

at 8 and 12h (p<0.001), a response which was again significantly higher in CFTE vs HTE (p<0.05). This indicates the presence of cross talk between TLRs.

To investigate if the loss of CFTR is involved in TLR4 dysregulation, HTE cells were treated with CFTR inhibitor (172Inhib) or CFTR siRNA. CFTR inhibition caused no change in basal TLR4 expression, but significantly increased TLR4 after LPS stimulation (24h, p<0.05 vs. untreated HTE); analyses of supernatants showed this treatment produced a modest increase in IL-8 over controls, but this did not reach significance. In CFTR silenced cells (siRNA) TLR4 expression was increased basally & increased further after stimulation (p<0.001 vs. scrambled), reaching similar levels as seen in the CFTE cells. This elevated TLR4 expression was associated with a higher again increase in IL-8 (vs. CFTR inhib) at baseline & after stimulation but was not significant.

In conclusion, the loss of CFTR is associated with (1) Accumulating evidence suggests that defective function/expression of CFTR contributes to the hyperactive innate immune response of respiratory epithelia in cystic fibrosis (CF). TMEM16A was recently identified as a calcium activated chloride channel (CaCC) at the plasma membrane. Based on the significant CaCC activity and the lack of lung disease in CF mouse, it was postulated that activation of the CaCC may alleviate the mucocilliary clearance defect and excessive innate immune response of respiratory epithelia in CF patients. To assess this possibility experimentally, we asked whether TMEM16A has the ability to suppress the inflammatory status of human respiratory epithelia model lacking CFTR. First we examined the consequence of CaCC activation on pro-inflammatory cytokines secretion in CFBE41o-epithelia derived from a CF patient with undetectable ∆F508CFTR expression. IL-4 has been shown to induce the endogenous expression of TMEM16A in primary epithelia. Alternatively, the channel can be transiently activated by P2Y receptor stimulation with UTP leading to calcium release from intracellular stores. Both of these methods were used to activate TMEM16A in CFBE41o-cells cultured on permeable filter supports and differentiated under air-liquid culture (ALC) conditions. Initial results suggest that neither UTP nor IL-4 treatment were able to reduce the hyper-inflammatory phenotype, marked by IL-8 secretion, of the CFBE41oepithelia. Furthermore, we were unable to suppress the pro-inflammatory phenotype by overexpression of exogenous TMEM16A alone or in combination with IL-4 treatment and UTP stimulation. Ongoing studies are aimed at assessing the extent of TMEM16A activation by IL-4 and UTP in CFBE41o-epithelia. In contrast to TMEM16A, overexpression of CFTR in CFBE41o-epithelia under a tetracycline-inducible promoter attenuated the pro-inflammatory cytokine secretion. Notably, wt CFTR inhibited the secretion of pro-inflammatory cytokines (IL-8, IL-6, MCP-1 and GROβ/γ) by >50%. Preliminary observations suggest that CFTR expression reciprocally influences the CaCC activity. Expression of wt CFTR reduced the UTP or ionophore stimulated halide secretion in CFBE41o-epithelia expressing endogenous or exogenous TMEM16A. While CFTR induction does not alter the global TMEM16A expression, ongoing studies suggest that it may attenuate the CaCC channel density at the apical plasma membrane. Although the overall abundance of the TMEM16A and CFTR channel density may be preserved by a presently unknown physiological mechanism, our results are consistent with the notion that TMEM16A is unable to compensate for the hyper-inflammatory phenotype of CFBE41o-model of CF bronchial epithelia.

Supported by NIH, CIHR and EMBO fellowship to GV. The thin liquid layer lining the alveolar space is essential for maintaining efficient gas exchange, surfactant homeostasis, and defense against inhaled toxins/pathogens. Airway epithelial cells actively secrete or absorb ions in response to local stimuli to coordinately regulate the liquid layer lining airway surfaces. However, it remains unclear whether the alveolar epithelium exhibits similar capacities, and if so, how the physiologies of the two regions are coordinated. Since it has been difficult to achieve a unified understanding of the surface liquid physiology within the intact alveolus, in vitro studies were performed using primary human alveolar type II (AT2) cells to observe how these cultures respond to specific challenges. First, our data revealed that AT2 cultures exhibited both epithelial Na + channel (ENaC)-mediated Na + absorption and cystic fibrosis transmembrane conductance regulator (CFTR)-mediated Clsecretion, both regulated by extracellular nucleotides. Moreover, measurements of alveolar surface liquid (AvSL) height by confocal microscopy revealed that human AT2 cells: 1) achieved an extracellular nucleotide concentration-dependent steady-state AvSL height of approximately 4 µm; 2) absorbed liquid when the lumen of the AT2 culture was flooded; and 3) secreted liquid when treated with forskolin, uridine 5' triphosphate (UTP), or subjected to cyclic compressive stresses mimicking tidal breathing. The secretory responses mediated by either nucleotides or cAMP were shown to be highly dependent on the function of CFTR, since pretreatment of normal AT2 cultures with the CFTR inhibitor (CFTRinh-172) significantly inhibited these changes in AvSL height. In addition, the increase in AvSL height/secretion was markedly inhibited in cultures pretreated with apyrase, demonstrating a nucleotidemediated regulation of AvSL height by AT2 cells exposed to UTP or subjected to continuous cyclic compressive stress. Overall, our studies of AvSL regulation demonstrate that human AT2 cells in vitro can respond to ambient needs, e.g., can absorb in response to the equivalent of alveolar flooding, but can secrete in response to addition of exogenous secretagogues. Our results suggest that in vivo, AT2 cells under baseline conditions can be secretory, thus, resulting in a bulk movement of liquid from AT2 cells to airway surfaces. Further studies will be required to test this hypothesis and analyze how the integrated system, including AT1 cells contribute to surface liquid transport regulated in response to physiologic stresses in health and in response to pathophysiologic stresses in disease. We have previously demonstrated that ENaC activity is regulated by the protease/protease inhibitor balance on the apical surface of cultured human bronchial epithelia (HBE). This balance is disturbed in CF, leading to constitutive proteolytic ENaC activation, Na + hyperabsorption, airway surface dehydration, and mucus plugging in the lung. Channel activating protease (CAP) expression appears to be increased in CF HBE and prostasin, a known CAP, appears to undergo excessive zymogen activation in these tissues. Recent evidence suggests that matriptase is an upstream protease that activates prostasin and matriptase is likely to be highly active in the acidic airway surface liquid (ASL) present in CF. Therefore, we hypothesized that matriptase activity in the airway is regulated by ASL pH and that excessive matriptase activation promotes Na + hyperabsorption in the CF airway.

Methods and Results: First, we examined the effect of pH on matriptase activation in primary cultures of CF and non-CF HBE. As assessed with the specific mAbs M24 (latent and active matriptase), M69 (active compounds applied apically may cross the cell membrane and inhibit basolateral channels. Therefore, cellular localization studies are underway. The pharmacological profile at the apical membrane of sensitivity to bupivacaine and quinidine but resistance to Ba 2+ is consistent with that of KNCK5. Therefore, K2P channels may be potential targets for pharmacological manipulation in CF. (O 3 ) and other air pollutants may have negative impacts on the respiratory health of both normal and children with compromised respiratory systems and that these negative effects may persist into adulthood. Only one epidemiologic study exists that suggests an association between ambient air pollution and increased risk of pulmonary exacerbations and decreased lung function in cystic fibrosis patients, a disease primarily attributed to mutation-mediated dysfunction of the cystic fibrosis transmembrane conductance regulator (CFTR) protein. The purpose of this study was to determine if CFTR knockout (CFTR-/-) mice respond to cyclic O3 exposure differently than do wild-type (CFTR+/+) mice.

Methods: CFTR+/+ and gut-corrected CFTR-/-mice were exposed to filtered air (FA) or O 3 {ages 4-22 days (weaning): 0.5 ppm O 3 8 h/d 5d/week, ages 22 to 70 days: every other week to 0.5 ppm O 3 8 h/d 5d/week}. At 70 days of age, the lungs were evaluated via histopathology (formalin fixed and paraffin embedded), antioxidant analysis (lavage fluid and lung tissue), gene expression or proteomics.

Results: O 3 exposure did not result in significant changes in overall growth, but in both CFTR+/+ and CFTR-/-mice, O 3 exposure resulted in distal airway epithelial hyperplasia. Tissue ascorbate, glutathione and urate levels were not different among genotype or exposure groups. Significant ozone-induced upregulation of 16 of 84 oxidative stress and antioxidant genes occurred in CFTR+/+ male mice, but not in CFTR-/-males or in female mice. Proteomic analysis of lung lavage fluid and tissue also indicates that O 3 -induced protein up-and down-regulation to be significantly different between CFTR+/+ and CFTR-/-mice, as well as between male and female mice.

Conclusions: Early life exposures to O3 results in airway epithelial remodeling in both CFTR+/+ and CFTR-/-mice, however, the chronic alterations in gene and protein expression indicate that CFTR-/-mice may respond to additional pulmonary challenges differently than CFTR+/+ mice.

Supported by Cystic Fibrosis Foundation grant R464-CR07 and NIEHS grant P01 ES-11617. Background: Nitric oxide (NO) modulates airway tone, inflammation and host defense against Pseudomonas aeruginosa, the predominant opportunistic pathogen in CF. Arginases are enzymes that compete with NOS synthases for L-arginine as common substrate and thereby regulate NO production. Airway NO deficiency is a characteristic of CF lung disease.

Aims: To measure changes in arginase and NOS activity following Pseudomonas infection of the lung using in-vivo, ex-vivo and in-vitro techniques, and to study alterations in NO production in Pseudomonas pneumonia in mice. matriptase), and M19 (latent and matriptase associated HAI-1), latent matriptase (70 kDa) is activated and forms a complex with HAI-1 (120 kDa) at a pH optimum of 6. Activation occurs rapidly, on the order of minutes, after which time the matripase/HAI-1 complex is shed from the apical HBE surface and is subsequently present in apical secretions. Preliminary results suggest that CF HBE cells express more active matriptase than non-CF control cultures. With the increased expression of the 120 kDa matriptase/HAI complex at pH6 and in the CF samples, we observed a corresponding decrease in latent matriptase. Conversely, we did not observe the corresponding decrease in unbound HAI-1 suggesting that HAI-1 is expressed in molar excess to matriptase. Next, we examined ENaC processing by matriptase in BEAS2B cells transiently transfected with epitope tagged ENaC subunits and human matriptase constructs. Matriptase appears to cleave the gamma subunit in a similar pattern to the processing previously observed with prostasin, CAP2, elastase and plasmin.

Conclusions and Future Directions: These results indicate that matriptase activation occurs under slightly acidic conditions in airway epithelia and leads to increased ENaC processing. As the CF ASL is slightly acidic and within the pH optimum for matriptase activation, excessive ENaC processing by matriptase may be an additional mechanism of Na + hyperabsorption in the CF airway. Further studies are currently underway in our laboratory to identify the matriptase cleavage site and to determine whether inhibition of matriptase activity by raising the ASL pH can mitigate Na + hyperabsorption. In human bronchial epithelial (HBE) cells, abnormal CFTR function results in absent chloride (Cl -) secretion and greater than normal sodium (Na + ) absorption with subsequent depletion of airway surface liquid (ASL) and mucus dehydration. These changes impair mucociliary clearance, resulting in increased infection and inflammation, the principal causes of CF lung disease. Therefore, compounds that either increase Clsecretion or decrease Na + absorption in HBE cells are of great therapeutic interest in CF.

Chloride secretion and Na + absorption in HBE cells depend on potassium (K + ) channels to provide an electrical driving force by maintaining a negative cell membrane potential. Most attention has been paid to the role of basolateral membrane K + channels, though data from Calu-3 cells (Davis KA and Cowley EA. Pflugers Arch. 2006 Feb; 451(5) :631-41) and human bronchial epithelial cells (Namkung et al., JBC. 2009; 284, 15916-15926) also support the presence of apical membrane K + channels, raising the possibility of inhaled K + channel modulators as therapeutics for CF patients.

We tested the hypothesis that two-pore K + (K2P or KCNK) channels regulate transepithelial ion transport in well-differentiated HBE cells. To test this hypothesis, the K2P channel blockers bupivacaine, quinidine, and barium (Ba 2+ ) were applied to either the basolateral or apical membrane of HBE cells mounted in Ussing chambers and short-circuit current (Isc) was measured under different conditions. Bupivacaine applied to either the basolateral or apical membrane inhibited forskolin-stimulated Isc, a measure of CFTR-dependent Clsecretion, in a dose-dependent manner with IC50 values of 211 and 23.5 µM, respectively, and maximal inhibitions of 82% and 100%, respectively. Both bupivacaine and quinidine applied to the apical membrane of HBE cells inhibited amiloride-sensitive Isc, a measure of Na + absorption, in a dose-dependent manner with IC50 values of 70 and 47 µM, respectively, and maximal inhibition of 70% for each. Barium applied to the basolateral membrane but not the apical membrane inhibited amiloride-sensitive Na + absorption. Reverse transcriptase -polymerase chain reaction demonstrated that KCNK1, KCNK2, KCNK3, KCNK5, and KCNK6 mRNA were expressed in our cells.

Our data demonstrate that inhibitors of KCNK channels inhibit both Na + absorption and Clsecretion in HBE cells. Furthermore, the finding of different IC50 values at the apical versus basolateral membranes supports the hypothesis that K2P channels reside in both the apical and basolateral membranes of HBE cells, though one must account for the possibility that Methods: Mice (C57BL/6) were infected by direct tracheal instillation of Pseudomonas (PAO-1)-coated agar beads using 2x10 6 CFU in 40 µl. On day 3 post infection anesthetized non-ventilated mice received a primed constant stable isotope infusion containing labeled L-arginine, L-ornithine and citrulline (0.1 ml/h). In a second set of experiments, mechanically ventilated isolated mouse lungs were constantly perfused with a stable isotope solution containing the same labeled amino acids. The isotopic enrichment of each of the infused isotopomers and its product amino acids were measured using liquid chromatography-tandem mass spectrometry. Rates of conversion were used to calculate enzyme activities, i.e. L-arginine conversion to citrulline for NOS and L-arginine to ornithine for arginase activity. In the whole mouse experiments arginase activity was quantified using atom percent excess (APE) ratios. Furthermore, arginase activity was measured invitro by L-arginine conversion to ornithine and NO-metabolites (NOx) were determined in lung homogenates by chemiluminescence.

Results: Infection resulted in a moderately increased APE ratio of Arg>Orn/Arg in lung (2.8±0.3 vs 1.8±0.1, p=0.01) using perfused stable isotopes in the whole mouse model. The APE ratio in liver was elevated in infected animals as well (116.3±22.1 vs 47.8±4.9, p =0.002), but no change in APE ratio was found in trachea, plasma or kidney. Lung NOS activity was significantly increased in Pseudomonas infected animals (254.6±36.3 vs 9.0±1.4 µmol/g/h, p<0.001), no difference in NOS activity was seen in other tissues or in plasma. In the isolated perfusion model in infected lungs the activity of arginase (42.1±8.2 vs 3.6±1.3 µmol/g/h) and NOS (16.3±7.3 vs 0.37±0.1 µmol/g/h) was found to be significantly higher compared to naïve controls (p < 0.001, respectively). Increased arginase activity was confirmed in-vitro (48.8±20.8 vs 13.9±1.4 mU/mg protein, p<0.05). NOx concentrations were increased significantly in infected lungs compared to non-infected controls (1.9±0.4 vs 1.0±0.3 nmol/mg protein, p=0.007), confirming an increase of NOS activity in Pseudomonas pneumonia in mice.

Conclusions: Pseudomonas infection of the lung results in increased arginase and NOS activity. We currently study the effect of systemic arginase inhibition on pulmonary NO production in Pseudomonas pneumonia.

Supported by the Lynne and Arnold Irwin Family Foundation and the Canadian CF Foundation. It is now well established that cAMP signaling is compartmentalized. The components of the cAMP signaling machinery, such as the GPCRs, adenylyl cyclases, and cAMP effectors such as PKA and its downstream targets are tethered by macromolecular signaling complexes, an arrangement that increases both efficiency and specificity of signaling. Cyclic nucleotide phosphodiesterases (PDEs), the enzymes that hydrolyze and inactivate cAMP, are critical components of these signaling complexes. These enzymes serve to restrict the free diffusion of cAMP throughout the cell and to finely tune cAMP levels within localized microdomains of signaling. PDEs represent a superfamily of isoenzymes that are divided into 11 PDE families on the basis of their kinetic and pharmacologic properties. Here, we investigated whether PDE isoenzymes are tethered to macromolecular signaling complexes involving the CFTR and may, thus, serve to control the cAMP-mediated regulation of this channel. Immunoprecipitation of endogenous CFTR from T84 epithelial cells resulted in co-IP of a significant amount of PDE activity that was sensitive to the PDE4-selective inhibitor, Rolipram. Co-IP of PDE4 activity was also observed in CFTR-IPs from Calu3 cells and mouse lung tissue suggesting that the formation of CFTR/PDE4 complexes is a conserved feature among different epithelia. The PDE4 family consists of four genes, PDE4A-D, which are each expressed as multiple variants. IP of PDE4A from Calu3 cells resulted in co-IP of a large amount of CFTR protein, whereas only a minor amount of CFTR was co-immunoprecipitated with PDE4D. The predominant interaction of CFTR with PDE4A rather than PDE4D is somewhat surprising as previous reports suggested a predominant functional and physical interaction of the PDE4D subtype with the CFTR. However, CFTR also interacted more strongly with PDE4A compared to PDE4B or PDE4D in the Hek293 overexpression system, thus, confirming a preferential interaction of CFTR with PDE4A. Deletion of the four C-terminal amino acids of CFTR encoding the PDZ-binding domain did not ablate the CFTR interaction with PDE4A. Thus, PDE4A interactions with CFTR are mechanistically different from the previously reported interaction of CFTR with PDE4D which is dependent on the PDZ domain scaffold protein Shank2. Furthermore, recombinant CFTR carrying the F508 deletion immunoprecipitates PDE4A with the same efficiency as wildtype CFTR. This may suggest that F508-CFTR still associates with and is controlled by PDE4A in CF patients and that inactivation or displacement of PDE4A may provide a means to elevate cAMP in the vicinity of the mutant channel and stimulate F508-CFTR activity.

Supported by the Cystic Fibrosis Foundation. Background: Primary ciliary dyskinesia (PCD) is an autosomal recessive condition leading to chronic bronchial infections and bronchiectasis. Infections are caused by a primary failure of mucociliary clearance. Once treatment has stabilized pulmonary function, there is no progressive deterioration.

In contrast, cystic fibrosis (CF) is characterised by secondary failure of ciliary function. Chronic endobronchial infection leads to bronchiectasis and ultimately respiratory failure, despite modern management. As both CF and PCD are neutrophilic airway diseases, but PCD generally has a milder clinical phenotype, a comparison of the inflammatory profile in stable patients was carried out. The hypothesis was that local and systemic inflammatory responses to bacterial infection differ between the two diseases and account for the differences in clinical severity.

Patients and Methods: PCD patients consented to a research visit where blood and sputum were collected for measurement of inflammatory markers (cell counts, CRP, IL1β, IL6, IL8, IL10, TNFα, calprotectin, neutrophil elastase, myeloperoxidase). PCD patients were matched for gender, age and bacterial infection with CF patients taking part in a longitudinal study of potential biomarkers.

Results: Twenty-two PCD and 22 CF patients were included. Systemic inflammation was significantly higher in CF: blood neutrophils (PCD 4.2x10 9 /L (2.0-9.5 x10 9 /L); CF 6.4 x10 9 /L (2.5-10.5 x10 9 /L) p<0.01), serum IL6 (PCD not detected, CF 3.6 pg/ml (3.6-15.5) p<0.01), IL8 (PCD 1.8 pg/ml (0.7-4.1); CF 3.1 pg/ml (1.7-40.3) p<0.01) and calprotectin (PCD 11435 ng/ml (2544-48957) ; CF 15394 ng/ml (5784-51962) p<0.05), although CRP was not different. Conversely, pulmonary pro-inflammatory cytokines were significantly greater in PCD: sputum IL8(PCD 25748 pg/ml (3247-74500); CF 8844 pg/ml (2400-23923) p=<0.01), IL6 (PCD 86.38 pg/ml(14.82-1320); CF 13.22 pg/ml (4.65-270.7) p<0.01). Only IL1β was higher in CF (PCD 154.9 pg/ml (11.5-2386); CF 414.2 pg/ml (41.38-2874) p<0.05). Other cytokines, sputum proteases and sputum cell counts did not differ between groups.

Conclusions: CF is characterised by a greater systemic but less sputum pro-inflammatory response compared with PCD, with paradoxically similar sputum neutrophil counts. The improved pulmonary status of PCD patients does not appear to be due to a milder local inflammatory milieu. We speculate that the increased systemic inflammatory response in CF may contribute to the extrapulmonary components of the disease. Further studies are needed to understand the mechanisms behind these findings. incurred due to antibiotic overuse and the deleterious side effects of steroids. Safe, but effective, compounds that enhance Cltransport, specifically those that activate the cystic fibrosis transmembrane conductance regulator (CFTR), could provide significant therapeutic advantages in this regard. The objectives of the present study were to investigate whether the natural bioflavonoid hesperidin (identified by compound library screening of CFTR potentiators): 1) increases transepithelial Clsecretion in vitro and in vivo, 2) enhances ciliary beat frequency (CBF), and 3) exerts its mechanistic effects through cAMP/PKA dependent pathways. Transepithelial Cltransport (Ussing chamber) and CBF were investigated in primary murine nasal septal (MNSE) and human sinonasal epithelial (HSNE) cultures. In vivo activity was also measured using the murine nasal potential difference (NPD) assay. CFTR R-domain phosphorylation and cAMP levels were examined following exposure to hesperidin as a further test of cAMP/PKA dependent activation.

Results: Hesperidin significantly increased CFTR-mediated Cltransport (change in short-circuit current, ∆I SC ) in both MNSE [13.51+/-0.77 vs. Conclusion: These in vitro and in vivo findings indicate hesperidin is a robust Clsecretagogue and CBF activator in sinonasal epithelium. The mechanism by which the compound augments CBF may involve changes in periciliary viscosity due to CFTR potentiation, or direct effects on pathways known to influence ciliary beating. The nasal culture model provides an excellent model system for elucidating the mechanisms by which compounds such as hesperidin activate MCC, including analysis of CFTR biochemistry, mucus propulsion in vitro, and cell signaling pathways that contribute to the mucociliary apparatus. Submucosal glands secrete mucus that is rich in antimicrobials and is thought to play a role in airway innate defense. Gland fluid secretion is stimulated by increases in [cAMP]i or [Ca 2+ ]i and both of these are probably elevated during neurally-stimulated secretion. In CF humans, pigs and ferrets, gland secretion to [cAMP] i is severely depressed because of CFTR dysfunction; secretion to [Ca 2+ ]i is partially spared. The phosphodiesterase inhibitors sildenafil (Dormer,Thorax,2005) and milrinone (Kelley,JCI,1996 , Kelley,PNAS,1997 as well as structural analogs of sildenafil (Robert,Mol Pharmacol,2008) partially correct function of ∆F508-CFTR. In the studies by Kelley, combinations of β-adrenergic agonists+milrinone (PDE3 inhibitor) stimulated cAMP-mediated chloride transport in CF-T43 human ∆F508 cell line, in primary nasal polyp cells from a patient homozygous for ∆F508, and in ∆F508 but not -/-mouse nasal epithelium. To determine if milrinone affected forskolin-stimulated fluid secretion from normal human glands from donor tracheas (CONTROL) and from CF airway glands(from lung transplant), we used in-situ optical monitoring to measure secretion rates from individual glands. A forskolin dose-response curve in CONTROL subjects (100nM~100µM) gave an EC50 of 2.5 ± 0.4 µM and Vmax of 0.57 ± 0.02 nL/min/gland (nmg) (36-57gl#,n=3), with response to 30 µM and 100 µM forskolin no different than to 10µM forskolin (P=0.9). Therefore 10 µM forskolin was used for all tests, milrinone was used at 100 µM, and all secretion rates are given as mean ± SEM nmg. In CONTROL, forskolin stimulated secretion of 0.31 ± 0.08 nmg (99gl#,n=9), milrinone induced secretion of 0.28 ±0.1 nmg (67gl#,n=7), and the combination of milrinone+forskolin increased secretion 1.4-fold to 0.4 ± 0.1 nmg(99gl#,n=9), which greater than the response to either agent alone(P<0.05) but was less than an additive effect. In CF, forskolin stimulated secretion of only 0.07± 0.02 nmg, significantly less than CONTROL(P=0.01,84gl#,n=8). Milrinone alone was not tested in CF, but Background: Dysfunctional mucociliary clearance (MCC) is a common pathophysiologic process in many airway diseases, including cystic fibrosis. Sinupret ® is a combination of natural bioflavonoids that is widely used as an alternative treatment for respiratory ailments, such as chronic sinusitis. However, the mechanism by which this formulation confers MCC improvement is not known. The present experiments utilized primary murine nasal septal epithelial (MNSE) [wild type (wt) and transgenic CFTR -/-] and human sinonasal epithelial (HSNE) cultures to investigate the mechanistic basis of this therapeutic intervention. Cultures were subjected to transepithelial ion transport measurements using pharmacologic manipulation in Ussing chambers, Fura-2 calcium (Ca 2+ ) imaging, and cAMP signaling. Clsecretion was also measured in murine models in vivo by nasal potential difference (NPD). Ciliary beat frequency (CBF) was analyzed with high speed digital imaging following apical and/or basal exposure to the drug formulation.

Results: The change in current (∆I SC -expressed as µA/cm 2 ), representing Clsecretion, was significantly increased by Sinupret in MNSE [19.04+/-1.67 vs. 4.55+/-1.33 (control), p<0.05]. ∆I SC decreased with basal membrane permeabilization (13.24+/-1.1), incubation with the PKA inhibitor, H89 (4.47+/-0.95) or addition of the CFTR inhibitor INH-172 (3.14+/-1.45, p<0.05). In CFTR -/-MNSE, ∆I SC was significantly amplified by the drug [7.65+/-1.86 vs. 0.23+/-0.04 (control), p<0.05]. CFTR -/cells pre-incubated with the calcium activated Clchannel inhibitor niflumic acid (2.94+/-0.36) or permeabilized with amphotericin (4.33+/-0.67) demonstrated significant reductions in the magnitude of ∆I SC stimulation (p<0.05). Sinupret increased Cltransport in HSNE cultures (17.42+/-1.72 vs. 4.6+/-0.53 (controls), p<0.05) confirming robust Clsecretagogue activity in human sinonasal epithelium. In vivo (murine NPD, n=42), Clsecretion was enhanced when compared to controls (-3.8mV vs. -0.9mV) and substantially more robust than forskolin -one of the strongest known activators of wt CFTR (-1.65mV) (p<0.05). CBF (fold-change) increased following either apical [2.05+/-0.15 vs. 1.52+/-0.10 (control)] or basal [1.37+/-0.09 vs. 0.9+/-0.10 (control)] exposure (p<0.05). There was no stimulation of cellular cAMP attributable to treatment.

Conclusion: The observation that Sinupret robustly stimulates CBF and transepithelial Clsecretion establishes a mechanism by which this formulation is likely to enhance MCC, and points to ways that agents utilizing the same therapeutically relevant pathway might be optimized in the future. Partial activation of Clsecretion in CFTR -/-MNSE indicates that a non-CFTR mechanism contributes (at least in part) to the Clsecretory response, but without requiring elevations in Ca 2+ as a second messenger. Our findings indicate that effects on CFTR underlie the majority of Clsecretion attributable to the drug, although enhanced electrochemical gradients for Cltransport, or CFTR independent pathways at the apical membrane, may also contribute. Background: Ineffective mucociliary clearance (MCC) is a common pathophysiologic process associated with airway inflammation and infection. Airway surface liquid (ASL) is a principal component of the mucociliary apparatus and is strongly influenced by the vectorial transport of ions, such as chloride (Cl -). Dysfunctional Cltransport results in dehydration of the ASL and mucus stasis as demonstrated in both the lower and upper respiratory tracts of individuals with cystic fibrosis (CF). Conventional interventions for chronic sinusitis have been limited by bacterial resistance milrinone+forskolin increased the secretion rate to 0.5±0.2nmg, 7.5-fold more than forskolin alone (P=0.03,84gl#,n=8). Remarkably, the mean secretion rate to milrinone+forskolin across subjects were equivalent for CON-TROL and CF subjects(P=0.69), although the control responses in this series were ~half as large as historical controls for unknown reasons. The mechanism of milrinone's effect is under investigation to determine if it is correcting or potentiating mutant CFTR or is activating an alternate channel. If rapid correction is occurring, phosphodiesterase inhibitors might be useful starting points for developing CFTR correctors.

Tirouvanziam, R.; Zirbes, J.M.; Milla, C. Pediatrics, Stanford University, Palo Alto, CA, USA Rationale: With the advent of CF newborn screening there is an urgent need to advance our knowledge of the earliest stages of CF airway disease to develop better informed effective interventions. In particular, there remains significant uncertainty about the origins and nature of airway inflammation in CF newborns and infants. Here, we tested the hypothesis that the characteristic pattern of active neutrophil dysfunction that we discovered in CF children and adults (Tirouvanziam et al. Proc Natl Acad Sci USA, 2008) may already be in effect at the inception of CF airway disease.

Methods: We used a minimally invasive airway fluid collection method, nasopharyngeal suction (NPS), to yield airway fluid from CF newborns and infants (N=12) and non-CF controls (including primary ciliary dyskynesia and lung hypoplasia subjects, N=5). An 8Fr suction catheter was advanced through a nare down to the level of the pharynx. Cough was stimulated and suction then applied to aspirate airway secretions. Suction was released and the catheter was immediately withdrawn and cleared with 0.5 mL sterile preservative-free 0.9% NS. The functional properties of NPS fluids, which are limited in volume (500 microliters or less), were assessed for cellular and molecular correlates by high-content analytical platforms (digital flow cytometry and mass spectrometry, respectively).

Results: NPS fluids from CF newborns and infants yielded significant numbers of live, non-apoptotic neutrophils. Compared to those from non-CF controls, NPS fluids from CF newborns and infants showed clear evidence of live airway neutrophils releasing their primary granules, which contain deleterious proteases ( e.g., neutrophil elastase, among others) and the oxidative enzyme myeloperoxidase. In particular, airway neutrophils from CF newborns and infants displayed characteristic surface upregulation of CD63 and concomitant downregulation of surface CD16. Furthermore, mass spectrometric analysis revealed the significant presence in CF NPS fluids of free amino acids and amino acid-containing molecules, indicative of ongoing proteolysis. Perhaps most strikingly, CF NPS fluids contained methionine sulfoxide, a molecular signature seen at later disease stages that is likely formed through proteolysis of methionine-rich airway mucins by neutrophil-derived proteases followed by oxidation by neutrophil-derived myeloperoxidase, thereby providing an independent proof of early airway neutrophil dysfunction.

Conclusions: This study suggests that NPS is a valid tool for early monitoring of CF airway disease. Furthermore, results from our cellular and molecular analyses of CF NPS fluids from newborns and infants suggest an early dysfunction of neutrophils recruited to the CF airway environment, similar to that seen at later disease stages, and consistent with a key role for live airway neutrophils in the establishment of the disease, as recently proposed by our group.

Patients with CF develop paranasal sinus hypoplasia, chronic sinusitis, craniofacial abnormalities, and dental enamel defects. However the pathophysiology of CFTR in craniofacial and dental development is difficult to study in humans because of the associated confounding factors of infection, inflammation, and antibiotic use. Moreover, longitudinal imaging studies in CF patients are limited because of the detrimental effects of repeated imaging from birth to adulthood. CFTR -/-pigs provide the opportunity to investigate the effect of CFTR loss at birth without confounding factors.

Our goal was to examine the craniofacial and dental phenotype of CFTR -/-pigs. We investigated CFTR expression, craniofacial morphology, paranasal sinus size, and dental enamel in litter-matched newborn wild-type and CFTR -/-pigs. Reverse-transcriptase PCR confirmed CFTR expression in wild-type but not in CFTR -/-sinus epithelium. Craniofacial morphology and paranasal sinus size was obtained from 3-dimensional CT-scan volumetric analysis. There were no craniofacial morphological differences between wild-type and CFTR -/-pig skulls. Individual paranasal sinuses were identified and radiologically scored for presence of infection. Newborn pigs, regardless of genotype had well developed maxillary and ethmoid sinuses free of disease. Erupted maxillary and mandibular canines and incisors were dissected from newborn wild-type and CFTR -/-pigs. The total sinus volume of the CFTR -/-pigs was smaller than that of wild-type pigs, even when their skull volumes were normalized. The gross phenotype of the dental enamel was recorded, microstructural analysis of the teeth performed with electron microscopy, and elemental analysis performed with scanning electron microscopy. The enamel of CFTR -/-pigs were grossly different than wild-type pigs at birth, with chipping of the enamel, and darkened pigmentation. There were no obvious changes noted on enamel composition or microstructural analysis.

CFTR is expressed in the sinus epithelium of newborn wild-type pigs. Lack of CFTR may account for hypoplastic sinus development, similar to the phenotype seen in humans. This finding is independent of sinus infection and inflammation. Lack of CFTR does affect the gross phenotype of dental enamel at birth, but the structural and elemental changes are yet to be determined. The CF pig may aid understanding of the effect of CFTR on craniofacial and dental development.

tified potential biomarkers and therapeutic targets for CF. Further characterization of these identified proteins may also lead to better understanding of molecular mechanisms underlying CF.

Introduction: A20 is an inducible gene responsible for the termination of NF-κB signalling through immune receptors including Toll-Like Receptors (TLRs). Mice deficient in A20 display persistent activation of NF-κB by TLRs and TNF-R, causing multiorgan inflammation (1) . A20 inhibits TLR induced NF-κB signalling at the level of TRAF6, but is reliant on the formation of a complex with Ring Finger protein (RNF)11, the E3 ligase Itch and an adaptor protein called TAX1BP1 to do so (2) . We propose that in the chronically inflamed CF epithelium, A20 and the A20 signalling complex is fibrosis transmembrane conductance regulator (CFTR) or are secondary to postnatal infection and inflammation. Recently, we developed a porcine model of CF. The lungs of newborn CFTR-targeted pigs lacked inflammation. However, with time they spontaneously developed characteristic features seen in the lungs of older patients with CF including inflammation, remodeling, mucus accumulation, and infection. In this study, we asked whether, at birth, the airways of CFTR-/-pigs show any abnormalities using CT scans, pathology and morphometry. Compared to littermate controls, the trachea and mainstem bronchi had a smaller caliber, tracheal cartilage rings were irregular and disrupted, and cross-sections of trachea were less circular than controls. Accompanying these changes, we found alterations in trachealis smooth muscle including altered bundle orientation and an increased trend in size. We also asked if these changes occur in young infants and children with CF. First, we retrospectively examined chest CT scans from non-CF and CF children two years of age and younger to evaluate the trachea. We found no significant difference in the tracheal lumen area between non-CF and CF, but CF tracheas were distinguished from controls by having less circular walls in cross sections. These changes were similar to those seen in neonatal CFTR-/-pigs. Second, we analyzed previously published morphometric data (Sturgess J and Imrie J. Am J Pathol 106: [303] [304] [305] [306] [307] [308] [309] [310] [311] 1982) and found reduced tracheal caliber in CF infants less than 2 weeks of age. These findings suggest that CF-related airway changes begin in utero and might contribute to the pathogenesis of CF lung disease. Rationale: There is significant evidence that the cystic fibrosis (CF) neutrophil is intrinsically abnormal. However, molecular mechanisms underlying dysregulated neutrophil activity and the actual presence of the CF transmembrane receptor (CFTR) in neutrophils remains controversial. We focused on the neutrophil membrane as it is involved in cell signaling and plays a central role in regulating neutrophil delivery, function, and clearance. The question that this project will address is: are CF neutrophil membranes intrinsically abnormal?

Methods: Western blot and flow cytometry were used to confirm or dispute the presence of the CFTR protein in peripheral blood neutrophils. Membranes from circulating neutrophils of CF patients during an exacerbation, CF individuals during a stable phase, non-CF bronchiectasis patients and normal healthy subjects (n=6 for each group) were prepared by sucrosedensity ultracentrifugation. Proteins (25µg) from each sample were separated in the first dimension before being resolved by two-dimensional difference gel electrophoresis and visualized with CyDye™ fluorescent labeling. DeCyder™ software was used to match, and analyze protein spots from multiplexed fluorescent images. Proteins were determined to be differentially expressed if there was a 1.5 fold difference in expression observed with a p value of <0.05 deemed to be statistically significant.

Results: By Western blot and flow cytometry the CFTR protein was confirmed as an integral neutrophil membrane component, absent in CF (delta508) neutrophil membranes. In turn, the normal and non-CF bronchiectasis membrane proteome profiles markedly differed from the profiles of CF patients. Results revealed 31 proteins to be differentially expressed, comprising both integral and membrane associated proteins. Classification of differentially expressed proteins revealed 42% to be integral membrane components, 25% cytoskeletal, 23% cytosolic and 10% secondary or tertiary granule derived proteins. Differentially expressed proteins were confirmed by Western blot analysis.

Conclusion: Our results confirm that the CF neutrophil membrane proteome is intrinsically abnormal. A number of proteins are differentially expressed between the normal healthy control and CF patients when stable, with a significant number of these differences only present in the CF cohort of patients and not present in the inflammatory control group. Our data iden-compromised and unable to act on target proteins (TRAF6), thereby contributing to persistent pro-inflammatory signalling through the NF-κB pathway.

Methods: Experiments used Non-CF (16HBE41o-) and CF (CFBE41o-) bronchial epithelial cell lines and primary nasal epithelial cells obtained from patients homozygous for F508del and age-matched controls. Cell lines were grown in submersion while primary cells were fully differentiated at air-liquid interface to represent the lung environment. All cells were stimulated with LPS (P. aeruginosa, Sigma) for 0-24h. Cell lines were treated with a p65 inhibitor (JSH-23, CalioBiochem) to determine which NF-κB subunit regulates IL-8 release (ELISA; Peprotech). RNF11, Itch and TAX1BP1 expression was assessed by qPCR (Roche LightCycler). Following stimulation with LPS, cell line lysates were immunoprecipitated with A20 (Santa Cruz) and probed for TRAF6 (Cell Signaling) by Western Blot.

Results and Discussion: Previously we reported that peak A20 expression is delayed in CFBE41o-compared to 16HBE41o-. Moreover, A20 expression in CFBE41o-fell below basal levels by 12h exposure and was accompanied by persistent p65 expression and release of IL-8 (3). IL-8 release in CFBE41o-was found to be p65 dependent, suggesting that the inflammatory response in CFBE41o-may be largely p65-driven. Since RNF11, Itch and TAX1BP1 are essential for proper A20 function the expression of these complex members was determined by qPCR. All three genes were upregulated in 16HBE41o-following LPS exposure. However, in CFBE41o-, RNF11 mRNA expression was significantly reduced while Itch and TAX1BP1 remained unchanged following exposure to LPS (12h). These findings were confirmed in primary epithelial cells where A20, RNF11, Itch and TAX1BP1 mRNA expression were substantially reduced in F508del homozygotes (compared with controls) basally and following exposure to LPS (24h). Furthermore, in 16HBE41o-A20 co-localizes with TRAF6 1h after LPS exposure, coinciding with peak A20 expression, but co-localization was not observed in CFBE41o-at any time-point (0, 1 and 4h).

Conclusion: Our findings from cell lines and primary cells support our proposal that in CF epithelium, the A20 signalling complex is compromised, preventing A20 from negatively regulating NF-κB and contributing to an exaggerated and prolonged NF-κB driven inflammatory response. This work is supported by the CF Trust UK (PJ541). Vitamin D insufficiency is common in CF patients. Such a problem may lead to a reduction in the innate immune defenses in the lung. The active form of vitamin D, 1,25-Dihydroxyvitamin D3 (1,25(OH) 2 D 3 ) plays an important role in the regulation of innate immunity through the induction of the gene expression of antimicrobial peptides such as the cathelicidin LL-37, which plays a role in defense against bacteria and other infectious agents. We earlier reported that LL-37 is induced by (1,25(OH) 2 D 3 ) in airway epithelial cells (AEC) with a resultant increase in antimicrobial activity against airway pathogens. We further examined AEC for the expression of other innate immune genes in response to (1, 25 (OH) 2 D 3 ). We observed that a novel mediator of innate immunity, Triggering Receptor Expressed on Myeloid Cells (TREM-1) is induced by 10 -8 M (1,25(OH) 2 D 3 ) in both CF and normal AEC. In general, lower levels of TREM-1 were induced in CF cells compared with the normal cell line. Inhibition of CFTR in the normal line results in a reduction in TREM-1 levels to those found in the CF cells. Furthermore, a characterization of SNPs in the Vitamin D Receptor (VDR) gene showed differences at four common polymorphisms between the normal and CF cells. Activation of TREM-1 in normal cells by a cross-linking antibody results in the induction of human β-defensin 2 expression and an increase in antibacterial activity in the airway surface fluid. An examination of the TREM-1 promoter region demonstrated the stimulation promoter activity with (1,25(OH) 2 D 3 ), which is inhibited by the transcription factor PU.1. Since TREM-1 has been demonstrated to be involved in the early defense against bacterial infection in the lung, our data suggest that differences may exist in the vitamin D-regulation of innate immunity in CF patients, and that regulation of innate immune mediators by vitamin D may be able to enhance the natural antibacterial defense in the CF lung.

Supported by a grant from the Cystic Fibrosis Foundation. CF patients suffer more severe consequences of respiratory virus infection than non-CF individuals leading to exacerbation of airway disease. Our goals are to determine the impact of virus infection on CF lung disease. We used an in vitro model of human ciliated airway epithelium (HAE) and respiratory syncytial virus (RSV) as model systems as this virus commonly causes exacerbation of CF lung disease.

We demonstrate that RSV infection of non-CF HAE results in increased airway surface liquid (ASL) height measured by XZ confocal microscopy. This effect maximized at ~3 days post-inoculation (pi) (ctrl 7.5±0.3 µm, RSV 15±1 µm, n=6). Parallel experiments using CF HAE revealed that RSV infection failed to increase ASL height over the same time period (ctrl 5.8±0.1 µm, RSV 5.5±0.2 µm, n=6). These data demonstrate that non-CF HAE, but not CF HAE, respond to RSV infection by increasing mucosal fluid secretion and suggest that CFTR plays a critical role in this response. Using pharmacological reagents we confirmed a role for CFTR in RSVinduced fluid secretion. Bumetanide (Bum), an inhibitor of chloride secretion, significantly inhibited RSV-induced ASL secretion (ctrl, 7.2±1 µm, RSV 14.5±1 µm; RSV+bum 9±0.5 µm, ctrl+bum 5.9±0.5 µm, n=3). To determine the specific contribution of CFTR-mediated Clsecretion to this response we used CFTRinh172 which significantly inhibited RSV-induced ASL height (ctrl, 6.8±0.3 µm, RSV 11 ±1.8 µm, RSV+172 8.2± 0.2 µm, ctrl+172 6.8±0.5 µm, n=3). Thus, our data indicate that CFTR activation during RSV infection is critical for the increased ASL height. Previous studies have shown that regulation of fluid secretion in non-CF HAE is governed by extracellular nucleotide concentrations. We found that RSV infection of either non-CF HAE or CF HAE significantly increased levels of ATP and adenosine (ADO) in the ASL, suggesting that increased nucleotides mediate increased fluid secretion when CFTR is present during RSV infection. Furthermore, the adenosine receptor antagonist, 8-SPT, significantly attenuated RSV-induced fluid secretion in non-CF HAE (RSV 12.7±0.3 µm, RSV+8SPT 9.0±0.1 µm, n=3). To determine the source of ATP/ADO after RSV infection we assessed ATP release from the epithelium vs. cells shed after RSV infection and found that ATP was derived from the shed, infected ciliated cells and that increased ASL height correlated with the numbers of shed cells. We hypothesize that RSV infected ciliated cells shed from the epithelium undergo apoptosis releasing ATP into the ASL. Breakdown of ATP to adenosine by ectonucleotidases then stimulates CFTR-mediated fluid secretion.

We conclude that RSV infection increases adenosine levels in the ASL sufficient to activate CFTR-dependent increases in ASL. We speculate that increased fluid secretion after RSV represents a host defense mechanism to facilitate virus clearance and shed cells from the airways. The dependency on CFTR for this innate defense suggests that mechanical clearance of virus/virus-infected cells is diminished in CF airways possibly resulting in prolonged RSV infection and increased inflammatory consequences.

Supported by the CFF and NIH.

reduced pulmonary exacerbations and improved weight gain in CF patients 6-18 years of age uninfected with P. aeruginosa (Saiman et al. JAMA 2010) . While the mechanism of action by which azithromycin exerts positive effects in CF patients remains unclear, evidence suggests that azithromycin may act as an anti-inflammatory agent. We here report results on serum inflammatory markers in patients participating in the RCT. Methods: Serum myeloperoxidase (MPO), high sensitivity C reactive protein (CRP), intracellular adhesion molecule 1 (ICAM 1) and interleukin 6 (IL-6) were measured at baseline, after 28 days as well as on day 168 in patients receiving with either oral azithromycin (n= 126) or placebo (n=125) using sensitive ELISA methods.

Results: Concentrations for ICAM 1 and IL-6 were below the detection limit in the majority of patients and did not demonstrate any significant differences between the two groups at all time points (data not shown). MPO and CRP levels were similar in both groups at baseline. MPO concentrations significantly decreased in the azithromycin group at day 28 compared to the placebo group. This treatment effect was less apparent at day 168. Similarly, CRP concentrations decreased in treated patients at day 28; this difference was no longer significant at day 168.

Conclusions: In patients not infected with P. aeruginosa, treatment with oral azithromycin significantly reduced neutrophils and serum inflammatory markers within 28 days after initiation of treatment providing indirect evidence for an anti-inflammatory effect of azithromycin in this patient population.

Supported by CFF. Purpose: Cystic fibrosis is characterized by an intrinsic hyper-immune state. This primary inflammatory condition prior to bacterial infection is supported by clinical observations in CF neonates and young children. Mounting evidence also suggests an exaggerated and prolonged inflammatory response to foreign invaders of the airways. The aim of this work was to evaluate the anti-inflammatory and anti-proteolytic activity of doxycycline on the CF phenotypic bronchial airway epithelium.

Methods: The CF compound heterozygote IB3-1 and rescued isogenic S9 cell lines were used to characterize doxycycline's effects on lung parenchyma inflammation and proteolysis. Biological markers IL-8 and MMP-9 were analyzed by ELISA and qRT-PCR following stimulation with IL-1β (10 ηg/mL), TNF-α (10 ηg/mL), and TGF-β1 (10 ηg/mL) in the presence or absence of doxycycline at 24 hours for various studies. Cytotoxicity of doxycycline was measured by MTT. Immunoblotting was incorporated to examine the presence of constituitive MMP-9 and future mechanistic details.

Results: Diminution of IL-8 protein release upon doxycycline treatment in a dose dependent manner was observed in both IB3 and S9 cell lines The majority of CF patients show a slow progressive loss of pulmonary function as a consequence of smoldering chronic infection with P. aeruginosa and inflammation. This is punctuated by episodes of acute exacerbations due to infection or acquisition of new infectious agents. Respiratory viruses are detected in approximately 28-48% of patients with pulmonary exacerbations, and hence viruses are believed to be important triggers of exacerbation in CF. We hypothesized that persistent exposure to inflammatory milieu, bacteria and/or bacterial products may alter the innate immune responses of airway epithelial cells to secondary viral infection.

Methods: Primary well-differentiated or immortalized CF cells were infected with mucoid P. aeruginosa (MPA) at MOI of 0.01 or sham-infected with PBS and incubated for 24 h. Cells were shifted to fresh media containing gentamicin and superinfected with intact RV39 or replication-deficient UV-irradiated RV39 and incubated at 33°C for 24 h. Viral titer, type I and type II interferon (IFN)s, and IFN-stimulated gene, RSAD2 were measured. Phosphorylation of IRF3 and Akt was determined by Western blot analysis. In some experiments, cells were pretreated with diphenylene iodonium (DPI), or N-acetyl cysteine (NAC) to determine the effects of antioxidants on the IFN responses of CF airway epithelial cells to coinfection with MPA and RV.

Results: CF airway epithelial cells coinfected with MPA and RV showed significantly more viral titer compared to RV39-infected cells, suggesting increased persistence/replication of RV39 in MPA-infected cells. MPA and RV39 co-infected cells showed significantly less mRNA expression of IFN-β, IFN-λ1, IFN-λ2/3, and RSAD2. Further, MPA and RV39 coinfected cells showed reduced activation of Akt and IRF3, each of which are required for maximal IFN response to viral infection. A chemical inhibitor of PI-3 kinase, LY290042 inhibited RV-induced IFN response similar to MPA infection. In contrast to CF airway epithelial cells, normal cells co-infected with MPA and RV showed increased expression of IFNS and RSAD2. CF airway epithelial cells pretreated with antioxidants DPI or NAC showed IFN response to co-infection with MPA and RV similar to normal airway epithelial cells.

Conclusions: Our results indicate that prior MPA infection increases persistence/replication of RV by attenuating expression of IFNs only in CF airway epithelial cells, but not in normal cells. Secondly, MPA appears to inhibit IFN response by inhibiting PI-3 kinase activity that is required for maximal IRF3 activation and IFN response. Finally, treatment with antioxidants corrects the IFN response to coinfection in CF airway epithelial cells, indicating that preexisting oxidative stress in CF cells may alter the innate immune responses to secondary viral infection.

Supported by the Cystic Fibrosis Foundation (SAJJA08G0). MTT assays showed minimal cytotoxicity, 11% and 25% in IB3 and S9 respectively, with doxycycline up to 100 µg/mL. TC 50 could not be determined because of insolubility issues greater than 500 µg/mL in cell culture medium. Western analysis revealed the translation of MMP-9 message into constituitive protein and further work to elucidate the molecular signaling of doxycycline is forthcoming. Conclusion: Doxycycline exhibited immune-modulatory activity, without serious toxicity concerns, as measured by the inflammatory marker IL-8, airway proteolysis marker MMP-9, and the metabolic viability agent MTT in a CF in vitro model. Defective CFTR results in a variety of changes in airway clearance and lung milieu contributing to inefficient resolution of lung infection and inflammation in cystic fibrosis (CF). The inability to resolve infection/inflammation suggests defects in immunity; however whether this is primary or secondary to the defective epithelium is unknown. We hypothesize that deficient Cftr expression in myeloid cells directly impacts the phenotype and response to infection in the lung ultimately contributing to the pro-inflammatory phenotype of CF airways. To determine the role of myeloid Cftr in chronic infection to Pseudomonas aeruginosa, myeloid specific Cftr KO mice (generated using the Cre+/LoxP system and the lys-M promoter), controls (Cre-), congenic Cftr null (B6.129P-2 Cftrtm1Unc ) and C57Bl/6 mice were treated with agar beads impregnated with Pseudomonas aeruginosa and evaluated for 10 days (10-12 mice/group, n=3). Outcome measures included survival, weight change, clinical and lung pathology scores. Surviving animals were euthanized for bronchoalveolar lavage (BAL) which was used to measure cytokines, total cell count and differential. The myeloid specific Cftr KO mice had decreased ability to resolve infection and inflammation relative to the controls (p<0.05). The infection resolution in the myeloid specific Cftr KO (40±17% survival) was similar to the congenic Cftr null mice (55±20% survival). The myeloid specific Cftr KO also had increased values for clinical score (p=0.003) and lung pathology (p=0.04) relative to the controls, which correlated with the increased inflammation (r 2 = 0.963, p≤0.01) and loss of weight (r 2 =0.978; p≤0.01). The BAL total cell counts increased (p=0.07) with elevated numbers of neutrophils. Controlling for the variability in the infection model, two sets of studies were done infecting both Cftr null (n=14) and myeloid specific Cftr KO mice (n=15) and controls at the same time. In these studies, 5 mice were euthanized at 3 days post-infection, while the remaining mice continued to day 10. The myeloid specific Cftr KO mice had an elevated total cell response when compared to the myeloid specific control (p<0.05) similar to Cftr null mice (p<0.05). We then compared the relative neutrophilic response in the myeloid specific Cftr KO and Cftr null mice along with controls at both the inflammation time point (day 3) and inflammation resolution time point (day 10). There was no statistical difference in the neutrophilic response to infection between the Cftr nulls, myeloid specific Cftr KO versus controls at day 3. Differences in the neutrophilic response was observed at day 10 for both the Cftr null (p=0.053) and myeloid specific Cftr KO (p=0.07) . Further, at day 10 the BAL of the myeloid specific Cftr KO mice had elevated cytokines IL-6 and KC relative to controls (p<0.05) consistent with observations in the Cftr null. When evaluating the myeloid specific Cftr KO BAL cells directly, we found an enhanced responsiveness to LPS stimulation resulting in increased IL-6 and KC gene expression (n=4, p≤0.05). These studies suggest that myeloid cells in CF are intrinsically different in their ability to response and resolve inflammation in CF.

This work was funded by NIH DK027651 and the Cystic Fibrosis Foundation. Recently, we demonstrated that this hyperinflammatory response to LPS is due to abnormal spatio-temporal localization of TLR4 during signaling. Specifically, CF macrophages have more rapid TLR4 internalization and prolonged TLR4 retention in the endosomal compartment compared to WT cells. This abnormal TLR4 trafficking leads to increased LPS-induced activation of the NF-kb, MAPK and IRF-3 pathways (Bruscia et al., submitted) . We hypothesized that the prolonged retention of TLR4 in endosomal compartments and the increased signal transduction were concurrent with impaired TLR4 degradation. To test this hypothesis, WT and CF macrophages, pretreated with the protein synthesis inhibitor cycloheximide, were exposed to LPS (3h and 6h) and total TLR4 protein was assessed by Western blot. We found that during LPS stimulation, the rate of TLR4 degradation was slower in CF cells than WT controls. These data suggest that the retention of the receptor in the endosomal compartments in CFTR-/-macrophages delays TLR4 degradation, which is fundamental for resolution of inflammatory stimuli.

Next, we examined if primary CF human macrophages responded similarly to murine cells. Human peripheral blood mononuclear cells from three healthy donors (HD) and five CF patients carrying at least one deltaF508 allele, were cultured for 21 days with M-CSF, resulting in a population of adherent cells with macrophage-like morphology and immunophenotype (CD45 and CD14 positive). Cells from HD expressed low levels of CFTR. There was a modest increase in plasma membrane TLR4 expression in CF compared to WT cells. Moreover, the intracellular expression of TLR4 was significantly higher in CF compared to HD macrophages (p<0.05). The supernatants of untreated and LPS treated macrophages were analyzed for cytokine content by Luminex. Similar to our observations in murine cells, human CF macrophages had significantly (p<0.05) higher secretion of TNFalpha (6h, 24h), IL-6 (6h, 24h), MCP-1 (6h), MIP-1alpha (3h), G-CSF (6h, 24h), GM-CSF (3h, 6h, 24h), IP-10 (3h, 6h, 24h), RANTES (6h), and IL-1beta (6h, 24h). The basal level of IL-8 in the supernatant of both the WT and CF macrophage-like cells was abundant, but was significantly higher in CF compared to HD controls. With activation, IL8 secretion increased similarly in both Wt and CF cells. Thus, as we have shown for murine cells, human CF macrophages have abnormal TLR4 expression and signaling, reinforcing the hypothesis that immune cells contribute directly to the development of CF lung disease in humans.

Supported by the CFF and NIH. Introduction: Increased and prolonged Pseudomonas aeruginosa (PA)induced signalling via activation of Toll-Like Receptors (TLR2 and 4) contributes to chronic pulmonary inflammation in cystic fibrosis (CF). Internalisation of the receptor complex leads to activation of NF-kB and AP-1 pathways and subsequent pro-inflammatory response (eg. IL-8). Signalling is in cytosol, membrane, nuclear, and cytoskeletal fractions of CF matched cell line pairs (16 HBEo-Sense/Antisense and 9HTEo-PCEP/PCEP-R), and primary CF and non-CF HBE cells (obtained from Chantest Inc.). Furthermore, we treated cells with compounds that enhance Nrf2 release from Keap1 (CDDO and CDDO-Me) and a compound that increases CREB binding protein (CBP) availability for interaction with Nrf2 in the nucleus (Rp-cAMPS). We used proteomics to assay for the activity of Nrf2 by measuring the expression of PRDX1, 3, and 6, GST1, and TRX1.

We found that the amount of whole-cell unphosphorylated Nrf2 and phosphorylated Nrf2 were significantly decreased in CF cell lines compared to non-CF matched pairs (by 57.2 ± 12.4% for unphosphorylated and 32.8 ± 6.4% for phosphorylated Nrf2). Similar results were observed in CF primary HBE cells grown at an air-liquid interface with a decrease in CF cells of 38.4 ± 12.5% and 39.4 ± 5.7% in unphosphorylated and phosphorylated Nrf2, respectively. No significant differences were found in Keap1 levels in any of our models. Fractionation studies revealed significant decreases in Nrf2 in the cytosol and nuclear fractions of CF models. Conversely, Nrf2 levels in membrane and cytoskeletal fractions were significantly increased. These results are consistent with the decreased activation of Nrf2 in CF models, which we confirmed by proteomic analysis. Treatment with activators of Nrf2 did not increase whole cell or cytosol levels of the transcription factor in CF cells versus controls, but it did significantly increase levels in the nucleus, demonstrating potential approaches for the correction of Nrf2 dysfunction.

This work was supported by the Cystic Fibrosis Foundation. The novel receptor protein family T-cell immunoglobulin and mucin-domain containing molecules (TIMs) has been described as a critical regulator of the immune response. However, the role of TIMs has not been studied in CF despite the abnormal immune response in the CF lung. We therefore investigated the expression and function of TIMs in the CF airway epithelium.

Methods: TIM expression in normal (HBE) and CF bronchial epithelial cells (CFBE) was examined using real-time reverse transcriptase-polymerase chain reaction (RT-PCR), Western blotting and laser scanning microscopy. TIM cell surface localisation was analysed by membrane surface biotinylation and immunofluorescence confocal microscopy. Overexpression of TIMs in vivo was confirmed by RT-PCR in bronchial brushings from CF patients and control subjects. We also evaluated the effect of lipopolysaccharide (LPS) on TIM expression by laser scanning microscopy and RT-PCR. Cleavage of TIM-3 by neutrophil proteases was confirmed by Western blotting and laser scanning microscopy. Neutrophil elastase and proteinase-3 cleavage sites were identified by N-terminal sequencing. Levels of IL-8, galectin-9 and TIM-3 in bronchioalveolar lavage (BAL) from CF and non-CF bronchiectasis patients were measured by ELISA.

Results: We report for the first time that TIM-1 and TIM-3 are expressed in human bronchial epithelial cells. Furthermore, TIM-1 and TIM-3 are significantly upregulated in quiescent CF cells (p<0.05, n=3), implicating a role for CFTR in TIM expression. TIM-3 ligand, galectin-9 is also upregulated in CF (4-fold increase vs control, p<0.001, n=3). We also showed that TIM-3 expression can be modulated by pro-inflammatory stimuli. LPS (10 µg/ml) induced TIM-3 expression after 24 h (3 fold-increase in CFBE, n=3 vs. 2-fold increase in HBE, n=3, p<0.05). Increased TIM-3 expression was confirmed in bronchial brushings from CF patients (1.6-fold increase in CF, n=3 vs controls, n=3, p<0.05). However, TIM-3 and galectin-9 undergo rapid proteolytic degradation in the CF lung, primarily due to neutrophil elastase and proteinase 3 activity. Galectin-9 was not detected in CF BAL (n =10) but was detected in 6 out of 10 non-CF bronchiectasis BAL samples (13.36 -960.07 pg/ml). Importantly, levels of IL-8 correlated negatively with galectin-9 in non-CF bronchiectasis patients (Spearman r= -7629, p<0.05). stopped by targeting the receptor-LPS complex to lysosomal degradation or to the Golgi-apparatus for recycling. Our study aims to investigate the role of endo-lysosomal degradation in prolonged TLR2 and 4 signaling in CF epithelium.

Methods: Non-CF (16HBE41o-) and CF (CFBE410-) bronchial epithelial cell lines were grown to 80% confluency on collagen-coated plates prior to stimulation with 50 mg/ml PA-lipopolysaccharide (LPS; Sigma) for 0-24h. TLR2, TLR4 and NF-kB (p65) levels were measured by flow cytometry after stimulation. Nuclear extracts for NF-kB staining were obtained using a DNA CycleTest kit (BD Biosciences). IL-8 and IL-10 release was measured by ELISA (Peprotech) using supernatants from cells treated with PA-LPS. Endosomal markers (EEA1 (early) and RAB7 (late)), the lysosomal marker LAMP1 and the 58K Golgi protein were assessed by Western Blot. Statistics were calculated by ANOVA.

Results: CFBE cells showed prolonged intracellular expression of TLR2 and 4 following LPS stimulation. In CFBE cells, TLR2 and 4 expression was increased significantly after 4h (p<0.01 vs. 0h) and remained elevated up to 12h after stimulation. Following stimulation, peak TLR2 and 4 expression was observed after 1 and 4h respectively (p<0.05) but was not sustained in HBE cells. This was accompanied by a time-dependent increase in IL-8 release cells and a reduction in the release of the anti-inflammatory cytokine IL-10 in CFBE cells but not in HBE cells. Persistent expression of the NF-kB subunit p65 was observed in CFBE cells, while in HBE cells, p65 levels peaked following 2h exposure to LPS before returning to basal levels. CFBE cells showed prolonged expression of the early endosomal marker EEA1 (peak expression observed 4h post stimulation) and high expression of the late endosomal marker Rab7 at all time points (up to 24h). HBE cells showed peak expression of EEA1 and Rab7 at 1h stimulation with LPS and pronounced expression of LAMP1 (up to 24h), but failed to express 58K Golgi protein. Moreover, CFBE cells poorly expressed the lysosomal marker LAMP1 but showed notable expression of the 58K Golgi protein up to 24h post stimulation.

Conclusion: Our data suggest that in HBE cells LPS challenge results in formation of signaling endosomes [1] , which are subsequently targeted for lysosomal degradation [2] . However, in CF cells there is sustained endosome formation which is accompanied by increased pro-inflammatory IL-8 release. Moreover, upon stimulation with LPS CF cells show no ability to form degradative lysosomes, but display persistent 58K Golgi protein expression suggesting TLR recycling. Both sustained formation of signaling endosomes and rapid recycling of TLR contribute to prolonged pro-inflammatory responses in CF epithelium. This work is supported by CF Trust UK (PJ541 A strong body of evidence demonstrates a marked dysfunction of the resolution of inflammation in CF, in patients and experimental models. Previously, we reported that CF epithelial cell lines, primary cells treated with the CFTRinh-172, and airway epithelial tissues from R117H mutant mice fail to properly eliminate proinflammatory oxidants [Dysfunction of Nrf-2 in CF epithelia leads to excess intracellular H 2 O 2 and inflammatory cytokine production. Chen J, Kinter M, Shank S, Cotton C, Kelley TJ, Ziady AG. PLoS One. 2008;3(10):e3367]. Central to this dysfunction is a decrease in the activity of the transcription factor Nrf2 in CF epithelia. We showed that the failure to properly activate Nrf2 contributed to an increased inflammatory response, while correction of Nrf2 function significantly reduced it. To determine the mechanism by which Nrf2 is dysregulated in CF epithelia, we examined the Nrf2 activation cascade. In the inactive state Nrf2 is sequestered in the cytoplasm by its interaction with the cytoskeleton-associated protein, Keap1. When oxidant levels increase, Keap1 is oxidized and Nrf2 is released, then phosphorylated, and either enters the nucleus or is ubiquitinated and targeted for degradation. In the nucleus, Nrf2's interaction with a number of binding partners, such as CREB binding protein (CBP), modulates its activity. We used Western blot analysis to probe for unphosphorylated Nrf2, Keap1, and phosphorylated Nrf-2 in whole cell as well as Conclusions: Our work demonstrates the dysregulation of the TIM-3/galectin-9 signalling axis in the CF lung, due to proteolytic cleavage. This mechanism represents a novel intrinsic defect that contributes to the neutrophil-dominated immune response in the CF airway. Up until recently, neutrophil dysfunction in the CF airways has been mainly linked with necrosis and release of proteolytic enzymes and DNA. However, scientists and clinicians are now considering the blood CF neutrophil as inherently abnormal. Consistent with these views the circulating CF neutrophil has been shown to release greater levels of primary granule proteins, including neutrophil elastase and myeloperoxidase. In spite of this, recent in vitro studies have shown that CF neutrophils kill bacteria including Pseudomonas aeruginosa inefficiently. To address this anomaly, the aim of this study was to determine whether CF circulating neutrophils illustrated altered degranulation of secondary and tertiary granules when compared to healthy control cells.

Purified neutrophils were exposed to TNFα (10ng/2x10 7 cells) for 0, 5, 10 or 20 minutes. The level of degranulated proteins in surrounding supernatants was determined by Western blot analysis and ELISA. Proteins of interest included LL-37 and lactoferrin as markers for secondary granule releases and matrix metalloprotease 9 (MMP-9) as a marker for tertiary granules. Additionally stimulated neutrophils were probed for CD66b, a membrane marker for successful secondary and tertiary granule degranulation, analysed by flow cytometry. The p value was determined with Student's t test.

CF neutrophils released 50% less lactoferrin and LL-37 after 10 and 20 min of TNFα stimulation when compared to neutrophils of healthy controls (p<0.05). Moreover, release of MMP-9 from tertiary granules of CF neutrophils was decreased by 75% after 20 min stimulation (p<0.005). In addition, surface expression of CD66b, a marker for secondary and tertiary granules, was significantly reduced in CF cells as measured by flow cytometry (p<0.05 for 5 and 10 min, p<0.005 for 20 min post stimulation). Impaired response to TNFα may cause decreased degranulation of antimicrobial proteins from secondary and tertiary granules by CF circulating neutrophils and impact upon the killing ability of these innate immune cells. The underlying disease mechanism of CF remains unclear. It is generally believed that impaired chloride transport due to reduced CFTR function impairs transepithelial salt/water balance and dehydrates the airway surface. As a consequence, viscous mucus inhibits antimicrobial activity and promotes recurrent bacterial infection. The continued bacterial colonization in the CF lung leads to a hyperinflammatory response with persistent infiltration of immune cells. However, the CF lung is not able to suppress bacterial infections, even with increased inflammatory response, suggesting the possibility of innate immune suppression. Recent data has indicated that Pseudomonas aeruginosa secretes a quorum sensing homoserine lactone (3OC12-HSL) that represses innate immune function by disrupting critical transcriptional responses mediated by nuclear factor-κB (NF-κB). As such, P. aeruginosa may actively establish and maintain infections in CF, rather than merely flourishing in a favorable niche. We developed a high-throughput screen to identify small molecules that block HSL-mediated inhibition of NF-κB signalling to be used as a research tool to investigate the immune response repression in P. aeruginosa colonization of the lung and as potential drug candidates. A stable Fischer rat thyroid cell line was generated that expresses luciferase under the control of NF-κB-response elements (FRT-NF-κB cells). Stimulation of FRT-NF-κB cells with lipopolysaccharide (LPS, 0.5 µg/ml for 4 hours) resulted in a robust luminescence response (~5times background signal), which was suppressed completely by 3OC12-HSL (at 25-50 µM). This cell-based assay was adapted for high-throughput screening, giving a robust Z'-factor > 0.7. Primary screening of 25,600 drug-like synthetic small molecules gave several classes of active compounds that blocked 3OC12-HSL suppression of NF-κB signalling by >70 %. Secondary screening verified the activity of six compound classes in FRT-NF-κB cells, including benzothiazoles and isoxazoles. Current efforts are focused on testing of these compounds in additional relevant cell types, including macrophages, optimization of leads, and testing in animal models of inflammation. Small-molecule 3OC12-HSL inhibitors may be useful in studying mechanisms of P. aeruginosa colonization and lung inflammation, and may provide a novel approach to treat P. aeruginosa infection in CF.

Funded by the CFF and NIH. Background: Cystic fibrosis (CF) patients develop chronic bacterial infections reflecting a dysfunctional innate immunity. Neutrophil gelatinaseassociated lipocalin (NGAL) is released from neutrophils and epithelial cells and functions as a cell growth and differentiation factor by donating iron to cells. NGAL also mediates an innate immune response to bacterial infection by scavenging bacterial iron-chelating siderophores. NGAL is upregulated following bacterial infection in a TLR4-dependent manner. Due to heavy bacterial colonization in CF lungs, LPS and other bacterial ligands that induce innate immunity are shed in large quantities. The aim of this study was to determine NGAL levels in serum from CF subjects compared to healthy donors and the effect of circulating NGAL on macrophage innate immune responses.

Methods: NGAL levels were measured by ELISA method in sera from 33 CF subjects, 33 septic patients and 21 healthy controls.

Results: Serum NGAL levels were highly elevated in CF patients when compared to healthy controls (p<0.0001). The mean serum NGAL level measured in CF was found to be 79.7 ± 24.6 ng/ml (range 40 -135 ng/ml), compared to septic patients 130.7 ± 34.68 ng/ml (range 68 -211 ng/ml), or healthy donor controls 47.93 ± 11.5 ng/ml (range 22 -66 ng/ml). We then investigated the effect of elevated NGAL on macrophage innate immune function using THP-1 human macrophage-like cells treated with recombinant NGAL protein (100 ng/ml) overnight prior to stimulation with Pseudomonas aeruginosa LPS (10 pmol/ml). We found that NGAL suppressed LPS-induced cytokine production in THP-1 macrophages as demonstrated by decreased production of the inflammatory cytokine TNFα and inhibition of the pro-inflammatory chemokine IP-10 (CXCL10) release. We also investigated reactive oxygen species (ROS) release from macrophages as a defense mechanism that is important for the oxidant killing of invading pathogens; however, excessive ROS release also contributes to tissue damage. We observed that exposure to CF serum primed macrophages for massive ROS release, reflecting a pathological oxidative stress in CF host. ROS release was detected using the chemiluminescent probe lucigenin in THP-1 cells primed with either CF serum (50 µl /10 million cells in 10 ml volume), or the TLR2 ligand Pam3CSK4 (1µg/ml), TLR4 ligand P. aeruginosa LPS (10 pmol/ml), or recombinant TNFα (1ng/ml). Taken together, the data suggest that under the pressure of persisting bacterial infections, NGAL levels continue to rise in CF patients leading to suppressed macrophage innate immune responses.

Conclusions: Elevated NGAL levels contribute to dysfunctional innate immunity in CF and NGAL could be a biomarker of progressive CF disease. clinical strains) with or without selective EGFR antagonists. VEGF-A culture medium concentrations were measured by ELISA. C57/Bl6 mice (n=8/group) were instilled intratracheally with sterile or PAO1-coated (5.105 CFU/animal) agarose beads with the EGFR inhibitor AG1478 (12.5 mg/kg/d, i.p.) or vehicle. Mice were sacrificed 7 days after instillation. Immunostaining for VEGF-A, for phosphorylated (activated) EGFR and for the endothelial marker von Willebrand factor (vWF) was performed and analyzed using quantitative morphometric analysis. Western blotting for VEGF-A was performed on lung homogenates.

Results: In NCI-H292 cells, all PA supernatants induced dose-dependent VEGF synthesis, which was prevented by EGFR inhibition. In control mice, immunostaining for VEGF was weak and localized to airway epithelium, and no staining was observed for phosphorylated EGFR. Instillation of PAO1 coated beads resulted in rapid increase in VEGF immunostaining in epithelium and in alveolar macrophages. Western blot analysis confirmed that PAO1 infection induced VEGF synthesis that was present at day 1 after instillation and peaked at day 7, whereas sterile beads were without effect. At day 7, peribronchial vascularity was doubled in PAO1-infected mice, due to an increase in small vessels. PAO1 infection also induced marked staining for phosphorylated EGFR in airway epithelium, and treatment of mice with AG1478 abolished EGFR phosphorylation and PAO1-induced VEGF expression in airway epithelium, and reduced peribronchial vascularity.

Conclusion: Our results suggest that PA infection induces abnormal bronchial angiogenesis at least in part via EGFR-dependent VEGF synthesis in airway epithelium.

Supported by Vaincre la Mucoviscidose, Chancellerie de l'Université de Paris (Legs Poix).

Seegmiller, A.; Katrangi, W.; Laposata, M. Department of Pathology, Vanderbilt University Medical Center, Nashville, TN, USA Objectives: Patients with cystic fibrosis (CF) exhibit reproducible abnormalities in essential fatty acid (EFA) composition, including decreased linoleate (LA) and docosahexaenoate (DHA), and variably increased arachidonate (AA) and Mead acid (MA). These changes are independent of nutritional status and correlate with the severity of disease symptoms and have been replicated in animal and in vitro cell culture models. Work in these systems suggests a role for EFA alterations in CF pathogenesis. However, the underlying mechanisms of these changes have not been well understood. Recent work shows that EFA alterations are due to increased expression and activity of fatty acid ∆5and ∆6-desaturases in human respiratory epithelial cells that lack CFTR expression. These enzymatic changes can be corrected by exogenous supplementation with DHA, which also corrects the EFA abnormalities. Expression of both desaturases is regulated, in part, by sterol regulatory element binding proteins (SREBPs), particularly SREBP-1. The objective of this study is to determine if SREBP-1 plays a role in the generation of EFA abnormalities in CF.

Methods: Experiments were performed in cultured 16HBE human bronchial epithelial cells stably transfected with plasmids containing the first 131 nucleotides of CFTR in the sense or antisense orientations, resulting in a wild type phenotype that expresses normal CFTR protein, or a CFlike phenotype that does not express CFTR protein, respectively. Expression of SREBP isoform mRNAs was measured by quantitative RT-PCR using RNA from sense and antisense cells in the presence or absence of DHA. SREBP-1 expression was blocked by transfecting cells with specific siR-NAs. The effect of this treatment was determined by measuring expression of desaturase mRNAs by RT-PCR and EFA composition by gas chromatography-mass spectrometry.

Results: Expression of both SREBP-1a and -1c isoform mRNAs is significantly increased in antisense versus sense cells. DHA supplementation results in a marked suppression of SREBP-1a and a lesser diminution of SREBP-1c, which correlates with decreased expression of fatty acid desaturases. Blocking SREBP-1 expression by transfection of specific siRNAs also results in significant decreases in desaturase expression as well as increased LA and decreased AA and MA in antisense cells. Background: Persistent airway inflammation and infection are key contributors in the disease pathogenesis of cystic fibrosis (CF) lung disease. Gamma delta T cells are an immunoregulatory subset of T cells exhibiting diverse functions, however they have been implicated in the disease pathogenesis of several chronic lung disorders. The relationship between bacterial infection, gamma delta T cell cytokines and clinical status in children with CF remains unexplored.

Aim: To study the relationship between bacterial colonization, gamma delta T cell cytokines (IL-2, IL-8, IL-4 and IL-10) and clinical status in children with CF.

Material and Methods: Twenty children with CF (7-14 years), in a stable clinical condition were enrolled from the Outpatient Clinic of the Department of Pediatrics PGIMER, Chandigarh, India. Sputum induction was carried out according to standard procedure. Gamma delta T cell cytokines (IL-2, IL-8, IL-4 and IL-10) (FITC or PE labeled) were estimated using flow cytometry upon stimulation with PMA -Io. Sputum microbiology was performed according to standard guidelines. Clinical status was ascertained by Shwachman scoring.

Results: Fourteen children were infected with Pseudomonas aeruginosa (n=14, CFi+), six children were free from any detectable bacterial infection (n= 6, CFni) at the time of the study. We observed a high percentage of gamma delta IL-2 CFi (28.9+ 3.9) vs CFni (22.1+ 3.1) vs controls (12.4+ 2.2) p<0.05, gamma delta IL-8 CFi (24.4+ 3.6) vs CFni (16.8+ 2.4) vs controls (10.1+ 2.5) p<0.05. No significant difference between gamma delta IL-4 in both the groups was observed. Gamma delta IL-10 percentage was significantly lower in the infected group CFi (11.0+ 2.4) as compared with non-infected CFni (15.1+4.1) vs controls (19.9+ 3.2) p<0.05. We observed a significant positive correlation between gamma delta IL-8 and Pseudomonas colonization (p<0.05), an inverse correlation between gamma delta IL-8, IL-2 and Shwachman score in the subgroup CFi (p<0.05). A significant positive correlation between IL-10 and Shwachman score (p<0.05) was observed.

Conclusion: Our findings demonstrate a significant correlation between bacterial colonization, gamma delta T cell cytokines and clinical status in children with CF, thereby highlighting the role of gamma delta T cells in perpetuating the airway inflammation in children with CF. Background: Abnormal angiogenesis in peribronchial blood vessels presumably contributes to pathophysiology of hemoptysis in cystic fibrosis (CF) airways. We have previously described that increased numbers of peribronchial blood vessels in CF bronchi were associated with increased expression of VEGF-A in airway epithelium. Pseudomonas aeruginosa (PA) infection in airways is a feature of CF and transgenic expression of VEGF-A in airway epithelium induces peribronchial angiogenesis. We hypothesized that PA infection induces VEGF epithelial synthesis and peribronchial angiogenesis. Because EGF receptor (EGFR) activation regulates VEGF synthesis in cancer, we also evaluated the role of EGFR in vitro and in mice.

Methods: Airway epithelial (NCI-H292) cells were incubated 24h with bacteria-free PA supernatants (laboratory strain PAO1 and 6 different CF Conclusions: Alterations in EFA composition and enzyme expression in a CF cell culture system correlate with increases in SREBP-1 expression. The changes can be reversed, in part, by blocking SREBP-1 expression either by DHA or by transfection of specific siRNAs. These findings suggest that SREBP-1 may play an important role in the production of EFA abnormalities in CF and in their correction by DHA. CF lung pathology is dominated by a huge recruitment of neutrophils in the bronchial lumina, in which IL-8, mainly released by respiratory epithelial cells, plays a key role. IL-8 transcription is regulated by several transcription factors (TFs) such as NF-kB, AP-1 and NF-IL6. Recently, it has been reported that also the TF CHOP (GADD153) is involved in IL-1β-and PGE2-dependent IL-8 transcription in CF bronchial epithelial cells (IB3-1) [1] . The aim of this investigation is to detect the possible involvement of CHOP also in P. aeruginosa-dependent IL-8 transcription and gain further insights in the transmembrane signalling leading to CHOP activation. Decoy oligodeoxynucleotides (ODNs) strategy [2] has been applied in IB3-1 cells pre-incubated for 24 hrs with ODNs mimicking the CHOP consensus sequence localized in the IL-8 promoter region, before exposure to P. aeruginosa PAO1 strain, IL-1β or TNF-α for further 4 hrs, before quantifying IL-8 mRNA. We confirm that CHOP is implicated in IL-1β-dependent IL-8 transcription [1] , and we suggest its involvement also in P. aeruginosadependent pro-inflammatory signalling. Interestingly, CHOP does not seem to participate in IL-8 transcription mediated by TNF-α. It has been previously reported that CHOP is induced by the p38 MAPK and HSP27 pathways in human melanoma cells [3] . Therefore we tested the pattern of phosphorylation of 21 different phosphokinases with an immunoenzymatic nitrocellulose membrane assay and we found that: a) P. aeruginosa leads to the phosphorylation of the MAPK p38 isoforms α, δ and γ, RSK-1, GSK-3α/β and HSP27; b) IL-1β of MAPK p38 isoform α, δ and γ, RSK1, ERK2, MSK2, and HSP27; c) TNF-α of MAPK p38 isoform δ, JNK2 and JNK3. Thus, in our model, we confirm that CHOP is involved in the IL-8 transcription when induced by P. aeruginosa and IL-1β in human bronchial epithelial cells and this seems associated with the activation of HSP27 and possibly the MAPK p38 isoforms alpha and gamma, but not delta. Previously, we reported that extracts from N. arvensis seeds, which are widely used as anti-inflammatory remedies for skin and pulmonary inflammatory diseases in traditional medicine of Northern Africa, were able to inhibit selectively the expression of the pro-inflammatory chemokine IL-8 in IB3-1 CF bronchial cells exposed to the P. aeruginosa laboratory strain PAO1 [1] . In this study we analyzed the chemical composition of the extracts and identified four major compounds: beta-sitosterol, stigmasterol, campesterol and thymoquinone. The first three molecules are phytosterols, cholesterol-like molecules found in plant material with the highest concentrations occurring in vegetable oils. Phytosterols act as a structural components in the vegetal cell membrane, a role which is played by cholesterol in mammalian cells. The molecules found in N. arvensis seeds were purchased and tested in IB3-1 CF bronchial cells exposed to the P. aeruginosa laboratory strain PAO1 for 4 h. IB3-1 cells were pre-treated for 20 h with increasing doses, ranging from 1 nM to 200 µM for beta-sitosterol; to 100 µM for stigmasterol; to 10 µM for campesterol; to 100 µM for thymoquinone. Total RNA was extracted and the neutrophil chemokine inflammatory marker IL-8 transcript was quantified. Beta-sitosterol was the only compound that significantly inhibited the PAO1-dependent transcription of IL-8 starting from the dose of 100 nM. By now, limited data has suggested that phytosterols may help reducing inflammation [2] . These data provide encouraging results supporting the concept of developing bioactive compounds following screening active leads purified from medicinal plants to identify safe and innovative pharmaceutical molecules to control lung inflammation in CF patients. Objectives: Innate defence regulator (IDR) peptides are synthetic derivatives of endogenous host defence peptides, and are currently under investigation as novel anti-infective agents. IDRs lack effective antimicrobial activity; rather, they enhance bacterial clearance by increasing leukocyte recruitment to infection sites while simultaneously suppressing harmful inflammation. As cystic fibrosis (CF) pathology arises from chronic bacterial infections which in turn drive chronic lung inflammation, we sought to investigate the ability of a lead IDR (IDR-1018) to modulate human bronchial epithelial (HBE) cell responses to pro-inflammatory stimuli.

Methods: Primary normal HBE cells, CF HBE cell lines (IB3-1, CuFi) or control HBE cell lines (C38, NuLi, 16HBE) were stimulated for 24 hr with flagellin or heat-killed Pseudomonas aeruginosa PAK (MOI=50), with or without 1 hr pretreatment with IDR-1018, and ELISA was used to measure IL-6 and IL-8 in supernatants. For microarray experiments, IB3-1 and C38 were stimulated for 2 hr with flagellin, with or without 1 hr pretreatment with IDR-1018. RNA was harvested, converted to cDNA, and hybridized to Illumina HumanHT-12 chips. Systems biology and network analyses were carried out using our new systems biology platforms: Innat-eDB, Cerebral (Cytoscape) and MetaGEX. Pathways were validated by stimulating IB3-1 for 24 hr with flagellin, with or without 1 hr pre-treatment with metformin, Akt Inhibitor VII, or GSK-3β inhibitor, and ELISA was used to measure IL-6 and IL-8 in supernatants.

Results: IDR-1018 was found to enhance IL-6 and IL-8 secretion induced by heat-killed P. aeruginosa or flagellin in primary HBEs and normal cell lines but in contrast had an inhibitory effect on cytokine secretion in CF cell lines. Cytokine modulation by IDR-1018 was not observed in cells stimulated with PAK ∆fliC, suggesting that immunomodulation in HBEs was flagellin-dependent. Systems biology and network analyses identified multiple dysregulated genes, pathways, and transcription factors medi- Background: Clinical studies have linked increased sputum and peripheral blood neutrophil MPO activity with increased airflow obstruction in cystic fibrosis (CF) patients of the same age, gender, airway bacterial flora, and CFTR genotype. Variations in the TGF-b1 gene associated with increased TGF-b1 production have been linked to worse airflow obstruction in CF patients of similar age, gender and CFTR genotype.

Objectives: We hypothesized that in the presence of MPO, TGF-b1 production would increase in airway epithelial cells and this would be modified by the CFTR function of these cells and by variation in the TGF-b1 promoter region (-509C-T, c.-1347C>T) and in codon 10 (+869T-C, c.+29T>C).

Methods: We obtained normal and CFTR mutant (del508/W1282X) human airway epithelial cells (hAEC), and cultured them in collagen coated dishes, with or without CFTR inhibitor 172. TGF-b1 mRNA was measured by PCR, and TGF-b1 protein was detected by immunofluorescence staining and Western blot.

Results: We found increased TGF-b1 mRNA after hAEC were exposed to MPO (at activities found in CF sputum) normalized to mRNA of constitutive beta actin. Under identical conditions, TGF-b1 protein expression was increased in hAEC. The magnitude of the TGF-b1 mRNA response of bronchiole derived hAEC to MPO differed depending on their TGF-b1 promoter and codon 10 genotypes. Reduced CFTR function resulted in increased TGF-b1 mRNA, but this increased response, using bronchial derived hAEC of the same TGF-b1 promoter and codon 10 genotypes, did not show any further increase on exposure to MPO.

Conclusion: Neutrophil MPO present on the apical surface of cultured hAEC induces transcription of TGF-b1 that is modulated by variation in their TGF-b1 promoter, codon 10 genotypes and airway site of origin, and is upregulated by CFTR dysfunction. 1. Physiology, Dartmouth Medical School, Hanover, NH, USA; 2. Microbiology and Immunology, Dartmouth Medical School, Lebanon, NH, USA; 3. Immunology, Roswell Park Cancer Institute, Buffalo, NY, USA We previously reported that a secreted toxin from P. aeruginosa, Cif (PA2934), reduces CFTR-mediated chloride secretion by human airway epithelial cells by enhancing the amount of ubiquitinated CFTR and its degradation via the lysosomal pathway (Swiatecka-Urban, et al. 2006; Bomberger et al. 2009 ). The aim of the current study was to determine if Cif regulates other ABC transporters, specifically the transporter associated with antigen processing (TAP). Patients with TAP deficiency, like CF patients, have recurrent bacterial infections, nasal polyps, and chronic purulent rhinitis in the first six years of life, and chronic bacterial infections, bacterial pneumonia, and bronchiectasis later in life. Thus, we hypothesized that Cif facilitates the degradation of the TAP complex, resulting in reduced viral antigen presentation via the major histocompatibility complex (MHC) class I molecules and therefore, diminished immune response to viral pathogens. Recombinant Cif protein reduced TAP1, but not TAP2, protein abundance in polarized human airway epithelial cells (CFBE41o-cells) in a time-dependent manner. The Cif-mediated reduction in TAP1 was more dramatic in CFBE41o-cells, as compared to WT-CFTR complemented CFBE41o-cells. Cif inhibited ubiquitin specific protease 10 (USP10), a deubiquitinating enzyme, resulting in increased poly-ubiquitination and proteasomal degradation of TAP1. The inhibition of USP10 by Cif was more profound in CFBE41o-cells as compared to WT-CFTR complemented CFBE41o-cells. The Cif-mediated decrease in TAP1 expression reduced ating the differences between normal and CF cells, including the AMPK-Akt-GSK-3β axis. Pretreating IB3-1 cells with metformin (AMPK activator) or inhibitors of Akt or GSK-3β had negligible or subtle inhibitory effects on flagellin-specific responses, but completely reversed the ability of IDR-1018 to inhibit airway cell cytokine secretion.

Conclusions: Taken together, these data implicate the AMPK-Akt-GSK-3β axis in mediating dysfunctional responses of CF cells to immunomodulatory therapeutics, and indicate the utility of systems biology in identifying signalling pathways which underlie abnormal pro-inflammatory cytokine secretion by CF cells. Mechanical clearance of mucus constitutes the primary airway defense mechanism and without which, the mucus layer is not only ineffective in airway defense, but the accumulated mucus contributes to airway pathophysiology. Traditionally, two primary mechanical mechanisms are known to be important in airway mucus clearance; 1) mucociliary clearance (MCC) and 2) cough clearance. While both MCC and cough clearance have been extensively studied, there have been almost no studies on a purported third mechanism of mucus clearance; cilia-independent two phase gas-liquid transport (GLT). This clearance mechanism was proposed over two decades ago using mechanical models to suggest airflow-mediated mucus clearance at airflow velocities of normal breathing. Despite the potential importance of gas-liquid transport in airway clearance, studies directly investigating this clearance mechanism using in vivo systems are completely absent. To investigate the role of GLT on airway clearance, we developed a mouse model that was devoid of ciliated cells (by SO 2 exposure, 500 ppm for 3 hrs) and studied the effect of breathing on radiotracer clearance. To assess mucus transport by both these mechanisms, we used radiolabeled sulfur colloid particles (99mTc-SC) focally deposited (150 nl of radiotracer) in the main stem bronchus and imaged clearance by gamma camera scintigraphy. At day 3 post-SO 2 treatment, radiotracer clearance was measured on SO 2 -treated mice as well as controls. While the rate of clearance in the SO 2 -treated mice was reduced about 2 fold from the control mice (from 13.5±1.0 to 8.7±3.0 %/min), these mice were still able to clear the tracer in the absence of cilia. We allowed the mice to recover for one day following the first imaging procedure, and then again anesthetized the mice, delivered the isotope and immediately killed the mice by exsanguination and repeated our imaging protocol. Because cilia will still continue to beat effectively for many hours following death, the control mice still exhibited a robust clearance (12.2±1.0 %/min) which was not significantly different from alive, control, mice. This suggests that when cilia are capable of beating effectively, MCC represents the dominant mechanism of mucus clearance. However, after death and cessation of breathing, the SO 2 -treated mice exhibited virtually no airway clearance (0.4±0.3 %/min). These data indicate that in the absence of beating cilia, the movement of air in live SO 2 -treated mice is capable of resulting in substantial airway clearance. However, without airflow or the mechanical forces associated with breathing, airway clearance dropped to zero. These data demonstrate for the first time that GLT is an important mechanism of airway clearance, providing significant clearance when mucociliary clearance is impaired. We predict that gas-liquid transport is an important mechanism for airway clearance, and identifying variables that improve GLT may be important in optimizing this method of airway clearance in vivo including in CF patients. Because CF airways exhibit fluid hyperabsorption, resulting in PCL collapse and mucus thickening, we speculate that this transport system is defective in CF.

Funded by the CFFT Mucus Clearance Consortium.

peptide uptake into the endoplasmic reticulum and decreased MHC class I molecule abundance at the plasma membrane, effects that were enhanced in CFBE41o-cells compared to WT-CFTR complemented CFBE41o-cells. Furthermore, Cif decreased MHC class I presentation of peptide in a timedependent manner and eliminated CD8 T cell recognition of influenza Ainfected cells. We propose that P. aeruginosa, by secreting the Cif toxin, inhibits the ability of CD8 T cells to recognize and eliminate viral infections, severely compromising the immune defenses in the lungs of CF patients. This is the first bacterial toxin that inhibits the viral immune response of the host, potentially illuminating a mechanism for the clinical observation that viral-bacterial co-infections dramatically increase the morbidity of CF patients chronically colonized by P. aeruginosa. This work was supported by grants from the CF Foundation, BOMBER08F0 and Research Development Program (RDP: STANTO7R0), and the NIH (R01-HL074175 and P20 RR-018787). A structured-based Virtual Screening (VS) of differently substituted furocoumarins and analogues has been carried out against NF-κB to select molecules for their ability to inhibit the binding of this transcription factor with the DNA. The focus library was developed starting from known chemical structures from literature as well as retrieving compounds from suitable commercial databases on the network. The first step of the present study was the construction of a customized small molecules database. The prepared library of 1710 compounds contained 54 structures of angelicin derivatives (molecular weight range: 186-478), 1384 of psolaren derivatives (molecular weight range: 186-619), 51 of Furo(3,2-c) chromen-4one structures (molecular weight range: 268-439) and 166 of benzoquinolizin-5-one analogues (molecular weight range: 258-499). Molecular docking approach was followed against NF-κB in both p50 aggregation states using Glide software. The docking studies were carried out by using the extra precision (XP) method on a subset of 1000 molecules selected by preceding Standard Precision (SP) procedure as implemented in Glide software. The selection of interesting docked compounds was based on the docking GlideScore (G-Score) values in each target. All the individuated best ranked ligand poses in both dimer and monomer, were localized in the DNA binding surface of the N-terminal domains involved in the formation of the solvent-filled cavity as in case of active cellular dimeric aggregation process. Among the ten highest-scored ligands selected from docking studies, five different molecules were commercially available and investigated in further experimental analysis. Four furocoumarin derivatives showed IC50 values in the range of 40-100 µM in electrophoretic mobility shift assay (EMSA). Three compounds significantly inhibited NF-κB dependent biological functions in cystic fibrosis IB3-1 cells infected with Pseudomonas aeruginosa (expression of IL-8). These findings on one hand validated the virtual screening approach here presented and reinforce the successful results of our previous computational studies to a difficult target as NF-κB. On the other hand, the discovered four novel compounds could be of relevance to identify more potent inhibitors of NF-κB dependent biological functions which could be beneficial to control lung inflammation of CF patients. Background: CF is characterized by chronic pulmonary inflammation causing damage and repair mechanisms that lead to fibrosis. The alternatively activated macrophage (AAM) phenotype is an important subtype of cell involved in tissue remodeling-the role of AAM in the CF lung has not been investigated. Previous work in our lab has associated AAM markers with pulmonary function decline in a CF patient population. We have also demonstrated that azithromycin (AZM), a common therapy for CF, induces markers of alternative activation. AAM are important in repair mechanisms and orchestrate the build-up of the extracellular matrix (ECM) through production of key mediators such as transforming growth factor-β (TGFβ). Conversely, proteins involved in turnover of the ECM, such as matrix metalloprotease-9 (MMP-9), are down-regulated. Both TGFβ and MMP-9 have been shown to be increased in CF patients.

Objective: Because AZM induces an AAM-like phenotype, we hypothesized that AZM would increase TGFβ and decrease MMP-9 both in vitro and in sputum samples of human subjects.

Methods: In vitro, the macrophage cell line J774 was cultured with the fibroblast cell line NIH3T3. Cells were treated with interferon gamma (INFγ), infected with Pseudomonas aeruginosa, and exposed to AZM. Cells and supernatants were collected to analyze mRNA and protein levels of MMP-9 and TGFβ. For our human subjects, sputum samples were collected from stable CF patients with and without AZM. Gene transcription analysis was performed to analyze fibrotic mediators including TGFβ and MMP-9. Clinical data at the time of the sputum sample was also collected.

Results: In vitro, AZM increased gene transcripts of MMP-9 and TGFβ compared to cells that were not treated with AZM. MMP-9 mRNA was increased at 30-60min after AZM treatment, while protein concentrations were doubled in AZM treated cells from 12 hours after AZM treatment, with a p value of < 0.001. Furthermore, neutralizing TGFβ activity did not affect MMP-9 concentrations, suggesting that AZM has a different mechanism for increasing MMP-9. MMP-9 and TGFβ transcripts are currently being analyzed in 5 CF patients with recruitment ongoing. MMP9 and TGFβ levels will be correlated to lung function to determine if AZM alters this relationship.

Conclusions: AZM increases both TGFβ and MMP-9 at the mRNA level as well as the protein level. We confirmed part of our initial hypothesis that AZM would increase TGFβ but disproved the hypothesis regarding AZM's negative effect on MMP9 gene and protein expression. Further, the effects on MMP-9 do not appear to be dependent on TGFβ activity. These data indicate that AZM may have a more complex role in ECM buildup and turnover than initially predicted, and may help explain the efficacy of chronic AZM treatment in patients with CF. 1 1. Physiology, Dartmouth Medical School, Hanover, NH, USA; 2. Microbiology and Immunology, Dartmouth Medical School, Hanover, NH, USA Recent studies (7, 8) suggest that the ∆F508 mutation in CFTR reduces the proinflammatory response to infection in airway epithelia. We therefore applied a novel metanalytical approach (1) to microarray studies that measured the effect of 1 to 4 hour P. aeruginosa (PA) exposures on human airways cell lines expressing either ∆F508-CFTR or wt-CFTR (2) (3) (4) (5) (6) . These included two ∆F508-CFTR cell lines (IB3 and CFBE41o-) and four wt-CFTR (S9, A549, CALU3 and BEAS2B) lines. We found that PA exposure activated a suite of inflammatory genes including IL-6, IL-8, and NFκB1 in all 6 lines. Notably, PA exposure caused less robust cytokine gene activation in ∆F508-CFTR lines than in wt-CFTR lines (P<0.05). As many would predict stronger activation by ∆F508-CFTR cell lines, we performed microar-p<0.05). In contrast, no relationship was observed between serum vD and BAL LL-37. This molecule is known to be sensitive to proteolysis and can be degraded by both neutrophil elastase and MMPs, which could therefore mask any effect of vD on LL-37 expression. Levels of proteolytic enzymes are currently being measured in these samples.

Children with CF are at risk of low vD levels even if supplements are being prescribed. In addition to the potential detrimental effects on bone metabolism, this may increase the susceptibility of the airway to infection by reducing local levels of innate defence molecules. Objectives: Patients with cystic fibrosis (CF) consistently demonstrate selective abnormalities in essential fatty acid (EFA) concentrations, including decreased linoleate (LA) and docosahexaenoate (DHA). Increased arachidonate (AA) is also commonly found. These changes are independent of absorption status and appear important for the pathophysiology of the disease. However, the mechanisms of these alterations are not clearly understood. The aim of this study was to determine whether changes in the expression and/or activity of specific EFA metabolic pathway enzymes could account for the fatty acid changes observed in CF.

Methods: Experiments were performed in cultured human bronchial epithelial cells (16HBE) transfected with plasmids expressing either wildtype (sense) or antisense cftr fragments, resulting in the presence or absence of functional CFTR protein, respectively. To evaluate fatty acid metabolism, the cells were grown to confluence and then incubated for 4 hours with one of several 14 C-labeled fatty acid substrates. Metabolism of the radiolabeled substrates to downstream products was assessed by HPLC. Gene expression was determined using quantitative RT-PCR and immunoblotting. Culture medium eicosanoid levels were measured using ELISA.

Results: Antisense (CFTR-) cells demonstrated increased metabolism of LA to AA (n-6 pathway) compared to sense cells. There was a parallel increase in metabolism of linolenate (LNA) to eicosapentaenoate (EPA; n-3 pathway) in antisense cells. These changes correlated with increased expression of fatty acid ∆5and ∆6-desaturases, key enzymes in these metabolic pathways. In contrast, the antisense cells showed decreased metabolism of AA and EPA to docosapentaenoate (DPA) and DHA despite increased or unchanged expression of the relevant metabolic enzymes, suggesting that these precursors are metabolized by an alternative pathway. Indeed, the expression of both cyclooxygenase-2 and lipoxygenase-5, as well as the production of prostaglandin E 2 and leukotriene B 4 was significantly increased in antisense cells compared to sense cells.

Conclusions: These findings demonstrate that the diminished LA and increased AA in CF cells may result from increased metabolism of LA, due to increased expression and activity of fatty acid desaturases. In addition, they suggest that the observed decrease in DHA levels may be due, in part, to consumption of its precursor, EPA, in the increased production of oxygenated fatty acid products.

Njoroge, S.W.; Laposata, M.; Seegmiller, A. Department of Pathology, Vanderbilt University Medical Center, Nashville, TN, USA Objectives: Significant alterations in essential fatty acid (EFA) levels have been observed in the blood and tissue of cystic fibrosis (CF) patients, as well as in mouse and cell culture models of CF. These alterations include decreased linoleic acid (LA), decreased docosahexaenoic acid (DHA) and increased arachidonic acid (AA). Oral administration of DHA to cftr -/mice has been shown to correct these fatty acid changes and reverse the pancreat-ray, qPCR and ELISA measurements in matched CFBE41o-cells (∆F508/∆F508 versus CFBE41o-cells complemented with wt-CFTR). CFBE41o-∆F508/∆F508 cells produced 40% as much IL-8 than CFBE41ocells complemented with wt-CFTR as measured by ELISA (P<0.05) in response to PA. Moreover, PA induced IL-8 gene expression in CFBE41o-∆F508/∆F508 cells was also 40% of CFBE41o-cells complemented with wt-CFTR, consistent with previous experiments (7,8) in CFBE41o-and CFT1 cells. Thus, in CF, the loss of wt-CFTR reduces the proinflammatory response of airway epithelial cells to P. aeruginosa, which may contribute to the chronic infection with P. aeruginosa in CF patients.

1. Hampton & Stanton. A novel approach to analyze gene expression data demonstrates that the DF508 mutation in CFTR downregulates the antigen presentation pathway. Am J Physiol Lung Cell Mol Physiol (2010) vol. 298 (4) Vitamin D (vD) levels have been reported to correlate with lung function in healthy populations and disease severity in pulmonary TB, COPD and asthma. One proposed mechanism, supported by in vitro studies, is that there are important vD response elements present in the promoter regions of genes encoding certain host defence molecules including members of the defensin family and cathelicidin (LL-37). As patients with CF are at risk of deficiency of fat-soluble vitamins despite supplementation, we sought to explore this hypothesis in vivo in a cohort of CF children.

Frozen serum and bronchoalveolar (BAL) fluid samples, which had been donated for research at the time of a clinically-indicated bronchoscopy were available from 49 children with CF. Mean age at the time of the procedure was 6.8 years (range 0.03-15.99). Forty-five (92%) were biochemically pancreatic insufficient and were receiving pancreatic enzyme supplementation and supplemental fat-soluble mulitvitamin preparations. Serum 25OH vD3 was measured using HPLC and mass spectrometry. BALF human β defensin-2 (hβD2) and LL-37 were quantified using ELISA. VD deficiency was defined as <20 ng/ ml based on internationally-accepted criteria. Deficient and sufficient groups were compared with Mann Whitney tests and Spearman's correlations were performed.

Sixteen (33%) children were vD-deficient. This group had significantly lower median (range) levels of BALF hβD2 than the vD sufficient group (186.29 (43.82-615.69 ) pg/ml and 390.53 (59.6-939.7) pg/ml respectively ic, ileal, and pulmonary pathology of these mice. However, the mechanism is unknown. Recent data suggests that increased expression and activity of fatty acid ∆5and ∆6-desaturase enzymes plays a role in EFA abnormalities in a CF cultured cell model by increasing conversion of LA to AA. The aim of this study was to determine if the effect of DHA on EFA metabolism is due to suppression of these enzymes.

Methods: Experiments were performed in cultured 16HBE human bronchial epithelial cells stably transfected with expression plasmids containing partial cftr sequences in the sense or antisense orientations, resulting in a CFTR(+) or CFTR(-) phenotype, respectively. To assess EFA metabolic activities, these cells were grown to confluence, incubated in the presence or absence of varying concentrations of DHA for 24 hours, then labeled with 14 C-LA or 14 C-linolenate (LNA). Conversion of these substrates to various products was assessed by HPLC. Gene expression was determined using quantitative RT-PCR.

Results: In the absence of DHA, antisense cells demonstrate increased conversion of LA to AA when compared to sense cells, indicating increased desaturase activity in the n-6 pathway. However, in cells supplemented with 5 µM DHA, there is no difference in AA production between sense and antisense cells. This is due to a reduction in activity in the antisense cells. Similar findings are noted when cells are labeled with LNA, which exhibits increased conversion to EPA in antisense cells by the same desaturase enzymes (n-3 pathway). Supplementation with 5 µM DHA results in partial suppression of EPA production. Complete suppression of metabolic activity to the level of sense cells requires at least 10 µM DHA. Gene expression analysis demonstrates a dose-dependent decrease in both ∆5-and ∆6-desaturase mRNAs in antisense cells with DHA administration.

Conclusions: Supplementation of CFTR(-) (antisense) cells with DHA results in a dose-dependent decrease in the expression and activity of fatty acid desaturase enzymes to the level of sense cells. This suggests that correction of EFA abnormalities, and by extension, CF-related pathology, is mediated through regulation of these enzymes. Background: Cystic fibrosis (CF) airway disease is characterized by exuberant inflammation. Pharmacological modulation of this inflammation without compromising antibacterial defenses is a desired therapeutic goal in CF. Current drugs (systemic corticosteroids, high-dose ibuprofen) are effective clinically, however, their serious adverse effects necessitate a search for novel, safer medications. Production of many inflammatory cytokines, such as IL-8, are regulated transcriptionally (NF-κB, JNK and p38 MAPKs) and post-transcriptionally (p38 MAPK). Identification of molecular mechanisms of current drugs should facilitate development of prospective therapies. Recently, we ruled out NF-κB as a potential molecular target to decrease IL-8 because both systemic corticosteroid dexamethasone (Dex) and high-dose ibuprofen (Ibu) suppress NF-κB transactivation, but only Dex down-regulates IL-8 production {Dauletbaev N et al., Respiration 2010;79:234}. Specific objective of this study was to identify molecular targets responsible for the decrease of stimulated IL-8 production in CF respiratory epithelium. MKP-1 is the major negative regulator of JNK and p38 MAPKs, and a recently recognized molecular target for corticosteroids. We hypothesized that MKP-1 is the principal molecular target of Dex responsible for the decrease in stimulated IL-8.

Methods and Results: CFTE29o-respiratory epithelial cells (∆F508 homozygous) were stimulated with TNF-α or IL-1β (10 ng/ml for 1 hr) ± pre-incubation with Dex or Ibu (both drugs used at clinically relevant concentrations of 0.1-10 nM or 480 mM, respectively). Only Dex decreased IL-8 production in CFTE29o-cells (IC50 ~4 nM). Further, only Dex up-regulated MKP-1 mRNA expression (2.3-fold increase; p<0.01). MKP-1 silencing by siRNA resulted in hyperproduction of IL-8, with a greater magnitude on IL-1β stimulation (>2-fold increase; p<0.001). This IL-8 hyperproduction was due to transcriptional mechanisms and did not involve increased mRNA stability (Actinomycin D pulse-chase). Further, IL-8 hyperproduction in MKP-1 silenced cells was completely reversed by the p38 MAPK inhibitor SB203580 (IC50 ~0.2 µM), whereas the JNK inhibitor SP600125 did not show any effects at equimolar concentrations. The ability of Dex to down-regulate IL-8 was significantly (>25%; p<0.01) diminished by MKP-1 siRNA silencing. Dex decreased IL-8 mRNA stability, and this effect was recapitulated by p38 MAPK inhibition. By contrast, neither MKP-1 siRNA silencing nor treatment with Ibu had any effects on mRNA stability.

Conclusions: MKP-1 is a major molecular target for inhibition of stimulated IL-8 production in CF respiratory epithelium. MKP-1 suppresses IL-8 transcriptionally, while p38 MAPK appears to control IL-8 both transcriptionally and post-transcriptionally. A suppression of IL-8 at both the transcriptional and post-transcriptional level is required to achieve maximum inhibition.

Supported by the BREATHE Initiative of the Canadian Cystic Fibrosis Foundation.

Oxidative stress has been implicated in the pathogenesis and progression of CF pulmonary disease. The increased oxidant stress burden of the CF lung arises from the release of reactive oxygen species (ROS) from activated neutrophils recruited to infections sites. Furthermore, there are decreased anti-oxidant levels due to decreased absorption of fat soluble antioxidant vitamins from the diet, as well as decreased glutathione transport via CFTR. Using the model human epithelial airway cell lines, 16HBE14o-and CFBE41o-, we have confirmed that non-CF epithelial cells are indeed much more susceptible to oxidant stress. To do this, we utilized the stable oxidant stressor tert-butyl hydroperoxide (t-BOOH) to induce oxidant stress. We exposed both 16HBE14o-and CFBE41o-cells to increasing concentrations of t-BOOH and monitored cell viability using an MTT assay. No significant cell death was apparent in the 16HBE14o-cells at t-BOOH concentrations between 50 -500µM, while 1000 µM was toxic to nearly all of the cells. In contrast, the CFBE41o-cells succumbed to t-BOOH at much lower concentrations; for example, application of 200 µM t-BOOH resulted in a cell viability of 98.9 ± 2.1 % for the 16HBE14o-cells versus 4.2 ± 1.3 % for the CFBE41o-cells, suggesting that CF cells are inherently less able to deal with exposure to oxidant stress. As a result of these findings, we hypothesised that the heme-oxygenase (HO) system may be dysfunctional in the CF cells, and this may play some role in the differential effects of oxidant stress on CF and normal cells. HO is a member of the heat shock family, of which HO-1 is induced by a number of stresses related to oxidative stress and inflammation. HO-1 is reportedly upregulated in CF patients, likely as a cytoprotective mechanism. HO-2 is expressed constitutively in a number of cell types, though little is known about its role in the airways. Initially, we confirmed via RT-PCR, that both 16HBE14o-and CFBE41o-cells express HO-1 and HO-2. We additionally investigated expression of mPGES-1 (microsomal prostaglandin E synthase-1) in both cell types, since previous reports indicate that HO-1 induction affects mPGES-1 expression and consequent PGE2 protection. Again, via RT-PCR, we determined mPGES-1 mRNA expression in both cell lines. Using quantitative PCR, we next investigated the effects of short term oxidant stress on HO-1, HO-2 and mPGES-1 expression. Exposure of cells to 600 µM H 2 O 2 for 2 hours resulted in significant decreases in HO-1, HO-2 and mPGES-1 levels in the 16HBE14ocells, with no significant difference in expression in the CFBE41o-cells. This concentration of H 2 O 2 caused no cytotoxicity. These findings are in contrast to our predicted results, since oxidant stress generally induces HO-1 expression. Further experiments are underway to delineate the role of HO and oxidant stress in airway epithelial cells.

Supported by the Canadian Cystic Fibrosis Foundation. expression values. Normalized expression was compared between CF patients and healthy controls using multiple linear regression adjusting for age, gender, and leukocyte count. Results: We measured gene expression in 72 CF patients and 45 healthy controls. We found higher expression of SOCS1 and SHIP1 in patients with CF relative to healthy controls (p=0.02 and p=0.04 respectively). In contrast, levels of TOLLIP were lower (p=0.008) in patients with CF. Finally, levels of TLR4 were greater in CF patients (p=0.04).

Conclusions: A specific subset of negative regulators of TLR-mediated signaling (SOCS1 and SHIP1) is expressed at higher levels in leukocytes of CF patients relative to healthy controls and TLR expression is preserved. SOCS1 and SHIP1 could play a role in the down-regulation of TLR-mediated innate immune responses in CF patients. The objective of this study was to establish the role of NETs in the response by peripheral blood neutrophils from healthy donors or patients with CF to Pseudomonas aeruginosa.

Methods: Freshly isolated human neutrophils from healthy volunteers or patients with CF were tested for their ability to form NETs in vitro in response to P. aeruginosa strain PAO1, and fluorescence microscopy used to investigate physical association between P. aeruginosa and NETs. An in vitro killing assay, developed to maximize NETs killing and reflect the constant motion of the lung, was utilized to investigate killing of P. aeruginosa PAO1 by normal and CF neutrophils. In addition, the ability of non-mucoid and mucoid strains of P. aeruginosa, isolated from the airways of a patient with CF, to be killed by NETs was investigated.

Results: P. aeruginosa stimulates NETs formation by neutrophils from healthy volunteers in a dose-dependent fashion and physically associates with NETs. In addition, NETs were the predominant mechanism of killing under conditions that reduce close association between the bacteria and neutrophils. Under these circumstances, NETs effectively killed P. aeruginosa across a wide range of bacterial concentrations. To investigate the relevance of NETs to P. aeruginosa infection in the CF airway, peripheral blood neutrophils isolated from patients with CF were tested. Although a variety of processes function abnormally in CF neutrophils, their ability to form NETs in response to P. aeruginosa was equal to that of neutrophils from healthy controls. Finally, isogenic clinical isolates of P. aeruginosa obtained from a patient early and late in the course of CF were killed by NETs under conditions of constant motion, though some feature of P. aeruginosa adaptation in the CF airway, perhaps the development of mucoidy, resulted in partial protection of planktonic P. aeruginosa against this mechanism of killing.

Conclusions: Together, these results demonstrate that normal and CF neutrophils have equal capacity to kill P. aeruginosa via NET formation, but that strains may evolve in the CF airway, possibly via the development of mucoidy, to resist NETs-mediated killing. Support: Cystic Fibrosis Foundation, National Jewish Health Pulmonary Division Pilot Grant Program, Rebecca Runyon Bryan Chair for CF, Max and Yetta Karasik Foundation, NIH 1R01HL090991. Airway surface dehydration in βENaC-overexpressing transgenic mice (Scnn1b-Tg) impairs mucus clearance, resulting in neonatal mortality and development of complex lung pathology in surviving mice. Spontaneous neonatal bacterial colonization previously demonstrated in Scnn1b-Tg mice suggests that pathogen associated molecular patterns accumulating in static mucus may trigger an innate inflammatory response by signaling through pattern recognition receptors, such as the Toll-like receptors (TLRs). We have also previously noted that the combination of poor mucus clearance and absence of the TLR downstream signaling in Scnn1b-Tg mice lacking the TLR adapter molecule MyD88, enhances bacterial colonization in neonatal mice and decreases bacterial clearance in older mice, as compared to MyD88-sufficient Scnn1b-Tg mice. Among the TLRs upstream of MyD88, TLR2 and TLR4 recognize bacterial wall components and are known to function in respiratory epithelial and inflammatory cells, respectively. To test the hypothesis that TLR2 and TLR4 signaling contribute to host defense and the development and progression of lung pathology due to airway surface dehydration, we bred Scnn1b-Tg mice (hemizygous, inbred line C57BL/6N-6608, average survival 75%) with mice deficient in TLR2 or TLR4 or both in combination. Although redundancy in TLR signaling is thought to achieve effective airway immune surveillance in the absence of a single TLR, we found that absence of either TLR2 or TLR4 in Scnn1b-Tg mice caused a severe drop in neonatal survival in comparison to TLR-sufficient Scnn1b-Tg mice (TLR2 -/-, Scnn1b-Tg 58 ± 7%, n=43 vs. TLR2 +/-, Scnn1b-Tg 81 ± 7%, n=33; TLR4 -/-, Scnn1b-Tg 46 ± 14%, n=13 vs. TLR4 +/-, Scnn1b-Tg 65 ± 8%, n=32). Moreover, preliminary data suggest that the effect of combined TLR2 and TLR4 deletion on survival is additive. Absence of TLR2 or TLR4 did not eliminate neutrophil and eosinophil infiltration in surviving mice, phenotyped at 5 weeks of age, and did not significantly modify the lung pathology characteristic of Scnn1b-Tg mice, namely mucus plugs, mucous secretory cell hyperplasia, airway inflammation and emphysema. Assessment of, bacterial burden, sepsis and lung pathology in neonatal mice from all TLR -/-Scnn1b-Tg crosses are in progress. These results highlight the importance of innate defense mechanisms in the neonatal period and suggest that while TLR2 and TLR4 per se are not essential for developing or sustaining chronic lung pathology in the Scnn1b-Tg mouse model, they are non-redundantly critical for host defense that enables neonatal survival in the context of defective mucus clearance. Background: Patients with cystic fibrosis who have identical CFTR genotypes can have marked variation in pulmonary phenotype. We have previously demonstrated high inter-individual variation in Toll-like receptor (TLR)-mediated innate immune responses in humans and have found that these responses are down-regulated in patients with cystic fibrosis (CF) relative to healthy controls (ref). The mechanism underlying this down-regulation of TLR-mediated responses is not known.

Hypothesis: Intracellular suppressors of TLR-mediated innate immune inflammatory responses are differentially up-regulated in the peripheral leukocytes of CF patients.

Methods: We isolated total RNA from peripheral leukocytes obtained from CF patients who had not received intravenous antibiotics for an exacerbation for at least 3 weeks and who were not taking immunosuppressive medications. RNA was also purified from leukocytes from healthy controls frequency-matched for age and gender. We performed quantitative real-time PCR to measure the level of mRNA expression for SOCS1, TOLLIP, IRAKM, SARM1, SHIP1(INPP5D), AKT1, as well as TLR4 and TLR7. Levels of beta-actin and GAPDH mRNA were used to normalize the gene It is well known that pulmonary acquisition of P. aeruginosa (Ps. a.) by individuals with cystic fibrosis (CF) portends declining lung function. Ps. a. lung infection is characterized by destructive neutrophilic inflammation thought to result in part from airway epithelial cells responding to Ps. a. by secreting pro-inflammatory chemotactic factors. We have previously shown that acute stimulation of well-differentiated human bronchial epithelial (W-D hBE) cells with sterile Ps. a. culture filtrates strongly stimulates CXCL1, CXCL5, CXCL8 and CCL20 production. After chronic Ps. a. stimulation, secretion of these cytokines was somewhat reduced, but was sustained at significantly higher levels than the original baseline. To uncover the molecular mechanisms underlying this new pro-inflammatory state, we treated W-D hBE cultures with control, bacterial growth medium or with Ps. a. conditioned medium for three consecutive days, referred to as naïve and Ps. a.-adapted cells, respectively. We obtained mRNA expression profiles using microarrays for both naïve and adapted cells at baseline and four hours after control or Ps. a. challenge using cells from six individuals at each time point. We found higher steady state mRNA levels for the above cytokines in Ps. a.-adapted cells, which was validated by qRT-PCR. Higher cytokine protein production in Ps. a.-adapted cells correlated with elevated mRNA levels, suggesting transcriptional regulation. To understand altered cytokine mRNA transcription in W-D hBE cells, the full-length CXCL8 promoter driving luciferase expression was cloned into a lentiviral vector, which was used to infect primary hBE and growth extended UNCN3T cells. After infection, the cells were differentiated for 21-28 days at an air-liquid interface and assayed for luciferase activity in response to Ps. a. or control challenge. We observed enhanced basal CXCL8 promoter activity in Ps. a.adapted cells without further stimulation by subsequent rounds of Ps. a. challenge. Via mutation of the CXCL8 promoter, we determined that specific proximal NF-κB and C/EBP transcription factor binding sites regulated the enhanced baseline expression after adaptation to Ps. a. Each site had a specific role, i.e. the NF-κB promoter binding site was essential for both inducible and enhanced basal activity of the CXCL8 promoter in naïve and Ps. a.-adapted cells. However, Ps. a. did not stimulate a minimal NF-κB binding site reporter construct in adapted cells, suggesting that the NF-κB alone was not sufficient to drive the enhanced baseline transcription in Ps. a.-adapted cells. Mutation of the C/EBP element adjacent to the NF-κB site in the CXCL8 promoter similarly decreased the response of both naïve and Ps. a.-adapted cells to Ps. a. stimulation. However, inhibition of C/EBPβ with shRNA increased basal CXCL8 promoter activity, suggesting that chronic exposure to Ps. a. releases C/EBPβ repression. These results suggest that pharmacologic manipulation of the pathways resulting in C/EBPβ derepression may be useful to reduce sustained neutrophilic inflammation in the lungs of CF patients. Introduction/Rationale: Cystic fibrosis (CF) is marked by a robust inflammatory response. Macrophages are key regulators of innate immunity as they are responsible for bacterial phagocytosis and recruitment of neutrophils. Previous studies demonstrate that changes in membrane potential accompany phagocytosis in macrophages and that chloride (Cl -) channels may be important in this process. We hypothesize that Clflux is present in macrophages and that multiple Clpathways contribute to this flux.

Methods: Clefflux was measured in bone marrow derived macrophages cultured from cftr -/-(Cftr tm1UNC ) and ∆F508/∆F508 (Cftr tm1Kt ) mice, using N-[ethoxycarbonylmethyl]-6-methoxy-quinolinium bromide (MQAE), a Cl --specific fluorescent dye. Clefflux was quantified as the initial change in MQAE fluorescence over time (slope) as the cells transition from Cl --containing to Cl --free solutions. The effects of NPPB (100µM) or CFTR inh -172 (20µM) on Clefflux were assayed to confirm the presence of Clefflux and to assess the contribution of CFTR to total flux. The effects of extracellular calcium (Ca 2+ ) on Clefflux were measured by examining for changes in flux as cells transitioned from Clto Cl --free solutions with low=0.1 mM, normal/control=0.5mM and high=2mM extracellular Ca 2+ concentrations. In addition, the effects of increased intracellular Ca 2+ concentrations on Clefflux were assessed by examining efflux after carbachol (100µM) treatment.

Results: Clefflux was present in all macrophages studied. WT macrophages exhibit greater Clefflux than ∆F508 or cftr -/macrophages (p<0.001, n=83-137). There is a significant reduction in Clefflux with addition of NPPB (p<0.001, n= 79-140) in all macrophages. In the presence of CFTR inh -172, Clefflux in WT macrophages is significantly reduced (p=<0.001, n=27-91), but no effect is observed on CF affected macrophages. With decreases in extracellular Ca 2+ concentration, there is a significant reduction in Clefflux observed in WT and cftr -/macrophages (p<0.001, n=38-111). With increases in extracellular Ca 2+ concentration, ∆F508 and cftr -/macrophages show a significant increase in Clefflux (p<0.001, n= 94-136) that is not observed in WT macrophages. A significant increase in Clefflux was also observed in ∆F508 and cftr -/macrophages (p<0.001, n=71-73) with transition to Clfree solution after carbachol therapy that was not observed in WT macrophages. These findings suggest that there is an upregulation in a Clefflux pathway that is sensitive to increased intracellular Ca 2+ concentrations in CF affected cells when compared with WT cells.

Conclusions: Taken together, these results suggest that CFTR is functional in WT macrophages and may contribute to Clefflux. Additionally, Ca 2+ modulated Clpathways likely contribute to Clefflux in WT, ∆F508 and cftr -/macrophages. However, the Clpathways utilized by CF affected macrophages are different than those utilized by WT macrophages. Presently we are examining if similar pathways are present in human macrophages.

Funded by the NIH and CFF. The clinical course of cystic fibrosis (CF) varies considerably among Caucasians carrying the same CFTR mutation suggesting that additional genetic modifiers may contribute to this variability. The highly conserved ancestral haplotype 8.1AH contains a cassette of linked genes and is carried by up to one third of Caucasians, comprising linked gene polymorphisms that play a key role in the inflammatory response: LTA +252A/G (Lymphotoxin A), TNF -308G/A (Tumor necrosis factor), HSPA1L +1267A/G (Heat shock protein A1 Like) and RAGE -429T/C (Receptor for Avanced Glycation Endproducts). Moreover, 8.1AH has recently been associated with delayed onset of lung bacterial colonization in CF patients in some cohorts. As airway inflammation is a key component inducing CF lung damage, we investigated whether the 8.1AH represents such modifier in CF patients.

To test whether the 8.1AH modified lung disease CF severity, we analyzed a cohort of 436 European CF patients from France (n=230), UK (n=111) and Germany (n=95). Lung function was evaluated by forced expiratory volume in one second (FEV 1 ) measurements. Age adjusted FEV 1 dif- Chronic pancreatitis (CP) is a complex disorder with multiple genetic and environmental risks. As a CF associated disease, a significant fraction of CP patients are either heterozygous or compound carriers of CF causing mutations but no one has yet characterized the specific mutations that cause idiopathic chronic pancreatitis.

Our mathematical model of the pancreatic duct physiology predicted that CFTR variants that inhibit bicarbonate conductance will impair pancreatic juice secretion and predispose to recurrent acute pancreatitis through a trypsin-dependent mechanism. Mutations in the trypsin inhibitor (Serine protease inhibitor Kazal type 1, SPINK1) are predicted to further increase the risk of CP in patients with susceptibility for premature trypsin activation. We predicted that CFTR mutations of class I and II are likely to be associated with CP as well as CF, while there may be mutations that cause CP and not necessarily CF due to specific disruption of CFTR bicarbonate conductance without significant change in chloride conductance.

To test if common CFTR mutations in CP that do not cause classic CF have defective bicarbonate secretion, CP patients and controls were screened for CFTR variants in all 27 exons. We also screened for SPINK1 variants to identify complex risk patients. The final study group included idiopathic CP (n=53), probands of familial idiopathic pancreatitis (n=27) and unrelated controls (n=150). SPINK1 variants were identified in 3% of controls and 36% of subjects (OR 16.5, CI 6.1-44.9).

The class II CF mutation, F508del was identified in 8.75% of subjects and 1.3% of controls (OR=7.1, CI 1.4-35.0). Co-inheritance of F508del plus a SPINK1 mutation was identified in 5% of patients and no controls (OR=131.5, CI 32. 2-535.6 ). An atypical CFTR variant was found in 16% of subjects and 7% of controls (OR 2.5, CI 1.0-5.8). Co-inheritance of both the atypical CFTR plus SPINK1 variant was identified in 8.75% of CP patients (OR 40.0, CI 16. 6-95.4) and no controls.

The atypical CFTR variant was cloned into wild-type (wt) CFTR and expressed in HEK293 cells to measure the relative conductances of CFTR wt and variant proteins to HCO 3 and Cl -. Patch-clamp recordings of atypical variant CFTR demonstrated normal chloride currents but significantly reduced bicarbonate current (p=0.0001).

This study validates the prediction of our pancreatic duct mathematical model of bicarbonate secretion and development of pancreatitis. While the isolated risk is 2 to 3-fold, the multiplicative risk in patients with SPINK1 mutations confirms a trypsin-dependent mechanism. Future studies are warranted to determine if other CFTR-variants that are not associated with CF will lead to an increased risk for the development of CP through impairment of bicarbonate flow. This work was supported by NIH grants DK061451 (DCW) and DK54709 (DCW), and postdoctoral scholarships from the University of Heidelberg at Mannheim (AS) and the NIH (JL). ferences were calculated in each country, and a random effect meta-analysis was used to pool the results.

Distribution of the 8.1AH carriers was similar: 22% in the French; 28% in the UK and 29% in the German CF patients. We found 8.1AH was significantly associated with lower FEV 1 in all 3 cohorts (meta-analysis: p=0.03). These findings support the concept that the 8.1AH appears to be an important genetic modifier of lung disease in CF. We suggest that examination of multiple linked genes provides an alternative avenue of research into genetic variability in CF outcome. The level of CFTR function needed to ameliorate the CF phenotype has been a matter of debate. Thresholds have been estimated in the context of an individual mutation or small groups of mutations and have defined function using the amount of correctly spliced CFTR mRNA, mature CFTR protein quantification or ion conductance in limited in vitro cell lines, or using in vivo current analysis. Thresholds of 5%, 10% and 25% of wild-type (WT) CFTR function have been proposed. The Clinical and Functional Translation of CFTR (CFTR2) project has undertaken a systematic analysis of CFTR function using a standardized in vitro assay paired with extensive clinical information on CF patients carrying specific CFTR mutations.

The CFTR2 project has assembled cross sectional clinical and genotype information for 39,614 CF patients collected through national registries and large clinics' databases. Mutations created using site directed mutagenesis were stably incorporated into Fischer Rat Thyroid (FRT) cells using the Invitrogen Flp-In system. In a standardized fashion, short circuit chloride current (ISC) was measured in these cell lines following forskolin (10µM) administration. All currents are listed as a percentage of the ISC for WT CFTR expressed in FRT cells. CFTR mRNA levels for all mutations were within two fold of WT.

A total of 161 mutations were found to have an allele frequency > 0.01%. Of these, 57 were missense mutations which were evaluated functionally; 33 of the 57 missense mutations had an ISC < 5% of WT. Patients (2904) carrying at least one of these mutations consistently had a mean sweat chloride measurement > 60 mEq/L (33/33 mutations) and frequently had an incidence of pancreatic insufficiency (PI) > 50% (25/33). The eight mutations (254 patients carrying these mutations) that had an ISC between 5-10% WT function were associated with sweat chloride > 60 mEq/L in 4/8, but none (0/8) had a rate of PI > 50%. Three additional mutations (44 patients) had ISC between 10-25% WT function. Two of the three were associated with a sweat chloride > 60 mEq/L and 1/3 with a rate of PI > 50%. Finally, 13 mutations had an ISC >25% WT; 8 of 13 mutations were associated with sweat chloride < 60 mEq/L and PI < 50%. The remaining five mutations were associated with mean sweat > 60 mEq/L, but are suspicious for incomplete or incorrect genotyping. That group includes a mutation previously identified as a neutral variant (p.Met470Val), mutations that have primarily been associated with an obstructive azoospermia phenotype (p.Asp1152His, p.Asp1270Asn), or mutations that are part of a complex allele (p.Ile148Thr, p.Ile1027Thr).

Mutations associated with ISC < 5% WT function are deleterious and have a high penetrance for CF with PI. Testing of additional mutations is being performed to refine the relationship between function and disease liability for mutations that generate ISC between 5-25% WT function. CFTR mutations purported to cause CF that generate ISC > 25% WT function may be complexed with deleterious alleles elsewhere in the same CFTR gene.

Supported by the US CFF. Introduction: Using data from the North American Cystic Fibrosis Gene Modifier Consortium, we conducted a genome-wide association study (GWAS) for Meconium Ileus (MI), using 556,445 SNPs typed on 3,655 CF patients, 615 of whom had MI. We did not observe any genome-wide significant results, based on a p-value criterion of 5x10 -8 or a q-value (FDR adjusted p-value) of 0.05. It is common for a GWAS to result in suggestive but not overwhelming evidence for association. Recent methodological work suggests that such single-SNP analysis approaches are a preliminary step in the gene identification process, where methods that help to prioritize observed results should then be used. Pathway analysis is a popular method to prioritize single-SNP results (referred to as GWAS-PA), however an interactive pathway is but one hypothesis for disease mechanism and can be restrictive, since contributing genes/proteins can relate to residual or compensating activities in other ways that could span multiple processes. We apply a more general approach to our MI GWAS results, that we refer to as the Hypothesis-driven GWAS (GWAS-HD), highlighting that any hypothesis can be tested within this framework. Given that loss of CFTR leads to CF with perturbed epithelial transport, we considered that coincident phenotypes including MI reflect residual or adapted transport capability. The polarization of epithelial tissue involves distinct cell localization for many transport-relevant proteins and we hypothesized that constituents of the apical plasma membrane, where CFTR normally resides, may be contributors.

Methods: To test this hypothesis in our MI GWAS data we applied our GWAS-HD methodology by: (1) prioritizing the genome by generating a list of proteins present in the apical plasma membrane (based on the GO consortium with AmiGO version 1.7); (2) using the stratified FDR (SFDR) to weight our GWAS p-values and determine genome-wide significance for given loci; and (3) performing simulation studies to determine the statistical significance of our hypothesis and provide informative graphical displays.

Results: Our list consisted of 151 genes spanning 3,734 GWAS SNPs, although eight were not tagged by any of the ~556K GWAS SNPs. Using the SFDR, we identified SNPs from four genes within the apical plasma membrane list that reached genome-wide significance (q-value < 0.05), with q-values as low as 6x10 -4 . Permuting the case-control status and re-conducting the association analyses of the 3,734 SNPs, provided for assessments of the significance of the whole AmiGO list, itself, while retaining the LD structure of the SNPs within the genes (p < 0.05).

Conclusion: The GWAS-HD provided significant evidence that multiple genes present at the apical plasma membrane may contribute to the MI phenotype, and we were able to prioritize four genes for further study. Cystic fibrosis (CF) is due to mutations in the chloride channel CFTR gene causing impairment of chloride secretion in the apical membrane of epithelial cells. The most common CFTR mutation in CF is the deletion of the phenylalanine at the position 508 of the protein (∆F508), leading to the retention of CFTR in the ER and its rapid degradation via ERAD pathway. More than 1700 mutations have been identified so far, variably affecting CFTR activity but none of them have been clearly linked to a certain pathophysiology. In other words, even if carrying the same genotype, CF patients might face a different development in their disease. Clearly, CF disease also depends on other genetic and/or environmental factors. Lately, a number of studies focus on modifier genes that may determine severity of lung disease, as e.g. Mannose Binding Lectin and Transforming Growth Factor beta. Recently, Wright et al. performed whole-genome microarrays assays of nasal cells from non-CF individuals, mild CF and homozygous ∆F508-CFTR patients and found a total of 652 out of 1187 genes differentially expressed in these three groups (1) .

MicroRNAs (miRNAs) are single-stranded RNA molecules of about 21-23 nucleotides in length which regulate gene expression. First described in 1993, miRNAs are non-coding RNA; instead each primary transcript (a pri-miRNA) is processed into a short stem-loop structure called a pre-miRNA and finally into a functional miRNA. Mature miRNA molecules are partially complementary to one or more messenger RNA (mRNA) molecules, and thus down-regulate gene expression by inducing mRNA degradation or protein expression by translation inhibition. Differential expression in miRNAs has been shown to influence disease development in Alzheimer's, cancer, heart failure and COPD. No systematic analysis of miRNA expression in CF has been reported to date, although one study showed a down-regulation of miR-126 in CF airway epithelial cells (2).

We performed microarray analysis of miRNAs expression using primary bronchial epithelial cells from three CF donors (∆F/∆F) and from three non-CF donors. Among the 856 miRNAs identified in the human genome at the time, 94 were expressed in HBE cells with 16 of them differentially expressed between CF and non-CF patients. The following table summarizes the microarrays results.

Real-time PCR experiments partially confirmed microarray results, finding significant differences in expression of 11 of the 16 miRNAs. Further target analysis of the differentially expressed miRNAs is expected to identify new therapeutic targets in CF.

Supported by the Cystic Fibrosis Foundation. The cystic fibrosis (CF) airway is characterized by a proinflammatory phenotype, characterized by massive overexpression of IL-8 and other proinflammatory chemokines. However, it is poorly understood how mutations in CFTR actually result in the disease-related proinflammatory expression program in CF airway. A growing body of evidence supports a role for post-transcriptional regulation and RNA processing in different pathologies which include chemokine production.

Hypothesis: Our preliminary MS-based studies on the nuclear extracts from CF lung IB3-1 epithelial cell line, compared with daughter IB3-1/S9 cells which have been repaired with [wildtype]CFTR, have revealed the differential binding of ribonucleoproteins hnRNP[A2B1] and hnRNP[A1] to IL-8 promoter. We have therefore hypothesized that CF-causing mutations in CFTR may be associated with large scale changes in RNA control, including alternative splicing and differential expression of various gene transcripts.

Methods: To test this hypothesis we have utilized the Illumina platform to analyze gene transcript profiles in CF lung IB3-1 epithelial cells versus IB3-1/S9 cells with [wildtype]CFTR. For data analysis and interpretation, CF carrier screening, the majority of patients diagnosed through MN NBS have at least one F508del mutation (85.1%), showing the mutation remains the most frequent. According to CFF Registry data, about 2% of individuals meeting criteria for having CF have atypical CF. Since NBS, the percentage of patients followed in our institution with 2 mutations and negative/intermediate sweat tests is significantly greater. Most patients are compound heterozygotes for a severe and mild allele, or two mild alleles; many meet criteria for CRMS. This has implications for treatment and management of patients and psychosocial implications for families coping with the uncertainty regarding if or when their child may develop symptoms. Introduction: New treatments are needed to improve the health of people with cystic fibrosis (CF). Controlling lung-damaging inflammation is likely to be beneficial but specific anti-inflammatory targets have not been identified. Recently we generated in vitro evidence that TLR5 is a major mediator of P. aeruginosa-induced inflammation, identifying TLR5 as a biologically-plausible candidate gene that may modify CF lung disease.

Objective: To examine if the TLR5 c.1174C>T polymorphism (rs5744168) modifies the clinical outcome in patients with CF.

Methods: By combining cellular immunology approaches with a large population-based genetic modifier study, we examined TLR5 as an antiinflammatory target and modifier gene in CF. Using two pairs of human CF and control airway epithelial cells, we demonstrated that the TLR5-flagellin interaction is a major mediator of inflammation following exposure to P. aeruginosa. To validate TLR5 as an anti-inflammatory target, we analyzed the disease modifying effects of the TLR5 c.1174C>T SNP (rs5744168) in a large representative cohort of CF patients (n=2219). rs5744168 encodes a premature stop codon and the T allele is associated with a 45.5-76.3% reduction in flagellin responsiveness (P<0.0001). To test the hypothesis that reduced TLR5 responsiveness would be associated with improved health in CF patients, we examined the relationship between rs5744168 and two clinical phenotypes: lung function and body weight.

Results: Adults with CF carrying the TLR5 premature stop codon (CT or TT genotype) had higher body-mass index than CF patients homozygous for the fully functional allele (CC genotype) (P=0.044); however, similar improvements in lung function associated with the T allele were not statistically significant.

Conclusion: Although follow up studies are needed to confirm the impact of TLR5 on nutritional status, this translational research provides preclinical evidence that strategies to inhibit TLR5 may improve the health of people with CF.

we employed a Systems Biology approach using GenePattern (the Broad Institute at MIT and Harvard), Genomatix and Ingenuity Pathways Analysis.

Results: Using the Comparative Marker Selection from the GenePattern suite, we find that the CF IB3-1 cells are marked by downregulation of RNA-regulating genes such as the regulator of nonsense transcripts UPF1, multiple splicing factors (SF2A2/3, U2AF1), and ribonucleoproteins (SNRP/hnRNPs), which are involved in the spliceosome functioning. Unlike repaired cells with higher ribonuclease H (RNASEH2) levels, the CF cells possess relative higher expression of the RNASEL and RNASET2, which have been implicated in carcinogenesis. As a result of the aberrant expression profile of multiple RNA controlling genes, the CF cells demonstrate distinct expression profiles for the gene transcripts which correspond to different isoforms, thereby implying alternative RNA splicing. Among various genes that appeared to be alternatively spliced in CF cells, we identified the RNA-binding MSI2 which further suggested alterations in posttranscriptional RNA control. In addition, differentially expressed transcripts for multiple genes involved in cell death/proliferation (PDCD2, JADE1, BAIAP2, CDCA7, and CDK10) indicated dysregulation of the cell survival associated transcriptional control in CF-inflammation.

Significance and Impact: The status of CFTR in the CF lung epithelial cell appears to have a profound impact on processes associated with RNA processing and splicing. It is plausible to suggest that these alterations impact the balance between pro-and anti-inflammatory programs resulting in the excessive production of IL-8 and other proinflammatory cytokines in dysfunctional CF airway. However, whether these changes are causal or compensating is yet to be discerned. Background: Some studies report a decrease in incidence of CF while others observe a consistent incidence. The Minnesota (MN) Newborn Screening (NBS) Program, initiated in March 2006, measures immunoreactive trypsinogen (IRT) followed by DNA analysis for infants with IRT values above the 96th percentile, or >170 ng/ml. According to CF Foundation guidelines, infants with two CFTR disease-causing mutations and normal sweat tests have CF or CFTR-related metabolic syndrome (CRMS), and evaluation is recommended. With the implementation of NBS, and CF carrier testing recommendations, milder, atypical cases may be identified earlier.

Aims : Jerusalem, Israel; 2. Pediatric Gastroenterology Unit, Hadassah University Hospital, Jerusalem, Israel; 3. CF Center, Carmel Medical Center, Haifa, Israel; 4. CF Center, Rambam Medical Center, Haifa, Israel; 5. CF Center, Shaare Zedek Medical Center, Jerusalem, Israel; 6. CF Center, Soroka Medical Center, Beer Sheva, Israel; 7. Gastroenterology Unit, Hadassah University Hospital, Jerusalem, Israel Background: Like the nasal potential difference (NPD) test, intestinal current measurements (ICM) may be useful especially in the diagnosis of non-classic cases of cystic fibrosis (CF). However, ICM is easily applicable at all ages. The aim of this study is to evaluate a new diagnostic scoring system of ICM in CF patients and controls and its application to a large cohort of patients whose diagnosis is questionable.

Methods: Rectal biopsies were taken from 3 groups: known CF patients, healthy control patients, and patients who are suspected for CF. The last group (mean age: 4.2±5.3 years) had a variety of symptoms suggestive of CF including recurrent pneumonia, unexplained bronchiectasis, chronic diarrhea and failure to thrive. ICMs were performed using standard protocols by mounting the rectal biopsy in an Ussing chamber and sequentially adding secretagogues while recording current changes.

Results: Ten known CF patients (aged 2±1.9 years) and 16 control patients (aged 15.6±14.4 years) were examined and have remarkably different results (all results are presented as µA/cm 2 ): carbachol 16±7, histamine 13±9, and forskolin 4.8±4 for healthy control group; and carbachol −3.7±6.8 (p<0.0001), histamine −3.1±2.7 (p<0.0001), and forskolin 0.2±0.4 (p=0.0004) for the CF group. The suggested reference values are: +3.75, +0.25, +1.32 for carbachol, histamine and forskolin, respectively. The combination score (the sum of carbachol, histamine and forskolin) of +4.09 of all 3 secretagogues differentiates normal from abnormal (ROC Curve analysis, area under the curve =1.00, both sensitivity and specificity are 100%). This statistical model was applied to 70 patients suspected for CF and revealed that 59 patients had normal ICM results (combination>4.09) and 11 patients had abnormal ICM results (combination<4.09).

Conclusion: In this study we have shown that ICM tests may be useful to differentiate between CF and non-CF patients and may be included in diagnostic algorithms. Larger studies are needed to confirm these results. This score may be used in CF treatment protocols. Fetal bowel anomalies may reveal cystic fibrosis (CF) and the search for CFTR gene mutations is part of diagnostic investigations in such at-risk pregnancies, according to European recommendations. We collated data on 694 referrals related to suspicion of CF in fetuses from 1992 to 2009 to better document CFTR genotypes and correlations with ultrasound patterns.

Methods: CFTR gene analysis was performed in a multistep process, including a search for frequent mutations in the parents and subsequent indepth search for rare mutations, mostly when a mutation was present in one parent and/or the fetus. Ultrasound patterns were correlated with the genotypes. Cases were discriminated depending on whether they were referred directly to our laboratory or secondly after a first genetic study in another laboratory.

Results: A total of 30 CF fetuses and 8 cases compatible with CFTRrelated disorders were identified. CFTR rearrangements accounted for 15.0% of CF alleles. Of fetuses carrying a frequent mutation, 21.2% had a second rare mutation, indicative of CF. The frequency of CF among fetuses with no frequent mutation was 0.43%. Correlation with ultrasound pattern revealed a significant association of bowel anomalies in CF fetuses.

The results emphasize the need to search for rearrangements in the diagnosis strategy of fetal bowel anomalies. The diagnostic value of ultrasound patterns combining hyperechogenic bowel, loop dilatation and/or non-visualized gallbladder incite to revise the current strategy and offer an extensive CFTR gene study in case of associated anomalies, even when no frequent mutation is found in the first-step analysis. Objectives: We aimed to clinically validate an operating protocol for early and non-invasive prenatal diagnosis (NI-PND) of cystic fibrosis (CF) and Spinal Muscular Atrophy (SMA) avoiding chorionic-villous sampling (CVS) which carries a 2% risk of miscarriage.

Methods: We first developed a new approach to rapidly target circulating epithelial cells after their enrichment by ISET (Isolation by Size of Epithelial Tumor/Trophoblastic cells) and then performed a phase III clinical validation study of a protocol for NI-PND of CF and SMA. Circulating epithelial cells enriched from the blood of 63 pregnant women at 25% risk of having a child affected by CF or SMA were laser microdissected and tested individually by using short tandem repeat (STR) genotyping to identify circulating fetal trophoblastic cells (CFTCs). Individual CFTCs were subsequently blindly and independently used, in number of 5 or 10 per mother, to perform NI-PND of CF or SMA. In parallel, we tested weekly (4th to 12th week of gestation (WG)), the blood of 14 women after in vitro fertilization (IVF), and counted the number of CFTCs per ml, in order to assess the earlier appearance and kinetics of CFTCs circulation in blood.

Results: A total number of 960 CFTCs were identified in the 77 pregnant women. NI-PND was independently performed on 475 individual CFTCs isolated from the 63 pregnant women. Results were demonstrated to be identical to those obtained by the invasive CVS approach. Sensitivity, as computed on the 105 CFTCs with a mutated genome, was 100% (CI95%: 96.5-100%). Specificity, as computed on the 370 CFTCs without a mutated genome, was also 100% (CI95% 99.0-100%). Finally, we identified 485 CFTCs in 14 women after IVF. CFTCs were found in all of them from the 5th WG through the 12th WG ranging from 1 to 2 CFTCs per ml.

Conclusions: These results show that a reliable NI-PND of genetic diseaeses is feasible and should open the way to implementation of an early and reliable protocol for NI-PND of CF and SMA.

To determine this we collected exhaled breathe condensate in 16 CF subjects, 9 wild type (WT, AA at amino acid 663) and 7 non-wild type (NWT, AT or TT at amino acid 663) to compare exhaled Na + concentrations (age=24±9 vs. 20±5 yrs., ht.=166±9 vs. 169±5 cm., wt=67±14 vs. 56±8 kg., BMI=24±3 vs. 20±2 kg/m 2 , BSA=1.73±0.21 vs. 1.64±0.13, VO 2peak =58±13% vs. 58±26% predicted, FEV 1 =73±28% vs. 71±28% predicted, mean±SD, for WT and NWT respectively, p<0.05 for BMI) at baseline and 30, 60 and 90 minutes following the administration of a nebulized short-acting β-agonist (albuterol sulfate, 2.5 mg diluted in 3 ml normal saline).

Results: Although not statistically significant, we found that the WT group had a higher baseline level of exhaled Na + and a reduction in exhaled Na + in response to a β-agonist compared to the NWT group which had no change in exhaled Na + (baseline=2.4±1.2 vs. 2.0±0.7 mmol/l; 30 min post=2.1±1.0 vs. 2.1±0.8 mmol/l; 60 min post=1.7±0.5 vs. 2.0±1.1 mmol/l; 90 min post=1.9±0.8 vs. 1.9±1.1 mmol/l, mean±SD, for WT and NWT, respectively).

Conclusion: These findings demonstrate that genetic variation of the alpha subunit of ENaC influences β 2 mediated Na + clearance from the airways. These results suggest that CF subjects with the WT variant of ENaC would likely benefit more from ENaC inhibition therapy to attenuate the enhanced Na + influx from β 2 stimulation. In CF, the delta F508 mutation is the most common in Caucasian population but is less common in South Asians (1) . A high incidence of novel mutations has been reported in South Asians (1) . We present two patients with V456A that have significant disease. An 11-yr old female of Pakistani descent with consanguineous parents was admitted to the hospital for severe obstructive lung disease, respiratory distress and hypoxia. FEV1 was 32% predicted and FEF25-75 was 33% predicted. Initial sweat test was negative. Immunological evaluation, including HIV, was normal. Sputum cultures were positive for Haemophilus influenzae and Pseudomonas aeruginosa. A repeat sweat test by quantitative ionotrophic electrophoresis was borderline (40 mEq/L and 39 mEq/L). Bronchoscopy with lavage and biopsy showed diffuse mucoid secretions and normal ciliary architecture. CT of the sinuses showed pan sinus disease and CT scan of chest showed multiple areas of atelectasis and bronchiectasis. Subsequent CF genetic analysis including gross deletions revealed that the patient was homozygous for V456 A. We have also followed a 6-year old female of Indian descent who presented with positive newborn screening and failure to thrive. She was a compound heterozygote for V456A with mild lung disease and a borderline sweat test. Both of these patients were negative for the most common CFTR mutations. V456A was initially called a novel variation but now is described as mild pathogenic mutation (2) . Patient # 1 who is homozygous for V456A presented with a severe phenotype. These cases provide evidence that V456A can Objective: Evaluation of the functional cystic fibrosis transmembrane conductance regulator (CFTR) to assess new therapies and define diagnosis of atypical cases of cystic fibrosis (CF) is cumbersome and practicable only for selected subjects in few centers. It is known that leukocytes express detectable levels of CFTR. Our goal was to develop methods suitable to assess CFTR expression and activity in leukocytes obtained from healthy subjects or patients with cystic fibrosis ex vivo.

Methods: Western blotting and flow cytometry, and cell membrane depolarization, by single-cell fluorescence imaging, using the potential-sensitive probe bis-(1,3-diethylthiobarbituric acid) trimethine oxonol (DiS-BAC2), were used.

Results: We detected expression of a CFTR isoform in monocytes whose specificity is supported by the lack of expression in cells derived from a patient homozygous for nonsense mutations. The isoform detected in monocytes is different from the pattern observed in epithelial bronchial and pancreatic cells suggesting a different post-translational processing. Functional analysis has shown that, upon stimulation, only non-CF donors and, to a lesser extent, obligate CFTR heterozygous carriers (HTZ) monocytes showed an increase in the fluorescent signal, while monocytes from CF patients failed to respond and presented a spontaneous depolarization signal higher than controls and HTZ monocytes. This feature allows CF patients to be distinguished from all other individuals with high sensitivity and specificity (>95%). As a reference assay, Nasal Potential Difference was measured in selected subjects in parallel to the monocytes assay.

Conclusions: These findings suggest that evaluation of CFTR expression by Western blotting and flow cytometry and the measurement of its activity in monocytes by optical techniques may represent a surrogate biomarker suitable for assessment of CFTR expression and function in both basic and translational research, including drug development and diagnostic applications.

Supported by Lega Italiana Fibrosi Cistica Associazione Veneta Onlus, Italy and Italian Cystic Fibrosis Research Foundation, Italy (Grant: FFC 05/09). Rationale: Epithelial Na + Channels (ENaC) are located on the apical membrane of airway cells, playing an important role in airway surface liquid (ASL) maintainence through regulation of Na + influx. Individuals with CF suffer from hyperabsorption of Na + and consequently thick, viscous mucus and reduced ASL. Genetic knock-out models have shown that disruption of the alpha subunit results in the greatest attenuation in channel activity in alveolar cells. Previous work in humans has suggested that the A663T genotype of the alpha subunit of ENaC may help protect against hypertension, suggesting reduced ENaC function with the T polymorphism at amino acid 663 compared to the wild-type (A663A). Since stimulation of the β 2adrenergic receptors results in an increase in number and activity of ENaC, we sought to determine the influence of genetic variation at position 663 of ENaC on exhaled Na + in individuals with CF following β 2 stimulation. cause significant disease and support the theory that novel mutations are prevalent in South Asian populations. Background: CFTR mutations can be classified according to loss of function, with class 1, 2 and 3 mutations (including dF508) resulting in severely reduced CFTR function, while class 4 and 5 have some residual function. While CFTR functional classes are known to predict pancreatic status and survival, their association with clinical characteristics in young CF patients is poorly understood.

Objective: To evaluate the association between CFTR mutation functional class and characteristics of CF patients ≤12 years of age with a negative lifetime history of Pseudomonas aeruginosa (Pa) respiratory cultures.

Methods: Subjects were participants in the EPIC Observational Study, an ongoing longitudinal multicenter cohort study of early CF lung disease that enrolled 1700 CF patients at 59 centers from 2004-06. Subjects included in the current analysis fulfilled: 1) age ≤12 years at enrollment, 2) no prior isolation of Pa from respiratory cultures; 4) genotype data available; 5) clinical data available within 28 days of enrollment. The CFTR functional status of patients were classified as: Group A, both mutations in class 1, 2 or 3; Group B, 1 or both in class 4 or 5; Group U, 1 or both could not be assigned to a class. We describe the association between CFTR functional group and participant characteristics at enrollment.

Results: A total of 917 subjects were included in the analysis, of which 689 (75%) were in Group A, 13% Group B, and 12% Group U. Mean (SD) age at enrollment was 4.5 (SD 3.5) years. Patients in Group B were less likely to have been diagnosed with meconium ileus (3%) than those in Group A (25%) or U (16%) (p < 0.0001). Patients in Group B were less likely to use pancreatic enzymes (38%) than those in Group A (95%) or U (74%) (p < 0.0001). Patients in Group B had higher mean (SD) height (45(29)) and weight (50(32)) percentiles than those in Group A (height 35 (27), weight (38(28)) or U (height 35 (27), weight (42(30)) (p ≤ 0.0006 for each). BMI and lung function were not significantly different between groups. Rates of S. aureus and methicillin-resistant S. aureus colonization at enrollment were not significantly different between groups A, B and U (for Staph, 43%, 33%, 40%, p=.19; for MRSA, 6%, 1%, 2%, p = 0.06). In linear regression models, functional CFTR group was significantly associated with weight and height percentile but not CF-specific FEV1 percentile.

Conclusions: In a cohort of 917 Pa-negative CF patients at a mean age of 4.5 years, CFTR functional group was associated with nutritional status but not lung function or respiratory microbiology, suggesting a potentially large role of modifier genes and/or environment factors on pulmonary phenotype.

Supported by the CFF (CFF EPIC09K0). Background: Topiramate and zonisamide are used as anticonvulsant medications in children, and Topiramate is also used for migraine prophylaxis. There are reports of topiramate and zonisamide causing oligohydrosis as a side-effect (Clin Neuropharmacol 2008; 31: 339-346, Pediatr Neurol 2003; 28: 184-189) , but their effect on sweat chloride levels has not been studied systematically (Indian Pediatr 2008; 45: 238-240) .

Hypothesis: Topiramate and zonisamide not only reduce sweat production, but they can also cause elevation of sweat electrolytes.

Methods: Eighteen patients who were receiving topiramate or zonisamide therapy for at least 6 months were enrolled either from Neurology outpatient clinic or Epilepsy Monitoring Unit, and were only included in the study if they had no signs or symptoms and a negative family history of cystic fibrosis. Patients underwent pilocarpine iontopheresis and sweat collection via Macroduct system (Wescor Inc., Logan, UT) as per Clinical and Laboratory Standards Institute (CLSI) and Cystic Fibrosis Foundation (CFF) guidelines. The samples were analyzed the same day using a chloridometer and the mean sweat chloride values were compared to published normative data (J Pediatr 2008; 153: 758-63) by Student's t-test.

Results: Of the 18 pts (age range 3-22 years, mean age 9.94 years, sex ratio M:F = 0.44) that were enrolled for this study, 14 adequate samples (>15 µL volume) could be obtained from left arm, and 13 from right arm. There were 14 patients on topiramate and 4 were receiving zonisamide. The mean sweat chloride level was 41.77 ± 18.71 mEq/L from the left arm and 46.475 ± 19.29 mEq/L from the right arm for patients receiving topiramate. For patients on Zonisamide therapy, the mean sweat chloride level was 29.83 ± 10.00 mEq/L from the left arm, and 36.5 ± 3.54 mEq/L from the right arm. Sweat chloride values and volumes did not correlate with dose of either anticonvulsant (p=0.5).

Overall 7/14 (50%) patients on topiramate or zonisamide had either a borderline (> 40 mEq/L but < 60 mEq/L) or elevated (> 60 mEq/L) sweat chloride test result. Comparing these results to the standard reference values for sweat chloride in the normal population, the mean sweat chloride level of patients on topiramate was noted to be significantly higher (p<0.001) for both arms. There was no significant difference in the mean values for sweat chloride between the topiramate and zonisamide groups for both sides (p>0.05).

The insufficient sweat collection rate was 3/18 (17%) for both arms, and another 3/18 had insufficient sweat collected from one of the arms. This rate is much higher than the goal of <5% inadequate sweat collection rate that is established by CFF for accredited sweat testing laboratories, and suggests that drug-induced oligohydrosis is operative in this patient population.

Conclusions: Topiramate and zonisamide can not only cause oligohydrosis, but they can also elevate sweat chloride levels in patients with no clinical features of cystic fibrosis. Further studies to explore whether this is due to interaction with CFTR or via a different mechanism, are underway.

Hebestreit, H.; Syrée, C.; Hebestreit, A. Pediatrics, University of Würzburg, Würzburg, Germany Moderate to heavy physical exercise can change nasal potential difference (NPD) in healthy individuals and patients with cystic fibrosis (CF). The proposed mechanisms include an inhibition of epithelial sodium channels and an activation of chloride channels mediated by local effects of the exercise-induced increased ventilation such as a shear stress-induced ATP release or hyperosmolality of airway lining fluid caused by an epithelial fluid loss during exercise. Other mechanisms not related to local airflow changes may include a local secretion of NO from non-adrenergic noncholinergic nerves or effects of natriuretic factors released during exercise.

Since centrioles originate basal bodies, which then nucleate cilia, we also investigated the intracellular localization of TMEM45a in ciliated cells. In MDCK-1 cells bearing a primary cilium, TMEM45a was localized to the basal body. In ciliated native human nasal epithelial cells, TMEM45a expression was strongly associated with the layer of basal bodies underlying the apical membrane. It also co-localized to the basal region of the cilia themselves and with an intermediate fibrous structure lying between the Golgi and the apical membrane.

Our data document the co-localization of TMEM45a in both native nasal epithelial cells and cell lines with a variety of markers, and suggest its function as a regulator of or component in the development of cilia. We are currently attempting to strengthen the evidence for such a role by knocking down TMEM45a expression in a cell model of ciliogenesis. With a carrier frequency of 1/30 to 1/40, the most common mutation of the CFTR gene, F508del, is present in the whole population of central and western Europe and in their descendants residing elsewhere such as in North America. The origin and the age of the F508del is still a matter of debate and controversy. Although Serre et al (1) estimated F508del to be 3000-6000 years old, Morral et al (2) reported an "age" (duration of prevalence) approximating 52000 years. In order to gain new insight in this debate and improve our understanding of the origin and distribution of F508del, we selected homozygous CF patients and their parents (trios) in three current European populations (Irish, Austrian, Breton) and used the likelihoodbased method developed by Genin et al (3) to estimate the age of the most recent common ancestors of F508del mutation carriers. Unrelated CF patients and their relatives from Ireland (20), Austria (27), and Brittany, France (16) were included in the study. Microsatellites around the CFTR gene (7q31) were screened as previously described by Fichou et al (4) . These sequences were amplified by multiplex PCR using universal fluorescent labelling. Haplotypes at the 10 microsatellites loci were reconstructed, and the age of the most recent common ancestor was estimated using the estiage program (3). The age of the most common ancestor was estimated with 95% confidence interval: for the samples from Brittany, the result was 3750 years (2550-5625), for Austria 3650 years (2725-4950), and for Ireland 3925 years (2775-5675). These results indicate that the age of the F508del is quite similar in these three areas of Europe.

Our estimate corresponds to the Bronze Age and could be related to Indo European people who were Celtic ancestors and who migrated to and within Europe during this period and thereafter. These data also confirm that F508del is probably the oldest mutation of the CFTR gene in the white European population, compared to the age of the G551D mutation, the W846X or the 1078delT alleles which were estimated to arise more recently in the population of Brittany (600 to 1000 years old). Determining when the F508del mutation first became prevalent in Europe, coupled to historic information on environmental exposures, may provide a better understanding of the currently high prevalence of F508del CF.

Supported by NIH (USA) and INSERM (France The objective of this study was to assess the effects of a maximal cycling exercise with and without inhibiting nasal breathing on NPD to confirm a possible role of these latter non-local factors. Three separate studies were performed (study I-III). In all experiments, subjects completed an all-out cycling exercise following the Godfrey protocol. NPD was measured before and 11-25 min after the exercise bout. Study I evaluated the changes in NPD from baseline to post-exercise in 12 controls and 8 patients with CF wearing a nose clip during exercise. Nasal epithelium was superfused during NPD measurements only with NaCl 0.9% to confirm steady state post-exercise NPD values for minutes 11-25. Study II repeated study I with 8 controls and 4 patients with CF superfusing the respiratory epithelium at baseline and post-exercise with different solutions to estimate the effects of exercise on apical ion channel activity (Hebestreit et al., AJRCCM 2001) . Study III repeated study II in 7 controls exercising with and without a noseclip to assess the impact of nasal breathing on NPD and ion channel action.

All exercise tests were performed with maximal effort as indicated by high peak exercise heart rate values and respiratory exchange ratios, and high post exercise lactate levels. Five min post exercise, plasma levels of lactate and inorganic phosphate but not Na + , K + , and Ca 2+ were significantly increased over baseline in controls and patients with CF. Study I confirmed a stable NPD between 11 and 25 min post exercise in controls and patients with CF. The change in NPD between rest and the post exercise period was -6.5±6.2 mV in controls (p<0.001) and +1.5±7.0 mV in CF. Study II indicated a stronger effect of amiloride on NPD following exercise in controls (7.2±2.3 mV vs. 3.4±1.4 mV, p<0.01) and patients with CF (28.5±12.3 mV vs. 24.2±16.1 mV, p<0.01) while isoprenaline had a stronger effect after exercise only in controls (7.2±2.3 mV vs. 3.4±1.4 mV, p<0.01). Exercising with or without a nose clip had no effect on NPD and, thus, inferred ion channel activity (Study III).

In conclusion, maximal exercise induces a post exercise activation of amiloride sensitive sodium channels and increases the effects of isoprenaline on CFTR channels after exertion. The mechanisms involved are most likely not due to changes in airflow. Involvement of non-adrenergic non-cholinergic nerves or atrial natriuretic factors is unlikely since these should lead to an inhibition of epithelial sodium channels. Possibly, catecholamines released during exercise may -at least in part -be responsible for the observed findings. In a microarray study of differential gene expression in native nasal epithelial cells we found TMEM45a to be regulated in CF compared to healthy subjects. This 32 kDa transmembrane protein of unknown function has been found to be upregulated following hypoxia [1] and ER stress [2] , and downregulated by asthma and smoking [3] , and is thereby implicated in pathways relevant to CF pathophysiology. Our goal is to define the functional significance of differential TMEM45a expression in CF.

Firstly, we performed qRT-PCR in human bronchial epithelial cell lines, and found that TMEM45a expression was 2-fold down-regulated in CFBE41o-(F508del/F508del) vs 16-HBE (WT/WT) cells (p<0.01). We also confirmed that in these cells TMEM45a was up-regulated by hypoxia (4fold; p<0.0001). Next, using a polyclonal antibody against TMEM45a (C-13, Santa Cruz) we determined the intracellular localization of this protein by immunocytofluorescence. We found a punctate expression pattern for TMEM45a in several different cell types (HBE, CFBE, 293-HEK) corresponding to that expected for a centrosomal protein. Partial co-localization with αand γ-tubulin further suggested that TMEM45a is present in the pericentriolar matrix, a mass of proteinaceous material that organizes microtubules in the cytoplasm and mitotic spindle [4] . Our data also showed that the TMEM45a stained structure is dynamic, sometimes ring-shaped, sometimes taking the form of a pair of centrioles, and sometimes dispersed, e.g., during spindle formation and mitosis. When condensed, the structure is associated with the Golgi apparatus, as shown by co-localization with Golgi-P58 protein. The CFTR gene has been cloned 2 decennia ago, and over 1500 CFTR mutations have been identified. However, in certain genotypes we are still uncertain about the accompanying phenotype and their prognosis. Although one can divide CFTR mutations by type and phenotypic result (Castellani 2008 , Farrell 2008 , some gene variations like R117H remain difficult to classify due to the large variety in clinical outcome. This makes it difficult to decide whether this mutation should be tested in the NBS protocol.

Aim: To gain further insight into the genotype -phenotype relationship of the R117H CFTR mutation including determining CFTR function by NPD, intestinal current measurement (ICM), sweat test and faecal elastase.

Methods: We studied our group of CF patients which consists of approximately 230 pediatric and 168 adult CF patients. In the cohort of patients visiting our CF centre, we compared clinical presentation of the homozygous R117H patients.

Results: Two patients were homozygous R117H (7T/7T), one male (age 35) and one female (age 33). The male patient presented with infertility by CBAVD. The female person was without clinical symptoms and only tested because her son was identified with F508del/R117H by the newborn screening test. Function tests like NPD, ICM, and faecal elastase were normal in both subjects. Sweat tests were borderline in both patients.

With the newborn screening test by IRT, we found a relatively large proportion of R117H positive patients in the group with an elevated IRT. In our present CF patient population we see around 1-5% of patients with a R117H mutation. In the group found by the neonatal screening using IRT, PAP and DNA analysis we found R117H in 40% of patients (7 out of 18).

Conclusion: We showed that homozygosity for the R117H mutation does not predict clinical presentation. Our female patient was without any disease phenotype and produced offspring, whereas our male patient presented with CBAVD. Therefore, this single nucleotide polymorphism (SNP) is possibly associated with vas deferens dysfunction, but not with classical CF characterized by pulmonary or intestinal involvement. However, the presence of two CF mutations including R117H-7T on one allele can be associated with late presentation and severe lung disease (Peckham et al 2006) . The borderline sweat test in both patients does suggest a degree of CFTR dysfunction at least in this organ system. Therefore, we think it is too early to exclude this mutation from NBS. We will continue our analysis by including the phenotype and function tests of the R117H compound heterozygous patients to gain further knowledge on the penetration of the R117H mutation and the justification of its presence in the NBS panel. The phenotype of homozygous-∆F508 human subjects varies considerably, suggesting that genes other than CFTR modulate CF disease expression. The Yeast Oligomycin Resistance gene is an ABC transporter analo-gous to human CFTR. yor1-∆F670 (containing a mutation corresponding to ∆F508 in human CFTR) displays a CFTR-∆F508-like phenotype with regard to protein misfolding as judged by limited proteolysis and defective biogenesis. Sixty-three percent of genes described previously for the human CFTR Proteome (hCFTRP, Balch, et al. Cell. 2006 ) have yeast orthologs (as compared to only 23% human-yeast orthologs overall). A yeast model of ABC protein maturation therefore offers a powerful means to identify, prioritize, and determine the epistatic relationships among genes that influence ABC protein trafficking. In the present study, we are applying a novel method, Quantitative High Throughput Cellular Phenotyping (Q-HTCP), in combination with a molecular genetic protocol called SGA (synthetic genetic array) for introducing a tetracycline regulatable allele of yor1-∆F into the complete genomic collection of yeast deletion strains. This strategy has allowed us to identify enhancers and suppressors of oligomycin resistance in order to investigate their potential roles in biogenesis and trafficking to the plasma membrane of yor1-∆F. By analyzing kinetic growth curves for 4850 double mutant strains at multiple concentrations of oligomycin, we have defined several modifier gene categories relevant to yor1-∆F maturational processing. These functional categories include well known chaperones, protein trafficking genes and components of the ubiquitination and proteasome pathways previously shown to mediate defective biogenesis of CFTR-∆F508. Based on encouraging results from the primary screens, secondary genetic and computational studies were performed to prioritize hits. Biochemical and cell biological validation (QT-RT-PCR of yor1-∆F mRNA, cellular localization of yor1-∆F, pulse chase analysis) and computational screens (phenomic clustering of yor1-∆F modifiers, overlap of hits with related genetic screens) will be presented. We have also confirmed that a novel, evolutionarily conserved protein folding complex, recently identified by another laboratory and with relevance to CFTR (Weissman, et al. Science. 2009 ), participates in the biogenesis of yor1-∆F. Reduced growth is a common manifestation in cystic fibrosis. Understanding the origins of the growth reduction observed in CF patients is important because of its correlation with severity of lung function and reduced survival. CFTR is present in many tissue types other than epithelia, and evidence is mounting that it is functional in these tissues and, in certain tissues, may even contribute to normal growth. CFTR is present in the brain and is hypothesized to play a role in acidification of synaptic vesicles in neurons. CFTR's loss in neurons may inhibit secretion of important neuropeptides and possibly disrupt growth. However, given the systemic nature of CF it is difficult to identify if loss of Cftr in the brain directly contributes to poor growth in CF. CF mouse models, which display reduced growth similar to humans with CF, provide a tool to identify factors that contribute to growth reduction in CF. We have created a conditional Cftr knockout (KO) mouse that provides the ability to inactivate Cftr in specific tissues and cell types. To test the hypothesis that Cftr function in the brain may affect growth, we specifically inactivated Cftr in all neurons using our mouse model system. In these neuronal specific Cftr KO mice, we observed normal Cftr function in nasal and intestinal epithelium but complete absence of Cftr expression in the brain supporting neuron specific deletion of Cftr. We were surprised to find that in contrast to CF mice in which poor growth is evident within days of birth, the neuronal specific Cftr KO mice were indistinguishable from littermates early in life but by 15 days of age were significantly smaller than control littermates (p<0.05). At 6 weeks of age growth reduction was still evident (p<0.05). Interestingly, body mass index (BMI) of these mice was similar to control mice and significantly increased compared to CF mice (p<0.05). These findings may be due to an increase in inguinal fat in neuronal specific Cftr KO mice (p<0.05) compared to CF mice which have very little inguinal fat. In addition, Insulin Like Growth Factor-1 (Igf-1), an important regulator of growth, was significantly reduced in neuronal specific Cftr KO mice and CF mice compared to controls (p< 0.005). Our findings in these neuronal specific Cftr KO mice are similar to findings observed in Cystic fibrosis (CF) is an autosomal recessive disease in which the majority of the morbidity and mortality is associated with progressive pulmonary infection and inflammation. In the past, studies exploring the causes of CF pulmonary disease have been hindered by the absence of an animal model that develops lung disease similar to humans. The development of the pig model for CF provided us with the opportunity to study the early features of pulmonary disease, including the timing of the onset of infection and inflammation. The development of spontaneous pulmonary disease in older CF pigs and the documentation of a defect in the ability of newborn CFTR-/-pigs to eradicate bacteria, enabled us to investigate early temporal events regarding the onset of pulmonary inflammation in this model. Here we summarize results from experiments done in newborn pigs (within 6-12 hrs post birth). Histopathological examination of the airways and parenchyma revealed no evidence of cellular inflammation, with the newborn CFTR-/-lungs appearing very similar to that of their CFTR+/+ littermates. Bronchoalveolar lavage (BAL) performed on several newborn animals to measure total cell counts, neutrophil percentage and IL-8 concentrations in BAL fluid found no statistically significant differences between genotypes. We also evaluated gene transcript profiles in trachea, bronchi, and distal lung in newborn CFTR+/+ and CFTR-/-pigs. Unsupervised hierarchical clustering segregated samples into two clusters: airway (trachea + bronchi) and distal lung; but failed to segregate samples based on CFTR genotype. A very small number of genes were identified by ANOVA to be differentially expressed between the two genotypes. However, neither did these genes exhibit consistent change in expression across the three tissues, nor were they involved in any pathway or gene network based on analysis done using Ingenuity Pathway Analysis (IPA) and Database for Annotaton, Visualization and Integrated Discovery (DAVID). To explore for the presence of enriched pathways, we queried the microarray data using Gene Set Enrichment Analysis (GSEA). No evidence of a pathway, gene network, or signaling cascade consistent with inflammation in the lungs of newborn CF pigs was identified. To specifically investigate the enrichment of pathways involved in innate immunity, host defense, fluid and electrolyte transport, and other systems relevant to CF pathogenesis, we made several custom gene sets for GSEA based on published literature. No statistically significant enrichment was observed in the CFTR-/-samples. Finally, on looking for intersections between our data and previously published expression profiling in CF, we found no genes common between our work and previous studies that were relevant to CF pathogenesis and consistently changed in the porcine data. These data indicate that newborn CFTR-/-pigs manifest a lung host defense defect before the onset of inflammation. Normal airway epithelial cells have the capacity to absorb and secrete salt via active ion transport mechanisms, with a resultant movement of water in response to the generated osmotic gradient. However, as a consequence of defective CFTR in CF cells, the control of secretion of Clions and water flux is disturbed, leading to dehydrated airway mucosa and cessation of mucociliary clearance (MCC). Hypertonic saline (HS) is an ionic rehydrating agent that increases the osmotic pressure on the surface of the airway epithelium and draws water down an imposed osmotic gradient into the airway surface liquid (ASL). Aerosolized HS has been clinically proven to be an effective therapy in CF, resulting in improved MCC and FEV1. Despite mouse models in which the growth hormone pathway is disrupted due to specific gene mutations. Similarities between these mouse models suggest that the growth hormone pathway may be disrupted in CF due to loss of Cftr in the brain. These results also suggest that loss of Cftr in the brain may contribute to an overall endocrine dysfunction that could contribute to endocrine abnormalities observed in CF patients. Further investigation is necessary on the specific role Cftr plays in the brain and how its loss may contribute to growth. These studies suggest that therapeutic strategies for CFTR correction should include the targeting of non-epithelial tissues to facilitate correction of CF manifestations. Nearly all male cystic fibrosis (CF) patients exhibit tissue abnormalities in the reproductive tract, a condition that renders them azoospermic and infertile, a phenotype not present in CF mouse models. Progress in defining a causal link between CFTR mutations and duct atresia has been hindered by the lack of an animal model that includes CBAVD. Two porcine CF models have been reported recently that include respiratory and digestive manifestations that are comparable to human CF. The goal of this study was to determine whether a male reproductive tract phenotype is present in the porcine CF models. Tracts from newborn CFTR -/-, CFTR ∆F508/∆F508 and wildtype (WT) littermates were isolated, placed in ice-cold buffer and studied in 24 hours. Inspections of CFTR -/tracts revealed that 14 of 14 samples exhibited vas deferens and/or epididymis abnormalities, while 28 of 28 WT littermates presented normal anatomy. Vas deferens atresia was often localized to the distal region of the duct (prostatic end), with a variable proportion of the proximal vasa being present. Epididymal atresia was present in 13 of 14 CFTR -/tracts, and was consistently localized to the corpus epididymis. In some cases, only rudimentary caput and cauda epididymes were present. Among CFTR ∆F508/∆F508 , 2 revealed disruptions in the corpus epididymis and no overt atresia was observed in 2 tissues, although the vasa appeared more fragile and thinner. Histology from the CFTR -/vas deferens often revealed the presence of a lumen lined by epithelial cells. Epithelial cells delineated round areas of empty space adjacent to the basement membrane, suggesting the formation of cysts. Signs of epithelial cell degeneration and luminal debris were present. Epithelial cells lining patent vas deferens were isolated and grown on permeable supports for electrophysiological and immunochemical analysis. Western blots revealed CFTR immunoreactivity only in WT, whereas signals were detected for SLC26A3 and SLC26A6 in all samples. When examined in Ussing chambers, WT exhibited transepithelial resistance that was 26% greater than CFTR -/and CFTR ∆F508/∆F508 , which suggests that the epithelial barrier is compromised in the absence of CFTR. WT responded to forskolin, norepinephrine, adenosine and PGE2 with increases in current (I SC ) that were subsequently sensitive to DASU-02 or bumetanide, as reported for adult porcine and human vas deferens epithelia. CFTR -/exhibited no discernable Isc responses, while CFTR ∆F508/∆F508 presented a degree of forskolin-induced Isc. When cultured in the presence of dexamethasone, all genotypes exhibited greater basal I SC that was sensitive to amiloride, again consistent with data reported for adult epithelia. Taken together, these results show that the male reproductive duct is fully formed during fetal development, but that CFTR -/and CFTR ∆F508/∆F508 epididymes and vas deferens undergo atresia in late gestation. Epithelial cells lining the reproductive ducts can absorb Na + via ENaC, but CFTR -/are unable to secrete either Clor HCO 3 -, likely resulting in reduced duct luminal fluid volume, with abnormal characteristics. Mechanisms linking CFTR expression and ion transport abnormalities to duct atresia remain to be determined, and these porcine models appear to be ideal tools for such studies. [Supported by NIH HD058398.] its clinical use, however, little is known about the mechanism of action of HS in vitro. Previous studies evaluating the effects of HS have used large, non-physiologic volumes applied directly to the surface of airway epithelia. The goal of this study is to utilize an in vitro nebulizer-based delivery system to examine the effects of a nebulized dose of HS on ASL kinetics during and after treatment to human CF airway epithelial cultures.

Using CF human bronchial airway epithelial (HBE) cells cultured on an air-liquid interface and a novel nebulizer delivery system mimicking the delivery of an ultrafine mist of nebulized medications, we delivered 7% HS to the apical surface of the cells in a clinically relevant (nanoliter) rate/volume. Changes in ASL height during and after delivery were measured by confocal microscopy. The resulting effects on the ASL height were mathematically quantified using a specially designed computer program (MAT-LAB).

Application of HS (at 66 nL/min) to the surface of airway epithelial cells over a 15 min nebulization period resulted in a predictable rise in ASL height. The average rate of ASL fluid increase during the nebulization process was 4.11 µm/min (range 2.03-7.74 µm/min). The maximal height of ASL varied by culture but did seemingly continue to rise throughout the treatment to a maximal height of 46.2 µm (range 35.3-58 µm) from a starting height of ~10 µm. The reabsorption of the ASL began immediately after termination of nebulization and returned to baseline ASL height within an average of 11 min. ASL reabsorption rate was -2.93 µm/min (range -2.27 to -3.26 µm/min).

Nebulized HS applied to the surface of CF HBE results in a relatively rapid rise in ASL height and an equally rapid resorption of the ASL once nebulization ceases. Our initial data indicates that there is at least a four-fold increase in ASL height with HS nebulization. Interestingly, the effect is not sustained past the time of the treatment, since ASL falls to baseline within minutes of stopping nebulization. Additional studies are planned to evaluate the effect of deposition rate or % HS on sustaining the increase in ASL height. Since this technique models the deposition rate delivered to the lower airways during nebulization, we anticipate this approach will be useful in studying therapeutics or combinations of pharmacologic agents purported to affect ASL height. 

Rationale: Cystic fibrosis is a classic "monogenic" disorder; however, there is great phenotypic heterogeneity in CF lung disease, reflecting genetic and environmental influences. Polymorphisms in genes other than the CFTR gene have been shown to modify severity of CF pulmonary disease by the North American CF Genetic Modifiers Consortium via candidate gene approach, and are now being studied using GWAS technology (Illumina 610 Quad array) (Pediatric Pulmonology 2009 Sept;Suppl 32:165). A complementary approach will test variation in gene expression versus severity of lung disease, as defined by a quantitative lung phenotype (Pediatric Pulmonology 2009 Sept; Suppl 32: 266). In addition to testing expression from RNA from transformed lymphocytes (n=1000), nasal epithelial RNA is being collected for gene expression studies in a subset of 200 patients with a range of mild to severe lung disease. These gene expression studies require estimates of existing inflammation in nasal mucosa to evaluate for confounding influence.

Methods: Each patient has assessment of the degree of underlying inflammation via characterization of inflammatory cell types found on inferior turbinate sampling by nasal scrape biopsy (DiffQuik® stain), and by cytokine profile from nasal lavage (Invitrogen® Multiplex assay). Nasal epithelia are obtained by nasal scrape biopsy, and RNA is isolated by standard purification techniques, using QIAzol® Lysis Buffer. Gene expression will be quantified using the Affymetrix® gene array (minimum of 5 µg of highly purified RNA).

Results: Initial studies in 23 patients (DF508 homozygotes) with a range of lung disease severity have demonstrated that nasal epithelial sampling identifies respiratory epithelial and inflammatory cells on stained cytospin preparations). We identified an average 16% neutrophils (+/-21%), 1% lymphocytes (+/-1%) and 83% epithelial cells (+/-21%). Cytokine profiles from nasal lavage are pending (data to be shown). Pilot studies demonstrated robust signals from IL-8 (645.4 pg/mL +/-152.4) indicating variability for analysis of degree of inflammation. RNA obtained in 20 of these patients had (mean +/-SEM) an average concentration of RNA isolated 467 ng/µL (+/-185.8). The average total quantity obtained was 14.0 µg (+/-5.57). The mean 260:280 ratio was 1.96 (+/-0.05), and mean 260:230 ratio was 1.88 (+/-0.12), indicating high purity. Finally, the average RNA Integrity Number was 7.98 (+/-0.6) further reflecting high quality RNA for gene expression analysis.

Conclusions: Our techniques yield adequate nasal lavage and mucosa for characterization of inflammation by cell counts and by cytokine profile characterization. We obtain a sufficient quantity of high quality RNA for Affymetrix® expression profiles. Our analyses of nasal epithelial gene expression can be correlated with expression profiles of transformed lymphocytes, and allow eQTL analyses using GWAS data to better inform analyses of genomic variation.

Supported by: 5R01HL095396-02. In respiratory diseases such as CF, the small airways are the major site of airway epithelium injury and remodelling. The bronchiolar epithelium has then to rapidly regenerate its structure in order to restore its defense functions. The aim of our study was to determine and characterize the sequence of events leading to the reconstitution of the human bronchiolar epithelium. Bronchioles were microdissected from lung samples and processed for histology, immunohistochemical characterization and epithelial cell dissociation. After two rounds of enzymatic dissociation, bronchiolar epithelial cells (BECs) were cultured at the air-liquid interface (ALI), and cultures were collected at different steps of the regeneration process for characterization of epithelial cell polarization, differentiation and functionality. Our results demonstrated that, at day 6, BECs formed a polarized monolayer expressing intercellular junction proteins. BECs started to differentiate at day 15, and at day 25, the cultures were well-differentiated, similar to the native human bronchiolar epithelium, and composed of basal, ciliated, goblet and Clara cells. The CC10 protein was detected in the secretory products collected from ALI cultures. The Ussing chamber technique showed that cAMP-induced chloride effluxes were detected at, and increased from the step of ciliated cells appearance (day 15) whereas CFTR mRNAs were expressed as early as day 5. Finally, the ability of BECs to kill S. aureus progressively increased during the course of regeneration, and in the well-differentiated cultures the ciliary beating frequency was similar to that measured for native bronchiolar epithelium sheets. In conclusion, we have developed an in vitro model allowing reconstitution of a fully mature and functional human bronchiolar epithelium. This model represents a tool for the analysis of the molecular events associated with CF and non-CF bronchiolar epithelium regeneration.

Supported by the French Association Vaincre La Mucoviscidose and by the Agence Nationale pour la Recherche.

Jolly, T. 1 ; Le Naour, R. 3 ; Roux, J. 1 ; Collet, S. 1 ; Birembaut, P. 1,2 ; Coraux, C. 1 1. UMRS 903, INSERM, Reims, France; 2. CHU Maison Blanche, Laboratoire Pol Bouin, Reims, France; 3. EA 4303, URCA, Reims, France In constant contact with the external environment and in chronic inflammatory respiratory diseases such as cystic fibrosis, the bronchiolar epithelium is frequently injured, remodelled and has to regenerate its structure for the restoration of its integrity and defence functions. Previous studies per-decreased mineral appositional rate in F508del/F508del mice compared to that for Cftr+/+ mice. Surprisingly, most of abnormalities observed were more pronounced in male compared to female F508del/F508del mice. Our results on histomorphometric and dynamic parameters indicate a significant reduction in bone formation and a concomitant increase in bone resorption. Studies on proliferation, osteoblast gene expression and matrix mineralization in cultured bone marrow stromal cells from F508del/F508del and Cftr+/+ mice are currently in progress.

Conclusion: Severe osteopenia and structural abnormalities in F508del/F508del mice suggest that F508del mutation has a direct impact on bone metabolism in F508del/F508del mice that is not sex specific.

Supported by grants from Inserm, French Cystic Fibrosis association (Vaincre La Mucoviscidose, contract n°0905) and the Region Champagne-Ardenne (contract n°D200911226). The human CFTR mutation, ∆F508, results in a misfolded protein product which traffics improperly. However, recent studies utilizing expression systems show that this mutation in mouse Cftr and pig CFTR may not disrupt trafficking to the same extent as in humans (1, 2) . The availability of transgenic CFTR pigs, specifically those with the CFTR ∆F508/∆F508 genotype, allows measurement of the functional impact of this defect in native tissue and hence permits us to test this hypothesis (3).

Subclinical hypothyroidism has been associated with CF, suggesting a role of CFTR in thyroid function (4) . To assess CFTR transport function in thyroid, we obtained newborn CFTR ∆F508/∆F508 (∆F) pig thyroids, grew primary thyroid epithelial cultures (pThECs) on permeable supports and measured short-circuit current (I sc ). We evaluated 1) inhibition of Na + absorption with mucosal amiloride (AMIL; 10 -5 M); and 2) stimulation of Clsecretion in response to serosal application of the cAMP agonist, isoproterenol (ISO; 10 -5 M). CFTR +/+ (WT) and CFTR -/-(KO) thyroid cultures were used as positive and negative controls. We also performed surface biotinylation, probing the input and eluted fractions with mouse monoclonal anti-human CFTR antibodies which we determined to cross-react with pig CFTR (596 and 660; Cystic Fibrosis Foundation Therapeutics) (5) .

Monolayers grown from ∆F thyroids showed baseline I sc of 6.54 ± 0.577 µA . cm 2 (n=12 from 4 thyroids), not significantly different from WT (p>0.05). AMIL inhibited I sc by ~ 78 ± 15.5% (5.12 ± 0.507 µA . cm 2 ; 12 monolayers, 4 thyroids). Interestingly, ISO addition transiently increased I sc (peak of transient ~ 2.24 ± 0.298 µA . cm 2 ). This finding is consistent with previous work on cells heterologously expressing porcine ∆F508. In contrast, pThECs grown from KO thyroids never responded to ISO with a transient increase in I sc .

Control biotinylation studies of WT pThECs indicate that 23.9% of the total protein is present at the cell surface. Astonishingly, our preliminary work shows that 23% of the total ∆F508 was detected at the surface as well, hinting at a negligible trafficking defect.

The present findings support the idea that pig ∆F508 CFTR traffics to the apical surface of epithelial cells, a conclusion stemming from work using heterologous expression of exogenous mouse and pig ∆F508 (1, 2). To our knowledge, the present study is the first demonstration of this concept in a native system endogenously expressing physiological levels of ∆F508 CFTR. formed in animal models, have proposed that Clara cells could be the bronchiolar epithelial stem cells. Recently, we have shown that in human, adult bronchial epithelial basal cells are able to reconstitute a mucociliary differentiated and functional epithelium, demonstrating that the epithelial basal cells could be considered, at least, as progenitor cells of the human adult bronchial epithelium (Stem Cells, 2007, 25 (1):139-48). The aim of the present study was to determine the stem/progenitor potential of basal cells of the human adult bronchiolar epithelium.

Human bronchioles were microdissected from lung pieces obtained from surgery and processed either for histological examination and immunohistochemical detection of potential basal cell markers, or for bronchiolar epithelial cells isolation. After enzymatic dissociation, bronchiolar epithelial cells were prepared for analysis or cell sorting by flow cytometry. We evaluated the sorted basal cells capacity to regenerate a human bronchiolar epithelium, in vitro, in Air-Liquid Interface (ALI) culture. Finally, we measured the ciliary beating frequency of the in vitro regenerated epithelia, and the bioelectric properties in Ussing chamber.

Our results demonstrate that the human bronchiolar epithelium shows a basal cell layer. Bronchiolar basal cells, expressing specifically the cytokeratin 13 (CK13), exhibit a specific CD151 staining on their plasma membrane. Analysis by flow cytometry confirms the CD151 and CK13 expression by bronchiolar epithelial basal cells. Based on the CD151 expression, CD151+ and CD151-cells were sorted by flow cytometry from total human bronchiolar epithelial cells. While the CD151-sorted bronchiolar epithelial cells, consisting of ciliated, goblet and Clara cells, do not adhere on porous membranes, the CD151+ sorted basal cells adhere in vitro, proliferate then regenerate a differentiated bronchiolar epithelium. The bronchiolar epithelium, regenerated from CD151+ sorted basal cells, shows bioelectric properties and ciliary beating frequency similar to those of a bronchiolar epithelium regenerated from total bronchiolar epithelial cells.

Our results demonstrated that the human bronchiolar basal cells are able to regenerate a differentiated and functional bronchiolar epithelium in vitro in ALI culture. They can be then considered as progenitor cells of the human adult bronchiolar epithelium. This work is supported by the French Association Vaincre La Mucoviscidose and by the Association des Amis de l'American Memorial Hospital de Reims. Background: Recent clinical studies have suggested that the origin of CF bone disease in early childhood may be independent of nutritional status or disease severity. However, the direct contribution of F508del mutation in CF bone disease remains unknown. We recently showed that in normal primary human osteoblasts, the inhibition of CFTR-mediated chloride function caused a marked dysregulation in the production of osteoprotegerin and prostaglandin E2, two key regulators of the bone formation-resorption (Le Heron et al. J Cystic Fibrosis, 2010) .

Objective: To evaluate whether severe osteopenia is directly linked to the F508del mutation, we analyzed histomorphometric and dynamic parameters in bones of F508del-CFTR homozygous mice aged 10-12 weeks compared to normal Cftr+/+ littermates.

Results: Bone mineral density (BMD) and bone mineral content (BMC) were measured by DEXA (using a PIXImus instrument) in groups of F508del/F508del mutant and Cftr+/+ 10-week-old gender-matched mice. In F508del/F508del mice, the whole body BMC and BMD were significantly reduced (-22 % and -11%, respectively) compared to gender-matched Cftr+/+. BMC and BMD reductions were also observed in individual bones (femur, -27% and -19%, respectively; tibia, -20% and -10%, respectively; caudal vertebrae, -18% and -10%, respectively). Histomorphometric measurements made on femur sagittal sections revealed significantly reduced trabecular bone volume from F508del/F508del mice. Trabeculae number and thickness were decreased significantly in F508del/F508del femoral bone whereas trabeculae separation was increased. Quantitative analysis of TRAP-positive osteoclasts showed a 2 and 4-fold increase (female and male mice, respectively) in femoral trabecular bone from F508del/F508del mice compared to controls. Dual tetracycline and calcein labeling showed a

Bielemeier, A.; Marshall, L. Infection and Immunity, Aston University, Birmingham, United Kingdom Introduction: Understanding the disease pathogenesis of CF currently relies on animal and/or single cell culture models. Whilst these approaches partially enable the study of CF, the models have inherent limitations.

Objectives: We intend to overcome some of these limitations by developing a multi-cellular co-culture model of human airways in vitro. Our coculture model for non-CF airways uses human pulmonary fibroblasts (HPF) and human bronchial epithelial cells (Calu-3) grown at air liquid interface (ALI). This model is adapted to mimic CF by replacement of Calu-3 with IB3-1 (the CF cell line bearing F508del/W1282X).

Cell Viability: Optimal conditions for co-culture growth were defined by light microscopy, flow cytometry and a fluorescence-based assay after exposure to fresh and conditioned medium.

Cell Morphology: Antibodies to the epithelial cell specific markers cytokeratin (CK) 5 and 8 were used to stain mono-and co-cultures. The presence and location of fibroblasts was confirmed with 1B10, an antibody specific for fibroblast surface antigens. Zonulae occludens-1 (ZO-1) staining and z-stack analysis provided evidence of tight junction formation. Scanning electron microscopy (SEM) was used to verify the presence of cilia in long-term, differentiated mono-cultures of normal and CF epithelial cells.

Cell Physiology: Differentiation to a "tight" epithelial cell barrier was monitored by measuring transepithelial electrical resistance (TER). Apical fluid secretions were collected from mono-and co-cultures using sterile PBS and analysed for the presence of mucin using a monoclonal MUC5AC antibody.

Results and Discussion: Cell Viability: Suitable growth medium has been defined for the co-culture models, which maintains typical morphology, viability and growth characteristics of each cell type.

Cell Morphology: CK staining demonstrates basal (CK5 positive) and differentiated (CK8 positive) epithelial cell populations after growth at ALI for mono-and co-cultures. Z stack analysis of ZO-1 staining indicates that tight junctions are present apically; confirming that a polarised epithelial cell layer is formed. Using SEM, cilia can be seen on the apical surface of 3 month old mono-cultures.

Cell Physiology: Increasing TER indicates formation of a barrier by epithelial cell mono-cultures and epithelial-HPF co-cultures but not by HPF in mono-culture. Mucin is secreted apically by co-cultures and preliminary results suggest mucus hypersecretion in CF models compared to non-CF. Currently we are working on analysing the responses of our cell models to CF relevant pathogens and characterising the spectrum of antibacterial activity of apical secretions.

Conclusions: These results are an invaluable starting point for developing an in vitro model employing multiple, human cell types for further investigation of CF. We envisage that our models will be useful for comparing the response of human CF and non-CF airways to infection and inflammation. Submucosal glands (SMG) are recognized as an important tissue in the pathogenesis of cystic fibrosis (CF) lung disease. Some important functions include secretion of bulk liquid, electrolytes, mucin, and antimicrobial peptides and proteins. However, morphologic features of SMGs early in life are limited and often biased by multiple variables. We adopted a morphometric approach to characterize SMGs in fixed cross sections of neonatal CFTR-/-(CF) and CFTR+/+ (WT) pig trachea. The number of SMG ducts was reduced in CF pigs, however when normalized to lumen circumference, duct numbers were similar between CF and WT trachea suggesting similar frequency of SMG units. We then stereologically assessed SMG units (including ducts, acini, tubules, lumens). CF pigs had reduced SMG volume compared to control pigs, and this was also true even after normalization to lumen circumference and inner wall area. We found preferential SMG distribution along the posterior wall, followed by the anterior and then lateral walls. While this distribution pattern was similar for both groups, CF pigs had reduced SMG volume compared to WT pigs at each wall location. We assessed surface epithelium mucus with Periodic acid-Schiff and immunohistochemistry for MUC5AC and MUC5B; they showed similar mucus staining patterns in both genotypes. In contrast, periodic acid-Schiff and MUC5B staining were significantly reduced as a percentage of SMG volume in CF pigs, suggesting reduced mucous cell differentiation. This was confirmed by reduced cytokeratin 18 staining in CF SMGs, consistent with reduced mucous cell differentiation. SMG lumen size as a percent of SMG unit volume was the same in both genotypes, indicating that they were not dilated. We examined a proliferation marker (Ki-67) and found similar proliferation in CF and WT surface epithelium, but reduced proliferation in CF SMG. We then asked if there were changes in the global transcripts of SMG and surface epithelium. We found no enrichment in the surface epithelium gene set, but CF pigs had significant global reduction in SMG gene sets. Our data are consistent with immature and hypoplastic SMGs in the neonatal CF pig trachea. This is contrary to the hypertrophic and hyperplastic SMG changes often described in advanced CF disease. Furthermore, immature and small SMGs could have altered secretions, which could influence mucociliary clearance and antimicrobials levels in airway fluid, both potentially contributing to CF pathogenesis. Deletion of F508 in CFTR is the most common mutation that causes CF. In vitro studies have shown that the ∆F508 mutation disrupts protein processing and decreases chloride channel function. However, the relationship between in vitro and in vivo behavior of CFTR∆F508 protein remains uncertain. In addition, lack of a CFTR∆F508/∆F508 animal model has limited insight into pathophysiology and possible treatments. Therefore, we developed CFTR∆F508/∆F508 pigs (hereafter called ∆F pigs). All ∆F pigs had meconium ileus, and microgallbladder was common. Focal biliary cirrhosis was detected in some livers. The pancreas showed significant destruction, although detectably less severe than in CFTR-/-pigs. The ∆F lung showed no evidence of inflammation and the morphology was similar to that of the CFTR-/-pigs. Thus, the histopathological findings in ∆F pigs were strikingly similar to those in the null pig. We asked how CFTR mRNA and protein are altered in ∆F pigs. Preliminary results indicated that mRNA levels in ∆F intestine were approximately half those in wild-type littermates, consistent with our earlier results in CFTR+/∆F508 pigs. We also found a striking reduction in the amount of CFTR∆F508 protein in cultured and excised airway epithelia and in intestine. CFTR isolated from wild-type pigs was nearly all mature band C protein. In contrast, CFTR isolated from ∆F pigs was present as both band B and band C. This pattern was the same in cultured and excised airway epithelia and in intestine. Immunocytochemistry showed wild-type CFTR in intestinal crypts and the apical membrane of airway epithelia. CFTR∆F508 was expressed in the same cells, but at greatly reduced levels, and much of the mutant protein was intracellular. These data are consistent with our earlier studies of in vitro recombinant porcine CFTR∆F508. To test CFTR∆F508 function, we assayed Clcurrent in intestine mounted in Ussing chambers. We found little to no forskolin-stimulated current in intestine with the exception of ileum where we also recovered The Ift88 TG737RpW polycystic kidney disease (PKD) mouse was developed as the result of a large-scale, transgene-induced insertional mutagenesis screen. An insertion disrupted the intraflagellar transport 88 (Ift88) locus, resulting in a hypomorphic murine allele (Ift88 TG737RpW ). Homozygous mice with this mutation not only exhibit renal cyst abnormalities, but also pancreatic and hepatic damage (including cyst formation) morphologically reminiscent of human cystic fibrosis. Cysts in PKD have been reported to develop in a fashion dependent upon CFTR expression. Moreover, Ift88 may have important ramifications with regard to the CF phenotype. For example, Ift88 is necessary for normal development of both immotile and motile cilia in mammalian epithelial tissues. Abnormalities attributable to motile cilia dysfunction have been noted previously in Ift88 TG737RpW mice, and defects in Clsecretion have been reported in certain tissues of the Ift88 mouse model. Taken together, these observations suggest a shared mechanistic pathway linking two very common congenital diseases defined by cyst formation. The present study was therefore intended to determine whether partial or total loss of CFTR would influence the phenotype of Ift88 TG737RpW mice, or if disruption of Ift88 (with disturbed epithelial ciliary activity) could negatively impact a CFTR-/-murine model. We bred CFTR tm1Unc/J mice onto the Ift88 TG737RpW strain (both strains congenic on either C57BL/6 or Balb/c backgrounds), and genotyped a colony of more than 60 individuals over several generations. Measurement of nasal potential difference, intestinal Isc, ciliary function (in murine airway cell monolayers), and histopathology will be presented for heterozygous and double homozygous mice to determine whether mutations in both genes confer physiologic or morphologic additivity. The study will also provide a formal test of the model in which PKD severity is dependent upon functional CFTR.

Supported by CFF and NIH. Small molecules and gene products that are capable of partially restoring the ∆F508-CFTR folding and processing defect have been, predominantly, identified by function-based high-throughput screening (HTS) assay that measures the PKA-activated halide permeability of the plasma membrane. As a direct biochemical measurement, we developed a cell surface ELISA, which detects the extracellular 3HA-tag of the ∆F508-CFTR, however, it entails multi-steps immunoperoxidase detection, which is not optimal for the HTS. To overcome this limitation we designed a robust, sensitive and economical biochemical assay that would permit performing chemical and genetic screens to identify additional corrector molecules and potential drug targets. The assay is based on the assumption that inserting the horseradish peroxidase (HRP) enzyme in the exofacial surface of CFTR would provide an enzyme activity in the extracellular compartment following the mutant translocation to the plasma membrane. To ensure that the CFTR-HRP preserves the biochemical and functional properties of the channel, the biosynthetic processing and chloride transport characteristics of the chimera were compared to that of CFTR-3HA. Processing of wt CFTR-HRP and the low temperature rescued ∆F508-CFTR-HRP were confirmed by immunoblot analysis, as well as by glycosidase digestion in stably transfected BHK cells. Preliminary metabolic pulse chase and cell surface stability measurements indicate that the turnover of the wt and rescued ∆F508-CFTR-HRP variants are comparable to that of their 3HA-tagged counter-the highest amount of CFTR∆F508 protein. Excised and cultured airway epithelia studied in Ussing chambers showed a reduced forskolin-stimulated Clcurrent that was slightly greater than that in CFTR-/-epithelia consistent with the reduced amount of CFTR∆F508 at the apical membrane. Measurements of in vivo nasal voltage revealed a pattern like that in CFTR-/pigs. These studies suggest the following interpretations: First, the clinical phenotype of ∆F pigs is similar to that of CFTR null pigs. Second, because disease in ∆F pigs was not more severe than that in null pigs, it is likely that loss of functional CFTR rather than a detrimental effect of mutant CFTR∆F508 caused disease. Third, the small amount of function conferred by the CFTR∆F508 protein was not sufficient to prevent disease. Fourth, in vitro studies of recombinant pig CFTR∆F508 may predict the in vivo behavior of the mutant protein. NFκB is a major transcription factor that regulates genes responsible for both the innate and adaptive immune response and controls many genes involved in inflammation. NFκB forms a complex that includes p65 (aka RelA). In unstimulated cells, NFκB complexes are sequestered in the cytoplasm by a family of inhibitors, called IκBs (Inhibitor of κB) and upon stimulation, IκB is degraded allowing p65 to translocate into the nucleus where it then regulates transcription. Common methods for measuring NFκB activation include luciferase-based assays and Western blots, neither of which are ideal for making repeat measurements in live cells over time. We therefore set out to design an NFκB reporter assay that could be used in live cells, and was suitable for use in a microplate reader/96 well plate format. To this end, we designed a FRET pair that incorporated p65 N-terminally fused to enhanced green fluorescent protein (eGFP) and IκB that was N-terminally labeled with monomeric cherry fluorescent protein (mCherry). Confocal microscopy analysis demonstrated that both constructs were situated in the cytoplasm of HEK293 cells under basal conditions, and that eGFP-p65 translocated to the nucleus upon stimulation with either TNFα or NECA, which activates adenosine receptors. In contrast, the fluorescence intensity of mCherry-IkB was markedly reduced upon stimulation consistent with the degradation of this protein that accompanies p65 activation. Confocal measurements of FRET revealed a ~3 fold decrease in FRET as these proteins dissociated (n=10). We next measured FRET in a 96 well plate using a Tecan Infinite plate reader. Non-transfected wells served to correct for background and wells transfected with eGFP-p65 alone served to control for any spectral bleed through. We then excited eGFP at 488 ± 5 nm and collected emission at 510 ± 15 nm (donor channel). We also collected FRET fluorescence by exciting at 488 ± 5 nm and collecting mCherry emission at 610 ± 15 nm (FRET channel). Under these conditions, we could detect significant emission in both channel and divided donor emission by FRET emission to calculate the FRET ratio. Five min after addition of 10 µM NECA, the FRET ratio significantly increased 3.2 fold, a change which persisted for 30 min before falling towards baseline (n=6). Similarly, TNFα exposure elicited a 50% increase in the FRET ratio which also waned over time (n=6). We conclude that the eGFP-p65/mCherry-IκB FRET pair is suitably sensitive to be used to measure FRET in an automated fashion in a microplate reader. The effect of CFTR on NFκB FRET remains to be determined.

Funded by BAT and the NIH.

parts in BHK cells. Iodide efflux assays suggest that the wt CFTR-HRP chimera is functional. To more closely mimic the in vivo situation, CFTR-HRP was also expressed in human CFBE41o-bronchial epithelial cells. Initial cell surface density measurements indicate that low temperature rescues the ∆F508-CFTR-HRP processing defect, manifesting more than ten-fold higher fluorescence or luminescence signal over the background. Further validation of the assay is underway using the corrector panel provided by CFFT and Dr. R. Bridges. Initial results suggest that the correction profiles, detected by the cell surface ELISA and the CFTR-HRP-assay are comparable. Optimization of the assay in 96 and 384 well format will facilitate largescale biochemical corrector screens to identify potent lead compounds and drug targets in CF. Supported by CFFT and NIH as well as EMBO and CFF fellowships to GV and TO, respectively. Airway mucus clearance is an essential innate defense mechanism and its failure contributes to pathology of obstructive lung diseases, such as cystic fibrosis (CF). In Scnn1b-transgenic (Scnn1b-Tg) mice, airway-targeted overexpression of the epithelial Na + channel β subunit (βENaC, Scnn1b gene) causes airway surface dehydration and results in airway mucus obstruction and inflammation that strikingly resemble CF lung pathology, except for the lack of chronic bacterial infections. To determine if bacteria play a role in the pathology of Scnn1b-Tg mice, we performed a time course analysis of bronchoalveolar lavages (BAL) from neonatal (5-7 day-old), juvenile (8-12 day-old), and adult (1-6 month-old) congenic C57BL6/N Scnn1b-Tg and wild-type (WT) littermates. The majority (13/15) of neonatal Scnn1b-Tg mice harbored bacteria (2.46 ± 0.33 log CFU/mouse), whereas WT littermates did not (n=14), linking inadequate mucus clearance to increased susceptibility to bacterial infection/colonization. Preliminary results suggest that these bacteria are part of the oropharyngeal flora. There was an inverse relationship between age and total CFU (juvenile, n=3/8, 0.94 ± 0.46 log CFU/mouse; adults, n=2/11, 0.29 ± 0.19 log CFU/mouse), indicating the ability of the mice to clear these bacterial species. CFU directly correlated with number of neutrophils/ml BAL fluid, which was highest in neonatal mice (182 ± 69 × 10 3 cells/ml BALF, n=8) and lower at older ages (40-60 × 10 3 cells/ml BALF, 1-6 months). We have previously reported that Scnn1b-Tg mice lacking the Toll-like receptor adapter molecule MyD88, had defective neutrophil recruitment. CFU quantification confirmed that neonatal MyD88-deficient Scnn1b-Tg mice had significantly more bacteria than Scnn1b-Tg mice (MyD88 -/-, Scnn1b-Tg n=11/11, 5.65 ± 0.39 vs. MyD88 +/-, Scnn1b-Tg, n=14/14, 3.80 ± 0.31 log CFU/mouse). MyD88 deficiency was associated with higher CFU in Scnn1b-Tg mice at all time points, but CFU did decline with age (adult MyD88 -/-, Scnn1b-Tg n=6/8, 2.86 ± 0.80 log CFU/mouse), indicating that MyD88-independent mechanisms aid in bacterial clearance. We also tested whether higher bacterial burden was responsible for the severe lung pathology observed in CFTR-deficient Scnn1b-Tg mice. Although not significantly, CFU counts were higher in ∆F508 homozygous as compared to ∆F508 heterozygous Scnn1b-Tg mice (3.87 ± 0.28 log CFU/mouse, n=8/8 vs. 2.84 ± 0.36 log CFU/mouse, n=13/15, respectively, p=0.068), suggesting that CFTR deficiency further impairs mucus clearance. Finally, to test the contribution of neonatal bacterial infection to the development of chronic lung pathology, we have derived Scnn1b-Tg mice in a germ-free (GF) environment. As expected, no bacteria were detected in BAL from GF neonatal mice (n=5), but surprisingly, adult GF Scnn1b-Tg mice exhibited airway inflammation and mucus obstruction comparable to those of mice reared under non-GF conditions. This finding indicates that the bacteria normally present in the lungs of neonatal Scnn1b-tg mice are not necessary for the development of lung disease, and that other triggers can drive the development of lung pathology due to airway surface dehydration. MUC5AC is a gel-forming mucin secreted by airway goblet cells which plays a key role in the viscoelastic properties of the mucus layer. Although mucus is necessary for airway clearance, epidemiological and pathological studies have suggested that excessive mucus secretion may be accountable for airway obstruction and/or inflammation. In obstructive pulmonary diseases such as cystic fibrosis (CF), MUC5AC expression is upregulated as a result of goblet cell metaplasia, which is believed to contribute to the CF pathogenesis. To determine whether increased Muc5ac production alone is sufficient to produce airway obstruction and inflammation, we generated a mouse model that overexpresses Muc5ac in the lungs using the Clara cell (rCCSP) promoter. Muc5ac cDNA was cloned from mouse lungs and tagged internally with a GFP for tracking purposes. The transgenic line was tested by quantitative PCR for transgene expression and showed a 20-fold mRNA increase in the lungs. Bronchoalveolar lavages were analyzed by agarose Western blots and revealed an 18-fold increased protein expression. In addition, Western blotting confirmed the proper maturation and polymerization of Muc5ac-GFP as depicted by the typical ladder-like migration pattern. Despite overexpression of a functional mucin, no incidence of lung pathology was revealed in transgenic animals. Survival curve was unchanged. Histopathology was similar to wild-type (WT) mice, with few AB/PAS-positive cells and no mucosubstance accumulated in the lumen. Immunohistochemistry revealed that GFP was mainly localized to the airways in domeshape-like cells, which correlates with the predicted CCSP expression. Inflammatory cell count was not increased in transgenic mice. The absence of pro-inflammatory phenotype led us to explore the response of these mice to influenza virus challenge. Thus, mice were instilled intranasally with 700 PFU PR8 (H1N1). Virus titer, lung histology and inflammation were monitored at day 2 and day 4 post-infection. Virus grew to titers 10-fold lower in Tg-Muc5ac than in WT littermate controls, suggesting a protective role of the mucin. No significant difference in inflammation was detected. To demonstrate that Muc5ac alone was responsible for the reduced infection, we purified Muc5ac-GFP by density gradient and presented the mucin to increasing concentrations of viruses for plaque assay measurements. The results showed decreased virus infection when incubated with Muc5ac-rich fractions, supporting the hypothesis of a direct interaction between mucin and viruses. In conclusion, we have demonstrated that the overexpressed Muc5ac-GFP was a high-molecular-weight protein, packaged in Clara cells and formed polymers upon secretion. The lack of discernable phenotype in the transgenic lungs suggests that mucin overproduction alone is not sufficient to trigger luminal mucus plugging, airway inflammation and goblet cell metaplasia. Conversely, this work demonstrates that Muc5ac acts as a shield to prevent viral infection, supporting a protective role for mucins in the absence of other triggering agents.

Supported by the CF Foundation, EHRE08I0. Model systems provide an excellent platform for studying mucociliary clearance, yet there is a trade-off between physiological relevance and the ability to measure fundamental aspects of the system. Cell culture models makes it possible to make measurements at the cellular level and understand those elements in the context of the larger clearance process. The access afforded by cell culture, however, is also a liability: in most culture systems the cultures are in a closed geometry and can't truly clear the airway surface liquid (ASL) that they generate. Moreover, the in vitro environment lacks the forces found in vivo; forces that have proven to play an essential role in cell phenotype and tissue behavior. We present a mucus clearance assay capable of the powerful analytical techniques used in cell culture while moving further towards physiological relevance. Our clearance assay grows cells in a bilayered fluidics channel, which allows us to perfuse the cells basolaterally while maintaining an air-liquid interface (ALI) on the apical surface. We have demonstrated the ability to grow well-differentiated, wellciliated cultures inside of our system. Furthermore the fluidics channel presents a geometry that enables the system to truly clear the ASL. Additionally we have integrated apical ports into our system. The ports facilitate several key features: (1) Exploring ciliary entrainment via the application of external forces, (2) Sampling and dosing the ASL, (3) Studying clearance when challenged by external flows e.g. airflow simulating tidal breathing. Finally we have built our system so that all of the aforementioned studies can be performed with high-resolution microscopy: we can investigate ciliary beat frequency and ASL properties on the cell length scale. Ultimately our system provides a platform to investigate the effects of several parameters (e.g. external forces, disease, pharmacological agents) on mucus clearance while retaining the ability to measure the underlying phenomena. Spectral domain optical coherence tomography (SD-OCT) is an emerging technique capable of providing reflectance images with subcellular resolution at video rates. Previously we demonstrated discrimination of respiratory epithelia, cilia, mucus, and gland anatomy in three dimensional images of formalin-fixed porcine and human airways. Subsequently, we have developed an SD-OCT system for video-rate microscopy with an axial resolution of 1 µm and a transverse resolution of 2 µm. Cross-sectional images are acquired at a speed of 32 frames per second with 512 axial lines (reflectance as a function of depth) per image.

In fully-differentiated primary human airway epithelial cells (HAE) derived from CF and non-CF individuals, airway surface liquid depth, cilia beat frequency (CBF), and mucociliary transport (MCT) were readily quantified. In comparison to WT epithelia, CF epithelia exhibited depleted ASL depth (3.6 ± 1.3 µm vs. 17.0 ± 3.9, P<0.01, n=8 filters/condition), reduced CBF (3.3 ± 0.2 Hz vs. 5.3 ± 0.3, P<0.001), and delayed MCT (0.06 ± 0.02 mm/min vs. 0.58 ± 0.24 mm/min, P<0.05), each imaged after 5 days of mucus accumulation. We tested the effect of CFTR modulation using SD-OCT imaging of HBE cells derived from a G551D/F508del donor. HBE were treated with forskolin (10 µM) + VX-770 (10 µM), VX-770 alone, Linux server (kernel version 2.6.18), using the Apache 2.0 web server, and PHP 5.2.6 for the server-side code. The FX.php library is used for communication between the PHP code and the database server. FileMaker Server 10.0 is used for the data, while images, PDFs, and other user-submitted files are stored on an ITS SPA-managed static media SAN called Collections which runs Windows Server 2003. In COLLARS, piglets are as listed as individuals with their unique identifications and biometric data. Treatments for any physiologic or infectious conditions, procedures performed with results from blood profiles, X-rays, CT scans, histopathology, and microbiologic sampling of the nasal cavity, lungs, and other tissues can be all recorded and saved. Blood and tissue specimens collected from pigs are cataloged for future use and reference. COLLARS has been tested on a number of platforms and browsers, and is known to work in the latest versions of Firefox, Google Chrome, Opera, Internet Explorer, and Safari. It has been tested under recent operating systems, including Windows XP/Vista/7 and Mac OS X. It has also been found to work on recent versions of the iPhone, as well as Android 2.0+-based mobile devices. In summary, COLLARS will provide a powerful informatics tool for CF investigators to track the clinical, microbiological, and pathological records associated with CF pigs.

Supported with funds from the Institute for Clinical and Translational Science (ICTS).

Joseloff, E. 1 ; Lee, D. 2 ; Wetmore, D. 1 ; Guo, L. 2 1. Cystic Fibrosis Foundation Therapeutics, Inc, Bethesda, MD, USA; 2. Metabolon, Durham, NC, USA Aim: Human bronchial epithelial (HBE) cells are currently being used as a pre-clinical model system to understand cystic fibrosis (CF) disease and for the development of investigational new drugs. In order to gain understanding of the epithelial dysfunction associated with CF mutations and delineate the mechanism of action of CFTR modulator compounds, untargeted metabolomic analysis was performed on cultured primary human airway epithelial cells isolated from explant lung tissue of CF and non-CF subjects treated in the presence and absence of compound. This study identified metabolic differences between CF and non-CF cells and assessed the effect of CFTR modulators in this model system.

Method: CF and non-CF HBE cells were grown on collagen membranes at the interface of air and liquid culture medium in the presence of two CFTR modulators and vehicle control for 15 and 30 hours. The cells were extracted for untargeted global metabolomic analysis using a combination of GC-MS, LC+MS/MS, and LC-MS/MS. The metabolites in the samples were identified using a proprietary software and library (Metabolon®). Statistical analysis between the treatment and control groups were performed.

Results: Significant differences in metabolism were found between CF and non-CF cells. The CF cells are impaired in several key metabolic pathways and nutrient update, such as decreased nucleotide levels, reduced energy metabolism, increased oxidative stress, and decreased cellular osmolytes. Treatment with CFTR modulators, particularly at 30 hours, restored some of the altered metabolites in CF cells toward their levels in non-CF cells.

Discussion: These results reveal novel key metabolic pathways that may be impaired in CF epithelia and identify a panel of potential biomarkers. The impact of the CFTR modulators on these altered metabolic pathways in CF suggests that metabolomics may be an avenue to investigate the mechanism of action of CF drug candidates to support drug discovery and development programs.

Supported by Cystic Fibrosis Foundation Therapeutics, Inc. The UK CF Gene Therapy Consortium will be undertaking a Phase IIB clinical trial of repeated application of liposome-based gene therapy over a one year period in approximately 100 CF patients (Multidose Trial). In preparation for this, we sought to address two key questions. Firstly, could we define the optimal set of patients in which the therapy could both be delivered (good access to the airways via nebulisation), and in whom any therapeutic effect was measurable (one or more abnormal measures of lung disease). Secondly, in this set of "can deliver -can measure" patients, which biomarker(s) could be powered to be the primary outcome measure for the trial. To address both questions, we undertook a study (Run-in), cross-sectionally assessing "can deliver" and longitudinally assessing a large set of candidate biomarkers for "can measure." One hundred ninety-two patients from age 10 upwards, with FEV1 >40% were enrolled at two clinical centres; 154 of these remained in the study after four visits spaced at approximately 4-5 month intervals. Biomarkers assessed cross-sectionally included radionucleotide deposition scans, CT and mucocilary clearance. Longitudinal biomarkers included a large series of serum, sputum and exhaled breath inflammatory markers, lung physiology, exercise-related assays and quality of life assessment. Twelve patients were judged too severe for adequate delivery and were excluded. A shortlist of 4 biomarkers was generated based on a) showing a CF/non-CF difference, b) response to course of intravenous antibiotics, and c) coefficients of variation. These four were matched against the remaining 142 patients, and a further 7 patients excluded in whom none of these shortlisted biomarkers was abnormal. Eighty-nine patients (3 or 4 biomarkers abnormal) have been definitely included to progress into the Multidose Trial, and a further 46 (1 or 2 biomarkers abnormal) are awaiting the final primary outcome selection. The Run-in study has, therefore, been able to a) select a cohort of "optimal" patients in which to assess gene therapy, and b) provide an indication of which may be the more useful biomarkers to use in phase IIB clinical trials of novel therapeutic agents.

Funded by the UK Cystic Fibrosis Trust. F508del CFTR is the most common cause of cystic fibrosis, found in approximately 70% of CFTR alleles and 90% of CF patients. Development of small molecules capable of correcting defective F508del CFTR maturation, gating, and plasma membrane half-life could be of potential clinical benefit to the majority of CF patients. In the current study, 89 adult CF subjects (median age: 26 years, gender: 60% male) homozygous for the F508del CFTR mutation were each randomized to one of four VX-809 dose groups or matching placebo to evaluate the safety, tolerability and CFTR activity of once daily VX-809 compared with placebo for 28 days. Dose levels evaluated included 25 mg, 50 mg, 100 mg, and 200 mg, and the primary measure of CFTR activity was sweat chloride (Wescor macroduct). Optional nasal potential difference (NPD) and rectal biopsy to detect C Band CFTR by Western blot were each performed in a subset of subjects (75% and 37%, respectively). In addition, lung function (FEV 1 , FVC, FEF 25%-75% ), and Cystic Fibrosis-Questionnaire-Revised (CFQ-R) scores were measured. Safety was assessed by adverse events, physical exams, standard laboratory testing, ECGs, and vital signs. The study was performed in a two part, dose-ascending format with two blinded, independent interim assessments for safety. Safety data were analyzed using descriptive statistics only, and efficacy data using an analysis of covariance model. All randomized subjects who received at least one dose of study drug were included in the safety and pharmacodynamic analyses. VX-809 was generally well-tolerated across the four dose groups compared to placebo. Overall, the type and rate of AEs were similar among treatment and placebo groups and respiratory events were the most common AEs reported. AEs leading to withdrawal were seen in one subject in each active treatment arm and were related to respiratory symptoms. Pulmonary exacerbations were seen in 17% of VX-809 subjects and 12% of placebo subjects (p=0.62). Sweat Clvalues were reduced in a dose-dependent manner (p=0.0013) during the period of treatment. The treatment difference between the placebo response at day 28 in the 25-mg, 50-mg, 100-mg, and 200-mg groups were +0.10, -4.61, -6.13, and -8.21, respectively. These differences reached statistical significance for the 100mg (p<0.05) and 200-mg (p<0.01) groups. Sweat Clvalues returned to pretreatment values following 7 days of drug washout. Lung function, CFQ-R and NPD parameters did not change significantly during the study in any dose groups, and no C band CFTR was detected in any subjects. The results provide evidence of the tolerability and biologic activity of VX-809 in F508del CFTR homozygous patients, and provide support for future clinical trials.

Sponsored by Vertex Pharmaceuticals Inc.

impaired lung function and on limited background pharmacotherapy in a Phase 3, placebo-controlled clinical trial (TIGER-1) with a 24-week doubleblind treatment phase followed by a 24-week open-label phase.

The study enrolled patients ≥5 years old with FEV 1 ≥75% predicted. Data from the subgroup of patients on 0 to 2 concurrent pharmacotherapies (n=37 denufosol, n=34 placebo) were analyzed.

Results: In the subgroup analysis, % predicted FEV 1 at baseline was 93% in the denufosol group and 91% in the placebo group. During doubleblind treatment, FEV 1 improved with denufosol over placebo by 5.7% (p = 0.028) or 100 ml (p = 0.059) with a net improvement of 6.4% over placebo in % predicted FEV 1 (p = 0.011). Patients on 48 weeks of denufosol improved versus baseline by 12% (183 ml) for FEV 1 and 4% for % predicted FEV 1 . These improvements exceeded those in the intent-to-treat population, in which >50% of patients used ≥4 pharmacotherapies. Denufosol was well tolerated with a safety profile comparable to that of placebo.

Conclusions: Patients on minimal background pharmacotherapy and with minimally impaired baseline lung function (93 % predicted FEV 1 ) showed significant improvement with denufosol versus placebo and versus baseline in a 24-week, double-blind trial with a 24-week open-label extension. This evidence supports denufosol's potential as an early intervention treatment of CF lung disease.

Supported by Inspire Pharmaceuticals. Objective: Arikace™ is a sustained-release lipid formulation of amikacin for inhalation, being developed for the treatment of chronic P. aeruginosa (Pa) infection in CF patients (pts). The primary objective of the Phase 2 program was to evaluate the safety, tolerability, pharmacokinetics (PK) and efficacy of 28 day (d) exposure to once daily Arikace™.

Methods: CF pts were enrolled in two similar dose-ranging Phase 2 studies conducted at 18 US and 15 European centers. Pts were stratified by FEV1% predicted at baseline and randomized to receive either Arikace™ or placebo (1.5% NaCl). Four dose cohorts comprising 70 mg, 140 mg, 280 mg and 560 mg were evaluated. Pts received study drug once daily for 28 d by inhalation with eFlow® nebulizer system (PARI Pharma GmbH) and were followed for 28 d off study drug. Safety, PK, lung function, Pa sputum density, Patient Reported Outcome (CFQR) and time to pulmonary exacerbation were evaluated weekly during the study period of 56 d. DSMB review of interim safety and efficacy data resulted in stopping enrollment in the two lower groups, and the US study was amended to extend evaluation of the 560 mg dose for a total study period of 84 days.

Results: One hundred five pts were enrolled: 19 adults in the 70 mg and 140 mg groups; 21 pts in the 280 mg group; 36 pts in the 560 mg and 34 pts in the combined placebo group. Mean baseline FEV1% predicted was 67% (FEV1% >75% in 32% of pts; FEV1% <=75% in 68% of pts). Mean age was 21 yr (range= 6-68 yr); and ~85% of pts grew Pa (mucoid phenotype). Acute tolerability was similar across the placebo and treatment groups, and adverse effects were consistent with those expected in a population of CF pts receiving inhalation medicines. Twenty-eight days after stopping drug, pts receiving 560 mg Arikace™ showed a ~13.0% relative improvement in FEV1 (p=0.003), with treatment effects of 14.6% and 8.5% for subjects with FEV1%<=75% and >75%, respectively. Reduction in Pa density, including mucoid strains was statistically superior in the 560 mg treatment group compared to placebo (p=0.007). Also observed was clinically significant improvement in respiratory symptoms of CFQR-Respiratory Scale (67% on Arikace™ improving versus 36% on placebo, p=0.006).

Conclusion: Arikace™ given once daily for 28 d at doses of 70 mg, 140 mg, 280 mg, and 560 mg was well tolerated in subjects with CF. The Arikace™ arms vs placebo group demonstrate dose dependent treatment 3. Nemours Childrens Clinic, Orlando, FL, USA; 4. Massachusetts General Hospital, Boston, MA, USA; 5. Maine Medical Group, Portland, ME, USA; 6. Pharmaxis Ltd, Sydney, NSW, Australia Background: Airway clearance strategies are an essential component of a daily regimen for CF patients in order to maintain lung health. Inhaled dry powder Mannitol is an osmotic agent that increases water content of the airway surface liquid and improves mucociliary and cough clearance in patients with CF (1). A 6-month double blind (DB) phase III trial of inhaled Mannitol (400 mg bid) in 295 CF patients aged 6 -56 years, mean 23 ± SD 11.3 years demonstrated a sustained improvement in FEV 1 of 6.5% at 26 weeks from baseline (2, 3) . A similar improvement in FEV 1 was seen in the subset of patients on concurrent rhDNase, (5.1% change compared to control, p=0.006) (2, 3) . This sister trial was conducted in the U.S., Canada, Argentina and Europe to further assess the efficacy and safety of inhaled Mannitol.

Methods: The study was a randomized, DB, parallel, controlled design where patients inhaled 400 mg dry powder Mannitol (40 mg capsules x 10) or control, twice daily for 6 months, with an optional additional open label phase continuing for 6 months. Patients had confirmed CF, were ≥ 6 years with an FEV 1 ≥ 40 and < 90% predicted, tolerant to mannitol and were not using hypertonic saline.

Demographics: Preliminary baseline data were available for the 305 randomized patients. Patients were aged 6-53 (mean 20.0 yrs) with 48.5% female and mean BMI of 20.0 kg/m 2 . The baseline % predicted FEV 1 was 63.8 ± SD 15.9 and for the subset of 229 patients concurrently taking rhD-Nase was 63.2 ± SD 15.8.

Study Objectives: The primary endpoint was change in FEV 1 . Secondary endpoints included pulmonary exacerbations, quality of life and antibiotic treatments.

Results: This 6-month DB study completed in April 2010. The database lock is set for June 2010 and results will be presented for the primary and secondary endpoints. This will be the first presentation of the U.S., Canada, Argentina and Europe study results. Objective: Patients with cystic fibrosis (CF) often develop detectable structural airway damage before reaching 5 years of age. Disease modifying therapies have the potential to prevent or reduce such damage when used as early intervention. Denufosol, a novel ion channel regulator, stimulates Cltransport, inhibits Na + absorption, and increases ciliary beat frequency independently of CFTR genotype. The potential of denufosol as an early intervention agent was assessed in a subgroup of patients with minimally effects, with the 280 mg dose appearing to be the minimum effective dose; and the 560 mg dose showing statistically significant improvements in clinical symptoms, microbiologic activity, and sustained improvement in lung function for up to 28 days post-study drug. These data warrant confirmation of safety and efficacy of Arikace™ in Phase 3 trials of CF patients chronically infected with Pa. The methodology for measuring mucociliary (MCC) and cough clearance (CC) needs to be standardized between CF centers to allow for its future use as an effective non-invasive, outcome measure of CF therapeutics. A standardization protocol has been developed at the University of North Carolina at Chapel Hill, NC (UNC), Johns Hopkins University (JHU), and the University of Pittsburgh (UP).

Methods: Healthy adult subjects inhaled an aerosol (mass median aerodynamic diameter = 5.0 µm) generated by a Devilbiss #646 nebulizer from a solution containing the radioisotope 99mtechnetium sulfur colloid in 0.9% saline. Subjects inhaled the radiolabeled aerosol at 500 ml/sec by following a metronome and flow signal from a commercial dosimeter. Following inhalation, subjects sat in front of a gamma camera, as sequential lung images were acquired for 60 min. Subjects then coughed 60 times and images were acquired after each set of 20 coughs, i.e. at 70, 80 and 90 min. Subjects returned to the laboratory 24 hours later for a final image of residual lung activity. Initial aerosol distribution was measured as a central/peripheral (C/P) ratio of activity. MCC for the right lung was expressed as the average of % clearance over the 60 min (Ave60Clr). MCC/CC was expressed as the average of % clearance over the 90 min (Ave90Clr).

Results: A multivariate analysis of clearance vs. time with site and C/P as covariates showed no significant site-specific differences. Ave60Clr (± standard deviation) was similar at the three sites with 8±8%, 11±6% and 10±9% at UNC (n=30), JHU (n=10) and UP (n=10), respectively. Ave90Clr was also similar with 11±9%, 16±8% and 14±10% at UNC, JHU and UP, respectively. Mean clearance at 24 hours was 39±14%, 36±18% and 45±25% at UNC, JHU and UP, respectively. The initial aerosol distribution was similar at all three sites with C/P ratios averaging 1.63±0.43, 1.81±0.62 and 2.04±0.55 at UNC, JHU and UP, respectively. Interestingly, MCC and MCC/CC was greater in females (n=19) vs. males (n=31), with Ave60Clr =13±10% and 7±5%, respectively (p=0.01) and Ave90Clr 16±11% and 10±7%, respectively (p=0.03). There were no gender differences for either C/P ratio or 24-hour clearance.

Conclusions: Using these standardized methods, we found that MCC and MCC/CC were similar in healthy adult subjects studied at the three sites. In addition, clearance was significantly faster in females over the 60 and 90 min periods suggesting a gender difference in large airway clearance. Though the reason for this gender difference is not clear, it has implications for the design of future studies, e.g. the number of female and male subjects should be matched in all study cohorts. This standardization protocol may prove beneficial in multi-center trials for testing new therapies that are designed to improve MCC and MCC/CC, allowing more subjects to be studied in a shorter period of time and thus bringing new therapies to patients more rapidly.

This study was funded by the Cystic Fibrosis Foundation. Background: Validated biomarkers are urgently needed to advance care and research in cystic fibrosis (CF) lung disease. Most validated biomarkers in use are circulating proteins. However, due to the technological challenges of profiling the circulating proteome, particularly inaccessibility of low abundance proteins, the discovery of new protein biomarkers has been limited. Proteomic analysis from the airways should reflect pathways important to lung disease. SOMAmers are specific high affinity nucleic acids-based reagents selected to recognize human protein targets that provide access to the low abundance proteome developed to study the circulating proteome.

Here we show this analytical platform is capable of measuring protein levels in bronchoalveolar lavage fluid (BALF).

Methods: BALF was collected for clinical indications in patients with CF (n=5), Surfactant dysfunction mutations (SDM; n=5), Neuroendocrine Cell Hyperplasia of Infancy (NEHI; n=5), and disease controls (DC; n=8). A pilot project was performed to determine if a high-throughput multiplexed SOMAmer-based proteomics platform, measuring 811 proteins, could be applied to BALF. An analysis of variance model was fit to each of the 811 proteins individually to test the association across disease groups. The molecular function ontology for the significant proteins was investigated for over-represented groups. Principal component analysis was used to reduce the dimensionality of the protein profile for the 5 CF patients.

Results: There were 393 proteins significantly increased in the CF group compared to DC (p< 0.01). The 457 proteins that were significantly different in CF from the other disease groups were over-representative for signaling molecule and growth factor molecular functions. Proteins commonly studied in CF (including IL-6, IL-8, TIMP-1 and elastase) are distinguishable across disease groups, however, they are ranked below other proteins that better distinguish CF. The 811 proteins were reduced to the first two principal components, which captured 80% of the variability in the protein profiles across the CF patient samples. Based on this information, 2 of the 5 CF patients had a distinctive protein profile from the other 3. Interestingly, the two divergent samples, with lower protein levels, did not grow bacterial pathogens. Determining whether this difference in protein profiles is clinically meaningful, such as predicting future FEV decline would be very valuable.

Conclusion: The pathogenesis of CF is dramatically different from other pediatric lung diseases and even within CF, as determined by profiles of 811 proteins. This technology expands the pool of candidate biomarkers which may prove to be more useful than commonly measured proteins. These data may provide insight into the variations of disease prognosis and suggest novel therapeutic targets. INSA, Lisboa, Portugal; 2. DQB, FCUL, Lisboa, Portugal; 3. FCM, Unicamp, Campinas, Brazil; 4. IP, UR, Regensburg, Germany Recent high-throughput screens identified several novel small molecules with potential to treat the basic defect in CF. These include correctors (like VRT-325 & C4) that partially rescue the trafficking defect of F508del-CFTR and potentiators (like VRT-532 & VX-770) that correct the gating defect of G551D-CFTR [1] [2] [3] . However, the action and efficacy of these compounds on native human tissues remains to be determined.

We recently screened a 250 small-molecule library by quantitative Western blot in CFBE cells and F508del-NBD1 folding by differential scanning fluorimetry [4] . Among the 4 primary lead "hit" corrector compounds is TS-01-02-A3.

Analyses were performed on pooled data from Parts 1 and 2 of the trial, at baseline (pre-treatment) and study Day 14. The detection of VX-770 effects on Clconductance was generally similar across all analyses. The best nostril response identified pre-treatment analysis demonstrated significant within-subject and vs placebo changes in Cltransport in the 75, 150, and 250mg groups (P<0.05), whereas changes from baseline using the best nostril response per visit and the mean of nostril response analyses were significant within-group only (P<0.05 for the 75, 150, and 250-mg groups). In general, the predicted pharmacologic effect of VX-770 on Na + conductance was most evident when data were analyzed using the best nostril response per visit, then the best nostril response identified pre-treatment, followed by the mean nostril response. Dose-dependent improvements in maximal basal PD were observed only in the best nostril response per visit analysis (P<0.05 within group and vs placebo for the 75, 150, and 250-mg groups). Based on these preliminary analyses examining the best nostril response per visit, treatment arms of <20 subjects should have 85% power to detect withingroup changes in zero Clplus isoproterenol -4 mV (SD=4.0), 90% power to detect 10 mV changes in mean basal PD (SD=9.0), and 90% power to detect 15 mV changes in maximal basal PD and Ringer's PD (SD=13). This analysis may provide more power than traditional NPD analyses based on mean values per subject, in which 24 subjects would be required for detection of -4 mV improvement (SD=4.5). These results may help clarify what aspects of the NPD measurement are best matched to particular analytic approaches.

This study was sponsored by Vertex Pharmaceuticals Inc. with support from CFFT. Introduction: New treatment options for inhaled antimicrobials are needed for CF patients to reduce the morbidity caused by chronic airway infections. This study assessed the efficacy and safety of MP-376 (Aeroquin), a novel levofloxacin formulation for inhalation, in CF patients with extensive previous use of inhaled antimicrobials for chronic PA lung infection.

Methods: Randomized, double-blind, placebo controlled trial of 3 dose groups of MP-376 (120 mg QD, 240 mg QD, 240 mg BID) vs. placebo for 28 days, delivered by a customized investigational PARI eFlow nebulizer, with an additional 28 days of follow up. Inclusion criteria: age >/= 16 years, chronic PA airways infection, FEV1 between 25-85% predicted, and >/= 3 courses of inhaled antibiotics over the past 12 months. The protocol-defined endpoint of needing an anti-pseudomonal antimicrobial was satisfied if a patient had at least one of 4 symptoms commonly seen with exacerbations at the time of initiation of anti-pseudomonal antimicrobials. Results are presented for the MITT patient population.

Results: One hundred fifty-one patients were enrolled with mean baseline characteristics including age 29 years, FEV1 52% of predicted, and 4.8 courses of inhaled antibiotics over the prior 12 months. Concomitant respiratory medications included dornase-alfa (78%), azithromycin (74%), and hypertonic saline (46%).

MP-376 treatment was associated with a reduction in the need for additional anti-pseudomonal antimicrobials during the treatment and follow up periods compared to placebo. Hazard ratios estimated using Cox proportional modeling showed a statistically significant reduction in the need for antipseudomonal antimicrobials in all 3 MP-376 groups vs. placebo (120 mg QD HR 0.29, P=0.007; 240 mg QD HR 0.39; p=0.021; 240 mg BID HR 0.21; p=0.0007), with the greatest effect seen in the 240 mg BID group.

Here we performed ion transport studies on native colonic biopsies collected from CF patients and mounted on Ussing chambers. Our aim is twofold: to quantify CFTR-mediated Clsecretion in rectal biopsies from CF patients with different CFTR genotypes for clinical diagnosis and prognosis; and to assess the efficacy of the above correctors directly in human native tissues ex vivo.

CFTR cAMP-mediated Clsecretion was measured in freshly excised rectal biopsies from 68 individuals, as previously [5] [6] , allowing their classification into 3 groups. In the non-CF individuals group (n=31), we observed cAMP-dependent Clsecretion (∆Isc=-92.7 ± 26.7 µA/cm 2 ) upon stimulation with IBMX/Forskolin (I/F) and co-colinergic activation with carbachol (CCH). In contrast, in one group (I) of CF patients (n=29), the cAMP-induced Clsecretion was absent and lumen-positive responses (∆Isc=32.3 ± 5.3 µA/cm 2 ), reflecting K + secretion, were detected. In another group (II) of CF patients with milder clinical phenotypes (n=8), residual Clsecretory responses were detected upon I/F/CCH stimulation (∆Isc=-19.1 ± 4.2 µA/cm 2 ).

The same biopsies used to confirm/exclude a CF diagnosis, were used to test the effects of 6.7 µM VRT-325, 10 µM C4 and 10 µM TS-01-02-A3 after 8 h incubation. Preliminary results in biopsies from F508del/F508del patients did not evidence a clear effect for VRT-325 and C4 correctors, but showed a modest effect for TS-01-02-A3. Ongoing work involves testing of efficacy of the above primary lead correctors on native tissues with longer incubation times (up to 16 h) in parallel with primary cultures of human airway cells.

These results help to further establish rectal biopsy measurements (namely CFTR-mediated Clsecretion) as a sensitive method for the diagnosis and prognosis of CF disease. Moreover, this approach can also identify CF patients with residual CFTR activity who can thus benefit from potentiators therapy.

Work supported by TargetScreen2 The recent phase IIa study of VX-770, a CFTR potentiator, in subjects with G551D-CFTR cystic fibrosis demonstrated improvements in NPD, sweat Cl -, and lung function during drug treatment compared with placebo. This dataset afforded the opportunity to re-analyze specific NPD parameters in order to identify the optimal outcome measurements for future clinical trials. NPD measurements performed included a parameter of Clion transport, the zero Clplus isoproterenol response; as well as parameters of Na + ion conductance, such as basal PD (average of five PD measurements within the inferior meatus), maximal (most-polarizing) basal PD, and Ringer's PD (obtained following 1 minute of Ringer's perfusion). Pre-specified NPD analyses were based on changes from baseline using mean values from both nostrils (per subject). Data were reanalyzed post-hoc to evaluate outcomes on the basis of the best nostril response per visit (defined as the most polarized NPD value for each individual measure), or the best nostril response identified pre-treatment (at baseline) carried forward throughout the trial.

At day 28, patients in the 240 mg BID group also had the greatest improvement vs. placebo in lung function (relative change in % predicted FEV1 10.94, P = 0.0008, percent change in FEF25-75 22.30, P = < 0.0001) and the greatest reduction in log 10 sputum PA density (0.96, P = 0.0014). The percent of patients with AEs was similar across all groups, with no evidence of increasing incidence or severity of AEs with increasing MP-376 dose.

Conclusion: Nebulized MP-376 (Aeroquin) was well tolerated and demonstrated statistically significant reductions in the need for antipseudomonal antimicrobials and improvement in other efficacy endpoints compared to placebo in this heavily-treated CF patient population. Phase 3 studies are planned. Background: FTI is an inhaled antibiotic with activity against Gram positive, Gram negative, and anaerobic pathogens. AZLI is bactericidal against Gram negative pathogens and substantially increases FEV 1 during treatment, but FEV 1 returns toward (pre-AZLI) baseline after therapy is withdrawn. To test whether the broad-spectrum antibiotic FTI could maintain improvements in FEV 1 achieved with AZLI, this study enrolled subjects with CF age ≥ 18 years with pulmonary PA and FEV 1 ≥ 25% and ≤ 75% predicted. Following 28 days of AZLI, subjects received FTI or placebo administered via the Investigational eFlow ® Nebulizer. The primary endpoint was relative change in FEV 1 % predicted from Day 0 to Day 28 for FTI 160/40 mg compared to placebo; other endpoints were change in the Cystic Fibrosis Questionnaire-Revised (CFQ-R) Respiratory Symptoms scale and log 10 PA CFUs.

Results: One hundred thirty-five subjects with CF from 33 US centers received open-label AZLI; 119 were then randomized to FTI or placebo. Subjects were 57% male, had a mean age of 32 years, and mean FEV 1 of 49% predicted prior to AZLI. Results are summarized in the table below.

Conclusion: Both FTI doses maintained the significant improvement in FEV 1 % predicted achieved with AZLI, while lung function in placebo subjects returned toward pre-AZLI levels. CFQ-R respiratory symptoms scores declined only slightly in subjects receiving FTI 80/20 mg, with a larger decline in the FTI 160/40 mg group likely reflecting decreased airway tolerability of the higher dose. FTI 80/20 mg was superior to placebo in further reducing log 10 PA CFUs, with the lower FTI dose being more effective than the 160/40 mg dose. Background: To test whether the broad-spectrum combination antibiotic FTI could safely maintain improvements in FEV 1 following an effective inhaled anti-pseudomonal antibiotic (AZLI), this study enrolled subjects with CF age ≥ 18 years with pulmonary PA and FEV 1 ≥ 25% and ≤ 75% predicted. FTI was administered using the Investigational eFlow ® Nebulizer. Safety endpoints included clinical and laboratory adverse events. Airway reactivity (defined as an acute decrease in FEV 1 [L] of ≥ 15%) was assessed 30 minutes post-dose on Days 0, 14 and 28.

Results: One hundred thirty-five subjects with CF from 33 U.S. centers received open-label AZLI; 119 were then randomized to FTI or placebo. Subjects were 57% male, had a mean age of 32 years, and mean FEV 1 of 49% predicted prior to AZLI. Adverse events (AEs) were consistent with CF lung disease, primarily cough (see table) . Respiratory AEs, including dyspnea and wheezing, were lower in the FTI 80/20 mg group compared to the 160/40 mg group. One subject in the placebo group had an acute FEV 1 [L] decrease ≥ 15% compared to 2 subjects in the FTI 160/40 mg group and none in the FTI 80/20 group. Serious AEs and hospitalizations were similar across treatment groups. No clinically significant changes in laboratory values were reported.

Conclusion: FTI was generally safe and well-tolerated in subjects with CF. Subjects in the FTI 80/20 mg group were more likely to complete treatment and reported fewer respiratory adverse events than those in the FTI 160/40 mg or placebo groups.

Acknowledgements: Submitted for the FTI Phase 2 Study Investigators. Supported by Gilead Sciences. a Listed events are those occurring in >20% of subjects in either FTI group renal clearance of amiloride together with formulation considerations, limited the dose of drug that could be administered in these studies. Novel, high potency ENaC blockers, designed specifically for inhaled formulation and that avoid renal clearance and thus hyperkalaemia will enable the role of ENaC in CF lung disease to be further tested with a view to therapeutic intervention.

A series of potent amiloride containing dimers were designed with properties consistent with the desired pharmacokinetic and physicochemical properties for an inhaled ENaC blocker. An example, Compound 1, inhibited the amiloride-sensitive short circuit current (ISC) in primary human bronchial epithelial cells (HBE) and Fisher Rat Thyroid cells transiently expressing either rat or guinea pig ENaC (αβγ) with IC50 values of 8, 14 and 22 nM respectively. Incubation of Compound 1 (10x IC50) on the apical side of HBE for 10 minutes resulted in a sustained block of ISC on washout (>90% inhibition), that was in contrast to a full recovery of ISC in amiloride (10x IC50) treated cells. The potent monomeric amiloride derivative, 552-02 (10x IC50), showed a partial recovery of ISC (<50% residual inhibition) in this assay.

Intra-tracheal dosing of Compound 1 to anesthetized guinea pigs induced an attenuation of the tracheal potential difference measured at both 1 and 4 hours after dosing with ED50 values of 1.3 and 388 nmole/kg. Compound 552-02 showed a similar potency for inhibition of airway ENaC in this model with ED50 values of 0.3 and 244 nmole/kg at 1 and 4 hours after dosing respectively. Blood K+ levels were significantly elevated by 552-02 at doses ≥114 nmole/kg. Blood K+ levels were not elevated by Compound 1 at doses up to and including 2239 nmole/kg, the highest dose studied. Intra-tracheal dosing of amiloride or 552-02 to rats likewise induced hyperkalaemia that was in contrast to Compound 1. Compound 1 showed a 20fold reduction in renal clearance in rat compared with 552-02 consistent with the reduced potential for renal-based side effects.

Dimeric amiloride-derivatives also enhanced MCC in sheep at doses that could be formulated as dry powder lactose blends. Inhalation of crystalline compounds using a breath-actuated inhaler device enhanced the rate and magnitude of 99mTc whole lung clearance. The efficacy and duration of action observed with these compounds in the sheep model was superior to that observed with inhaled hypertonic saline, a therapeutic intervention demonstrated to improve lung function and to reduce exacerbation frequency in CF patients.

Together, these pre-clinical data highlight the feasibility of designing potent ENaC blocker compounds that enhance MCC but that can minimize renal-based side effects upon the inevitable absorption of drug from the lung. Background: Measurement of mucociliary and cough clearance (MCC/CC) by gamma scintigraphy following inhalation of a radiotracer has been used as a biomarker in CF. Administration of inhaled hypertonic saline (HS) to CF patients results in acute and sustained improvements in MCC (Donaldson, et al 2006) , while long-term administration of HS is associated with improvements in lung function and pulmonary exacerbations (Elkins, et al 2006) . Nevertheless, the duration of action for a single dose of HS remains unknown. This knowledge would be useful in the development of novel inhaled therapies designed to achieve similar clinical benefits as HS. We hypothesize that MCC/CC can be used to determine the duration of action of agents like HS, and is an excellent choice as a Proof of Concept endpoint for exploratory clinical studies in the development of new inhaled therapies for CF.

Objectives of the Study: To assess the duration of action on MCC/CC out to 4 hours for a single dose of inhaled HS (7%) in adult CF patients.

Methods: In this ongoing study, 16 CF patients from 2 institutions, age 18 or older with an FEV 1 ≥50%, will be studied. Three MCC/CC studies are performed on separate visits: one at baseline, one immediately after HS (3:2) to TIP (112 mg tobramycin/4 capsules) via the T-326 dry powder inhaler or TOBI® (300 mg tobramycin/5 mL) via PARI LC® PLUS nebuliser twice-daily for 3 treatment cycles. Each treatment cycle consisted of 28 days on-drug followed by 28 days off-drug. Safety and efficacy outcome measures and treatment satisfaction were evaluated.

Results: A total of 517 patients (TIP=308, TOBI®=209; 55.3% male; mean age 25.6 y, mean baseline FEV1% predicted 53.0) received the study drug. Eighty-eight percent of patients experienced adverse events (AEs) that were mostly mild or moderate in severity. Treatment-related AEs more commonly found with TIP than TOBI® were cough (25.3 vs 4.3%), dysgeusia (3.9 vs 0.5%) and dysphonia (12.7 vs 3.3%). A similar proportion of patients in TIP (27.4%) and TOBI® (29.2%) groups experienced serious adverse events (SAEs). The most common SAE was lung disorders (pulmonary/CF exacerbations) (TIP: 19.5%; TOBI®: 18.7%). The frequency of AEs and SAEs decreased with each successive treatment cycle. There was no significant difference in the change in FEV1% predicted from baseline to end of Cycle 3 between TIP and TOBI® [least square mean difference: 1.1%; lower bound of 1-sided 85% confidence interval = -0.67] confirming the non-inferiority hypothesis. The mean change in sputum density of Pa biotypes (log 10 CFU/g) was comparable between treatments across the 3 cycles and greater with TIP than TOBI® on Day 28 of Cycle 3 (mucoid: -1.6 vs -0.92; dry: -1.77 vs -0.73). On Day 28 of Cycle 3, 11.6% TIP and 9.9% TOBI® patients had negative Pa cultures. The distribution of tobramycin MIC of Pa isolates was relatively consistent over the course of study. The number of patients with MIC values >8 µg/ml for all Pa biotypes in TIP and TOBI® groups at the end of Cycle 3 was 19.1% and 13.4%, respectively. Treatment satisfaction, assessed by the Treatment Satisfaction Questionnaire for Medication (TSQM), was significantly higher for TIP than TOBI® at all visits for effectiveness (P<0.0001), convenience (P<0.0001) and global satisfaction (P=0.0018), while there were no statistically significant differences in side effects ratings. The overall administration time for TIP (5.6 min) was 14.1 min less than TOBI® (19.7 min).

Conclusion: The overall TIP safety profile was similar to TOBI®, except for some expected local tolerability effects which may be related to the relatively high amount of powder with TIP. FEV1 response was similar, while the microbiologic response was better for TIP. CF patients rated TIP more convenient and satisfying than TOBI®, which may be of benefit in clinical practice.

Study supported by Novartis Pharma AG, Basel, Switzerland. Blockers of the epithelial sodium channel, ENaC, represent a therapeutic opportunity for the treatment of CF lung disease. Inhaled amiloride, a low potency ENaC blocker, enhanced mucociliary clearance (MCC) in the CF lung but showed equivocal clinical efficacy. The mechanism-based side effect of hyperkalaemia, caused by ENaC blockade in the kidney following treatment (0 hr), and one at 1 or 4 hours post-treatment, according to a randomization schedule. A 72-hour wash-out period from HS and DNase use is required. Treatment consists of 4 mL of 7% HS from a Pari LC Star nebulizer. Clearance of radiotracer is analyzed over a 90-minute period (Ave90Clear) to assess large airway effects, and through 24 hours (24hrClear) to assess combined large and small airway clearance effects. An absolute increase of ≥5% in either Ave90Clr or 24HrClr is considered a meaningful response.

Results: Eleven subjects have completed the study (55% female, median age 28 years, mean FEV 1 82% of predicted). Forty-five percent of subjects are regular HS users. Four of 11 CF subjects had enhancement of Ave90Clear immediately following a single dose of 7% HS. Of those who had a response, all had evidence of persistent augmentation of Ave90Clear at their later randomized time point (1 or 4 hours) . None of the patients who failed to respond to 7% HS with an immediate enhancement in Ave90Clear, had an increase at 1 or 4 hours post-HS dosing. The improvement in clearance was more evident for 24hrClear, with 6/11 patients having enhanced 24hrClear immediately and at their 1 or 4 hour post HS dosing (p≤0.05 for both 1 and 4 hour delays vs. baseline).

Conclusions: In response to a single dose of inhaled 7% HS, there appears to be a subset of patients with a robust increase in MCC/CC immediately post-dose. In those responders, MCC/CC was also augmented at 1 and 4 hours post dose, suggesting a pharmacodynamic response in each of these subjects lasting at least 4 hours. Nevertheless, there is still a group of patients who demonstrate no significant change in MCC/CC in response to HS. These patients may benefit from novel therapies designed to improve MCC/CC by mechanisms different than HS.

Supported by Novartis and the CF Foundation. 

Dehydration of airway surfaces and reduced mucus clearance due to increased airway Na + absorption via epithelial Na + channels (ENaC) plays an important role in the pathogenesis of cystic fibrosis (CF) lung disease in patients and causes CF-like lung disease in βENaC-transgenic (βENaC-Tg) mice. Hypertonic saline (HS, NaCl 7%) improves airway surface hydration by inducing osmotic flux of water into the airway lumen. Previous clinical trials demonstrated that inhalation of HS improved mucus clearance and lung function and reduced pulmonary exacerbations in CF patients with chronic lung disease (Donaldson SH et al., N Engl J Med., 2006; Elkins MR et al., N Engl J Med., 2006) , and recent pilot studies demonstrated that inhalation of HS is well tolerated by infants and young children with CF (Dellon EP et al., Pediatr Pulmonol., 2008) . However, effects of HS therapy on other characteristic features of CF lung disease, i.e. airway mucus obstruction and airway inflammation, and the benefits of preventive HS treatment have not been studied. In the present study, we used βENaC-Tg mice as a murine model of CF lung disease, and determined the effects of preventive and late HS therapy on mortality, airway mucus obstruction and airway inflammation. Newborn or 4-week old βENaC-Tg mice and wildtype littermate controls were treated by intrapulmonary instillation of HS (NaCl 7%; 1µl/g body weight) or vehicle (H 2 O) alone three times per day for a period of 2 weeks. During HS therapy, growth and survival were monitored. After the 14 day treatment period, mice were euthanized, bronchoalveolar lavage (BAL) was performed to determine inflammatory cell counts, and lungs were processed for histology and morphometry to quantitate airway mucus obstruction. We demonstrate that preventive HS therapy significantly reduced pulmonary mortality, improved growth, and reduced airway mucus content in HS-treated compared to vehicle-treated βENaC-Tg neonates. In adult βENaC-Tg mice with chronic lung disease, HS also conferred significant reduction in airway mucus content compared to vehicle treatment. BAL studies demonstrated that HS treatment was associated with a significant increase in BAL inflammatory cell counts in wild-type and βENaC-Tg neonates, as well as in adult wild-type mice. Our preclinical studies indicate that both preventive and late HS treatment provide an effective mucolytic therapy for CF-like lung disease. However, HS promoted air-way inflammation in neonates and had no therapeutic effect on airway inflammation in adult βENaC-Tg mice with established CF-like lung disease. These results suggest that potential proinflammatory side effects of early HS treatment should be evaluated before using HS as treatment for CF infants. Further, our results suggest that combined mucolytic and antiinflammatory therapeutic strategies may be necessary for the treatment of CF lung disease.

Supported by EC (MEXT-2004-013666 We have previously shown that simian immunodeficiency virus pseudotyped with F and HN protein from Sendai virus (F/HN-SIV) transduces murine nasal epithelium efficiently (3-5% of respiratory epithelial cells after perfusion with 5x10 8 TU) and importantly can be repeatedly administered. We now show that expression in the nasal epithelium persists for the lifetime of the animals (16-25 months, 9 of 9 C57Bl/6 mice). The vast majority of published studies have assessed lentiviral vectors in mouse nasal epithelium, and transduction of lung epithelium, particularly without preconditioning through polidocanol treatment or tissue damage, remains challenging. Here, we show that F/HN-SIV transduces lung epithelium efficiently and dose-dependently. In contrast to other studies, we show that lentivirus-mediated gene expression in the lung is stable for at least 13 months (latest time-point currently analyzed) (month 2: 362660±63922 photons/sec, month 13: 431454±65647 photons/sec, n=5-8 mice). We also show, for the first time, that lentivirus can be repeatedly administered to the lung (2 doses of F/HN-SIV-GFP followed by a third dose of F/HN-SIV-Lux at monthly interval) without loss of activity compared to a single dose of F/HN-SIV-Lux (1 Dose: 40178±4843, 3 Doses: 39080±7490 RLU/mg protein, n=21/group). Importantly, we also show that gene expression increases with increasing number of doses (10 doses at daily interval) (Dose 1: 4693± 899, Dose 10: 239212±48362 RLU/mg protein, n=8/group, p< 0.0005). We have not observed acute or chronic toxicity in 12 months follow-up studies in mice. We have also assessed the performance of F/HN-SIV in various airway ex vivo models. (a) The virus transduces human air liquid interface cultures efficiently with expression persisting for at least 4 months. (b) The virus transduces freshly obtained human nasal brushings dose-dependently. (c) The virus transduces human lung slices efficiently and expression persists for the life-span of the slices. These data suggest that F/HN-SIV may be a suitable vector for cystic fibrosis gene therapy. The nuclear membrane is likely to be a significant barrier to non-viral gene transfer into non-dividing differentiated airway epithelial cells. Transcyclohexane 1,2 diol (TCHD) is an amphipathic alcohol which has been shown to temporarily disrupt the permeability of the nuclear pore. Previous studies have shown that TCHD can increase lipid-mediated transfection in vitro (Vandenbroucke et al, 2007) . Here, we aimed to reproduce these in vitro studies using the cationic lipid GL67A which we are currently assessing in cystic fibrosis gene therapy trials and, more importantly, to assess the effects of TCHD on transfection efficiency in fully differentiated airway epithelium ex vivo and in vivo.

To reproduce previous in vitro studies with a clinically relevant lipid we first transfected T293 HEK cells with GL67A complexed to pCIKLux for 6 ery that controls both activity and levels of neutrophilic inflammation. Our data verifies the targeted delivery of the proposed delivery system.

We have developed a novel targeted nano-therapeutic vehicle for site directed sustained release of CF anti-inflammatory and "corrector" drugs. #Contact: nvij1@jhmi.edu Support: Accelerated Translational Incubator Program (NIH R025-CR07), NIH R03HL096931, CFF & FAMRI (NV). Rationale: Cystic fibrosis (CF) results from the mutation(s) in the CF transmembrane conductance regulator (CFTR) gene. The most common mutation, ∆F508-CFTR, is a temperature-sensitive, trafficking mutant with reduced chloride transport and exaggerated immune response. The neutrophil-dominated airway inflammation results from NFκB mediated, proinflammatory cytokine, IL-8, hyper-secretion. Class II histone-deacetylase (HDAC) inhibitors like subenorylanilide hydroxomic acid (SAHA) modulate NFκB signaling and neutrophil chemotaxis.

Objective: To evaluate the efficacy of selective HDAC inhibitors to control CF lung disease.

Methods: HEK293 cells were transfected with HDAC6shRNA, ∆F508-pCEP or control plasmids and treated with SAHA or TSA (Trichostatin A) for 24 hrs followed by immunoblotting for HDAC6 and β-actin or metabolic 35 S-labeling, CFTR immunoprecipitation and radiography. The age and sex matched CFKO mice (n=3, each group) were injected (i.p.) or treated intratracheally (i.t.) with SAHA (25 mg/kg bw) and/or P. aeruginosa LPS (15 mg/kg bw, i.p. or 20 µg, i.t.) for 24 (i.t.) or 36 (i.p.) hrs. Inflammatory state of murine CF lung disease was quantified by measuring serum or BAL (bronchoalveolar lavage) cytokine (IL-6) and/or neutrophil (MPO, myeloperoxidase) activity by sandwich ELISAs (R&D and Hycult Biotechnology), and immunostaining and immunoblotting for inflammatory markers in lung tissues.

Results: We found that selective HDAC inhibition by HDAC6shRNA, SAHA or TSA inhibits NFκB and ER stress activities. In addition, we identified by metabolic 35 S-labeling that SAHA induces expression and post-Golgi transport of mutant ∆F508-CFTR. Our in vivo studies demonstrate that SAHA significantly decreases LPS induced IL-6 cytokine secretion (i.p and i.t.) and MPO (i.p.) levels, an indicator of activated neutrophils. Moreover, the immunostaining data verified that SAHA inhibits LPS induced NFκB expression and nuclear localization, and neutrophil (marker, NIMP-R14) chemotaxis. We confirmed by immunoblotting that SAHA inhibits LPS induced NFκB signaling, HDAC2 dependent glucocorticoid sensitivity and Nrf2 mediated oxidative stress response. In addition to controlling neutrophilic inflammation, we identified by immunoblotting and immunostaining that SAHA induces the recruitment of regulatory T cells (Tregs) to the airway to control the pathogenesis of chronic CF lung disease.

Conclusion: Selective inhibition of HDAC activity controls pathogenesis of chronic CF lung disease by modulating NFκB signaling, neutrophil chemotaxis and Treg recruitment. Our data suggests that selective HDAC inhibition can be developed as a novel therapeutic strategy to ameliorate chronic CF lung disease. Support: NIH (R025-CR07, R03HL096931), CFF & FAMRI (NV). #Contact: nvij1@jhmi.edu hrs (1 µg DNA/well at a 6:8 lipid:DNA molar ratio) followed by a 1 hr exposure to TCHD (0.5 to 2% w/v in medium). TCHD exposure increased luciferase expression dose-dependently up to 100-fold compared to no TCHD treated controls (n=6/group, p<0.001). We next assessed the effects of TCHD in fully differentiated human air liquid interface (ALI) cultures (Epithelix Sàrl, Switzerland). ALIs were transfected with GL67A complexed to pCIKGLux (10 µg/insert at a 6:8 lipid:DNA molar ratio) for 6 hr followed by a 1 hr exposure to TCHD (0.5 to 2% w/v in medium). The effect of TCHD in ALIs was inconsistent with only 1 out of 4 ALI batches (n=3 ALIs/batch/dose) showing a dose-dependent increase in gene expression 48 hrs after transfection (0%: 1585±801; 0.5%: 17446±3595 RLU/µl apical medium). Different batches of ALIs are generated from different donor material and the inconsistent effect of TCHD in ALIs warrants further investigation. We then assessed the effects of TCHD in vivo. GL67A/pCIKLux complexes were administered to the lungs of mice by nebulisation and various concentrations of TCHD (0.4 to 10% w/v) were administered by nasal sniffing immediately pre and post nebulisation to ensure maximal exposure of the mice (n=8/group). In a separate set of experiments mice (n=8/group) were nebulised with GL67A/pCIKLux complexes followed immediately by nebulisation of 10% TCHD (maximum concentration that could be dissolved) for 1 hr. In a third set of experiments GL67A/pCIKLux was administered to the nasal epithelium by nasal perfusion over 15 min followed by nasal perfusion of 100 µl of 10% TCHD over 15 min (n=8 mice/group). This protocol ensured prolonged contact time between the airway epithelium and the compounds. None of the TCHD treated mice had higher gene expression when compared to no TCHD controls. In conclusion, although the amphipathic alcohol TCHD significantly increases gene transfer in cell lines, the effects are lost when moving into more relevant ex vivo and in vivo models of non-dividing fully differentiated airway epithelium. Rationale: Cystic fibrosis (CF) results from the mutation(s) in the CF transmembrane conductance regulator (CFTR) gene. The most common mutation, ∆F508-CFTR, is a temperature-sensitive, trafficking mutant with reduced chloride transport and exaggerated immune response. The neutrophil-dominated airway inflammation results from NFκB mediated, proinflammatory cytokine, IL-8, hyper-secretion. The recent studies have identified several novel "correctors" and molecular targets for functional rescue of misfolded protein or chronic inflammatory state of CF but the challenge is to bypass the mucus and selectively deliver these drug(s) to the airway.

Objective: Our data suggest that design and development of novel nano-based biodegradable therapeutic vehicles capable of bypassing the mucus barrier will provide a therapeutic tool for controlled drug delivery and selective modulation of disease causing homeostatic processes to rescue the CF pathophysiology.

Methods: The WT and CFTR-/-(FABP CFTR gut corrected)-mice and cells (airway and inflammatory) were used to evaluate the release kinetics and efficacy of drug delivery by microscopy, in vivo live animal imaging, flow cytometry, spectrophotometry, immunoblotting and immunostaining.

Results: We have standardized the PLGA-PEG nanoparticle (PNP) based drug delivery to CF knock out (CFKO, FABP CFTR gut corrected) murine lungs based on its ability to control Pseudomonas aeruginosa (Pa) LPS induced chronic lung disease and drug activity. Next, we standarized the efficacy of PINP (PEGylated Immunoparticle)-mediated drug delivery by quantifying the activity of non-steroidal anti-inflammatory drug (ibuprofen) in HBE-cells based on its ability to control TNFα induced NFκB levels (~ 2 fold) by immunoblotting. Our data demonstrates the in vitro efficacy of PINP mediated drug delivery and activity. We also standardized the in vitro and in vivo release kinetics of PINP by quantifying the release of fluorescent (Rhodamine-6G/Nile-Red) dye over time (day 1-10). The dispersion and qualitative analysis of marker dye loaded PINP as compared to PNP was carried using the FACsCalibur (BD Inc) flow cytometer and cell quest software. We verified the targeted neutrophil delivery of PINP NIMP in murine lungs by microscopy and FACS analysis of bronchoalveolar lavage (BAL) from mice. We observed the specific binding of PINP NIMP-R6G to the BAL neutrophils as compared to PNP R6G that did not show any binding. In addition, we observed the clearance of GFP tagged Pa PAO1 bacteria by PINP NIMP (neutrophil tagged PNPs) mediated anti-inflammatory drug deliv- Objective: Arikace™ is an aqueous liposomal dispersion of amikacin for inhalation being developed for the treatment of P. aeruginosa (Pa) infection in CF patients (pts). The primary objective of the study was to evaluate the long-term effects of multiple cycles (cy) of treatment with Arikace™.

Methods: This is a multi-cycle, open-label study being conducted in 11 centers in Europe. Pts 6 years (yr) and above who were previously enrolled in the placebo controlled study of Arikace™ were consented to participate in this 6 cy, 18 month (mo) study. Each cy comprises 28 days (d) of treatment with 560 mg of Arikace™ by inhalation with PARI eFlow® nebulizer followed by 56 d off drug. Safety, pharmacokinetics, change in lung function, Pa sputum density, quality of life (CFQ-R) and exacerbation rate are evaluated at regular intervals.

Results: Forty-nine eligible pts were enrolled over several months. To date, 47 pts have completed 2 cycles. Forty-six pts, 39 pts, 25 pts, and 9 pts have completed 3, 4, 5, and 6 cycles respectively. Baseline Mean age is 17.3 yr; FEV1(% pred) is 60.1%; male 41.7%; female 58.3%. Overall, Arikace™ 560 mg administered once daily for 6 cy is well tolerated and demonstrates adverse effects consistent with those expected in a population of CF pts. Available data from 4 cy show an increase above baseline in FEV1(%) of 8.5%-12.0% during the treatment period and sustained increase of 5%-10% at the end of the 2 months off drug for each cy. Up to 1.3 log reduction in Pa was also seen with no appreciable shift in MIC. Updated data from additional cycles and patients will be available at NACFC.

Conclusion: Arikace™ 560 mg given once daily for 28 d for multiple cy is well tolerated in CF pts. An increase in FEV1(%) significantly above baseline is observed during treatment period of each cy which is sustained during the 2 mo off study drug indicating patient benefit. Introduction: Lung damage in CF is thought to begin by mucus impaction mainly localized in the peripheral airways. To prevent lung damage from occurring sputum mobilization is important in the treatment of CF. Daily nebulization of rhDNase facilitates mobilisation of mucus from the airways. However, relatively little of the inhaled drug is deposited in the peripheral airways. Targeting rhDNase to the peripheral airways might improve mucus clearance from these airways. Therefore we hypothesized that peripheral deposition of rhDNase can improve lung function in children with CF.

Aim: To investigate the effect of treatment with nebulized rhDNase targeted to the peripheral airways compared to rhDNase targeted to the central airways on lung function in children with CF.

Methods: Multi-centre, randomized controlled clinical trial. Study population: children with CF aged 6-18 years, on maintenance treatment with rhDNase and in a stable clinical condition. Patients were randomized to either treatment with inhaled rhDNase targeted to the peripheral airways or rhDNase targeted to the central airways for four weeks, using the AKITA® 2 APIXNEB inhalation system that uses TouchSpray® vibrating mesh technology. Study measurements included spirometry and Lung Clearance Index (LCI) and were performed at t = 0 weeks, t = 2 weeks and t = 4 weeks. Primary endpoint: FEF 75 .

Results: Forty-nine children were included. There were no significant differences in age, height and lung function parameters between the two groups at start of the study.

Lung function improved significantly in both treatment groups: FEF 75 increased by 5.2 %predicted (p = 0.03) in the central group and by 9.2 %predicted (p < 0.001) in the peripheral group after 4 weeks of treatment with the AKITA® 2 . There was no significant difference in effect between the two treatment groups: difference was 3.6 %predicted in favor of the peripheral group(p = 0.27), 95% CI: (-2.9 -10.1). For FEV 1 and MMEF similar results were found. FVC and LCI did not change significantly in both groups.

Conclusions: The study results indicate a benefit of inhaled rhDNase in the medium and small airways. Since all patients already used rhDNase as maintenance medication before start of the study, it is interesting to see that lung function improved significantly with the use of the AKITA® 2 APIXNEB device. From this study we cannot conclude what might have caused the improvement in lung function. Multiple factors might have contributed: increased lung dose due to the higher efficiency of the AKITA® 2 , improved peripheral deposition in both groups due to the controlled breathing or improved adherence during the study.

This study could not prove our hypothesis that peripheral deposition of rhDNase improves lung function more than central deposition in children with CF. However, based on this study we cannot exclude a possible beneficial effect of peripheral deposition, since the 95% CI is relatively wide. Objectives: To develop a population pharmacokinetic (PK) model for tobramycin inhalation powder (TIP) in cystic fibrosis (CF) patients, and to characterize the effect of the covariates age, body mass index, creatinine clearance, gender, lung function (as FEV1% predicted), and weight on the model parameters, if any.

Experimental methods: TIP-containing arms of three clinical studies (one Phase 1 and two Phase 3) were pooled to generate a PK analysis dataset for modeling. In total, 931 concentration-time observations from 154 CF patients dosed with 28 mg, 56 mg, 84 mg, 112 mg once daily, and 112 mg twice daily were included.

Data analysis and modeling methods: A population PK model was developed using the software program NONMEM and used for investigating covariate effects on area under the curve at steady state (AUCT), maximum concentration (Cmax), and minimum concentration (Ctrough).

Results: The final population PK model developed was a two-compartment model with first-order elimination and first order absorption. For a typical CF patient, the apparent clearance (CL/F) was estimated to be 14.1 L/h, apparent peripheral volume of distribution (Vd/F) 84.1 L, apparent peripheral volume of distribution (Vp/F) 162 L, and apparent intercompartmental clearance (Q/F) 5.68 L/h. Interindividual variability (IIV) of CL/F had a variance of 0.122, IIV of Vd/F had a variance of 0.149 and IIV of the absorption rate constant (ka) had a variance of 0.169. Interoccasion variability (IOV) on CL/F was estimated to have a variance of 0.159.

No significant covariate was identified for CL/F, hence no covariate effect on AUCτ was noted. Vd/F was found to increase nonlinearly with increasing body mass index (BMI), and decrease with increasing FEV1% predicted. Ka was found to decrease nonlinearly with increasing FEV1% predicted. Simulations showed that: 1) with BMI decreasing from 90th percentile (25.4 kg/m 2 ), to 5th percentile of the study population distribution (13.3 kg/m2), Cmax increases 21% from 1.19 µg/mL to 1.44 µg/mL, while Ctrough decreases 19% from 0.31 µg/mL to 0.25 µg/mL; 2) with FEV1% predicted increasing from 5th percentile (29.7% predicted), to 90th percentile (100% predicted) of the study population distribution, Cmax increas-

NPD is an important biomarker of CFTR activity and diagnostic test for CF. In preparation for a number of international clinical trials, we standardized NPD performance using an electronic data capture system (AD Instruments), KCl calomel electrodes (Fisher) and agar two-lumen human interface. As part of our standard operating procedure, each operator was required to submit 5 CF and 5 normal tracings to achieve qualification by the CFF-TDN Center for CFTR Detection National Resource Center. Criteria for qualification included 1) completion of the standardized process and sequence, 2) absence of sustained breaks in the tracing, 3) stable and interpretable values at respective 10 sec scoring interval, and 4) >3 mV change in amiloride and ATP PD. All tracings were reviewed and qualified by a single analyst. We received 380 agar tracings from 54 operators and 29 centers across 9 countries; 217 were non-CF and 163 were CF tracings. All PD values were the mean from two nostrils following sequential perfusion of Ringer's, Ringer's + amiloride (100 µM), zero chloride gluconate + amil, zero chloride + amil + isoproterenol (10 µM), and zero chloride + amil +iso + ATP (100 µM). Eighty-seven percent of normal subject tracings and 93% of CF tracings were qualified. The analysis cohort included 144 normal and 135 CF tracings. For normal subjects, the mean PD value (mV ± SD) for Ringer's, Amiloride, Zero Cl -, Iso, and ATP were -16.10 ± 9.32, -7.33 ± 6.65, -19.50 ± 10.53, -30.00 ± 13.45, and -37.30 ± 15.58 respectively. For CF, these values were -35.17 ± 12.53, -16.52 ± 8.43, -15.09 ± 8.70, -15.33 ± 9.14, and -27.45 ± 11.93 respectively. All P values were < 0.0001. Derived NPD parameters also demonstrated clear differences between groups. Total CFTR-dependent Clconductance in normal and CF individuals was -22.41 ± 13.09 and 1.18 ± 3.94, respectively (P<0.0001). The Max Basal PD (most polarized of 5 measures within the inferior meatus, averaged between each nostril) for normal and CF were -21.66 ± 10.69 and -41.03 ± 15.49 respectively. The Mean Basal PD for normal and CF were -14.68 ± 10.14 and -27.55 ± 12.38 respectively. All P values were < 0.0001. Based on these data, we have composed the 95% CI for normal and CF subjects for this international dataset and the ROC curves for chloride and sodium conductance. For change Zero Cl -+ Iso, the 95% percentile range is between -57.16 and -5.54 mV for normal (not normally distributed), and between -8.96 and +8.18 for CF subjects. The results suggest that values between -8.96 and -5.54 mV are indeterminate. ROC curves were determined for both chloride and sodium conductance. Using a cutoff of -6.62 mV, change Zero Cl -+ Iso is 94.8% sensitive and 96.5% specific to diagnose CF. The 95% percentile range for Max Basal PD for normal and CF was -48.99 to -6.80 mV and -75.31 to -18.36 mV, respectively. Prospective validation of diagnostic thresholds is ongoing (~10 tracings analyzed/week). These data demonstrate NPD is a useful diagnostic measure for CF, and provide the first report of normative values from a large international dataset.

Supported by CFFT & NIH. Defective CFTR-mediated chloride transport in airway epithelium is a prominent feature in cystic fibrosis (CF). Sensitive bioassays of CFTR function are important in CF drug development for evaluation of small-molecule drug candidates. The purpose of this study was to develop and evaluate a non-electrophysiological, fluorescence assay of chloride-specific ion transport in human airway epithelia for preclinical testing of drug correctors of defective CFTR in CF. Our assay utilized N,N-dimethyl-9,9-bisacridinium es 16% from 1.17 µg/mL to 1.36 µg/mL, while Ctrough decreases 24% from 0.33 µg/mL to 0.25 µg/mL.

Conclusions: A population PK model was successfully developed to adequately describe the tobramycin serum concentration time-course in the CF patient following inhalation of TIP. BMI and FEV1% predicted were found to be statistically significant covariates for Vd/F, as well as FEV1% predicted for ka. However, none of the covariate effects would likely be clinically significant based on the simulation results. Recombinant adeno-associated virus (rAAV) represents a promising potential gene therapy vector for CF. However, the most commonly-studied rAAV serotype (rAAV2) is inefficient at transducing polarized human airway epithelia (HAE) from the apical membrane, and CF clinical trials using this vector failed to produce efficient gene expression in vivo. Interestingly, rAAV2 transduction from the basolateral membrane is highly efficient. In contrast, rAAV1 is efficient at transducing HAE with no polarity bias. Differences in intracellular trafficking of rAAV virions have been thought to underlie the transduction differences observed with these two serotypes. We hypothesized that rAAV2 trafficking from the apical membrane results in accumulation of the virus in an endosomal compartment that is unable to process the virion, and that this compartment would differ from that used during apical trafficking of rAAV1. Co-infection experiments with differently-labeled fluorescent rAAV1 and rAAV2 particles suggested that whereas rAAV1 trafficking pathways from both membranes converge, rAAV2 utilizes two completely different endosomal trafficking pathways from the apical and basolateral membranes. In order to identify the specific apical and basolateral trafficking pathways of rAAV2, we performed staining for various endosomal markers on polarized HAE infected with fluorescentlylabeled virions. Entry through the apical membrane was found to be markedly slower than through the basolateral membrane. We determined that rAAV2 particles entering through the apical membrane move to the early endosome, and then a majority of these particles move rapidly to the late endosome/lysosome, where they remain for long periods of time. We are currently evaluating the endosomal compartments utilized by rAAV1. Treatment with proteasome inhibitors has been shown to augment rAAV2 transduction by enhancing viral trafficking to the nucleus. We hypothesized that proteasome inhibitors might rescue rAAV2 from the lysosome following apical entry. We found that fluorescently-labeled rAAV2 particles had trafficked significantly farther towards the nucleus following proteasome inhibitor treatment. Furthermore, treatment with proteasome inhibitors two weeks following apical infection of HAE rescued transduction to levels seen with basolateral infection, which is highly efficient. We conclude that inefficient rAAV2 apical transduction is caused by the routing of viral particles into the late endosome/lysosome, where they are unable to be effectively processed and remain dormant for long periods of time. Treatment with proteasome inhibitors rescues functional virions from this compartment, allowing them to move into the nucleus and transduce the cell. These results provide promising evidence that rAAV2 gene therapy in the airway may be successful if the basic trafficking biology of the virus is altered.

(lucigenin), a membrane impermeable and selective chloride-sensing dye, whose fluorescence is quenched by 50% at 3 mM chloride. Chloride efflux from airway epithelia into a chloride-free mucosal buffer solution containing lucigenin was quantified from the kinetics of lucigenin fluorescence. Large, robust fluorescence signal changes, reflecting CFTR function, were found in primary cultures of human bronchial epithelial cells grown for 21 days at an air-liquid interface. In a protocol in which cells were first incubated with a chloride-free buffer, addition of CFTR agonists (forskolin/IBMX in presence of indomethacin) to a 0 mM chloride basolateral perfusate greatly increased chloride efflux into the apical solution in non-CF but not in CF cells. Subsequent addition of chloride (10-120 mM) to the basolateral perfusate greatly increased rates of chloride efflux into the apical solution, with higher chloride efflux rates in the presence of CFTR activators, allowing the resolution of transcellular and paracellular components of chloride transport. In preliminary studies we applied the method to measure CFTRdependent chloride transport in pig trachea, suggesting its potential utility in native ex vivo airway tissue. Studies using ion channel inhibitors and ion substitution maneuvers are underway to characterize the ion transporter determinants of the fluorescence signal. This newly established optical assay of CFTR-dependent chloride transport may have utility in preclinical testing of candidate CF drugs as an adjunct or alternative to conventional electrophysiological assays.

This work was supported by CFF, NIH and CFRI. Cystic fibrosis (CF) is the most common life-limiting recessive disease in the U.S. and Europe, and is due to mutations in the cystic fibrosis transmembrane-conductance regulator (CFTR) gene. CF mutations, of which the most common is ∆F508-CFTR, cause a massive pro-inflammatory phenotype in the lung arising from profound expression of inflammatory genes including interleukin-8 (IL-8). Recently, miRNAs have been proven to be key regulators of gene expression by directing their target mRNAs toward degradation and/or translational repression. Based on partial homology between the 3′-UTRs of TNFα and IL-8, we hypothesized that miR-16, which is known to reduce TNFα mRNA stability, might also reduce IL-8 mRNA stability in CF cells. We found that miR-16 indeed suppressed IL-8 expression in CF lung epithelial cells by interacting with the AU-rich sequences in the 3′-UTR of IL-8 mRNA. To more generally identify CFspecific miRNAs, we screened a miRNA library for differential expression in [∆F508] CFTR and [wildtype] CFTR lung epithelial cell lines. We found 18 miRNAs that are up-regulated in CF cells and four miRNAs that are down-regulated. Of these, miR-155 is the miRNA most significantly elevated in CF cells. Consistently, treating the CF cell with the stable antisense antagomir for miR-155 potently suppresses expression of IL-8 mRNA, as well as IL-8 protein. These data suggest a regulatory role for miR-155, and a few other miRNAs, in the CF airway, and identify the antagomir-155 as a novel candidate anti-inflammatory therapeutic for CF. Introduction: Lubiprostone, a bicyclic fatty acid, is in clinical use for treatment of chronic constipation. It alleviates obstruction by stimulating intestinal fluid secretion, purportedly through activation of CLC-2-type chloride channels. Intestinal obstruction caused by the loss of CFTR-mediated chloride secretion is also a severe pathological symptom in a subpopulation of CF patients. To find out whether lubiprostone, through activation of CLC-2, could compensate for the chloride secretory defect in CF, we investigated its mode of action in both mouse and human intestine.

Methods: Transepithelial chloride currents were measured in Ussing chambers, in three model systems: (1) human T84 colonocytes; (2) intestinal epithelium of wild-type mice, Cftr-null mice (Cftr tm1Cam ), and F508del/F508del mice (Cftr tm1Eur ); (3) resected ileum and rectal biopsies of CF patients (homozygous F508del) and healthy controls. The expression and tissue distribution of ClC-2 in mouse intestine was monitored by immunostaining using a C-terminal CLC-2 specific antibody (ACL-002, Alomone Labs).

Results: In T84 monolayers, nystatin-permeabilization of the basolateral membrane revealed that lubiprostone activated an apical chloride conductance. The chloride secretory response was attenuated by H89, an inhibitor of cAMP-dependent protein kinase (PKA), and lubiprostone pre-treatment strongly diminished the responsiveness to the cAMP agonist forskolin. CFTR blockage by CFTRinh172, but not ClC-2 blockage by Cd 2+ , inhibited the lubiprostone response. Lubiprostone also induced a Cd 2+ -insensitive secretory response in the intestine of WT mice, but failed to induce intestinal chloride secretion in Cftr-null mice. Similarly, lubiprostone induced a secretory response in human rectal biopsies and resected ileum of controls, but not of CF patients. The EP4-type prostanoid receptor antagonist L-161,982 blocked the lubiprostone response in all three models studied. In T84 cells, lubiprostone induced a rise in cellular cAMP levels that was blocked by L-161,982. In mouse colon and ileum, CLC-2 was localized in the basolateral membrane of surface cells, but not in the crypt compartment. Interestingly, CLC-2 expression was reduced but not abolished in Cftr-null mice.

Conclusions: Activation of the EP4 receptor/ cAMP/PKA/CFTR pathway can fully account for the pro-secretory action of lubiprostone in mouse and human intestine. Lubiprostone-induced chloride secretion is lost in CF intestine, and is not compensated by CLC-2 channel activation. Most plausibly, CLC-2 functions as an absorptive chloride channel at the basolateral membrane of intestinal surface cells, implying that its downregulation in CF may serve to reduce NaCl and fluid absorption and to restrict luminal dehydration. Lubiprostone may further promote luminal hydration in CF through an anti-absorptive rather than pro-secretory action, based on the known inhibitory effect of cAMP/PKA signaling on salt-and water absorption at the level of the sodium-proton exchanger NHE3. Acknowledgement: Supported by the Dutch Digestive Disease Foundation (MLDS-WO 06-11). Background: Inflammation plays a critical role in lung disease development and progression in CF. Antiinflammatory drugs, interfering with a vicious circle that amplifies inflammation and perpetuates infection, could represent an important strategy for CF lung pathology. Phosphodiesterase type 5 inhibitors, such as vardenafil, are used for the treatment of male erectile dysfunction and are able to activate chloride transport in CF mice homozygous for the F508del mutation (Lubamba et al. 2008) , although the underlying mechanisms of action are poorly understood.

We tested the hypothesis that vardenafil modulates the inflammatory response in CF airways.

Methods: We monitored molecular inflammatory markers in lungs of CF-F508del mice, either in baseline conditions or after induction of acute inflammation by intratracheal instillation of lipopolysaccharide from Pseudomonas aeruginosa (LPS). The effect of vardenafil pretreatment (0.14 mg/kg/day), given by a single intraperitoneal injection, was evaluated.

Verhoeven, R.S. 1 ; Lambert, A. 2 ; Cowlen, M.S. 1 1. Inspire Pharmaceuticals, Durham, NC, USA; 2. Charles River Laboratories Preclinical Services Montreal, Senneville, QC, Canada Objective: Denufosol, an investigational early-intervention therapy for cystic fibrosis (CF), is a novel ion channel regulator that restores chloride transport via calcium-activated chloride channels, inhibits sodium absorption via epithelial sodium channels, and stimulates ciliary beat frequency. Denufosol has been demonstrated to be effective and well tolerated in the treatment of CF in Phase 2 and 3 trials and has Fast Track designation and orphan drug status in the United States and orphan drug designation in Europe for the treatment of CF. This research assessed the effect of denufosol at doses exceeding the human clinical dose on background lung inflammation after chronic inhalation in rats and dogs.

Methods: Denufosol was administered as an isotonic, aerosol formulation using Pari eFlow nebulizers. Rats were treated daily for 2 years at doses up to 2.36 mg/g lung weight/day; dogs were treated daily for 1 year at doses up to 7.27 mg/g lung weight/day. Control animals were treated with nebulized isotonic saline. The potential for denufosol to induce inflammatory responses in respiratory tract tissues was assessed histopathologically by board-certified veterinary pathologists.

Results: No denufosol-related inflammatory microscopic changes were observed following daily administration for 104 weeks to rats or for 52 weeks to dogs. Findings in denufosol-treated animals were considered to be unrelated to denufosol exposure based on similar findings in control animals and a lack of dose-dependency.

Conclusions: Denufosol did not increase background levels of pulmonary inflammation compared with saline control following daily inhalation for up to 2 years in rats at doses up to 6 times the clinical dose for children and 13 times the clinical dose for adults with CF, or for up to 1 year in dogs at doses up to 18 times the clinical dose for children and 40 times the clinical dose for adults with CF.

Supported by Inspire Pharmaceuticals.

Verhoeven, R.S.; Cowlen, M.S. Inspire Pharmaceuticals, Durham, NC, USA Objective: Denufosol, an inhaled investigational early-intervention therapy for cystic fibrosis (CF), is a novel ion channel regulator that corrects the ion transport defects in patients with all cystic fibrosis genotypes. Denufosol enhances airway hydration and mucociliary clearance by increasing chloride secretion, inhibiting sodium absorption, and stimulating cilia beat frequency in respiratory epithelial cells. These integrated pharmacologic actions provide the basis for therapeutic efficacy in CF lung disease. In addition, the pharmacokinetics of denufosol is an important attribute that contributes to the drug's safety and efficacy profile. The purpose of this study was to characterize the pharmacokinetic profile of denufosol in the lung and plasma of Sprague Dawley rats.

Methods: The pulmonary distribution of denufosol was examined in rat lung following acute and chronic daily inhalation of a nebulized aerosol for approximately 1 year at doses that exceeded the human clinical dose (up to 109 mg/kg/day vs. 9 mg/kg/day in children). Lung tissue and bronchoalevolar lavage fluid (BALF) was collected at time points before and after treatment with denufosol. Denufosol was quantified by a sensitive LC/MS/MS to assess airway surface and lung tissue distribution and the potential for denufosol to accumulate in the lung during chronic administration. The plasma pharmacokinetics of denufosol was also evaluated following inhalation and intravenous injection.

Results: Denufosol was localized primarily on the airway surface as determined by BALF analysis following a single inhaled dose or after chronic inhalation for 1 year, with much lower levels detectable in postlavage lung tissue. Elimination of denufosol from lung tissue and the airway surface occurred with half-lives of 14 min and 53 min, respectively. Denufosol was generally undetectable or present at very low levels in BALF and Results: Vardenafil significantly inhibited chemokine C-C motif ligand-(CCL)-2 and macrophage inflammatory protein-2 (MIP2) release in LPS-induced inflammation. Vardenafil reversed these inflammatory parameters nearly back to control values.

Conclusion: Our findings support the concept that vardenafil reduces lung inflammation outcome measures in CF-F508del mice. The drug represents an interesting research tool to investigate immune responses of airway cells and a promising candidate for pharmacotherapy of CF.

Supported by the French CF Foundation (Vaincre la Mucoviscidose) and by Fonds Scientifique de la Recherche (Univ. cathol. Louvain). Methods: Randomized trials of TSI vs. placebo (PBO) [1] and AZLI vs. PBO [2] were selected as having similar inclusion criteria (PA infection and naïve to inhaled antibiotics) and designs (28-days on treatment followed by at least 2 weeks off treatment). To adjust for baseline differences between trials, individual patient data available from the TSI trial were re-weighted to exactly match the published AZLI trial's mean FEV1 %-predicted, mean age relative to CF population median survival in the baseline year, and proportions with FEV1 %-predicted ≤ 50%, dornase alpha use and female gender. After matching, mean % change in FEV1 (L) was compared at postbaseline visit weeks 2, 4 and 6, and on average across these weeks. To further adjust for potential differences between trials, outcomes were compared relative to placebo arms (TSI -PBO vs. AZLI -PBO) after matching. Simulations (based on estimated means and standard errors [SE] for effects on FEV1) were used to assess differences in drug costs and % increase in FEV1 while accounting for statistical variation in the cross-trial comparison. Wholesale acquisition prices (2010) were obtained from Red Book®.

Results: Before matching, patients in the TSI trial (N=528) vs. those in the AZLI trial (N=164) had similar adjusted ages and similar proportions of females, but significantly greater use of dornase alpha (78 vs. 65%), lower mean FEV1 %-predicted (50 vs. 55%) and a greater percentage with FEV1 %-predicted ≤ 50% (53 vs. 37%) (all p < 0.05). After matching, these baseline characteristics were balanced for patients treated with TSI vs. AZLI. During the outcome period, adjusted differences in mean (SE) % increases in FEV1 between TSI and AZLI were 3.8 (2.6) at week 2, 1.9 (2.5) at week 4, 2.5 (2.5) at week 6 and 2.7 (1.9) on average across weeks. In 93% of simulation scenarios based on adjusted outcomes, TSI was associated with lower drug costs per cycle (by $324) and greater mean % FEV1 improvements vs. AZLI (by 3.0% on average); in the remaining 7% of scenarios AZLI was associated with greater drug costs per cycle (by $324) and greater mean % FEV1 improvements (by 0.8% on average).

Conclusion: Adjusting for differences between trials, TSI was associated with lower drug cost and greater FEV1 improvement during 6 weeks of treatment in the large majority of scenarios compared to ALZI. Limitations include short follow-up and the assumption of adequate adjustment for differences between trials, including the ten-year difference in study years. Further outcomes research is needed to assess longer-term cost-effectiveness of TSI or AZLI incorporating costs for medical services and other drugs.

Funding provided by Novartis Pharmaceutical Corporation. lung tissue collected prior to dosing on study day 359, confirming a lack of accumulation during chronic daily inhalation for 1 year. Plasma denufosol concentrations were often below the assay lower limit of quantification (LLOQ, 1 ng/mL) at the highest doses tested in males and females after a single dose or chronic inhalation. When data was available above the LLOQ, the plasma concentration profiles exhibited a rapid decline post-administration with an elimination half-life (t1/2) of approximately 2 minutes. Therefore, no significant or sustainable systemic exposure to denufosol was achieved during chronic inhalation. Plasma denufosol concentration following intravenous injection declined rapidly with initial phase and final elimination half-lives of approximately 16 seconds and 5 minutes, respectively. Conclusions: Denufosol is localized primarily on the airway surface after inhalation, with much lower levels associated with lung tissue and plasma, and does not accumulate on the airway surface, in lung tissue, or in plasma during chronic administration. Thus, the pharmacokinetic attributes of denufosol are ideally suited for an early intervention therapy for CF lung disease.

Supported by Inspire Pharmaceuticals. Pulmonary disease in CF is characterised by an exaggerated neutrophilic inflammatory response to infection. There is also an important but not fully understand role for platelets. This study investigates platelet-neutrophil-endothelial cell interactions in the recruitment of neutrophils across the blood vessel wall. We have investigated the effect of the aminoglycoside antibiotic, tobramycin and the copper-tobramycin complex on transendothelial migration (TEM) of neutrophils.

Polymorphonuclear cells (PMN) and platelets were isolated from normal human blood and TEM was measured across human lung microvascular endothelial cells (HLMVEC) grown in Transwells. PMN respiratory burst was stimulated with serum-opsonised zymosan and superoxide ions measured using cytochrome C at 550 nm. The effect of tobramycin and copper-tobramycin (0.01-0.5 mM) on TEM and respiratory burst was investigated. Cytotoxicity was tested using flow cytometric analysis of annexin V binding on PMN and caspase assay of HLMVEC. NAP-2 release from thrombin activated platelets was measured by ELISA.

A low level of spontaneous PMN migration (2.0±0.9%) was observed on unactivated EC, which was significantly increased (23±3.5%) by stimulating the cell layers with TNF-α (10 ng/ml). The addition of unstimulated platelets to the lower compartment induced a further significant (p<0.01) increase in TEM (50.3±8.5%). Thrombin activation induced a further significant (p<0.001) increase in TEM (75.0±5.7%). Platelets stimulated with thrombin released high concentrations of NAP-2, but undetectable IL-8. Anti-NAP-2 antibody (10 ng/ml) significantly (p<0.05) inhibited TEM, indicating NAP-2-dependent PMN TEM. Copper-tobramycin (0.5 mM) significantly (p<0.05) inhibited TEM of PMN (from 71.0±5.3% to 51.0±8.5%) across activated EC in the presence of activated platelets. PMN respiratory burst was also significantly (p<0.01) inhibited by copper-tobramycin, but tobramycin had no effect. In addition, copper-tobramycin, but not tobramycin, has superoxide dismutase activity. Copper-tobramycin and tobramycin were equally effective against Pseudomonas aeruginosa and neither was cytotoxic to PMN or EC.

These studies suggest a link between platelet activation and progressive impairment of CF lung function. Platelets are necessary in recruitment of leukocytes to areas of inflammation. Previous studies established that patients with CF have an increased number of circulating platelets, increased platelet activation in response to Gram-negative infection and increased release of platelet-derived mediators, causing negative pulmonary consequences. Here, platelets were shown to release NAP-2 and induce TEM of PMN indicating a novel therapeutic target in CF.

In conclusion, copper-tobramycin has anti-inflammatory properties that may be associated with SOD-like activity and may represent a new therapeutic strategy.

Supported by Institute of Biomedical and Biomolecular Sciences, University of Portsmouth. Objective: There is an unmet need for therapies to prevent or reduce the accelerated loss of lung function that occurs during adolescence in patients with cystic fibrosis (CF). Denufosol tetrasodium is a novel ion channel regulator that restores Cltransport via calcium-activated Clchannels, inhibits Na + absorption via epithelial Na + channels, and stimulates ciliary beat frequency. The efficacy and tolerability of denufosol in adolescent patients with CF was assessed in a subgroup analysis of data from a Phase 3 clinical trial (TIGER-1).

Methods: TIGER-1 was a 24-week randomized, double-blind, parallelgroup, placebo-controlled trial followed by a 24-week open-label extension. Eligible patients (n=352) were ≥5 years old with a confirmed CF diagnosis who were clinically stable at entry with normal to mildly impaired lung function (FEV1 ≥75% predicted) and who were receiving standard care. Efficacy and safety data were analyzed for an adolescent subgroup (12 to 18 years old; n=59 denufosol, n=64 placebo).

Results: In the adolescent subgroup, mean (SD) age was 14.0 (1.7) for denufosol and 14.3 (2.0) for placebo; and, as in the intent-to-treat population, approximately half of patients were male (56% denufosol, 50% placebo) and most were white (92% denufosol, 95% placebo). Change in FEV1 from baseline to Week 24/endpoint was +112 mL (+4.0%; n=59) for denufosol compared with -10 mL (-0.6%) for placebo (p=0.013). Change in % predicted FEV1 from baseline to Week 24/endpoint was -1.1% for denufosol compared with -4.1% for placebo (p=0.055). Change in FEF25-75 from baseline to endpoint was +115 mL/sec for denufosol compared with -112 mL/sec for placebo (p=0.036). These results compared favorably with those in the intent-to-treat population (n=352), where denufosol treatment led to an FEV1 change from baseline to Week 24/endpoint of +45 mL (+2%; p=0.047) vs placebo. At Week 48 in adolescents, the change from baseline in FEV1 for patients who were continuously on denufosol for 48 weeks was +226 mL (+8.8%, -0.9% predicted), and the change from baseline in FEF25-25 was +192 mL/sec (+10.4%). Patients who switched from placebo to denufosol at Week 24 and then received denufosol through Week 48 had a change in FEV1 from baseline to Week 48 of +83 mL (+3.4%) and a stabilized decline in % predicted FEV1 (-4.5%). No significant differences were observed between treatment groups in pulmonary exacerbations or quality of life scores. Denufosol was well tolerated in adolescent patients with an adverse-event profile comparable to that of placebo.

Conclusions: Denufosol, which enhances airway hydration and mucociliary clearance through multimodal actions (restoration of Cltransport, inhibition of Na + absorption, stimulation of ciliary beat frequency), significantly improved lung function vs placebo in adolescent patients and was well tolerated when used with standard care. Denufosol may prove to be useful for ameliorating the accelerated loss of lung function during adolescence.

Supported by Inspire Pharmaceuticals. Introduction: Chronic inflammatory response in the airway tract of patients affected by cystic fibrosis (CF) is characterized by an excessive recruitment of neutrophils to the bronchial lumina, driven by the chemokine Interleukin (IL)-8. Modulation of IL-8 is considered to be one of the key strategies to reduce the progressive pulmonary deterioration of these patients. In this context, we previously identified 5-methoxypsoralen as a relevant molecule able to reduce the P.aeruginosa (Pa)-dependent transcription of IL-8 mRNA ( Nicolis, 2009) . Extending the investigation to linear and angular trimethylated analogues (e.g. psoralen (PSR), angelicin (ANG), 4,5',8-trimethylpsoralen (TMP), 4,6,4'-trimethylangelicin (TMA)) identified a selective reduction of IL-8 mRNA upon exposure of the cells to psoralens, with TMA being the most potent. We observed that these compounds are also able to potentiate CFTR-mediated ion transport not only in w/t but also F508del CFTR expressing cells, where the mutant protein was rescued to the apical membrane. The above findings led us to investigate TMA in murine models of lung infection following exposure to Pa.

Methods: C57BL/6 mice were pre-treated with TMA (20 mg/Kg/200 µl) by intraesophageal gavage one hour before the intratracheal inoculum with Pa strain PAO1. Bronchoalveolar lavage fluid (BALF) was collected 4 hours post PAO1 infection and the inflammatory response was analysed by leukocyte recruitment and MPO activity in the airways.

Results: TMA significantly reduced the amount of neutrophils recruited in BALF and MPO activity in PAO1 infected mice. We, however, did not observe a significant increase of the amount of PAO1 in BALF, suggesting that TMA treatment attenuated the excessive inflammation without reducing the natural anti-microbial defenses. Levels of pro-inflammatory cytokines in the BAL and lung homogenates are under investigation.

Conclusions: Altogether, these results indicate that TMA is a promising molecule as a modulator of Pa-dependent lung inflammation.

Background: Pseudomonas aeruginosa (PA) is a ubiquitous Gramnegative bacterium. It is an increasingly prevalent source of nosocomial infections as well as the major cause of morbidity and mortality in CF patients. Such infections are notoriously difficult to clear from the lung with conventional therapies due to rapid evolution of antibiotic resistance and presence of PA in biofilm.

Bacteriophages (phages) have been used extensively to successfully treat many bacterial infections including those caused by PA however little, robust, peer-reviewed experimental data exists in the public domain. Recent trials have supported the safety of this approach and the first modern clinical efficacy trial was published in 2009. This indicated efficacy against PA ear infections.

Here we assessed the ability of eight lytic phages (BioPhage-PR) to kill a diverse range of clinical PA isolates and we evaluated the survivability of this following aerosolisation -a necessary step prior to delivery to the CF lung. The ability of our phage to penetrate and kill bacteria in experimental biofilms was also investigated.

to measure the effect of Auranofin on CFTR in the human colonic epithelial cells (T84) and Fisher rat thyroid epithelial cells (FRT) expressing wild-type CFTR and several CF-associated mutants. Auranofin activated CFTR mediated currents in both cell lines. The Auranofin-stimulated Isc was blocked by two specific CFTR inhibitors, GlyH-101 and Inh-172, indicating that Auranofin is an activator of wt-CFTR. Auranofin stimulated electrogenic Cltransport across the apical membrane in a concentration-dependent manner with an EC 50 of ~50 µM and ~30 µM, and its maximal response reached 80% and ~60% of that observed with 2 µM forskolin in T84 and FRT cells, respectively. Unlike forskolin, our data demonstrated that Auranofin did not increase cellular cAMP and cGMP levels in T84 and FRT cells. Thus, Auranofin is not an inhibitor of phosphodiesterase or an activator of adenylate cyclase. In addition, the PKA inhibitor H89, the PKC inhibitor BisX and cGMP-dependent protein kinase inhibitor KT5823 failed to abolish the I sc response to Auranofin. However, activation of CFTR by Auranofin was suppressed by staurosporine, a nonselective protein kinase inhibitor, suggesting that activation occurs through an unknown kinase rather than protein kinase A. Moreover, our data revealed that Auranofin significantly increased the forskolin-stimulated I sc in temperature-corrected F508del-CFTR expressing FRT cells (27°C, 48 hr) with an EC 50 of ~20 µM. When F508del-CFTR-FRT cells were incubated with correctors, overnight at 37°C, Auranofin without forskolin directly stimulated Isc. When pretreated with a combination of correctors, this direct stimulating effect of Auranofin was enhanced ~4-fold compared to the effect on cells treated with a single corrector. Although Auranofin was a weak activator of G551D-CFTR in the absence of forskolin, it significantly increased I sc of forskolin-stimulated G551D-CFTR by ~10-fold with an EC 50 of ~20 µM. A similar effect of Auranofin was also observed for the G1349D-CFTR mutant. In conclusion, these results indicate that Auranofin is an activator of wt-CFTR by a cAMPindependent mechanism and a potentiator of CF-associated CFTR mutants including F508del-CFTR, G551D-CFTR and G1349D-CFTR. Therefore, Auranofin, acting as an anti-inflammatory compound and a modifier of electrolyte transport, may be beneficial in the treatment of CF. Current efforts are underway to assess Auranofin's effects in primary cultures of human bronchial epithelial cells. The pyrazinoylguanidines (e.g. amiloride, benzamil) currently represent the only class of potent blockers of the epithelial sodium channel (ENaC). Amiloride has been demonstrated to enhance mucociliary clearance in cystic fibrosis (CF) when delivered into the airways as an aerosol, but with a short duration of action. An inhaled ENaC blocker with improved duration of action would be a promising candidate for CF therapy. In this study we focused on exploring bioisosteres of the acyl guanidine portion of amiloride, which when protonated acts as the sodium mimetic. We postulated that quaternary amines being formally positively charged would also be effective sodium cation mimetics whilst their charged nature might aid lung retention by reducing transcellular absorption thus enhancing duration of action following inhaled delivery. Simple amide linked quaternary amines were designed and synthesised which exhibited similar potency to amiloride in blocking ENaC in human bronchial epithelial cells (HBEC, IC 50~3 00 nM). A rapid expansion of the structure activity relationships (SAR) with this chemical scaffold afforded analogues with excellent in vitro potency (HBEC IC 50 < 5nM), limited membrane permeation and ENaC block in vivo as measured using a Guinea Pig tracheal potential difference (TPD) model (ED 50 44 -66 µg/kg). These studies provide the first reported potent alternatives to the pyrazinoyl guanidine core in the search to discover novel ENaC blocking compounds as candidate therapies for the treatment of CF lung disease.

One hundred PA sputum isolates from United Kingdom (n=50) and United States (n=50) CF patients were analysed in this study.

These included examples of the major UK epidemic CF clones and multiply antibiotic resistant isolates.

Pure phage cultures were isolated from UK sewage facilities and maintained at -80 o C in 50% glycerol prior to testing. Lytic activity of eight separate Podoviridae isolates was assessed by overlaying cultures on plated bacteria before overnight incubation at 37 o C and assessment of plaque formation.

PA biofilms were established in a modified Calgary peg and plate model using flat-bottomed 96-well microtiter plates with polystyrene inserts. The ability of phage to kill biofilm-associated bacteria was tested using six PA isolates.

Phage suspensions were aerosolised using a commercial nebuliser attached to a compressor. Samples were collected using a standard glass impinger containing collection buffer. Phage collected from impinger buffer and washes of the apparatus were assayed and survivability expressed as a proportion of the original suspensions.

Results: Forty-five of 50 UK and 45/50 US isolates were lysed by at least one of the eight phages used in this study giving an overall effectiveness of 90%. After exposure to BioPhage-PR the proportion of viable PA cells was reduced by 94.83% after 24 h and to 99.98% after a further 48 h, relative to untreated controls.

The proportion of phage recovered following aerosolisation varied from isolate to isolate within the range 0.5-1.5 log.

Conclusions: The eight phage in the BioPhage-PR formulation show a wide spectrum of activity against diverse isolates of PA from the UK and USA and these effectively penetrated model biofilms. BioPhage-PR phages were active against PA isolates of noted virulence such as UK epidemic clones. Our data show that aerosolisation of phage cultures causes a loss of viability, probably due to mechanical shearing. This varied widely between phage types but were within a range that should permit potentially effective therapeutic mixtures to be selected.

Möller, W. 2 ; Schuschnig, U. 1 ; Khadem Saba, G. 2 ; Häussinger, K. 3 ; Knoch, M. 1 1. PARI Pharma GmbH, Gräfelfing, Germany; 2. Helmholtz Zentrum München, München, Neuherberg, Germany; 3. Asklepios Hospital Gauting, Gauting, Germany Introduction: Chronic rhinosinusitis (CRS) is a common disease in CF and recent studies demonstrate that the upper airways are colonized with bacteria of the same genotypes as found in the lower airways in CF patients. These findings are underlining the necessity of a concomitant treatment of upper and lower airways in order to disrupt the reinfection of the lungs from the nose. The Vibrent® (PARI Pharma, Germany) is a novel device using pulsating airflow to enable nasal aerosolized drug delivery to the sites of infection: the osteomeatal area and the paranasal sinuses.

Methods: Sinus ventilation, nasal and paranasal aerosol deposition as well as nasal clearance of aerosols delivered by pulsating airflow in comparison to nasal pump sprays was studied in 16 healthy volunteers using gamma camera imaging with 81mKr-gas and aerosolized 99mTc-DTPA. Masking techniques were used to visualize and quantify nasal and sinus aerosol deposition and retention.

Results: Using pulsating airflows ventilation of the paranasal sinuses was confirmed in all volunteers. We found 71.5 ± 13% of the emitted aerosol deposited in the nose and 7.1 ± 1.4% in the sinuses. Nasal masking with a lead shield enabled the clear confirmation of activity in the maxillary sinuses while no sinus deposition was observed for nasal sprays. Compared to nasal sprays, nasal retention kinetics were prolonged by about four-fold with the Vibrent.

Conclusions: Our data provide evidence that ventilation and topical drug delivery to the posterior nose and osteomeatal area, including paranasal sinuses, are possible using pulsating airflows. Currently, this deposition study is being continued in up to 30 CRS patients. The long pentraxin PTX3 is a soluble pattern recognition receptor (PRR) which plays non-redundant roles in innate immune response and inflammatory process. Previous works demonstrated that PTX3 selectively recognized P. aeruginosa and PTX3 deficiency is associated to increased susceptibility to P. aeruginosa lung infection (Garlanda et al, 2002) . However, the role of PTX3 in the contest of cystic fibrosis (CF) disease and long term chronic P. aeruginosa infection remain to be established.

Objectives: Here, we evaluate the role and the therapeutic potential of the humoral PRR PTX3 in the CF contest as preclinical evaluation.

Methods: Sputum samples from CF and non-CF patients were isolated and characterized for PTX3 localization by immunofluorescence analysis. Therapeutic potential of PTX3 was assessed in the agar beads mouse model during persistent infection with the P. aeruginosa RP73 clinical strains isolated from a CF patient. C57Bl/6 Cftr tm1Unc TgN(FABPCFTR) and C57Bl/6 wt mice, were treated daily with human PTX3 and, after 14 days of infection, analyzed for infection by CFU count, inflammation by cytokines and total protein content measurements and lung damage by histology.

Results: In the sputum of CF patients, PTX3 co-localized with neutrophils and macrocolonies of bacteria. In murine model, PTX3 had a protective role against P. aeruginosa chronic lung infection. Mice treated with PTX3 showed a significant reduction of lung colonization after 14 days from infection when compared with non-treated mice both in wt and CF mice (5.3X10 5 vs 1.3X10 5 CFU and 9.3X10 4 vs 4.2X10 4 CFU non-treated vs PTX3 treated wt and CF mice respectively; p<0.02). Treatment with PTX3 also reduced inflammation after 14 days of infection with a significant reduction in the percentage of recruited neutrophils (72.1% vs 42.7% and 72.9% vs 58.3%, non-treated vs PTX3 treated wt and CF mice respectively; p<0.01) and total protein content in BAL fluid (1546 vs 1354 µg/ml and 256 vs 156 µg/ml, non-treated vs PTX3 treated wt and CF mice respectively; p=0.04). Protection was also associated with decreased levels of proinflammatory cytokines (IL-1β, IL-17, CCL2, CXCL1 and CXCL2) and higher level of anti-inflammatory cytokine TGF-β1 in wt PTX3-treated mice compared to non-treated mice. Analysis in CF mice is under investigation. In addition, the pathology of murine airways after 14 days of chronic infection and PTX3 treatment displayed less severe histopathological lesions and reduced inflammatory cell infiltration compared to non-treated mice both in wt and CF mice.

Conclusions: Our results indicate that PTX3 is a potential therapeutic tool in the control of P. aeruginosa chronic lung infection, inflammation and lung damage.

Supported by Telethon (GGP05095) and Italian FFC (#14/2008). examined under voltage clamp conditions, we observed similar findings. G418 treatment (250 µg/mL x 48 hrs) increased stimulated Isc following forskolin (10 µM) to 7.8 ± 0.2 µA/cm 2 compared to 1.5 ± 0.1 with vehicle control (p<0.005). Sequential addition of VRT-532 (5 µM) increased Isc further (4.5 ± 0.3) and greater than VRT-532 alone without G418 pretreatment (3.6 ± 0.5, p<0.05). In preparation for in vivo studies using the human G542X transgenic mouse (on a Cftr -/-background) treated with combination therapy, we next evaluated whether potentiation of human CFTR could be observed in mice intestine examined ex vivo in the Ussing chamber. For these studies, we examined excised intestine from Cftr -/-mice with gut-specific expression of human WT-CFTR under the FABP promoter. In the presence of Clsecretory gradient, addition of VX-770 (10 µM) increased Isc to 53.8 ± 1.0 µA/cm 2 compared to 44.4 ± 0.3 with forskolin (100 nM) alone (p<0.005). No potentiation by VX-770 was observed in hG542X Tg Cftr -/mice without gut correction or AG treatment (1.8 vs. 1.8 µA/cm 2 ). VX-770 also strongly potentiated WT-CFTR from gut-corrected mice when examined without a Clgradient, further supporting the approach (29.1 ± 2.9 µA/cm 2 with VX-770 vs. 14.7 ± 4.3 control, p<0.001). Applications of a CFTR potentiator to excised murine intestine following treatment with AGs are currently in progress. Results suggest CFTR potentiators augment rescue of CFTR nonsense mutations in vitro and provide a rational basis for testing the approach in the clinic. Funded by the NIH and the CFF. Inhibition of histone deacetylases (HDACs) has been found to significantly improve the trafficking of ∆F508-CFTR to the plasma membrane (Hutt et al., Nat. Chem. Biol. 6:25-33, 2010) . Therefore, a histone deacetylase inhibitor (HDACi) may be effective as a ∆F508 corrector. We have evaluated the efficacy of SAHA, a HDACi, in cell lines and in primary CF bronchial epithelial cells. Incubation for 24 hours of CFBE41o-and A549 cells expressing ∆F508-CFTR with SAHA (5-10 µM) caused a dramatic increase in cAMP-dependent anion transport, as measured by YFP fluorescence assay or short-circuit current experiments. The effect of SAHA was larger than that of known ∆F508 correctors such as corr-4a. SAHA, but not corr-4a, also increased the CFTR mRNA levels. We also analyzed the maturation of ∆F508-CFTR protein by immunoprecipitation/Western blot analysis. SAHA markedly increased the levels of the partially-glycosylated form of ∆F508-CFTR (band B) but not the mature version of the protein (band C). We tested the efficacy of SAHA and other treatments on bronchial epithelial cells from CF patients carrying the ∆F508 mutation. Cells were plated on porous membranes (Snapwell) to generate polarized epithelia and studied with the short-circuit current technique. Under control conditions, cell stimulation with forskolin and genistein activated a small current that was inhibited by the selective CFTR inhibitor 172. The amplitude of this current was not affected by treatment with SAHA (5-10 µM). Our results indicate that SAHA is particularly effective in transfected cell lines because of a possible transcriptional mechanism. The upregulation of ∆F508-CFTR expression in heterologous expression systems may cause an increased trafficking of the mutant protein to the plasma membrane by mass action. The lack of effect in primary cells needs further investigation.

Supported by CFFT and Italian Foundation for Cystic Fibrosis. The UK Cystic Fibrosis (CF) Gene Therapy Consortium is interested in non-viral gene therapy for CF. It is widely accepted that in addition to extracellular barriers responsible for inefficient uptake, there are key intracellular obstacles to the nuclear delivery of the therapeutic plasmid DNA (pDNA). Thus, we are investigating the intracellular fate of pDNA following transfection, using the clinically relevant Genzyme cationic lipid (GL67A) formulation and three-dimensional Spinning-Disk real-time confocal, combined with transmission electron microscopy (TEM) to track, quantitate and provide high resolution "snapshots" of pDNA at the single molecule level in transfected primary human airway epithelial cells (AECs) grown at the air-liquid interface (hALI). The pDNA was tagged with fluorescent, photostable semiconductor quantum dots (Qdot-pDNA) or 1.4 nm gold nanoparticles (Au-pDNA) for use in fluorescence or TEM studies, respectively. Both confocal microscopy and TEM experiments demonstrate that GL67A was able to transfect AECs with Qdot-and Au-pDNA. The number of gold spots in the nuclei of Au-pDNA-transfected AECs compared with those in unconjugated-pDNA-transfected control cells was significantly higher (p<0.05, n=5 independent experiments). Approximately 50% of the total internalised pDNA localised to nuclei within one hour posttransfection in both confocal (123 AECs, 8 independent experiments) and TEM (40 AECs, 5 independent experiments) studies. Thus, within one hour pDNA is equally distributed between the cytoplasm and the nucleus in well differentiated hALIs following non-viral based gene transfer. Experiments are now underway to track the intracellular trafficking of the pDNA at earlier time points.

Funded by the UK CF Trust. Suppression of premature termination codons (PTCs) represents a genetic oriented therapy with the potential to treat the underlying cause of cystic fibrosis (CF) in affected patients. Certain aminoglycosides (AGs) and other small molecules (e.g. ataluren (PTC124)) have been shown to induce translational readthrough of PTCs in CFTR in a number of preclinical and clinical settings. Efficacy has been limited in some experimental models and human trials. High-throughput drug screening has discovered novel compounds that strongly potentiate mutant CFTR localized to the surface, including VX-770, a CFTR potentiator efficacious in G551D subjects in a Phase 2 trial. Although VX-770 was designed for plasma membrane-localized mutations, it also activates WT-CFTR, suggesting VX-770 can enhance function following translational readthrough of PTCs and provide an augmented response to nonsense codon repair. To evaluate, we exposed HeLa cells transduced with W1282X CFTR with the AG G418 (500 µg/mL) or vehicle control for 24 hours, then stimulated CFTR with forskolin (20 µM), genistein (50 µM) and the CFTR potentiator VRT-532 (5 µM) or vehicle control while monitoring increase in fluorescence-based halide efflux (SPQ) over time (DF/DT). Treatment with either G418 (6.2 ± 0.3 DF/DT) or VRT-532 (4.6 ± 0.4 DF/DT) augmented efflux compared to vehicle control (3.4 ± 0.2 DF/DT), whereas combination therapy exhibited superior stimulation (8.5± 0.5; p<0.005 for all comparisons). When confirmed in primary human bronchial epithelial monolayers derived from a G542X/F508del donor

The most frequent mutation in cystic fibrosis (CF), F508del, causes the arrest of the CFTR protein in the endoplasmic reticulum (ER) and its rapid degradation by the proteasome. Drug-like molecules called correctors are able to partially rescue the F508del-CFTR protein from the ER in heterologous expression systems. Cell treatment with combinations of correctors or with a corrector plus low temperature incubation often leads to a strong F508del-CFTR rescue. These findings suggest the existence of multiple quality control checkpoints for mutant CFTR. However, corrector effect is strongly affected by cell background, and in native airway epithelial cells, the efficacy of the correctors identified so far is low or absent. A better understanding of CFTR processing will allow identification of the key molecular targets for a more focused approach.

To identify key proteins involved in the processing of F508del-CFTR we adopted a functional genomics approach, based on the screening of a siRNA library. As a preliminary step, we generated a list of ∼ 150 relevant targets to be silenced with siRNAs. The list of targets came from information existing in the literature. For example, we included proteins with known interaction with CFTR like RMA1, CHIP, Derlin-1, p97/VCP, and CAL. The list was further expanded by adding other macromolecules of the quality control system and trafficking like other ubiquitin E3 ligases, heat shock proteins, and proteins of the vesicular transport system. The initial screening has been performed on A549 and CFBE41o-cell lines coexpressing F508del-CFTR and the YFP.

The results have shown that the extent of functional correction achieved by silencing of known targets is quite modest, with a strong dependence on cell background. Indeed, inhibition of RMA1 or proteasome subunits is partially effective on CFBE41o-cells. On the contrary, the effect of modulation of proteins involved in the sumoylation process seem to be more evident in A549 cells. The study will now be extended to the screening of a druggable genome-wide siRNA library to identify new key proteins involved in F508del-CFTR maturation and trafficking to the plasma membrane which possibly constitute the targets for novel approaches for CF therapy.

Supported by Italian Foundation for Cystic Fibrosis and CFFT. Introduction: Biofilm mode of growth is a main reason that the Pseudomonas aeruginosa lung infection in cystic fibrosis (CF) is rarely eradicated. The purpose of this study was to evaluate the killing activity of colistin on non-mucoid and mucoid P. aeruginosa biofilm on different time points in vitro.

Methods: Killing curves of colistin on biofilm growing non-mucoid PAO1, PAO579, and mucoid PDO300 were made on days 1, 3 and 7. The microtiter plate method was used to test minimal inhibitory concentration (MIC) and minimal bactericidal concentration (MBC) for planktonic bacteria, and the minimal biofilm inhibition concentration (MBIC) and minimal biofilm eradication concentration (MBEC) were detected on days 1, 3 and 7.

Results: Colistin showed concentration-dependent killing activity for biofilms of non-mucoid PAO1, PAO579, and mucoid PDO300 on days 1, 3 and 7. The MIC and MBC for planktonic PAO1, PAO579, PDO300 were 4, 4, 4 µg/ml and 8, 8, 16 µg/ml respectively. MBIC of PAO1, PAO579, PDO300 was 8, 8, 16 µg/ml (day 1 biofilm); 16, 16, 32 µg/ml (day 3 biofilm) and 32, 16, 32 µg/ml (day 7 biofilm) respectively. MBEC of PAO1, PAO579, PDO300 was 64, 64, 64 µg/ml (day 1 biofilm); 128, 128, 256 µg/ml (day 3 biofilm) and 256, 128, 256 µg/ml (day 7 biofilm) respectively.

Conclusions: The killing activity of colistin was found to be concentration-dependent on both biofilm and planktonic growing P. aeruginosa of non-mucoid PAO1, PAO579, and mucoid PDO300. MBIC and MBEC of colistin were much higher than the MIC and MBC of the planktonic P. aeruginosa. MBIC and MBEC on day 3 and day 7 were higher than the value on day 1. The results suggest the importance of early treatment on biofilm infection of P. aeruginosa in CF patients. Background: With improved nutritional intervention and new medical treatments, the median predicted survival among CF patients is now 37 years. However, CF patients incur high medical expenditures which may impact future clinical decision-making. In this analysis, we plan to compare healthcare costs and utilization in CF patients infected with PA treated with AZLI (Cayston®, Gilead Sciences) or TIS using 6-month data from Study GS-US-205-0110 (ClinicalTrials.gov NCT00757237). New treatments such as AZLI that improve lung function and respiratory symptoms may result in reduced utilization and lower total medical costs. eration clinical plasmids (pCF1-CFTR: 3/7 samples positive at day 2, 0/7 at day 7). For clinical use, GMP production of high purity plasmid was established (<2% contaminating RNA, DNA, Protein; <5 EU/mg endotoxin). Multi-gram scale production of pGM169 was achieved via host cell fermentation at the 100 L scale, coupled with bubble lift lysis, anion exchange membrane capture, HIC chromatography purification and TFF formulation and concentration (VGXI, Houston Texas). Reproducibility was assured via validated master and working cell banks. Pharmaceutical stability of pGM169 was assessed via real-time and accelerated ICH conforming stability studies; the current shelf-life of pGM169 using our production process exceeds 4 years. The ability of the GMP lots to express hCFTR mRNA in target human airway cells was confirmed following ex vivo transfection of human nasal brushings (mean 24 %VE; n=5). The performance of a single aerosol dose of plasmid pGM169 is currently being evaluated in the lungs of CF patients. The physiological etiology of cystic fibrosis (CF) lung disease involves increased Na + absorption and decreased Clsecretion in the apical airway epithelium, resulting in dehydration of airway surface liquid and impaired mucociliary clearance. The disruption of this mechanical clearance system leads to chronic bacterial infections, pulmonary inflammation, bronchiectesis, and death in severely affected CF patients. Denufosol is an investigational early-intervention therapy for CF shown to be effective and well tolerated in the treatment of CF in Phase 2 and 3 clinical trials. The objective of this research was to further elucidate denufosol's mechanism of action by investigating the effects of denufosol on Na + absorption through the epithelial sodium channel (ENaC), Clsecretion through the calciumactivated chloride channel (CACC), airway surface liquid height, and cilia beat frequency in well-differentiated primary human airway epithelial cells isolated from normal and CF respiratory tract tissue.

Methods: Human airway epithelial cells were isolated from normal or CF respiratory tract tissue and grown in transwells under conditions that resulted in a well-differentiated, polarized primary epithelial cell culture. Na + absorption and Clsecretion were assessed electrophysiologically in Ussing chambers by measuring short circuit current under selective conditions. Amiloride, an inhibitor of ENaC, was added before or after the addition of denufosol to determine the effect of denufosol on ENaC-and CACCmediated ion transport. Airway surface liquid height was quantified by confocal fluorecence microscopy, and cilia beat frequency was measured by videomicrography.

Results: Denufosol markedly inhibited the amiloride-sensitive short circuit current in primary human airway epithelial cells, indicating inhibition of Na + absorption via ENaC, and stimulated short circuit current in the presence of amiloride, indicating stimulation of Clsecretion via CACC. Denufosol also increased airway surface liquid height and stimulated cilia beat frequency in a dose-dependent manner. Plasma denufosol levels are very low after an inhaled dose due to rapid metabolism in blood. Therefore, denufosol does not alter systemic ion channel activity or electrolyte balance as indicated by clinical laboratory evaluation following long-term inhalation in CF patients.

Conclusions: Denufosol, a novel ion channel regulator and investigational early-intervention therapy for CF, inhibits Na + absorption via ENaC, restores Cltransport via CACC, increases airway surface liquid height, and stimulates cilia beat frequency in human airway epithelial cells. These findings further elucidate denufosol's primary mechanisms of action and suggest a net in vivo effect of enhanced mucociliary clearance, thereby providing a pharmacological basis for therapeutic efficacy in the treatment of CF lung disease.

Supported by Inspire Pharmaceuticals.

We conducted a literature review to estimate US medical costs over 6 months among CF patients with PA. These estimated costs will be compared to utilization data for patients treated with either AZLI or TIS from Study GS-US-205-0110, a 6-month trial of 3 on/off courses of AZLI (N=136) vs. TIS (N=132). Specifically, respiratory hospitalizations, IV and inhaled antibiotic use and concurrent medication use will be compared. US medication costs were based on 2010 wholesale acquisition costs while hospitalization and physician visit costs were based on peer reviewed published literature. This cost utilization analysis calculates cost savings based on key utilization differences between an average AZLI-treated patient and an average TIS-treated patient over 6 months. Day 28 efficacy results presented from study GS-US-205-0110 are based on ANCOVA methods.

Results: In the US, the total average annual direct medical cost for one CF patient is $48,098. Of that total cost, 51% or $24,318 is due to one inpatient admission. In addition, average annual expenditures for outpatient visits and prescription medications are higher for those with CF than those without CF. The total average annual physician visit cost per patient is $1,578. Clinical studies are underway for the aerosol delivery of Genzyme lipid GL67A/plasmid DNA to the lungs of patients with CF. A key requirement for successful non-viral gene therapy is a plasmid with the ability to direct long-term hCFTR gene expression in target cell types. To achieve this, we developed a series of novel, fourth generation plasmids that were free of CpGs to mimimise inflammatory responses after delivery, and tested the utility of a range of promoter, enhancer and terminator sequences in various models. We evaluated the plasmids for the expression of CpG-free, codonoptimized human CFTR (soCFTR2). Ultimately, a synthetic CpG-free promoter, hCEFI, was selected for long-term soCFTR2 expression. Successful expression of functional hCFTR protein was confirmed following transient transfection of HEK293T cells, using Western blotting, iodide efflux and patch clamp studies (Hyde 2008 Nature Biotechnology 26:549). To evaluate the performance of pGM169 in vivo we complexed the plasmid with GL67A, and aerosolised to the lungs of mice. To quantify plasmid deposition and hCFTR mRNA expression in the lungs we used TaqMan PCR and RT-PCR assays with sensitivities in the near single copy number range. After aerosol delivery, bulk pGM169 DNA levels in the lung fell by approximately three orders of magnitude from a peak at day 1 (27.8±12.0 x10 3 copies per lung cell genome) to a steady state level by d14 onwards (P<0.05). Importantly, hCFTR mRNA was expressed in the lung at >5% endogenous (%VE) levels for at least 2 months (range 9.04-20.64%) and was not significantly different at d1, d14, d28 and d56 (P>0.33). Expression was also measured in human ex vivo cultures of nasal cells grown at an air liquid interface (Epithelix, Switzerland) (pGM169: 8/8 samples positive at d2; 5/8 at day 7), at higher levels (median 28.6 %VE at day 2) than first gen-

Cowlen, M.S. Inspire Pharmaceuticals, Inc, Durham, NC, USA Objective: Denufosol, a novel ion channel regulator, is an investigational early-intervention therapy that corrects the ion transport defects in patients with all cystic fibrosis genotypes. Denufosol enhances airway hydration and mucociliary clearance by increasing Clsecretion, inhibiting Na + absorption, and stimulating cilia beat frequency in respiratory epithelial cells. These integrated pharmacologic actions provide the basis for therapeutic efficacy in CF lung disease. As is customary for the development of new drugs intended for inhalation, denufosol has been tested in a variety of nonclinical safety studies. The objective of this presentation is to describe these studies in the context of the clinical development of denufosol.

Methods: A series of in vitro and in vivo toxicology and safety pharmacology studies were performed to evaluate the nonclinical safety profile of denufosol. The studies were designed to meet specific guidelines promulgated by the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) and requirements of the US Food and Drug Administration, the European Medicines Agency, and Health Canada. Toxicology studies performed in rats, rabbits, and/or dogs included acute, subchronic and chronic inhalation toxicity studies, acute and subchronic intravenous toxicity studies, fertility and early embryonic development, embryo-fetal development, and pre-and postnatal development studies, and a 2-year inhalation carcinogenicity study. The potential for denufosol to induce genetic damage was evaluated in vitro and in vivo. Safety pharmacology studies were performed to assess central nervous system and cardiovascular effects.

Results: The battery of required nonclinical studies have revealed no clinically relevant pulmonary, systemic, reproductive or developmental toxicity or adverse pharmacological effects in rats, rabbits or dogs treated with inhaled or IV injected denufosol at significantly higher doses than the investigational human inhaled dose. No toxicity was observed even when systemic exposure to denufosol was much greater than human clinical exposure based on plasma denufosol concentrations. No genetic toxicity was observed in standard in vitro and in vivo tests for mutagenicity and chromosome damage, and denufosol was noncarcinogenic in rats treated with inhaled denufosol for 2-years. These studies are required to support the clinical development of denufosol through Phase 3, and are expected to support marketing approval in the US and abroad based on compliance with ICH guidelines and guidance from international regulatory agencies.

Conclusions: There have been no clinically relevant adverse effects observed in any of the required toxicology or safety pharmacology studies that were attributable to denufosol, a novel ion channel regulator and investigational early-intervention therapy for CF, at significantly higher doses and/or plasma concentrations compared to the clinical dose and exposure levels anticipated in children and adults.

Supported by Inspire Pharmaceuticals. Methods: This two site, open labeled, unblinded non-randomized study is a Phase 1 safety and pharmacokinetic (PK) study using a 5 day IV infusion of gallium nitrate at two different doses (cohort 1: 100 mg/m 2 /day and cohort 2: 200 mg/m 2 /day) in patients with CF infected with P. aeruginosa. A total of 18 CF subjects (9 in each cohort) will be enrolled. Inclusion criteria include age 18 years and older, FEV 1 > 30% of predicted, no evidence of renal disease (creatinine < 2.0 mg/dl or < 1.5 the upper limit of normal), nor osteoporosis and an ionized calcium level > the lower limit of normal. The primary endpoints of the study are to evaluate both safety and pharmacokinetics of IV gallium in CF. Safety is evaluated by laboratory assessments and collection of adverse events. PK is assessed to establish steady state levels and also to evaluate sputum gallium levels. Secondary endpoints include lung function, quantitative sputum microbiology and respiratory symptoms as assessed using the CF Respiratory Symptom Diary.

Results: Preliminary safety results from cohort 1 will be presented.

Conclusions: IV gallium, if found safe with favorable PK, offers a novel therapeutic agent for the treatment of chronic lower airway infection with P. aeruginosa in CF. It also could provide an additional agent to treat multi-drug resistant infections that have been shown to be associated with a poorer prognosis in CF. Adeno-associated viruses (AAV) are promising vectors for use in gene therapy due to their high safety profile in human patients and long term transgene expression. We have employed a novel evolution and selection strategy to create and identify new AAV capsid variants which successfully infect a polarized human airway model system (Excoffon et al. PNAS 2009 ). The resulting capsid (AAV2.5T) was a chimera of AAV2 and AAV5, with amino acids 1-128 from AAV2 and amino acids 129-725 from AAV5, and a single additional point mutation (A581T). We hypothesized that AAV2.5T would have increased sialic acid-dependent binding on human airway epithelia. We found that AAV2.5T binding required sialic acid on the apical surface of human airway epithelia, and that AAV2.5T bound ~30 times better than AAV5. AAV2.5T binding reached saturation with a K d of 3.2x10 5 vg/cell and 7.7x10 3 receptors/cell. To further investigate the sialic acid dependence, we examined AAV2.5T binding on parental and sialic acid-deficient CHO cells. Interestingly, AAV2.5T bound equivalently in the presence or absence of sialic acid on the surface of CHO cells, but sialic acid was required for transduction by the virus. Thus, we hypothesized that sialic acid would be important downstream from the initial binding of the virus. We found that the sialic acid interaction was required for internalization of the virus. Restoration of α2-3 N-linked sialic acid on mutant CHO cells restored internalization and transduction by the virus. Based on the distinct sialylation patterns between humans and pigs, we hypothesized that AAV2.5T may not infect airway epithelia cultured from pigs. Although viral binding was only slightly reduced in pig epithelia, viral internalization was significantly reduced. Thus, we found that AAV2.5T transduction correlated with the amount of viral internalization and not with the level of viral binding, suggesting that internalization was the rate limiting step in virus transduction of primary airway epithelia. We concluded that a novel interaction between the virus and sialic acid is required for internalization and transduction of AAV2.5T in human airway epithelia. This work was supported by a PPG grant from the National Institutes of Health (HL51670-11), a grant from the Cystic Fibrosis Foundation (EXCOFF07G0), and the National Institutes of Health R01HL081527. modifying therapy with multimodal action against the basic defects in CF lung disease across all CFTR genotypes: denufosol restores Cltransport via calcium-activated Clchannels, inhibits Na + absorption via epithelial Na + channels, and stimulates ciliary beat frequency. This work assessed the suitability of denufosol as early intervention based on its ability to target small airways and its systemic bioavailability in patients with CF.

Methods: Denufosol droplet size generated by PARI LC STAR was measured with the Next Generation Impactor and laser diffraction. Systemic exposure was evaluated in patients with CF by measuring denufosol plasma concentrations in a Phase 3, double-blind, placebo-controlled, parallelgroup trial (TIGER-1) that enrolled patients ≥5 years old with FEV 1 ≥75% predicted. One of the efficacy parameters assessed in TIGER-1 was FEF 25-75 , a measure of small airways function.

Results: Denufosol had a small median droplet size (3.4 to 3.8 µm, 2.4 µm mass median aerodynamic diameter [MMAD]) suitable for reaching small airways. Following inhalation in the TIGER-1 trial (n=178 denufosol, n=174 placebo), systemic exposure to denufosol was low to none with no evidence of accumulation with chronic dosing. Denufosol was significantly more effective than placebo in improving FEF 25-75 in patients with ≤110% predicted FEV 1 (p=0.025). Consistent with the demonstrated lack of systemic exposure after acute or chronic administration, the adverse event profile of denufosol was comparable to that of placebo. No evidence of systemic effects including no changes in liver enzymes, blood biochemistry, or differential blood cell counts was observed.

Conclusions: Denufosol, which acts multimodally to restore Cltransport, inhibit Na + absorption, and stimulate ciliary beat frequency, targets the small airways and improves small airways function with low risk of systemic exposure. Due to these properties denufosol has potential as an early intervention agent for the treatment of CF lung disease.

Supported by Inspire Pharmaceuticals. Materials and Methods: Five B. cepacia strains with levofloxacin MICs ranging between 0.25 -8 mg/L were tested in a mouse model of pulmonary infection. Female BALB/c mice were injected with 150 mg/kg of cyclophosphamide 3 days prior to infection. On day 4, the mice were infected with 50 µl of bacterial suspension (~10 6 CFU/ml) using a curved beadtipped oral gavage syringe under isoflurane anesthesia. Treatment with MP-376 (60 mg/kg BID) or saline was initiated 72 hours post-infection and continued twice daily for four days using a microspray aerosol device. Sixteen hours after the last treatment, mice were sacrificed, lungs harvested, homogenized, and plated to determine colony counts (CFU).

As part of a Phase 2b trial of MP-376 in CF patients, a 16 year old male patient infected with B. cepacia (levofloxacin MIC = >128 mg/L) received MP-376 240 mg QD for 28 days.

Results: MP-376 produced at least one log CFU of bacterial killing for all strains in the mouse infection model (see Table) . In the CF patient, a 1.7 log CFU decrease in bacterial counts was observed over 28 days.

Conclusion/Discussion: Aerosol administration of MP-376 produced significant bacterial killing in strains with a wide range of MICs. The nonclinical and clinical data support the future clinical evaluation of MP-376 in the management of chronic pulmonary infections due to B. cepacia.

Hyde, S.C. 1,2 ; Davies, L.A. 1,2 ; Nunez-Alonso, G.A. 1,2 ; Gill, D.R. 1, 2 1. Gene Medicine Research Group, University of Oxford, Oxford, United Kingdom; 2. UK Cystic Fibrosis Gene Therapy Consortium, Edinburgh, London, Oxford, United Kingdom Aerosol delivery of gene transfer agents is the most practical route for CFTR based gene therapy for CF lung disease. When complexed with plasmid DNA (pDNA), the cationic polymer, 25kDa branched polyethyleneimine (PEI) retains transfection efficiency following aerosolisation and consequently is a popular non-viral choice for lung gene transfer. Until recently, progress towards the clinical use of PEI was limited by the low pDNA concentration achievable in standard PEI formulations. However, we have shown that concentrated PEI (cPEI) formulations containing up to 8 mg/ml pDNA can be prepared by controlled ultrafiltration of pDNA/PEI complexes, leading to robust levels of gene expression in the lungs of mice and sheep following aerosol delivery. We have capitalised on this important development to further exploit the potential of cPEI aerosol delivery for use in the clinic, focusing on formulations that are physically stable and suitable for long-term storage following manufacture, an absolute requirement for clinical utility. Plasmid DNA encoding firefly luciferase reporter gene was complexed with branched 25 kDa PEI (N:P of 10:1) at an initial pDNA concentration of 0.2 mg/ml in sterile water and concentrated to 2.0±0.2 mg/ml using a pressurised ultrafiltration cell incorporating a 100 kDa cut-off cellulose membrane. Complexes were filter sterilised and aliquots stored under nitrogen in glass vials at 4±2°C for 0, 1, 2, 4, 8, or 21 weeks. Visual inspection revealed no evidence of aggregation or precipitation at any time-point. A small increase in particle size from 62.5±0.3 nm (mean±SEM) at 0 weeks to 65.1±0.3 nm at 21 weeks (p<0.05 ANOVA) was identified using dynamic light scattering. No significant deviations from the initial particle zeta potential (+62.8±0.5 mV) were observed over the same period. Following storage, cPEI complexes were aerosolised to the lungs of BALB/c mice via whole body aerosol exposure and 24 hrs later luciferase reporter gene expression was measured in lung homogenates at 426±32 RLU/mg protein compared to 0.05±0.01 RLU/mg in untreated mice (p<0.0001, Student's ttest). Storage of cPEI formulations for 1 week was associated with a minor reduction in luciferase reporter gene expression (307±37 RLU/mg), but no further loss of biological efficacy was observed upon continued storage, with equivalent expression levels observed for at least 21 weeks (p>0.05 ANOVA and Tukey's multiple comparison test).

In conclusion, we have utilised physical characterisation and in vivo expression studies to investigate the stability of cPEI complexes, demonstrating that pDNA/cPEI formulations prepared by ultrafiltration are stable and can be stored as a sterile, ready to use, single vial formulation for extended periods at +4°C with minimal loss of biological efficacy. This provides further evidence to support the potential of cPEI for clinical aerosol gene therapy to treat CF lung disease.

Funded by a grant from the Cystic Fibrosis Trust to the UK CF Gene Therapy Consortium. Background: Many patients with cystic fibrosis (CF) receive repeated courses of aminoglycosides, leading to an increased risk of acute kidney injury (AKI) and chronic kidney disease (CKD). Since markers such as blood urea nitrogen and serum creatinine (SCr) lack adequate sensitivity and specificity for the diagnosis of AKI, other biomarkers may be useful to promptly and accurately identify AKI. Early diagnosis and intervention may modulate AKI and improve clinical outcome in patients who experience repeated subclinical renal insults as a result of chronic nephrotoxin exposure. We sought to characterize Kidney Injury Molecule-1 (KIM-1), Nacetyl-β-glucosaminidase (NAG) and total protein as urinary biomarkers (UBMs) of kidney injury in our CF patient population.

Methods: We conducted a cross-sectional analysis of urinary KIM-1, NAG, total protein and creatinine in 119 patients (98 with CF and 21 non-CF pulmonary controls). Clinical characteristics of interest include cumulative lifetime aminoglycoside exposure, current topical or IV aminoglycoside therapy, history of AKI and CKD, FEV1, sex, age, bacterial colonization, baseline serum creatinine, the presence of CF related diabetes, NSAID use, and vancomycin exposure. Clinical data was obtained by electronic chart review.

Results: Among the 98 CF patients measured (mean age of 23.8 years; 8-63 years); 54% were female, 15% were diabetic; 64% culture positive for Pseudomonas aeruginosa. Twelve percent were classified as severe lung disease, (FEV1 < 40%), while 63% had mild lung disease (FEV1 > 75%). Among the non-CF patients (mean age of 15 years; 8-28 years); 44% were female, none were diabetic; and 94% had mild lung disease such as asthma. Mean baseline SCr was 0.66 mg/dL (range: 0.5-1.5 mg/dL) among all CF patients and all within normal range for age. Five of 8 patients actively receiving IV aminoglycosides had elevated KIM-1 levels (> 1.0 ng/mg creatinine) and all of these subjects had significant proteinuria (>0.2 mg/mg creatinine). KIM-1 levels were significantly increased compared to non-CF controls for all patients (p<0.001) and though it did not achieve statistical significance, this trend was reversed when comparing aminoglycoside naïve patients with non-CF controls. Elevated KIM-1, NAG and total protein levels in urine were significantly associated with disease severity (p <0.001) in the CF patients. Only NAG levels were significantly elevated in females in comparison to males (p<0.001), though mean levels for all three UBMs were increased in women. There were no age related differences in UBM levels.

Conclusions: Our data shows that urinary KIM-1, NAG, and total protein are detectable UBM's in pediatric and adult CF patients. Urinary KIM-1 and total protein levels were elevated in patients receiving aminoglycosides, even in the absence of elevation of serum creatinine. In addition, we have demonstrated that random KIM-1 and NAG levels are directly associated with disease severity. We expect that further characterization of UBMs will clarify their utility in detecting kidney injury in CF. Research and Treatment Center, UNC at Chapel Hill, Chapel Hill, NC, USA In CF, adhesion between dehydrated luminal secretions and surface epithelia is thought to underlie the formation of mucus plaques and plugs. As shown in CF epithelial cultures, releasing adhered mucus plaques from epithelial surfaces is extremely difficult with rehydration alone. The use of a therapeutic aerosol with detergent-like properties, therefore, was postulated to reverse mucus adhesion by interrupting the molecular interactions that bind mucus secretions to airway surfaces. Inhaled surfactants are a therapeutic candidate that may fulfill this need. Previous studies with surfactant preparations have yielded improvements in mucus transport in a variety of model systems. Based on these findings, an ongoing single center clinical trial was designed to test the efficacy and safety of nebulized Lucinactant (Discovery Laboratories, Inc.). Lucinactant is a wholly synthetic preparation that includes the lipid constituents and the KL4 peptide, which mimics SP-B functions.

Study Design: In this double blind, placebo controlled crossover study, eligible CF patients are at least 14 years of age and have an FEV 1 of ≥40% of predicted. Subjects must have stable lung disease and be able to discontinue use of hypertonic saline and dornase alfa for at least 3 days prior to and during each treatment period (TP).

After passing screening, baseline measures of mucociliary clearance and cough clearance (MCC/CC), lung function and safety assessments are performed. Subjects then receive, in random order, either 5 doses of Lucinactant (6 ml of 20 mg/ml total phospholipid content) or placebo (6 mL of 0.9% NaCl) during a 24 hour period. All study medications are delivered via the eFlow® nebulizer system (PARI Pharma). After a 2-week washout period, subjects cross over to the opposite treatment arm in the second treatment period (TP). To evaluate efficacy, MCC/CC measured via gamma scintigraphy provides the primary outcome measures. Other efficacy assessments include spirometry, Lung clearance index via multiple breath N 2 washout technique, quality of life (CFQ-R) and sputum rheology.

Results: To date, 12 subjects (6 females), have been enrolled and received at least one dose of study drug. The mean age is 26.9 yrs. The mean FEV 1 is 77% pred. Most patients were chronic users of hypertonic saline (75%) and dornase alfa (67%). One patient was withdrawn after the first dose of study medication due to an asymptomatic, 18.6% drop in FEV 1 , according to protocol stopping criteria. No other treatment emergent adverse events attributable to study medication have been encountered, and there have been no serious adverse events. A blinded assessment of drug tolerability reveals little change in FEV 1 15 minutes after the first dose of study medication (-5.84 ± 1.72% in TP1; -5.63 ± 3.56% in TP2). The FEV 1 change measured 3 hours after the 5th dose of study medication during TP1 and TP2 were +0.94 ± 1.41% and -0.68 ± 1.34%, respectively.

Four additional patients will be enrolled in the trial, for a total of 16 patients receiving study drug. We anticipate completing the trial and unblinding data in July 2010.

Supported by the CFF. Cystic fibrosis (CF) is an autosomal recessive disease caused by mutations in the CF transmembrane conductance regulator (CFTR) gene, which encodes an epithelial chloride channel that regulates ion and fluid transport in multiple organs, including the lung. The F508del-CFTR mutation is the most common CF-disease causing mutation and is present in over 70% of CFTR alleles. This mutation results in impaired intracellular trafficking and reduced cell-surface expression of the F508del-CFTR protein. Such mutations are known as Class II mutations. Class I mutations, such as W1282X and G542X, are less common and result in a total lack of protein synthesis due to premature stop codons. Class III mutations (e.g. G551D) lead to the pared to 10.6% of normal in the placebo group (P<0.05 for the 250-mg group vs. placebo). In the 150-and 250-mg groups, Ringer's PD improved to 97.9% and 127.3% of the PD in normal subjects, compared to 182.9% of normal in the placebo group (P<0.05 for 150-mg group vs. placebo). In the 150-mg group, amiloride response was 89.3% of normal compared to 151% in placebo (P<0.05). This study demonstrated that Na + and Clconductance in HBEs treated with VX-770 closely corresponded to the in vivo response to VX-770 measured by NPD. HBE derived from CF subjects may be a useful model to predict the efficacy of CFTR ion transport modulators in development.

This study was sponsored by Vertex Pharmaceuticals, Inc. with support from CFFT. FoldRx Pharmaceuticals has developed a novel yeast-based, highthroughput platform designed to identify small-molecule modulators of protein trafficking. Yeast is a well-recognized model system, and many of the proteins involved in protein trafficking and quality control in yeast have well-conserved counterparts in humans. An endoplasmic reticulum (ER) to Golgi transport assay in yeast was used to screen a diverse compound library and the screening hits were tested for their ability to correct and potentiate F508del CFTR in FRT cells.

Several families of dual-active compounds having both corrector and potentiator activities emerged from the FRT assays, as well as pure correctors and potentiators. Three families of compounds have shown both corrector and potentiator activities in primary bronchial epithelial cells from CF patients (CF hBE), demonstrating the successful discovery of trafficking activators in yeast and the translation of the activity into CF hBE cells.

Several different mammalian cell lines have been used to develop assays for F508del CFTR trafficking and have identified a variety of corrector compounds (1, 2) . However, the efficiency of finding relevant correctors in high throughput screens using any one mammalian cell line is hindered by two issues. First, many active correctors in these cell lines fail to show activity in CF hBE cells. Second, different cell lines identify different sets of correctors (3) . Because of these discrepancies, we sought an additional cell line to test our hits from our primary screen (yeast ER to Golgi trafficking) with the goal of identifying novel compounds not detected by the FRT assay.

Baby Hamster Kidney (BHK) cells have been used for screening for modulators of F508del trafficking (4) . Iodide flux assays using halide-sensitive yellow fluorescent protein (YFP) have been adapted for several cell types (3) and have been used for high throughput screening to identify correctors and potentiators (2) . Here we report an adaptation of the iodide flux assay for use in BHK cells. Cells were transiently co-transfected with plasmids encoding for F508del CFTR and halide-sensitive YFP, transfected cells were treated with test compounds overnight, and the extent of iodide influx was measured the next day. Good assay characteristics were achieved, with Z' values > 0.5. Several correctors showed activity, including VRT-325 and corrector 4a, as well as FoldRx correctors. Some correctors not detected in the FRT assay were active in the BHK assay, including Daunorubicin, Doxorubicin and Ciclopirox, demonstrating proof-of-concept for the discovery of different correctors than the FRT assay. We conclude that the BHK iodide flux assay offers a new opportunity to discover novel modulators of F508del CFTR trafficking and activity, and we are currently moving forward with testing hits from our yeast primary screen in this assay.

This work was supported by CFFT. We thank Dr. L. Galietta & Prof. W. Balch for providing reagents and cells. production of proteins which reach the plasma membrane; however, their regulation is defective. Class IV mutations (e.g. R117H) are associated with altered channel conductance. Class V mutations (e.g. 2789 + 5G Ç A) lead to the production of normal proteins, but in reduced levels. VX-770 is an orally bioavailable agent that aims to potentiate the gating of CFTR channels. Previous studies demonstrated that VX-770 increased the gating activity of the G551D and F508del forms of CFTR. The current in-vitro study aimed to evaluate whether VX-770 could potentiate the activity of more than 50 different mutant forms of CFTR expressed in recombinant Fischer rat thyroid (FRT) cells and in cultured human bronchial epithelial (HBE) isolated from the bronchi of CF donors carrying different mutant CFTR forms. VX-770 was evaluated in FRT cells expressing 8 Class III mutations (e.g., G178R, G551D, G551S, G1349D). In this group, forskolin (10 µM; EC 99 )-stimulated chloride transport was increased by more than 10-fold following acute addition of VX-770. In FRT expressing Class IV mutations (e.g., R117H, D1152H, I148T), VX-770 potentiated 10 µM forskolin-stimulated chloride transport to a lesser degree than in the Class III mutations. Cystic fibrosis (CF) is caused by mutations in the CFTR gene that encodes the CFTR protein, an epithelial Clion channel. VX-770, an investigational CFTR potentiator, aims to increase the ion channel activity of CFTR at the cell surface. In in-vitro studies, forskolin-stimulated primary F508del/G551D-HBE cells were treated with VX-770 10 µM. Cltransport increased by a mean of 4.3 mV and potential difference (PD) prior to amiloride decreased by a mean of 21.9 mV, suggesting reduced hyperabsorption of Na + . In CF patients, Na + is hyperabsorbed through ENaC in comparison to non-CF individuals. In the presence of forskolin, the baseline PD improved from 300% to 210% of wild-type with application of VX-770 (P<0.01). A Phase 2 randomized, double-blind, placebo-controlled study of VX-770 was conducted in 39 CF subjects with a G551D-CFTR allele. Safety and efficacy results were reported previously. Statistically significant improvements from baseline in Clconductance as measured by NPD were seen after 14 days. In the 150-mg group (n=16), the mean change in zero Clplus isoproterenol response was -4.6 mV (P<0.01 within-subject, P<0.05 vs placebo). The mean change was -7.6 mV in the 250-mg group (n=7; P<0.05 within-subject and vs placebo). A significant increase in Ringer's PD was observed in the 150-mg group, suggesting a reduced hyperabsorption of Na + . In the 150-mg group (n=16), the starting basal PD increased by 13.0 mV (P<0.001 within-subject) compared to a 2.7 mV decrease in the placebo group (P=not significant within-subject for placebo and P<0.05 for 150mg vs. placebo). This suggests CFTR functional improvements were accompanied by reduced ENaC activity. As previously reported by Van Goor et al., the restoration of Cltransport to ~50% wild-type activity in primary F508del/G551D-HBE was similar to the change in Clconductance, measured by NPD, observed in the study of VX-770 in CF subjects with a G551D mutation. At day 14 in the 150-mg and 250-mg groups, the zero Clplus isoproterenol response was 29.7% (n=16) and 40.6% (n=7), respectively, of the normal (non-CF) response derived from a multicenter NPD dataset, com-

Rietschel, E. 1 ; Staab, D. 2 ; Merkel, N. 3 ; van Konigsbruggen, S. 1 ; Posselt, H. 4 1. CF Center, University Children's Hospital, Cologne, Germany; 2. CF Center, Children's Hospital Charité, Berlin, Germany; 3. University Children's Hospital, Halle, Germany; 4. University Children's Hospital, Frankfurt, Germany Introduction: Twice daily (b.i.d.) inhalation of TOBI™ (300 mg/5 mL) in a 28-day on/off-dosing regimen has previously proven its safety and efficacy for the treatment of Pseudomonas aeruginosa (P.a.) in CF subjects.

Method: In this randomized multicenter, open label, two period crossover study pharmacokinetic (PK) data of 8 weeks continuous treatment were assessed in CF patients ≥ 6 years chronically infected with P.a. Following Screening patients were randomized to the two possible sequences with inhalation of 1 x 300 mg/d and 2 x 300 mg/d TOBI™ via PARI eFlow™ rapid for 8 weeks. The primary endpoint was serum PK of tobramycin (AUC0-90'). Further objectives were safety, change in MIC of P.a. and lung function.

Results: Twenty-nine patients were randomized, 24 patients completed both treatment regimens. Patients were between 8 and 35 years old, mean (SD) age was 19.8 (6.3) Rationale: Inhaled hypertonic saline (HTS) has been shown to improve quality of life and reduce pulmonary exacerbations when given long term in stable patients with cystic fibrosis (CF). While it is increasingly being offered for acute pulmonary exacerbations, little is known about the efficacy in this setting. In this retrospective study, we examined the prevalence of HTS use among a cohort of adult patients hospitalized with a CF lung disease exacerbation and examined its tolerability and efficacy.

Methods: Using retrospective chart review, we examined clinical outcomes and compliance with HTS in 45 subjects who were admitted to the inpatient service for acute pulmonary exacerbation between January 2006 and December 2007 (visit 2). In a subset of this group consisting of 18 subjects who were also admitted the previous year (visit 1) when HTS was not offered, we compared changes from baseline in FEV 1 and weight, and time to next exacerbation after each of the 2 visits.

Results: Mean age of the cohort was 32.5 years and mean length of stay was 11.5 days. HTS was offered to 33 subjects and was well tolerated for a total use of 336 days out of 364 days of hospital stay. Wheezing and hemoptysis shortly after initiation of treatment were the reasons for early discontinuation in 2 subjects. Baseline demographics, lung function and sputum culture results were comparable in both visits. Use of HTS was not associated with an improvement in FEV 1 (p=0.1), weight gain (p=0.24) or in the time to next admission (p=0.08).

The data in this study suggest HTS is well tolerated during CF pulmonary exacerbation but offers no clear outcome benefits. It is possible that HTS may not have much advantage above and beyond intensive rehabilitation and intravenous antibiotics and may just add to hospital costs. Further prospective studies are needed to justify HTS in the inpatient setting. Oxylipins are a group of oxidized metabolites of polyunsaturated fatty acids including the cyclooxygenase (COX), lipoxygenase (LOX) and cytochrome P450 (CYP) derived metabolites of polyunsaturated fatty acids. These oxylipins play an important role in regulating cell proliferation, apoptosis, tissue repair, blood clotting, blood vessel permeability, inflammation, immune cell behavior and numerous other biopathologies. Therefore, a comprehensive and robust oxylipin profiling method is vital to advance research in the field of regulatory lipids. In previous studies, we have developed a sensitive quantitative measurement method for determining the oxylipins from plasma and tissues using liquid chromatography tandem mass spectrometry (LC/MS/MS) and extraction methods.

Sputum is a very useful, noninvasive and feasible specimen for characterization of the respiratory tract inflammation. Since there are intense airway inflammatory processes in CF, determinations of sputum lipid mediators should represent a useful methodology for quantitating the rate of bioactive lipids in CF and possibly suggest novel therapeutic approaches.

In the present studies, we have developed an effective extraction method for oxylipins from CF sputum and used this technique to assess the oxylipin profile of 18 adult CF patients' sputum samples.

Methods and Results: Compared to plasma, BALF and urine samples, sputum samples have a considerably more complex matrix effect to overcome. To assess recoveries, we spiked 10 deuterated oxylipins to freshly acquired sputum samples. In our initial attempts following a plasma extraction protocol, most of the deuterated oxylipins recovery rates were around 10%. With improved modified methods for extraction we have improved recoveries to approximately 60%.

Using the current extraction protocol and oxylipin profiling method, many biologically active pro-inflammatory and anti-inflammatory lipid mediators were detected. They included PGE2, PGD2, TXB2, LTB4, 6trans LTB4, 20 OH LTB4, 20 COOH LTB4, 20 HETE, 15 HETE, 11 HETE, 12 HETE, 8 HETE, 9 HETE, 5 HETE, EpETrEs, diols, Resolvin, LXA4 . Most of them have not been reported previously in CF sputum.

Conclusions: A promising oxylipins extraction method for sputum samples has been developed. Comparison of the different extraction methods indicates most of oxylipins stay in the pellet fraction of sputum. Liquid-liquid extraction has the highest extraction efficiency. Using the current extraction protocol and oxylipin profiling methodology, many biologically active pro-inflammatory and anti-inflammatory lipid mediators have been detected. Most of them have not been reported previously in CF sputum. Further characterizations and quantitation of bioactive lipids in CF secretions will be helpful towards further understanding of inflammation in CF and have potential diagnostic and therapeutic implications. Acknowledgement: This work was supported in part by NIEHS SBRP Grant P42 ES004699, NIEHS Grant R37 ES02710 and NIH/NIEHS R01 ES013933. Jun Yang was supported by the Elizabeth Nash Memorial fellowship from the Cystic Fibrosis Research, Inc. AAV is the most promising viral vector to develop successful gene therapy for CF, since feasibility and safety have been demonstrated without pathogenicity. Inflammation, known to be minimal, is not absent: in fact the various serotypes have been recognized as potentially impacting gene expression. The viral capsid is thought to trigger the innate immune response through pattern recognition receptors, prior to induction of the adaptive immune response. We hypothesized that aerosolized gene delivery would perturb gene expression, allowing the identification of those pathways involved in the modification of gene expression.

In a preclinical trial of which AAV serotypes were best suited for human airway epithelium, doses of 1x10 13 vg of AAV1 firefly luciferase and rAAV5 renilla luciferase were simultaneously administered to the airway lumen of adult chimpanzees (N=4) under general anesthesia utilizing a microsprayer via fiberoptic bronchoscope. Pre-and post-treatment serum samples were taken at sequential time points, day -2, 0, 8 hours, and 24 hours following gene delivery. Samples were pooled and run on duplicate monoclonal antibody arrays, Clontech AB Microarray 500, covering 507 proteins. Proteins used in the analysis were both significantly different at p ≤ 0.05 and having 20% or greater change in expression. At 8 hr, 29 proteins were significantly different and at 24 hr, 55 proteins were significantly differentially expressed; 16 proteins remained common to both time points. Gene Ontology, protein interaction network and pathway analysis was performed with MetaMiner(CF). ELISA detection of multiple cytokines was also performed to generate an immune profile of each animal.

There was an enrichment of apoptosis controlling cell cycle checkpoint proteins without overt indices of inflammation. Gene Ontology analysis indicated a significant enrichment of proteins involved in cell cycle regulation, specifically the G1/S transition checkpoint at both 8 and 24 hour; apoptosis regulating proteins were significantly enriched at 8 hr. At 24 hr IL1 and CASP4 were downregulated while TNF and RAB24 were upregulated.

Overall, the antibody microarray expression pattern of known secreted proteins at 8 hr suggested a mild innate immune response directed toward apoptosis or cell cycle inhibition without an overt change in the expression of traditionally recognized immune inflammatory markers such as IL6 or IL8. Both cell cycle inhibition and apoptosis are known effects of AAV therapy; such pathways can potentially modify the efficacy of gene expression.

Funded by the CFF and NHLBI. Medicine, Royal Brompton Hospital, London, United Kingdom; 2. Royal Children's Hospital, Melbourne, VIC, Australia; 3. Children's Hospital Westmead, Sydney, NSW, Australia; 4. Respiratory Services, Auckland City Hospital, Auckland, New Zealand; 5. St.Vincent's University Hospital, Dublin, Ireland; 6. Pharmaxis, Sydney, NSW, Australia There continues to be an unmet need for effective long term treatments in patients with CF. Inhaled dry powder mannitol (Bronchitol™) is an osmotic agent formulated to deliver a high osmotic load to the airways in a short time. The initial 26 wk double blind (DB) trial of mannitol (400 mg bid) in 295 CF patients (6-56 yrs) showed an improvement in FEV 1 by 6 wks which was maintained to 26 wks (118 mL ∆ baseline to wk 26, p<0.001) (1, 2) . The benefit in FEV 1 was seen irrespective of concurrent rhDNase (108.8 mL ∆ compared to control at wk 26, p=0.002) (1, 2) . The open label (OL) study was designed to further assess the safety and efficacy of mannitol (M) up to 18 mo.

Methods Results: At wk 52 a significant (p<0.001) improvement of 148.5 mL (from baseline) in FEV 1 was achieved by those C patients commencing on M in the OLP, while those on M during the DB phase continued to maintain an improvement in FEV 1 out to 18 mo (Table) .

One hundred fifty-two subjects (89.4%) experienced an adverse event (AE) during the OL phases (M 85.6%, ex-control (C) 94.5%). Nine subjects experienced a treatment related serious AE (M 4.1%, C 6.8%). The most commonly reported treatment related AEs (>5%) for mannitol (M and C respectively) included: condition aggravated (5.2%, 2.7%), cough (3.1%, 8.2%), haemoptysis (3.1%, 8.2%). Haemoptysis and bronchospasm led to early withdrawal of 2 subjects and 1 subject respectively.

Conclusions: The benefits seen with M during the 26 wk DB phase are sustained to at least 18 mo. The AE rate for the first 6 mo on mannitol was similar between the DB phase and the patients randomised to control who received mannitol in the OLP. The OL phases reinforce the efficacy and acceptable safety profile of mannitol structures containing palmitate (C16 fatty acid) and aminoarabinose indicating unique recognition of the CF airway environment. These modifications are associated with resistance to cationic antimicrobial peptides and increased inflammatory responses, indicating that they are likely to be involved in airway disease.

The microenvironment in the CF airway has been shown to be oxygenlimited within mucus plugs. Therefore, this environment may apply a selective pressure to PA that promotes the synthesis of CF-specific lipid A modifications, which include the addition of palmitate. To determine if this environmental condition had the ability to regulate specific LPS structural modification in PA, we examined the lipid A profiles from a variety of PA strains after growth under microaerophilic and anaerobic conditions. We report that under these oxygen-limited growth conditions, laboratory adapted and environmental strains and acute disease clinical isolates were induced to add palmitate similar to those constitutively present in CF clinical isolates. Additionally, we showed that this is a PA specific process, as other bacterial species such as E. coli, S. typhimurium or Yersinia subspecies strains do not add palmitate after growth under oxygen-limited conditions. Finally, we show that the addition of palmitate is regulated by the PhoP/PhoQ two-component regulatory system. PhoP/PhoQ regulates lipid A modifications in many bacterial backgrounds in response to limitation of divalent cation, antimicrobial peptides, and low pH. This is the first report of this regulator responding specifically to oxygen limitation. This work was supported by grants from the National Institutes of Health (AI47938) and the CF Foundation. Introduction and Aims: The majority of patients with cystic fibrosis (CF) die from chronic opportunistic infection of the lungs. Numerous studies have shown that CF airways are infected with complex microbial communities and that pathogens persist through multiple rounds of antibiotics. As many microbes are not detected by conventional culture methods, molecular detection techniques offer a more complete view of the nature of CF infections. The objective of this study was to characterize the CF lung microbiome and to determine the effect of antibiotic treatment on bacterial diversity and abundance during exacerbation.

Methods: Sputum and oral lavage samples were collected from 23 CF patients at the beginning and completion of antibiotic treatment of exacerbations and when clinically stable. Pyrosequencing and Terminal-Restriction Fragment Length Polymorphism (T-RFLP) analysis were used to assess microbial community diversity within the samples. Total bacterial abundance was measured by quantitative PCR. Microbial community diversity was compared with markers of disease severity such as lung function (FEV 1 ).

Results: Both pyrosequencing and T-RFLP analysis produced a similar view of the micrbobial communities; however, pyrosequencing was far more sensitive and allowed a more detailed examination of microbial community changes. Our data indicate that: (i) CF airway microbial communities clustered into two distinct groups, those dominated by Burkholderia and those dominated by Pseudomonas and Streptococci. Additional species were associated with the distinct communities, suggesting that there are common subtypes of infection that are shared across patients. (ii) Treatment with antibiotics does little to perturb the structure of CF airway microbial communities; however, the relative abundance of specific taxa are temporarily reduced. (iii) Microbial community diversity varies between CF patients and reduced diversity correlates strongly with decreased lung function. In contrast, changes in total bacterial abundance do not account for the changes in lung function. (iv) Reduced microbial diversity in oral lavage samples also tend-ed to be associated with reduced lung function, suggesting that such samples may be a useful biomarker of health.

Conclusions: Our results suggest that exposure to antibiotics provides temporary relief of symptoms associated with exacerbation but at the possible cost of decreasing overall microbial diversity. Over time a small number of taxa predominate causing more severe clinical symptoms. While the presence of Burkholderia was usually associated with a low FEV 1 , patients with Burkholderia that had higher overall microbial diversity tended to have a higher FEV 1 . Similar trends were seen with Pseudomonas. These data argue that microbial diversity may be a useful clinical parameter in addition to the usual assignment of pathogen identity. Many bacterial species can cause either acute or chronic infection, depending on host conditions. Pseudomonas aeruginosa is a prime example. P. aeruginosa can cause life-threatening infections in some hosts, while in other conditions, such as cystic fibrosis (CF), bacteria can persist within the airway for decades and never produce a disseminated infection. In chronic infections, P. aeruginosa becomes extremely tolerant to antibiotics and host defenses, even though the bacteria are susceptible when tested ex vivo. We hypothesized that the physical and biochemical properties of sputum may promote P. aeruginosa chronic infection by making the bacterium resistant to killing and less injurious to the host.

Because sputum is a gel-like environment, we assessed the growth of P. aeruginosa in agar gels of varying viscosity by adjusting agar concentrations. At low agar concentrations, P. aeruginosa grew diffusely throughout the gel. When agar concentrations were increased, P. aeruginosa grew as aggregates within the gel. Importantly, organisms grown in aggregates werẽ 100-fold more tolerant to killing by several anti-pseudomonal antibiotics. To determine if limited motility could produce aggregated growth and tolerance to killing in the absence of high gel viscosity, we tested isogenic P. aeruginosa mutants defective in swimming motility. These mutants formed aggregates in low viscosity gels that were tolerant to antibiotics and host defenses like human neutrophils and neutrophil-derived oxidants.

To investigate whether sputum components could contribute to aggregate formation, we added supernatants prepared from CF sputum to gels, and found that aggregates tolerant to antibiotics, human neutrophils, and oxidants formed even if gels were composed of low viscosity agar. We further determined that aggregate formation required the activity of neutrophil serine proteases, which limited P. aeruginosa swimming motility.

Because bacteria causing chronic infections generally exhibit reduced invasiveness, we investigated whether aggregated growth would impair the ability of bacteria to disrupt an epithelial barrier. P. aeruginosa growing diffusely in gels lacking sputum components rapidly injured airway epithelia as determined by transepithelial resistance measurements. In contrast, P. aeruginosa aggregates grown in CF sputum gels produced far less epithelial injury. These data show that both the physical and biochemical properties of CF sputum may contribute to growth of P. aeruginosa in aggregates. Formation of aggregates markedly increases tolerance to host defenses and antibiotics, reduces injury to epithelia, and possibly promotes the development of chronic airway infections in CF patients. To determine whether individual amino acids could influence the levels of c-di-GMP, a hyper-swarming strain with mutations in two diguanylate cyclases (sadC and roeA), involved in the synthesis of c-di-GMP, was tested on arginine. The sadC roeA mutant strain was able to restore swarming on arginine unlike the wild type, indicating that repression of swarming by arginine requires the synthesis of c-di-GMP and further suggesting that arginine can modulate c-di-GMP levels. To further confirm this observation, the double sadC roeA mutant was shown to act as a hyper-swarmer for every single amino acid, including methionine and phenylalanine, two amino acids that normally repressed swarming. Furthermore, the sadC roeA mutant, known to be deficient in biofilm formation due to its lower level of c-di-GMP, was also reduced in biofilm when grown with every individual amino acid tested. Finally, our initial biochemical analysis demonstrated that P. aeruginosa PA14 cells grown in the presence of arginine had elevated levels of intracellular c-di-GMP compared to cells grown with casamino acids in which swarming occurs.

In conclusion, our data suggest that individual amino acids can modulate the intracellular levels of c-di-GMP, and therefore impact P. aeruginosa behaviors directly regulated by this signaling molecule. Mortality in cystic fibrosis is most commonly a result of persistent bacterial infection in the lungs due to reduced mucociliary clearance. As the number of options for the treatment of persistent infection has grown, it has become unclear which therapy is optimal. This uncertainty partly reflects the absence of a good model of chronic airways infection consequent to reduced mucociliary clearance in which to test potential therapies. To address this problem, we have developed a technique in which an infection with Pseudomonas aeruginosa, an organism commonly found in CF lungs, can be established and chronically maintained in the lungs of mice overexpressing the β subunit of the epithelial sodium channel scnn1 (βENaC). βENaC and wild type (wt) mice were intratracheally challenged with a small volume (15 µL) of low melting point agarose containing a low dose (1 x 10 4 cfu) of Pseudomonas aeruginosa. By 21 days after the challenge, nearly 90% of wt mice had completely cleared the Pseudomonas challenge, and those wt mice still infected exhibited an average bacterial load of less than 50 cfu/mouse. However, more than 85% of the challenged βENaC mice maintained an active infection with Pseudomonas throughout the course of the challenge, with an average load of 6 x 10 3 cfu/mouse. The bacterial load in βENaC mice did not significantly change through the course of the experiment. The infection in βENaC mice led to a chronic mucopurulent obstructive inflammatory response in the airways characterized by neutrophil and lymphocyte infiltration, as well as mucus overproduction leading to obstruction of the airways. This model will allow the investigation and comparison of clinical treatment regimens that clear mucus from obstructed airways, treat bacterial infections, or combine the two strategies. The hemolytic phospholipase C, PlcH, of Pseudomonas aeruginosa (P.a.) is an important virulence factor that is highly induced during lung infection in both animal models and P.a. infected CF sputum. In animal models this induction is dependent on the transcription factor GbdR. We have previously reported that PlcH directly alters pulmonary surfactant leading to closure of the small airways. While directly targeting PlcH with therapeutic agents is under study, we have data that stimulation of GbdR also induces other potential virulence factors. In addition, catabolism of host choline or carnitine provides glycine betaine (GB) which is a potent osmo-exposed to different selective pressures, genetic diversity may evolve in vivo in the infecting population. The emergence of extensive diversity is made more likely by the fact that infecting populations are large, mutator strains arise, and environmental conditions within the CF lung are heterogeneous.

To test the hypothesis that P. aeruginosa infecting populations are diverse in CF, we isolated several hundred randomly selected colonies from sputum samples submitted by three CF patients over five consecutive days. We confirmed that the isolates diverged from a common ancestral strain using random amplification of polymorphic DNA fingerprinting. Patients were studied during periods of disease quiescence (i.e. not during pulmonary exacerbations). The P. aeruginosa isolates were subjected to a battery of phenotypic tests that measured heritable traits. This analysis showed that extensive genetic diversity exists in the P. aeruginosa populations present in each sputum sample, and that abundance of subpopulations differed within samples collected on consecutive days.

In natural ecosystems, spatial heterogeneity of environmental conditions and the geographic isolation of organisms are key driving forces for the evolution of diversity. Spatial heterogeneity certainly exists in CF lungs as areas of severe, moderate, and mild disease co-exist in most patients. In addition, lung topography, the inactivation of motility genes, and constraints on bacterial movement imposed by viscous secretions could lead to subpopulations being geographically isolated. These points led us to hypothesize that the composition of P. aeruginosa populations varies in different lung regions. To test this, we sampled explanted lungs from three CF patients removed at the time of lung transplantation. Several hundred P. aeruginosa colonies from both peripheral lung parenchyma and major lobar airways were isolated and subjected to the phenotypic tests described above. These experiments showed that different anatomic regions harbored distinct P. aeruginosa subpopulations.

This study demonstrates that a diverse infecting population of P. aeruginosa evolves within the lungs of CF patients. Furthermore, analysis of explanted lungs suggests that regional differences in environmental conditions and/or geographic isolation of subpopulations may contribute to the evolution of the population diversity. Our findings have implications for CF lung disease as diversity generally increases the stress resistance of biological populations, and their ability to exploit available resources. If these principles apply to the P. aeruginosa populations infecting CF patients, diversity may enhance the ability of bacteria to persist in the face of host defenses and clinical treatments. Bacteria can sense a diversity of signals regulating many of their functions allowing them to adapt to changing environments. Pseudomonas aeruginosa, the most important bacterial pathogen encountered by the cystic fibrosis (CF) lungs, faces a wealth of signaling molecules in this environment. The CF sputum is an ideal niche to support P. aeruginosa growth allowing the establishment of chronic infections. To support high-density growth, the CF sputum contains abundant amino acids that can be used as carbon sources. Although P. aeruginosa can catabolize only a few amino acids, it was shown that it could sense and respond to a variety of individual amino acids. However, such studies often used non-physiological amount of amino acids, making conclusions difficult to extrapolate to the host environment. However, the concentrations of free amino acids in CF sputum were recently quantified allowing for more relevant in vitro studies.

In this study, we first determined the impact of amino acids on two P. aeruginosa surface behaviors: biofilm formation and swarming motility. Using relevant in vivo concentrations of 19 amino acids, we demonstrated that each amino acid differentially impacted the ability of P. aeruginosa strain PA14 to form biofilms when used as a sole carbon source. Although some amino acids, such as arginine, did not support robust growth, they still promoted the formation of biofilms in a dose-dependent manner. On the other hand, most amino acids could support swarming with the exception of arginine, methionine, and phenylalanine. We previously demonstrated that the level of the intracellular signaling molecule c-di-GMP had opposing effects on biofilm formation and swarming motility. In particular, high lev-protectant and stress protectant that may increase P.a. survival in the lung. In this study, we report a rationally designed small library of choline analogues to dissect the roles of the quaternary amines in regulation of P.a. virulence and to generate lead compounds that inhibit one or more steps in the catabolic pathway.

Most of the choline analogues were synthesized by addition of an appropriate bromoalkane to dimethylethanolamine; other syntheses are detailed in the presentation. We exposed P.a. to the analogues in the presence and absence of choline and measured growth, osmoprotection, and plcH induction. RNA was isolated from P.a. treated with the choline analogues and processed for Affymetrix microarray.

The choline analogues possessed specific functions consistent with the length and bond structure of the added functional group. Addition of two carbons altered but did not eliminate the osmoprotectant function of the analogue. Addition of a two or three carbon group did not impact transport, but these analogues could not be used as sole carbon sources. We generated two specific inhibitors of GB catabolism, propargylcholine and aminocholine. We also generated compounds that could induce BetI and/or GbdR-dependent transcripts but were not catabolized as carbon sources, including diethylcholine, ethylcholine, and pyrrolidinylcholine. Using these compounds in a chemical genetic approach, we have dissected the multiple roles of choline and its catabolites on the P.a. transcriptome.

Choline analogues have enabled us to dissociate plcH induction from choline catabolism, as well as dissociate the effects of osmoprotection from cellular energetics. These molecules are valuable tools to probe the mechanisms underlying the regulation of P.a. virulence by choline and its catabolites.

This work was supported by NIH NCRR CoBRE P20-RR021905 and the Cystic Fibrosis Foundation (WARGO09F0). Objective: To determine the bacterial community present in oropharyngeal (OP) samples from children with CF during periods of clinical stability and pulmonary exacerbation, and to compare these findings to the bacterial community detected in healthy children.

Methods: Oropharyngeal samples were obtained from children with CF and healthy children. Samples from CF subjects were collected during clinical stability and at treatment onset with oral or intravenous antibiotics for a pulmonary exacerbation. All samples underwent microbial culture and quantitative real-time PCR (qPCR) assays for total bacterial load, three CFassociated pathogens (Pseudomonas aeruginosa, Staphylococcus aureus and Haemophilus influenzae), and five anaerobic bacteria (Prevotella denticola, Prevotella melaninogenica, Prevotella oris, Peptostreptococcus and Fusobacterium). A subset of samples was analyzed by parallel pyrosequencing. Diversity indices were determined for each sample analyzed by parallel pyrosequencing. Results were compared between three groups consisting of stable CF, exacerbated CF, and healthy controls (HC).

Results: Twenty-one CF subjects (10 male, median age 2.7 yrs, range 0.3-4.9 yrs) and 20 HC (11 male, median age 1.0 yr, range 0.4-3.4 yrs) were enrolled. Fifty-one OP samples (21 stable CF, 10 exacerbated CF, and 20 HC) were collected. Microbial culture and qPCR assays were performed on 51 samples, and parallel pyrosequencing was performed on 35 samples. The most common bacteria detected by culture were S. aureus (38% stable CF, 50% exacerbated CF and 50% HC) and H. influenzae (24% stable CF, 30% exacerbated CF and 35% HC). P. aeruginosa was detected in one HC sample. Sequencing revealed that HC samples were 1.4 times more diverse than exacerbated CF and 1.1 times more diverse than stable CF, although these differences were not statistically significant. Sequence counts of Capnocytophaga, Fusobacterium and Prevotella were increased in exacerbated CF compared to stable CF, whereas Haemophilus, Leptotrichia and Moraxella were decreased in exacerbated CF samples (p< 0.05). There was no differ-ence in total bacterial load by group but qPCR counts for H. influenzae, P. denticola and P. melaninogenica were lower in both CF groups compared to HC samples while P. aeruginosa and S. aureus were higher (p < 0.05). Quantities of Peptostreptococcus and P. aeruginosa measured by qPCR were increased in exacerbated CF compared to stable CF samples (p <0.05). P. aeruginosa was detected in eight CF samples by qPCR despite negative microbial cultures.

Conclusions: Molecular analyses using qPCR and sequencing reveal qualitative and quantitative differences in oropharyngeal bacteria between stable and exacerbated CF and healthy children. In particular, there are shifts in the anaerobic community within the CF group that deserve further exploration. In addition, qPCR may be more sensitive for identifying P. aeruginosa. The role of Candida albicans in the cystic fibrosis (CF) airway remains undetermined. Considered a colonizer, few question its pathogenic potential despite high isolation frequencies from sputum culture. We evaluated the frequency and identified the strongest predictors of C. albicans colonization in CF. Independent association of colonization with clinical outcomes were determined and the longitudinal effects of C. albicans acquisition on body mass index (BMI) and forced expiratory volume in the first second (FEV1) evaluated.

Methods: A prospective observational study of 89 CF patients was performed (3,916 sputum samples over 11 years). Frequency of C. albicans growth in sputum allowed classification of the cohort into colonizers and non-colonizers. BMI, FEV1, hospital-treated exacerbations and other clinical parameters were followed throughout the study to determine association with colonization status. Multivariate regression determined the strongest predictors of colonization and for clinical effects after adjustment for confounders. Repeated measure ANOVA assessed the longitudinal effect of colonization on BMI and FEV1.

Results: Colonization with C. albicans was frequent (49.4%) and best predicted by pancreatic insufficiency (p=0.014), osteopenia (p=0.03) and co-colonization with Pseudomonas spp. (p=0.002). C. albicans colonization significantly predicted hospital-treated exacerbations (p=0.004) after adjustment for confounders. Exacerbation rate significantly increased in the chronically or intermittently colonized following first acquisition of C. albicans. Colonization accelerated rates of decline for BMI (p<0.0001) and FEV1 (p<0.001).

Conclusion: Airway colonization with C. albicans presaged a greater rate of FEV1 decline and hospital-treated exacerbations in CF. Cystic fibrosis (CF) is the most lethal genetic disease among the Caucasian population affecting over 4,000 Canadians. The genetic defect results in the production of dehydrated thick mucous at many epithelial surfaces. In addition to other clinical complications, this leads to chronic colonization of the lower airways of CF patients. It is chronic airway infection and the associated loss of lung function that is responsible for more than 90% of mortality in CF. CF microbiology is usually associated with well characterized pathogens such as Pseudomonas aeruginosa, but recently other organisms such as the Streptococcus milleri group have been implicated as potentially common but overlooked pathogens in CF. Furthermore, both culture dependent and independent approaches have demonstrated that the CF airway infections are polymicrobial. These communities include organisms which in animal models are avirulent on their own by enhancing the virulence of pathogens such as P. aeruginosa. These syergens include many streptococci that do not fall into known species. In this study, we screened our CF cultured microbiome collection (greater than 1400 strains) for potential novel streptococci by partial 16S rRNA sequencing. One hundred isolates did not correspond to any typed strains. Genotypic analysis, including full 16S sequencing and multilocus sequence typing with the sodA, rpoB, and ddl genes, yielded 48 novel strains in 7 clades. Phenotypic assays were conducted in parallel. Results showed that 86% of strains were positive for at least one exoenzyme virulence factor. There was poor correlation between genotypic and phenotypic properties highlighting the variability among this genera and the limitations to current speciation. Antibiotic susceptibility assays were also completed using the disc diffusion method with a high percentage of strains displaying resistance to macrolides, such as azithromycin (61%) and erythromycin (71%), drugs commonly used in CF treatment. These strains may serve as reservoirs of antibiotic resistance genes within the airway microbiome. The characterization of these novel Streptococcus spp. highlights their potential to be both primary pathogens as well as synergens within the CF microbiota. Results: Forty-one subjects isolating M. abscessus were identified (31 multiple isolates/10 single isolates). One subject was excluded from evaluation as insufficient clinical details were available. The mean age at first isolation was 27.6 years. Concerning concomitant diagnoses, ABPA was present in 38% of subjects who had multiple isolates and CF related diabetes in 10%.

Seventeen subjects had received some form of treatment directed against M. abscessus and the mean time from first isolation to treatment was 14.5 months. Lung function had undergone a mean decline of 18.6%. Of the group that were treated, 47% suffered an adverse event that was serious enough to warrant a treatment regimen change.

Case controlled data was available for 29 matched pairs (22 multiple isolates and 7 single isolates). If data was missing for a particular parameter the matched pair were excluded from that analysis.

When comparing the multiple isolates and control group there was no significant difference between FEV1 (1.80L vs 2.01L), BMI (21.4 vs 22.3) and CRP (20 vs 24) at baseline. The prevalence of S. aureus chronic infection was similar (50%) in both groups. There was no difference between groups regarding symptoms of haemoptysis, fever and sweats at first isolation.

Concerning concomitant drugs at first isolation there was no significant difference between groups for azithromycin (5 vs 9) and oral steroid (5 vs 3) use.

Conclusion: This study demonstrates that there was no difference at first isolation between baseline clinical parameters of lung function, nutrition and systemic inflammation for those subjects with multiple isolates of M. abscessus and those without. This would suggest at present universal screening needs to continue.

A possible association of M. abscessus with ABPA and absence of regular azithromycin use requires further investigation.

A prospective study is needed to establish adverse features associated with clinical deterioration and furthermore to establish which patients to treat and optimal timing for this. Conclusion: Viruses were detected in CF patients even when they were well. The data suggested a trend to increased virus detection in sick CF samples, but results were not statistically significant. Partial sequencing of rhinovirus positive samples will be informative to determine whether specific clades of rhinoviruses are associated with CF exacerbations. Further studies are necessary to determine if a difference does exist between the responses to viral presence of CF patients compared to people without CF.

This study is supported in part by Vanderbilt Institute for Clinical and Translational Research Vanderbilt CTSA grant UL1 RR024975 from NCRR/NIH. Our overall goal is to rapidly distinguish between bacterial genera, species, and strains as well as the inflammation state of the lung during bacterial infection via analysis of patient breath using advanced mass spectrometry (MS). We show we can discern between genera, species, and using specific biomarkers, strains of bacteria of importance to the CF community. Additionally, we are building a library of metabolic fingerprints and biomarkers for Pseudomonas aeruginosa (Pa), including its response to different oxygen levels, growth stages, carbon sources, and aminoglycoside dose. We show it may be possible to determine bacterium identity and state of exacerbation in a lung infection via breath analysis for CF patients.

Background: Infection by bacteria and other microorganisms in the lung of patients with CF increase their morbidity and mortality. Rapid detection of bacteria and the inflammation state of the lung would aid patient treatment.

Objectives: To use secondary electrospray ionization MS (SESI-MS) to: 1) distinguish between bacterial lung pathogen genera and species via a volatile metabolic fingerprint in the flask; 2) determine the metabolic status of Pa via a volatile metabolic fingerprint in the flask for different carbon source and tobramycin dosing time points; and 3) distinguish between dominant pathogens present in CF sputum and lavage samples.

Methods: An acidic electrospray solution charged volatile headspace molecules (e.g., from Pa, Sa, and Legionella pneumophila (Lp) cultures and clinical samples), passaged through the electrospray cloud via an ultrapure CO 2 carrier gas. Volatiles were separated based on their mass to charge ratio in an interface-modified Applied Biosystems API3000 triple quadrupole MS. Putative biomarkers were identified via comparison of fragmentation spectra with ultrapure chemical standards.

Results: SESI-MS spectra can be used to distinguish between Pa, Sa, and Lp and other bacteria grown under complex and defined media conditions via their volatile metabolic spectra using principle component analyses. Fifteen metabolite peaks were identified, including the diagnostic grape-like odor (2-aminoacetophenone) for Pa. The volatile fingerprint of Pa varies when the bacterium is grown under different conditions. Whole spectra as well as biomarkers associated with metabolic pathways (such as a consequence of different carbon source catabolism and cell response to antibiotic stress) can be distinguished using SESI-MS. Preliminary data on the global volatile fingerprint of patient sputum and lavage samples suggests that differentiation of bacterial genus dominating sputum samples may be possible.

Conclusions: SESI-MS can be used to distinguish between bacterial genera as well as Pseudomonas species, and Pa strains. Differences in the volatile metabolome are revealed in the global fingerprint as well as the presence/absence and proportion of biomarkers. This research indicates that a rapid, non-invasive, and sensitive breath test for bacterial infection diagnosis and exacerbation state may be possible for CF patients.

Supported Background: S. aureus was originally considered the most important pathogen in CF and was the cause of mortality in CF infants and children in the 1940s-1960s. Antibiotic treatment of any relevant microorganism, ignoring presence or absence of respiratory symptoms has therefore been our policy since establishment of our CF center in 1968 to prevent subsequent infection and lung tissue damage (1) . However, the short term influence on lung function and BMI of such early aggressive treatment has not been thoroughly evaluated.

Materials and Methods: All patients are scheduled for monthly visits including growth measures, microbiological surveillance of sputum or secretion collected by nasopharyngeal suctioning, and lung function. Simultaneous finding of bacterial pathogens by microscopy and culture leads to a course of oral antibiotics for 14 days. In this study, we focused entirely on S. aureus colonization during 2009. Data on FEV1%predicted and FEF50%predicted within 7 days before and a maximum of 14 days after end of treatment for S. aureus in otherwise clinically stable patients were retrospectively retrieved from our CF register.

Results: First, 305 S. aureus positive sputum cultures in 106 CF patients were evaluated. Second, of these 145 unique courses against monocultures of S. aureus were identified in 75 patients, median (range) age 19 years. Both FEV1% and FEF50% increased highly significantly between measurements; 80.5-84.8% and 79-86.5% (p<0.0001), respectively, while zscore BMI exhibited a corresponding tendency in positive direction, -0.21 to -0.18 (p=0.07).

Conclusion: Demonstration of a positive impact on lung function by elective treatment strategy of colonization of lower respiratory tract in CF supports the rationale for antibiotic treatment of S. aureus even in clinically stable CF patients. Several studies have shown successful eradication of initial P. aeruginosa (Pa) infection but the optimal regimen to prevent recurrence is unclear. The 18-month EPIC trial compared cycled antibiotic therapy (administered quarterly regardless of culture findings) with culturebased therapy (administered only in response to Pa positivity). Participants received tobramycin inhalation solution (TIS) with either oral ciprofloxacin or placebo during treatment cycles. No differences between groups were identified in the primary outcomes, time to pulmonary exacerbation and overall proportion of Pa positive cultures. The objectives of this study were to evaluate key secondary outcomes to (1) determine whether cycled therapy (with or without ciprofloxacin) is more effective than culture-based therapy in delaying Pa recurrence, and (2) Conclusions: PE are frequent events in CF. Approximately one quarter of events fail to respond adequately to initial management attempts. Both patient demographic and episode specific clinical information can be used to identify individuals at increased risk of initial management failure. Antibacterial therapy based on in vitro susceptibilities did predict treatment outcome in this study. Methods: Phase 3 open label, randomized, active comparator trial evaluating safety and efficacy of AZLI vs. TIS in CF patients (pts) with PA infection, conducted at 93 CF centers in the EU and US. CF pts age ≥6 yrs, with baseline FEV1 ≤75% predicted, chronic PA infection and stable pulmonary disease were included. Patients were randomized 1:1 to receive AZLI 75 mg TID via Pari Investigational eFlow® Nebulizer System or TIS 300 mg BID via Pari LC Plus nebulizer for 3 cycles (28 days on/28 days off). Co-primary endpoints are: Non-inferiority of AZLI for mean %change in FEV1 %predicted at Day 28 and Superiority of AZLI for mean actual change in FEV1 %predicted across three treatment cycles. ANCOVA methods were used for efficacy analyses. Safety data included adverse events and hospitalization rates.

Results: A total of 273 pts were randomized, 268 pts treated (AZLI=136; TIS=132). The mean age is 25.5 yrs with 59 pts (22%) <18 yrs; mean baseline FEV1 %predicted is 52%. The mean % changes in FEV1 % predicted at Day 28 were AZLI 8.35, TIS 0.55. Treatment difference 7.8 (p=0.0001; 95% CI: 3.86, 11.73). Mean changes in CFQ-R respiratory symptoms score at Day 28 were AZLI 8.02, TIS 2.49. Treatment difference 5.53 (p=0.005; 95%CI: 1.65, 9.41) .

For 228 pts with TIS use ≥84 days in the 12 months prior to randomization, mean % changes in FEV1 %Predicted at Day 28 were AZLI 10.04, TIS 0.54. Treatment difference 9.5 (p <0.0001; 95% CI: 5.14, 13.86). Mean changes in CFQ-R respiratory symptoms score at Day 28 were AZLI 9.95, TIS 2.64. Treatment difference 7.30 (p=0.0005; 95%CI: 3.26, 11.35). Safety data is consistent with published studies of AZLI, and AE rates were similar to or lower than the TIS treated group.

Results of 6 months/3 treatment cycles will be presented. Conclusions: AZLI met its primary non-inferiority endpoint and demonstrated superiority to TIS at 28 days in both lung function and respiratory symptoms improvements. AZLI was safe and well tolerated with adverse events consistent with previous clinical trial experiences.

Supported by Gilead Sciences.

resistant phenotype, presence of other microorganisms, and prior usage of anti-pseudomonal antibiotics. Methods: Children with CF ages 1-12 years with new onset Pa were randomized to one of four antibiotic regimens: cycled therapy received TIS for 28 days (300 mg BID) with either oral ciprofloxacin or placebo (15-20 mg/kg BID) for the first 14 days on a quarterly basis, and culture-based therapy received the same antibiotic therapies only during the quarters that their respiratory cultures were positive for Pa. Time to recurrence was analyzed using Cox proportional hazards regression.

Results: The risk of recurrent Pa was increased among those Pa positive at baseline as compared to those negative (hazard ratio [HR]: 3.01, 95% CI: 2.02, 4.48). Among those positive for Pa at baseline (n=118/295), no significant difference was seen in the risk of recurrence in the cycled group versus the culture-based group (HR 1.03, 95% CI: 0.63,1.69) or in those who received ciprofloxacin versus those who received placebo (HR 0.89, 95% CI: 0.54,1.46 ). Among those negative for Pa at baseline (n=177/295), the cycled group had less risk of recurrence as compared to the culture-based group (HR 0.27, 95% CI: 0.13,0.56) and there was no difference in those who received ciprofloxacin versus those receiving placebo (HR: 1.09, 95% CI: 0.59,2.03). Among the 65/150 (43%) who had recurrent Pa in the culture-based group, 38/65 (58%) were negative for the remainder of the study after a second course of treatment; 61% of those Pa negative at baseline and 55% of those positive. No significant predictors of recurrence were identified.

Conclusion: Although cycled therapy may have prevented Pa recurrence in those Pa negative at baseline, the majority with recurrent Pa in the culture-based group responded successfully to subsequent treatment. No predictors of recurrence were identified, indicating the importance of ongoing studies evaluating Pa phenotypic characteristics and serology.

Supported by CFF, NIH U01HL080310 and ULIRR025014-03. There is little information on the frequency with which PE management is considered a clinical success or failure. Furthermore, factors that increase the risk of PE treatment failure remain relatively unknown.

Methods: All patients chronically infected with P. aeruginosa attending the Adult Cystic Fibrosis Centre of Northern Ireland were identified and their records reviewed. All PE treated with parenteral antibiotics were reviewed from 2006-2009. Treatment failures were a priori categorized as; antibiotic regimen change during therapy (ARC) (not for adverse reaction), early repeated exacerbation (<45 days) (ERE), failure to recover lung function (>90% of baseline FEV 1 ) (FTR) and prolongation of therapy (>20 days) (POT). Patients were excluded if they were concomitantly infected with Burkholderia cepacia complex, or from the year their lung function fell below 30% FEV 1 and/or were listed for lung transplantation.

Results: During the four year study period, 101 patients chronically infected with P. aeruginosa met inclusion criteria and had 452 evaluable PE. Twenty patients had no PE during the study period, 33 had 1-3PE, 24 had 4-6 PE, 8 had 7-9 PE, 12 had 10-15, 4 had 15-20 and one patient had >20 PE. Treatment failures were observed in 125 (27.7%) of PE episodes which were classified as: ARC 27 (6%), ERE 63 (14%), FTR 66 (15%) and POT 29 (6%). Thirty-nine (48%) patients experienced treatment failures. Of patients who experienced a mean one treatment failure/year demographic risk factors included: receipt of enteric feeds, CF-related diabetes and CFliver disease but did not include female sex or F508del homozygosity. Increased risk of treatment failure was associated with lower admission FEV 1 and increased admission white cell count (WCC) and C-reactive protein (CRP). At therapy completion an elevated WCC and CRP increased risk of treatment failure. Therapy failure risk with individual antibiotics were as follows: piperacillin-tazobactam odds ratio (OR) 0.6, ceftazadime OR 0.9, aztreonam OR 1.24, meropenem OR 1.43, ciprofloxacin without beta-lac- 6 ; Ratjen, F. 7 ; for the AZ0004 Study, G. 6 1. Columbia Univ, New York, NY, USA; 2. Univ Kentucky, Lexington, KY, USA; 3. Univ Washington, Seatte, WA, USA; 4. McGill Univ, Montreal, QC, Canada; Seattle, WA, USA; 6. CFF, Bethesda, MD, USA; 7. Univ Toronto, Toronto, ON, Canada Background: We previously reported the results of a randomized clinical trial (RCT) performed in CF patients 6-18 years of age uninfected with P. aeruginosa in which oral azithromcyin was not associated with an improvement in lung function, but was associated with reduced pulmonary exacerbations, reduced initiation of new oral antibiotics, and increased weight gain (Saiman et al. JAMA 2010) . We now report the results of the open-label, follow-on study that directly followed the RCT.

Methods: Eligible subjects from 31 selected sites who completed the RCT by December 2008 were invited to participate in the open-label, follow-on study on the final day of the RCT (Day 168). Detection of nontuberculous mycobacteria or Burkholderia cepacia complex during the RCT were exclusion criteria. During the 24-week open-label study, subjects initially randomized to either azithromycin or placebo received the same dose of azithromycin used during the RCT (250 mg for subjects 18-35.9 kg and 500 mg for subjects > 36 kg on Monday-Wednesday-Friday). Study visits corresponded with quarterly CF clinic visits and monthly telephone calls assessed adverse events (AEs) and compliance.

Results: In all, 77 (95%) of 81 eligible subjects initially randomized to azithromycin (azithromycin-azithromycin group) and 69 (91%) of 76 eligible subjects initially randomized to placebo (placebo-azithromycin group) agreed to participate in the open-label study. The baseline characteristics of subjects in the randomized study and subjects in the open-label study, AEs and subject retention are shown in the Table. Changes in pulmonary function; rates of pulmonary exacerbations, initiation of oral, intravenous, and inhaled antibiotics, and hospitalization; microbiology; and changes in weight will be presented.

Conclusions Results: Between 1996 and 2003, the annual incidence of MARPA conversion in patients with chronic PA infection ranged from 6.9% to 9.2% (mean = 7.5%), but it did not increase significantly over time (P = .72). A total of 4,349 patients with chronic PA and adequate PFT data were identified; 1111 subsequently developed MARPA, while 3238 patients were PA+ but MARPA−. Compared with patients who remained MARPA free, MARPA+ patients had lower FEV1 and received more antibiotic therapy. In the MARPA+ group, the mean FEV1 decline in the 2 years prior to MARPA detection was −2.22% predicted/year. After MARPA detection, it was − 2.46%, which was not a significant change (P = .39). Persistent MARPA detection was found in only 11.1% of patients. Of patients with at least 1 MARPA+ culture, 36.3% were MARPA− for all subsequent cultures, with the balance of patients being intermittently MARPA+. There was no relationship between persistent infection and FEV1 decline. MARPA prevalence was higher at ESCF sites in the upper FEV1 quartile, but analysis of FEV1 decline based on site quartile yielded similar results regardless of quartile.

Conclusions: MARPA acquisition in patients with CF is associated with lower FEV1 and increased antibiotic use. However, MARPA acquisition is not associated with a significant change in FEV1 decline. These results suggest that MARPA is more likely to be a marker of more severe disease and more intensive therapy, and it does not independently contribute to more rapid lung function decline. Multi-year surveillance of bacterial pathogens in respiratory cultures of nearly 300 pediatric patients with CF and subsequent molecular typing of MRPA isolates revealed the continued prevalence of the endemic strain, or dominant clone. A retrospective study was performed to ed throughout the length of the study. Eleven of the 82 patients were under evaluation for lung transplantation, but the remaining 71 patients were further analyzed based on clonal groupings. Objectives included identification of risk factors for acquisition of the dominant MRPA clone and patient outcomes related to infection with the endemic MRPA strain. Initial results prompted further evaluation of infection control processes including stricter guidelines for contact isolation. Multiple improvements related to infection control in the CF patient population led to the reduced incidence of the endemic MRPA strain.

Conclusions: Molecular typing of MRPA in a pediatric CF center identified the initial presence, documented the ongoing prevalence, and tracked the continued incidence of infections with the endemic strain. Molecular epidemiological surveillance enabled clinical comparisons based on clonal groupings, provided evidence for the need for stricter contact isolation guidelines, and served as a metric for the effectiveness of improved infection control practices. Purpose: Colistimethate is an antimicrobial agent active against Gramnegative multidrug-resistant (MDR) pathogens often colonizing the lungs of patients with cystic fibrosis (CF); however, its use is limited by concerns with associated nephrotoxicity and neurotoxicity. The purpose of this study was to evaluate the safety and effectiveness of IV colistimethate treatment for CF exacerbations.

Methods: A retrospective chart review was conducted including patients with CF admitted to Cincinnati Children's Hospital Medical Center with acute pulmonary exacerbation, and treated with IV colistimethate or IV tobramycin between July 1, 2007 and January 31, 2010. Effectiveness of treatment with IV colistimethate was evaluated by analyzing changes in forced expiratory volume in one second (FEV 1 ), as well as physician-reported symptom improvement. Adverse effects were evaluated in the IV colistimethate group, and the incidence of nephrotoxicity was extracted and compared to similar endpoints in the IV tobramycin group.

Results: Thirty episodes of IV colistimethate use (mean age=17.9 years [range 16-20]; mean baseline FEV 1 =58.5% [range 30-79]; mean baseline serum creatinine (SCr)=0.63 mg/dL [range 0.5-1]), and 30 episodes of IV tobramycin use (mean age=13.6 years [range 2-20]; mean baseline FEV 1 =81% [range 34-119]; mean baseline SCr=0.59 mg/dL [range 0.31-1]) were evaluated. All pulmonary exacerbation episodes were treated with antibiotic combination therapy. The mean change in FEV 1 from admission to completion of colistimethate therapy was 12.6% [range -3 to 30], and the incidences of nephrotoxicity and neurotoxicity were each 16.7% (5/30). One patient discontinued treatment with IV colistimethate due to perioral tingling and numbness, which resolved upon discontinuation. No nephrotoxic episodes were reported in the tobramycin group based on study definitions. Of the patients who were able to perform pulmonary function tests in the colistimethate group, 61.5% (16/26) reached ≥90% of their baseline FEV 1 upon completion of antibiotic therapy.

Conclusions: IV colistimethate therapy achieved overall clinical improvement and increased FEV 1 per patient episode. There was a greater incidence of nephrotoxicity with the use of IV colistimethate versus IV tobramycin; however, this may have been due to the significantly older colistimethate sample population (p=0.0001), and the higher cumulative lifetime exposure to nephrotoxic medications associated with increasing age in patients with CF. The change in SCr between the two groups was not significantly different. Episodes of neurotoxicity were mild and self-resolving. Overall, IV colistimethate appears to be a safe and effective option for treating pulmonary exacerbations in patients with CF in the context of appropriate dose titration and monitoring. identify risk factors for acquisition of the clone as well as determination of differences in patient outcome associated with subsequent infection.

Methods: The study included 71 patients with CF and documented MRPA infection. During the study timeframe, a total of 278 patients were seen at the CF Care Center, with 82 patients experiencing an MRPA infection. Eleven patients were excluded due to transplant evaluation status. Patients were followed for an average of three years. Designation of patients as members of the dominant MRPA clone group (32 patients) or a nondominant MRPA clone group (39 patients) was based on molecular typing by rep-PCR of the patient's initial MRPA isolate obtained from respiratory cultures. Demographic information and clinical parameters prior to MRPA infection were analyzed by logistic regression as potential risk factors for MRPA acquisition. Differences in patient outcome based on infection with the dominant or a non-dominant MRPA clone, including change in BMI, change in FEV 1 , and hospitalization rate, were evaluated by MANOVA.

Results: Recent hospitalization (≤ 90 days) was a statistically significant (p = 0.035) risk factor for acquisition of the dominant MRPA clone as opposed to a non-dominant MRPA clone. Patients hospitalized ≤ 180 days prior to MRPA acquisition were three to four times more likely to be infected with the dominant MRPA clone. Patients in the dominant MRPA clone group were hospitalized more frequently both pre-and post-infection as compared to the non-dominant MRPA clone group. Significant overlap in hospitalization of patients known to be infected with the dominant MRPA clone and patients subsequently infected with the dominant MRPA clone was observed. While differences in patient outcomes were not statistically significant, trends toward decline in lung function and nutritional status were suggestive of adverse effects associated with infection with the dominant MRPA clone.

Conclusions: Recent hospitalization was a significant risk factor for infection with the dominant MRPA clone, and following infection, patients infected with the endemic MRPA strain exhibited declines in outcome parameters, suggesting potentially increased virulence of this strain. Molecular typing of MRPA in a large pediatric CF center facilitated the identification of the endemic strain, allowed for multi-year molecular epidemiological surveillance, and provided evidence of reduced transmission of the endemic strain following the implementation of stricter contact isolation guidelines. Methods: Respiratory specimens were collected from patients with CF during clinic visits and inpatient episodes. Conventional microbiology culture identified P. aeruginosa isolates that met the antibiotic resistance criteria for classification as MRPA. Bacterial DNA was extracted, and the DNA was amplified by rep-PCR. Amplified products were processed using microfluidics-based electrophoresis, and analysis of relative similarities of resulting molecular profiles was performed using a web-based application. Molecular profiles were archived in libraries for ongoing comparisons with new MRPA isolates.

Results: More than 500 MRPA isolates from 82 patients with CF at TCH were typed from 2004-2009. An endemic MRPA strain, termed the dominant MRPA clone, was identified in the initial typing batch and persist- Objectives: To determine if antimicrobial susceptibility testing based on biofilm growth of P. aeruginosa (PA) leads to improved cystic fibrosis (CF) patient outcomes.

Methods: This is a prospective randomized double blind controlled trial of 14 days of 2 intravenous antibiotics based on biofilm versus conventional (Sensititre) antimicrobial susceptibilities of CF patients with chronic PA infection with an acute pulmonary exacerbation at the Hospital for Sick Children and St Michael's Hospital in Toronto. The primary outcome is the proportion of patients who have a > 3 log drop in colony forming units (CFUs) of PA in sputum. Secondary outcomes include change in pulmonary function tests, quality of life scores, sputum inflammatory markers and time to subsequent pulmonary exacerbation.

Results: To date, 91 patients have been enrolled in this ongoing study, of which 79 passed the sputum density screening (>10 5 CFU/ml). Eleven patients have had 17 acute pulmonary exacerbations and were randomized to one of the two antibiotic treatment arms. Biofilm and Sensititre testing were done on 302 patient strains of PA. Significantly fewer susceptible strains across all antimicrobial classes were observed by biofilm compared to Sensititre testing (Table 1) . Time required to achieve proper inoculum for susceptibility testing ranged from 3.5-24 hours, but the majority of strains (72%) achieved this within 7 hours. Of 196 sputum antimicrobial susceptibility pairs, 56 (29%) had concordant and 140 (71%) had discordant antibiotic choices (p<0.0001). Of those with discordant choices, one drug was different in 96 (69%) and both drugs were different in 44 (31%) of them.

Conclusions: Cross-sectional analysis of PA strains from patients enrolled in the study demonstrate that these strains are significantly less susceptible to all antimicrobial classes by biofilm compared to Sensititre testing. These antimicrobial susceptibility differences lead to significantly different antibiotic choices. Data on the primary and secondary outcome continues to be blinded as study recruitment is ongoing.

Funded by an unrestricted grant from Innovotech Inc. 

There is a clear unmet need for novel antibiotics or combinations of antibiotics that possess activity against CF pathogens such as P. aeruginosa (PA) and methicillin resistant Staphylococcus aureus (MRSA). Furthermore, given that anaerobic conditions exist in the lungs of CF patients with persistent respiratory infection, antibiotics with activity under anaerobic conditions would be particularly beneficial. One antibiotic currently in development for treatment of bacterial respiratory infections is a 4:1 (w/w) combination of fosfomycin/tobramycin for inhalation (FTI). Significantly, fosfomycin activity has been shown to be greater under anaerobic conditions. Therefore, the aim of this study was to determine the in vitro susceptibility of MRSA and PA isolates to fosfomycin and tobramycin, and a combination of fosfomycin and tobramycin in a 4:1 ratio (F:T) under both aerobic and anaerobic conditions.

Methods: The susceptibility of clinical CF isolates (MRSA, n=25; PA, n=100) to F:T, fosfomycin and tobramycin was determined by the agar dilution method under aerobic and anaerobic conditions. The MIC was read as the lowest concentration of both antibiotics alone and the combination that completely inhibited visible growth of the isolate. This data was used to calculate geometric means and the concentration of the antibiotic combination and of each antibiotic alone required to inhibit 50% (MIC 50 ) and 90% (MIC 90 ) of the isolates.

Results: Both fosfomycin and F:T had high activity against MRSA isolates under both aerobic and anaerobic conditions with fosfomycin showing better activity under anaerobic conditions (Table) . Anaerobic incubation had no effect on F:T activity with the MIC 50 and MIC 90 values similar under aerobic and anaerobic conditions. In contrast, tobramycin was much less active against MRSA under anaerobic conditions. Tobramycin was the most active Conclusion: This study has shown that F:T has excellent activity against both Gram-positive and Gram-negative CF respiratory pathogens under anaerobic conditions, similar to those found in sputum in the lungs of CF patients.

Supported by Gilead Sciences, Inc.

Methods: All Copenhagen CF patients (N=278) were included in the study. Numbers of sputum cultures positive for S. aureus from 1.1.2008 to 31.12.2009 were counted. Spa-types of S. aureus were compared before and after treatment to decide whether the treatment had eradicated the bacteria. Standard indication for treatment was detection of S. aureus by culture and microscopy in secretions from the lower respiratory tract regardless of the patient's clinical condition. Our standard treatment was two weeks oral fusidic acid in combination with dicloxacillin or with amoxicillin + clavulanic acid. The treatment was regarded prolonged if antibiotic was prescribed for more than 14 days, and most often consisted of one month of fusidic acid in combination with dicloxacillin followed by two month of dicloxacillin. If the patient was allergic to penicillins, a combination of clindamycin and fusidic acid or azithromycin/clarithromycin and rifampicin for fourteen days were prescribed. These combinations were also used in case of failure of the standard regimens. No prophylactic antibiotic chemotherapy was used.

Results: A total of 142 patients in 2008 and 141 in 2009 had at least one sputum sample positive for S. aureus. In total 1200 positive S. aureus cultures were found in 168 patients, median age 16 years (range 0 -47). There were 540/1200 (45%) positive samples confirmed by microscopy in 136 patients with a mean of three positive microscopy samples per patient (range 1-13). Thirty-nine patients (14%) had been infected with the same spa-type for six months or more during the study time, and were therefore considered chronically infected. A total of 170 different S. aureus spa-types were found in 168 patients. Of these, 156 spa-types were resistant to penicillin, 53 spatypes were resistant to clindamycin and azithromycin, six spa-types were resistant to fucidic acid and 19 spa-types were resistant to rifampicin. Only two MRSA positive sputum cultures were found, and these were not the same spa-type. There were 546 treatments given in 131 patients with eradication at 60% of the initial S. aureus spa-types. The initial spa-type was eradicated in 62% of the treatments when standard treatment was used (200 treatments), in 50% when the treatment was prolonged (145 treatments), in 71% if penicillin allergy regimen was used (28 treatments) and in 64% when other treatments were used (173 treatments).

Conclusion: In general, we have very few problems with S. aureus. MRSA are very seldom cultured from our CF population, and resistance to other drugs have no clinical implications on treatment of S. aureus infection. We do not use prophylactic chemotherapy but treat when S. aureus is detected by microscopy and culture. The number of eradications was about 60% and only 14% of the patients were chronically infected during the two years. The type I interferon (IFN) signaling cascade is increasingly being recognized as an important part of host defense against bacterial as well as viral infections, contributing to both pro-and anti-inflammatory signaling in the airway. In this study we tested the hypothesis that induction of type I IFN signaling by Pseudomonas aeruginosa is reduced in cystic fibrosis (CF) airway epithelial cells and that type I IFN signaling expedites clearance of P. aeruginosa from the airways. We used a human CF bronchial epithelial cell line IB3 (∆F508/W1282X) and corrected line C38 for in vitro comparison of type I IFN signaling following exposure to live P. aeruginosa, the TLR4 ligand lipopolysaccharide (LPS), or the TLR3 ligand Poly:IC. CF epithelial cells were significantly attenuated in their ability to produce type I IFN in response to P. aeruginosa as compared to corrected epithelial cells. Interestingly, no defect was observed in CFTR knockout dendritic cells. Phosphorylated IRF3, a kinase immediately upstream of the type I IFN promoter, and type I IFN production was reduced in CF cells following LPS (TLR4) but not Poly:IC (TLR3) treatment compared to corrected cells. To elucidate a role for type I IFN in clearance of P. aeruginosa from the airways, wild type C57BL/6J mice were pretreated with Poly:IC to stimulate IFN production 24 hours prior to inoculation with P. aeruginosa (~1×10 7 CFU). Bacterial counts in bronchial alveolar lavage fluid (BALF), lung homogenate, and spleen samples were determined after 24 hours. Immune cell populations present in the BALF were analyzed by FACS. Mouse lungs infected with P. aeruginosa showed significant induction of type I IFN as measured by quan- Background: Respiratory syncytial virus (RSV) causes acute lung infection in children worldwide, resulting in considerable morbidity and mortality. Children with cystic fibrosis (CF) are at risk for severe RSV infections because they are prone to lung inflammation, bacterial colonization and airway disease. No treatment currently exists, hence prevention is important. Palivizumab (Synagis®) has been shown to reduce RSV hospitalization rates and is recommended for prophylaxis in high-risk children with other conditions. It is unclear if palivizumab is effective in preventing RSV hospitalizations in children with CF. We conducted a systematic review to determine the efficacy and safety of palivizumab in preventing hospitalization and mortality due to RSV infection in children (18 years and younger) with CF.

Methods: In February 2009, we searched the Cochrane Cystic Fibrosis and Genetic Disorders Group Trials Register for studies and contacted the drug manufacturer to inquire about ongoing or unpublished trials. One eligible trial, only available as a poster presentation, was identified from literature received from the drug manufacturer. We also searched the reference list of the eligible trial and existing review articles on palivizumab. Two review authors independently abstracted data and assessed the risk of bias in the trial.

Results: Currently no randomized controlled trial has been published on the efficacy of palivizumab in CF children with RSV infections. One unpublished randomized controlled trial including 186 infants (all 2 years of age and younger) was identified. Trial participants were randomized to receive palivizumab (N=92) or placebo (N=94) monthly over five months of one RSV season. At 6 months follow-up, 1 participant in each group was hospitalized due to RSV infection. In the palivizumab group 89 children had any adverse event and of these 5 were related adverse events and 19 were serious adverse events. In the placebo group 90 children had any adverse event and of these 4 were related adverse events, 16 were serious adverse events and 2 were related serious adverse events. No deaths were reported in either treatment group.

Conclusions: While the overall incidence of adverse events was similar in the palivizumab and placebo groups, it is not possible to draw conclusions on the safety and tolerability of RSV prophylaxis with palivizumab in infants with cystic fibrosis because the included trial did not specify the criteria used for classifying adverse events. Six months after treatment completion, the authors reported that there were no clinically meaningful differences in outcomes, however no data were provided. The report of only one participant in each group being hospitalized due to RSV infection is consistent with the findings of non-randomized studies of CF patients in the literature. Full publication of the identified trial and additional randomized trials are needed to establish the safety and efficacy of this vaccine in children with CF.

Partially supported by the Cystic Fibrosis Foundation, USA. We acknowledge the contributions of Olaide A. Odelola. titative real-time PCR. Pretreatment of mice with Poly:IC greatly improved clearance of P. aeruginosa and increased the level of CD86+ airway macrophages and dendritic cells compared to mice treated with P. aeruginosa alone. We conclude that CF epithelial cells produce less type I IFN in response to P. aeruginosa, potentially due to a specific defect in TLR4 signaling, as type I IFN production is not reduced in these cells when treated with Poly:IC, a TLR3 agonist. Furthermore, type I IFN production in vivo facilitates the clearance of P. aeruginosa from the airways, suggesting that a reduction in type I IFN production in CF patients may promote colonization of the lung by this opportunistic pathogen. The majority of cystic fibrosis (CF) patients acquire chronic lung infection with Pseudomonas aeruginosa due to biofilm formation. The inflammatory response may depend on whether the biofilm infection is established in the larger airways in the conducting zone (airflow) or the smaller airways in the respiratory zone (oxygen and carbon dioxide gas exchange) of the lungs. Due to localization of alveolar macrophages in the respiratory zone, we propose that biofilm infection in the smaller airways is more important, due to increased recognition. As a consequence, increased tissue damage in the respiratory zone would be expected.

Using two different sizes of seaweed alginate beads (small beads (SB) diameter 35 µm, range 14-62 vs big beads (BB) diameter 122 µm, range 74-175) containing P. aeruginosa (PAO 579) we experimentally simulated infection in the respiratory versus the conductive zones of the left lung in BALB/c mice (n=36). The Nisco Encapsulation Unit VARJ 30 was used for seaweed alginate bead production. All mice received the similar amount of alginate and number of P. aeruginosa (0.66 x 109 CFU/ml (SB) vs 0.71 x 109 CFU/ml (BB)). The study was evaluated by quantitative bacteriology, pulmonary MIP-2 production (chemo-attractant) and histopatology of the entire left lung (alcian blue staining and PNA FISH staining) at day one after challenge. MIP-2 was measured in homogenized lung tissue and determined by a specific ELISA.

In the group infected with SB more bacteria as compared to mice infected with BB was observed (7.45-9.12 log CFU/lung vs 6.09-7.79 log CFU/lung, p=0.0001). Furthermore, mice in the group infected with SB had significantly higher concentration of MIP-2 as compared to the mice in the group infected with BB (1135-1370 pg/ml vs 391-1181 pg/ml, p<0.0001).

The bronchial diameter where biofilms were present was significantly smaller in the SB group (118 µm, range 39-203 µm) as compared to the BB group (186.5 µm, range 112-295 µm) (p<0.0001). The biofilm sizes were 36.3 µm (range 13-78 µm) in the SB group and 71.4 µm (range 40-140 µm) in the BB group (p<0.0001). P. aeruginosa specific PNA-FISH in general confirmed the presence of P. aeruginosa in the alcian blue areas of the lungs. Moreover, the total numbers of P. aeruginosa containing biofilms in the absolute periphery of the lungs were 42 (2-27 in four mice) in the SB group as compared to 6 (0-4 in four mice) in the BB group (p=0.058).

In conclusion, an experimental model making it possible to investigate the significance of biofilm growing P. aeruginosa in the two major pulmonary compartments was established. The present study shows the significance of determining localisation of the pulmonary infection, since the biofilm infection of the smaller airways was more difficult to eradicate and generated a more severe inflammation. CF has been increasing. The objective of this study was to describe the epidemiology of Staphylococcus aureus infections in pediatric patients with CF cared by a center located in a highly prevalent geographic location for MRSA.

Method: Arkansas Children's Hospital (ACH) is the only accredited pediatric CF center in the state where 170 patients receive regular care. Clinical and epidemiologic data was obtained by review of the hospital's medical records, the microbiology laboratory database, and the Cystic Fibrosis Foundation Patient Registry (CFFPR).

The number of positive cultures for MRSA and MSSA per year (1995-2009) from individuals with CF were identified. The prevalence of MRSA in non-CF isolates was calculated (2002) (2003) (2004) (2005) (2006) (2007) (2008) (2009) ). The annual prevalence of MRSA and the mean age for MRSA acquisition in CF patients were calculated. Risk factors were identified for CF patients infected with MRSA early in life. Antimicrobial sensitivities of the S. aureus isolates were recorded.

Results: The prevalence of MRSA in non-CF isolates obtained in both the inpatient and outpatient settings increased dramatically. In 2002, MRSA constituted only 3% of non-CF S. aureus isolates, however, the number increased to 64% in 2009. Similarly, in patients with CF, the prevalence of MRSA increased from 0% in 1995 to 45% in 2009 with a peak of 46% in 2007. Out of 314 patients with CF, 135 were at least one time infected/colonized with MRSA during the follow up period. According to 2007 CFFPR data, our center had the highest MRSA rate in the nation (42% at our center versus 22% national average).

The mean age of acquisition of MRSA in pediatric patients with CF decreased significantly from 13 years in 1997 to 5 years in 2009. The majority of CF MRSA isolates was sensitive to bactrim and tetracycline and remained so. Clindamycin sensitivity was 43% in 2003 and 33% in 2009 with a peak of 56% in 2006. No vancomycin resistance was noted; however, 4 patients developed linezolid resistance. There were 12 CF patients colonized with MRSA during their first year of life (mean 5.4 months; median 7 months). The risk factors for early acquisition of MRSA included an older sibling with CF, prior hospitalization, surgical intervention, prolonged NICU stay, and symptomatic diagnosis.

Conclusions: Mirroring changes seen in the general population, the prevalence of MRSA has increased in our CF center patients over the last 15 years. The age of acquision of MRSA by our CF patients has decreased dramatically. Recently, the linear increase in MRSA prevalence seems to have leveled off in both the non-CF and CF populations. This could be due to saturation of the community with MRSA or secondary to strict infection control practices at our center.

Despite the intensive usage of antibiotics, the majority of our CF MRSA strains are sensitive to common antibiotics. The development of linezolid resistance within a short period of time is worrisome and suggests that this antibiotic should be used judiciously. Lung disease in cystic fibrosis (CF) is characterized by chronic microbial colonization and repeated acute exacerbations of pulmonary infection. Vitamin D has been previously shown to induce expression of the glycoprotein CD14, thus enhancing innate immunity. Present in a variety of cells including epithelial cells, CD14 is implicated in the recognition of Aspergillus fumigatus . Containing sites for the vitamin D receptor (vitamin D response element VDRE); the CD14 gene is positively regulated by 1α, 25 dihydroxyvitamin-D3. As fungal colonization by the opportunistic fungal pathogen A. fumigatus in CF individuals leads to clinical deterioration of the lungs, the aim of this study was to investigate the effect of Aspergillus secreted toxins on vitamin D receptor (VDR) expression.

The A. fumigatus clinical strains employed in this study were ATCC 26933 and CAF1. The effect of A. fumigatus on VDR expression was investigated in vitro in human CF bronchial and tracheal epithelial cells (CFBE 41o-, CFTE 29o-) and in vivo, in CF bronchial brushings of A. fumigatus colonized individuals. VDR expression was evaluated on both the gene and practitioners. Secondary objectives include time to reach the AG PK/PD target, number of dosing adjustments, length of stay (LOS), number of levels ordered, changes in pulmonary function tests, drug waste and the related cost of therapy.

Methods: This is a retrospective cohort study beginning January 1, 2007. Any CF patient receiving an AG and having at least two AG serum concentrations, which could be used to determine PK/PD parameters, was included. To detect a 25% difference in the number of CF patients achieving AG PK/PD targets with an α = 0.05 and a power of 80%, 40 treatment courses are required per group. A χ2 test and Students t-test will be used to analyze nominal and continuous variables, respectively.

Results: Baseline demographic data was similar between each group. Twenty-nine patients accounting for 52 courses of AG therapy were included in the P group, and 22 patients, accounting for 42 courses, were included in the NP group. Ninety-eight percent of patients in the P group reached AG PK/PD targets compared with 71% in the NP group, p < 0.001, χ2 = 13.8. Patients in the P group reached the AG PK/PD parameter in a mean of 1.9 days (range 1-3) compared with 4.8 days (range 1-15) in the NP group, p < 0.0001. The average LOS in the P group was 9 days (range 2-16) compared with 12 days (range 4-21) in the NP group, p = 0.033. The mean number of levels per patient was 2.7 (range 2-4) in the P group compared with 5.2 (range of 2-20) in the NP group, p < 0.001. The mean change in FEV1 during treatment was an increase of 0.36 % (95% CI 0.24-0.49) in the P group and an increase of 0.38 % (95% CI 0.26-0.51) in the NP group, p > 0.05. The mean change in FVC during treatment was an increase of 0.41 % (95% CI 0.27-0.55) in the P group and increase of 0.43 % (95% CI 0.30-0.56) in the NP group, p > 0.05. The costs (reflected as charges) associated with drug levels, dosing adjustments and LOS were $26,549, $14,069, and $1,680,000 in the P group as compared with $40,683, $27,812, and $1,940,000, respectively, in the NP group, resulting in a cost reduction of $287,877.

Conclusion: Pharmacist-managed TDM resulted in a higher percentage of pediatric CF patients achieving AG PK/PD targets. Furthermore, patients in the P group attained the AG PK/PD an average of 3 days sooner and had an average LOS that was 2 days shorter. Pharmacist managed TDM resulted in fewer dosage adjustments, drug levels, and cost associated with serum sampling, drug wastage, and LOS. Introduction: CF is characterized by chronic pulmonary infection and intermittent acute exacerbations. Pseudomonas aeruginosa is the most common pathogen known to affect patients with CF, and by age 18, almost 80% of individuals are chronically infected with P. aeruginosa. In 2003, Saiman et al. published the results of a randomized controlled trial demonstrating that treatment for 6 months with azithromycin in patients with CF and P. aeruginosa was beneficial. Specifically, there was a statistically significant increase in FEV 1 , decrease in pulmonary exacerbations, and an increase in weight compared to control patients. These data have led many CF clinicians to prescribe chronic macrolide therapy to patients with CF who are chronically infected with P. aeruginosa. According to the 2008 Cystic Fibrosis Foundation (CFF) Patient Registry Report, only 65.6% of people who fit the criteria are receiving macrolide therapy. However, there have been no large scale long term follow-up studies on this patient population.

Objective: To determine if chronic macrolide therapy is associated with a persistent reduction in pulmonary exacerbations requiring hospitalization in individuals with CF infected with P. aeruginosa.

Methods: Chronic macrolide use was first recorded in the CFF patient registry in 2006, and this was a 3-year cohort study using the registry from 2006-2008. We studied individuals infected with P. aeruginosa who had been on macrolide therapy for at least two years during that time period. Multiple Poisson regression for repeated measures was performed to adjust for age, sex, FEV 1 , pancreatic insufficiency, TOBI use, pulmozyme use, hypertonic saline use, methicillin-resistant Staphylococcus aureus (MRSA), and Burkholderia cepacia.

protein level by real-time reverse transcriptase-polymerase chain reaction (RT-PCR) and Western blotting respectively. The effect of the Aspergillus secondary metabolite gliotoxin, on VDR expression was also examined.

A. fumigatus culture filtrates down-regulated VDR mRNA expression and reduced the level of VDR protein by 50% in both CFTE (P=0.005) and CFBE (p=0.009) cell lines. Culture filtrates from the Aspergillus strain CAF1 did not contain the immunosuppressive mycotoxin gliotoxin and did not affect VDR expression. Gliotoxin levels in fungal filtrates was measured by reverse-phase HPLC and a concentration of 0.8 µM significantly downregulated VDR expression in both CFTE (p=0.002) and CFBE (p=0.02) cells. In vivo, analysis of bronchial brushing of Aspergillus colonized versus non-colonized CF patients, revealed reduced levels of VDR (p=0.02) and CD14 gene expression in fungal colonized individuals.

This data indicates that in CF bronchial and tracheal epithelial cells, A. fumigatus secreted gliotoxin dose-dependently regulates VDR mRNA and protein expression. Aspergillus colonized CF patients were characterised by distinct down-regulation of VDR and CD14 gene expression. Via inhibition of VDR expression, we propose that Aspergillus may directly impact upon CD14 expression levels thereby altering the inflammatory response within the CF lung. Results of this study indicate gliotoxin as a therapeutic target.

Funding The prevalence of methicillin resistant Staphylococcus aureus (MRSA) is increasing in patients with cystic fibrosis. (CF) According to the CF registry, in 1995 less than 1% of patients had a respiratory culture positive for MRSA. Today that number has risen to greater than 20%. Several studies have recently shown a deleterious effect of persistent MRSA infection in patients with CF. For example, new persistent infection with MRSA in these patients is associated with an increased rate of decline in lung function (FEV1). There is no strong consensus among CF experts regarding the management of MRSA infection and even less published on its eradication. Therefore the University of Wisconsin-Madison American Family Children's Hospital CF Center developed a protocol to eradicate MRSA upon first acquisition. The aim of this study was to evaluate the efficacy of the triple antibiotic protocol using trimethoprim/sulfamethoxazole, rifampin and topical mupirocin. Three sets of surveillance cultures were obtained using standard technique 2 weeks following the completion of the protocol, as well as sputum cultures at 6 and 12 months post eradication protocol. From 1/10/2005 to 7/31/2009, 22 patients cultured positive for MRSA. Of those, 5 received the eradication protocol due to first acquisition of MRSA. All (100%) of the protocol patients remain free of MRSA 1 year post eradication protocol. Of the remaining 17 patients who did not receive the protocol due to persistent MRSA infection, 3 (18%) cleared MRSA on their own, and 14 (82%) remain colonized with MRSA. The overall results reveal MRSA clearance rates pre-protocol of 18% and post protocol of 100%, with a P value of 0.002. Based on these findings, we conclude that utilization of our protocol is positively associated with eradication of first acquisition of MRSA. Purpose: Cystic fibrosis (CF) patients are often treated with aminoglycoside (AG) antibiotics during an infective pulmonary exacerbation caused by Pseudomonas aeruginosa. Achieving pharmacokinetic and pharmacodynamic (PK/PD) targets to improve outcomes and counteract increasing rates of resistance is paramount. The primary objective of this study was to compare the number of pediatric CF patients achieving AG PK/PD targets when therapy is managed by a pharmacist (P) compared with nonpharmacist (NP)

The study cohort included 1,276 individuals. Mean age was 22.4 ± 10.4 years; 47.2% were female. Mean BMI was 22.2 ± 4.3 and mean height was 63.4" ± 5.7". In the study cohort, 46.8% were delF508/delF508 and 94.1% were pancreatic insufficient. Mean FEV 1 was 2.3 ± 0.9L (72.3% ± 24.3% predicted). Mean FVC was 3.2 ± 1.1L (85.0% ± 20.0% predicted). The cohort contained 79.6% on TOBI, 85.0% on Pulmozyme and 57.4% on hypertonic saline. Finally, 31.0% had MRSA and 4.6% had B.cepacia.

There was a reduction in pulmonary exacerbations requiring hospitalization in individuals with P. aeruginosa on chronic macrolide therapy, incidence rate ratio (IRR) of 0.85 (p-value 0.043).

Conclusions: In an observational cohort of individuals with CF with P. aeruginosa, the rate of hospitalizations for pulmonary exacerbations decreased for patients treated for two years with chronic macrolide therapy. This is consistent with prior studies showing short term benefits of macrolide therapy, and macrolide therapy should be considered in individuals older than six with CF and P. aeruginosa. 

Introduction: Improved survival in cystic fibrosis (CF) has been seen with aggressive antibiotic usage. However, identification of new bacteria and fungi from sputa of CF patients is increasingly being reported. Whether this is the result of more sensitive microbiological techniques allowing for the isolation of organisms that have always been present makes the implications of these isolates unclear. Recently Exophiala (Wangiella) dermatitidis (Ed) has been identified in sputum of CF pts. The purpose of this study was to report our recent experience of this pathogen and determine the clinical significance in CF pts.

Methods: All respiratory cultures of CF pts attending clinic from 2008-2010 were screened. Charts for all pts who isolated Ed ≥1 occasion were reviewed for information on clinical status, co-infection and management.

Results: Seven of 120 (6%) pts isolated Ed ≥1 occasion during the 2 yr period. The age range of pts was 10-18 yrs (median 14 yrs; male 4). All Ed isolates are being typed to determine the strain of Ed. Table 1 summarises the clinical features and progress of all 7 pts from 6 months prior to 6 months post first positive Ed culture. The median change in FEV 1 % predicted during this time period was -9% (range -16 to +1%) and the median change in absolute BMI value was -0.2. Six of 7 (86%) pts isolated atypical mycobacteria during the same time period as having Ed. Four co-isolated both organisms simultaneously and 2 isolated Ed within 4 months after culturing atypical mycobacteria. Six of 7 (86%) of pts had ≥1 hospital admission during the time Ed was isolated. Treatment with oral voriconazole was given to 3 pts. One pt received voriconazole for 6 months and has not regrown Ed. One pt was treated for 3 months and regrew Ed 5 months later. One pt was unable to tolerate voriconazole due to visual disturbance but continues to grow Ed and is on nebulised amphotericin.

The clinical significance of Ed infection and response to treatment is difficult to determine in this small series. However it is possible that Ed contributes to clinical findings in CF pts and may require specific therapy. The high rate of co-infection with atypical mycobacteria seen in our population may be significant and suggests the need to actively search for this pathogen in pts isolating Ed failing to respond to usual CF treatments. Previous reports suggest Ed infection responds to treatment with antifungal agents and although this has been attempted in selected pts from our case series we cannot as yet determine the effectiveness of such therapy. A major cause of chronic airway infection in cystic fibrosis is biofilm formation by P. aeruginosa. Biofilms impede antibiotic penetration and increase antibiotic resistance. A small molecule that inhibits biofilm formation or disrupts/weakens existing biofilms may reduce P. aeruginosa colonization and recurrent lung infection in cystic fibrosis. Available iron chelators to inhibit biofilm formation are too toxic for clinical use. We conducted high-throughput screens to identify compounds that inhibit P. aeruginosa biofilm formation and/or growth. P. aeruginosa strain PA01 at mid-log phase was grown in 96-well plates (initial concentration 1 million cfu/ml) for 9 h at 37 o C in the presence of test compounds or vehicle (DMSO) alone. Growth was measured by optical density, and biofilm formation by crystal violet staining after washing to remove non-biofilm-associated bacteria. The screen was robust with Z'-factors greater than 0.5-0.6 for identification of antimicrobials and biofilm inhibitors. We screened 100,000 synthetic small molecules at 50 and 100 µM concentration, as well as smaller collections of approved and investigational drugs, as well purified plant-and microbederived natural compounds. Screening revealed known antimicrobials that inhibited P. aeruginosa growth, as well as several novel small-molecule antimicrobials, the most potent of which, a quinolone derivative, had IC50 6.3 µM. Two classes of compounds were identified that inhibited biofilm formation without affecting P. aeruginosa growth. After preliminary optimization by structure-activity analysis of more than 300 chemical analogs, the most potent class of compounds, benzothienopyrimidines, inhibited biofilm formation with IC50 ~ 3.1 µM. Biofilm inhibition was confirmed using alternative peg-lid and flow-cell methods, and by confocal imaging of fluorescently labeled bacteria. Following further optimization and compound prioritization, and testing on clinical isolates, the best compound(s) will be evaluated in a mouse model of P. aeruginosa infection. Non-toxic inhibitors of P. aeruginosa biofilms may be useful for therapy of chronic lung infection in cystic fibrosis. This work was supported by the CFF and NIH.

tutive auto-inducer production. The speed of biofilm spread is greater with higher auto-inducer production by planktonic bacteria and bacteria in the biofilm.

Conclusions: Our mathematical model shows that low rates of diffusion of planktonic bacteria and auto-inducer molecules are essential for biofilm formation. Higher rates of bacterial auto-inducer production increase both biofilm formation and biofilm spread. Reducing the viscosity of mucus in the mucociliary tract may discourage the formation of biofilms, although the necessary level of reduction depends on the properties of the colonizing bacteria.

Acknowledgements: This study was supported in part by the CF Foundation, the Ben B and Ira M Margolis Family Foundation of Utah, the Modeling the Dynamics of Life Fund and the NIH (HL-084341). The EPIC trial is a recently completed study that randomized 304 participants with cystic fibrosis ages 1-12 years to one of two treatment regimen frequencies using tobramycin inhalation solution with or without oral ciprofloxacin. To qualify for the study, participants had to have new onset Pseudomonas aeruginosa (Pa) infection defined either as a first lifetime documented positive respiratory culture for Pa or a positive Pa culture after at least a two year absence of Pa. Thus, all 304 participants were positive for Pa at screening. At the subsequent baseline visit, only 40% percent of the participants were found to be Pa positive. By protocol, the screening culture had to be obtained within six months prior to baseline, during which time participants were allowed limited antibiotic use (no more than one course of intravenous or inhaled anti-pseudomonal antibiotics). The aim of this study was to examine differences in antibiotic usage and other clinical parameters prior to the start of the study amongst those who were Pa positive and Pa negative at baseline.

Methods: Participants were grouped by their Pa status at baseline and by various clinical parameters including age, genotype, Pa history, and time between the screening culture and baseline visit. Prior antibiotic use was examined in the period between the screening and baseline culture.

Results: Of the 295 participants with a baseline respiratory culture, 177/295 (60%) became Pa negative at baseline and 118/295 (40%) remained Pa positive. Participants who were Pa negative at baseline were slightly older in age than those who were Pa positive (Mean age: 5.8yrs [SD=3.5] vs. 5.5yrs [SD=3.6], p=0.38). Median duration between screening culture and baseline visit was 67 days among Pa negative participants and 19 days for those Pa positive (p<0.01). Neither history of Pa positivity nor genotype was significantly different between Pa positive and Pa negative participants at baseline (p=0.20 and p=0.92 respectively). Amongst the Pa negative participants, 98/177 (55%) had at least one course of antibiotics between their screening and baseline visits as compared to 26/118 (22%) of Pa positive participants (p<0.01). Overall, 79/295 (27%) of participants became Pa negative at the baseline visit without having used antibiotics. In contrast, 26/295 (9%) were treatment failures, defined as those participants that remained Pa positive despite having taken antibiotics.

Conclusion: Antibiotic use between screening and baseline visit was the key factor defining Pa positivity at baseline. A significant proportion of participants became Pa negative without the aid of antibiotics, indicating possible spontaneous eradication of the organism in some participants with new onset Pa. Additionally, a small proportion of participants had persistent Pa positivity in spite of antibiotic use prior to entering the trial.

Supported by the CFF and NIH U01-HL080310 and ULI-RR025014-03.

Bean, H.D.; Hill, J.E. School of Engineering, University of Vermont, Burlington, VT, USA Background and Objectives: A primary cause of morbidity and mortality for cystic fibrosis (CF) patients is chronic Pseudomonas aeruginosa (P.a.) lung infection. In the lung, P.a. acquires mutations that allow it to persist in its new environment. Some of these mutations, including mucoid conversion, cause or are correlated with metabolic changes in the bacteria, which may be observed as changes in the volatile metabolites produced by the bacteria, and exhaled in human breath. We are using secondary electrospray ionization-mass spectrometry (SESI-MS) to identify the volatiles that are indicative of P.a. adaptation to the CF lung. Our goals are to develop a rapid test to detect, track, and characterize P.a. infection and mutations in situ in order to initiate early, targeted treatment, which is essential to managing infection and maintaining healthy lung function in CF patients.

Methods: Two non-mucoid P. aeruginosa strains (PAO1 and PA14) and one mucoid CF-derived strain (FRD1) were cultured aerobically for 24 h at 37°C in 50 ml of a complex medium (LB or synthetic CF sputum medium (SCFM)), or a defined medium with single or dual carbon sources (MOPS with succinate, glucose, and/or pyruvate). Mass spectra of the headspace volatiles in the mass range of 20 -500 Da were collected over two minutes (40 scans) in positive ion-mode.

Results: In all of the media tested, each strain produced unique combinations of volatiles that made it possible to distinguish between the nonmucoid non-CF strains PAO1 and PA14 and the mucoid CF-derived strain FRD1.

Discussion and Conclusions: Our data demonstrate the utility of SESI-MS in real-time identification in vitro of P.a. mutants that are characteristic to chronic infections of the CF lung. We are in the process of identifying the unique compounds produced by FRD1 and correlating those compounds to metabolic pathways, which will yield specific biomarkers for CF-adapted mucoid P.a. Tracking the biomarkers for P.a. mucoidy via breath analysis will allow physicians to detect mucoid conversion in the lung prior to the successful culture from sputum or BAL isolates, which have the risk of false negatives. This potential for early detection will facilitate the implementation of earlier, better-targeted treatments for P.a. infection.

Funding: This study was supported by NIH P20 RR021905-01. Objective: In cystic fibrosis, viscous mucus slows the removal of inhaled pathogens via the mucociliary tract, providing bacteria with more time to infect the airway and allowing sufficient time for bacterial biofilm formation. Previous researchers have suggested that the reduced diffusion of bacteria, particularly Pseudomonas aeruginosa, and auto-inducer in CF mucus promotes biofilm formation. We address the following questions regarding the effect of mucus viscosity on biofilm formation:

What conditions promote biofilm formation? When will the biofilm formation occur? How fast does the biofilm continue to grow once formed? Methods: We propose a system of nonlinear partial differential equations to describe a bacterial population that relies on an auto-inducing signal to create a biofilm. These equations include a medium, like mucus, for the bacteria in which the diffusion rate of both auto-inducer and planktonic bacteria can be altered. Results were obtained through analytical methods and simulations.

Results: When the auto-inducer level remains below a threshold, a wave of planktonic bacteria moves through the medium with no biofilm production. If the threshold is surpassed, a biofilm forms and spreads. Just above this threshold, the biofilm contribution to auto-inducer production is low and spreading planktonic bacteria leave a succession of small biofilms in their wake. The time taken for biofilm initiation decreases with the rate of consti-Background: Various CF centres from Europe and North America have reported increasing prevalences of non-tuberculous mycobacterial colonisation/infection over the last years. The aim of the present study was to analyse the frequency of non-tuberculous mycobacterial detection among patients at the Copenhagen CF Centre from 1991-2010.

Method: From the central mycobacterial laboratory at Statens Serum Institut all samples positive for non-tuberculous mycobacteria from CF patients were retrospectively identified. The patient files were scrutinized for clinical manifestations and treatments.

During the study period sputum samples for mycobacterial culture were not routinely done. Samples were collected when clinically indicated. Furthermore, when clinically indicated mycobacterial analyses were performed on bronchoalveolar lavages.

Results: During the study period between 241 and 302 patients were seen annually at the Centre. A total of 34 patients had non-tuberculous bacterias detected м1 time, mainly from respiratory samples. Mycobacterium abscessus complex, M. avium and M. gordonae were isolated from 22, 11 and one patient respectively. Ten patients had only one positive sample. The median number of positive samples was 3,5 (range 1-72). During the study period the number of samples analysed increased with a four-fold increase from 2001 to 2009. No increase in non-tuberculous mycobacterial detection rate was found. Twenty-three patients received anti-mycobacterial treatment, including three patients with only one positive culture. The regimens and duration of treatment varied greatly.

Conclusions: Over a 19-year period the non-tuberculous mycobacteria detection rate among CF patients at the Copenhagen CF centre was stable, with a predominance of Mycobactium abscessus complex infections. USA300 is the most common community-acquired S. aureus pulsotype and carries staphylococcal cassette chromosome (SCC) mec type IV and Panton-Valentine leukocidin (pvl), a cytotoxin associated with invasive infections in otherwise normal children. The USA100 pulsotype typically carries SCCmec type II, lacks pvl, and has been associated with hospital-acquired MRSA infections. No comprehensive study has examined the role of various S. aureus virulence factors and their effects on outcomes in children with CF. The purpose of this study was to describe the current epidemiology of S. aureus in CF patients and to identify specific S. aureus virulence factors associated with selected outcomes. We hypothesized that pvl would be associated with lower lung function and BMI or weight-for-length (W/L) in pediatric patients with CF.

Methods: A prospective study of all pediatric CF patients between May 2008-May 2009 at Texas Children's Hospital. Sputum isolates of S. aureus from routine out-patient clinic visits were tested for antibiotic susceptibility and analyzed by pulsed field gel electrophoresis and PCR to identify pulsotype, pvl genes, and SCCmec type. Patients who did not grow S. aureus during the study period served as controls. Demographic and clinical data was collected from corresponding visits from the CF Registry. Unpaired ttests were used to compare means.

A total of 287 patients were enrolled with a mean age of 10.2 years and mean FEV 1 84.2% predicted. S. aureus was isolated in 71.8% of patients (n=206). Of these, 39.3% (n=81) had MRSA. Presence of MRSA was associated with lower FEV 1 compared to MSSA (79.9% vs. 89.3% predicted, p=0.018). Among patients with MRSA, the USA100 pulsotype (48.1%, n=39) and USA300 pulsotype (39.5%, n=32) were most commonly identified. The ages of children with USA100 v. USA300 or with MRSA v. MSSA were not different. The majority of USA100-related MRSA strains were clindamycin-resistant (92.3%, n=36), pvl-(94.7%, n=37), and SCCmec II (97.4%, n=38), whereas the majority of USA300-related MRSA strains were clindamycin-susceptible (75%, n=24), pvl+ (93.8%, n=30), and SCCmec IV (100%, n=32). FEV 1 was substantially lower for USA100 v. USA300 (67.7% v. 93.1% predicted, p<0.001). Among patients with MSSA (n=125), pvl was present in 8.8% of the isolates (n=11). Unique pulsotypes (57.6%, n=72 ) and USA200 (18.4%, n=23) were most prevalent. FEV 1 was not different among MSSA patients with regards to pulsotype, antibioticresistance, and presence of pvl. BMI or W/L % predicted did not correlate with any S. aureus virulence factor.

Conclusions: The prevalence of S. aureus, particularly MRSA and the USA100 and USA300 pulsotypes, was high in our pediatric CF patients. MRSA was associated with lower lung function than MSSA. Moreover, specific virulence factors, including the USA100-related pulsotype, clindamycin-resistance, the lack of pvl, and SCCmec II, were associated with lower lung function. Pseudomonas aeruginosa uses a number of strategies to colonize and persist in the CF lung. One recently discovered strategy is secretion of a virulence factor called Cif. Cif is capable of reducing trafficking of CFTR to the apical membrane of epithelial cells, thus diminishing the amount of CFTR present in infected lungs. Expression of this virulence factor is variable, with highest levels of expression seen in early-colonizing clinical isolates of Pseudomonas aeruginosa, suggesting its importance in early infection. Our goal is to decipher the regulation of this virulence factor to better understand its role in Pseudomonas infection. One previously identified regulator of Cif, CifR, is a TetR family repressor of gene expression. We know that this regulator directly binds to the promoter upstream of the cif gene to repress its expression, however we do not know the location of its binding site or if there are any other regulators that modify Cif expression. Using electrophoretic mobility shift assays (EMSA) and promoter mapping, we locate two distinct binding sites of CifR upstream of cif. Through mutational and overexpression analysis we also characterize another regulator of cif expression, CifA which is a LysR family repressor of cif expression. We further show, via expression analysis of these two regulators, that the regulation of this system is complex, with a dynamic interplay of repression. From these analyses we can conclude that CifR and CifA tightly regulate cif expression and that they are important for the regulation of Pseudomonas response to the host environment. Although the controls received no antibiotic treatment, the percentage of resistant strains in their periodontal pockets was similar to that of CF patients. Obviously, the frequent treatment with antibiotics in CF patients does not increase the percentage of resistant strains in their oral cavity. However, identical facultative and obligate anaerobic bacteria in CF periodontal pockets and sputum samples suggest the oral cavity as a possible source for lung infection with these bacteria. 

Background: Chronic P. aeruginosa (Pa) infection is an unfavorable event for CF patients. Eradication at first/new isolation could be a simple way to improve prognosis. There are no publications comparing the efficacy of the various treatments.

Aims: The main objective of this randomized trial was to analyze the efficacy of two early Pa infection-eradicating treatment protocols: inhaled (inh) colistin/oral cipro (arm A) versus inh tobramycin/oral cipro (arm B) for 28 days in a sizeable number of CF patients. Patients who were Pa-free (never colonized or Pa-free with at least three cultures negative in the last 6 months) and older than 1 year were considered eligible to participate in this trial.

The treatment was considered efficacious if patients were infection-free with at least 3 cultures negative over 6 months.

Results: Thirteen Italian centers participated in this trial and 215 patients were randomized. Of the 215 patients, 103 (47.9%) were treated with inh colistin/oral cipro; 112 (52.1%) out of 215 patients were treated with inh tobramycin/oral cipro.

Pa first/new infection occurred in 36.1% of patients aged 1-5 years; in 38.2% of patients aged 5-12 years and in 25.7% of patients older than 12 years. A total of 101 patients, free from Pa infection and eligible for the study, were previously infected by Pa, whereas it was the first ever infection for the remaining 114 patients. Mean age at first Pa infection was 8.2±7 yrs. We did not observe any difference (p=0.8) in age at first colonization between dF508 homozygotes or dF508 compound heterozygotes with stop mutations (median 7 years, range 1-32) and patients with another genotype (median 5 years,range 1-36).

At present a total of 155/215 (72%) patients have completed the 6month follow-up, 69 patients in arm A and 86 in arm B; 26 patients withdrew from the study (11 in arm A and 15 in arm B).

Using one single cycle of antibiotic treatment Pa eradication was achieved in 36/69 patients in arm A and in 49/86 in arm B. According to the 4th interim analysis, no statistically significant difference was observed between the two arms and boundaries (O'Brian-Fleming's method) were not exceeded.

Conclusions: We have analyzed the demographic characteristics of a large group of patients with first/new Pa infections as part of a multicenter, randomized, controlled trial investigating the efficacy of two eradicating treatment protocols.

According to the 4th interim analysis, 24 months after the beginning of the study, we have found similar treatment efficacy by using one single cycle of antibiotic treatment for 28 days, but at the moment we cannot reject the null hypothesis regarding absence of differences between the two types of treatment in relation to primary outcome. preparation of tobramycin in young children who were culture-positive for P. aeruginosa (the IT004 trial [1] ) led to lower airway culture-negativity in most, but not all, treated study subjects. We sought to determine whether in vitro bacterial characteristics would correlate with eradication. We examined (1) biofilm formation without tobramycin, (2) inducibility of biofilm formation by tobramycin, and tobramycin inhibitory concentrations of the isolates grown suspended in (3) liquid culture (MIC) and (4) when grown as a biofilm (BIC), and defined the relationship between these characteristics and the occurrence of P. aeruginosa eradication in source patients.

Methods: P. aeruginosa isolates used for this project were from the IT004 trial that included 75 subjects: 38 with a P. aeruginosa-positive baseline bronchoalveolar lavage (BAL) culture who were assigned to one of four cohorts treated with inhaled tobramycin, and 37 subjects with a P. aeruginosa-negative baseline BAL culture, of whom 26 had a positive oropharyngeal culture during the study. The resulting isolate set included 300 P. aeruginosa isolates. Biofilm formation and induction were analyzed using standard crystal violet staining procedures in polystyrene 96-well plates as described [2] [3] [4] . Tobramycin MICs were determined by broth microdilution. Tobramycin BICs were measured using a 96-well polystyrene peg system as described previously [4] . These bacterial characteristics were examined using descriptive statistics for their relationship to eradication status during the IT004 trial.

Results: No significant relationship between biofilm inducibility or suppression by tobramycin, biofilm formation, or tobramycin MIC and eradication was found. However, P. aeruginosa isolates from subjects in whom eradication failed to occur did tend to have higher BICs than did isolates from subjects in whom eradication occurred. In post-hoc analyses, when considering the most tobramycin-resistant BAL strain at baseline, among the 31 subjects evaluable for eradication, isolates from 12/23 (52%) subjects in whom eradication occurred had BICs less than 4 µg/mL. In subjects with persistent infection, 0/8 (p=0.01) had BICs in the susceptible range.

Conclusions: Standard MICs did not predict eradication by inhaled tobramycin. Similarly, no relationship could be demonstrated between eradication and biofilm formation with or without tobramycin treatment. There appeared to be a relationship between tobramycin BIC and likelihood of eradication with inhaled tobramycin alone. This observation must be investigated in a larger, directed study. Facultative and obligate anaerobic bacteria induce chronic lung infections in CF patients. The frequent use of antibiotics may lead to increased bacterial resistance. Here, we investigated if periodontal pockets may represent sources of CF lung infections, and if bacteria in the oral cavity of CF patients show higher percentages of resistant strains compared to healthy controls.

Facultative and obligate anaerobes were identified using the Crystal® (Becton Dickinson, Heidelberg, Germany), the Vitek® (BioMérieux) and the Rapid AnaII® (remel, Lenexa, KS, USA) identification systems in 9 CF sputum samples and swabs from periodontal pockets of 11 CF patients and 11 healthy controls. Antibiotic sensitivity profiles were obtained using etests (AB Biodisk, Solna, Sweden) for ceftazidime (CAZ), piperacillin/tazobactam (TZP), meropenem (MRP), azithromycin (AZM), colistin (CS), metronidazole (LZ), and clindamycin (CD).

In all CF patients (100%), identical obligate anaerobic genera were found in periodontal pockets and in sputum. In 6 out of these patients (66.6%), 10 pairs of identical obligate anaerobic species (Staphylococcus Background: Non-tuberculous mycobacteria (NTM), ubiquitous environmental acid-fast bacteria, are potential respiratory pathogens in cystic fibrosis (CF). Their prevalence has been increasingly reported (about 13%) and varies greatly depending on the center. Isolation of NTM may be a problem because sputum decontamination is sometimes unable to eradicate overgrowing bacteria, particularly Pseudomonas aeruginosa, so that realistic epidemiological data are difficult to achieve.

The clinical impact on lung disease in CF is difficult to establish: ATS criteria do not apply to CF patients (pts) because of the overlap of clinical and radiographic findings between NTM disease and underlying CF. Therapeutic options, when lung damage is demonstrated, are often ineffective and NTM eradication may be hard due to the presence of multiresistant strains.

Aim: Carry out a systematic NTM infection surveillance strategy among a cohort of adult CF pts.

Methods: A total of 190 CF adult pts attending the CF Adult Center of Milan (76 females, 114 males, mean age: 36.2 yrs, range: 21-69.5 yrs) were studied. In the period January 2009-April 2010, pts able to produce sputum were investigated for NTM pulmonary infection. Sputum samples analysis included smear microscopy and culture.

Results: Of the 190 pts, 110 (57.8%) pts were screened for NTM and provided at least one sputum specimen for analysis. Multiple samples in 26/110 (23.6%) pts were undiagnostic due to unsuccessful decontamination: 15/26 of them (57.8%) were chronically infected by Pseudomonas aeruginosa, 4/26 (15.4%) by Achromobacter xylosoxidans, 1/26 (3.8%) by Burkholderia cepacia, 4/26 (15.4%) by Pseudomonas aeruginosa and Achromobacter xylosoxidans, 1/26 (3.8%) by Pseudomonas aeruginosa and MRSA, 1/26 (3.8%) by Pseudomonas aeruginosa and Stenotrophomonas maltophilia. A total of 71/110 (64.5%) were negative and 13/110 (11.8%) pts were positive for NTM (Table) .

All NTM positive pts were submitted to a clinical and radiological follow up (clinical evaluation every two months, microbiologic analysis twice in a year and annual TC chest scan) to promptly detect any significant change. At present only one patient with a pulmonary lesion consistent with NTM infection is on specific therapy for M. avium; another patient was unsuccessfully treated for M. abscessus in November 2007. Negative patients include a woman who was eradicated for M. chelonae in 2005 with adequate chemotherapy.

Conclusions: An extensive microbiologic surveillance for NTM pulmonary infection is necessary to clarify the pathogenic role of NTM and prevent progressive lung damage. Background: FTI is a combination antibiotic with antibacterial activity against gram positive and gram negative pathogens. To test whether FTI could augment or maintain decreases in PA sputum density achieved by AZLI therapy, this study examined changes in PA sputum density in 119 subjects age ≥18 years with CF, pulmonary PA, and FEV1 ≥25% and ≤75% predicted.

Methods: At each study visit (Day -28, 0, 14, 28, 42 and 56) sputum samples were collected, processed according to the CF Foundation standards, and cultured for PA and other CF-associated pathogens. Changes in CFU sputum density of pathogens were evaluated. The antibiotic susceptibility of PA isolates to fosfomycin, tobramycin and a combination of fosfomycin and tobramycin in a 4:1 ratio (F:T) were tested using an agar plate dilution technique.

Results: The reduction in PA sputum density achieved by AZLI from Day -28 to Day 0 (-0.7 log 10 CFUs/g) was maintained through Day 56 in the FTI 80/20 group but not in the FTI 160/40 or placebo groups (Table) . The MIC50 and MIC90 of all PA isolates remained unchanged (≤2 fold increases) for tobramycin and F:T in all three treatment groups from Day 0 to Day 56. A ≥4-fold increase in fosfomycin MIC90 was seen at Days 14-56 in the FTI 80/20 mg group. At Day 0, 26 patients (21.8%) were co-infected with PA and MRSA. A >2 log reduction in MRSA sputum density was observed in subjects with MRSA in the FTI groups (table) . Emergence of other pathogens was rare and similar across groups.

Conclusion: FTI demonstrated activity against PA and MRSA and antibiotic effects were maintained past the 28-day treatment period. No changes in susceptibility to F:T were observed in PA isolates during or after FTI therapy. Treatment-related emergence of other CF pathogens was infrequent and similar between treatment groups during the 56-day study period.

Supported by Gilead Sciences, Inc.

* p ≤ 0.05 versus placebo (ANOVA); + p ≤ 0.05 versus placebo (Wilcoxon) detect PVL encoding genes. Selected isolates were grown overnight under both aerobic and anaerobic conditions, the supernatant heat treated, and the production of delta-haemolysin (d-hly), a marker of agr function, measured by incubation with red blood cells to determine levels of haemolysis. The expression of protein A (SpA) was visualised by Western Blotting of outer membrane protein extracts. Results: Isolates from UNC and BPCF centre formed two genetically distinct pulsotypes, with approximately 40% homology in PFGE patterns between groups. Two isolates from 2 patients in UNC, and no BPCF isolates, were PVL positive. MRSA isolates within each pulsotype displayed highly similar banding patterns (>80% homology), and in some cases were considered identical. Twenty of 22 BPCF isolates were SCCmec IV, spa type t032, similar to EMRSA-15. UNC isolates were either SCCmec II spa type t992/t002 (5/11) or SCCmec IV [spa type t992 (1/11); spa type t681 (USA 300; 2/11)]. In 19/28 (68%) isolates, d-hly production was increased when the isolates were grown under anaerobic conditions with no significant difference observed for 8/28 (29%) isolates and a decrease observed for 1/28 (3%) isolates. This difference was marked in isolates expressing agr II (predominantly UNC isolates). In contrast, SpA was detected in isolates grown under aerobic conditions but was either not detected, or detected at lower levels in isolates grown anaerobically. Heterogeneity in the molecular weight of SpA was observed between isolates.

Conclusions: MRSA cultured from paediatric CF patients appears to reflect strains in circulation in the respective hospitals and form closely related groups within each centre, in terms of PFGE, agr and SpA type. Growth in an anaerobic environment significantly effects the expression of staphylococcal virulence factors. D-hly levels are increased in isolates grown anaerobically, indicating increased levels of RNA III, the key effector molecule of the agr system. Consistent with this increased expression, observed levels of SpA are reduced under anaerobic conditions. This differential expression of key staphylococcal virulence factors under aerobic and anaerobic conditions and its potential immunomodulatory effect, together with further measurement of upstream virulence regulators, is currently under investigation. Methods: Data for the analyses were obtained from a Phase 2 study evaluating three dosage regimens of MP-376 (120 mg QD, 240 mg QD and 240 mg BID) given for 28 days to stable CF patients. MP-376 doses were administered using a customized investigational configuration of a PARI eFlow nebulizer. All PPK analyses were conducted using Monte-Carlo parametric expectation maximization implemented in S-ADAPT 1.5.6. PK exposures were predicted by simulating serum and sputum PK profiles using the final PPK parameter estimates for each patient.

Results: The PPK dataset contained 527 serum and 191 sputum concentrations from 112 patients. The mean sputum LVX PK values for all doses are shown in the Table. Mean exposures in serum and sputum increased proportionately with dose. All doses produced LVX concentrations that were high in sputum, resulting in significant dose-related reductions in P. aeruginosa density, and relatively low in serum. MIC 50 and MIC 90 did not change significantly after exposure to MP-376 for 28 days.

Conclusions: A relatively simple PPK model adequately described serum and sputum levofloxacin concentrations in a Phase 2 clinical trial. All doses of MP-376 produced high sputum concentrations resulting in PK-PD The intracellular signaling molecule, cyclic-di-GMP, has been shown to influence surface-associated behaviors of Pseudomonas aeruginosa, including biofilm formation and swarming motility. Given that P. aeruginosa is able to chronically colonize the lungs of cystic fibrosis patients leading to an increase in morbidity and mortality, we would like to better understand the mechanisms by which the c-di-GMP signaling pathway impacts surface behaviors of this bacterium.

Previously, we reported a role for the bifA gene in the regulation of biofilm formation and swarming motility of P. aeruginosa. The bifA gene encodes a c-di-GMP phosphodiesterase and the ᭝bifA mutant exhibits increased cellular pools of c-di-GMP, forms hyper-biofilms and is unable to swarm.

To better understand how c-di-GMP represses swarming motility, we isolated suppressors of the ᭝bifA swarming defect. Several suppressor strains had mutations that mapped to the pilY1 gene, previously implicated in twitching motility (another form of surface motility used by P. aeruginosa) and epithelial cell adhesion. Mutations in the pilY1 gene, but not the pilin subunit pilA gene, show robust suppression of the swarming defect of the ᭝bifA mutant, as well as its hyper-biofilm phenotype. We also showed that enhanced expression of the pilY1 gene represses swarming motility of the WT strain and this swarming repression specifically requires the diguanylate cyclase (DGC) SadC. Epistasis analysis indicates that PilY1 functions upstream of SadC.

In this study, we sought to identify and characterize additional components that function together with pilY1 in the repression of swarming motility. The pilY1 gene maps to a region containing other pili-related genes and has been shown to be co-transcribed as an operon with the members of this locus (fimUpilVWXY1E). The fimU, pilV, pilW, pilX and pilE genes encode pilin-like proteins (termed "minor" pilins)) that are important for assembly of a functional pilus. We generated mutations in each of the minor pilin genes in the ᭝bifA mutant background and found that mutations in pilV, pilW and pilX show robust suppression of the swarming and hyper-biofilm defects. We are currently working towards identifying the mechanism by which these minor pilins participate with PilY1 in repression of swarming motility.

In conclusion, our results indicate that members of the pilY1 locus participate in the c-di-GMP-mediated repression of swarming motility. Therefore, this locus impacts all three surface-associated behaviors of P. aeruginosa -twitching, swarming and biofilm formation and thus, might impact the ability of this bacterium to colonize a variety of niches, including the CF lung environment.

MRSA is recognized as a potential contributor to declining lung function in CF patients but its precise contribution to disease progression remains unclear. Since oxygen can be limited in the CF lung, this study aimed to determine the impact of growth in an anaerobic environment on the expression of key virulence factors in MRSA isolates from paediatric and adult CF patients.

Methods: MRSA isolates were cultured from nasal and cough swabs (22 isolates from 21 patients in the Belfast Paediatric CF centre [BPCF] ) and bronchoalveolar lavage fluid (11 isolates from 7 patients in the University of North Carolina [UNC]). Isolates were characterized by pulsed field gel electrophoresis (PFGE) and SCCmec, spa and agr typing. PCR was used to indices (Cmax/MIC and AUC/MIC) consistent with the dose-related reduction in bacterial density in sputum and the absence of significant changes in LVX MIC observed during the trial. Low serum levofloxacin levels at all dose levels were associated with good tolerability. Phase 3 clinical trials are planned for the 240 mg BID dose.

Supported by a grant from Mpex Pharmaceuticals. Introduction: Burkholderia cepacia (BC) complex infects approximately 2.9% of CF patients. Infection with BC is associated with rapid decline in lung function, increased mortality and rarely, fulminate sepsis known as cepacia syndrome. Various virulence factors and anti-microbial resistance have historically made this organism difficult to treat. While no standard treatment strategy exists, reports of successful eradication in the literature support the attempted eradication of this pathogen.

Purpose: This study was conducted to evaluate the efficacy of a treatment regimen to eradicate BC.

Methods: Patients at the Pediatric Cystic Fibrosis Center of the University of Alabama at Birmingham/Children's Health System (UAB/CHS CF center) with positive cultures of BC from January 2006 to June 2009 were included in this study. Inclusion criteria: age less than 21 years with first positive culture for BC from sputum, BAL, or oropharyngeal swab who underwent antimicrobial therapy for eradication. Data collected included: patient demographics, source of sample, quantity of BC in samples, species of BC, antibiotic susceptibility, co-infections, treatment, number of days on therapy, treatment failure, number of inpatient days 1 year prior and 1 year post BC acquisition. Treatment regimen consisted of two empiric IV antibiotics (aminoglycoside and a beta lactam) given concurrently with an inhaled antibiotic (colistin or tobramycin) for three weeks as well as continuation of therapy with one oral antibiotic (sulfamethoxazole/trimethoprim or ciprofloxacin) and continued use of the inhaled antibiotic after IV therapy was completed. Some regimens also included chronic azithromycin therapy. Eradication was defined as one year, of at least quarterly cultures, with cultures negative for BC. Treatment failure was defined as positive cultures within one year.

Results: Eight patients were included in the study. Patients cultured B. cenocepacia (50%), B. multivorans (25%) and B. gladioli (25%). Five patients met the definition of successful eradication. There was no significant difference observed related to sputum microbiology, duration of therapy, or antimicrobial combinations. Following successful eradication, patients had a significant reduction in the number of inpatient days during the year following therapy (p = 0.04). The mean age at first culture was 43.6 months in the eradicated group compared to 139.3 months in the non-eradicated group. The mean age at the time of CF diagnosis and age at first culture for BC was younger in the group achieving eradication.

Our findings suggest that the combination of systemic and inhaled antibiotics with activity against BC over an extended duration can successfully eradicate the organism. Limitations of this retrospective study include a small sample size, non-standardized antimicrobial use and the absence of lower airway cultures after therapy. In this study we were able to achieve successful eradication in 62% of subjects under a reasonable definition of eradication. infection is increasingly found in patients with CF. We aimed to determine the prevalence and clinical impact of MRSA in a cohort of CF patients. We also studied the dynamics of MRSA strains in patients and the possibility for cross colonization between patients.

Patients and methods: This retrospective cohort study included 419 patients older than 7 years attending our pediatric or adult CF centers in 2005 and 2006. Patients had not received lung transplant and had several sputum samples available over those 2 years. They were classified in 3 groups of respiratory insufficiency according to their FEV1: severe (<40% pred.), moderate (40-60% pred.) or mild (>60% pred.).

The clonal distribution of the different strains of MRSA isolated from patients' sputum was analyzed: by antibiotic susceptibility testing and by pulsed-field electrophoresis (PFGE) after Smal digestion of chromosomal DNA. Strains harboring minor differences in the banding pattern (>80% similarity as assessed by the Dice coefficient) were considered clonal.

Results: Among the 838 patient years (PY), 127 (15.1%) were colonized with MRSA. Their mean age was 21.5±9.0 yrs, 56.7% were males, 84.1% had exocrine pancreatic insufficiency, 15.9% diabetes and 10.3% hepatic cirrhosis. Associated colonization was seen with Pseudomonas aeruginosa (PA) in 67.7%, methicillin-susceptible SA (MSSA) in 30.7%, Aspergillus fumigatus in 29.9%, Haemophilus influenzae in 22.0%, Stenotrophomonas maltophilia in 7.9%, Achromobacter xylosoxidans in 7.9% and Burkholderia cepacia in 2.4%.

A total of 109 MRSA strains from 63 patients were studied; 68% were resistant to more than three different antibiotic families. PFGE analysis on 97 strains revealed that MRSA were grouped in 12 clusters observed only in 27 patients. The other MRSA isolates seemed to be sporadic. Ninety-five percent of the patients were colonized with the same clone during the two years of follow-up.

Mean FEV1 was 57.7±27.0% pred. in all MRSA patients. We did not find any difference in the distribution of MRSA strains in our CF patients among the 3 classes of respiratory function. The yearly decline of FEV1 was -2.4±12.3% in all MRSA patients. Mean FEV1 and yearly decline in FEV1 were compared according to bronchial colonization (Table) .

Conclusion: A possible clonal identity of MRSA was observed for some patients but cross infectivity was low. Respiratory function of patients with MRSA was similar to that of patients with MSSA, but the association with PA worsened their respiratory status.

Supported by Vaincre la Mucoviscidose.

ial communities called biofilms. In previous studies, we showed that CF airway cells contain more iron than their WT corrected counterpart (1). Consequently, we developed a novel co-culture model and showed that enhanced iron release by CF airway cells facilitates biofilm formation by P. aeruginosa growing on airway epithelial cells (1) . In order to identify the mechanisms involved in the iron imbalance observed in CF cells, we investigated the expression of DMT1, the major divalent metal transporter involved in human iron metabolism. The DMT1 gene is known to produce 4 isoforms by alternative splicing of its N-and/or C-terminus ends. The presence of an Iron Response Element (IRE) in the 3'-UTR of the mRNA differentiates the C-terminus spliced variants and is thought to stabilize and regulate the (+)IRE species of DMT1. Here we examined the expression of DMT1 in a human airway cell line originally isolated from a CF patient (CFBE+∆F508) and in its isogenic WT corrected counterpart (CFBE+WT). We show that both cell lines constitutively express the DMT1 (-)IRE protein. Surprisingly, CFBE+∆F508 cells also express DMT1 (+)IRE constitutively, whereas this isoform was undetectable in CFBE+WT under the conditions tested. Using a commercially available DMT1 (+)IRE cDNA, protein expression of the (+)IRE isoform was achieved in CFBE+WT cells at levels similar to the ones found in CF cells. Overexpression of the (+)IRE isoform in CFBE+WT cells resulted in enhanced biofilm formation at the apical surface of these cells, with a biomass significantly higher than on CFBE+WT cells transformed with an empty vector (1.59 ± 0.10 vs 1.10 ± 0.09 µm 3 /µm 2 respectively, P=0.0009) and similar to the biomass observed on CFBE+∆F508 cells (1.40 ± 0.07 µm 3 /µm 2 , P=0.119). Intracellular iron levels were determined with the calcein assay, in which the degree of calcein fluorescence quenching measures the chelated iron inside cells. Both CFBE+∆F508 and CFBE+WT cells incorporated calcein-AM at the same rate. However fluorescence quenching was more pronounced in CFBE+∆F508 cells suggesting that they have a higher (~1.8-fold) intracellular iron concentration. This result is in accordance with our previously published ICP-MS data (1) , and validates the use of the calcein assay. Overexpression of the (+)IRE isoform of DMT1 in CFBE+WT cells enhanced calcein fluorescence quenching (i.e., enhanced iron accumulation) compared with CFBE+WT cells transformed with an empty vector. In conclusion, expression of the (+)IRE isoform of DMT1 in CFBE+∆F508 cells increases iron uptake and is responsible, in part, for increased biofilm formation by P. aeruginosa at the apical surface of airway epithelial cells.

Supported by the NIH (RO1 AI51360, RO1-HL-074175, P20-RR018787, T32-DK-07301) and the CFF (STANTO97RO). 

Background: The lungs of persons with cystic fibrosis become chronically colonized with microorganisms, significantly contributing to the morbidity and mortality of the disease. The microorganisms of primary importance are bacteria such as Staphylococcus aureus, Pseudomonas aeruginosa and Burkholderia cepacia complex. The prevalence of airway colonization with Aspergillus species varies widely depending on the study population but rates over 50% have been reported. Allergic bronchopulmonary aspergillosis is relatively common but despite high rates of Aspergillus spp. colonization, invasive Aspergillus spp. infection is very uncommon in those who have not been transplanted. Colonization with Aspergillus spp. is a predictor for worsening lung function, hospitalization and infection post-lung transplant. The role of antifungal therapy in persons with cystic fibrosis and Aspergillus spp. colonization remains unclear. A galactomannan assay (Bio-Rad Laboratories Platelia™ Aspergillus EIA) is approved for use in conjunction with other diagnostic procedures in the diagnosis of invasive aspergillosis in adults by detecting circulating galactomannan in human serum. There are studies in pediatric oncology patients suggesting serum galactomannan may also be useful in children but this has not been established. We sought to establish whether the galactomannan assay could detect circulating galactomannan in CF patients with chronic colonization and aid in identifying a subgroup that may need potential antifungal intervention. In an effort to identify breakpoints relevant to the clinical efficacy of inhaled antibiotics in CF, a retrospective analysis of clinical response to aztreonam for inhalation solution (AZLI; Cayston ® ) or tobramycin inhalation solution (TIS) therapy was conducted in relation to Pseudomonas aeruginosa (PA) susceptibility to aztreonam and tobramycin, respectively. Data from patients treated with 75 mg AZLI or placebo three times daily (TID) for 28 days in two Phase 3 double-blind, multicenter, randomized, placebo-controlled studies (AIR-CF1 and AIR-CF2) were pooled for analysis. TIS data are presented from patients who used ≥3 courses of TIS in the previous 12 months and received a 28-day course of TIS prior to AZLI or placebo in AIR-CF2. Three efficacy endpoints were used to categorize patients as improved, worsened or maintained at the end of 28-day therapy: CFQ-R respiratory symptoms score (RSS), pulmonary function (FEV 1 ) and sputum log 10 PA CFU.

Results: The highest MIC value analyzed was 64 µg/mL, as the number of patients with less susceptible PA isolates at study baseline was limited. Following a 28-day placebo course, approximately 25% of patients improved or worsened while 50% experienced no change, regardless of PA susceptibility to aztreonam. Following a 28-day course of AZLI TID, approximately 50% of patients improved, regardless of PA susceptibility to aztreonam. In a separate analysis following a 28-day course of TIS, fewer patients with highest tobramycin MIC for PA ≥4 µg/mL improved compared to patients with highest tobramycin MIC for PA <4 µg/mL. In patients with highest tobramycin MIC for PA ≥4 µg/mL, approximately 25% of patients improved or worsened while 50% experienced no change.

Conclusion: A susceptibility breakpoint was not identified for AZLI. In contrast, patients with highest tobramycin MIC for PA ≥4 µg/mL appeared to have a less robust clinical response to TIS than those with MIC for PA <4 µg/mL. Since standard clinically available sensitivity testing is usually the disk diffusion method, the parenteral breakpoint is not predictive of clinical efficacy for AZLI therapy but may be for patients on chronic TIS therapy.

Supported by Gilead Sciences, Inc. and stored at -20 o C. We included sera from patients chronically colonized with Aspergillus spp. as well as those who had never had Aspergillus spp. cultured from a clinical specimen. The frozen specimens were thawed and thoroughly mixed prior to testing. The sera were tested for galactomannan using a one-stage immunoenzymatic sandwich microplate assay (Bio-Rad Laboratories Platelia™ Aspergillus EIA), following manufacturer recommendations. An optical density index of ≥0.5 was considered positive.

Results: Seventy-five serum samples from 58 pediatric patients were tested, 23 males and 35 females. The average age at time of serum blood sample collection was 12 years 3 months (range 2 years 6 months to 17 years 8 months). Forty-one patients (71%) had at least one previous respiratory culture positive for Aspergillus spp. Two samples from one patient were collected after lung transplantation. All 75 serum samples had an optical density index < 0.5 and were therefore considered negative for galactomannan. There was no difference (p=0.53) in the mean optical density index between patients with at least one respiratory culture positive for Aspergillus spp. (mean optical density index 0.225) and those without a respiratory culture positive for Aspergillus spp. (mean optical density index 0.215).

Conclusions: In our study, detection of serum galactomannan in cystic fibrosis patients was uniformly negative, in both patients colonized with Aspergillus spp. and those not colonized. Background: Pseudomonas aeruginosa and Burkholderia cepacia complex are frequent colonizing bacteria in the lungs of adults with cystic fibrosis (CF) and are also responsible for acute pulmonary exacerbations. These bacteria are often multidrug resistant and methods to identify synergistic antibiotic combinations are needed. Results from multiple combination bactericidal testing (MCBT), a checkerboard titration technique performed elsewhere, are currently used at our institution to determine optimal antibiotic combinations. However, MCBT is labour and time intensive and not routinely done in the diagnostic medical microbiology laboratory. Recent reports suggest that using Etest for combination testing can be employed in the routine microbiology laboratory with good correlation to MCBT.

Methods: Twenty-six multidrug resistant P. aeruginosa and 21 multidrug resistant B. cepacia complex isolates from CF patients were each tested against three triple antibiotic combinations using Etest combination testing. Etest combination testing was performed as outlined in the Etest combination testing procedure as described by AB BIODISK. Synergy, antagonism and additive/indifference are defined as per the AB BIODISK Etest combination testing protocol.

Etest synergy testing results were correlated to MCBT results when MCBT results were available. Correlation was defined as: i) Etest results for a given antibiotic combination shows synergy AND MCBT results for the same antibiotic combination shows bactericidal activity; OR ii) Etest results show indifference or antagonism AND MCBT results do not show bactericidal activity.

Results of Etest synergy testing and MCBT results were not considered to correlate if: i) Etest results showed synergy AND MCBT did not show bactericidal activity; OR ii) Etest results showed indifference or antagonism AND MCBT showed bactericidal activity.

Results: Results are listed in Table 1 .

Etest combination antibiotic testing identified in vitro synergistic activity against P. aeruginosa and B. cepacia complex isolates from CF patients. There was poor correlation of Etest synergy testing results and MCBT results. Persistence of Pseudomonas aeruginosa in the lungs of cystic fibrosis patients is due to its biofilm forming capacity. The biofilm mode of growth contributes to the tolerance against treatment with antibiotics as well as to the cellular components of the host response especially the polymorphonuclear leukocytes (PMNs). In addition, the persistent biofilms stimulate immune complex mediated tissue damage resulting in the demise of the patient. We have previously demonstrated that P. aeruginosa biofilm engages quorum sensing regulated production of cytolytic amounts of rhamnolipid when interacting with PMNs in vitro. Therefore we hypothesized that P. aeruginosa biofilm also employs rhamnolipid as a cytolytic defence against PMNs during infection in vivo. To study the dynamics of the interplay between bacterial biofilms and the innate immune system, an interplay which has only been described in vitro, we developed an animal model using hollow tubes which are implanted in the peritoneal cavity of mice. We followed the pathogen-host interaction by means of scanning electron microscopy (SEM) at different time points to elucidate the role of rhamnolipids which causes the PMNs to undergo a necrotic death in vitro.

The implant tubes were colonized with wild-type (WT) P. aeruginosa (PAO1) or its corresponding rhlA mutant for 20 hours before installation in the peritoneal cavity of BALB/c mice. Tubes were removed from euthanized mice and prepared for evaluation by Scanning electrone microscopy (SEM).

The SEM images revealed that already 6 hours post-insertion an abundant number of PMNs had been attracted for both strains in order to clear the infection. Even more PMNs were seen in contact with the WT strain on day 1, 2 and 3, however, most of them appeared to be damaged and embedded in the biofilm matrix. The rhlA mutant, which is unable to produce rhamnolipid, were cleared from the tubes since hardly any bacteria were found day 1 post-insertion and intact PMNs were visualized.

The SEM images clearly emphasised the WT enabled destruction of the PMNs confirming rhamnolipids as one of P. aeruginosa's main virulence factors. This also confirms our previous findings, which propose that a rhlA mutant is cleared faster than the WT due to lack of rhamnolipid production. The observed interaction between the biofilm and PMNs may not only involve biofilms on implants but most likely also biofilms in the CF lung. been substantial effort made to understand the mechanism of signal reception.

It is generally believed that acyl-HSL dependent QS regulators require the presence of an appropriate acyl-HSL during polypeptide synthesis to fold into a functional conformation. In vitro experiments have led to the conclusion that once the folding is completed, the regulators do not release signal. However, there are some exceptions. For example, we previously found that purified his-tagged QscR was dependent on exogenous addition of signal for DNA binding. Therefore we concluded that QscR released signal during purification. We report here on the interaction of QscR and signal molecules in more detail.

We overexpressed native QscR in recombinant Escherichia coli grown in the presence of 3OC12-HSL, and purified it using buffers without any added acyl-HSL. Unexpectedly, QscR completely retained 3OC12-HSL throughout the purification. Furthermore, a significant portion of QscR from E. coli grown in the presence of N-(β-ketocaproyl)-homoserine lactone (3OC6-HSL) aggregated after each purification step. The aggregated QscR contained very little 3OC6-HSL whereas soluble QscR retained 3OC6-HSL. We conclude that QscR releases signal, but if the protein is at a sufficient concentration, in the nanomolar range, there will be sufficient signal with which to bind. Signal-free QscR is extremely unstable and aggregates or denatures quickly. Based on our findings, we propose a model whereby functional QscR molecules are synthesized in the absence of an acyl-HSL but they do not accumulate under those conditions because they are unstable and are eliminated from cells by aggregation or proteolysis. We do not believe that this model is specific to QscR. It likely applies to QscR homologs. We have a related poster showing that LasR can bind its cognate signal reversibly and it can fold into an active conformation in the absence of signal. These findings are relevant to efforts towards development of anti-QS inhibitors.

This work was supported by USPHS grant GM-59026. K.O. received a Postdoctoral Fellowship for Research Abroad from the Japan Society for the Promotion of Science. Biosynthesis of hypochlorous acid (HOCl), a potent anti-microbial oxidant, in phagosomes is one of the chief mechanisms employed by polymorphonuclear neutrophils (PMNs) to combat infections. This reaction, catalyzed by myeloperoxidase, requires chloride anion (Cl -) as a substrate. Thus, Clavailability is a rate-limiting factor that affects neutrophil microbicidal function. Our previous research demonstrated that defective CFTR, a cAMP-activated chloride channel, present in CF patients leads to defective chloride transport to neutrophil phagosomes and impaired bacterial killing (Painter et al., 2008 (Painter et al., & 2010 . In this report, we successfully achieved lentiviral vector-mediated expression of the small interference RNA against CFTR in HL60 cells, a human promyelocytic cell line. After DMSOinduced differentiation, CFTR expression in the neutrophil-like cells was abated by ~50-80% when compared to that of the control. The created CFTR deficiency in the HL60-derived neutrophils resulted in impaired bactericidal capability, thereby recapitulating the phenotype seen in CF patient cells. The results provide further evidence suggesting that CFTR plays an important role in phagocytic host defense. Pseudomonas aeruginosa (PA) lung infections are main causes of morbidity and mortality in CF. Repeated courses of antibiotics give fewer infections and improve prognosis, but the threat of serious side effects by the use of antibiotics is a reality -especially the rapidly increasing antibiotic resistance; but also the toxic effects on kidneys and hearing, the detrimental effect on the normal intestinal flora and the emergence of new Gram negative bacteria in the lungs. Alternatives and/or complements to antibiotics are urgently needed.

Passive immunization is a well-known alternative to fight infections. Hens produce large amounts of antibodies, IgY, which can be retrieved from the egg yolk. Hens that are vaccinated with PA produce specific Anti-Pseudomonas IgY (Anti-PA IgY).

Anti-PA IgY has now been in clinical use for more than 15 years as a treatment to CF-patients to avoid chronic PA infections. The intermittent results have earlier been reported to NACFC. These results show statistically significant reduction of PA positive cultures and a clear tendency to delay the onset of chronic PA infections. For these patients the need of antibiotics has been greatly reduced. Humans do not produce anti-IgY antibodies to orally given IgY. Thus, the risk that bacteria should develop resistance to IgY is extremely small. After clinical use with more than 50,000 doses there have been no adverse events. Gram negative bacteria have appeared only sporadically and no cultures have been positive for B. cepacia. BMI has been close to normal and pulmonary function has been preserved at the end of the study (Nilsson et al, Pediatric Pulmonology, 2008) .

Due to these results Anti-PA IgY has been granted orphan drug designation by the European Medical Agency (EMA) for treatment of cystic fibrosis. EMA's motivation: "…justifications have been provided that avian polyclonal IgY antibody against PA may be of significant benefit to those affected by the condition."

We have now designed a multicentre, randomized, double blind, placebo-controlled clinical phase III study to be performed in 7-10 European countries given that we get financial support from EC or others. This study will go on for about four years. Inclusion criteria are: Pancreatic insufficient CF-patients who have had one or several PA positive sputum cultures but still are not chronically infected. An ongoing PA infection has to be eradicated initially with a short course of antibiotics. The patients will gargle anti-PA IgY or placebo every evening after tooth brushing. The patients will be in the study until they have got a new PA infection or for up to two years. When they are in the study, they will receive all their usual treatment except antibiotics against PA. When they leave the study due to PA infection they will get an antibiotic course against PA according to guidelines. All visits for the study will be the same as for the usual follow up of these patients -i.e. once every third month. The study protocol is at present under scrutiny by EMA. At the time of the congress we will know if financial support has been received from EC.

Oinuma, K.; Greenberg, E.P. Microbiology, University of Washington, Seattle, WA, USA Pseudomonas aeruginosa, a major colonizer in the lungs of cystic fibrosis patients, uses quorum sensing (QS) to control virulence in a cell density-dependent manner. QS requires acyl-homoserine lactone (acyl-HSL) synthases and transcription factors that are activated by the binding of specific acyl-HSLs. In P. aeruginosa LasI and RhlI produce N-3-oxododecanoylhomoserine lactone (3OC12-HSL) and N-butanoyl-homoserine lactone (C4-HSL), respectively, and LasR and RhlR respond to these signals. There is a third acyl-HSL receptor, QscR, which has no cognate signal synthase but can respond to LasI-produced 3OC12-HSL and several other acyl-HSLs. Because QS is a potential target for anti-pseudomonal therapeutics there has Results: Among 47 patients followed in these CF centers, 47 symptomatic cases were reported. Mean age of the patients was 15 years (0-55 yrs). Chronic Pseudomonas aeruginosa sputum colonisation was present in 38%. Mean FEV1 (% predicted) was 82% (25-114%). Three patients had undergone lung transplantation. The most common symptoms were fever (100% of the patients), increased cough and sputum production (82%), asthenia (58%), and worsening of dyspnea (47%). A large proportion of the patients (90%) received oseltamivir and 70% antibiotics. Twelve patients (25%) required hospitalizion. The most common reason for admission was dyspnea and hypoxemia. Length of stay ranged from 2 days to 2 months. Three patients were admitted in to the intensive care unit. Two were put on lung transplant list because of severe respiratory failure, one of those being transplanted 1 month after influenza infection.

At follow-up, 4 patients demonstrated a colonization with P. aeruginosa. About 50% of the patients whose respiratory function tests were available, had a significant decrease in FEV1 in comparison to baseline (at 3 months : -12% (-9% to -29%) and at 6 months : -14% (-9% to -18%)).

Conclusions: This case series evidences the pronounced morbidity of H1N1/09 in worsening CF disease and predisposing to subsequent bacterial infections. As the disease progresses, the cystic fibrosis (CF) lung demonstrates marked anatomic heterogeneity characterized by bronchiectasis, airway mucus plugging, cyst formation, and atelectasis. An inability to clear a chronic microbial biofilm infection and the subsequent inflammatory response results in this airway remodeling. The objective of this study was to determine the effects of the spatial heterogeneity on the airway microbiota and to identify the extent to which microbial or viral infections are distributed throughout the CF lung.

Methods: Studies of CF microbes are generally conducted using sputum and/or bronchoalveolar lavage samples, which may not accurately represent the nature of infections in the peripheral parenchyma. Using a metagenomic approach, we directly analyzed the viral and microbial communities in individual lobes of the lungs from two CF patients: one explant from a lung transplant recipient and the other from an autopsy specimen from a deceased CF patient who had experienced a severe exacerbation. The major clinical isolate from both specimens was mucoid Pseudomonas aeruginosa. Samples were collected from the six major anatomical divisions of the lungs (right upper, middle, and lower; left upper, lower, and lingula), and viral and microbial DNA were extracted. The viral DNA and bacterial 16S rRNA PCR products were sequenced with a Roche/454 Life Sciences GS-FLX pyrosequencer using Titanium chemistry.

Results: The resulting metagenomes show a distinct pattern of spatial heterogeneity in the explant lung, while distribution in the autopsy lung is more uniform. In general, lobes that are physically closer share a greater number of microbiota than those that are more distant. A number of eukary-otic viral and phage taxa identified in the lung samples were confined to a single lobe. Several of these phage were types capable of modulating host metabolism, such as those that cause the development of small colony variants and seroconversion of P. aeruginosa. The majority of bacterial species identified were P. aeruginosa, but there were also significant percentages of Streptococcus and Staphylococcus species in the autopsy lung, suggesting a change in the bacterial community associated with exacerbation and mortality.

Conclusion: This study provides strong evidence for spatial heterogeneity of microbial and viral infections in the CF lung, and suggests that the degree of heterogeneity may depend on the progression of the disease. 

In the first few months after transplantation, primary infections are dominated by nosocomial organisms, largely Aspergillus and Candida spp. associated with a high mortality. Aerosolized liposomal CsA, even targeted delivered and locally high concentrated, may have the impact for both -the prevention of the development of BOS as well as the prevention of some opportunistic lung infections.

Enriched sheep blood agar without di-and monosaccharides was chosen to simulate the anastomoses site which are the primary site of clinically apparent infection. Each Petri dish thus simulates 0.254 m 2 of wound surface. The test strains represented yeasts (Candida albicans, ATCC 10231) and molds (Aspergillus niger, ATCC 16404, Aspergillus fumigatus ATCC 9197), Pseudomonas (Pseudomonas aeruginosa ATCC 27853) and Streptococcus (Streptococcus pneumoniae ATCC 33400). Agar plates were overlaid with L-CsA and microorganisms (100 µl of suspension with about 3.5 x10 3 CFU), incubated at 36°C +/-1°C and colony counts were observed and documented at day 1, 2, 3, 4 and 5.

Results: The L-CsA inhibited the harmful pathogen Aspergillus niger in this simulation and slowed the evident growth of A. fumigatus. The yeast Candida albicans showed less colony counts. Streptococcus pneumoniae was inhibited over time und L-CsA showed classical dose-response relationships.

Conclusion: Cyclosporine as a calcineurin-inhibitor has shown some activity towards opportunistic pathogens such as Aspergillus niger and fumigatus, Cryptococcus neoformans, and some Candida strains as well as in combination with azoles.

Locally high concentrations of liposomal CsA may exceed the MIC/MBC of various opportunistic microorganisms preventing colonization as well as infection of the lung of transplanted patients in combination with systemic antimicrobial drugs. Background: Sinonasal involvement in cystic fibrosis (CF) has been proposed as a gateway and reservoir for subsequent pulmonary infection with P. aeruginosa. The aim of the present study was to investigate the phenotypes and the gene expression profiles of clonally related mucoid (M) and non-mucoid (NM) P. aeruginosa isolates from sinuses and lungs in order to reveal adaptive mechanisms specific to the different compartments of the "united airways." otic demonstrated it was a polyketide with a structure that correlated to the mapped biosynthetic pathway. The potent activity of the B. ambifaria antibiotic on B. multivorans and other Burkholderia that cause cystic fibrosis infection suggest it may have potential as a future therapeutic agent. Design and use of the custom B. ambifaria AMMD microarray was supported by the Cystic Fibrosis Foundation Therapeutics Burkholderia microarray program (Grant MAHENT0V06). Achromobacter xylosoxidans, formerly known as Alcaligenes xylosoxidans, is an environmental non-lactose fermenting aerobic motile gram-negative rod characterized in 1971. It is most often found in relation to nosocomial infections targeting immunocompromised patients suffering in particularly from cancer, advanced HIV, diabetes mellitus or chronic renal failure. In 1985 the first description of A. xylosoxidans in relation to pulmonary infection in cystic fibrosis (CF) was published. A. xylosoxidans is increasingly found in CF clinics world wide contributing to the problematic treatment of these patients due to its extremely elevated resistance to a broad range of antibiotics. These circumstances are making it an emerging CF pathogen of great interest. Several investigations have documented chronic infections by A. xylosoxidans with a general incidence of approximately 6-10% of the CF population and there is an overall perception from studies that it can spread by cross-infection. The majority of the literature about A. xylosoxidans is case reports, whereas publications documenting important pathogenic properties and resistance mechanism are scarce.

The aim of this study was to elucidate the complete genomic sequence of A. xylosoxidans NH44784-1996 and to resolve possible functionality of important genes by the program Rapid Annotations using Subsystems Technology (RAST). The strain for sequencing was obtained from a CF patient at the Copenhagen CF centre. Furthermore, we have investigated the ability of this strain to form biofilm in vitro under different growth conditions and its susceptibility to a range of antibiotics (MIC).

The complete genome sequence was determined by pyrosequencing on the GS FLX Titanium platform and assembly using the newbler assembler software. The complete genome of A. xylosoxidans NH44784-1996 is 6.9 kb. The sequence revealed the presence of a range of antibiotic resistance genes encoding efflux pump systems and antibiotic modifying enzymes. Regarding virulence, there was no resemblance to known genes encoding virulence factors or type III secretion system used to secrete toxins from the bacteria into the cells of the infected organism. The clinical isolate of A. xylosoxidans was investigated for biofilm formation capacity under stationary conditions in a suspension and in a flow cell system with a continuous flow of media. Scanning electron microscopy and confocal laser scanning microscopy techniques showed that the investigated bacteria aggregate in biofilms.

The highly developed antibiotic resistance and the capability of biofilm formation make treatment of A. xylosoxidans infections extremely difficult. Furthermore, it underlines the importance of developing antibiotic treatment guidelines.

In total, 24 P. aeruginosa isolates representing M and MN pairs from sinus and lung from six chronically infected CF patients were included in the study. The isolates from the sinuses were obtained by Functional Endoscopic Sinus Surgery (FESS) and the lung isolates were obtained from sputum samples. All isolates were typed by pulsed-field-gel electrophoresis. The isolates were characterized phenotypically (protease, motility, antibiotic resistance and hypermutability), genotypically (mucA, algT, lasR and rhlR sequence variability) and the gene transcription profiles were investigated by microarray analysis (Affymetrix).

Results: The isolates from the upper and lower airways of each patient were clonally related and the six patients were infected by six different clones. The same mucA and lasR mutations were identified in the sino-pulmonary pairs of M or NM isolates in five of the six patients. No significant differences were found between the phenotypes of the sino-pulmonary pairs of M or NM isolates. The number of differentially expressed genes (more than four fold changes, p<0.05) between the upper and lower airways varied from 0 to 49 for the M isolates and from 4 to 58 for the NM isolates and included pyochelin operon; genes involved in response to oxidative stress; and genes responsible for the metabolism of amino acids, inorganic sulfur and carbon; suggesting differences in the inflammation and growth conditions between the two locations.

Conclusions: These data suggest similar evolution of P. aeruginosa in the sinuses and the lungs, which could be explained by a bi-directional communication between these two locations. A difference in the inflammatory response between the infection in the sinuses and the lungs is suggested by transcriptome data. Burkholderia bacteria are known to produce many antibiotics of which those with antifungal activity have been well characterised. Pyrrolnitrin, xylocandin, the pseudanes, 4-quinolones, cepalycin, cepaciamide, maculosin, banegasin, and non-ribosomal peptide antibiotics are among the many antifungal Burkholderia compounds that have been identified. Few Burkholderia antibiotics with anti-bacterial activity have been characterised with the cepacins and the recently characterized polyketide, bactobolin, being examples of those with anti-Gram positive activity. The capistruin peptide of Burkholderia thailandensis is also one of the few antibiotics secreted that is active on another Burkholderia species, Burkholderia caledonica; such inhibition between species from the Burkholderia cepacia complex (Bcc) has not been observed. To survey antimicrobial secondary metabolite secretion by species within the Burkholderia cepacia complex, a collection of 267 strains representative of the 13 current species and several novel species groups was screened. Overlay assays with Staphylococcus aureus, Bulkholderia multivorans, Pseudomonas aeruginosa and Candida albicans were performed to detect anti-Gram positive, anti-Gram negative, and antifungal activities. lata were all resistant to the B. ambifaria antimicrobial. Antibiotic production was genetically localized using transposon mutagenesis to high GC genomic island encoding a novel cluster of polyketide biosynthesis genes within the genome of B. ambifaria strain AMMD. Mutations in quorum sensing genes, the general protein section pathway, exopolysaccharide biosynthesis, and novel hypothetical genes were also found to alter or prevent antibiotic production. The global gene expression linked to the production of the antibiotic was determined using a custom genomic microarray designed B. ambifaria strain AMMD and demonstrated that multiple genes within the polyketide biosynthesis island were upregulated during antibiotic production. Purification and structure analysis of the B. ambifaria antibi- 

For a successful infection to occur, infecting bacteria must tolerate conditions such as low levels of iron, oxidative stress, macerating enzymes, phagocytic cells and other defence mechanisms and administered antimicrobials. Much evidence has accumulated over the years showing that surface attached biofilms to a great extent tolerate the above antimicrobial properties met during infections.

However, images from sites of infections have shown that bacteria do not need to be attached to surfaces in order to establish chronic infections. Instead, the bacteria generate non-attached microcolonies by aggregating to their fellow bacteria through matrix components, and they appear to establish an impenetrable barrier to the host, for example, phagocytic cells.

In this study we found that non-attached microcolonies from cultures of Pseudomonas aeruginosa resemble the phenotype of biofilms both with regards to internal structures, matrix components, tolerance to antibiotics and phagocytes.

We employed scanning electron microscopy, confocal microscopy, live dead staining as well as traditional culturing techniques to characterize these non-attached microcolonies.

We found that the non-attached microcolonies were equally as tolerant to various antimicrobials as the attached biofilms. To test whether the tolerance is reversible we treated the aggregates with antibiotics for 24 hours and then disrupted the microcolonies (harbouring tolerant bacteria) mechanically before another 24 h treatment of the bacteria. Interestingly all bacteria were killed by the second round of antibiotic, indicating that the tolerance towards antibiotics is reversible and thus likely to be caused by the physiological states of the aggregates rather than mutations. These findings should be useful for the development of novel treatment regimens.

Cigana, C. 1 ; Lorè, N. 1 ; Schwager, S. 2 ; Juhas, M. 2 ; Eberl, L. 2 ; Bragonzi, A. 1 1. Infections and Cystic Fibrosis Unit, San Raffaele Scientific Institute, Milan, Italy; 2. Department of Microbiology, University of Zurich, Zurich, Switzerland Background: CF lung disease is characterized by transient airway P. aeruginosa infections and excessive neutrophil-dominated inflammation early in life followed by permanent chronic infection that causes persistent respiratory symptoms and a decline in lung functions. The pathogen itself may actively cause a great deal of damage, rather than damage caused by the host response. In an earlier study, we showed that P. aeruginosa is subjected to PAMPs modification as a strategy to hijack genes involved in innate immune responses. Comparing lipid A and peptidoglycan of sequential strains isolated from CF patients during a period of up to 7.5 years, we established for the first time that P. aeruginosa has evolved the capacity to evade immune system detection (Cigana et al, PlosOne 2009; Bragonzi et al, AJRCCM 2009 ). However, the "pathological drift" by P. aeruginosa adapted strains as a mechanism to sustain chronic infections remains to be established.

Methods: We analysed the host response to early and late clonal strains of P. aeruginosa in a multihost pathogenesis system including four different models, namely, Caenorhabditis elegans, Galleria melonella, Drosophila melanogaster and mouse. In addition, the epithelial bronchial cells of CF origin IB3-1, their wt-like isogenic cells C38 and macrophage-like cells THP-1 were used to sustain the ability to provoke inflammation or damage during early or late phases of P. aeruginosa infection.

Results: P. aeruginosa strains at the onset of infection are more pathogenic than late isolates from the same patient when tested in C. elegans, G. melonella and D. melanogaster. In murine model of acute infection, the early P. aeruginosa strain induced higher mortality than late clonal strains. Although attenuated in mortality, P. aeruginosa late isolates retained their capacity to persist in mouse models of chronic infection. H&E, PAS-staining and Tunnel assay of lung tissue sections showed that early strains induced pronounced leukocyte recruitment indicating strong inflammatory response while late strains increased numbers of mucin-positive goblet cells and apoptotic cells, a typical hallmark of damage in the airway chronic diseases.

To establish the "pathological drift," IB3-1 cells were infected with early and late P. aeruginosa clonal strains from CF patients and subjected to microarray analysis to evaluate inflammatory cytokines, tissue remodelling factors, leukocyte receptors and adhesion molecules, and PPRs. Preliminary results indicated a decreased inflammatory response, including down-regulation of leukocyte receptors and adhesion molecules, and an increased damage mediated by tissue remodelling, due to late strains compared to early strains.

Conclusions: Our findings suggest that during long-term infection P. aeruginosa revises its interaction with CF host by activating alternative pathways including evasion of the immune response, non-inflammatory cell death and those relevant for tissue damage and remodelling process to ultimately result in chronic disease and decline in lung functions.

Supported by the European Union 7th Framework Programme (project NABATIVI). Aim: MRSA strains isolated from a large population of Italian CF patients are collected and their genetic backgrounds are analyzed in order to assess whether they represent HA-or CA-MRSA. The antibiotic profiles of MRSA strains representing CA-or HA-MRSA were also compared.

Methods: During a multicenter study, 77 MRSA strains (one/patient) were collected from CF patients attending eight CF Italian Centers over a period of two years (2008) (2009) . All collected bacterial isolates were identified as MRSA isolates with commercial methods and SCCmec typing was performed as previously described. The antibiotic susceptibility profiles of MRSA strains were tested by disk diffusion test. For isolates that tested resistant to erythromycin but susceptible to clindamycin by single-agent testing, a D-zone test to detect inducible clindamycin resistance was performed.

Results: Forty-eight (62%) out of 77 MRSA isolates belonged to HA-MRSA (SCCmec I,II, III), the remaining 29 (38%) strains were CA-MRSA (SCCmec IV). Antibiotic susceptibility was evaluated by assessing the activity of a large panel of drugs including tobramycin, gentamicin, ciprofloxacin, levofloxacin, rifampin, minocycline, doxycycline, erythromycin, clindamycin, trimethoprim/sulfamethoxazole, linezolid, vancomycin, teicoplanin. The resistance rates of the CA-MRSA and HA-MRSA strains are compared: no resistance against glycopeptides (vancomycin and teicoplanin) and oxazolidinones (linezolid) was found, the most active antibiotics among the other classes were tetracyclines (minocycline and doxycycline), trimethoprim/sulfamethoxazole, and rifampin. CA-MRSA, as previously demonstrated, were more susceptible to gentamicin, doxycycline, and trimethoprim/sulfamethoxazole. CA-MRSA usually have good susceptibility to clindamycin, however clindamycin resistance, including inducible resistance, was high in CA-MRSA isolated from CF Italian

The lungs of chronically infected cystic fibrosis (CF) patients have been intensively studied. Pseudomonas aeruginosa is reported to be the predominant pathogen in CF (Baurnfeind et al. 1987 , Koch 2002 , however lately a debate of the true bacterial load has occurred. The diagnostics for CF rely mainly on culture based techniques performed on expectorated sputum samples, and most studies are centered on this sample type. Using molecular methods researchers have detected many more bacteria, especially many patients and was similar for CA-MRSA and HA-MRSA strains (76% and 85% respectively). HA-MRSA demonstrated a higher resistance for gentamicin, doxycycline and trimethoprim/sulfamethoxazole.

The results of SCCmec typing confirmed that, in spite of the high risk of HA-MRSA acquisition for frequently hospitalized CF patients, a high prevalence of SCCmec IV, associated with CA-MRSA (38%) has been found. The diffusion in the CF population of highly resistant and potentially pathogenic CA-MRSA strains could have a severe impact on the clinical status of CF patients, and therefore careful surveillance and close monitoring of MRSA infections is essential. Background: Cystic fibrosis (CF) patients with initial Pseudomonas aeruginosa (Pa) infection are usually treated with an early eradication therapy protocol. Several serological assays determining anti-Pa antibodies have been used to monitor the course of infection. In addition to respiratory cultures, increased antibody levels may suggest lack of response to treatment with evolution to chronic infection. However the clinical relevance of the immunological response against specific antigens is not fully understood.

Aim: To evaluate longitudinally the anti-Pa immune response using two different methods in patients undergoing early eradication treatment.

Methods: Patients in regular follow-up under early eradication treatment in two Italian CF centers were evaluated immunologically. Patients Pafree (never colonized or Pa-free with at least three negative cultures in the last 6 months) and older than 1 year participated in this study. Treatment was considered efficacious if patients were infection-free with at least three cultures negative for Pa over 6 months.

Sera of 81 patients were collected. Mean antibody titers, assessed 8 months (median) range 1-12 months before detection of Pa in respiratory samples (T-1), were compared with mean antibody values at the time of the Pa isolation (T0) and after antibiotic therapy (T1).

We used two ELISA methods to test the presence of IgG against Pa lipopolysaccharide (LPS) and against Pa sonicated whole cells (CF-IgG) respectively using the manufacturer's cut-off (6 U/ml for anti-LPS-IgG and 2.96 ELISA units for CF-IgG).

Results: There were 22% of patients with IgG levels against Pa LPS and 27% against CF-IgG over the cut-off at T-1 and 29% and 30% at T1, respectively. At the time of Pa isolation, the mean LPS antibody was significantly different in patients who had not undergone previous eradicating therapy (3.66 ± 2.4 U/ml) compared to those who had (8.48 ± 6.85 U/ml) (p=0.002). The mean antibody titer against Pa sonicated whole cells (CF-IgG) was also significantly different in the two groups of patients (1.18 ± 1.5 vs 2.15 ± 1.39) (p=0.001). At T0 the mean anti-LPS antibody titer (5.32 ±3.86) showed a significant increase compared to T-1 (3.17 ± 2.63) (p=0.01). This increase was not evidenced with the other method used.

There were no significant differences in antibody titers in patients treated with success and those who continued to colonize Pa according to both methods used at T1 (mean time of follow-up 6 months).

Conclusions: A significant percentage of patients present anti-Pa antibodies beyond the cut-off of both methods at T-1. Pa isolation is accompanied by an increase in mean antibody titer when using the anti-Pa LPS IgG method. We found that 6-month follow-up is not sufficient to reach a statistically significant difference in antibody titers between those patients successfully treated and those who are not.

anaerobes in sputum and BAL samples. It is however problematic since it is possible that the samples may have been contaminated by oral flora during expectoration and BAL sampling. We have in a recent publication only detected P. aeruginosa by PNA FISH (Bjarnsholt 2009) in tissue samples of explanted CF lungs. To study the true bacterial load in CF patients at the Copenhagen CF Center we investigated direct sampled material from explanted lungs (n=4) of CF patients by both culturing and PCR.

Methods: The samples (3-5 from each lung) were examined by standard culturing techniques including expectorated sputum pre-explantation, including aerobic and anaerobic growth. These findings were compared to results obtained by 16S rRNA gene analysis (16S rRNA gene amplification, cloning, sequencing and phylogenetic analysis) performed blinded of the growth results.

Results: P. aeruginosa were the only bacteria cultured from 3 of the lungs and Achromobacter xylosoxidans were the only cultured from the 4th lung. The results from the expectorated sputum were identical. The 16S rRNA gene analysis of the samples corresponded largely to these findings, but also revealed a minor fraction (1-8%) of other bacterial species, with a 3% maximum of a single species.

Conclusion: The bacteria identified by standard culturing are the predominant bacteria, which is also our clinical experience. These clearly reveal that the bacteriology of CF patients can be performed on expectorated sputum, and that we do not overlook any important bacteria doing this. This study included only material from 4 patients, we are currently including more lungs to increase the impact of our findings. Torresani, E. 1 ; Colombo, C. 1 ; Taccetti, G. 2 1. CF Centre and Microbiology Lab, Fond. IRCCS Ca' Granda, Osp. M. Policlinico, , Milan, Italy; 2. CF Center, Department of Pediatrics, Meyer Hospital, Florence, Italy Rep-PCR is a semi-automated repetitive DNA sequence-based polymerase chain reaction, to assess strain relatedness.

Aim: To follow CF patients (pts) at first isolation of Pseudomonas aeruginosa (PA), undergoing eradication therapy, and evaluate the genotype correlation with a subsequent isolate of PA during a followed period of 6 months.

Method: Starting from January 2008, 75 pts attending the CF centres of Milan and Florence were eligible for enrolment at the time of their first lung infection by PA (T0). These patients were treated for 28 days with two different schedules (oral ciprofloxacin plus inhaled tobramycin or inhaled colistin) and assessed by microbiological evaluations during a 6 month follow up time. All strains isolated from sputum culture at T0 and isolates acquired during the 6 months following antibiotic therapy (T1) were stored at -80°C. Treatment efficacy was determined on the basis of the comparison of rep-PCR genotypes of the PA isolates between T0 and T1 using DiversiLab® system (BioMerieux S.p.A.).

Results: Seventy-five pts who were PA positive underwent eradication therapy and 29/75 were shown to be re-infected by PA during the follow up. Rep-PCR was performed on 19 of them. The results are summarized in the Table. Thirteen pts carried the same PA genotype (A/A with a similarity 94-99%), 4 pts presented a different PA genotype (A/B), while 2 pts showed a simultaneous presence of more than one genotype at T1.

Conclusions: In spite of the quite small study population, it is suggested that, when a first PA colonization episode is concerned, the presence of a genotypically identical PA in a subsequence isolate may be an additional tool to predict impending chronic colonization. Furthermore the ability of rep-PCR to investigate infra-specific genetic relationship, indicates that it can be a valuable tool in monitoring the efficacy of eradication therapy. Background: Streptococcus pneumoniae is the most common cause of severe infections, many of which may have strong epidemiological impacts, such as meningitis, otitis media, pneumonia and sinusitis. In patients (pts) with cystic fibrosis (CF), S. pneumoniae has been reported as the fourth most common bacterial organism isolated from sputa, after Pseudomonas aeruginosa, Staphylococcus aureus and Haemophilus influenzae. Its prevalence may be underestimated, as S. pneumoniae is difficult to isolate from sputa often heavily colonized with the other organisms. The pathogenic role of S. pneumoniae in CF is controversial and only few reports document its potential involvement in pulmonary damage. The isolation of S. pneumoniae was associated with episodes of acute exacerbation.

Aim: The aim of this study is the first to analyze prevalence of serotypes and patterns the distribution of antibiotic susceptibility of S. pneumoniae.

Materials and Methods: From June 2009 to April 2010 a total of 86 strains of S. pneumoniae, isolated from lower airway of 674 CF pts, were studied. The collected respiratory samples were cultured following standard laboratory procedures. Isolates of S. pneumoniae were identified by bile solubility and optochin susceptibility. MICs were determined by E test Method (AB BIODISCK Sweden). The results were interpreted according to the CLSI criteria. ATCC 49619 was used for Quality Control. The serotyping was performed with the Quellung reaction using antisera from the Statens Serum Institut, Copenhagen, Denmark.

Results: The yearly incidence of pts colonized with S. pneumoniae was 86/674 (12.8%). Microbiological data showed that S. pneumoniae isolates in pts were recovered without the concomitant presence of other CF pathogens in 19 cases (19/86, 22.1%). S. pneumoniae was recovered with other pathogens in 67 cases (67/86, 77.9%). Among 86 strains of S. pneumoniae, 9 (10.5 %) of the isolates were penicillin-resistant and 6 (7.0%) penicillinintermediate. All isolates were susceptible to levofloxacin. The prevalence of multidrug resistant pneumococci, defined by resistance to ≥ 3 drug classes, represented by penicillin and erythromycin, and/or tetracycline and/or SXT, was 9.3%. The serotypes involved were 18, the prevalent serogroups were: 3 (17.4%), 19F (11.6%), 19A (8.1%). The multiresistant strains of S. pneumoniae were included in the serogroups 14 (37.5%), 19F (25%), 23F (25%) and 19A (12.5%).

Conclusions: The knowledge of serotype distribution in S. pneumoniae isolates may have epidemiological and clinical implications. It has been recommended by American Academy of Pediatrics, that all CF pts older than 2 years should receive the 23-valent polysaccharide vaccine against S. pneumoniae. Recent data from the Pneumococcal Reference Laboratory in Lombardy (Italy) show that the most prevalent serotypes are 14 (11.7%), 3 (11.3%), 19A (9.9%), instead of our CF series, 3 (17.4%), 19F (11.6%), 19A (8.1%). The serotypes 3 and 19A are not included in 7-valent pneumococcal conjugate vaccine. The rate of penicillin resistance is higher in CF pts than in non-CF pts. the majority of CF patients are colonized by P. aeruginosa, Cif represents a potential roadblock for the effectiveness of CFTR targeted therapies in a clinical setting. Additionally, elucidation of the Cif mechanism may illuminate novel pathways regulating CFTR trafficking.

We have determined the structure of Cif and shown it to be an epoxide hydrolase (EH) with unique substrate selectivity (3) . Mammalian EHs function as regulators of blood pressure and inflammation, while bacterial EHs are thought to detoxify xenobiotic compounds. Cif is the first example of an EH serving as a virulence factor. Based on structural comparison, it appears that the enzyme utilizes a catalytic triad of residues Asp129, Glu153 and His297, with accessory residues His177 and Tyr239 coordinating the epoxide oxygen during ring opening. His207 sits along the substrate access tunnel to the active site (see Figure) . We have generated mutations in the catalytic triad (E153Q), the ring coordinating residues (H177A), and the substrate tunnel (H207A). These mutations all prevent EH activity and also eliminate the CFTR inhibitory effect seen on airway epithelial cells. Thus, it appears that EH activity is strictly required for Cif mediated mislocalization of CFTR, suggesting that Cif may act on a physiological epoxide with a role in regulating CFTR trafficking. Mutation of the Glu153 side chain in the catalytic triad prevents Cif from releasing products of epoxide hydrolysis, but does not obstruct the formation of a covalent enzyme-substrate intermediate complex. This supports our mechanistic hypothesis and provides a basis for substrate-trapping experiments designed to identify candidate physiological targets. This work was supported by grants from the CFF (MADDEN08G0) and the NIH (P20-RR018787 and T32AI007519). Viral respiratory infections are a major source of morbidity in patients with CF, increasing lung inflammation, reducing Cltransport, and contributing to acute and chronic pulmonary decline. Respiratory syncytial virus (RSV) is the most common cause of bronchiolitis in young children, with higher incidences of disease complications in those with CF. Previously, we have reported that RSV is a potent stimulus of pro-MMP-9 expression and release in human airway cells. MMP-9 activity demonstrates a negative correlation with lung function in CF patients [1] , and is a critical enzyme in the production of chemotactic proline-glycine-proline (PGP) trimers from types I and II collagen [2] . PGP has been shown to bind CXCRI receptors on neutrophils, recruiting them into the pulmonary compartment. MMP-9 activity and PGP are extremely high in CF lower airway secretions and may represent an important component of excess levels of neutrophils and activity in the CF airway [3] . In the current study, we examined the effect of RSV infection on the MMP-9 axis in polarized human airway cell monolayers (16HBE cells and CFBE410-cells stably transduced with wtCFTR and F508del CFTR). Measurements were taken of RSV replication and MMP-9 and TIMP-1 expression and release. RSV infection at an MOI of 0.1 -0.2 was sufficient to stimulate pro-MMP-9 expression and release from the apical and basolateral membranes, with maximal expression at days six to nine, post-infection. Control virus experiments with heat-killed RSV, and replication competent and incompetent adenovirus (serotype 5) failed to illicit MMP-9 expression and release. TIMP-1 expression was low or undetectable following RSV infection, indicating that MMP-9 was produced and released independently of its natural inhibitor. Inhibition of MMP-9 expression with transfection of siRNA directed against MMP-9 (Assays on Demand) and inhibition of activity through addition of a small molecule inhibitor of MMP-9 (doxycycline, 100 µg/mL) abrogated RSV replication in CFBE41o-wtCFTR+ cells, also reducing syncytia formation over the time course of RSV infection (p<0.001). Our results confirm that RSV is a potent stimulus of MMP-9 expression and release in human airway cell monolayers and identifies a critical role for MMP-9 in RSV replication. Use of MMP-9 inhibitory molecules may represent a novel means to reduce RSV replication and morbidity in CF patients infected with RSV.

Supported by KPRI and CFFT. 

Bahl, C. 1 ; Gerber, S.A. 2 ; Madden, D.R. 1 1. Dept. of Biochemistry, Dartmouth Medical School, Hanover, NH, USA; 2. Dept. of Genetics, Dartmouth Medical School, Hanover, NH, USA Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen that causes chronic infection in the lungs of CF patients. P. aeruginosa has been shown to secrete a soluble protein, termed Cif (CFTR inhibitory factor), which inhibits the endocytic recycling of CFTR. The result of Cif treatment on airway epithelial cells is a loss of CFTR mediated chloride efflux across the apical membranes. This effect is due to mislocalization of the transporter to endosomes, and by preventing trafficking back to the plasma membrane, CFTR is shunted to the lysosomes for degradation (1, 2) . Since isolate. All but 1 patient was chronically infected with Pseudomonas aeruginosa.

In three patients, the first isolate was in the last 12 m, and it is too early to report on eradication. Of the remaining 13, eradication has been achieved in 5 (38%): 4 cases of B. multivorans and 1 of B. vietnamiensis. Although antibiotics for BCC eradication were prescribed in 9 cases, and included a range of 2-4 antibiotics for a median of 28 d, only three of these patients successfully cleared the infection, whilst 2 cleared the infection without receiving specific therapy. Of the successful eradications, 3 isolates were only on a single occasion and 2 persisted for 1-2 m.

Of the chronically infected patients, the median maximum time interval between BCC isolates in sputum was 10 m (max 13 m), with up to 9 negative samples between isolates. Only one patient had over 12 m of negative cultures between isolates, but 4/8 had intervals of over 6 m between isolates.

All except the B. latens case were infections due to individual sporadic strains.

There have been no cases of cepacia syndrome during this period, and no patients with chronic BCC have died. A single patient has been transplanted.

Conclusions:

1. The success of patient segregation is borne out by the low rates of new B. cenocepacia infection.

2. The majority of new BCC infections are B. multivorans, a sporadic infection acquired from the environment, which can be cleared either spontaneously or with treatment in around one third to one half of cases.

3. Time interval between positive isolates can be prolonged, and emphasises the need for continued surveillance and segregation for at least 12 months after last isolate.

Nguyen, D. 1 ; Joshi Datar, A. 2 ; Buerle, E. 2 ; Singh, P.K. 2 1. McGill U., Montreal, QC, Canada; 2. U. Washington, Seattle, WA, USA Existing antimicrobials fail to eradicate chronic P. aeruginosa (Pa) infections, once Pa biofilms are established in the CF airways. Biofilm bacteria are remarkably tolerant to antimicrobials and host defenses and this tolerance is a major factor causing persistent infection. Several mechanisms likely contribute to the antimicrobial tolerance of biofilms including the extracellullar matrix and induction of efflux pumps, but its mechanism remains elusive. Further understanding of tolerance may uncover new targets that could sensitize biofilms against antimicrobials.

Biofilm bacteria are subject to gradients that limit nutrient supply, and nutrient-limited bacteria are notoriously difficult to kill. Thus, the starvation of subpopulations is thought to be a principal cause of antimicrobial tolerance in biofilms. Here we hypothesized that antimicrobial tolerance is mediated by the active induction of starvation responses in biofilms, and that interfering with these responses could reverse tolerance. We tested this by manipulating the stringent response (SR). This global regulatory mechanism is induced when bacteria sense stress and starvation, and the relA and spoT genes synthesize (p)ppGpp. This alarmone regulates many functions needed in starvation and stress conditions. We compared the antimicrobial tolerance of the relA spoT mutant (unable to induce the SR) to the wild type strain. In log-phase planktonic cells, serine starvation caused growth arrest in both strains but only conferred ofloxacin tolerance to the wild type. This indicates that starvation alone (without induction of the SR) was not sufficient to confer tolerance. Inactivation of the SR also increased the killing of biofilm bacteria by a broad range of antimicrobials (ofloxacin, meropenem, colistin) by 1000 to 10000 fold, and oxidants by 100 to 10000 fold.

We have made progress on understanding the mechanism by which SR inactivation reduces tolerance. Recent data suggest that antimicrobials kill bacteria through generation of reactive oxygen species (ROS). Inactivation of the SR caused increased levels of ROS. Increased ROS and impaired antimicrobial tolerance were due to increased HAQ molecules (4-hydroxy-2-alkylquinolines), as inactivation of HAQ biosynthesis (via pqsA mutation) restored tolerance and ROS levels in relA spoT mutant biofilms to wild type levels.

We tested the in-vivo relevance of our findings using animal infection models. Mice infected with intra-peritoneal stationary phase wildtype bac-teria died despite ofloxacin treatment, but were rescued when infected with the relA spoT mutant. Ofloxacin also had a 60 fold greater killing of the relA spoT mutant biofilm bacteria compared to wildtype in a subcutaneous biofilm implant.

These data show that the stringent response mediates the antimicrobial tolerance of Pa in conditions in which bacterial starvation occurs. Inactivation of the stringent response greatly enhanced antibiotic activity against both nutrient limited planktonic bacteria and biofilm bacteria, and in-vivo infection models. These results raise the possibility that antimicrobial tolerance in biofilms and other nutrient limited conditions may depend upon orderly and regulated response to starvation that could be targeted using new therapeutic approaches. Infection with the Gram-negative rod Pseudomonas aeruginosa causes considerable morbidity in patients with cystic fibrosis and other causes of bronchiectasis. P. aeruginosa often forms airway biofilms in such patients by production of a battery of exoproducts and virulence factors, in a cell population density dependent fashion, a process called quorum sensing. In Pseudomonas, quorum sensing is mediated in part by the transcription factor LasR. This protein has previously been shown to have a very high affinity for its cognate signal, N-3-(oxododecanoyl)-L-homoserine lactone (3OC12-HSL). Prior studies of LasR-signal interactions, most of which were conducted in vitro, demonstrated functionally irreversible binding of LasR to 3OC12-HSL. In addition they suggested that 3OC12-HSL serves as a scaffold for LasR folding and as a consequence the absence of signal results in a misfolded, inactive protein. These results were consistent with prevailing models for members of the LasR family from other bacteria. Because of the extremely high affinity of LasR for 3OC12-HSL, it has been regarded as a difficult target for drug inhibition, although targeting quorum sensing may result in significant benefits to patients with pseudomonal colonization.

In our studies, we demonstrate that removal of signal from cells quenches LasR activity, consistent with reversible binding of signal to LasR in vivo. This occurred in either recombinant E. coli or P. aeruginosa. We subsequently showed that the function of these LasR molecules can be rescued by re-addition of 3OC12-HSL, suggesting that this transcription factor retains an active conformation even in the absence of bound signal. We show that LasR is functional when synthesized in the absence of signal and, finally, that it can bind to a specific DNA target in an electrophoretic mobility shift assay when 3OC12-HSL is subsequently added. Together, these data demonstrate that LasR can be produced intracellularly in a properly folded activatable form, even if signal is not present, and also that the removal of cognate signal from the milieu results in cessation of LasR activity. These findings change our general view of the transcription factors that serve as signal receptors in acyl-HSL-mediated quorum sensing. They further suggest that the interaction of LasR and 3OC12-HSL may be more amenable to pharmacological disruption than previously believed. This work was supported by USPHS grant GM-59026. KO received a Postdoctoral Fellowship for Research Abroad from the Japan Society for the Promotion of Science.

Background: Inhaled hypertonic saline decreases pulmonary exacerbations in patients with CF in part by reversing dehydration at the epithelial cell surface and improving mucociliary clearance. The effect on P. aeruginosa of short term exposure to high salt as might be expected from daily hypertonic saline treatments has not been studied. We hypothesize that intermittent exposure to high salt will impair P. aeruginosa surface colonization.

MIC against 8 other bacteria associated with lung infections. Pharmacokinetics and acute inhalation toxicology data of Panaecin will also be presented.

Methods: MIC testing of Panaecin against bacterial clinical isolates was done under standard CLSI methodology by IHMA testing lab. Acute inhalation exposures were performed in mice using a whole-body exposure system. Response markers evaluated include: gallium concentrations in the lungs and blood, cellularity and cytokine levels in BAL, hematology and blood chemistry parameters, lung histopathology, and pulmonary mechanics.

Results: The antimicrobial activity of Panaecin against 102 Pa clinical isolates showed that the MICs for Panaecin were lower than either tobramycin or aztreonam with >80 strains having an MIC ≤1 µg/ml compared to 2 or 20 for tobramycin and aztreonam, respectively. Broad spectrum antimicrobial activity was also demonstrated against both Gram-negative and Gram-positive bacteria. Pharmacokinetic analysis of mice exposed to 4 different doses of gallium nitrate exhibited a half-life in the lung of 20-35 hr, which is in agreement with previous studies in rats and humans. The amount of gallium deposited in mouse lungs at the lowest inhalation dose (resulting in ~7 µg Ga/gram lung tissue) was ~100 fold higher than the MIC for Pa, and based on the PK analysis will remain ~25 fold above the MIC for over 48 hrs. These acute exposure studies also indicated that Panaecin was well tolerated at doses approximately greater than 2 orders of magnitude above the MIC for Pa. At such dose, there was no evidence of inflammatory response, and no increase in inflammatory cells in BAL fluid, or histopathologic changes noted.

Conclusion: Panaecin is effective in killing recent and antibiotic-resistant P. aeruginosa clinical isolates. The bactericidal potency of Panaecin on Pa was approximately 10-fold higher than current standard of care drugs tobramycin and aztreonam. The antimicrobial activity of gallium extends to other Gram-negative as well as Gram-positive bacteria. Animal studies demonstrated that inhaled gallium can be deposited into the lung at ~100 fold higher doses than the MIC for P. aeruginosa and can be maintained at levels at least 25 fold above the MIC for over 48 hrs. Acute inhalation of Panaecin at such dose failed to produce toxicity in mice. Aims of the study are to analyze the correlation between chronic lung infection by S. aureus (MSSA and MRSA) and pneumothorax (PNX), and the PNX impact on clinical aspects and pulmonary function in CF pts.

Methods: We studied 8 CF pts (4 females, 4 males) with a history of PNX (mean age 25.8 y). We selected 8 CF pts matched for age, sex, BMI and clinical conditions and no history of PNX as control group. We analyzed frequency of exacerbations, respiratory function, BMI, SatO 2 and sputum cultures (S. aureus, MRSA, P. aeruginosa, mucoid P. aeruginosa, B. cepacia, S. maltophilia) before and after PNX in the studied population and in the controls.

Results: The comparison of clinical data collected before and after PNX has highlighted ( Objectives: Determine the effect of short-term hypertonic saline treatment on in vitro P. aeruginosa surface colonization and virulence factor production.

Methods: The laboratory strain PAO1 was grown at 37 degrees C without shaking in plastic wells with M9 minimal medium supplemented with sodium glutamate. The bacteria were exposed to continuous 1% NaCl, continuous 6% NaCl, or 3 hr of 1% NaCl followed by 3 hr of 6% NaCl, and then a return to 1% NaCl for overnight incubation. Swimming motility, bacterial attachment, and microcolony formation were visualized by microscopy. Bacterial biofilms were stained with crystal violet, the crystal violet dissolved in ethanol and the A590 measured. The bacterial supernatant was harvested and filtered for rhamnolipid-dependent surfactant activity and pyocyanin production. Drop collapse assays of bacterial supernatant diluted with water determined the presence of rhamnolipids. Pyocyanin was measured after chloroform extraction and acidification of the supernatant by recording the A512.

Results: A single 3 hr exposure to 6% hypertonic saline did not reduce growth as determined by the OD 600 of the supernatant but was sufficient to alter attachment and reduce biofilm formation and the production of associated virulence factors compared with continuous growth in 1% saline (data not shown and Figure 1) .

Conclusions: Short term exposure to hypertonic saline alters biofilm formation and virulence factor production by P. aeruginosa. Current work is underway to examine the expression of genes required for exopolysaccharide synthesis and the production of biofilm-associated virulence factors.

Supported by K08 AI071074 from NIAID and an award from the Yale Child Health Research Center. Figure 1 . Short term exposure to high salt alters P. aeruginosa PAO1 in vitro virulence properties. A 3 hr exposure to 6% NaCl results in microcolony formation shown by the arrows (A). Either continuous exposure to 6% saline or a 3 hr exposure to 6% saline resulted in decreased drop collapse activity (B), biofilm formation (C), pyocyanin production (D). Background: Panaecin TM is a new anti-infective with demonstrated efficacy against Gram-negative and Gram-positive bacteria grown under planktonic or biofilm conditions. Gallium competes for iron binding and exploits the bacterial dependency on iron (Fe) for growth and pathogenicity. An inhalable nebulized liquid formulation for delivery using a commercially available vibrating mesh nebulizer has been developed. We report the bactericidal activities of Panaecin against >100 P. aeruginosa (Pa) strains from CF patients as compared to tobramycin and aztreonam, as well as the Microbiological analysis of sputum isolates, showed the following incidences of microbiological isolations stable over the years: PNX+: S. aureus 100%, MRSA 62.5%, PA 75.0%, PAm 50.0%, B. cepacia 12.5%, S. maltophilia 25.0% PNX-: S. aureus 75.0%, MRSA 12.5%, PA 50.0%, PAm 37.5%, B. cepacia 12.5%, S. maltophilia 12. 5% We also analyzed data on the frequency of co-infections: -S. aureus + P. aeruginosa: 75% in PNX pts; 37.5% in No Pnx pts -MRSA + P. aeruginosa: 50% in PNX pts; 12.5% in No PNX pts. Conclusions: Prevalence of chronic infection by S. aureus in the PNX group was 100%, with a MRSA prevalence of 75%. Prevalence of chronic S. aureus in the control population was 75%, with a MRSA colonization in only 12.5% pts. Moreover, co-infection with S. aureus and P. aeruginosa was more frequent in pts with a history of PNX (75%) compared to controls (37.5%),and even more frequent in pts with MRSA and P. aeruginosa coinfection (50% vs 12.5%). There was a stable decline in SatO 2 in pts with PNX and a decrease in FVC, compared to controls. Our data revealed a correlation between chronic infection by S. aureus and PNX in CF pts. Given the small population, more studies are needed to confirm this conclusion. Pseudomonas aeruginosa is a pathogenic, Gram negative bacterium implicated in lung infections of cystic fibrosis patients. P. aeruginosa can employ several different mechanisms of virulence, including formation of biofilms, which are bacterial aggregates with unique phenotypic properties. Biofilms of P. aeruginosa generally display a reduction in expression of toxins. Using a novel model of P. aeruginosa biofilm formation on cultured human airway epithelial cells, we have found that antibiotic treatment decreases bacterial toxicity toward the mammalian cells. We further found that this effect could be mediated, in part, by the magnesium transporter MgtE; expression of MgtE decreased toxicity, and deletion of the mgtE gene increased toxicity. Tobramycin treatment of P. aeruginosa biofilms grown on CF airway cells upregulated expression of the mgtE gene. In an attempt to further understand the role of MgtE in P. aeruginosa virulence, we are investigating the effects of other antibiotics on MgtE regulation. Toward that end, transcriptional assays and RT-PCR are being used to elucidate the response of mgtE after treatment of P. aeruginosa biofilms with several different antibiotics. Since MgtE plays a role in biofilm cytotoxicity and biofilms are an integral part of P. aeruginosa infections, this research will lead to a better understanding of the role MgtE plays in biofilm virulence, the impact of different antibiotics on biofilm formation, and, ultimately, P. aeruginosa virulence. Chronic lung infection with the bacterium Pseudomonas aeruginosa is a major cause of death for individuals with CF. This infection process involves the formation of P. aeruginosa communities called biofilms. While the production of some toxins and other virulence determinants generally decreases as P. aeruginosa forms biofilms, the bacteria upregulate expression of factors that promote chronic survival, including alginate (an exopolysaccharide) and genes involved in antibiotic resistance. Using a tissue culture model system for developing P. aeruginosa biofilms on human airway epithelial cells, we have previously found that expression of the bacterial magnesium transporter MgtE decreases cytotoxicity of the bacteria by decreasing transcription of the type III secretion system (T3SS). We also previously found that cytotoxicity could still be inhibited if magnesium transport was disrupted, suggesting a dual function for MgtE. In order to more clearly define these functions, we performed a structure/function analysis of MgtE. Point mutations, as well as C-terminal truncations, were created in the mgtE gene, and the resultant clones were tested for the ability to transport magnesium and inhibit cytotoxicity. Thus, we are identifying which regions of the protein are involved in which function. On the other hand, deletion of the mgtE gene leads to increased cytotoxicity by enhanced transcription of T3SS genes. In order to clarify the molecular details of this interaction, we have performed a genetic screen of random transposon mutants in an mgtE deletion mutant background, looking for suppressor mutations. In this manner, we have identified several genes that might be involved in MgtE-mediated control of T3SS transcription, including a gene for a transcription repressor that influences expression of antibiotic resistance genes. Our studies show that MgtE can affect the virulence of P. aeruginosa grown as biofilms on CF airway cells. These data further suggest that it might be possible to induce P. aeruginosa to switch to a planktonic (nonbiofilm) state while maintaining low expression of toxicity factors. Future therapies might target MgtE to disrupt biofilms and decrease bacterial toxicity. Methods: SP isolates were collected from sputum of CF patients at quarterly visits or during exacerbations and from pediatric patient sputa without CF (NCF) between 2002 and 2008 at Children's Hospital, Birmingham, AL, where 300 CF patients are followed. Patient demographics, antibiotic susceptibility, and co-infecting species were recorded. The isolates were serotyped by multiplex bead assay which allows identification of 51 serotypes including those contained in current pneumococcal vaccines (PCV7, PCV13 (Pfizer), and Pneumovax). Isolates not recognized by the assay were labeled nontypeable (NT).

Results: Eighty-seven SP isolates were collected during the study period from 65 CF patients, which represent 22% of all (300) CF patients. During the same period SP was isolated from 49 NCF sputum samples. The predominant serotypes identified in each group were 3, 19A, 19F, and NT. The frequencies of these serotypes in each of the two groups were as follows: Type 3-CF 15%, NCF 6%; Type 19A-CF 11.5%, NCF 16%; Type 19F-CF 10%, NCF 16%; and NT-CF 24%, NCF 36.7%. Only 13% of the isolates had serotypes included in the heptavalent pneumococcal conjugate vaccine, PCV7. The new 13-valent PCV contains serotypes for only 46% of the serotypes isolated from CF patients and 41% of serotypes from NCF patients.

Among the CF SP isolates 15 /87 (17.2%) isolates were mucoid: 13/15 (86.6%) of these were serotype 3; 2/15 (13.3%) were NT. Of 49 isolates from NCF patients, only 1 (2%) was mucoid (serotype 3) (P=0.0105).

Pseudomonas aeruginosa (PA) was co-isolated with SP in the CF patients 19.5 % of the time. For serotype 3 CF isolates only 1/15 was co-isolated with PA and this patient had PA prior to isolation of SP. Staphylococcus aureus (SA) was co-isolated with SP in CF 38% of the time, but only coisolated with SP in 22% of NCF sputa. There was no difference in antibiotic susceptibility between CF and NCF isolates to the individual antibiotics erythromycin, rifampin, and cotrimoxazole.

Methods: STAR-CF is a prospective multi-center cohort study being performed in 7 pediatric CF centers with high rates of MRSA to assess clinical outcomes associated with different microbiologic characteristics of MRSA. Demographic and clinical data are obtained from Port CF for individual patients and MRSA isolates are typed for SCCmec and pvl status.

Results: The mean age of participants is 10.8 ± 0.5 years and 55% are male. To date 263 MRSA isolates from 358 patients have been received and typed. Overall, 189 (72%) isolates are SCCmec II / pvl negative (consistent with HA strains) and 74 (28%) isolates contain SCCmec IV (consistent with CA strains) and 41/74 (55%) are pvl positive. The proportion of SCCmec IV vs. SCCmec II does not differ between centers, including after adjusting for age. From 2007-2009 P. aeruginosa was isolated from 30% of those with HA-MRSA and from 21% of those with CA-MRSA. A comparable proportion of subjects with HA-MRSA (9%) vs. CA-MRSA (11%) had BMI < 5%.

Antibiotic sensitivities have been done on 91 MRSA and thus far, none were resistant to vancomycin or linezolid. Sensitivities to clindamycin and TMP-SMX, which may distinguish between HA and CA, are shown in Table 1 ; these are similar for TMP-SMX in SCCmec IV and SCCmec II pvl positive strains but higher for SCCmec II negative strains.

Conclusion: The majority of MRSA isolates in the pediatric population appear to be HA strains. Thus far, the clinical presentations associated with CA and HA strains appear similar. Susceptibilities do not reflect patterns described in non-CF MRSA isolates. The study is ongoing to assess longitudinally the frequency of exacerbations.

Acknowledgment: This study is funded by Cystic Fibrosis Foundation Therapeutics, Inc. We thank all study sites for their contribution. Introduction: Central to the management of pulmonary exacerbations in cystic fibrosis (CF) is the intravenous use of a variety of potentially protein allergy stimulating antibiotics. There is gathering evidence of increasing antibody formation associated with the progressive lung disease in CF in response to numerous inflammatory stimuli. Eosinophilia has not previously been identified as a contributing factor to this inflammatory cascade.

Methods: We reviewed the records of 124 consecutive inpatient hospital admissions. All patients receiving intravenous antibiotics for more than four days and in whom paired differential white blood counts were available were analyzed and correlated with the various antibiotics received and whether prednisone was utilized.

Results: Eosinophils on admission ranged from 0.0-7.7% and from 0.2-15.9% at discharge with an average increase of 2.51% (median 2%). Two 33% outliers were discarded. There was an increase in eosinophils directly related to the duration of treatment with an average increase of 1.59% at 4-6 days of treatment progressing to 5.54% by 28 days. Patients who had repeated admissions were noted to have lost the previous discharge eosinophilia and started anew at the prior admission level both individually and as a group. Prednisone had a dampening effect on the increment of change with a 1.74% increase for the prednisone treated patients compared to a 4.35% for the non-prednisone patients. The aminoglycoside had the most effect (∆ 3.19%), followed by the cephalosporins (∆ 2.87%), tricyclic glycopeptide (∆ 2.57%), and carbapenems (∆ 2.45%). Because of the use of multiple intravenous antibiotics in the same patient the possibility of additive or cross-reaction could not be determined.

Discussion: We determined that peripheral eosinophilia increase is greater than expected in patients with CF who are receiving IV antibiotic treatment. This increase is felt related to the use of intravenous sensitizing antibiotics especially tobramicin and cephalosporins. The eosinophilia is Conclusions: Our findings suggest that the CF lung provides favorable conditions for selection of mucoid SP, compared to lung of NCF patients. The preponderance of mucoid CF isolates was serotype 3. The incidence of PA at the time SP isolation was less (P<0.0001) than the expected 50% and suggests a possible inhibition of PA by SP.

Supported by University of Alabama at Birmingham Gregory Fleming James Cystic Fibrosis Research Center Pilot and Feasibility grant, AI021458 (DEB), and NIH AI-3002 (MHN). Secondary or late acylation is carried out by the acyltransferase enzyme, htrB (lpxL). Mining of the laboratory-adapted strain PAO1 genome revealed the presence of two homologues to the S. typhimurium htrB gene, htrB1 (PA0011) and htrB2 (PA3242). These genes share 29% homology to each other. In this study, clean deletions were generated in these genes and their effect on lipid A structure was analyzed. To assess potential alterations in the fatty acid content from these mutant strains, lipid A was subjected to gas chromatography with flame ionization detection (GC-FID). Subsequently, to determine the location of the individual acylation reactions, complete lipid A structural analysis was examined by both matrix-assisted laser desorption ionization-time of flight mass spectroscopy (MALDI-TOF MS) and electrospray ionization mass spectroscopy (ESI-MS n ). Interestingly, the two genes were not redundant but each carried a site-specific function. Disruption of htrB function may decrease resistance to antimicrobial peptides, making further study of these genes important for the improvement of treatment of PA in CF patients. This work was supported by grants from the National Institutes of Health (AI47938) and the CF Foundation. 4 1. Pediatrics, UNC, Chapel Hill, NC, USA; 2. Microbiology, UNC, Chapel Hill, NC, USA; 3. Biostatistics, UNC, Chapel Hill, NC, USA; 4. Columbia University, NY, NY, USA Introduction: The prevalence of methicillin resistant Staphylococcus aureus (MRSA) in American CF patients is increasing. In 2008, the overall prevalence was ~ 23%, ranging from 4-41% between CF centers. The highest age-specific prevalence was reported in adolescents. MRSA complicates care (isolation, medication choices) and may affect clinical outcomes. Molecular analysis of MRSA shows different staphylococcal cassette chromosomes (SCCmec) types that carry the resistance genes. Epidemiologically SCCmec II has been associated with hospital-associated [HA] strains and SCCmec IV with community associated (CA) infections. Additionally the pvl gene is one of the important virulence factors, which is more frequently but not uniformly seen in SCCmec IV strains. It is unknown if different molecular types of MRSA, as have been reported in non-CF patients, are present in CF patients and are associated with different outcomes. directly related to the duration of exposure but does not appear to be a fixed sensitization as eosinophil values on subsequent admissions have returned to their prior baseline. The effect of prednisone as a modifier was noted but no dose relationship was established. However, consideration should be given to the concurrent use of low-dose prednisone to blunt the potentially deleterious effect of the release of unspecified cytokine agents from the degranulation of the excessive circulating eosinophils.

More precise studies by serologic or skin testing to identify the antibiotic agents felt to be most directly responsible should be sought in hopes of modifying the potential long-term damage of specific antibiotics. Whether the alternative use of aerosol routes of administration would cause a less eosinophilogenic response should be investigated. 

Introduction: Over the last several years, the prevalence of methicillinresistant Staphylococcus aureus (MRSA) has increased among CF patients. Although the impact of MRSA on disease progression in CF is not clearly established, some studies have shown an association between persistent infection with MRSA and a more rapid rate of decline in lung function.

Objectives: To determine the effects of persistent MRSA on the clinical status and the lung function in CF children followed at our CF clinic. To characterize the MRSA strains in this population.

Methods: A retrospective study comparing patients before and after MRSA colonization was performed. All subjects with persistent MRSA followed at our clinic between years 1990 and 2008 were included. Persistent MRSA was defined as three or more MRSA positive cultures during the study period. Outcome measures included FVC % predicted, FEV1 % predicted, FEF 25/75 % predicted, numbers of severe respiratory exacerbations and days of intravenous antibiotics treatment and nutritional status. Data were collected 2 years prior to and 2 years after MRSA acquisition. The first isolate from each patient was further characterized by molecular analysis.

Results: Twenty-three patients were identified with persistent MRSA colonization, with the first case identified in 1998. The median age of MRSA acquisition was 12.2 years (0.7-16.9). All patients were pancreatic insufficient and 63.6 % were ∆F508 homozygous. Thirteen patients were colonized with chronic Pseudomonas aeruginosa and 1 patient with Mycobacterium abscessus by the time of MRSA acquisition. Methicillin sensitive Staphylococcus aureus was isolated in 21 out of 23 patients. No significant difference was observed for the rate of lung function decline prior to and after MRSA colonization p=0.375 ) and for the numbers of days of intravenous antibiotics treatment (19.5 days/yr versus 26.7 days/yr; p=0.168). MRSA colonization was associated with an increased number of severe respiratory exacerbations (0.84/yr before compared to 1.32/yr after MRSA acquisition, p=0.024). Seventeen out of 19 patients were CMRSA-2 (USA 100) and 2 isolates were undetermined strains. All isolates were Panton-Valentine-Leukocidin negative.

Conclusion: Persistent MRSA is associated with an increased number of respiratory exacerbations requiring hospitalization in CF children. No short-term effect on lung function was observed. The predominant MRSA isolate in our population is the epidemic USA-100 strain.

Bruinenberg, P. 1 ; Serisier, D. 2 ; Cipolla, D. 1 ; Blanchard, J. 1 1. Aradigm, Hayward, CA, USA; 2. Mater Adult Hospital, Brisbane, QLD, Australia Rationale: Oral and IV ciprofloxacin is widely prescribed to treat lung infections frequently experienced by CF patients. Ciprofloxacin is often preferred because of its broad-spectrum antibacterial action. Inhaled liposomal formulations of ciprofloxacin are being developed to increase the concentration and duration of action of the antibiotic in the lung, the site of P. aeruginosa (PA) infections, while reducing the systemic drug exposure and thus reduce or eliminate the potential for side effects.

Methods: In healthy volunteers three doses of an inhaled liposomal ciprofloxacin formulation (ARD3100) were studied to determine the safety, tolerability and pharmacokinetics (PK), as single doses of 3, 6 and 9 mL of the 50 mg/mL formulation using a PARI LC Sprint nebulizer. Additionally 8 subjects inhaled the 6 mL dose once daily for 7 consecutive days. In a phase 2 study in 22 adult CF patients, the 6 mL dose was further evaluated for effects on safety, tolerability, PK and sputum PA density. In subsequent studies in healthy subjects, another formulation (ARD3150) with a release profile different from ARD3100 and new delivery systems have been evaluated.

Results: ARD3100 and ARD3150 were found safe and well tolerated. Plasma PK profiles were consistent with sustained release of ciprofloxacin from the liposomes over 24 hours with an absorption-limited half life of 1 0 hours, compatible with a single daily dosing regimen (Table 1) . ARD3150 both in healthy subjects and patients exhibited a higher Cmax compared to ARD3100. The PK parameters observed in the patients were similar to those in healthy volunteers. High concentrations of ciprofloxacin were found in sputum. In the phase 2 studies in CF patients, there was a significant reduction in PA colony forming units (CFU) in sputum after 14 days (p<0.01). Higher efficiency inhalation systems have been investigated to reduce the overall administration time and increase patient convenience. Based on the in vitro data, twelve inhalations from the AERx inhaler have the potential to deliver the same dose to lung as nebulizer loaded with a 2 mL dose.

The sustained release profile of inhaled liposomal ciprofloxacin and its effect in reducing PA sputum density support further clinical development.

Supported by Aradigm Corporation. µg/ml) is still rare, growing evidence suggests that even small increases in MIC, known as "MIC creep," are associated with dramatic increases in Vanco treatment failure. In patients with MRSA bacteremia, MIC > 1 is associated with up to 7-fold increase in localization of the calcium-sensitive transcription factor NFAT, which could contribute to the enhanced secretion of IL-13 and other cytokines. In summary, our data identified that CFTR dysfunction in T cells can lead directly to aberrant immune responses. Specifically, Cftr deficient lymphocytes directed skewed responses to Aspergillus fumigatus, leading to a higher than normal IgE response. These findings implicate the lymphocyte population as a potentially important target for therapeutics directed to the treatment of CF lung disease. Background: Patients (pts) with cystic fibrosis (CF) are at increased risk for severe illness with influenza. Annual influenza vaccination is therefore commonly recommended for people with CF. During the 2009 influenza A/H1N1v pandemic, vaccination against A/H1N1v was strongly recommended for CF pts. The aim of this study was to evaluate the immunogenicity of influenza A/H1N1v vaccination in CF pts.

Methods: In a prospective, multicenter study, 120 CF pts vaccinated against pandemic influenza A/H1N1 according to the French recommendations (i.e. one dose of AS03-adjuvanted vaccine containing 3.75 µg haemagglutinin (HA) (Pandemrix®), excepted for children < 2 years or lung transplanted patients for whom it was recommended 2 doses of a non-adjuvanted vaccine containing 15 µg HA (Panenza®) administered 21 days apart) between 13/11/2009 and 13/01/2010 were included. Serum hemagglutination-inhibition (HI) antibody titers were measured before and 21 days after vaccination.

Results: Median age of the pts was 20 years (min, max: 1-53), 13 pts had received lung transplantation. At baseline, 21 (17.5%) pts had titers of vaccine strain -specific HI antibodies equal to or higher than 1:40, the geometric mean titre (GMT) of HI antibodies was 15.3. By day 21 after vaccination, higher immune response was observed in pts receiving Pandemrix® (see Table) . Only 2 of the 13 lung transplanted patients exhibited seroconversion. No vaccine related serious adverse events have been reported.

Conclusions: One dose of adjuvanted-influenza A/H1N1v 2009 vaccine yielded a higher immune response that the non-adjuvanted one in CF pts. The vaccines were well tolerated.

Seroprotection rate: percentage of pts with HI antibody titre ≥1:40. Seroconversion rate: percentage of pts with prevaccination HI antibody titre <1:10 and post-vaccination titre ≥1:40, or a prevaccination titre ≥1:10 and at least a fourfold increase in post-vaccination titre.

Ohman, D. 1,2 ; Rowe, W.J. 1 ; Lebman, D.A. 1 1. Microbiology & Immunology, Virginia Commonwealth Univ Med Ctr, Richmond, VA, USA; 2. McGuire Veterans Affairs Med Ctr, Richmond, VA, USA Cystic fibrosis patients often have chronic pulmonary infections with mucoid strains of Pseudomonas aeruginosa, which leads to lung inflammation and increased morbidity. The mucoid phenotype is due to the production of alginate, a viscous exopolysaccharide. Others have shown that alginate production prevents killing by IFN-γ activated human peripheral macrophages. The goal of this study was to better understand the mecha-Vanco failure. This combined with the known poor penetration of Vanco into the lung, suggests this could be a significant issue in CF as well. Therefore, we decided to investigate the epidemiology of Vanco MICs at our center and determine factors associated with MIC creep and its clinical significance.

Methods: We retrospectively reviewed medical records for 52 nontransplant patients with a confirmed diagnosis of CF who have grown MRSA at our laboratory between 2007-09. After abstracting data such that all unique patient identifiers were removed, we looked at estimated odds ratios (OR) of the associations between MIC and patient age, co-infection with Pseudomonas aeruginosa (Pa) and best/mean FEV1% predicted in the past year.

Results: Nine (17%) patients had an MIC ≤ 0.5, 17 (33%) had an 0.5 < MIC < 1, and 26 (50%) patients had an MIC ≥ 1, with 3 (6%) having MIC ≥ 2. There was no significant association between MIC and age category (< 6, ≥ 6 and < 18, ≥ 18). Forty of 52 (77%) patients had a documented history of Pa colonization in the past 2 years. In patients with a history of Pa, there was a non-significant trend towards higher MIC (MIC ≥ 1 v. MIC < 1), (OR: 2.44; CI 0.63-9.45). Forty-three patients had spirometry. The mean FEV1 % predicted for patients with an MIC ≤ 0.5, > 0.5 and < 1, and ≥ 1 were 77.9%, 75.1% and 67.9%, respectively. However, when stratified by age, patients 6-18 with MIC ≥ 1 were more likely to have lower FEV1 (68.0% vs. 89.6%; p=0.03). Of note, 2 of the patients at our center were linezolid resistant.

Discussion: Our results indicate that almost half our patients have Vanco MICs which have been associated with increased rates of clinical failure in other disease states, suggesting that the MIC creep observed in our CF patients may lead to worse outcomes. Colonization with Pa may be a risk factor for MIC creep. Increased MICs were associated with decreased lung function in patients 6-18. With an increased sample size, we believe that these associations would become more clearly defined with consideration of their confounding effects. Future analysis is also warranted to look at MIC changes with time as well as to uncover possible associations between MRSA resistance patterns and gender, chronic azithromycin use, and the incidence of treatment failure when Vanco is used to treat CF pulmonary exacerbation.

Muller, C. 1 ; Braag, S. 1 ; Keeler, A. 1 ; Hodges, C.A. 2 ; Drumm, M.L. 2 ; Flotte, T.R. 1 1. Pediatrics, UMass Medical School, Worcester, MA, USA; 2. Pediatrics, Case Western Reserve University, Cleveland, OH, USA Cystic fibrosis (CF) remains the most common fatal monogenic disease in the US, affecting 1 in 3,300 live births. CF is the result of mutations in CFTR, a chloride channel and regulator of other ion channels. The mechanisms by which CFTR mutations cause chronic lung disease in CF are not fully defined, but may include the combined effects of altered ion and water transport across the airway epithelium and aberrant inflammatory and immune responses to pathogens within the airways. We have shown that Cftr-/-mice mount an exaggerated IgE response toward Aspergillus fumigatus (Af) when compared to Cftr+/+ mice. Along with the increased IgE levels, the Cftr-/-mice had higher levels of IL-13 and IL-4, mimicking both the Th-2 biased immune responses and predilection to mounting Af-specifc IgE seen in CF patients. Herein we hypothesize that these immune aberrations are primarily due to the lack of Cftr expression in lymphocytes rather than with Cftr deficiency in the epithelium. Our results indicate that adoptive transfer experiments with CF splenocytes confer higher IgE response to Af in host mice as compared to hosts receiving wild-type splenocytes. The predilection of Cftr-deficient lymphocytes to mount Th2 responses was confirmed by in vitro antigen recall experiments, where higher levels of IL-13 and IL-4 where seen only in the presence of Cftr-deficient lymphocytes. Conclusive data on this phenomenon were obtained with conditional Cftr knockout mice, where mice lacking Cftr in T-cell lineages developed higher IgE titers as compared to their wild-type littermate controls. Further analysis of Cftr-deficient lymphocytes revealed an enhanced intracellular Ca 2+ flux in response to T cell receptor activation as compared to normal lymphocytes. This was accompanied by a significant increase in nuclear nisms by which alginate confers resistance to complement-independent phagocytic killing. To accomplish this, we developed a model to investigate the early stages of bacterial phagocytosis, specifically adherence and uptake. Phorbol ester was used to induce THP-1 cells, an acute monocytic leukemia derived cell line, to become adherent and highly phagocytic macrophages. Isogenic Pseudomonas strains carrying a plasmid constitutively expressing GFP or mCherry were visualized by fluorescent microscopy. Mucoid P. aeruginosa FRD1, a frequently studied CF isolate and alginate overproducing (Alg+) strain, was used. When phagocytic uptake of an algD::Tn501 Alg-mutant, FRD1131, was compared to the Alg+ parent strain, the proportion of macrophages containing the Alg-strain greatly exceeded that of the parental strain. To verify that the bacteria had been phagocytosed, the infected cells were treated with gentamicin to remove the extracellular bacteria and lysed, after which cells were plated to assess the number of viable P. aeruginosa. Interestingly, in contrast to E. coli (which served as a positive control), intracellular P. aeruginosa were still viable after 6 hours of culture. Using flow cytometry, both the kinetics and extent of uptake were measured. Alg+ and Alg-bacteria were both taken up, but phagocytosis of the Algstrain was more rapid and efficient. Exogenous alginate (from Pseudomonas or seaweed) did not appear to protect the Alg-strain from uptake. Currently, we are examining the relationship between phagocytic uptake and the induction of signaling.

Supported by grants from the VA Medical Center and the NIH-NIAID. The cystic fibrosis transmembrane conductance regulator (CFTR) represents the predominant Cltransport conduit within respiratory epithelial cells. Activation of secondary Cltransport pathways through calcium-activated chloride channels (CaCCs) provides a potential mechanism to circumvent CF-related airway defects in mucus clearance. While much is known regarding CFTR structure and function, the Clchannel TMEM16A has only recently been identified as a major pulmonary CaCC. Because TMEM16A may serve as a respiratory target for CF therapy, studies to improve understanding of CaCC in CF airway disease are indicated. In the present experiments, primary murine nasal septal epithelial (MNSE) cultures (wild type (wt) and transgenic CFTR -/-) were exposed to lipopolysaccharide (LPS) or an ultrafiltrate of PAO1 Pseudomonas aeruginosa (bacteria-free preparation from 20 hour log-phase growth). Basal media was collected from airway cell monolayers and analyzed for murine CXCL1/KC (human IL-8 analogue) by ELISA to confirm activation of NFKB mediated inflammatory signaling. Cultures were mounted in Ussing chambers for ion transport measurements. The presence of TMEM16A mRNA and protein in primary cell culture was determined using RT-PCR and immunofluorescence, respectively.

Results: MNSE cultures incubated with PAO1 filtrate or LPS (100 nM) for 24 hours (i.e. exposures substantially below physiologic levels present in CF sinuses) produced significantly elevated CXCL1/KC (PAO1, 1267.4 ± 54.3 pg/ml and LPS, 1774 ± 159.4 pg/ml) when compared to controls (660 ± 139.5 pg/ml) (p<0.05). CaCC-mediated Cltransport [change in short-circuit current, ∆I SC (µA/cm 2 )] measured using the purinergic agonist UTP (150 µM) was significantly decreased in wt MNSE compared to controls (PAO1, 8.6±0.72; LPS, 10.6±0.91; control, 14.1±0.87, p<0.05) . Results were confirmed in the absence of CFTR (CFTR -/-MNSE), which also demonstrated significant inhibition of ∆I SC (PAO1, 12.3±0.34; LPS, 13.2±0.96; control, 19.2±2.1, p<0.05) . Expression of TMEM16A mRNA and protein were qualitatively verified in MNSE cultures.

Conclusions: Exposure to LPS or PAO1 extract in primary airway epithelial cells led to decreased CaCC-mediated Cltransport. Because we have verified TMEM16A expression and cell surface localization, quantitative studies of mRNA and protein are in progress to establish the mechanisms underlying CaCC repression described here. The findings have broad implications for rescue strategies aimed at restoring Cltransport via TMEM16A in CF airway disease. In particular, pharmacologic interventions intended to stimulate TMEM16A will need to overcome suppression of the CaCC in the setting of airway infection and inflammation.

Planet, P.J. 1, 2 ; Harasym, M. 1, 3 ; Quittell, L. 1 ; Saiman, L. 1 ; Prince, A. 1 1. Pediatric Infectious Diseases, Columbia University, New York, NY, USA; 2. Sackler Institute of Comparative Genomics, American Museum of Natural History, New York, NY, USA; 3. Biology, College of New Rochelle, New Rochelle, NY, USA Background: Molecular, culture-independent techniques have greatly enhanced the characterization of the microbial communities in the lungs of patients with cystic fibrosis (CF), and have helped to identify new potential pathogens. We hypothesized that specific changes in microbial diversity, evenness, and species abundance would correlate with changes in clinical status.

Methods: Respiratory samples (swab, sputum, and bronchoalveolar lavage) from CF patients ages 0 to 19 years are being collected prospectively at every clinical encounter. Data on the current clinical status including FEV1, use of chronic medications such as azithromycin, TOBI, aztreonam, Pulmozyme and acute antibiotic treatment are documented. We use three different small subunit rDNA-based, culture-independent techniques (T-RFLP, Sanger sequencing, and pyrosequencing) to carefully characterize, and cross-validate the microbial diversity in samples. Every sample is also compared to duplicate culture data obtained from our clinical microbiology laboratory.

Results: To date 95 patients have been enrolled in the study with 95 initial isolates and 134 repeat specimens. We expect to enroll 150 patients who will be followed for 3 years with a projected total of >900 specimens. Preliminary data suggests that respiratory samples from CF patients may be more diverse than previously thought, with many samples having more than 20 bacterial phylotypes present. In addition, our results underscore the importance of anaerobic bacteria (eg., Prevotella spp.), as abundant and important members of the CF respiratory microbiota. This ongoing pilot study's aim was to screen a variety of essential oils (EOs) for antimicrobial activity toward cystic fibrosis (CF)-adapted pathogens, toward their potential use in the battle of persistent and polymicrobial infections in CF. Essential oils (EOs) have long been known to possess healing effects, with many providing antimicrobial, antioxidant, and/or anti-inflammatory activities. Individuals with CF suffering from chronic respiratory tract infections are frequently co-colonized with one or more phenotypes of Pseudomonas aeruginosa (PA), and/or Staphylococcus aureus (SA), along with other bacterial and fungal species. The EO screening study presented here tested laboratory strains and clinical isolates of CF pathogens PA and SA. EO inhibition of respiratory pathogen growth was evaluated by disk diffusion and vapor exposure assays. To date,159 EOs have been evaluated. Generally SA was more sensitive to contact and vapor inhibition by EOs than PA. Fifty-three oils were effective on 1 or more PA strains. PA zones of inhibition (ZoI) ranged from 6 mm (w/6mm disk) to over 20 mm in disk diffusion tests, employing consistent volumes of "neat" essential oils. Where no growth was seen in the zone, inhibition was considered complete; where growth was less within a zone surrounding the disk, it was judged as incomplete. One CF isolate was generally the most sensitive and lab strain ATCC 39018 was generally the least sensitive PA based on zone size and completeness of growth inhibition. The top performers evaluated thus far are bay (Pimenta racemosa), cinnamon bark (Cinnamomum verum), oregano (Origainum compactum), thyme white (Thymus vulgaris), cassia (Cinnamomum cassia) and wintergreen (Gaultheria fragrantissima), with disk ZoI of 9 -35 mm. Cassia was the only oil to significantly inhibit all PA tested (avg ~20 mm for these top 6 oils). Ongoing studies include To better understand the complex bacterial/inflammatory cell interactions occurring in the CF lung, we have chosen to study the arginine metabolite, agmatine. Agmatine is the product of the arginine decarboxylase pathway present in both bacteria and mammals. In mammals agmatine acts as a neurotransmitter, a regulator of nitric oxide and a ligand of the α2-adrenoreceptor, which is implicated in lung inflammation. In most bacteria, agmatine serves as the major intermediate in the production of polyamines, which are essential for bacterial growth. We believe both the host and pathogen agmatine pathways contribute to the agmatine pools we have found in CF sputum. We have previously shown that sputum agmatine is positively correlated with inflammatory markers and that agmatine is capable of NF-κB mediated inflammatory signaling in both macrophage cell culture and in mouse lungs as measured in a transgenic NF-κB reporter mouse. Our new data confirms this signaling occurs through the α2-adrenoreceptor as systemic administration of a specific α2-adrenoreceptor blocker in mice inhibits agmatine induced NF-κB stimulation in the lung.

In a separate line of recently published work we discovered a novel operon that induces P. aeruginosa to form a biofilm in the presence of agmatine. As agmatine appears to be an inflammatory mediator of the lung, P. aeruginosa may have evolved this mechanism to detect a potential inflammatory milieu and retreat to a biofilm in its presence. We have generated a number of mutants of the agmatine pathways in P. aeruginosa and our preliminary results in an acute pneumonia model suggests strains that do not metabolize agmatine are more lethal than those that cannot make agmatine. This suggests unchecked bacterial agmatine production may overdrive the inflammatory response, or prevent bacterial killing by a possible attempt to form a biofilm. We have also generated agmatine-induced, bioluminescent reporter P. aeruginosa to measure bacterial transcription of the agmatine operons while in the live mouse lung or in cell culture models. These reporter strains have revealed real-time detection of agmatine pools within the mouse lung during an acute pneumonia model. Future studies will seek to determine the contribution of bacterial, or host cell derived, agmatine pools in lung inflammation and potentially biofilm formation in vivo.

This work supported by NIH P30 HL 101311-01 and NIH RO1 HL 61419. Rationale: Human coronaviruses have been associated with upper respiratory illness, but newly described species may be involved in lower respiratory tract manifestations. The aim of this study was to investigate the presence of human coronaviruses (OC43, 229E, HKU1 and NL63) in pediatric patients with cystic fibrosis.

Methods: One hundred and three patients with cystic fibrosis (48F:55M), age range 3.8 months to 17.9 years, median 8.9y, were enrolled in the study. They were submitted to clinical examination, spirometry, pulse oximetry and collection of nasopharyngeal aspirate (nasal blow) and sputum during routine and unscheduled clinical visits to the outpatient clinic. Nucleic acid extraction was performed by Qiamp Viral RNA kit and synthesis of cDNA by High Capacity cDNA kit. Human coronaviruses were identified by RT-PCR with generic primers and amplicon detection obtained by capillary electrophoresis. Determination of human coronavirus species was performed by amplicon sequencing using an ABI377 automated sequencer.

Results: A total of 408 samples (1-9 samples /patient, median 4 samples/patient) were obtained from September 2006 to August 2007. Respiratory exacerbations were identified in 142 occasions, and hospital admissions occurred in 31 occasions. Rhinovirus was the main viral agent identified (140 samples, 34.3%), but was not associated with respiratory exacerbations. Human coronavirus (hCoV) was detected in 19/408 samples (4.65%), and was also not associated with respiratory exacerbations. Amplicon sequencing was performed in 16 samples. Human coronavirus OC43 was identified in 5 (31.25%) samples, 229E in 1 sample (6.25%), HKU1 in 4 (25%) and the NL63 in 6 samples (37.5%). Age distribution was similar among different hCoV species. Co-infections were identified in 6 samples chemical analyses of effective oils, antimicrobial assays of the individual components of the most effective oils in initial screening, chemical modification of key constituents to improve their effectiveness and reduce potential adverse effects, and screening for possible synergistic effects of EOs with antibiotics used in CF. As CF infections often consist of multiple bacteria and yeasts and/or fungi, and present in the context of the thick, heterogeneous mucus environment in which the opportunistic pathogens persist, the effectiveness of EOs against organisms in CF sputum directly is also being addressed. In initial testing, cassia was found to effectively inhibit growth of all species when a typical polymicrobial CF sputum specimen with mucoid PA and C. albicans was plated directly and challenged with EO by disk diffusion or vapor. This preliminary study suggests that further screening, chemical analyses, and evaluation in cell culture systems, of specific essential oils may indeed lead to an additional and complementary, botanicals-based therapeutic tool for the treatment of complex airway infections in CF. Studies utilizing culture-independent methods suggest that the airways of chronically infected cystic fibrosis (CF) patients harbor a diverse collection of bacterial species. The organisms detected include those not considered to be typical pathogens, and recent work raises the possibility that these organisms may contribute to disease manifestations in CF. However, the oropharynx is known to be colonized with a vast array of microbial species and these organisms could confound microbial diversity studies if oropharngeal bacteria contaminate expectorated sputum. To mitigate this potential problem, we studied CF lung explants removed at the time of lung transplantation using a culture-independent method to measure microbial diversity. To date, endobronchial secretion samples from five explanted lungs were evaluated using bacterial tag-encoded FLX-Titanium amplicon pyrosequencing. Three patients also provided preoperative upper airway sputum samples for bacterial comparison.

Direct lung sampling revealed that conventional CF pathogens dominated the lower airways. Furthermore, a similar spectrum of colonizing species was detected regardless of whether culture-independent or culture-dependent methods were used. These data raise the possibility that in end-stage CF lung disease, culture-independent analysis of samples obtained through the upper airway overestimates bacterial diversity present in the lungs. Cystic fibrosis (CF) lung disease is characterized by chronic airways infections that are nearly impossible to eradicate once established. The bacterial infections, particularly due to P. aeruginosa, exist as a biofilm, a form of growth that is resistant to phagocytosis and antibiotic therapy. Another hallmark of CF lung disease is an intense neutrophilic infiltration and activation within the airways. Despite the recognition of the bacterial biofilm as a pathologic nidus in the airway, there is very little knowledge of its in vivo regulation or the signals it releases to generate such a robust, but ineffective, inflammatory response.

(31.5%), mainly with rhinovirus (4 cases). Exacerbation of respiratory disease was diagnosed in 142/408 occasions (6/19 occasions when hCoV was identified). Hospital admission occurred in 31 occasions (2 with hCoV +). Main symptoms in hCoV positive patients were coryza (47%), nasal obstruction (31%), dry cough (21%) and sore throat (16%). A respiratory exacerbation was diagnosed among 50% of patients with identification of hCoV NL63.

Conclusions: We found the newly described species of hCoV among our cystic fibrosis children. Symptoms are mainly related to the upper airway and we did not find an association of all hCoV or a given hCoV species with an increased risk of respiratory exacerbation or hospital admission. The development of antibiotic resistance of Pseudomonas aeruginosa (PA) is a significant clinical concern in CF patients, particularly adults. One mechanism of resistance involves antibiotic efflux pumps which have previously been associated with resistance in non-CF samples. They have been suggested to play a major role in PA infections in patients with CF. Resistance to aminoglycosides has specifically been associated with increased expression of the mexXY efflux pump in both CF and non-CF isolates, however little is known about the time course of change in the CF lung. This study aims to evaluate whether the level of mRNA expression of mexY, a subunit of the mexXY efflux pump, changes in CF patients at different durations of infection and its association with development of antibiotic resistance.

Methods: Clinical isolates of PA have previously been obtained from both pediatric and adult CF patients for evaluation in this study. Antibiotic sensitivities were determined by conventional clinical microbiology testing for each sample. The patient cohorts were divided into adults (> 18yr) and children with either their first positive PA culture or those chronically infected (> 2 sputum PA (+) >1 year duration). PA isolates were evaluated using quantitative real-time PCR to compare mexXY mRNA expression in clinical isolates compared to wild-type strain PAO1.

Results: Isolates were randomly selected from a large collection of clinical PA isolates obtained from CF patients at various stages of infection. A total of 55 isolates have been analyzed to date divided between the three patient cohorts. Mean fold increases in mexY mRNA expression compared to wild-type were 4.54±7.67, 16.62±16.20, and 46.83±27.94 in the first infected, chronically infected children and adults respectively. The expression levels in isolates obtained from adult patients were significantly higher compared to first infected and chronically infected children (p<0.001). Evaluation of individual data points suggests a bimodal distribution of levels with an increased proportion of high level expression in isolates acquired from adult patients compared to children, although isolates with high levels were seen in all patient cohorts. Gentamicin resistance was noted to be 0%, 25%, and 75% in first infected, chronically infected children, and adults respectively. However evidence of antibiotic resistance did not universally correlate with increased mexY mRNA expression.

Conclusions: Our data suggests that mexY mRNA expression may play a role in aminoglycoside resistance in PA infections in the CF lung over time. These increased levels of expression may be associated with increased antibiotic resistance, however the association remains unclear and further evaluation is ongoing. Jassem, A.N. 1, 4 ; Zlosnik, J.E. 2, 4 ; Henry, D.A. 2, 4 ; Ernst, R.K. 3 ; Speert, D.P. 2, 4 1. Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada; 2. Pediatrics, University of British Columbia, Vancouver, BC, Canada; 3. Department of Microbial Pathogenesis, University of Maryland, Baltimore, MD, USA; 4. Centre for Understanding and Preventing Infection in Children, Vancouver, BC, Canada The Burkholderia cepacia complex (BCC) group of Gram-negative bacteria are highly virulent, opportunistic pathogens in cystic fibrosis (CF) patients. It is the current dogma that all species of the complex are highly and intrinsically resistant to polycationic antimicrobials, and that characteristics of the lipopolysaccharide (LPS) are responsible for the resistance. Cationic agents enter Gram-negative bacteria through LPS-mediated selfpromoted uptake, relying specifically on anionic lipid A binding sites. Here we observed that most environmental and clinical isolates of B. vietnamiensis, unlike other BCC species, were intrinsically susceptible to aminoglycosides, but resistant to other cationic agents (natural and synthetic cationic antimicrobial peptides and polymyxin B). Furthermore, some strains acquired aminoglycoside resistance during chronic CF infection. The objective of this study was to characterize aminoglycoside susceptibility in B. vietnamiensis and gain insights into the potential mechanisms involved in aminoglycoside resistance. Gentamicin and tobramycin time-kill assays revealed that high drug concentrations were required for significant killing of a susceptible B. vietnamiensis CF strain. Susceptible B. vietnamiensis CF isolates accumulated 5 to 6 times more [ 3 H]gentamicin than a resistant isolate, but permeability to the fluorescent hydrophobic probe 1-N-phenylnapthylamine (NPN), a measure of self-promoted uptake, was not enhanced following incubation with gentamicin or tobramycin. MALDI-TOF mass spectroscopy revealed the presence of lipid A-associated aminoarabinose moieties in aminoglycoside-susceptible and -resistant B. vietnamiensis isolates. These data suggest that in B. vietnamiensis decreased cellular accumulation is involved in aminoglycoside resistance and aminoglycoside entry does not occur via self-promoted uptake. We are currently investigating the basis of the observed differential gentamicin accumulation. This work was supported by the Michael Smith Foundation for Health Research and the Canadian Cystic Fibrosis Foundation. Human influenza A viruses bind host cell sialic acids terminating in an α(2,6)-linkage, while hybridized avian and animal strains adapted to infect humans may bind both α(2,3)and α(2,6)-linkages. Also, many candidate adeno-associated virus (AAV) gene therapy vectors transduce via sialic acid receptors. Human airways are commonly colonized with non-typeable Haemophilus influenzae (NTHi), a gram negative bacterium. NTHi inhabits the nasopharynx of asymptomatic carriers and the conducting airways of cystic fibrosis patients. NTHi scavenges sialic acids, incorporates these sugars onto its surface and forms sialic acid-rich biofilms.

We hypothesized that NTHi would inhibit epithelial binding/infection of influenza A viruses by either removing sialic acids from host cell surfaces or competitively blocking binding through sialic acid expression, thus protecting airway epithelia from viral infection. AAV 2.5T, a novel serotype developed by directed evolution, was used to model α(2,3)-linkage receptor specific viruses, and AAV 1 was selected for its α(2,6)and α(2,3)-linkage specificity. Seasonal and 2009 Seasonal and H1N1 live-those diagnosed in 1994 Seasonal and -2000 Seasonal and and 1986 , p< 0.0001. These findings imply that being diagnosed more recently shortened the time to first PA infection to a lesser extent in children with earlier onset of PA infection compared those with later onset of PA infection. With regard to gender, across all quantiles in the range of 10th-65th, boys and girls had the similar age at the first PA infection, i.e., age 2.8 yrs at the first quintile (20th percentile) and 8.3 yrs at the third quintile (60th percentile).

Conclusion: Our analysis using the quantile regression model discovered a heterogeneous association of the effect of calendar year on the onset of PA infection in children with CF.

Supported by NSF DMS-0706985 and NIH R01 DK 072126. The ethnic distribution of CF is changing due to an increasing number of Hispanic births in many U.S. cities. Some studies suggest that CF outcomes are worse in Hispanic patients. However, the underlying causes of poor outcomes in this population are not well understood. The objective of this study was to compare the age at initial positive Pseudomonas Aeruginosa (PSA) culture in Hispanic-White (HW) patients with CF compared to Non-Hispanic White (NHW) patients and to determine risk factors associated with an early PSA culture in individuals with CF who are HW to those who are NHW.

Methods: Children with CF identified through newborn screening or meconium ileus in Colorado and Wyoming between 1982 and 2007 were included in this study. Demographic, diagnostic, and clinical outcome data were obtained from an IRB approved longitudinal study of newborn screening outcomes. Hispanic ethnicity was based on self-report at time of diagnosis. T-tests and Kruskal-Wallis tests were used to compare continuous data, and chi-square tests were used for frequencies. A time-to-event Cox proportional hazards models was used to test PSA free time in the HW compared to the NHW; hazards ratios (HR) and 95% confidence intervals are reported.

Results: The age at diagnosis was higher in the 50 HW children (median 32 days, IQR 26-51) compared to the 356 NHW (median 25 days, IQR 18-40 p<0.01). No differences were found in the CFTR genotype high risk classification (HW: 44/50, 88%; NHW: 284/356, 80%, p = 0.10). HW children with CF were significantly younger than NHW when they had their first positive PSA culture (3.8± 3.3 years vs. 5.4 ± 4.9 years, p 0.02). Evaluation for previously described risk factors for PSA acquisition showed: 1) No significant difference in the number of visits to the clinic, although the HW were younger at the most recent visit (7.74± 4.93 years vs. NHW 10.0±6.3 years, p<0.01); 2) HW had lower socioeconomic status as determined by percent of children on State funded health insurance (HW: 25/50, 50%; NHW: 59/356,17%, p<0.0001) and the highest level of maternal education (HW 11.2±4.1 years vs. NHW 14.7±8.6 years, p<0.0001); 3) No significant difference in smoking in the household. Further analysis of the PSA culture data showed that HW are at higher risk for an early PSA positive culture than NHW; unadjusted HR for first positive PSA culture in HW was 1.66 (95% CI 1. 2-2.4) . After adjusting for location where culture was performed (new vs. old facility), severity of mutation, and maternal education, HR=1.51 (95% CI 1.0-2.3).

Conclusion: HW children were at higher risk for early positive PSA culture than NHW children, and this relationship was significant after adjusting for socioeconomic status and CFTR mutation. Early positive cultures and colonization with PSA is shown to be associated with poor clinical outcomes in CF. Further investigation needs to be done to evaluate whether HW children with CF are at risk for worse outcomes due to early positive cultures with PSA. This project was supported by NIH/NCRR Colorado CTSI Grant Number UL1 RR025780. attenuated influenza virus (LAIV) vaccines were used for their α(2,6)-linkage receptor specificity. Calu-3 human airway epithelial cells were grown at an air-liquid interface, in the presence or absence of NTHi. Quantitativepolymerase chain reaction, trans-epithelial conductance measurements, and multiple imaging techniques were utilized to evaluate viral binding, cytotoxicity, and infection.

When colonized with NTHi, human airway epithelial transduction by α(2,3)-specific AAV 2.5T was reduced compared to non-colonized airway cultures. Control, non-sialic acid binding AAV serotype (AAV 2) transduction was unaffected by NTHi. No NTHi protection was observed upon infection of NTHi colonized epithelia with α(2,6)-specific seasonal LAIV, suggesting NTHi's protective effect is α(2,3)-linkage specific. AAV 2.5T did not bind to NTHi's surface. Moreover, binding to the matrix of NTHi's biofilm was minimal, and not different to AAV 2. These data suggest that viral binding to sialic acids on NTHi's surface or biofilm is not a mechanism of protection. Finally, sialic acid-specific lectin staining showed that NTHi reduced α(2,3)but enhanced α(2,6)epithelial surface sialic acid expression.

To our knowledge, this is the first time sialic acid interaction in NTHi and viral infection has been studied. These data suggest NTHi colonization may offer a novel defense mechanism from hybridized avian and animal influenza A viruses. Moreover, colonization may act as a barrier to gene therapy using α(2, 3)-linkage specific vectors and to vaccination in future reassortant influenza A pandemics. The synergistic cytotoxicity of NTHi and seasonal LAIV, which we speculate is α(2, 6)-linkage mediated, sheds light on potential therapeutic targets for influenza A induced acute exacerbations in CF.

Background: Pseudomonas aeruginosa (PA) accelerates lung disease and reduces survival. Our previous studies show that factors influencing PA infection included gender, calendar year of CF diagnosis and method of diagnosis (Am J Epidemiol 2004 , Biometrics 2009 ). However, the statistical methods used in these studies are limited to modeling the hazard of PA event (proportional hazard model) or comparing mean/median prevalence (temporal regression model) by predictors.

Objective: We utilized a novel statistical model, namely, the quantile regression model (JASA, 2008) , to investigate PA infection in children with CF during the first decade of life. This method adds a new dimension to existing analyses because it can model PA at any given quantile of interest. Therefore, the quantile model provides a more complete picture of the association between predictors and onset of PA infection. For example, it allows us to test a new hypothesis that the effect of any predictor of interest may be different in patients acquiring PA early (e.g., the 1st quintile) versus late (e.g., the 3rd quintile).

Methods: Data from 11375 children documented in 1986-2005 CFF Patient Registry were utilized. Among them, 2735 children (24%) were left censored, meaning that the first PA infection occurred prior to first observation; 2284 (20%) were right censored, meaning that they did not acquire any PA infection at the end of the observation period. No subjects were excluded since our method can accommodate double censoring.

Results: Our analyses showed that calendar year of diagnosis was significantly associated with the onset of the first PA infection but gender was not. Compared to children diagnosed in 1986-1989, children diagnosed in a more recent cohort (1994) (1995) (1996) (1997) (1998) (1999) (2000) CO, USA; 2. CFF, Therapeutics Development Network, Seattle, WA, USA; 3. CFF Therapeutics Inc, Bethesda, MD, USA; 4. Case Western Reserve University, Cleveland, OH, USA; 5. University of Washington, Seattle, WA, USA; 6. Indiana University, Indianapolis, IN, USA; 7. University of Michigan, Ann Arbor, MI, USA; 8. Vanderbilt University, Nashville, TN, USA Background: There is an urgent need for development and validation of biomarkers that more quickly predict the usefulness of potential drugs in CF.

Objective: Develop a CF biomarker panel that quickly demonstrates a response to treatment by determining changes in candidate protein biomarkers in blood before and after intravenous (IV) antibiotic treatment for pulmonary exacerbations.

Methods: CF subjects ≥10 years of age were enrolled at the onset of a pulmonary exacerbation, diagnosed by symptom-based criteria (adapted from Fuchs, NEnglJMed, 1994) . Blood was collected within 24 hours of starting IV antibiotic therapy (V1) and approximately 2 weeks into the treatment course of IV antibiotics, generally between days 10-21 (V2). We have measured 8 protein biomarkers (G-CSF, IL-1ra, IL-6, IL-8, TNF-α, TGF-β1, NEAPC, hsCRP) in paired pre and post IV antibiotic plasma samples and are currently measuring 7 additional proteins (α1-antitrypsin, haptoglobin, ceruloplasmin, arginase-1, calprotectin, serum amyloid A, sCD40L). Serum aliquots collected both pre and post IV antibiotic therapy were submitted for complete blood counts. Additionally, spirometry, a CF Respiratory Symptom Questionnaire, and sputum for bacterial culture were collected at both time points. Changes in individual and combinations of biomarkers will be correlated to changes in these clinical parameters.

Results: One hundred subjects with CF (70 females, age 22±8 years) were enrolled in the study and 85 had samples collected and analyzed at both time points. Values of 4 plasma biomarkers (IL-8, TGF-β1, NEAPC, hsCRP) were all above the lower limits of detection (LOD) of the respective assays, while 13% of G-CSF values, 76% of IL-1ra values, 28% of IL-6 values, and 26% of TNF-α were at or below the LOD. Antibiotic treatment led to significant reductions in 6/8 plasma biomarkers (G-CSF, IL-1ra, IL-6, TGF-β1, NEAPC, hsCRP). Mean log 10 transformed biomarker measurements at V1 and V2 and the difference between visits are reported below.

Conclusions: Circulating proteins demonstrate a response to treatment of CF pulmonary exacerbations. G-CSF and hsCRP are the most responsive of the 8 biomarkers assessed thus far. A CF biomarker panel could speed up drug discovery, leading to a quicker, more efficient drug development process for the CF community.

Supported by CF Foundation (#SAGEL07A0) & Colorado CTSA #1 UL1RR014780 (NCRR/NIH).

Griesenbach, U. 1 ; Soussi, S. 1 ; Newman, N. 1 ; Donovan, J. 1 ; Leiton, J. 2 ; Campbell, K. 2 ; Gibson, J. 2 ; On behalf of the UK CF Gene Therapy Consortium, .. 1, 2 1. Gene Therapy, Imperial College, London, United Kingdom; 2. Medical Genetics, Western General Hospital, Edinburgh, United Kingdom Inflammatory markers in sputum and serum have been used with variable success as outcome measures in interventional studies. Limited data are available on reproducibility of such assays in cystic fibrosis (CF) particularly over a long time period. The UK CF Gene Therapy Consortium Run-in Study was designed to address this; stable patients (FEV1>40, >10yrs age) attending the Royal Brompton Hospital (London), Western General Hospital (Edinburgh) and the Royal Hospital for Sick Children (Edinburgh) were recruited into the study which ran over an 18 month period with up to 4 hospital visits.

Patients provided a 24 hr sputum collection, for weighing at each visit. Spontaneous sputum was collected at the beginning of each visit; if insufficient sample was obtained, sputum was induced with hypertonic saline. Inflammatory markers were measured in dithiothreitol-processed sputum (total and differential cell counts, IL8 (and other cytokines), calprotectin, neutrophil elastase, myeloperoxidase and extracellular DNA). Blood was collected at each visit for cytokines (IL1β, IL6, IL8, IL10, IL12(p40) and TNFα) and calprotectin.

Data are available from 189 patients at 655 visits. Adequate sputum for analysis was obtained at only 60% of visits. Sputum induction accounted for only 16% of adequate samples.

Median and range for each detectable assay are shown below. Serum cytokines IL1β, IL10, IL12(p40) and TNFα were undetectable at each visit, and IL6 was only detectable in 17% of samples. To assess intra-individual variability the coefficient of variation of results across each visit for each patient is presented. Both sputum and serum assays showed a large range of results at each visit, but the variation for each individual was much higher than the ideal 10%. Serum assays were not able to discriminate between CF and non-CF, apart from calprotectin (CF 10.1(0.8-72.5) vs non-CF 0.40 (0.2-1.12) p<0.001).

Due to the difficulty in obtaining sputum samples reliably and the large variability of results between visits in these stable patients, we consider it unlikely that a change due to a new therapy would be detectable. As such, we are not considering sputum inflammatory markers as primary or secondary efficacy endpoints in our multidose gene therapy trial. Serum markers also appear to be of limited use is assessing efficacy but will still be useful for toxicology and safety studies.

Funded by the UK CF Trust. ment had the largest increase in airway NO production and pulmonary function.

Conclusions: This is the first study to show that both ADMA, an endogenous inhibitor of NOS as well as SDMA, a L-arginine derivate that reduces substrate availability for NOS, are increased in CF airways. Antibiotic treatment is associated with a decrease in ADMA and SDMA sputum concentrations which results in increased L-arginine availability, increased airway NO production and improved pulmonary function.

Supported by the Lynne and Arnold Irwin Family Foundation and the Canadian CF Foundation.

Amin, R. 1 ; Subbarao, P. 1 ; Lou, W. 2 ; Jabar, A. 1 ; Balkovec, S. 1 ; Jensen, R. 1 ; Kerrigan, S. 1 ; Gustafsson, P. 3 ; Ratjen, F. 1 1. Pediatrics, Sick Kids Hospital, Toronto, ON, Canada; 2. Biostatistics, Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada; 3. Pediatrics, Central Hospital, Skovde, Sweden Rationale: Sensitive outcome measures to assess the efficacy of therapeutic interventions in cystic fibrosis patients with mild lung disease are currently lacking.

Objectives: Our objective was to study the ability of the Lung Clearance Index (LCI), a measure of ventilation inhomogeneity to detect a treatment response to dornase alfa inhalation in paediatric CF patients with normal spirometry.

Methods: Paediatric CF patients ranging between the ages of 6 and 18 years with FEV1% predicted ≥ 80% were eligible for the study. In a crossover design, seventeen CF patients received 4 weeks of dornase alfa and placebo in a randomized sequence separated by a 4 week washout period. The primary endpoint was the change in LCI from dornase alfa versus placebo. Secondary outcome measures included spirometry and quality of life as measured by the CFQ-R. The mixed model approach with fixed subject effects incorporating period-dependent baselines was used to examine the outcomes. This trial was registered with ClinicalTrials.gov, number NCT00557089.

Measurements and Main Results: The mean age ± SD was 10.32 ± 3.35 years. Dornase alfa significantly improved LCI compared to placebo (0.90 ± 1.44, p=0.022). Small airways flows as measured by forced expiratory flow at 25% to 75% expired volume (FEF25-75) as measured in liters, percent predicted and z-scores also improved in subjects on dornase alpha compared to those on placebo (0.175 L ± 0.07, p=0.02; 6.1% ± 2.5, p=0.03; and 0.28 z-score ± 0.11, p =0.03). The other secondary outcomes were not significant.

Conclusions: Dornase alfa significantly improved LCI in paediatric CF patients with mild lung disease. Therefore the LCI may be a suitable tool to assess early intervention strategies in this patient population.

Ives, A. 1 ; Irving, S. 1 ; Hogg, C. 1 ; Edey, A. 1 ; Hansell, D. 1 ; Alton, E.W. 1,2 ; Davies, J.C. 1,2 ; Bush, A. 1 1. Paediatric Respiratory Medicine, Royal Brompton Hospital, London, United Kingdom; 2. Gene Therapy, Imperial College, London, United Kingdom Aims: Cystic fibrosis (CF) and primary ciliary dyskinesia (PCD) are both neutrophilic airway diseases that are characterised by mucociliary dysfunction, chronic infection and bronchiectasis. There are significant differences however in the age at acquisition of infection, rate of lung function decline and life expectancy. Current PCD management has largely been extrapolated from that used in CF, including methods of measuring disease progression.

LCI is a reliable, repeatable test which in CF becomes abnormal before conventional tests such as spirometry deteriorate [Am J Respir Crit Care Med 2005; 171:249-256] . It is at least as sensitive as CT in the detection of abnormal structure [Thorax 2008; 63:129-134] . PCD is a neutrophilic dis- Background: Cystic fibrosis is associated with reduced airway nitric oxide (NO) production. Reasons for low NO production in CF include limitation of substrate (L-arginine) availability for NOS caused by increased arginase activity. The naturally occurring competitive NOS inhibitor asymmetric dimethylarginine (ADMA) and symmetric dimethylarginine (SDMA) also compete with L-arginine for transmembrane transport and could affect NO production. The role of ADMA and SDMA for CF airway NO production has not been previously studied.

Methods: Thirty-four children (8-17 years of age) with a confirmed diagnosis of CF were included in this study. Sputum was collected in 16 clinically stable patients and eighteen patients before and after IV antibiotic treatment for a pulmonary exacerbation. Comparison was made to sputum samples from eleven healthy non-smoking controls. The concentrations of L-arginine, citrulline and ornithine as well as ADMA and SDMA were measured in sputum samples using liquid chromatography-tandem mass spectrometry.

Results: Mean (±SEM) FEV1 was 80.4 (±4.5)% predicted in stable CF patients and increased from 52.5 (±2.5)% to 65.8 (±3.4)% predicted after antibiotic treatment (p<0.001, paired t-test) in CF patients with pulmonary exacerbations. Sputum ADMA was increased in stable CF patients compared to control (0.04±0.006 vs. 1.5±0.4 µmol/L) as was SDMA (0.12±0.003 vs. 0.28±0.05 µmol/L) (p < 0.01, respectively). The ratio of Larginine/ADMA was decreased in stable CF patients (628.3±74.0 vs. 393.6±68.4, p<0.05). Antibiotic treatment for a pulmonary exacerbation resulted in a significant decrease in sputum arginase activity (216.7±35.6 vs 75.1±13.2 mU/mg protein, p=0.0005). In contrast, the concentrations of sputum NO metabolites (NOx) increased (432.3±83.6 vs 893.6±143.7 µmol/L, p=0.0004), confirming previous results from our group. The absolute concentrations of L-arginine, ADMA and SDMA decreased with treatment (p < 0.01, respectively). However, the L-arginine/ADMA ratio in CF sputum increased significantly with treatment (194.7±30.9 vs 311.9±47.7, p=0.01) suggesting an increase in the ratio of substrate over inhibitor for NOS.The changes in sputum ADMA level showed a significant correlation with both changes in sputum NOx concentrations (r=-0.63) and changes in FEV1 (r=-0.66), as did the change in SDMA (r=-0.49 and -0.63, respectively) suggesting that CF patients who showed the greatest degree of decrease in sputum ADMA and SDMA concentrations with antibiotic treat-ease often compared to CF, but little is known about the relationships between spirometry, LCI and HRCT in PCD. We hypothesised that the worse prognosis for CF would be reflected by disproportionately greater lung destruction as a given level of functional impairment than in PCD.

Materials and Methods: Twenty-two CF patients (mean age 28.0, range 10.0-49.4 years) were matched by age, sex and FEV1 to 22 patients with PCD (mean age 30.3, range 10.8-62.2 years). All underwent LCI, FEV1 and HRCT. HRCT was scored by two blinded experienced thoracic radiologists in the following: extent and severity of bronchiectasis, airway wall thickness, small and large mucus plugs, air trapping, and consolidation. The means of the lobar scores were used to calculate an overall score in each category.

Results: There were 10 males and 12 females in each patient group. There was no difference in height, weight or BMI. Mean (SD) FEV1 was 64.2% (14.2) and 64.7% (18.0) for CF and PCD respectively (p=0.93). Four PCD and 2 CF patients had a normal FEV1 (>80%). All patients had an abnormal LCI (>7.1): CF 12.6 (2.7), PCD 11.7 (3.2) (p=0.34).

In CF, CT scores for several parameters correlated significantly with FEV1 and LCI. Extent of bronchiectasis (3.29 v 1.75, p=0 .0011), severity of bronchiectasis (1.93 v 1.32, p=0 .02), airway wall thickness (1.98 v 1.47, p=0 .0001)and large mucus plug score (0.63 v 0.46, p=0.039) were all significantly higher in CF.

There was a significant negative correlation between LCI and FEV1 in the CF group (r2 0.298, p=<0.01)as well as between LCI and most measures of structural lung disease bu these were not seen in PCD.

Conclusion: 1. For any given level of large and distal airway dysfunction, as shown by FEV1 and LCI respectively, there is greater structural impairment in CF than PCD.

2. The relationships between function and structure cannot be extrapolated between diseases.

3. We speculate that the greater structural impairment in CF reflects the greater airway burden of neutrophil elastase in CF [Pediatr Asthma Allergy Immunol 2009; 22; 163-169] . 1 1. Pulmonology and Radiology, Erasmus MC, Rotterdam, Netherlands; 2. Radiology, University "Sapienza", Rome, Italy; 3. Pediatrics, Ca' Foncello Hospital, Treviso, Italy; 4. Radiology, Ca' Foncello Hospital, Treviso, Italy; 5. Verona CF Center, Azienda Ospedaliera di Verona, Verona, Italy Background: The most important components of CF lung disease are bronchiectasis (BE) and trapped air (TA). Chest-CT is the current gold standard to diagnose and monitor BE, TA and other structural lung alterations in CF. Chest-MRI has been suggested as a radiation free alternative for CT, but is limited by a lower spatial resolution, especially in the lung periphery. To date no head to head comparison has been done to compare the sensitivity of chest-CT and chest-MRI to detect BE and TA in CF.

Aim: To compare the sensitivity of chest-CT and chest-MRI to detect BE and TA in CF.

Methods: This 2 center study was approved by the institutional review boards. Thirty-eight stable CF patients (22 females; mean age 22.7 years, range 6-51 years) had a low-dose chest-CT and -MRI performed on the same day. MRI studies were acquired with a 1.5 T scanner (Avanto Siemens, Enlargen, Germany) with the following acquisition protocol: respiratory triggered T2w BLADE sequence (TR: 2000; TE = 27; FOV 400 mm; flip angle: 150; slice thickness 5 mm) acquired on an axial and coronal plane; perfusion-weighted MRI (PWI) and Diffusion-weighted imaging (DWI). CT was performed on a 64-row scanner (Somatom Siemens, Enlargen, Germany): KV 120; mAs 20-40; collimation: 1 mm; slice thickness: 1 mm; end inspiration and end expiration. CTs and MRIs were anonymized and scored in random order by 2 independent observers using the validated CF-CT score (Brody II) and an equivalent CF-MRI scoring system. BE, TA and other structural components (airway wall thickening, atelectasis/consolidation, bullae/cysts, ground glass opacities) were scored. All scores were expressed as % of the maximal score. Mean scores of both observers were used for comparisons between CT and MRI scores. Statistical analysis:

Pearson; Intra class coefficient (ICC); Bland-Altman plots; results mean (range).

Results: ICC: CT-BE (0,931); MRI-BE (0,892); CT-TA (0,917); MRI-TA (0,677) Correlation between CT-BE score and MRI-BE score (R=0,940, p=0,0001) ; between CT-TA and MRI-TA (r=0,511, p=0,005). Bland-Altman plots showed that MRI systematically underestimated severity of BE relative to CT.

Discussion: This head to head comparison between chest-CT and chest-MRI shows that there is good agreement between observers in the identification of BE and TA both on CT as on MRI and that CT-BE and MRI-BE correlated well. However, the sensitivity of MRI to detect BE was inferior to that of CT. The measurement of TA by MRI was not reliable.

Conclusion: Chest-MRI represents a promising radiation-free alternative to CT to monitor CF lung disease. Our study demonstrated as a previous study that MRI is less sensitive compared to CT in the detection of BE and TA. We are currently testing alternative image analysis techniques to improve the sensitivity of MRI to monitor CF lung disease. Results: The survey was completed by a minimum of 58 sites per year, by the site PI at ~75% of sites and by the research coordinator at ~25% (varied slightly by year). Infant pulmonary function testing was used to monitor lung disease in 2006 at 20/58 (34%) sites and in 2008 at 25/58 (43%). In 2008, performance was routine (76%) and for clinical indications (64%, not mutually exclusive). Chest CT scans were used to monitor lung disease in 2006 at 34/58 (58%) sites and in 2008 at 50/58 (86%) sites. In 2008, CT scans were performed more for clinical indications (96%) rather than routine use (14%, not mutually exclusive). As CF newborn screening has become more widespread, cohorting for infection control purposes has increased: scheduling infants and young children on different days or in different areas increased from 5 (8%) in 2005 to 17 (29%) in 2008. In terms of treatment practices, in 2008 56/58 (97%) sites always or often treated the initial isolation of Pseudomonas (Pa) in patients ≤12 years of age independent of symptoms.

Inhaled tobramycin (98%) and oral ciprofloxacin (89%) were the most frequently prescribed antibiotics in 2008, though 36% of sites prescribed IV antibiotics and 9% oral azithromycin (not mutually exclusive). In 2008, 18/58 (31%) sites always or often treated the initial isolation of mucoid Pa. Frequently, a more aggressive or intensive regimen was used, most commonly inhaled tobramycin (89%), oral ciprofloxacin (84%), IV antibiotics (62%), and oral azithromycin (43%, not mutually exclusive).

Conclusions: Among the 58 US sites providing survey data during the EPIC study, there is substantial variability in the use of CT scans and infant lung function testing in the monitoring of early CF lung disease. Treatment with anti-pseudomonal antibiotics at first isolation of P. aeruginosa from respiratory cultures is the norm, though there is variability in the antibiotics prescribed. Further research regarding outcomes associated with surveillance infant lung functions tests and CT scans as well as the relative efficacy of different Pseudomonas eradication regimens would provide a basis for guidelines to standardize care.

A random coefficient model was created to determine average annual FEV 1 % slope CF care center (rate of decline) for patients ages 6-17 years. FEV 1 % was modeled against Center and Age (Center) with patient as a random effect (intercept and slopes).

Model 3: FEV 1 % predicted was modeled against Group and Age (Group) with random intercepts and slopes for patient and CF care center. Patients born 1988-1992 were assigned to birth cohort 1. Patients born during years 1993-1997 were assigned to birth cohort 2. FEV 1 % predicted slopes for ages 9-15 years for these birth cohorts were compared.

For each model, data from the CFF Patient Registry from 2001-2008 was used. Patients were included in the analysis if they had at least three FEV 1 % measurements. CF care centers were included in the analysis if they had at least five patients per year in the Registry for at least six of the eight years analyzed.

Results: Model 1: Slope analysis was peformed for 13,337 patients. From 2001-2008, the mean FEV 1 % predicted rate of decline was -0.95%/year (SD 5.88) with a median of -0.88%/year (IQR -3.22 to +1.22). The FEV 1 % predicted mean (+0.54) and median (+0.23) rate of decline were significantly better in patients 6-12 years of age when compared to patients 13-17 years of age (-1.82 and -2.05 respectively).

Model 2: Slope analysis was peformed for 208 CF care centers. From 2001-2008, the mean FEV 1 % rate of decline was -0.89%/year (SD 4.79) with a median of -1.05%/year (IQR -1.69; -0.48). The FEV 1 % mean (-0.68) and median (-0.07) rate of decline were significantly better for CF care centers in patients 6-12 years of age when compared to patients 13-17 years of age (-2.17 and -2.40 respectively).

Model 3: From ages 9-15 years, patients in birth cohort 2 (born from 1993-1997; n = 4,381) had a significantly better mean FEV 1 % rate of decline (-0.65%/year) when compared to patients in birth cohort 1 (born from 1988-1992; n = 4,249) (-1.36%/year) (p< .0001).

Conclusions: CF Registry data reveal improvements in the rate of decline in lung function in children over time. The current rate of decline in FEV 1 % predicted in children ages 6-12 years borders zero. The rate of decline in lung function is better in patients ages 6-12 years when compared to patients ages 13-17 years in this cross-sectional analysis. Longitudinal analysis is needed to evaluate whether there is a disparity in lung function slope between younger children with CF and these same children when they become adolescents, or whether this is a birth cohort effect.

Sawicki, G.S. 1 ; Signorovitch, J. 2 ; Latremouille-Viau, D. 2 ; von Wartburg, M. 2 ; Wu, E. 2 ; Zhang, J. 3 ; Shi, L. 4 1. Division of Respiratory Diseases, Children's Hospital Boston, Boston, MA, USA; 2. Analysis Group, Inc., Boston, MA, USA; 3. Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA; 4. Health Systems Management, Tulane University, New Orleans, LA, USA Objective: To assess the association between use of inhaled tobramycin and mortality in patients with cystic fibrosis and chronic Pseudomonas aeruginosa (PA) infection.

Methods: Patients aged ≥ 6 years with FEV1 %-predicted 25-75% and ≥ 4 cultures with PA were identified from the Cystic Fibrosis Foundation Patient Registry (1996) (1997) (1998) (1999) (2000) (2001) (2002) (2003) (2004) (2005) (2006) (2007) (2008) . Patients were followed from the first year in which all these criteria were met (index year) to death or end of continuous data availability. Reported use of inhaled tobramycin in each year (yes/no) was obtained from annual questionnaires. Longitudinal logistic regression was used to assess the association between current-year reported use of inhaled tobramycin and subsequent-year mortality, with adjustment for demographics, comorbidities, reported use of dornase alpha or high-dose ibuprofen, FEV1 %-predicted and medical resource use. Generalized estimating equations accounted for within-patient correlation across multiple follow-up years.

Results: Among 11,151 patients meeting the inclusion criteria, 2,267 deaths were observed during a median follow-up of 6 years. At index, mean age was 21 years, mean FEV1 %-predicted was 62%, 16% were underweight for age and 72% reported dornase alpha use. Inhaled tobramycin use was reported in 64% of index years and 67% of all follow-up years. After adjustment, inhaled tobramycin use was associated with a significant 15%

McPhail, G.L.; Fenchel, M.C.; VanDyke, R.; Amin, R.S.; Seid, M. Pulmonary ML 2021, University of Cincinnati, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA Background: Children with rapid rates of decline in FEV 1 % predicted are important to study. The goal of this project was to identify risk factors associated with the rate of decline in FEV 1 % predicted in this "rapid early decliner" group.

Methods: CF Registry data was used from 1994-2007 to calculate individual patient FEV 1 % predicted slopes in patients ages 6-17 years prescribed dornase alfa. Slopes were calculated from the first three years of patient data available. Patients were divided into quartiles based on their FEV 1 % slopes. Patients with extreme outlier values for slopes (the highest and lowest 5 percent) were excluded. A random coefficient model was applied, modeling subsequent FEV 1 % (after the first three modeled years) as the response variable with covariates of sex, BMI percentile, baseline FEV 1 % predicted, MRSA status, Pseudomonas aeruginosa status, hospitalizations, insurance status, and age of dornase alfa initiation (before age 6 years or at age 6 years and older). Linear terms for time-from-baseline crossed with gender were included as predictors of FEV 1 %. Linear terms for time-from-baseline crossed with Group*BaseFEV 1 % (<60, 60-79, 80-99, =>100) strata were also included as a predictors. Random intercepts and linear slopes for subject and CF care center were also included. Models were then run for subsequent FEV 1 % (after the first three modeled years) for patient ages 9-17 years.

Results: Results are reported as (mean +/-SE, p-value). Analysis was performed for 4,451 patients, with 1,113 patients in slope quartile one (rapid early decliners). P. aeruginosa infection (p = 0.05) and government insurance status (p = 0.03) were associated with lower lung function in this group but were not associated with differences in subsequent FEV 1 % predicted slope. Risk factors associated with faster rates of decline in FEV 1 % predicted were female sex (-2.5 +/-0.45%/year faster; p< 0.0001), lower BMI percentile (-0.14 +/-0.006%/year faster/centile; p< 0.0001) and number of hospitalizations (-0.64 +/-0.09% per year faster/hospitalization; p < 0.0001). Initiation of dornase alfa before age 6 years in rapid early decliners was associated with subsequent improvements in FEV 1 % predicted slope from ages 9-17 years when compared to initiation of dornase alfa at age 6 years and older. These improvements in the rates of decline in lung function were noted in each strata of baseline FEV 1 >60% predicted. Improvements in the rates of decline in lung function with initiation of dornase alfa before age 6 years were not seen in patients in the other quartiles of FEV1% predicted slopes.

Conclusions: Children with rapid early rates of decline in FEV 1 % predicted are a unique group and should be further characterized. Initiation of dornase alfa before age 6 years in children with a rapid early decline in lung function is associated with subsequent improvements in FEV 1 % predicted slope. Background: Advances in pulmonary care in cystic fibrosis in recent decades include inhaled mucolytics, inhaled antibiotics and oral anti-inflammatory therapies. Registry analysis is needed to assess whether patient outcomes have improved over time.

Methods: Model 1: CF Registry FEV 1 % predicted data was used to create random coefficient models to determine the average annual FEV 1 % patient slope (rate of decline) for patients ages 6-17 years. FEV 1 % was modeled against Subject and Age (Subject) with CF care center as a random effect (intercept and slopes). reduction in the odds of subsequent year mortality (odds ratio [OR]: 0.85, 95% confidence interval [CI]: 0.76-0.94, P=0.003). Use of dornase alpha was similarly associated with reduced mortality (OR: 0.86, CI: 0.76-0.97, P=0.016). In the same model, factors associated with increased mortality (P<0.05) included (in order of decreasing OR) cancer, renal dysfunction, Burkholderia cepacia infection, liver disease, hypertension, transplantation, underweight for age, depression, CF-related diabetes, prior hospitalizations, female gender, prior home intravenous antibiotics, earlier calendar year, lower FEV1 %-predicted and additional cultures with PA.

Conclusion: After adjusting for clinical status and other treatments, inhaled tobramycin use was associated with significantly reduced mortality among cystic fibrosis patients meeting recommended criteria for inhaled tobramycin use.

Funding provided by Novartis Pharmaceutical Corporation and The Tulane Educational Fund. (Ped Pulm 2004; 37:400-6) and FEV 1 recovery is a goal of antibiotic (ABX) treatment (Am Rev Respir Dis 1990; 141:914-21) . ABX are often administered until a satisfactory FEV 1 response is observed. However, patients can fail to completely recover lost FEV 1 (Ped Pulm 2010;45:127-34), complicating choice of treatment duration. We analyzed data from a previous CF exacerbation study to characterize FEV 1 responses to ABX over time.

Methods: Data were from a comparison of IV ceftazidime + tobramycin (C/T) and meropenem + tobramycin (M/T) for treatment of CF pulmonary exacerbations (Chest 2005; 128:2336-46) . Subjects from the blinded, randomized portion of the study who received 4+ days of ABX treatment were included. Baseline lung disease stage (early, FEV 1 > 70% pred; intermediate, FEV 1 40%-69% pred; advanced, FEV 1 < 40% pred), ABX treatments, and spirometric measures were studied. Relative FEV 1 changes from baseline on each treatment day were calculated. Peak FEV 1 change was the largest observed relative increase from baseline.

Results: Fifty C/T and 45 M/T subjects were studied. Eleven (11.6%), 46 (48.4%), and 38 (40.0%) subjects had early, intermediate, or advanced lung disease at baseline, respectively. ABX treatments averaged 12.6 days (median, 13; SD, 3.2; range, 23); 76 subjects (80%) completed ABX by 14 days and 96.8% completed by 16 days. No significant differences in mean treatment duration by ABX treatment or lung disease stage were seen. Mean time from ABX initiation to peak FEV 1 was 8.7 days (median, 10; SD, 4.0; range, 19); 2 subjects did not improve over baseline. Peak FEV 1 for 12 subjects (12.6%) was on their final treatment day, which averaged 11.5 days (SD, 3.1). Subjects were treated an average of 3.9 days after peak FEV 1 (median, 3; SD, 3.8). Peak FEV 1 for 70 subjects (73.7%) was seen by 11 treatment days; 89 (93.7%) peaked by 13 days. There was no significant difference in mean time to peak FEV 1 between C/T and M/T groups (8.9 vs. 8.5 days, respectively; P=0.62). Baseline lung disease stage impacted time to peak FEV 1 : late stage disease subjects required an average 9.9 days to reach peak FEV 1 compared to 8.0 days for those with intermediate disease (p = 0.025). Mean peak FEV 1 increase was 0.55 liters, 47.2% over baseline. M/T subjects experienced a greater average FEV 1 increase than C/T subjects (0.65 versus 0.46 liters; p = 0.033), but relative increases were not significantly different (55.9% versus 39.3%, p = 0.056). Mean peak FEV 1 increases were not significantly different across lung disease stage, ranging from 0.51 to 0.70 liters.

Conclusions: Mean times to peak FEV 1 response by ABX group were indistinguishable despite differences in peak FEV 1 magnitude, suggesting that treatment duration and response magnitude may be independent. Peak FEV 1 for 93.7% of subjects occurred on or before day 13, suggesting that 14 day study protocols would be both ethical and likely to detect true differences between ABX treatments. Finally, differences in time to peak FEV 1 associated with baseline lung function suggest that lung disease stage should be accounted for in prospective CF exacerbation study designs.

VanDevanter, D. 1 ; Yegin, A. 2 ; Elkin, E. 3 ; Pasta, D. 3 ; Konstan, M.W. 1 1. Case Western Reserve University School of Medicine, Cleveland, OH, USA; 2. Genentech, South San Francisco, CA, USA; 3. ICON Clinical Research, San Francisco, CA, USA Background: Pulmonary exacerbation (PEx) incidence as measured by IV antibiotic (ABX) treatment inversely correlates with lung-disease stage (1) . Average lung function of the CF cohort has steadily improved over recent years (2) . We therefore studied ESCF patients from 1995 to 2005 to determine trends in incidence of and the clinical criteria preceding IV-ABX treatment of PEx during the same period.

Methods: The status of four age-specific criteria (3) associated with IV-ABX treatment of PEx were collected for each IV-ABX visit and the most recent prior stable visit. Criteria studied were: increased cough (all ages), new crackles (all ages), increased sputum production (age <6 years), decreased weight-for-age (<6 years), >15% drop in FEV 1 (≥6 years), new P. aeruginosa (6-12 years) , and increased hemoptysis (≥13 years). That PEx with the highest number of PEx-associated criteria present within a calendar year was included in the analysis for patients with more than one PEx during the year. For comparison, ESCF patients who had received no ABX (by any route) for PEx in the same year were studied using the visit with the highest number of PEx-associated criteria. Overall IV-ABX treatment incidence and the distribution of exacerbation criteria were summarized for each calendar year 1995-2005.

Results: The percentage of ESCF patients treated with IV-ABX remained fairly constant from 1995 to 2005, at about 32%. As the number of age-specific treatment-associated criteria increased in patients, their probability of receiving IV-ABX increased, with 45.0% of patients presenting with 4 criteria treated versus only 8.3% with no criteria (29.4% with 3 criteria were treated, as were 15.9% with 2 criteria and 9.7% with 1 criterion). However, a majority of IV-ABX treated patients (69.3%) had only 0 or 1 age-specific criteria present, versus only 8.0% that had 3 or 4 criteria present. These relationships were reasonably constant throughout the study period. Prevalence of criteria in treated patients <6 years old were increased cough, 44.2%; increased sputum production, 38.1%; new crackles, 28.0%; decreased weight for age, 25.9%. In older treated patients, a >15% drop in FEV 1 was consistently the criterion most commonly associated with IV-ABX treatment (46.4% in 6-12 year olds, 43.6% in 13-17 year olds, and 40.3% in patients 18+ years old).

Conclusions: Despite overall improvements in lung function, the incidence of IV-ABX-treated PEx has been relatively constant in patients with CF from 1995 to 2005. As shown previously by Rabin et al., (3) the presence of 3 or more age-specific treatment-associated criteria was highly predictive of IV-ABX treatment. However, a majority of IV-ABX-treated patients had fewer than two of these age-specific criteria present. In patients ≥6 years old, a sudden drop in FEV 1 of 15% or greater was the most common criterion associated with IV-ABX treatment. Introduction: While the role of intensive care units in CF care remains a controversial subject, recent studies have highlighted survival rates of up to 76% after ICU admissions. These studies include patients who are managed with non-invasive ventilation. Data regarding patients who require endotracheal intubation and mechanical ventilation remains sparse. This study reviews the outcomes of CF patients requiring intubation in a single Background: Lung function in children with CF < 6 years of age is incompletely characterized. Sedated infant pulmonary function testing (IPFT) can be performed until about 3 years of age, and preschool spirometry is possible in cooperative children ages 3 to 6. This study investigates the agreement between measures assessed by infant and preschool lung function testing, and describes trends in lung function from infancy through preschool in a cohort of children with CF.

Methods: Children with CF participated in longitudinal studies of both IPFT (raised volume rapid thoracic compression technique) and preschool spirometry. Both studies utilized standardized equipment, rigorous site training, quality control, and an expert panel for determining acceptability of measures. Participants were scheduled for 4 IPFTs over one year and up to 5 preschool spirometry procedures over up to 2 years. Respiratory cultures were collected as part of standard care. Acceptable infant and preschool measurements of FEF25-75, FEV0.5, and FVC and corresponding z-scores, which standardize the measurements to a control population based on published reference equations, were analyzed with mixed effects models.

Results: Forty-five children had a total of 252 acceptable lung function measurements (137 IPFT, 115 preschool spirometry), obtained at ages 0.3 to 6.5 years and lengths of 57 to 125 centimeters. The mean number of measurements per participant was 6 (range: 3, 9) . The correlation between participants' z-score at the last IPFT and first spirometry was 0.4 (p = 0.02) for FEV0.5, 0.3 (p=0.03) for FEF25-75, and 0.3 (p=0.02) for FVC. The mean elapsed time between these measurements was 2.4 years (range: 0.9, 4.2). Average z-scores decreased by 0.2/year (95% CI: 0.1, 0.3) for FEV0.5, 0.1 (0, 0.2) for FEF25-75, and 0.2 (0.1, 0.3) for FVC over the observation period. However, there was significant variability among individual slopes (standard deviation of slopes was 0.2 z-scores/year for FEV0.5, 0.3 for FEF25-75, and 0.2 for FVC; range of slopes was -0.5 to 0.1 z-scores/year for FEV0.5, -0.6 to 0.4 for FEF25-75, and -0.5 to 0.1 for FVC). A recent culture positive for Pseudomonas aeruginosa (Pa) was associated with a 5% (95% CI: 0, 10) lower FEV0.5 and 16% (7, 24) lower FEF25-75 on average, after adjustment for length and test (IPFT or preschool spirometry). A recent Papositive culture was also associated with a 0.3 (95% CI: 0, 0.6) lower Zscore for FEV0. 5 and 0.8 (0.4, 1. 3) lower Z-score for FEF25-75 on average, after adjustment for age.

Conclusions: Our results suggest that lung function in infancy predicts lung function in the preschool years, and that Pa is associated with lower lung function in infants and preschoolers with CF. Our study is the first to report the individual variability in lung function trends from infancy through the preschool years among children with CF.

Supported by Cystic Fibrosis Foundation Therapeutics.

centre (2002) (2003) (2004) (2005) (2006) (2007) (2008) (2009) to determine potential prognostic markers and survival data.

Methods: A retrospective case note review of patients admitted to ICU and receiving endotracheal intubation and mechanical ventilation was undertaken. Lung function, clinical features and co-morbidity data was collected from 2 years prior to intubation. Patients who had undergone lung transplantation were excluded from the study.

Results: Thirty CF patients (34 admissions in total) were admitted to the ICU and underwent endotracheal intubation. The main indications for intubation were haemoptysis, pneumothorax or infective exacerbation. Mean FEV1 % predicted on intubation was 28%. Eleven patients died on ICU (37% mortality) and 7 after ICU but before hospital discharge (overall mortality 52.9%). This accounted for 7.9% of deaths in the CF unit over the time period. For patients discharged home after ICU admission, mean survival was 447 days.

There were no significant relationships between ICU survival and reason for admission, lung function (nor decline in lung function over preceding 2 years), sputum microbiology or frequency of infective exacerbations. Conventional ICU admission scoring systems were not significantly different between groups. However, in non-survivors there was a significantly higher prevalence of osteoporosis and a greater drop in BMI over the preceding 24 months.

Conclusions: This study, examining only patients undergoing endotracheal intubation, shows similar mortality results to other studies of CF patients on intensive care units. It highlights that neither current lung function nor rate of decline in lung function should be regarded as a bar to invasive ventilation and that other indicators of chronic disease such as preceding changes in BMI or co-existent osteoporosis may be more relevant prognostic markers.

Background: Allergic bronchopulmonary aspergillosis (ABPA), a dual type hypersensitivity reaction with specific Aspergillus IgE and IgG antibodies, causes frequent exacerbations and deterioration in lung function in approximately 10% of CF patients. Oral glucocorticosteroids is gold standard therapy but associated with severe side-effects especially in children at growth. High-dose iv pulse methylprednisolone (HDIVPM) has previously been proven effective in interstitial lung diseases as well as cases of ABPA.

Methods: Patients with CF and ABPA (according to CFF Consensus criteria) were treated with pulse methylprednisolone intravenously (15 mg/kg/day) for 3 consecutive days per month, in combination with an oral azole (itraconazole, voriconazole or posaconazole) until clinical and laboratory resolution of ABPA. Study period: January 2007-March 2010 (39 months).

Results: Of a total of 301 patients (181 adults and 120 children) in the Copenhagen CF Center 25 patients (9F/16M, average age 19.9 (10-34)) with CF and ABPA were treated with HDIVPM. Pulmonary symptoms, lung function and total IgE were used to monitor treatment response. No serious side effects were observed. Treatment was discontinued after 3-7 pulses (average 3.5) (1 patient stopped after 1 month).

For duration of treatment and time to exacerbation in the 25 patients, see Table. The number of ABPA exacerbations were: 1 in 11 (44%) patients, 2 in 10 (40%) patients, 3 in 3 (12%) and 4 in 1 (4%) patient. The majority of the patients (>80%) had radiological changes in the upper lobes of the lungs.

Conclusion: ABPA was successfully treated with HDIVPM as alternative treatment to systemic glucocorticosteroids. Generally, the treatment postpones the next exacerbation of ABPA at least close to a year. Only few and transient side-effects were reported. Serious long-term side effects especially in children at growth remain to be evaluated. Just above half of the CF patients who experience ABPA are likely to have another exacerbation within a 3-year period. A comparative study with systemic glucocorticosteroids is needed. 9 1. University of Rochester, Rochester, NY, USA; 2. Children's Hospital, Aurora, CO, USA; 3. University of Washington, Seattle, WA, USA; 4. Children's Hospital of Philadelphia, Philadelphia, PA, USA; 5. Nationwide Children's Hospital, Columbus, OH, USA; 6. Rainbow Babies Hospital, Cleveland, OH, USA; 7. Texas Children's Hospital, Houston, TX, USA; 8. CF TDN Coordinating Center, Seattle, WA, USA; 9. University of North Carolina, Chapel Hill, NC, USA Background: We recently completed a multicenter longitudinal observational cohort study of lung function in preschool children with CF. The goal of the present analysis was to assess the associations between clinical characteristics and lung function at the enrollment visit.

Methods: Spirometry, forced oscillometry (FO), respiratory inductive plethysmography (IP), and clinical data from the enrollment visit were used for this analysis. Observed values were converted to Z scores using published normal reference equations. Spirometry (FEV0.5, FEF25-75), FO (R8, X8), and IP (phase angle and phase RTB) measures were compared between subjects with different clinical characteristics.

Results: Ninety-three total subjects were enrolled at 8 sites; research quality data at enrollment were available from 48 subjects for spirometry measurements, 33 for FO, and 40 for IP. We were unable to detect a difference in the distribution of any PFT parameter based on Pseudomonas status, CFTR mutation class, Wisconsin cough score, Shwachman score, environmental tobacco smoke exposure, family history of asthma, or nutritional indices.

Conclusions: We previously reported that spirometry distinguishes CF patients from healthy controls in this cohort (Pediatr Pulmonol S32:293) . In this analysis we did not find any association between lung function at enrollment and selected clinical characterics. This observation may reflect the relatively mild reductions in lung function in our study cohort as well as the limited sample size. Planned longitudinal analysis of the association between the change in lung function and clinical characteristics in this cohort may provide additional insight. Rotterdam, Netherlands; 2. Radiology, Erasmus MC, Rotterdam, Netherlands; 3. Biostatistics, Erasmus MC, Rotterdam, Netherlands Introduction: Close monitoring of young CF patients is necessary to optimize treatment and to prevent lung damage. For children aged 0-4 years, currently used routine lung function tests are often cumbersome and variable. It is therefore important to develop alternative methods to assess CF lung disease.

Lung Clearance Index (LCI), nightly oxygen saturation (SpO 2 ), and/or cough audiometry might be valuable tools in the monitoring young CFpatients.

Aim: To compare LCI, overnight SpO 2 and cough in CF patients aged 0-4 years with that in a group of healthy controls.

Methods: Prospective cross sectional study. All children with CF were recruited from the outpatient clinic of the Erasmus MC -Sophia Children's Hospital. Healthy children were recruited from child day care facilities. LCI was measured by the Exhalyzer®D at the outpatient clinic. Nightly oxygen saturation was measured by the Novametrix Model 2001 MARS pulse oximeter and nightly cough was measured with an audiometer. Both were measured at home during a normal night sleep. For cough, the cough seconds per hour (cough sec./hr) were calculated: the total number of cough seconds devided by the total hours of registered sleep.

Results: Twenty children with CF and 30 healthy children aged 0-4 years were included. Mean age of children in the study was 2.3 years. Age, sex, height and weight were not significantly different between both groups.

LCI measurement was measured succesfully in 15 out of 30 healthy children (50%) and in 15 out of 20 CF patients (75%).

LCI was significantly higher in CF patients than in healthy controls: mean LCI was 9.26 in the CF patients and 7.06 in the healthy children (p < 0.001). For nightly SpO 2 and for nightly cough no significant differences were found. Median SpO 2 was 97.0 in children with CF and 97.7 in healthy children. Median cough sec/hr was 0.20 in children with CF and 0.45 in healthy children.

Conclusions: LCI showed a significant difference between healthy children and CF patients at a young age. Our study supports previous studies that suggest LCI is possibly a sensitive test to detect pulmonary changes in early stage lung disease. Further validation against chest computer tomography in our cohort is ongoing.

Sullivan, S.M. 1 ; Powers, S.W. 1 ; Brody, A.S. 2 1. Behavioral Medicine and Clinical Psychology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA; 2. Radiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA Introduction: The toddler and preschool years are a challenging time in development to evaluate CF lung disease. At the same time, there is great interest in studying new interventions in these younger children with less advanced CF lung disease. CT scanning has been shown to be a sensitive method for evaluating CF lung disease and CT scans of young children have revealed evidence of lung disease earlier than x-rays or PFTs. Yet, limited data from CT in CF exist on the prevalence of morphologic abnormalities in this age range. We examined the prevalence and severity of lung abnormalities in a group of 2-6 year olds with CF.

Methods: Subjects were recruited from a larger behavioral nutrition clinical trial of 2-6 year olds with CF. Entry criteria for the larger study allowed for an inclusive sample of participants. Subjects positive and negative for Pseudomonas infection and with weight for age z-scores up to +1.0 were included. All subjects were determined pancreatic insufficient on fecal elastase-1. Entry criteria excluded very few of the children in the age range. After obtaining separate consent for the CT substudy, CT scans were performed near the time of the baseline visit for the larger trial. CT scans were performed under general anesthesia (GA) (ages 2-5) or with voluntary lung volume control (ages 5-6). The method for 5 year olds depended on the level of cooperation with breathing instructions. Forty-two CT scans were obtained at 2 sites. Each subject's CT scan was evaluated for the presence/absence of each of 7 features of CF lung disease (bronchiectasis, peribronchial thickening, mucous plugging, air trapping, ground glass, opacity, and consolidation) in the 6 lobes (including the lingula as a lobe).

Results: Mean ± SD age was 44.1 ± 15.9 months, range 25.6-74.1 months. 40% of subjects were male and 69% homozygous for ∆F508 genotype. A Pearson correlation was conducted and the relationship between age and total abnormalities was not significant, r (42) = .11, P = ns. Percent of children with bronchiectasis by year of life was 2=31% (N=16), 3=37% (N=11), 4=50% (N=4), 5=44% (N=9), 6=0% (N=2).

Discussion: CT scanning has been shown to be a sensitive method for evaluating CF lung disease. Many investigators believe motion free images at TLC (total lung capacity) and near RV (residual volume) are important to obtain accurate information. This has often led to limiting imaging trials to children 6 years old or older, when cooperation with breathing instructions becomes possible. We have previously shown that GA in children ages 2-5 is a safe and effective way to obtain high-quality CT images. In this crosssectional study we used GA in 2-5 year olds and voluntary lung volume control in 5-6 year olds to evaluate the presence and severity of CF lung abnormalities on CT. No differences in lung disease across ages were demonstrated. This finding has been reported by others, and suggests that CF lung disease may progress relatively slowly in this age range. A longitudinal study is ongoing to further characterize lung disease in toddlers and preschoolers with CF.

Supported by NIH Grant R01 DK054915 and CFF (POWERS05A0). 31, 2010. Nasopharyngeal (NP) swabs were obtained on all enrolled patients and tested by direct immunofluorescence (IF) for respiratory syncytial virus (RSV) A and B, influenza A and B, human metapneumovirus, parainfluenza 1-3 and adenovirus A and B, as well as by a multiplex molecular diagnostic assay (ResPlex II v2, Qiagen) that detects the above plus rhinovirus, coxsackie/echovirus, parainfluenza 4 and coronavirus. Demographic data, microbiological results, pulmonary function tests, antibiotic use, hospitalizations, quality of life (CF Questionnaire-Revised) and severity scores (CF Clinical Score) were recorded. Sputum inflammatory markers, cell counts and bacterial density were also measured. Results: Forty-three patients were enrolled and had NP swabs for viral testing by IF and multiplex assay. Nineteen were negative for any virus and 24 were positive for at least 1 virus (10 positive by IF and multiplex; 14 positive by multiplex alone). In 9 of 24 patients (37%), more than 1 virus was detected. The most common viruses detected were RSV (28% of specimens), coxsackie/echovirus (22%), rhinovirus (13%), parainfluenza virus (10%) and pandemic H1N1 influenza virus (9%). Baseline patient characteristics demonstrated that virus-positive patients were more likely to be younger than virus-negative patients (7.8 yrs vs 10.8 yrs, p=0.05) and male (p=0.03). Other baseline characteristics including genotype, comorbidities and colonizing respiratory flora were similar between both groups. A greater percentage of virus-negative compared to virus-positive patients received the seasonal (31% vs 25%) and the pandemic H1N1 influenza vaccine (63% vs 37%). At the time of pulmonary exacerbation, a greater percentage of virus-positive compared to virus-negative patients received antibiotics (79% vs 68%) and were hospitalized (33% vs 16%), but this was not statistically significant. At exacerbation, virus-positive patients had a lower mean FEV1% predicted (58.8 vs 69.4, p=0.17), higher severity score (27 vs 23, p=0.02) and lower quality of life (physical) score (54 vs 74, p=0.02) compared to virus-negative patients. The severity score inversely correlated with the quality of life score (p=0.0014). Sputum producing virus-positive patients had similar levels of 27 sputum inflammatory markers (including IL-8, 6, 10 and TNF-α) and sputum neutrophil counts (99.45% vs 99.5%) compared to virus-negative patients despite lower sputum bacterial densities (3x10 7 vs 1x10 9 CFUs/ml).

Conclusions: More than half of pulmonary exacerbations are associated with a respiratory virus (most commonly RSV and coxsackie/echovirus). Molecular techniques enable the detection of more viruses than immunofluorescence. Pulmonary exacerbations with viruses are associated with higher severity scores, lower quality of life (physical) scores and similar levels of pulmonary inflammation compared to exacerbations without viruses.

Kuhn, R.J. 1 ; Wertz, D.A. 2 ; Zhang, J. 3 ; Chang, C. 2 ; Stephenson, J.J. 2 1. University of Kentucky School of Pharmacy, Lexington, KY, USA; 2. HealthCore, Wilmington, DE, USA; 3. Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA Background: CF guidelines recommend chronic use of TSI to improve lung function in patients with moderate-to-severe lung disease and persistent Pseudomonas aeruginosa in airways. This study evaluated the cost impact of TSI in CF patients in a managed care population.

Methods: CF patients, aged 0 to 64 years, were identified from the HealthCore Integrated Research Database. Patients were required to have ≥ 2 claims of CF diagnosis (ICD-9=277.0x) between 01/01/04 and 3/31/08. TSI users were categorized by the number of TSI filled in the 12-month post-index period as follows: low users (<2 fills), medium users (2 to <4 fills) and high users (≥4 fills). For TSI users, the first claim for TSI between 03/01/04 and 3/31/07 was considered the index date; for non-users, the index date was assigned based on the distribution of the duration between eligibility start and index date in the TSI user group. Patients were excluded if they had a history of lung transplant or <3 months pre-index and <12 months post-index continuous eligibility. Change of CF-related and allcause per member per month (PMPM) costs from pre-index to post-index was analyzed using paired t-tests.

Results: In all, 832 patients met the study criteria (444 non-users/388 users). Mean age was 19 years for users and 30 years for non-users with approximately 50% female. TSI users PMPM total healthcare cost The 2009 CFF infant care guidelines recommend breastmilk as the initial type of feeding for CF infants. However, our previous analysis showed that CF infants exclusively breastfed longer than one month achieved lower weight and length z-scores during the first 2 years of life compared to those exclusively formula-fed (NACFC 2008) . Despite attenuated growth, breastfeeding may provide pulmonary benefits compared to formula-feeding.

Objective: To characterize infection patterns of Pseudomonas aeruginosa (PA) and Staphylococcus aureus (SA), the two most common pathogens found in young children with CF, and the associations of PA and SA with breastfeeding during the first 2 years of life in infants diagnosed through the Routine Wisconsin Newborn Screening (NBS) program.

Methods: The study population included 70 CF infants enrolled in the Wisconsin Routine CF NBS Study who were born during 1994-2006, had no meconium ileus (MI) and no known mutations associated with pancreatic sufficiency. Infant feeding and infections data from diagnosis (range 0-2.4 wk) to age 2 years were obtained through extensive medical record review and the CFF Patient Registry. The infants were first classified into 2 feeding groups: ever breastfed (BF, n=34) and never breastfed, i.e., exclusively formula-fed (FM, n=36). The BF group was further classified into exclusively breastfed <1 month (ExBF1mo-, n=16) and exclusively breastfed ≥1 mo (ExBF1mo+, n=18).

Results: Forty-seven percent (n=16) of the BF group had at least one positive PA compared to 58% (n=21) in the FM group. The prevalence of PA infections in the BF group ranged from 3% at age 3 mo, 6-9% at ages 6-12 mo, and 9-15% at ages 15-24 mo, which were lower than the FM group (5% at age 3 mo, 19-26% at ages 6-12 mo, 9-29% at ages 15-24 mo). Consistent with this observation, logistic regression demonstrated that the BF group was associated with a significantly lower likelihood of positive PA compared to the FM group (OR=0.38, 95% CI=0.22, 0.65; p=0.011). Additionally, the cumulative number of PA infections were significantly lower in BF infants (53% never positive, 35% one positive, and 12% two or more positive) compared to FM infants (42% never positive, 19% one positive, and 39% two or more positive), p=0.027. Further analysis separating the BF group by the duration of exclusive breastfeeding revealed that the BF1mo+ group had the fewest PA infections, 61% never, 39% once, and none with two or more, followed by the BF1mo-group (44% never, 31% once, and 25% two or more), both were better than the FM group, p=0.019. The prevalence of SA infections during the first two years of life was higher than PA infections at all ages (range 31-43%), but no significant differences were observed between BF and FM groups.

Conclusion: Breastfeeding is associated with fewer PA but not SA infections during the first two years of life compared to formula-feeding. Exclusive breastfeeding longer than one month is associated with reduced growth (NACFC 2008) but fewer PA infections compared to exclusive formula feeding.

Supported by NIH R01 DK072126. Objectives: To determine if respiratory viruses, detected by molecular techniques, increase the severity of pulmonary exacerbations in pediatric cystic fibrosis (CF) patients.

Methods: This was a single-centre, observational, cross-sectional prospective study of children with CF who presented to the Hospital for Sick Children, Toronto with pulmonary exacerbation November 1, 2009-March decreased $959 after starting TSI therapy over the 12-month follow up period compared to an increase PMPM cost of $77 of non TSI users. TSI users CF-related PMPM cost decreased $113 compared to an increase of $552 for non-users. The biggest change was seen with CF-related inpatient cost with a decrease of $1171 PMPM (P=0.01) in TSI users compared to an increase of $480 PMPM in non-users. CF-related prescription cost increased $992 PMPM in the TSI group while non-users increased $25 PMPM. Both TSI medium users and high users had a greater decrease in PMPM CF-related inpatient cost than low users, The decrease of CF-related inpatient PMPM costs were $381 for low users, $1,425 for mid users and $1,829 (P=0.02) for high users.

Conclusion: Both total and CF-related PMPM cost decreased after initiation of TSI therapy. This effect was attributed to a larger decrease in inpatient cost compared to increase in prescription cost. A trend of greater inpatient PMPM cost decrease was observed with better TSI adherence. Background: Angiogenesis is an important mechanism of airway remodeling in pulmonary disease. We previously demonstrated that serum VEGF is elevated in CF and declines with therapy for pulmonary exacerbation. We hypothesized that VEGF is elevated early in the course of CF, is independent of inflammation and is associated with tissue level hypoxia.

Methods: Thirty stable infants and children with CF were recruited. Serum was analyzed for VEGF and for other markers of tissue hypoxia (erythropoietin (EPO), insulin-like growth factor binding protein-1 (IGFBP-1)) and for inflammatory mediators (IL-1 beta, IL-6, IL-8, and tumor necrosis factor alpha (TNF alpha)) using Meso Scale multi-spot serum immunoassays. This technique utilizes antibodies with an electrochemiluminescence tag to measure protein levels and allows for measurement of proteins over a linear range of 0.1-10,000 pg/10µL sample. Measurements were correlated between assay groups; and with age in months and lung function (FEV0.5 or FEV1). Pearson correlations were performed for normally distributed variables and Spearman correlations were used for nonparametric distributions.

Results: Twenty patients aged 6-12 years and 10 patients <3 years of age were recruited. For the entire population VEGF, EPO, TNF alpha and IL-8 were elevated compared to published normative values. VEGF levels were not significantly correlated with any inflammatory mediators. However, VEGF correlated with EPO (r=0.414; p<0.05). Inflammatory mediators were correlated as follows: IL-6 and TNF alpha (r=0.574; p<0.05), IL-8 and IL-1 beta (r=0.456; p<0.05). IGFBP-1 correlated with TNF alpha and IL-1 beta (r=0.468; p<0.05 and r= 0.395; p<0.05 respectively) and with age in months (r= -0.428; p<0.05). There was no correlation between lung function and markers of inflammation or tissue hypoxia.

Conclusions: VEGF is elevated in young, stable infants and children suggesting angiogenesis as a contributing mechanism for early lung disease in CF. VEGF levels are not associated with inflammatory mediators known to be important in CF, but rather associated with EPO levels. We propose that VEGF elevation and angiogenesis may result from a direct tissue hypoxia pathway in CF. Inhaled aerosol drugs can deliver substantial doses of medication directly to the lungs while minimizing the systemic concentrations that are often associated with side effects. Antibiotics are often administered through this route for the treatment of the pulmonary infections associated with cystic fibrosis. The altered patterns of ventilation associated with this and other lung diseases can cause inhaled drugs to deposit non-uniformly so that some lung regions receive very high local doses of medication while other regions go untreated. In an infected lung, this could result in a reservoir of infection that is never effectively treated by the therapy. Preliminary studies suggested that the addition of surfactant to the aerosol formulation may improve the uniformity of drug distribution by creating surface tension driven or Marangoni flows that will disperse the formulation over the surface of the airways after aerosol deposition.

To determine the optimal characteristics of surfactant formulations to be used in aerosol drug vehicles, we have investigated the spreading of aqueous solutions of soluble surfactants over entangled aqueous mucin solutions that mimic the airway surface liquid within the lung. The surfactant solutions contain dyes that not only serve as tracers but also as drug mimics, modeling the use of surfactant formulations as self-dispersing drug carriers for pulmonary drug delivery. Anionic, cationic and nonionic surfactants were compared in order to determine the potential impact of surfactant-mucin binding on the extent of spreading. For all surfactants examined, we find that Marangoni stresses initiate flow over the rheologically complex mucin solutions and that the convective spreading terminates with the spread solution confined to a well-defined static area. Further motion of the spread solution progresses by much slower diffusive processes. Similar behaviors were observed on entangled aqueous polyacrylamide solutions, despite the significant difference in chemical complexity and surfactant-binding tendencies of the two macromolecules. The independence of spreading behaviors on the chemistry of the surfactant and subphase, as well as the solubility of the tracer dye, indicates that the spreading process is controlled by capillary and hydrodynamic phenomena and not specific chemical interactions among the components of the system, indicating that optimal formulations can be designed based on those fluid phenomena alone. We find that the final extent of spreading -the key parameter is determining the efficacy of a self-dispersing formulation -is not controlled by the same surface tension gradients that drive the initial spreading. Rather, this area is optimized by choosing surfactants that remain confined to the spreading solvent layer and formulations that inhibit the movement of solvent into the mucus subphase.

We acknowledge grants from NIH 5P30DK072506, NSF CBET-0931057, and Cystic Fibrosis Foundation Therapeutics CORCOR07A0 for support of this work.

Sheikh, S.; Quake, J.; McCoy, K. Pulmonary Medicine -Pediatrics, Ohio State University College of Medicine/Nationwide Children's Hospital, Columbus, OH, USA Background: Patients with cystic fibrosis (CF) have progressive lung disease. GER is common in patients with CF and can contribute to respiratory symptoms.

Objective: To evaluate efficacy of surgical management of GERD (Nissen fundoplication) in patients with CF and GERD and with worsening respiratory symptoms who fail medical therapy of GERD. Improvement, if any, was quantified by comparing changes in lung function and pulmonary disease activity within two years before and after surgical management of GERD in children and adults with CF.

Methods: We retrospectively reviewed charts of all patients with CF and GERD followed at the CF Center at Nationwide Children's Hospital, Columbus, OH, and, who underwent Nissen fundoplication over the last 20 years . Patients who were followed for at least one year before and after their surgery were included in this study (n=48). We compared number of hospital admissions, emergency room visits, pulmonary clinic visits, pulmonary exacerbations requiring intravenous antibiotics and/or oral/nebulized antibiotics, weight and PFT. PFT measurements done within one month before and/or after exacerbations were not included in the study and % predicted values are compared. Statistical Analysis: Outcomes were compared using non parametric tests, Wilcoxon signed ranks test. A p-value of < 0.05 was considered significant.

Results: Among our cohort of CF patients (n=495), 70 patients with GERD underwent Nissen fundoplication in the last 20 years. Forty-eight patients were included in this study. Median age at the time of surgery was 14 years (range 1-36), adult 19: children 29, male 23: female 25. Paired t- Rationale: Multiple variables impact outcomes in cystic fibrosis including genotype and modifier genes, microbiology, nutritional status, response to therapies, environmental exposures, socioeconomic factors and adherence to prescribed regimens. Our inability to measure adherence to prescribed medications and therapies impairs our ability to evaluate the effects of specific interventions. No objective measurements of adherence are available, and we rely on subjective patient and parent histories for its assessment. We aimed to compare the number of dispensed doses of dornase alfa with the number of prescribed doses using primary insurance prescription records, and to seek any possible relationship between dornase alfa adherence and clinical outcomes.

Objectives: To utilize primary insurance records to calculate percent adherence to prescribed dornase alfa among pediatric CF patients at our center in 2009, and to explore for associations between adherence and multiple clinical outcomes (including pulmonary function, BMI, treatment with antibiotics, and hospitalizations).

Methods: Prescription information for dornase alfa was obtained from 3 major primary insurance companies (Highmark, BlueCross, and Medicaid) for 42 CF patients for whom dornase alfa was prescribed at our center in 2009. Percent adherence was calculated for each patient by dividing the number of dornase alfa doses actually dispensed by the total number of doses prescribed in 2009. Patients with percent adherence at or above 70% were compared to those with adherence less than 70%. Outcomes measured included the year's best percent predicted FEV1, the year's best BMI percentile, number of days of oral and IV antibiotics, number of hospitalizations, and number of hospital days.

Measurements and Main Results: There were 20/42 patients with a percent adherence at or above 70%. Compared to those with adherence less than 70%, the more adherent group was more likely to have a better percent predicted FEV1 (OR 5.41, CI 0.98 to 32.73), significantly fewer hospital days (OR 10.23, CI 1.05 to 484.69), and significantly fewer days of IV antibiotics (OR 4.89, CI 1.01 to 26.53). There was no significant difference seen in age, gender, BMI, number of hospitalizations, number of oral antibiotic courses, or number of days on oral antibiotics.

We were able to demonstrate that the level of adherence to dornase alfa can be partially determined by using primary insurance records, and that the regular use of dornase alfa is associated with key clinical outcomes such as pulmonary function, days of hospitalization, and days of IV antibiotics. The study is unable to measure actual adherence as we cannot be sure that the doses were actually administered once dispensed from the pharmacy. We speculate that improved adherence to dornase alfa and other therapies may greatly improve outcomes in CF, and that the percent adherence to prescribed dornase alfa may prove to be a useful surrogate marker for the level of overall adherence to prescribed therapies. Effective modes of improving adherence in CF patients need to be developed and validated. Background: A growing number of adolescent CF patients are receiving courses of intravenous (IV) antibiotics at home as treatment for pulmonary exacerbations. There are limited studies evaluating patient-reported outcomes in this setting. At the Pediatric CF program at Children's Hospital Boston, patients are admitted for 3-5 days and discharged home to complete their therapies if deemed eligible for home-based IV therapy by their care team.

tests showed significant improvement in mean weight, in kg, at one year (41±17.5) and 2 year (43.6±15.8) after Nissen (p<0.05 for each) when compared to weight at Nissen (40 ± 18) and at one year (38.5±19) or two years (38±19.6) before Nissen. There was also a significant decrease in CF exacerbations requiring intravenous antibiotics during the year after surgery (1.38±1.4) compared to one year before surgery (2.7±1.5) but no difference was evident by the end of year (1.7±2). PFT were compared using Friedman tests. Compared to PFT at one year before Nissen, measured as % predicted FVC, FEV1 and FEV1/FVC (81.4±18.9, 69±24.2, 73.6 ±16.6 respectively), PFT at the time of surgery revealed significant deterioration in all three tests (75. 6 ± 17.8, 61±24.3, 70.7±17.6) according to Friedman tests (p<0.05 for all). PFT measured as % predicted FVC, FEV1 and FEV1/FVC at one year after surgery (79.6±19.6, 66.4±23.6, 74.6±15.2 respectively), when compared to PFT at surgery, revealed significant improvement in all of these tests (p<0.05) but improvement was lost by the end of second year (74.9±20.6, 57.2±27, 67.9±17.6). There was no significant improvement in any other measures.

Conclusion: In patients with CF and GERD who fail medical management of GERD and develop progressively worsening lung disease, surgical management of GERD (Nissen fundoplication) not only decreased decline in lung function but led to significant improvement in weight, PFT and a significant decrease in need for intravenous antibiotics for at least one year after surgery. Background: HFCC is an effective therapy for CF. There is, however, little knowledge regarding what frequency (freq) setting provides optimal clearance. Impulse oscillometry (IOS) is a validated technique for assessing airflow obstruction in children with obstructive lung diseases. One measurement obtained by IOS is the resonant frequency (f res ), the freq at which reactance equals zero. Resonance is the tendency of a system to oscillate at larger amplitude at some freqs (f res ) than at others. At f res , small periodic driving forces can produce large amplitude oscillations.

Objectives:

To evaluate whether f res can be used to optimize settings of HFCC vests for pediatric CF patients and to characterize: f res , the HFCC freq used by subjects (f used ), and the HFCC freq that generated the greatest volume (f vol ) and flow (f flow ) changes, measured at the mouth.

Methods: We studied CF patients 6-21 years old who used HFCC routinely and were clinically stable. All subjects underwent IOS for three 30sec measurements without obvious disturbances. F res for all 3 trials was averaged. Subjects who also elected to undergo HFCC frequency tuning wore an inCourage (RespirTech) HFCC vest while breathing tidally through a pneumotachometer to assess flow and volume changes. Data was collected for 30 sec at 13 different frequencies (6Hz-30Hz, in 2 Hz increments) using a digital signal-processing program that sampled flow and volume at 200Hz. This data was processed using cumulative trapezoidal integration and a high-pass filter. Spectral analysis showed peaks corresponding to the freq of imposed pressure variations. A band-pass filter was then used to extract the amplitude of flow and volume harmonics, which allowed determination of f flow and f vol for each subject. The relationships among f res , f vol , f flow and f used were analyzed using Pearson's correlation.

Results: Forty-five subjects (51% male, median age 12 years) were enrolled with 20 subjects undergoing HFCC tuning. F used data was available for 87% subjects, of which 18% reported using more than one freq and 82% using a single freq. Median f used was 14Hz (range, 6-30 Hz). Median f res was 20.23 Hz (range, 7.85-33.65 Hz). For subjects undergoing frequency tuning, median f flow was 8Hz (range, 6-30 Hz) and f vol was 26Hz (range, 8-30Hz). There was no correlation between f res and f flow (r2=0.19) or f vol (r2=0.023). Similarly, there was no correlation between f used and: f res (r2=0.045), f flow (r2=0.12), or f vol (r2=0.031).

Conclusions: f vol and f flow typically occurred at frequencies not used by most patients at home, and could not be predicted by f res as measured by IOS. Frequency tuning may be required to optimize clinical utility of HFCC.

We prospectively studied CF patients who received IV antibiotic therapy for pulmonary exacerbations. Participants completed the CFQR (Cystic Fibrosis Questionnaire-Revised) and the MSAS (Memorial Symptom Assessment Scale) prior to discharge and at clinic follow-up. Participants also completed a daily treatment diary checklist during therapy and a satisfaction survey at follow-up visit. Outcomes evaluated included the changes in CFQ-R and MSAS Scores, FEV1 between pre-and post-therapy, and patient-reported satisfaction of the entire home IV process.

Results: Over an 8-month period, 16 patients (mean age 16.4 ±3.7) with a mean pre-admit FEV1 (68 ± 21% pred) participated in the study. The mean LOS and length of therapy were 4.3±1.3 and 15.1±3.1 days, respectively. Participants completed a total of 91 days of the treatment diary and reported 68% (62 days) of the time spending greater than 120 minutes per day to complete all IV related treatments. There were no baseline differences in lung function, symptom burden or QOL based on age, gender or admission lung function. Change in lung function positively correlated with improvements in respiratory QOL scores (r=0.31) and health perception QOL scores (r=0.27). There were decreases from baseline to follow-up in the mean symptom burden scores for: coughing, weight loss, pain, vomiting, worrying, sinus discharge and/or pain, and difficult sleeping. The preparation for home IV therapy by health care providers and home care companies were rated as excellent or very good by 94% and 86%, respectively. Finally, 94% of participants reported feeling comfortable or very comfortable with the overall home IV therapy process.

Conclusion: In our cohort, home-based IV antibiotic therapy for CF pulmonary led to improvement in lung function, symptom burden and quality of life. Frequently patients reported spending greater than 2 hours on IVrelated therapies. However, patient-family satisfaction with home-based IV therapy is high.

Scotten, M.S. Pediatrics, university of kansas medical center, Kansas City, KS, USA Background: The use of telemonitoring is emerging as a cost effective and safe way for medical teams to follow their patients between office and hospital visits. Home lung monitors, measuring FEV1, have been developed and been shown to be effective in providing useful data on the state of pulmonary health via remote access. Studies using home spirometry for tracking chronic pulmonary conditions such as chronic obstructive pulmonary disease and asthma are numerous throughout the literature. There are virtually no studies, however, assessing the utility and feasibility of using home spirometry in pediatric patients with CF.

Methods: This is a twelve month study following seven patients, ages eight to thirteen, with CF. We are reporting on phase one of a larger study examining the use of home telemonitoring at our CF Center. Each patient received a home spirometer and was asked to record daily FEV1 readings after their evening respiratory treatment. Results were then downloaded via telephone or computer to a secure website. Clinicians and families have 24 hour access to FEV1 data and study participants can communicate directly with the CF Center for questions regarding the use of the home spirometry device. Feasibility of home monitor use has been assessed by length of time from when families received devices to when data was downloaded and via patient satisfaction surveys. Adherence to device use has been monitored by recording the number of days with FEV1 recordings.

Results: All 7 patients enrolled in the study recorded FEV1 data which have been downloaded to the study website. Average time from study enrollment and device distribution to downloading data to the website was 21.8 days (range: 8-60). Families reported high satisfaction with using the device and accessing support from the help desk. Families reported that use of the device assists them in knowing how sick their child is and that use of the device is quick and does not greatly extend treatment time. Several families have also stated that they use the recorded numbers in a "game fashion" to encourage their children to do a better job with respiratory treatments. Adherence to daily device readings was calculated by using the percentage of days with recorded FEV1 data. Average percentage of days with recorded FEV1 data in our sample was 58% (range: 3-92%).

Conclusions: Implementation of a home telemonitoring program is a feasible endeavor in the pediatric CF population. The majority of families find that use of a home spirometer fits in easily with their respiratory treatment regimen. Barriers to adherence need further investigation. Background: Cumulative exposure to intravenous aminoglycosides along with comorbid conditions (e.g. diabetes, hypertension) that have emerged with improved survival of patients in CF have contributed to the development of chronic kidney disease (CKD) in this population. The goals of this study are to determine 1) the prevalence and relative contribution of these risk factors to development of CKD, and 2) the level of agreement among 3 commonly used equations to estimate GFR (glomerular filtration rate).

Methods: All patients (n=157) attending the adult CF clinic were evaluated by review of medical charts to screen for CKD. Pertinent laboratory studies including urinary albumin/creatinine, serum creatinine, C-reactive protein (CRP), and cystatin c (CysC) which were measured routinely during annual evaluation visit, comorbid conditions, and medication history were obtained from review of medical charts. CKD was defined by the presence of proteinuria and/or estimated GFR using the MDRD (Modified Diet in Renal Disease) equation according to the NKF KDOQI guidelines. GFR was also estimated using the CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) and with an equation incorporating serum cystatin c (CysC). The degree of agreement between the 3 predictive equations was assessed based on their ability to classify patients into the GFR ranges defined by the NKF KDOQI guidelines.

Results: The overall prevalence of CKD in this adult CF population (n=157) was 12.1% (19/157); stage 1 (14/19), stage 2 (2/19) and stage 3 (3/19 ). The population was characterized by mean age of 30 y (±8.7), cystic fibrosis related diabetes (CFRD) in 38%, history of nephrolithiasis or IgA nephropathy in 5%, and lung transplant in 5%, and hypertension in 2.5% of patients. CFRD (86.7% vs. 31.5%; p<0.01), underlying kidney pathology (15.8% vs. 2.9%; p=0.04), and lung transplant recipient (15.8% vs. 2.9%; p=0.04) were identified as significant risk factors for CKD by univariate analyses . Mean GFR estimates using the MDRD (114±30), CKD EPI (117±19), and CysC (123±33) did not differ among the 3 predictive equations. As expected there was reasonable agreement (Kappa coefficient=0.64) between the two creatinine-based equations (MDRD and CKD EPI), however, there was relatively poor agreement between the creatinine-based and cystatin C equations (Kappa coefficient=0.22 and 0.25) when used to classify patients into guideline-specific GFR ranges. The ability to discern agreement between equations is limited by the small number of patients with more significant kidney disease (stages 2 and 3). In contrast to a previous report we found no significant correlation between serum CysC and CRP (r=0.04, p=0.68).

Conclusions: CKD is relatively common in adult patients with CF. Although CFRD appears to be the principal cause of CKD in CF, annual screening for proteinuria may identify additional patients with CKD. Additional studies are needed to determine which predictive equation provides the greatest diagnostic accuracy in identifying early CKD. Background: Chronic P. aeruginosa infection causes an intense inflammatory response in the airways of cystic fibrosis patients. Neutrophil accumulation and subsequent release of proteolytic enzymes (e.g. matrix metal-tricuspid ring, peak systolic strain rate from the mid free wall of the left and right ventricle (longitudinal wall motion).

Results: Patients and controls were matched for age (Table) . All CF children had normal left ventricular systolic function by conventional parameters (fractional shortening over 28%, mean 37.9±6.7%) and by TDE. Left ventricular mitral ring peak systolic velocities and peak systolic strain rate were comparable in both groups (Table) . CF patients had diminished tricuspid ring TDE velocities compared to controls (Table) . Peak systolic strain rate was also significantly diminished in CF patients compared to normal controls (Table) .

Conclusions: This study demonstrated subclinical changes of right ventricular wall motion in children with CF without overt signs of right heart failure during routine outpatient clinic visits. Strain rate imaging appears to be a sensitive tool to monitor the heart in CF. Myocardial velocities are also changed but are more difficult to interpret in individual children due to their strong age-dependence. The prognostic significance of these abnormalities will have to be determined. Rationale: Spirometry has been identified as the best predictor of lung disease and is utilized as the standard to evaluate disease severity and the requirement of additional treatment and/or hospitalization in individuals with CF. To date measures of diffusion of the lungs for carbon monoxide and nitric oxide (DLCO/DLNO) and its components alveolar-capillary membrane conductance (DM) and pulmonary-capillary blood volume (Vc) in CF subjects during maximal exercise using the rebreathe technique have not been studied as predictors of changes in outpatient therapy secondary to acute exacerbation. We hypothesized that these gas diffusion measures may explain more variability than resting spirometry measures alone. Resting spirometry has previously been criticized as lacking sensitivity to pulmonary status and early inflammation, especially in early stages of the disease.

Methods: To explore the potential contribution of gas diffusion measures (DLCO, DLNO, DM, Vc, DM/Vc) at rest and peak exercise, along with resting spirometry measures (FVC, FEV 1 , FEF 25-75 ) and lung clearance index (LCI), we studied a cohort of 9 cystic fibrosis subjects (age=26±8 yrs, ht.=164.7±7.7 cm, wt.=61.0±9.8 kg, BMI=23±3 kg/m 2 , VO 2peak =67±23 %predicted, mean±SD) using regressional analysis for predictors of changes in outpatient therapy which was determined from subjects' medical records one year prior to completion of maximal exercise testing in our laboratory.

Results: The factors predicting changes in outpatient therapy included resting LCI and DM at peak exercise, which although not significant explained 30% of the variability, F (2, 5) =1.07, p=0.41, R=0.55, R 2 =0.30. DM at peak exercise accounted for the largest portion of the explained variation (β=-0.66, p=0.33) and LCI accounted for the smaller portion of the variation (β=-0.15, p=0.82). The factors did not show evidence for multicolinearity (Tolerance=0.38). Analysis using only resting measures to predict changes loproteinase-9) lead to airway remodeling and loss of lung function. Doxycycline (DOX) has immunomodulatory activity, specifically against MMP-9. The goal of this study is to characterize the short-term safety, pharmacokinetics, and effect of DOX on inflammatory biomarkers.

A prospective, open-label, randomized controlled study was conducted to determine the effect of DOX on inflammatory biomarkers. Twenty-four patients will be stratified according to pulmonary function (FEV 1 >70%) or moderate (FEV 1 40-70%) and randomized in block to receive no drug, 40 mg, 100 mg, or 200 mg of DOX daily for 28 days. The first dose was administered intravenously and all subsequent doses were given orally. Induced sputum and spirometry were obtained during the 28day treatment period and at a two-week follow-up visit. Sputum MMP-9 and serum CRP levels were determined by ELISA. Serum DOX levels were determined by high-pressure liquid chromatography.

Results: Data for the first 14 patients (11 subjects on DOX, 3 controls are presented. The mean baseline age, FEV 1 , and BMI were 33.82 ± 10.82, 77.545 ± 7.771, and BMI 21.68 ± 2.3 respectively. Serum DOX levels increased in proportion to the dose: Cmax/Cmin (mg/L) were 1.3/0.61, 2.3/0.65, and 3.6/1.6, for 40 mg, 100 mg, and 200 mg respectively. The Tmax for the three doses ranged from 1-2 hours. At the end of treatment and at two weeks following discontinuation of DOX the median change in FEV 1 % was -1.22. and -4.348 respectively. For the control patients, the median change in FEV1% from baseline to the end of treatment was 10.152 and at the six weeks follow up period, it was 4.859. For the DOX-treated patients, the median % change in MMP-9 in sputum from baseline was -16.750 and at two weeks following discontinuation of DOX, was -6.503. For controls the median % change in MMP-9 in sputum from baseline to the end of treatment of 1.950 and -2.165 at the 6-week followup. At the end of treatment, the median % change in hsCRP for the DOX-treated patients was -6.206 (upper and lower limits of 108.654 and -82.609, respectively). At the two week follow up period, the median % change in hsCRP for the DOXtreated patients was 1.231. For the control patients, at the end of treatment, the median % change in CRP was -30.563 and 16.404 at 6-weeks follow-up. No patients experienced any significant adverse effects. Two of the patients on DOX did not complete the study (intolerance during the IV administration of DOX and patient withdrawal).

Conclusions: Preliminary results indicate short-term administration of DOX is well tolerated. Pharmacokinetics are favorable and support once daily administration. No statistically significant differences in measures of airway inflammation or pulmonary function were demonstrated.

Pauliks, L.B.; Vender, R.; Kitch, D.; Allwein, L.; Graff, G. Pediatrics, Penn State Hershey Medical College, Hershey, PA, USA Background: Cor pulmonale is an important complication of cystic fibrosis (CF). However, noninvasive monitoring of right heart function has been difficult as conventional echocardiographic methods only permit a qualitative assessment and magnetic resonance imaging may not always be available. A recent paper in adults with CF using tissue Doppler echocardiography (TDE) showed diminished right heart function in advanced lung disease. TDE tricuspid ring velocities permitted quantitation of RV dysfunction in CF. The objective of this study was to study the right heart in children with CF using tricuspid ring velocities and peak systolic strain rate derived from TDE. Strain rate was used because it is less age-dependent than TDE velocities (which increase during childhood and are therefore difficult to interpret in the individual child).

Methods: Twelve patients (age 8 months-18 years, 4 males) were enrolled during routine outpatient clinic visit when they were in their usual state of health. Controls with a normal cardiac work up (n=32) were recruited in Cardiology clinic. For CF patients, echocardiographic images were obtained with a portable machine (Vivid i, GE Healthcare, Milwaukee, WI) from parasternal and apical standard views and stored as digital echocardiographic raw data. Frame rates for TDE were >100 frames per second. TDE images were analyzed off-line using commercially available software (Echopac, GE Healthcare, Milwaukee, WI). The following parameters were obtained: Peak systolic velocity of the myocardium just below the mitral and in outpatient therapy included LCI and FEV 1 , which accounted for 32% of the unexplained variability though not significant, F (2, 5) =1.17, p=0.38, R=0.57, R 2 =0.32. FEV 1 accounted for the largest portion of the variation (β=-0.77, p=0.30) and LCI accounted for the smaller portion of variation (β=-0.27, p=0.70). The factors did not show evidence of multicolinearity (Tolerance=0.31). The addition of LCI to the model from FEV 1 alone improved the R 2 value from 17% to 32%.

Conclusion: These findings suggest that gas diffusion measures at rest and during exercise do not provide additional information beyond what can be determined from conducting resting spirometry measures. This study suggests that the inclusion of LCI along with standard spirometry can provide addition information regarding the variability in changes to outpatient therapy secondary to acute exacerbations in patients with CF. Due to small sample size in this initial analysis, further investigation is necessary to confirm the findings of this exploratory study. In patients with cystic fibrosis (CF) forced expiratory volume in one second (FEV1) is the currently accepted method to predict morbidity and mortality. Body plethysmography and impulse oscillometry (IOS) may reflect better the physiological situation as manoeuvres require only quiet breathing for a short period of time. In addition, IOS can be used in very young children (>2 years old).

Objectives: This study was designed to compare the long-term results of these three lung function tests with clinical outcome in CF patients.

Methods: This prospective observational study included 109 adult CF patients and 1028 visits. Over a period of 4.5 years, IOS, body plethysmography and spirometry were performed at routine visits. Clinical data (cough, sputum, overall condition, C-reactive protein, IgE, IgG) were available for all visits and were scored as stable, improved, worsened or exacerbation in comparison to the previous visit. We analysed the parameters FEV1, specific and total airway resistance (sRtot, Rtot), total impedance Z at 5 Hz (Z5), resistance at 5 Hz (R5), reactance at 5 Hz (X5) and area under the curve of reactance (AX) in association with the clinical outcome. As statistical model multivariate analysis and generalized estimated equations were used.

Main Results: FEV1 was significantly correlated with Rtot (r=-0.88), sRtot (r=-0.80) and X5 (r=0.83). Rtot was significantly correlated with R5 (r=0.76) and X5 (r=-0.85). Rtot and R5 were found to give interchangeable readings for resistance values < 0.8 kPa*L-1*sec. Above this cut-off, R5 underestimated airway obstruction compared to Rtot. When using the generalized estimating equation to evaluate time dependent changes, Rtot, sRtot and X5 were equally correlated to FEV1.

The best predictor of decrease of clinical condition were previous fluctuations of AX.

Conclusions: Spirometry, body plethysmography and IOS were highly correlated in a large cohort of adult CF patients over 4.5 years. While FEV1 had a higher discriminatory capacity in mild to moderate lung disease, sRtot and X5 are more reliable to reveal changes over time in patients with severe lung disease. AX has the potential to predict subsequent worsening of clinical conditions. We propose the use of IOS as alternative trend measurement in CF patients. Background: ATS standards for spirometry require a minimum of 3 maneuvers, and suggest no more than 8 maneuvers be performed. Testing can be concluded if the 2 highest maneuvers have FVC and FEV 1 that vary by less than 150mL. Our laboratory has allowed and encouraged patients to perform additional maneuvers even after meeting reproducibility criteria, while many adult laboratories do not. We hypothesized that forced exhala-tions during spirometric maneuvers may provide airway clearance to patients with CF, resulting in further improvement in lung function.

Methods: Retrospective review of PFT database. We randomly selected 50 spirometry studies from patients with CF who were over 18 years of age and who performed 6 or more efforts during their testing session. We compared the maximum FVC and FEV 1 in the first 3 efforts to the maximum FVC and FEV 1 in any subsequent efforts.

Results: FVC increased (mean ± sd) 2.72 % ± 3.4% and FEV 1 increased 2.41% ± 4.6% in efforts 4-8 compared to efforts 1-3. Twelve patients (24%) had an increase in FVC or FEV 1 more than 5% with the maximum change in FVC and FEV 1 being 12.6% and 19.6%, respectively. When analysis was restricted to testing that met strict reproducibility criteria for the first 3 efforts, 7/38 patients (18%) had an increase in FVC or FEV 1 more than 5%. Interestingly, 7/38 (18%) experienced a decrease in FVC and 11/38 (29%) experienced a decrease in FEV 1 with additional efforts. Improvement with subsequent efforts was not correlated with baseline lung function, gender, or age.

Conclusion: Most patients did not have large changes in FVC or FEV 1 after the initial 3 efforts. However, a significant minority of patients may demonstrate improvement in lung function after the initial 3 spirometric maneuvers, even if these maneuvers meet ATS reproducibility criteria. It is possible that this is due to airway clearance during the testing. A similar percentage of patients may experience decline in their lung function, which may reflect fatigue. Introduction: Cystic fibrosis (CF) lung disease is characterised by progressive destruction and dilatation of the airways. This occurs in the setting of chronic infection, particularly with Pseudomonas aeruginosa (Pa), and is associated with loss of elastin and cartilage, with progressive submucosal fibrosis. Matrix Metalloproteinases (MMPs) can degrade all components of the extracellular matrix, and are inhibited by the Tissue Inhibitors of Metalloproteinases (TIMPs). High MMP-9 activity has been identified in CF sputum, and correlates inversely with lung function (FEV1). The source and regulation of MMP-9 in CF is unknown. In other chronic suppurative lung diseases, for example tuberculosis, direct infection of inflammatory cells drives MMP-9 production, but inflammatory-parenchymal cell networks further augment MMP-9 secretion. We hypothesised that Pa infection could drive MMP-9 secretion by inflammatory and epithelial cells in the CF airway.

Methods: Neutrophils (PMNs) were isolated by dextran-sedimentation followed by Ficoll separation. Monocytes were isolated by density centrifugation across a Ficoll-Paque gradient followed by adherence, and were matured in GM-CSF for 5 days to produce monocyte derived macrophages (MDMs). PMNs, monocytes and MDMs were stimulated for 4 hours with the laboratory Pa strain, PAO-1. To investigate the ability of infected monocytic/granulocytic cells to increase MMP-9 production by the epithelium, conditioned medium from Pa infected PMNs (coNPa), monocytes (coMPa), monocyte derived macrophages (coMDMPa) was used to stimulate normal bronchial epithelial cells (HBE) and CF bronchial epithelial cells (CFBE). MMP-9 was measured by gelatin zymography and ELISA. TIMP-1, the major inhibitor of MMP-9, was measured by ELISA.

Results: Direct live infection with Pa increased MMP-9 in the culture supernatants of PMNs, monocytes and MDMs. In contrast, Pa infection at MOI of 100 caused a significant reduction in TIMP-1 in culture supernatants of PMNs (1.6+/-0.19 ng/ml vs 2.5+/-0.19 ng/ml, p<0.033), monocytes (124.3 +/-10.95 pg/ml vs 384.1+/-39.9 pg/ml, p<0.001), and MDMs (313+/-21.7 pg/ml vs 447.5+/-42.12 pg/ml, p=0.038). Conditioned media from infected PMNs (coNPa) significantly increased MMP-9 production from HBE (coNPa 17.8+/-0.83 ng/ml vs 4.5+/-0.19 ng/ml, p=0.001) and CFBE cells (25+/-0.5 ng/ml vs 9.5+/-0.3 ng/ml, p=0.001). Similarly CoMPA increased MMP-9 production from HBE (coMPA 10.29+/-0.6 ng/ml vs 5.2+/-0.3 ng/ml, p=0.001) and CFBE cells (coMPA 16.6+/-2 ng/ml vs 6.0+/-0.8 ng/ml, p=0.001) while coMDMPa increased MMP-9 production by 3 fold from both normal and CF epithelial cell lines (p=0.001). In

Vermeulen, F.; Ophoff, J.; Proesmans, M.; De Boeck, C. CF centre, University Hospital Leuven, Leuven, Belgium Aim: Lung clearance index (LCI) is a sensitive marker of early CF lung disease. After washin (WI) of a tracer gas to equilibrium, gas mixing efficiency is evaluated during the washout (WO). Theoretically, equivalent results could be obtained from the washin, with the potential to shorten the measurement time. Aim of the present study is to compare results obtained from the WI and the WO phase in CF children and healthy controls.

Methods: Functional residual capacity (FRC) and LCI were measured according to ERS standards using a commercially available device, quantifying flow and molar mass with an ultrasonic sensor and using helium as the tracer gas (Exhalyzer D, Ecomedics, Zurich, Switzerland). Fifty-nine healthy controls (25 males, median age 10 y, range 5-17) and 47 CF patients (21 males, median age 11 y, range 5-17) performed three tests. Two tests with an FRC <10% apart were required for further analysis. Spirometry was measured according to ATS/ERS standards after the multiple breath manoeuvres.

Results: In controls as well as in CF patients, more WO than WI tracings reached reproducibility criteria (87/106 vs 76/106). In 32 CF and 35 control subjects, both WI and WO were interpretable.

Mean LCI WO was 6.90 (SEM 0.13) in controls and 8.79 (SEM 0.25) in CF patients (difference of means 1.89, p<0.001). Mean LCI WI was 6.53 (SEM 0.15) in controls and 7.71 (SEM 0.19) in CF patients (difference of means 1.18, p<0.001).

The mean coefficient of variation for LCI was similar for WI and WO (respectively 7.84% and 6.05%, p=0.079).

LCI WO was higher than LCI WI with a mean difference of 0.71 (95% limits of agreement -1.41 to 2.83). This difference was larger in CF patients than in controls (0.37 vs 1.07, p=0.028), and increased with higher LCI values (R=-0.323, p=0.008).

Conclusion: LCI WI and LCI WO are both significantly higher in CF than in controls but reproducibility is better for LCI WO than for LCI WI . These data show that values calculated from multiple breath WI and WO are not equivalent. Ways to improve reproducibility of WI should be explored. In vitro data shows that Gallium (Ga) kills P. aeruginosa (including multi-drug resistant strains), prevents biofilm formation, and effectively treats P. aeruginosa infections in animal models. Data for inflammatory lung diseases (eg, sarcoidosis) or difficult-to-find infectious sites indicate that Ga accumulates in regions of inflammation or infection. We hypothesized that the antimicrobial activity of systemically administered Ga might be enhanced if it accumulates within areas of infection, an advantage not typical of antimicrobial agents. In this study we explored whether acute CF pulmonary exacerbation resulted in high concentration of 67-Ga in the lung.

Methods: We performed 67-Ga Single Photon Emission Computed Tomography (SPECT) scans at the start of an acute exacerbation, then 2-4 weeks post completion of antibiotics. We also collected lung function data and correlates of inflammation (serum: ESR, CRP, PMNs; sputum IL-1β, IL-8, PMNs).

Results: Six CF patients had complete data. Sputum from all six grew mucoid P. aeruginosa although two were co-infected with MRSA, one with B. multivorans. Paired 67-GaSPECT scans results were positive for 2 of 6 contrast TIMP-1 was reduced in HBE and CFBE following coNPa, coMPA, and coMDMPa stimulation.

Conclusion: Direct Pa infection of inflammatory cells increases MMP-9 in culture supernatants and is accompanied by a reduction in its major inhibitor TIMP-1. Pa also induces functionally uninhibited MMP-9 secretion from airway epithelial cells by neutrophil and monocytic-dependent networks. These data suggest that both the infiltrating leucocytes and the lung parenchyma are potential sources of excessive MMP-9 activity in CF, and may contribute to the airway destruction that occurs in response to Pa infection. Background: Progression of lung disease in cystic fibrosis (CF) is characterized by the development of bronchiectasis (BE) and trapped air (TA). This can be detected and monitored by chest computed tomography (CT). BE is a potentially important surrogate endpoint for CF lung disease. It is unknown whether BE influences the Health Related Quality of Life (HRQoL). HRQoL can be assessed by the validated Cystic Fibrosis Questionnaire-Revised (CFQ-R).

Objective: To test the hypothesis: the severity of BE has a negative effect on the HRQoL as established by the respiratory domain of the CFQ-R.

Methods: Cross sectional study. Inclusion criteria: CF, clinical stability, availability of chest CT and CFQ-R obtained on the same day as part of the routine annual examination between July 2007 and January 2010. The CFQ-R was completed by children 6-13 years, their parents (to assure a multi-informant approach) and adolescents (≥ 14 years). The CFQ-R (6-13 years) contains 8 domains, the CFQ-R (≥ 14 years) includes 12 domains. These domains include physical, vitality, emotion, social, role, body image, eating, treatment burden, respiratory symptoms, digestion, health and weight. For each domain, a score between 0-100 can be obtained. Higher scores indicate better HRQoL. CT scans were anonymized and scored in random order using the CF-CT score, an upgraded and validated version of the Brody II score. The CF-CT score includes the following component scores: BE, TA, airway wall thickening, mucus plugging and parenchyma. Component scores are expressed as a percentage of the maximum scores. Correlations between the respiratory domain of the CFQ-R and BE were calculated by using the Spearman's correlation coefficient.

Results: Seventy-two patients with CF (35 males) were included. We collected a total of 109 CFQ-R questionnaires, consisting of 40 CFQ-R (6-13 years), 32 CFQ-R (≥ 14 years) and 37 parental CFQ-R. Preliminary results showed a median (range) CFQ-R respiratory symptom score of 83 (11-100) points. For children aged 6-13 years a significant correlation between the respiratory domain of the CFQ-R and BE was found, using the parent report (rs = -0.35, p = 0.03). The respiratory domain of the CFQ-R ≥ 14 years strongly correlated with the BE score (rs = -0.46, p < 0.01). In general the respiratory domain of the CFQ-R correlated with the CF-CT BE score (rs = -0.38, p = 0.00).

Conclusion: Higher CF-CT BE scores are associated with lower CFQ-R respiratory symptom scores. This study is supported by an unconditional grant by Gilead.

patients at time of exacerbation and strikingly correlated with maximal area of disease on lung CT. The serum inflammatory markers comparing onset of exacerbation to 2-4 weeks post antibiotics trended towards improvement, but not significantly. Similar trends were seen for the FEV1 and sputum data although the difference in PMN counts was significant (table below) . Additional work will be needed to determine why Ga accumulation localized in areas of marked CT scan changes in some patients but not others. Possibilities include poor perfusion to affected areas and poor Ga distribution due the small doses delivered during 67-GaSPECT scans. It is also possible that the severity of CT findings do not correlate well with lung inflammation in CF.

Conclusions: 1. 67-GaSPECT scan results for CF lung disease show results disparate from markers of inflammation or infection; 2. Ga may accumulate in areas of CF lung associated with inflammation or infection in some situations; 3. It seems reasonable to proceed with studies that will examine the efficacy of IV Gallium nitrate as anti-P aeruginosa treatment in CF.

Funded by CFF-Iowa RDP.

*P = 0.009 (Paired t test) Objective: With the increasing life expectancy for patients with cystic fibrosis, the importance of the cumulative iatrogenic radiation exposure is becoming more significant. This study explores the estimated cumulative effective dose (CED) from all thoracic and extra-thoracic imaging modalities and interventional procedures, over a 17 year period at a nationally designated Irish CF centre. Additionally the temporal patterns of imaging modality use are explored over the same period.

Methods: The estimated CED from all thoracic and extra-thoracic imaging modalities (X-ray, CT, interventional and fluoroscopic imaging) for all pediatric and adult patients with CF (PWCF) who exclusively attended a nationally designated tertiary Irish CF centre for >1 year from 1st July 1992 to either time of lung transplantation, death or the study end date (1st May 2009) was determined. The study period was divided into 3 equal tertiles (tertile 1: Jul 1992 -Feb 1998 , tertile 2: Mar 1998 -Nov 2003 and tertile 3: Dec 2003 -May 2009 ) and estimated CED for all thoracic and extra thoracic radiological procedures, per tertile, per person year of follow up was calculated.

Results: In all, 230 patients met inclusion criteria with 2,240 person years of follow up and a total of 5596 radiological procedures. One patient had a CED >75 mSv (0.43%), 36 patients (15.6%) had a CED between 20-75 mSV, 56 patients (24.3%) had a CED between 5-20 mSv and 138 patients (60%) had a CED between 0-5 mSv over the study period. The mean annual CED per patient increased consecutively from 0.42 mSv/yr to 0.50 mSv/yr to 1.58 mSv/yr over tertiles 1-3 respectively (p=0.001). CT scanning use increased 5.7 fold over the 3 tertiles adjusted per person years of follow up (p=0.001); 46.7% of CED related to thoracic imaging and procedures and 53.3% of CED was from extra thoracic imaging and procedures. Of the extra-thoracic imaging 80.1% related to abdominal imaging and abdominal interventional procedures, representing 42.7% of total radiation exposure for PWCF.

This study highlights a progressive increase in iatrogenic ionizing radiation exposure to PWCF, and a significant increased use of CT scanning across a 17 year period. The need to examine closely each indication for CT scanning in PWCF is highlighted, and where indicated, a need to develop appropriate low dose CT thorax strategies. Furthermore, extrathoracic low dose imaging protocols and radiation protection strategies are crucial to decrease cumulative effective radiation dose. "Sapienza", Rome, Italy; 2. Pulmonology and Radiology, Erasmus MC, Rotterdam, Netherlands; 3. Pediatrics, University of Rome Sapienza, Rome, Italy; 4. Radiology, "Ca' Foncello Hospital", Treviso, Italy Background: CF is characterized by chronic airways infection and inflammation and progressive lung disease from early childhood. Currently there are no sensitive, radiation free methods available to localize and quantify lung inflammation. PET-CT has been proposed for the localization and quantification of active inflammation in the lung, but its use is restricted by high ionizing radiation exposure and costs. DWI is a promising tool for the identification of active foci of inflammation in lung parenchyma. DWI signal intensity is strictly dependent to the free movement of water molecules which depends on the changes occurring in the extra cellular environment. DWI signal will be higher where an increased number of cells and macromolecules limit water movements (as it happens in inflammation). DWI accuracy in the identification of inflammation has been proven in organs such as: brain, liver, and intestines.

Objective: To investigate DWI in the lungs in a cohort of CF patients.

Methods: This 2 center study was approved by the institutional review boards. Thirty-three stable CF patients (14 males) had a chest MR including a DWI sequence with a 1.5 T scanner (Avanto Siemens,Germany); for a morphological evaluation, a respiratory triggered T2 weighted BLADE sequence was performed (TR: 2000; TE = 27; FOV 400 mm; flip angle: 150; slice thickness 5 mm) followed by DWI (TR: 5632.12; TE: 83; flip angle: 90; slice thickness 5 mm). A chest CT was performed on the same day. Scans were anonymized and randomly scored by 2 independent observers. DWI were scored using a newly developed semi-quantitative scoring method, taking in account the number of areas with altered signal intensity (hot spots) in lung parenchyma. CTs and morphological MRIs were scored using a validated CF-CT score (upgraded Brody II). Overlay techniques were used to study overlap between DWI and morphological MRI. DWI scores were correlated to clinical and radiological parameters reflecting disease severity: FEV1 % predicted, Body Mass Index (BMI), CF-CT and MRI bronchiectasis (BE) scores. Patients without DWI hot spots were compared to those with hot spots. Scores are expressed as % of the maximal possible score.

Results: Mean age 24.6 years (range 8-51 years); FEV1 % predicted 78.64 (38 to 113); DWI score 9.89 (0 to 37); total CT scores 22.88 (3 to 49); total MRI scores 20.65 (7 to 40). DWI score correlated positively to total CT score (r=0.66, p=0.0001), total MRI score (r=0.599, p=0.001) and CT-BE score (r=0.615, p=0.001). FEV1 and BMI of patients without DWI hotspots (n=10) was higher (all p<0.0001) compared to patients with hot spots. DWI pattern only in part overlapped that of structural abnormalities on morphological MRI or CT.

Discussion: DWI could potentially be used to localize and quantify active lung inflammation in CF. Further validation is needed to specify the exact nature of the hot spots.

Conclusion: In CF lung disease DWI is able to distinguish between hot and cold spots that overlap in part the structural abnormalities as can be seen on MRI and CT.

Srivastava, S.A. 1,2 ; Burgess, J.C. 2 ; Bilton, D. 2 ; Hodson, M.E. 2 ; Gyi, K.M. 2 ; Baker, E.H. 1 1. Centre for Clinical Pharmacology, St George's, University of London, London, United Kingdom; 2. Department of Cystic Fibrosis, Royal Brompton Hospital, London, United Kingdom Introduction: In cystic fibrosis (CF), CF-related diabetes (CFRD) and impaired glucose tolerance (IGT) affect up to 25% of patients. Hyperglycaemia in CF is mainly due to insulin deficiency but is confounded by insulin resistance which increases during exacerbations. The development of CFRD is associated with accelerated pulmonary decline, weight loss and early death. The relationship between "glycaemic load" and pulmonary parameters in patients without CFRD or IGT is not known.

Aims: To investigate the relationship between HbA 1C , a marker of "glycaemic load," and pulmonary disease in CF patients without CFRD or IGT.

Methods: Adult CF outpatients (≥16 yrs) with stable disease were recruited from a single tertiary CF centre. All gave written informed consent and the study had ethical approval. Stability was defined as ≥ 6 weeks without any new pulmonary symptoms, new antibiotics or steroids. All had clinical parameters recorded and HbA 1C measured. Patients were designated as: CFRD if on treatment or by oral glucose tolerance test (OGTT) according to WHO guidelines; IGT by OGTT or fasting hyperglycaemia; or normal if normal OGTT or diabetes not clinically suspected on selective screening policy (HbA 1C < 6.1%, random blood glucose < 11.1 mmol/l, no new unexplained clinical symptoms). Patients were followed for 52 weeks to determine exacerbation frequency defined by either the patient or the treating physician.

Results: Of 254 patients, 17% had CFRD (HbA 1C 8.1±1.8% (mean±SD)), 10% had IGT (HbA 1C 6.5±0.4%) and 73% had no known diabetes (HbA 1C 5.8±0.5%) (p=0.000). The following results apply to the 177 patients with no known diabetes.

HbA 1C was significantly higher in patients: taking steroids (prednisolone 6.1±0.6%, no prednisolone 5.8±0.4%, p=0.007); with P. aeruginosa or B. cepacia on sputum culture (Ps. or cepacia 5.9±0.4%, no Ps. or cepacia 5.7±0.5%, p=0.015) and with pancreatic insufficiency (PI) (PI 5.9±0.5%, no PI 5.7±0.4%, p=0.032).

HbA 1C was significantly correlated with: FEV 1 %predicted (R=-0.401, p=0.000); oxygen saturation on air (R=-0.333, p=0.000); chest xray score (R=-0.349, p=0.000); BMI (R=-0.234, p=0.002); WCC (R=0.200, p=0.008); and exacerbation number by 52 weeks (R=0.199, p=0.026).

On univariate analysis of variance, only FEV 1 %predicted remained independently associated with HbA 1C (p=0.046).

Conclusion: One third of CF patients with no known diabetes have elevated HbA 1C , which is independently associated with FEV 1 %predicted. Mechanisms could include insulin deficiency and insulin resistance induced by pulmonary inflammation or infection. The use of HbA 1C in the CF patient without diabetes merits further investigation.

Acknowledgements: This study was funded by Cystic Fibrosis Foundation Therapeutics, Inc. The effectiveness (in actual practice) of many CF treatments is unknown. The CFF registry is a tool for testing effectiveness, but has not been used because of the treatment-by-condition confound (more care is associated with worse outcomes because sicker patients need more care and, usually have worse outcomes). Using the CFF registry, we sought to demonstrate that an instrumental variables (IV) approach can address this con- Introduction: As patients with cystic fibrosis (CF) live longer, lung function decline is no longer a sensitive marker of pulmonary disease, and there is a need for other non-invasive biomarkers. In adult patients with CF, we have previously found breath glucose concentrations to be elevated 4-8 times compared to healthy volunteers [Baker 2007; 102:1969-75] . Elevated glucose concentrations in lung secretions could therefore be a marker for CF lung disease.

Objectives: To determine the use of breath glucose as a clinical biomarker of pulmonary disease in CF.

Methods: Adult CF outpatients (≥16 yrs) with stable disease were recruited from a single tertiary CF centre. All participants gave written informed consent and the study had ethical approval. All had a diagnosis of CF based on two of: positive sweat test; clinical phenotype; diagnostic genotype. Stability was defined as ≥ 6 weeks without any new pulmonary symptoms, new antibiotics or steroids.

All participants underwent 10 minute exhaled breath condensate collection and blood glucose measurement. Clinical data was simultaneously collected. Patients were then contacted at 12, 26 and 52 weeks to record the number of pulmonary exacerbations. Exacerbations were defined by either the patient or the treating physician.

Condensate samples were lyophilised, resuspended and glucose concentration measured by high performance anion exchange chromatography. Dilution correction was performed using total cation concentration, calculated from sample conductivity, to obtain breath glucose, an estimate of glucose concentration within lung secretions. All statistical tests were performed on SPSS 16.0.

Results: A total of 254 patients were enrolled over 18 months from 2007-2009; 227 patients (90%) had exhaled breath condensate samples analysed for glucose. The mean (± SD) condensate glucose value was 0.72 (± 1.32) µmol/L, with median 0.03 µmol/L (IQR 0.02, 0.08). A histogram of breath condensate values for this cohort demonstrates skewed data with 82% of patients having breath glucose values of <1 µmol/L.

When compared to baseline clinical parameters, there was no statistically significant association between breath glucose and the following: sex, age, lung function as measured by FEV 1 , body mass index, sputum microbiology, chest xray northern score or inflammatory markers (CRP or WCC). There was no association between breath glucose and paired blood glucose or diabetic status. There was no association between breath glucose and time to next pulmonary exacerbation or number of exacerbations by either 12, 26 or 52 weeks.

Conclusion: Breath glucose is unlikely to be a useful clinical marker in its present state. The problems of dilution and concurrent metabolic activity make it difficult to quantify, and the use of assays working to the limit of detection flaw the accuracy of these measurements.

The breath glucose values obtained are lower than those obtained in our earlier work, which confirms that this method is currently unreliable for clinical use.

Acknowledgements: This study was funded by Cystic Fibrosis Foundation Therapeutics, Inc. found, using the example of tobramycin in slowing rate of decline in lung function.

Methods: We used 2001-2008 data from the CFF patient registry. Patients 6-21 y with >2 years observed FEV1%p were included. FEV1%p rate of decline was the outcome variable. Patients were eligible for tobramycin in each clinical encounter if they had chronic Pseudomonas aeruginosa (PA) (>50% of prior cultures in past year positive for PA). Patient tobramycin usage (TU) was calculated as the ratio of number of times the patient was listed as taking the drug to number of eligible times during the study period.

IVs are variables correlated with treatment but unrelated to severity of illness. Random allocation to experimental condition is the perfect IV, but others exist: We used CF center tobramycin prescribing rates as our IV, reasoning that patients of equal severity treated at CF centers with different prescribing patterns had a non-equal likelihood of being prescribed tobramycin. We accounted for observed burden of illness by including baseline FEV1%p, proportion of time spent with CF-related diabetes, B. cepacia, concomitant medication use, hospitalization days due to pulmonary exacerbation, and gender in all models. We conducted ordinary least squares (OLS) regression to calculate results for the "naïve" regression and for the IV analysis.

Results: Results are reported as regression coefficient or mean +/-SE, p-value (p) and percentage. There were 22,448 patients with sufficient data who met previously defined constraints. The "naïve" OLS model indicated that increased TU was associated with faster rate of decline (0.27 +/-0.05%p per year; p<0.001). The second-stage regression, which uses the instrumented version of patient TU, indicated that increased usage is associated with slower rate of decline (-0.61 +/-0.21%p per year; p=0.0040).

Conclusion: An IV approach can correct for the treatment-by-condition bias in registry data. Because CF care is organized by center, using centerspecific prescription rates as an IV is a feasible approach to determining the effectiveness of treatments using the CFF data registry. These data indicate that tobramycin is effective in clinical practice. 4 ; on behalf of the UK CF Gene Therapy Consortium, .. 5 1. Imperial College, London, United Kingdom; 2. Royal Brompton Hospital, London, United Kingdom; 3. Southampton University, Southampton, United Kingdom; 4. Western General Hospital, Edinburgh, United Kingdom; 5. UK CF Gene Therapy Consortium, Edinburgh, London, Oxford, United Kingdom The UK CFGTC is conducting a large longitudinal study (the Run-in) to assess outcome measures and identify optimal patients for a multidose trial of non-viral gene therapy. Two imaging modalities are being employed: radioisotope deposition and clearance scans and CT.

In contrast to the repeated measures performed at each visit and in view of the radiation involved, stable subjects underwent these scans on a single occasion only. The purpose was: Deposition scan, to determine the optimal patients for topical drug delivery; and CT, to assess its suitability as an efficacy outcome measure.

99mTc-labelled human serum albumin was inhaled via a Pari LC and Akita system to enhance conducting airway deposition. Anterior and posterior planar gamma camera images and SPECT were used to assess 3-D deposition. Images were scored digitally and, after coding, they were visually graded (VG) I-IV by a group of respiratory physicians: I-no defects; IIpatchy deposition; III-patchy deposition with large defects; IV-grossly abnormal. HRCT scans were scored by two radiologists on a lobar basis for the following features: extent and severity of bronchiectasis, airway wall thickening, small and large mucus plugs and gas trapping.

Deposition scans were available on 147 subjects; digital indices (DI) ranged from 34 (best) to 150 (severely abnormal). Visual grading was well correlated with DI (R2 0.63; p<0.001) and both were significantly negatively correlated with FEV1% (p<0.01). Nine subjects in Grade IV had a mean (SD) FEV1 over all study visits of 43.9 (4.3)%, which was significantly lower than the means of groups I-III (p<0.01). They were also the most severely affected group for every CT score. On the basis of these deposition scan appearances, this group was considered unsuitable to progress to the gene therapy trial. Patients in the other groups have deposition scans which suggest that the gene therapy product could be delivered at least moderately well to their airways and will now be filtered through other inclusion/exclusion criteria.

Potentially reversible components of the HRCT scores are being considered as efficacy outcome measures. As an example, whilst bronchiectasis per se is irreversible, airway wall thickening was closely correlated with it in our cohort (R2 0.52; p<0.001) and was reduced in our previous study by IV antibiotics. Although our group has a wide range of FEV1 severity, only 7.2% of scans had no or trivial wall thickening. Power calculations based on these cross sectional data suggest that our anticipated study group of 100 subjects would have 80% power to detect a change in wall thickness half that seen with IV antibiotics. A subgroup of patients with near normal CT scans and other likely outcome measures are being identified; this subgroup may be deemed unsuitable to progress due to having no "measurable" outcome measures.

Lung imaging techniques have aided us in the identification of patients to take through into our Multidose Trial and are currently under consideration as efficacy outcomes.

Funded by the UK CF Trust. Background: Pulmonary exacerbations in cystic fibrosis (CF) are associated with increased pulmonary decline and reduced life expectancy. Exacerbations are usually defined by symptoms, signs and investigation results. In the absence of a consensus definition, need for treatment with intravenous (IV) antibiotics is a common trial endpoint. Milder CF exacerbations not requiring IV antibiotics are less well described. We therefore aimed to characterise all exacerbations requiring antibiotic treatment in an adult CF population.

Methods: 232 CF patients (30±9 years) underwent baseline assessment and follow up for 1 year. Patients were contacted at 12, 26 and 52 weeks to document exacerbations, defined as increased or new antibiotics for: change in respiratory (shortness of breath, cough, sputum volume and colour, wheeze, haemoptysis, chest pain, coryzal) or general (tired, fever, time off work/education, decreased appetite or weight) symptoms; fall in FEV 1 ; new sputum microbiology. The study had ethical approval and all participants gave written informed consent.

Results: In 1 year 48% patients required IV antibiotics, 72% patients required oral antibiotics and 82% had antibiotics by either route. Of 629 exacerbations, 63% were treated with oral (13.9+4.2 days) and 37% treated with IV (14.7+5.2 days) antibiotics (p=0.053). Exacerbations requiring IV antibiotics were more likely to be diagnosed in the specialist CF centre (IV 89%, oral 64%, p=0.000) and present with combined chest and general symptoms (IV 32%, oral 18%, p=0.000).

In individuals, the proportion of antibiotics given IV was inversely related to FEV 1 %predicted (R= -0.478, p=0.000) and body mass index (R= -0.270, p=0.000), and was greater in patients with pancreatic insufficiency (35±33%) vs pancreatic sufficiency (22±25%, p=0.046) and with diabetes (43±35%) vs no diabetes (28±33%, p=0.012). Sputum culture determined the proportion of IV use as follows: no pathogen 21±30%; S. aureus 17±29%; Pseudomonas spp 38±35%; Burkholderia spp 47±35%; non-fermenting Gram negative 53±25%, p=0.001. FEV 1 %predicted (partial Eta 2 =0.145, p=0.000) and an interaction between Pseudomonas and diabetes (partial Eta 2 =0.045, p=0.017) were independent predictors of IV requirement.

Time to next exacerbation was: oral, 49 (25-91) days (median (interquartile range)); IV, 60 (30-93) days, p=0.256. Time to first IV administration was prolonged by increasing courses of oral antibiotics: 0 courses, 66 (35-150) days; 1 course, 126 (69-240) days; 2 courses, 184 (163-272) days; 3 courses, 207 (115-312) days; 4 courses, 294 (178-234) days, log rank p=0.002.

Conclusion: Adult CF patients use oral antibiotics twice as often as IV antibiotics for exacerbations and do this 36% of the time without consulta-A cohort of 36 β-lactam-allergic and 12 sulfamethoxazole-allergic patients, were studied using skin testing (skin prick and intradermal test) and the in vitro lymphocyte transformation test, which was modified for the sulfamethoxazole samples to increase its sensitivity.

Skin tests were positive in 3 piperacillin-allergic individuals and 1 sulfamethoxazole-allergic patient. Drug-specific lymphocyte proliferative responses were detected in a dose-dependent manner in 19/28 piperacillinallergic (68%), 3/11 meropenem-allergic (27%) and 1/15 aztreonam-allergic (7%) patients. Ceftazidime-specific lymphocyte responses were not detected. Lymphocytes from 2/12 sulfamethoxazole-allergic patients (17%) were stimulated to proliferate strongly, in a concentration-dependent fashion, with the SMX metabolite nitroso (SMX-NO). In contrast, responses to SMX were weak and only detected at one concentration. Using the modified proliferation assay, SMX-NO, but not SMX, specific responses were detected with lymphocytes from 7/7 tested patients with cutaneous reactions (100%). Lymphocytes from patients with thrombocytopenia and periorbital edema were not stimulated. Piperacillin-stimulated lymphocytes secreted an array of cytokines including TNFα, IL1β, IL6, IL13, and MIP-1α. TNFα, IFNγ, IL4, IL5, IL10 and IL13 were secreted by allergic patients lymphocytes in response to SMX-NO, but not SMX.

These findings show the presence of antigen-specific lymphocytes in blood of drug-allergic patients with cystic fibrosis. The introduction of an antigen-driven T-cell enrichment step prior to the analysis of proliferation increased the sensitivity of the assay. Background: Pulmonary exacerbations (PEs) in CF patients have been associated with increased health care costs, reduced quality of life, and persistent decline in lung function in some patients. Recently we reported the results of a randomized clinical trial (RCT) in which oral azithromycin was associated with a 50% reduction in PEs and a 27% reduction in the initiation of new antibiotics in CF patients 6-18 years of age uninfected with P. aeruginosa (Pa) without a significant improvement in lung function (Saiman et al. JAMA 2010 ). An a priori definition for PEs was utilized in the RCT designed to capture the incidence of PEs in a CF patient group with mild lung disease. This definition was based on previously accepted definitions of a pulmonary exacerbation but utilized a shorter symptom duration (3 vs. 5 days) to meet the criteria for an exacerbation in this mild patient group. The characteristics and impact of PEs in this group are therefore important to characterize.

Methods: A multicenter, randomized, double-blind placebo-controlled trial was conducted from February 2007 to July 2009 at 40 CF care centers in the United States and Canada as recently published. Eligibility criteria included age 6-18 years, FEV1 > 50% predicted, and negative respiratory tract cultures for Pa for at least one year. Two hundred sixty participants were randomized and received study drug. The azithromycin group (n=131) received 250 mg (weight 18-35.9 kg) or 500 mg (weight > 36 kg) of azithromycin Monday-Wednesday-Friday for 168 days and the placebo group (n=129) received identically packaged placebo tablets. Signs and symptom data were collected at the start of each acute antibiotic course, and used to assess whether a protocol defined exacerbation had occurred. The decision to utilize antibiotics was however independent of whether the event met the protocol defined criteria for PE.

Results: In all, 38 and 65 PEs occurred in the azithromycin vs. the placebo participants, respectively. Increased cough and increased chest congestion were reported as the predominant symptoms in both groups and occurred in 92/103 (89%) and 68/103 (66%) of PEs, respectively. Overall, 17/38 (45%) of the PE's in the azithromycin group and 23/65 (35%) in the placebo group were associated with a decline in FEV1 of ≥ 10%. Of the tion with a CF specialist. Oral antibiotic use was more likely in patients with less severe disease or exacerbations without systemic symptoms and delayed time to first IV antibiotic use. The effect of milder exacerbations treated with oral antibiotics on outcomes for CF patients requires further elucidation.

Colistin due to its activity against key pathogens in cystic fibrosis (CF), such as Pseudomonas, is an important tool in the management of CF respiratory infections. Colistin can be associated with dose-dependent nephrotoxicity and neurotoxicity, and, uncommonly, with allergic reactions. The aim of this study was to evaluate adverse reactions to colistin in a cohort of CF patients, and to characterize the presence of colistin-specific lymphocyte in reactive patients.

In our cohort, 180 adult patients had received intravenous colistin, 51 (28%) of which had reactions leading to treatment discontinuation. These reactions consisted of rash (14), headache (11), paraesthesia (9), swollen lips (7), dizziness (6), chest tightness (3), and arthralgia (1). Out of these 51 "sensitive" patients, 11 took part in the study, reactions consisted of rash (7), headache (3), and paraesthesia (1), in parallel to 7 "tolerant" CF patients and 5 non-CF naive controls. Patients were skin tested (both pinprick and intradermal) with colistin. Peripheral blood mononuclear cells were isolated for the ex vivo lymphocyte transformation tests (LTT), and the supernatants of these cultures were analyzed for cytokine expression.

Skin tests were negative in all colistin-sensitive patients. Colistin-specific proliferation was detected in a dose-dependent manner in lymphocytes from 9/12 sensitive patients (75%), with either skin reactions or neurological symptoms. Compared to other antibacterials, colisitin was more toxic against lymphocytes ex vivo, but the dose toxicity profile was similar in CF ("sensitive" or "tolerant") and "naïve" cells. IFNγ, TNFα, MIP-1α, and IL10 secretions were increased by colistin in lymphocytes from "sensitive" compared to "tolerant" CF patients. Colistin increased the secretion of IL1a, RANTES, MIP-1β, and IL1β, in cells from all CF patients. IL6 secretion was increased by colisitin in all lymphocyte cultures, including those from naïve volunteers.

These preliminary findings show that drug-specific lymphocytes might play a role in colistin reactions that involve neurological signs, such as headaches. It also appears that colistin has a dose-dependent direct effect on lymphocytes.

28/131 (21%) and 50/129 (39%) in the azithromycin and placebo groups who experienced a PE, only 7/28 (25%) and 12/50 (24%) respectively experienced a subsequent PE.

Conclusions: PEs occurred frequently in this group of mild CF patients not chronically infected with PA and could be associated with a significant decline in lung function. Few patients in either group experienced more than one PE during the 6 month study. The impact of PEs in this patient group will be further evaluated by determining if the occurrence of a PE is significantly associated with changes in FEV1 and weight, and whether this association differs by treatment group.

Introduction: There is evidence that 30-50% of patients with idiopathic bronchiectasis carry CFTR mutations. The aim of this study was to evaluate if there is a high prevalence of CFTR mutations in patients with idiopathic chronic sinopulmonary disease who have been referred to a tertiary centre. Further, we tested the hypothesis that a significant subset of these patients may be diagnosed with CF based upon evidence of abnormal CFTR-mediated ion channel function, which would be missed if testing was restricted to genotyping.

Methods: Prospectively enrolled patients with idiopathic chronic sinopulmonary disease, including patients with idiopathic bronchiectasis, were extensively evaluated for evidence of CFTR-mediated abnormalities, by CFTR genotyping, sweat test and nasal potential difference (NPD). The results were analysed to determine the prevalence of CF and CFTR-related pulmonary disease.

Results: Nine of 81 patients with another primary cause of sinopulmonary disease were excluded from our analysis. Thirty-three of 72 (49%) patients with idiopathic sinopulmonary disease (age range 6.2 to 66.7 years) carried at least one CFTR mutation and 43 of 72 (60%) showed abnormalities in at least one of the CF diagnostic tests. A CF diagnosis was established in 23 (32%) patients; only 3 (13%) of these patients were diagnosed by genotyping alone, whereas an abnormal sweat test identified CF in 14 (61%) and NPD in 21 (91%) of these patients. CF was more common in patients with idiopathic bronchiectasis (16/50 [32%]), of whom 41/50 (82%) were female. No differences were observed in lung function parameters, pulmonary pathogens and nutritional status between patients with CF, CFTRrelated disease or idiopathic disease.

Conclusion: A significant portion of patients presenting with idiopathic chronic sinopulmonary disease in a tertiary referral centre carry a diagnosis of CF. Since clinical parameters do not help to identify sinopulmonary patients at risk for CF, a low threshold for testing for evidence of CF in this patient population is recommended. Though CFTR gene analysis is important, functional rather than genetic testing is the preferred test to evaluate for CF disease. Background: Scedosporium apiospermum is the third most common colonizing fungi encountered in patients with cystic fibrosis. Despite its increasing identification and prevalence, the effect of Scedosporium apiospermum on pulmonary function is not well understood.

Objective: To evaluate the effect of Scedosporium apiopsermum on FEV 1 % predicted in adult patients with CF.

Methods: Patients attending the University of Southern California Adult Cystic Fibrosis Center were retrospectively screened from 2005 to 2009 for the presence of Scedosporium apiospermum in sputum cultures. Fourteen of the 157 patients in our center were colonized with Scedosporium apiospermum. Patients who had Scedosporium apiospermum at the time of transferring their care from another center during the study period were excluded. FEV 1 % predicted trends were analyzed from one year prior to (baseline) and one year following colonization. Clinical data and comorbid conditions were reviewed through medical records. Decline in FEV 1 % predicted trends were evaluated by linear regression pre and post Scedosporium apiospermum colonization. A paired T-test (2-tailed) was used to assess statistical significance.

Results: The overall prevalence of Scedosporium in our adult CF population was 8.9% (14/157 patients). The 4 men and 10 women had a mean age of 33 years with an average BMI of 22 kg/m 2 . Four of the fourteen patients had CF related diabetes. The mean decline in FEV 1 % prior to and after colonization was -0.26 and -0.16 respectively. Scedosporium colonization had no significant effect on FEV 1 % predicted trends (p-value = 0.7).

Conclusion: This preliminary retrospective cohort study suggests that Scedosporium apiospermum colonization does not appear to significantly impact the rate of decline in FEV 1 % in adult patients with cystic fibrosis. Additional studies are needed to determine if Scedosporium has a long-term impact on pulmonary function as well as its potential for pathogenicity.

Kazmerski, T.; Orenstein, D. Children's Hospital of Pittsburgh, Pittsburgh, PA, USA Background: Dyspnea, or the uncomfortable sensation of breathlessness, is a common and major clinical symptom of many disease processes, yet it is difficult to quantify, especially in children. This study attempts to provide information on the use of the Ease of Breathing Test (EOB test), a previously developed objective measure of dyspnea, as a method to assess the response of acutely ill pediatric pulmonary patients to a therapeutic intervention.

Methods: The EOB was administered before and after treatment for twelve episodes of pulmonary exacerbation in eleven subjects with cystic fibrosis (CF), age 6-21 years old (median age 15.3 years). Each subject performed the two measures (Sustained Phonation and Single Breath Counting) of the EOB Test both at the initiation and conclusion of treatment. Each measure was performed both at rest and after three minutes of stair-stepping exercise on a single 15 cm step at a cadence of 30 complete steps per minute (STEP 30). Heart rate (HR), respiratory rate (RR), oxyhemoglobin saturation (SaO2), and a Visual Analog Scale (VAS) score were also measured.

Results: The Sustained Phonation Score was found to be significantly different pre-and post-stair-stepping exercise, with lower scores (shorter sustained phonation) after stepping, at both the initial (Z = 2.51, p = 0.012) and follow-up visits (Z = 2.35, p = 0.019). The Sustained Phonation Score was also significantly better (longer phonation) at the pre-exercise time point after treatment than it was pre-exercise at the initial visit (Z = -2.20, p = 0.028); and the Sustained Phonation Score post-exercise was longer at follow-up (after the course of antibiotics) than pre-antibiotics, with the difference approaching statistical significance (Z = -1.91, p = 0.056). Changes in the Sustained Phonation Score also significantly tracked and correlated with

Berlinski, A. 1,2 ; Hayden, J.B. 1,2 1. Pediatrics, UAMS COM, Little Rock, AR, USA; 2. Pediatric Aerosol Research Laboratory, ACHRI, Little Rock, AR, USA Background: CF patients perform airway clearance techniques and receive bronchodilators on regular basis. Some positive expiratory pressure (PEP) devices allow concomitant administration of nebulized albuterol. We hypothesize that this practice may alter the aerosol characteristics of nebulized albuterol. We compared the aerosol characteristics of nebulized albuterol generated by 2 different types of nebulizers alone and when connected to different PEP devices.

Materials and Methods: Three units of a continuously operated nebulizer (UP-DRAFT II® Optineb Nebulizer, HUDSON ) and 3 units of a breath enhanced nebulizer (PARI LC® Plus, PARI) were tested in duplicate (n=6 for each configuration). HUDSON was tested alone and connected to Acapella® Choice (resistance setting of 1), Acapella® Duet (resistance setting of 1), EzPAP® (pressure of 5 cm H 2 O). PARI was tested alone and with PARI PEP™ system (resistance setting of 1.5 and 4.5) and PARI PEP™ S system (resistance setting of 1.5). Nebulizers were loaded with 2.5 mg/3 mL albuterol solution and run for 4 minutes at 6 L/min of central air. Aerosol characteristics were determined by cascade impaction [cooled (4°C) Next Generation Impactor technique operated at 15 L/min]. Mass median aerodynamic diameter (MMAD), geometric standard deviation (GSD), percentage of particles < 5 µm (%<5) were calculated with CITDAS V3 software according to US and European Pharmacopeia recommendations. Albuterol was diluted with ultrapure water and assayed via spectrophotometer at 276 nm. Analysis of variance for repeated measures followed by post-hoc analysis with Dunnet's test were used to compare data. Significance level was set at 0.05.

Results: Results expressed as mean ± SD of 6 measurements (Table) . Conclusions: MMAD of albuterol significantly decreased when HUD-SON nebulizer was connected to either an Acapella® device (Choice or Duet) or EzPAP®. The percentage of particles smaller than 5 µm increased when HUDSON was connected to an Acapella® device (Choice or Duet). These changes may affect drug deposition. PARI PEP™ systems did not change nebulized albuterol characteristics.

Disclosure: Supported, in part, by the University of Arkansas for Medical Sciences College of Medicine Children's University Medical Group Fund Grant Program.

*p < 0.05 when compared to HUDSON alone. # p < 0.01 when compared to HUDSON alone.

Mckoy, N.A.; Saldanha, I.; Robinson, K.A. Department of Medicine, Johns Hopkins University, Baltimore, MD, USA Introduction: Chronic infections are common in cystic fibrosis (CF), and repeated infections can cause lung damage and disease. People with CF use airway clearance therapies (ACTs) to clear secretions and improve lung function. ACTs that are self-administered, such as the active cycle of breathing technique (ACBT), may be preferred by people with CF and their fami-changes in subjects' pulmonary function (FEV1) before and after antibiotic treatment (0.659, p= 0.020).

Conclusions: The Sustained Phonation portion of the EOB test was validated as a tool to track improvement of pediatric CF patients brought about by antibiotic therapy for a pulmonary exacerbation. The further development of this test could have benefits as both a clinical tool to track changes in pulmonary health and a research tool to gauge efficacy of new CF therapies.

Supported by Antonio J. and Janet Palumbo Cystic Fibrosis Chair. 

Background: Long expiration through a fixed resistive load was shown to double mucus clearance in cystic fibrosis (CF) patients, but short-term improvement effect on lung function is small or insignificant. The study aims to evaluate if expiratory resistive breathing though "volumetric incentive spirometer" (VISex), newly introduced to the market, can improve lung function in the short term in CF patients.

Methods: Forty CF patients (age 8-44 yrs) were studied during their routine visits to the clinic. Patients performed, with the VISex, exhalation against chosen resistance (estimated 1:3 inspiratory/expiratory ratios). Visual feedback encouraged maintaining target volume. After 8-10 cycles of breathing, huffs followed by spontaneous cough were performed until mucus was expectorated. Exercise session lasted 20-30 minutes. Forced expiratory maneuvers were measured before and 15 minutes after exercising. A change 10% from baseline values of FVC, FEV1, PEF or 20% in FEF25-75 was considered significant.

Results: All patients finished the exercise session successfully and willingly. Twenty-eight patients (70%) improved FVC and FEV1 10% above baseline values. PEF improved in 22 (65%) of the patients and FEF25-75 in 14 (35%) of the patients. Improvements did not correlate with baseline lung function or body mass index. The incidence of improvements in FVC was higher in females (P=0.048) and FEV1 incidence of improvements was higher in young children (<18 years) than in adults (p=0.0728).

Conclusions: The new VISex which combines long expiration through chosen resistance, and visual feedback to withhold constant expiratory volume, followed by huffs, significantly improves short-term lung function in most CF patients. Such modality should be further explored.

lies. We conducted a systematic review of the effectiveness of ACBT compared to other ACTs.

Methods: In June 2009, we searched the Cochrane Cystic Fibrosis and Genetic Disorders Group Trials Register and the reference lists of articles and reviews for relevant studies. Two review authors independently screened each article, abstracted data, and assessed the risk of bias in each study.

Results: Seventeen studies were identified; of these, 13 studies were crossover trials with data published in insufficient detail. Thus four studies (98 participants) were included in the meta-analysis: one randomized controlled trial and three randomized crossover trials. ACBT was compared to conventional chest physiotherapy, airway oscillating devices, autogenic drainage, and high frequency chest compression devices. When assessing patient preference the results varied: more patients preferred autogenic drainage over ACBT, more preferred ACBT over airway oscillating devices, and more were comfortable with ACBT versus high frequency chest compression devices. When assessing sputum weight, no statistically significant differences were seen between ACBT and autogenic drainage or between ACBT and airway oscillating devices. When assessing lung function and the number of pulmonary exacerbations, there was no statistically significant difference between ACBT and ACBT + conventional chest physiotherapy.

Conclusions: There is insufficient evidence to support or reject the use of ACBT over any other ACT. Among the four studies included in the metaanalysis each had different comparators and the sample sizes of three studies were small (<18 participants). ACBT was comparable to other therapies in outcomes such as patient preference, lung function, sputum weight, oxygen saturation, and number of pulmonary exacerbations. Longer-term trials are needed to more adequately assess outcomes important for patients such as quality of life and treatment preference.

Partially supported by the Cystic Fibrosis Foundation, USA. We acknowledge the contributions of Olaide A. Odelola. Background: Over the years, many airway clearance techniques (ACT) have been developed and used in cystic fibrosis. These range from sometimes expensive machines to inexpensive hand-held devices as well as breathing techniques. Equipment of any size is difficult to bring from place to place during the course of the day. HVPEP may help individuals with CF to perform ACT throughout the day. Since no equipment is used, it can be done with Active Cycle of Breathing (ACB), Autogenic Drainage (AD) or Huff Coughs (FET). HVPEP is accomplished by the following: 1) hold the hand in a fist; 2) place the thumb-side of the hand in front of the mouth; 3) take a deep breath in through the mouth with a three second hold; 4) blow out through the closed fist at the point of the thumb and index finger to end of expiration; 5) as expiration is occurring, the fingers can be held tighter to increase positive expiratory pressure (PEP). In addition, the fingers can be squeezed intermittently to create vibrations up to 1-6 Hz.

Objective: This is a pilot study to determine the potential effectiveness of HVPEP.

Method: Ten participants with cystic fibrosis were asked to perform HVPEP each day whenever time and circumstances allowed in addition to their regular ACT regimen. FEV1 was measured at the first visit and then again at the next visit. (Range 1-4 months depending on center schedule visit.)

Results: Center visit one: mean age= 30 (range 19-43), mean FEV1 = 1.46 L, mean percent predicted FEV1= 43.9%. Center visit two: mean age= 30, mean FEV1 = 1.58 L, mean percent predicted FEV1 = 48.1%. An 8.2% improvement in FEV1 was noted in the group after using HVPEP (range of improvement 3% to 19%). No one using the technique reported any adverse effects. One participant had a 6% decline in FEV1. This patient had a large improvement in FEV1 in the prior center visit after a course of IV antibiotics. Positive expiratory pressure (PEP) ranges from as low as 20 cm and up to 45 cm H 2 O pressure were obtained when measured with a manometer. In addition, oscillations of 1-6 Hz can be performed as well.

Conclusion: HVPEP may be an effective adjunct to the ACT regimen in cystic fibrosis. It is easy to comprehend and teach. Subjectively, the patients that used the technique reported HVPEP as useful and that they would use it as part of their regimen. Based on these findings, HVPEP could be studied on a wider population and could potentially be added as a part of the ACT regimen. Because it is always "at hand," HVPEP allows patients to do a form of oscillatory PEP with no need of equipment at any time during the course of the day.

Nikander, K. 1 ; Geller, D.E. 2 ; Kesser, K. 2 ; Potter, R. 3 ; Metcalf, A. 3 ; Hardaker, L. 3 1. Respiratory Drug Delivery, Philips Respironics, Parsippany, NJ, USA; 2. Nemours Children's Clinic, Orlando, FL, USA; 3. Respiratory Drug Delivery, Philips Respironics, Chichester, United Kingdom Introduction: Inhaled drugs for the treatment of CF are often approved for use with the specific nebulizers used in the approval process. However, in practice a patient may be issued a different nebulizer to deliver a range of drugs; therefore, the performance characteristics of these nebulizers for use with CF drugs are of interest. We compared the particle sizes, delivered doses and treatment durations of SideStream Plus (Philips Respironics, Parsippany, NJ) and LC Plus (Pari Respiratory Equipment Inc., Midlothian, VA) nebulizers when used to deliver three drugs commonly used in the treatment of patients with CF. Surface tension and viscosity of the drugs was also examined.

Method: Five SideStream Plus nebulizers and five LC Plus nebulizers were tested with each of the three drugs: Dornase alfa (Pulmozyme 1 mg/mL), Tobramycin Inhalation Solution (TOBI 60 mg/mL), and hypertonic saline (Hyper Sal 7%). Inspiration Elite compressors (Philips Respironics) were used to run the nebulizers with dornase alfa and hypertonic saline, a Pulmo-Aide compressor (DeVilbiss Healthcare, PA) was used for tobramycin. Tests were conducted across two laboratories. Delivered dose was assessed using simulated breathing (Vt=500 mL, f=20 bpm, I:E=1:1.5) by collecting aerosol onto a filter. Aerosol was collected until one minute past the onset of sputter (treatment duration), and aerosol deposits were assayed using: spectroscopy (dornase alfa), High Performance Liquid Chromotography (tobramycin), and freezing point depression (hypertonic saline). Particle size was determined using laser diffraction (Spraytec, Malvern Instruments Ltd, Worcestershire, UK), surface tension using a handheld tensiometer (SITA t60, Messtechnik GmbH, Germany), and viscosity using a microviscometer (AMVn, Anton Paar GmbH, Austria).

Results: The performance characteristics were similar between the two nebulizers for all but the delivered dose of dornase alfa (Table) . We would not expect this small difference to be clinically relevant, given the flat doseresponse curve for dornase alfa. The surface tension and viscosity of the drugs was such that no correlation could be made to particle size.

Conclusions: Though in vitro results can not be directly translated to clinical efficacy, the results of this study suggest that the two nebulizers would yield similar clinical outcomes when delivering the drugs tested.

Rationale: While improved treatments for cystic fibrosis (CF) have led to better survival, the disease remains characterized by a steady rate of decline in lung function.

Purpose: To evaluate the longitudinal relationship between habitual physical activity (HPA) and rate of decline of lung health in patients with CF.

Methods: We measured HPA and lung function (forced expiratory volume in 1-sec, FEV1) over a 7-yr period (n=212, 7-18 yrs at baseline; females n=107). Patients performed pulmonary function testing and completed a Habitual Activity Estimation Scale (HAES) at clinical visits (every 12-14 wks, while participating in their habitual activity, excluding visits associated with acute illness).

Results: The rate of overall increase in weekday total activity was 0.27 hours/day/year (p<0.0001) and the rate of decline in FEV1 was -1.7%/year (p<0.0001). The individual rates of change of HPA were further classified into two groups, above (HIGH) or below (LOW) the overall average rate of change. We found a significant correlation between rates of change of HPA

Browning, G. 1 ; Conway, E. 1 ; Wooldridge, J.L. 2 1. Cincinnati Children's, Cincinnati, OH, USA; 2. Pulmonary Medicine, Cincinnati Children's, Cincinnati, OH, USA Background: Airway clearance techniques (ACT) encompasses many techniques from use of mechanical devices to teaching patients breathing techniques which allows for movement of secretions. Literature has shown each technique works effectively as long as the technique is performed correctly. Cincinnati Children's revamped their practice of airway clearance through writing of protocols and education of their respiratory therapist and patients. Through these efforts we achieved a better program for our patients with cystic fibrosis. We recognized some patients needed more support then others to achieve best practice of ACT along with the need for consistent communication of ACT between the in-patient and out-patient units.

The bedside RT found it challenging to give extra time to patients who clearly needed extra support with airway clearance. We explored the possibilities of hiring an Airway Clearance Specialist which will allow extra time with our patients and provide a consistency of communication from our inpatient to out-patient setting.

Method: A proposal was given to our Senior Leadership of our CF unit for the position of Airway Clearance Specialist. The position was approved and hired in July of 2008. We filled the position with an RT who worked with our CF patient population. The Airway Clearance Specialist role is utilized to facilitate input in regards to patient ACT with our Pulmonary Team, is an expert in ACT modalities and finding the best modality for each patient in the in-patient and out-patient setting, provides educational support for patients and families to ensure ongoing proper techniques, and provides the much needed communication between our in-patient and out-patient setting. The Airway Clearance Specialist is utilized to assess patient's adherence to ACT through bi-annual assessments of ACT modalities. Families and patients now have a consistent resource for ACT.

Outcome: Cincinnati Children's CF Center has two hundred patients and our Airway Clearance Specialist met with each patient and their families twice during the first year. During these visits 184 patients were introduced to a new modality of ACT and 16 patients reviewed their current ACT modality. During the first year a comparison of FEV1 was calculated to see the impact of this role. Of the 151 patients who performed a PFT, 44% showed an improvement or stayed the same with an average increase of 12%. A survey asking our patients if they felt having our Airway Clearance Specialist review their ACT twice a year was beneficial showed 83% thought it was beneficial, 17% felt it was not beneficial or were not sure. We also asked if they felt their knowledge of ACT improved: 73% thought their knowledge improved, 17% felt their knowledge didn't improve or were not sure.

We feel the addition of our Airway Clearance Specialist has significantly improved our center's goal of providing best practice in airway clearance. In addition to seeing our CF patients, physicians are utilizing the expertise in other patient populations.

Sheridan, H.S. 1, 4 ; Davies, G. 2, 4 ; Bell, N.J. 3, 4 ; Reid, P.A. 3, 4 ; McLeod, F. 2, 4 ; Saunders, C. 2, 4 ; on behalf of the UK CF Gene Therapy Consortium, .. 4 1. Royal Hospital for Sick Children, Edinburgh, United Kingdom; 2. Department of Gene Therapy, Imperial College, London, United Kingdom; 3. Western General Hospital, Edinburgh, United Kingdom; 4. UK CF Gene Therapy Consortium, London, United Kingdom Introduction: Exercise capacity is predictive of mortality in CF. Objective measurement of daily physical activity may be related to exercise capacity. Participants in the UK CF Gene Therapy Consortium Run-In study had a field-based estimate of exercise capacity and objective measurement of physical activity around 4 study visits performed at times of clinical stability at intervals of [3] [4] [5] [6] p<0 .0001, mixed model analysis), with a higher increase in activity associated with a less steep rate of rate of decline of FEV1. In addition, the rate of decline in FEV1 was less steep for HIGH, when adjusted for the disease-specific factors of gender, age, mucoid Pseudomonas aeruginosa and cystic fibrosis related diabetes (p=0.0074). When the analysis was performed with HPA as the outcome and change in FEV1 as the covariate, the effects were less significant, supporting our hypothesis that a change in activity has an effect on FEV1, rather than vice-versa.

Conclusion: Patients with an above average rate of change in HPA demonstrated the slowest rate of decline in FEV1 over the study period, offering the CF clinical team a valuable opportunity to intervene with the prescription of regular physical activity, to positively impact lung function decline in patients with CF.

Ruf, K. 1 ; Fehn, S. 1 ; Bachmann, M. 2 ; Möller, A. 2 ; Hebestreit, H. Wuerzburg, Germany; 2. Kinderspital, Zürich, Switzerland Regular physical activity (PA) has become part of the treatment of cystic fibrosis (CF). Despite the importance of PA for patients' health and quality of life in CF, the amount and intensities of the patients' activities are not regularly assessed in the clinical setting and PA has not yet been incorporated as a variable in CF registries. This is probably due to the fact that there are almost no validated means to quickly, easily and reliably assess the level of PA in the CF population. Therefore, the main study objective was to validate questionnaires with the help of accelerometry and exercise testing in a group of patients with CF.

A population of 41 patients with CF (age 12-42 years, FEV1 31-126%) underwent an incremental exercise test according to the Godfrey protocol and wore an accelerometer (GT1M) for 7 days on the right hip. Time spent in moderate to vigorous physical activity (MVPA) was determined by accelerometry as minutes per day with an accelerometer reading above 1000 counts per minute. Furthermore, the patients completed three activity questionnaires: the Habitual Activity Estimation Scale (HAES), the 7 Day Physical Activity Recall questionnaire (7D-PAR) and the Lipid Research Clinics questionnaire (LRC). Data were expressed in minutes spent in a certain activity category per day (7D-PAR and HAES) or the individuals activity level (LRC). The results of the accelerometry were related to the results of the questionnaires by calculating intraclass correlation coefficients (ICC) (for HAES and 7D-PAR) or Spearman rank correlation coefficents (LRC). Furthermore the results of the exercise test (maximal power in %predicted and peak oxygen uptake in %predicted) were related to the results of the questionnaires by using Spearman rank correlation coefficents.

In the ICC analyses, the categories "active+very active" as well as the category "very active" of the 7D-PAR were significantly related to MVPA (ICC=.404, ICC=.408 respectively, both p=.004) whereas no significant intraclass correlations were seen between the HAES categories "somewhat active+active" or "active" and MVPA. The results of the LRC were also not significantly related to MVPA. When relating the exercise test results to the activity levels assessed by questionnaire, significant correlations were only seen for the LRC (maximal power: r=.466, p=.002, peak oxygen uptake: r=.032, p=.041).

In conclusion, the 7D-PAR questionnaire best reflected MVPA in an adolescent and adult population with CF. However, the 7D-PAR can only be used as a means to describe PA within a population and, thus, for epidemiological studies based on CF registries but not to generate valid PA data for the individual patient, since the ICC between accelerometry and 7D-PAR was not strong enough. Likewise, the LRC can be employed to estimate the physical fitness within a population but similar to the results of the 7D-PAR the correlation is too weak for assessing individuals.

Button, B.M. 1, 4 ; Rasekaba, T. 2 ; Wilson, J.W. 1, 4 ; Holland, A. 2, 3 1. Department of AIRMED, The Alfred Hospital, Melbourne, VIC, Australia; 2. Department of Physiotherapy, The Alfred Hospital, Melbourne, VIC, Australia; 3. School of Physiotherapy, La Trobe University, Melbourne, VIC, Australia; 4. Department of Medicine, Monash University, Melbourne, VIC, Australia Physical activity and exercise are vital components of ongoing therapy for people with cystic fibrosis. Measurement of these factors forms an important feature of ongoing self-management and research of exercise rehabilitation in cystic fibrosis. We aimed to assess the validity of the International Physical Activity Questionnaire (IPAQ) in the adult CF population by comparing energy expenditure measured by the IPAQ versus the accelerometer.

Methods: With Alfred Hospital Ethics Office approval we prospectively recruited suitable successive adult patients with CF attending the CF Outpatient Clinic at the Alfred Hospital, Melbourne Australia. Patients were required to be at baseline state and function. Participants wore an accelerometer (Actigraph GT1M) secured around the waist for seven days, at the end of which they completed the IPAQ. Criterion validity of the IPAQ was assessed by comparing the IPAQ weekly energy expenditure (EE) in kilocalories (kCal) with weekly EE (kCal) from the accelerometer using Spearman correlations and Bland-Altman procedures.

Results: Thirty participants (53% females) completed the assessment: mean (SD); age = 29 (7) yrs, FEV1 %predicted = 61 (25)%, BMI = 22 (3) kg.m -2 . All participants wore the accelerometer for 7 days of awake time. The median (range) EE: IPAQ = 3695 (113, 19573) kCal, GT1M = 1669 (355, 5443) kCal. Spearman correlations of FEV1 %predicted with EE were GT1M EE r = 0.68, p<0.001; IPAQ EE r = 0.28, p>0.05. Correlation of the IPAQ EE with accelerometer EE was moderate (r = 0.46, p = 0.010). There was a trend towards higher EE measured by the IPAQ than measured by the accelerometer (Wilcoxon signed ranks test: z = -3.4, p=0.001). Bland-Altman plot showed poor agreement between energy expenditure from the two measures, with limits of agreement from to -11423 to 5550 kcal.

Conclusion: The IPAQ appears to underestimate physical activity for patients with lower energy expenditure activities and overestimate for those with higher energy expenditure activities in adults with CF. The IPAQ would be a useful clinical screening tool for individualized exercise prescription and monitoring of physical activity longitudinally, but assessment of exercise participation for research needs to be informed by more quantifiable methods such as the accelerometer.

Discussion: This program conformed to the components of physical WCXR scores, FEV1, and FEV1/FVC. In particular, WCXR scores were negatively related to parents' reports of Health Perceptions, Respiratory Symptoms, and Physical Functioning (p<0.003). FEV1 was significantly related to parent report of CFQ domains of Health Perceptions, Respiratory Symptoms, Physical Functioning, Weight, and Body Image (p<0.008). FEV1/FVC was only significantly related to parent report of CFQ Physical Functioning (p=.001).

In children < 14 years, significant relationships were found between self-reported CFQ Physical Functioning and Respiratory Symptoms and WCXR scores (p<0.007) as well as FEV1/FVC (p<0.006). In adolescents > 14 years, WCXR scores were significant related to self-reported CFQ Weight and Respiratory Symptoms (p<0.001). FEV1 was significantly related to self-reported CFQ Weight, Respiratory Symptoms, and Physical Functioning (p<0.003).

There were more significant associations between objective measures of the child's pulmonary health and parent-reported CFQ domains than those self-reported by children. Parent and child reports showed agreement on all but two domains of the CFQ Child-version: Body Image and Digestive Symptoms (p<0.0001). However, parent and child reports disagreed on all but 3 domains of the CFQ-Adolescent/Adult version. Differences between parents and their adolescent children were significant for Treatment Burden, Eating, Emotional State, Physical Functioning, Body Image, Digestive Symptoms, and Weight domains (p<.001). Patients > 14 years consistently rated themselves higher than their parents. Based on objective measures. most patients had mild lung disease. Patients and their parents tended to report high levels of HRQOL.

Conclusion: Maturation affects children's self-perceptions of HRQOL. The increased difference between parents and their older children on measures of HRQOL is consistent with normative developmental task of adolescent individuation. Parents' reports tended to be more closely aligned with objective measures of pulmonary health. Findings underscore the value of understanding the perspectives of parents as well as their children and adolescents regarding HRQOL.

Supported by CFF #A001-5-01, NIH #DK 34108, #M01 RR03186, #1UL1RR025011, and Children's Hospital of Wisconsin # M01RR00058. Aim: To examine the effect of a 12-week pilot training program on muscle mass and strength, lung function and exercise capacity in an adult CF population.

Method: Patients completed the 12-week training which included three 30 minute exercise sessions per week. One of the three sessions was supervised and included an additional 30 minute educational component on various types of exercise including resistance and cardio-vascular training. Training was based on multiples of metabolic equivalents of the resting energy expenditure (METs) dependent on FEV1 percentage predicted. Those with over 70% predicted FEV1 completed 8-14 METs activity; those with 40-69%, 4-9METs; and those patients with FEV1 <40%, 2-3 METs in short intervals totalling 30 minutes. Fat free mass (FFM), body mass index (BMI), spirometry (FEV1), multi-stage shuttle walk test (MSSWT), handgrip strength and exercise self efficacy scores (ESES) were recorded at the start and end of the intervention. Inpatient days were recorded for each patient in the preceding and subsequent 6 months of the intervention period.

Results: Seventeen patients started the program. Nine patients dropped out during the 12-weeks. Eight patients completed the program. In those who completed the training FFM increased by an average of 4.3%, BMI increased by 3.2% (average BMI=23.9), FEV1 increased by 11%, MSSWT by 0.3%, grip strength by 25%, press up to failure by 21% and ESES by 14%. Hospital days reduced from 80 days in the 6 months prior to the intervention for the entire group to 40 in the subsequent 6 months following the training period. Reasons given by patients for dropping out were recurrence of illness requiring IV antibiotics, financial and time constraints, depression and one patient died (due to an unrelated cause). Ankle joint muscles play an important role in maintaining balance and performing activities of daily living. Peripheral muscle function is diminished in adults with cystic fibrosis (CF), however studies in children, hampered by small cohorts which did not account for differences in body habitus, are equivocal.

Objectives: To compare plantar flexor strength, adjusted for body size and composition, in 5-to-18-yr-old Caucasian children with and without CF and pancreatic insufficiency (PI).

Methods: Anthropometry and DXA for body composition were measured and associated Z-scores generated. Plantar flexion isometric maximal voluntary contraction (MVC) torques of the left ankle at 4 angles (-10, 0, 10, 20°) were assessed with the Biodex Multi-Joint System 3 Pro (Biodex Medical Systems, Inc, Shirley, NY) .

Results: Sixty-one children with CF (FEV 1 = 99 ± 22% predicted) and 54 healthy control children did not differ by age (10 ± 2 vs. 9 ± 2 yrs), sex (male, female: 37, 24 vs. 24, 30) or maturation stage (Tanner 1, 2: 43, 18 vs. 36, 18) . Children with CF had significantly lower (all P < 0.04) height Z (-0.5 ± 0.9 vs. 0.3 ± 0.8), weight Z (-0.4 ± 0.8 vs. 0.3 ± 0.8), BMI Z (-0.2 ± 0.8 vs. 0.2 ± 0.9), arm circumference Z (-0.4 ± 0.9 vs. 0.2 ± 1.0), upper arm muscle area Z (-0.5 ± 0.9 vs. 0.3 ± 0.9), whole body lean mass (21 ± 6 vs. 24 ± 6 kg) and whole body lean mass-for-height Z (-1.3 ± 0.9 vs. -0.07 ± 0.9) compared to healthy controls. Unadjusted MVC torques were significantly lower in children with CF compared to healthy controls at each angle (-10°: 43 ± 22 vs. 56 ± 19; 0°: 39 ± 19 vs. 50 ± 20; 10°: 33 ± 15 vs. 42 ± 16; 20°: 26 ± 11 vs. 34 ± 14 Nm; all P < 0.01). Group differences persisted when the highest MVC torques (at -10°angle) were adjusted for left leg length, lean and fat mass using multiple regression, together explaining 65% of the variance.

Conclusions: After adjusting for growth and body composition deficits, plantar flexor strength is attenuated in children with CF compared to healthy children. These results highlight the importance of improving growth status and body composition in children with CF and suggest that additional factors contribute to the decreased performance. Background: Conflicting literature exists about the relative benefits of Hospital in the Home (HITH) programs for management of acute respiratory exacerbation in children with cystic fibrosis (CF). A CF HITH program which includes nurse managed intravenous antibiotics (IVABs) and twice daily chest physiotherapy, to fully mirror conventional hospital care, has not been evaluated. This study evaluated outcomes of HITH versus conventional admission for acute respiratory exacerbation in the paediatric CF population. The Mater Children's Hospital (MCH) has delivered HITH since June 2008, providing twice daily physiotherapy and daily delivery of IVABs. Outcome measures of length of stay (LOS), spirometry and weight change were compared with spirometry data and previous admission data for the 12 months prior.

Methods: A retrospective audit over 18 months following the introduction of the paediatric CF HITH program. The HITH admission comprised an initial hospital stay for intravenous access, antibiotic levels and physiotherapy, followed by home treatment. Eligible children were transferred to home for care when medically stable, within geographic proximity, and parental consent gained. Once home, the CF medical team assessed weekly, with daily home visits by nursing staff for intravenous antibiotics and twice daily home visits by a physiotherapist for airway clearance and exercise therapy. Carers had access to nursing staff 24 hours a day via an on-call system.

Results: There were 30 children who underwent 50 HITH admissions over the timeframe, aged 7.86 years (mean, range 1-17years, 15 girls). Hospital re-admission from HITH occurred in 2 children (one admission each) due to onset of new symptoms and non-compliance. HITH LOS comprised 6 days hospital (mean +/-3.02sd) followed by 9.4 days at home (mean +/-4.22sd). Analysing the first HITH admission for each child, FEV 1 (%predicted) improved over the admission, with no difference in level of improvement compared with prior hospital admission. There was no difference between FEV 1 on HITH discharge and best value for the 12 months prior (n=21, 82.35% +/-14.15sd, p=0.113) .

Conclusions: The HITH treatment for respiratory exacerbation in children with cystic fibrosis was equivalent to hospital care (twice daily physiotherapy, and nurse supervised intravenous antibiotics). This resulted in outcomes which were equivalent to hospital-based care, with improvement in FEV 1 to usual baseline levels. Objective: To compare the effect of hypertonic saline (HS) inhalation before, during or after airway clearance techniques (ACT) on lung function and subjective outcomes in adults with CF.

Methods: Forty-four adults, mean (SD) age 31 (9) y, FEV1 55 (21) %pred, were randomised to three timing regimens of HS inhalation over three days. Each day participants performed baseline spirometry (FEV1 and FVC), had a bronchodilator, and then inhaled HS either before, during, or after ACT, in random order. Two hours later, participants recorded ease of expectoration, tolerability and satisfaction on a visual analogue scale (VAS). In a subset of 27 participants, spirometry was repeated at this time. Another subset of 10 participants repeated the entire 3-day study on a second occasion, recording VAS data and their preferred timing regimen, to determine repeatability.

Results: No outcomes significantly differed when inhalation before ACT was compared to during ACT. Ease of expectoration was significantly worse when HS was inhaled after ACT than before or during ACT (mean differences 11 [95%CI 2 to 20] and 16 [6 to 26] , respectively). Similarly, satisfaction was significantly worse when HS was inhaled after ACT than before or during ACT (mean differences 20 [10 to 30] and 25 [3 to 25] , respectively). Lung function did not significantly differ between any of the three timing regimens. Among the 10 participants who repeated the study, VAS data had wide limits of agreement. Despite this, 9 of the 10 participants were consistent in their selection of their preferred timing regimen. In 7 of these, inhalation of HS before ACT was preferred.

Conclusion: There is no significant impact on lung function due to the timing of HS inhalation with respect to ACT. Subjective perceptions of treatment efficacy and satisfaction were significantly lower with HS inhalation after ACT than the other timing regimens. It is likely that a patient's preferred timing regimen remains constant over time.

in a positive relation to ∆lactate (6-3min) (r=0.4 p<0.01), but not in significant relation to SPO2% at 6 min exercise (r=-0.01).

Conclusion: Impaired O 2 delivery to skeletal muscle may slow VO2 kinetics during heavy exercise in hypoxemic PWCF. Alterations in skeletal muscle metabolism resulting in increased blood lactate accumulation may help to explain the difference in VO2 slow component amplitude between PWCF and healthy control. Exercise training may reduce VO2 slow component amplitude in PWCF.

Introduction: Peak oxygen uptake (VO2peak) is considered to be the most accurate parameter to assess exercise capacity in healthy people and cystic fibrosis patients (CF). Numerous equations have been published to predict VO2peak in healthy people. Currently there are insufficient normative data to predict VO2peak in CF. The purpose of this study was to create a CF-specific equation to predict VO2peak.

Methods and Results: A total number of 590 cardiopulmonary exercise tests (CPET, n=250 female aged from 8 to 49 and n= 340 male CF patients, aged from 9 to 46 years) were used to create a CF-specific equation to calculate VO2peak for females and males, respectively. All CPET were performed on a cycle ergometer using Godfrey protocol. Multiple stepwise regression analyses identified FEV1, VC, MEF25 and age, weight and height to explain 62% and 58% of the variance of VO2peak in females and males, respectively. The so built prediction models were subsequently evaluated with a different cohort of CF patients (n= 103 females aged from 10 to 43 and n=97 male aged from 10 to 41 years). A significant correlation (p<0.05) was found between predicted and measured VO2 peak in females (r=0.779, p<0.01) and for males (r=0.720, p<0.01) with CF.

Conclusion: The present study was the first to provide and evaluate prediction equations for VO2peak in female and male CF patients based on anthropometric and pulmonary function measures. The tested patients reflected a wide range of severity of disease and age. The regression models identified in the present study still has limitations with respect to an accurate prediction in selected individuals. However, it predicted VO2peak in CF patients equally well as corresponding procedures specifically developed for healthy people or patients with selected other chronic diseases. 3 1. School of Health and Human Performance, Dublin City University, Dublin, Ireland; 2. School of Sport and Health Sciences, University of Exeter, Exeter, United Kingdom; 3. Department of Medicine Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin, Ireland Background: There are many mechanisms involved in slowing overall oxygen uptake (VO2) kinetics and increasing the O 2 deficit and amplitude of the VO2 slow component in individuals with lung disease. Measurements of VO2 kinetics and lactate concentration during heavy constant load exercise may provide important information about O 2 transport and utilization in people with cystic fibrosis (PWCF). The purpose of this study is to evaluate VO2 mean response time (MRT), slow component amplitude, O 2 deficit, and blood lactate accumulation during heavy cycle exercise in PWCF compared to healthy controls.

Methods: A total of 26 people with PWCF (25±5 yr, f=7), with wide range of disease severity (FEV1% predicted=64±25) and 17 healthy controls (21±2 yr, f=8) participated. All subjects performed a maximal exercise test on a cycle ergometer. The PWCF performed a high-intensity constant load exercise test at 70% VO2peak (100±41 W). Healthy subjects performed a submaximal constant load test at the same absolute work rate corresponding to 70%VO2peak of the PWCF (100 W). Peripheral blood lactate concentration was measured from the ear lobe every minute during exercise.

Results: VO2peak was higher in healthy controls than PWCF. MRT and O 2 deficit were higher in PWCF than healthy controls. For the slow VO2 component phase, ∆VO2 (6-3min) and ∆lactate (6-3min) were higher in PWCF than healthy controls. For all the subjects combined, there was both an inverse relation between MRT and SPO2% at minute 3 during exercise (r=-0.43 p<0.01), and a positive relation between MRT and ∆lactate (6-3min) (r=0.45 p<0.01). For all the subjects combined, ∆VO2 (6-3min) was * p <0.05, ** p <0.001, vs. healthy controls, p <0.05 vs. mild CF.

Morgan, G.E.; Lord, V.; Hunt, C.; Parrott, H.; Agent, P.; Simmonds, N.; Bilton, D. Adult CF Unit, The Royal Brompton & Harefield NHS Foundation Trust, London, United Kingdom Introduction: Exercise is an essential component in the long term management of CF. It is an adjunct to airway clearance, and benefits the cardiovascular system and psychological well being. Exercise is encouraged as part of lifestyle, but usual regimens can be disrupted by hospitalisation. We assessed our inpatient exercise regimens to evaluate the training effect achieved.

Aim: To evaluate the training effect created by inpatient exercise programmes using the "Firstbeat SPORTS" software.

Method: Adults with CF admitted to our unit were offered the opportunity to use the Firstbeat SPORTS software (consists of a thoracic belt or ECG style monitor). Subjects were stratified for analysis depending on baseline FEV 1 % predicted on admission (severe <40%, moderate 40-70%, mild >70%). Exercise regimes were prescribed and included treadmill, cycling, walking or the Wii games console. The programme analysed heart rate data to estimate EPOC (Excess Post-Exercise Oxygen Consumption), and translated this into training effect. The programme scales this into 5 categories from 1-5 (minor to overreaching), depending on how much the exercise has improved maximal aerobic capacity and fatigue resistance. The closer to 5 the training effect is, the more demanding the training has been.

Results: Twenty-one patients (11 female, 10 male) completed 50 exercise episodes (range 1-7 episodes per patient). Mean age was 24 years (16-48yrs), mean FEV 1 1.39L (39%) (0.48-3.84L). Mean exercise time 20 minutes (2 to 63 minutes).

Conclusion: Although limited by small subject numbers, this study illustrates that our exercise service does offer a "fitness maintaining effect" to inpatients as defined by the programme. While this is to be encouraged to help patients continue with everyday life, for some patients an "improving effect" is desirable. The data indicates that use of the treadmill or bike has a greater training effect than the Wii or walking. However, it is clinically important to note that similar effects can be achieved with static bikes or Wii, as these are useful for patients in isolation. The data indicates that our patients in the severe group achieved a "fitness maintenance effect" while more of those with moderate disease were achieving an "improving effect." Interestingly, our mild group achieved the lowest training effect, indicating they need to be pushed more. The Firstbeat system increased patient interest in exercise as a new way of showing them fitness improvements and reasons for modifications. This software utilising EPOC allows us to produce a specific exercise prescription. Background: As patients (pts) with CF grow older, they develop chronic airway infection and often take nebulized medications. Home nebulizer equipment can become contaminated with bacteria and this may lead to more pulmonary exacerbations in pts with CF. There is evidence that pts with CF do not disinfect nebulizers routinely.

Hypothesis: We hypothesize that a significant portion of our adult CF population are improperly disinfecting home nebulizer equipment. We implemented a program to assess and re-educate pt knowledge of home nebulizer equipment disinfection.

Method: In February 2010, our RT staff began the assessment of pt home nebulizer disinfection in our outpatient adult CF clinic setting. Pts are asked the following two questions 1) Does the pt clean his/her nebulizers? 2) How does the pt clean his/her nebulizers?

The following six disinfection techniques were deemed appropriate: 1) Boiling for 5 minutes; 2) Microwaving in water for 5 minutes; 3) Dishwasher for 30 minutes (if water temperature is > 158 ᭺ F); 4) Soak in 1:50 solutions of bleach and water for 3 minutes; 5) Soak in 70% isopropyl alcohol for 5 minutes; or 6) Soak in 3% hydrogen peroxide for 30 minutes. Based on individual pt interview results, pts were divided into two groups: use of a proper disinfection technique or an improper cleaning technique.

All pts using an improper cleaning technique are given written and verbal cleaning instructions. Also barriers to proper disinfection are identified and a problem-solving approach utilized to overcome barriers. Pts are asked to choose a technique to implement at home. Pts are assessed at each clinic visit and re-educated.

Results: To date, 57 adult patients with CF have been interviewed. Five patients were excluded from further analysis, as they were not taking nebulized medication. Preliminary data analysis of the remaining 52 patient interviews show that 42.3% (n=22) of our adult CF patients taking nebulized medication are using proper home nebulizer equipment disinfection.

Conclusion: Adult patients with CF have significant knowledge deficit in the proper techniques of home nebulizer equipment.

Future Goals: This study is ongoing. We will continue to assess and educate this knowledge deficit in these adult patients with CF. Further analysis will include FEV1 assessment and exacerbations. Ultimately, we will examine if addressing this knowledge deficit will lead to a higher FEV1 and fewer pulmonary exacerbations in our adult CF program. whether exercise has also an effect on inflammation and can improve phagocytosis of Pseudomonas aeruginosa in patients with CF. Methods: Blood was taken before and after performance of a maximal cardiopulmonary exercise test (CPET). Total leukocyte and subset counts, CRP, BSE, IgG, and the capacity of leukocytes to phagocytose Pseudomonas aeruginosa were determined. Exercise capacity was expressed as maximal oxygen consumption (VO2max) and lung function as the forced expiratory volume in 1 second (FEV1). Differences were calculated with the Wilcoxon Signed Rank Test.

Results: Seventeen patients with CF (7 boys; 8 girls, all dF508/dF508) with a mean (SD) age of 14.6 (1.8) years performed a CPET. Mean (SD) for FEV1%pred was 80.7 (22.5) % and for VO2max %pred 75.8 (21.0) %. Exercise induced a statistically significant increase in the leukocyte content (p=0.001) from a median (IQR) of 6.1 (4.9) to 11.2 (7.8) x10 9 /L. Monocytes showed the largest increase (103.9%), followed by lymphocytes (52.7%) of which NK cells increased most (97.9%). Statistically significant increases were also seen for IgG total (3.2%), IgG1 (4.1%), IgG2 (0.7%) and IgG3 (6.8%), but not for CRP, BSE and IgG4. Results of the phagocytosis assays will follow.

Conclusion: In patients with CF exercise leads to a leukocytosis dominated by monocytes and NK cells. Small increases were also seen for IgG. Whether exercise also increases phagocytosis of Pseudomonas aeruginosa by leukocytes will be further investigated. Background: Musculoskeletal impairments are common in patients with cystic fibrosis (CF), including postural abnormalities and impaired peripheral muscle performance. However, there are no standard, reliable outcome measures to assess these impairments. The purpose of this study was to determine the reliability of two clinical tests for posture and muscle performance, the acromial angle distance and the vertical jump, respectively.

Materials and Methods: Five subjects without known pulmonary disease participated in this initial, pilot-feasibility study. Acromial angle distance was measured as the distance from the acromial angle (the prominent angle at the junction of the posterior and lateral borders of the acromion of the scapula) to the wall with a flexible tape measure while subjects stood erect with heels, buttocks, thorax, and head against the wall. Measures were taken by 2 raters on day 1 to assess interrater reliability. Two days later the measures were repeated to assess intrarater reliability. Vertical jump was assessed on the "Just Jump" system (a ~90 cm 2 mat connected to a timer; time off the mat (seconds) was converted to vertical jump (cm)) using both a controlled (90°) and uncontrolled knee angle on day 1. Test-retest reliability was then assessed with repeat measures later that week. Pearson correlation coefficient was used to assess reliability for all tests.

Results: Interrater reliability for acromial angle distance was 0.9764 (p=0.0008) and the intrarater reliability for each investigator was 0.9975 (p=0.0451) and 0.9998 (p=0.0119). Vertical jump (cm) was no different with regard to whether knee angle was controlled or not (43.18 (SD 7.87) vs. 43.94 (SD 9.14), respectively; p=0.5619). Test-retest reliability of the vertical jump was 0.9688 (p=0.0066).

Conclusion: Both the acromial angle distance and vertical jump were highly reliable, required minimal time and equipment (a tape measure and the Just Jump system). These tests could easily be incorporated into the physical therapist's examination in the hospital or outpatient clinic and could prove to be a valuable outcome measure.

Christiansen, J.; Thompson, L.; McNamara, J.; Johnson, M.; Fenlon, K. Cystic Fibrosis Research, Children's Hospitals and Clinics of Minnesota, Minneapolis, MN, USA Background: Children with cystic fibrosis (CF) may have decreased bone mineral density, core muscle weakness, overuse of their respiratory muscles secondary to the disease process, and poor posture that may result in chronic pain. These factors may lead to impairments in their musculoskeletal and neuromuscular systems. Few studies address how core strengthening and respiratory muscle exercises may improve these impairments.

Aims: This pilot study examines the effects of an outpatient physical therapy intervention program with a clinic and at-home component using core strengthening and respiratory muscle exercises in five children with CF.

Methods: Children with CF ages 10-21 were eligible for this study. Two participants have completed the study; three are still receiving study treatment; and one was removed due to compliance issues. Participants receive core strengthening and respiratory exercise interventions weekly for six weeks and then monthly for four months in the clinic. They also complete a home exercise program a minimum of three times per week independently for six months. All participants are evaluated at baseline and at the six month follow-up using these outcome measures: PFTs including MIPS and MEPS, VO2max, rib cage mobility (circumference measurements of upper, middle, and lower ribcage during inspiration), posture (lordosis, kyphosis, forward shoulders, winged scapulae, forward head, and scapular position), and core strength (plank position, v-up position, and Pilates hundred in table top position as measured by seconds held; and isometric abdominal test, isometric internal/external abdominal obliques (IAO) test, and isometric dynamic horizontal (IDH) side support test as measured by a 1-5 manual muscle testing (MMT) scale). Study analysis assesses whether at least three of the outcome measures show a clinically significant change at six months when compared to baseline. A significant clinical change occurs if 1) PFTs increase by 5%; 2) VO2max increases by 5%; 3) rib cage mobility increases by 10%; 4) posture increases by two points on a 1-5 qualitative postural analysis (QPA) scale; and 5) core strength exercise holding time increases by 20% and by a 0.5 increase on a 1-5 MMT scale.

Results: Significant clinical improvements occurred in the first two participants as evidenced by baseline/follow-up measurements of the subsequent variables: Participant 1: PFT (pre bronchodilators FEV1: 2.53/2.77 L); posture by QPA scale (lumbar lordosis: 1/3, forward shoulder: 3/5, and forward head: 3/5); and core strengthen (plank: 30/90 sec, Pilates hundred: 56/100 sec, and IDH: 4/5 by MMT scale). Participant 2: PFT (post bronchodilators FEV1: 2.0/2.15 L and MIPS -73/-108 cm H 2 O); rib cage mobility (upper: 2/6 mm, middle: 7/10 mm, and lower: 9/12 mm); and core strengthen (plank: 10/60 sec, IAO: 2+/3, and IDH: 3/4).

Conclusion: Results suggest that musculoskeletal and neuromuscular impairments common in children with CF may improve with a comprehensive outpatient physical therapy intervention and a home exercise program. Based on these results, we will continue data collection with current participants. If these participants demonstrate similar clinical improvements, we will propose a larger controlled study.

Weert -van Leeuwen, P.V.; Beekman, J.; Ent, K.V.; Hulzebos, E.; Arets, B. Pediatrics, University Medical Centre Utrecht, Utrecht, Netherlands Introduction: Exercise induces immunological responses that mimics those induced by other physical stressors. Postexercise a transient leukocytosis and increased phagocytic capacity of neutrophils is seen in healthy people. Patients with cystic fibrosis (CF) suffer from pulmonary infections predominantly caused by Pseudomonas aeruginosa, which causes tissue damage leading to a decline in lung function and exercise capacity. Regular exercise has been shown to improve lung function, physical fitness, morbidity and survival in patients with CF. The aim of this study is to investigate Wielgus, E.; Cullina, J.; Prickett, M.; Lam, A.; Jain, M. Northwestern University, Chicago, IL, USA Patients with cystic fibrosis have significant treatment burden. Therefore good patient understanding of prescribed therapies and medications has the potential to improve patient perceptions and treatment adherence.

This study aims to evaluate the correlation between patient comprehension of the purpose of routinely prescribed pulmonary medications/therapeutics for cystic fibrosis, patient perceived efficacy, and the level of adherence in our adult cystic fibrosis population. This single-center, prospective study examined patients with cystic fibrosis age >18 years seen at routine clinic visits. From the period of June 2009-November 2009, consecutive patients seen at our CF center were provided with a questionnaire regarding eleven drugs/therapies typically prescribed for pulmonary complications in CF. On the questionnaire, patients were asked, in a multiple-choice format, to choose the mechanism(s) of action for each drug/therapy, how well they felt each drug/therapy worked, and the percentage of time they used the drug/therapy. Approximately 42 questionnaires were completed and collected. Patient compliance correlated highly with patient perceived efficacy for recombinant human dornase-alpha (R squared 0.985), airway clearance (R squared 0.950), and albuterol (R squared 0.970) but not for hypertonic saline (R squared 0.605) or exercise (R squared 0.0178). However there appeared to be no association between knowledge of the correct mechanisms of action for therapies and higher patient compliance. These results suggest that patient's perception of medication/treatment efficacy may affect patient compliance with the prescribed therapy. Certainly, patients should be informed regarding the purpose of their medications, but this knowledge seems to affect adherence less than patient perception of the intervention's efficacy. Further work will be performed to improve patient perception of efficacy and adherence to prescribed therapies.

Bongers, B.C.; Hulzebos, E.H.; Helders, P.J.; Takken, T. Child Development & Exercise Center, University Medical Center Utrecht, Utrecht, Netherlands Objective: In children with cystic fibrosis (CF), it would be more appropriate to perform sub-maximal exercise tests in order to examine cardiopulmonary fitness. As a substitute for maximal exercise tests, the oxygen uptake efficiency slope (OUES) has been proposed as an objective measure of cardiopulmonary fitness, which is independent of exercise duration. The aim of this study was to investigate whether the OUES is a valid sub-maximal parameter of cardiopulmonary fitness in children with CF.

Methods: Bicycle ergometry exercise tests with gas-analyses were performed in 12 patients with CF with a mean age of 15.24±0.82 years (range 13-17 years) and mild to moderate lung disease (mean FEV1 84.80±13.04% of predicted, mean RV/TLC ratio 32.64±5.06%). Peak oxygen uptake (VO2peak), peak minute ventilation (VEpeak), ventilatory threshold (VT), and, at different exercise intensities (using exercise data up to the VT, using the first 75% of the exercise data, and using 100% of the exercise data), the OUES were determined. OUES values were normalized for body surface area (BSA) in order to compensate for individual anthropometric differ-ences due to growth and maturation. Ventilatory equivalents for O 2 (VE/VO2-slope) and CO 2 (VE/VCO2-slope) were determined as well. Values were compared with 12 sex and age-matched healthy children with a mean age of 15.18±0.99 years.

Results: The VO2peak was significantly lower in children with CF with a significantly higher VE/VO2-slope and VE/VCO2-slope compared with healthy participants (see Table) . In children with CF, the OUES values determined at different exercise intensities differed significantly from each other. A significant difference between OUES values in healthy children and children with CF was only detected using exercise data up to the VT.

Conclusion: Due to its nonlinearity throughout progressive exercise, the OUES may not be a valid sub-maximal parameter of cardiopulmonary fitness in children with CF and mild to moderate lung disease. Whether the OUES is a valid indicator of cardiopulmonary fitness in a large sample of patients with CF with more severe lung disease needs additional research. 

Dellon, E.P. 1 ; Sawicki, G.S. 2 ; Wolfe, J. 3 ; Hanson, L.C. 4 1. Pediatrics, University of North Carolina, Chapel Hill, NC, USA; 2. Medicine, Children's Hospital Boston, Boston, MA, USA; 3. Psychosocial Oncology, Dana Farber Cancer Institute, Boston, MA, USA; 4. Medicine, University of North Carolina, Chapel Hill, NC, USA Background: Many patients with advanced CF lung disease receive life-sustaining treatments with little communication with physicians about these treatments preceding their use. Family caregivers of patients dying from complications of CF have identified limited communication with physicians as a major barrier to making informed choices about the use of life-sustaining treatments.

Objectives: We examined physicians' perspectives on current communication with patients with advanced CF lung disease and their caregivers about treatment preferences, and ways to improve communication about the use of life-sustaining treatments.

Methods: We surveyed CF physicians at two major CF care centers about barriers to discussing treatment preferences with patients and caregivers. Survey items addressed patient/caregiver, physician, and system/institutional factors, with Likert scale response options. We conducted follow-up interviews to explore these barriers and to elicit recommendations for improving communication. Summary statistics were used to analyze survey responses. Predominant themes from interview responses were identified using qualitative content analysis. patients). One more patient reached group A at the two years follow-up, but 2 were demoted to group B. The linear model used to explain the evolution of HbA1c found a non-significant average downward trend over time. Univariate analysis showed that an uncontrolled CFRD prior to transplant (OR=15, p=0.025) and a long delay between diagnosis of CFRD and transplant (OR=1.3, p=0.043) were significantly predictive of uncontrolled CFRD one year after transplant.

Conclusion: At one and two years of follow-up, lung transplant does not seem to worsen CFRD. Patients who would benefit from a simultaneous lung and pancreatic islet transplant are probably few and should be selected with criteria which remain to be determined. Background: As of 2010, screening newborns to identify infants at risk for cystic fibrosis (CF) is underway nationwide in the U.S. About two-thirds of states use high values of immunoreactive trypsinogen (IRT) and DNA testing (IRT/DNA) to screen newborns, while the other one-third repeats the IRT test in a second blood sample (IRT/IRT). The primary goal of newborn screening (NBS) is to diagnose, and thus initiate treatment of, CF in the neonatal period, i.e., before 28 days of life. However, it is unclear if IRT/DNA and IRT/IRT methods achieve this goal. Our hypothesis was that infants born in IRT/DNA states are diagnosed at younger ages as compared to those born in IRT/IRT states.

Methods: Data from the 2001-08 CFF Patient Registry were used to identify infants born in states that implemented CF NBS before January 2008. A total of 1611 infants from 36 states (23 IRT/DNA, 13 IRT/IRT) were identified. Infants diagnosed through prenatal screening (n=56, 3.5%) and those born with meconium ileus (n=267, 16.7%) were excluded, as their CF diagnoses were unlikely resultant from NBS.

Results: Overall, 968 infants diagnosed with CF in IRT/DNA states and 320 in IRT/IRT states were included in this study. Infants with CF born in IRT/DNA states had their initial visit to a CF Center at a younger age (median [interquartile range] 41 days [21, 106] ) than those in IRT/IRT states (54 [28, 97] ), p=0.008. Wisconsin had an even younger age at first CF Center visit (30 days [17, 85] vs. 43 days [22, 108] in other IRT/DNA states and 54 days [28, 97) in IRT/IRT states, p=0.015). Not all infants' CF diagnosis was reported to be as a result of NBS: in IRT/DNA states, 82% of infants had their CF diagnosis "suggested by NBS," versus only 62% in IRT/IRT states, p < 0.001. Of infants with CF diagnosis suggested by NBS, proportionately more were sweat tested in IRT/DNA than IRT/IRT states (88% vs. 82%, p = 0.02). For infants whose diagnosis was suggested by NBS and who were sweat tested, age at first CF Center visit was younger in IRT/DNA than IRT/IRT states (38 days [21, 86] vs. 46 days [29, 75] , p=0.02), although age at sweat test was not significantly different. Also, proportionately more of these infants were genotyped in IRT/DNA than IRT/IRT states (97% vs. 87%, p<0.0001), and age at genotype was younger (5 days [2, 29] vs. 35 days [28, 58] , p<0.0001). Of infants whose CF diagnoses were not suggested by NBS and who were not sweat tested, 92% were genotyped; age at first CF Center visit was younger in IRT/DNA states (n=40, 33 days [15, 110] than IRT/IRT states (n=50, 54 days, [21, 213] ), but this difference was not significant (p=0.67).

Conclusions: Infants with CF in states that use IRT/DNA to screen newborns for CF are more likely to have their diagnosis suggested by NBS, be sweat tested, and visit CF Centers at younger ages than those in states that use IRT/IRT. There are quality improvement opportunities in all states to shorten the time between birth and the initial CF Center visit. NBS has not achieved the goal of initiating CF treatment during the neonatal period.

Supported by NIH R01 DK072126 and DK34108.

Results: Surveys were completed by 30 of 34 (88%) eligible physicians, and interviews by 26 (76%). Physicians report a mean of 16 years experience in CF care, with half caring for >20 CF patients. Four of the 5 major barriers identified were patient/caregiver factors, including overly optimistic expectations about prognosis, overly optimistic expectations about effectiveness of life-sustaining treatments, severity of patient illness limiting participation in discussions, and disagreement within families about the use of these treatments; concern for taking away hope was the only physician factor identified as a major barrier. Other physician factors such as uncertainty about when to discuss treatment preferences, uncertainty about effectiveness of treatments, and insufficient training in communication were also identified as barriers but were felt to be less important. System/institutional factors were felt to be minimal barriers. Recommendations for improving communication encompassed 4 themes: developing standards or consensus about timing and content of communication, offering specific training for physicians in communicating with patients and caregivers about life-sustaining treatments, creating educational tools for patients and caregivers, and utilizing members of the multidisciplinary CF care team to facilitate communication.

Conclusions: While family caregivers feel physician factors limit communication about treatment preferences, CF physicians identify patient/caregiver factors as the major barriers to effective communication. Despite this perception by CF physicians, their recommendations for improving communication focus on changing physician factors which act as barriers to communicating with patients and caregivers about treatment preferences.

Supported by the National Palliative Care Research Center. Background: Cystic fibrosis (CF) related diabetes (CFRD) occurs in 15% of pancreatic insufficient patients, with prevalence increasing with age. Many mechanisms could lead to glucose intolerance, such as decreased insulin secretion caused by pancreatic islets fibrosis, and increased insulin resistance, impaired by chronic pulmonary sepsis. Evolution of CFRD after lung transplant has not been described. We hypothesized that in some patients, CFRD control is improved after lung transplant, possibly because of the suppression of a chronic inflammatory focus.

Methods: In a retrospective observational study including CF patients who underwent lung transplant, CFRD control was assessed by daily insulin dose, number of injections, fasting blood glucose level and HbA1c. Data were collected in the 3 months preceding transplant and during routine consultations at 1 and 2 years post-transplant. Diabetes control was considered satisfactory if HbA1c < 7% and insulin dose < 1 UI/kg/day (group A). Others were considered as uncontrolled (group B). A linear mixed regression was used to estimate HbA1c level mean. A logistic regression was performed to adjust uncontrolled CFRD at one year according to sex, age at transplant, weight, time between diagnosis of diabetes and transplant, corticosteroids and tacrolimus dosages.

Results: Data from 59 CF patients who underwent lung transplant were collected. Data from 4 patients were missed. Thirty (53.6%) were diabetic and received insulin. Two underwent simultaneous lung and pancreatic islet transplant and were excluded. At time of transplant, median age of the 28 included patients was 29 years old. Median time between diagnosis of CFRD and transplant was 2 years. Eighteen of 28 (64.3%) patients were in group A prior to transplant. Twenty-six patients were included at the one year follow-up (1 patient died in the first year following transplant, and transplant was too recent in 1 patient). Five of 10 (50%) patients who were in group B prior to transplant reached group A one year after. Sixteen patients were included at the two years follow-up (3 patients died in the second year following transplant, and transplant was too recent in 7 other

Thia, L.P. 1, 2 ; Stocks, J. 1 ; Hoo, A. 1, 2 ; Chudleigh, J. 1, 2 ; Prasad, A. 2 ; Lum, S. 1 ; Bush, A. 3 ; Wallis, C. 1, 2 1. Portex Respiratory Unit, UCL Institute of Child Health, London, United Kingdom; 2. Respiratory Unit, Great Ormond Street Hospital for Children NHS Trust, London, United Kingdom; 3. Respiratory Unit, Royal Brompton and Harefield Hospitals NHS Trust, London, United Kingdom Background: CF infants diagnosed by NBS are likely to be healthy. It is essential that treatment is started early to maintain lung health; novel therapies aimed at early intervention are now in the pipeline. Therefore, there is an urgent need for objective outcome measures to detect early lung disease in these infants.

Aims: To a) establish the nature and magnitude of early CF-related lung disease in order to inform power calculations for future studies; b) assess the feasibility of recruiting infants diagnosed by NBS into trials using lung function, bronchoalveolar lavage (BAL) and chest computed tomography (CT) as outcome measures.

Methods: Infants with CF diagnosed by NBS referred to centres participating in the London CF collaboration (East Surrey Teaching Hospital, Great Ormond Street Hospital, Kings College Hospital, Lewisham University Hospital, Royal Brompton Hospital and Royal London Hospital) are potentially eligible for this study; healthy controls (HC) are being recruited from the local community. Infant lung function tests (LFT); multiple breath inert gas washout and raised volume technique are performed at around 3 months and repeated at around 1 year of age in all infants. CT and BAL are performed in CF infants at 1 year.

Results: To date, initial LFT have been performed in 38 of 44 eligible infants with CF and 24 HC at (mean [SD]) 10.9 [2.4] and 12.8 [2.2] postnatal weeks respectively. Compared to HC, lung clearance index (LCI) was higher (mean difference [95%CI]) 0.38 [0.005; 0.76], p<0.05 and forced expiratory volume (FEV 0.5 ) z score lower -1.15 [-1.63; -0 .67], p<0.0001 in infants with CF. Of the 9 CF infants with a raised LCI (>8.3) by 3 months, 6 also had a reduced FEV 0.5 z score (<1.96). These preliminary results differ from those of the AREST-CF study (Linnane BM et al, AJRCCM 2008) . The second LFT has been performed in the 10 CF infants and 9 HC, who have now reached around 1 year of age, with CT and BAL also being successfully completed in those with CF and no loss to follow-up.

Conclusions: Despite some reviewers' concerns regarding parental reluctance to participate in such a study, our results to date suggest that recruitment of NBS CF infants to a multicentre clinical trial with invasive outcome measures (CT and BAL) is feasible. By 3 months, despite early diagnosis and treatment, lung function was significantly reduced in some infants with CF. Ongoing recruitment and follow up will allow us to investigate the evolution of lung function in NBS CF infants and its relationship with lung structure and inflammation.

Acknowledgements: This study is being undertaken by the London CF Collaboration and is funded by the GOSH Special Trustees and the CF Trust. D. 1 ; Hoffman, G. 2 ; Nugent, M. 3 ; Schneck, K. 4 ; K. 2 ; Rock, M. 5 ; Farrell, P.M. 5 ; Simpson, P. 3 ; Levy, H. 4 1. Genetics, Children's Hospital of Wisconsin, Milwaukee, WI, USA; 2. Wisconsin State Lab of Hygiene, Madison, WI, USA; 3. Quantitative Health Sciences, Medical College of Wisconsin, Milwaukee, WI, USA; 4. Pulmonary and Sleep Medicine, Children's Hospital of Wisconsin, Milwaukee, WI, USA; 5. Pediatrics, UW School of Medicine and Public Health, Madison, WI, USA Background: The potential impact of CF NBS on the incidence of CF has been reported by the Massachusetts NBS program. They found a decrease in the number of newborns with CF from 1999-2006. In contrast, the Colorado NBS program noted year-to-year variation, but no significant change from 1983-2006. To clarify the discrepancy, we evaluated the incidence of CF in Wisconsin (WI) over a 15 year period as detected by NBS.

Methods: NBS for CF using IRT/DNA testing has been performed in WI since 1994 with DNA testing of the top 4% of daily IRT results. The DNA mutation panel consisted of dF508 from 1994-2002 and the ACMG 23 panel since 2002. The total number of CF patients diagnosed by 2 CFTR mutations and sweat chloride of 60 or greater mEq/L (thus excluding CRMS) was collected by the WI NBS program from 1994-2008. WI birth rate data was obtained from the WI Department of Health Services. All data was categorized by ethnic background of the birth mother: white non-Hispanic (WNH), black non-Hispanic (BNH), Hispanic (HPC) and other. The incidence of CF births was adjusted for ethnic background. Loess smooth plot was used to fit the data and Pearson linear correlations were calculated.

Results: The incidence of CF in the WNH population and overall population decreased from 1994-2008. WNH diagnoses ranged from 2.9 per 10,000 births to 1.9 per 10,000 births with significant linear decline (r= -0.524, P< 0.046); overall it declined from 2.7 to 2.0 (r=-0.522, P< 0.046). There were too few cases in other ethnic backgrounds to detect a pattern.

Conclusion: There has been a decrease in new CF cases diagnosed by NBS over the past 15 years in the WNH population screened in WI. This is not due to changes in overall birth population demographics with regard to ethnic background or number of births. With the advent of universal NBS, this finding suggests a change in the composition of the CF cohort, influencing care and outcome. In the last 20 years, there has been great advancement in life expectancy of patients with cystic fibrosis (CF). However, improvements in CF outcomes may not be distributed to all demographics. Hispanic CF patients have worse pulmonary disease than non-Hispanic patients. Notably, in California, Hispanic CF patients make up one fifth of the CF population. There are few studies looking at the hospitalization and outcome patterns of the Hispanic patients compared to non-Hispanic CF patients.

Objective: To test the hypothesis that, in California, Hispanic CF pediatric patients are hospitalized more frequently and have greater rates of death than non-Hispanic CF patients.

Methods: This is a retrospective population study using the California Office of Statewide Health Planning and Development (OSHPD) hospital discharge database for the period of 1999-2007. Analysis was limited to Conclusions: In multivariate regression, warmer temperatures and public insurance were associated with worse lung function. Given that insurance status was the strongest SES predictor of lung CF lung function suggests that nationwide healthcare may have potential implications for CF and other chronic diseases. The lack of seasonal variation in lung function suggests that temperature may be serving as a proxy for non-seasonal, regionally based factors, which may include P. aeruginosa. Additionally, it is likely that other geographic determinants of health exist, warranting future research. 4 1. University of Rochester School of Nursing, Penn Yan, NY, USA; 2. Emory University and Children's Healthcare of Atlanta, Atlanta, GA, USA; 3. ICON, San Francisco, CA, USA; 4. University of Miami, Coral Gables, FL, USA Background: Previous studies have shown that women with CF who become pregnant have better lung function than those never pregnant and that pregnancy is not associated with short-term decline in clinical status or survival (1, 2) . Less is known about longer-term physiologic and functional outcomes for these women that might reflect competing demands of disease self-management and parenting.

Objective: To determine if there are differences in long-term physiologic and functional outcomes for women experiencing a pregnancy compared with those who never become pregnant.

Methods: This analysis was conducted using ESCF data from 1994 to 2005. Women reporting a pregnancy that could have spanned ≥30 weeks between 1994 and 2003 were each matched with 10 never-pregnant women using a propensity score that included age, race, genotype, pulmonary function, nutritional measures, reported complications, clinical symptoms, intensity of therapies, and measures of healthcare utilization. In a subgroup of 260 women (54 pregnant and 206 never pregnant), health-related quality of life (HRQoL) was measured using the CFQ-R (3). Group differences for FEV 1 , pulmonary exacerbations, and healthcare utilization in the endpoint period of 2004-2005 were assessed using repeated-measures models adjusting for matched sets. Group differences in clinical status and treatment were analyzed in the same period using conditional logit models, also adjusting for matched sets.

Results: Women (n=19) who met inclusion criteria for pregnancy were compared with 1190 matched controls. Median time between pregnancy and the endpoint period was 6.0 years (range 1.8-11.1). No significant differences were found in annualized change in lung function or BMI z-score between women who had been pregnant and controls. Report of cough but no other signs or symptoms was greater in women post-pregnancy (p<.05). Women experiencing pregnancy were treated for pulmonary exacerbations more frequently (p<.01) and had more illness-related clinic visits (p=.024), but there was no significant difference in routine therapies prescribed. There was a borderline increased risk of diabetes in previously pregnant women (OR 1.46; , and post-pregnancy, women reported lower HRQoL scores in physical functioning, vitality, health perceptions, and respiratory symptoms compared to controls (p<.01).

Conclusion: In the years following pregnancy, women did not experience worsening pulmonary function or nutritional status compared with women who were never pregnant, but did have increased cough and required more illness-related treatments. They also reported worse HRQoL in physical functioning, vitality, health perceptions, and respiratory symptoms. These differences may reflect the impact that the physical and emotional demands of parenting have on mothers' abilities to meet the daily challenges of an arduous CF-related treatment regimen. Our findings have implications for pre-pregnancy counseling. 1 McMullen et al. Chest 2006; 129:706-11. 2 Goss et al. Chest 2003; 124:1460 -8. 3 Quittner et al. Chest 2005 128:2347-54. pediatric patients who were 0-18 years old at the time of admission. CFrelated discharges were identified by the ICD-9 code, 277. Ethnicity and discharge disposition data were obtained for CF discharges. The US Cystic Fibrosis Foundation (CFF) registry was queried to obtain the number of Hispanic CF patients in California during the study period. The rate of death for CF patients by ethnicity was calculated using the discharge disposition data in OSHPD and the CF population in the CFF registry by year. The discharge percentages by ethnicity were compared by χ 2 and rate of death by two-tailed t-test. The Stanford University and the State of California committees for the Protection of Human Subjects approved this study.

Results: Using the OSHPD database, 14,550 CF-related pediatric discharges with ethnicity data were identified for 1999 through 2007. Although Hispanic CF patients make up 21% of the CF patient population in California, discharges of Hispanic CF patients make up 39% (n=5735) of the CF discharges (p<0.0001). Compared to non-Hispanics, hospitalized Hispanic CF patients are younger and more likely to have public insurance and lower household incomes. There were 161 deaths of hospitalized non-Hispanic CF patients and 145 deaths of Hispanic CF patients during the study period. The rate of death was 13 deaths per 1000 non-Hispanic CF patients compared to 43 deaths per 1000 Hispanic CF patients (p< 0.0001).

Conclusions: In California, there are differential outcomes observed between hospitalized Hispanic and non-Hispanic CF pediatric patients. Hispanic CF patients are hospitalized disproportionately to their representation in the CF population. They also have 3 times the rate of death compared to non-Hispanic CF patients. These findings may have multifactorial causes including greater severity of disease and/or differential provision of care. Rationale: As U.S. healthcare transitions to an outcomes-based system, understanding sources of variation in outcomes assumes increasing importance. Specifically, there may be geographic factors that may impact health outcomes independent of the quality of care provided, suggesting that rankings of healthcare providers and the design of multicenter clinical trials may have to account for these environmental differences. Delineating specific geographic factors that modify lung disease may provide information for future therapeutic targets as well as allow for correction of rankings of CF centers for factors not within their control.

Methods: Primary analysis was performed on 1378 individuals from the CF Twin-Sibling Study (CFTSS) and key results were replicated in an independent group of 7189 CFTR F508del homozygous, Caucasian subjects from the CF Foundation Data Registry (CFF). The primary outcome was cross-sectional CF-specific measures of FEV1 (Kulich et al.) . Environmental variables were derived from questionnaires and home zip codes, and included secondhand smoke exposure, maternal education level, median household income, insurance coverage, housing density, air pollution (PM2.5), elevation, distance from CF care center, population density, ambient temperature, and humidity. Statistical analyses were performed using student's T-tests, chi-square tests, and generalized estimating equations (GEE) regression. Final multivariate GEE models were constructed using all significant factors from univariate models, then stepwise removal of nonsignificant variables.

Results: In the final multivariate GEE model, age (p:<0.001), insurance status (p:<0.001), and ambient temperature (p:0.005) were found to be associated with lung function in the CFTSS population and replicated in the CFF population. Although the Kulich lung function percentiles do correct for age, there exist cohort and survivor effects that are not accounted for. On average, individuals with public insurance had lung function that was 6.4 percentile points lower than those with private insurance. Lastly, for every 10°F increase in mean annual temperature, lung function was found to decrease by 3.4 percentile points, thus for example, centers in New York and Florida may have a 7 percentile point difference in lung function based on temperature alone. Using data collected in both January and July demonstrated that the association of temperature and lung function is independent of season in both the CFTSS and CFF populations. Finally, secondary analyses determined that the prevalence of Pseudomonas aeruginosa was higher and the age of its acquisition younger in regions with warmer temperatures.

Ooi, C.Y. 1, 2 ; Keenan, K. 1 ; Gonska, T. 1, 2 ; Taylor, L. 1 ; Tam, K. 1 ; Ratjen, F. 1, 2 ; Solomon, M. 1, 2 ; Steele, L. 1 ; Ray, P. 1 ; Matteo Corral, D. 3 ; Price, A. 3 ; Morgan, L. 4 ; Zielinski, D. 5 ; Van Wylick, R. 5 ; Chakraborty, P. 6 ; Durie, P. 1, 2 1. SickKids, Toronto, ON, Canada; 2. University of Toronto, Toronto, ON, Canada; 3. University of Western Ontario, London, ON, Canada; 4. Windsor Regional Hospital, Windsor, ON, Canada; 5. Queen's University, Kingston, ON, Canada; 6. CHEO, Ottawa, ON, Canada While newborn screening (NBS) for CF identifies infants with an uncertain CF diagnosis (QUERY), their incidence and natural history has not been evaluated. In April 2008 when NBS was introduced in Ontario, Canada using an IRT/DNA strategy, we established a formal protocol to ascertain, assess and monitor QUERY patients prospectively. We describe our preliminary findings.

Methods: The number of NBS-positive infants classified as QUERY and confirmed CF was collected by 3 of 4 Ontario CF Clinics participating in NBS. QUERY infants included those with: (1) One disease-causing mutation and a borderline sweat chloride (30-60 mmol/L); or (2) CFTR mutations on each allele (non-CF causing on ≥1 allele) and a normal/borderline sweat chloride; or (3) Very high immunoreactive trypsinogen (IRT), no mutations and a borderline sweat chloride. Baseline and longitudinal clinical and genetic data were ascertained in the majority of QUERY infants that provided consent. For comparison, we enrolled NBS-positive infants with confirmed CF from the Toronto CF clinic during the same time period.

Results: Of a total of 652 NBS-positive infants, 25 (3.8%) QUERY and 37 (5.7%) CF infants were identified from April 2008 to April 2010; the QUERY:CF ratio was 1:1.5. Data from 21 QUERY and 16 CF infants were ascertained. At baseline, 4 (19%) QUERY infants had 1 mutation and borderline sweat chloride, 11 (52%) carried 2 mutations and borderline sweat chloride, and 6 (29%) had 2 mutations and normal sweat chloride. None, so far, has fulfilled the criteria for CF. Among 20 QUERY infants who had serial sweat tests, there was a significant fall in mean sweat chloride (SD) from 38.3 (10.8) to 27.7 (10.7) mmol/L (P=0.0007) over a mean (SD) period of 8.7 (6.8) months. All QUERY infants were pancreatic sufficient (PS) at baseline and at mean (SD) follow-up age of 10.4 (6.4) months old. All except 2 CF infants were PI at diagnosis; they remained PS at 5.7 and 15.7 months old respectively. Although all NBS-positive infants had IRT values exceeding the 96th percentile, QUERY infants had significantly lower IRT compared to CF infants (87.4 (33.7) vs. 153.0 (82.7) ng/mL; P=0.003). Serial IRT values demonstrated that QUERY infants had significantly lower serum trypsinogen than CF infants at age-intervals of 1-6 months (72.3 (36.1) vs. 274.6 (114.3) ng/mL; P<0.0001) and 6-24 months (40.2 (12.8) vs. 81.9 (21.6) ng/mL; P=0.006); 7/9 (78%) QUERY infants had elevated IRT (>46.5 ng/mL) at 1-6 months old while only 2/8 (25%) had raised IRT between 6-24 months old. In contrast, IRT remained high, at both periods, in all CF infants who had levels measured.

Conclusions: The relative incidence of QUERY compared to confirmed CF infants identified by NBS was high (ratio of 1:1.5). Sweat chloride and IRT concentrations may remain elevated in a subset of QUERY infants although sweat chloride, generally, decreased over time in QUERY group. Whether this represents a subset of those at risk of CF-like PS phenotype at an older age remains to be determined.

Jackson, A. 1,2 ; Kelleher, C. 1 ; Fitzpatrick, P. 1 ; Fletcher, G. 2 ; Harrington, M. 2 ; Zhou, S. 2 ; Daly, L. 1 1. School of Public Health, Physiotherapy and Population Science, University College Dublin, Dublin, Ireland; 2. Cystic Fibrosis Registry of Ireland, University College Dublin, Dublin, Ireland Background: An important attribute of any disease registry is the extent to which it covers the entire population. Complete registration of the affected population is required to avoid the introduction of selection bias should registered cases differ from those missed by the registry. Many disease registries use capture-recapture methods to estimate the size of affected popu-lations. This approach uses information on patient ascertainment by multiple incomplete sources to estimate the total population. Few cystic fibrosis registries have used such methods to determine their coverage rate.

Objective: Using capture-recapture methods, we examined ascertainment of persons with CF by the Cystic Fibrosis Registry of Ireland (CFRI), which was established in the Republic of Ireland (RoI) in 2002.

Methods: Persons with CF in the RoI were identified using three sources; the CFRI (CFRI staff invite patients to participate during outpatient clinics or inpatient stays), Central Statistics Office Ireland (CSO) certified deaths, and a survey of CF centre attendance (survey). Initially, three pairwise comparisons (CFRI-CSO, CSO-survey, CFRI-survey) were undertaken to obtain preliminary estimates for combined source pairs. Differences in pairwise population estimates suggested that data sources were not independent, violating an underlying assumption of the capture-recapture method. To take account of such dependencies, log-linear models of capture probabilities were developed to estimate the number of patients missing from the 3 sources. Models were compared using Akaike's Information Criterion and the model with the lowest value was chosen.

Results: A total of 1317 patients were identified by 3 data sources; the CFRI, survey and CSO reported 1117 (84.5%), 1187(90.1%) and 146 (11.1%) cases respectively. Of the 1317 patients, 18 were identified by all 3 sources, 1115 by 2 sources, and the CFRI, survey and CSO each identified patients reported by no other source (5, 130 and 67 patients respectively). The best fitting log-linear model estimated that only 5.8 patients had not been identified by any source, therefore the three-source capture-recapture abundance estimate was 1322.8 (95% CI: 1317.1, 1331.1).

The difference between the observed population (1317) and the capture-recapture abundance estimate (1322.8) was small, suggesting that nearly all the CF population in the RoI can be identified using a combination of registry data, CF centre attendance lists and certified death information. CF healthcare services are provided free of charge in the RoI, therefore population ascertainment by the CF centre survey is likely to be high, and the probability of missing individuals from our capture-recapture abundance estimate is low.

Conclusions: CF registry completeness can be improved by using information on patient ascertainment from other sources such as CF centre attendance lists and certified death data. Adopting a standard approach of population estimation can enhance the validity of registry data analyses.

Acknowledgements: This work was funded by the Health Research Board, Ireland.

Barr, H. 1 ; Britton, J. 1 ; Smyth, A. 2 ; Fogarty, A. 1 1. Nottingham Respiratory Biomedical Research Unit, Division of Epidemiology and Public Health, University of Nottingham, Nottingham, United Kingdom; 2. Nottingham Respiratory Biomedical Research Unit, Division of Child Health, University of Nottingham, Nottingham, United Kingdom Aims: To determine the impact of socioeconomic status (SES) and gender on median age at death from cystic fibrosis (CF) in England and Wales, from 1959 to 2008.

Methods: Mortality data for CF were obtained from the Office for National Statistics. Cause of death was coded by the International Classification of Diseases category applying to CF.

From 1959-2000, SES was classified as manual or non-manual. From 2001-2008, SES was divided into 3 categories: managerial/professional, intermediate, and routine/manual. SES of those who died in childhood was classified according to parental status.

Age at death was provided in 5-year age bands. Odds ratios for death above the estimated median age at death by SES and gender were calculated by logistic regression using STATA statistical software.

Results: Between 1959 and 2008, 6750 deaths were attributed to CF and the number of deaths per year declined. The median age at death increased from age band 0-4 years to 25-29 years during this period.

Overall, females were significantly less likely to die at an age above the median age at death than males (OR, 0.91; 95% CI 0.82 to 1.00; P <0.05).

From 1959-2000, median age at death was significantly lower in the manual group compared to the non manual group at all time points. From 2001-2008, median age at death was significantly lower in the routine/man-Sweat testing is consistently warranted given the clinical suspicion of CF despite negative screening results and in older siblings of CF-screened infants.

We acknowledge all the physicians from the CF care centers: Amiens, Angers, Besancon, Bordeaux, Caen, Clermont Ferrand, Giens, Grenoble, Lyon, Marseille, Montpellier, Nancy, Nice, Paris (Necker, Robert Debré, Trousseau) , Reims, Roscoff, Rouen, Saint Pierre La Réunion, Toulouse, Tours.

Renard, E. 1 ; Munck, A. 3 ; Guittet, L. 2 ; Roussey, M. 3 ; Travert, G. 1 1. Neonatal screening Lab, University Hospital Caen, Caen, France; 2. ERI3 INSERM, University Hospital Caen, Caen, France; 3. AFDPHE, Paris, France In the French CF Newborn Screening (NBS) program, immunoreactive trypsinogen (IRT) measurement is performed in 22 laboratories; 8 laboratories using a radio-immunoassay (RIA) kit (CIS bio International -Gif sur Yvette, France), 11 the AutoDELFIA (AD) system and 3 DELFIA (D) kits (PerkinElmer, Turku, Finland). The NBS algorithm is IRT/DNA (CF30 Elucigene kit). Two fixed IRT cutoff values are used: 55 µg/L (whole blood) for retesting in duplicate and 65 µg/L, after retesting, for CFTR gene analysis. A threshold generating 0.5% of positive IRT tests was deliberately targeted.

Objectives: To identify some of the main variables that may influence IRT measurement, which is the first and most crucial step in CF NBS.

Methods: Analyzed parameters were the percentage of newborns submitted to CFTR mutation detection and the proportions of newborns with initial IRT levels above 60 and 70 µg/L, that flank the 65 µg/L cutoff value. Among variables that could influence IRT measurements and thus the positive rate, we considered: 1) IRT assay technology ; 2) season of birth ; 3) region of birth. Positive rates were compared using Poisson regression.

Results: IRT data on 4,794,811 neonates, born December 1, 2003 -November 30, 2009 , were anonymously collected in the national NBS IRT databases.

1) Over a 6-year period, the percentage of newborns submitted to CFTR gene analysis was 0.58% (n = 27 688) and this percentage was slightly smaller using D (0.55%) than RIA or AD (0.58% for both) (p < 0.01). However, it is quite obvious that lot number of kits and/or improvement in laboratories practice influence IRT data, since the positive rate was higher during the first half of the time period than in the second one. Very similar patterns were logically observed with the proportions of IRT levels above 60 and 70 µg/L.

2) The overall positive rate varied from 0.67% in winter to 0.50% during summer (p < 0.001), with intermediate values in spring and autumn. When related to IRT assay methodologies, these features were: RIA, winter 0.71% versus summer 0.49% (p < 0.001); D, 0.66 vs 0.44% (p < 0.001) and AD, 0.62% vs 0.54% (p < 0.001). So, the seasonal effect was less sensitive with AD than with the other two methods.

3) On a map of France, a straight line could be drawn that separates the half (11) of laboratories located more to the west and south, from the other half located to the east and north-east. Thus, a very significant difference between the positive rates was observed with 0.45% (9,424/2,092,823) on the west side of the line and 0.68% (18,264/2,701,988) on the east side (p < 0.001).

Conclusions: IRT concentrations were significantly lower during summer suggesting that some part of the IRT molecule(s) may be unstable at high temperature. Also, it might be questionable whether the cutoff values should vary according to seasons and/or regions of birth. Close attention should be paid to assess if CF babies who were born during summer or in the west and south of France were more likely to be false negative cases. Further analysis will focus on this point. ual, intermediate and unclassified groups compared to the managerial/professional group (Table 1) .

Conclusions: Socioeconomic status and gender are important potential determinants of age at death with a diagnosis of cystic fibrosis. The strong association of socioeconomic status on median age at death does not appear to have changed substantially over the last 40 years.

Acknowledgments: This study was funded by the Medical Research Council and the University of Nottingham. We thank Brian Johnson from the Office of National Statistics for providing the raw data. The Office of National Statistics bears no responsibility for the analysis or interpretation of the data provided. Methods: data on NBS were centrally analyzed. A yearly prospective questionnaire collecting all newly identified FN cases was sent to the 22 regional associations and 37 CF centers respectively in charge of IRT/DNA and sweat testing (ST) along with conclusions concerning infant status.

Results: Over a 6-year period, among 4,331,539 newborns, CF was diagnosed in 1012 (including meconium ileus). In addition, over an 8-year period up until 31/12/2009, 21 CF centers reported 48 FN cases (23M, 25F) diagnosed based on clinical symptoms (n=41) or after a CF NBS sibling diagnosis (n=7). The overall sensitivity of the program was 95.5% with no statistical difference over time (94.4, 97.4, and 94.9%) . Initial and failsafe IRT were below the cut-off level in 40 and 4 patients respectively; a technical laboratory error was reported in 4 cases (3 IRT, 1 ST) . Median age at diagnosis was 1y 2m [4.9m-2y1m] with no significant delay between those presenting only respiratory symptoms (n=15, 8 cases with severe pulmonary involvement (1 death)) and those with dehydration, gastrointestinal or combined symptoms (n=25). Pancreatic insufficiency was found in 69%. Sweat test values were in the abnormal range in 40/46 (1 death, 1 epidermolysis bullosa) with a median value of 97 mmol/L [77-109] and 4 patients in the intermediate range (30-59 mmol/L). Mutation analysis found a CF-causing mutation in one or both alleles in 44 patients, while expanded genetic analysis remained negative for 5 alleles in 3 patients; in 4 cases a R117H was identified with one 5T.

Conclusion: The sensitivity of our nationwide NBS program properly assessed by a minimum of two-year follow-up period and centralized data management is in agreement with the a priori rate of about 5% FN cases. of children born in the Netherlands yearly, in which about 38 CF patients are born. For all strategies, implementation caused savings compared to no screening, as the costs of screening were smaller than the savings resulting from reduced costs of diagnosis among CF and non-CF patients. Reduction of life long costs of treatment that can be obtained by detection of CF patients through neonatal screening compared to clinical detection of CF patients appeared to be the most important parameter determining the variance in cost effectiveness. Costs presented were based on 0% reduction, but if 5% is assumed, implementing neonatal screening for CF may lead to additional financial savings of about 1.1 million Euros yearly, depending on the screening strategy used.

Conclusions: Neonatal screening for CF leads to financial savings in comparison to a situation without screening. Costs are in the same range for the three strategies. Therefore, the choice of which strategy to implement also needs consideration of other factors, like the number of false-positives and carriers detected.

Duguépéroux, I. 1, 2 ; Scotet, V. 1, 2 ; Audrézet, M. 1, 2 ; Saliou, A. 3 ; Blayau, M. 4 ; Schmitt, S. 5 ; Kitzis, A. 6 ; Férec, C. 1, 2 1. Inserm U 613, Brest, France; 2. Lab of Molecular Genetics, C.H.R.U., Brest, France; C.H.R.U., Brest, France; 4. Lab of Molecular Genetics, C.H.R.U., Rennes, France; 5. Lab of Molecular Genetics, C.H.R.U., Nantes, France; 6. Lab of Molecular Genetics, C.H.R.U., Poitiers, France Introduction: Cystic fibrosis (CF) can be diagnosed in several ways. In utero, several clues may suggest it, the most common being the presence of a foetal echogenic bowel (FEB) detected at ultrasound during pregnancy. Other elements, as non-visualized foetal gallbladder (NVFGB) are also possible evidences for CF.

Aim: The aim of this study was to describe and characterize the weight of the diagnosis of CF in a cohort of foetuses with NVFGB, and to compare it to the global population.

Method: This study is based on the experience of 4 university hospitals (Brest, Rennes, Nantes, Poitiers) of western France. We reviewed all the consecutive cases of NVFGB who were referred, in one of the 4 hospitals, for analysis of the gene responsible for CF (CFTR gene) over the 2002-2009 period. Results: Over that period, 37 NVFGB were recorded. We identified 5 CF foetuses (F508del / F508del n=2, F508del/937insT n=1, F508del/4005+1G>A n=1, F508del/R553X n=1), leading to an incidence of the disease of 13.5% (1/7.4) . This calculated rate was significantly higher than in the global population (1/2 928 p<0.0001). All CF affected foetuses had NVFGB associated with FEB, and all affected pregnancies were terminated. We also identified 3 heterozygous foetuses, leading to a carrier rate of 8.1%. All were sharing a single copy of the F508del mutation and they showed only NVFGB. All these pregnancies resulted in births.

Moreover, in our population, 6 foetuses without CFTR gene mutation had both NVFGB and hyperechogenic bowel. When these 2 ultrasonographic evidences are combined, the calculated rate of CF was 45.5%, i.e. 5/11. This rate was significantly higher than in our FEB population (1/10 p<0.005).

Some foetuses without CFTR mutations presented other pathologies, such as chromosomal abnormalities, abdominal, cardiac, cerebral or polymalformatif pathologies, and 2 pregnancies were terminated. Atrophy or agenesis of the gallbladder was evidenced after birth in few foetuses.

Conclusion: We highlighted the importance of visualizing the gallbladder during pregnancy. In our study, NVFGB seems to be an essential element to consider in achieving a diagnosis of CF, especially when associated with FEB (all the CF foetuses have both). Support: Programme Hospitalier de Recherche Clinique.

persists and is less understood in the first two years of life as compared to later in childhood. With NBS now universal in the US, interventional trials are on the horizon for this population which necessitates better understanding of nutritional failure outcomes. The objectives of this retrospective study are to: (1) characterize growth in CF infants and determine factors most associated with differences in growth trajectory; and (2) evaluate different definitions of nutritional failure for their sensitivity and potential use as endpoints for clinical trials.

Methods: A retrospective cohort study was conducted using all infants entering the US CF National Patient Registry before 4 months of age in [2004] [2005] [2006] [2007] [2008] . Records up to 2 years of age are used. Risk factors evaluated for poor growth included diagnosis by newborn screening, pancreatic sufficiency (PS or PI), genotype, and meconium illeus (MI). Growth outcomes included weight, length, and weight for length standardized into percentiles and z-scores using the Centers for Disease Control (CDC) Growth Charts; weight and length velocity were calculated and benchmarked to the World Health Organization's velocity percentiles. Mixed models were implemented to estimate growth as a function of age and risk factors. Categorical definitions of nutritional failure were analyzed as rates and as time-dependent events modeled using Cox proportional hazards regression.

Results: Cohort criteria were met by 1988 CF infants and 1524 (76.7%) had follow-up records to at least 9 months of age. Half (49.4%) the cohort was diagnosed via NBS; 89.3% of the cohort used enzymes (PI) while 7.4% were PS based on genotype. PI and MI were factors most highly associated with weight percentiles averaged over the first year of life (p<0.0001 and p<0.001, respectively). Additional risk factors are under evaluation. Approximately 50% of all infants were defined as having nutritional failure by 7 months of life under multiple definitions, including both those based on growth velocity percentiles and those based on crossing major CDC percentiles. Assuming 90% of subjects will experience nutritional failure by one year of age, 100 subjects per group (total N=200) using a log-rank test for time to failure at the 2-sided 0.05 level, there would be 80% power to detect a 36% reduction in events in a therapeutic arm compared to a placebo arm.

Conclusions: This study identifies factors most strongly associated with growth and nutritional failure in CF infants. While causality can not be confirmed, the observed effect of one or more factors may approximate that seen by a successful intervention. Surprisingly, a high proportion of infants reach failure within the first year of life indicating the need for nutritional interventions. Our analysis gives a practical assessment to commonly used anthropometric outcomes in early childhood. While the scientific question should drive endpoint selection in a trial, statistical efficiency is equally important in a vulnerable infant population with an orphan disease. Kier, C. 1 ; Soultan, Z.N. 2 ; Giusti, R. 3 ; Sharp, J. 4 ; Fiato, K. 5 ; Caggana, M. 5 ; New York State, C. 6 1. Stony Brook University Medical Center, Stony Brook, NY, USA; 2. SUNY Upstate Medical University, Syracuse, NY, USA; 3. NYU Langone Medical Center, New York, NY, USA; 4. Women and Children's Hospital, Buffalo, NY, USA; 5. New York State Department of Health, Albany, NY, USA; 6. New York State, New York, NY, USA Background: Newborn screening (NBS) for CF in New York State (NYS) started on October 2002. Immunoreactive trypsinogen (IRT) is used to determine top 5% of daily specimens, which are then tested by mutation panel. Screen-positive (2 mutations, 1 mutation, the high IRT no mutation: IRT>170 ng/ml and the top 0.2%) are referred for sweat testing. Preliminary report in Australia showed only a small proportion of babies with IRT in the 99th percentile have CF. Additional CF testing has an extremely low yield (Masse. Arch Dis Child 2006; 91:222-225) . The implications may be different in an ethnically diverse population.

Objective: Demographics, mutations outside of NYS mutation panel, and clinical status in the high IRT no mutation group infants will be described.

Methods: NY Department of Health records were reviewed October 2002-October 2009. CF center charts of CF infants in this group who had positive sweat chloride tests (≥30 mmol/L) were reviewed to obtain demographic/clinical data. dichotomized age at test and QNS rates. Cochran-Armitage test for trend was used to examine the association between the categorized birthweight and the QNS rates. Logistic regression was used to model the probability of QNS with methods (GCT and MCS), adjusting for age at test.

Results: A total of 568 infants were referred for 616 sweat tests from March 2006-January 2010. The infants from each site had similar demographic profiles. The mean age was 32.8 days (6-285 days) and mean weight was 4.0 kg (2.3-7.3kg) at the time of the first test. The GCT had a significantly higher QNS rate compared to the MCS, 15.4% vs. 2.1%, respectively (p〈.0001). There was no association between age at test (〈 or 〉 30 days) and the probability of QNS. The probability of having a QNS sweat test decreased as birthweight increased for both methods (p=0.02). After adjusting for age at test, the odds of QNS using the GCT was estimated to be 8.34 ) times that of the MCS.

The MCS had a significantly lower QNS rate compared to GCT in all infants referred to the Minnesota CF Center for diagnostic sweat testing after a positive NBS.

Supported by: CFF C061-09.

Results: There were 21 CF infants in the high IRT no mutation group. Positive predictive value (PPV) for the high IRT no mutation group is extremely low at 0.4% (21 CF in 5557 referrals). In comparison, PPV for 2 mutation referrals is 42% (106 CF in 259 referrals), PPV for 1 mutation referrals is 4.2% (202 CF in 4, 896 referrals) . IRT range is 134-262 ng/ml (mean + SD=194 + 39). Sweat chloride range is 34-112 mmol/L (87 + 20) . Ethnicity showed Hispanic 9/21 (43%); Caucasian 6/21 (28%); Asian 4/21 (19%) , only one each of African-American (AA) and mixed (Hispanic and AA) origin. Eight out of 21 CF infants have 2 mutations each (outside of the NYS mutation panel). Two infants have only one mutation each. The remaining 11 infants had no reported mutations, either unknown or not done, but all have positive sweat tests, range = 71-112 mmol/L. A third of the CF infants (7/21; 33%) reported as pancreatic insufficient based on fecal elastase values, 5/21 (24%) as pancreatic sufficient, 9/21 (43%) as unknown pancreatic status.

Conclusions: CF in the high IRT no mutation group is extremely low (PPV=0.4%), even in this ethnically diverse population. CF infants in this group are predominantly of Hispanic origin (43%), and were not representative of the ethnic distribution in the total CF positive disease of this screening program (80% Caucasian, 10% Hispanic, 4% AA, 0.4% Asian and 5% mixed/other origins). Interestingly, only one CF infant is AA in spite of a two-fold increased incidence of an elevated IRT in AA infants, as recently reported from our NYS program (Guisti. Ped Pulm 2008; 43:638-641) . Implications: Extremely low yield for CF in this group, but yet with a different ethnic distribution, necessitates careful assessment of the algorithm process of the screening program. NYS has expanded its mutation panel to 40 mutations after 5 years (2007), and recently (2010) changed algorithm to establish the 0.1% cutoff IRT value. Careful algorithm changes will decrease referrals yet will minimize missing CF diagnoses. Data will be presented in the near future. Background: Colorado adopted newborn screening (NBS) for cystic fibrosis in 1982, using an algorithm of two immunoreactive trypsinogen determinations (IRT/IRT). We have previously reported a sensitivity of 94.6% (90.0-96.2%). Reports from other programs suggest sensitivity of this approach with initial cutoffs >100 ng/ml may be as low as 80%. We identified individuals in the Cystic Fibrosis Foundation (CFF) Registry who may have been missed by the Colorado NBS in order to further refine our sensitivity estimates.

Methods: Names and current CF centers were requested from the CFF on individuals born in Colorado between 1982 and 2007 that have CF but not seen by the Colorado Center. Local CF centers were contacted and asked to confirm state of birth with families. The additional missed cases were added to the number of missed cases previously known. Infants with meconium ileus (n=67) were not included. False negative rate and 95% confidence intervals are reported.

Results: Twenty infants were previously known missed cases followed in Colorado, (median age of diagnosis 1 year: range 0. 1-13.4 years) , and 332 non-meconium ileus infants were identified by NBS (missed case rate 6.0%, 3.9-9.1%). Eight missed cases (40%) had genotypes associated with pancreatic sufficiency. An additional 27 patients were identified from the CFF query, however only 2 have been confirmed as missed cases on the Colorado NBS. Of the remaining infants, 21 were confirmed as not missed on NBS, and 4 individuals have not yet been located. The 2 confirmed missed cases were diagnosed in other states: one diagnosed at 6 months due to failure to thrive and electrolyte imbalance (Cl=120, F508del/ F508del) and one diagnosed at 3 years of age (Cl=74, not genotyped). Three babies were reported who have CF identified by Colorado NBS. One infant moved following an elevated IRT, sweat test performed in infancy in another state was negative, and the child was diagnosed at 14 years (Cl=111, F508del/ F508del). One moved to another state after birth and diagnosis at 1.2 months was prompted by elevated IRT; another lived in a neighboring state and diagnosis was prompted by elevated IRT at 1.4 months. Four infants had been diagnosed by the Colorado NBS but not reported to the CFF. Eight infants were a duplication of records (names changed/misspelled). The birth state was misreported in four cases. One case was concluded to be Munchausen's by proxy (CFTR mutations not identified) and one child had a heart transplant, with no further mention of CF (CFTR mutations not identified). The missed case rate using infants with known state of birth is 22/337 (6.5%, 4.4-9.7% ). Assuming all four infants who have not been located were born in Colorado and missed on NBS, the missed case rate is 26/341 (7.6% 5.6-10.9%).

Conclusions: A systematic review of national data reflecting potential missed cases from the Colorado NBS program resulted in only two additional cases missed by NBS, and one case missed by a negative sweat test in another state. These data suggest the sensitivity of IRT/IRT in Colorado was 92.3% (89. 1-94.7% ). This systematic approach should be considered in evaluating other CF NBS protocols. Methods: Twenty-one designated CF follow-up professionals (genetic counselors and nurses) from CF Clinic Practices in Illinois and St. Louis were invited to participate in a 26-item web-based survey. The survey consisted of four sections that addressed the following issues: 1) disclosure of positive CF newborn screen results; 2) the sweat test process; 3) follow-up genetic counseling; and 4) background information. Fourteen genetic counselors and three nurses completed the survey (17 of 20 eligible participants), resulting in a response rate of 85%.

Results: Results indicated inconsistencies and variability in: 1) the information provided to parents at the time of the positive newborn screen results disclosure; 2) the information provided to parents during follow-up genetic counseling; and 3) the number of parents who are referred to and attend follow-up genetic counseling. As part of this study, barriers and facilitators to genetic counseling were also assessed. Participants indicated the following barriers to genetic counseling: 1) lack of physician referral; 2) parent priority on other issues; 3) lack of availability of a genetic counselor; and 4) limited access to transportation and working telephone. Study participants also offered suggestions as to ways to overcome these barriers and facilitate the follow-up process, including: 1) scheduling the genetic counseling appointment at the time of the sweat test; 2) educating pediatricians and parents that genetic counseling is part of the standard protocol for positive CF newborn screens; 3) providing parents with the opportunity to speak with a genetic counselor in advance of the sweat test and genetic counseling appointment; 4) encouraging good communication between clinic staff, lab staff, and genetic counselors; and 5) notifying the genetic counselor of all scheduled sweat tests.

Conclusions: This study was successful in furthering our understanding of current follow-up practices to positive CF newborn screens in Illinois. Findings from this study could serve as a basis for discussion regarding current follow-up practices to positive CF newborn screens. In order to facilitate this type of discussion, efforts will be made to share these findings with genetic counselors and other health care providers involved in follow-up care for positive CF newborn screens. Hopefully, efforts will be aimed at developing strategies to overcome the inconsistencies and barriers identified in this study and ultimately improve and facilitate follow-up services for positive CF newborn screens.

Supported by Northwestern University's Graduate Program in Genetic Counseling.

Objective: This study aimed to determine the prevalence of vitamin D insufficiency in infants with CF identified by NBS and identify potential risk factors.

Methods: Infants diagnosed with CF by NBS between 2002-2010 were included if a 25-OH vitamin D level was obtained in the first 2 years of life. Because older infants were more likely to eat table food and be treated with pancreatic enzyme and vitamin supplements, the patient data was divided into 2 groups: less than 10 months of age and 10-24 months of age. Serum levels of vitamin D were assessed in relationship to potential modulating factors including pancreatic status (sufficient vs. insufficient), vitamin supplementation, skin tone, and season of birth. Vitamin D insufficiency was defined as less than 30 ng/ml and vitamin D deficiency as less than 20 ng/ml.

Results: Twelve of 23 (52%) infants had vitamin D levels below 30 ng/ml in the first 10 months of life and of those patients 5/23 (22%) had levels below 20 ng/ml. Of infants tested between 10-24 months of age, 7/28 (25%) had levels below 30 ng/ml and 2/28 (7%) were below 20 ng/ml. From the younger cohort, 14/23 infants had repeat levels after 10 months of age and although there was a mean increase of 6.3 ng/ml this was not significant. Of infants tested in the first 10 months, 16/23 (70%) were pancreatic insufficient (PI). Within this group of infants with PI, 10/16 had vitamin D levels less than 30 ng/ml and when compared to the PS infants the relative risk was 2.2 with a 95% CI of 0.63 and 7.5. In the older group, there were 19/28 (68%) patients with PI of which 7/19 had a vitamin D level less than 30 ng/ml and none of the infants with pancreatic sufficiency had low vitamin D levels. Of the 23 infants tested in the first ten months of life, 12 were born during the summer. The relative risk of being vitamin D insufficient or deficient for those born during the summer compared to those born in the winter was 1.83 with a 95% CI of 0.76 and 4.41. In the older cohort, the seasonal effect decreased and the relative risk was 0.72 with a 95% CI of 0.17 and 3.06. There was insufficient data to analyze the effect of vitamin supplementation or skin tone.

Conclusion: This study suggests that vitamin D insufficiency occurs as early as the first year of life and, although there is some improvement with age, the deficit is sustained. Although this study was limited by a small sample size, the data suggests that PI and being born in the summer are potential risk factors for vitamin D insufficiency in the first ten months of life. These results, if corroborated by larger studies, support early nutritional intervention with attention to vitamin D supplementation to avoid vitamin D insufficiency and its inherent complications. Barker, H.; Floto, A.; Haworth, C.S. Adult Cystic Fibrosis Unit, Papworth Hospital, Cambridge, United Kingdom Background: The Department of Health in the UK recommends influenza vaccination to people of clinical risk groups including patient with cystic fibrosis. Influenza infection can lead to significant morbidity and mortality in this group of patients [1] and vaccination remains the most effective measure in preventing this illness. In addition, there has been evidence to show that influenza vaccination in CF can prevent its subsequent acquisition [2] .

Aims: To evaluate the influenza vaccine uptake rate for the 2008-09 season in our CF centre and the factors associated with non-adherence.

Methods: Retrospective questionnaire was given to all patients who attended the CF outpatient department and CF unit between June and September 2009.

Results: Of 174 patients participating. 131 (75%) patients were vaccinated, 41 (24%) were not vaccinated and 2 (1%) could not remember if they were vaccinated.

The vaccinated group has significantly better FEV1 (p<0.05) and body mass index (p<0.05) compared to the non-vaccinated group. Transplant, diabetes and CF related liver disease status do not appear to influence the vaccination uptake rate.

For those who were not vaccinated, 10 patients forgot, 4 were scared of needles, 4 were worried of side-effects, 6 were too unwell and 6 previously did not find vaccination effective; other reasons for not receiving vaccination include not being able to make an appointment with primary care physicians to receive vaccination; patients did not feel they require vaccination; some were not reminded about vaccination and some were not offered.

Discussion: The vaccination coverage in this group of patients is much higher than the overall uptake rate in the clinical risk groups in England (75% vs 47.1%). However, the uptake rate may be further enhanced by providing reminders and patient education. The recent Swine flu pandemic may also indirectly heighten patient's perception of future seasonal flu vaccination uptake. Patients with better lung health and nutritional status are more likely to take up vaccination. Phinizy, P.; Quittell, L. Pediatrics, Columbia University, New York City, NY, USA Introduction: Vitamin D deficiency is a recognized manifestation of pancreatic insufficiency in people with CF and is increasingly associated with osteopenia and osteoporosis. In addition to its role in bone health, vitamin D plays a role in promoting immune function in the lung. Early intervention and nutritional support of infants diagnosed by newborn screening (NBS) should potentially decrease the risk of vitamin D insufficiency.

the Prague CF Center. Mutation analysis and sweat testing were carried out simultaneously whereas PAP concentration examinations utilized the Muco-PAPTm assay (Dynabio) followed by absorbance analysis using the "Plate reader" from Bio-Rad. Within the routine nationwide scheme, mutation testing was performed sequentially with the same assay in samples with IRT concentrations above 65 mg/L (99.0 percentile). From the 42,688 newborns tested thus far, 87 (0.2 %) had a "suspect" result within the IRT/PAP/DNA protocol, while 6 CF patients and 5 carriers were diagnosed. In comparison, within the routine IRT/DNA protocol we tested 31,603 newborns, 318 (1.0 %) had a "suspect" result and among those 5 CF patients were identified, including 12 carriers. Interestingly, we detected 1 case of false negativity in the IRT/PAP/DNA protocol, but this patient was unambiguously detected in the nationwide scheme. Although the numbers of the IRT/PAP/DNA "parallel" scheme are still relatively small, our pilot data indicate that further studies are needed to compare both protocols. Updated results will be presented at the meeting. 2 1. Dept. of Pulmonology, University Children's Hospital III, Heidelberg, Germany; 2. Dept. of Metabolic Diseases, University Children's Hospital I, Heidelberg, Germany Objective: Ethical concerns and disadvantages of newborn screening (NBS) for cystic fibrosis (CF) by genetic testing have raised controversies and impeded implementation of CF NBS in some countries. To circumvent these problems, we are studying a sequential and pure biochemical strategy with immunoreactive trypsinogen (IRT) as first tier and pancreatitis associated protein (PAP) as second tier, and validated this approach (IRT/PAP protocol) against the widely used IRT/DNA protocol in a population-based NBS study in Southwest Germany.

Methods: Prospective quantitation of PAP and genetic analysis for the presence of 4 mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene most prevalent in Southwest Germany (p.F508del, p.R553X, p.G551D, p.G542X) was performed in all newborns with IRT > 99.0 percentile. NBS was rated positive when either PAP was > 1.0 ng/mL and/or at least one CFTR mutation was detected. Positive rating lead to referral to a CF centre for sweat testing (sweat Cl-concentration). In addition, IRT > 99.9 percentile was considered a positive rating.

Findings: The study started in April 2008. Until March 2010 84857 newborns were screened for CF. Of these, 130 were positive in IRT/PAP, 74 in IRT/DNA, and 22 due to an IRT > 99.9 percentile. After sweat testing of 226 CF NBS positive infants, 19 were diagnosed with CF. Three out of 19 diagnosed CF newborns had a PAP concentration below the cut-off of >1.0 ng/mL.

Conclusions: Sequential measurement of IRT/PAP allows reliable and cost-effective CF NBS and avoids genetic testing. Our results indicate that lowering the previously suggested PAP cut-off level from >1.0 ng/mL to > 0.9 ng/mL may further improve the sensitivity and should therefore be further evaluated. This concept should substantially promote the implementation of CF NBS in countries with legal restrictions for genetic testing for population based screening. The opportunity to develop new treatments for cystic fibrosis and the search for a cure are more promising than ever before. Thus, there is an increased need for patients to participate in clinical trials. A national survey conducted by the CF Foundation revealed that the main reason people do not participate in clinical trials is because they were never asked. In 2006, the CFF launched the "I am the Key" initiative to promote a culture of research within CF care centers. Since its implementation, there has been an increased awareness and more people report being asked to participate in a clinical trial. However, many still do not participate for a variety of reasons. Efforts are needed to continue to improve communication and help overcome barriers to participation. Since people with CF and their families can relate to others with CF, we began a center specific campaign called "Faces of a Cure" featuring pictures of many of our research participants (different ages, genders, ethnic backgrounds, etc). The research staff contacted individuals who had participated in research and collected an array of photos of them in their normal activities. These photos were used in a collage to help make it personal and illustrate the variety of people that have and can participate in clinical trials. The hospital marketing department was contacted to help with the layout and make sure it was consistent with their standards. The IRB approved the campaign which was launched in May of 2009. Posters were made and hung in the patient exam rooms and other prominent places in the hospital. These were laminated so they could be cleaned in compliance with our infection control guidelines. Postcards were also printed and are used to notify a patient that they potentially qualify for participation in a clinical trial. The postcards "congratulate" patients on their eligibility and invite them to inquire for more information.

Data from the two years prior to the program (May 2007-April 2009) revealed an average enrollment rate in clinical trials of 1.4 subjects per month (range: 0-6; median: 1).

Since implementing the "Faces of a Cure" program in May of 2009 (12 month period), enrollment in clinical trials at our center has more than doubled with an average enrollment of 3.6 subjects/month (range: 0-10; median: 3).

While a number of variables can influence study recruitment at any given time, the "Faces of a Cure" campaign appears to have helped raise awareness of the importance of participating in clinical trials. In order to sustain the improvement, we plan to build on this initiative and fine-tune the process of prescreening to identify potential subjects.

Callahan, K.A. 1 ; Curtin-Brosnan, J. 3 ; Chapman, C.G. 1 ; Felling, E.M. 2 ; Mogayzel, P.J. 1 ; Zeitlin, P.L. 1 ; Boyle, M.P. 2 1. Pediatric Pulmonary, Johns Hopkins University, Baltimore, MD, USA; 2. Adult Pulmonary and Critical Care Medicine, Johns Hopkins University, Baltimore, MD, USA; 3. Pediatric Allergy and Immunology, Johns Hopkins University, Baltimore, MD, USA Background: Johns Hopkins is a large cystic fibrosis (CF) care center with 471 pediatric and adult patients and over 30 currently active research protocols. While a significant portion of patients at the center participate in research, many do not. The purpose of this study was to assess patients' and parents' perceptions regarding access, willingness, and barriers to participation in CF research at Johns Hopkins.

Methods: A 10-item, multiple choice survey was sent to CF adults and parents of children with CF who provided an email address. Demographics including gender, race, clinic, and perceived severity were collected. Participation in CF research studies was assessed by 1) types of research done in the past; 2) interest in future research; 3) barriers to participation; 4) access to the research team; and 5) sources of information about CF studies. Analyses were conducted using Cuzick trend test.

Results: There were 113 responses to the 338 surveys distributed (33%). Sixty-three percent of the survey respondents were adult patients, 53% were female, and 95% were Caucasian. Patient and parent perceived severity of disease was very mild 24%, mild 37%, moderate 33%, and severe 6%. Perceived severity did not vary significantly by age group (p=0.31). Twenty-one percent of respondents had never participated in CF research. Of the 79% that had participated, 83% had done so in the last 5 years. While questionnaires (89%) were identified as the type of study in which respondents were most interested in participating, 68% expressed interest in participating in specimen collection studies, 58% in drug trials, and 57% in procedure studies (e.g., NPD, sweat testing). Interestingly, 74% expressed interest in participating in trials specifically focused on nutrition or exercise. Several barriers to participating in research were identified, with the most common being too busy (39%) and concern about missing school or work (35%). A separate analysis of parents' responses however revealed their primary barrier was how studies might affect their child's health (56%). Other identified barriers included not always being aware of studies (26%) and living too far from the center (24%). When considering whether to participate in research, 66% reported that talking to their CF physician in clinic would be the most important step. Contacting the Johns Hopkins research team was second (54%).

Conclusion: CF patients report interest in participating in CF research but also identify being too busy and unable to miss school or work as significant obstacles. To increase research participation, we must work to minimize time off from work/school, be considerate of time commitment of study visits, discuss with parents the potential impact of study on their child's health, and perhaps most importantly increase access to physicians and research coordinators in clinic to specifically discuss ongoing research, as patients look to them as the key sources of information about CF studies and potential research participation. Research is an exponentially growing field in the CF world. Therefore, integration and collaboration of the clinical and research teams is an essential part of creating and sustaining a successful research program. The University of Alabama at Birmingham (UAB) CF center provides care for 450 total patients -150 adult patients followed by 4 physicians at UAB hospital and 300 pediatric patients managed by 5 physicians at the Children's Hospital of Alabama. Four research coordinators are responsible for the coordination of clinical and observational trials across both populations.

Three components have proved integral parts of creating a culture of research at the UAB CF center. Weekly research team meetings engage pediatric and adult investigators and research staff in subject identification, participant enrollment and study execution. The presence of a research coordinator at pediatric pre-clinic meetings which gather all disciplines and present the opportunity for individual patient discussion each week has established research as a connected discipline in patient care. Finally, an engaged pediatric CF coordinator providing daily interaction between clinic and research teams maintains an updated connection among the two branches.

These methods of integration and collaboration among clinic and research teams have resulted in high performance outcomes of the UAB CF center in the Cystic Fibrosis Foundation Therapeutics Development Network's (CFF-TDN) April 2010 12-month enrollment update categories. The UAB CF center ranked 32 of 77 TDN centers for interventional study enrollment (11.7% of total patient population) and ranked 4 of 62 centers for observational study enrollment (42.9% of total patient population). When the enrollment of these studies is weighted to consider complexity and duration, UAB ranked 8 of 77 centers (29.3% of total patient population) and ranked 12 of 77 centers for weighted interventional and outcome measurement study enrollment (24.1% of total patient population).

Plans for recruitment and enrollment sustainability and improvement include:

1. maintenance of present team integration practices 2. the presence of at least one research coordinator at monthly pediatric CF team updates 3. regular education of the clinical team regarding active research 4. increased involvement of the research team into investigator-generated research development among all professional CF care center disciplines. Given the need to increase overall cystic fibrosis (CF) research participation, the Stanford CF Research Team queried our research subjects, who had participated in at least one research study since 1992, in order to pinpoint which areas of the CF research program could be improved to facilitate increased study participation. A questionnaire was developed to investigate 5 broad categories of inquiry: the overall research experience, the research process, research-related procedures, research facilities, and research coordinator professionalism. General comments were encouraged with free space at the end of the questionnaire.

In all, 107 patients were identified through the Stanford database as having participated in either an observational or interventional trial. A total of 84 (79%) completed the questionnaire: 25 responders were the parents of young children (<10 years of age) and 59 responders were adolescents or adults. The remaining 23 patients were unavailable despite repeated attempts to contact them via phone and e-mail.

A non-biased interviewer (the Stanford Database Coordinator) asked the questions either in person at a clinic visit or over the phone, and the answers were entered verbatim into a Lenovo ThinkPad. Results: 94% of respondents were satisfied with their research experience, 100% felt safe during their research participation, 100% felt that they were well cared for during their research experience, and 95% felt that their research experience was worthwhile. Areas that we identified for improvement included: (1) providing adequate compensation for time spent in studies, (2) the need for clinic staff to further foster research awareness, (3) new studies to open for enrollment in a timely manner, and (4) some recruitment strategies such as T-shirts, newsletters were not found to be helpful in fostering the research mission. All of these areas can be improved by research-focused education of CF patients, their families, and clinic staff.

McDonald, C.M. 1 ; Haberman, D. 1 ; Christensen, N.K. 2 ; Brackner, E. 1 1. CF Clinic, Intermountain Primary Children's Medical Center, Salt Lake City, UT, USA; 2. Food and Nutrition Science, Utah State University, Logan, UT, USA level of interest in 12 different topics. A point value for interest was assigned: none = 0, slightly = 1, moderately = 2, very = 3. Topics of greatest interest for future "Tips" were: infection control (48), nutrition (45), general resources for CF (43), managing medications (42), respiratory treatment/equipment (42), and recipes (42). Transition to adult care (19) was the topic of least interest. Respondents' interest in specific types of recipes was: Healthful recipes with instructions for adapting to either high calorie or low calorie for different family members (45), recipes that other families with CF use and recommend (42), recipes that are easy to prepare (42), and high calorie, high fat recipes (39). Recipes using "from scratch" ingredients was the item of least interest (32).

The CF PAB reviewed website progress, "Tips," and results of the survey. They identified the following for ongoing quality improvement: 1) increase family awareness of CF PAB website, 2) elicit greater participation in surveys by distributing them at parent education events as well as electronically, and 3) use the preferences identified in the survey to generate future "Tips."

This project received funding from a CFF Quality Improvement Grant. 1. The increasing need for consultation with the CF team leads to an increased burden of care for the patient.

2. The increasingly multidisciplinary nature of care can lead to loss of direction within care through communication between caregivers.

3. The increasingly complex drug therapy and outpatient pharmacy leads to coordination problems and errors.

Aim: The project aims to improve care (measured by SNS-care indicators, including lung function and nutritional status), improve communication between caregivers and patient, increase the quality of life and reduce medication errors.

Method: To help solve these problems we developed an electronic transmural patient dossier with patient-access (E-portal), connected to our Electronic Patient Dossier (EPD). The patient can accces this portal with his own secure login code.

Since 2007, the CF Center Utrecht has used completely electronic files for all children with CF. The file contains the patient's medical history, physical examination, diagnostics, correspondence with the 1st and 2nd line, and prescribed medication. Through his personal E-portal, the patient has the opportunity to check a part of his patient file at home. He can also get in touch with his caregivers through e-consulting.

The E-portal includes access to: -CF-physician formulated treatment -current medication prescriptions -laboratory results -microbiology results -growth data -lung function results -correspondence between caregivers -records containing personal data and caregivers data -planning of outpatient visits, etc. Possibility for -e-consulting with the CF-Nurse Practitioners -electronically renewing prescriptions and required statements for insurance, school, etc.

-entering home data -home measured lung function -home sputum culture -preparing for hospital consultations through questionnaires on line

The E-portal is a part of the site of the CF Center Utrecht (www.CF-Centrum.nl), where general information about CF and the CF Center Utrecht can be found.

Results: The first patients received their personal logincode in February 2010. The first version of the e-portal is operational and will be evaluated and extended in the coming months. We aim to include 200 children before July 2010. The effect of the introduction of the E-portal on improving patient care will be researched extensively in the coming years and will be presented at a later stage. At this stage we would like to report about the aim, content and implementation of the E-portal, the first experiences of the users (both patients and caregivers) and the first results of the study on improvement of care and communication. The USC University Hospital, Center for Cystic Fibrosis Patient Advisory Committee (PAC) is actively involved in quality improvement at the Center. One of their goals is to develop tools to empower patients to take a more active role in their care. As part of this initiative two new tools have been developed: One tool is the Wallet ID & Medication Card, a wallet-sized card that carries a summary of significant medical emergency information, diagnoses, and medications. The second tool is the Yearly Diagnostic Test Sheet, constructed to help patients keep track of their annual tests including laboratory analyses, Xrays, immunizations and their anticipated due dates. A pilot study is in progress to evaluate the utility of these new patient tools through two surveys.

Each IRB-approved, 14-question survey was built on the Survey Monkey website. Patients were sent the website link or provided a paper copy. The anonymous survey responses were collected and analyzed. A survey domain requests the patient to report their perspective of each tool's utility on a Likert scale from "not useful" to "completely useful."

In 22 respondents completing the Wallet Card survey to date, findings suggest that over 70% find the Wallet Card very or completely useful in the following categories: Overall (86%), organization (95%), size (86%), use during CF clinic (70%), non-CF clinic (89%) and hospitalizations (83%). Nearly half (47%) feel it reduces time spent in clinic (data shown in Table) .

Similarly, nearly 80% of 23 respondents find the Yearly Diagnostic Test Sheet very or completely useful: Overall (76%), in organization (80%), tracking dates of completed tests (90%) and knowing when tests are due (95%). Over half (58%) find the Yearly Diagnostic Test Sheet useful during clinic and 35% feel it reduces time in clinic (data not shown).

In summary, adults with CF feel that the two data tools are well-organized and beneficial for clinic visits and hospitalizations. Initial evidence suggests the majority of patients agree that use of the tools reduces time spent in clinic. Findings are likely attributable to a patient's readily available knowledge of test dates and medications. Recruitment is in progress for 100 patients to further investigate whether perception varies by patient demographics and item specific measures of design and utility. These preliminary results indicate that the integration of hospital EMR, the data tools, and Port CF into one system will result in greater valuable time savings for CF staff and patients. 1.20, CI: 0.411-3.525) were not significant in our sampling. Of note, 5 patients who had documentation of refusing the recommendation of hospital admission all were ultimately hospitalized.

Discussion: Our preliminary analysis suggests that there are indeed patient-associated factors predictive of oral antibiotic failure in pediatric CF patients seen in clinic with pulmonary exacerbations. This could be useful when counseling patients and families and provides opportunity for prospective evaluation. Our limited sample size suggests a larger sampling period will be required to evaluate the true strength of these factors.

Supported by CF Foundation and NICHD K12HD057588. The Toronto Adult Cystic Fibrosis (CF) Program takes care of over 400 patients with CF. Typically 25-30 patients attend the ambulatory clinic. The duration of patients' clinic visits is usually long. The objective of this project was to examine the service provided and operations of this clinic, and to develop recommendations to improve the efficiency of the clinic.

Methods: Data were recorded on patient's arrival and appointment times, and the length of time spent with each team member. Interviews were conducted with the medical director, manager, team members and patients.

Results: Over a period of seven clinic days, data on 149/160 (93%) patient visits were collected. The peak arrival time occurred between 9:30 and 11am. Patients arrived 11 ± 52 mins (mean ± SD) before their scheduled appointment time. Thirteen patients had unscheduled clinic visits over the seven clinic days. They typically arrived between 10 and 11am. Patients spent a total of 3 ± 1.5 hrs in the clinic; with 1.5 ± 0.75 hrs spent waiting for lab tests and/or to be seen by health professionals. For those patients requiring lab testing, a total of 47 ± 27 mins was spent in the outpatient laboratory. Another contributing factor to patient wait times was waiting to see the staff physician (26 ± 38 mins). Waiting for spirometry did not significantly prolong the clinic visit.

Conclusions: Given the constraints in financial, physical and human resources, a few recommendations were proposed based on our findings. The data on average times spent with health professionals, including physicians, can be used to set up an appointment template for maximal clinic efficiency. Patients needing annual bloodwork, medical imaging and/or exercise testing need to be identified and flagged "before" the preceding clinic visit, so that they will be scheduled at the appropriate timeslots. To reduce disruption to clinic flow, we recommend educating patients to minimize drop-in visits and to improve adherence to clinic appointment times. Booking the follow-up clinic visit prior to departure from the preceding clinic visit and improving access to booking appointments through email may also improve efficiency. Patients needing annual bloodwork should have their appointments clustered together to allow on-site phlebotomy. This will minimize the time spent waiting in the outpatient laboratory. In order to facilitate clinic flow and minimize delays, if an unanticipated prolonged visit with the physician is required the patient will be asked to wait until later in the clinic or have an appointment scheduled at a later time to allow scheduled patients to be seen in a timely fashion.

Chaudhry, S.R.; Critz, S.; Filbrun, A.; Nasr, S.Z. University of Michigan, Ann Arbor, MI, USA Introduction: In an effort to improve quality of care at our CF center, we developed a post-clinic follow-up program to identify issues for follow up after a clinic visit.

Methods: Our pre-clinic review process is conducted in the week prior to the clinic visit, to help identify patient needs in the upcoming clinics. This includes the yearly laboratory work and CXR, support staff that the patient needs to see, the psychological issues that need follow up and the patient's nutritional concerns. These issues are summarized in a list that is distributed to all care team members and is displayed in the clinic as a guide for the patient visit.

We have recently instituted a post-clinic meeting in the week following the patient clinic visit. CF clinic nurses, a social worker, and the CF Center Director and Associate Director are present at this meeting. The team reviews the pre-clinic list and identifies patients who did not attend their visit. We review all patients that came to the clinic visit, and verify that planned testing was completed and all staff follow up was done. In addition, we review patient nutritional status (weight, height, BMI trends), PFTs, symptoms, throat/sputum culture, other test results, and treatment plan and follow up. Problems, concerns or suggestions in any area are noted and sent to appropriate team members for follow up.

Results: Over an 18 month period (10/15/08-4/15/10) our CF clinic held 1,477 patient visits. At the post-clinic meetings, at least one issue was identified in 16% of these patient visits. The Table summarizes our results and categorizes the identified issues.

Discussion:

• Of the 237 issues identified, the category with the largest number of cases was: follow-up to no show for appointment (20.3%). This is particularly important as we aim to adhere to the CF Clinical Practice Guidelines.

• The next most common categories were: downward trend in spirometry (17.3%), and follow-up to culture, lab or radiology results (16.9%). Attention to these areas provides a safeguard to ensure that patient care issues are addressed in a timely manner.

• The category with the fewest number of cases was: referral to a research study (1.3%) . After further analysis of this category, it was noted that the research coordinators and the team identify patients that are candidates for studies prior to clinic visits.

• Our post-clinic follow-up program has allowed us to identify issues that might have been missed otherwise, and are central to patient care. We believe that this program will enhance delivery of care to our patients.

Total Number of Patient Visits = 1477 using nutrition risk zones, each labeled by color code. Initial nutrition zones were as follows: Green-acceptable (BMI >25th percentile); Yellow-nutrition risk (BMI 10-25th percentile); Red-urgent nutrition risk (BMI <10th percentile). A nutrition education booklet was created as a tool to help patients and families understand the NRP and motivate them to partner with the CF team for improved nutrition status.

NRP Progression: 1.) Nutrition "zones" were revised in 2008 in concurrence with CF Foundation (CFF) goals for BMI >50th percentile. Current nutrition zones are as follows: Green-acceptable (BMI >50 percentile); Yellow-potential nutrition risk (BMI 26-49th percentile); Orange-nutrition risk (BMI 10-25th percentile); Red-urgent nutrition risk (BMI <10th percentile). 2.) The NRP was standardized to include any patient <50th percentile, and patient data is tracked until patient maintains BMI >50th percentile for five consecutive visits. Prior to this change, NRP initiation was subjective based on physician discretion. 3.) The nutrition education booklet was revised to include the new nutrition zones as well as increased information regarding gastrostomy (GT) placement. 4.) The NRP format was changed to increase utilization by the CF team by creating a document readily accessible in the patient's electronic medical record.

Results: Because of change in zone classification, data was compared using number of patients with BMI above 25th percentile. This number increased from 77 percent in 2006 to 84 percent in 2010. According to the CFF Care Center Data, average BMI has progressively increased from 46.7th percentile in 2004 to 53.6th percentile in 2008, with average BMI now exceeding the national goal of 50th percentile.

Conclusions: Using the CQI process, our center achieved improvements in nutrition status. Three elements contributed to our success: 1.) Periodic evaluation of processes and readiness to make changes allowed quality of care to remain relevant and effective for our patient population. 2.) Involvement of the entire CF team, rather than the CF dietitian alone, provided a cohesive approach in achieving nutrition goals. 3.) Frequent updates of progress to team members prompted members to maintain quality care of patients and offered incentive to continue reaching positive nutrition outcomes. Background: Due in part to infection control, many CF patients complete IV antibiotic therapy at home to treat pulmonary exacerbations, with an expected treatment duration of 2-3 weeks. A PNP role was developed to optimize the CF home IV program by initiating patient care referrals, providing education, collaborating with community agencies, organizing follow-up, and assessing patients throughout treatment.

Methods: The PNP collaborates with the multidisciplinary care team to determine a patient's eligibility for home IV therapy based on standardized criteria. To meet one goal of decreasing hospital length of stay (LOS), the PNP developed a patient-family guide for home IV therapy, coordinated discharge processes in the hospital, and evaluated patients in the ambulatory setting throughout the course of their therapy.

Results: During 2005-2010, an increasing number of patients were discharged on home IV therapy. Since the inception of our home IV program in 2007, 759 patient admissions (199 distinct patients) ages 3-52 years were admitted for CF pulmonary exacerbations. Since 2007, 345 (45.5%) exacerbations were treated with home IV therapy. On average, 94-100% of eligible patients per quarter were discharged to complete home therapy. The average total length of therapy was 20.3 days with an average LOS in the hospital of 6.2 days. Patient-family satisfaction with the program was high.

Conclusion: Integrating the PNP has been an innovative approach to improving the quality of care for CF patients discharged on home IV therapy. Our PNP-led home IV discharge plan has led to improvements in patient/family education and communication with home care networks. The Kam, K.; McConnell, A. University of Saskatchewan, Saskatoon, SK, Canada Background: Seasonal influenza results in significant morbidity and mortality amongst certain high risk groups including healthy children less than 2 years of age (yoa) and those with chronic medical conditions, such as cystic fibrosis (CF). In order to reduce transmission within the household to high risk groups, both the Canadian National Advisory Committee on Immunization and the American Academy of Pediatrics recommend annual vaccination of household contacts, but there is little information in the literature to document that this practice is followed.

Objective: To assess and compare influenza vaccination practices and beliefs of households of CF children with those of children less than 2 yoa and at least 2 yoa.

Methods: Following the 2008-2009 influenza season, caregivers of children seen in a Saskatoon tertiary care center completed a 20-question survey on influenza vaccination including vaccination status of the child (CF, under 2 yoa, at least 2 yoa) and the rest of the household. The effect of various factors in the decision to have household members vaccinated was assessed using a 5-point Likert scale from 1=Strongly negatively influenced to 5=Strongly influenced, and 0=Not a factor. Differences between groups was analyzed using SPSS® statistical software, with the level of statistical difference set at α=0.05.

Results: Eighty-two questionnaires were completed (28 by caregivers of children with CF, 24 of children under 2 yoa and 30 of children at least 2 yoa). Reported vaccination rates were 64.3%, 45.8%, and 16.7% amongst children with CF, children under 2 yoa and children at least 2 yoa respectively. Reported vaccination rates were 21.4%, 25.0%, and 6.7% amongst household contacts of children with CF, children under 2 yoa and children at least 2 yoa respectively. Advice from their physician was a significant influence in household contacts of children with CF for receiving the vaccine (p=0.001). Conversely, belief that they were too healthy was a major reason for the same group to refuse influenza vaccination (p=0.01). Inconvenient times and location of vaccination centers was a significant deterrent for household contacts of healthy children compared to those of children with CF (p=0.001). Other main barriers to vaccination were beliefs that the vaccine does not prevent influenza, and that its side effects were greater than its benefits.

Conclusion: Influenza vaccination rates in CF children and their household contacts are higher than that in children under 2 yoa and at least 2 yoa. Nonetheless, there is room for improvement. By understanding the motivations and barriers to vaccination, especially pertaining to household contacts of children with CF, more effective strategies may be implemented to improve coverage against influenza. is a process which emphasizes the need for frequent assessment of improvement efforts as well as the methods used to achieve quality outcomes. The Arkansas CF Care Center has utilized the CQI process to achieve improved outcomes in patients since initiation of our Nutritional Risk Pathway (NRP). The objective of this study is to assess the long-term nutrition outcomes that resulted from utilization of the CQI process. Method: A genetic counseling consortium reviewed research, identified evidence-based interventions, selected theoretical frameworks, delineated objectives, and selected specific counseling psychology micro skills.

Results: This innovative intervention draws from theories of emotion regulation, person-centered paradigms, and adult learning. It also incorporates empirically validated education and mental health techniques. Pilot testing showed the approach to be feasible. Parent feedback confirmed satisfaction with the approach designed to a) minimize parents' distress, b) facilitate parents' understanding, c) increase parents' capacities to use genetic information, and d) enhance parents' experiences with genetic counseling. To achieve these objectives, counselors engage in a highly interactive process of evaluating parents' needs and using specific counseling psychology micro skills to tailor responses accordingly. Corresponding assessment and intervention domains include the therapeutic environment, parents' and infants' emotional needs, parents' informational needs, and a follow-up plan. Presentation will include preliminary efficacy results and videotaped examples of counseling micro skills.

Conclusion: This promising new model is the first known attempt to establish a theory-driven, evidence-based standard for genetic counseling following an abnormal newborn screen for CF. Additional research will evaluate the efficacy of the model in clinical practice.

Supported by National Human Genome Institute (1R21-HG4252), National Institute of Nursing Research (P20-0NR008987), the Institute for Translational Research (1UL1RR025011) Background: Based on the CFF Nutrition Guidelines, nutritional status is based on BMI percentile in patients 2 to 20 years and BMI in patients > 20 years. In keeping with our trend to improve patient outcomes through Quality Improvement Initiatives, our Center embarked on a nutritional theme project beginning in 2007. The global aim of our project was to improve the median BMI outcomes in our patients. The specific aim of the project focused on the calculation and adjustment of pancreatic enzyme doses utilizing the measurement of lipase units per kilogram per meal. Our goal was to optimize enzyme doses in all of our pancreatic insufficient patients to meet the guidelines of 2,000 to 2,500 lipase units per kilogram per meal.

Methods: At each visit, our patients provide us with a medication recall list. The list includes brand, dosage and number of pancreatic enzymes consumed with each meal and snack. Our center dietitian meets with all patients and calculates the lipase units per kilogram taken at each meal and snack. Our goal is to adjust the enzyme dosages to maintain patients in the desired range of 2,000 to 2,500 lipase units per kilogram per meal and snack. A history of poor weight gain, excessive stools or flatulence prompts the often needed adjustment of enzymes. As patients gain weight, the pre-emptive adjustment is also offered to meet the desired goals. At each visit, our dietitian also reviews caloric intake and the importance of adhering to pancreatic enzyme use with all meals and snacks. Patients are reminded to check medication expiration dates and the importance of proper storage of enzymes to prevent loss of medication potency. Once the enzyme calculations are reviewed by the dietitian and physician, the patients are given instructions as to what their adjusted dose will be.

Results Conclusions: A standardized surveillance program for enzyme adjustment to improve nutrition and growth has improved BMI percentiles in our patient population from 2-20 years. More time may be needed to assess the impact on patients > 20 years. We believe a proactive approach of adjusting pancreatic enzymes contributes to improving nutritional outcomes. Background: The Children's Hospital/ University of Alabama at Birmingham (CHS/UAB) pediatric CF care center is a large regional center that has adopted structured, interdisciplinary distribution of care. Significant improvements in care quality and heath related outcomes have been recognized as a result of this care delivery model. However, the care team has grown larger and the burden of care provision has increased significantly. In 2009 outpatient clinic sessions were increased from 2 to 5 per week with a goal of improving care efficiency, delivery and segregating Pseudomonas aeruginosa negative patients to an alternate clinic site.

Methods: Following implementation of these changes three electronic web-based surveys were administered to the entire CF care team to assess team satisfaction and stress as a result of the increased care commitment.

Results: The survey revealed that increased CF clinic sessions did not result in a significant increase in workload or stress for the majority of the team (58%). Most team members remained satisfied with team communication (89%), received respect (94%) and felt recognized for achievements (100%). Negative comments revealed important weaknesses to interdisciplinary care at our outpatient planning meeting.

A second survey was distributed to assess the outpatient planning meeting. This survey revealed that the team did not believe the meeting supported interdisciplinary discussion of patients at an appropriate level. Some team members expressed concern that they were not allowed to express their opinions in meetings or lacked confidence to voice care related concerns.

The third survey attempted to assess team cohesiveness as described in "The Five Dysfunction of a Team: A Leadership Fable" by Patrick Lencioni. The survey revealed the team to be at risk of four of the five dysfunctions: absence of trust, fear of conflict, lack of commitment, and inattention to results. The fifth dysfunction was identified as a probable dysfunction of the team: avoidance of accountability.

Conclusions: Based on the results of these 3 surveys, center leadership concluded that restructuring of the outpatient meeting was necessary. Inter-health and can even affect respiratory function. The overall prevalence of vitamin D insufficiency in the CF population has been documented as high as 90% from large CF centers [1] . In 2002, a CF Foundation Consensus Committee established guidelines for optimizing bone health in CF individuals, including screening for vitamin D insufficiency and treatment protocols for achieving and maintaining normal vitamin D concentrations.

Objective: The purpose of this quality improvement (QI) initiative was to evaluate the prevalence of vitamin D insufficiency and the effects of vitamin D supplementation at one CF center that cares for pediatric and adult patients.

Methods: A needs assessment was done to characterize the problem of Vitamin D insufficiency in our patient population and to develop an action plan. A review of vitamin D levels was collected from medical records of 75 CF patients (35 pediatric and 40 adult patients). Data included patient demographics, serum 25-hydroxyvitamin D [25(OH)D] levels, and the use of vitamin supplements. A vitamin D clinical protocol was also established and began immediately once a patient was identified as deficient.

Results: The prevalence of vitamin D insufficiency [25(OH)D < 30 mg/dl] was 50% in the pediatric population and 57% in the adults. With intervention, 42% of these patients achieved vitamin D levels >30 ng/mL.

Conclusions: This QI initiative was effective in better understanding the prevalence of vitamin D insufficiency in our CF population and in optimizing vitamin D levels. The tracking mechanism and protocols developed as a part of this initiative are now ongoing. The incidence of cystic fibrosis related diabetes (CFRD) increases with age. The onset is often insidious and delays in diagnosis can result in an avoidable deterioration in both pulmonary function and clinical status. An oral glucose tolerance test (OGTT) is recommended annually (UK CF Trust and CFF guidelines). At our Centre, prior to 2007, OGTT was offered annually on an ad hoc basis. Patients who had long distances to travel to the Centre were frequently noted to decline this test. From mid 2007 the OGTT was included as part of the out-patient annual review, but in addition a pilot scheme was introduced offering patients an OGTT at home.

Aim: To determine whether the addition of home OGTT testing to a structured annual review clinic improves uptake of annual OGTT screening.

Method: The hospital electronic patient records and community nursing records of all patients attending the Adult CF Centre at King's College Hospital between January 2007 and December 2009 were reviewed to identify those patients who had an OGTT performed at home or as part of an annual review in hospital. Patients known to be diabetic were not included.

Results: Patient numbers at our Centre have increased annually. Patients eligible for OGTT screening, after exclusion of those with CFRD, are shown (Table) . Overall there has been an increase in OGTT uptake from 42 to 80%. By offering a home OGTT service more patients have received annual screening. Everyone offered an OGTT at home accepted this.

Conclusion: The inclusion of the OGTT as part of the annual review has increased the number of patients screened for CFRD at our Centre. Offering home-based testing has further increased the uptake for OGTT screening. The service has been well received by our patient group, with no additional resource implications. As a result we plan to continue offering OGTT at home for patients who have to travel long distances.

ventions include: re-introduction of meeting skills at team meetings, reorganizing the outpatient planning meeting to request focused interdisciplinary input and enrollment of inexperienced team members into the online QI collaborative soon to be available through the CF Foundation. A team retreat has been planned which will focus on team discovery exercises such as personality assessment and discussion of team strengths and weaknesses. Introduction: Pulmonary exacerbations in cystic fibrosis (CF) patients contribute to accelerated lung function decline. Improved detection of exacerbations through patient self-monitoring could improve clinical outcomes and slow lung function decline via earlier therapeutic interventions. Therefore, we developed a multifaceted home spirometry program in a population of adult CF patients. We hypothesized that increased home surveillance would facilitate earlier reporting and more expeditious and effective treatment of pulmonary exacerbations.

Methods: Ninety adult CF patients (mean age, 30 yrs; 60% female; mean FEV 1 , 60% predicted) at UNC received portable spirometers (PiKo-6, nSpire Health) and logbooks containing tracking sheets and treatment algorithms. Patients were encouraged to record FEV 1 values, respiratory symptom scores (CFQ-R) and weight at least weekly when they were at their baseline state of health. An algorithm for reporting signs of an exacerbation was provided. During periods of perceived illness, patients were instructed to self-monitor daily and increase airway clearance. FEV 1 data was downloaded at each return clinic visit. One year into the program, a survey was mailed to participating patients in order to assess their perceptions of the program.

Results: To date, 71 patients (84%) have returned to clinic after enrollment. Of these returning patients, 48 (68%) have returned with their spirometer at least once. The median proportion of visits in which patients returned with their spirometer (the return rate) was 60%. The median return rate of the 48 patients who have returned with their spirometer at least once was 93% (mean, 78%). The median number of documented spirometer uses was 0.52/week (mean, 0.76/week; range, 0-6.4/week).

Thirty-eight of the 88 patient surveys (43%) have been returned. Of those who found it difficult to adhere to the weekly recordings, the most frequently reported reason was forgetfulness (86%), while 23% reported that they did not want to see their lung function numbers. Regarding the program's utility, the following percentages of respondents agreed or strongly agreed with the following statements: The program is useful (84%); Obtaining a reproducible FEV 1 is easy (68%); Using the program allows me to detect when I'm getting sicker earlier (78%); The program prompted me to call my caregiver at a time when I typically would not have (43%); Use of the program helps me maintain my lung health (67%).

Conclusions: Overall, administration of this program was feasible, as evidenced by the high percent usage of those returning to clinic with their spirometers. Additionally, survey respondents overwhelmingly believed in the value of the program for helping them detect exacerbations earlier and preserve their lung function. Based on survey responses and feedback from non-users, an aversion to seeing FEV 1 values appears to be a prominent psychological barrier to use. Despite the subset of never-users, we believe that continued emphasis on patient self-monitoring at home, including the use of home spirometry, should be integrated into the standard of care for most adult CF patients. Although not statistically significant, those who failed eradication or had an early recurrence were older (mean age 13.6 versus 7.3 yrs, p=0.05), were more likely to have a prior history of Pa (71% versus 33%, p=.17) and were more likely to have a mucoid strain of Pa (43% versus 7%, p=.28). In the 9 months prior to protocol implementation, 13 patients (median age 9.1 yrs, range 1.1-17.4 yrs) had a new Pa infection. Eight patients were treated with TOBI®, three with TOBI® plus ciprofloxacin, one with IV antibiotics followed by oral ciprofloxacin, and one was untreated. Pa eradication rate was similar (77%), but of the four patients who failed eradication or had a recurrence, 1 was untreated and 1 did not receive TOBI®.

Conclusions: These results suggest that a consistent, standardized approach to new Pa infections can be successfully implemented at a large pediatric CF center. Our initial eradication rate was 75%. Eradication failures were more likely in older patients with a prior history of Pa infection, particularly the mucoid phenotype. Education about therapies is a key element for patients/families to consistently use them. Patient and/or family understanding, acceptance, and compliance with specific medications and treatments were surveyed as one part of an ongoing quality improvement project.

Methods: Surveys were distributed to patients and/or families during their routine three month visits in the pediatric CF clinic at the University of Iowa. Surveys were anonymous and targeted specific best practices for treatments/medications and had a wide variety of topics ranging from nutritional practices to chronic medication use. For each specific therapy, we asked three questions on a 1-10 scale with "never/poor" being the lowest and "always/good" as the highest. Questions included: 1) How well do you understand why you (or your child) are on this medication/treatment? 2) How much do you agree with and/or accept that this medication/treatment is needed?

3) Have you (your child) been consistent with regularly using this medication/treatment? Results: For all therapies and medication surveyed, understanding tended to be scored the highest and compliance the lowest. Understanding and agreement were generally highly correlated. For inhaled medications, twice a day treatments had lower patient reported compliance compared to once a day treatments. Azithromycin MWF had the lowest understanding and poorer compliance than was expected.

Conclusions: Patients and families generally understood why they were on specific medications and treatments and agreed that they should be on these. Compliance was consistently the lowest measure and was especially poor when understanding was not high. Inhaled medications with increased frequency of administration generally had the poorest compliance. Agents with less intuitive rationales for use, such as azithromycin, showed poorer understanding, acceptance, and compliance. Education about why azithromycin is used in CF may be an avenue to increase compliance with this medication and improve outcomes.

Supported by a grant from the CFF (STARNE09QI).

drove recommendations for antibiotic route: oral/inhaled (OI) antibiotics (mild-mod), outpatient IV antibiotics (mod), and inpatient IV antibiotics (mod-severe). Follow-up visits were standardized to ensure re-assessment at set intervals based on severity. PES at follow-up drove subsequent treatment duration and algorithms for alternative diagnostic evaluation for patients not improving as expected. For each PEx, initial PES, antibiotic route, PES at follow-up, and time to follow-up were recorded. Clinicians completed PExT satisfaction surveys and a newsletter describing the PExT was distributed to patients and families. Results: PEx treatment data was compared from the 6 months prior to implementation of the PExT with treatment data post-use. PEx rate in both time periods were similar (29% v 33%). The use of the PES and PExT were 98% and 88%, respectively. The use of OI (51% v 55%) and IV (49% v 45%) were similar for both periods. Mean PES for patients treated with OI and IV antibiotics were similar in both periods (OI: 6.9 v 6.4 and IV: 11.2 v 9.4) . Mean times to follow-up for IV antibiotics were similar (16+/-12.5 v 14+/-12 days). However, mean time to follow-up for OI was reduced after implementation with less variability (27+/-21.4 v 20+/-8.9 days). Clinicians reported overall satisfaction with the PExT for ease of use and quality of therapy.

Discussion: We describe a standardized PEx treatment algorithm that utilizes the PES to define severity of a PEx, guide treatment strategy, and follow-up. Similar, standardized treatment algorithms for other conditions such as myocardial infarction, acute stroke, and asthma exacerbation improve outcomes and reduce cost. Decreasing variability in clinical practice rapidly identifies patients who do not respond to therapy for further diagnostic evaluation. Use of the PExT reduced variability in time to followup for OI treated patients. There were no other changes in our approach to treating a PEx, largely because the algorithm was based on current practice. However, with uniform, aggressive follow-up and re-assessment, we hypothesize that eliminating variability in practice will result in improved lung function. Importantly, the PExT was well-accepted by clinicians. We propose that the PExT is a critical step in the development of national, standardized PEx treatment guidelines to improve clinical outcomes.

Zemanick, E.T.; Towler, E.; Accurso, F.J.; DeVoogd, R.; Wagener, J.; Sagel, S.D. Pediatrics, University of Colorado Denver, Aurora, CO, USA Background: Chronic infection with Pseudomonas aeruginosa (Pa) is associated with more rapid pulmonary function decline and increased morbidity and mortality in cystic fibrosis (CF). Early eradication regimens consisting of antibiotics directed against Pa are effective in clearing bacteria and delaying the development of chronic infection.

Objective: To provide a rapid response to new isolation of Pa in airway cultures of patients with CF.

Methods: We developed a standardized protocol for treatment of new Pa infections. Individuals with new isolation of Pa, defined as either first positive culture or positive culture after at least 1 year and 2 negative cultures, were identified for this quality improvement (QI) initiative. Patients who were not ill or had only mild symptoms were treated with inhaled tobramycin (TOBI®) for 1 month and oral ciprofloxacin (if ≥1 year old) for 2 weeks. Patients experiencing a pulmonary exacerbation were treated with intravenous (IV) antibiotics followed by 1 month of TOBI®. Follow-up cultures were obtained 1-2 months after completion of TOBI® to evaluate for eradication. Patients with a second positive culture were treated with 10-14 days of IV therapy followed by 1 month of inhaled TOBI®. Patients remaining culture positive after two attempts of eradication, or with a subsequent positive culture within 1 year, were considered chronically infected and treated with alternate month TOBI® and thrice weekly azithromycin. We tracked patients who qualified for this QI initiative, determined eradication and recurrence rates, and examined factors associated with eradication failures. We also compared eradication rate to that of patients with new Pa infections detected in the 9 months prior to protocol implementation.

Results: Twenty patients (12 male, median age 6.9 yrs, range 2 mo-20 yrs) were identified with new Pa infection during the first 6 months of this QI initiative. Fourteen were treated with TOBI® plus ciprofloxacin, 1 was treated with TOBI® alone, and 5 were treated with IV antibiotics followed Background: It is widely accepted that body mass index (BMI) directly correlates to pulmonary function tests in CF patients. Our center BMI for 2007 measured below the national average. As a quality improvement process, our multidisciplinary team created a nutrition and psychosocial algorithm to assess all patients' nutritional risk and identify barriers to improve BMI. In this pilot study, the goal was to assess the effectiveness of the revised algorithm at improving BMI of individuals at our center.

Methods: We retrospectively reviewed data of all adult CF patients seen at the clinic from December 1, 2008 through December 1, 2009. Psychological assessments reviewed the presence of depression, anxiety, and body dysmorphic disorder. Nutritional assessments reviewed the utilization of oral supplements, enzyme dose adjustment, appetite stimulants and enteral tube feedings. Invariant, patient-specific parameters, such as age, gender, family income, education, and psychological assessments were controlled for with patient-specific dummy variables in the fixed effects model. A Hausman test was performed to eliminate potential random bias.

Results: In this pilot study, the study population includes 164 patients; complete data was available on 40 patients thus far. The population was characterized by: mean age of 30 years (±8.8), 16 females (40%), baseline BMI of 21.7 (±3.6), baseline FEV 1 of 62% (±25.5), and baseline hemoglobin A1c of 6 (±1.2). When controlling for patient-specific parameters, the BMI of the 40 patients following the nutrition and psychosocial algorithm increased by an average of 0.35 kg/m 2 in one year (p=0.057). Further analysis showed that in patients with a starting BMI of 20 kg/m 2 or less, (12/40 patients), placement on the nutrition algorithm and psychosocial algorithm increased BMI by an average of 1.22 kg/m 2 (0.58 -1.86; 95% CI; p<0.001) in one year. Neither the starting BMI nor the starting FEV 1 significantly affected the rate of BMI gain. For the subset of patients starting with a BMI above 20 kg/m 2 , the nutrition and psychosocial algorithm did not significantly change the BMI.

Conclusion: A significant improvement is seen in the CF patients with BMI of 20 kg/m 2 or less with the utilization of the nutrition and psychosocial algorithm. This has served as an effective tool at identifying key therapies to improve patient BMI. Further analysis is underway to identify which intervention of the nutritional and psychosocial assessments treatment phase has the most significant impact on BMI. Although the sampling size is small, we believe the impact of the nutrition and psychosocial algorithm will reach significance when the data from the entire center is analyzed.

Background: Developments in national CF data registries are creating opportunities to compare patient outcomes and clinical practice between treatment centres. The patient registry of the US CFF has been used effectively for such purpose (Gawande 2004, Schechter and Margolis 2005) . Smaller national registries do not have the numbers to deliver benefits on the same scale. For this reason, international benchmarking at centre level is of interest. Arrangements for sharing a limited range of indicators, which have been in place between the US and Australian registries in recent years, illustrate the benefits.

Objective: To extend to an international level the benefits obtainable from centre-level benchmarking of CF patient outcomes and treatments.

Methods: A short list of indicators that are widely used for monitoring patient outcomes has been identified, as well as standard methods of construction and common disaggegations for risk adjustment (Sims 2009). The indicators include FEV1 % predicted, BMI for adults and, for chldren and adolescents, BMI percentiles derived using population-based reference values. Median values at centre level are calculated from registry data, excluding centres with small numbers of patients in any category.

Data: Column charts for selected indicators show median values in 2008 for US and Australian centres, overlaid in sequence from highest to lowest values, with Australian centres distinguished and identifed by a code. Better nutrition outcomes for Australian children are suggested, supported by evidence of improved outcomes reported at the national (Australian) level over recent years.

Conclusions: Benchmarking typically raises questions that suggest worthwhile supplementary investigation. Results presented here indicate that further study of child nutrition outcomes may be valuable. Opportunities for potential benefits on a broader scale could be realised through international CF benchmarking data distributed through a website. This would require support from national CF registries along with agreed governance arrangements to ensure data integrity and access security.

Patient/family involvement: A proposal for an international CF data resource website is being developed in consultation with Cystic Fibrosis Worldwide.

Gawande A (2004) It is easy to identify malnourished patients whose Body Mass Index (BMI) is very low. However, a patient whose BMI is slightly low or normal may be at risk for nutritional failure due to other factors and is more difficult to identify. We propose using a standardized scoring sheet to identify patients at risk for declining nutrition status during clinic visits will improve BMI. To test this hypothesis, we developed a nutritional scoring system, employed it for 1 year and determined its ability to identify patients at nutrition risk and prompt an intervention.

Methods: A scoring tool was developed using 6 criteria related to the nutrition status of a patient: (1) current BMI of the patient, (2) weight change since last visit, (3) current appetite, (4) stool quality, (5) energy level, and (6) the presence of CFRD. All possible answers for each parameter were given certain point values and a total score was calculated. A "positive score" was ≥ 3 points. Any patient with a positive score received at least one intervention aimed at improving their nutrition status. In addition, patients who scored a 3 or 4 received phone or email follow-up prior to their next clinic visit, or referral to a subspecialist. Patients who scored ≥ 5 were asked to return to clinic in 4-8 weeks versus 3 months for closer monitoring.

Results: Data collection occurred between February 2009 and February 2010. In our initial review, three outcomes were evaluated: (1) how often scoring sheets were completed, (2) how often they yielded a "positive" result, and (3) whether the provider agreed with the result. Ninety-eight percent of the scoring sheets were completed, and 32% of the time a "positive" score was obtained. The providers agreed with the result of the scoring sheet 95% of the time. The BMI for the patients with scores ≥ 3 are found in the Table. For each instance of a positive score, one or more of the following interventions were employed: dietary changes (47%), treatment of pulmonary exacerbation/other (23%), enzyme/PPI changes or GI referral (15%), and insulin changes or endocrine referral (10%). Five percent of interventions were not listed. According to the 2009 PortCF registry data, the median BMI percentile trend in our pediatric patients increased over four quarters (45.4, 51.4, 48.0, and 49 .0 for quarters 1-4 respectively Methods: We initiated a QI program including: 1) assessment of BMI at each clinic visit; 2) flag charts for patients with BMI<50%; 3) BMI reminder flyers in clinic; 4) redesign of clinic note; 5) nutrition treatment protocol; and 6) develop a nutrition education role for each provider during clinic visits.

Program analysis involved collection of data on clinician behavior, nutrition treatment plans, and patient BMI. Evaluation utilized a pretestposttest design with multiple observations to determine behavior maintenance throughout the study period. Nutrition status was tabulated every 6 months over 2 years, and then reassessed after 2 more years to determine program maintenance.

Results: At baseline, the mean BMI/age was 38.1. Four years later, mean BMI/age increased to 43.7 among patients who attended the clinic at least annually. Provider involvement remained high, with RD involvement rising from 70% of patients with RD visit to 92.3%. MD and SW visits related to nutrition also increased in frequency. Use of supplements increased from 47% to 87.3%. All changes were significant to the p <0.002 level.

Conclusions: The QI project increased rates of adherence to nutrition recommendations, including the use of BMI as the primary nutrition marker and recommendation of aggressive nutrition treatment. The QI program demonstrated that simple behavior cues can produce changes in clinician behavior, which, in turn, impacts patient outcomes. Large-scale QI projects can show ongoing improvements even after the immediate intervention ends if the intervention has significantly impacted clinic culture. The need for BMI/age >50th percentile and high caloric intake are well-documented in CF, but influential factors continue to need further analysis, especially as CF patient demographics grow more diverse. The following study analyzed factors associated with BMI/age in non-CF research to determine impact on CF populations.

Methods: A total of 135 patients at the CHLA CF Center were studied. Data was collected via chart review on various factors with potential to adult patients, the median BMI values trend did not change (21. 9, 21.3, 21.3, and 21 .8 for quarters 1-4 respectively).

Discussion: It appears that the scoring tool was most beneficial in the pediatric clinic with 24% of patients being identified at risk that might have been missed based on one criterion alone. Most importantly, an increase was seen in BMI percentiles in our pediatric patients during the course of this study. The results do not seem as promising for our adult patients, possibly because they are less willing to make changes to their diet. CFF support. The positive relationship between lung function, measured as FEV1, and BMI percentile is well documented in the cystic fibrosis (CF) medical literature. The CF Foundation has established a goal of maintaining BMI greater than the 50th percentile for age. As part of our CF Center's participation in the CFF Quality Allies project, a quality improvement committee was formed to review current nutritional outcomes at our center and implement a process to increase BMI percentiles, focusing on decreasing the percent of patients under the 25th percentile BMI for age.

Hypothesis: Nutritional outcomes, as indicated by BMI percentile for age, can be improved by initiating an aggressive program of nutritional intervention through case management.

Methods: A multidisciplinary committee was formed, including physicians, nurses, dietitians, social workers, a psychologist and the CF Center Coordinator. The group was tasked with reviewing our center's nutritional outcome data, identifying patients in or at highest risk for nutritional failure and developing a comprehensive plan to improve their nutritional status. Barriers specific to each patient attaining adequate nutritional status were identified and individualized intervention plans were implemented. Progress was measured by monitoring BMI percentiles over time with a goal of reducing the percent of patients less than the 25th percentile BMI for age for all CF patients in the CF Registry at the Baylor Pediatric CF Center. The study initially reviewed patients eight years of age and younger and was later expanded to twelve years of age and younger.

Results: Over an eight month period of time in the population eight years of age and younger, the percent of patients with a BMI 10th percentile for age or less was reduced from 8.1% to 4.6% and the percent of patients with a BMI 10th to 25th percentile for age increased from 10.8% to 12.8%. The scope of the committee was then expanded to patients twelve years of age and younger over a ten month period of time. The percent of patients less than or equal to the 10th percentile BMI for age was reduced from 5% to 3.5% and patients with a BMI for age between 10th to 25th percentile increased from 12.2% to 13.4%. Overall, the percent of patients with a BMI at the 10th percentile or less who are at the highest nutritional risk was reduced from 8.1% to 3.5% and the percent of patients at the 25th percentile or less BMI for age was reduced from 18.9% to 16.9% during the study.

Conclusions: Aggressive case management of individual CF patients in or at highest risk for nutritional failure with a multidisciplinary committee and individualized interventions can improve nutritional outcomes.

impact BMI: ethnicity (Hispanic vs. Non-Hispanic), home location (urban vs. suburban), type of nutrition therapy (Diet alone, diet + supplements, diet + GT feeds), nutrition-relevant discussions with providers (MD, RD, MSW), and medical care payor. Patient BMI was tracked over 18 months. Data was analyzed using paired t-test and multiple regression to determine association with BMI change over time. All factors were controlled for age as age results in a natural increase in BMI over time.

Results: Ethnicity was not a statistically significant factor for prediction of BMI change over time nor did BMI differ significantly between various ethnic groups. Home location did predict BMI outcomes as patients in urban home locations (defined as population density greater than 5,000 persons per square mile) had a higher BMI/age than those in rural areas. Type of nutrition therapy did not independently predict BMI change. Provider involvement was predictive of BMI increase over time with those patients who interacted with all three providers (MD, RD, SW) regarding nutrition status showing the most improvement over time. Medical care payor was not predictive of BMI outcome; however, it should be noted that attendance at clinic and RD visits was impacted by payor and only those patients who had a payor who permitted participation in the full scope of CF treatment had statistically significant change in BMI over time.

Conclusions: Although it is often assumed that caloric intake alone impacts nutrition status, a variety of factors have been demonstrated in research to influence BMI in the general population. This study examined several of these factors in a diverse CF population. The most influential factors related to BMI increase over time were access to healthcare, provider involvement in nutrition treatment, and home location. The best type of nutrition therapy did not emerge in this study. It may be that any intervention that increases caloric intake is likely to influence BMI to some extent. Ongoing investigation will continue to illuminate controllable factors that can be manipulated to aid CF patients to achieve BMI/age at recommended percentiles. Background: Pancreatic sufficiency (PS) is present in 10-15% of patients with cystic fibrosis(CF). These patients have sufficient pancreatic enzyme secretion to prevent overt steatorrhea. Pancreatic sufficient (PS) patients are known to have a milder CF phenotype than pancreatic insufficient(PI)patients and prior studies have shown absence of complications such as diabetes, liver disease, rectal prolapse or vitamin deficiencies. Studies published decades ago showed excellent growth in pancreatic sufficient patients. After noting that several of our PS CF patients persistently fell into the nutritional "at risk" or "urgent need" categories, we undertook a study to assess nutritional status in this population.

Objective: To compare the nutritional and gastrointestinal status of pancreatic sufficient and pancreatic insufficient children with CF at a single center that has met the national goal for CF nutrition (mean BMI> 50th percentile) since 2005.

Methods: This was a cross-sectional, retrospective study performed at a single point in time. A total of 129 children with CF aged 2-19 years, from Children's Memorial Hospital CF center were categorized into PS (N=20) and PI (N=109) groups. All PS patients had normal fecal elastase levels on at least one occasion after the first birthday. Patients of each group were divided into 4 categories based on their BMI percentiles: <10th, 11-24th, 25-49th and >50th percentile. The presence of G-tube, constipation and laxative usage was compared between the PS and PI patients with < 50th and > 50th percentile BMI. Fisher exact test was used for testing statistical significance.

Results: In the PI group (n=109) 5.5% of children were in the BMI <10th percentile, 10% in the 11th-24th percentile, 26.6% in 25th-49th percentile and 57.7% in >50th percentile. In the PS group (n=20) 5% had BMI <10th percentile, 10% in 11th-24th percentile, 25% in 25th-49th percentile and 60% in >50th percentile. The p value was statistically insignificant for all groups.

Conclusion: In a cystic fibrosis center that achieves the national goal for nutrition, there was no difference in the nutritional status of pancreatic sufficient and enzyme supplemented pancreatic insufficient children with CF. Constipation was higher in the PS group with BMI <50th percentile. These findings demonstrate that with good nutritional practice, PI patients are nutritionally indistinguishable from PS patients. Moreover, a significant proportion of PS CF patients have significant nutritional challenges. Tube feedings (TF) are frequently recommended for patients with CF when oral intake is not sufficient to promote adequate weight gain and/or growth. Most research related to TF and CF review nutrition and pulmonary outcomes after tube placement. Little has been published regarding practice standards related to placement and management of enteral feeds, or use of enzyme replacement therapy in CF patients receiving TF. We hypothesize that there is considerable variability in patient assessment and education prior to tube placement as well as TF management among CF care centers in the U.S. and internationally.

Methods: An electronic, 43 question survey was developed to assess issues pre and post tube placement and TF practices at U.S. and international CF care centers. The survey link was sent to members of the CF Registered Dietitian (RD) nutrition listserv (CF-NUTRITION@listserv.dartmouth.edu).

A total of 109 RDs and 1 physician responded: 93% from the U.S. and 40% caring for both children and adult with CF.

Results: An upper GI was the most common test performed prior to tube placement (44%) however 25% of the RDs were unsure of the type of testing performed at their center. Contraindications for the placement of a tube included: adherence problems (28%); severe liver disease (17%) and unknown (28%). The type of formula chosen for TF varied (i.e. semi-elemental vs. intact, 1 kcal/ml to 2 kcal/ml).

Enzyme recommendations ranged from enteric coated enzymes by mouth or via tube, and/or non-enteric coated enzymes added to formula. Enzymes were given before, during, and/or after tube feeds. There was also variability in dose of enzymes prescribed for TF.

Significant pain (69%), feeding intolerance (68%), leaking (35%), and reflux (25%)were common complications reported immediately after tube placement. However, there were many RDs who were unsure of the estimated frequency of TF complications.

Most RDs reported an important role in discussion of tube placement with patients and families.

Conclusions: There is variation in pre-surgical testing, formula selection, enzyme use, and RD knowledge of TF care across CF care centers. These findings indicate a need to develop evidence-based clinical care guidelines for TF in CF. They also indicate that if RDs have a role in educating patients and families about TF, efforts must be made to broaden their comprehensive understanding of the process, including testing and risk of complications, to provide appropriate information to patients and families to assist them in making informed care decisions regarding tube placement and TF.

Recommendations: Further research is needed in the following areas: 1. Patient/family perspectives of TF; 2. Identification of patients likely to benefit from and have potential complications from TF; 3. Pre-placement workup; and 4. Optimal enzyme administration and type of formula with TF.

Results: Control CF mice had increased intestinal permeability as compared to control WT mice, demonstrated by a significantly greater level of plasma rhodamine-dextran at 90 min post-gavage and more serum albumin in the intestinal lumen. In the control CF intestine, histochemical detection of IAP was noticeably less than in WT. There was a 60% reduction in Akp3 mRNA and a 70% reduction in IAP enzyme activity in the control CF intestine. The control CF mouse lung exhibited mild thickening of the epithelium, increased neutrophils, and increased MPO activity. Treatment of CF mice with laxative or oral antibiotics decreased intestinal permeability by about half, and restored IAP mRNA and enzyme activity to WT levels. Analysis of lung parameters in treated CF mice is under way.

Conclusions: The CF mouse intestine has impaired mucosal barrier function as shown by increased permeability to macromolecules and reduced IAP activity. There is also a mild inflammatory state in the uninfected CF mouse lung. Treatments that eradicate bacterial overgrowth of the small intestine improve barrier function and normalize IAP expression. These data suggest that bacterial overgrowth contributes to the impaired mucosal defenses in the CF intestine. Future work will determine the extent to which these experimental manipulations that improve intestinal barrier function also reduce inflammation of the CF mouse lung.

Supported by NIH grants R21 AI083479 and P20 RR024214. The n-6 and n-3 polyunsaturated fatty acids are required in diet and are crucial for numerous metabolic and physiological processes. The major dietary n-6 and n-3 fatty acids are the 18 carbon n-6 linoleic acid (LA,18:2n-6) and n-3 alpha linolenic acid (ALA,18:3n-3). LA is metabolized by delta 6 and 5 desaturases to dihomo-γ-linolenic acid (DGLA, 20:3n-6) and arachidonic acid (ARA, 20:4n-6), while ALA is metabolized to eicosapentaenoic acid (EPA, 20:5n-3). Metabolism of EPA via 22:5n-3 to docosahexaenoic acid (DHA, 22:6n-3) involves delta 6 desaturation and chain shortening in peroxisomes; 22:5n-6 synthesis from ARA via 22:4n-6 uses a similar pathway. The functions of n-6 and n-3 fatty acids are related to their roles in phospholipids (PL), but PL are diverse, differ in fatty acids and have complex, inter-related metabolism. Abnormal n-6 and n-3 fatty acid patterns occur in CF, but their etiology and relevance to CF complications is incompletely understood. We used quaternary gradient HPLC resolution of lipids and GLC to separate and analyze lipids from erythrocyte membranes of children with and without CF. We separated and quantified 1-alkenyl-2 diacyl ethanolamine PL (EPG, plasmalogen), and diacyl ethanolamine (PE) and choline (PC) and their fatty acids. EPG are major PL on the inner membrane surface, are synthesized in peroxisomes, and contain high DHA and ARA, but low LA. PE and PC are synthesized from diglycerides in the endoplasmic reticulum, and PC can be formed by methylation of PE. PC is higher in LA and lower in ARA and DHA than PE. Changes in PL and their synthesis lead to changes in n-6 and n-3 fatty acids. Diet fat composition also modifies cell membrane fatty acids.

Children with CF had lower EPG (172.4 ± 8.69 and 180.6 ± 8.26, µg/mL cells) but higher PE in red cell membranes than children without CF (301.3 ± 11.72 and 281.3 ± 12.97 µg/mL, respectively), n=20/group. Although choline and the PC/PE ratio is decreased in plasma of children with CF, the red cell PC/total PE ratio was not different between the CF children and control children without CF. DHA was decreased and 22:5n-3, 22:5n-3/DHA and 22:4n-6/22:5n-6 were increased in EPG, elevated 20:3n-6/ARA was present in all PL of children with CF. The results suggest impaired peroxisomal lipid metabolism, as impaired fatty acid desaturation. However, dietary fatty acid intake is an important determinant of cell membrane fatty acids. Dietary records were collected for 75 children with CF and 45 children without CF. CF children consumed as mean +SD percent total energy, 48+7.2 total fat, 26+13.4 saturates, 12.6+4.3 monounsaturates, 5.6+1.3 n-6 and 1.1+0.6 n-3 fatty acids, with 110+68 mg ARA, 42+55 mg EPA and 88+86 mg DHA/day. High fat intakes were due to foods high in saturated fatty acids, and this gave an unbalanced dietary fat profile; some children consumed <10 mg/day EPA and DHA. In conclusion, while the lipidomics confirms and extends the altered fatty acids in CF, interpretation is complicated by the deviations in dietary fat profile from that of age and gender matched reference children. The metabolic significance of high proportions of dietary fat from saturated fat should be considered.

Supported by a grant from CFF. Liquid chromatography combined with electrospray ionization mass spectrometry (LC-MS) enables detection of hundreds of metabolites, sometimes termed metabolomics, which is useful for broader understanding of disturbances across multiple pathways and discovery of previously unknown metabolic changes that predict or result from disease. Our focus is the biochemical basis for and implications of disturbed methyl and thiol metabolism among children with cystic fibrosis. Using this approach, we first analyzed urine from children with CF and healthy children without CF, and identified consistently lower hippurate levels in urine from the children with CF. Hippuric acid is the hepatic conjugation product of benzoic acid and glycine, thus implicating metabolic limitation of conjugation due to limiting glycine. To elucidate possible changes in amino acid metabolism, including glycine, we next developed a method for liquid-chromatography and fluorescence detection in combination with electrospray mass spectrometry to probe ortho-phthaldialdehyde and mercaptopropionic acid derivatives of primary amines found in biofluids and tissues. This method detects amino acids and other biologically relevant primary amines such as glutathione and polyamines. To date, with this approach analysis of 36 urine samples from children with CF showed lower excretion of glycine and serine compared to children without CF, but elevated N-acetylputrescine and more specifically elevated N-acetylcadaverine (p<0.05). Putrescine and cadaverine are polyamines which are important in numerous cell events, and which also require S-adenosylmethioninamine, the decarboxylated product of S-adenosyl methionine (SAM) for further metabolism to spermidine and aminopropylcadaverine, respectively. Using LC-MS, we analyzed plasma from 84 children with CF and 45 healthy control children and found reduced methionine (30.9+1.43 and 25.8+1.55 µMol, respectively). A major path of glycine synthesis is via serine hydroxymethyltransferase using SAM and serine as substrates, but our analysis showed not only decreased methionine but also reduced urinary glycine and serine in children with CF. In conclusion, integrating our results across metabolic pathways shows consistent evidence of disturbed methylation and low glycine with effects that extend to multiple pathways.

Supported by CFF. Methods: In a group of 75 pediatric patients with CF (mean age: 14.7 ± 3.1y), fat-free mass (FFM) and fat mass (FM) were measured using Dual-energy X-ray absorptiometry (DXA). Nutritional failure was defined as BMI percentile < 10th (often referred as "red zone") and FFM depletion as FFM index (FFMI) percentile < 5th according to commonly used standards. In addition, bone mineral density (in z-score) and lung function (as reflected by forced expiratory volume in 1 second (FEV1)) of each patient was recorded and the relationship with BMI percentile and body composition was evaluated. Statistics was done using ANOVA and t-tests when appropriate.

Results: Of the total group, 29% was characterized by nutritional failure according to the criteria BMI percentile < 10th and/or FFM depletion. Of the studied CF subjects, 13% had a BMI percentile < 10th, 25% was FFM depleted, and 15% was characterized by "hidden" depletion (BMI percentile > 10th and FFMI percentile < 5th). Of the FFM depleted patients only 42% was also characterized by BMI percentile < 10th. A stronger relationship was found between FFMI and lung function in boys than girls (r: 0.43, p<0.001 vs. r: 0.30, p=0.058, respectively). Stratification of the group into BMI categories of 10 percentiles showed a gradual reduction of FFMI and FEV1 values below BMI percentile < 50th with a significant drop in FFMI and FEV1 at 20th BMI percentile (p<0.05). In contrast, FMI was unchanged between 10th and 50th BMI percentile. We observed that the 20th BMI percentile was the threshold associated with a significant increase in the percentage of patients with FFM depletion (50% at 20th BMI percentile vs. 18% at 30th BMI percentile), a reduction in bone mineral density z-score and FEV1 values below 80% predicted.

Conclusion: BMI percentile below the critical threshold of 20th needs to be considered as "red zone" that predicts loss of muscle mass, bone mineral density and compromised lung function in pediatric patients with CF. [NaCl] , may underestimate their fluid needs during exercise in the heat, resulting in a state of "involuntary dehydration." Physically-active CF patients were compared to healthy physically-active individuals with "normal" sweat NaCl loss to determine the effect of dehydration on physiological responses and fluid/electrolyte balance during exercise. It was hypothesized that CF subjects would have attenuated increases in blood osmolality and thus decreased osmotic stimulation of thirst relative to change in hydration status compared to individuals with normal sweat [NaCl] .

Methods: Six physically-active cystic fibrosis patients (CF) and 8 physically-active healthy individuals (Control) performed cycling in the heat (32-33°C, 35% relative humidity) at 50% of VO 2 max until 3% body weight loss.

Results: Mean (± SD) regional sweat [Na + ] (132.6 ± 6.4 mM) and [Cl -] (127.0 ± 12.1 mM) were higher (p<0.001) in CF compared to Control (43.7 ± 9.9, and 41.9 ± 6.9 mM, respectively). Whole body sweat rate relative to body mass (12.0 ± 2.0 mL/kg/hr), exercise time required to reach 3% dehydration (125.7 ± 19.9 min), and core temperature at exercise termination (38.3 ± 0.4°C), did not differ between CF and Control (p>0.05). Compared to Control, CF had smaller changes in dehydration-induced serum osmolality (6.1 ± 4.3 vs. 14.8 ± 3.5 mOsm/kg H 2 O, p=0.002), serum [Na + ] (2.9 ± 1.7 vs. 7.5 ± 1.0 mM, p<0.001), and serum [Cl -] (-0.8 ± 0.7 vs. 4.9 ± 1.0 mM, p<0.001). In contrast, percent change in plasma volume from baseline was greater (p=0.03) in CF (-19.3 ± 4.5%) than Control (-14.3 ± 2.3%). Rating of thirst relative to dehydration and changes in plasma concentrations of dipsogenic hormones aldosterone, vasopressin, and angiotensin II were not different (p>0.05) between CF and Control. During recovery with ad libitum sports beverage, serum [Na + ] (by 2.8 ± 2.1 mM) and [Cl -] (by 4.8 ± 1.3 mM) decreased below resting values in CF despite a 30% lower (p=0.04) fluid intake (relative to body weight) compared to Control.

Conclusion: Individuals with CF exhibit attenuated blood osmolality but greater relative plasma fluid loss at moderate levels of dehydration,

White, H. 1,2 ; Conway, S. 1 ; Peckham, D. 1 1. Adult Cystic Fibrosis Unit, Leeds Teaching Hospitals, Leeds, United Kingdom; 2. Nutrition and Dietetic Group, Leeds Metropolitan University, Leeds, United Kingdom Aim: To determine the impact of enteral tube feeding (ETF), decline of ETF or treatment optimisation on nutritional status and lung function at 1, 2 and 3 years.

Method: Retrospective analysis of patient records at the adult cystic fibrosis unit, Leeds, UK (n=350), yielded 31 patients [16 male, mean age 21.7 (3.4) years] who met the UK standard criteria (Cystic Fibrosis Trust, 2002) for initiation of ETF from Jan 2004 -2010. Each was categorised according to acceptance of ETF (n=17), decline of ETF (n=5) or had treatment optimisation (n=9). Feeding route was noted as gastrostomy (n=9), nasogastric tube (n=7), jejunostomy (n=1). Treatment consisted of commencement of pancreatic enzyme replacement therapy (n=2), adjustment of pancreatic enzyme replacement therapy and commencement of oral supplements (n=4) and social and psychological interventions (n=3). Mean time taken for patient decision to undertake ETF after initial discussion was noted and mean weight (kg), BMI (kg/m 2 ) and FEV1(%) was recorded at baseline, 1, 2 and 3 years. Trends at each time point were analysed using a general linear model (repeated measures).

Results: Time taken to commence ETF following initial discussion is 10.6 (12.0) months. Patients accepting ETF or treatment optimisation had significant weight gain at 1 and 3 years. Those declining ETF had significant weight loss at the same time points (See Table) . There was no significant decline in lung function for patients who chose ETF or treatment optimisation but a significant reduction at 3 years for those who declined ETF (see table) .

Conclusion: Patients took a mean of 10.6 months in deciding to undergo ETF, indicating their difficulty in acceptance and decision making regarding invasive nutritional intervention. ETF or treatment optimisation both resulted in sustained long term weight gains of 13% and 14% respectively at 3 years and BMI within the normal range (BMI >19kg/m 2 ). Declining ETF resulted in 15% weight loss at 3 years and accompanying reduction of 47% in lung function. Positive outcomes in patients who underwent treatment optimisation indicates that clinical judgement in non-discussion of ETF is justified in some patients.

Nutritional and pulmonary outcome in patient groups fulfilling the criteria for ETF * P < 0.05, **P < 0.001 resulting in similar ratings as Control subjects for thirst during exercise in the heat. However, "involuntary dehydration" reported to occur in CF and observed via the blunted appetite for hypotonic fluids in the recovery phase of this study, may reflect physiological cues directing behavioral responses to preserve salt balance over volume restoration. Therefore, fluid replacement recommendations that focus primarily upon restoration based on body weight loss may be contraindicated for CF patients and further research is needed to identify optimal fluid and electrolyte replacement strategies for this population. Objectives and Study: Cirrhotic cystic fibrosis liver disease (CCFLD) develops in 5-10% of CF patients. CCFLD is mostly diagnosed based on physical examination in combination with ultrasound (US) of liver and spleen. Unfortunately, these diagnostics cannot predict which patients will develop CCFLD in the future. Currently, ursodeoxycholic acid (UDCA) is the only treatment employed to prevent CCFLD. It might be beneficial if UDCA could be administered in an early phase to patients with a high risk of CCFLD. We aimed to determine if liver biochemistry, in particular a solitary serum γ-glutamyltransferase (GGT) elevation, could predict the development of CCFLD.

Methods: We retrospectively analyzed the records of all 280 CF patients, 2-18 years of age on 1-1-2007, from our two CF centers. CCFLD was defined as a heterogeneous and nodular aspect of the liver on periodical US in combination with clinical or US proven splenomegaly. The date that patients first met this definition was used as date of diagnosis. GGT results 2 years prior to the diagnosis were evaluated. We used as controls historical GGT results of CF patients with normal liver US and no signs of splenomegaly at the age of 15 years (n=50 patients, available GGT results: mean: 9, range 1-27). We excluded elevated GGT values if additionally alanine transaminase (ALT) was increased (2xULN).

Results: Fifteen patients met our definition of CCFLD (4.5%). All had established CCFLD before the age of 15 years. We calculated mean, median and peak GGT values. At least 2 GGT results were available in the 2-year period before the diagnosis of CCFLD in 12/15 patients (80%). Using ROC analysis, the mean GGT value of the available results was identified as the best predictor for CCFLD (AUC: 0.89). Consecutive cross tab analysis showed that a cutoff value GGT above 30 IU/L (mean based on at least 2 measurements), in the absence of ALT elevation was a significant independent predictor for the development of CCFLD in the period 0-2 years prior to the diagnosis. A GGT > 30 IU/l could predict CCFLD with a sensitivity of 83%, specificity of 92% and a PPV and NPV of 71% and 96%, respectively. Based on these results the odds ratio for a CF patient age 2-15 years with a persistent GGT above 30 U/l, in the absence of ALT elevation, for developing CCFLD in the following 2 years was 60 (95%CI 10-374).

Conclusion: Our results in a large cohort of CF patients strongly indicate that isolated serum GGT elevations could predict the development of CCFLD. Our results provide a perspective for targeted prophylactic treatment for CCFLD. Recent advances in treatment have increased the lifespan of individuals with cystic fibrosis (CF). A major factor contributing to increased survival has been identifying the relationship between optimal nutrition and overall health. The CF Foundation (CFF) has established consensus reports that guide nutrition professionals' care of patients. However, the guidelines do not encompass all aspects of care. Therefore, uncertain areas are left to clinical judgment and center-specific preferences. The aim of this study is to assess nutrition practices of CF centers and to identify areas where more research might lead to evidence-based guidelines.

Methods: A 25-question web-based survey was created by a multidisciplinary team and distributed to nutrition professionals via the Dartmouth listserv after Institutional Review Board approval. Questions were grouped into categories and included infant feeding practices, pancreatic enzyme replacement therapy (PERT), and bone health. Respondents were allowed to choose more than one option for multiple choice questions and were given the opportunity of free text on 16 of the 25 questions. Responses were entered anonymously from July-October 2009.

Results: There were 76 respondents to the survey, with a response rate of 21% based on a total of 361 listserv members from the USA, Canada, Australia, and the UK. The survey provided insight into many areas of professional practice. Some are highlighted below:

Infant Feeding Practices: 87.9% of respondents stated that administration of enteric coated beads in acidic foods is important. However, when asked the rationale for this practice, only 14 of 47 respondents (29.8%) mentioned that acidic food prevents the beads' enteric coating from disintegrating before reaching the duodenum.

PERT: Results were most varied regarding both the timing and dose of enzymes administered with gastrostomy (GT) feeds. Most respondents (66.7%) give enzymes before and after GT feeds, using meal dose for each. Although most respondents use enteric coated enzymes given by mouth (89.5%), many also use non-enteric coated enzyme powders for GT feeds (46.1%).

Bone Health: Although CFF guidelines suggest assessing vitamin D levels based on season, most respondents (86.8%) do not follow this guideline. The survey found that 71.1% of respondents supplement with vitamin D3, but dosing is extremely varied. DXA scans are performed routinely at most centers (64.5%); however, only 4 respondents stated that bisphosphonates are frequently used. Only 34% of respondents recommended calcium supplements for adolescent patients.

Conclusion: Although the CFF has provided excellent guidelines for patient care, this survey identified areas where more research might result in evidence-based guidelines that could improve patient care through standardized strategies. As well as identifying potential research areas, the survey also suggests that additional education regarding rationale for certain practices may be beneficial to professionals.

Grujic, D. 1 ; Harrison, A. 2,3 ; Piedra, J. 3, 5 ; Szymanczyk, S. 3, 5 ; Szwiec, K. 3, 4 ; Bala, T. 3 Standard therapy for EPI in humans, including patients with CF, is porcine-derived pancreatic enzyme replacement therapy (PERT). However, nutrient absorption remains low and potential for viral contamination is high. Thus, we posited a new approach for treatment of EPI in children and adults with a microbial product, Lipromatase, rapid disintegrating tablet (LipRDT), which contains 3 highly purified and stable enzymes: crystalline protease, amorphous amylase, and crystalline, cross-linked lipase. We tested the product in young pigs with EPI, in which pancreatic duct ligation causes a total lack of pancreatic enzymes and diminished levels of bicarbonate that results in poor nutrient absorption, steatorrhea and arrested growth. We used young EPI pigs to demonstrate an effect of LipRDT on fat and protein digestion (%CFA, %CNA), absorption (growth, TG and NEFA muscle strength) and overall health.

Methods: Two doses of Liprotamase, high (HRDT:290,000 U Lipase, 24,000 U Amylase, 194,400 U Protease), low (LRDT:146,00 U Lipase, 12,000 U Amylase, 102,000 U Protease) or placebo were given bid with a high fat diet (HFD) as 4 tablets/meal (~40 g fat) to 8 EPI pigs. The cross over study included 1 control and 2 treatment periods (each 1 wk) with a washout in between. As a comparator, 5 healthy pigs were maintained on the HFD.

Results: One week of treatment with either dose of RDT improved fat and protein digestion, and normalized stool appearance and weight. Treatment CFA was 84% and 79% with HRDT and LRDT respectively, from basal of 24%, p<0.05 vs. 92% in healthy. Similarly, treatment CNA was 60% and 58%, from basal of 34%, p<0.05 for each vs. 82% in healthy. Adequate food digestion with HRDT and LRDT resulted in normalization of postprandial blood lipids; TG doubled 3h post-meal, reaching 0.71±0.5 and 0.75±0.4 mmol/L at 2h and 3h respectively, similar to 0.72±0.2 mmol/L in healthy. Blood NEFA peaked 6h after meal and were 0.25± 0.11 and 0.24 ± 0.01 mmol/L, again similar to 0.26 ± 0.06 mmol/L in healthy. As a result, body weight gain was 6.0% and 8.0% with HRDT and LRDT, respectively. Surface electromyography (sEMG, skeletal muscle performance parameter) demonstrated that RDT supplementation restored muscle contractility to levels recorded in healthy pigs fed HFD. L. dorsi muscle sEMG response upon stimulation of 200Hz tripled from non-treatment period (from 0.37±0.08 to 0.95±0.31 and 1.17±0.31mV/kg bw with LRDT and HRDT respectively, similar to 1.43±0.58mV/kg bw in healthy) suggesting proper neuronal control and nutrient supply.

Conclusions: LipRDT treatment facilitated digestion and absorption of nutrients, leading to normalization of postprandial TG and NEFA levels, improved muscle contractility and growth of EPI pigs. Thus, Liprotamase represents an effective alternative (fewer tablets and virus risk free) to traditional PERT for EPI in humans. Background: In the last 18 months we have identified a number of patients attending our unit with symptoms of classical acute gout. There are no previous reports of elevated uric acid or gout in CF. Here we describe 7 cases with a diagnosis of gout, and report elevated serum uric acid in a randomly selected population of patients.

Cases: Seven patients at our regional centre have been diagnosed with acute gout in the last 18 months (2% total patients). All patients are Caucasian males, pancreatic insufficient with normal renal function. One patient was diabetic and 5 are DF508 homozygotes. Median age was 36 yrs (range 21-54 yrs). All presented with classical podagra (painful, hot swelling of the 1st metatarsal), though the knee was also involved in one case. Joint aspiration was only performed in one case, and confirmed the presence of uric acid crystals. Symptoms were managed with NSAIDs (4/7), allopurinol (3/7), colchicine (1/7) or codeine (1/7). Mean (SD) highest serum uric acid in these patients was 9.9 (1.2) (normal range up to 7 mg/dL). No patients were on any medications known to increase uric acid or precipitate gout.

CF population: Most recent uric acid levels were reviewed from 200 randomly selected CF patients. Patients who had described gout symptoms were excluded. Uric acid levels were available in 182 (108 male).

Mean (SD) uric acid was 6.6 (1.5) in males and 5.2 (1.4) mg/dL in females. This is significantly greater than in historical controls: 5.1 (1.1) in males (n=2283, p<0.0001), 4.0 (0.9) in females (n=2844, p<0.0001). Mean serum creatinine was 65 µmol/L (max 108 µmol/L, normal range 60-120 µmol/L).

Taking the upper limit of normal as 7 mg/dL (males) and 5.7 mg/dL (females), 40 men (37%) and 27 women (36%) had elevated levels. Maximum recorded level was 10.9 mg/dL. Twenty-two patients (12%, 1 female) had levels >8 mg/dL, historically associated with a high risk of developing gout (>33%).

Clinical significance: Elevated serum urate is common in patients with CF, and may be responsible for a significant number of cases of gout. The reasons for this are unclear, but may be related to diet (high fructose intake) or metabolic dysfunction. Gout may be unrecognised in some and should be considered in all patients with single joint or recurrent arthritis, particularly if there is a history of elevated urate. Introduction: Pancreatic enzyme replacement therapy (PERT) is essential to achieve adequate nutrition and appropriate weight-/height-for-age in subjects with exocrine pancreatic insufficiency (EPI) due to cystic fibrosis (CF). PERT studies have focused on fat absorption; however, nitrogen absorption is equally important as deficiencies can lead to decreased protein stores and metabolic deficiencies.

Objective: To investigate the relationship between malabsorption severity and magnitude of improvement in protein absorption with pancrelipase assess the efficacy of Ultrase® MT12 (UMT12) in the control of steatorrhea in children 2 to 6 years of age with CF and PI. Secondary objectives were to assess the signs and symptoms of malabsorption and the safety of UMT12. Eligible subjects entered a 9-day baseline phase and maintained their prescribed PEP along with a high-fat diet of approximately 4 g fat/kg/day. Their prescribed PEP was then replaced by an equivalent dose of lipase from UMT12, after which subjects entered a 19-day treatment phase. Up to 10 stool samples were collected in the last 5 days of each phase and were pooled for analysis of stool fat by Nuclear Magnetic Resonance (NMR). Steatorrhea was defined as stool fat ≥ 30%. Subjects with stool fat < 30% were categorized as "responders." Efficacy and safety measures were compared between the baseline and treatment phases.

Forty-nine children (30 boys and 19 girls) with a mean (SD) age of 3.8 (1.43) years were enrolled from 14 US sites, and included in the ITT population. The mean (SD) lipase intake during the stool collection period while on UMT12 was 2112 (682) UI/kg/meal or 2482 (858) UI/g fat ingested. In the primary efficacy population (ITT using 50th percentile imputation for missing data for 4 subjects), mean (SD) stool fat with prescribed PEP was 30.6% (9.1) and with UMT12 was 31.8% (8.1); the proportion of responders with prescribed PEP was 46.9% and with UMT12 was 42.9%. For the 45 subjects with complete data, mean (SD) stool fat with prescribed PEP was 30.6% (9.2) and with UMT12 was 31. 8% (8.4) , and the proportion of responders in the two arms was 47.9% and 46.7% respectively.

The mean (SD) daily stool frequency with prescribed PEP was 1.54 (0.72) and with UMT12 was 1.43 (0.57). Approximately 75% of stools during both phases were of normal consistency. Other secondary efficacy parameters (abdominal distention, abdominal pain and flatulence) were similar between prescribed PEP and UMT12, with a positive trend toward more days without abdominal complaints with UMT12. There was no significant change in mean body weight throughout the study. The most common AEs were gastro-intestinal or respiratory disorders, consistent with CF. Most AEs were mild and assessed as not related to PEP. No serious adverse event occurred, and no subject discontinued due to an AE possibly related to PEP.

In conclusion, Ultrase® MT12 is safe and effective in controlling the signs and symptoms of malabsorption. Based on the percentage of fat in the stool, the proportion of responders while on prescribed PEP and on UMT12 was similar.

Russo, M.; Soultan, Z.N.; Anbar, R.D. Pediatrics, Upstate Medical University, Syracuse, NY, USA Introduction: Distal intestinal obstructive syndrome (DIOS) has been reported to occur in 10-15% of children with cystic fibrosis (CF). It occurs almost exclusively in pancreatic insufficient patients, and manifests by partial or complete intestinal obstruction at the level of distal ileum, cecum, or proximal colon. Contributing factors are thought to include abnormal intestinal mucins, undigested food residues, prolonged intestinal transit time, bowel dilatation, and abnormal intestinal electrolyte and water transport. We report regarding the anatomy of the colon in children with CF and DIOS, which was characterized during radiological studies at the SUNY Upstate Medical University CF Center, Syracuse, NY.

Method: We reviewed the records of children with CF who were diagnosed with DIOS from 2003 through 2010, at our CF Center. The criteria used to diagnose these patients included a history of abdominal pain and constipation with evidence of bowel obstruction on physical examination and/or abdominal x-ray.

Results: During the study period, 16 of 143 pediatric patients followed at the Center (11%) were diagnosed with DIOS. Four of the 16 patients had recurrent DIOS. Fifteen of the 16 patients were pancreatic insufficient. Of the 16, 10 underwent gastrografin enema in order to treat their DIOS, which also confirmed the diagnosis. Of these 10 patients, 5 were found to have anatomic abnormalities during the radiographic study, including colonic tortuosity and redundancy. Two of 5 children with colonic abnormalities had a history of recurrent DIOS.

Conclusions: Half of the children with CF and DIOS at our Center who underwent a gastrografin enema were documented with colonic abnormalities. We speculate that the prolonged oro-rectal transit time in these patients as a result of their anatomic abnormality is an additional risk factor for the (pancreatin) delayed-release capsules (CREON ® ) using the coefficient of nitrogen absorption (CNA).

Methods: This was an analysis of individual patient data from two double-blind, randomized, multicenter, placebo-controlled, cross-over studies (NCT00690820, NCT00510484) of pancrelipase in subjects with CF and EPI. Subjects were randomized to one of two sequences, pancrelipase then placebo or placebo then pancrelipase (5 days each with 3-14 day intervening washout on subjects' usual PERT). Individual differences in CNA according to age group on pancrelipase vs placebo were compared with CNA values on placebo (lower values indicate more severe malabsorption).

Results: Overall, 16 subjects aged 7-11 years (mean dose 4472 ± 743 lipase units/g fat/day) and 31 subjects aged ≥12 years (mean dose 4189 ± 732 lipase units/g fat/day) were included. The LS mean CNA on pancrelipase vs placebo was 80.3% vs 45.0% (7-11 years) and 85.1% vs 49.9% ( ≥12 years); both p<0.001. Individual CNA values on pancrelipase were similar regardless of protein malabsorption severity; thus, subjects with more severe malabsorption had greater improvements in CNA on pancrelipase vs those with less severe malabsorption (Fig). There were no obvious differences according to age (Fig) . Similar results were seen when on-treatment changes in CFA were plotted by CFA values on placebo; the CFA data correlated well with the CNA data.

Conclusions: Subjects with EPI due to CF with more severe malabsorption had greater improvements in CNA on pancrelipase than those with less severe malabsorption, giving similar on-treatment values in all subjects. Improvements in CNA are relevant to nutritional status. The data imply high predictability of improvement in protein absorption during treatment with pancrelipase, regardless of malabsorption severity.

Funded by Abbott, Marietta, GA. Background: Calorie supplements are often prescribed to patients with cystic fibrosis (CF) to increase energy intake for the purpose of weight gain. However, few studies have evaluated whether supplementation elevates total energy intake to the recommended level of 120% of energy requirement.

Objective: The objective of this study is to describe calorie supplementation relative to estimated energy requirement in children diagnosed by routine newborn screening (NBS) in Wisconsin.

Methods: The study population includes a subgroup of 62 children enrolled in the Wisconsin Routine CF NBS Study (WI-Routine Study) who completed at least one 3-day food diary. The WI-Routine Study is an ongoing, observational investigation initiated in 2006 to evaluate the benefits of routine NBS in Wisconsin, which was implemented in 1994. Food diaries were distributed to families bi-annually. Infants (N=9) were excluded from analysis leaving 84 food diaries completed for 53 children, aged 1-14 years. Average daily energy intake was expressed as a percent of the estimated energy requirement (EER) based on the 2005 Dietary Reference Intakes. When calculating EER for those over 3 years, an active physical activity fac-tor was assumed. Descriptive statistics are from the first food diary; regression model included all data and accounted for repeated measures.

Results: Overall, median dietary energy intake (food only), as percent of EER (pctEER), was 116%. Twenty-two subjects (42%) reported calorie supplement intake. Median pctEER from diet was similar between supplement (118%) and non-supplement (114%) users (non-parametric p=0.9). Supplemental calories contributed, on average, 14% (SD=15%) to total pctEER. However, median total energy intake (food plus supplements) as pctEER was not significantly higher in supplement users (126%) compared to non-users (114%, non-parametric p=0.4). Fifty-nine percent of supplement users had energy intake greater than 120% of the EER, compared to 39% in the non-supplemented group (p=0.17). Supplement use varied by phenotype (p=0.048), with the greatest use in those born with meconium ileus (MI) (6 out of 9 subjects, 67%), followed by those who were pancreatic insufficient (PI) without MI (12 out of 25 subjects, 48%) and those who were pancreatic sufficient (PS) (4 out of 19 subjects, 21%). In a logistic regression model which included supplement use, phenotype, age and sex, the odds ratios (OR) of total energy intake greater than 120% of EER were significant for MI (OR=4.07, 95% confidence interval (CI)=1.18, 14.1; p=0.03) and PS (OR=0.27, 95% CI=0.01, 0.77; p=0.02), compared to PI. The OR for supplement users compared to non-users was 1.17 (95% CI: 0.44-3.10, p=0.75). Age and sex were not associated with high calorie intake.

Conclusions: Supplement use is highest in MI subjects and lowest in PS subjects. Total energy intake relative to estimated energy requirement does not differ between calorie supplement users and non-users. Further analysis will examine the association of supplement use to nutritional status.

( Celiac disease (CD, gluten induced enteropathy) is a severe chronic gastrointestinal disease. believed to occur in up to 1% of Caucasians. The common clinical features of CD (such as steatorrhoea, malnutrition, failure to thrive) are indistinguishable from common clinical features of cystic fibrosis (CF).

The coincidence of CF and CD has been reported previously, and one study based on clinical and biochemical, but not serological, evaluation followed by small bowel biopsy (SBB) confirmed the presence of CD in 5 of 1100 CF patients (1) .

In non-CF individuals, use of serological screening tests, particularly tissue trans-glutaminase (tTG) has increased the frequency of ascertainment of CD. We report results from the first survey of CF patients using modern serological testing (tTG) for ascertainment of CD followed by confirmation of diagnosis with SBB.

We screened 114 CF patients aged 1-18 years who attended the CF clinic at BC Children's Hospital in Vancouver. Seven patients had elevated tTG although they were clinically not distinguishable from other CF patients. Of these, 5 were positive for HLADQ2 or DQ8. SBB was positive for CD in 4 and negative in 1 (modified Marsh criteria (2,3)), while 2 patients await SBB. This is the first report of screening for CD in CF patients using tTG.

A serological prevalence in CF of ≥6% (7/114) and confirmed SBB diagnosis in ≥3.5% (4/114) is greater than expected for the general population.

Our results should be confirmed by other studies, but suggest that routine serological tTG screening for CD is indicated in all CF patients. 1 Valetta E A, and Mastella G. Acta Paediatr. Scand. 1989; 78:784-785. 2 Oberhuber G, Granditsch G, Vogelsang H. Eur. J. Gastroenterol. Hepatol. 1999; 11:1185- Introduction: Essential fatty acid deficiency (EFAD) has been well documented in cystic fibrosis (CF) and pancreatic insufficiency, and may be involved in inflammatory response. Several studies indicated that these deficiencies are independent of nutritional status, but they may be related to CF regulated tissues.

Objective: To determine the essential fatty acid profile in nasal epithelial and erythrocyte membrane in patients with CF controlled in our Center.

Patients and Methods: Cross sectional study. We enrolled patients with CF (n:11) > 6 years old, with genotype and mild to moderate lung disease and pancreatic insufficiency, and a healthy control group (n:10). The mean age was 10.1 years for both groups and Z score BMI for CF group -0.57. None of the CF patients had liver disease, signs of infection or nutritional supplements. Blood samples and nasal epithelial scrapings were drawn. Total lipids of plasma and nasal mucosa were extracted by chloroformmethanol 2:1 v/v (Folch method). Fatty acid methyl esters were analyzed by gas-liquid chromatography in a capillary column, and the fatty acid pattern was determined. Statistical Method: ANOVA test.

Results: Children with CF showed lower level of essential fatty acids compared to control, especially in Σn-6 and in total polyunsaturated fatty acids in nasal mucosa and erythrocyte membrane.

Conclusion: EFAD seems apparent in CF children when compared to a healthy control sample. The importance of these findings related to pulmonary disease should be determined, as well as the role of mutant CFTR in essential fatty acid deficiency.

Values are mean ± SEM. * p < 0.05 relative to control Background and Aims: After more than 50 years, there is still little consensus for a common cause of mucus accumulation in the lung, intestine, and various other organs affected in cystic fibrosis (CF). Our recent report showed that normal mouse small intestine mucus release requires cystic fibrosis transmembrane conductance regulator (CFTR) -dependent bicarbonate secretion, which may serve to prevent the formation of aggregated mucus. But how HCO 3 secretion functions in normal mucus release is still unknown. The aim of the present study is to better define the process of HCO 3 dependent mucus release. Methods: Segments of distal small intestines from CL57BL/6 wildtype mice were continuously perfused with glucose-free Ringers solution. Serial samples of luminal perfusates were collected and assayed for discharged mucus content by using WGA lectin binding reaction (Dot Blot) assay. For goblet cell exocytosis measurements, the periodic acid-Schiff The genetic potential for height is commonly estimated from mid-parental stature. However, our recent study (JCF 2010; 9:135) revealed that the validity of using mid-parental height adjustment is unclear in CF children of parents with discordant heights, i.e., tall father and short mother or vice versa, because the genetic contribution to a child's stature from both parents might not be equal.

Objective: The aim of this study is to examine parent-child height correlations according to degree of discordance between parent heights.

Methods: Examination of parent/child height correlations would ideally be conducted using data from a disease-free population. However, these data are not readily available. As an alternative, children with mild CF, i.e. pancreatic sufficiency (PS), were utilized because their height status was similar to healthy peers, shown below. Data from 682 PS patients aged 2-20 years with both parental heights documented in the 1986-2008 CFF Registry were used. PS was defined by not using pancreatic enzymes and having no PS-associated mutations. The most recent height and weight were normal in girls (n=326, height 50th ± 29 percentile; BMI 58th ± 28 percentile) and boys (n=365, height 46th ± 28 percentile; BMI 59th ± 28 percentile). In parent/child height analyses, heights were expressed as sex-and age-specific percentiles and z-scores (haz) by using the 2000 CDC growth reference, with reference values at age 20 applied to parental heights. Discordance in parental heights was defined by the difference between maternal and paternal height percentiles; namely, ≥50 percentile points as "very discordant," 25-50 as "discordant," and ≤25 as "concordant."

Results: Very discordant heights were observed in 164 pairs of parents (24%). Parent-child haz correlation coefficients are reported in the table. In general, mid-parent-child correlations were higher than mother-child or father-child correlations. However, the strength of mid-parent-child correlations lessened with greater parental height discordance, especially in girls. In fact, the correlations were statistically insignificant in girls of parents with very discordant heights.

Conclusions: Correlations between mid-parental heights and child heights are reduced in children of parents with discordant heights, especially in girls. This implies that in children having parents with very discordant height,~25% estimated from our study, new methods for adjusting genetic potential need to be developed.

(Supported by NIH grant DK072126.)

#p>0.05 for Pearson correlation test secretion, segments of distal small intestine were placed in Ussing chambers and HCO 3 secretion was measured via pH stat. Results: In the wild type mouse small intestine, carbachol (100 µM, an intracellular Ca 2+ mediated activator) induced goblet cell exocytosis and mucus discharge in the presence and absence of HCO 3 -, but did not stimulate sustained HCO 3 secretion. Basolateral HCO 3 -Ringers solution did not enhance carbachol induced goblet cell exocytosis or mucus release. In contrast, cAMP-mediated activators (VIP 100 nM or isoproterenol 10 µM) did not stimulate significant goblet cell exocytosis or mucus discharge, even with HCO 3 in the bath solution, but did induce a marked increase in HCO 3 secretion. With HCO 3 in the basolateral solution, carbachol plus VIP or isoproterenol induced significantly more mucus discharge into the perfusate compared to carbachol alone ( Figure 1A) . Interestingly, the enhancing effects of cAMP activators on carbachol-induced mucus discharge were absent when we removed the HCO 3 from the bath solution ( Figure 1B) . Conlusions: These results suggest that cAMP-mediated activators potentiate Ca 2+ -dependent mucin exocytosis in mouse small intestine by stimulating HCO 3 secretion. The authors' work is supported by Elizabeth Nash Memorial Fellowship, CFRI; MCCC-Cystic Fibrosis Foundation; NIH-RO1 HL084042. The authors thank Dr. Hui Dong for the use of his pH stat equipment and helpful discussions. Objective: To describe serum vitamin B 12 status and the predictors in a current cohort of children with CF and PI following current clinical care practices.

Methods: Baseline measurements were obtained from subjects with CF and PI (aged 5-17 yrs) with FEV 1 > 40% predicted and no known liver or metabolic disease participating in a multicenter nutrition intervention study. Fasting serum B 12 samples were analyzed by immunoassay. Height (HAZ), weight (WAZ), and BMI (BMIZ) Z scores were calculated. FEV 1 % predicted, liver enzymes, serum lipids, high sensitivity C reactive protein (HS-CRP), genotype, current medications, PERT use, and dietary B 12 intake from 3-day weighed food records were also assessed. Daily B 12 intake from CFspecific and over-the-counter vitamins was reported. Student's t-tests or Wilcoxon rank sum tests were used to compare outcomes for subjects with high serum B 12 (HI-B12) with those within the reference range for age and sex (NORM-B12). Significant predictors of B 12 status were determined using multiple regression and multiple logistic regression models.

Results: Serum vitamin B 12 ranges were assessed in 88 subjects (50 males, 10.8±3.0 yrs). Mean B 12 level was 1228±539 pg/mL (range; 387, 3455); 56% subjects were in the HI-B12 group, based upon lab reference ranges. Dietary B 12 intake was 6.9±3.4 mcg/d (range:1.5,18.1); ~400% the RDA, similar to age and gender peers (NHANES 2005 (NHANES -2006 . Supplemental B 12 intake was 15.5±18.6 mcg/d (range: 0, 150),~900% RDA. Compared to subjects with NORM-B12, those with HI-B12 were older (11.5±3.1 vs. 9.9±2.5, p=0.009), had greater supplemental B 12 intake (19.2±23.6 vs. 10.8±6.8, p=0.03), higher serum cholesterol (136±21 vs. 126±23, p=0.04), and lower FEV 1 (88±22 vs. 103±22, p=0.002). They were more likely to be using inhaled tobramycin (43% vs. 21%, p=0.03), and anti-fungal medications (14% vs. 0%, p=0.01). In multiple logistic models, the best predictors of HI-B12 status were FEV 1 (-), supplemental B 12 intake >12 mcg/d (+), and serum cholesterol (+) (R2=0.20, p<0.001). In multiple regression models, FEV 1 (-), serum cholesterol (+), and HS-CRP (+) (R2=0.21, p=0.001) were significant predictors of serum B 12 .

Conclusions: More than half of the subjects had elevated B 12 levels; none were deficient. Dietary plus supplemental intake of B 12 provided an average of 13x the recommended intake for healthy children. Elevated serum B 12 levels were associated with greater B 12 supplementation, higher serum cholesterol and poorer pulmonary function. These findings suggest serum status of B 12 is associated with excess intake and/or possibly better absorption. Elevated B 12 may be a biomarker of poorer respiratory status, or, alternatively, reflects an unknown mechanism.

Supported by NIDDK (R44DK060302) Introduction: Pancreatic enzyme replacement therapy (PERT) is essential for the treatment of patients with exocrine pancreatic insufficiency (EPI). The key efficacy measure of PERT in the treatment of EPI is the coefficient of fat absorption (CFA). Although rarely performed in clinical practice, CFA measurement involves a 72-hour stool collection, which presents a logistical challenge due to hospitalization to ensure collection requirements in clinical studies.

Objective: To demonstrate suitability of sparse stool fat evaluation as an alternative to complete 72-hour collection and measurement of CFA in subjects with EPI due to cystic fibrosis (CF).

Methods: This was a data analysis of a double-blind, randomized, placebo-controlled, two-period cross-over trial in subjects aged 7-11 years with EPI due to CF. Although the primary endpoint was CFA, informed consent was given to permit measurement of percent fat (PF) in complete bowel movements. CFA was calculated from known fat intake and fecal fat contained in stool samples collected over 72 hours. PF was determined by nuclear magnetic resonance from each complete bowel movement during the collection period. Four values for each subject were derived and analyzed: mean PF from the first and last bowel movements on each day; mean PF from three randomly selected bowel movements (each on a different day); PF from a single random sample selected irrespective of day. The PF data were compared to CFA as a dichotomous variable and the analysis included evaluation of sensitivity, specificity, and positive predictive value (PPV). Area under the curve (AUC) values (1 = 100% accuracy) were obtained from receiver operating characteristic plots of sensitivity vs 1-specificity. Correlation analyses were performed to evaluate the association between PF and CFA as a continuous variable with three groups of samples: active treatment, placebo, and both combined. Scatter plots were used to verify the linear relationship between the two variables. A negative correlation and slope would indicate that as one variable (e.g. CFA) increases, the other (e.g. PF) decreases, and vice versa.

Results: Twelve subjects provided samples (10 in both treatment periods and 2 in only one period). For the multiple-sample methods, PF values below a 30% threshold were highly predictive for CFA values ≥80%, as shown by PPV, sensitivity, and specificity values ≥0.89, with high accuracy (AUCs >0.92). PPV, sensitivity, and specificity values were lower for the single sample method (≥0.78) with an AUC of 0.91. Correlations between PF and CFA as a continuous variable were high for the multiple sample methods (Pearson values >0.81/Spearman values >0.79 for active and combined samples), had a negative slope in the scatter plots, and were statistically significant (p<0.001) for active and combined samples but not for placebo samples.

Conclusion: Sparse stool sampling to measure PF is a possible method to determine adequate response to PERT; however, a larger study is required to establish broad use in subjects with EPI.

Supported by Abbott, Marietta, GA, USA. g/cm 3 (SD±0.30. Patients in the treatment group had worse lung function, more chest infections, courses of antibiotics and oral steroids, lower energy, protein, vitamin D, calcium intakes and BMI as well as fall in D 3 levels over the year. In non-CF peak BMD is thought to be achieved around 25 yrs. Mean change in Z score at the femoral hip for placebo group ≤ 25 yrs was +0.150 g/cm 3 (SD±0.17) compared with -0.33 g/cm 3 (SD±0.33) in the treatment group. Mean change in Z score at the femoral hip for the placebo group ≤25 yrs was -0.086 g/cm 3 (SD±0.13) compared with those 〉25 yrs having no mean change. Mean change in Z score at the lumbar spine for the placebo group ≤25 yrs was +0.09 g.cm 3 (SD±0.17) compared with -0.089 g/cm 3 (SD±0.25) in the treatment group. Participants 〉25yrs in the placebo and treatment groups had mean changes at the lumbar spine of -0.029 g 3 (SD±0.11) and -0.05 g/cm 3 (SD±0.39) respectively.

Conclusion: At baseline approximately 30% were osteopaenic/osteoporotic and 60% vit K deficient despite standard CF nutritional therapy. Supplementation with vit K for one year improved carboxylation of osteocalcin but had no significant effect on BMD, except in those aged 〉25yrs vit K may have a protective effect against decline in BMD. Background: Pancreatic insufficiency (PI) is observed in 85-100% of patients with CF.The probability of pancreatic insufficiency can depend on the underlying gene defect.The gold standard for determining pancreatic status is the faecal elastase test (normal range 〉200 mcg/g, mild/moderate PI 100-200 mcg/g,severe insufficiency〈100 mcg/g). All PS patients have annual faecal elastase tests as PS patients can change to PI. However not all PS patients report normal bowel habits despite normal faecal elastase levels and report avoiding high fat foods because of effects on bowel motions.

Aim: To assess the use of PERT in PS patients and determine any benefit on bowel motions, weight, BMI and food tolerance.

Method: A retrospective review of case-notes of PS CF patients with normal faecal elastase and abnormal bowel habits trialled with low dose PERT. Anthropometric data and bowel habits were noted before and at least 3 months after initiation of low dose PERT.

Results: One hundred and forty five case-notes were reviewed over the period [2004] [2005] [2006] [2007] [2008] [2009] . Eight PS patients were found to have started low dose PERT with Creon®10,000 of which 7 patients continued to take PERT in 2009. All 8 patients had milder clinical manifestations of CF; some were diagnosed as adults. Three out of 8 patients had a body mass index (BMI) above 25 kg/m 2 (normal range 20-25 kg/m 2 ).

Mean BMI for all patients was 26.8 kg/m 2 . Five out of 8 patients had faecal elastase levels 〉500 mcg/g. three had levels between 200 and 500 mcg/g (range 286 to 〉500 mcg/g).

Seven out of 8 patients had pale, loose, frequent bowel motions (up to 6x/day) with difficult to flush stools, abdominal pains, faecal urgency and all had manipulated their dietary intake & avoided high fat foods.

One patient with repeated episodes of pancreatitis had problems with constipation and elastase level fell year-on-year (latest level 286 mcg/g).

Mean quantity of Creon®10,000 taken per day was 12.5 capsules (range 8-35 capsules) with a mean of 2,601 units of lipase per kg/day (range 99-11,824 units).

In patients whose BMI 〉25 before PERT was started mean change in weight was -3.9 kgs(range 0 kgs to -6.1 kg) and in BMI was -2.15 kg/m 2 (range 0 to -2.2 kg/m 2 ).

In patients with normal or low BMI before PERT was started mean change in weight was +2.5 kgs (range -3.3 to 15.5 kg) and in BMI was +3.5kg/m 2 (range -1.3 to + 4.3 kg/m 2 ).

Seven out of 8 patients reported a reduction in bowel motions to approximately 2x/day and improvement in stool consistency to formed, flushable stools.

One patient with repeated constipation reported an improvement with regular bowel motions on PERT.

Wouthuyzen, M. 1 ; Bijvelds, M. 2 ; de Jonge, H. 2 ; Verkade, H. UMC Groningen, Groningen, Netherlands; 2. Department of Biochemistry, Erasmus University Medical Center Rotterdam, Rotterdam, Netherlands In cystic fibrosis (CF) patients, pancreatic enzyme replacement therapy does not completely correct fat malabsorption. Non-pancreatic mechanisms are suggested to play a role in the remaining fat malabsorption, like small intestinal bacterial overgrowth (SIBO). We aimed to determine the contribution of SIBO to fat malabsorption under CF conditions. Homozygous ∆F508 mice and wild type littermates received antibiotic treatment (ciprofloxacin and metronidazol) or placebo for three weeks. Before and after treatment, fat malabsorption was quantified by a 72 hours fat balance test. The competence of fat digestion (triglyceride hydrolysis) and of fatty acid uptake were assessed separately by determining plasma appearance of stable isotopelabelled fats, originating from intragastrically administered tri-1-13C-tripalmitin and 1-13C-stearate. After termination, bacterial load in the small intestine was quantified via qPCR, and faecal bile salts by gas chromatography. Bacterial load did not differ between ∆F508 and wild type mice and was not reduced in either genotype after antibiotic treatment. In ∆F508 mice, antibiotic treatment improved the absorption of saturated fatty acids. Plasma concentrations of 13C-fats from tri-1-13C-palmitin and 1-13Cstearate were non-significantly higher in the antibiotic treated ∆F508 mice. Antibiotic treatment reduced faecal excretion of primary and secondary bile salts in ∆F508 mice compared to placebo treatment (primary, 11±5 vs. 19±3 µmol/100 gram BW/day, p<0.05; secondary, 0.4±0.2 vs. 7.0±0.9 µmol/100 gram BW/day, p<0.01, resp.). We conclude that antibiotics improve fat absorption in homozygous ∆F508 mice, but the mechanism does not seem to involve treatment of SIBO. Introduction: A large body of evidence links cystic fibrosis with abnormal content and metabolism of certain lipids in plasma and tissues from patients and animal models. The cftr exon 10-knockout mouse shows intestinal obstruction, malabsorption and inflammation, and a fatty acid imbalance in intestinal mucosa. The aim of this study was to perform a lipid cartography of mouse colonic mucosa and to search for any lipid alterations in the cftr knockout model with regard to intestinal morphology.

Methods: Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS) imaging was performed on colonic mucosa from wild-type (WT) and cftr knockout (CF) mice. Ten µm-thick sections were obtained from 3 WT and 3 CF mice, and analyzed by a TOF-SIMS IV mass spectrometer (Ion-Tof GmbH, Muenster, Germany), fitted with a bismuth (Bi3+) source. Secondary ions were collected in the negative mode at a 2µm resolution. Data were further processed by multivariate statistical analyses, including principal component analysis (PCA), partitioning clustering, hierarchical clustering, and genetic algorithm analysis (GAA).

Results: TOF-SIMS data showed a particular localization for cholesterol, cholesteryl sulfate and vitamin E at the epithelial border, C16 fatty acids in Lieberkühn glands, and C18:0 in lamina propria and submucosa. When comparing the spectra of these three regions from WT vs. CF tissues, slight but statistically significant increases in vitamin E and C16:0 in the epithelial border of CF colon were found. PCA and partitioning clustering on this selection resulted in a thorough correlation of clusters of lipids with the morphological structures of colonic mucosa. Lieberkühn glands were characterized by increased levels of C16:0, C14:0, C16:1 and phosphatidylethanolamine; lamina propria and submucosa were correlated with higher C16:0, C18:2 and phosphocholine; the internal epithelial border with C16:0, C18:0, C18:1, C18:2 and cholesteryl sulfate; and the external epithelial border with C14:0, C16:1, C18:1, phosphatidylinositol and phosphatidylethanolamine. In addition, hierarchical clustering analysis by principal components on spectra from the Lieberkühn gland region -where CFTR is mainly expressed-allowed a good separation of CF and WT individuals, mostly based on variations in C18:2, cholesteryl sulfate, C16:1, phosphatidylinositol and sulfatide ions. This result shows for the first time the spatial distribution of lipids in colonic mucosa and TOF-SIMS imaging plus multivariate analyses as a powerful tool in the investigation of tissue spatial lipid signatures in cystic fibrosis.

Supported by ANR-07-MRAR-006-01, Associations Vaincre la Mucoviscidose. CF patients typically experience reduced growth; both body weight and height are retarded, and consequently result in lower BMI. Malnutrition, as a consequence of exocrine pancreatic dysfunction, and increased energy expenditure are typically implicated to explain growth reduction in CF. There is a strong, positive correlation with morphometric indices and lung function (FEV1) as well as CF patient survival. Konstan et al found that changes in body weight predicted future lung function in CF children, suggesting a causal link between body weight and lung health. Thus, current nutritional recommendations of higher energy intakes (high-fat diets) coupled with pancreatic enzymes are commonly implemented. While this approach makes sense superficially, our preliminary data indicate that partitioning and utilization of dietary fuels is quantitatively and qualitatively different in CF mice. We previously presented qualitative evidence of lack of fatty acid malabsorption (NACFC 2009 ) in CF mice. The objective of this work was to study processing and metabolism of dietary lipids in CF mouse models to elucidate the effectiveness of increased dietary fat on body weight and body composition. To study lipid processing and absorption, we fed mice with diets containing increasing lipid content, i.e. regular chow (6% fat kcals, P3000) and "breeder's" diet (10% fat kcals, 7960). We studied the effects of diets in wild-type, S489X null, and ∆F508 mice. Lipids were extracted from feces and analyzed using gas chromatography-mass spectrometry. Analyses of fecal lipid composition revealed that essential fatty acids (18:1, oleic and 18:2 linoleic) were completely absorbed with either diet (Table 1) . We did not find any indication of malabsorption of saturated fatty acids on P3000 diet. The 7960 diet with higher fat content caused increased excretion of saturated fatty acids. To reduce obstruction of the gastrointestinal tract, Colyte (PEG-3350+electrolytes) was added to drinking water for ∆F508 mice. This may partially explain the increase in saturated lipids in the feces of ∆F508 mice. In addition, we find that ∆F508 mice on Colyte and P3000 diet had higher excretion of fecal cholesterol (Chol); this was true for S489X mice on 7960 diet, but not for ∆F508 mice on 7960 diet. We demonstrate that CF mice absorb fatty acids differently than non-CF mice and that this is a selective process. While the absorption of essential fatty acids appears similar between CF and non-CF mice, regardless of diet, the absorption of saturated fatty acids is handled much differently between CF and non-CF and is further influenced by the dietary context as well. Our current work focuses on the analyses of a larger spectrum of fatty acids and other lipid components and their processing and metabolism.

These studies are supported by National Institutes of Health and Cystic Fibrosis Foundation. Background: Pancreatic insufficiency is quantitated using coefficient of fat absorption (CFA). A higher CFA is thought to be associated with better growth though CFA has never been correlated with nutritional outcomes. We performed a year-long, open-label study of the clinical activity and safety of liprotamase, a non-porcine pancreatic enzyme replacement therapy (PERT) containing highly purified crystalline protease, amylase, and crystalline, cross-linked lipase in subjects with CF ("safety study"), some of whom participated in a Phase III study of a minimally effective dose of liprotamase ("efficacy study").

Methods: In the efficacy study, subjects had a baseline CFA (bCFA) off PERT and then took liprotamase in an open-label phase prior to double blind randomization to placebo or continued liprotamase with a previously identified minimally effective dose of liprotamase (1 capsule with each of 3 meals and 2 snacks per day). Dose adjustment was not allowed. Subjects with a bCFA ≥ to 80% were not randomized. A tCFA was done 22 -31 days after bCFA. In the safety study, subjects took an initial dose of 1 capsule of liprotamase (32,500 units lipase activity, 25,000 units protease activity and 3,750 units amylase activity) in the middle of 3 meals and 2 snacks each day. Subjects could adjust their dose based on clinical parameters. Weight and height were measured at each study visit and body mass index (BMI) was calculated. These were transformed to z scores to enable comparison of subjects of all ages.

Results: A total of 214 subjects (58% male) were treated in the safety study; 88 subjects in the efficacy study rolled over into this protocol; placebo =44, liprotamase =36 (in the randomized period), and not randomized =8. Sixty-seven subjects completed 12 months of treatment. The dose adjustment took place predominantly over the first 3 months during which the average dose was ~ 5 capsules/day. Average dose for the period after 3 months was ~ 6.5 capsules/day. ANOVA for mean change in weight and BMI z scores after dose adjustment from 3 months until the 12 month visit stratified by bCFA or tCFA was not significantly different among groups.

Conclusion: After dose adjustment at an average liprotamase dose of 6 .5 capsules/day, weight and BMI z-scores were stable over a nine month period. Based upon this data there was no apparent association between nutritional status and baseline CFA or on treatment CFA. Background: Liver disease is a known complication in people with CF; hepatic steatosis occurs in 20-73%. Choline deficiency is associated with steatosis and supplementation with either choline or its oxidative metabolite betaine has been demonstrated to improve, or reverse the hepatic steatosis. Differences between subjects with CF and PI and healthy controls in liver fat (LF) content and calf muscle fat (MF) and choline (MC) content assessed using magnetic resonance spectroscopy (MRS) has not previously been reported.

Objectives: To compare liver fat and muscle fat and choline content using baseline MRS in subjects with CF and PI enrolled in a prospective nutrition intervention study, to healthy controls matched for age, sex, body size and composition.

Methods: Subjects with CF and healthy controls (10 to 17 yrs) underwent MRS scan (1.5T Siemens Avanto Whole Body Scanner) after a standard 11 hour fast. Muscle/liver Proton MRS were acquired with a Siemens Extremity coil/body array. Localized water-unsuppressed spectra were acquired from various volumes of interest (VOI, 18cc) selected in the liver using STEAM pulse sequence. Chemical Shift Imaging with and without water suppression was acquired from various regions in the calf (soleus) muscle. Percent LF, calf MF and MC were calculated. For all subjects, height (HAZ), weight (WAZ), BMI (BMIZ), upper arm muscle (UAMAZ) and fat (UAFAZ) Z scores were calculated. Whole body fat % (%BF) were estimated using the four skinfold method. For subjects with CF, FEV1, serum cholesterol (CHOL), liver enzymes, HS-CRP, genotype, current medications, dietary fat and choline intake were also assessed. Student's t-tests or Wilcoxon rank sum tests were used for group comparisons, and multiple regression for predictors of LF%, MF% and MC%.

Results: Compared to healthy controls (n = 10, 7 male, 14.0 ± 3.0 yrs.), subjects with CF and PI (n = 25, 17 male, 13.4 ± 2.2 yrs), had higher LF% (7.5 ± 8.3 vs. 0.6 ± 0.4, p=0.014), lower MC% (0.09 ± 0.03 vs. 0.13 ± 0.03, p=0.005), but similar MF% (4.1 ± 1.4 vs. 3.6 ± 1.1). Only group (CF vs. controls) predicted LF% (F=6.8, R 2 =0.17, p=0.013), UAMAZ predicted MF% (F=2.9, R 2 =0.16, p=0.028) and group, age and MF% predicted MC% (F=10.5, R 2 =0.51, p<.001). In subjects with CF, there were no significant Our research has established the potential of fenretinide, a semi-synthetic retinoid, as a novel therapy for cystic fibrosis. We have previously demonstrated that fenretinide has the ability to improve clearance of Pseudomonas pulmonary infections in CF mice, attenuate the expression of inflammatory genes (IL-1β and S100A8), reduce the levels of arachidonic acid (AA) and increase the levels of docosahexaenoic acid (DHA) and ceramide (Guilbault 2008 (Guilbault , 2009 .

In April 2010, fenretinide received Orphan Drug status from the FDA for treatment of CF patients infected with Pseudomonas.

In preparation for clinical trials, we have conducted pharmacokinetics of fenretinide using a range of doses in mice heterozygous for the CFTR mutation. Additionally, we monitored the changes in lipids in plasma and lung samples. Drug concentrations were analysed by HPLC and mass spectrometry (MS). Lipids were extracted using a chloroform/methanol method. AA and DHA were assessed using gas GC/MS. Phospholipids were isolated by TLC and ceramide was quantified by ELISA.

The peak concentration of fenretinide in plasma following treatment with a single 5 mg/kg dose of fenretinide was observed at 4 hrs with an average of 1.11 µM ± 0.27. Highest levels of DHA occurred at 24 hrs while AA continued to drop even at 90 hrs post-treatment. Ceramide levels peaked at 48 hrs. In the lungs, DHA levels were highest and AA levels lowest after 72 hrs. Ceramide levels peaked at 48 hrs.

We also conducted similar pharmacokinetic studies in non-human primates (Macaca fascicularis). At 4 hrs, the concentration of fenretinide in plasma was 1.44 µM ± 0.25. There was a statistically significant reduction in the AA/DHA ratio 6 hrs post-treatment (p<0.01).

We hypothesized that treatment with fenretinide will have a direct impact on blood cells. We isolated leukocytes from 10 patients with CF and 10 healthy controls. Following treatment with either 1 µM or 2.5 µM, AA levels were reduced (p=0.002 and p<0.0001, respectively), DHA levels were increased (p=0.0444 and p<0.0001, respectively) and the overall ceramide levels were increased (p=0.0012 and p<0.001, respectively).

In summary, these results demonstrate that treatment with fenretinide results in a systemic improvement of the AA/DHA ratio and the ceramide imbalance which has been associated with favourable control of inflammatory genes in mice. Treatment of non-human primates with the same low dose of the fenretinide resulted in similar effects on the plasma lipid profiles with comparable plasma bioavailability of the drug without needing a dose adjustment between species. Interestingly, we have shown that very low concentrations of fenretinide have direct effects on lipid levels in leukocytes of CF patients. 1 1. Northwestern University, Chicago, IL, USA; 2. Children's Memorial Hospital, Chicago, IL, USA Pancreatic enzyme replacement and proton pump inhibitors are valuable tools in optimizing growth and nutrition in pediatric cystic fibrosis patients. However, less is understood of their impact in CF adults. It had been our practice to titrate pancreatic enzyme dosing and proton pump inhibitor initiation based on patient reported malabsorption symptoms. As part of a quality improvement initiative our clinic developed an algorithm to optimize nutritional status. This algorithm included more frequent clinic visits for CF patients aged 18 and older determined to be "at risk" with BMIs of <21 for females and <22 for males, increasing pancreatic enzyme dosing to achieve optimal levels based on weight and initiation of PPI if patients had not achieved weight goals. Here we report the effects of implementing this algorithm using a retrospective review of all "at risk" adult patients. A total of 28 adults were identified as meeting criteria as of July 2008. After one year of implementation, enzyme dosing was statistically increased from a mean of 1731±641 units lipase/kg/meal to 1920±512, p<0.005. Similarly, the use of proton pump inhibitors increased from 32% to 62%. Subjectively these adult patients reported appreciation for the additional educational steps utilized with this aggressive intervention. At the end of one year, 3 of 28 (11%) had moved out of the "at risk" category and a total of 18 of 28 (64%) patients had improved their BMI. For the overall population, the BMI increased from 19.54 to 19.83 which was not statistically significant on paired analysis.

These results suggest that aggressive measures to improve utilization of enzyme replacement and proton pump inhibitor use can be achieved in the adult population with a significant impact on BMI in a minority of patients. Sustained interventions may be needed to impact a larger population and longer-term evaluation is ongoing.

increased 16S rRNA in mesenteric lymph nodes. A likely mechanism for these findings is the increased expression of claudin 10, a "leaky" claudin, in intestinal and colonic epithelium in CFTR-/-mice; claudin 10 is expressed at very low levels in the normal gut epithelium and is known to alter the functional integrity of tight junctions. Finally, we detected a marked increase in the number of mast cells in the intestinal as well as colonic lamina propria in CFTR-/-mice, and in the human CF lung, indicating that bacterial translocation in CF was biologically-significant.

Conclusions: We show that loss of CFTR in the gastrointestinal tract is associated with primary abnormalities in gut barrier function, and identify claudin10 as a plausible mechanism. A generalized epithelial defect attributable to absent CFTR could have significant implications for a number of tissues affected by CF, including intestine, pancreas, and lung. An increased ability for bacteria (within the intestine, or lung) and digestive enzymes (in the pancreas) to traverse the CF epithelial barrier and elicit submucosal injury, inflammation, and tissue remodeling could represent an important, and previously unappreciated, aspect of CF pathophysiology. Background: Previous correlative studies on lung function and serum levels of vitamins A and E have been conducted in children and adults with cystic fibrosis (CF). Adult CF patients have higher micronutrient requirements and have variability in lung function compared to children with CF. Due to such variances, adult CF patients require more in depth analysis. The purpose of this study was to investigate the relationship between serum vitamin A and E levels in adult CF patients and markers of lung function, particularly FEV1 percent predicted values and the number of infective exacerbations.

Methods: A cross-sectional retrospective study was conducted on patients attending our adult CF center. All 180 adult patients were screened for inclusion. Patients were included if they met the following criteria: exocrine pancreatic insufficiency, did not receive a lung transplant, normal liver function tests and albumin levels, and good adherence to pancreatic enzymes and fat-soluble vitamin supplementation. Patients who did not meet laboratory and pulmonary function criteria, supplementation defined as less than 5 days per week, or did not have complete laboratory and pulmonary function test data available were excluded. Clinical data were collected from the medical records without patient identifiers. Data analysis was completed using SPSS to run the Pearson product-moment correlation coefficient and multiple regression analysis. The significance level was set at a p-value of <0.05.

Results: Of 180 patients screened, 62 met inclusion criteria. Thirty-two males and 30 females ranged in age from 18-57 years. The mean BMI of the sample was 22.1 kg/m 2 and the mean FEV1 was 63.5% predicted. Serum vitamin A levels significantly correlated to FEV1 percent predicted values directly (r = .341, p = .007). The regression analysis showed that serum vitamin A levels had a significant impact on FEV1 percent predicted values (β = .004, R2 = .116, p = .007). However, there was no significant correlation between serum vitamin A levels and number of acute infective exacerbations (r = -.161, p=.211). Serum vitamin E levels and lung function or rates of acute infective exacerbations did not significantly correlate.

Conclusion: There is a direct, positive relationship between serum vitamin A levels and FEV1 percent predicted values, but not rates of acute infective exacerbations. Vitamin E levels did not correlate to either lung function values or exacerbation rates. The results of this study are in agreement with the majority of previously conducted studies in adults and children with CF. Larger sample size investigations are needed to determine therapeutic target vitamin A levels in adults with CF that correlate to pulmonary function values and potential rates of acute infective exacerbation. Furthermore, it yet remains unclear the dosage and type of vitamin A supplementation to achieve this targeted goal. that CFTR-/-pigs develop significant pancreatic disease that progresses postnatally. The presence of significant pancreatic disease in newborn CFTR-/-pigs indicates that organ involvement begins in utero.

Hypothesis: Pancreatic lesions begin in utero in a porcine model of CF. Assessing the histopathological progression of pancreatic lesions may provide insights to disease pathogenesis.

Methods: We studied the pancreatic histology through examination of serial stages of fetal development of CF (CFTR-/-), WT (CFTR+/+) and heterozygous (CFTR+/-) pigs at 54, 60, and 69 days gestational age (gestational length ~114 days).

Results: There were no differences between WT and heterozygous animals. In non-CF animals, pancreata were detectable but small at gross examination on d54. Histopathological examination of non-CF animals showed poorly formed pancreatic lobules composed of some ducts and immature acini lacking zymogen material. Acini and ducts within lobules were individually separated by a thin myxomatous matrix. By d60, the acini were more compact, reducing interacinar spaces. Approximately half of the acinar cells had zymogen granules in the apical cytoplasm. By d69, almost all acini had zymogen granules in the cytoplasm. In some pancreata (d54-69), scattered acini and ducts were partially dilated by clear space and/or wispy eosinophilic material. Mitotic figures and apoptotic bodies were variably detected along with rare solitary leukocytes. In contrast, CF pancreata were morphologically distinguishable from WT at all time points, but there was variation in the extent of changes for each CF pig. Changes were initially distinguished from WT at d54: there were solitary to patchy acini and ducts with mild to moderate lumen dilation by dense, colloid-like eosinophilic material with rare cellular debris and rare solitary neutrophils. By d60, the distribution of affected acini was slightly expanded with affected lumens showing similar filling. Infrequent acinar lumens had moderate to severe filling by eosinophilic material and cellular debris with rare dystrophic calcification and uncommon neutrophils. By d69, there was mild progression, with more distinct dilation of affected acini; interestingly cellular debris and scattered neutrophils were detected only in some acini and ducts.

Conclusion: As in humans, pancreatic disease starts in utero in CF pigs. Plugging of acini and ducts with eosinophilic material is one of the earliest lesions detected in the CF pig pancreas, suggesting that it plays an important role in disease pathogenesis. Background: Existing information regarding the development of gut mucosal inflammation in CF emphasizes a primary role for bacterial overgrowth in the gut lumen. However, bacterial overgrowth has not been previously implicated as an isolated factor to cause intestinal inflammation in conditions other than CF. The normal gut epithelium provides an effective barrier against bacterial translocation and is also profoundly downregulated for inflammatory responsiveness to bacterial products. In this study, we investigated whether CF is associated with primary abnormalities in gut barrier function, which would permit the translocation of luminal bacteria across the epithelium and cause local and systemic immune activation.

Methods: Gut mucosal permeability in WT (CFTR+/+) and CFTR-/-C57B6 mice was compared by enteral administration of FITC dextran followed by the measurement of FITC fluorescence signal in plasma. Bacterial translocation across the gut mucosa was measured by PCR detection of bacterial 16S ribosomal gene RNA in mesenteric lymph nodes and by specific staining of tissue sections for bacteria (Brown and Brenn method). Expression of tight junction genes was measured by real-time PCR, immunohistochemistry, and Western blots. To investigate whether bacterial translocation we detected in the CF intestine was biologically-relevant, we enumerated c-kit+ mast cell populations in the intestinal and colonic lamina propria by immunohistochemistry. To confirm these findings, we also stained a few archived sections of human CF lung.

Results: CFTR-/-C57B6 mice had significantly higher gut mucosal permeability than WT mice (147±68 vs. 58±4 fluorescence units in plasma 2 hrs after enteral administration of FITC dextran; p<0.05). We also detect- Method: Descriptive retrospective study of children with cystic fibrosis diagnosed from November 2005 to November 2007. The data collection was gotten from a research done from hospital medical records, involving clinical data from the first hospital appointment and anthropometric measures gotten after 6 and 12 months. This study was approved by the Ethics Committee in Research from HIJG (068/2008).

Results: Thirty patients were analyzed (56,7% female). The median age at diagnosis was 2.6 months old. The average age was significantly lower in patients diagnosed by the NS (2.2 mol (SD 1.4mo)) in relation to the ones diagnosed later (85.6 mo (SD 52.9))(p<0,05). The NS was regarded as the initial screening test in 70% of the cases. The research of the mutation DF508 was done in 77% of the patients (homozygosity in 7% of the cases and heterozygosity in 37%). Ten percent of the patients presented meconial ileum at birth. At diagnosis, 80% had pancreatic insufficiency. The average serum albumin was 3.5 g/dl, 43.5% of the patients had lower concentrations. From the total of oropharynx swab culture, 26% were positive for S. aureus, 5% for P. aeruginosa and 1% for B. cepacia. The patients identified by the NS presented better nutritional outcomes, with Z-score -2 for BMI and -2.2 for weight-for-age at admission, and respectively -0.09 and -1.4 after one year (p<0.05; OMS). There was a significant difference regarding the nutritional classification in the curves from WHO/2006 and CDC/2000, being respectively, the Z-score for weight-to-age at admission, after six months and one year of -2, -1.4 e -0.8 by WHO, while by CDC it was -1.8, -1.7, -1.5 (p<0.05). The Z-scores for height-to-age by WHO were -1.7, -1.3, -1.1; and by CDC, -1.4, -1.1, -0.9 (p<0.05).

Conclusion: Most patients diagnosed by the neonatal screen present pancreatic insufficiency, hypoalbuminemia and positive culture for Staphylococcus aureus at diagnosis. The ones who were diagnosed by the NS had a better nutritional outcome than the ones diagnosed by the sweat test. One year after diagnosis, the patients presented significant improvement in their nutritional state. There is a significant difference between WHO/2006 and CDC/2006 nutritional classification. In the first six months of life, regarding weight, WHO would diagnose more malnourished patients, inverting this situation after this age. Regarding height, in all ages, WHO would diagnose as more with growth failure. Objective: Specific fatty acid abnormalities have been identified in the plasma and tissues of cystic fibrosis (CF) animal models, human patients, and in cultured cell models. These abnormalities include increased concentrations of palmitoleate (16:1n-7; POA), oleate (18:1n-9; OA), and eicosatrienoate (20:3n-9; ETA). These fatty acids are of particular interest in CF pathophysiology because of their potential roles as lipokines and as mediators of the inflammatory process. The aim of this study was to evalu-ate the hypothesis that increased expression and activity of fatty acid metabolizing enzymes is responsible for these fatty acid alterations in CF. The conversion of radiolabeled palmitate (16:0; PA), stearate (18:0; SA), and OA to downstream fatty acid products was measured in human bronchial epithelial (HBE) cells expressing a normal CFTR protein and compared to conversion in cells expressing a CF phenotype. These results were correlated with the expression of desaturase and elongase enzymes.

Study Design: HBE cells with either a wild type phenotype that expresses normal CFTR protein (sense cells), or a CF phenotype that does not express CFTR protein (antisense cells) were grown to confluence. The cells were incubated with media supplemented with radiolabelled PA, SA, or OA for 4 hours or 24 hours and then harvested. Metabolism of the radiolabeled substrates to downstream products was measured by HPLC coupled to a scintillation detector. Metabolic enzyme gene expression was measured by quantitative RT-PCR.

Results: At 4 hours, there was significantly increased conversion of PA to downstream products POA, SA, and OA in antisense cells measured by both DPM/10 6 cells and by percent of total counts. At 24 hours, the per cell effect had largely normalized, though conversion to POA and OA was still significantly increased in antisense cells. There was a significant increase in SA uptake in antisense versus sense cells (total DPM/10 6 cells = 810±140 vs. 320±20 at 4 hours; 6860±870 vs. 1880±110 at 24 hours; P=0.05 and 0.01, respectively). There was also increased enzymatic conversion of SA to OA and ETA demonstrated by increased fractions of these metabolites in antisense cells at 4 and 24 hours. When cells were incubated with OA, there was a marked increase in uptake in antisense versus sense cells at 4 hours (total DPM/10 6 cells = 4150±120 vs. 650±80; P<0.001). Only a small fraction of OA is converted to ETA, however, this fraction is increased by 20% and 44% in antisense cells at 4 and 24 hours, respectively. Taken together, these data indicate that antisense cells exhibit increased enzymatic activity catalyzing the conversion of PA to POA and from PA through SA to OA and ETA. We measured a significant increase in the relative mRNA levels corresponding to both ∆9-desaturase and elongase 6 metabolic enzymes.

Conclusions: CFTR-negative cells show increased metabolism of fatty acids to POA, OA and ETA. These changes correlate with increased expression of desaturase and elongase enzymes, suggesting that changes in mRNA levels are responsible for the increased activity observed in antisense cells. Background: Diabetes is common in cystic fibrosis (CF) and is associated with increased mortality and prediabetic clinical decline prior to diagnosis, which is potentially reversible with insulin.

The current UK gold standard for diagnosis includes oral glucose tolerance testing (OGTT).

Little is documented about day to day glucose control in adults with CF who have a normal OGTT result.

Continual glucose monitoring (CGM), a novel method of assessing glycaemia, may have a meaningful role reflecting "real life" glycaemic control.

Objective: We evaluated whether the OGTT reflects "real life" glycaemic control and whether subjects with normal OGTT results had abnormal continual glucose monitoring.

Methods: Forty-eight stable adults with CF and 9 healthy non-CF control subjects had concurrent CGM (Medtronic, USA) and OGTT (WHO standard criteria). CF patients were "stable" i.e. no symptoms of infective exacerbation, no management change including additional IV/oral antibiotics, for 2 weeks prior.

Participants completed a 72 hour CGM study (interstitial glucose sensor reads glucose every 5 mins, returning 288 glucose values per 24 hours) collected 4 capillary blood glucoses for calibration each day, and recorded food With the increasing life expectancy of individuals with cystic fibrosis (CF), we are detecting new contributors to CF bone disease. This multifactoral problem continues to grow in incidence. Current efforts are exploring systemic origins (from endocrine abnormalities and the hyper-inflammatory state), as well as cellular contributions in the bone-forming (osteoblast) and bone-resorbing (osteoclast) cells. We previously demonstrated reduced bone density in the Cftr total knockout (KO) mouse model, however the development of conditional KO mice (via loxP-flanked Cftr) allows a more specific identification of factors associated with bone homeostasis. Previous reports reveal the neuronal specific Cftr KO mouse (Nestin-Cre) has reduced growth, and more recent data implies an endocrinologic origin. To test this particular aspect of bone health, we compared bone parameters of adult neuronal specific Cftr KO mice against control littermates.

We collected femurs and calvaria from 14-week-old neuronal specific Cftr KO mice (n=6 males and 5 females) and control littermates (n=6 males and 6 females). Micro computed tomography (micro CT) was used to measure samples for quantitative bone parameters and sex-specific comparisons were performed between neuronal Cftr KO and controls.

Bone size was significantly reduced in both male and female neuronal Cftr KO mice compared to controls (femur length p<0.0001, total bone volume p<0.005, cortical bone area p<0.05, and cortical thickness p<0.05). Total femoral bone density was reduced in neuronal Cftr KO males (p<0.05, females p=0.08). For cortical bone, female neuronal Cftr KO mice had reduced sub-periosteal area and density, with increased porosity (p<0.05 for all), while male neuronal Cftr KO mice had reduced sub-periosteal area and increased porosity (p<0.0001 for both). At the trabecular bone, male neuronal Cftr KO mice had reduced trabecular number and increased trabecular spacing (p<0.05 for both). No difference was detected in calvaria bone density between neuronal Cftr KO and controls.

Neuronal specific Cftr KO mice have reduced femur length consistent with previous growth findings in this disease model. Additionally, bone mineral density is moderately reduced, implying a neuroendocrine contribution to bone formation in CF. These influences are in the absence of not only lung disease, but also other hyperinflammatory confounders. Use of conditional Cftr KO models will assist further studies of cellular components and interactions, as well as other systemic clinical aspects, necessary to evaluate underlying mechanisms CF bone disease. The incidence of cystic fibrosis (CF) bone disease is growing as the life expectancy continues to increase. Cftr-null mice (Cftr S489X-/-mice, further identified as Cftr-/-) have reduced bone density, despite the lack of lung or pancreatic disease, suggesting an inherent defect in bone metabolism. Previous reports reveal CFTR is expressed in both osteoblasts and osteoclasts. Additionally, a defect in bone homeostasis, in CF individuals and mouse models, implies reduced bone formation with increased bone resorption. To pursue a mechanism for reduced bone mineral density, we tested effects of CFTR, utilizing ex vivo osteoblast differentiation of bone marrow stromal cells (BMSCs) from Cftr-/-, heterozygote (Cftr+/-) and wildtype (Cftr+/+) mice.

We collected BMSCs from femurs of Cftr-/-, Cftr+/-and Cftr+/+ mice at 8-9 wks of age. Adherent osteoblast precursor cells were selected and osteoblast differentiation was induced in culture by addition of osteogenic medium. Cellular growth was documented at day 4, 6, 8, 10, 12 , and 14 by imaging. The staining for alkaline phosphatase activity was performed at 7 and 14 days to examine the extent of osteoblast differentiation. Alkaline phosphatase staining was quantified by histochemical analysis in the 7-day cell cultures (at 3.2X), utilizing the ImageJ software (v1.43) with color counter. Additionally RNA was collected from the cultures at 7, 14, 21 and 28 days, for gene expression analyses.

Cftr-/-and Cftr+/-BMSCs displayed reduced osteoblast proliferation at 6, 8 and 10 days, compared to Cftr+/+ cells. Alkaline phosphatase staining was also reduced in Cftr-/-and Cftr+/-cultures compared to Cftr+/+ on both the time points, although it did not achieve quantifiable statistical significance. (ImageJ analysis demonstrated as mean pixel count ± SEM; Cftr-/-12,834 ± 3562, Cftr+/-12,706 ± 5844, and Cftr+/+ 22,400 ± 8088.)Osteoblasts with reduced CFTR have a slower growth rate and when they reach confluency do not express optimal levels of alkaline phosphatase, a bone phenotype protein required for mineralization of the calcified matrix. Gene expression profiling is in progress to expand understanding of altered components of osteoblast activity for normal bone formation, by analysis of bone specific marker genes.

Homozygote and heterozygote CFTR-deficient osteoblasts exhibit reduced growth rate and osteogenic differentiation when compared to wildtype cells. Further mechanistic information is to be determined through gene expression analysis. However, these preliminary data suggest a contribution of the CFTR to control osteoblast proliferation and activity, and potentially a direct association with metabolic bone disease in CF.

Supported by the Cystic Fibrosis Foundation.

study was to review all male patients attending a Regional Centre who had become fathers and assess the impact of paternity on disease course and management. Methods: Case notes of all male patients who had become fathers between 1990 and 2009 were reviewed retrospectively. Data on modality of conception and health status (lung function, weight, number of clinic visits and days of intravenous antibiotics) were recorded 4 years pre and 4 years post the child's date of birth. For each child data on DOB, gender, outcome, CF status and singleton/multiple birth were obtained.

Results: Twenty men (median age 31.7 years, FEV1 66%, BMI 24.3) became fathers to 32 children (20 boys, six sets of twins)during the study period. Two births occurred in period 1990-1994, 4 in 1995-1999, 8 in 2000-2004 and 18 between 2005 -2009 . Successful fertilisation was achieved by ICSI (n=20, 63%), AID (n= 5, 15%) and natural conception (n=4, 13%; genotype DF508/3849+10kB C>T). One child was adopted and 2 children were conceived outside of the relationship. All fathers are still alive. One child died at 0.3 yrs (congenital heart defect). No child has CF; median (range) age of 31 children 3.69 (0.3-16) yrs. Twelve of 20 men had MESA followed by ICSI; 9/12 successful at 1st attempt, 2/12 at 2nd and 1/12 at 4th attempt. No significant change in weight, number of clinic visits or days of intravenous antibiotics/yr pre vs. post. Significantly increased rate of decline in FEV1 post vs. pre birth (180 ml/yr vs. 60 ml/yr, p=0.015) but not with FVC (108 ml/yr vs. 43 ml/yr, p=0.128).

Conclusion: Biological fatherhood, AID and adoption are realistic options for men with CF. The additional responsibility of caring for a child/children may have a negative impact on the father's health. The significant fall in FEV1 with new parental status suggests that these men require heightened observation and more intensive treatment. Introduction: Cystic fibrosis (CF) is the most common lethal autosomal recessive disease among Caucasians. CF is due to a mutation in the Cystic Fibrosis Transmembrane Regulator (CFTR). The main mutation is a deletion at amino acid 508 (∆F508). The life expectancy of CF patients has increased and consequently new complications such as Cystic Fibrosis Related Diabetes (CFRD) have emerged. CFRD is primarily due to a defect in insulin secretion. Studies have shown that women and subjects homozygotes for the ∆F508 mutation have a higher risk of developing CFRD. However, very few studies have examined the effect of gender and genotype on insulin secretion. The main purpose of this study was to compare insulin secretion between CF men and women with different genotypes.

Subjects and Methods: One hundred ninety-six non diabetic adult subjects were recruited from the CF clinic of the Centre Hospitalier de l'Université de Montréal. Genotype status was extracted from the medical files. All subjects underwent a 2-h Oral Glucose Tolerance Test (OGTT) to determine their glucose tolerance: normal glucose tolerance (NGT), impaired glucose tolerance (IGT) or CFRD. Blood samples were taken every 30 min for 2h to measure plasma glucose and insulin. From these samples, insulin secretion and sensitivity were evaluated using Stumvoll indices.

Results: For a comparable glucose excursion there were gender and genetic differences in insulin values during the OGTT. Heterozygous patients have higher insulin secretion than those with the homozygous mutation (p=0.027 for men and p=0.046 for women). Whatever their glucose tolerance (NGT vs IGT vs de novo CFRD) and genotype, women present a higher insulin secretion than men. Calculation of the disposition index (insulin secretion adjusted for the degree of insulin sensitivity) confirmed the higher insulin secretion in women.

Conclusion: In a large group of CF patients without previous diagnosis of diabetes we observe an association between higher insulin secretion and heterozygote status for ∆F508 mutation and/or in women.

Etherington, C.; Peckham, D.; Clifton, I.; Conway, S. St James's University Hospital, Regional Adult Cystic Fibrosis Centre, Leeds, United Kingdom Introduction: As the median survival for CF continues to improve, most women now reach reproductive age in a relatively healthy state and face decisions regarding pregnancy. Pre-pregnancy lung function has been suggested as the most important factor influencing maternal and fetal outcome. The aim of this study was to review the experience of pregnancy over a 20 year period in a large Regional UK Centre and assess maternal and fetal outcomes.

Methods: A retrospective review of the case notes of women who became pregnant between 1990 and 2009 was performed. Data on modality of conception, baseline demographic data, pregnancy and mode of delivery were obtained. Maternal lung function, weight, number of clinic visits and days of intravenous antibiotics were recorded for 4 years pre pregnancy, during pregnancy and 4 years post pregnancy. For each child data on DOB, gender, singleton/multiple births, CF status and outcome were obtained. Subgroup analysis was performed with regard to baseline FEV1 <60% predicted.

Results: Thirty-five women (median age 24yrs, FEV1 62% and BMI 20.9) became pregnant (48/50 natural conception, 1/50 IVF and 1/50 PGD) and delivered 50 children (25 boys, 1 set of twins). Sixty-one percent of women had one child, 26% two and 13% three children. Only 40% of pregnancies were planned and 79% of partners screened for carrier status. Thirteen of 35(37%) had pre-existing CFRD and 6 women developed gestational diabetes. Median (range) maternal weight gain during pregnancy was 6.5 (-2.9 to 15.3) kg, gestation 36 (31-40) weeks and birth weight 2.637kg (37% <2.5kg). Mode of delivery -44% SVD, 42% elective LSCS, 9% assisted and 5% emergency LSCS. Eighty-four percent of children have had a sweat test; no child has CF; median (range) age of children 7.8 (0. 1-17.5) yrs. There was no significant change in weight or lung function post vs. pre pregnancy but a significant increase in number of clinic visits (p=0.009) and requirement for iv antibiotics during pregnancy and up to 2 yrs post delivery (p=0.013). Women with CFRD had lower weight gain during pregnancy (5.3 vs. 6.8kg, p=0.024), lower birth weight babies (2.1kg vs. 2.96kg, p=0 .002) and shorter gestation (34 vs. 37 weeks, p=0.001). A pre-pregnancy FEV1 less than 60% predicted was associated with lower maternal weight during pregnancy (49.9 vs. 59.4kg, p=0.04), shorter gestation (34 vs. 36 weeks, p=0.018) lower birth weight (2.35 vs. 2.67kg, p=0 .022) and a significant decline in FEV1 post pregnancy (p=0.03) compared to women with baseline FEV1 >60% predicted. Seven women died post pregnancy -3/19 women with FEV1 >60% died vs. 4/13 with FEV1 <60%, p=0.001. Ten children have lost their mothers; median (range) age of child at death 6.3 (1.1-13.4 

Conclusions: With an aggressive MDT approach and meticulous antenatal care most women with CF can become pregnant, deliver healthy children and live to raise them with minimal impact on health status. The course of pregnancy cannot be predicted in any individual and all women should anticipate more intensive antepartum and postpartum treatment. The presence of pre-existing CFRD and a pre-pregnancy FEV1 of <60% predicted are associated with significantly worse outcomes. Introduction: Improved clinical care has led to a dramatic improvement in survival rates for patients born with CF and a child born in 2000 is now predicted to live into their 50s. Our understanding of diabetes has changed over the last 20 years. CFRD was initially thought to be mild and uncomplicated because patients were not predicted to live long enough to develop complications. It is well recognised that CFRD has a negative impact on pulmonary function, BMI and mortality, as well as causing microvascular complications. However, the observations previously recorded in younger adults may not apply to more mature adults with CF. Within the Manchester CFRD clinic we have become aware that several patients were overweight.

Method: The nutritional impact of CFRD was re-evaluated in a population based review of the patients who attended the Manchester CFRD Clinic in 2009. BMI, age, lung function, BP and genotype were retrospectively analysed.

Results: CFRD is present in 115 (33%) patients attending MACFC. Mean BMI for the diabetic population was 21.4. Two (1.7%) patients with CFRD were obese (BMI >30), 12 (10.4%) overweight (BMI 25-29.9), 65 (56.6%) normal (BMI 20-24.9) and 36 (31.3%) were underweight (<20). Patients who were obese (39.5yr±0.7) or overweight (36.3yr±8) were older than those who were underweight (29 yr±9.2)(p 0.035). Lung function was significantly lower in the group who were underweight (p=0.004). There was no difference in the proportion of patients homozygous for ∆F508 between the groups. Systolic hypertension was present in both patients who were obese and 33% patients who were overweight.

Discussion: Patients with CFRD have traditionally been regarded as having poor nutrition and reduced pulmonary function, and the risk of macrovascular disease has been considered to be low. However, 12.1% of our diabetic population is overweight or obese. This group is older and hypertension is present in a significant proportion. Mature adults with CF may be susceptible to diseases that are common in other adults such as cardiovascular disease. Four cases of IHD in patients with CFRD have now been reported in the literature. This stresses the importance of regular diabetic screening programs. Advice for diet and lifestyle may need to be personalised with previous fat, salt and calorie intake being reconsidered.

Anand, A. 1, 2 ; Vedam, H. 2 ; Conway, A. 3 ; Moriarty, C. 2 ; Culling, B. 4 ; Torzillo, P. 2 ; Yozghatlian, V. 2 ; Persson, J. 5 ; Smith, H. 6 ; Bye, P.T. 2 1. Respiratory and Sleep Medicine, St. Michael's Hospital, Toronto, ON, Canada; 2. Respiratory and Sleep Medicine, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 3. Respiratory Medicine, Concord Repatriation Hospital, Sydney, NSW, Australia; 4. Molecular and Clinical Genetics, Royal Prince Alfred Hospital, Sydney, NSW, Australia; 5. Reproductive Medicine, Sydney IVF Clinic, Sydney, NSW, Australia; 6. Reproductive Medicine, Westmead Hospital, Sydney, NSW, Australia In recent years the prognosis of patients with cystic fibrosis (CF) has improved, with the majority surviving into adulthood. As a consequence many people with CF are having families and it is estimated in Australia, nearly 1/7 people with CF have children. Given that greater than 98% of males with CF are infertile, owing to absent vas deferens, they face unique challenges in having children using assisted reproductive technologies (ART). Congenital bilateral absence of the vas deferens (CBAVD) in CF men obstructs the transport of spermatozoa from testicular or epididymal structures to the outer genital tract resulting in azoospermia. However, in general, spermatogenesis is normal. On account that their sperm is normal, sperm harvesting and ART allow these patients to have their own biological children. The aims of this study were to assess the success rate of these tech-niques, and review the respiratory status of patients utilising these techniques, and paternal survival rates over this time period.

A retrospective audit of patients from the RPAH adult CF clinic treated at specialised local fertility clinics, from 1997 to 2009 was conducted. Age, lung function (measured as FEV1), body mass index (BMI) and CFTR genetic mutation analyses were noted. Success rates, lung transplantation and paternal survival were reviewed.

Twenty-four men with a mean age of 34 years, mean FEV1 2.23 L (52% predicted) and mean BMI of 22.3 kg/m 2 were assessed. Twenty-three of the twenty-four couples (96%) were treated successfully. However, information on the number of cycles of intracytoplasmic sperm injection (ICSI) was not available. Twenty-three patients were treated with sperm retrieval and ICSI. One of the patients utilised artificial insemination by donor (AID) sperm for harvesting on two separate occasions. Four couples used ART on 2 occasions. Twin births occurred in 4 of the pregnancies and a fifth couple unfortunately lost a twin in utero at 16 weeks. Seventy-one percent of men who utilised these techniques prior to lung transplantation had an FEV1 <55 % predicted. Nine of 24 men subsequently underwent lung transplantation with 5/9 being alive to date, after a follow-up period varying from 2-13 years.

Advances in ART have been very successful and have enabled men with CF, even those with moderately severe lung disease to father children. However, 4 fathers died despite undergoing lung transplantation leaving 6 children fatherless. Increased patient awareness and education at an early age about male infertility, access to genetic counselling and when appropriate discussion about availability and success rates of ART must be provided to men with CF and their partners. This information together with discussion from their physician about the state of their disease and likely prognosis should enable couples to make timely and informed decisions about the option of parenthood. Purpose: We aimed to determine if there is a correlation between serum 25(OH)D levels and the development of CFRD in adolescents with CF.

Methods: Approval was obtained from the UAB Institutional Review Board. A retrospective chart review during the spring of 2010 at the UAB/CHS CF Care Center was conducted to gather data from the years 2008 and 2009. All patients age 14 to 20 years with outpatient, fasting oral glucose tolerance tests and serum 25(OH)D measurements within two months of each other were included. Insulin sensitivity, insulin resistance, and beta cell function were measured using the homeostatic model assessment 2 (HOMA2). Insulin sensitivity was also assessed using ISI 0,120 (2). The American Diabetes Association criteria were used to classify patients as having impaired glucose tolerance or CFRD. SPSS was used for statistical analysis to obtain Pearson correlation coefficients.

Results: Twelve males and seven females were included. Fifteen subjects were Caucasian and four were African American. Five subjects had impaired glucose tolerance and one had CFRD. The mean body mass index (BMI) z-score was .167. Serum 25(OH)D levels ranged from 14-47 ng/ml with a mean of 29.47 ng/ml. A significant positive correlation was shown between serum 25(OH)D levels and two hour blood glucose (p=.011) as well as between HOMA2-insulin resistance and BMI z-score (p=.016). A significant negative correlation was found between HOMA2-insulin sensitivity (IS) and BMI z-score (p=.014). A negative trend was identified between serum 25(OH)D levels and HOMA2-IS (p=.087) and ISI 0,120 (p=.068). A positive trend was found between HOMA2-beta-cell and BMI z-score (p=.079). Oral glucose tolerance testing (OGTT) is the gold standard screening test for CFRD. Continuous glucose monitoring (CGM) may offer advantages compared to OGTT, but the utility of this assay in CFRD screening has not been established.

Hypothesis: CGM is more sensitive than OGTT to screen adult patients for CFRD.

Methods: A prospective, single center trial of 20 patients recruited over a 2-year period. Simultaneous 3 day CGM and 2 hour 75 gram OGTT were performed. Interpretation of the CGM was made, excluding the 3 hour duration of the OGTT, while blinded to the OGTT results. One Touch glucometer and Medtronic CGM system were used to obtain CGM data. The primary endpoint is correlation between testing methods. Secondary endpoints compare clinical outcomes.

Results: Twelve patients have completed both OGTT and CGM. Complete data is available for OGTT in all patients. Two patients did not have adequate data for interpretation by CGM due to failure of the device or data retrieval. Of 10 patients with complete data (Table 1) , a significant disparity was seen between results from OGTT and CGM, as only 3/10 yielded the same result (p=0.05). In 6 of 7 subjects where results differed between the two tests, CGM indicated a greater level of glucose impairment than OGTT.

Conclusions: In preliminary results from an ongoing prospective trial, a significant difference between the methods was present, as CGM defined a greater number of patients as having CFRD or pre-diabetic conditions. Failure rate of CGM may represent a consideration in adopting this method. The validity of this approach is currently being tested through clinical response of subjects started on insulin based on this testing. Future: By the NACFC in October 2010, enrollment for this trial will be nearly complete. Additional data and patient follow-up parameters will be included for the meeting.

OGTT:Oral glucose tolerance testing CGM:Continuous glucose monitoring IGT:Impaired glucose tolerance IFG:Impaired fasting glucose Discussion: These results are not consistent with the literature. This study indicates that increased serum 25(OH)D promotes higher two hour blood glucose levels and suggests a decrease in insulin sensitivity. It also implies that as BMI z-score increases, insulin resistance increases and insulin sensitivity decreases which is inconsistent with weight loss being a symptom of CFRD. A trend was seen indicating that an increase in BMI is associated with improved beta-cell function. Limitations of this study are that the sample size was small, confounding variables were not considered, the range of serum 25(OH)D levels had little variation, and only one patient had a serum 25(OH)D level less than 15 ng/ml. A recently published study suggested that insulin resistance increases with a serum 25(OH)D less than 15 ng/ml (3) . Further research on this topic is needed.

Burgess, J.C. 1,2 ; Bilton, D. 1 ; Hodson, M.E. 2 ; Simmonds, N. 1 1. Cystic Fibrosis, Royal Brompton & Harefield NHS Foundation Trust, London, United Kingdom; 2. Imperial College, London, United Kingdom Introduction: Guidelines issued by the UK CF Trust recommend an annual OGTT for all adult cystic fibrosis (CF) patients. Yung et al. (1999) devised a selective approach to determine which patients require an OGTT (based on 1 or more of the following: random blood glucose (RBG) >11.0mmol/l; glycosylated haemoglobin (HbA 1C ) >6.1%; unexplained weight loss >5% per annum; FEV 1 decline >10% per annum and symptoms (polyuria, polydypsia and nocturia)), thus reducing the need for a universal OGTT. We performed the present study to: (1) validate the accuracy of the selective screening approach; and (2) if indicated, modify the selective screening criteria to optimise its sensitivity and specificity. Our overall aim was to develop a CFRD screening tool that could be used reliably at Annual Review -the Brompton Screening Tool (BST).

Methods: 1. The first 100 CF patients that were clinically stable (without CFRD) attending Annual Review were recruited in this prospective study (March-December 2009) . All subjects underwent an OGTT and were screened by two experienced CF doctors using the selective approach (both blinded to the OGTT results). The sensitivity and specificity of the selective approach were then calculated.

2. The parameters of the selective approach were then modified to improve its accuracy with the aim of being 100% sensitive (and at least 60% specific).

Results: Data from 94 patients was included in the analysis (6 patients had incomplete data due to laboratory error) -mean age=31±9 (SD), 38% female, 79% pancreatic insufficient. Of the whole group, 6.4% (n=6) were diabetic (OGTT at 2 hours >11.1mmol/l) and 8.5% (n=8) had impaired glucose tolerance (OGTT at 2 hours 7.8-11.1mmol/l). Forty-two patients were selected for OGTT by the selective approach (inter-reviewer agreement 100%). Reason/s for OGTT: RBG >11.0mmol/l, n=6; HbA 1C >6.1%, n=30; ÑFEV 1 , n=8; ÑWeight, n= 6; symptoms, n=0. One patient in the diabetic range (subsequent glucose monitoring was normal) and 2 in the IGT range were not selected for OGTT: sensitivity=83.3% and specificity=58.0% (positive predictive value=11.9% and negative predictive value=98.1%). With systematic adjustment of the parameters for the selective approach, we achieved 100% sensitivity and 61.4% specificity by the following: HbA 1C ≥ 6.0%, FEV 1 decline >13%, RBG >11.0mmol/l, and the removal of symptoms and weight loss from the criteria (as these were not present in isolation). Using this modified method, no patients with CFRD would have been missed and only 39 (41%) patients would have been called for an OGTT.

Conclusion: The Brompton Screening Tool (BST) is a highly accurate method to screen for CFRD, providing an excellent negative predictive value and avoiding the burden of an OGTT (a time and resource consuming test) for the majority of patients. Introduction: Previous studies reported a high prevalence of low bone mineral density (BMD) in adult patients affected by cystic fibrosis (CF). It is currently unclear if the altered bone metabolism is related to an abnormal vitamin D metabolism or rather due to the complications of advanced disease.

Aim: We evaluated the relationship between BMD, serum vitamin D levels and pulmonary function in adults with CF.

Subjects and methods: In 43 (23 males) CF adult outpatients, BMD was measured by dual energy X-ray absorptiometry (DEXA) and expressed as T score. Seventeen out of the 43 patients (39%) had DeltaF508 homozygous genotype, 38 (88%) pancreatic insufficiency (PI), 8 (19%) CF related diabetes (CFRD), and 6 (14%) regular steroid therapy, respectively. Body mass index (BMI), respiratory function tests (FEV1), serum levels and supplementation of vitamin D were recorded and compared in patients with and without osteopenia (defined as T score < -1).

Results: Patient characteristics are shown in the table 1.

Patients with osteopenia were 19/43 (44%). The mean FEV1 value was significantly lower (61% ± 24) in patients with osteopenia than in patients with normal BMD (77% ± 18) (p< 0.01). No differences were found in BMI, serum levels and supplementation per day of vitamin D between the two groups. By means of ROC curve analysis, the FEV1 cut-off point, which better identified patients with osteopenia was 58% of predicted value (0.87 sensitivity and 0.53 specificity; AUC 0.716, p<0.01).

Conclusions: Our results show that a reduction of BMD is related to poor respiratory function, but not to the vitamin D metabolism. These findings suggest that improvement in ventilatory function may have a beneficial effect on BMD in CF adults. Background: Osteopenia and osteoporosis are common in CF and begin early, resulting ultimately in debilitating fractures. Etiology is multifactorial and complex. Measurement of bone mineral density (BMD) using dual energy x-ray absorptiometry (DXA) at the lumbar spine (LS) is recommended every 2 yrs but has pitfalls, particularly in children, where 2-dimensional areal density may be misleading.

Ultrasound attenuation appears to reflect bone strength more accurately, is simple to use in the clinic and is radiation free.

Aim: To evaluate bone strength in CF using a quantitative ultrasound technique (Omnisense Bone Sonometer, Israel) measuring the speed of sound (SOS) at the calcaneus and to compare this to other parameters of bone status.

Methods: An open label, cross sectional study of clinically stable CF patients, comparing SOS to DXA-LS, and to serum bone markers: specific alkaline phosphatase (BSAP) and C-terminal propeptide of type 1 procoallagen (PICP) reflecting bone formation, and cross-linked carboxy-terminal telopeptides of type 1 collagen (CTx) reflecting bone resorption. Results were correlated with patient demographics, clinical features, 25(OH)vitamin D (D3), intact PTH, and circulating markers of inflammation. Questionnaires were completed for calcium intake and habitual physical activity. Correlations between parameters were determined by Spearman's rho.

Results: Seventy-seven CF patients, median age 16.9 y (range 4-59 y) were evaluated. Twenty-eight were pancreatic sufficient (PS) and 49 pancreatic insufficient (PI). FEV1 was 81.5+ 21.2 % predicted (mean + sd); BMI sds was 0.09+ 0.68. SOS z-score was 0.1 + 0.88 (range -2.2 to +1.9). Three of 77 patients had SOS <-1.6 (high risk for osteoporosis) and 6/77 had SOS -0.1 to -1.6.

DXA-LS was -0.57 + 1.35 (range -4.4 to +4.3). Nine of 61 had DXA-LS <-2 consistent with osteoporosis, whereas 14/61 had osteopenia with DXA-LS between -1 to -2.

There was no correlation between SOS and DXA-LS or serum bone markers. add clinically useful information regarding bone integrity in a small CF cohort.

Methods: Prior DXA scans were used to select study participants. Enrolled patients were stratified across WHO defined categories of bone density: normal (<-1.0), osteopenia (-1.0 to -2.5) and osteoporosis (>-2.5). Patients underwent two limb HR-pQCT (XtremeCT, Scanco Medical, Switzerland) which were correlated with routine DXA performed within one month.

Results: Fourteen non-transplanted CF patients were recruited. There were 5 patients in the normal BMD group: 3 male, FEV 1 range 32-82% predicted, four were pancreatic insufficient and two had CFRD. The HR-pQCT measurements showed an average total bone density (TBD) of 364 mg/cm 3 and 333 mg/cm 3 , trabecular bone density (TrBD) of 196 mg/cm 3 and 187 mg/cm 3 and cortical bone density (CBD) of 881 mg/cm 3 and 896 mg/cm 3 for the radius and tibia, respectively. The osteopenic group had 2 females, 2 males, FEV 1 range from 53-104% predicted, three were pancreatic insufficient and no CFRD. The HR-pQCT measurements showed an average TBD of 248 mg/cm 3 and 257 mg/cm 3 , TrBD of 118 mg/cm 3 and 137mg/cm 3 and CBD of 812 mg/cm 3 and 870 mg/cm 3 for the radius and tibia, respectively. The osteoporotic group had 3 females, 2 males, FEV 1 range from 20-33% predicted, all were pancreatic insufficient and 2 had CFRD. All patients in this group were on anti-resorptive therapy for at least 1 year prior to study participation. The HR-pQCT measurements showed an average TBD of 290 mg/cm 3 and 247 mg/cm 3 , TrBD of 141 mg/cm 3 and 121 mg/cm 3 and CBD of 821 mg/cm 3 and 845 mg/cm 3 for the radius and tibia, respectively. In the majority of patients, there was good correlation between results from DXA and HR-pQCT (normal values and standard deviations based on data from the Canadian Multicenter Osteoporosis Study). There were three patients that showed discordance between DXA and HR-pQCT BMD: two whose BMD was underestimated by DXA and one whose BMD was overvalued.

Conclusions: In general there was good correlation between HR-pQCT and DXA BMD measurements. However, for three patients (21% of study population), threshold for treatment using the HR-pQCT data would result in a change in clinical management. Larger studies are needed to expand on the role adjunctive imaging modalities may play in the assessment of bone health in CF.

Rand-Giovannetti, D.; Eakin, M.N.; Ridge, A.; Riekert, K.A. Pulmonary, Johns Hopkins University, Baltimore, MD, USA Rationale: Despite evidence of the importance of CF treatments in health outcomes, many patients do not perform their treatments as prescribed. One possible factor affecting an individual's decision to adhere to treatments is disease disclosure. If the individual does not feel comfortable telling others about their CF diagnosis, s/he may be less likely to perform treatments in their presence. The purpose of this study was to assess who individuals with CF feel comfortable disclosing to and how that decision affects their treatment adherence.

Methods: Thirty-nine adults and adolescents with CF age 16 and older (51% male; 92% Caucasian; mean age=26.7 years (SD=7.7); mean FEV1 % predicted= 73.2% (SD=25.8)) were recruited from the Johns Hopkins Adult and Pediatric CF clinics. Nineteen participants were single (49%), 19 married/with a partner (49%), and 1 divorced (3%). Participants completed a questionnaire assessing their disease disclosure to romantic partners, close friends, and casual acquaintances. For each audience, participants were asked whether they had told all, some or none of this group about their CF. Participants were also asked to rate how comfortable they felt discussing their CF experience with and doing treatments in front of this audience on a 1-10 scale where 1=not at all comfortable and 10=completely comfortable. A composite medication adherence score was calculated for each participant based on their pharmacy refills of dornase alpha, inhaled tobramycin, azithromycin, and hypertonic saline.

Results: As expected, participants were significantly more likely to share their CF diagnosis with romantic partners than close friends (p<0.01) or casual acquaintances (p<0.01). All (100%) of respondents with a roman-SOS was lower in those with Vitamin D3 < 30ng/ml compared to > 30 ng/ml (ANOVA, p<0.05) and correlated with MMP12 and inversely with iPTH (p<0.05).

DXA-LS but not SOS correlated with FEV1 (p<0.01), BMIsds (p<0.01) and bone markers (BSAP, PICP and CTx) (p<0.05) and decreased with age, chronic P. aeruginosa and diabetes (p< 0.05).

BSAP, PICP and CTx correlated with FEV1 and inversely with age, chronic P. aeruginosa and diabetes. iPTH level increased with IL4, IL6 and IL1β.

There was no correlation between either SOS or DXA-LS and calcium intake, activity level, serum calcium and phosphate, pancreatic status, oral steroid use or inflammatory markers IL4, IL6, IL8, TNFα.

Conclusion: CF bone disease is multifactorial and complex. Evaluation by SOS is accessible, radiation free and appears to provide information which is different compared to bone markers and DXA-LS, probably reflecting bone strength. Longitudinal follow up is proposed.

Supported by Grants from Schneider Children's Medical Center of Israel and from Sackler Faculty of Medicine, Tel Aviv University. People with cystic fibrosis (CF) are smaller than humans without CF. That observation has been attributed, in part, to decreased insulin-like growth factor 1 (IGF-1) levels, and this reduction has been blamed on malnutrition and pulmonary inflammation. We found that like humans, CF pigs were smaller than non-CF littermates and had lower IGF-1 levels. To better understand the basis of IGF-1 reduction, we studied newborn pigs and found low IGF-1 levels within 12 hours of birth. Moreover, humerus length and bone mineral content were decreased, consistent with less IGF-1 activity in utero. In exploring possible mechanisms, we discovered reduced growth hormone (GH) release in organotypic pituitary slice cultures of newborn CF pigs. These findings led us to test newborn humans with CF, and we found that they also had reduced IGF-1 levels. Discovering lower IGF-1 levels in newborns indicates that the decrease is not solely a consequence of malnutrition or pulmonary inflammation, and that loss of CFTR function has a more direct effect. These findings may explain the long-standing observation that CF newborns are smaller than non-CF babies and why some patients with good clinical status fail to reach their growth potential. The results also suggest that measuring IGF-1 levels might be of value as a biomarker. Finally, they raise the possibility that IGF-1 or GH supplementation beginning in infants might be beneficial in CF. Objectives: Metabolic bone disease in cystic fibrosis (CF) patients is highly prevalent. The standard method of assessment for bone mineral density is dual x-ray absorptiometry (DXA) which provides a two dimensional assessment of bone quantity. DXA scans have not been shown to correlate with fracture risk in CF. High resolution peripheral quantitative computed tomography (HR-pQCT) is a new technology that provides information on both the quantity and quality of bone and has not been studied in the CF population to date. Our pilot study sought to determine if HR-pQCT would Raise Expectations), a clustered randomized controlled trial, is the first study to translate an evidence-based, behavioral intervention to improve adherence in to a clinic setting. The success of these programs may depend, in part, on the commitment of team members across disciplines (i.e, physicians, nurses, etc.). The current study evaluated the participation of multidisciplinary team members in a structured adherence intervention training.

Methods: The iCARE study was designed to compare the effectiveness of provision pharmacy data vs. a Comprehensive Adherence Program (CAP) for improving adherence in adolescents with CF. This longitudinal study is currently being conducted at 21 CF care centers. Prior to implementing the CAP program, the sites receive training in problem-solving. This intervention involves a member of the multidisciplinary team working with the adolescent to identify barriers to daily treatments and brainstorm solutions to those barriers with their parent. Participation in the CAP training consisted of 1) an adherence overview, 2) training in the protocol, and 3) practice in role-playing. All team members were invited to the full CAP training; those who attended the protocol and role-play session were identified as "Adherence Champions." Data were collected using an attendance records.

Results: Seven of 11 CAP sites completed the attendance forms for the CAP training. Additional sites have been scheduled for training. Participation ranged from 8 to 15 people for the adherence overview and from 3 to 12 people for the role-play session across sites. Between 38 and 100% of those who attended the adherence overview also attended the role-play training (see Table for participation by discipline).

Conclusions: Overall, 78% of those who attended the adherence overview also participated in a role-play session to practice problem-solving techniques. Results of this study suggested a strong commitment to improving adherence by the multidisciplinary care team. Future directions include evaluation of the effects of team member participation on study outcome variables. It is hypothesized that sites with higher team participation will have patients with better adherence and health outcomes. Introduction: As a result of prolonged survival, more patients with cystic fibrosis (CF) participate in the labor force. The aim of our study was to evaluate social & demographic characteristics, education & occupation levels among adults with CF & to determine risk factors for work disability.

Patients & Methods: A total of 207 patients (100 males & 107 females -mean age: 30.7± 8.7 years) answered a self-administered questionnaire at our adult CF center between January & June 2009. We looked for their demographic data, educational level & occupational status. Independently, we reviewed medical records for illness severity indicators.

Results: Exocrine pancreatic insufficiency was present in 83.4% of the patients and 31.4% had diabetes. Severe respiratory insufficiency (FEV1<40%) was present in 28.5% of the patients, whereas 24.6% had a mild disease (FEV1>70%). Sixty-three percent had at least one antibiotic IV course per year, 66% received aerosol therapy, 9.7% oxygen therapy, 38% had oral nutritional supplements and 7% enteral feeding.

Fifty-nine percent of the patients were single, 37.6% were married or living together and 22.7% had children. A total of 18.8% were students, 56.5% had a job and 24.7% did not work (2 retired, 13 unemployed, 11 homemakers and 25 disabled). Among the 51 patients who did not work, The educational level among non-student patients was as follows: 71.7% had at least high school diploma, 56.7% had studied at least 2 years at college and 16.9% had a Master's degree. The job levels of the 117 patients in paid employment were as follows: professionals 25%, intermediate 37.1%, clerks 33.6% and workers 4.3%. A total of 64.6% of the patients worked full-time and 78.9% had permanent contracts. Sixty percent worked in a private company and 40% in the public sector. Only 33% had revealed they had CF at job interviews.

In our study, 70.2% had had at least one sick leave during the previous year, mainly during the IV antibiotic courses; 55.2% considered they had limitations in their job due to CF, 66.7% thought that CF prevented a career and 37.2% that it was a cause for a lower income. The study showed 36.7% of the patients had had an adjustment of workstation, mainly through a decrease of work duration.

With a univariate analysis, gender, age, marital status did not account for the difference between subjects who were working & those who were not. Patients living with their parents worked less (p=0.0002) and those with high educational level worked more (p=0.004). Age at diagnosis, CFTR genotype, pancreatic status, diabetes, BMI & bacterial colonization had no impact on employment status. Work disability was enhanced in patients with IV antibiotic courses (p<0.01), oxygen treatment (p<0.01), oral nutritional supplements (p<0.02), FVC<50% (p<0.05) and FEV1<70% (p<0.01). The multivariate analysis showed that the level of education was the most predictive factor of employment status.

Conclusion: These data show that many CF patients have access to professional life despite the severity of the disease. Higher educational level improves the chance to reach employment. Their employability can be enhanced by working part time.

Sawicki, G.S. 1 ; Kimberg, C. 2 ; Marciel, K.K. 2 ; Riekert, K. 3 ; Zhang, J. 4 ; Quittner, A.L. 2 1. Children's Hospital Boston, Boston, MA, USA; 2. University of Miami, Miami, FL, USA; 3. Johns Hopkins University, Baltimore, MD, USA; 4. Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA Background: Adherence to routine therapies among adolescents with cystic fibrosis (CF) requires acquisition of disease knowledge and self-management skills. Parental knowledge passed down to the adolescent is an important source of such information.

Methods: We examined CF-specific parental knowledge using the Knowledge of Disease Management in Parents of Adolescents with CF Questionnaire. Parents of children enrolled in the ongoing multi-center iCARE (I Change Adherence and Raise Expectations) study were surveyed using this 45-item multiple choice questionnaire as part of the iCARE baseline assessment. The questionnaire assesses knowledge in 4 domains (General Health, Lung Health, Nutrition, and Treatment) and is scored as the percentage of questions answered correctly. We tested the association of knowledge scores with parent and child characteristics using t-tests and Pearson correlations.

Results: As of the abstract deadline 41 parents from 5 iCARE sites have completed the questionnaire. Demographics were available from 32 parents (78% mothers, 56% with at least some college education). The mean age of their children was 14.2 years, 56% of their children are male, and the mean FEV1 of their children was 84% predicted. The mean overall parental disease knowledge score was 82.5%. Highest mean knowledge scores were in the Lung Health domain (93.6%), followed by Treatment (85.8%), General Health (78.4%), and Nutrition (75.2%). General Health and Nutrition scores were moderately correlated (r=0.43, p=0.005), but we found no other significant correlations among the 4 knowledge domains. There were no differences in knowledge scores based on parental gender, child gender, or parental education level. Scores on the nutrition knowledge domain were negatively correlated with child body mass index (BMI) percentile (r=-0.45, p=0.01). There appears to be a negative correlation between overall knowledge score and child age (r=-0.31, p=0.09).

Conclusion: Overall, CF-specific disease knowledge among parents appears high, and is highest for lung health and lowest for nutrition. Prelim-inary analysis reveals some differences in parental knowledge based on their child's age and health status. Parents of children with lower BMI percentile had higher nutritional knowledge scores, likely reflecting an increased involvement in their child's nutritional care. Improving parental knowledge in all aspects of CF care, and designing programs to facilitate transferring such knowledge from parents to their children, should be integrated into programs promoting adherence in CF.

Funding: Funding for iCARE is provided by Novartis Pharmaceuticals Corporation, Genentech, Inc., & Cystic Fibrosis Foundation. Methods: Qualitative, semi-structured paired interviews of adolescents with CF (ages 17-21) and one of their parents are being conducted with the goal of understanding current experiences and perceptions of health care transition.

Results: Thus far, 4 adolescent-parent pairs have been interviewed (8 one-hour interviews). Emerging themes from adolescents include: 1) the difficulty of balancing the development of autonomy with reliance on parental support, 2) uncertainty about the actual process of transition, and 3) realization that transition of CF care needs to be coupled with other life transitions. Themes from parents include: 1) anxiety over a lessening role in their child's daily health management as they get older, 2) concerns over future health insurance, and 3) uncertainty about the processes of care at an adult CF center. Adolescents and parents voice support for early discussions of transition, written transition care plans, and development of electronic medical records. No adolescents or parents interviewed were aware if anyone among their CF care team was currently responsible for coordinating transition.

Conclusions: Adolescents with CF and their parents express concerns over the health care transition process that intersect with the developmental challenges of balancing independence and continued reliance on family caregivers for CF care. Support for structured transition programs including written care plans and portable electronic medical records is common. Future interviews will further elucidate these emerging themes and aid in the design of patient and family centered CF transition programs. Background: Recent advances in the treatment of chronic, pediatric diseases have dramatically increased rates of survival. Genetic discoveries, for example, have led to the development of new medications for cystic fibrosis (CF). Newborn screening and more aggressive treatments have increased the median life expectancy for CF from school-age in 1955 to 37.4 years in 2008 (Cystic Fibrosis Foundation, 2008 . Transitioning to adult clinics is now an important milestone for most CF patients. Despite these advances, CF is still a life-shortening disease. Thus, there is a great need for palliative care services for pediatric patients and their families. Palliative care, originally developed as end of life care, has recently expanded to include patients with life-limiting conditions (IOM, 2003) . With the aims of reducing symptom severity and increasing health-related quality of life (HRQoL), palliative care can benefit patients with CF and their families.

Methods: The current study examined a subset (n=146) of a larger sample of healthcare providers who completed the Survey on Pediatric Palliative Care for Healthcare Providers. This sample included healthcare providers on CF teams (i.e. doctors, nurses, social workers, psychologists, and respiratory therapists). The sample came from an ethnically and racially diverse hospital in South Florida, primarily serving low SES and uninsured patients. The sample included 36.4% White Non-Hispanic, 28.0% White Hispanic, research provides a mechanism to identify individuals with anxiety and depression and to offer them psychological services.

Participants. Caregivers of children 17 and younger and patients 12 and older were invited to participate. To date, 34 individuals have participated in the research, including 16 parents (11 mothers, 5 fathers) and 18 patients (age 12-52 years, M=23.4; 6 males).

Measures. 1) Hospital Anxiety and Depression Scale: a 14-item instrument for screening depression and anxiety in patients with chronic medical conditions. 2) Center for Epidemiological Studies-Depression Scale: a 20item self-report measure designed to assess depression in community samples. 3) Cystic Fibrosis Questionnaire-Revised: a health-related quality of life (HRQOL) measure designed for use with CF patients and caregivers; measures physical, emotional, and social impact of CF on patients. 4) McMaster Family Assessment Device -General Functioning: a measure of overall health and pathology of family functioning.

Procedure. Families and adult patients attending routine clinic visits are approached in the lobby before any clinic procedures (e.g., weight, PFT) and invited to participate. After providing informed consent and assent, participants complete the measures above. Anxiety and depression screening measures are scored immediately; clinical elevation of any anxiety or depression score automatically triggers a consultation by the pediatric psychology doctoral student. In addition to consultation, families and individuals are offered the opportunity for outpatient psychotherapy with the doctoral student or referrals to outside mental health providers.

Results: Of the 34 participants recruited to date, 38% (N=13) met at least one clinical cutoff score for anxiety and/or depression symptoms, including 28% of patients and 44% of parents. Of families screened using the current measures, eight went on to seek outpatient mental health services with the doctoral student as a direct result of their study experience.

Discussion and Implications: This poster presents data demonstrating the use of psychosocial support services provided by the pediatric psychology doctoral student by patients and families identified via the screening study. Further, this poster presents preliminary data from the ongoing study including prevalence of mental health symptoms, HRQOL, and family functioning. These data demonstrate the clinical utility of the partnership between formal psychological screening measures to identify patients and caregivers who may benefit from psychological services and the provision of services via a pediatric psychology pre-doctoral trainee. Implications for improved health outcomes for patients and families will be discussed, along with feasibility implications for implementation using free or low-cost psychological services. Background: Managing a serious, chronic disease such as cystic fibrosis (CF) requires effective coping behaviors. Prior research has utilized generic measures of coping which are not specific to the disease or context. Given that adolescents with CF face unique challenges, such as performing daily treatments and disclosing their diagnosis, it is important to evaluate their ability to cope with these stressors. We have developed a context-specific measure of coping efficacy, the Role-Play Inventory of Situations and Coping Strategies (RISCS) (DiGirolamo et al., 1997) for adolescents with CF to evaluate coping in the following domains: Treatment Adherence, Parent-Teen Relationship, Eating/Weight Gain, Future Issues/Long-term Health, Medical Procedures/Symptoms, Friends, and School Issues. The purpose of this study was to examine correlates of effective coping, such as age, gender, health status, and parent-teen conflict.

Methods: In the context of a larger intervention study, 106 adolescents ages 10-17 completed the Adolescent RISCS, consisting of 25 vignettes that describe frequently occurring, problematic situations reported by adolescents with CF. For each situation, adolescents reported how frequently it occurred and how difficult it was. They then generated a coping response which was rated for efficacy on a 4-point scale using empirically-derived criteria. Adolescents also completed the Conflict Behavior Questionnaire (CBQ) (Robin & Foster, 1989) , which measured parent-teen conflict. In addition, demographic and medical information was gathered from parents and confirmed by chart review.

Analyses revealed significant within-subjects differences in coping efficacy across domains (p < .001). Overall, adolescents' scores for coping efficacy were highest in the Treatment Adherence domain and lowest in the Future Issues domain. Girls reported higher levels of overall problem difficulty and demonstrated higher coping efficacy in the Friends and Future Issues domains than boys. Effective coping in the Treatment Adherence domain was associated with: 1) younger age (p < .05), 2) lower conflict with parents (p < .05), and 3) lower perceived difficulty of the situation (p < .01). Coping efficacy in the Parent-Teen Relationship domain was associated with lower conflict with parents (p < .01), but not with age. Effective coping in the Friends domain was positively correlated with BMI percentile (p < .05); however, this was the only relationship found between coping and health status indicators.

Conclusions: These findings suggest that effective coping in adolescents with CF is situation-dependent, rather than uniform across situations. Further, there is no single set of factors that predicts effective coping across contexts. The Adolescent RISCS examines coping efficacy across contexts and is useful for researchers and clinicians in highlighting strengths and weaknesses in coping. Future analyses will examine the effect of a familybased intervention on these coping skills.

Funding was provided by NIH (HL #47064). Background: Social cognitive theory constructs such as motivation and perceived importance have been shown to be related to a patient's health behavior. However, these constructs have not been well studied in CF. One recent study found that self reported nonadherence to Chest Physical Therapy and antibiotics was related to lower ratings of medication necessity.

Objective: The purpose of this project was to evaluate the relation between a patient's motivation and perceived importance of taking medications and objectively measured medication adherence.

Methods: We recruited 39 older adolescents and adults (mean age =26.8, range 16-55 years, 51% male; 93% Caucasian) seen in a CF care center and prescribed at least one of the following pulmonary medications for the past 12 months: azithromycin, dornase alpha, inhaled tobramycin, and hypertonic saline. Participants completed a baseline assessment, which included self-report on motivation and importance of taking pulmonary medications on a scale of 1 -10, with 10 being the highest. Pulmonary medication adherence was measured using pharmacy refill records for the past 12 months and a composite medication possession ratio (MPR) was calculated according to the participant's regimen. Baseline health outcome data including FEV1 % predicted (FEV1 %) and BMI were abstracted from the medical records. Correlations were conducted to examine the relation between health beliefs and adherence. Multiple regressions were also conducted to control for health status on the relation between health beliefs and adherence.

Results: Preliminary results indicated that participants who reported higher motivation to take medications (r = 0.36; p = 0.02) had higher medication adherence. Furthermore, patients who rated the importance of taking medications higher had higher medication adherence also (r = 0.32; p = 0.05). There results were consistent even when controlling for baseline FEV1% and BMI.

Conclusions: Results indicate that patient ratings of importance of medication and motivation to take medications predicted their objectively measured adherence at baseline. Interventions are needed to target a patient's motivation and perceived importance of the medication to improve adherence. Clinicians should focus on discussing the patient's motivation and perceived importance of medication to improve adherence.

Supported by NHLBI-HL087997. Successful management of CF requires support from family and friends. Understanding how these individuals facilitate or interfere with disease management is particularly important during adolescence when adherence to treatments typically declines. This study identified common supportive and non-supportive behaviors provided by family and friends and explored which behaviors were related to treatment adherence and health outcomes.

Method: Data included 85 adolescents (mean age=14.34 yrs; 45% female) from 7 CF centers who completed the Baseline assessment of the iCARE study. Youths completed the Perceived Adolescent Social Support-CF questionnaire by rating the frequency and supportiveness of 24 family and 27 friend behaviors. They also documented the frequency and duration of treatments during the last week by completing the Treatment Adherence Questionnaire. Youths' responses to these questionnaires were linked with CFF registry data to extract the most recent BMI%ile, FEV 1 %pred, and pulmonary exacerbations during the last year.

Results: Analyses of the frequency and supportiveness of family and friend support and their relationship to disease management and health outcomes were conducted. The most frequent family behaviors included, treated me like a normal person, showed me they cared about my health, and helped me remember to do my treatments; these behaviors were also rated as the most supportive. The most frequent friend behaviors included, complained about my CF, left me out of activities because of my CF, and nagged me about exercising, while the most supportive behaviors were accepted my CF, showed me they cared about my health, and asked me how I was feeling. Significant relationships from the exploratory analyses included the following. Families' reminders, encouragement, rewards for treatments, and flexibility were related to enzyme adherence (r=.22 to .35). Families' naggings about treatments were related to fewer exacerbations (r=-.29), but treatment interference was related to more exacerbations (r=.23) and lower FEV 1 %Pred (r=-.24). Friends' acceptance, encouragement, flexibility, and reminders were related to fewer exacerbations (r=-.22 to -.33) and higher BMI%iles (r=.23 to .37), but friends' presence during treatments was related to lower adherence with airway clearance and aerosol medications (r=-.22 to -.30). Importantly, the strongest relationships were between behaviors that closely matched the outcome (e.g., friends' nagging about exercise and time spent exercising, r=.43).

Conclusions: Results highlight the complexities of managing CF in the context of family and peer relationships. Some behaviors that are perceived as less supportive (nagging) appear to relate to better outcomes, while others that are perceived as supportive (being present during treatments) appear to interfere with disease management. Moreover, family and friends appear to have both positive and negative influences on teens' health. Understanding the role of family and friends in managing CF will require more precise matching of specific behaviors to particular tasks of disease management.

Sponsored by an investigator-initiated grant to Dr. Quittner by Novartis Pharmaceuticals. Introduction: Erectile dysfunction (ED) is the inability to achieve and maintain an erection adequate for intercourse to the mutual satisfaction of the man and his partner. It has an important negative impact on self-confidence, interpersonal relationships and male quality of life. The prevalence of ED increases with age and with increasing life expectancy in CF, ED is possible in this population but its prevalence has never been sought.

Aim: To determine the prevalence of ED in the CF population in our centre.

Methods: Anonymous self-reporting validated questionnaire using the International Index of Erectile Function (IIEF-5) [1] was sent out to all male Children and adolescents with cystic fibrosis (CF) develop a personal meaning of their chronic illness and prognosis from their experiences of growing up with CF. Parents, families, and healthcare providers contribute to these impressions, which children filter through their limited cognitive ability. However, children's perceptions about CF may be discordant with reality and those of their parents and healthcare providers. To minimize differences from peers, children manage their perceptions through selective disclosure and concealment to enhance their quality of life.

Objectives: The purpose is to explore the perceptions, understanding, and personal meaning of chronic illness of children and adolescents with CF by synthesizing the results of three grounded theory qualitative studies.

Methods: Three grounded theory qualitative studies of children and adolescents with CF were synthesized to explore perceptions and misconstructions about chronic illness. Bronfenbrenner's ecological developmental framework was used to guide this research. A purposeful sample of 55 children and adolescents (6-22 years) with CF was obtained during three studies; theoretical saturation was achieved. In-depth interviews were used to explore perceptions, understanding, and personal meaning of CF. Strategies of reflection and paraphrasing were used to clarify meaning, confirm accuracy, and credibility. The constant-comparative method with open and axial coding was used to identify core categories, dimensions, and relationships.

Results: Children and adolescents with CF characterized their perceptions and understanding of their chronic illness over time as Discovering the Chronic Illness Trajectory: (a) discovering a sense of difference (school-age children), (b) reducing a sense of difference (early-to-middle adolescence), and (c) discovering the chronic illness course (late adolescence). These impressions revealed hidden truths, discordant views, and multiple realities (including expectation of early death), that were embedded in their perceptions about growing up with CF. As children and adolescents with CF struggled to understand the meaning of chronic illness, they experienced problems with peer relationships as they tried to appear normal in comparison to healthy peers. Moreover, their psychosocial and developmental demands conflicted with CF management and adherence to treatment, as they progressed along the chronic illness trajectory.

Conclusions: Understanding how children and adolescents with CF perceive their chronic illness is essential for developing interventions to improve CF management and adherence to treatment. Moreover, to improve quality of life for children and adolescents with CF, it is critical to understand the interplay between the demands of CF with psychosocial and developmental demands, as well as the social consequences of their chronic illness.

Funding: Center for Research on Chronic Illness, NINR, NIH (#2P30NR03962); Biomedical Research Grant, NIH; University Research Grant.

Dewar, M. 1, 5 ; Madge, S. 2 ; Voase, N.W. 3, 5 ; Sheridan, H. 4, 5 ; Bell, N.J. 1, 5 ; Reid, P.A. 1, 5 ; Quittner, A.L. 6 ; Davies, G. 3, 5 ; On behalf of the UK CF Gene Therapy Consortium, .. 5 1. Respiratory Unit, Western General Hospital, Edinburgh, United Kingdom; 2. Royal Brompton Hospital, London, United Kingdom; 3. Dept. Gene Therapy, Imperial College, London, United Kingdom; 4. The Royal Hospital for Sick Children, Edinburgh, United Kingdom; 5. UK CF Gene Therapy Consortium, Edinburgh, London, Oxford, United Kingdom; 6. University of Miami, Miami, FL, USA Introduction: Quality of life measures are increasingly used as outcome measures in CF clinical trials. The UK CF Gene Therapy Consortium Run-In study is a longitudinal observational study that has followed patients aged ≥10 years from London and Edinburgh at regular intervals during periods of clinical stability. Tests performed at each visit include physiological markers of severity, sputum and serum markers, and functional impact of disease. This abstract describes use of the CFQ-R UK Quality of Life questionnaire in our study population, to determine its role as a potential key outcome measure in our future multi-dose gene therapy clinical trial.

Method: For each of the 4 study visits, patients completed the appropriate CFQ-R UK (adult version for subjects 14 years and above; child for those <14 years). A score between 0 and 100 for each questionnaire domain (e.g. Respiratory, Weight) was obtained with higher scores reflecting better quality of life. Adult and child versions of the questionnaire were analysed separately. Results analysis concentrated on the Respiratory domain. Comparisons between groups were analysed using t-tests or Chi-squared analysis.

Results: At enrolment the mean (SD) age of patients in London was 24 (11.9) years and in Edinburgh 20.8 (9.9) years (p=0.052). London FEV 1 was 67 (18) % and Edinburgh 79 (19) % (p<0.001). The mean CFQ-R UK score in the Respiratory domain for all adult patients was 76 (SD 16.2) and for all children 80 (SD 14) at visit 1. Broken down according to clinical site, adult patients in London had a mean score of 73 (SD 16.7) and in Edinburgh of 82 (SD 13.5) at visit 1 (p<0.001). In children, there was no difference between sites (p=0.5). No longitudinal trends in scores were seen across visits. The coefficient of variation in Respiratory score between visits was 11% for adults and 12% for children. Results of other questionnaire domains were in keeping with expected values for our CF study population. Near maximal scores were obtained in a large proportion of patients in the domains of Role, Body Image, Weight, and Eating Disturbance. Children were more likely than adults to have a Respiratory domain score above 90 (Chi square test p = 0.01) and therefore would have less opportunity to detect change in response to any proposed therapeutic intervention.

Conclusions: The Respiratory domain of the CFQ-R UK may be a useful outcome measure in our population the majority of whom have suboptimal and reproducible scores. An arbitrary comparison of scores between clinical sites which differ in disease severity suggests a benchmark of score difference on which power calculations can be made. Other CFQ-R UK domains were characterised by either high baseline scores (and therefore limited potential to detect improvement in any clinical trial) or likely limited utility to discriminate change in CF subjects.

Funded by the UK CF Trust. Cystic fibrosis (CF) is a disease that places a significant burden on patients and families. However, assessments of burden of CF disease do not typically include the out-of-pocket costs and unpaid caregiving time associ-ment Adherence, Parent-Teen Relationship, Growing Up and Medical Procedures domains. Results indicated that caregivers within the same family demonstrated similar coping efficacy on the Financial Issues domain.

Conclusions: Caregivers and their adolescents exhibited different coping skills when faced with contextually similar challenges. Parents who demonstrated effective skills when managing future concerns had adolescents who displayed better coping skills with eating and school issues. Higher parent ratings of frequency and difficulty were related to better adolescent health outcomes, perhaps indicating greater parent involvement and supervision. Future directions include determining whether a behavioral family systems intervention improves parental coping behavior across different situations.

Funding was provided by NIH (HL #47064). Background: Knowledge of disease management is an important precursor to the development of successful adherence behaviors; it is necessary, but not sufficient, for behavioral change. Therefore, gaps in knowledge may lead to decreased adherence behaviors and worse health outcomes. Clinical care teams are often unaware of adolescents' knowledge deficits because many patients are diagnosed in infancy and care teams may assume that adolescents have received sufficient education. The current study assessed gaps in knowledge of disease management in adolescents with CF, identified factors associated with adolescent knowledge and evaluated the relationships between knowledge and health outcomes.

Methods: The iCARE (I Change Adherence and Raise Expectations) study is a multi-site, randomized controlled trial designed to compare the effectiveness of two adherence interventions: provision of pharmacy refill data and the Comprehensive Adherence Program (CAP). CAP is a multicomponent treatment which includes measurement and remediation of gaps in knowledge and skills, as well as a problem-solving intervention. Adolescents completed a 43-item multiple-choice questionnaire that evaluated Knowledge of Disease Management across four domains: General Health, Lung Health, Nutrition, and Treatments. Percent correct was calculated for each domain; a total correct percentage score was also derived.

Results: Preliminary results from 83 adolescents (M = 14.39 years, SD = 2.51) at 7 CF Centers are reported. Adolescents were most knowledgeable about Lung Health (% correct M = 86.7, SD = 14.7), followed by General Health (% correct M = 80.0, SD = 18.0), Treatments (% correct M = 79.7, SD = 16.1), and Nutrition (% correct M = 65.6, SD = 14.9). Total Knowledge ranged from 34.9% to 95.3% correct, with the three most commonly missed questions falling within the Nutrition domain. Adolescent gender and family income were not associated with adolescent knowledge scores; however, higher maternal education was associated with greater adolescent knowledge for Lung Health and Nutrition (r = .43, p < .01, r = .27, p < .05, respectively). Older age was significantly associated with better knowledge (General Health domain r = .35, p < .01; Lung Health domain r = .22, p < .05; Nutrition domain r = .50, p < .01; Treatment domain r = .33, p < .01). General Health knowledge was also positively related to a higher BMI percentile (r = .31, p < .01); the Nutrition domain was positively associated with better observed enzyme skills (r = .27, p < .05).

Conclusions: Adolescents with CF were generally knowledgeable about their illness except for the nutrition domain. Knowledge was related to some treatment skills and BMI percentile. Although CF teams may be providing relevant education, it appears that younger adolescents with CF have major gaps in knowledge. These results can be applied clinically by measuring practical knowledge and remediating observed gaps. Future directions include determining whether remediation of knowledge gaps improves adherence behaviors and health outcomes.

Funding was provided by Novartis Pharmaceuticals Corporation, Genentech, Inc., and Cystic Fibrosis Foundation.

plished with a team approach. The Baylor College of Medicine Adult CF Center cares for 175 patients. Because of the large size of our center, information regarding hospital discharge planning of CF inpatients (IP) was not consistently being transmitted to the CF outpatient (OP) center. The OP center staff was often not notified of the discharge plan developed in the hospital. Rotations in CF Center medical personnel contributed to further breakdown in the communication chain. It thus became important to set up a new process.

Methods: A meeting was held to identify barriers and develop solutions to improve flow of information from IP to OP care. The meeting was attended by the CF hospital case manager and pharmacist, the CF center medical director, nurse practitioner, and center coordinator. After review of the current process and identification of deficiencies in communication, the "Hospital to Home Worksheet" was created. It included information regarding discharge date, discharge medications, home antibiotic dosages, schedule and length of therapies, required lab testing, home care agency name and number, and DME company if indicated. A phone call from an OP nurse to the patient made one day post discharge was added to confirm the discharge plan.

Results: The process was implemented as follows. The Hospital to Home Worksheet is initiated by the case manager at the onset of discharge planning. She/He completes the worksheet, discusses it with the patient, who is then given a copy. The pharmacist meets with the patient to review discharge medications, including changes and additions to the patient's previous regimen. The case manager notifies the CF coordinator by phone of the patient's discharge and the Worksheet is faxed to the CF center. This information is then incorporated into the patient's electronic medical record. The date and time of the hospital follow-up appointment is confirmed with the patient prior to discharge. The OP nurse makes a one day post discharge phone call to the patient.

Conclusion: Development and implementation of the "Hospital to Home Worksheet" has allowed information regarding hospitalization and discharge to be transmitted from IP to OP staffs. The OP center has the information needed to give complete follow-up care. Patients are returning more consistently for recommended hospital follow-up visits. Patients have a positive response to the phone call received a day after discharge. Coordination of care from the IP to the OP setting has been achieved. Our future plan is to review the last quarter of discharges and determine the impact of this plan. As Riley Hospital completes an expansion of beds, care delivery changes from age based to service based. This change has led to the need to provide structured, specific education related to CF care. Our center engaged in a quality project to address the needs of the inpatient care team.

The aim of this project was to offer CF care education and improve communication. The CF care team designed an educational offering to meet the assessed needs of the pulmonary unit. The project was initiated January 2010 and completed July 2010. Staff for the pulmonary inpatient unit was selected by nursing administration. Senior inpatient nursing staff with CF care experience was selected for the unit, however many took supervisory roles with less experienced staff remaining for daily care needs. Issues immediately identified by the team and inpatient management were medication administration, infection control, nutrition, airway clearance, new diagnosis teaching/care, patient/staff boundaries, and transition.

The team worked with the inpatient management to develop content for educational needs and completion of education in a timely manner, especially in regard to infection control issues."Hot topic" issues related to transition and boundaries that were time sensitive. There was no desire to extend educational offerings over several weeks to complete the 6 modules. The CF team and unit management proceeded with planning two identical weekend sessions. Borrowing from other CF teams we referred to the educational sessions as CF Bootcamp.The sessions were interactive and intended to promote the beginning of team cohesiveness. The sessions were taped and are now used for staff orientation. Anecdotally, the team has recognized the benefit of the educational offerings as the inpatient staff clearly are able to identify all members of the team and utilize them as resources. The team was able to ascertain through post education evaluations that the topics selected met the perceived needs of the staff. Attendance for two sessions was 93. Eighty-one evaluations were completed. CF Basics, Nutrition, Medications,and ACT scored over 90% when evaluated on relevancy to practice. Interestingly, Life and CF, focused on psychosocial and transition issues scored low with 56% of respondents evaluating the presentation as relevant to practice. The needs assessment, however, listed transition most frequently as a topic for which additional information was desired; 25% of responses. Other topics identified were: Advanced ACT 15%, PFT 15%.

To ascertain whether learning has been sustained the team designed and will administer a 6 month evaluation. The post test content is based upon content presented January 2010. The survey will be completed on line.The design of the post test reflects basic knowledge and can be completed quickly. Based upon the results of the education survey and the first round needs assessment, the CF team will plan additional education opportunities for the inpatient staff.

Finally, the needs related to transition have led to planning for the CF team and the Family Advisory Board. Two possible education sessions have been identified: family perception of transition, presented by families to the inpatient staff; staff perceptions of transition and how to implement a sound plan for transition. Background: For cystic fibrosis (CF) patients depression, anxiety, stress and/or pain management are common complications of the disease. A beneficial service service at Morristown Memorial Hospital is Integrative Medicine. They offer holistic treatments to assist with these issues. The purpose of this study was to measure the efficacy of the treatments given by Integrative Medicine on the somatic complains of depression, anxiety/stress, and pain management.

Methods: A pilot study was conducted between the Adult Cystic Fibrosis Center at Morristown Memorial Hospital and Atlantic Integrative Medicine to study the effects of holistic treatments on hospitalized and outpatient adult CF patients. Upon admission, a referral was made by the social worker to the Integrative Medicine coordinator. An appointment with a holistic nurse was then scheduled within 24-48 hours, and a bedside assessment and/or a treatment session was provided. A survey was given to patients after treatment and forwarded to the CF social worker. A similar protocol was used for outpatient referrals.

Results: Ten patients received treatments from Integrative Medicine in 2009. Four patients did not complete surveys. Six patients provided feedback by 10 surveys. Three of the patients who completed surveys had multiple sessions and completed surveys at the end of each session. Feedback on one patient was provided by holistic nurse providing the treatments. The predominant techniques used were: sinus release, CranioSacral Therapy, Jin Shin Jitsu, and relaxation/guided imagery exercises. Patients (both inpatient and outpatient) have stated, via survey, reductions in pain symptoms and stress, depression or anxiety.

Using the 10 point Likert scale measuring both pain and depression/anxiety/stress, an outpatient's pain reduction had ranged in a 0-7 point drop, with an average drop of 3.5 points. An outpatient's depression/anxiety/stress had dropped during treatment from 0-4 points, and after treatment dropped a range from 1-5 points. The average change in depression/anxiety/stress during treatment was a 1.75 point drop, and after treatment was a 2.5 point drop.

An inpatient's pain reduction had ranged from 0-4 points, with an average of a 2.3 point drop. An inpatient's depression/anxiety/stress reduction ranged from a 3 to a 5 point drop, for both during and after treatments. The average point drop for during treatments was a 3.7, and an average of a 3.8 point drop was obtained after treatments.

Discussion/Conclusion: The involvement of Atlantic Integrative Medicine's therapeutic techniques for pain management and the reduction of anxiety/depression/stress symptoms appears to have been beneficial for hospitalized and outpatient adult cystic fibrosis patients. These treatments appear to achieve the greatest level of efficacy for pain management in an outpatient setting, and depression/anxiety/stress in the inpatient setting. Results for pain management in the inpatient setting, and depression/anxiety/stress in the outpatient setting were still noteworthy. in answers among age groups. However, the question related to chest clearance approached significance with an increase in age associated with a higher incidence of incorrect answers.

These results suggest that patients are well educated regarding CF with noted gaps in knowledge regarding CFRD, nutrition in patients, and chest clearance in adolescents. This questionnaire has also achieved the goal of increased interactive educational opportunities between nurses and patients. Future plans are the continued use of this questionnaire with collection of data through 2010. Adaptation of the questionnaire for age specific needs is also in progress. In addition, the dieticians are developing a similar questionnaire for the patients in order to directly educate them about nutrition.

Introduction and Aim: Patients with cystic fibrosis are prone to dehydration, especially in the presence of adverse external factors like a rise in environmental temperature, fever, excess loss and/or inadequate intake of salt and fluids. If dehydration is severe and losses are not replaced, this may lead to hypovolemic shock with detrimental effect on vital organs. At the present study, we investigated the impact of dehydration on kidneys of infants with cystic fibrosis. This impact was evaluated with renal ultrasound scans after an episode of severe dehydration.

Patients and Methods: Over the last year, 12 infants with CF were admitted to the hospital with severe dehydration. There were 6 boys and 6 girls (mean age 17 months). All the infants were admitted during summer months (June to August). For 4/12 dehydration was the presenting symptom of cystic fibrosis. All infants were investigated with renal ultrasound scan while dehydrated. The severity of dehydration was correlated to scan findings.

Results: Four of 12 infants with CF and dehydration had abnormal renal ultrasound at the time of dehydration. To be more specific, the ultrasound examination demonstrated echogenic kidneys with preservation of the corticomedullary differentiation. The kidneys were equally involved. All 4 children had markedly abnormal electrolytes with chloride below 70 mmol/L and urea above 100 mg/dL. Conclusion: Severe dehydration alters renal morphology as detected by ultrasound scan. The clinical implication of this observation is very significant, so as to avoid concomitant use of nephrotoxic drugs, which may have an add-on effect on vulnerable kidneys and cause irreversible damage. nities to enhance nutritional intake were missed since few children supplemented their oral intake between breakfast and lunch and 20% of children admitted to missing enzymes on weekly basis. Privacy concerns, embarrassment over being different and time management were issues which surfaced when children were questioned about how enzymes were administered at school.

The study author wrote legislation which would allow children with CF to carry and self-administer pancreatic enzymes while at school. Working with the Jacksonville legislators, this bill was passed into law during the 2010 Florida legislative session. The law will take effect starting with the 2010-2011 school year. The second part of the study will be completed after the coming school year to determine the impact of self-administration of pancreatic enzymes while at school on perception of illness and empowerment along with FEV1 and BMI. Aim: Describe experiences of ward admissions in young adults transitioned to an adult centre.

Methods: Consent was obtained from 90 patients who were sent an online questionnaire. Fifty two closed questions on their rating of experience of transition, healthcare staff and ward services and environment. An optional open question was included at the end of the questionnaire inviting respondents' comments. Thematic analysis was the method chosen to treat the qualitative data from the open question.

Results: The response rate to the questionnaire was 78% (n=70) and mean age was 20.3 yrs (range 16-24 yrs). The number of responses to the open question was 63% (n=43). Analysis of qualitative data led to aggregation of data into three main themes (Table) .

Discussion: This study elicited the views of young adults undergoing transition and initial admission into an adult respiratory ward. The high response rate to the survey and the open question is evidence that patients with a chronic condition, such as CF, are interested in being active participants on the development of services regarding their care. Support around the first admission following transition to the adult ward was highlighted as an important area. There was also a demand for more knowledge, coordination, availability and positive attitudes from staff. Facilitation of patients' role in decision-making and the need to acknowledge their views and opinions was seen as improving satisfaction and adherence to treatment.

Conclusion: The importance of transition clinic/process, staff education concerning this age group and new ways of providing information to patients (e.g. DVD, Blog) is acknowledged. Methods: A prevalence-based model was developed incorporating agespecific PA prevalence rates, utilization of medical services, and productivity data. The 2008 U.S. direct (medical and non-medical) and indirect costs for CF patients infected with PA were calculated by clinical stage (i.e. acute PA infection, chronic PA without acute exacerbations, and chronic PA with acute exacerbations). Model parameters were obtained from medical literature (most recent 5 years of published information), practice guidelines, Medicare and managed care fee schedules, and expert panel input. Sensitivity analyses were conducted by varying selected model inputs by ± 25%.

Results: In 2008, the average annual per capita costs of PA among CF patients were $54,127 ($51,872 in direct medical, $652 in direct non-medical, and $1,604 in indirect costs), corresponding to approximately $634 million total costs in the U.S. population. Patients with acute PA infection and chronic PA with exacerbations used the most medical services. The average costs for acute PA infection decreased as patients aged. The sensitivity assessment did not appear to have a significant impact as results did not vary by more than 11%.

Conclusions: In the United States, direct medical costs contributed 96% of the total per capita and national costs for the care of CF patients with PA. Compared to previously published cost models for the United States, our costs are based on recently available data, real world age-and clinical stagespecific treatment patterns and costs, and a wider range of cost components.

Research funded by Bayer Schering Pharma AG, Germany.

Kinyon-Munch, K. Pulmonology, Nemours Children's Clinic, Jacksonville, FL, USA

The daily medical management for cystic fibrosis (CF) is time consuming and complicated. Rates of adherence to CF care are lowest for dietary recommendations with only 16-20% persons following the prescribed nutritional and gastrointestinal therapies (Modi, Lim, Yu, et al., 2006) . Adherence to pancreatic enzymes is estimated to be anywhere from 27% to 43% (Modi, Lim, Yu, et al.) . A prospective cohort study design involving 38 children and 46 parents was conducted. At a routine CF clinic visit, each participant was asked to complete the Brief Illness Perception Questionnaire which is an eight question Lichard scaled questionnaire. A chart review was completed after the visit to gather data on the child's FEV1 percentile, BMI percentile and enzyme use at school.

Results seemed to show a difference between child and parental perception of how cystic fibrosis affects the child's life, with parents perceiving a significantly higher severity effect than their children. However, the children had greater feelings of anger, sadness, fear, and frustration than their parents were aware. Children also felt their treatments were less effective than their parents. There was a tendency for children to perceive less control over their disease than their parents, with parents feeling they have a greater understanding of their child's CF.

Few children appeared to be manipulating their oral intake while at school despite the majority who were required to have their pancreatic enzymes administered by front office school personnel. However, opportu-it possible to literally navigate through the organs in the human body to specific genetic mutations in the DNA, whilst combining information from multiple databases together with experimental data from patients and new models supplied by researchers. This tool is linked with different databases related to diseases, organs, tissue, proteins, pathways, genes, and metabolites in a workflow model created by the user, and also provides an API in order to exploit your own data or to use the visualization tool without exploiting the data given by Activ8. Activ8 contains all the information related to cystic fibrosis as it was born from a project that was focused on it: ActivA.

Here we present a demonstration of the Activ8 system as an educational resource available from virtually any computer connected to the internet and needing just simple standard and free plugin installations that can be used anywhere. We show how access to the information related to CF in a multilayer fashion can be used to move access to the information and to education on CF from a complicated task to a visually interactive environment, which in turn links the patient with the top research that is available in public databases.

Weiland, J. Pulmonary Medicine, Cincinnati Children's Hospital, Cincinnati, OH, USA

The Cincinnati Children's Hospital CF Team had struggled to find the ideal format for working with families and patients at the time of diagnosis. An early approach was to bring the families to clinic for education, evaluation and to meet with the entire team. This visit lasted up to five hours. Several years ago the process was revised based upon feedback from families. An interdisciplinary group, including families, developed an inpatient process for education, support, assessment and family mentoring. This process was successful in that families felt confident in their ability to provide care to their children by the end of the brief admission. However, a review of respiratory cultures revealed an increased incidence of MRSA in the children that were hospitalized for initial education compared to those that had been educated in an outpatient setting. As a result, this process was stopped and the team slowly reverted to the old process of providing the education in the outpatient setting. Without a consistent approach, however, the team was concerned that there were gaps in the information and support provided to each family. In 2009, the CF Foundation published "Cystic Fibrosis Foundation evidence-based guidelines for management of infants with cystic fibrosis." Based upon the recommendations in the guidelines, the Cincinnati CF Team developed a system to ensure consistency in education. The team first designed a tool to gather information via telephone from parents of newly diagnosed infants. This tool provides data to the team that assists them in planning for the initial visit. Next, the team mapped out the itinerary for the initial or "CF Overview" visit. They specified which providers would see the patient and in which order. Time and content were also considered to avoid overwhelming families with too long of a visit or information overload. The input from the parents on the team was critical in determining content and structure. The team decided that there was a need for standardizing the information shared by the physician and ensuring the use of visual teaching aids. A Powerpoint® show was developed with a CF overview. It was placed on a flash drive so that parents could take it home with them to share with others. The itineraries for visits two and three were similarly developed. Based upon the recommendations of the Guidelines, the team included which testing would be completed at each visit. Power-point® shows for portions of visits two and three are in development. Measures of satisfaction with the process and confidence in ability to provide care will be completed by families at the end of the third visit. Results will be compared to data collected from families whose children were admitted for initial education.

Data collected included 25-OHD levels before and after treatment and vitamin D dose. Student's paired t-test was used to compare 25-OHD levels before and after treatment. CF patients who were pancreatic sufficient and infants were excluded from this study. Results: Among 41 CF patients (mean age 9.54±5.6 years; 21 females, 20 males) who had 25-OHD level checked, 26 patients (63%) had a vitamin D level < 30 ng/mL. 19 patients had a level <25 ng/mL. Mean 25-OHD level was 20.2±3.51 ng/mL. 16 patients were started on high dose vitamin D supplementation. Repeat 25-OHD levels were available for 14 patients. Mean 25-OHD level was 62.4±35.4 ng/mL. The 25-OHD levels were significantly higher after treatment (p-value 0.001). In 11 patients (79%), the 25-OHD level was normal, but it was still < 30 ng/mL in 3 patients (21%). Two patients had a level >100 ng/mL. Conclusion: Routine daily supplementation of vitamin D in CF patients provides inadequate amount of vitamin D and results in vitamin D deficiency in the majority of CF patients

The aim of this study was to assess the efficacy, tolerability and safety of the risedronate in adults with CF and low BMD. Patients with a lumbar spine (LS), total hip (TH) or femoral neck (FN) BMD Z-score of -1 or less were randomised to receive risedronate 35 mg weekly or identical placebo, and calcium (1g) + vitamin D3 (800 iU), in addition to their standard multivitamin supplements

3) p=0.48, respectively. In conclusion, after two years treatment there was a significant increase in lumbar spine BMD with weekly risedronate compared to placebo. Strategies to reduce the incidence of bone pain following bisphosphonate treatment need to be

Cystic Fibrosis Therapeutics, Bethesda, MD, USA ly but significantly related to weight z-score both at entry (r2=0.043, p<0.01)and at end follow-up (r2=0.048, p<0.01), whereas insulin sensitivity was not. In contrast, height z-score was significantly and inversely related to insulin sensitivity, both at entry and end follow-up (p<0.05). In addition, beta cell function was significantly related to FEV1 percent in each of the subsequent years of follow-up (p<0.01) but not to the rate of FEV1 decline during the whole observation period. In conclusion, beta cell function defects are significantly related to respiratory defects and to weight reduction over a subsequent four year period. However

EARLY INTRODUCTION OF ONCE-DAILY INSULIN DETEMIR IMPROVES WEIGHT & LUNG FUNCTION IN INSULIN-DEFICIENT CYSTIC FIBROSIS

1,2 ; Walker

Pancreatic destruction in cystic fibrosis (CF) causes insulin deficiency, progressing to diabetes. Declining weight and lung function, associated with early insulin deficiency precedes diabetes onset and predicts early mortality. Insulin treatment is standard care for diabetes in CF, but not yet for early insulin deficiency. Conventional multiple-daily insulin injections are poorly accepted by patients and carry significant hypoglycemia risk. Objectives: We proposed that a single daily dose of insulin detemir (Levemir) would be anabolic, improve lung function and would be well accepted. Methods: Eighteen CF-subjects (median age 12.5 yrs, range 7.2-18.1), newly diagnosed by oral glucose tolerance test with early insulin deficiency (n=12) or diabetes (n=6), were treated with once-daily, pre-breakfast insulin detemir (median dose 0.13 units/kg/day, IQR 0.12) adjusted for blood glucose target range of 4-8 mmol/L (72-144mg/dL). We compared changes after insulin treatment to changes prior to treatment (over a comparable period) in the following: weight standard deviation score (∆WtSDS), % predicted forced expiratory volume in 1 second (∆%FEV1) and % predicted forced vital capacity (∆%FVC)

The subgroup with early insulin deficiency (n=12), but without diabetes

During this 72 hours, subjects attended our centre for a 2 hour OGTT, having fasted overnight. For validation of CGM in CF patients, plasma glucose ("gold standard") at 0 hours and 2 hours during OGTT in CF subjects was compared with concurrent CGM sensor values. Deviation of sensor glucose from plasma glucose was compared with established acceptable ranges from non-CF validation studies. CGM results are described in terms of mean glucose value (over 72 hours) and variability of glucose value (SD of glucose values over 72 hours). Results: Mean (SD) deviation of CGM glucose from OGTT plasma glucose in CF subjects was 15.7% (14.3%), consistent with existing data in non-CF studies

CGM is a valid tool in adults with CF, its accuracy being consistent with that found in non-CF studies

OGTT results in adults with CF do not necessarily reflect their "real life" glycaemic control. Many CF subjects with normal OGTT have abnormally high or variable values on CGM

CGM may have a role in early recognition and treatment of abnormal glycaemic control in stable adults with CF, regardless of OGTT result

Methods: Prospective cross-sectional study to analyze fracture history in a large cohort (n=43) of juvenile patients with CF, aged 17.8 ± 6.20 (6.20-29.8) years. In addition, bone mass, density, geometry, and strength as well as forearm muscle size and strength were measured by peripheral quantitative computed tomography (pQCT) and DXA. Data were compared to a cohort of 700 healthy controls of the DONALD Study. In addition, we determined pertinent biochemical parameters of bone metabolism. Results: Eighteen of 43 patients (41.9%) had already experienced fractures predominantly of the peripheral skeleton, 8 of them more than one. Compared to the general German population in 2007 aged 0-30 years (26,370,006 persons with 127,903 fractures) with a fracture rate of 48.4 per 10,000 patient years (95% CI, 48.1 to 48.8), this CF cohort exhibited a 10-fold increased fracture rate of 445 fractures per 10,000 patient years (95% CI, 307 to 622). The probability of remaining free of any fracture was reduced in CF patients to 39

Referral for OGTT was indicated by the baseline or 1 hour BGL reading in 96% of patients, indicating this screening test could be shortened to 1 hour. Comparisons were made between the 2 hour BGL readings in both the GCT and OGTT, 11 of 22 (50%) patients showed agreement between glucose categories between the 2 tests. This suggests the GCT is not a suitable substitute for the formal OGTT. This study is on going, the 12 month collection period will end at the end of July this year. At this point, the prevalence of IGT and CFRD in our 10 -19 yr age group is 11% and 9% respectively

Diagnosis, screening and management of cystic fibrosis related diabetes mellitus. A consensus report. Diabetes Research and Clinical Practice

Pulmonary/Critical Care

Anxiety and depression in cystic fibrosis. Seminars in Respiratory and Critical Care Medicine

The Health Anxiety Questionnaire

Long-term study of one hundred five patients with cystic fibrosis; studies made over a five-to fourteen-year period

Background: Studies report reduced bone mineral density (BMD) in patients with CF. Vitamin K (vit K) is important for carboxylation of osteocalcin (OC). A lack of vit K results in undercarboxylated OC (Glu-OC) causing an imbalance in bone formation and breakdown. CF patients are at risk of vit K deficiency due to pancreatic insufficiency and liver disease.Aim: To determine if vit K supplementation over 12 months is effective in improving carboxylation of OC and bone formation in patients with CF.Method: Thirty-eight patients(age 16-43 yrs) not taking vit K recruited from CF clinics at Barts and the London, Kings College Hospital and Lewisham University NHS Trusts were randomised to receive either 10 mg vit K or placebo. Data collected pre and post supplementation included: bone markers (total OC, Glu-OC), serum vitamin D 3 (D 3 ), BMD Z scores (femoral hip and lumbar spine),spirometry, number of courses of antibiotics, and oral steroids, dietary energy, protein and vitamin A, D, K and calcium intake and body mass index (BMI). Primary outcomes were Glu-Oc to OC ratios and BMD Z scores.Results: Supplementation with vit K improved carboxylation of OC (p=0.007); mean reduction in Glu-OC to OC of 14.02 % (SD±15.94) in the treatment group compared to 3.69 % (SD±20.59) in the placebo group. Z scores at the study start indicated low BMD. Twenty percent in the placebo group and 16.7% in the treatment group had Z scores <-1.0 (femoral hip). Thirty percent in the placebo group and 33.3% of the treatment group had Z scores <-1.0 (lumbar spine). One participant had a Z score consistent with osteoporosis at the lumbar spine.At 12 months mean change in the placebo group at the femoral hip was +0.053 g/cm 3 (SD±0.19) and in the treatment group was -0.20 g/cm 3 (SD±0.31) Mean change in Z scores in the placebo group at the lumbar spine was +0.041 g/cm 3 (SD±0.15) and in the treatment group was -0.073 The effects of early derangements in glucose tolerance mechanisms on the nutritional status and respiratory function of cystic fibrosis patients are currently investigated. Starting from year 2003 patients who received routinely OGTT at the regional CF center of Milan were followed prospectively. The OGTT at recruitment was analysed for glucose tolerance and insulin secretory (beta cell glucose sensitivity) and insulin sensitivity parameters derived from modelling of glucose, insulin and c-peptide profiles.

One hundred thirty-five patients aged 8-31 yr (mean age 16.9±0.4yr) were followed for 590 patient-years (median length of follow-up 4.5 years) and their weight, height and BMI z-scores were evaluated at entry and at end follow-up, with the FEV1 % values during the observation period. These data were related to the insulin secretory and insulin sensitivity parameters evaluated at entry. Twenty healthy subjects with a comparable age and sex distribution received a similar OGTT protocol and served as controls.Height z-score at entry was -0.6±0.1 in the subjects before puberty and turned to -0.8±0.1 at end follow-up (p=n.s) whereas weight z-score in all subjects was-0.5±0.1 at entry and turned -0.9±0.1 (p<0.001) at end followup. Beta-cell glucose sensitivity was reduced in CF compared to controls (70.0±4.1 vs 117.9±11.6 pmolmin -1 m -2 mM -1 , p>0.001) whereas insulin sensitivity was almost superimposable. Beta-cell glucose sensitivity was weak-

There is a well-demonstrated correlation between BMI and lung function in both children and adults with CF. Previous studies have evaluated demographic and clinical characteristics as predictors of pulmonary outcomes in childhood, but the impact of early childhood nutritional status on outcomes in adulthood has not been thoroughly explored. The goal of this study is to examine the long-term clinical effects of malnutrition early in life.

Patients were selected from a retrospective review of the Emory University CF Registry data, and were included if visit data was recorded at both 12-36 months of age, and again at age 18-20 years. Presence of malnutrition at the childhood visit was defined as weight for age (WFA)< 10th percentile and at the adult visit as a body mass index (BMI) < 10th percentile. Initial comparisons were made using t-tests for linear and Chi square or Fisher's Exact test for categorical variables. Linear regression models were compiled to predict adult BMI percentile and adult FEV1. Childhood characteristics (gender, diagnosis suggested by meconium ileus (MI), pancreatic enzyme use, age at diagnosis, year of diagnosis, Pseudomonas infection, and Medicaid at diagnosis) were considered as potential predictors in a stepwise fashion with significance defined as p=0.10 for entry and p=0.05 to stay in the model. We also drew Kaplan-Meier survival curves to evaluate whether WFA < 10th percentile was a predictor of childhood death in order to rule out survivor bias in our primary analysis.Results: Data was analyzed for 91 patients, diagnosed between 1979-1994; 43% of patients were female, and average age of diagnosis was 6.7 months. At the childhood visit, 42% of patients were malnourished; 37% were malnourished at the adult visit. Diagnosis was suggested by MI in 19% of patients; the rate of MI was significantly higher in patients malnourished at the childhood visit (p=0.033) but not in patients malnourished at the adult visit. Although gender was similar between groups at the childhood visit, at the adult visit 26% of malnourished patients were female versus 60% of patients without malnutrition (p=0.002). In the linear model of adult BMI percentile, pancreatic insufficiency (p= 0.001) and female gender (p=0.011) were significant predictors; childhood WFA approached significance (p=0.060). In the model of adult FEV1, childhood WFA was the only significant predictor (p=0.011). In the survival analysis, Wilcoxon rank test of equality demonstrated a significant increase in mortality prior to age 20 in patients with malnutrition at the childhood visit (p=0.041).Conclusions: This study shows a significant association between early childhood WFA and adult FEV1. In our single-center analysis, the relationship with adult BMI is less clear-cut. This may be due to inadequate power. Since this and previous studies have suggested a causal relationship between nutritional status and later-in-life outcomes, further analysis is warranted. We plan to examine whether a critical age can be determined at which a more clear-cut predictive relationship exists between childhood status and adult nutritional and pulmonary outcomes. Furthermore, we believe that the question merits an analysis of the entire CF patient registry database.

Parkins, M.D. 1, 2 ; Ibrahim, M.A. 2 ; Rendall, J.C. 2 ; Elborn, J.S. 2 1. Department of Medicine, University of Calgary, Calgary, AB, Canada; 2. N Ireland Adult Cystic Fibrosis Centre, Belfast, United Kingdom Introduction: CFTR-related disease (CFTR-RD) represents clinical entities associated with CFTR dysfunction that do not fulfil diagnostic criteria for CF. Distinguishing between CFTR-RD and CF may require technologies unavailable to clinicians outside a few reference centres. We sought to compare individuals with intermediate sweat chlorides suspected of CFTR-RD to individuals with H-F508del CF to determine disease trajectory and complication rates.Methods: Patients with suspected CFTR-RD are followed at a specialized clinic at the Northern Ireland Centre for Cystic Fibrosis. Patients are defined as having CFTR-RD by the following criteria; a relevant phenotype, sweat chloride values between 30-60 mmol/L and 0-1 mutation associated with CF but not meeting diagnostic criteria for CF. Patient records were reviewed, complications for the last five years documented and compared to H-F508del cohort.Results: Patients (n=29; 62% male) with CFTR-RD were compared to 80 individuals with H-F508del CF (61% male). Patients were identified by the following; bronchiectasis 12 (43%), recurrent chest infections 6 (21%), genetic screening and infertility 9 (31%), and 1 (3%) for each of nasal polyposis and recurrent pancreatitis. Thirteen (45%) had a single F508del mutation, 6 (21%) another single mutation and 10 (34%) no mutations. Patients with CFTR-RD attained higher height (173 vs 166 cm, p=0.05) and BMI Powers, C.A. 1 ; McColley, S. 1 ; Delute, A. 1 ; Healy, E. 1 ; Jain, M. 2, 1 1. Children's Memorial Hospital, Chicago, IL, USA; 2. Northwestern Memorial Hospital, Chicago, IL, USA Objective: To measure patient and family awareness, preferences and barriers influencing clinical trial participation.

Methods: A total of 221 surveys were sent out to both adult patients (n=94) and parents of children (n=137) followed at our CF Center to assess their awareness and concerns with participation in clinical trials.Results: We received 88 (40%) responses from 57 parents (42%) and 31 patients (33%, 23 female, 8 male, mean age 29 years). Of these, 98% (56) of parents and 87% (27) of patients responded that they had received information and/or have been asked to participate in a clinical trial; 68% (21) of patients and 65% (37) of parents (on behalf of their child) stated they had participated in a clinical trial. The percentage of subjects who reported participating in a clinical trial was higher than the percentage actually enrolled.The top factors important when deciding whether or not to participate in a clinical trial included: the desire to help find a cure or better treatment (81%), understanding both the potential benefits and risks (45%), and the number and frequency of study visits required (41%). Patients valued compensation for time and travel (55%).The main reasons given for declining participation in a clinical trial included: concerns about the time commitment (42%), possible harm to health (22%), and possible side effects (20%). In addition, the need to stop a medication currently used to maintain health was a concern.Many preferred to learn about opportunities to participate in clinical trials from their CF doctor (80%) or the research coordinator (48%). In addition, 60% would like to receive updates on CF and research opportunities via e-mail, and 77% provided their e-mail addresses. Few of those responding relied on other venues (i.e., center newsletters or website) for information.Discussion: Over-reporting of participation in clinical trials may have occurred due to confusion with the large number of observational research studies conducted at our center (i.e., the CFF registry, EPIC). In addition, those who have actually participated in clinical trials may have been more likely to reply to the survey.As investigators and caregivers, CF physicians want to avoid a conflict of interest, but these results clearly indicate that patients and parents wish to hear about clinical trials directly from their doctor. We will continue our current practice which is to have the CF doctor discuss current trials with the patients in clinic, with the research coordinator following up to review the study details and participate in the informed consent process. In addition, patients and families prefer hearing about CF and research opportunities via e-mail; this should be utilized to disseminate information while adhering to regulatory and privacy standards. Distributing educational information such as newsletters or website links via e-mail also may increase the utilization and effectiveness of these resources.Conclusion: Patient and family concerns should be taken into account when designing a CF clinical research program. We believe that incorporating this feedback into practice will further improve recruitment into clinical trials.(25.2 vs 21.0 kg/m2, p<0.001). At baseline individuals with CFTR-RD had a higher percent predicted FEV 1 than H-F508del individuals (91 vs 59%, p<0.0001) but both groups deteriorated over time (-0.6 vs -1.32%/year, p=0.23) . Rate of pulmonary function decline did not differ in individuals with CFTR-RD by presenting complaint. CFTR-RD patients were more likely to have chronic respiratory track cultures positive for Haemophilus influenzae (OR 3.0) , only methicillin sensitive Staphylococcus aureus (OR 2.3) , or grow no specific pathogen (OR 7.6) and none were chronically infected with Pseudomonas aeruginosa or Burkholderia cepacia complex. Six patients did have transient positive cultures for P. aeruginosa which cleared with therapies of inhaled colistin and/or oral ciprofloxacin. There was an increased incidence of pancreatitis (17 vs 0%), asthma (28 vs 3%) and gastro-eosophageal reflux disease (24 vs 5%) in individuals with CFTR-RD relative to H-F508del. Complications such as osteopeneia (21 vs 27%), osteoporosis (30 vs 26%) and cancer (3 vs 0%) did not differ between groups. There was less complications of diabetes (0 vs 12%) and liver disease (0 vs 11%) in individuals with CFTR-RD. One patient developed pancreatic insufficiency after nine years of follow up confirmed by fecal elastase (<150 mg/g) and was subsequently re-classified as having CF.Conclusions: CFTR-RD represents a diverse collection of clinical phenotypes. While suffering a milder disease trajectory then individuals with CF, these patients do experience similar complications and a declining respiratory status albeit at a lower rate. Follow up at a CF centre remains warranted. Lack of dedicated structures seriously limits clinical research in the field of CF in Italy. Our CF Center in Verona (Italy) shared the same limitations: approximately 800 patients were followed up clinically, but enrolment in clinical trials was poor until a significant change was brought about by our decision to create a professional team fully dedicated to clinical research.

Our CF Clinical Research Center (CFCRC) was started in 2007 with its mission being to develop investigational methods and to conduct therapeutic trials and observational studies, in line with the aims of the Cystic Fibrosis Therapeutics Development Network (CF-TDN).The team, led by a physician, comprised two full-time pharmacistswith previous experience in clinical trials, acting as Research Coordinators (RC) -one full-time laboratory technician and another physician, part-time.Preliminarily, the whole team underwent Good Clinical Practice (GCP) training. Each member was then trained in the Standard Operating Procedures (SOPs) specific to his/her area of competence: RCs assembled the documentation to be submitted to our hospital Investigational Review Board (IRB) and managed the approval process; they then organized the clinical trials and managed the logistics (clinical supplies, recording materials, etc.). RCs were also responsible for patient pre-screening, for ensuring protocol adherence and inputting data onto eCRF. A data manager was involved, on an as-needed basis, to design database and entry forms for each study.The team was successful in developing and validating nasal potential difference (NPD) measurement methodology for both diagnostic and experimental application.Achievements: From April 2008 to December 2009, the Center participated in 7 RCT and 5 observational studies enrolling a total of 68 patients in RCTs and more than 400 patients in observational studies (Table) .Costs: Total costs for the CF Center were about 150,000 US $/year, mostly supported by the local patients association (Lega Italiana Fibrosi Cistica-Associazione Veneta ONLUS), and partially by a 3-year grant of the North American CF Foundation.Conclusions: Having a dedicated team for clinical research led to impressive improvements. Patient acceptance was very high. Our experience shows that the main limitations were of organizational nature dispelling our original suspicion that enrolment was hampered by scepticism that prevented patients to volunteer. Investing in such infrastructure is essential to improve the quality of clinical research and the recruitment capacity of CF centers.Immediately after its formation, the CF Parent Advisory Board (CF PAB) at this pediatric CF Center set a priority goal to create a local, independent, evidence-based and reliable website for parents of children with CF. Initially, the CF PAB launched a basic website with low cost selfdesignware, then obtained external funding for professional website design and increased functionality.Concurrent with the initial website, surveys and focus groups identified parents' interest in receiving frequent, short, electronic updates or reminders about CF nutrition care, especially suggestions or recipes from other CF parents. CF parents were invited to join a weekly "CF Nutrition Tip of the Week" ("Tips") email list hosted by the CF Center dietitian. Subject matter for "Tips" was generated by interviews with parents or patients in CF clinic, research, or CF clinical staff suggestions.Approximately one year after the initiation of email "Tips," the CF PAB website upgraded to include a newsletter function and "Tips" migrated to the website. With the migration, "Tips" expanded beyond nutrition to include more universal CF care. At present, weekly "Tips" are accompanied by a recipe and each is archived on the CF PAB website.Approximately 200 families (representing 250 children with CF) receive CF care at this center; 124 email addresses are registered on the CF PAB website for the newsletter. Eighteen individuals (14.5% of "Tips" recipients) responded anonymously to an electronic survey regarding preferences for future "Tip" topics. All child age groups were represented, but more parents of school age children responded (7 of 18) than other age groups. Of 18 respondents, 15 (83%) had read 10 or more "Tips" and 11 of 18 (61%) felt "many" or "most" contained useful information while 7 of 18 (39%) found "some" or "a few" of the "Tips" to be useful. Respondents indicated their

Crumb, T.L. 1 ; Bonnema, H. 1 ; Symington, T. 1 ; Peterson, M. 1 ; Millard, S. 2 ; Schuen, J. 2 ; Fitch, S. 3 ; McClelland, M. 3 1. Research Department, Helen DeVos Children's Hospital at Spectrum Health, Grand Rapids, MI, USA; 2. Pediatrics, Helen DeVos Children's Hospital, Grand Rapids, MI, USA; 3. Pulmonary and Critical Care, Spectrum Health Hospitals, Grand Rapids, MI, USA Background: CF Registry Data reveals a low percentage of patient participation in clinical research trials. As a Therapeutic Development Network Center, we reviewed our performance on clinical trials. In an effort to improve research at Helen DeVos Children's Hospital and Spectrum Health Hospital CF Centers, we developed strategies to increase our Education, Efficiency and Enrollment.Objective: To increase enrollment and exposure to research at our CF centers.Methods: Focusing on Education and Efficiency, we utilized the following methods:Education: 1. Our CF Research Director presents a lecture on research participation and we have a research booth at our annual CF Family Education Day. We hire videography and share the DVDs with all CF patients/families.2. We mail out a bi-annual newsletter focusing on research topics. 3. We hold research education meetings with physicians and staff. 4. Regular attendance in Adult clinic to introduce research, identify subjects and encourage participation.Efficiency: 1. We schedule regular 15/30 min blocks of time with physicians for ongoing review of subject data and regulatory documents.2. We blocked research time slots within the clinic schedule for easy access to appointments to target visit windows and avoid over booking.3. We implemented a new team approach to enrollment. Each new enrolled patient is assigned to a single coordinator, which we refer to as the patient's primary coordinator. 4. A weekly list of patients is sent to the CF team, identifying them as potential or planned for research visits.5. Earlier Port CF query to promote faster enrollment in upcoming studies.6. New collaboration with other CF centers in Michigan. 7. Improved systems for processing outside referrals of potential subjects.Results: After review of our research database: Enrollment -Our plan led to a large increase in our adult patient enrollment. This stimulated further interest of the Adult CF Center providers in clinical trials. In 2009, 91% of our study patients were new to research. In 2010, we continue to match that new patient enrollment rate. We have become active as a referral center for research patients (self/private office/CF center). We successfully approached and enrolled patients outside of our center population.Conclusion: We implemented a diversified approach to improving our CF Center's role in clinical research. We noted a trend that patients are seeking out participation in new studies. We speculate this is due to increased awareness of the need for participation through efforts from the CFF and our focus on education.Discussion: The one key element in each of the "three E's" is the partnership that has been established at our site. There is a partnership between the research staff and the clinical staff, between the research staff and the physicians, between the CF team and patients/families, and between the CF team and the institution. This comprehensive and unified approach to expand research is woven into every strategy and has resulted in our success.

After discussion with CF physicians, we developed a protocol to assess hepatitis immunizations. First, a letter was sent to families explaining the project. Second, families provided immunization records or signed medical releases for these. Third, immunization records were reviewed to determine patient immune status to HepA/HepB. Patients who received 2 previous HepA shots were considered protected. HepA shots were given to patients who needed complete vaccine courses. Patients who previously completed full courses of HepB shots had quantitative HepB surface antibody (anti-HBs) levels measured. Anti-HBs values >10 IU/mL indicate protective immunity to HepB, values <8 IU/mL indicate inadequate immunity, and values of 8-10 IU/mL are equivocal. Patients with anti-HBs <8 IU/mL received a booster HepB shot, and anti-HBs levels were repeated at least 1 month later. Patients with anti-HBs levels <8 IU/mL after booster completed a full HepB vaccine course, and if anti-HBs levels remained <8 IU/mL, they were considered vaccine non-responders.Results: Through review of our CF clinical database, that includes 310 pediatric patients, we found 41 pediatric patients on ursodiol for presumed CFALD. In the first 4 weeks of this ongoing project, 20 children (mean age 14 yrs, range 8-20) were screened for HepA/HepB immunity. One child had completed a full HepA vaccine course at screening. Eighteen (90%) had completed a full HepB course at screening; 5 (25%) are protected against HepB, 14 (70%) have inadequate immunity, and 1 (5%) has equivocal immunity. At time of abstract, only 1 child received post-booster anti-HBs levels, with a protected result. All patients have negative HepC antibody.Conclusion: In one CF center, children with presumed CFALD have low rates of protection against HepA/HepB. Only 1 child has received HepA vaccination, and the rest are unprotected against this virus. While the accepted non-responder rate to HepB vaccination is 5-10%, our HepB non-responder rate is higher despite good HepB vaccine compliance. With HepB boosters, these children may have increases in anti-HBs levels. With the increased risk of morbidity and mortality associated with preventable viral hepatitis in patients with existing liver disease, CF treatment protocols should include routine assessment of HepA/HepB vaccine status in CF children, particularly in those with presumed CFALD. Future expansion of this protocol to CF patients with repeat liver enzyme elevations, or to all CF patients in general, may be beneficial. Introduction: Parenteral antibiotic therapy remains the mainstay of management for CF pulmonary exacerbation which often requires hospitalization. This can be disruptive to patients and their families and is frequently undesired or even refused. Oral antimicrobials are recommended only for "mild" CF pulmonary exacerbations, but are often requested by families and patients. Our experience suggests that distinct patient characteristics are associated with failure of oral therapy leading to eventual parenteral treatment. We hypothesized that we could identify potential patient characteristics associated with oral antimicrobial failure by comparing cases of failed versus successful oral therapy management for CF pulmonary exacerbations seen in the CF clinic.

Methods: We retrospectively reviewed all the clinical encounters of 113 non-transplant patients with a confirmed diagnosis of CF, aged 6 to 21 years, who were evaluated by our CF center team over a four month period. Geography dictates that telemedicine is an essential practice at our center so we included phone "encounters" in which oral antibiotics were prescribed if the phone encounter was preceded or followed by at least one in-person clinical evaluation. After abstracting data such that all unique patient identifiers were removed, odds ratios (OR) were defined for several, readily available patient characteristics suspected of contributing to oral antibiotic failure using 2x2 contingency tables.Results: Sixty-two unique pulmonary exacerbations were identified from the encounters that we reviewed; 23 of these had no resolution of their pulmonary symptoms after oral therapy and went on to parenteral therapy, and were deemed oral antibiotic failures. By comparing patient characteristics between the 23 cases of failure and 39 successes, we found FEV1% predicted <75% (OR: 3.75, 95% CI of 1.285-10.929), FEV1% predicted drop 507ଙ CYSTIC FIBROSIS ACTION PLAN: CAREGIVERS AND HEALTHCARE TEAM PERSPECTIVE Willis, L.M. Arkansas Children's Hospital, Little Rock, AR, USA Introduction: The Cystic Fibrosis Foundation (CFF) recommends quarterly evaluation at accredited cystic fibrosis (CF) centers to monitor patient stability. The CF team at Arkansas Children's Hospital (ACH) developed and implemented a CF Action plan (CFAP) tool as part of routine clinic visits to help communicate health metrics to patients and caregivers. The CFAP lists patient's height, weight, BMI%, and FEV1% predicted, as applicable, at each clinic visit. It also provides information on airway clearance therapy, important reminders, and lists changes or new recommendations resulting from the clinic visit. After one year of use, we assessed caregiver/patient and CF healthcare team satisfaction with the implementation of the CFAP.Methods: This study was designed as a quality improvement initiative at the ACH CF Center (Little Rock, AR). Paper surveys, a letter of explanation, and a blank reference CFAP were given to caregivers (or patients if 18 years old) to complete during routine clinic appointments. A web-based survey was completed by CF team members. Both survey results were collected anonymously.

Caregiver Responses: Ninety-six caregiver responses were collected (one per family) from the 147 families followed at our center (65% return rate). Of these, 73% and 16% of caregivers stated they either always or most of the time received a CFAP at each clinic visit over the past year; 73% would like to continue to receive a CFAP at each clinic visit. Eighty-five percent strongly agreed and 14% agreed the CFAP was easy to read; 77% strongly agreed and 23% agreed the CFAP was useful; 87% strongly agreed and 13% agreed they understood the purpose of the tool; 75% strongly agreed and 21% agreed they refer back to the CFAP after clinic visits; 74% strongly agreed and 24% agreed the CFAP helped them better understand nutrition and lung status, respectively; 54% strongly agreed and 32% agreed the CFAP helped them take their medications, while 64% strongly agreed and 25% agreed the CFAP motivated them to take better care of themselves/their child.Team responses: Seventeen of 30 CF team members completed a survey (57% return rate). Forty-one percent and 59% of the CF team stated they either always or most of the time gave CFAPs to patients at each clinic visit over the past year; 71% felt CFAPs should continue to be given at each visit; 59% strongly agreed and 41% agreed the CFAP is easy to read; 53% strongly agreed and 29% agreed the CFAP was easy to complete; 35% strongly agreed and 47% agreed the CFAP was easy to incorporate into practice; 35% strongly agreed and 53% agreed the CFAP helped patients/caregivers understand lung status, while 47% strongly agreed and 53% agreed it helped understand nutrition status. Forty-one percent strongly agreed and 47% agreed the CFAP was overall a useful tool.Conclusion: CFAP is a useful tool to be used at each clinic visit. Both caregivers and the CF healthcare team expressed satisfaction with the implementation and use of the CFAP. Minor changes will be incorporated into the CFAP based on caregivers' and healthcare team suggestions.

Shapiro, A.J.; Leigh, M.; Dellon, E.P. Pediatrics, University of North Carolina, Chapel Hill, NC, USA Hepatitis A and Hepatitis B (HepA/HepB) infections are worldwide health problems. People with chronic liver disease, like cystic fibrosis associated liver disease (CFALD), are at increased risk of morbidity and mortality from these infections. HepA/HepB infections are preventable through universally recommended childhood vaccines. However, the HepA vaccine was not universally recommended until 2005, and pediatric HepA vaccine compliance is below 50% in many states. The HepB vaccine has good pediatric compliance rates, however research shows HepB antibody protection can diminish to unprotected levels 15 yrs after initial immunization.Objective: To ensure CF patients with liver disease are appropriately immunized with HepA/HepB vaccines. MANAGEMENT TABLE FOR CARE OF THE  INFANT WITH CF, UNCERTAIN DIAGNOSIS, OR 3. Identify & address barriers that prevent effective use of the Management Table. Method: Charts were reviewed for patients with CF, 0-24 months of age, who were seen in clinic from 11/1/09 to 12/31/09. The following data were obtained:

1. Presence of the Management Table. 2. Whether respiratory culture was a) due at that visit, b) done at that visit.3. Whether vitamin levels were a) due at that visit, b) done at that visit. A questionnaire regarding perceived barriers to usage of the Management 1 1. Health Systems Management, Tulane University, New Orleans, LA, USA; 2. Analysis Group Inc, Boston, MA, USA; 3. Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA Background: Between 1986 and 1994, 9. 3% of children with cystic fibrosis (CF) were continuously enrolled in Medicaid while 66.1% were never on Medicaid (Schechter et al. 2001 ) with very few uninsured. The State Children's Health Insurance Program (SCHIP) introduction and Medicaid expansions in late 1990s may have crowded out private insurance and affected healthcare utilization and health outcomes. The recent Patient Protection and Affordable Care Act will expand Medicaid coverage and more CF patients will likely shift to Medicaid.Objective: This study aimed to document children with CF who switched from private plans to Medicaid from 1996 to 2008 and to analyze the effects of this change on healthcare utilization and health outcomes of children with CF in comparison to those continuously enrolled in private plans.Methods: The sample was drawn from 19,899 CF children under age 20 years who were continuously observed in the National Cystic Fibrosis Foundation Patient Registry (NCFPR: 1996 (NCFPR: -2008 . Healthcare utilization was measured by hospitalization, home intravenous (IV) therapy use, number of outpatient visits and clinical encounters. Health outcomes were measured by forced expiratory volume in one second (FEV1) and mortality in subsequent year. We used fixed-effect logistic and negative binomial regressions and controlled for previous-period utilization and health outcomes.Results: Between 1996 and 2008, 27% of sampled CF children were continuously covered by Medicaid versus 38.6% who were never on Medicaid and only 0.5% percent continuously uninsured. Out of 12,987 children covered by private plans when first observed, 29% percent later switched to Medicaid for some period (38.3%) or switched permanently (61.7%) and 512 deaths occurred during a median follow-up of 7 years. CF children who switched to Medicaid had lower hospitalization rates than those who remained on private plans (odds ratio [OR] 0.868, 95% confidence interval [CI]:0.784-0.962, P=0.007) after adjusting for prior year FEV1, age, and year fixed effects. No differences were found in other types of utilization. Moreover, the switch to Medicaid had no immediate effect on the probability of having a normal FEV1 (OR: 0.922,CI: 0.693-1.226, P=0.576) or of dying (OR: 1.054, CI: 0.795-1.396, P=0.717).Conclusion: A reduction in hospitalization rates was found among the CF children who switched to Medicaid. Further research of long-term effects on healthcare utilization and health outcomes is needed along with other types of healthcare services utilization, quality of care and health outcomes, such as type of drug treatment, being underweight, frequency of infection and complications. In 2004 we adopted a standard definition for a Pulmonary Exacerbation (PEx), the Pulmonary Exacerbation Score (PES). The details of this assessment tool were previously described. Its use is associated with improved lung function at our center. Currently, there are no formal guidelines for the comprehensive treatment of a PEx that include route of antibiotics, treatment duration, and follow-up. We present a comprehensive, standardized treatment algorithm (PExT) that uses the PES to identify patients experiencing a PEx, determine severity, guide treatment strategy and duration to ensure complete PEx resolution.Methods: PES thresholds were determined for the PExT from current practice by reviewing treatment strategies for all patients with a PEx from 6/1-12/1/2009. The PExT algorithm was developed by defining PES thresholds for PEx severity: mild, moderate, and severe. Severity subsequently Introduction: Vitamin D deficiency is highly prevalent among CF centers in the USA and the world. We have observed this same finding at our CF center with a large percentage of our population being identified as vitamin D deficient. The current treatment paradigm for vitamin D deficiency is flawed in that we wait for vitamin D levels to become low before treating instead of focusing on prevention. We believe that the focus should be moved to the prevention of vitamin D deficiency, especially in the CF population where it is well established that vitamin D deficiency is very common.Methods: From November through December 2009 all patients that came in for appointments at the Emory Clinic outpatient CF program were offered high dose vitamin D3 (regardless of previous vitamin D status). The dose of the vitamin D3 was 50,000 IU capsules to be taken once per week for a 12 week period at home during the winter months. Fifty-one patients participated in this QI project.Results: At the end of winter, we found that only 32% of our patients were found to have vitamin D deficiency (25(OH)D <30 ng/mL). This represents a dramatic improvement where 76% of our patients in the past were reported to have a vitamin D deficiency (1) .Conclusions: Based on this preliminary data, we feel that supplementing patients with a high dose of vitamin D3 during the winter months will help in preventing the natural seasonal decline of vitamin D levels in the CF population. Since not all of our patients were sufficient at the end of winter, a higher dose or prolonged course of vitamin D may be necessary. We are optimistic that a wintertime supplementation strategy may translate to optimal vitamin D status year round. The mucosal barrier of the gut is important to protect the body from the microbes that inhabit the intestines and the proinflammatory molecules they release. Transepithelial permeability, controlled by the junctional complexes between epithelial cells, is a major factor of the barrier. Another more recently recognized factor is that intestinal alkaline phosphatase (IAP) detoxifies its natural substrate endotoxin (LPS). Loss of gut barrier function can allow LPS access to the circulation which may contribute to inflammation at distant sites in the body, such as the lung. In CF there is an excessive immune response to infection of the airways, and the airway damage that ensues is the eventual cause of mortality in CF. To assess intestinal mucosal barrier function in CF and its potential relationship to airway inflammation, we are using the Cftr knockout mouse model which has severe intestinal dysfunction, including small intestinal bacterial overgrowth.Methods: Cftr tm1UNC (CF) and wildtype (WT) littermate C57BL/6 mice were used. Three groups of WT and CF mice were used: (1) maintained on a liquid diet (Peptamen) as controls; (2) fed solid chow and given oral laxative to eradicate small intestinal bacterial overgrowth; and (3) treated with oral broad spectrum antibiotics (ciprofloxacin and metronidazole) to reduce bacterial load in the intestine. Intestinal permeability was assessed by the passage of gavaged fluorescent-dextran into the blood, and levels of serum protein in the gut lumen. IAP was investigated by histochemistry, by qRT-PCR measurement of mRNA for mouse IAP (Akp3), and by enzyme activity. Lung inflammation was assessed by histology, by immunohistochemistry for neutrophils, and by myeloperoxidase (MPO) activity.All patients reported a greater freedom to eat high fat foods without adverse events.Conclusion: In PS patients with CF who experience adverse bowel symptoms a trial period of low dose PERT should be considered. In our 8 patients it markedly improved bowel symptoms. Interestingly introduction of PERT resulted in weight loss in patients with BMIs 〉25 and weight gain or maintenance for patients with normal or low BMIs.Anecdotally, patients reported a greater freedom to eat without adverse bowel symptoms. Objective: Liver disease affects 20-50% of individuals with cystic fibrosis (CF). Cirrhotic liver disease with or without portal hypertension, affects approximately 7% of individuals, who may require liver transplantation. Several studies have examined survival for CF patients who have undergone liver transplantation, as well as post-transplant lung function which was improved after liver transplant. However, many of these studies are small single center experiences with a limited number of patients. The aim of our study was to examine the lung function of individuals in the U.S. CF Registry who had undergone an isolated liver transplant between 1989 and 2007, compared to CF patients with no diagnosis of liver disease and who had not undergone a liver transplant.

Methods: We performed a retrospective analysis of CF Foundation Patient Registry data, using mixed effect models to determine lung function in CF patients with liver transplant vs. those without transplant (controls) matched for age, gender, pancreatic status (defined by CFTR mutation class and pancreatic enzyme use), and Pseudomonas and B. cepacia status. Cases were matched to controls at age of transplant. We examined the change in FEV 1 (percent predicted) for 3 years prior and post age at matching. We compared the FEV 1 slope between cases and controls 3 years prior to transplant, and 3 years post-transplant. We also compared FEV 1 slope pre-vs. post-transplant in both cases and controls separately.Results: Between 1989 and 2007, there were 193 CF patients with isolated liver transplant (approximately 60% male; mean age 15.5 years, SD: 7.28, range: 1.0 -39.6 years). The median age at CF diagnosis was 0.5 years (range: 0 -30.3 years). CFTR genotype was high risk in 64.3%, low risk in 0.5% and unknown type in 35.2%. A total of 72.0% of patients had a positive Pseudomonas aeruginosa culture and 4.7% a positive Burkholderia cepacia culture in the past 3 years. Liver transplant cases had significantly lower percent predicted FEV 1 3 years pre-transplant compared to controls (71.0 vs. 75.4, p=0.012). Liver transplant cases had a significantly slower decline in FEV 1 (0.05%/year) compared to controls (-1.35%/year) in the pre-transplant period (p = 0.001). There was no difference in FEV 1 decline in the post-transplant period in cases (-1.53%/year) vs. controls (-2.20%/year, p = 0.125). Both cases and controls had a significantly slower decline in the pre-transplant (cases: -0.05%/year; controls: -1.35%/year) compared to the post-transplant period (cases:-1.53%/year; controls: -2.20%/year; p-value for difference: cases = 0.002; controls < 0.0001).Conclusions: CF patients receiving a liver transplant have worse lung function 3-years pre-transplant but a slower rate of decline in lung function, compared to controls. Liver-transplanted patients have similar 3-year posttransplant lung function and rate of decline of lung function, compared to matched controls. The slower rate of decline in lung function pre-transplant is a novel finding and merits further exploration. , an FDA-approved pancrelipase formulation with 100% labeled lipase content without overage, has previously been shown to significantly improve steatorrhea (CFA) in a randomized, double-blind, placebo-controlled, crossover study of 34 patients with exocrine pancreatic insufficiency (EPI) due to CF.

Objective: To follow symptom improvement with EUR-1008 after switching patients from their previous FDA-unapproved formulations of other pancreatic enzyme products (PEPs [e.g., Creon ® , Ultrase ® , Pancrecarb ® ]) in a post hoc analysis of 31 evaluable patients. In a subgroup analysis, symptom improvement was compared in patients previously taking, or not taking, proton pump inhibitors (PPIs)/H 2 receptor antagonists.Methods: Enrolled patients discontinued PPIs/H 2 antagonists and were switched from unapproved PEPs to EUR-1008 during a dose-titration and stabilization period according to symptom control of EPI for 6-9 days prior to randomization (EUR-1008 or placebo). A total symptom score index (TSI) was calculated from patient-reported signs and symptoms of stool consistency, bloating, flatulence, pain, and visible oil in stool.Results: At comparable doses, patients experienced significant and rapid symptom improvement during the stabilization period when switched from previous PEPs (mean dose 5,100 U lipase/kg/day) to EUR-1008 (mean dose 4,600 U lipase/kg/day). Mean TSI decreased significantly from 11.2 (±15.8) on previous PEPs to 4.3 (±4.0) after 7 days of EUR-1008 treatment (P<0.004). Symptom improvement was observed as early as day 2 and was significant at day 4 (P=0.017). The variability in the severity of symptoms was markedly decreased during EUR-1008 treatment. Twenty of 34 (58%) patients were taking PPIs or H 2 antagonists prior to enrollment; there was no significant impact on symptom improvement during the dose-titration and stabilization period based on previous exposure to PPIs/H 2 antagonists.Conclusions: EUR-1008, tested as a single agent, provided enhanced EPI symptom control after patients were switched from unapproved PEPs, without additional reliance on PPIs/H 2 antagonists, suggesting the potential for reduction of costs and pill burden.Sponsored by Eurand Pharmaceuticals, Inc., Yardley, PA, USA. It has been shown that many children with cystic fibrosis (CF) have low vitamin D levels (25-OHD) despite routine supplementation with multi-vitamins specially formulated for cystic fibrosis. Hence, the CF Foundation recommends the use of 50,000 IU vitamin D2 once per week for 8 weeks for children ≥5 years and 12,000 IU once per week for 8 weeks for children <5 years whose 25-OHD level <30 ng/mL. Some studies have suggested that this regimen is inadequate to normalize vitamin D levels. It has been proposed that high dose Vitamin D supplementation daily for 4 weeks would be more effective in normalizing vitamin D levels. As a quality improvement project, we developed a protocol to monitor vitamin D levels and treat children with vitamin D deficiency.

Objectives: Determine the prevalence of vitamin D deficiency in our patients with CF and evaluate the effectiveness of protocol based treatment in normalizing vitamin D levels.Methods: We conducted a retrospective chart review of all the pediatric CF patients who had their vitamin D level checked between Jan 01, 2010 Aim: With improved clinical outcomes in CF patients and development of CFTR-correctors novel biomarkers are necessary that detect subtle changes in CF related disease processes. The current study seeks to identify serum markers that differ between CF and non-CF disease control children.Method: Serum was obtained in a standardized manner from 31 CF and non-CF children matched by age and gender. Serum was stored frozen until processing. Samples were extracted to remove protein and dislodge any small molecules in the precipitant. Metabolic analyses were performed using GC-MS, LC+MS/MS and LC-MS/MS for fragmentation and mass determination. Targets were identified using a proprietary software and library (Metabolon®). Statistical analyses included group-wise and paired comparison.Results: Mean age of both groups was 9.6 ± 1 years, gender was 15 M/16 F. The main problems in Non-CF patients were cough, asthma, and lower respiratory infections. Only 1/31 children was on antibiotics at the time of serum collection and 25/31 had BMI% >25%. Lung disease in CF patients was mild to moderate (FEV 1 70 ± 5% measured in 20/31 subjects), genotype included 7 hetero-and 22 homozygotes for ∆F508.Fourteen patients were on antibiotics at time of serum collection. Nutritional status, as measured by BMI%, was >50% in 17, 25-50% in 9 and < 5% in 5 patients. Of note, 15/31 CF and 26/31 non-CF subjects were fasting (NPO 6-10 hours) at time of blood draw.Metabolic markers that differed between the two groups fell into 4 categories: 1) Decreased fatty acid β-oxidation and altered metabolism of branched chain amino acids. As a group, CF patients had lower levels of some ketone bodies deriving from metabolism of fatty acids and decreased products of amino acid metabolism. These findings could indicate overall worse nutritional status in CF or be a reflection of acute worsening of disease as 23/31 patients were sampled during an exacerbation. 2) Alteration in lipid profiles; and 3) Elevation of markers of oxidative stress. Such changes in lipid profile (e.g. essential fatty acid docohexaenoate = DHA) and increased oxidative stress (increased degradation products of glutathione and decreased γ-tocopherol) have been described in CF and thus confirm the validity of our findings. 4) Finally, metabolites known to originate from bacteria in the gut were elevated or decreased compared to the non-CF group; many of these bacteria were anaerobic bacteria. This finding may be related to ongoing infection, bacterial intestinal overgrowth or alterations in gut microbiology secondary to antibiotics.Discussion: These data provide a first indication of sensitive metabolomic biomarkers and show measurable differences in serum obtained from children with CF compared to age and gender matched controls. Overall metabolic rate is known to be increased in CF, however specific alterations in various metabolic pathways need to be further explored and may be also helpful in longitudinal analyses and following CFTR specific changes.Supported by Cystic Fibrosis Foundation Therapeutics, Inc. M. Muhlebach and L. Guo contributed equally to this work. predictors of LF%, UAMAZ and CHOL predicted MF% (F=5.97, R 2 =.37, p=0.008), and age, male gender, %BF, CHOL, and MF% strongly predicted MC% (F=13.4, R 2 =0.79, p<.001).

Conclusion: Compared to healthy matched children, subjects with CF and PI had significantly higher liver fat and lower calf muscle choline content after controlling for age and muscle fat. In subjects with CF, better nutritional status (%BF, UAMAZ) and higher serum cholesterol were associated with higher muscle fat and choline content, suggesting better dietary fat intake and absorption. Support: NIDDK (R44DK060302), CTRC (UL1RR024134) and Nutrition Center at CHOP, GCRC (M01RR000847) at UVA Health Systems and Avanti Polar Lipids Inc. Background: Patients with cystic fibrosis (CF) have impaired capacity to neutralize gastric acid due to decreased bicarbonate output from the pancreas, biliary tree and intestinal mucosa. An acidic gastrointestinal tract may compromise fat absorption in patients with CF. We hypothesized that a wireless motility capsule (WMC) could demonstrate reproducible pH profiles in patients with CF that distinguish them from matched controls.

Methods: A single-site, pilot study was conducted employing a convenience sample of 10 adult subjects with CF and 10 healthy subjects matched by age, gender and body mass index. Eligibility criteria specified that subjects could not be taking acid-suppressing medicines. Subjects on cycled inhaled antibiotics were mid-cycle during study periods. Eligible subjects ate a standardized meal bar (255 kcal), swallowed the WMC, and then went home with a receiver that stores data from the WMC. When WMC passed the subjects mailed the receiver back to the clinic where data was downloaded. All subjects repeated the study under the same conditions one month later. The area under the curve (AUC) of the pH profile in the small bowel was calculated for 5 minute increments from WMC gastric emptying (indicated by an abrupt rise in pH of 3 or greater units) until 60 minutes within a small bowel. pH AUC defines buffering of gastric acid in the proximal small bowel. Data points not captured by a receiver over a 5 minute period were excluded from the analysis. Data were analyzed using GIMS Data Viewer, Prism GraphPad and SAS version 9.1.3 statistical software.Results: Ten subjects (7 F) were recruited in each group. There was no difference between CF subjects and controls for age (mean=21.6 years vs. 22.5) or mean BMI (23 vs. 23.2). There was a statistically significant difference between CF subjects and controls for AUC of pH profile in the first hour after gastric emptying (p=0.03), with the most dramatic differences seen in the first 15 minutes (p=0.003). Analysis of small bowel 5 min pH AUC profiles did not demonstrate reproducibility in CF or control subjects.Conclusion: There were clear differences between subjects with CF and healthy controls in pH AUC in the first 15 minutes after gastric emptying, a critical time for microencapsulated enzyme dissolution and mixing of bile salts with digested triglycerides. Factors that affect reproducibility need to be explored further. WMC technology may be useful to study CF patients with difficult to control fat malabsorption.This study was sponsored by Cystic Fibrosis Foundation Therapeutics.Background: Pancreatic disease begins in utero in the majority of patients with cystic fibrosis (CF) and progresses over time to complete destruction of the organ. Although inflammatory cells have been welldescribed in the pancreas of humans with CF and the CFTR-/-pig model, it is not known whether inflammation plays a role in the destruction of pancreas in CF.Hypothesis: Understanding the immune response in CF pig pancreas may offer clues to disease pathogenesis.Methods: Using four-color flow cytometry, we analyzed the surface phenotype of leukocytes in the pancreas, blood and mesenteric lymph nodes (MLNs) of CF pigs (4 CFTR-/-, 1 CFTR∆F508/ ∆F508) and WT and heterozygous pigs (3 CFTR+/-, 3 CFTR+/+, 1 CFTR+/∆F508) immediately after birth. Cell suspensions were incubated with primary monoclonal antibodies against CD2, CD3, CD4, CD8, CD14, CD21, CD25, IgM, TCR γδ SWC1, SWC7, SWC8 and MHC-II antigens. Following incubation with secondary antibodies, measurements were made using a FACS Calibur flow cytometer. Microarray expression profiling was done to explore the differences in gene transcription between CF (4 CFTR-/-), WT (4 CFTR+/+) and heterozygous (4 CFTR+/-) pig pancreata using an Affymetrix Porcine GeneChip.Results: Leukocyte cell populations and microarray gene expression were not different between heterozygous and WT pig pancreas. However, CF pig pancreas had an increased proportion of B cells (IgM+); effector (MHC-II+) and cytotoxic (CD2+CD8+) γδ T cells; activated (MHC-II+ and/or CD25+) and effector (CD4+CD8+) αβ T helper cells; effector natural killer cells (MHC-II+CD3-CD8+); monocytes/macrophages (SWC8-CD14+) and neutrophils (SWC8+CD14+) compared to pigs without CF. Flow cytometry from blood and MLNs did not indicate that leukocyte populations differ between CF and non-CF pigs. Microarray gene expression profiling of newborn pigs showed significant upregulation of proinflammatory and complement pathway genes compared to non-CF pigs.Conclusion: Both the innate and adaptive immune systems were activated and proinflammatory and complement genes were upregulated in the newborn CF pig pancreas. Blood and MLN inflammatory cell populations were not different between CF and non-CF pigs, suggesting that the pancreatic immune response represents a localized rather than systemic inflammation. An activated immune response contributes to the pathogenesis of pancreatic lesions in CF. In patients with cystic fibrosis (CF), pancreatic involvement is common and over time leads to pancreatic insufficiency in most patients. Pancreatic disease is evident in infants, and lesions have been observed in fetuses. However, the pathophysiological processes that occur in utero are unknown. Investigating this process in humans is not feasible, and CF mice show little or no pancreatic disease before weaning. We have found Cystic fibrosis related diabetes (CFRD) is increasingly recognised as an important complication of CF. Effective screening protocols are essential for early detection of glucose impairment. Screening and early detection may help prevent a decline in weight and/or lung function. The oral glucose tolerance test (OGTT) is the "gold standard" for screening and diagnosis of CFRD and impaired glucose tolerance (IGT), however it is not always logistically feasible for large CF clinics serving large geographical areas. As a result we have used annual random blood glucose levels (BGL) and increased monitoring during admissions for acute exacerbations, as indicators for referral to OGTT. Based on this monitoring, our prevalence of CFRD and IGT in our 10-19 yr age group was 4% and 6% respectively when recently audited, well below previously published results. (Moran et al, 1999) .

Glucose challenge tests (GCT) are routinely performed as a screen for gestational diabetes. A recent paper comparing OGTT to the GCT in adult CF patients found that the GCT could be used effectively as a screen to reduce the number of OGTTs required. (Lee et al, 2007) . The utility of the GCT in the paediatric CF population is unclear. Nine months ago we introduced a modified GCT screening tool on all patients ≥10 yrs performed at clinic during the annual check. BGL were checked at baseline, (patients are non fasted) and at 1 and 2 hours during which time they were fasted, after an oral glucose load of 1.75 mg/kg to max 75g. Any patient with an elevated BGL at baseline and any patient with a BGL reading of ≥ 7.8mmol/L at the 1 or 2 hour reading are referred for OGTT.To date 49 of 66 (74%) patients aged ≥ 10 yrs who have attended clinic for their annual review, were approached for screening. Of those approached, 41 of 49 (84%) completed the GCT, 13 (32%) patients had a normal GCT, 28 (68%) were referred for a formal OGTT (this includes 4 patients who had a baseline elevated BGL). Eight patients did not complete the test, 3 (6%) could not tolerate the glucose drink and 5 (10%) refused the test. OGTT have been performed on 27 of 28 (96%) patients to date. Based on current accepted international criteria, 15 of 27 (55%) were normal, 11 of 27 (41%) had IGT, 1 of 27 (4%) had CFRD.tic partner had disclosed their diagnosis to their partner; 69% of respondents reported telling all their close friends about their diagnosis. Only 15% had told all casual acquaintances, while 28% reported that they had disclosed to no casual acquaintances. Women were significantly more likely to disclose to all casual acquaintances than men (p<0.01). Comfort discussing one's CF experience with close friends was negatively correlated with medication adherence (p=0.02; Table) .Conclusions: The relationship between disclosure and adherence is complex, likely moderated by age, gender, and living circumstances, and warrants further research. Clinicians should consider assessing their patients' comfort with disclosing their diagnosis to and performing treatments in front of various audiences. Background: Perceived social support (PSS) is a known buffer of psychological distress and adverse medical outcomes, including morbidity. Progress in the treatment of CF now allows patients to live into adulthood, wherein the sources of support expand from family to include friends and a significant other. Research has only begun to characterize the sources and predictors of social support among adults with CF.

Objective: The present study explored common demographic factors associated with PSS derived from friends (Fr), family (Fm) and a special person (SP), i.e.,significant other.Methods: Sixty-two individuals with a diagnosis of CF, 8 status post lung transplant, participated in this cross-sectional survey study; response rate was 67%. Participant demographics: 56.5% were men, mean age=32.8 years (range 22-54), 96.8% were Caucasian, and 56.5% had a significant other (SO). With respect to education, participants were distributed as follow: 4.8% high school or less, 22.6% high school diploma/GED, 6.5% vocational school, 22.6% some college, 33.9% college degree,9.7% professional/graduate degree. Participant scores on the three subscales of the Multidimensional Scale of Perceived Social Support (MSPSS), Fr, Fm, and SP were the primary outcome measures. We examined how these subscales were associated with gender, age, SO status and education. All reported results reflected statistically significant (p<.05) Pearson correlation tests or analysis of covariance (ANCOVA).Results: MSPSS Fr was positively correlated with higher education and not having a SO. MSPSS Fm was positively correlated with being female and higher education. MSPSS SP was positively associated with being female, having a SO, and higher education. Age was not associated with MSPSS Fr, Fm or SP. To further clarify the relationship of gender and partner status on PSS, a repeated measures (MSPSS: Fr, Fm, SP) by gender (M vs F) by SO status (yes vs. no) ANCOVA with education as a covariate was conducted. The ANCOVA revealed a significant 3-way interaction such that for: 1) MSPSS Fr, females rated PSS equally whether or not they had a SO, whereas men rated PSS higher than women if SO no and lower than women if SO yes; 2) MSPSS Fm, women rated PSS higher than men overall, irrespective of SO status, whereas men, rated PSS much lower if SO yes, and 3) MSPSS SP, women overall rated PSS higher than men, and overall those with an SO rated PSS higher.Conclusions: MSPSS scores were remarkably similar to those obtained in numerous other healthy and clinical populations, suggesting CF patients are not unique in their perceptions of social support. However, ratings of perceived sources of social support were differentiated: Women rated PSS higher then men and were less influenced by SO status. Men rated PSS of friends and family lower if they had a SO, possibly related to relatively greater reliance on social support from a SO. More education was associated with greater PSS for Fr, Fm and SP. Given the importance of PSS on health outcomes, these findings suggest who may be most vulnerable, i.e., especially men who are less educated and without a SO.Supported by CFF. Objectives: To characterize 1) the need adults with CF have to talk about CF and to talk for emotional support and help (ESaH), 2) satisfaction with meeting these two needs to talk, and 3) how these needs and satisfaction with meeting them are associated with quality of life.

Methods: Fifty-four non-transplanted adults with CF participated in this cross-sectional, survey study; response rate was 67%. Participant demographics: mean age=32 years (range 22-50), 57% men, 63% with some college education or more, and 56% with a partner. Participants rated their need to talk to another person during the last month from none (1) to extreme (5) and rated satisfaction with the talks they had from very dissatisfied (1) to very satisfied (7). To measure life satisfaction, participants completed the Cystic Fibrosis Questionnaire-Revised (CFQ-R) and the Satisfaction with Life Scale (SWLS). All reported findings reflect statistically significant (p<.05) Pearson correlation or χ 2 tests.Results: Neither need to talk about CF nor need to talk for ESaH was significantly associated with age, education, or partner status. Sixty-seven percent of people reported needing to talk about CF, with 37% reporting this need was moderate to extreme. Fifty-nine percent of people reported needing to talk for ESaH, with 28% reporting this need was moderate to extreme. Men and women did not differ in their need to talk about CF. Women reported greater need than men to talk for ESaH. Sixty percent of people who talked reported they were satisfied with their talks about CF; similarly, 74% reported they were satisfied with their talks for ESaH. Demographic variables were not associated with satisfaction with either CF or ESaH talks.Negative correlations were found between need to talk about CF and the CFQ-R measures (higher scores are better) of: physical, role, social, and emotional functioning; health perceptions; vitality; treatment burden; and respiratory symptoms. Need to talk for ESaH yielded a similar pattern of negative correlations, with the exception that treatment burden and respiratory symptoms were unrelated to ESaH. SWLS was not correlated with the need to talk about CF or to talk for ESaH. Satisfaction with talks about CF and talks for ESaH were both positively correlated with CFQ-R measures of emotional functioning and health as well as with the SWLS.Conclusions: A majority of CF patients endorsed a need to talk about their CF as well as to talk for ESaH. For many this need was moderate to extreme. A majority reported satisfaction with their talks. Whereas need to talk was motivated by general/disease-specific distress, satisfaction with talking was associated with greater health, emotional functioning and overall life satisfaction. These findings underscore the importance of socially interacting (i.e., talking about disease-specific as well as emotional support concerns) to facilitate quality of life of adults with CF.Supported by CFF.Background: Patients with cystic fibrosis (CF) require both outpatient and inpatient care. Therefore, frequent and consistent communication among inpatient and outpatient staff is critical. In efforts to improve patient care, we identified needs for improvement in this area and developed an initiative to address these needs.Methods: We began our initiative by identifying the following goals: 1. Improve communication between inpatient and outpatient personnel regarding CF patient care.2. Establish ways to include inpatient staff as part of the extended CF care team.3. Provide (at minimum) annual up-to-date CF education to inpatient staff to ensure quality continuity of patient care.We established the following plan to accomplish these goals: 1. Phase 1: Offered CF education opportunities to inpatient staff presented by the outpatient team members, beginning with the dietitian and respiratory therapist.2. Phase 2: Established quarterly patient safety meetings with all disciplines including physicians, nurses, nurse educators, infection control nurses, pharmacists, respiratory therapists, dietitians, social workers, case managers, palliative care team and child life specialists from the outpatient clinic and the pediatric and adult inpatient units. These meetings include an education component to address patient care topics, such as "Encouraging Adherence in the Patient with CF" and "Infection Control Specific to CF," and an open forum to discuss updates and concerns from each unit as well as future meeting topic ideas.3. Phase 3: Included the pediatric and adult inpatient units in our annual CF Education Day for our patients and families. This provided an opportunity for attendees to ask questions of the inpatient staff regarding hospitalization and gave each unit the opportunity to present what they are currently doing to improve patient care. The inpatient staff representatives present were introduced to attendees as part of the extended CF care team.4. Phase 4: Established a transition program, led by the outpatient clinic social worker and the adult inpatient nurse manager, to coordinate the transition of our teenage patients from the pediatric inpatient hospital setting to the adult inpatient hospital setting. This includes opportunities for patients to meet the adult inpatient staff and tour the adult hospital facility prior to their first adult admission.Results: Preliminary polling yielded reports of an increase in knowledge of CF care and improved communication and collaboration between inpatient and outpatient staff. A formal annual survey is scheduled for September 2010 of inpatient and outpatient staff involved in the initiative to assess progress in achieving the goals established. Patients currently involved in Phase 4 are scheduled to undergo a post-program interview upon completion of the transition program to gather feedback.Conclusion: We have developed a multi-component initiative to improve communication and patient care between inpatient and outpatient units. Preliminary feedback for the program has been positive. Our future plans include an extension of Phase 1 to initiate a multi-disciplinary continuing education presentation by the outpatient CF team for inpatient staff. Background: Intensive behavioral intervention with nutrition counseling is recommended as an evidence-based treatment for children with CF ages 1-12 years below the 50th %ile BMI (1). The guidelines are based on research that has demonstrated behavioral interventions for nutrition to be effective at increasing calorie intake and weight gain (2, 3, 4) . Positive nutrition and growth outcomes were present at post-treatment for a cohort of toddlers and maintained at the 2 year follow-up time point, however outcomes declined for several children between follow-up years 2 and 4 (5). The declines suggested that ongoing booster treatment is needed. In order to develop appropriate treatment targets, it is important to understand the challenges associated with achieving CF nutrition recommendations and the strategies that parents use to achieve them during this developmental transition.

Aims: To identify: 1) challenges that impact CF nutrition during this transition; and 2) behavior and nutrition strategies parents recall from past treatment and continue to use.Methods: Eight of nine families in the original Powers et al. trial (5) participated in a semi-structured interview asking parents about their experiences during and since the behavior-nutrition treatment regarding their child's eating. Interviews were audiotaped, transcribed verbatim, and coded in the approach described by Braun and Clarke (6) . Themes were independently identified by three coders and then confirmed via consensus. Consensus themes were compared to themes identified by an independent reliability coder.Results: The mean age of the children at the time of the interview was 8.2 years (SD=0.8). Sixty-three percent of the participants were male. Four domains were identified that represented ten major themes (see Table 1 ).Conclusions: This is the first investigation conducted in order to understand what families recall from a brief behavior and nutrition treatment delivered when their children were toddlers, and which of these strategies families continue to use to achieve CFF nutrition recommendations. Parents reported that they continue to use behavioral and nutrition recommendations delivered several years ago, and have been able to apply them as their children have grown. Results suggested that families would benefit from anticipatory guidance about the transition to starting school or daycare and how to best engage in transfer of treatment responsibility. In addition, addressing the issues of ongoing parental distress and behavioral issues of picky eating and noncompliance were noted by parents as needing attention.Supported by NIH grants K24 DK059973 & T32 DK063929.12.6% Black, 11.2% Black Caribbean, 6.3% Pacific Islander, 1.4% American Indian, 1.4% Black Hispanic, and 2.8% other. A quarter of the survey respondents were male (25.7%). Results: Results indicated that approximately half the sample felt inexperienced in the following areas: communicating with dying patients, communicating with dying patients' families, and managing dying patients' pain (43.8%, 45.1%, and 50.7% respectively). A larger number of healthcare providers indicated feeling inexperienced in discussing end of life care and DNR status with pediatric patients (63.2% and 65.3%). The top five barriers to providing pediatric palliative care included the family's prolonged inability to accept terminal diagnosis, the family's unrealistic expectations, the family's level of education, the patient's level of education, and the family's culture.The majority of staff indicated that they were "likely" or "very likely" to consult the newly established pediatric palliative care team at the hospital (67.1%), with only a small percentage indicating they were unlikely to use the service (15.5%). Healthcare providers reported losing approximately six patients in the last year who were, on average, 8.64 (6.17 ) years old at the time of death. Slightly more than half the sample reported feeling that the hospital did not provide adequate support for those staff working with children with a life-limiting illness (52.5%).Conclusions: Overall, these results highlighted the need for additional education and support for pediatric healthcare staff in providing palliative care. While attempting to find a cure, it is also important to recognize that children with CF continue to experience worsening health and premature death. Palliative care teams can aid patients with CF, families, and staff in improving HRQoL. The aim of this study was to understand the influence of patient preference on physician prescribing practices. A second aim was to evaluate whether physician characteristics & beliefs are associated with prescribing a drug based on patient preference.

Methods: Participants were solicited by an email sent to CF Center Directors by the CF Foundation; the email provided a weblink for survey completion. Survey respondents chose a treatment regimen for a clinical case vignette in which the guidelines would support prescribing azithromycin, dornase alpha, hypertonic saline, & inhaled tobramycin (e.g., over 6 years old, chronically infected with P. aeruginosa, moderate to severe lung disease). Patient & illness characteristics, (including patient "prefers not to take a lot of medication") were then individually varied to assess their impact on prescribing practices. Respondents completed questions about their beliefs & barriers to following treatment guidelines, perceptions of the efficacy of each medication to improve lung function & reduce exacerbations & personal demographics.Results: A total of 122 physicians responded to the survey (73% males, 71% directors, 72% from pediatric clinics, time since medical school graduation: M=23 years (SD=10)). In the initial vignette, 79% of respondents prescribed all four drugs (Table) . When patient preference was added, only 27% prescribed all four drugs; physicians were significantly less likely to prescribe hypertonic saline (p<0.005; 44% of those who initially prescribed it removed it from the regimen), inhaled tobramycin (p<0.05; 30% removed it) and dornase alpha (p<0.05; 18% removed it). There was no difference for azithromycin (p=0.12; 17% removed it). In multivariate analyses, factors that predicted removing a drug in the patient preference vignette after prescribing it initially: male gender & more recent medical school graduate (hypertonic saline), being a center director (inhaled tobramycin), and lower perception of the helpfulness of the drug in reducing exacerbations (dornase alpha). No variable was significantly associated with removing azithromycin.Conclusion: In the initial vignette, most physicians prescribed all four CF guideline-recommended pulmonary medications; physicians were less Introduction: The natural progression of CF lung disease is one of progressive impairment eventually leading to disability. An analysis of the CFF Patient Registry (CFFPR) has shown an increase in the number of adults characterized as disabled despite improved health status in those patients. There are also patients with severe impairment who continue to work full time. We wanted to investigate the health status of adult patients using two definitions of disability.Methods: We analyzed the CFFPR (1999-2008) comparing the clinical status and adult milestone achievement of disabled patients defined by two different methods: (1) clinical criteria (e.g. lung function, exacerbation rate) used in the Social Security definition of disability, and (2) disability as recorded in the CFFPR. We specifically wanted to assess those patients who met one criterion but not the other. Patients meeting criteria for neither or both definitions were excluded. We compared health parameters (e.g. lung function, nutritional status) and milestone achievement (e.g. education level, marital status and parenthood) between the groups.Results: Patients characterized as disabled in the CFFPR only were older, had better lung function (FEV1), and better nutritional status (fewer patients with BMI <20) than did those patients who met criteria for disability by clinical criteria only. These differences were noted in all years investigated (Table shows data from 2008) . Despite being healthier, in those patients 23 years or older, completion of college was lower in patients disabled by the CFFPR than in those who met disability by clinical criteria. In addition the percentage of patients characterized as disabled in the CFFPR only has nearly doubled since 2000.Discussion: There is a difference in overall health status and adult milestone achievement when comparing patients with disability as reported by the CFFPR and using criteria from a clinical definition of disability. Patients disabled by the CFFPR only may have better health because they are able to focus on their treatment regimen, or there may be other factors not captured in the CFFPR as to why patients are disabled. Although, some patients may benefit from improved transition programs to better prepare them for adulthood. These groups deserve further investigation. Introduction: Cystic fibrosis survival continues to improve, with an estimated median survival of 50 years if born in the year 2000. Consequently, reproductive and sexual health (RSH) issues and the relationship between lung function, hormone levels and sexual activity are of increasing importance.Objective: To investigate awareness and attitudes of adult male CF patients towards RSH issues in a designated adult CF centre in Ireland. Secondary objectives were to address the relationship between sexual well being, predicted FEV1% and testosterone levels in adult males with CF.Methods: A RSH survey was developed based on published literature and a descriptive cross sectional study was undertaken in the Cork Adult CF centre. Questionnaires were posted to all male CF patients over the age of 16 (n=62). Best FEV1% and testosterone levels in the previous year were recorded from the medical notes. Data was analysed using SPSS (version 17.0).Results: Thirty-five men replied with a mean age (SD) of 27.2 (6.84). Of these, 34 patients (97%) were aware that male fertility is affected in CF; 65.7% provided the correct explanation how. Men reported first hearing about fertility issues from a variety of sources: CF healthcare professional 62.1%, parent 13.8% and other 24%. Over half the patients felt "not bothered" on initial discussions about infertility; 57.6% reported that the knowledge of infertility affected them more with increasing age. Eleven patients (31.4%) have undergone semen analysis testing, at a mean age (SD) of 28.7 (5.3) . Assisted fertility techniques were attempted by 6 patients (17.1%), with 4 (66.67%) successful. There is an inverse relationship between selfreported impotency rates and testosterone levels (p<0.05). Lower FEV1 levels are associated with self reported affected sex life (p<0.01). Twenty patients (57.1%) desire further written information on RSH issues.Conclusion: Irish males are well informed about infertility. This study reinforces findings in previous studies, where feelings about infertility become more significant with increasing age. This was predominantly due to a desire to father children. The relatively low rates of semen analysis are not dissimilar to other male CF cohorts.Further information is required. Based on the effect of patient FEV1% predicted and testosterone levels on sexual health, testosterone replacement studies need to be considered.

Opipari-Arrigan, L.C.; Bergman, E.; Seid, M. Cincinnati Children's Hospital Medical Center, Cincinnati, OH, USA Objective: The clinical encounter is at the core of care delivery and thus has great potential as a context for adherence promotion over a lifetime of CF care. Little research exists, however, to guide providers in how to make the best use of scarce time to promote adherence. Our objective was to identify clinical behaviors related to adherence determinants (self-efficacy, agreement on treatment plan) in routine CF visits.Methods: Video recorded clinical encounters of 30 patients (3-12 y; mean=7.2) and parents with their pediatric pulmonologists (MD; 4 MDs in study). Each verbal behavior was coded and categorized as "relationship building," "information exchange," or "treatment planning." Surveys, following the encounter, measured parent self-efficacy to adhere, and parent and provider reports of the treatment plan (for enzymes, aerosol treatment, airway clearance). Agreement between parent and MD understanding of the treatment plan was calculated for airway clearance (whether prescribed, type of clearance, number of times per day, treatment length), aerosol treatments (whether prescribed, type of medication, times per day), and enzymes (whether prescribed, type, dose).Results: Clinical encounters averaged 31 minutes (range=16-55). While 58% of MD time talking was spent in information exchange (asking questions, giving information), parents spent very little time asking questions (average=12 sec). MDs asked, on average, 5.1 (sd=2.6) closed-ended

The CF team play a significant role in the lives of families of children with CF. They are recognized as invaluable by parents/carers and are the first port of call when faced with a challenge. Potentially, the PROM CLCF-Q may have an important role in the annual review process. As well as extracting clinical data it raises unidentified concerns which may alert the CF team to otherwise unrecognised issues. A favourable consequence of routinely introducing the CLCF-Q into the annual review process would be an increase in the parent's/carer's SE through improved communication between the family and the CF team.Supported by the Alder Hey Cystic Fibrosis Trust Fund. (Geist, Grdisa, & Otley, 2003) . Thus, extended absences from school may have both psychological and health consequences, including social and emotional difficulties, increased use of healthcare, and higher risk of long-term adjustment difficulties (Perquin, et al., 2000) . Although considerable research has examined the effects of school absence in other chronic illnesses (e.g., chronic pain, sickle cell, cancer), few studies have addressed this issue in children and adolescents with cystic fibrosis (CF). Thus, the purpose of the current study was to evaluate the extent of school absences among children and adolescents with CF in relation to physical functioning and quality of life.

Methods: The current study examined a subset of school-aged children from two multi-site, longitudinal adherence studies at several CF Centers in the U.S. (N=55). Mean age of participants was 10.36 years (SD = 3.83), 40.3% were female, mean FEV1% predicted was 88.9%, and mean BMI percentile was 47.4%. School attendance rates were obtained retrospectively from participants' schools.Results: Preliminary results indicated a wide range of absenteeism among children and adolescents with CF. The mean number of school days missed over one year in this sample was 18, with a range of 2-61 days. No significant difference in school attendance was found for gender or age. Although not statistically significant, a negative trend was found between school attendance and disease severity, as measured by FEV1% and BMI percentile, with more absences associated with worse lung function and nutritional status, respectively (r=-.23, p= .12 and r= -.20, p = .17).Conclusion: Findings from the current study indicated that school absences vary substantially in this population. Further, the number of missed days was negatively associated with disease severity. To our knowledge, this is the first study to assess school absenteeism, as measured by school attendance reports, in children and adolescents with CF. Future analyses will examine school absenteeism in relation to other outcomes, such as coping skills and health-related quality of life (HRQoL) as measured by the CFQ-R. As life span for children with CF increases, more attention should be paid to their pursuit of age-appropriate activities, such as school attendance. Mental health of CF patients and caregivers is closely related to quality of life and health outcomes. In an effort to improve patient care, a pediatric psychology practicum placement was developed at a mid-size CFF accredited care center, enabling this clinic to participate in a psychosocial screening study in partnership with Dr. Quittner and the University of Miami. This questions for every 1 open-ended question. And while 24% of MD time talking was spent on relationship building, most of that time was spent in social chit chat as opposed to more sophisticated strategies to build a working alliance. Only 1.5 minutes, on average (14% of MD time talking) was spent in treatment planning.

MD time in social chit chat was negatively correlated with parents' efficacy to manage airway clearance (r=-0.46, p<0.05) and aerosol treatments (r=-0.50, p<0.05) when their child was sick. Agreement on treatment plan scores were averaged across the 3 domains of treatments and a median split was used to form two groups: high (n=12) and low agreement (n=15). There was no difference in visit length between the high and low agreement groups. A higher ratio of closed-to open-ended questions and a higher number of closed-ended questions were associated with less agreement, as was more time spent in social chit chat. Despite the small amount of time spent in treatment planning overall, the more time spent by MDs proposing and clarifying treatment plans, the higher the likelihood of agreement.Discussion: This study demonstrates feasibility and utility of using video recording to empirically describe the CF clinical encounter. These data suggest that MD behavior during routine CF clinic visits is related to parent self-efficacy, and treatment plan concordance, both key determinants of adherence behavior. These data as a whole suggest that there may be missed opportunities to utilize the clinical encounter to promote adherence. In order to make the best use of the clinical encounter for adherence promotion, MDs should ask more open-ended questions (versus closed-ended questions), spend less time in social chit chat, and spend more time in discussing and clarifying treatment plans.

Patel, L.; Glasscoe, C.; Dixon, C.; Dyer, K.; Southern, K.W.

Background: Parent Reported Outcome Measures (PROM) have become an integral component of patient care yet there is little evidence supporting the role of PROM and their benefits. We are conducting a single centre study to assess the impact of a PROM on the Self Efficacy (SE) of parents/carers of children with cystic fibrosis (CF). The SE of a person is their expectation of personal success in a given situation. In this study it refers to the parent's/carer's faith in their ability to deal with the challenges that a child with CF brings to family life.Methods: The Challenges of Living with Cystic Fibrosis-Questionnaire (CLCF-Q) is a 10-part, 62-item questionnaire which measures the burden of care on a parent/carer. It was developed with the input of parents/carers and previously piloted for sensitivity. To explore the feasibility and effect of incorporating the CLCF-Q into the annual review process we are conducting a parallel randomised controlled pilot study using the SE-Questionnaire (SE-Q) as an outcome measure. All participants complete the SE-Q at 2 time points. Between these time points, participants in the intervention group complete the CLCF-Q. They receive feedback at their next clinic appointment from the team in the form of colour coded tables illustrating both the positive and negative issues raised in the CLCF-Q. They are invited to talk to members of the CF team and others (consultants, CF nurses, dieticians, physiotherapists, psychologists, pharmacists, family doctors and the school) about the issues raised in the CLCF-Q. Select participants (n=3+3) are also invited to participate in narrative interviews.Findings: Preliminary data from 10 cases; parents/carers of children aged 2-13 years (7˛, 3ˇ) are reported, (5 control, 5 intervention) . From the intervention group 20% were single parents and 80% were lone carers. Twenty percent were carriers of Aspergillus and 20% of Pseudomonas and 100% encountered hospital admissions in the last 3 months. Sixty percent were unhappy about the support they received from family and friends and 60% found the family doctor uncooperative and unhelpful. Forty percent were not coping with the added expenses that a child with chronic disease brings and 80% had admitted sacrificing paid employment to care for their child. Sixty percent were very worried about the future and 60% were overwhelmed by the burden of care. The SE score ranged from 26/40 to 35/40 at baseline. In the control group the SE score remained relatively static whilst the intervention group has shown an increase in SE. CF patients greater than 20 years of age (see table) . This identifies 5 domains of sexual health (erectile confidence, erection firmness, maintenance frequency, maintenance ability and intercourse satisfaction). The maximum score is 25 and ED is classified into 5 severity levels; none (22-25), mild (17-21), mild to moderate (12-16), moderate (8-11) and severe (5) (6) (7) .Results: A total of 109 questionnaires were sent out with 32 responses (29%). ED was found in 9 patients (28%). Of these 7 (78%) had mild ED, 3 of them (43%) attributed the lack of erectile confidence being the cause for ED but none reported the inability to maintain an erection. There were 2 patients (22%) with severe ED and they struggled with all domains of sexual health. Twenty-three patients (72%) did not report any erectile dysfunction.Discussion: This survey showed that ED is prevalent in the male CF cohort. The poor response rate may have underestimated its true prevalence. The high proportion of patients who lack erectile confidence indicates a possible psychological component. However, the aetiology may be multifactorial and could be secondary to chronic infection/inflammation, diabetes, medication side-effects and hormonal imbalance. A more detailed study is warranted to investigate the exact factors and manifestation of ED in CF.1. Rosen R et al. Development and evaluation of an abridged 5-item version of the International Index of Erectile Dysfunction (IIEF-5) as a diagnostic tool for erectile dysfunction. J Impot Res 1999; 11: 319-326.

Background: Daily medical treatment regimens for patients with cystic fibrosis (CF) are intensive and time-consuming. Unlike other chronic diseases, they involve multiple modalities that include oral and inhaled medications and mucous clearance techniques. Research shows that many patients with CF are poorly adherent to recommended medical treatment plans but does not explain why patients become non-adherent and offers little guidance into treatment of this issue. Motivation to follow prescribed treatment plans is likely due to internal influences that could be explained through an individual's personality. Recently, there has been an influx of research in education illustrating improved learning when teaching is tailored to personality type.Objective: Identify the correlation of Myers-Briggs personality type to treatment adherence in patients with cystic fibrosis.Methods: This ongoing, pilot study is a cross sectional survey of seventy patients with cystic fibrosis, ages 16 and older, recruited from our institution. Two standardized tools and a demographic survey are administered to each patient. Personality types are identified via the Self-Scoring Myers-Briggs Type Indicator (MBTI) and patients are given a four letter categorical personality type. The Medication Adherence Questionnaire (MAQ) is a self-reported measure of volitional and non-volitional adherence. Patients are asked to report on their adherence to inhaled medications, pancreatic enzymes and airway clearance. A numerical score is given, based on MAQ responses, and designates the degree of non-adherence. Standard statistical analysis will be used to assess for any significant correlations.Results: Twenty patients have been accrued since March 1, 2010 . Early data indicates a preliminary trend between several MBTI types and poor adherence. This data is early and may change as full recruitment is attained. The study team plans for a full data set to be completed and presented at conference.Conclusion: We believe our study is a novel way to identify a plausible cause and potential treatment strategy for non-adherence. A correlation between non-adherence and MBTI type could have large implications for the potential development of effective communication and education programs for patients with chronic disease.

Bennett, D. 1 Poor physical health is one factor that may increase anxiety symptoms among youths with a chronic illness. The current study sought to examine the relation between physical health and both health anxiety and trait anxiety in an outpatient sample of youths with CF.Methods: Participants (ages 7-23; current n = 40) attending a regularly scheduled CF clinic visit completed measures of health anxiety (Health Anxiety questionnaire, Lucock & Morley, 1996) and trait anxiety (State-Trait Anxiety Inventory for Children, Spielberger, 1973). Youths and their parents each completed the Modified Health Perception Questionnaire (Pridham, 1994) to assess their perceptions of child health. As youth and parent ratings of child health were significantly related, scores were standardized and summed to create a perceived health composite score. In addition, objective health was assessed by standardizing and summing measures of lung function (FEV1), nutrition (Shwachman & Kulcyzcki, 1958) , and pancreatic sufficiency.Results: Perceived child health was related to both health anxiety (r = -.33, p < .05) and trait anxiety (r = -.53, p < .001) such that youths with relatively poor physical status had more anxiety symptoms. Moreover, a trend was found for health anxiety to mediate the relation between perceived physical health and trait anxiety (sobel test = 1.81, p = .07). Although perceived health was related to objective health status (r = .31, p < .05), the objective health status score was unrelated to anxiety symptoms.Conclusion: Subjective, but not objective, ratings of physical health were related to increased anxiety symptoms. Longitudinal research is needed to examine whether relatively poor perceived health precedes increased health anxiety, which in turn leads to increased general, or trait, anxiety symptoms.HADS-Depression were designed for patients with chronic illnesses and do not assess somatic complaints. The CES-D, in contrast, was developed for community samples and includes these items. These measures were administered to CF patients 12 years of age and older and to parent caregivers. Data on 623 patients, 492 mothers, and 131 fathers were analyzed using established clinical cut-off scores.Results: The mean age of participants was 17.76 years and 55% were female; 92.9% Caucasian, 3.9% African American, and 4.0% other. Elevated symptoms of anxiety on the HADS were found in 35% of patients (ado-lescents= 29.6%; adults= 37.7%), whereas 8.5% had elevated depression scores on the HADS (adolescents= 3.5%; adults: 11.0%). On a measure including somatic items, (CES-D), 27.1% of patients had elevated depression scores (adolescents= 21.7%; adults= 29.2%). Among mothers, 51.4% had elevated anxiety scores, 12.8% had elevated depression scores on the HADS and 25.6% on the CESD. In fathers, 43.8% had elevated anxiety scores, 14.3% had elevated depression scores on the HADS and 20.0% on the CESD. Paired sample T-tests suggested that the CES-D was significantly more sensitive to symptoms of depression in patients and mothers compared to rates of depression on the HADS. Small, but statistically significant inverse correlations were also found between FEV1 % predicted and patients' HADS-Depression scores (r =-.18, p<.01), with worse lung function related to more depressive symptoms.Conclusions: Overall, clinically elevated symptoms of anxiety were found in about one-third of adolescent and adult patients, with parents reporting even more anxiety, affecting about one half of all parents. Elevations in depression were also found on both the HADS and CESD, however significantly higher clinical rates of depression were found when somatic items were included in the assessment. Given these prevalence rates, annual screening of these symptoms and referrals for those in the clinically elevated range are recommended. The CF Foundation Infection Control Guidelines acknowledge the paucity of studies on the guidelines' impact on socialization among patients with CF. Isolation of patients with CF from each other has led internet-savvy individuals to use new media to create social networks while avoiding the risk of person-to-person transmission of pathogens. CF-specific web-based opportunities remain scarce. The impetus for this survey was the development of a friendship between two patients as a result of our recognition of a common interest; these two never met. Their four-year email friendship was posted as a blog with the aim of inspiring young adults who feel they are struggling with illness alone. Our hypothesis was that adults with CF use newer social media to create and maintain CF friendships as a means to overcome social isolation.

Methods: We conducted a survey of our adults with CF (n=200) via email and standard mail, with the goal of determining use of and interest in CF specific social interaction.Results: In all, 104 patients responded, of whom 51 were male (age 18-59). Ninety-seven percent reported having easy access to the internet. The most frequent modalities were telephone/texting (2.38 hours per week), followed by Facebook/Twitter (1.91 hours per week). 63% use these media for social interaction. The most important reasons for using the newer social media were maintaining non-CF friendships (39%) and contacting the care team with questions (38%). Most commonly listed as "somewhat important" were creating non-CF friendships (41%) and seeking information about clinical trials (31%). Interestingly the least important reasons for using the newer media included "create CF friendships" (49%); "share thoughts and feelings with other CF patients" (47%); "maintain established CF friendships" (45%); "contact other CF patients with similar health issues" (42%). Forty percent indicated frustration by restrictions imposed by the guidelines and 60.3% indicated a desire to stay in contact with social networks during hospitalization. Only 42% wanted CF Families Day re-initiated.Discussion: The data suggest that adults with CF frequently utilized these internet-based social networks to create/maintain friendships which were not restricted to other patients with CF. Our population placed little Research has shown that individuals with CF living in households with no smoke exposure have consistently higher pulmonary function measurements than individuals of families who smoke. Further, there is some evidence to suggest that the incidence of parental smoking is higher in the CF population when compared to other chronic respiratory conditions. There is a paucity of research examining the attitudes of parents of children with CF about smoking, as well as parental level of education about smoking and CF. This kind of information is an important first step in helping to improve lung function in children with CF through decreased exposure to tobacco smoke.

Participants. Parents of 27 children with CF will complete a survey designed to assess smoking behavior, knowledge, and attitudes.Measures. A smoking questionnaire has been developed for use in this study. It measures parental smoking behavior, knowledge about the effects of tobacco smoke on respiratory functioning, and attitudes regarding smoking and CF. Disease-related variables and demographic information are collected for each patient.Procedure. Parents are approached about completing the survey during routine CF outpatient visits. Patients who will not be seen in the next month will be sent a letter, together with a consent form and the survey to complete. Parents complete the survey either in clinic or at home and return them to a research assistant.Data Analysis. Data are currently being collected, reviewed and analyzed. For this presentation, descriptive statistics will be reported for the sample. Pearson product-moment correlation coefficients will be computed to examine the relationship between disease severity and smoke exposure, parental knowledge of how smoke exposure impacts children with CF and smoking behaviors, and parental attitudes toward smoking and smoking behavior.Hypotheses.1. Based on prior research, CF patients who are exposed to tobacco smoke will have poorer respiratory function, as measured by FEV1.2. No significant relationship will be observed between parental knowledge of how smoke exposure impacts respiratory function in CF and smoking behaviors.3. A significant relationship will be observed between parental attitudes toward smoking and smoking behavior, such that parents who smoke will have less negative attitudes toward smoking. (Katon & Ciechanowski, 2002) . Little is known however, about the prevalence of these symptoms in patients with cystic fibrosis (CF) and parent caregivers (Quittner, et al., 2009) . To date, only small single center studies have measured these symptoms in patients with CF. The purpose of this study was to estimate the prevalence of anxiety and depression in a national sample of patients with cystic fibrosis (CF) and parent caregivers in the US, and to evaluate associations between these symptoms and health outcomes.

Methods: Two standardized screening measures, one which did not include somatic items (e.g. fatigue) and one which did, were administered at routine clinic visits at 28 CF centers in the US. The HADS-Anxiety and value on a traditional CF Families Day format. We were surprised that only a minority of our patients expressed feelings of social isolation even though most were aware of the CFF infection control guidelines. However, there was a strong desire to maintain contact with social networks while hospitalized. More research is required to determine if patients with a higher degree of social isolation find support through social media with other CF patients or if patients with difficult-to-treat organisms experience more isolation. This presentation/poster addresses the activities of an interdisciplinary collaborative to improve communication between the three worlds of families, healthcare providers and school personnel in the area of "transition," with particular focus on the implementation and evaluation of a graduate level course that integrated the three perspectives.

The Interdisciplinary Collaborative on Healthcare and Education Transition (ICHET) first assembled in the spring of 2009, in response to an increased need for communication between healthcare providers and educators about the transition processes in their respective fields. Through multiple experiences of miscommunication, faculty from the different fields came to realize that both education and healthcare have processes referred to as "transition." These processes are different within each field and frequently professionals in one field are unaware of the parallel process in the other. ICHET includes interdisciplinary representatives from the UF College of Medicine (physician, social worker, family partner), the UF College of Education and the UF Institute for Child Health Policy.With the realization that fundamental change in professional behavior is most often effected at the graduate student level, ICHET faculty developed and provided a semester long, interdisciplinary online graduate course on Healthcare Transition through the UF College of Education. The course was offered to UF and distance students and was primarily designed to help graduate students in education learn about healthcare transition needs of teens with chronic health conditions. However, the course was also taken by graduate students in health professions fields and served the purpose of having each profession learn about the others' definitions of and activities towards "transition."The course consisted of six, two-week modules: Young Adults with Special Health Care Needs; Education View of Transition; Healthcare View of Transition; Family View of Transition; Young Adult Transition; Action Plan. Each module included an online summary of material, structured discussions, readings and a collaborative activity. Post-course surveys indicate that, students moved from an average of 1.5 to 4.0 (on a five point scale) in their knowledge of education/healthcare transition. Students identified the main course strengths as the interdepartmental collaboration and the amount of information and knowledge provided and unanimously indicated that they would recommend this course to friends.Additional ICHET activities have included collaborative presentations at a national education conference (Division on Career Development and Transition) and submission of a book proposal. ICHET faculty members plan to publish their experiences and to apply for funding for a pilot research project in the fall of 2010. In spring 2010 we will re-teach the course and provide a one day collaborative transition conference for school personnel and healthcare providers.

Romero, S.L.; Blackwell, L.S.; Armenteros, E.; Quittner, A.L.

Objective: Family functioning has been found to influence the pulmonary health of children with cystic fibrosis Patterson et al., 1993) . This study examined both adolescent and parent perceptions of parent-teen conflict at an initial visit prior to beginning an adherence intervention. Asso-ciation between demographic characteristics, disease severity and conflict were also evaluated.Methods: Participants included 117 youths with CF (mean age = 13.03 years; 39% female) and their parents who were part of a larger randomized clinical trial. Pulmonary function (FEV1 % predicted; mean = 85.03%) and BMI percentile (mean = 45.76%) at the Baseline assessment were used as markers of disease severity. Participants reported their perceptions of parentadolescent conflict on the Conflict Behavior Questionnaire (CBQ; Robin & Foster, 1989) , which was rated from 0 to 20, with higher scores indicating greater conflict.Results: Preliminary analyses indicated that parent and teen conflict behaviors were significantly correlated (r = .39, p < .01) at Baseline. Parents tended to report more conflict (mean score= 7.32) than adolescents (mean score= 5.81; t(116)= -3.6, p < .01). There were no gender differences between boys and girls in the magnitude of their reported conflict with parents. However, boys reported significantly less conflict with their parents than girls (t(60)=-3.65, p<.01). Older adolescents also reported more conflict with parents than younger adolescents (r = .32, p < .01). Disease severity (pulmonary function and BMI percentile) was not associated with conflicts reported by teens and parents. However, when the relationship was examined by gender, we found that disease severity was inversely related to conflict only for males; adolescent males who had better lung function reported less conflict with their parents (r = -27, p < .05).Conclusions: Findings suggested that adolescents who have conflicted relationships with their parents may be at a greater risk for poor health outcomes. Interventions aimed at improving parent-teen relationships should be considered given the tremendous impact that CF and its associated treatments have on family functioning. Future analyses will examine associations between treatment adherence and extent of conflict between parents and adolescents, with a focus on adherence as a potential mediator of this relationship.Funding was provided by the National Institutes of Health (HL#47064). 

Background: Treatment adherence is a significant issue for adolescents with cystic fibrosis (CF). Research suggests that although CF team members provide education to increase adherence behaviors, adolescents often lack the skills to enact these recommendations. Parental coping has also been linked to adolescent adjustment and health outcomes in other chronic diseases; however, these studies have primarily used self-report measures that are not context or disease-specific, limiting the clinical utility of the results. Quittner and colleagues developed the Role-Play Inventory of Situations and Coping Strategies (RISCS) to evaluate the frequency and difficulty of common situations faced by parents of adolescents with CF and assess the efficacy of parent-generated coping strategies. The current study utilized the Parent RISCS to: 1) determine the association between parent and adolescent coping, 2) evaluate the relationship between parent coping strategies and adolescent health outcomes, and 3) identify whether caregivers within the same household demonstrated similar levels of coping efficacy.Methods: As part of a larger randomized trial, 161 caregivers of adolescents with CF, comprising 103 families, completed the RISCS at the Baseline assessment. Twenty-six common, problematic situations faced by families with CF were grouped into 5 scales: Treatment Adherence, Parent-Teen Relationship/Discipline, Growing Up/Long-term Health Issues, Medical Procedures/Symptoms, and Financial Issues. Adolescents completed the Adolescent RISCS. Health status variables were also collected.Results: Results suggested that parents and their adolescents do not demonstrate similar coping skills across equivalent domains; however, effective parent coping on the Growing Up domain (e.g. responding to fears that your teen may miss out on life due to CF) was positively related to adolescent coping on the Eating (e.g. frustration due to lack of weight gain) and School Issues (e.g. catching up on schoolwork) domains. Adolescent age was positively associated with the frequency of parent-reported Treatment Adherence problems (e.g. deciding to skip treatments to see friends). Better lung function was correlated with more parent-reported difficulty on the Treatment Adherence, Parent-Teen Relationship and Medical Procedures scales. Similarly, higher adolescent BMI percentile was associated with increased difficulty ratings by parents on situations comprising the Treat-ated with CF care. The objective of this two-phase study was to assess the overall economic burden of disease associated with CF for patients and their caregivers. The first, qualitative phase of the study explored the full range of possible CF costs and unpaid time. The second, quantitative phase measured the specific categories of out-of-pocket costs and time estimates for households, including such direct costs as prescription and physician co-pays, OTC medications, medical equipment, and specialized childcare. Administration of home therapy, traveling to healthcare providers, office, clinic, and hospital/ER time, financial management, and housekeeping were identified as categories of unpaid time. "Opportunity costs" included missed work days or school days, and vacation costs/losses. An expense diary and brief survey were distributed to 133 households, which included CF patients under 6 years of age (n=27), 6-12 years of age (n=18), 13-17 years of age (n=26), and patients 18 years of age or older (n=62). The diary covered a 16week period from November 2009-February 2010 and was timed to capture "big" and "small" expenses, such as vests and nebulizer tubings, respectively. A wide range of disease severities, geographies, and family structures were included and analyzed in the study. Most respondents fell within the "mild or moderate" range of severity based on their most recent self-reported FEV 1 status (40-89% predicted). The CF patients also represented a mix of mild and severe CFTR genotypes. Approximately 57% of CF households were covered under private insurance, although 18% had no insurance for either some or all CF-related expenses. In 24% of households, Medicaid, Medicare, or another state or federal program served as the primary source of coverage. In this study, all patients had relatively high out-of-pocket costs, although the costs for certain CF patients were higher than others. Significant amounts of time were spent caring for CF patients. These hours were converted into dollars using the market value replacement wage rate approach, and when monetized, increased the economic burden even further.

Sufian, B.S.; Passamano, J.; Sopchak, A.; Thanos, C. Sufian & Passamano, L.L.P., Houston, TX, USA Introduction: Children and adults with CF without health insurance coverage or inadequate coverage visit CF centers less frequently and are unable to obtain prescribed medication. Consequently, their health declines. Those with CF who lose coverage or have inadequate coverage need an immediate solution. Assistance with efforts to obtain coverage can result in improved access to medical treatment and improved health outcomes.Method: The CF Insurance Access Project ("IAP") was established in March 2009 to provide pro bono representation to people with CF who need insurance coverage. The CF Patient Assistance Foundation (CFPAF) provides grant funding. The IAP is staffed by attorneys in the law firm of Sufian & Passamano, LLP. The IAP represented 100 individuals between March 2009 and March 2010 whose income was at or below 400% of Federal Poverty Level and who met eligibility criteria for a government program that provides insurance coverage. Individuals were referred to the IAP by the CFPAF, the CF Legal Information Hotline or a CF center. IAP clients reported that loss of coverage meant no funds to pay for clinic visits or medication resulting in a substantial decline in health. IAP clients with insurance coverage indicated that high co-pays resulted in missing CF center visits and reducing the amount of medication taken each day. Medication assistance programs provided access to a limited number of medications but did not meet all of the IAP clients' medical needs.Results: Individuals represented by the IAP in their application for government benefits were approved for benefits 90% of the time. Nationally, only 10% of all initial applications for benefits are approved. Benefit appeals may take 6-24 months. Representation by an attorney familiar with CF and eligibility for government benefits led to strong applications which were approved within 1-3 months. Eligibility for government benefits and insurance coverage resulted in improved access to care for the client. Health outcomes improved after medication and clinic visits resumed.Discussion: CF centers need to concentrate their time addressing patients' complex medical issues. Many CF centers do not have a dedicated social worker or the time to assist with government benefit applications. While the assistance of an attorney makes a difference for applicants there are only a few attorneys who handle initial applications due to inability to receive a fee for services. One half of all IAP clients had limited or no cov-erage for the year prior to contacting the IAP and had an FEV1 below 40%. Obtaining government benefits resulted in coverage which allowed the IAP client to access medical care and medication. Individuals whose application for Social Security benefits was approved gained access to Medicaid or Medicare coverage and a monthly financial benefit which assisted in obtaining housing or purchasing food.Conclusion: Individuals who do not have health insurance coverage will not have access to the care they need to treat CF. Income disparities should not result in people with CF being denied access to care when the person may be eligible for government benefits. CF centers should identify patients who may be eligible for government benefits and refer them to the IAP for assistance. The theory of reasoned action (TRA) is frequently used to study health behaviors. In addition to elements of intentions to adhere, subjective norms, self-efficacy, and attitudes related to treatment, TRA acknowledges background factors' influence on adherence behaviors. This study's purpose was to develop preliminary data concerning the relationship between certain background factors (spiritual and religious constructs) and adherence determinants.

Method: A cross-sectional design employing N=30 parents of children aged 6-12 years with CF completed a series of adherence and spirituality and religion scales. Data distribution was non-parametric so Spearman's correlation coefficients between spiritual variables and adherence determinants were calculated.Results: The sample was 100% Caucasian; 43% male. The body sanctification scale Sacred Qualities of the body significantly correlated with self-efficacy in performing two daily treatments: airway clearance (rho=0.44; p<0.05) and aerosolized medications (rho=0.49;p<0.01) under numerous circumstances. Sacred Qualities and Manifestation of God in the Body scales showed a trend toward significance in their correlation with Attitudes (expected treatment value; rho=0.35; p=0.06 and rho=0.34; p=0.06, respectively); a large effect size (q=1.84) indicates these trends would likely have reached significance in a larger sample. No significant correlations were found between Sacred Qualities and self-efficacy for administering pancreatic enzymes, nor between the Manifestation of God in the body scale and self-efficacy for any of the three treatments.Conclusion: These data show a moderate relationship between parental sanctification of their child's body and parents' self-efficacy ratings to accomplish daily CF treatments. There may also be a correlation between parental sanctification of their child's body and attitudes towards the expected treatment value. This study is the first to link spirituality and self-efficacy in a pediatric chronic condition, expanding the potential importance of at least one of the TRA's background factors. Two other noteworthy aspects of this study are: first, Sacred Qualities significantly correlated with parental self-efficacy when the family was stressed in some way and not during "typical" weekdays or weekends. Spirituality appeared more operative when parents were stressed. Clinically, parental self-efficacy's correlation with intentions to adhere suggests it may be possible to influence adherence intentions by addressing a spiritual construct. Second, it was a spiritual and not religious construct that correlated with parental self-efficacy. This is important because religion denotes a set of corporately expressed, while spirituality denotes an individually-held set of internal beliefs and is therefore thought to be a more proximal determinant of adherence. Understanding parental sanctification (and perhaps other unexplored spiritual constructs) opens new and potentially important avenues to address adherence for many parents.

Taylor-Robinson, D. 1 ; Whitehead, M. 1 ; Diggle, P. 3 ; Smyth, R. 2 1. Division of Public Health, University of Liverpool, Liverpool, United Kingdom; 2. Division of Child Health, University of Liverpool, Liverpool, United Kingdom; 3. CHICAS, Lancaster University, Lancaster, United Kingdom Background: The rate of decline of FEV1 in people with cystic fibrosis is one of the most important predictors of death. Low socioeconomic status has been linked with poor outcomes in CF. We have explored, for the first time in a UK-wide cohort, longitudinal rate of decline in FEV1 and its relationship with socioeconomic status (SES) and other risk factors.Methods: We undertook a retrospective longitudinal cohort study of 3587 people with CF aged less than 20 years, and explored relationships between predicted FEV1 and SES, demographic characteristics, genotype and other clinical characteristics using the UK CF registry between 1995 and 2006. Mixed model linear regression analysis was used to estimate the effect of fixed and time-varying covariates on the mean FEV1 at age five (intercept) and the rate of lung function decline (slope). Census based indices of multiple deprivation (IMD) from the UK constituent counties were used as small area measures of SES.Results: The mean population change per year of % predicted FEV1 was -1.55 (95%CI -1.44 to -1.66), with a mean FEV1 at age five (intercept) of 93.1% (CI 92.2 to 94.1). Table 1 outlines significant risk factors in univariate analysis. All of these factors remained significant in multivariate analysis with the exception of screening and pancreatic enzyme supplementation. There was no difference by genotype comparing those carrying at least one F508 mutation to other genotypes. The most deprived SES quintile had a mean difference in population FEV1 at age five of -4.71% (95% CI -7.88 to -1.53), but SES did not have a significant effect on subsequent rate of decline of FEV1 in this population. This suggests that low socioeconomic status may have health damaging effects in CF in the first five years of life.Conclusions: Important risk factors for lung function decline are female sex, birth cohort pre 1990, P. aeruginosa or B. cepacia colonisation, receipt of supplemental feeding and CF related diabetes. Low SES is associated with a lower mean FEV1 at age five.We acknowledge the support of the MRC and the Cystic Fibrosis Trust. Background: Disparities in health outcomes such as lung function and BMI have been shown in cystic fibrosis patients. Previous studies have shown that socioeconomic status (SES), as defined by median income by zip code, is related to lung function, BMI, and health-related quality of life. However, median zip code is an imprecise measure of SES when compared to family-reported household income. Furthermore, most of the previous research has focused on maternal education but there is no data on the role of paternal education.Objective: The purpose of this study was to evaluate the relation between caregiver-reported income, maternal and paternal education level on health outcomes (FEV1% predicted (FEV1), BMI percentile (BMI %), and health related quality of life (HRQOL) in a pediatric CF patient sample.Methods: We recruited 50 adolescents and their caregivers (mean age =13.8, range 11-20 years, 56% male; 78% caucasian) from 5 CF care centers as part of a multi-site randomized trial (iCARE). Participants completed a baseline assessment, which included measures of baseline demographics and Cystic Fibrosis Quality of Life-Revised. Baseline health outcome data including FEV1 % predicted and BMI percentile were abstracted from the medical records. T-tests were conducted to examine mean differences in health outcomes by household income, maternal and paternal education level. A multivariate model was conducted to control for gender and race on the relation between SES and health outcomes.Results: Preliminary results indicated that participants who were below the median income, or had less maternal and paternal education had lower lung function results (p<0.05). Furthermore, paternal education was related to higher BMI percentile (p <0.05). Multivariate analyses that controlled for gender and minority status indicated that household income was the strongest predictor of both BMI percentile and FEV1 % predicted (p<0.01, accounting for 10% of the variance). There were no associations between household income, maternal and paternal education and any of the subscales of the CFQ-R.Conclusions: This study demonstrated that household income and parent education was related to lung function and BMI but not to measures of health related quality of life in this sample. This study is unique because it utilized caregiver report of income as opposed to median income by zip code, which may not be as sensitive a measure of SES and it included measures of paternal level of education. Furthermore, this study demonstrates that caregiver report of income was the strongest predictor of health outcomes compared to maternal and paternal level of education, even after controlling for minority status and gender.Supported by Novartis Pharmaceuticals Corporation, Genentech, Inc. and the Cystic Fibrosis Foundation.Introduction: An increasing number of patients with cystic fibrosis (CF) reach adulthood in relatively good health. As a result, many female patients become pregnant and have babies. Prior studies have found that women with CF who become pregnant are healthier than non-pregnant patients before pregnancy, and therefore tolerate pregnancy well.Methods: We performed a two center case-control study in women with CF who became pregnant and carried fetuses beyond the 2nd trimester from 2002 and 2008. All patients were matched to controls who were the same age and had similar lung function just before pregnancy. Lung function, body mass index (BMI), number of exacerbations, presence of diabetes and pancreatic insufficiency, demographic and microbiology data were collected. The Student t-test or its non-parametric equivalent was utilized to compare the groups. Multivariable linear regression was used to estimate predictors of loss of lung function in the cohort of patients.Results: During the study period, 36.7% of female patients in the adult clinic had prior pregnancies or became pregnant. There were 22 matched pairs of patients (pregnant vs. non-pregnant) 20.5±1.7), diabetes (31.8% vs. 50.0%) before pregnancy. There were no differences in the rate of exacerbations before, during and after pregnancy between the two groups. There was a trend towards decreased rates of pancreatic insufficiency in the pregnant group (66.7% vs. 90.9%; p=0.057). There were no differences in the change of BMI between the two groups at the end of pregnancy ( . On multivariable analysis pregnancy had no effect on change in lung function from pre-pregnancy to last follow-up. The only significant predictors of decline were higher lung function prior to pregnancy (p<0.001 and pancreatic insufficiency (p=0.036). There were no significant peripartum complications in this cohort.Conclusions: Pregnancy is common among adult cystic fibrosis patients. Patients that become pregnant tend to have mild to moderate disease. Pregnancy leads to no adverse events immediately or in the long-term for CF patients when compared to similar controls. Larger studies are needed to determine whether any subgroups of patients have increased morbidity with pregnancy. Limitations: Small differences between the groups could have been missed because of the small size of the study. The purpose of this study was to describe the total and Pseudomonas aeruginosa (PA) related annual healthcare costs and resource utilization among cystic fibrosis (CF) patients enrolled in commercial health plans.

Methods: Data for this study were derived from MarketScan claims database which captures person-specific direct medical utilization, expenditures, and enrollment from approximately 100 payers. A retrospective crosssectional study design was used. CF patients with an initial claim for a PA infection (ICD-9:482.1) were identified during 2005-2008 using international classification of diseases diagnosis codes from their medical claims of 277.0 and 482.1, respectively. First medical claim for PA infection served as index date and demographic information from administrative claims and healthcare utilization and costs from medical and pharmacy claims were extracted for 12 months pre and post this initial claim. Descriptive and pairwise non-parametric statistical analyses were conducted to describe the study sample and compare healthcare resource use and costs before and after PA infection using SAS version 9.2.Results: A total of 358 CF patients with PA infection met the study criteria with mean age 20.1 (SD:15.5) years and 48% females. Approximately 67% had pancreatic insufficiency, followed by chronic sinusitis (20%) and diabetes (15%) as comorbid conditions. Annual per-patient PA-related costs were estimated to be $18,167 (median: $9,305) for overall healthcare resource use, $3,113 (median: $478) for outpatient care, $10,123 (median: $0) for inpatient care and $4,928 (median: $443) for pharmacy services, respectively. Overall inpatient visits (0.5 vs. 1.3), outpatient visits (18.3 vs. 21.7) and unique pharmacy prescriptions (11.0 vs. 12.7) increased significantly (p<.01) 12-months before and 12-months after the initial claim for PA. Similarly, related healthcare costs also were statistically higher (p<.01) in the post period compared to baseline.Conclusions: PA infections place a significant economic burden to payers among cystic fibrosis patients. Future studies need to determine whether this PA-related burden varies across demographic and comorbid illnesses. In an effort to change the nursing annual visit education from a lecture with handouts to more interactive sessions, the nurses at the CHS/UAB pediatric CF center developed a multiple choice questionnaire to assess knowledge about CF. This questionnaire included 2 demographic questions regarding who completed the survey, parent or child, and age of child. The next 10 questions covered the topics of pathophysiology, lung health, infection control, nutrition, transition, CFRD, and knowledge of lung microbiological flora. The final question requested topics about which the patient or family wanted to learn more.

The data from this survey was collected over a 5-month time period. Results were analyzed using chi-square with a p value of less than 0.05 considered significant. Surveys were collected from 56 patients and parents, 39% were completed by the parent. The patient age was divided into ranges: 21% were 0-5 yrs; 24% were 6-9 years; 13% were 10 -12 yrs; 18% were 13-15yrs; 12% were 16-18 yrs; 12% were 19-21 yrs.More than 90% of the patients or parents were able to identify at least one of the correct answers in multiple-choice questions as correct. The question most commonly answered incorrectly was related to CFRD. A significant difference (p=0.002) between parent and patient answers was found only on the question related to nutrition. There was no significant difference The life expectancy of people with CF has steadily increased to more than 37 years of age. While the majority of adults with CF have reduced fertility, it is possible for them to have children with CF and become caregivers. Also, older siblings with CF may take on the role of caregiver for younger siblings. In 2009, we had 2 cases where the caregivers of admitted patients also had CF. Several concerns arose during these hospitalizations including 1) application of CF infection control practices to the caregiver; 2) how to legally intervene with symptomatic caregivers not under our medical care; 3) caregiver utilization of services intended for patients; and 4) staff confusion as to which person was the patient.

Case 1: The first case involved a mother with CF who stayed with her child. This was the mother's first role as caregiver during an admission due to her own health issues. The mother brought her own medications and equipment to perform her care which confused staff. A nurse tried to hook up mother's feeding pump to the patient. RTs were unsure of which vest unit belonged to the hospital. The staff was also unsure as to which person was the actual patient since the mother was symptomatic. Staff was uncertain how to apply the infection control policy to a caregiver with CF who needed access to the cafeteria and laundry facilities.Case 2: The second case involved a caregiver with CF who assumed the caregiver role of her younger sibling with CF. This situation was further complicated in that the caregiver was a pediatric patient herself only a year ago. It was confusing for staff to treat her as a caregiver since they provided medical care to her over the last 20 years of her life. The caregiver also had personal confusion in her role as she would utilize services intended for her sibling such as ordering extra food for herself or asking staff for assistance in her care. Staff was concerned about not providing food as they were aware of the caregiver's financial barriers and medical condition. As a previous patient the caregiver was familiar with infection control practices, but these were not enforced since she wasn't a patient.Discussion: The infection control team was consulted to address whether caregivers with CF would fall under our hospital's CF isolation policy. It was decided caregivers with CF may stay with children in the hospital unless they have contagious conditions, such as flu or TB, which are harmful to the general public. However, the CF infection control policy couldn't be applied to caregivers since 1) they aren't under our direct medical care; and 2) legally they don't have to disclose their medical condition. Also, we couldn't enforce room restriction since in-room amenities for caregivers aren't provided, such as meal or laundry services. Finally, caregivers were encouraged to practice good infection control measures including hand washing, staying in room as much as possible, traveling only to essential areas, and practicing good cough etiquette.Conclusion: New challenges arise when individuals with CF become caregivers. We encourage CF centers to review their current policies to be prepared to encounter similar scenarios.

Barto, T.L.; Flume, P.A. Pulmonary/Critical Care, Medical University of South Carolina, Charleston, SC, USA Introduction: CF patients are living longer and there is a growing adult population. We now have established adult programs to provide age-appropriate care to this growing population. Transition programs have been recommended to prepare patients not only for the movement from a pediatric to adult healthcare system but also to provide tools necessary for adult development. We would anticipate with implementation of these programs there would be a subsequent increase in achievement of adult milestones.Methods: We analyzed the CFF Patient Registry (CFFPR) comparing patients aged 25-30 years in 1999 and 2008. In 2000, the CFF mandated there be adult care programs at each CF center, therefore patients were analyzed prior to and after this mandate. These ages were chosen to allow for completion of education. We evaluated clinical parameters as well as adult milestone achievement (i.e. education, employment, and marital status).Results: There were differences between the two groups (1999 vs. 2008, respectively) in their employment status with a decrease in those working full-time (51% vs. 45%) and an increase in those characterized as disabled (14% vs. 22%). There were no differences in the milestones of education (completion of college; 40% vs. 41%) and married status (48% vs. 49%). Overall health had improved over time in all groups, with better lung function and nutritional status, including in those patients characterized as disabled (Table) .Discussion: Clinical outcomes of adult CF patients (i.e. nutrition and pulmonary function) have improved over the last decade. Despite the implementation of adult programs, there is no evidence of improved adult milestone achievement, and there has been an increase in the number of patients characterized as disabled. Patients characterized as disabled have also realized improvement in overall health, possibly because they are better able to attend to their healthcare needs. However, we should consider that the lack of an increase in milestone achievement may be because transition programs have not fully matured and require further development. The Johns Hopkins Adult CF Program cares for approximately 250 adults with CF. During hospital admissions for CF exacerbations, peak and trough tobramycin levels are obtained as part of routine monitoring to optimize effectiveness and limit toxicity. Because these levels require precise timing and close coordination between nursing and phlebotomy staff, inappropriately timed levels can be a frequent problem. A survey of CF adults who had recently been hospitalized identified efficiency of tobramycin monitoring as one of the main areas for potential inpatient care improvement. A QI project to optimize tobramycin monitoring using a combination of an established monitoring protocol, regular physician and staff education, and continual assessment of monitoring outcomes commenced and was led by the nurses of the Johns Hopkins Adult Cystic Fibrosis Inpatient Unit (Nelson 4).Methods: The QI initiative included the following components: 1) Establishment of a standardized order set for tobramycin dosing and monitoring; 2) Changing the dosing time for tobramycin to 9 a.m. to avoid competition with nurse report and with other medications usually given at 10 a.m.; 3) Development of a tobramycin monitoring education packet for house staff and nurses; 4) Monthly educational meetings with rotating house staff, led by nurse coordinators; 5) Transfer of responsibility for coordinating drug administration and timed blood draws from house staff to the inpatient nursing staff; 6) Redesign of tobramycin level tube labels to include both time of blood draw and time of drug administration; 7) Monthly meetings with nurse and physician representatives to review monitoring outcomes.Results: Retrospective review of two years of tobramycin monitoring data prior to these interventions revealed that 734 inpatient tobramycin levels had been drawn, of which 142 (19.3%) were deemed not interpretable due to inappropriate timing of the blood specimen or lack of exact details on the timing of tobramycin infusion, and therefore termed "futile sticks." In the six months following initiation of the QI project, only 11 of 204 (5.3%) of levels were futile sticks (p<0.0001, t-test of means using quarterly data). This resulted in elimination of approximately 28 futile patient needle sticks over the 6 month project period. In addition, the percentage of interpretable peak levels increased from 69% to 88% (p<0.0009), and interpretable troughs from 82% to 96% (p=0.03).Conclusion: A QI project focused on improving inpatient tobramycin monitoring by more actively involving nursing significantly improved quality of monitoring data and also reduced patient discomfort. Background: As the cystic fibrosis population ages, there will be more patients who should be candidates for screening as part of healthcare maintenance as done for the general population. This would include screening for cervical cancer, breast cancer, prostate cancer and colon cancer. Since CF clinics may play a major role in the primary care of adult CF patients, we wanted to investigate the attitudes and practices of the adult centers in the United States towards preventive health maintenance.

Methods: An internet based survey was created and all adult program directors were invited to participate. The survey first queried as to whether the CF center served as a primary care provider for some or all of their patients. We included questions regarding healthcare maintenance asking about cervical cancer, breast cancer, prostate and colon cancer screening. In addition, we asked the methods by which the centers screened for these cancers and if referrals were made to other specialists such as OB/GYNs or urologists. A descriptive analysis of the data was performed.Results: There were 33 responses from programs providing care for 4332 adult patients (51% male and 49% female). A large percentage (36.8%) was reported to be greater than 40 years of age. The adult clinics estimated they provide primary care for 45.4% of their patients. Female patients routinely screened for cervical cancer ranged from 25-100% with the average being 65.6%. No CF physicians reported performing cervical cancer screening and referred all female patients to a gynecologist. Breast cancer screening is also generally deferred to the gynecologists; however, several centers routinely discuss self breast examination with their patients. Male patients routinely screened for prostate cancer ranged from 5-100% with an average of 29.5%; all male patients are sent to urologists for prostate examinations. Thirteen (39%) centers order PSA screening for their male patients. For colon cancer screening, 17 centers (50%) order routine colonoscopy for patients over the age of 50 years of age, and 6 centers (18%) in addition order Hemoccult test for the patients. Interestingly, some centers begin colon cancer screening at ages 35 years (4 centers), 40 years (4 centers) and at 45 (1 center).Discussion: While many patients and centers indicate that the adult CF center is the primary care physician, the data indicate there is no uniform approach to routine preventive healthcare maintenance. Often the centers admitted being unaware of whether the patient was up-to-date on screening examinations.Conclusion: As many adult cystic fibrosis clinics assume a role as primary care provider for their patients, they should begin routine preventive health maintenance in appropriate patients. There are currently not any guidelines set for preventive health maintenance in the CF patient population. This may be an opportunity to investigate further what preventive health maintenance screening is appropriate for the cystic fibrosis population.We reviewed our Respiratory Care procedures and developed educational materials for proper use and care for each of them. During clinic we provided complete aerosol and airway clearance therapy, and then repeated the spirometry. We found significant improvement in some of them leading to a discussion in our quality initiative group. We recognized some inconsistencies in our care instruction and identified this area as an area of focus. We then developed a questionnaire that covered all areas of respiratory care such as aerosol use, secretion movement, device use, and care and cleaning of items. Within 2 months, we were able to employ this questionnaire to our entire CF population of 414. Using the response to questions, an educational plan was developed and implemented individually for each patient.Results: We are now starting to repeat the annual questionnaire and have seen some significant improvements on the results of the questionnaire. We feel that patients are more aware of the proper way to complete their therapy. Unfortunately, we have been able to identify a few patients who admit to non-adherence even with comprehensive education. These patients have been targeted for enhanced education.We also discovered that our team, both in the hospital and clinic have some variability in how we were teaching our patients. This program required us to review the new procedures with each team member to support consistency for our patients in all clinical settings by rotation of inpatient and outpatient personnel. Patient and parent satisfaction appears improved.Conclusion: In addition to informal reviews with each visit, we are now assessing each of our patients formally every year and documenting the results in our electronic medical record. The assessments are utilized to develop individualized education plans to meet the needs of each patient. We have also assured that each patient is presented with consistent education and re-training of Respiratory Care procedures that meet their specific needs.Clinical Implications: Providing consistent education and eliminating confusion should result in better outcomes, improvement in pulmonary function, and reduction in frequency of antibiotic use. We will continue to look for ways to improve the compliance of our non-adherence patients.

Fletcher, C.L.; Wilde, K.; Corrigan, M.; Cairns, A. MMP Pediatric Specialty Care, Maine Medical Center, Portland, ME, USA Background: Individuals with cystic fibrosis require a complex and well coordinated plan of care. After reviewing the 2007 center specific registry data, it was our belief that we could improve patient outcomes by focusing on the CF Foundation guidelines for care and then addressing targeted areas for improvement in outcomes. The CF team is responsible for providing tools and education to families in order to improve wellness and overall health as part of chronic illness care. Optimizing communication and providing clear, consistent expectations to families was thought to be a key factor in achieving our goals. By partnering with families and setting measurable goals, improved outcomes should follow.Methods: Our CF center director and program coordinator recognized the need to develop a plan to improve the quality of our patient care by encouraging patients to be seen in clinic four times annually, and ensuring that our patients received annual chest x-rays, pulmonary function tests, labs, and visits with members of our multidisciplinary CF team. A plan was developed to increase patient and family awareness of the CFF guidelines for outpatient care. A letter was written to each family, inviting them to participate in the "Wii® Can Win" initiative for 2009. The CF Foundation guidelines for care were clearly stated and a copy of CFF consumer fact sheet, "It's all about Sally" was included. A spread sheet was developed to track all patients at our center for 2009 clinic visits and annual requirements for care. Rewards were built into the program to encourage the kids to be involved. Educational materials including posters and written materials were used throughout the year to increase awareness of goals for the project. Patients were given cards that were stamped at each visit and when all requirements for the year were complete, they were entered in a contest to win a Wii® Game System and Wii® Fit game! Surveys were done at the conclusion of the project to obtain feedback from families. Transitioning from pediatric to adult cystic fibrosis (CF) care requires the active involvement of adolescent patients in the medical management of their chronic disease. We aim to improve our median body mass index (BMI) percentile and prepare adolescent patients for transition to the adult CF center by measuring CF-specific knowledge and providing targeted nutrition education. Based on 2008 CF Foundation patient registry data, our median BMI is less than the national average. We hypothesize that improving the median BMI in adolescents ages 13 through 19 could be achieved by providing targeted education on CF nutrition.

Methods: The Knowledge of Disease Management (Quittner et al., 2009) questionnaire consists of 46 questions divided into 4 sections: Nutrition, Pulmonary Health, General Health, and Treatments. Each study participant (n=25) was provided this questionnaire either during clinic or during their hospital admission. After completing the questionnaire, the dietitian or nutrition graduate assistant reviewed the 14 nutrition questions with each participant. The results of the questionnaire were shared with each participant by providing them with their answers marked on an answer key. The answer key included the survey questions, the correct answer in bold, and an explanation on why it is the correct answer. Study participants were then provided with a CF nutrition education two-page handout 4-6 weeks after their initial study visit via email or mail. Participants will be asked to repeat the Knowledge of Disease Management questionnaire at their follow-up CF clinic visit. Data collected on study participants included weight, height, BMI percentile, and forced expiratory volume in one second (FEV1) percent predicted.Results: There was no statistical significance between an adolescent's questionnaire test scores when compared to their baseline BMI percentile using the Pearson correlation. On average, adolescents scored fairly well on the Knowledge of Disease Management questionnaire (M=75%, SD=11). Based on individual sections, adolescents scored the highest in the Pulmonary (M=82%, SD=15) and Treatment (M=82%, SD=10), then Nutrition (M=70%, SD=15), followed by General Health (M=67%, SD=15). Results on the effect of the initial questionnaire, targeted education, and follow-up nutrition education have on the follow-up questionnaire, BMI percentile, weight, height, and FEV1 percent predicted are pending.Conclusions: By administering an initial and subsequent CF knowledge questionnaire, we seek to determine if knowledge regarding CF care and providing targeted nutrition education has a positive effect on an adolescent's BMI percentile, weight, height, or FEV1 percent predicted. Measuring knowledge and remediating gaps may strengthen our adolescent transition program to the adult clinic by providing participants with individualized education regarding CF care, improving adolescents' self-care independence. from 87% to 93%. The number of patient visits with CF team members also increased. The number of patients visiting at least one time with the nutritionist went from 89% to 99%, those seeing the social worker from 91% to 96%, and those seeing PT/RT from 88% to 93%.

Conclusions: In summary, our patients gained a greater awareness of the importance of adhering to the CF Foundation guidelines for care. Our staff also gained a heightened awareness of our need to coordinate patient visits to maximize patient time with team members. Our clinic registry data concluded that we showed improvement in all areas. Next steps included the development of an intensive nutrition program for 2010 to improve BMI outcomes. (CFRD) is the most common co-morbidity in CF. Successful management of CFRD with insulin therapy improves nutritional status which in turn improves pulmonary function. Extensive diabetes education is required to achieve this goal. We sought to assess the diabetes knowledge base of our CF patients.

Methods: We developed a novel Cystic Fibrosis Diabetes Knowledge Survey focusing on the following CFRD domains: pathophysiology, symptoms, management, and complications. Content was assessed by an expert panel in diabetes and CF. Surveys were administered to patients with and without CFRD ≥14 years of age in CF Clinic or on the inpatient service. The knowledge score was calculated as the percent of questions answered correctly. The relationship between the knowledge score and the most recent BMI, FEV1%, and HgbA1c was estimated using Pearson correlation coefficients.Results: Sixty-five patients completed the survey, 32 CFRD patients and 33 patients without CFRD. Gender, race, age, BMI, and FEV1% were similar between the groups. In CFRD patients average age was 32.9 ± 13 years, average BMI was 21.4 ± 3.9 kg/m2 and average FEV1% was 58 ± 24.1%. HgbA1c was significantly higher in patients with CFRD (8.51% vs. 5.96%, p<0.001). Patients with CFRD scored significantly higher on the knowledge survey than patients without CFRD (83% vs. 72.2%, p<0.001). For patients with CFRD, overall, higher knowledge scores were associated with lower age (r = -0.44, p=0.01) and lower BMI (r = -0.46, p=0.008). However, when stratified by gender, knowledge scores were negatively correlated with BMI only in females (r = -0.72, p=0.0005). Knowledge scores were not correlated with HgbA1c, FEV1%, or duration of CFRD. In patients without CFRD, no correlation was found between knowledge scores and age, BMI, FEV1% or HgbA1c. With regard to specific CFRD knowledge, 1 ⁄4 of CFRD patients did not recognize weight loss as a symptom of high blood glucose levels. Similarly, 1 ⁄4 of CFRD patients did not identify high blood glucose levels to be associated with acute pulmonary exacerbations. Nearly 1 ⁄3 of CFRD patients did not appreciate that declining PFTs could be related to poor CFRD management. All but one of the CFRD patients indicated they received formal diabetes education, the majority of which was performed by an endocrinologist (84%), an endocrine nurse (75%), and/or a CF dietitian (72%). While almost all patients received education of CFRD symptoms, management, and complications, 1 ⁄3 of patients with CFRD were not instructed in diagnosis of CFRD and 1 ⁄ 4 were not instructed in how CFRD relates to pulmonary status.Conclusions: We developed a novel survey to evaluate diabetes knowledge in CF and CFRD patients. We have identified gaps in diabetes knowledge, especially relating CFRD to weight and pulmonary status, which will guide improvements in CFRD education. Our survey identified that diabetes knowledge was not correlated with better glycemic control, suggesting that other factors impact CFRD management. Future studies will include multivariate models and will assess the impact of adherence to diabetes care and barriers to care on CFRD management. Background: Before newborn screening (NBS), cystic fibrosis (CF) was often diagnosed by clinical suspicion and frequently required hospitalization at time of diagnosis. Thus, initial "CF education" occurred as an inpatient, even for those patients who required no hospital care. However, NBS for CF often results in the diagnosis of asymptomatic infants. This shift offered the CF center the opportunity to assess the methods and efficacy of the education provided to new families.

Purpose: The objective of this study was to determine if 1-day outpatient education (OPE) was as effective as the standard 2-day inpatient education (IPE) for providing the basics about CF and its management for newly diagnosed (ND) families.Methods: Eligible subjects were parents with an infant diagnosed with CF through NBS and referred to the CF center. Participants were given the option of choosing the OPE or IPE tracks. The content of the education and team members that provided the education were the same for both tracks. Information was given in verbal, written and audiovisual formats for multiple learning styles. All information emphasized the benefits of early diagnosis and NBS. Upon completion of education, parents were given a 10-item questionnaire to assess knowledge retention and a 16-item questionnaire to assess satisfaction with the education process. Similarly, CF team members were given a survey to assess satisfaction with the education tracks.Results: The families of 12 ND babies consented to participate in the study. The participants were 11 mothers, 8 fathers, and 1 paternal aunt (N=20). Average age at diagnosis was 26.2 days (IPE) and 26.7 days (OPE); current average wt/ht percentile is 52% (IPE) and 59% (OPE). Five families completed the IPE track; 3 chose this track while the patient was asymptomatic and 2 received education during an admission for treatment of a pulmonary exacerbation. Seven families chose the OPE track. Statistical analysis (chi square) showed that knowledge retention and satisfaction with the education provided between the two groups were not statistically different. However, for the question: "If given the choice, would you choose this educational track or would you prefer another track?" families in the IPE group would have chosen the OPE track (p= 0.0307). Furthermore, families in the OPE group were more satisfied with their overall experience (p= 0.038). Eighty-six percent of the CF team was more satisfied with OPE. Finally, hospital charges were significantly higher in the IPE group: $4,572 vs. $1,324 for OPE (p <0.05; includes only infants who were asymptomatic).Discussion: In our experience, a structured OPE session achieves the same success as the traditional IPE method with educational goals and knowledge retention in families of ND, mostly asymptomatic infants with CF. The families and CF center staff clearly were more satisfied with the OPE track and costs of providing the required services were significantly lower. We speculate that OPE should be the preferred method for initial CF education and will decrease the distress that a hospitalization can add to a new diagnosis. Methods: A self-administered anonymous questionnaire was given to 199 CF patients followed in 13 CF Care Centres and to 1 physician of each centre. Questions concerned the characteristics of the patients and their nebulized treatments used at home (type of drugs used, duration of treatments, Children's Healthcare of Atlanta at Scottish Rite, Atlanta, GA, USA Background: Upon diagnosis, patients and families from our care center are seen by the cystic fibrosis (CF) nurse practitioner and a gastroenterologist or pulmonologist for an initial office visit and basic CF education. Since 2007, when CF newborn screening (NBS) was introduced in Georgia, 41 NBS patients have been assigned to our care center. An additional 14 patients (non-NBS) have been diagnosed with CF in that time. Due to the increasing number of new patients and limited time for education during a routine clinic visit, our care center staff developed a program called Building Blocks of Cystic Fibrosis (BBCF) as a supplement to basic CF education. Program objectives include assessing families' primary knowledge of CF, discussing concepts of CF care in more detail and increasing caregiver confidence in providing for their newly diagnosed child.

Methods: Each BBCF session is presented by the CF team registered dietitian, respiratory therapist, pharmacist and social worker and is two hours in duration. At the beginning of the class, caregivers are given a 10question pre-test related to CF care. The nutrition presentation describes the importance of nutrition and its relation to lung function, basics of CF digestion, absorption, treatment with enzymes, importance of fat soluble vitamins and high calorie, high protein diet. The respiratory therapist covers potential treatments, how inhaled medications are administered, the rationale for airway clearance, the importance of proper cleaning and maintenance of equipment, and an outline of pulmonary function testing. The pharmacist discusses CF medications, including bronchodilators, mucous thinners, antibiotics, anti-inflammatory agents, pancreatic enzymes, vitamins and acid suppression therapy. The need for brand name pancreatic enzymes is emphasized, as is the importance of avoiding cough suppressants. Finally, the social worker discusses team roles and clinic dynamics, as well as resources that are available for CF families, including resources on grief and loss. In order to determine the effectiveness of this multidisciplinary educational program, a 12-question satisfaction survey is given at the end of the class to evaluate both the presenters and the material provided.Results: Since October 2009, 20 caregivers representing 10 families have completed the class. All caregivers completed the pre-test, and the average score was 92.5 out of 100. Additionally, 12 caregivers completed the post-class survey; 97% of survey respondents indicated "strongly agree" or "agree" on the 12 satisfaction measures, indicating high caregiver satisfaction with the class and confidence in providing care for the newly diagnosed CF patient.Conclusion: The BBCF program has been well received by patient families and staff, and we have met our initial objectives. The pre-test results suggest that many caregivers understand basic CF care principles. BBCF builds on that knowledge and further prepares caregivers. Goals to further develop the caregiver education program include assessing retention of material taught during the class and developing a refresher course for existing patients called Stepping Stones of Cystic Fibrosis. frequency of delivery per day, efficacy, maintenance of nebulizers, and easyto-use feeling).Results: The mean age of the patients (49.5% males) was 26 years. Sixty-five percent of patients considered that they had moderate and stable CF. Ninety-three percent were daily treated with nebulizations.RhDNase wa used in 95%, 83% used inhaled colimycine, 80% inhaled tobramycine, 46% hypertonic saline. Most of them already stopped treatment because of a feeling of inefficiency, or deleterious effect, but treatment was re-initiated in 55% of them because of a decrease in lung function. Fifty-nine percent of patients described an ongoing benefit with their treatment, 28% no daily benefit, and 14% side effects.Eighty-five reported that the treatment was easy to manage except for colimycine, and during holidays. Nevertheless, 64% of the patients treated with RhDnase said they take their treatment every day versus 60% for inhaled antibiotics, and 43% for hypertonic saline serum.Seventy percent of the patients declared they clean their nebulizer every day, and 7% only once a month or never. Thirty percent of patients never used soap. Confusion was perceptible between cleaning and sterilization of the nebulizer.Patients considered they were well informed about inhaled treatments.Physicians were more optimistic about potential efficacy of nebulised therapies than the patients, but were pessimistic about compliance and ease of use.Conclusion: This study confirms the need for clear and unvarying information, an adequate education explaining the purpose of each treatment. This should improve the patient adherence and management of such a chronic treatment.Supported by Roche Inc., Neuilly, France. (1) regarding specific therapeutic recommendations for treating patients, 6 years of age or older, with cystic fibrosis (CF). The purpose of this study was to better understand the attitudes and perceptions of pulmonologists practicing at CF center facilities regarding these recommendations, specifically the chronic use of dornase alfa, inhaled tobramycin, hypertonic saline, and azithromycin.

A survey was developed within the model published by Cabana et al. (2) , a framework designed to assess how knowledge level and attitudes towards a given set of guidelines impact guideline adherence.Of the 159 responses received from physicians practicing in CF centers, 65% were directors of their respective centers. Directors see an average of 14 patients with CF per week; non-directors see an average of 10 per week.Findings from this study suggest that the guideline recommendations were generally well received by a majority of both director and non-director physicians treating patients with CF at CF centers. Outcomes expectancy varied with the specific recommendations; better outcomes were expected with the chronic use of inhaled tobramycin and dornase alfa than inhaled hypertonic saline or azithromycin. Neither group found it difficult to decide when to initiate any of the four therapies, although use of inhaled hypertonic saline was perceived as more difficult than the other recommendations. Respondents who did not fully agree with these recommendations were most likely to cite lack of clinical evidence as the reason for lack of agreement. Future efforts to disseminate evidence for the specific recommendations may be warranted to reinforce and improve adherence. Highlighting key recommendations in combination with underlying evidence-base and data demonstrating a favorable impact on outcomes may help overcome the limited resistance to using these valuable resources.This study was supported by Novartis and assisted by the CF Foundation. 1 Background: For 50 years, the Cystic Fibrosis Foundation (CFF) has worked to ensure that people with CF and their families benefit from the best possible clinical care, in addition to scientific and clinical research. In 2002, the CF Foundation launched a national strategic initiative to accelerate the rate of improvement in CF care. This initiative has resulted in the following: Learning and Leadership Collaboratives (LLC), Benchmarking, Data Transparency, and Mentoring Programs. Mentoring involves teaching and the application of learning to the clinical setting. It is a relationship that pairs an experienced colleague (mentor) and a new or new to role (apprentice) professional.Purpose: The purpose of the CF Respiratory Therapy (RT) Mentoring Program is to provide resources and information for respiratory therapists new to CF with the desired goal of improving the quality of care delivered.Development: The CF RT Mentoring Program began in 2008 with the development of the program's "RT 101" document and pilot program. The 101 document is a comprehensive review of the respiratory care delivered to CF patients. After successful completion of the pilot program, the Mentoring Metric was developed as a tool to measure the CF RT's knowledge base. It is used to quantify the benefits of partnering an apprentice with a well established RT and their CF center.Process: • The Metric, or knowledge base measuring tool, consists of 13 questions that are pulled from the RT 101 document and the answers are scored using the following modified Likert rating scale:Strongly disagree: -2 Disagree: -1 Agree: 1 Strongly agree: 2 Not covered: Blank • Apprentices complete a pre-visit metric and after visiting their mentor's CF center, they re-evaluate their knowledge base by completing a postvisit metric.• Mentors complete a post-visit metric that scores their perception of the apprentice's knowledge base after the visit. The apprentice's and mentor's post-visit metrics are compared for agreement on the level of acquired knowledge.Measurement definitions and scoring computations were developed for the above responses and ratings.Conclusions: To date, there have been 22 completed paired metrics and 5 remaining scheduled apprentice visits.• Apprentice pre-visit metric is used as a tool for identifying areas of strength and weakness, which can guide/aid in establishing improvement goals for the mentoring site visit.• Goal of visit: Try to make every apprentice pre-visit rating/assessment that is a Disagree (i.e., the apprentice feels they do not possess the skill) an Agree (a "New Competency").• We found a high degree of agreement between mentors and apprentices.• All of the apprentices demonstrated improvement in one or more areas. Eighty percent (182/227) of the time there was agreement about Improvements (including New Competencies).• Eighteen of 22 apprentices demonstrated New Competencies (79%, 75/95, of survey responses) that were agreed upon by mentor.• Apprentice-Mentor Competencies Disagreement is used as an action item(s) for follow-up. Cystic fibrosis (CF) is a demanding multi-system chronic disease. Optimal control requires patients and their families to become expert CF selfmanagers. Self-management (SM) includes monitoring and treatment behaviors (Bartholomew LK, et al. Patient Educ Counsel. 1993; 22:15-25) . Evaluation of the CF Family Education Program (CF FEP) showed positive results compared to usual care in health status, knowledge, SM behaviors, and coping (Bartholomew LK, et al. Health Educ Behav. 1997; 24(5) :652-66). The CF FEP is designed to provide training in CF SM with materials for parents, young children, elementary children and teens. It is a self-paced print curriculum based on social cognitive theory. This study describes updates to the CF FEP and its performance objectives in light of advances in CF medical care and to make use of web-based dissemination and better integration of SM concepts.

Methods: Performance objectives were updated and reviewed by a multi-disciplinary panel. Content revision included review of past user feedback and CF Foundation Clinical Practice Guidelines. Revisions included the addition of new therapies, newborn screening, infection control, and more content on objective monitoring such as pulmonary function and glucose tolerance tests. The Communication and Coping parent modules were integrated into a new Becoming a CF Manager module. Using Self-regulation Theory, the content is organized around a SM framework -Watch & Discover, Think & Act to guide monitoring, decision-making, problem-solving and communication. Watch & Discover to find CF related health or quality of life problems early and Think & Act to treat and manage problems. Interactive worksheets and the glossary were expanded and refined. More role model stories, CF written action plans, and use of a CF travel folder were added. New graphics have been added such as the relation between body mass index and lung function and rating exercise exertion. All materials are formatted to allow for web-based access and printing. All updates are reviewed by multi-disciplinary CF health care team members, CF parents, and the CF Foundation Education Committee. Developmentally appropriate patient materials for the three age groups will be reviewed and updated as well.Conclusion: The CF Family Education Program will continue to serve as a valuable tool to promote CF SM skill development for CF patients and families. Watch & Discover, Think & Act embodies principles of self regulation and can help guide monitoring and decision-making to address new CF related problems and to judge the effectiveness of treatment. The updated program's utility and effectiveness in clinical practice deserves further study.Supported by the CF Foundation, Robert Wood Johnson Foundation and in part by an unrestricted educational grant from Genentech, Inc. Original CF FEP funding by NIH NHLBI grant R1 HL38339.

Hernández, M. 2 ; Dalton, J. 1 ; Tortajada, J. 2 ; Mutlu, S. 1 ; Rejas, A. 2 ; Vilaró, J. 3,2 ; Lalinde, W. 2 ; Villà-Freixa, J. 1 1. Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, Barcelona, Spain; 2. O2 Health Link, Inc, Barcelona, Spain; 3. FCS Blanquerna, Universitat Ramon Llull, Barcelona, Spain The improvement of 3D graphics both on hardware (integration of 3D graphic card into CPU processor) and software (WebGL and HTML5) is making it possible for common computers to visualize complex realistic model representations. This, together with the ability of tools to visualize scientifically grounded biological data, makes it possible to create powerful tools to support clinicians and researchers, and also patients. Activ8 makes