key: cord-276850-tnlyk0wz
authors: Rodrigues, Anderson Messias; Fernandes, Geisa Ferreira; Araujo, Leticia Mendes; Della Terra, Paula Portella; dos Santos, Priscila Oliveira; Pereira, Sandro Antonio; Schubach, Tânia Maria Pacheco; Burger, Eva; Lopes-Bezerra, Leila Maria; de Camargo, Zoilo Pires
title: Proteomics-Based Characterization of the Humoral Immune Response in Sporotrichosis: Toward Discovery of Potential Diagnostic and Vaccine Antigens
date: 2015-08-25
journal: PLoS Negl Trop Dis
DOI: 10.1371/journal.pntd.0004016
sha: 
doc_id: 276850
cord_uid: tnlyk0wz

BACKGROUND: Sporothrix schenckii and associated species are agents of human and animal sporotrichosis that cause large sapronoses and zoonoses worldwide. Epidemiological surveillance has highlighted an overwhelming occurrence of the highly pathogenic fungus Sporothrix brasiliensis during feline outbreaks, leading to massive transmissions to humans. Early diagnosis of feline sporotrichosis by demonstrating the presence of a surrogate marker of infection can have a key role for selecting appropriate disease control measures and minimizing zoonotic transmission to humans. METHODOLOGY: We explored the presence and diversity of serum antibodies (IgG) specific against Sporothrix antigens in cats with sporotrichosis and evaluated the utility of these antibodies for serodiagnosis. Antigen profiling included protein extracts from the closest known relatives S. brasiliensis and S. schenckii. Enzyme-linked immunosorbent assays and immunoblotting enabled us to characterize the major antigens of feline sporotrichosis from sera from cats with sporotrichosis (n = 49), healthy cats (n = 19), and cats with other diseases (n = 20). PRINCIPAL FINDINGS: Enzyme-linked immunosorbent assay-based quantitation of anti-Sporothrix IgG exhibited high sensitivity and specificity in cats with sporotrichosis (area under the curve, 1.0; 95% confidence interval, 0.94–1; P<0.0001) versus controls. The two sets of Sporothrix antigens were remarkably cross-reactive, supporting the hypothesis that antigenic epitopes may be conserved among closely related agents. One-dimensional immunoblotting indicated that 3-carboxymuconate cyclase (a 60-kDa protein in S. brasiliensis and a 70-kDa protein in S. schenckii) is the immunodominant antigen in feline sporotrichosis. Two-dimensional immunoblotting revealed six IgG-reactive isoforms of gp60 in the S. brasiliensis proteome, similar to the humoral response found in human sporotrichosis. CONCLUSIONS: A convergent IgG-response in various hosts (mice, cats, and humans) has important implications for our understanding of the coevolution of Sporothrix and its warm-blooded hosts. We propose that 3-carboxymuconate cyclase has potential for the serological diagnosis of sporotrichosis and as target for the development of an effective multi-species vaccine against sporotrichosis in animals and humans.

Sporothrix schenckii was originally described in 1898 as the causal agent of a subcutaneous disease in humans in the Mid-Atlantic USA [1] . Subsequently, the disease was reported in rats naturally infected in southeastern Brazil [2] and later in a wide range of animals including dogs, cats, horses, cows, camels, dolphins, goats, mules, birds, pigs, and armadillos. Several Sporothrix spp., previously reported to be closely related to S. schenckii, are known to establish infections in various hosts, with dissimilar virulence traits [3, 4] and responsiveness to antifungal treatment [5] .

The S. schenckii complex consists of at least four closely-related species [6, 7] , ranging from geographically restricted agents such as S. brasiliensis [8, 9] to cosmopolitan pathogens such as S. schenckii s. str. and S. globosa [7, 10, 11] . Sporothrix spp. are endowed with an extraordinary ecological diversity [12] [13] [14] [15] ; they are frequently recovered from soil, plants debris, and insects (Coleoptera: Scolytidae). Phylogenetic data support a recent habitat shift within Sporothrix from plants to cats [9] that culminated in the largest epizootic transmission in southeastern Brazil [16] [17] [18] [19] . Feline sporotrichosis emerged in the 1990s, with S. brasiliensis recovered from many outbreaks [8, 20] . More recently, S. brasiliensis has been recognized as a threat to humans [21] [22] [23] due to the massive zoonotic transmission in southeastern Brazil that affects thousands of patients regardless of whether they are immunocompetent or immunocompromised [9, [24] [25] [26] .

Cats have been a source of Sporothrix spp. infection transmitted to humans and other animals [18, 19, 27] . Most human cases occurred in housewives and professionals who had contact with infected animals and a history of scratches or bites [21, 28] . The largest epidemic of sporotrichosis due to zoonotic transmission was reported in the State of Rio de Janeiro, Brazil [18, 19, 21, 23, 28] ; since then, the incidence of sporotrichosis among animals, particularly cats, has increased [8, 28, 29] . More than 4,000 humans and 4,124 cats were diagnosed at Instituto Nacional de Infectologia (INI) Evandro Chagas /Fundação Oswaldo Cruz by 2012 [30] . Pereira et al. [29] observed that the majority of cats with sporotrichosis in Rio de Janeiro between 2005 and 2011 were male, mongrel, and unneutered, had a median age of 24 months, and presented with three or more cutaneous lesions in non-adjacent locations. This mycosis in cats is hard to treat and generally requires a long period of daily care; these animals do not always respond well to antifungal treatment [30] .

Sporothrix is widely distributed in nature, and traumatic inoculation of plant organic matter is classically associated with infection [31] . Feline habits render cats more susceptible to the fungal agent because they are constantly in contact with soil, plant debris, and other cats that may be infected. In cats, a broad spectrum of clinical presentation ranges from single lesions to fatal systemic forms. After monitoring the feline epidemic for 4 years, Schubach et al. [18] observed that the lymphocutaneous form occurred less frequently than did involvement of the mucous membranes of the respiratory tract and upper digestive tract and multiple cutaneous lesions. Some animals present involvement of skin and various internal organs [32] .

Cats are susceptible to a variety of fungal infections, including blastomycosis [33] , histoplasmosis [34] , cryptococcosis [35] , candidiasis [36] , dermatophytosis [37] , aspergillosis [38] , coccidioidomycosis [39] , and sporotrichosis. Misdiagnosis results in ineffective treatment, which further worsens outcome. Major contributors toward misdiagnosis include the small number of affordable and effective treatment techniques as well as other social issues. Definitive diagnosis of feline sporotrichosis is based on mycological culture, micromorphological characterization, and mold-to-yeast conversion. Histopathological methods and cytopathological examination are useful tools for the presumptive diagnosis of Sporothrix infection in cats [40] . Detection of specific anti-Sporothrix antibodies offers a rapid alternative for accurate diagnosis [41] [42] [43] . We recently proposed a serological approach that employs an enzymelinked immunosorbent assay (ELISA) to diagnose feline sporotrichosis [44] using purified antigen (the S. schenckii ConA binding fraction) and crude antigen, with high sensitivity and specificity.

There is a lack of information about feline sporotrichosis and the antigenic components involved in infection; therefore, the present study aimed to explore the diversity of molecules expressed by closely related species (S. brasiliensis and S. schenckii) and that are recognized by immunoglobulin G (IgG) in sera from cats naturally infected with S. brasiliensis. We found remarkable cross-reactivity among S. brasiliensis and S. schenckii antigens, and we identified an immunodominant molecule that is an important biomarker in feline sporotrichosis, irrespective of clinical manifestation. Here, we show that, although S. brasiliensis and S. schenckii may infect different warm-blooded hosts, infection result in highly similar IgG-mediated response in cats compared to humans, what is important for serodiagnosis and to the development of prophylactic or therapeutic vaccine against the enormous health burden of sporotrichosis in endemic areas. This knowledge may enable selection of potential biomarkers that can be used in seroepidemiological studies, diagnosis, and vaccine development, and may contribute to understanding of the pathogenesis of this infection in cats and humans.

This study was performed in strict accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Ethics in Research Committee of the Fundação Oswaldo Cruz, Rio de Janeiro, Brazil, under license number L-041/06. This study was also approved by the Institutional Ethics in Research Committee of the Federal University of São Paulo under protocol number 0244/ 11.

S. brasiliensis and S. schenckii s. str. from cats and humans were used for protein extraction ( Table 1 ). The dimorphic nature of Sporothrix spp. was demonstrated by converting the fungus to its yeast form at 36°C in brain-heart infusion medium (Difco) and observing typical oval multibudding yeast cells. Molecular identification was performed and confirmed via DNA sequencing of the gene encoding calmodulin [25] .

Isolates were selected because they had been previously characterized at the molecular level [3, 9, 45, 46] ; crude exoantigen (CBS 132974 = Ss118; S. schenckii s. str.) was successfully used to diagnose feline sporotrichosis via ELISA [44] . Sporothrix spp. was grown on Sabouraud medium (Difco) agar slants at room temperature (20-25°C) for 7 days. Approximately 2x10 6 conidia (85% viable cells) were used to inoculate 500-mL flasks containing 150 mL of brainheart infusion broth. Cultures were incubated at 36°C in a rotary shaker (Multitron II, Infors HT) with constant orbital agitation at 110 rpm for 7 days.

Whole extracts of Sporothrix yeast cells were obtained as described elsewhere [47] . Briefly, yeast cells (4 mL of each culture) were collected via centrifugation at 5000 x g for 10 min at 4°C and washed three times in ultrapure water. Pellets were frozen in liquid nitrogen and disrupted by gridding. Cells were macerated with a pestle until a fine powder was obtained. This cellular powder was vortexed for 30 min at 4°C in Tris-Ca 2+ buffer (20 mM Tris-HCl pH 8.8, 2 mM CaCl 2 ) containing a commercial cocktail of protease inhibitors (1:1000; GE Healthcare), RNAse and DNAse (1:1000; GE Healthcare), and 600-μm glass beads (1:1; Sigma). Cell debris and glass beads were removed via centrifugation at 5000 x g for 10 min, and dithiothreitol (final concentration 20 mM) was added to the supernatant [48] . Protein concentrations were determined using the Bradford method [49] . Cell extracts were kept at -80°C until use.

Sera from cats with definitive diagnoses of sporotrichosis (via S. brasiliensis isolation in culture; n = 49) were obtained from INI/Fundação Oswaldo Cruz, Rio de Janeiro, Brazil. The distribution and number of skin lesions of the cats were classified as L1 (cutaneous lesion in only one place), L2 (cutaneous lesion in two non-adjacent places), or L3 (cutaneous lesions in three or more non-adjacent places). During examination, blood was collected via vein puncture and stored in an incubator for 1 h; serum was obtained via centrifugation and stored at -20°C until use. Sera from 19 cats with no evidence of sporotrichosis or other diseases (the control group) were obtained from São Paulo as described elsewhere [44] . Sera from 20 cats with other diseases were also studied to verify cross-reactions with feline infectious peritonitis/coronavirus (5 sera), feline leukemia virus (3 sera), feline immunodeficiency virus (2 sera), leptospirosis (3 sera), rickettsiosis (2 sera), erlichiosis (3 sera), and leishmaniasis (2 sera) as previously described [44] (S1 Diagram). All sera were stored at -20°C until use. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and 1D immunoblotting S. brasiliensis and S. schenckii protein extracts (2 μg) were analyzed via SDS-PAGE with 10% gels [50] and silver-stained [51] . The relative molecular weights of the fractions were estimated using standard broad-range molecular weight markers (Protein Benchmark, Invitrogen). For immunoblotting, proteins (10 μg) from strains CBS 132990, CBS 132021, CBS 132974, and CBS 132984 were resolved with SDS-PAGE and transferred onto 0.45-μm polyvinylidenedifluoride membranes (Bio-Rad) at 20 V for 30 min with transfer buffer (25 mM Tris base, 192 mM glycine, 20% methanol, pH 8.3) [52] using a Trans-Blot SD semi-dry device (Bio-Rad). Electrotransference was confirmed by staining with 0.15% Ponceau S and 1% acetic acid [vol/vol]. Membranes were destained and free binding sites were blocked overnight in phosphate-buffered saline blocking buffer (1% bovine serum albumin supplemented with 0.05% [vol/vol] Tween 20, 5% [wt/vol] skim milk, pH 7.6) at 4°C. To determine the best dilution of serum, one sample was tested at four dilutions (1:100, 1:200, 1:500, and 1:1000) against yeast extracts. Afterward, for all sera, membranes were probed individually with primary antibody diluted 1:500 at 25°C for 2 h. Membranes were washed three times with Tris-buffered saline (pH 7.5) containing 0.05% [vol/vol] Tween-20 for 10 min and incubated with horseradish peroxidase-conjugated goat anti-feline IgG (1:1000) for 2 h at room temperature. Membranes were then washed with Tris-buffered saline (pH 7.5) containing 0.05% [vol/vol] Tween-20 and signal was detected with an enhanced chemiluminescence detection kit (GE Healthcare). Blots were imaged in a transilluminator (Uvitec Cambridge). Allience 4.7 software was used to take several images at different time exposures, from 2 s each to a total of 10 images over 2 s.

Proteins were separated via 2D gel electrophoresis as previously described [45, 47] . Briefly, proteins (300 μg) were precipitated using the 2D Clean-up Kit (GE Healthcare) and resuspended in rehydration buffer (7 M urea, 2 M thiourea, 2% CHAPS, 1.2% DeStreak, 2% vol/vol isoelectric focusing buffer pH 4-7, and trace amounts of bromophenol blue) to a final volume of 250 μL. Immobilized pH gradient strips (pH 4-7, 13 cm; GE Healthcare) were rehydrated at 30 V for 12 h. Isoelectric focusing was performed at 20°C using a Multiphor III system (GE Healthcare) as follows: 200 V for 2 h, 500 V for 2 h, 1000 V for 5 h, and a gradient applied from 1000 to 5000 V for 2 h. Finally, the voltage was set to 5000 V for 60,000 Vhr. After 1D isoelectric focusing, the IPG strips were reduced for 15 min with 1.5% dithiothreitol and alkylated for 15 min with 2.5% iodocetamide in equilibration buffer (6 M urea, 50 mM Tris-HCl pH 6.8, 30% glycerol, and 2% sodium dodecyl sulfate). Second-dimension separation was carried out by placing equilibrated IPG strips onto 10 % polyacrylamide gels, sealing them with 0.5 % [wt/vol] low-melting-point agarose, and separating the proteins at 10°C using a Hoefer SE 600 unit (15 mA/gel for 30 min and then 23 mA/gel until the dye front reached the bottom of the gel). Proteins were developed with silver staining [51] or were directly transferred for immunoblotting.

For 2D immunoblotting, proteins were transferred onto 0.45-μm polyvinylidenedifluoride membranes at 25 V for 1 h with transfer buffer [52] using the Trans-Blot SD semi-dry system. The success of electrotransference was evaluated by staining with 0.15% Ponceau S and 1% acetic acid 1% [vol/vol]. Membranes were destained and free binding sites were blocked overnight in phosphate-buffered saline blocking buffer (1% bovine serum albumin supplemented with 0.05% [vol/vol] Tween 20, 5% [wt/vol] skim milk, pH 7.6) at 4°C.

Membranes obtained from 2D gels were probed with 1:500 primary antibody (gold standard pooled feline sera; n = 10) under the conditions used for 1D immunoblotting. Immunoreactive antigens were detected using an enhanced chemiluminescence detection kit (GE Healthcare). 2D immunoblots were imaged using the method used for 1D immunoblots.

Diagnostic values included sensitivity, specificity, positive predictive value, and negative predictive value. Receiver operating characteristic (ROC) curves were prepared and analyzed to determine the sensitivity and specificity of each antigen preparation for ELISA. The area under the curve (AUC) for ROC analysis was calculated to evaluate the diagnostic value of ELISA. We assumed that a test lacked diagnostic power when the ROC curve was linear with an AUC of 0.5 (the ROC curve coincided with the diagonal). A powerful test was assumed to yield an AUC of~1.0, indicating the absence of both false-positives and false-negatives (the ROC curve reached the upper left corner of the plot). To measure the degree of concordance of the results from preparations from strains CBS 132990, CBS 132021, CBS 132974, and CBS 132984, we calculated the kappa statistic and its 95% confidence interval (CI). Kappa values were interpreted as follows: 0.00-0.20, poor agreement; 0.21-0.40, fair agreement; 0.41-0.60, moderate agreement; 0.61-0.80, good agreement; 0.81-1.00, very good agreement [53] . P-values 0.05 were considered statistically significant. All calculations were performed with MedCalc Statistical Software version 14.8.1 (MedCalc Software bvba; http://www.medcalc.org). Findings are reported in line with the STARD checklist for studies of diagnostic accuracy (S1 Checklist).

We previously reported a high prevalence of S. brasiliensis in feline sporotrichosis outbreaks [8, 9, 20] . Based on this information, the main goal of the present investigation was to evaluate the presence and diversity of serum-derived antibodies against S. brasiliensis antigens in naturally infected cats. Further, we previously proposed the existence of a convergent humoral response in human sporotrichosis against antigens from S. brasiliensis, S. schenckii, and S. globosa [45] . To establish whether S. brasiliensis and S. schenckii express different antigens, we assessed whole cellular protein extracts from two strains of S. brasiliensis plus two strains of S. schenckii s. str. that were previously characterized by molecular [8, 9, 25, 54] and serological [3, [44] [45] [46] [47] methods. Remarkably, and in support of our hypothesis that immunological distance increases with phylogenetic distance, sera from these cats reacted similarly, with no significant differences in titer between ELISA plates coated with proteins from S. brasiliensis or S. schenckii (Fig 1) .

ELISA . Further information about ELISA statistics appears in S1 Table. doi:10.1371/journal.pntd.0004016.g001 S. schenckii CBS 132984, median 1.028 OD, 95% CI 1.027-1.294 OD (S1 Table) . When using the assay to diagnosis cats with sporotrichosis, the area under the ROC curve was 1.0 (95% CI 0.94-1.000; P<0.0001; Fig 2) , indicating excellent performance. The control group of 19 non-Sporothrix infected animals was associated with lower medians and smaller ranges: S. brasiliensis CBS 132990, median 0.2640 OD, 95% CI 0.2592-0.3098 OD; S. brasiliensis CBS 132021, median 0.2590 OD, 95% CI 0.2517-0.2942 OD; S. schenckii CBS 132974, median 0.2730 OD, 95% CI 0.2512-0.3136 OD; and S. schenckii CBS 132984, median 0.2670 OD, 95% CI 0.2567-0.2907 OD (S1 Table) . Differences between the absorbance values for the infected and noninfected groups were statistically significant (P<0.0001). Sera from cats with other infections were non-reactive. Similar cutoff values yielded 100% specificity and sensitivity: S. brasiliensis CBS 132990, 0.377 OD; S. brasiliensis CBS 132021, 0.363 OD; S. schenckii CBS 132974, 0.407 OD; and S. schenckii CBS 132984, 0.346 OD (S1 Fig). ELISA results showed very good agreement for the antigens assayed (kappa = 1.0). To diagnosis feline sporotrichosis via ELISA, we recommend the use of antigen preparations of S. brasiliensis, since this is the most prevalent species in feline sporotrichosis outbreaks.

Antigen diversity was assayed with 1D immunoblots using the four antigen preparations of S. brasiliensis and S. schenckii tested via ELISA. Proteins extracts were evaluated according to the amount of protein extracted, the diversity of bands, the integrity of the samples, and the reproducibility of extraction. Approximately 2 μg of Sporothrix yeast whole-cell extracts were resolved by SDS-PAGE; silver staining revealed numerous proteins ranging from 10 kDa to 160 kDa in size, with different intensities. The Tris-Ca 2+ extraction protocol [45, 47] was suitable for the study of Sporothrix antigenic molecules during feline sporotrichosis, yielding samples with high amounts of protein and no degradation.

As expected, antibodies from cats with sporotrichosis reacted with a wide variety of S. brasiliensis and S. schenckii proteins 20kDa to >160kDa in size (Fig 3) . Cat-to-cat variation resulted in characteristic banding patterns for each animal (Fig 3) ; supporting the hypothesis that in a genetically diverse population, the antibody repertoire is expected to vary among individual cats. On the other hand, we detected minor or no differences in IgG-reacting banding patterns between antigen preparations (Fig 3) , consistent with the close genetic distance between S. brasiliensis and S. schenckii [8, 9] . Despite this variation, all cats produced antibodies against a 60-kDa molecule in the S. brasiliensis proteome and a 70-kDa molecule in the S. schenckii proteome.

The major antigenic S. brasiliensis molecules (CBS 132990 and CBS 132021) recognized by feline IgG consisted of the following sizes: 60 kDa (100% and 100%, respectively), 90 kDa (92% and 92%, respectively), 100 kDa (86% and 86%, respectively), 38 kDa (60% and 56%, respectively), 40 kDa (56% and 58%, respectively), 45 kDa (44% and 42%, respectively), 30 kDa (36% and 26%, respectively), 52 kDa (30% and 32%, respectively), and 110 kDa (28% and 30%, respectively) (Fig 4A and 4C ). Minor molecules recognized by feline IgG had sizes of 80 kDa, 25 kDa, 28 kDa, 120 kDa, 160 kDa, 35 kDa, 20 kDa, 55 kDa, 85 kDa, and 23 kDa (Fig  4A and 4C) . The major antigenic S. schenckii molecules (CBS 132974 and CBS 132984) recognized by feline IgG had sizes of: 70 kDa (100% and 100%, respectively), 90 kDa (86% and 88%, Table. doi:10.1371/journal.pntd.0004016.g004 respectively), 100 kDa (76% and 82%, respectively), 38 kDa (74% and 62%, respectively), 40 kDa (64% and 56%, respectively), 52 kDa (58% and 56%, respectively), 30 kDa (50% and 34%, respectively), 55 kDa (48% and 48%, respectively), and 45 kDa (40% and 30%, respectively) (Fig 4B and 4D) . The minor S. schenckii molecules recognized by feline IgG had sizes of 25 kDa, 80 kDa, 28 kDa, 110 kDa, 120 kDa, 23 kDa, 35 kDa, 160 kDa, 85 kDa, and 20 kDa (Fig 4B  and 4D ). Sera from uninfected cats did not react with S. brasiliensis or S. schenckii antigens. Sera from cats with other infections were also non-reactive in the immunoblot assay. The frequencies at which Sporothrix molecules were recognized in the antigen preparations are presented in S2 Table. There was no association between the number of bands recognized by each serum and the distribution and number of skin lesions on cats with sporotrichosis.

We previously reported that the 60-kDa and 70-kDa proteins in S. brasiliensis and S. schenckii, respectively, are related to virulence profiles and are the main antigenic molecules during murine [3] and human [45] sporotrichosis. We also determined that this protein undergoes post-translational modification and is present as isoforms and glycoforms in the S. brasiliensis and S. schenckii proteomes [45] . We therefore investigated whether antibodies present in cat sera recognize all six isoforms in the S. brasiliensis proteome, as previously shown with human antibodies [45] . S. brasiliensis proteins were therefore resolved via 2D electrophoresis and immunoblotted with pooled sera from cats with sporotrichosis (n = 10) and optimal antibody titers according to ELISA. Serum-derived antibodies in naturally infected cats mainly recognized all six isoforms of gp60 (Fig 5) . The present results confirm that S. brasiliensis 3carboxymuconate cyclase is a highly polymorphic protein [45] with sizes of 55-62 kDa and with isoelectric points of 4.45-4.80. 2D immunoblotting revealed less diversity and more weakly reacting spots than 1D immunoblotting, perhaps due to protein loss during sample preparation for 2D electrophoresis (compared to the crude extracts used in 1D immunoblotting and ELISA) and serum dilution during pooling. 

In the past, sporotrichosis was reported in horses more frequently than in other animals [55, 56] . Although the disease has been reported in several animal species, cats are currently the most frequently affected domestic animal. Outdoor cats are exposed to the fungus through contact with natural environmental sources or other sick cats. Cats have gained importance in the zoonotic transmission of Sporothrix to humans [8, 28-30, 57, 58] . Presently, the all-time high number of feline sporotrichosis cases reported in Brazil has reached epidemic proportions [8, 20, 29, 59] . Unlike humans, cats are highly susceptible to this fungus due to the low frequency of granulomas and the richness of fungal elements observed during skin histopathology [18] . Moreover, unlike human-disseminated sporotrichosis, which classically affects immunocompromised individuals [24, 26] , systemic disease in cats occurs frequently and is not associated with immunodeficiency caused by feline immunodeficiency virus and/or feline leukemia virus [18] . In this scenario, absence of an efficient host immune response is a key factor in disease progression. The low frequency of granulomas and uncontrolled fungal growth suggest that a lack of adequate cellular immunity underlies disease severity and pathology.

ELISA has been achieved with various antigen preparations, thus enabling serodiagnosis of human [41-43, 60, 61] and feline [44] sporotrichosis. The present investigation suggests that ELISA-based quantitation of anti-S. brasiliensis IgG is remarkably sensitive for the detection of feline sporotrichosis (cutoffs of 0.346-0.407 OD); as expected, this strategy does not differentiate between S. brasiliensis or S. schenckii as the agent of infection. Despite its high prevalence in South and Southeast Brazil, S. brasiliensis infection in cats is not an exclusive host-pathogen association, since the sibling agent S. schenckii s. str. also occurs in cats in Brazil [8, 9, 20] and Malaysia [62] , albeit with significantly lower frequency. Our serology-based observations of a convergent antigenic response are similar to previous 2D immunoblotting results for human sporotrichosis [45] . Interestingly, in other thermally dimorphic fungi such as Paracoccidioides brasiliensis and P. lutzii (Onygenales), antigen composition seems to vary considerably between species, an observation that supported development of a differential diagnosis system based on titration of serum-derived antibodies from humans infected with distinct strains of Paracoccidioides [63] . It is likely that similarities in antigen profiling among clinical Sporothrix spp. (S. brasiliensis, S. schenckii, and S. globosa) could be related to a recent speciation event and therefore be less susceptible to variability; however, this hypothesis requires further investigation. In addition, factors related to environment or to host association may impose strong selection pressures on Sporothrix antigen profiles.

To date, no information about the humoral response in feline sporotrichosis has been reported in the literature; it is a completely unknown area in veterinary medicine that merits exploration. Here, sera from cats with sporotrichosis displayed immunoblotting patterns of Sporothrix-specific immunogenic molecules that were markedly different from patterns from sera from uninfected cats. These immunogenic components were 20-160 kDa in molecular weight. The main molecules recognized by cat sera were a 60-kDa protein in the S. brasiliensis proteome and a 70-kDa protein in S. schenckii s. str., followed by molecules weighing 100 kDa, 90 kDa, 40 kDa, and 38 kDa. The variety of antigenic components recognized by each serum may be due to specific antigens secreted by individual fungal strains [3, 64] as well as to different mechanisms of activation of each host's immune system [45, 65] . However, variation in antibody repertoire seems reasonable in a genetically diverse host population [66] .

We interpreted published data lacking taxonomic information or protein identification via matrix-assisted laser desorption/ionization time of flight (MALDI-ToF)/mass spectrometry (MS) and identified an immunodominant fraction that oscillated between 60 kDa and 70 kDa in various studies (Table 2 ). In this scenario, the humoral immune response may be influenced by the infection strain and the antigen preparation used to detect the humoral response. Regarding human sporotrichosis, Mendoza et al. [67] observed that 147-kDa, 90-kDa, 74-kDa, 55-kDa, and 40-kDa molecules are commonly recognized by sera from patients with sporotrichosis, and Scott & Muchmore [68] showed that 70-kDa, 40-kDa, 36-kDa, and 22-kDa molecules are immunodominant in human sporotrichosis. A 70-kDa molecule was previously highlighted as immunodominant during murine sporotrichosis [69] [70] [71] , with IgG1 and IgG3 predominant [71] . This 70-kDa molecule was originally described as an adhesin molecule for fibronectin and laminin in S. schenckii s. str. [69, 70] that localized to the fungal cell wall [46] . More recently, we identified this molecule via 2D immunoblotting followed by MALDI-ToF/ MS as 3-carboxymuconate cyclase (GenBank: KP233225), the major antigen of human sporotrichosis [45] . Based on 1D electrophoresis [3, 46] and 2D electrophoresis [45] , the molecular weight (55-73kDa) and isoelectric point (4.33-4.85) of 3-carboxymuconate cyclase vary intraand interspecifically [3, 45, 46, 64] . Variation in gp60/gp70 may be related to differential glycosylation patterns and amino-acid substitution along the protein core, since all glycoforms/isoforms display identical MALDI-ToF/MS spectra [45] (Table 2) .

Here, the intensity of the recognition of distinct molecules differed among sera from different cats, but serum from an individual cat displayed little variation when probing different antigen preparations. These data suggest that the antibody response differs between cats and that there are few qualitative variations in the expression of cellular antigens by S. brasiliensis and S. schenckii s. str. In this study, the clinical presentation of sporotrichosis in cats corresponded mainly to multiple skin lesions; we observed no association between the distribution and number of skins lesions and the number or type of molecules recognized by antibodies in the sera. Although it remains to be clarified whether the antibodies produced during active infection in feline sporotrichosis are protective, antibodies against 3-carboxymuconate cyclase appear to inhibit fungal adhesion to the host in a dose-dependent manner [70, 73] . In another dimorphic fungus, P. brasiliensis, passive administration of monoclonal antibodies against the immunodominant antigen gp43 or against the recombinant protein before and after intratracheal or intravenous infections reduced fungal burden and decreased pulmonary inflammation in mice [75, 76] .

Gaining insight into host-parasite interplay in the immunological context is essential for understanding the emergence of feline sporotrichosis and is critical to serodiagnosis and the development of vaccines. Here, we demonstrated that antigens derived from yeast cell extracts of S. brasiliensis and S. schenckii s. str. yielded excellent results in ELISA and immunoblotting. The variety of molecules recognized by sera may be related to certain characteristics of the isolate, such as virulence, or even related to immune-system activation in each individual host. During infection, Sporothrix antigens elicit an IgG-mediated response; 3-carboxymuconate cyclase (gp60 in S. brasiliensis and gp70 in S. schenckii) is the immunodominant molecule in feline sporotrichosis, similar to murine and human disease. Therefore, this molecule may also be useful as a marker in the diagnosis of feline sporotrichosis and is a promising candidate for the development of therapeutic vaccines to tackle sporotrichosis in highly endemic areas.

Supporting Information S1 Checklist. STARD checklist.

(DOC) S1 Table. Summary statistics for antigen detection via ELISA with four antigenic extracts.

(DOC)

S2 Table. Frequencies of antigen recognition on immunoblotting with serum-derived antibodies against S. brasiliensis and S. schenckii proteins in naturally infected cats (n = 49). 

On refractory subcutaneous abscesses caused by a fungus possibly related to the Sporotricha

On a mycosis observed in men and mice: Contribution to the knowledge of the socalled sporotrichosis

Characterization of virulence profile, protein secretion and immunogenicity of different Sporothrix schenckii sensu stricto isolates compared with S. globosa and S. brasiliensis species

Different virulence levels of the species of Sporothrix in a murine model

Genetic diversity and antifungal susceptibility profiles in causative agents of sporotrichosis

Molecular phylogeny of Sporothrix schenckii

Global ITS diversity in the Sporothrix schenckii complex

Phylogenetic analysis reveals a high prevalence of Sporothrix brasiliensis in feline sporotrichosis outbreaks

Emerging sporotrichosis is driven by clonal and recombinant Sporothrix species

Phylogeography and evolutionary patterns in Sporothrix spanning more than 14,000 human and animal case reports

Sporothrix globosa, a pathogenic fungus with widespread geographical distribution

Phylogeny of the Ophiostoma stenoceras-Sporothrix schenckii complex

Multi-gene phylogenies define Ceratocystiopsis and Grosmannia distinct from Ophiostoma

Emerging lineages in the Ophiostomatales

Sporothrix chilensis sp. nov. (Ascomycota: Ophiostomatales), a soil-borne agent of human sporotrichosis with mild-pathogenic potential to mammals

Sporotrichosis with widespread cutaneous lesions: report of 24 cases related to transmission by domestic cats in Rio de Janeiro, Brazil

Pathology of sporotrichosis in 10 cats in Rio de Janeiro

Evaluation of an epidemic of sporotrichosis in cats: 347 cases

Cat-transmitted sporotrichosis, Rio de Janeiro, Brazil

Feline sporotrichosis due to Sporothrix brasiliensis: an emerging animal infection in São Paulo, Brazil

Cat-transmitted sporotrichosis epidemic in Rio de Janeiro, Brazil: Description of a series of cases

Epidemic cat-transmitted sporotrichosis

Epidemic sporotrichosis

Disseminated Sporothrix brasiliensis infection with endocardial and ocular involvement in an HIV-infected patient

Emergence of pathogenicity in the Sporothrix schenckii complex

Sporotrichosis in Rio de Janeiro, Brazil: Sporothrix brasiliensis is associated with atypical clinical presentations

The epidemiological scenario of feline sporotrichosis in Rio de

Feline sporotrichosis: epidemiological and clinical aspects

Sporothrix schenckii sensu stricto isolated from soil in an armadillo's burrow

Haematogenous spread of Sporothrix schenckii in cats with naturally acquired sporotrichosis

Blastomycosis in indoor cats: suburban Chicago

Histoplasmosis infection in two cats from California

Cryptococcosis in cats: clinical and mycological assessment of 29 cases and evaluation of treatment using orally administered fluconazole

Candida spp. urinary tract infections in 13 dogs and seven cats: predisposing factors, treatment, and outcome

Exophthalmos secondary to aspergillosis in a cat

Coccidioidomycosis in dogs and cats: a review

Sensitivity of cytopathological examination in the diagnosis of feline sporotrichosis

Concanavalin A-binding cell wall antigens of Sporothrix schenckii: a serological study

Use of mycelial-phase Sporothrix schenckii exoantigens in an enzyme-linked immunosorbent assay for diagnosis of sporotrichosis by antibody detection

Immunoglobulins G, M, and A against Sporothrix schenckii exoantigens in patients with sporotrichosis before and during treatment with itraconazole

Serodiagnosis of sporotrichosis infection in cats by enzyme-linked immunosorbent assay using a specific antigen, SsCBF, and crude exoantigens

Immunoproteomic analysis reveals a convergent humoral response signature in the Sporothrix schenckii complex

Differences in cell morphometry, cell wall topography and Gp70 expression correlate with the virulence of Sporothrix brasiliensis clinical isolates

Two-dimensional gel electrophoresis data for proteomic profiling of Sporothrix yeast cells

Posttranslational, translational, and transcriptional responses to nitric oxide stress in Cryptococcus neoformans: Implications for virulence

A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding

Cleavage of structural proteins during the assembly of the head of bacteriophage T4

Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels

Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications

Practical statistics for medical research. London: Chapman and Hall

Genotyping species of the Sporothrix schenckii complex by PCR-RFLP of calmodulin

Systemic mycoses in dogs and cats

Sporotrichosis in a horse

Feline sporotrichosis: Transmission to man

Sporotrichosis in man and animal

Feline sporotrichosis in the southern region of rio grande do sul, Brazil: clinical, zoonotic and therapeutic aspects

A comparative serological study of the SsCBF antigenic fraction isolated from three Sporothrix schenckii strains

Development of an enzyme-linked immunosorbent assay for the serodiagnosis of several clinical forms of sporotrichosis

Molecular typing of Sporothrix schenckii isolates from cats in Malaysia

Serology of paracoccidioidomycosis due to Paracoccidioides lutzii

Heterogeneity of proteins expressed by Brazilian Sporothrix schenckii isolates

Differential induction of Th1-prone immunity by human dendritic cells activated with Sporothrix schenckii of cutaneous and visceral origins to determine their different virulence

Identification of the feline humoral immune response to Bartonella henselae infection by protein microarray

Production of culture filtrates of Sporothrix schenckii in diverse culture media

Immunoblot analysis of antibody responses to Sporothrix schenckii

Cell surface expression of adhesins for fibronectin correlates with virulence in Sporothrix schenckii

Passive immunization with monoclonal antibody against a 70-kDa putative adhesin of Sporothrix schenckii induces protection in murine sporotrichosis

Humoral immune response against soluble and fractionate antigens in experimental sporotrichosis

Host organism defense by a peptide-polysaccharide extracted from the fungus Sporothrix schenckii

Isolation and some properties of a glycoprotein of 70 kDa (Gp70) from the cell wall of Sporothrix schenckii involved in fungal adherence to dermal extracellular matrix

Detection of 2 immunoreactive antigens in the cell wall of Sporothrix brasiliensis and Sporothrix globosa

The monoclonal antibody against the major diagnostic antigen of Paracoccidioides brasiliensis mediates immune protection in infected BALB/c mice challenged intratracheally with the fungus

Saccharomyces cerevisiae expressing Gp43 protects mice against Paracoccidioides brasiliensis infection