key: cord-327855-txryqil7
authors: Kulka, M.; Chen, A.; Ngo, D.; Bhattacharya, S. S.; Cebula, T. A.; Goswami, B. B.
title: The cytopathic 18f strain of Hepatitis A virus induces RNA degradation in FrhK4 cells
date: 2003
journal: Arch Virol
DOI: 10.1007/s00705-003-0110-0
sha: 
doc_id: 327855
cord_uid: txryqil7

The mechanism responsible for the induction of apoptosis by the rapidly replicating HM175/18f strain of Hepatitis A virus (HAV) was investigated. Full length HAV RNA and viral capsid protein VP1 were detected in 18f infected cells at earlier times post-infection than in HM175/clone 1 infected cells. Analysis of total cellular RNA from HM175/18f infected FrhK4 cells by denaturing agarose gel electrophoresis and Northern blot hybridization revealed extensive degradation of both the 28S and 18S ribosomal RNA (rRNA) molecules. Similar degradation was observed when these cells were infected with Human coxsackievirus B1, a fast replicating enterovirus. In contrast, the parental strain of 18f, HM175/clone 1 did not induce RNA degradation. Inhibition of RNA degradation correlated with inhibition of virus replication. The pattern of rRNA degradation resembled degradation of rRNAs by RNase L, an enzyme activated in interferon-treated cells following infection with certain viruses. Ribosomal RNA degradation was accompanied by the reduction in the levels of several cellular RNAs including those for β-actin and glyceraldehyde-3-phosphate dehydrogenase, while the levels of c-myc and c-jun were higher. Interferon mRNAs could not be detected in either infected or mock-infected control cells, and STAT1, a key regulator of interferon action was not phosphorylated following virus infection. These results reveal a heretofore-undescribed pathway that involves the regulation of RNA degradation and apoptosis following HAV/18f replication in FrhK4 cells.

Hepatitis A virus (HAV), the only member of the genus Hepatovirus within the Picornaviridae family, is a major cause of infectious hepatitis worldwide [33] . The virus is characterized by its resistance to heat, acid pH, polar solvents, and common disinfectants [57] . It is primarily transmitted by the fecal-oral route and is a principal agent of localized food-borne outbreaks of infectious hepatitis in developed countries [17] . The virus contains a single (+) stranded RNA genome that is approximately 7.4 kb long and contains a 3 poly (A) tail. The 5 end is uncapped but covalently linked to the viral coded protein VPg [65] . The viral genome has been cloned and sequenced from a number of strains. It codes for a single polyprotein, which is subsequently cleaved by the viral protease 3C and possibly cellular proteases into mature viral proteins [43, 54, 56] .

Strains of HAV can be grouped into three categories based on their distinctive growth characteristics in permissive cells: i) wild-type (wt) strains, defined by a long incubation period between infection and detection of replication and a cytopathic effect (cpe) is not observed; ii) tissue culture adapted (cc) strains, defined by shorter incubation periods than for wt strains and cpe is not observed; and iii) tissue culture adapted (cp) strains, defined by shorter incubation periods than for cc strains and cpe is observed. The cp strains arise during repeated passage of cc strains in permissive cells [2, 7, 21, 25, 47, 62, 68] . Although the molecular basis for the development of the cp phenotype is poorly understood, it is well documented that both viral genome replication and viral protein expression can be detected much earlier in permissive cells infected with cp strains such as HM175/18f, HM175/24a and HM175/HB1.1, than in cells infected with the parental cc strains from which they were derived [7, 26, 29, 32, 42] . The rapid replication and viral gene expression observed for cp strains has been attributed to a more efficient initiation of translation from the IRES (internal ribosomal entry site) of cp strains due to mutations in this region [9, 28, 66] , though some have disputed this interpretation [32, 38] .

Apoptosis is observed following infection by a number of picornaviruses, especially when virus replication is restricted [1, 10, 60, 61] . While cell killing by apoptosis may aid in spreading progeny virus or escape cellular defense mechanisms [51] , premature induction of apoptosis may limit virus growth and spread [4, 10, 51] . The cytopathic effect that is observed in cells following infection with cp strains of HAV has been attributed to the induction of apoptosis [7, 29] . Apart from the demonstration of a link between rapid replication and cytopathogenicity of the cp strains of HAV, the actual molecular events leading to apoptosis following infection with cp strains are unknown. Expression of P2 region proteins 2B and 2C from either cp or cc strains in FrhK4 cells result in the accumulation of a tubularvesicular network [29, 37] . Expression of the P3 region protein 3AB has been shown to result in membrane alterations in bacterial cells with resultant changes in membrane permeability [50] . The lack of correlation between apoptosis and expression of either P2 or P3 region proteins prompted us to investigate alternative mechanisms of apoptosis induction by cp strains of HAV.

In this study, we report that infection of FrhK4 cells with the HAV cp strain HM175/18f results in the degradation of ribosomal RNA (rRNA) and the reduction of several cellular mRNAs including β-actin and GAPDH. Ribosomal RNA degradation was specific for the 18f strain and the fast replicating enterovirus coxsackievirus B1 (CBV1), while the parental cc strain of 18f did not induce rRNA cleavage. The effect of 18f or CBV1 infection on rRNA integrity was observed in the absence of interferon (IFN) induction. Ribosomal RNA degradation was detected early after virus infection and prior to induction of cpe. This degradation was blocked by inhibitors of virus replication. The pattern of rRNA degradation suggests the involvement of the dsRNA activated RNase L; the regulation of this enzyme could provide an alternative mechanism by which 18f infection induces apoptosis.

FrhK4 monkey kidney cell line was a kind gift of Dr. G. Kaplan (FDA, Center for Biologics Evaluation and Research, Bethesda, MD). The cells were grown in Eagles Minimal Essential Medium (EMEM, Life Technologies, Gaithersburg, MD) containing 5 % heat inactivated fetal bovine serum (FBS), MEM non-essential amino acids and sodium pyruvate (all from Life Technologies), with routine weekly sub-culturing. Under these conditions the cells increase in number about 20-fold in 6 to 7 days. The cell line is contact inhibited and can be kept in growth medium or maintenance medium (1 % FBS) for several weeks without degeneration of the monolayer. HAV strains HM175/18f and HM175/clone 1 [42, 68] , and Human coxsackievirus B1 (CBV1) were obtained from ATCC. All three viruses were grown in FrhK4 cells. Virus stocks were prepared essentially as described [7, 29] . 18f and CBV1 were titered by plaque assay, while clone 1 virus was titered by an enzyme immunoassay (EIA) in 96 well plates as described previously [31] using the HAVAB EIA Diagnostic Kit (Abbott Laboratories, Abbott Park, IL).

Confluent cultures of FrhK4 cells in 75 cm 2 flasks (5 × 10 6 cells) were infected with 5-10 pfu/cell of 18f or CBV, or 15 TCID 50 of clone 1 in 1.5 ml of MEM containing 1 % heat inactivated FBS or mock infected. After 1 h of adsorption, 13.5 ml of the same medium was added to each flask and incubation continued until the indicated times of cell harvest. Cells were harvested by scraping, centrifuged to remove medium, and the pellets washed once with PBS (Ca ++ and Mg ++ free) and either stored frozen at −70 • C or used immediately. For RNA isolation, cell pellets were dissolved in TRIZOL (Life Technologies) containing 100 mM β-mercaptoethanol (Sigma), following the protocol suggested by the manufacturer. All RNAs were routinely digested with RNase-free DNase (Promega, Madison, WI) in the presence of ribonuclease inhibitor (RNasin, Promega) for 30 min at 37 • C, followed by phenol/chloroform extraction and alcohol precipitation. RNAs were quantitated by UV absorption at 260 nm.

Infected or mock infected cells were treated with 50 µg/ml cycloheximide, 2.5 mM Guanidine hydrochloride, or 5 µg/ml of Brefeldin A in MEM containing 1 % FBS immediately following the virus absorption period. For measurement of virus replication, 12 well plates containing approximately 2.5 × 10 5 cells were used. Cells were fixed in methanol/PBS at 24 hpi and viral antigens detected by EIA [31] . For RNA isolation, 75 cm 2 flasks, containing approximately 5 × 10 6 cells at the time of infection were harvested at 24 or 48 hpi, and RNAs isolated as described above. 

A 25 cm 2 flask was seeded with FrhK4 cells and infected at approximately 80 % confluence with clone 1 virus as described above. The virus was allowed to replicate for 7 days and thereafter the cells were trypsinized and seeded into a 75 cm 2 flask. Subsequently, the flask was sub-cultured at weekly intervals by trypsinization and seeding a new flask with one tenth of the cell suspension. To monitor virus replication, periodically cells were seeded into 12 well plates and cultured for 24 to 48 h until the plates were confluent. Cells were fixed in 80 % methanol and viral antigens were measured by the HAVAB EIA procedure [31] . Uninfected cells were used as a negative control.

Samples of RNA (8 µg unless otherwise indicated) were denatured by heating at 55 • C for 15 min in 20 µl of buffer containing 1× MOPS buffer, 50 % formamide and 6.7 % formaldehyde. Samples were separated in a 1 % agarose gel containing 1× MOPS buffer, 0.41 M formaldehyde, and 1 µg/ml EtBr [16] . The gel was photographed under UV light and transferred overnight to Nytran N membranes (Schleicher & Schuell) in 20× SSC. The membrane was washed briefly in 2× SSC, baked under vacuum and stored at 4 • C prior to hybridization.

Probes for β-actin and GAPDH were labeled by random priming of DECAtemplate TM -βactin-mouse and DECAtemplate TM -GAPDH-mouse respectively (AMBION, Austin, TX), using α-32 P dATP (3000 Ci/mmol, ICN, Costa Mesa, CA). Probe for 18S rRNA was DECAtemplate TM -18S-mouse, also obtained from AMBION and was similarly labeled. The anti-sense RNA probe for the 28S rRNA (pTRI RNA 28S, Ambion) was labeled using SP6 polymerase and α-32 P UTP. The probe for HAV genomic RNA was obtained by Xba 1 digestion of the plasmid pHAV/7 (kindly provided by Dr. S. Emerson NIH, Bethesda, MD). Following gel purification, the large Xba 1 fragment, representing almost all of the genomic RNA (except the IRES), was labeled by nick translation using α-32 P dATP. Blots were prehybridized at 65 • C in Hyb-9 (Gentra) supplemented with denatured salmon sperm DNA or yeast tRNA (for the RNA probe) for 3-4 h with continuous agitation and hybridized with probe (2-3 × 10 6 cpm/ml) overnight. Filters were washed once in 2× SSC/1 % SDS for 30 min and then 3 times in 0.1× SSC/0.1 % SDS for 30 min each time at 65 • C, prior to exposure to film or phosphorimager screen (Molecular Dynamics).

Anti-VP1 antiserum was obtained from rabbits immunized with a synthetic peptide CVPETF PELKPGESRHT, corresponding to amino acids 64-79 of HAV viral capsid protein VP1, covalently linked to KLH. Synthetic peptide, free and KLH-conjugated, was purchased from New England Peptide, Inc (Fitchburg, MA). Pre-immune serum was collected from New Zealand white rabbits 1-2 days prior to immunization and stored at −70 • C. Rabbits were immunized with KLH-conjugated peptide emulsified in Freund's complete adjuvant and challenged at regular intervals, over a 14 to 19 week period, with KLH-conjugated peptide emulsified in incomplete Freund's adjuvant. The animals were hyperimmunized using unconjugated peptide resuspended in sterile PBS. The serum was collected and stored at −70 • C. All animal studies were conducted in compliance with the Guide for the Care and Use of Laboratory Animals under an approved IACUC protocol.

Culture media was removed from mock or virus infected cell cultures and the monolayer were scraped into Ca ++ and Mg ++ free phosphate-buffered saline (PBS). Cells were pelleted (1500 × g, 4 • C) and washed twice with cold PBS. To obtain a total cell lysate, the cell pellet (while on ice) was resuspended in cold RIPA buffer (50 mM Tris-Cl, pH 8.0, 150 mM NaCl, 1 % NP-40, 0.5 % sodium deoxycholate, 0.1 % SDS) containing 1 mM PMSF, and 1:100 dilution of protease inhibitor cocktail (Sigma St. Louis, MO) and phosphatase inhibitor cocktails I and II (Sigma). Additional PMSF was added (1 mM) to the lysate prior to its passage through a 20 gauge needle (8-10 times). Following 30 min of incubation on ice, the sample was centrifuged at 10,000 × g for 10 min 4 • C to pellet insoluble material. The extracts were stored as aliquots at −20 • C. Protein concentrations of lysates were estimated using the BCA-200 Protein Assay (Pierce Chemical Co., Rockford, IL) according to the manufacturer's instructions.

Sixty five µg of each protein extract were adjusted to equivalent volumes in denaturing sample buffer and heated at 95 • C for 5 min and subjected to electrophoresis in SDS-10 % polyacrylamide gels followed by electrophoretic transfer onto supported nitrocellulose membranes (Optitran, BA-S 83, Schleicher and Schuell Inc.) in transfer buffer (25 mM Tris base, 192 mM glycine) containing 20 % methanol. The membranes were blocked with TBS-T (Tris buffered saline/0.1 % Tween 20) containing 5 % nonfat dried milk (NFDM) for 90 min at room temperature and washed four times with TBS-T. Using a mini-blotter manifold, membranes were incubated either overnight at 4 • C with anti-STAT1 and anti-P-STAT1 antibodies (Cell Signaling Technologies, Beverly, MA) diluted 1:1000 in TBS-T containing 5 % BSA or 1 h at room temperature with anti-actin antibody (Sigma) and anti-VP1 antibody diluted 1:1000 in TBS-T containing 5 % NFDM. After four washes in TBS-T, the blots were incubated with goat anti-rabbit IgG-HRP antibody conjugate (Cell Signaling) diluted 1:2,000 in TBS-T containing 5 % NFDM. Following four washes at room temperature in TBS-T, the SuperSignal West Pico Chemiluminescent Substrate kit (Pierce Chemical Co. Rockford, IL) was used according to manufacturer's instructions followed by exposure to X-OMAT-XAR (Kodak, Inc.) for the detection of bound antibody-HRP conjugates.

Cells were grown in 25 cm 2 flasks and infected when confluent with 18f as above and labeled at 72 hpi for 2 h with 50 µCi/ml of a mixture of 35 S-methionine and 35 S cysteine (ICN) in methionine and cysteine-free medium supplemented with 5 % dialyzed FBS (ICN) following a 30 min pre-incubation in methionine and cysteine-free medium. CBV1 infected cells and clone 1 persistently infected cells were labeled similarly at 7 h or 60 days post-infection (pi), respectively. Cells were collected by scraping, washed once with PBS and the pellets lysed in 200 µl RIPA buffer containing protease inhibitors cocktail (Sigma). TCA precipitable incorporation was measured using 5 µl of clarified lysate and protein concentration was measured with 2 µl lysate. Labeled proteins were separated in a 4-20 % Novex pre-cast gel (Invitrogen, Carlsbad, CA), after heating the samples at 95 • C for 3 min in 1× sample buffer containing dithiothreitol. Following electrophoresis, the gel was fixed and stained with Brilliant Blue G (Sigma) in 25 % methanol, 5 % acetic acid and destained in 25 % methanol, 5 % acetic acid. The gel was then treated with Amplify (Amersham Biosciences, Piscataway, NJ) for 30 min, dried and exposed to a phosphorimager screen, which was scanned in a Typhoon scanner (Molecular Dynamics).

Total RNA preparations from mock infected, 18f infected (48 hpi), CBV1 infected (6 hpi), and HM175 infected (60 dpi) FrhK4 cells were labeled with α-32 P dATP using the CDS primers for human 1.2 array (Clontech, Palo Alto, CA), Clontech enzyme and reagents provided with the array. The labeled cDNA preparations were purified using the Nucleospin extraction kit (Clontech). Four identical arrays were pre-hybridized at 68 • C in hybridization bottles in a HYBAID oven (National Labnet, Woodbridge, NJ) in 10 ml of ExpressHyb (Clontech) containing 100 µg/ml denatured sheared salmon sperm DNA (Eppendorf Scientific, Westbury, NY). Hybridization was performed overnight at 68 • C by the addition of denatured probe (1.5 × 10 6 cpm/ml) and 10 µg human Cot-1 DNA (Clontech) directly to the pre-hybridization solution. The membranes were washed twice (30 min each) with 2× SSC-1 % SDS and twice (30 min each) with 0.1× SSC−0.5 % SDS at 68 • C, followed by a final wash at room temperature with 2× SSC for 5 min. The membranes were wrapped in Saran wrap and exposed to phosphorimager screens at room temperature for 24 to 48 h. The screens were scanned in a Molecular Dynamics Typhoon scanner at 100-micron resolution and analyzed using the AtlasImage 1.5 software (Clontech).

1 µgRNA samples were reverse transcribed in 20 µl reactions using oligo (dT) 15 as the primer and AMV reverse transcriptase and 5 µl of each cDNA pool were PCR amplified using primers BG 7 and BG 8 as described [31] . Ten µl samples were withdrawn after 27 and 30 cycles and analyzed by agarose gel electrophoresis. Bands were quantitated by the ImageQuant program from Molecular Dynamics following scanning of the EtBr stained gel by a Typhoon phosphorimager.

At a moi of 5-10 pfu/cell, the rapidly replicating cp strain HM175/18f of HAV (18f) induced cell rounding in FrhK4 cells within 48 h post-infection (pi) in approximately 10 to 15 % cells. The percentage of cells exhibiting cpe increased for the next 48 h, so that by 96 h, 70 to 80 % of the cells were rounded and detached from the surface of the culture flask. Thus, even with the 18f virus, infection is characterized by a slow and asynchronous recruitment of cells into supporting viral replication. Similar effects have been reported for other cp strains of HAV, namely HM175/24a, and HB1.1 [7, 29] . In agreement with these results, we confirmed the induction of apoptosis in 18f infected FrhK4 cells by a colorimetric TUNEL assay (data not shown). The parental HM175/clone 1 virus (clone 1), also used in the present study, produced no changes in cell morphology after several months of continuous culture of infected cells, although viral antigens could be detected by EIA after four weeks of weekly subculture following the initial infection. The coxsackie virus strain B1 (CBV1) used in this study replicated very rapidly in FrhK4 cells, producing marked cpe within 7 hpi and all cells were floating by 14 to 15 hpi.

We observed the appearance of RNA bands in the region between the 28S and 18S species, as well as below the 18S rRNA band, upon denaturing agarose gel electrophoresis of total RNA preparations isolated from 18f-infected cells (Fig. 1) . The apparent degradation products were easily detectable after ethidium bromide (EtBr) staining of formaldehyde-agarose gels (Fig. 1A) . RNA degradation was observed only in 18f infected cells at 48 and 96 hpi (lanes 3 and 4), while no degradation was observed at 96 hpi in clone 1 infected cells (lane 2) or mockinfected cells (lane 1). Degradation of rRNA has not been reported previously in HAV infected cells, therefore, the rRNA origin of the observed degradation The probe for 18S RNA was a 1.2 kb fragment of the mouse 18S RNA gene (nucleotides 505 to 1694), labeled by random priming with α-32 P dATP. The probe for HAV genomic RNA was the large Xba I fragment from a genomic clone pHAV/7, labeled by nick translation with α-32 P dATP. An RNA ladder (0.24 to 9.5 kb) was included in the analysis and the position of the 1.35 to 7.4 kb fragments identified in panel A. An arrow is used to identify the putative 18S rRNA degradation product in A and the position of the 18S RNA and its degradation products in B. Full length HAV RNA is 7.5 kb (C) products was confirmed by Northern blot analysis. Duplicate RNA samples were electrophoresed and transferred to Nytran N membranes. One half of the blot was hybridized to a probe for 18S RNA (Fig. 1B) , and the other half to a probe for HAV RNA (Fig. 1C) . The 18S probe hybridized to intact 18S RNA and to two additional species of RNA (indicated by arrows) from 18f infected cells, while these species were not detected in RNA preparations from either mock or clone 1 infected cells.

Apparently full length HAV RNA (7.4 kb) was detected in 18f infected cells at both 48 and 96 hpi, while no viral RNA was detected in clone 1 infected cells at 96 hpi, confirming that the 18f virus indeed was a faster replicating strain.

To further investigate whether RNA degradation was specific for 18f infection, and to determine the temporal relationship between RNA degradation and onset of cpe, we isolated total RNA from cells infected with CBV1, HM175/clone 1, and HM175/18f at different times, and analyzed them by formaldehyde-agarose gel electrophoresis (Fig. 2) . In CBV1 infected cells, degradation of both species of rRNA was observed after CBV1 infection at 6-7 hpi, when almost all the cells in the monolayer were rounded but still attached to the surface ( Fig (Fig. 2B, lane 3) , though small amounts of degradation products may be detected at 60 days following infection (Fig. 2B,  lane 5) . The levels of viral RNA and protein in clone 1 infected cells at 35 and 60 days pi were comparable to those observed in 18f infected cells, as measured by RT-PCR and Western immunoblotting (Fig. 5 ) and EIA (data not shown). We have continuously cultured the persistently clone 1 infected cells for over five months without significant changes in cell morphology or rRNA integrity. When RNA was obtained from 18f-infected monolayers following separation of the rounded cells from cells with a normal morphology by vigorous agitation, rRNA degradation was apparent in both populations of cells (Fig. 2B, lanes 6 and 7) . In lane 6, only 3 µg of RNA was applied, since the percentage of floating cells at 48 hpi was low and the yield of RNA much less compared to the cells that remained attached (Fig. 2B, lane 7) . In lane 6, no intact 28S or 18S bands were visible, indicating that almost complete degradation of the rRNA species has occurred in these cells. It should be noted that CBV1 infection proceeds much faster and synchronously in FrhK4 cells. Since we did not examine rRNAs from CBV1 infected cells between 4 and 6 hpi, it is quite possible that RNA degradation occurs in CBV infection prior to the onset of morphological changes associated with cpe.

To confirm that 28S rRNA degradation was occurring after infection with 18f or CBV1, a Northern blot of the RNA gel shown in Fig. 2 (panel A) was hybridized to a probe specific for 28S rRNA (Fig. 3) . The short probe (115 bases from the 3 end of the 28S RNA) hybridized strongly to a fragment migrating faster than the 1.35 kb marker, as well as to intact 28S rRNA, but did not hybridize to apparent Fig. 2A were subjected to denaturing agarose gel electrophoresis. The gel was Northern blotted and hybridized to a labeled anti-sense RNA probe specific for 28S rRNA (nucleotides 4515 to 4400). The probe was synthesized from pTRIRNA-28S by transcription with the SP6 polymerase. The arrow indicates the position of a 28S RNA degradation product. B An identical Northern blot was hybridized to a nick translated DNA probe for HAV as described in Fig. 1 . Arrows indicate position of full-length HAV RNA (7.5 kb) and putative degradation products degradation products seen in EtBr stained gels (Figs. 1 and 2 ) that migrated to the region between the 28S and 18S species. It is likely that these latter products were derived from the 5 end of the 28S rRNA and could not be detected by a probe targeted to the 3 end. Nevertheless, it is clear that both the large and small rRNA molecules were degraded in CBV1 and 18f infected cells, and that this degradation coincided with the appearance of cpe in CBV1 infected cells but preceded the appearance of cpe in 18f-infected cells. In contrast, infection of permissive cells with mouse hepatitis virus (a coronavirus) resulted in the degradation of only the 28S species [3] . Apparently intact HAV RNA was detected in 18f-infected cells at 48 hpi and 96 hpi, and, albeit at lower levels in clone 1 infected cells at 10 dpi (Fig. 3B ). This blot was exposed longer to reveal the existence of HAV RNA in clone 1 infected cells. While overexposure of the blot showed that some degradation of viral RNA might also be taking place, the bulk of the viral RNA remains intact, possibly protected from degradation by its association with capsid proteins. On the other hand, the smaller virus specific RNA molecules could have resulted from premature termination of transcription [48] .

To investigate whether the degradation of rRNAs observed in 18f infected cells required replication of the virus, we studied the effects of several inhibitors of virus replication on RNA degradation. Cyclohexamide (CH) is a strong inhibitor of protein synthesis in many cultured cell lines and produces a marked cpe in treated cells that is attributed to induction of apoptosis [1, 7] . In agreement with the results of Brack et al. [7] , we observed 18f-like cpe after treatment of FrhK4 cells with CH at concentrations ranging from 10 to 50 µg/ml. Despite the induction of cpe/apoptosis, CH treatment failed to elicit RNA degradation in uninfected FrhK4 cells (Fig. 2B, lane 8) , and inhibited both virus replication (Fig. 4A) , and degradation of RNA in 18f infected cells (Fig. 2B, lane 9 ). The inhibition of virus replication was not unexpected since CH inhibits picornavirus genome replication by preventing translation of input genome to produce viral RNA dependent RNA polymerase [1] . Brefeldin A (BFA) has been shown to induce apoptosis in other cell lines through the activation of caspase and disruption of Golgi complex function, and to inhibit replication of HAV and some picornaviruses, as well as viral RNA synthesis [5] . Treatment of FrhK4 cells with 5 µg/ml BFA produced marked apoptosis within 24 h while suppressing 18f replication and 18f induced RNA degradation (Fig. 4 , panels A and B respectively). Thus, RNA degradation in 18f-infected cells was prevented by both CH and BFA. CH and BFA induced apoptosis in FrhK4 cells was synchronous and was complete by 24 h as evidenced by cell rounding and eventual detachment from the growth surface. However, these changes were not accompanied by RNA degradation following treatment with either CH (Fig. 2B) or Brefeldin A (data not shown). Treatment of 18f-infected cells with 2.5 mM guanidine hydrochloride (GuHCl), on the other hand, had a marginal effect on 18f replication at high moi, and a somewhat larger effect at low moi (Fig. 4A) . However, GuHCl treatment did not prevent RNA degradation in 18f-infected cells (Fig. 4B ). It has been reported that GuHCl, an inhibitor of poliovirus replication, is ineffective against HAV due to a mutation in the genome [18] , but GuHCl was found to be a strong inhibitor of HM175/p39 virus [15] . These experiments clearly show that in FrhK4 cells, replication of the 18f virus is required for RNA degradation, and that degradation of RNA is not a consequence of apoptosis.

Degradation of rRNA is a feature of virus infection in interferon (IFN) treated cells and is believed to be due to the availability of double stranded RNA (dsRNA) during replication or transcription of the viral genome, resulting in the activation of the RNase L pathway [58, 59] . However, in the absence of IFN pretreatment, dsRNA can activate the RNase L pathway by activating pre-existing 2-5A synthetase, which ultimately results in cell death [35, 36, 52] . Since viral RNA replication and expression occurs much earlier post infection in 18f-infected cells compared to clone 1 infected cells, the degradation of rRNAs could result from an early dsRNA dependent activation of RNase L in 18f-infected cells. We further investigated whether the lack of RNA degradation and cpe in clone 1 infected cells observed at later times pi may correlate with diminished levels of viral RNA or protein. FrhK4 cells were infected with HM175/clone 1, and the cells were serially passaged at weekly intervals until viral antigens were detectable by EIA (data not shown). After 35 days, the level of viral antigens was comparable to that observed in 18f infected cells after an acute infection. Total cellular RNA, and proteins were isolated from the persistently infected cells. The level of viral RNA in the persistently infected cells was measured at 35 and 60 days of continuous culture by RT-PCR (Fig. 5A) . The level of viral RNA was somewhat lower in clone 1 infected cells after 35 days, but at 60 days the levels were almost the same as in 18f infected cells at 48 hpi. PCR amplification was carried out for 27 and 30 cycles to ensure that the signal strengths of the PCR products were obtained before the reactions reached saturation (27 cycles) . Western blot analysis with anti-VP1 antibody also showed lesser but easily detectable accumulation of the viral capsid protein VP1 in persistently clone 1-infected cells (Fig. 5B) . However, the cells did not show any discernable morphological change after more than 120 days of weekly subculture and produced viral antigens and RNA as determined by EIA and RT-PCR. The data indicate that, unlike 18f infection, clone 1 replication and expression does not induce rRNA degradation.

A reduction in cellular RNA and protein levels during virus infection may aid in the replication of viruses possessing a cp phenotype (e.g., HAV HM175/18f), or a rapid growth/cpe characteristic (e.g., CBV) in cell lines permissive for virus growth. For example, translation of HAV RNA in infected cells requires interaction of trans-acting cellular protein factors with cis-acting elements present at the 5 [13, 40] . Indeed, a competition between GAPDH and polypyrimidine tract-binding protein (PTB) for stem-loop IIIa of HAV IRES has been reported by Yi et al. [67] . While PTB functions as a positive effector of HAV RNA translation, the binding of GAPDH to stem-loop IIIa of HAV IRES has been shown to destabilize RNA secondary structure and inhibit translation [55, 67] . These results prompted us to investigate the levels of GAPDH mRNA in 18f-infected cells since decreased levels of this mRNA may indicate a functional relationship between 18f replication and regulation of effector proteins such as GAPDH. As a control, we also investigated whether virus infection affects the level of another cellular housekeeping gene β-actin. Northern blot hybridizations to β-actin and GAPDH specific probes were used to assess the levels of these two cellular mRNAs (Fig. 6) . The levels of both mRNAs were reduced in a time dependent manner in CBV1 and 18f infected cells while there was little effect of clone 1 infection on these molecules. However, unlike the rRNAs, we did not find any degradation products of these RNAs. Two reasons for this could be that these mRNAs are transcriptionally repressed and/or the degradation products are too small to detect by this analysis. 

The degradation of rRNA into discrete products is a hallmark of the antiviral effect of interferon. Since the cell line used for these experiments was not treated with interferon prior to virus infection, it was important to determine whether the FrhK4 cell line is a natural producer of interferon, or if interferon is induced as a result of virus infection. To investigate if interferon mRNAs were present in FrhK4 cells prior to or following virus infection, nylon arrays, each containing 1176 human genes including those for interferons α, β, and γ were hybridized to cDNA probes synthesized using total RNA from uninfected and virus infected cells as templates. Surprisingly, we did not find significant hybridization signals Table 1 . Induced and down-regulated genes during 18f infection of FrhK4 cells. Array hybridizations were carried out in duplicate with cDNA probes prepared using two independent RNA preparations each from mock and virus infected FrhK4 cells. Independent phosphorimager scans for each duplicate sample (control or infected) were averaged using AtlasImage program to produce composite arrays for control and infected cells. The composite arrays were then compared to detect differences in signal levels using AtlasImage software.

Gene signal values that were less than two-fold over background values were considered insignificant. Ratios of less than one represent down-regulation in the infected cells. Ratios of higher than one represent up-regulation in the infected cells. A less than two-fold difference of signal intensity between control and infected samples are considered insignificant. ND means the pixel value for that gene was equivalent to background corresponding to the interferon genes in either uninfected or HAV infected cells (Table 1) . These results are supported by the data from Pietiäinen et al. [49] suggesting that IFN mRNAs are neither present nor induced during PV1, EV1 and CBV4 infection of HOS and HeLa cells. However, low levels of IFN RNAs, and any comparative differences between uninfected and infected cells, may not be detectable by array hybridization. RT-PCR analysis of total RNA from virus infected FrhK4 cells did not show increased level of β-IFN mRNA compared to uninfected cells. However, treatment of these cells with dsRNA resulted in β-IFN mRNA induction (data not shown). We investigated the possibility that low levels of IFN may be present in control or infected FrhK4 cells by looking at the phosphorylation status of STAT1 by Western immunoblotting (Fig. 7) . Signal transducer and activator of transcription (STAT) 1 and 2 are regulators of IFNinduced gene expression. In cells treated with IFN, both STAT1 and STAT2 are phosphorylated by Janus kinase (JAK), which is a prerequisite for dimerization and translocation to the nucleus [58, 59] . The antibody to STAT1 readily detected the two forms of this protein in both control and virus infected cells, as well Volumes of extracts containing equal number of cpm or protein were heated in SDS sample buffer containing reducing agent prior to loading on a 4-20 % polyacrylamide gel that was run in MOPS-SDS buffer along with molecular weight markers. The gel was stained, destained and, and treated with Amplify (Amersham) prior to drying according to standard protocol and exposed to phosphorimager screen, which was scanned by a Typhoon scanner as in control and interferon treated HeLa cell extracts. Both STAT1 proteins in HeLa migrate as slightly larger molecules. Using an antibody specific for the phosphorylated STAT1, STAT1 phosphorylation was not detected in either control or 18f-infected FrhK4 cells (Fig. 7) , confirming the results obtained by array hybridization. In addition, STAT1 phosphorylation was not observed in CBV1 infected FrhK4 cells while phosphorylated STAT1 was readily detected in IFN treated HeLa cells (Fig. 7, panel A) , and FrhK4 cells (Fig. 7, panel B) . These results are in agreement with those obtained by array hybridization, indicating that IFN is not expressed and, therefore, is not mediating RNA degradation in 18f-infected cells.

Unlike most enteroviruses, even cp strains of HAV lack the ability to shut off host protein synthesis. This inability to cause host shut-off is most likely due to a non-functional 2A protease, and is probably a major reason for the slow growth phenotype of HAV [34] . While it would contradict published reports, it seemed possible that 18f induced RNA degradation could result in host shut-off triggered by events peculiar to FrhK4 cells. Therefore, we investigated the effect of 18f infection on overall protein synthesis by pulse labeling of cells for 2 h and analysis of the products by SDS-polyacrylamide gel electrophoresis (SDS-PAGE). As shown in Fig. 8 , CBV1 infection resulted in a dramatic reduction of cellular protein synthesis at 7 hpi, and viral proteins could be clearly identified. In contrast, the effect of 18f infection was much less pronounced, and comparable to the effect of the parental cc (clone 1) strain. Thus no positive identification of HAV specific proteins could be made from such an analysis. It should be noted that the clone 1 infected cells used for protein labeling in this experiment have been routinely sub-cultured for more than two months, and were actively producing viral antigens as detected by a colorimetric EIA procedure (data not shown) as well as viral RNA and VP1 capsid protein (Fig. 5) . However, RNA degradation was not evident in these cells.

In conclusion, rRNA degradation alone is not sufficient for host shut-off.

Cytopathic (cp) strains of HAV have been isolated in several laboratories during the past few years. The cp strains arise during repeated passage of cc strains in permissive cells and the genetic basis for the development of the cp phenotype is poorly understood [2, 7, 21, 25, 47, 62, 68] . There is no evidence to indicate conclusively which nucleotide changes arising in the cp genotype are responsible for both the rapid growth and the cp phenotype [23, 26, 29, 38, 42, 68] . These characteristics may be the result of different mutational events that have occurred during cc strain replication. The data presented here identify a previously unreported event that occurs during infection with a cp strain of HAV, namely the degradation of rRNAs and the reduction in the level of cellular RNAs. Similar effects on rRNA, and β-actin and GAPDH mRNAs were also observed after CBV1 infection. However, the non-cpe inducing parent strain of 18f, HM175/clone 1 had only a marginal effect on rRNA integrity during long-term persistent infection. Similar to the effects of enteroviruses, such as PV1, CBV4 and EV1, on the expression of certain cellular genes involved in apoptotic pathways, c-myc and c-jun levels were moderately increased in 18f-infected cells, while c-jun level was markedly reduced in clone 1 infected cells (Table 1) . Full length viral RNA was detectable much earlier in 18f-infected cells compared to cells acutely infected with the clone 1 virus (Figs. 1 and 3) . Significantly, viral RNA and viral capsid protein VP1 were present at comparable levels following extended culture of clone 1-infected cells in the absence of significant RNA degradation (Fig. 5) .

Degradation of rRNA following infection of interferon treated cells has been reported for a number of DNA and RNA viruses [58, 59] . Moreover, dsRNA, a likely intermediate of RNA virus replication or transcription, has been shown to promote rRNA degradation without interferon treatment in some cell lines [6, 12, 35, 36, 45, 52, 53, 64, 69] . The function of dsRNA as an effector of the interferon response is dependent on both the cell type and the nature of the virus [58, 59] . For example, vaccinia virus is known to evade the effects of IFN treatment, but in suspension cultures of mouse L929 cells treated with IFN, rapid degradation of rRNA was observed [30] . Activation of the 2-5A/RNase L pathway has been reported during estrogen withdrawal from chick oviducts [19] . The enzyme responsible for the synthesis of 2-5A, 2 -5 -oligoadenylate synthetase can apparently be also induced by hepatitis C virus coat protein [46] . While the role of a viral protein in the activation of 2-5A/RNase L pathway in 18f or CBV1 infected FrhK4 cells cannot be ruled out, previous reports with EMC virus and the studies involving dsRNA clearly suggests a similar mechanism of rRNA degradation reported here. It remains unclear why rRNA degradation and apoptosis are absent in clone 1 infected cells despite the presence of viral RNA and protein. Perhaps cellular inhibitors of RNase L play a role in blocking RNA degradation in clone 1 infected cells [44] . It is also possible that a mechanism, analogous to the adenovirus VAI RNA mediated inhibition of PKR activity is used by the clone 1 virus to disarm the RNase L pathway [58, 59] . Clearly, accumulation of sufficient viral RNA or proteins cannot account for either rRNA degradation or the cp phenotype, since cc strains accumulate enough viral RNA and proteins after an initial lag (Fig. 5) . It is interesting to note that while the precursor of mature VP1 protein VP1-2A is present in equal amounts in both 18f and clone 1 infected cells, VP1 itself accumulates to higher levels in 18f infected cells, suggesting a slower processing of the precursor in clone 1 infection. Surprisingly, we did not see a general inhibition of protein synthesis in FrhK4 cells infected with the cp (18f) strain of HAV despite massive RNA degradation, although an almost total shutoff of host protein synthesis was observed in CBV infected FrhK4 cells (Fig. 8) . This is in contrast to the general inhibition of protein synthesis observed upon activation of the RNase L pathway in some cell lines [11, 12, 30, 35, 36, 52, 69] . The effect of CBV infection on protein synthesis is most likely a combination of rRNA cleavage (this study) and cleavage of eIF4G, poly (A)-binding protein (PBP), and other translation factors mediated by an active 2A protease [10, 39, 41] .

It has been shown that both caspase activation and apoptosis induction can occur when 2 -5 A synthetase is activated and rRNA degradation is induced [11, 12, 24, 52, 69] . While we have not directly assessed the level of 2-5A following virus infection, no other RNase currently known produces the characteristic degradation pattern of rRNA that we have observed. It is apparent from our data, however, that RNA degradation is dependent upon cp virus strain replication, since compounds, such as cycloheximide and Brefeldin A, that inhibited 18f replication also abolished rRNA degradation in 18f infected cells (Figs. 2 and 5) . Treatment in the absence of infection did not induce RNA degradation. Additionally, the treatment of uninfected cells with these compounds induced apoptosis in the absence of RNA degradation (Figs. 2 and 4) . In view of the temporal relationship between onset of rRNA degradation and induction of cpe/apoptosis, these results suggest that RNA degradation following 18f (or CBV1) infection is the cause, rather than the effect, of apoptosis. On the other hand, an apoptotic reaction may also be induced by the preferential synthesis of apoptosis inducing proteins such as c-jun [20] or c-myc [22] . As shown in Table 1 , several gene transcripts linked to apoptosis are present in higher levels in 18f-infected cells compared to clone 1 infected cells. Particularly intriguing are lack of induction of c-myc and downregulation of c-jun in clone 1 infected cells. We cannot rule out the possibility, therefore, that apoptosis in 18f-infected cells may be induced via regulation of relevant proteins such as c-jun or c-myc.

Another interesting feature of the cp strains of HAV is that all apoptosis inducing cp strains are derived from persistently infected monkey kidney cells and induction of apoptosis (or cpe) is restricted to monkey kidney cells. Therefore, the nature of the host cell is clearly as important as the genetic makeup of the virus. It is interesting to note that a 14 base insertion is present in the IRES of cp strains HM175/18f as well as HM175/24a. It would be tempting to speculate that conformational changes brought about by this or other mutations result in a better inducer of 2 -5 A synthetase, a possibility that is currently under investigation.

Activation of RNase L resulting in rRNA cleavage is usually accompanied by inhibition of protein synthesis [6, 12, 30, 52, 69] and picornavirus growth [14, 53, 69] . We believe that the apparent lack of inhibition of host protein synthesis in 18f-infected cells is a result of asynchronous RNA degradation, which masks the effect of RNA degradation on protein synthesis. The other possibility is the absence of activation of dsRNA stimulated PKR, which may be lacking in these cells. However, several different mechanisms exist, through which viruses can disarm the PKR response [58, 59] . Further studies are required to assess the level of activity of this enzyme in virus-infected cells.

Recent studies using DNA arrays have revealed that genes involved in the interferon pathway (interferon response genes, ISGs) are the major targets of a diverse group of viruses [27] . A direct effect of interferon in inducing 2 -5 A synthetase during 18f or CBV infection is unlikely because we failed to detect IFN mRNA or STAT1 phosphorylation in either uninfected or virus infected cells. These results are supported by previous findings that cell culture adapted HAV does not induce IFN in cultured cells [63] . Moreover, recent studies also have shown that a non-cytopathic cc strain of HAV down regulates dsRNA-induced transcription of β-IFN [8] . However, there are now several studies to show that ISGs can be regulated in the absence of IFN by viral dsRNA or viral proteins [6, 45, 64] . While many viruses have evolved mechanisms to disarm various facets of the IFN inducible antiviral response, the usurpation of the IFN response by the cp strain 18f to facilitate it's spread is unique. The fact that only a few cell lines of primate kidney origin are susceptible to apoptosis in response to rapidly replicating HAV variants suggests that both cellular processes and genotypic changes in the viral genome are responsible for the establishment of a cp phenotype. In order to define the contributions of these two processes, it will be necessary to examine strain differences at the genetic level as they relate to substantive changes in viral protein expression/function, and to investigate the connection between viral and cellular events at the level of cellular protein expression.

In conclusion, rRNA degradation in combination with lower levels of cellular mRNAs such as β-actin, GAPDH and others could confer on the 18f virus a replicative advantage over the parental cc strain. Clearly, reduction in the levels of cellular mRNAs is advantageous to the virus, even if it leads to a small reduction rather than a dramatic shut-off of host protein synthesis.

Competing death programs in poliovirus-infected cells: commitment switch in the middle of the infectious cycle

Cytopathology, plaque assay, and heat inactivation of hepatitis A virus strain HM175

RNase-L-independent specific 28S rRNA cleavage in murine coronavirus-infected cells

Poliovirus protease 3C pro kills cells by apoptosis

Infection of polarized cultures of human intestinal epithelial cells with hepatitis A virus: vectoral release of progeny virions through apical cellular membranes

Differential kinetics of polypeptide expression and different biological activities in the human fibroblast response to dsRNA or interferon treatment

A cytopathogenic, apoptosis inducing variant of hepatitis A virus

Hepatitis A virus inhibits cellular antiviral defense mechanisms induced by double stranded RNA

In vitro characterization of an internal ribosomal entry site (IRES) present within the 5 nontranslated region of the hepatitis A virus RNA: comparison with the IRES of encephalomyocarditis virus

Caspase activation and specific cleavage of substrates after coxsackievirus B3-induced cytopathic effect in Hela cells

A study of the interferon antiviral mechanism: Apoptosis activation by the 2-5A system

The role of 2 -5 oligoadenylate-activated ribonuclease L in apoptosis

Cell type-specific proteins which interact with the 5 nontranslated region of hepatitis A virus RNA

Constitutive expression of (2 -5 ) oligo A synthetase confers resistance to picornavirus infection

Rapid completion of the replication cycle of hepatitis A virus subsequent to reversal of guanidine inhibition

One-hour downward capillary blotting of RNA at neutral pH

Vehicular transmission of hepatitis A

Complete nucleotide sequence of an attenuated hepatitis A virus: comparison with wildtype virus

Occurance of 2-5A and RNA degradation in chick oviduct during rapid estrogen withdrawal

Expression and involvement of c-fos and c-jun protooncogenes in programmed cell death induced by growth factor deprivation in lymphoid cell lines

Development of plaque assay for a cytopathic, rapidly replicating isolate of hepatitis A virus

C-Myc target genes involved in cell growth, apoptosis, and metabolism

Replicative events in hepatitis A virus infected MRC-5 cells

Activation of the interferon-inducible enzyme Rnase L causes apoptosis of animal cells

Preliminary characterization of a fast-growing strain of human hepatitis A virus

2B and 2C mutations are essential but mutations throughout the genome of HAV contributes to adaptation to cell culture

Virogenomics: a novel approach to antiviral drug discovery

Hepatitis A virus translation is rate limiting for virus replication in MRC-5 cells

A cytopathic and a cell culture adapted hepatitis A virus strain differ in cell killing but not in intracellular membrane rearrangement

Degradation of rRNA in interferon-treated vaccinia virus infected cells

A polymerase chain reaction-based method for the detection of hepatitis A virus in produce and shellfish

Coding sequences enhance internal initiation of translation by hepatitis A virus RNA in vitro

Prevention and control of hepatitis A

Hepatitis A viruses with deletions in the 2A gene are infectious in cultured cells and marmosets

Activation of P38 mitogen-activated protein kinase and cjun NH2-terminal kinase by double-stranded RNA and encephalomyocarditis virus: involvement of RNase L, protein kinase R, and alternative pathways

Activation of NF-κB by double-stranded RNA (dsRNA) in the absence of protein kinase R and RNase L demonstrates the existence of two separate dsRNA triggered antiviral programs

Membrane permeability induced by hepatitis A virus proteins 2B and 2BC and proteolytic processing of HAV 2BC

Replication of hepatitis A viruses with chimeric 5 nontranslated regions

Cleavage of poly (A)-binding protein by coxsackievirus 2A protease in vitro and in vivo: another mechanism for host protein synthesis shutoff?

RNA-protein interaction at the 3 end of the hepatitis A virus RNA

Cellular factors in the transcription and replication of viral genomes: a parallel to DNA-dependent RNA transcription

Antigenic and genetic variation in cytopathic hepatitis A virus variants arising during persistent infection:evidence for genetic recombination

Maturation of the hepatitis A virus capsid protein VP1 is not dependent on processing by the 3C pro proteinase

The RNase L inhibitor (RLI) is induced by double-stranded RNA

Direct induction of interferongamma-and interferon-alpha/beta inducible genes by double-stranded RNA

Activation of the interferon inducible 2 -5 -oligoadenylate synthetase gene by hepatitis C virus coat protein

Production of cytopathology in FrhK4 cells by BS-C-1 passaged hepatitis A virus

Detection of defective genomes in hepatitis A virus particles present in clinical specimens

Effects of echovirus 1 infection on cellular gene expression

Recombinant expression of hepatitis A virus protein 3A: interaction with membranes

Viruses and Apoptosis

Caspase-dependent apoptosis by 2 -5 -oligoadenylate activation of RNase L is enhanced by IFN-β

Constitutive expression of a 2 ,5 -oligoadenylate synthetase cDNA results in increased antiviral activity and growth suppression

Proteinase 3C of hepatitis A virus (HAV) cleaves the HAV polyprotein P2-P3 at all sites including VP1/2A and 2A/2B

Specific interaction of glyceraldehydes 3-phosphate dehydrogenase with the 5 nontranslated RNA of hepatitis A virus

The structures of picornaviral proteinases

Stability of hepatitis A virus

Viruses and interferons

How cells respond to Interferons

Apoptosis-inducing and apoptosis-preventing functions of poliovirus

Apoptosis in acute and chronic central nervous system disease induced by Theiler's murine encephalomyelitis virus

Isolation and molecular cloning of a fast growing strain of human hepatitis A virus from its double stranded replicative form

Hepatitis A virus infection and the interferon system

HAV 18f induces RNA degradation

Regulation of two interferon inducible human genes by interferon, poly(rI):poly(rC) and viruses

Detection of a genome-linked protein (VPg) of hepatitis A Virus and its comparison with other picornaviral VPgs

Low efficiency of the 5 nontranslated region of hepatitis A RNA in directing cap-independent translation in permissive monkey kidney cells

Functional significance of the interaction of hepatitis A virus RNA with glyceraldehydes 3-phosphate dehydrogenase (GAPDH): opposing effects of GAPDH and polypyrimidine tract binding protein on internal ribosome entry site function

An infectious cDNA clone of a cytopathic hepatitis A virus: genomic regions associated with rapid replication and cytopathic effect

Impact of RNase L overexpression on viral and cellular growth and death

Division of Molecular Biology

We thank Dr. Darcy Hanes for help in preparing the anti-VP1 antiserum, Dr. Joseph E. LeClerc and Dr. Dan Levy for critically reading the manuscript, Dr. C.D. Atreya and Dr.Stephen Feinstone for helpful discussions, and Marcia Meltzer for help with the graphics.