Effect of ribavirin on hepatitis A virus replication in vitro

The effect of ribavirin on fetal Rhesus monkey kidney cells (FRhK-4) acutely or chronically infected with hepatitis A virus was studied. The effect of ribavirin on hepatitis A virus yield as detected by radioimmunoassay in acutely infected FRhK-4 cells was dependent on hepatitis A virus inoculum dose. Treatment with 100 μg/mL ribavirin completely inhibited hepatitis A virus growth in cultures infected with 100 to 800 tissue culture infectious dose 50 (TCID(50)) hepatitis A virus, but inocula of 800 to 1600 TCID(50) resulted in limited production of virus. The effect was time dependent and required more than 96 h of treatment to inhibit the virus completely. Ribavirin was less effective in treating cells persistently infected with hepatitis A virus, although there was significant inhibition of hepatitis A virus (82%) in persistently infected cells as well. Ribavirin had some inhibitory effect on cell growth; treatment with 25, 50 or 100 μg/mL ribavirin reduced cell growth by approximately 0, 20 and 40%, respectively.


T HE CHANG ING EPIDEMIOLOGY OF HEPATITIS A INFECTION
in Western countries has resulted in a lack of naturally acquired immunity in many adults.Infection with hepatitis A virus is becoming known as 'the traveller"s d isease• and is associated with considerab le morbidity (1.2) .The occurrence of more clinically severe forms of the illness stresses the need for an effective antiviral treatment.Recently, ribavi1in, which has already found clinical application in treatment of respiratory syncytial virus infection (3).was reported to reduce hepatitis A virus growth in tissue culture (2,4).The effect of ribavirin on chronically infected cells has not b een reported.In this report the authors describe tl1e inhibitory effect of ribavirin on replication of h epatitis A virus in acutely or chronically infected fetal Rhesus monkey kidney cells (FRhK-4) .

MATERIALS AND METHODS
A continuous cell line (FRhK-4) was obtained from the American Type Culture Collection.The cells were grown in minimum essential medium (Gibco) containing 10% fetal calf serum at 37°C.The FRhK-4 cells persistently infected with hepatitis A virus were maintained in the same manner as normal cultures.Cells persistently infected wiili hepatitis A virus were developed by serial passage and were used for ribavirin treatment at the 40tl1 passage.Viable cell counts were determined by ilie trypan-blue exclusion test.
Hepatitis A virus was isolated from a stool sample of a Chinese patient (from the People's Republic of China) in FRhK-4 cells.Infectivity was titrated in FRhK-4 cells and TC!Dso/ mL was calculated by ilie Karber method (5).
A modified Havab-M raclioin1munoassay (6) was used for ilie detection of hepatitis A virus antigen harvested from infected cell cultures.A ratio of san1ple to n egative control mean (S/N ratio) of at least 2 . 1 was considered positive.
Freshly seeded or confluent monolayers of FRhK-4 cells in 25 cm 2 flasks were treated with different concentrations of ribavirin (25 , 50 and 100 pg/mL).and untreated cells were kept as control.The flasks were incubated at 37°C for seven days and observed microscopically for visible cytotoxic effects .At the end of the incubation period the monolayers were trypsinized and viable cells counted.Relative growth was eA.lJressed as the ratio of ilie number of viable cells in ribavirintreated versus control cultures.
FRhK-4 cells (in 25 cm 2 flasks) were inoculated witl1 100, 200.400, 800 or 1600 tissue culture infectious dose 50 (TCIDso)/mL of hepatitis A virus in 0.1 mL and a dsorbed for 4 hat 37°C.The monolayers were washed and maintained in medium containing different con-68

RIBAVIR IN (,ug/ml)
Figure 1) Effectofribavirinoncellgrowth.The cells were treated when a monolayer was formed (-) or at the time of splitting (---).The cells were counted and relative growth was calculated by d ivid ing the number of cells from the ribavirin-treated bottle by number of cells from the untreated bottles centrations of ribavirin (0, 50 or 100 pg/mL).All flasks were incubated for seven days at 37°C. and tl1en cells were harvested by three cycles of freezing and thawing in 0.4 mL phosphate-buffered saline containing 0 .1% nonionic detergent 40.Preparations were centrifuged and supernatants tested for the presence of hepatitis A virus antigen by radioimmunoassay.The effect of short term treatment witl1 ribavirin on hepatitis A virus was determined by inoculating cells with 100 TCIDso of virus and then exposing them to 100 ~tg/mL ofribavirin for 0, 48, 72 and 96 h.At ilie end of treatment the ribavirin was replaced by minimum essential medium containing 3% fetal calf serum and the cells were further incubated for seven days.Cells were harvested and tested for hepatitis A virus .All experiments were done in triplicate.
Two sets of flasks (25 cm 2 ) witl1 a complete monolayer of chronically infected FRhK-4 were treated witl1 each dosage of ribavirin (0, 25, 50 or 100 ~tg/mL) and incubated for seven days.One set was then harvested as in the case of acute infection, while the second set was trypsinized and subcultured.When new monolayers were formed, tl1ey were treated again with the same doses of ribavirin.Tl;J.e presence of hepatitis A virus antigen in treated and control cells was determined by radioin1munoassay as above.

RESULTS
The results in terms of relative growth are illustrated in Figure 1.The doses of ribavirin used caused only minor morphological changes in FRhK-4 cells.Some rounded cells were detected in cultures trea ted \Vitl1 100 ~tg/mL ribavirin.Morphological changes were not observed in cultures treated with 25 or 50 ~tg/mL.
Treatment with ribavirin at 100 J.lg/ m L at tl1e lime of seeding and wh en the m onolayer had formed resulted in relative growth of 61 and 57% .respectively.Relative growth u nder the two treatment regim ns was 84 and 71 % when doses of 50 J.lg/mL were used.Th e grov.rthrate was not affected by treatment wi tl1 25 po'jmL.
The effect of ribavirin on the expression of h epatitis A virus antigen in acutely infected cells is shown in Figure 2. Cultures inocu lated \vith 100, 200 or 400 TCIDso and treated with 50 J.lg/mL ribavirin produced S/N ratios 6.1, 7.3 and 12.7.respectively.ie, significantly less than the yields from drug-free hepatitis A virus control cultures (S/N ratios 41 to 47).Treatment of cells with 100 pg/mL ribavirin completely inhibited the expression of hepatitis A virus antigen in cultures inocu lated with 100 to 800 TCID so.When the cultures were inoculated with 1600 TCIDso a small amount of residual hepatitis A virus antigen remained de tectable by radioimmunoassay.The hepatitis A virus yield from untreated cells inoculated with 800 or 1600 TCIDso is not shown in the figure .The inhibition of hepatitis A virus by ribavirin was lime dependent.The shortest period of treatment (48 h) with 100 pg/mL of tibavirin inhib ited the exrpression of hepatitis A virus antigen by 83.2% .whereas with the longest period of treatment (96 h) the inhibition was 95.1 % .
The results in terms of persistently infected cells are shown in Figure 3. FRhK-4 cells persistently infected with hepatitis A virus were less responsive to treatment with ribavirin than acutely infected cells.Treatment with 25.50 and 100 pg/mL ribavirin reduced tl1e hepatitis A virus antigen yield by 34, 60 and 82%.respectively.When tl1e ribavirtn -treated cells were passaged and re-treated \vith the same dose of drug.the inhibitory effect was more p rono u nced but not statistically significant.

DISCUSSION
Hepatitis A virus .though considered a picornavirus.has proven to be resistant to inhibition by guanidine and 2 -(alpha-hydroxybenzyl)-benzimidazole (7) .Among other antivirals tested so far for antih epalilis A virus activity.ribavirin seems promising (2.4) .more so because it has been well characterized clinically (3.8) .
The data presented here indicate that ribavirin at a dosage of 50 J..lg/mL significantly reduced the yield of hepatitis A virus antigen in acutely infected cells .and when 100 pg/mL was used no residual antigen could be detected in infected cells.even when tl1e dose of hepatitis A virus was increased from 100 to 800 TCIDso.These results do not rule out the presence of a very mall amount of infectious viru which could not be detected by radioimmunoassay.The radioin1munoa ssay detects hepatitis A virus protein antigen.not infectious virus.The effect of ribavirin at a dosage of 100 pg/mL was not likely due to cytoxicity.as most of where SIN ratio is the ratio of sample to negative control mean hepatitis A virus antigen harvested from treated or untreated infected cell cultures the cells were viable and the inhibition of cell growth was not sufficient to produce this effect.Hepatitis A virus antigen production was also inh ibited by ribavirin in persistently infected cells.The expression of hepatitis A virus antigen was significantly reduced (82%) at a ribivarin dosage of 100 pg/mL.The anti -hepatitis A virus effect of ribavirin on persistently infected cells has. to the authors• lmowledge, not been reported befo re.Ribavirin was less effective in persistently infected cells .which is similar to results obtained for the mumps virus (9).On tl1e other hand.cells persistently infected with 'foot and mouth disease• virus were cured by ribavirin (10).
These findings indicate that ribavirin has an inhib itory effect on hepatitis A virus in vitro and should be further evaluated for tl1erapeutic activity.

CRIBAVIRINFigure 2 )Figure 3 )
Figure2) Effect of ribavirin on hepatitis A vims (HA V) replication infetalRhesus monkey kidney cells.The SIN value indicates the ratio of sample counts per minute to negative control by radioimmunoassay.TCIDso Tissue culture infectious dose 50