ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Jan. 1992, p. 100-107
Vol. 36, No. 1
0066-48041921010100-08$02.00I/ Copyright C 1992, American Society for Microbiology
In Vitro Activity of Pirodavir (R 77975), a Substituted PhenoxyPyridazinamine with Broad-Spectrum Antipicornaviral Activity K. ANDRIES,* B. DEWINDT, J. SNOEKS, R. WILLEBRORDS, K. VAN EEMEREN, R. STOKBROEKX, AND P. A. J. JANSSEN Janssen Research Foundation, B-2340 Beerse, Belgium Received 15 July 1991/Accepted 31 October 1991
Pirodavir (R 77975) is the prototype of a novel class of broad-spectrum antipicornavirus compounds. Although its predecessor, R 61837, a substituted phenyl-pyridazinamine, was effective in inhibiting 80% of 100 serotypes tested (EC.) at concentrations above 32 ,ug/ml, pirodavir inhibits the same percentage of viruses at 0.064 ,ug/ml. Whereas R 61837 was active almost exclusively against rhinovirus serotypes of antiviral group B, pirodavir is broad spectrum in that it is highly active against both group A and group B rhinovirus serotypes. Pirodavir is also effective in inhibiting 16 enteroviruses, with an ECgo of 1.3 ,ug/ml. Susceptible rhinovirus serotypes were rendered noninfectious by direct contact with the antiviral compound. Their infectivity was not restored by dilution of virus-drug complexes, but was regained by organic solvent extraction of the compound for most serotypes. Neutralized viruses became stabilized to acid and heat, strongly suggesting a direct interaction of the compounds with viral capsid proteins. Mutants resistant to R 61837 (up to 85 times the MIC) were shown to bear some cross-resistance (up to 23 times the MIC) to the new compound, indicating that pirodavir also binds into the hydrophobic pocket beneath the canyon floor of rhinoviruses. Pirodavir acts at an early stage of the viral replication cycle (up to 40 min after infection) and reduces the yield of selected rhinoviruses 1,000- to 100,000-fold in a single round of replication. The mode of action appears to be serotype specific, since pirodavir was able to inhibit the adsorption of human rhinovirus 9 but not that of human rhinovirus 1A. Pirodavir is a novel capsid-binding antipicornavirus agent with potent in vitro activity against both group A and group B rhinovirus serotypes.
Pirodavir (R 77975) is a substituted phenoxy-pyridazinamine, distantly related to its predecessor, R 61837, a substituted phenyl-pyridazinamine (2) (Fig. 1). The present work compares the activity of the two compounds against rhinoviruses and enteroviruses in vitro and provides information about the mechanism of action of pirodavir.
Rhinoviruses are a major cause of mild upper respiratory infections in humans (11). Several synthetic antiviral agents with widely different structures have been shown to possess high in vitro activity against a selection of rhinovirus serotypes. All of them seem to bind into a specific hydrophobic pocket within the capsid protein VP1, beneath the canyon floor of rhinoviruses (24). This binding apparently stabilizes the capsid proteins and prevents viral attachment and/or uncoating, depending on the viral serotype involved (21). We previously reported that rhinoviruses can be divided into two distinct groups (designated A and B) on the basis of their susceptibility profile for capsid-binding antiviral compounds (5). Compounds such as dichloroflavan and R 61837 are active almost exclusively against serotypes from antiviral group B. Other compounds such as WIN 51711 have a complementary spectrum in that they are preferentially active against serotypes from antiviral group A only. In a first approach to achieve broad-spectrum activity, we tried to synthesize compounds with activity against serotypes from antiviral group A. The use of such a compound with another one active against serotypes from antiviral group B (such as R 61837) would result in a mixture of compounds with broad-spectrum activity. By chemically combining substructures of R 61837 and WIN 51711, we found a lead compound with weak but specific activity against antiviral group A rhinoviruses. A rational screening program (4) involving the testing of about 600 compounds against 17 representative serotypes was used to increase its potency and eventually led to the synthesis of compounds having high activity against serotypes from both antiviral groups. *
MATERIALS AND METHODS Viruses and cells. The origin of the rhinoviruses used in this study has been described previously (5). Polioviruses (type 1 Mahoney and Sabin, type 2 Mefie, and type 3 Leon) were provided by B. Rombaut, V. U. B., Brussels, Belgium; the other enteroviruses were provided by the American Type Culture Collection, Rockville, Md.; foot-and-mouth disease virus (A5 Leffinge 1959 and Cl Loupoigne 1953) was provided by K. De Clercq, N. I. D. O., Brussels, Belgium; human coronavirus 229E was provided by D. A. J. Tyrrell, M. R. C., Salisbury, United Kingdom; and influenza virus A/Taiwan/1/86 (HlNl) was provided by F. G. Hayden, University of Virginia, Charlottesville, Va. Echoviruses were grown in WI-38 cells; foot-and-mouth disease viruses were grown in SK-6 cells; all other piconaviruses were grown in mycoplasma-free HeLa/M cells; influenza virus were grown in MDCK cells; and coronavirus were grown in MRC16 cells. Cells were maintained in Eagle's basal medium supplemented with 5% fetal calf serum. Compounds. R 61837, or 3-methoxy-6-[4-(3-methylphenyl)-1-piperazinyl]pyridazine (M.G. 284.36), and pirodavir (R 77975), or ethyl 4-[2-[1-(6-methyl-3-pyridazinyl)-4-piperidinyl]ethoxy]benzoate (M.G. 369.46), were synthesized in the Janssen Laboratories by methods described elsewhere (24a, 24b). Compounds were dissolved in dimethyl sulfoxide (10 mg/ml) and diluted in growth medium to achieve the final concentration needed.
Corresponding author. 100
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
VOL. 36, 1992
100
CH3
CH300\. N\- XN
101
2
10 -
C 0
R 61837
0 el 0
.1 0.
CH3 N4
D
CH2-CH2-Oo
C-O-CH2-CH3
32
.01 -0 0
CD
-A-0--
.001
R 77975 FIG. 1. Chemical structures of R 61837 (3-methoxy-6-[4-(3-methylphenyl)-1-piperazinyl]pyridazine) and pirodavir, or R 77975 (ethyl 4-[2-[1-(6-methyl-3-pyridazinyl)-4-piperidinyl]ethoxy]benzoate).
Cytotoxicity assay. HeLa cells were seeded at a concentration of approximately 180,000 cells per dish in six-well plates (Falcon) containing 4 ml of growth medium. Growth medium consisted of Eagle's basal medium, supplemented with 5% fetal calf serum, 2% sodium bicarbonate, and 1% glutamine. After 24 h of incubation at 37°C in a humidified CO2 atmosphere, the growth medium was removed and replaced by the test solutions (fresh growth medium with or without various concentrations of the antiviral compounds). To assess the cytotoxicity of the antiviral compounds, we determined the number of living cells present in triplicate cultures at the time of drug addition and every 24 h for 3 days. Following trypsinization, the number of viable cells for each drug concentration was counted in triplicate with a Coulter Counter (model ELT8; Ortho). The data presented in this paper represent results obtained in two experiments that were shown to be reproducible, and each result represents the average of three replicates. Antiviral assay. An automated tetrazolium-based colorimetric assay was used for determination of MICs of test compounds. Compounds were added to a series of triplicate wells in microtiter trays. Serial fivefold dilutions were made directly in the microtiter trays with a robot system (Zymate II). Pipette tips were changed after every three dilutions. Untreated control human rhinovirus (HRV)-infected and mock-infected cell samples were included in each test. Approximately 100 50% tissue culture infective doses of virus were added to two of the three rows. After 2 h, a HeLa cell suspension (5 x 105 cells per ml) was added to all wells. The cultures were incubated at 33°C in a C02-free atmosphere. Three days after infection the viability of mock- and virus-infected cells was quantitated spectrophotometrically by a tetrazolium colorimetric method, the MTT assay (20). The MIC was defined as the 50% inhibitory concentration for
cytopathicity. Virus yield reduction assay. HRV1A, HRV9, HRV14, HRV39, HRV89, or HRV Hank's (at a multiplicity of 1 PFU per cell) were preincubated with pirodavir at the indicated concentrations (Fig. 2) for 30 min and then added to HeLa cells. After 1 h of adsorption at 33°C, the inoculum was removed and fresh maintenance medium containing the same concentration of pirodavir was added. The virus yield was determined by plaque assays after an additional 13 h of incubation at 33°C. The data presented represent results obtained in experiments that were shown to be reproducible
.0001
_
_
_
HRV1A HRV9 HRV14 HRV39 HRV89 HRV Hanks
_
.01 .1 1 10 100 100010000 Ratio of R 77975 concentration to the MIC
FIG. 2. Effect of pirodavir (R 77975) on the yield of virus from a single round of replication. HeLa cells were infected with the indicated serotypes, which had been preincubated with the appropriate concentration of antiviral compound. The cells were incubated for 14 h at 330C and frozen and thawed, and the virus yield was plaque assayed.
in at least two occasions, and each result represents the average of three replicates. Mechanism-of-action studies. (i) Virus neutralization. The direct virus-neutralizing effect of pirodavir was studied by using the same set of rhinovirus serotypes. Approximately 107 PFU/ml was incubated with or without different concentrations of the test compound for 60 min at 330C. After this incubation period, the reversibility of the binding between the virus and the compound was assessed by two approaches. In the first approach, 10-fold dilutions of the virus-drug mixtures were made to obtain noninhibitory concentrations of free compound (as established in the MIC test). These dilutions were plaque titrated in HeLa cells for any remaining infectious virus. Samples treated with compound and having a lower virus titer than the untreated controls were considered to contain virus, neutralized by the compound, in a way that is irreversible by dilution. In the second approach, we tried to restore the infectivity of neutralized viruses by extraction of the compound with organic solvents. In this case, the incubated virus-drug mixtures were mixed with an equal volume of dichloromethane. This mixture was vortexed vigorously for 1 min at room temperature and centrifuged at 1,000 x g for 5 min. The aqueous phase of the supernatant was removed, diluted 10-fold to obtain noninhibitory concentrations of free compound, and plaque titrated in HeLa cells. Virus samples treated with compound, subjected to dichloromethane extraction, and having a lower virus titer than the controls (exposed to dichloromethane extraction but not to antiviral compound) were considered to contain virus, neutralized by the compound, in a manner that was irreversible by extraction and dilution. (ii) Stabilization against heat and acid inactivation. Approximately 107 PFU of HRV1A, HRV70, and HRV9 per ml were incubated at 330C for 60 min in minimal essential medium (MEM) containing 2% fetal calf serum with or without different concentrations (lx, 1Ox, or 100x the MIC) of the test compound. The virus-drug mixtures were then diluted (1:10) in drug-free buffer, either 0.1 M acetate buffer or 0.1 M citrate buffer (pH 4 for HRV70 and pH 5 for HRV9
102
ANDRIES ET AL.
and HRV1A), or MEM-N-2-hydroxyethylpiperazine-N'-2ethanesulfonic acid (HEPES). The exposure times and temperatures were 30 min at room temperature for virus exposed to acid and 6 min at 56°C for heat-treated virus. Acid and heat treatments were stopped by further 10-fold dilutions in roomtemperature MEM-HEPES. Only the dilutions beyond the MIC of the drug for the virus were plaque assayed. Stabilization to acid or heat was assessed by comparing the measured PFU titer with the titer of controls, consisting of virus not preincubated with a drug but exposed to acid or heat under the same conditions. (iii) Effect of time of addition of pirodavir on virus inhibition. Pirodavir at 100 ng/ml was added at different times (0, 15, 30, 60, 120, 300, and 540 min) after infection of parallel cultures of HeLa cells with HRV1A or HRV9 at a multiplicity of infection of >1. After 60 min of incubation at 33°C, cells were washed to remove unadsorbed virus and then medium with the same drug concentration was added. At 9 h postinfection, cultures were frozen'and thawed three times and plaque titrated. (iv) Inhibition of virus adsorption. The effect of pirodavir on the adsorption step of HRV1A and HRV9 was investigated by means of an infectious-center assay. A HeLa cell suspension was obtained by treatment of cell monolayers with 0.02% EDTA. The cells were washed, resuspended to a final concentration of 106/ml, and cooled at 4°C for at least 2 h. HRV1A and HRV9 were incubated with different concentrations of the antiviral compounds (lx, 1Ox, or 100x the MIC) at 33°C for 60 min, cooled to 4°C, and subsequently added to the HeLa cell suspension. Cells were incubated with the virus-drug mixtures for 60 min at 4°C to prevent the virus from entering the cells. After the adsorption period, unadsorbed virus and free compound were removed by washing three times with cold phosphate-buffered saline. The cells were then diluted serially at room temperature and plaque assayed for cell-associated viral activity. (v) Cross-resistance with R 61837. HRV9 was cloned three times by end dilution. The virus was then exposed to R 61837 at 30 or 150 ng/ml (i.e., approximately 3 and 12 times the MIC) during three passages in HeLa cells grown in 24-well plates. At the end of each passage (lasting 1 week), virus from one end dilution showing cytopathic effect was harvested and used to initiate the subsequent passage. Two isolates were studied. The isolate coded 9H was exposed to R 61837 at 30, 30, and 0 ng/ml and the isolate coded 9M was exposed to R 61837 at 150, 150, and 150 ng/ml in the first, second, and third passages, respectively. After completion of the third passage, the extent of resistance to R 61837 and the presence of'cross-resistance against pirodavir were assessed by the antiviral assay described above. RESULTS Cytotoxicity assay. The cytotoxicity of pirodavir in vitro was determined by assessing the effect of the compound on logarithmic cell growth. The HeLa cell number for the untreated controls increased approximately 19-fold compared with the day 0 level. Pirodavir concentrations of 16 and 4 pLg/ml reduced cell growth by 66% (s.e.m. 0.75) and 28% (s.e.m. 0.25), respectively. Lower concentrations (1 ,ug/ml) of the drug were not inhibitory for cell growth. The 50% cytotoxic concentration of pirodavir for logarithmic cell growth at 37°C was 7 ,ug/ml. Under the conditions of the antiviral assay (confluent HeLa cells at 33°C), the 50% cytotoxic concentration was >50 ,ug/ml.
ANTIMICROB. AGENTS CHEMOTHER.
Antiviral spectrum. The MICs of R 61837 and pirodavir were determined for 100 typed and 1 untyped (Hank's) rhinovirus strain (Table 1). The MICs ranged from 0.001 to >32 ,ug/ml. The concentrations needed to inhibit 50 or 80% of the rhinoviruses (EC50 and EC80, respectively) were 4.4 and >32 ,ug/ml for R 61837 and 0.010 and 0.064 jig/ml for pirodavir, respectively. Pirodavir is therefore about 500 times more active than R 61837. Whereas R 61837 was active almost exclusively against serotypes of antiviral group B, pirodavir was highly active against serotypes of antiviral groups A and B (Table 1). All enteroviruses tested were found to be susceptible to pirodavir (Table 2), which had an EC50 and EC80 of 0.15 and 1.3 ,ug/ml, respectively. Mengovirus, foot-and-mouth disease virus types A and C, influenza A virus, and a human coronavirus were not susceptible to the compound. Virus yield reduction assay. The effect of pirodavir on the replication of selected rhinoviruses in a single round of replication was determined (Fig. 2). The compound reduced the viral yield by 0 (HRV Hank's) to 80% (HRV9) at concentrations equal to the MIC. Doses 10 times this concentration reduced the virus yield obtained in one replication cycle by 102- to 104-fold. On a double-logarithmic basis, it was possible to predict more than 70% of the variation in virus yield reduction from the variation in the ratio of R 77975 concentration to the MIC (r = 0.85). Virus neutralization. Pirodavir neutralized the infectivity of the majority of serotypes studied in a manner that was irreversible by dilution. Exceptions were serotypes HRV1A, which was not neutralized irreversibly, and HRV89, which was only partially neutralized by high concentrations of pirodavir. A classical S-shaped dose-response curve between the extent of neutralization and the ratio of concentration to MIC could be observed (Fig. 3). Exposure of neutralized viruses to dichloromethane prior to the dilution step resulted in complete recovery of the viral infectivity for HRV Hank's. The infectivity of drug-treated HRV9 or HRV39 could be restored to 30 to 60% of the virus controls, respectively (results not shown). HRV14 and HRV89 were sensitive to the dichloromethane extraction itself, with a loss of viral infectivity of about 50%. The effect of incubation time and drug concentration in HRV Hank's was studied (Fig. 4). Virus neutralization was rapid at higher drug concentrations (completed within 15 min) and slower at lower drug concentrations (completed within 30 min). Low concentrations of the drug (0.1 x and 1 x the MIC) also had an effect on the infectivity of the virus, but this became clear only after very long exposure periods. Stabilization against heat and acid inactivation. Pirodavir at the MIC was able to protect HRV9 partially against inactivation by acetate, citrate, and heat (Table 3). When the virus had been preincubated in lOx the MIC, the extent of protection obtained was difficult to assess because the extent of virus neutralization by pirodavir (-2.6 log10) was equal to the extent of inactivation by acid or heat. A concentration of 1Ox the MIC was necessary to protect HRV1A and HRV70 from acetate or citrate inactivation, whereas 100X the MIC was needed to provide some protection for HRV1A against heat inactivation. The same concentration of pirodavir was insufficient to protect HRV70 from heat inactivation. Effect of time of addition of pirodavir on virus inhibition. Maximal inhibition by pirodavir was obtained when the drug was added to the cells together with the virus (results not shown). The viral yield was also reduced when the compound was added 20 or 40 min after infection, but no more so
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
VOL. 36, 1992
103
TABLE 1. Antirhinovirus activity of R 61837 and pirodavir HRV serotype
Antiviral group
MIC (,ug/ml)a of: R 61837
Pirodavir
HRV serotype
Antiviral group
MIC (p.g/ml)a of: Pirodavir
R 61837
0.003 0.062 B 30 0.014 >32 >32 0.153 8.7 B 31 0.044 0.009 0.053 0.009 B 6.4 33 0.002 0.253 B 34 0.012 >32 >32 >32 0.002 1.9 B 36 >32 0.007 B 0.210 3.0 38 >32 0.006 1.0 B 0.011 39 >32 0.001 0.347 0.049 B 40 >32 0.291 1.9 0.005 B 41 >32 0.002 B 44 1.3 0.373 >32 0.016 1.8 B 46 4.2 7.1 0.111 5.4 47 0.005 B >32 0.002 0.056 0.006 B 49 >32 >32 0.008 0.075 B 50 >32 0.005 B 51 6.8 1.7 >32 >32 A 0.040 B 1.6 53 >32 B 14.4 55 A 0.010 0.013 >32 B A 0.001 0.009 56 1.3 >32 >32 A B 0.132 0.386 57 >32 A 0.002 5.9 B 0.045 58 >32 A 0.002 0.007 5.7 59 B >32 A 0.052 1.5 B 0.143 60 A 79 B 0.002 61 4.4 0.013 0.023 >32 A 83 0.002 0.855 2.1 B 62 >32 A 84 0.055 0.003 5.6 B 63 >32 A 86 0.011 0.005 B 64 0.180 >32 >32 87 A >32 B 65 0.031 >32 A 91 4.5 B 0.010 66 0.007 >32 A 92 0.011 0.044 B 0.002 67 >32 A 0.956 93 0.004 B 0.013 68 >32 >32 A 95 0.001 B 0.873 71 A 97 B 0.598 0.006 0.014 73 0.076 99 >32 A B 0.823 74 0.008 0.127 B 1A 0.168 0.038 8.1 B 0.011 75 B 1B 0.009 B 0.009 76 0.886 0.011 2 >32 B 0.201 B 0.002 77 0.057 >32 B B 7 0.064 0.011 78 3.4 B 9 0.012 B 0.012 0.267 0.026 80 B 10 0.009 0.039 B 0.003 81 4.8 11 0.012 B 0.003 0.025 B 0.064 82 12 5.0 B 0.002 1.6 B 0.010 85 B 15 0.003 1.1 B 0.031 88 0.251 16 0.002 2.7 B B 0.040 89 0.042 0.009 18 0.017 B 0.011 90 B 1.2 0.014 0.115 B 94 B 19 0.035 0.263 0.010 0.101 B B 20 2.0 0.003 96 >32 0.049 21 B 98 B 0.002 0.010 >32 0.015 B B 0.008 100 22 1.6 0.002 0.009 B HA B 0.008 23 0.068 0.003 B 24 0.148 0.026 0.012 B 1.4 9 B 25 6.2 0.054 0.239 B 9MH 0.010 B 28 9.1 B 0.587 1.019 0.003 B 9m 0.070 29 a Defined as the concentration required to reduce the inhibition of the formation of formazan from A4TT by 50%. Values are the mediah of at least three separate assays. On the average, the upper and lower limits of the 95% confidence intervals ranged between 0.6 and 2.8 times the median MICs. 3 4 S 6 8 13 14 17 26 27 32 35 37 42 43 45 48 52 54 69 70 72
A A A A A A A A A A A A A A A
60 min after infection, providing a rationale for studying the effect of the drug on viral adsorption. As shown in Fig. SA, pirodavir at the MIC inhibited the adsorption of HRV9, but not that of HRV1A. Indeed, concentrations of pirodavir exceeding the MIC by over 100-fold had only a marginal effect on the adsorption of HRV1A (Fig. SB). Cross-resistance with R 61837. Two mutants of HRV9, designated 9H and 9M, were selected for resistance against R 61837. The MICs of R 61837 for these mutants were
increased 20- and 85-fold, respectively, over those for the wild type (Table 1). The MICs of pirodavir against mutants 9H and 9M were increased 2-fold (not significant) and 23-fold, respectively, over those for the wild type.
DISCUSSION The substituted phenoxy-pyridazinamines represent a new class of broad-spectrum antiviral drugs with significant in
ANDRIES ET AL.
104
ANTIMICROB. AGENTS CHEMOTHER.
TABLE 2. MICs of R 61837 and pirodavir against enteroviruses. Virus
Coxsackievirus
Poliovirus
MIC' (jig/ml) of:
Serotype
A13 A15 A18 A21 B1 B3 B4 1 1 2 3
Mahoney Sabin Mefie Leon
R 61837
Pirodavir
>50 >50 >50 3.5 >50 >50 >50
0.150 0.018 0.150 0.021 0.780 8.5 4.0
>50 >50 >50 >50
0.120 0.085 0.270 0.025
Echovirus
9 11 12
>50 >50 >50
0.019 1.3 0.620
Enterovirus
68 70
>50 3.0
3.5 0.002
a See Table 1, footnote a.
vitro efficacy against the rhinovirus and enterovirus genera of the picornavirus family. Pirodavir is especially active against rhinoviruses, with MICs as low as 0.001 ,ug/ml (0.004 FiM) for some serotypes. Compared with its predecessor, R 61837, a substituted phenyl-pyridazinamine, a greater than 500-fold improvement in potency was obtained, as shown by the drop in EC80 from >32 ,ug/ml for R 61837 to 0.064 ,ug/ml for R 77975. The increase in potency was accompanied by a marked broadening of the spectrum. Whereas R 61837 was active almost exclusively against rhinoviruses from antiviral group B, pirodavir was highly active against rhinoviruses from both antiviral groups (Table 1). Pirodavir, but not R 61837, was also effective in inhibiting all tested enteroviruses, although it had to be used at higher concentrations (EC80 = 1.3) than those necessary to inhibit rhinoviruses
2
a1) 0
Q a 0 am
-0_U
>D
---0--
HRVIA HRV9 HRV14 HRV39 HRV89 HRV Hank's
.1 1 .01 10 100 1000 Ratio of R 77975 concentration to the MIC
FIG. 3. Effect of virus serotype and drug concentration on virus neutralization. Approximately 106 PFU of the indicated serotypes was incubated for 30 min at 33°C with various concentrations of pirodavir (R 77975). The mixtures were then serially diluted until noninhibitory drug levels were reached and plaque assayed.
120
-
0
u 0 0 0. 0E CD
0
-C
0)
--. 0 C.
-C
>
5
15 30 60 240 Incubation time (minutes)
0.01 MIC R 77975 0.1 MIC R 77975 1 MIC R 77975 10MIC R77975 100 MIC R 77975
1440
FIG. 4. Effect of time and drug concentration on virus neutralization.
(Table 2). Close chemical analogs of pirodavir are better inhibitors of enteroviruses such as poliovirus (22), but are less potent inhibitors of rhinoviruses (6). Members of the genera cardiovirus and aphthovirus were not susceptible to pirodavir, nor were other viruses that are important in upper respiratory tract infections, such as influenza virus and human coronavirus. The testing of a set of 17 representative serotypes (4) provided predictive information for the results subsequently obtained after testing all rhinoviruses. A pirodavir concentration of 0.143 ,ug/ml, sufficient to inhibit 14 of 17 screening serotypes (82.3%), also inhibited 84 of 101 serotypes (83.1%). Two of the 17 screening serotypes (HRV42 and HRV45) were not susceptible to pirodavir (MIC > 1 jig/ml). Both serotypes appear at the same region (the right edge of antiviral group A) on the spectral map of serotypes and antiviral compounds (5). Upon testing the other serotypes, it was interesting to see that most other insensitive rhinoviruses identified mapped as a cluster into the same region of antiviral group A. This finding shows that from the results obtained with the 17 serotypes, a prediction can be made about which serotypes are most likely to be resistant to a given compound. Previously described capsid-binding molecules were shown to bind to rhinoviruses into a hydrophobic pocket beneath the canyon floor of VP1 (8, 24). Their differential activity toward serotypes from antiviral groups A and B can be explained by their putative binding to particular amino acids lining this antiviral pocket, which are different in rhinoviruses belonging to different antiviral groups (5). Pirodavir, on the other hand, is active against rhinoviruses from both antiviral groups. Despite having this markedly different antiviral profile, pirodavir apparently binds to the same antiviral target. Indeed, a HRV9 mutant that was resistant to R 61837 was cross-resistant to pirodavir (Table 1). Furthermore, preliminary results of X-ray crystallographic studies indicate that pirodavir binds into the hydrophobic pocket of HRV14 (23). The different antiviral profile, together with the sharing of the antiviral binding site, suggests that the broadspectrum activity of the compound is achieved by its binding to amino acids from the antiviral pocket that are conserved throughout rhinoviruses belonging to both antiviral groups and enteroviruses. An alignment for amino acids lining the pocket in HRV14 reveals that two tyrosines are conserved in all 11 sequenced rhinoviruses and enteroviruses (5). The wide range of susceptibilities of different HRV sero-
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
VOL. 36, 1992
105
TABLE 3. Stabilization of rhinoviruses by pirodavir PFU titer" (loglo PFU/ml) for: Drug concn (xMIC)
HRV1A (MIC, 0.011 uLg/ml) 0 1 10 100
HRV9 (MIC, 0.026 0 1 10 100
4Lmgl)
Virus neutralization (pH 7, 20°C)
Acetate treatment (pH 4 or 5, 20°C)
Citrate treatment (pH 4 or 5, 20°C)
Heat treatment (pH 7, 56°C)
0.0 0.0 -0.1 -0.1
>-3.8 >-3.8 -0.5 -0.3
>-3.8 >-3.8 -0.7 -0.4
>-3.3 >-3.3 -3.3 -1.1
0.0 0.0 -2.6 -3.4
-2.5 -1.1 -2.7 -3.4
>-4.4
-3.0 -2.6 -3.4
-2.9 -1.4 -3.1 -3.8
0.0 -0.2 -0.9 -1.0
-3.3 -3.3 -1.3 -1.2
-3.3 -3.3 -1.5 -1.3
>-3.2 >-3.2 > -3.2 >-3.2
HRV70 (MIC, 0.002 ug/ml) 0 1 10 100
aPFU titers expressed in reference to the virus controls (value 0.0).
types is a common feature of all capsid-binding compounds (4). Slight variations in the amino acid composition of the binding site on the viral capsid of different serotypes are likely to affect the strength of the virus-drug bond and therefore the MIC. Three of four rhinoviruses that are completely resistant to pirodavir (HRV8, HRV42, and HRV87) can be blocked by other capsid-binding compounds (4, 6). This indicates that these viruses do have a drugbinding pocket, but that their hydrophobic pockets are aberrant in amino acid composition and/or shape. Pirodavir binds to most rhinoviruses in a way that depends on serotype, drug concentration, and time. For some serotypes (e.g., HRV1A), binding is almost completely reversible by dilution. For others (e.g., HRV Hank's), a stronger binding results in neutralization of the viral infectivity, which can be reversed only by organic solvent extraction, as is the case with neutralization by antibodies. For a third group of rhinoviruses (e.g., HRV9), the binding results in a neutralization that is not reversible by organic solvent extraction. The properties of pirodavir are consistent with earlier observations obtained with R 61837 (2) and slightly different from those described for chalcone, dichloroflavan, and isoflavans, which can be removed from neutralized serotypes by organic-solvent extraction (10, 17). A classical S-shaped dose-response curve between the extent of neutralization and the ratio of pirodavir concentration to MIC could be observed (Fig. 3). The binding of pirodavir to viral particles results in their stabilization. Several susceptible rhinoviruses could be protected against inactivation by mild acidification or heat (Table 3). This observation supports the concept that the drug exerts its antiviral activity by a direct interaction with the viral protein capsid. The minimal concentrations necessary to inhibit either acetate, citrate, or heat inactivation were not correlated with the MICs, suggesting that inhibition of replication and stabilization are independent events, both resulting from the binding of the drug to the viral capsid. A similar conclusion was obtained in tests with R 61837 against five rhinovirus serotypes (3) and in tests with R 78206 against four poliovirus strains (22). The absence of a correlation
between the MICs of a given compound and the concentrations needed for stabilization of different serotypes does not exclude the possibility that such a correlation exists when several compounds of the same chemical family are being studied within one particular serotype (12). A time-of-addition study indicated that pirodavir had an effect on an early event in the replication of both HRV1A and HRV9. In HRV9, the adsorption step was inhibited by pirodavir concentrations closely related to the MIC (Fig. SA). On the other hand, the adsorption of HRV1A was not inhibited at concentrations greatly exceeding the MIC, indicating that the mode of action is serotype specific. A similar differential effect on rhinovirus serotypes has also been described for WIN 54954, another potent capsid-binding compound (21). The extent of conformational changes induced in the putative receptor-binding region for these viruses, the canyon structure (9), may be responsible for the observed differences in the mode of action. During the past decade, several compounds from different chemical classes with potent in vitro activity against several members of the picornavirus family have been described. They include flavans (7), isoflavans (10), chalcone (14), chalcone amides (18), pyrano-pyridines (15), and isoxazoles (19, 25). All of them are supposed to exert their antiviral activity by binding to virions, inducing conformational and flexibility changes in the capsid proteins and thus inhibiting the adsorption and/or uncoating event of the replication cycle (21). Drug-resistant mutants raised against some of these compounds usually exhibit cross-resistance to others, strongly suggesting a similar mode of action and a sharing of binding sites (3, 17, 18). Few of the compounds mentioned have been advanced to in vivo studies. Several compounds of the WIN series have proved to be efficacious in various animal models of enteroviral disease (16, 25). R 61837, the prototype compound of the pyridazinamine series, has shown clinical efficacy in humans (1). A pilot clinical trial of pirodavir under double-blind, placebo-controlled conditions indicated that the compound has a clinical effect in humans (13). Although pirodavir is inactive against common cold viruses other than rhinoviruses, it can be considered a
ANDRIES ET AL.
106
ANTIMICROB. AGENTS CHEMOTHER.
Q la
LI)
a
-0--
CC
HRV I
9 control
MIC R 77975
10 MIC R 77975
100MICR77975
.2
20
0
40
Time 0
a
60
(minutes)
c
b
0
10U0 U)
CL
150.
0
--0-
HRV IA control
1 MIC R 77975 10 MIC R 77975 -*- 100MICR77975
100.
20
0
40
60
Time (minutes) FIG. 5. Effect of pirodavir (R 77975) on the adsorption of HRV9 (a) and HRV1A (b). Results are expressed as the mean standard ±
error of the mean.
serious new candidate for prophylactic and/or therapeutic control of rhinovirus infections in humans. Members of the same chemical series may also be useful as stabilizers of enteroviral vaccines (22). ACKNOWLEDGMENT
We gratefully acknowledge K. De Clercq, National Institute for Veterinary Research, Brussels, Belgium, for testing pirodavir against foot-and-mouth disease virus.
REFERENCES 1.
Al-Nakib, W., P. G. Higgins, G. I. Barrow, D. A. J. Tyrrell, K. Andries, G. Vanden Bussche, N. Taylor, and P. A. J. Janssen. 1989. Suppression of colds in human volunteers challenged with rhinovirus by a new synthetic drug (R 61837). Antimicrob. Agents Chemother. 33:522-525.
2.
Andries, K.,
B.
Dewindt,
M. De
Brabander, R. Stokbroekx, and
P. A. J. Janssen. 1988. In vitro activity of R 61837, a new antirhinovirus compound. Arch. Virol. 101:155-167. 3. Andries, K., B. Dewindt, J. Snoeks, and R. Willebrords. 1989. Lack of quantitative correlation between inhibition of replication of rhinoviruses by an antiviral drug and their stabilization. Arch. Virol. 106:51-61. 4. Andries, K., B. Dewindt, J. Snoeks, R. Willebrords, R. Stokbroekx, and P. J. Lewi. 1991. A comparative test of fifteen compounds against all known human rhinovirus serotypes as a basis for a more rational screening program. Antiviral Res. 16:213-225. 5. Andries, K., B. Dewindt, J. Snoeks, L. Wouters, H. Moereels, P. J. Lewi, and P. A. J. Janssen. 1990. Two groups of rhinoviruses revealed by a panel of antiviral compounds present sequence divergence and differential pathogenicity. J. Virol. 64:1117-1123. 6. Andries, K., J. Snoeks, and R. Willebrords. Unpublished data. 7. Bauer, D. J., J. W. T. Selway, J. F. Batchelor, M. Tisdale, I. C. Caldwell, and D. A. B. Young. 1981. 4',6-Dichloroflavan (BW683C), a new anti-rhinovirus compound. Nature (London) 292:369-370. 8. Chapman, M. S., I. Minor, M. G. Rossmann, G. D. Diana, and K. Andries. 1991. Human rhinovirus 14 complexed with antiviral compound R 61837. J. Mol. Biol. 217:454-463. 9. Colonno, R. J., J. H. Condra, S. Mitzutani, P. L. Callahan, M. Davies, and M. Murcko. 1988. Evidence for the direct involvement of the rhinoviruses canyon in receptor binding. Proc. NatI. Acad. Sci. USA 85:5449-5453. 10. Conti, C., N. Orsi, and M. L. Stein. 1988. Effect of isoflavans and isoflavenes on rhinovirus lB and its replication in HeLa cells. Antiviral Res. 10:117-127. 11. Couch, R. B. 1984. The common cold: control? J. Infect. Dis. 150:167-173. 12. Gruenberger, M., D. Pevear, G. D. Diana, E. Kuechler, and D. Blaas. 1991. Stabilization of human rhinovirus serotype 2 against pH-induced conformational change by antiviral compounds. J. Gen. Virol. 72:431-433. 13. Hayden, F. G., K. Andries, and P. A. J. Janssen. 1990. Efficacy of intranasal R 77975 for the prevention of experimentally induced rhinovirus infection and illness. Antiviral Res. 13(Suppl. 1):103. 14. Ishitsuka, H., Y. T. Ninomiya, C. Ohsawa, M. Fujiu, and Y. Suhara. 1982. Direct and specific inactivation of rhinovirus by chalcone. Antimicrob. Agents Chemother. 22:617-621. 15. Kenny, M. T., J. K. Dulworth, T. M. Bargar, H. L. Torney, M. C. Graham, and A. M. Manelli. 1986. In vitro antiviral activity of the 6-substituted 2-(3',4'-dichlorophenoxy)-2H-pyranol[2,3-b]pyridines MDL 20,610, MDL 20,646, and MDL 20,957. Antimicrob. Agents Chemother. 30:516-518. 16. McKinlay, M. A., and B. A. Steinberg. 1986. Oral efficacy of WIN 51711 in mice infected with human poliovirus. Antimicrob. Agents Chemother. 29:30-32. 17. Ninomiya, Y., M. Aoyama, I. Umeda, Y. Suhara, and H. Ishitsuka. 1985. Comparative studies on the modes of action of the antirhinovirus agents Ro 09-410, Ro 09-0179, RMI-15,731, 4',6-dichloroflavan, and enviroxime. Antimicrob. Agents Chemother. 27:595-599. 18. Ninomiya, Y., N. Shimma, and H. Ishitsuka. 1990. Comparative studies on the antirhinovirus activity and the mode of action of the rhinovirus capsid binding agents, chalcone amides. Antiviral Res. 13:61-74. 19. Otto, M. J., M. P. Fox, M. J. Fancher, M. F. Kuhrt, G. D. Diana, and M. A. McKinlay. 1985. In vitro activity of WIN 51711, a new broad-spectrum antipicornavirus drug. Antimicrob. Agents Chemother. 27:883-886. 20. Pauwels, R., J. Balzarini, M. Baba, R. Snoeck, D. Schols, P. Herdewijn, J. Desmyter, and E. De Clercq. 1988. Rapid and automated tetrazolium-based colorimetric assay for the detection of anti-HIV compounds. J. Virol. Methods 20:309-321. 21. Pevear, D. C., M. J. Fancher, P. J. Felock, M. G. Rossmann, M. S. Miller, G. D. Diana, A. M. Treasurywala, M. A. McKinlay, and F. J. Dutko. 1989. Conformational change in the floor of the human rhinovirus canyon blocks adsorption to HeLa cell
VOL. 36, 1992
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
receptor. J. Virol. 63:2002-2007. 22. Rombaut B., K. Andries, and A. Boeye. 1991. A comparison of WIN 51711 and R 78206 as stabilizers of poliovirions and procapsids. J. Gen. Virol. 72:2153-2157. 23. Rossmann, M. G. (Purdue University). 1991. Personal communication. 24. Smith, T. J., M. J. Kremer, M. Lou, G. Vriend, E. Arnold, G. Kamer, M. G. Rossmann, M. A. McKinlay, G. D. Diana, and M. J. Otto. 1986. The site of attachment in human rhinovirus 14 for antiviral agents that inhibit uncoating. Science 233:12861293.
107
24a.Stokbroekx, R. A., M. J. M. Van der Aa, M. G. M. Luyckx, and G. A. J. Grauwels. 1989. Pyridazine derivatives as antivirals. Chem. Abstr. 112(5):35876y. 24b.Stokbroekx, R. A., M. J. M. Van der Aa, J. J. M. Willems, and M. G. M. Luyckx. 1985. Anti-virally active pyridazinamines. Chem. Abstr. 104(15):129918a. 25. Woods, M. G., G. D. Diana, M. C. Rogge, M. J. Otto, F. J. Dutko, and M. A. McKinlay. 1989. In vitro and in vivo activities of WIN 54954, a new broad-spectrum antipicornavirus drug. Antimicrob. Agents Chemother. 33:2069-2074.
Vol. 36, No. 1
0066-48041921010100-08$02.00I/ Copyright C 1992, American Society for Microbiology
In Vitro Activity of Pirodavir (R 77975), a Substituted PhenoxyPyridazinamine with Broad-Spectrum Antipicornaviral Activity K. ANDRIES,* B. DEWINDT, J. SNOEKS, R. WILLEBRORDS, K. VAN EEMEREN, R. STOKBROEKX, AND P. A. J. JANSSEN Janssen Research Foundation, B-2340 Beerse, Belgium Received 15 July 1991/Accepted 31 October 1991
Pirodavir (R 77975) is the prototype of a novel class of broad-spectrum antipicornavirus compounds. Although its predecessor, R 61837, a substituted phenyl-pyridazinamine, was effective in inhibiting 80% of 100 serotypes tested (EC.) at concentrations above 32 ,ug/ml, pirodavir inhibits the same percentage of viruses at 0.064 ,ug/ml. Whereas R 61837 was active almost exclusively against rhinovirus serotypes of antiviral group B, pirodavir is broad spectrum in that it is highly active against both group A and group B rhinovirus serotypes. Pirodavir is also effective in inhibiting 16 enteroviruses, with an ECgo of 1.3 ,ug/ml. Susceptible rhinovirus serotypes were rendered noninfectious by direct contact with the antiviral compound. Their infectivity was not restored by dilution of virus-drug complexes, but was regained by organic solvent extraction of the compound for most serotypes. Neutralized viruses became stabilized to acid and heat, strongly suggesting a direct interaction of the compounds with viral capsid proteins. Mutants resistant to R 61837 (up to 85 times the MIC) were shown to bear some cross-resistance (up to 23 times the MIC) to the new compound, indicating that pirodavir also binds into the hydrophobic pocket beneath the canyon floor of rhinoviruses. Pirodavir acts at an early stage of the viral replication cycle (up to 40 min after infection) and reduces the yield of selected rhinoviruses 1,000- to 100,000-fold in a single round of replication. The mode of action appears to be serotype specific, since pirodavir was able to inhibit the adsorption of human rhinovirus 9 but not that of human rhinovirus 1A. Pirodavir is a novel capsid-binding antipicornavirus agent with potent in vitro activity against both group A and group B rhinovirus serotypes.
Pirodavir (R 77975) is a substituted phenoxy-pyridazinamine, distantly related to its predecessor, R 61837, a substituted phenyl-pyridazinamine (2) (Fig. 1). The present work compares the activity of the two compounds against rhinoviruses and enteroviruses in vitro and provides information about the mechanism of action of pirodavir.
Rhinoviruses are a major cause of mild upper respiratory infections in humans (11). Several synthetic antiviral agents with widely different structures have been shown to possess high in vitro activity against a selection of rhinovirus serotypes. All of them seem to bind into a specific hydrophobic pocket within the capsid protein VP1, beneath the canyon floor of rhinoviruses (24). This binding apparently stabilizes the capsid proteins and prevents viral attachment and/or uncoating, depending on the viral serotype involved (21). We previously reported that rhinoviruses can be divided into two distinct groups (designated A and B) on the basis of their susceptibility profile for capsid-binding antiviral compounds (5). Compounds such as dichloroflavan and R 61837 are active almost exclusively against serotypes from antiviral group B. Other compounds such as WIN 51711 have a complementary spectrum in that they are preferentially active against serotypes from antiviral group A only. In a first approach to achieve broad-spectrum activity, we tried to synthesize compounds with activity against serotypes from antiviral group A. The use of such a compound with another one active against serotypes from antiviral group B (such as R 61837) would result in a mixture of compounds with broad-spectrum activity. By chemically combining substructures of R 61837 and WIN 51711, we found a lead compound with weak but specific activity against antiviral group A rhinoviruses. A rational screening program (4) involving the testing of about 600 compounds against 17 representative serotypes was used to increase its potency and eventually led to the synthesis of compounds having high activity against serotypes from both antiviral groups. *
MATERIALS AND METHODS Viruses and cells. The origin of the rhinoviruses used in this study has been described previously (5). Polioviruses (type 1 Mahoney and Sabin, type 2 Mefie, and type 3 Leon) were provided by B. Rombaut, V. U. B., Brussels, Belgium; the other enteroviruses were provided by the American Type Culture Collection, Rockville, Md.; foot-and-mouth disease virus (A5 Leffinge 1959 and Cl Loupoigne 1953) was provided by K. De Clercq, N. I. D. O., Brussels, Belgium; human coronavirus 229E was provided by D. A. J. Tyrrell, M. R. C., Salisbury, United Kingdom; and influenza virus A/Taiwan/1/86 (HlNl) was provided by F. G. Hayden, University of Virginia, Charlottesville, Va. Echoviruses were grown in WI-38 cells; foot-and-mouth disease viruses were grown in SK-6 cells; all other piconaviruses were grown in mycoplasma-free HeLa/M cells; influenza virus were grown in MDCK cells; and coronavirus were grown in MRC16 cells. Cells were maintained in Eagle's basal medium supplemented with 5% fetal calf serum. Compounds. R 61837, or 3-methoxy-6-[4-(3-methylphenyl)-1-piperazinyl]pyridazine (M.G. 284.36), and pirodavir (R 77975), or ethyl 4-[2-[1-(6-methyl-3-pyridazinyl)-4-piperidinyl]ethoxy]benzoate (M.G. 369.46), were synthesized in the Janssen Laboratories by methods described elsewhere (24a, 24b). Compounds were dissolved in dimethyl sulfoxide (10 mg/ml) and diluted in growth medium to achieve the final concentration needed.
Corresponding author. 100
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
VOL. 36, 1992
100
CH3
CH300\. N\- XN
101
2
10 -
C 0
R 61837
0 el 0
.1 0.
CH3 N4
D
CH2-CH2-Oo
C-O-CH2-CH3
32
.01 -0 0
CD
-A-0--
.001
R 77975 FIG. 1. Chemical structures of R 61837 (3-methoxy-6-[4-(3-methylphenyl)-1-piperazinyl]pyridazine) and pirodavir, or R 77975 (ethyl 4-[2-[1-(6-methyl-3-pyridazinyl)-4-piperidinyl]ethoxy]benzoate).
Cytotoxicity assay. HeLa cells were seeded at a concentration of approximately 180,000 cells per dish in six-well plates (Falcon) containing 4 ml of growth medium. Growth medium consisted of Eagle's basal medium, supplemented with 5% fetal calf serum, 2% sodium bicarbonate, and 1% glutamine. After 24 h of incubation at 37°C in a humidified CO2 atmosphere, the growth medium was removed and replaced by the test solutions (fresh growth medium with or without various concentrations of the antiviral compounds). To assess the cytotoxicity of the antiviral compounds, we determined the number of living cells present in triplicate cultures at the time of drug addition and every 24 h for 3 days. Following trypsinization, the number of viable cells for each drug concentration was counted in triplicate with a Coulter Counter (model ELT8; Ortho). The data presented in this paper represent results obtained in two experiments that were shown to be reproducible, and each result represents the average of three replicates. Antiviral assay. An automated tetrazolium-based colorimetric assay was used for determination of MICs of test compounds. Compounds were added to a series of triplicate wells in microtiter trays. Serial fivefold dilutions were made directly in the microtiter trays with a robot system (Zymate II). Pipette tips were changed after every three dilutions. Untreated control human rhinovirus (HRV)-infected and mock-infected cell samples were included in each test. Approximately 100 50% tissue culture infective doses of virus were added to two of the three rows. After 2 h, a HeLa cell suspension (5 x 105 cells per ml) was added to all wells. The cultures were incubated at 33°C in a C02-free atmosphere. Three days after infection the viability of mock- and virus-infected cells was quantitated spectrophotometrically by a tetrazolium colorimetric method, the MTT assay (20). The MIC was defined as the 50% inhibitory concentration for
cytopathicity. Virus yield reduction assay. HRV1A, HRV9, HRV14, HRV39, HRV89, or HRV Hank's (at a multiplicity of 1 PFU per cell) were preincubated with pirodavir at the indicated concentrations (Fig. 2) for 30 min and then added to HeLa cells. After 1 h of adsorption at 33°C, the inoculum was removed and fresh maintenance medium containing the same concentration of pirodavir was added. The virus yield was determined by plaque assays after an additional 13 h of incubation at 33°C. The data presented represent results obtained in experiments that were shown to be reproducible
.0001
_
_
_
HRV1A HRV9 HRV14 HRV39 HRV89 HRV Hanks
_
.01 .1 1 10 100 100010000 Ratio of R 77975 concentration to the MIC
FIG. 2. Effect of pirodavir (R 77975) on the yield of virus from a single round of replication. HeLa cells were infected with the indicated serotypes, which had been preincubated with the appropriate concentration of antiviral compound. The cells were incubated for 14 h at 330C and frozen and thawed, and the virus yield was plaque assayed.
in at least two occasions, and each result represents the average of three replicates. Mechanism-of-action studies. (i) Virus neutralization. The direct virus-neutralizing effect of pirodavir was studied by using the same set of rhinovirus serotypes. Approximately 107 PFU/ml was incubated with or without different concentrations of the test compound for 60 min at 330C. After this incubation period, the reversibility of the binding between the virus and the compound was assessed by two approaches. In the first approach, 10-fold dilutions of the virus-drug mixtures were made to obtain noninhibitory concentrations of free compound (as established in the MIC test). These dilutions were plaque titrated in HeLa cells for any remaining infectious virus. Samples treated with compound and having a lower virus titer than the untreated controls were considered to contain virus, neutralized by the compound, in a way that is irreversible by dilution. In the second approach, we tried to restore the infectivity of neutralized viruses by extraction of the compound with organic solvents. In this case, the incubated virus-drug mixtures were mixed with an equal volume of dichloromethane. This mixture was vortexed vigorously for 1 min at room temperature and centrifuged at 1,000 x g for 5 min. The aqueous phase of the supernatant was removed, diluted 10-fold to obtain noninhibitory concentrations of free compound, and plaque titrated in HeLa cells. Virus samples treated with compound, subjected to dichloromethane extraction, and having a lower virus titer than the controls (exposed to dichloromethane extraction but not to antiviral compound) were considered to contain virus, neutralized by the compound, in a manner that was irreversible by extraction and dilution. (ii) Stabilization against heat and acid inactivation. Approximately 107 PFU of HRV1A, HRV70, and HRV9 per ml were incubated at 330C for 60 min in minimal essential medium (MEM) containing 2% fetal calf serum with or without different concentrations (lx, 1Ox, or 100x the MIC) of the test compound. The virus-drug mixtures were then diluted (1:10) in drug-free buffer, either 0.1 M acetate buffer or 0.1 M citrate buffer (pH 4 for HRV70 and pH 5 for HRV9
102
ANDRIES ET AL.
and HRV1A), or MEM-N-2-hydroxyethylpiperazine-N'-2ethanesulfonic acid (HEPES). The exposure times and temperatures were 30 min at room temperature for virus exposed to acid and 6 min at 56°C for heat-treated virus. Acid and heat treatments were stopped by further 10-fold dilutions in roomtemperature MEM-HEPES. Only the dilutions beyond the MIC of the drug for the virus were plaque assayed. Stabilization to acid or heat was assessed by comparing the measured PFU titer with the titer of controls, consisting of virus not preincubated with a drug but exposed to acid or heat under the same conditions. (iii) Effect of time of addition of pirodavir on virus inhibition. Pirodavir at 100 ng/ml was added at different times (0, 15, 30, 60, 120, 300, and 540 min) after infection of parallel cultures of HeLa cells with HRV1A or HRV9 at a multiplicity of infection of >1. After 60 min of incubation at 33°C, cells were washed to remove unadsorbed virus and then medium with the same drug concentration was added. At 9 h postinfection, cultures were frozen'and thawed three times and plaque titrated. (iv) Inhibition of virus adsorption. The effect of pirodavir on the adsorption step of HRV1A and HRV9 was investigated by means of an infectious-center assay. A HeLa cell suspension was obtained by treatment of cell monolayers with 0.02% EDTA. The cells were washed, resuspended to a final concentration of 106/ml, and cooled at 4°C for at least 2 h. HRV1A and HRV9 were incubated with different concentrations of the antiviral compounds (lx, 1Ox, or 100x the MIC) at 33°C for 60 min, cooled to 4°C, and subsequently added to the HeLa cell suspension. Cells were incubated with the virus-drug mixtures for 60 min at 4°C to prevent the virus from entering the cells. After the adsorption period, unadsorbed virus and free compound were removed by washing three times with cold phosphate-buffered saline. The cells were then diluted serially at room temperature and plaque assayed for cell-associated viral activity. (v) Cross-resistance with R 61837. HRV9 was cloned three times by end dilution. The virus was then exposed to R 61837 at 30 or 150 ng/ml (i.e., approximately 3 and 12 times the MIC) during three passages in HeLa cells grown in 24-well plates. At the end of each passage (lasting 1 week), virus from one end dilution showing cytopathic effect was harvested and used to initiate the subsequent passage. Two isolates were studied. The isolate coded 9H was exposed to R 61837 at 30, 30, and 0 ng/ml and the isolate coded 9M was exposed to R 61837 at 150, 150, and 150 ng/ml in the first, second, and third passages, respectively. After completion of the third passage, the extent of resistance to R 61837 and the presence of'cross-resistance against pirodavir were assessed by the antiviral assay described above. RESULTS Cytotoxicity assay. The cytotoxicity of pirodavir in vitro was determined by assessing the effect of the compound on logarithmic cell growth. The HeLa cell number for the untreated controls increased approximately 19-fold compared with the day 0 level. Pirodavir concentrations of 16 and 4 pLg/ml reduced cell growth by 66% (s.e.m. 0.75) and 28% (s.e.m. 0.25), respectively. Lower concentrations (1 ,ug/ml) of the drug were not inhibitory for cell growth. The 50% cytotoxic concentration of pirodavir for logarithmic cell growth at 37°C was 7 ,ug/ml. Under the conditions of the antiviral assay (confluent HeLa cells at 33°C), the 50% cytotoxic concentration was >50 ,ug/ml.
ANTIMICROB. AGENTS CHEMOTHER.
Antiviral spectrum. The MICs of R 61837 and pirodavir were determined for 100 typed and 1 untyped (Hank's) rhinovirus strain (Table 1). The MICs ranged from 0.001 to >32 ,ug/ml. The concentrations needed to inhibit 50 or 80% of the rhinoviruses (EC50 and EC80, respectively) were 4.4 and >32 ,ug/ml for R 61837 and 0.010 and 0.064 jig/ml for pirodavir, respectively. Pirodavir is therefore about 500 times more active than R 61837. Whereas R 61837 was active almost exclusively against serotypes of antiviral group B, pirodavir was highly active against serotypes of antiviral groups A and B (Table 1). All enteroviruses tested were found to be susceptible to pirodavir (Table 2), which had an EC50 and EC80 of 0.15 and 1.3 ,ug/ml, respectively. Mengovirus, foot-and-mouth disease virus types A and C, influenza A virus, and a human coronavirus were not susceptible to the compound. Virus yield reduction assay. The effect of pirodavir on the replication of selected rhinoviruses in a single round of replication was determined (Fig. 2). The compound reduced the viral yield by 0 (HRV Hank's) to 80% (HRV9) at concentrations equal to the MIC. Doses 10 times this concentration reduced the virus yield obtained in one replication cycle by 102- to 104-fold. On a double-logarithmic basis, it was possible to predict more than 70% of the variation in virus yield reduction from the variation in the ratio of R 77975 concentration to the MIC (r = 0.85). Virus neutralization. Pirodavir neutralized the infectivity of the majority of serotypes studied in a manner that was irreversible by dilution. Exceptions were serotypes HRV1A, which was not neutralized irreversibly, and HRV89, which was only partially neutralized by high concentrations of pirodavir. A classical S-shaped dose-response curve between the extent of neutralization and the ratio of concentration to MIC could be observed (Fig. 3). Exposure of neutralized viruses to dichloromethane prior to the dilution step resulted in complete recovery of the viral infectivity for HRV Hank's. The infectivity of drug-treated HRV9 or HRV39 could be restored to 30 to 60% of the virus controls, respectively (results not shown). HRV14 and HRV89 were sensitive to the dichloromethane extraction itself, with a loss of viral infectivity of about 50%. The effect of incubation time and drug concentration in HRV Hank's was studied (Fig. 4). Virus neutralization was rapid at higher drug concentrations (completed within 15 min) and slower at lower drug concentrations (completed within 30 min). Low concentrations of the drug (0.1 x and 1 x the MIC) also had an effect on the infectivity of the virus, but this became clear only after very long exposure periods. Stabilization against heat and acid inactivation. Pirodavir at the MIC was able to protect HRV9 partially against inactivation by acetate, citrate, and heat (Table 3). When the virus had been preincubated in lOx the MIC, the extent of protection obtained was difficult to assess because the extent of virus neutralization by pirodavir (-2.6 log10) was equal to the extent of inactivation by acid or heat. A concentration of 1Ox the MIC was necessary to protect HRV1A and HRV70 from acetate or citrate inactivation, whereas 100X the MIC was needed to provide some protection for HRV1A against heat inactivation. The same concentration of pirodavir was insufficient to protect HRV70 from heat inactivation. Effect of time of addition of pirodavir on virus inhibition. Maximal inhibition by pirodavir was obtained when the drug was added to the cells together with the virus (results not shown). The viral yield was also reduced when the compound was added 20 or 40 min after infection, but no more so
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
VOL. 36, 1992
103
TABLE 1. Antirhinovirus activity of R 61837 and pirodavir HRV serotype
Antiviral group
MIC (,ug/ml)a of: R 61837
Pirodavir
HRV serotype
Antiviral group
MIC (p.g/ml)a of: Pirodavir
R 61837
0.003 0.062 B 30 0.014 >32 >32 0.153 8.7 B 31 0.044 0.009 0.053 0.009 B 6.4 33 0.002 0.253 B 34 0.012 >32 >32 >32 0.002 1.9 B 36 >32 0.007 B 0.210 3.0 38 >32 0.006 1.0 B 0.011 39 >32 0.001 0.347 0.049 B 40 >32 0.291 1.9 0.005 B 41 >32 0.002 B 44 1.3 0.373 >32 0.016 1.8 B 46 4.2 7.1 0.111 5.4 47 0.005 B >32 0.002 0.056 0.006 B 49 >32 >32 0.008 0.075 B 50 >32 0.005 B 51 6.8 1.7 >32 >32 A 0.040 B 1.6 53 >32 B 14.4 55 A 0.010 0.013 >32 B A 0.001 0.009 56 1.3 >32 >32 A B 0.132 0.386 57 >32 A 0.002 5.9 B 0.045 58 >32 A 0.002 0.007 5.7 59 B >32 A 0.052 1.5 B 0.143 60 A 79 B 0.002 61 4.4 0.013 0.023 >32 A 83 0.002 0.855 2.1 B 62 >32 A 84 0.055 0.003 5.6 B 63 >32 A 86 0.011 0.005 B 64 0.180 >32 >32 87 A >32 B 65 0.031 >32 A 91 4.5 B 0.010 66 0.007 >32 A 92 0.011 0.044 B 0.002 67 >32 A 0.956 93 0.004 B 0.013 68 >32 >32 A 95 0.001 B 0.873 71 A 97 B 0.598 0.006 0.014 73 0.076 99 >32 A B 0.823 74 0.008 0.127 B 1A 0.168 0.038 8.1 B 0.011 75 B 1B 0.009 B 0.009 76 0.886 0.011 2 >32 B 0.201 B 0.002 77 0.057 >32 B B 7 0.064 0.011 78 3.4 B 9 0.012 B 0.012 0.267 0.026 80 B 10 0.009 0.039 B 0.003 81 4.8 11 0.012 B 0.003 0.025 B 0.064 82 12 5.0 B 0.002 1.6 B 0.010 85 B 15 0.003 1.1 B 0.031 88 0.251 16 0.002 2.7 B B 0.040 89 0.042 0.009 18 0.017 B 0.011 90 B 1.2 0.014 0.115 B 94 B 19 0.035 0.263 0.010 0.101 B B 20 2.0 0.003 96 >32 0.049 21 B 98 B 0.002 0.010 >32 0.015 B B 0.008 100 22 1.6 0.002 0.009 B HA B 0.008 23 0.068 0.003 B 24 0.148 0.026 0.012 B 1.4 9 B 25 6.2 0.054 0.239 B 9MH 0.010 B 28 9.1 B 0.587 1.019 0.003 B 9m 0.070 29 a Defined as the concentration required to reduce the inhibition of the formation of formazan from A4TT by 50%. Values are the mediah of at least three separate assays. On the average, the upper and lower limits of the 95% confidence intervals ranged between 0.6 and 2.8 times the median MICs. 3 4 S 6 8 13 14 17 26 27 32 35 37 42 43 45 48 52 54 69 70 72
A A A A A A A A A A A A A A A
60 min after infection, providing a rationale for studying the effect of the drug on viral adsorption. As shown in Fig. SA, pirodavir at the MIC inhibited the adsorption of HRV9, but not that of HRV1A. Indeed, concentrations of pirodavir exceeding the MIC by over 100-fold had only a marginal effect on the adsorption of HRV1A (Fig. SB). Cross-resistance with R 61837. Two mutants of HRV9, designated 9H and 9M, were selected for resistance against R 61837. The MICs of R 61837 for these mutants were
increased 20- and 85-fold, respectively, over those for the wild type (Table 1). The MICs of pirodavir against mutants 9H and 9M were increased 2-fold (not significant) and 23-fold, respectively, over those for the wild type.
DISCUSSION The substituted phenoxy-pyridazinamines represent a new class of broad-spectrum antiviral drugs with significant in
ANDRIES ET AL.
104
ANTIMICROB. AGENTS CHEMOTHER.
TABLE 2. MICs of R 61837 and pirodavir against enteroviruses. Virus
Coxsackievirus
Poliovirus
MIC' (jig/ml) of:
Serotype
A13 A15 A18 A21 B1 B3 B4 1 1 2 3
Mahoney Sabin Mefie Leon
R 61837
Pirodavir
>50 >50 >50 3.5 >50 >50 >50
0.150 0.018 0.150 0.021 0.780 8.5 4.0
>50 >50 >50 >50
0.120 0.085 0.270 0.025
Echovirus
9 11 12
>50 >50 >50
0.019 1.3 0.620
Enterovirus
68 70
>50 3.0
3.5 0.002
a See Table 1, footnote a.
vitro efficacy against the rhinovirus and enterovirus genera of the picornavirus family. Pirodavir is especially active against rhinoviruses, with MICs as low as 0.001 ,ug/ml (0.004 FiM) for some serotypes. Compared with its predecessor, R 61837, a substituted phenyl-pyridazinamine, a greater than 500-fold improvement in potency was obtained, as shown by the drop in EC80 from >32 ,ug/ml for R 61837 to 0.064 ,ug/ml for R 77975. The increase in potency was accompanied by a marked broadening of the spectrum. Whereas R 61837 was active almost exclusively against rhinoviruses from antiviral group B, pirodavir was highly active against rhinoviruses from both antiviral groups (Table 1). Pirodavir, but not R 61837, was also effective in inhibiting all tested enteroviruses, although it had to be used at higher concentrations (EC80 = 1.3) than those necessary to inhibit rhinoviruses
2
a1) 0
Q a 0 am
-0_U
>D
---0--
HRVIA HRV9 HRV14 HRV39 HRV89 HRV Hank's
.1 1 .01 10 100 1000 Ratio of R 77975 concentration to the MIC
FIG. 3. Effect of virus serotype and drug concentration on virus neutralization. Approximately 106 PFU of the indicated serotypes was incubated for 30 min at 33°C with various concentrations of pirodavir (R 77975). The mixtures were then serially diluted until noninhibitory drug levels were reached and plaque assayed.
120
-
0
u 0 0 0. 0E CD
0
-C
0)
--. 0 C.
-C
>
5
15 30 60 240 Incubation time (minutes)
0.01 MIC R 77975 0.1 MIC R 77975 1 MIC R 77975 10MIC R77975 100 MIC R 77975
1440
FIG. 4. Effect of time and drug concentration on virus neutralization.
(Table 2). Close chemical analogs of pirodavir are better inhibitors of enteroviruses such as poliovirus (22), but are less potent inhibitors of rhinoviruses (6). Members of the genera cardiovirus and aphthovirus were not susceptible to pirodavir, nor were other viruses that are important in upper respiratory tract infections, such as influenza virus and human coronavirus. The testing of a set of 17 representative serotypes (4) provided predictive information for the results subsequently obtained after testing all rhinoviruses. A pirodavir concentration of 0.143 ,ug/ml, sufficient to inhibit 14 of 17 screening serotypes (82.3%), also inhibited 84 of 101 serotypes (83.1%). Two of the 17 screening serotypes (HRV42 and HRV45) were not susceptible to pirodavir (MIC > 1 jig/ml). Both serotypes appear at the same region (the right edge of antiviral group A) on the spectral map of serotypes and antiviral compounds (5). Upon testing the other serotypes, it was interesting to see that most other insensitive rhinoviruses identified mapped as a cluster into the same region of antiviral group A. This finding shows that from the results obtained with the 17 serotypes, a prediction can be made about which serotypes are most likely to be resistant to a given compound. Previously described capsid-binding molecules were shown to bind to rhinoviruses into a hydrophobic pocket beneath the canyon floor of VP1 (8, 24). Their differential activity toward serotypes from antiviral groups A and B can be explained by their putative binding to particular amino acids lining this antiviral pocket, which are different in rhinoviruses belonging to different antiviral groups (5). Pirodavir, on the other hand, is active against rhinoviruses from both antiviral groups. Despite having this markedly different antiviral profile, pirodavir apparently binds to the same antiviral target. Indeed, a HRV9 mutant that was resistant to R 61837 was cross-resistant to pirodavir (Table 1). Furthermore, preliminary results of X-ray crystallographic studies indicate that pirodavir binds into the hydrophobic pocket of HRV14 (23). The different antiviral profile, together with the sharing of the antiviral binding site, suggests that the broadspectrum activity of the compound is achieved by its binding to amino acids from the antiviral pocket that are conserved throughout rhinoviruses belonging to both antiviral groups and enteroviruses. An alignment for amino acids lining the pocket in HRV14 reveals that two tyrosines are conserved in all 11 sequenced rhinoviruses and enteroviruses (5). The wide range of susceptibilities of different HRV sero-
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
VOL. 36, 1992
105
TABLE 3. Stabilization of rhinoviruses by pirodavir PFU titer" (loglo PFU/ml) for: Drug concn (xMIC)
HRV1A (MIC, 0.011 uLg/ml) 0 1 10 100
HRV9 (MIC, 0.026 0 1 10 100
4Lmgl)
Virus neutralization (pH 7, 20°C)
Acetate treatment (pH 4 or 5, 20°C)
Citrate treatment (pH 4 or 5, 20°C)
Heat treatment (pH 7, 56°C)
0.0 0.0 -0.1 -0.1
>-3.8 >-3.8 -0.5 -0.3
>-3.8 >-3.8 -0.7 -0.4
>-3.3 >-3.3 -3.3 -1.1
0.0 0.0 -2.6 -3.4
-2.5 -1.1 -2.7 -3.4
>-4.4
-3.0 -2.6 -3.4
-2.9 -1.4 -3.1 -3.8
0.0 -0.2 -0.9 -1.0
-3.3 -3.3 -1.3 -1.2
-3.3 -3.3 -1.5 -1.3
>-3.2 >-3.2 > -3.2 >-3.2
HRV70 (MIC, 0.002 ug/ml) 0 1 10 100
aPFU titers expressed in reference to the virus controls (value 0.0).
types is a common feature of all capsid-binding compounds (4). Slight variations in the amino acid composition of the binding site on the viral capsid of different serotypes are likely to affect the strength of the virus-drug bond and therefore the MIC. Three of four rhinoviruses that are completely resistant to pirodavir (HRV8, HRV42, and HRV87) can be blocked by other capsid-binding compounds (4, 6). This indicates that these viruses do have a drugbinding pocket, but that their hydrophobic pockets are aberrant in amino acid composition and/or shape. Pirodavir binds to most rhinoviruses in a way that depends on serotype, drug concentration, and time. For some serotypes (e.g., HRV1A), binding is almost completely reversible by dilution. For others (e.g., HRV Hank's), a stronger binding results in neutralization of the viral infectivity, which can be reversed only by organic solvent extraction, as is the case with neutralization by antibodies. For a third group of rhinoviruses (e.g., HRV9), the binding results in a neutralization that is not reversible by organic solvent extraction. The properties of pirodavir are consistent with earlier observations obtained with R 61837 (2) and slightly different from those described for chalcone, dichloroflavan, and isoflavans, which can be removed from neutralized serotypes by organic-solvent extraction (10, 17). A classical S-shaped dose-response curve between the extent of neutralization and the ratio of pirodavir concentration to MIC could be observed (Fig. 3). The binding of pirodavir to viral particles results in their stabilization. Several susceptible rhinoviruses could be protected against inactivation by mild acidification or heat (Table 3). This observation supports the concept that the drug exerts its antiviral activity by a direct interaction with the viral protein capsid. The minimal concentrations necessary to inhibit either acetate, citrate, or heat inactivation were not correlated with the MICs, suggesting that inhibition of replication and stabilization are independent events, both resulting from the binding of the drug to the viral capsid. A similar conclusion was obtained in tests with R 61837 against five rhinovirus serotypes (3) and in tests with R 78206 against four poliovirus strains (22). The absence of a correlation
between the MICs of a given compound and the concentrations needed for stabilization of different serotypes does not exclude the possibility that such a correlation exists when several compounds of the same chemical family are being studied within one particular serotype (12). A time-of-addition study indicated that pirodavir had an effect on an early event in the replication of both HRV1A and HRV9. In HRV9, the adsorption step was inhibited by pirodavir concentrations closely related to the MIC (Fig. SA). On the other hand, the adsorption of HRV1A was not inhibited at concentrations greatly exceeding the MIC, indicating that the mode of action is serotype specific. A similar differential effect on rhinovirus serotypes has also been described for WIN 54954, another potent capsid-binding compound (21). The extent of conformational changes induced in the putative receptor-binding region for these viruses, the canyon structure (9), may be responsible for the observed differences in the mode of action. During the past decade, several compounds from different chemical classes with potent in vitro activity against several members of the picornavirus family have been described. They include flavans (7), isoflavans (10), chalcone (14), chalcone amides (18), pyrano-pyridines (15), and isoxazoles (19, 25). All of them are supposed to exert their antiviral activity by binding to virions, inducing conformational and flexibility changes in the capsid proteins and thus inhibiting the adsorption and/or uncoating event of the replication cycle (21). Drug-resistant mutants raised against some of these compounds usually exhibit cross-resistance to others, strongly suggesting a similar mode of action and a sharing of binding sites (3, 17, 18). Few of the compounds mentioned have been advanced to in vivo studies. Several compounds of the WIN series have proved to be efficacious in various animal models of enteroviral disease (16, 25). R 61837, the prototype compound of the pyridazinamine series, has shown clinical efficacy in humans (1). A pilot clinical trial of pirodavir under double-blind, placebo-controlled conditions indicated that the compound has a clinical effect in humans (13). Although pirodavir is inactive against common cold viruses other than rhinoviruses, it can be considered a
ANDRIES ET AL.
106
ANTIMICROB. AGENTS CHEMOTHER.
Q la
LI)
a
-0--
CC
HRV I
9 control
MIC R 77975
10 MIC R 77975
100MICR77975
.2
20
0
40
Time 0
a
60
(minutes)
c
b
0
10U0 U)
CL
150.
0
--0-
HRV IA control
1 MIC R 77975 10 MIC R 77975 -*- 100MICR77975
100.
20
0
40
60
Time (minutes) FIG. 5. Effect of pirodavir (R 77975) on the adsorption of HRV9 (a) and HRV1A (b). Results are expressed as the mean standard ±
error of the mean.
serious new candidate for prophylactic and/or therapeutic control of rhinovirus infections in humans. Members of the same chemical series may also be useful as stabilizers of enteroviral vaccines (22). ACKNOWLEDGMENT
We gratefully acknowledge K. De Clercq, National Institute for Veterinary Research, Brussels, Belgium, for testing pirodavir against foot-and-mouth disease virus.
REFERENCES 1.
Al-Nakib, W., P. G. Higgins, G. I. Barrow, D. A. J. Tyrrell, K. Andries, G. Vanden Bussche, N. Taylor, and P. A. J. Janssen. 1989. Suppression of colds in human volunteers challenged with rhinovirus by a new synthetic drug (R 61837). Antimicrob. Agents Chemother. 33:522-525.
2.
Andries, K.,
B.
Dewindt,
M. De
Brabander, R. Stokbroekx, and
P. A. J. Janssen. 1988. In vitro activity of R 61837, a new antirhinovirus compound. Arch. Virol. 101:155-167. 3. Andries, K., B. Dewindt, J. Snoeks, and R. Willebrords. 1989. Lack of quantitative correlation between inhibition of replication of rhinoviruses by an antiviral drug and their stabilization. Arch. Virol. 106:51-61. 4. Andries, K., B. Dewindt, J. Snoeks, R. Willebrords, R. Stokbroekx, and P. J. Lewi. 1991. A comparative test of fifteen compounds against all known human rhinovirus serotypes as a basis for a more rational screening program. Antiviral Res. 16:213-225. 5. Andries, K., B. Dewindt, J. Snoeks, L. Wouters, H. Moereels, P. J. Lewi, and P. A. J. Janssen. 1990. Two groups of rhinoviruses revealed by a panel of antiviral compounds present sequence divergence and differential pathogenicity. J. Virol. 64:1117-1123. 6. Andries, K., J. Snoeks, and R. Willebrords. Unpublished data. 7. Bauer, D. J., J. W. T. Selway, J. F. Batchelor, M. Tisdale, I. C. Caldwell, and D. A. B. Young. 1981. 4',6-Dichloroflavan (BW683C), a new anti-rhinovirus compound. Nature (London) 292:369-370. 8. Chapman, M. S., I. Minor, M. G. Rossmann, G. D. Diana, and K. Andries. 1991. Human rhinovirus 14 complexed with antiviral compound R 61837. J. Mol. Biol. 217:454-463. 9. Colonno, R. J., J. H. Condra, S. Mitzutani, P. L. Callahan, M. Davies, and M. Murcko. 1988. Evidence for the direct involvement of the rhinoviruses canyon in receptor binding. Proc. NatI. Acad. Sci. USA 85:5449-5453. 10. Conti, C., N. Orsi, and M. L. Stein. 1988. Effect of isoflavans and isoflavenes on rhinovirus lB and its replication in HeLa cells. Antiviral Res. 10:117-127. 11. Couch, R. B. 1984. The common cold: control? J. Infect. Dis. 150:167-173. 12. Gruenberger, M., D. Pevear, G. D. Diana, E. Kuechler, and D. Blaas. 1991. Stabilization of human rhinovirus serotype 2 against pH-induced conformational change by antiviral compounds. J. Gen. Virol. 72:431-433. 13. Hayden, F. G., K. Andries, and P. A. J. Janssen. 1990. Efficacy of intranasal R 77975 for the prevention of experimentally induced rhinovirus infection and illness. Antiviral Res. 13(Suppl. 1):103. 14. Ishitsuka, H., Y. T. Ninomiya, C. Ohsawa, M. Fujiu, and Y. Suhara. 1982. Direct and specific inactivation of rhinovirus by chalcone. Antimicrob. Agents Chemother. 22:617-621. 15. Kenny, M. T., J. K. Dulworth, T. M. Bargar, H. L. Torney, M. C. Graham, and A. M. Manelli. 1986. In vitro antiviral activity of the 6-substituted 2-(3',4'-dichlorophenoxy)-2H-pyranol[2,3-b]pyridines MDL 20,610, MDL 20,646, and MDL 20,957. Antimicrob. Agents Chemother. 30:516-518. 16. McKinlay, M. A., and B. A. Steinberg. 1986. Oral efficacy of WIN 51711 in mice infected with human poliovirus. Antimicrob. Agents Chemother. 29:30-32. 17. Ninomiya, Y., M. Aoyama, I. Umeda, Y. Suhara, and H. Ishitsuka. 1985. Comparative studies on the modes of action of the antirhinovirus agents Ro 09-410, Ro 09-0179, RMI-15,731, 4',6-dichloroflavan, and enviroxime. Antimicrob. Agents Chemother. 27:595-599. 18. Ninomiya, Y., N. Shimma, and H. Ishitsuka. 1990. Comparative studies on the antirhinovirus activity and the mode of action of the rhinovirus capsid binding agents, chalcone amides. Antiviral Res. 13:61-74. 19. Otto, M. J., M. P. Fox, M. J. Fancher, M. F. Kuhrt, G. D. Diana, and M. A. McKinlay. 1985. In vitro activity of WIN 51711, a new broad-spectrum antipicornavirus drug. Antimicrob. Agents Chemother. 27:883-886. 20. Pauwels, R., J. Balzarini, M. Baba, R. Snoeck, D. Schols, P. Herdewijn, J. Desmyter, and E. De Clercq. 1988. Rapid and automated tetrazolium-based colorimetric assay for the detection of anti-HIV compounds. J. Virol. Methods 20:309-321. 21. Pevear, D. C., M. J. Fancher, P. J. Felock, M. G. Rossmann, M. S. Miller, G. D. Diana, A. M. Treasurywala, M. A. McKinlay, and F. J. Dutko. 1989. Conformational change in the floor of the human rhinovirus canyon blocks adsorption to HeLa cell
VOL. 36, 1992
PIRODAVIR, A NEW BROAD-SPECTRUM ANTIPICORNAVIRUS DRUG
receptor. J. Virol. 63:2002-2007. 22. Rombaut B., K. Andries, and A. Boeye. 1991. A comparison of WIN 51711 and R 78206 as stabilizers of poliovirions and procapsids. J. Gen. Virol. 72:2153-2157. 23. Rossmann, M. G. (Purdue University). 1991. Personal communication. 24. Smith, T. J., M. J. Kremer, M. Lou, G. Vriend, E. Arnold, G. Kamer, M. G. Rossmann, M. A. McKinlay, G. D. Diana, and M. J. Otto. 1986. The site of attachment in human rhinovirus 14 for antiviral agents that inhibit uncoating. Science 233:12861293.
107
24a.Stokbroekx, R. A., M. J. M. Van der Aa, M. G. M. Luyckx, and G. A. J. Grauwels. 1989. Pyridazine derivatives as antivirals. Chem. Abstr. 112(5):35876y. 24b.Stokbroekx, R. A., M. J. M. Van der Aa, J. J. M. Willems, and M. G. M. Luyckx. 1985. Anti-virally active pyridazinamines. Chem. Abstr. 104(15):129918a. 25. Woods, M. G., G. D. Diana, M. C. Rogge, M. J. Otto, F. J. Dutko, and M. A. McKinlay. 1989. In vitro and in vivo activities of WIN 54954, a new broad-spectrum antipicornavirus drug. Antimicrob. Agents Chemother. 33:2069-2074.
