Mutation in Enterovirus 71 Capsid Protein VP1 Confers Resistance to the Inhibitory Effects of Pyridyl Imidazolidinone

Abstract

Enterovirus 71 is one of the most important pathogens in the family of Picornaviridae that can cause severe complications in the postpoliovirus era, such as encephalitis, pulmonary edema, and even death. Pyridyl imidazolidinone is a novel class of potent and selective human enterovirus 71 inhibitor. Pyridyl imidazolidinone was identified by using computer-assisted drug design. This virologic investigation demonstrates that BPR0Z-194, one of the pyridyl imidazolidinones, targets enterovirus 71 capsid protein VP1. Time course experiments revealed that BPR0Z-194 effectively inhibited virus replication in the early stages, implying that the compound can inhibit viral adsorption and/or viral RNA uncoating. BPR0Z-194 was used to select and characterize the drug-resistant viruses. Sequence analysis of the VP1 region showed that the resistant variants differed consistently by seven amino acids in VP1 region from their parental drug-sensitive strains. Site-directed mutagenesis of enterovirus 71 infectious cDNA revealed that a single amino acid alteration at the position 192 of VP1 can confer resistance to the inhibitory effects of BPR0Z-194.

Enterovirus 71 (EV71) was first isolated in 1969 in California (J. Blomberg, E. Lycke, K. Ahlfors, T. Johnsson, S. Wolontis, and G. von Zeipel, Letter, Lancet i:112). Two adults and 18 children were infected in that outbreak and one 5-year-old child died. Thereafter, mortalities caused by EV71 were reported in Bulgaria (6), Hungary (23), and Malaysia. In 1998, an EV71 epidemic occurred in Taiwan, with the virus infecting over 120,000 people and killing 78 children (1, 5, 13, 14). The central nervous system is the most vulnerable target of EV71 infection. After the central nervous system is virus infected, a patient can die very quickly from severe complications, such as encephalitis (19) and pulmonary edema (2, 16).

EV71, like other viruses in the family of Picornaviridae, is a small, nonenveloped, spherical particle with a diameter of ∼30 nm. The virus has a single-stranded positive-sense RNA enclosed by the capsid proteins VP1, VP2, VP3, and VP4. The capsid contains 60 structural proteins symmetrically arranged into an icosahedral lattice (15, 31, 32). In addition to protecting the viral RNA from nuclease cleavage, the capsid recognizes the receptors on the surface of the specific host cells (3, 9, 18) and displays antigenicity (20, 38). The surface of the virion has a prominent star-shaped plateau at the fivefold axis of symmetry, surrounded by a deep depression (“canyon”). The canyon has been shown to serve as a receptor-binding site in poliovirus and rhinovirus (29, 30).

Pleconaril, one of the WIN compounds with capsid-binding capability targeting VP1, is a novel agent for treating picornavirus infections. Pleconaril has passed the last stage of clinical trials (4, 26) and has shown excellent antiviral effects for most of the enteroviruses and rhinoviruses (27, 28, 33, 34). However, pleconaril did not neutralize the cytopathic effect (CPE) induced by EV71 (37). Therefore, a series of imidazolidinones based on WIN compound templates were developed and screened for antiviral activities to enteroviruses (37). BPR0Z-194, one of the imidazolidinones, can effectively inhibit the activity of EV71. Since BPR0Z-194 was discovered by computer simulation by using WIN compounds as templates, it is worthwhile in testing biologically whether the viral capsid VP1 protein is indeed a molecular target for BPR0Z-194. The present study proved that BPR0Z-194 targets VP1 by genetically analyzing the drug-resistant EV71 variants. Using mutagenesis in the infectious clone of EV71 revealed that altering a single amino acid in the EV71 VP1 capsid protein can confer resistance to the inhibitory effects of BPR0Z-194.

MATERIALS AND METHODS

Cell culture.

Vero cells (African green monkey kidney cells; the American Type Culture Collection [ATCC] accession no. CCL-81) and MRC-5 cells (normal fetal lung cells; ATCC accession no. CCL-171) were grown in modified Eagle medium; RD cells (rhabdomyosarcoma cells; ATCC accession no. CCL-136) were grown in Dulbecco modified Eagle medium (DMEM). These cell lines were used for virus isolation and propagation. The two kinds of medium were supplemented with 10% fetal bovine serum (FBS), 0.026 M sodium bicarbonate, 100 U of penicillin/ml, 100 U of streptomycin/ml, 0.25 μg of amphotericin B/ml, 2 mM l-glutamine, and 0.1 mM nonessential amino acids. The cells were cultured in a T150 flask at 37°C in an incubator that contained 5% CO₂. The cells were observed daily using a microscope. When the confluence was at an 80% level, the cells were treated at 37°C with 0.025% Trypsin-EDTA for 5 min in an incubator that contained 5% CO₂. A medium containing 10% FBS, 10 times the volume of trypsin-EDTA, was then added, and an optimal volume of cells was separated out.

Virus isolation and propagation.

EV71 (TW/1743/98, TW/2086/98, and TW/2231/98), EV68, coxsackieviruses A9, A10, A16, and A24, coxsackieviruses B1, B2, B3, B4, B5, and B6, echoviruses 9 and 29, and herpes simplex virus type 1 (HSV-1) were isolated from clinical specimens in Clinical Virology Laboratory of Chang Gung Memorial Hospital (Linkou, Taiwan). EV71-BrCr, the prototype of EV71 (ATCC accession no. VR 784), and human rhinovirus type 2 (HRV2) and HRV14 (ATCC accession no. VR482 and VR284) were obtained from the ATCC. EV68 and HSV-1 were amplified by using Vero cells. EV71, coxsackieviruses A9, A10, A16, and A24, coxsackieviruses B1, B2, B3, B4, B5, and B6, and echoviruses 9 and 29 were amplified by using RD cells. HRV2 and HRV14 were amplified by using MRC-5 cells; influenza A virus (A/WSN/33) and influenza B virus (B/HK/72) were amplified by using MDCK cells. When the cells were grown to 80% confluence in a T150 flask, the old medium was discarded, and the cells were washed twice with phosphate-buffered saline (PBS). Each of the viruses was diluted by serum-free minimal essential medium (MEM) and added to susceptible cells for adsorption at 4°C for 1 h. During adsorption, the flask was gently agitated at 15- to 20-min intervals. After the adsorption, the inocula were aspirated and replaced with a fresh medium containing 2% FBS, followed by incubation at 37°C (rhinovirus-infected cells were incubated at 33°C) in an incubator that contained CO₂. Once 90% of the cells showed CPE, the viral supernatants were gathered and centrifuged at 4,000 rpm for 5 min. The clear supernatant was then transferred to a new tube, and the debris was frozen and thawed three times before centrifugation. Finally, the supernatants were pooled and stored at −80°C.

Determining the virus titer via plaque assay.

The Vero cells cultured in a T150 flask were treated with 0.025% trypsin-EDTA and counted with a hemocytometer. The cell concentration was adjusted to 4 × 10⁵/ml, and 1 ml of the cell suspension was placed in six-well tissue culture plates. Then, 1 ml of MEM containing 10% FBS was placed into each well, and the plates were incubated overnight at 37°C in an incubator that contained 5% CO₂. After 20 to 24 h, the cells were washed twice with 1 ml of PBS in six-well plates. EV71 dilutions of 10⁻², 10⁻³, 10⁻⁴, 10⁻⁵, and 10⁻⁶ were prepared in a serum-free MEM. Next, 1 ml of viral dilution fluid was added to the cells on ice for 1 h (viral adsorption). After the 1-h adsorption, the inocula were aspirated, and the six-well tissue culture plates were washed twice with PBS. Then, 3 ml of MEM containing 2% FBS mixed with 3% agarose gel (agarose gel-medium [1:9]) was added to each well, and these six-well tissue culture plates were incubated for 4 days. At 4 days after infection, each well was fixed with 10% formaldehyde for 1 to 2 h. The formaldehyde was then removed, and the plate was stained with crystal violet dye for 1 to 2 min. After the plates were washed and air dried, the plaques were counted, and the virus titer was measured in PFU/milliliter.

Neutralization test.

This assay measured the ability of a test compound to inhibit CPE induced by viruses. The 96-well tissue culture plates were seeded with 200 μl of cells at a concentration of 3 × 10⁵ cells/ml in DMEM with 10% FBS. The cell types used for this antiviral assay were RD cells for enteroviruses, MRC-5 cells for rhinoviruses and HSV, and MDCK cells for influenza viruses. The plates were incubated for 24 to 30 h at 37°C and were used at ca. 90% confluence. Virus (100 50% tissue culture infective doses) mixed with different concentrations of test compounds was added to the cells, followed by incubation at 4°C for 1 h. After adsorption, the infected cell plates were overlaid with 50 μl of DMEM plus 5% FBS and 2% dimethyl sulfoxide (DMSO). The plate was wrapped in Parafilm and incubated at 37°C for 64 h (rhinovirus-infected cells were incubated at 33°C). At the end of incubation, the plates were fixed by the addition of 100 μl of 0.5% formaldehyde for 1 h at room temperature. After the removal of formaldehyde, the plates were stained with 0.1% crystal violet for 15 min at room temperature. The plates were washed and dried, and the density of the well was measured at 570 nm. The concentration required for a test compound to reduce the virus-induced CPE by 50% relative to the virus control was expressed as the 50% infective dose (IC₅₀). All assays were performed in triplicate and at least twice.

Time course study of the antiviral activity of BPR0Z-194.

Monolayers of Vero cells were infected with EV71 at a multiplicity of infection (MOI) of 1. BPR0Z-194 at 10 μM (previously diluted with DMSO to a final concentration of <0.1%) was added to the culture medium at specified times. At −1 h, the cells were incubated with BPR0Z-194 for 1 h, and then the medium was replaced by a new medium containing 1 MOI. EV71 and BPR0Z-194 in equal concentration. After adsorption had occurred for 1 h, the cells were washed and the medium was placed again in 10 μM BPR0Z-194. At time 0 h, BPR0Z-194 was added at the time of adsorption and present at all time. At time 1 h, BPR0Z-194 was added 1 h after viral adsorption and present at all times.

Generation and selection of the resistant viruses.

Approximately 85% of EV71 was observed to be inhibited by BPR0Z-194 at a concentration of 1 μM. EV71 was fully inhibited at a BPR0Z-194 concentration of 2.5 μM. Therefore, 1 μM BPR0Z-194 was used to select the potential resistant viruses, whereas a 2.5 μM concentration was used to isolate the truly resistant viruses from the variant viruses that survived at a lower concentration of BPR0Z-194. The concentration of Vero cells was adjusted to 4 × 10⁵/ml, and 1 ml of cell suspension was put into six-well tissue culture plates. Simultaneously, 200 PFU of EV71/well and 1 μM BPR0Z-194 was added to each well (BPR0Z-194 was diluted with DMSO to a final concentration of <0.1%). These six-well plates were then placed on ice for 1 h and, after the adsorption, the plates were twice washed with PBS. A total of 3 ml of MEM containing 2% FBS and 1 μM BPR0Z-194 was then added to each well, and the plates were placed for 4 days in an incubator that contained 5% CO₂ at 37°C. After the 4-day infection, the supernatants from the six-well plates were pooled and centrifuged at 4,000 rpm and 4°C for 10 min. The debris was then freeze-thawed three times and centrifuged again. Finally, the clear supernatant was termed passage 1. Passage 2 contained the supernatant collected from the Vero cells that were infected by 0.5 ml/well passage 1 and 1 μM BPR0Z-194 by the same method that passage 1 was obtained. Passages 3 and 4 were done with 1 μM BPR0Z-194, whereas passages 5 and 6 were performed with 2.5 μM BPR0Z-194. After passage 6, the viruses were passaged again in the absence of 2.5 μM BPR0Z-194, which was termed “BRV” (for BPR0Z-194-resistant variant) virus here. Figure 1 outlines the strategy for this selection.

FIG. 1.

Scheme for the generation and selection of BRV viruses. We used 1 μM BPR0Z-194 to select the potentially resistant viruses (passages 1 to 4) and 2.5 μM BPR0Z-194 to isolate truly resistant viruses from the viruses that survived at a lower BPR0Z-194 concentration (passages 5 and 6). After passage 6, the viruses were passaged again in the absence of 2.5 μM BPR0Z-194 and called BRV viruses.

Plaque purification and viral RNA isolation.

The BRV virus was then isolated by plaque purification. Seven clear plaques, BRV-1 to BRV-7, were selected and amplified. With >90% of the Vero cells displaying the CPE, the viral supernatant was collected from each cell, and the clear supernatant was gathered via centrifugation. These supernatants were the source of viral RNA isolation for sequencing.

Total RNA was isolated from the supernatant by using TRIZOL LS reagent. The supernatant was lysed directly with TRIZOL LS reagent, homogenized by pipetting, and incubated for 5 min at room temperature. Phase separation was achieved by adding chloroform to the TRIZOL LS suspension in a 1:4 ratio, which was then vortexed and incubated at room temperature for 5 to 10 min. The suspension was centrifuged at 14,000 rpm for 15 min at 4°C, and the RNA-containing upper colorless aqueous phase was collected. The amount of the aqueous phase was ca. 70% of the volume of TRIZOL LS reagent used for homogenization. Meanwhile, the aqueous phase was precipitated by mixing it with isopropyl alcohol in a 1:1 ratio, collecting the aqueous phase volume, and adding 5 to 10 μg of RNase-free glycogen as a carrier to the aqueous phase, followed by final mixing and incubation at −20°C overnight. After a 15-min centrifugation at 14,000 rpm and 4°C, a visible gel-like pellet of RNA precipitate formed. This pellet was washed with 75% ethanol, centrifuged at 14,000 rpm for 15 min at 4°C, and air dried. Finally, the pellet was dissolved in RNase-free water and stored at −20°C.

RT-PCR and sequence analysis.

The VP1 gene of EV71 was located between the VP3 and 2A genes. The isolated viral RNA served as the template for reverse transcription (RT), and a specific primer VP1-RR (5′-AAGGTTTGCCCAGTCATTA) that was complementary to the upstream of the 2A gene (nucleotides 3404 to 3422) was used. The reversed-transcribed product was further used as a cDNA template for polymerization chain reaction amplification, where the specific primers VP1-FF (5′-GCGGCAGCCCAGAAGAA) downstream of the VP3 gene (nucleotides 2364 to 2380) and VP1-RR were involved. The amplified RT-PCR product contains the full 891 nucleotides of EV71 VP1. The RT-PCR used a Reverse-iT One-Step RT-PCR kit (Abgene, Surry, United Kingdom) that included Thermoprime Plus DNA polymerase, optimized reaction buffer, deoxynucleoside triphosphate mix, and MgCl₂ (final concentration, 1.5 μM), and Reverse-iT RTase Blend (including RNase inhibitor). The reaction program included RT at 42°C for 30 min, followed by PCR amplification for 30 cycles, denaturing at 94°C for 1 min, annealing at 55°C for 1 min, and elongation at 72°C for 2 min. The PCR products were sequenced by using the same primers and aligned by using DNASTAR/Lasergene software.

One-step growth curve of resistant virus.

The concentration of Vero cells was adjusted to 2.5 × 10⁵/ml in a 12-well tissue culture plate and incubated at 37°C. After 20 h, the cells were washed twice with 1× PBS. The cells were then infected at an MOI of 5 with BRV virus, which were diluted with serum-free MEM medium and then adsorbed on ice for 1 h. After adsorption, the inocula were discarded and the plate was washed twice with 1× PBS. Each well was then supplied with 1 ml of MEM medium that contained 2% FBS before being incubated at 37°C. The supernatant and debris from each well were collected separately at 1-h intervals from 1 to 12 h. The collected debris was repeatedly freeze-thawed three times before being centrifuged at 1,000 rpm for 10 min. The virus titer was determined by plaque assay.

Mutagenesis in the full-length cDNA of EV71.

The pCR-XL-Topo plasmid (Invitrogen), including the full-length genome of EV71 wild-type stock (TW/2231/98), was constructed, and the VP1 mutants were generated by using a QuikChange site-directed mutagenesis kit (Stratagene). Mutations were made at positions 116 (Tyr to His; pEV71/Y116H), 167 (Glu to Asp; pEV71/E167D), 192 (Val to Met; pEV71/V192 M), and 243 (Ser to Pro; pEV71/S243P).

RESULTS

Antiviral activity of BPR0Z-194.

With WIN compounds as templates, computer-assisted drug design led to the identification of a novel series of imidazolidinone derivatives with significant antiviral activity against EV71 (37). The present study further investigated the antiviral mechanism of a pyridyl imidazolidinone derivative, BPR0Z-194. Figure 2 presents the chemical structure of BPR0Z-194. The activity of BPR0Z-194 against EV71, other enteroviruses, influenza A and B viruses, HSV-1, and HRV2 and HRV14 was tested. As shown in Table 1, BPR0Z-194 inhibited all EV71 of various genotypes (IC₅₀ = 0.32 to 1.21 μM). It also inhibited EV68 (IC₅₀ = 7.78 μM), coxsackievirus A9 (IC₅₀ = 0.38 μM), coxsackievirus A10 (IC₅₀ = 10.34 μM), coxsackievirus A24 (IC₅₀ = 0.1 μM), and echovirus 9 (IC₅₀ = 0.81 μM). No inhibitory activity against coxsackievirus A16, coxsackievirus B1, B2, B3, B4, B5, and B6, echovirus 29, influenza A and B viruses, HSV-1, HRV2, and HRV14 was detected.

FIG. 2.

Chemical structure of BPR0Z-194.

TABLE 1.

Antiviral activity of BPR0Z-194 against various viruses

Virus	IC50a (μM) of BPR0Z-194
EV71, TW/ 1743/98 (genotype B)	0.32 ± 0.027
EV71, TW/ 2086/98 (genotype C)	0.61 ± 0.01
EV71, TW/ 2231/98 (genotype C)	1.18 ± 0.4
EV71, BrCr (genotype A)	1.21 ± 0.11
EV 68	7.78 ± 2.53
Coxsackievirus A9	0.38 ± 0.068
Coxsackievirus A10	10.34 ± 1.22
Coxsackievirus A16	>25
Coxsackievirus A24	0.1 ± 0.05
Coxsackievirus B1	>25
Coxsackievirus B2	>25
Coxsackievirus B3	>25
Coxsackievirus B4	>25
Coxsackievirus B5	>25
Coxsackievirus B6	>25
Echovirus 9	0.81 ± 0.3
Echovirus 29	>25
HRV2	>25
HRV14	>25
Influenza A virus (A/WSN/33)	>25
Influenza B virus (B/HK/5/72)	>25
HSV-1	>25

^aMeans of triplicate well values. All experiments were performed at least twice. A neutralization test was used.

Time course study of the antiviral activity of BPR0Z-194.

A computer simulation suggested that BPR0Z-194 was a capsid binder. A capsid enables viral attachment to the host cells and uncoating during the viral replication cycle. A time course study was therefore performed in measuring the antiviral activity at different times after viral infection. As shown in Fig. 3, an inhibitory effect of >98% was observed when the compound was added before or at the stage of viral adsorption. This compound had a nearly 80 or 50% inhibitory effect when added at 1 or 2 h after adsorption, respectively. Almost no inhibition was observed when this compound was added after 4 h postadsorption. This antiviral activity of BPR0Z-194 was apparent during the early stage of viral infection, suggesting that BPR0Z-194 may interact with the capsid during viral attachment and uncoating.

FIG. 3.

Time course study of the antiviral activity of BPR0Z-194. Vero cells were infected with EV71 at an MOI of 1. First, 10 μM BPR0Z-194 was added to the culture medium at the specified times. Then, for the 0-h time point, BPR0Z-194 was added at the time of adsorption. Virus adsorption was determined at 4°C.

EV71 variants resistant to BPR0Z-194.

EV71 variants resistant to BPR0Z-194 were selected. The plaque morphology of BRV virus was similar to that of wild-type EV71 (data not shown). The one-step growth curve of wild-type EV71 was compared to that of BRV virus (Fig. 4). Vero cells were infected with viruses at an MOI of 5. After adsorption at different times postinfection, the cells were repeatedly freeze-thawed, and the supernatants were collected to determine the virus titer. As shown in Fig. 4, the growth of wild-type EV71 exceeded that of BRV virus in the first 8 h. However, the BRV virus titer caught up with the wild-type EV71 at 12 h postinfection.

FIG. 4.

One-step growth of BRV virus. Vero cells (2.5 × 10⁵/ml) were infected with BRV virus at an MOI of 5. The supernatant and debris from each well were collected separately at 1-h intervals from 1 to 12 h. The collected debris was freeze-thawed three times and then centrifuged at 1,000 rpm for 10 min. The virus titer was determined by plaque assay.

Sequence analysis of BRV virus VP1 gene.

As mentioned, BPR0Z-194 is a pyridyl imidazolidinone, which was found to exhibit anti-EV71 activity based on computer-assisted drug design by using the WIN compounds as skeleton. WIN compounds are considered to be capsid binders and to target capsid protein VP1, so BPR0Z-194 is presumably also able to interact with VP1. The BRV viruses were therefore plaque purified, and the VP1 sequences of six plaque-purified isolates were analyzed in terms of translated amino acids (Fig. 5). Sixteen spots were found to differ from their parental strain. BRV-1, -2, -3, and -4 were found to have 11 identical mutations, whereas BRV-5 had 13 mutations and BRV-6 had 9 mutations. Of these, mutations were found in seven common spots for all six drug-resistant viruses: Arg-22 to Gln, Asp-31 to Asn or Glu, Glu-98 to Lys, Tyr-116 to His, Glu-167 to Asp, Val-192 to Met, and Thr-240 to Ser.

FIG. 5.

Sequence analysis of the BRV virus VP1 gene.

A single amino acid alteration at position 192 in enterovirus 71 VP1 can confer resistance to the inhibitory effects of BPR0Z-194.

In order to prove that the drug-resistant phenotype is caused by mutations in VP1 region, we introduced nucleotide substitutions into an infectious cDNA clone derived from wild-type EV71 (TW/2231/98). A single amino acid mutation in VP1 region of infectious cDNA clone was made at selected positions of 116, 167, 192, and 243, respectively. After each transcribed RNA was transfected into Vero cell, viruses derived from pEV71/Y116H, pEV71/E167D, pEV71/V192M, and pEV71/S243P were treated with BPR0Z-194. There were several reasons we introduced mutations at positions 116, 167, and 192. First, all of the six plaque-purified BRV viruses have the same mutations at these positions. Second, these positions are likely to be located in the hydrophobic region of EV71 VP1, which has been considered a major region interacting with WIN-related compounds. We also made the mutation of Ser-243 to Pro for comparison because only five BRV viruses have a Ser-to-Pro change at this position, whereas BRV-5 retains Ser as the wild type does. As expected, the single mutation of Ser-243 to Pro is still sensitive to BPR0Z-194 treatment (Table 2). For positions 116, 167, and 192, only the virus derived from pEV71-V192M was recovered in the presence of BPR0Z-194. The genome of the resistant viruses was sequenced again to confirm that there was only one single amino acid mutation at position 192 (Val to Met). The drug susceptibility for these mutated viruses or clones was calculated in Table 2. These results demonstrate that mutation in EV71 VP1 can indeed change virus susceptibility to BPR0Z-194.

TABLE 2.

Drug susceptibility of mutant viruses

Virus mutant	Origin	Mutation(s) located in VP1		IC50 (μM)
		cDNA	Amino acid
BRV-1	Spontaneous		-a	>2.5
BRV-5	Spontaneous		-b	>2.5
BRV-6	Spontaneous		-c	>2.5
pEV71/WT	cDNA derived			0.022 ± 0.017
pEV71/Y116H	cDNA derived	TTA→TGA	Tyr116→His	0.21 ± 0.014
pEV71/E167D	cDNA derived	AGT→ACT	Glu167→Asp	0.031 ± 0.004
pEV71/V192M	cDNA derived	GTT→ATG	Val192→Met	>2.5
pEV71/S243P	cDNA derived	AGT→AGC	Ser243→Pro	0.007 ± 0.001

^aBRV-1 contains 11 mutations sites on VP1: Arg18→Lys, Arg22→Gln, Asp31→Asn, Glu98→Lys, Tyr116→His, Glu167→Asp, Val192→Met, Thr240→Ser, Ser243→Pro, Ala275→Ser, and Ala289→Thr. (BRV-2 to -4 contain the identical 11 mutation sites).

^bBRV-5 contains 13 mutation sites on VP1: Val16→Arg, Arg18→Lys, Arg22→Gln, Ala23→Thr, Asp31→Glu, His37→Leu, Val44→Ser, Gln48→Pro, Glu98→Lys, Tyr116→His, Glu167→Asp, Val192→Met, and Thr240→Ser.

^cBRV-6 contains nine mutation sites on VP1: Arg22→Gln, Asp31→Glu, Glu98→Lys, Tyr116→His, Glu167→Asp, Val192→Met, Thr240→Ser, Ser243→Pro, and Ala275→Ser.

DISCUSSION

With the rapid advances in computational and structural biology, computer-assisted drug design has become a useful tool for drug discovery, especially for newly emerging infectious diseases (37). Capsid-binding molecules are able to block viral uncoating and/or attachment to host cell receptors, which represents a good beginning point for developing potential antiviral compounds against enteroviruses and rhinoviruses. The site to which the capsid-binding compounds bind appears to be a hydrophobic pocket inside VP1 located in the canyon floor. EV71 is the most virulent enterovirus in the postpolio era. Computer simulation was used to design a series of pyridyl imidazolidinone compounds that act as capsid binders to inhibit EV71 replication. BPR0Z-194, one of these imidazolidinones, was found to effectively inhibit EV71 activity. The present study proves by an analysis of the resistant EV71 that BPR0Z-194 directly targets the VP1 protein.

The capsid binders, WIN compounds, play a remarkable role in the development of antiviral agents against both rhinovirus and enterovirus infections. Disoxaril, also known as WIN 51711, was the first in this compound family with satisfactory biological profiles to undergo a series of clinical trials (7, 8, 17, 21, 24, 25). It showed oral efficacy in preventing poliovirus type 2- and echovirus type 9-induced paralysis in mice. The clinical studies for disoxaril were discontinued due to the appearance of crystalluria in healthy volunteers at high dosages. Its successors, WIN 54954 (35, 36, 39) and WIN 61605, were then developed and evaluated. However, neither compound was researched further because of toxicity or low efficacy. Continuation of the search for structurally related bioisosteric molecules with a significant reduced liver toxicity finally yielded WIN 63843, also referred to as pleconaril, as a promising new drug candidate for the treatment of human enterovirus infections (4, 27, 28, 33, 34). In addition to much greater metabolic stability in the monkey liver microsomal assay, this newly developed 5-methyl-oxadiazole analogue was shown to be more potent than its oxazoline (WIN 54954) and tetrazole (WIN 61605) predecessors against a variety of rhinoviruses and enteroviruses. Pleconaril is orally administered and is currently being developed by ViroPharma to treat diseases associated with picornavirus infections. This drug candidate is in late-stage clinical trials for treating viral respiratory infections and viral meningitis. Pleconaril can neutralize the CPEs induced by many enteroviruses; however, it seems to have no effect on EV71-induced CPE (37).

Using the skeleton of WIN compounds as structural templates, a structure-based drug design group at the National Health Research Institutes (NHRI) in Taiwan has recently reported a library of virtual compounds whose minimum-energy conformations are highly similar to those of the VP1 pocket of HRVs and may fit well into this pocket. Preliminary studies identified a series of imidazolidinone derivatives that showed potent activity against various enteroviruses, including EV71 (IC₅₀ = 0.35 to 0.04 μM), coxsackievirus A9 (IC₅₀ = 0.47 to 0.55 μM), and coxsackievirus A24 (IC₅₀ = 0.47 to 0.55 μM). The specificity for EV71 renders this series of compounds extremely significant and useful for developing potential anti-EV71 agents.

The study of mutant viruses acquiring drug resistance would explicate the molecular mechanism of the antiviral effect of this compound. The WIN compound-resistant rhinoviruses have been studied for some time (11, 12). Genetic and structural analysis of the rhinovirus 14 mutants resistant to WIN compound revealed that the binding site for the compound was located in the hydrophobic pocket of capsid protein VP1. The pleconaril-resistant strain of coxsackievirus B3 was selected and tested with the murine model. The variants contained two genetic changes in VP1 at position 92 from Ile to Leu or Met and at position 207 from Ile to Val (10). In the present study we selected and analyzed the EV71 mutants that are resistant to the newly discovered capsid-binder BPR0Z-194. We showed that the mutation at position 192 of VP1 rendered the virus resistant to BPR0Z-194. It seems that the interaction between WIN-related compounds and virus particles is located within or near the hydrophobic pocket of VP1. However, different viruses exhibit distinguishing spots for the binding to these compounds.

Pyridyl imidazolidinone is the first compound reported to have the anti-EV71 activity in vitro. Our finding here demonstrated that BPR0Z-194 targets to capsid protein VP1. This anti-EV71 drug can inhibit viral adsorption and/or viral uncoating, which occurs in the early stages of viral replication. The target of the WIN-related compound is located in the hydrophobic pocket of VP1. The locations of the VP1 hydrophobic pocket of HRV14, poliovirus type 1 (Mahoney strain), and coxsackievirus B3 have been determined based on the crystal structures (22). For example, the middle region of VP1 (residues 100 to 250 in HRV14) that combines with multiple noncontiguous amino acids within this ∼150-amino-acid stretch could form the hydrophobic pocket in VP1. Subtle differences exist in the numbers and positions of VP1 residues that comprise the hydrophobic pocket among HRV14, poliovirus type 1, and coxsackievirus B3 (22). The crystal structure of EV71 has not been obtained thus far. However, pyridyl imidazolidinone was generated by recombinatory chemistry based on that the fact that it can fit into the hydrophobic pocket of VP1. Our findings here may provide useful information regarding the location of the hydrophobic pocket of EV71. Position 192 in VP1 of EV71, which was found to play an important role in interacting with BPR0Z-194, is probably a part of the hydrophobic pocket. From sequence analysis of the resistant BRV viruses (Fig. 5), it appears that positions 22, 31, 98, and 240 may also have some impact on interaction with BPR0Z-194, which implies that these positions might be contained within or have strong interaction with the hydrophobic pocket of EV71.

EV71 is the most virulent enterovirus to emerge since the global eradication of polioviruses. Development of an antiviral agent for EV71 is thus significant and urgent. Previous studies of WIN compounds by other scientists have led to the successful design of the novel anti-EV71 agent, BPR0Z-194. The mechanistic and virologic investigation referred to here provides a good understanding of this drug design, which can be used as a model strategy in further developing other antiviral agents for control of the newly emerging viruses in the Picornaviridae family.