INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The Picornaviridae family consists of more than 200 different viruses, which are associated with a wide variety of medically important diseases including the common cold, aseptic meningitis, conjunctivitis, encephalitis, and respiratory disease (reviewed in references 7, 39, and 44). Human rhinoviruses (HRV), which include over 100 different virus serotypes, are the most important etiological agents of the common cold. Although HRV-induced upper respiratory illness is often mild and self-limiting, HRV infection may ultimately result in sinusitis, otitis media, and lower respiratory tract illnesses including exacerbations of asthma, cystic fibrosis, and bronchitis in individuals with underlying respiratory disorders (3). Although no effective antiviral therapies for either the prevention or treatment of diseases caused by HRV infection are currently available, considerable progress in the discovery and development of new antirhinoviral drugs directed towards a novel target, the HRV 3C protease, has recently been made (11-15, 19, 22, 26, 36, 41, 47, 51-53).

The HRV 3C protease is responsible for the cleavage of viral precursor polyproteins into structural and enzymatic proteins which are essential for viral replication. DNA sequence comparisons among HRV serotypes, and even among several related picornaviruses, have identified a significant degree of homology within the 3C coding region including the strict conservation of the active-site residues, thus providing an additional rationale for targeting drug discovery efforts (8, 16, 20, 21, 29, 31, 38, 45, 48, 49, 55). AG7088 is a potent, irreversible inhibitor of HRV 3C protease that was discovered by protein structure-based drug design methodologies (12, 35, 36) and is currently undergoing evaluation in phase I clinical studies with humans. In this study, we describe the in vitro activity of AG7088 against a variety of different HRV serotypes as well as related picornaviruses in different cell-based systems, as well as its cytotoxicity in these systems.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Compounds. AG7088 and pleconaril (17) were synthesized at Agouron Pharmaceuticals, Inc. Pirodavir (1) was kindly provided by Janssen Research Foundation (Beerse, Belgium), and WIN 51711 (40) was kindly provided by Sterling Winthrop Research Institute (Collegeville, Pa.). Ganciclovir (Syntex Corp., Palo Alto, Calif.) was obtained from a local pharmacy, and acyclovir was purchased from Sigma (St. Louis, Mo.).

Cells and virus strains. All numbered HRV serotypes, echovirus type 11 (EV 11), enterovirus type 70 (ETV 70), coxsackievirus types A21 (CAV 21) and B3 strain Nancy (CVB 3), human cytomegalovirus (HCMV) strain AD169, and herpes simplex virus type 1 (HSV-1) strain McIntyre were purchased from the American Type Culture Collection (ATCC; Manassas, Va.). HRV Hanks and a nasal lavage from a patient challenged with HRV Hanks were kindly provided by Ronald Turner from the Medical University of South Carolina, Charleston, S.C. HRV and coxsackievirus stocks were propagated, and antiviral assays were performed, in H1-HeLa cells (ATCC) incubated at 34 and 37°C, respectively. ETV 70, EV 11, and HCMV stocks were propagated, and antiviral assays were performed, in MRC-5 (ATCC) cells at 37°C. HSV-1 stocks were propagated, and antiviral assays were performed, in Vero (ATCC) cells incubated at 37°C. Vero cells were grown in minimal essential medium (Life Technologies, Gaithersburg, Md.) supplemented with 5% fetal bovine serum (Hyclone, Logan, Utah). H1-HeLa cells and MRC-5 cells were grown in minimal essential medium supplemented with 10% fetal bovine serum.

Enzyme assays. The proteolytic activity of HRV 14 3C protease was measured by a continuous fluorescence resonance energy transfer assay as described previously (11-15). In brief, cleavage of the substrate peptide was monitored by the appearance of fluorescent emission at 490 nm (following excitation at 336 nm) in a Perkin-Elmer LS50-B spectrophotometer. Data were analyzed with the nonlinear regression analysis program ENZFITTER, which calculates a first-order rate constant for the inactivation of HRV 14 3C protease. Protease selectivity assays were performed with commercially available proteases (at approximately 10 nM concentrations) essentially as described by the supplier. Human liver cathepsin B, porcine erythrocyte calpain I, and human neutrophil elastase were purchased from Calbiochem (San Diego, Calif.), bovine chymotrypsin and human thrombin were purchased from Boehringer Mannheim (Indianapolis, Ind.), and bovine trypsin was purchased from Sigma.

Cell protection assay. The ability of compounds to protect cells against infection was measured by a dye reduction method (54). Briefly, H1-HeLa and MRC-5 cells were resuspended at 2 × 10⁵ and 5 × 10⁴ cells per ml, respectively, in medium containing appropriate concentrations of compound or medium only. In some experiments, assays were performed in the presence or absence of either human ₁-acid glycoprotein (AAG) or type 1-S bovine submaxillary gland bovine mucin (Sigma), both at a final concentration of 1 mg/ml in medium containing 10% fetal bovine serum. Cells were infected with HRV, CAV 21, and CVB 3 at a multiplicity of infection (MOI) of 0.004 to 0.5, 0.03, and 0.08, respectively, or mock-infected with medium only. An MOI of 2.0 to 6.0 was used in assays utilizing HRV 25. MRC-5 cells were infected with EV 11 or ETV 70 at an MOI of 0.003 or 0.004, respectively, or mock-infected with medium only. Vero cells were resuspended at 1.5 × 10⁵ cells per ml and infected with HSV-1 at an MOI of 0.05 or mock-infected with medium only. One to five days later, 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT; Sigma) with phenazine methosulfate (Sigma) was added to the test plates and the amount of formazan produced was quantified spectrophotometrically at a test reference of 450 nm and a reference wavelength of 650 nm. Data were expressed as the percentage of formazan produced in compound-treated cells compared to formazan produced in wells of uninfected, compound-free cells. The 50% effective concentration (EC₅₀) was calculated as the concentration of compound that increased the percentage of formazan production in infected, compound-treated cells to 50% of that produced by uninfected, compound-free cells. The 50% cytotoxicity concentration (CC₅₀) was calculated as the concentration of compound that decreased the percentage of formazan produced in uninfected, compound-treated cells to 50% of that produced in uninfected, compound-free cells. The therapeutic index was calculated by dividing the cytotoxicity (CC₅₀) by the antiviral activity (EC₅₀).

Time-of-addition assay. Subconfluent monolayers of H1-HeLa cells in six-well plates were infected with HRV 14 at an MOI of 15. After 1 h of adsorption, cell monolayers were washed three times with phosphate-buffered saline (PBS) and replenished with medium. AG7088 (0.5 µM) or WIN 51711 (3.0 µM) was added at concentrations 20-fold above the EC₅₀ (as determined by the cell protection assay) at the time of infection and at various times thereafter. Eight hours after infection, samples were processed by three freeze-thaw cycles followed by sonication for 15 s and clarification by centrifugation (5 min at 15,000 × g at 4°C). Clarified cell and supernatant lysates were stored at 70°C for subsequent analysis for infectious virus.

Virus yield assay. Infectious virus titers were determined by a virus plaque assay. Briefly, 0.2 ml of serial 10-fold dilutions of virus were allowed to adsorb onto monolayers of H1-HeLa cells. After 1 h of adsorption, the cell monolayers were washed twice with PBS and overlayed with medium containing 0.5% SeaPlaque agarose (FMC Bioproducts, Rockland, Maine). After 3 days of incubation at 34°C, the cell monolayers were fixed with EAF (65% ethanol, 22% acetic acid, and 4% formaldehyde) and stained with 1% crystal violet, and virus plaques were enumerated. Data were expressed as PFU per milliliter.

Analysis of proteolytic processing. The ability of AG7088 to inhibit HRV 14 3C-mediated proteolytic processing was assessed by polyacrylamide gel electrophoresis (PAGE) of radiolabeled sodium dodecyl sulfate (SDS)-solubilized lysates of HRV 14-infected cells. Initially, H1-HeLa cells were infected with HRV 14 at an MOI of 10. Eight and one-half hours after infection, the cells were washed with PBS and the medium was replaced with methionine- and cysteine-deficient medium (Life Technologies). At 9 h after infection, appropriate concentrations of compounds were added. After a 30-min exposure to compounds, 50 µCi of [³⁵S]Met-[³⁵S]Cys (Expre³⁵S ³⁵S protein label; New England Nuclear, Boston, Mass.) was added. One hour later, the monolayers were washed twice with cold PBS and lysed in 250 µl of radioimmunoprecipitation assay buffer (50 mM Tris-Cl [pH 8.0], 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS), sonicated, and stored at 70°C for subsequent analysis. Proteins present in the solubilized cell lysates were resolved by 12% PAGE. Following electrophoresis, gels were stained with Coomassie brilliant blue, destained, and treated with Amplify (Amersham, Arlington Heights, Ill.). Gels were air dried overnight in cellulose sheets and exposed to film at 80°C.

HCMV antiviral assay. The antiviral activity of AG7088 against HCMV AD169 replication in MRC-5 cells was determined by an enzyme-linked immunosorbent assay (ELISA) using a monoclonal antibody MAb directed against the HCMV major immediate-early gene product (MAb 810; Chemicon, Temecula, Calif.). Briefly, following a 2-h virus adsorption, the inoculum was removed and medium containing the appropriate concentrations of compound was added. Five days after infection, the MRC-5 monolayers were incubated with MAb 810, followed by goat anti-mouse antibody conjugated with horseradish peroxidase (Bio-Rad, Hercules, Calif.), and viral antigen was then detected spectrophotometrically at 650 nm with the tetramethylbenzidine liquid substrate system (Sigma). The EC₅₀ was calculated as the concentration of compound that reduced the optical density to 50% of that of the virus control. The CC₅₀ was measured by the XTT reduction method as described above.

Statistical analyses. Determination of statistical significance was made by using Student's unpaired t test with the Statview (SAS Institute, Inc., Cary, N.C.) software program.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Activity against HRV 3C protease. AG7088 is a ketomethylene-containing peptidomimetic compound which incorporates an unsaturated ethyl ester Michael acceptor (Fig. 1). AG7088 has demonstrated potent and irreversible inhibition of HRV 3C protease with an inactivation rate constant (k_obs/[I]) of 1,470,000 ± 440,000 M¹ s¹ for HRV 14. Enzyme inhibition was shown to be specific for the viral 3C protease since 10 µM AG7088 produced no significant inhibition against a variety of serine or cysteine proteases, e.g., human elastase, human thrombin, bovine trypsin, bovine chymotrypsin, human and bovine cathepsin B, and porcine calpain. In addition, lack of reactivity with nonenzymatic thiols was demonstrated by AG7088's stability when incubated in the presence of dithiothreitol (5 mM) (12).

View larger version (9K): [in this window] [in a new window]   FIG. 1.   Chemical structure of AG7088.

Activity against HRV serotypes and cytotoxicity of AG7088. The efficacy of AG7088 against a panel of 48 different HRV serotypes was evaluated in a cell protection assay utilizing H1-HeLa cells (Fig. 2; Table 1). These included representative virus strains derived from minor and major receptor groups (50) as well as from two antiviral groups (A and B) previously defined based on differing susceptibilities to capsid-binding molecules (2). Pirodavir and pleconaril, compounds which inhibit virus capsid attachment and/or uncoating (1, 17), were included for comparison. Results indicate that AG7088 was active against all HRV serotypes (48 of 48) tested with a mean EC₅₀ of 0.023 µM (range, 0.003 to 0.081 µM) and a mean EC₉₀ of 0.082 µM (range, 0.018 to 0.261 µM). These values were comparable to or significantly better than those obtained with both pirodavir and pleconaril; pirodavir inhibited the replication of 42 of 47 (89%) HRV serotypes tested with a mean EC₅₀ of 0.329 µM (range, 0.003 to 4.770 µM), and pleconaril inhibited the replication of 42 of 45 (93%) HRV serotypes tested with a mean EC₅₀ of 0.822 µM (range, 0.003 to 8.122 µM). Furthermore, although AG7088 was able to inhibit approximately 80% of the HRV serotypes tested with EC₅₀ and EC₉₀ of less than or equal to 0.038 and 0.110 µM, respectively, EC₅₀ of less than or equal to 0.579 and 0.862 µM and EC₉₀ of less than or equal to 5.84 and 3.96 µM were necessary for pleconaril and pirodavir, respectively, to inhibit this same percentage of HRV serotypes (Fig. 2). Consistent with results obtained with pirodavir, AG7088 demonstrated comparable levels of activity against HRV serotypes derived from either major or minor receptor groups as well as from antiviral groups A or B (Table 1). In contrast, significant reductions in activity against HRV serotypes classified in antiviral group A were observed for pleconaril.

View larger version (23K): [in this window] [in a new window]   FIG. 2.   In vitro activity of AG7088 against HRV serotypes. EC50 (A) and EC90 (B) of AG7088, pleconaril, and pirodovir for 48, 45, and 47 HRV serotypes, respectively, were determined by measuring XTT dye reduction following 2 to 5 days of infection of H1-HeLa cells as described in Materials and Methods.

Antiviral activity of AG7088 after HRV infection. A time-of-addition assay was performed to determine the efficacy of AG7088 when added at various times after virus infection. For this purpose, H1-HeLa cells were infected with HRV 14 at a high MOI to achieve a single cycle of virus replication and levels of infectious virus were determined 8 h later. Results indicated that the addition of AG7088 to infected cells could be delayed up to 6 h after infection without a significant loss of in antiviral activity (Fig. 3). In contrast, a compound (WIN 51711) which binds to virus capsids and acts by inhibiting virus uncoating (40) was inhibitory only when provided at the initiation of the virus life cycle; a loss of virus suppression was observed when addition of the compound was delayed until 2 h after infection.

View larger version (16K): [in this window] [in a new window]   FIG. 3.   Antiviral activity of AG7088 when added at various times after virus infection. H1-HeLa cells were infected with HRV 14 at an MOI of 15. AG7088 (0.5 µM) and WIN 51711 (3.0 µM) were added at various times after infection. Virus control represents infected cells incubated with medium only. Eight hours after infection, cell lysates and supernatants were collected and the level of infectious virus was determined as described in Materials and Methods.

Inhibition of HRV 3C proteolytic processing. To confirm that the in vitro antiviral activity of AG7088 was derived from a direct inhibition of HRV 3C-mediated proteolytic processing, HRV14-infected H1-HeLa cells were treated with AG7088 (2.0 to 0.5 µM) and radiolabeled polyproteins were resolved by SDS-PAGE. Pleconaril was included as a negative control. SDS-PAGE analysis (Fig. 4) indicated a dose-dependent accumulation of large HRV 14 precursor polyproteins with a concomitant reduction of low-molecular-weight cleavage products in cells treated with AG7088 but not pleconaril. Polyprotein cleavage products predicted to accumulate following inhibition of 3C-mediated proteolytic processing, e.g., P1 (97 kDa) and P2-P3 (146 kDa) were observed. Likewise, a predicted reduction in the polyprotein cleavage products, e.g., VP1 (33 kDa), VP0 (37 kDa), 2C (38 kDa), and 3CD (72 kDa), was also observed.

View larger version (73K): [in this window] [in a new window]   FIG. 4.   Inhibition of HRV 14 3C-mediated proteolytic processing by AG7088. SDS-solubilized lysates were prepared from uninfected cells (cell ctrl; 0.1× indicates 1/10 the amount of uninfected cell lysate analyzed in lane designated 1×) or infected cells treated with AG7088, pleconaril, or medium only (virus ctrl), and equal amounts of protein were analyzed by PAGE as described in Materials and Methods. P2-P3, P1, 3CD, VP0, 2C, and VP1 designate HRV-specific polypeptides. MWT, molecular weight (in thousands).

Effect of protein on the antiviral activity of AG7088. Since evaluation of AG7088 in human clinical studies involves delivery by the intranasal route, it was of interest to evaluate the potential effects of proteins present in nasal secretions, AAG and mucin, on the in vitro antiviral activity of AG7088 in a H1-HeLa cell protection assay. Although the physiological concentration of AAG or mucin in nasal fluid is not known with certainty, AAG at a concentration of 1 mg/ml, which represents a physiologically relevant concentration in plasma (30), and mucin at a concentration of 1 mg/ml, which represents the maximum soluble concentration, were resuspended in medium containing 10% fetal bovine serum for these experiments. Under these conditions, no statistically significant differences between the antiviral activity of AG7088 in the presence of either AAG or mucin (P > 0.05; Table 2) and the antiviral activity of AG7088 in the absence of AAG or mucin were observed. The mean EC₅₀ and EC₉₀ for AG7088 against HRV 14 in the absence of AAG were 0.021 and 0.040 µM, respectively, while in the presence of AAG, the mean EC₅₀ and EC₉₀ were 0.043 and 0.108 µM, respectively. Similarly, the EC₅₀ and EC₉₀ in the absence of mucin were 0.059 and 0.160 µM, respectively, while in the presence of mucin the EC₅₀ and EC₉₀ were 0.030 µM and 0.074 µM, respectively. No cytotoxicity was observed up to concentrations of 1 and 10 µM AG7088 in the presence of either AAG or mucin, respectively.

Activity of AG7088 against related picornaviruses. The efficacy of AG7088 against four related picornaviruses was also examined. Pirodavir and pleconaril were included for comparison. In H1-HeLa or MRC-5 cell protection assays, AG7088 was active against all four picornaviruses tested, with EC₅₀ ranging from 0.007 to 0.183 µM and EC₉₀ ranging from 0.033 to 0.340 µM (Table 3). These values were comparable to or significantly better than those obtained with both pirodavir and pleconaril; pirodavir inhibited the replication of two of the four (50%) picornaviruses tested with EC₅₀ of 4.833 and 0.443 µM, and pleconaril inhibited the replication of three of the four (75%) picornaviruses tested with EC₅₀ ranging from 0.037 to 1.012 µM. SDS-PAGE analysis of CAV 21-infected cells treated with AG7088 indicated a dose-dependent accumulation of large viral precursor polyproteins with a concomitant reduction of low-molecular-weight cleavage products, confirming that the in vitro antiviral efficacy of AG7088 is due to a direct inhibition of picornavirus 3C protease (data not shown).

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

As members of one of the largest families of medically important human pathogens, HRVs are the single major cause of the common cold. HRV-induced upper respiratory illness has also been associated with serious medical complications in individuals with underlying respiratory disorders (3). Although potent in vitro antirhinoviral activity has been described for numerous compounds to date (reviewed in references 5, 9, and 37), in only a few instances have reductions in clinical symptoms and/or virus infection been achieved in clinical trials (3, 9). The vast majority of these compounds act by binding to virus capsids and inhibiting either virus attachment or subsequent uncoating. However, recent reports have described novel inhibitors of 3C protease. The latter include peptide aldehydes (19, 22, 47, 53), isatins (52), and homophthalimides (51). Recently a class of irreversible inhibitors incorporating Michael acceptors, which exhibits potent cell-based antiviral activity with little to no cellular cytotoxicity, has been described (11-15, 26). In this study, we describe the antiviral activity and cytotoxicity of AG7088, a potent, peptidomimetic inhibitor of HRV 3C protease that has recently begun to be evaluated in human clinical studies.

In cell-based assays, AG7088 demonstrated comparable antiviral potencies (27-fold range in EC₅₀) against all 48 HRV serotypes tested. These results were in contrast to those observed with the capsid-binding compounds tested. The latter compounds demonstrated a significantly wider range in potency (1,590- to 2,707-fold range in EC₅₀) and activity against most but not all HRV serotypes tested. The finding of comparable ranges in the activity of AG7088 against numerous HRV serotypes is consistent with DNA sequence analyses performed on 3C protease-coding gene regions, which demonstrate a significant level of homology in substrate/inhibitor binding regions (8, 16, 20, 21, 29, 31, 36, 38, 45, 48, 49, 55). This same level of homology, reflected in the 3C protease-coding regions derived from other picornaviruses, is also consistent with levels of activity demonstrated against four related picornaviruses.

Nasal secretions are biochemically complex and contain many serum proteins and mucous glycoproteins (33, 43, 46) including acid glycoproteins (34, 42) (unpublished observations). Since human clinical studies of AG7088 involve delivery by the intranasal route, it was of interest to determine the potential effects of mucin and AAG, proteins that are present in nasal washings. AAG is a major serum glycoprotein that has been shown to reduce the in vitro antiviral activity of several human immunodeficiency virus protease inhibitors (4, 28, 32). AAG has also been detected in various concentrations in nasal lavages (unpublished data). No significant reductions, however, in the antiviral efficacy of AG7088 in the presence of either AAG or mucin were observed.

HRVs, as members of the picornavirus family, encode a single large polyprotein precursor (reviewed in references 7, 10, 23, 27, 39, and 44) that depends on the virally encoded 3C protease for all posttranslational cleavages with the exception of an initial autocatalytic cleavage by the 2A protease to release P1, the precursor to the viral capsid proteins, an alternative cleavage by 2A to generate 3C' and 3D' products, and a late autocatalytic cleavage of VP0, which occurs during final viral assembly. The necessity of the 3C protease throughout the virus life cycle was confirmed in experiments that demonstrated that the ability of AG7088 to suppress virus replication when added throughout a single cycle of virus replication in a time-of-addition assay. That the antiviral activity of AG7088 was due directly to its inhibition of 3C protease was demonstrated by SDS-PAGE analysis of infected cell lysates, which indicated a dose-dependent accumulation of large HRV 14 precursor polyproteins with a concomitant reduction of low-molecular-weight cleavage products. Profiles of polyproteins predicted to accumulate or be reduced were consistent with the cleavage profile observed when certain amino acid substitutions are introduced into the 3C protease catalytic site (6, 18, 24, 25).

These studies describe the antiviral activity and cytotoxicity of a novel inhibitor of HRV 3C protease. As an irreversible inhibitor, AG7088 may be capable of forming covalent interactions with other proteins and inducing possible toxicities in vivo. The potential for these types of interactions have, however, been minimized by selection of a Michael acceptor with only mild chemical reactivity (12) and certainly can be explored in appropriate clinical trial studies. In summary, the potent activity against all HRV serotypes tested combined with the ability to inhibit virus replication throughout a single cycle of virus replication indicates that AG7088 is a promising new clinical candidate.