INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Influenza virus infections, which have plagued humans throughout history (14), continue to be a serious health concern in terms of both morbidity and mortality (1). Current options for the control of influenza virus infections have limitations. Vaccines provide only partial protection due to their underutilization and the variability in the antigenic determinants of the surface glycoproteins (5, 29). Amantadine and rimantadine, the only antiviral agents approved for use for the treatment of influenza A virus infections, are not active against influenza B viruses, and virulent strains resistant to these agents develop quickly in the clinical setting (2, 7). Thus, there remains a need to identify new antiviral agents that can be used to prevent and treat influenza virus infections.

Recently, there has been a great deal of interest in the influenza virus neuraminidase (sialidase) as a potential antiviral target. This enzyme, which is expressed on the surfaces of influenza A and B viruses, hydrolyzes terminal sialic acid residues from glycoproteins, glycolipids, and oligosaccharides. The influenza virus neuraminidase is thought to be required for the elution of newly synthesized virions from infected cells and thus is essential for virus replication (17, 21, 22). In addition, the neuraminidase may facilitate movement of the virus through the mucus of the respiratory tract (4, 13).

Several sialic acid-based neuraminidase inhibitors have been shown to inhibit influenza A and B virus replication in vitro (22, 32, 35). Zanamivir (GG167) (Fig. 1), the most potent of these sialic acid-based inhibitors, has demonstrated efficacy in animal models of influenza virus infection (26, 27, 32, 35) and in studies with humans (9, 10), and it is under clinical development for the treatment of influenza A and B virus infections. However, due to its poor oral bioavailability, zanamivir is applied topically to the respiratory tract as an intranasal spray or inhalant (9, 10, 26, 27).

View larger version (10K): [in this window] [in a new window]   FIG. 1.   Structures of GS 4071, GS 4104, GS 4116, GS 4109, and zanamivir (GG167). Ac, acetyl.

In an attempt to identify potentially orally bioavailable influenza virus neuraminidase inhibitors, we have designed and synthesized a series of carbocyclic transition-state analog inhibitors of the influenza virus neuraminidases in which lipophilic side chains replace the polar glycerol moiety of the sialic acid-based inhibitors (12). GS 4071 (Fig. 1), the lead candidate from this series, is comparable to zanamivir in terms of its ability to inhibit influenza virus neuraminidase activity (K_i, ~1 nM) and virus replication when tested in vitro (12, 20). However, because GS 4071 lacks the polar guanidino and glycerol groups present in zanamivir, we postulated that GS 4071 or a prodrug of GS 4071 might be orally bioavailable.

In this study we have investigated the oral bioavailability of GS 4071 and, for comparison, its more potent guanidino analog, GS 4116 (12, 19) (Fig. 1). On the basis of the observation that esterification of carboxylic groups has increased the oral bioavailabilities of many compounds (30), we have also determined the bioavailabilities of GS 4071 and GS 4116 following oral administration of their ethyl ester prodrugs.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Compounds and reagents. The bioavailabilities of the following five compounds (Fig. 1) were determined in this study: GS 4071; GS 4104, an ethyl ester prodrug of GS 4071; GS 4116, the guanidino analog of GS 4071; GS 4109, an ethyl ester prodrug of GS 4116; and zanamivir (GG167). These compounds were synthesized at Gilead Sciences by previously published procedures (12, 33). Rat plasma and pooled human plasma were purchased from Pel-Freez Biologicals (Rogers, Ark.) and George King Bio-medicals (Overland Park, Kans.), respectively. Influenza A/PR/8/34 (H1N1) virus was from the American Type Culture Collection (Rockville, Md.). The fluorescent neuraminidase substrate 2'-(4-methylumbelliferyl)--D-N-acetylneuraminic acid (MUN) and all other chemicals were purchased from Sigma Chemical Co. (St. Louis, Mo.) unless indicated otherwise. Microfluor "W" white, flat-bottom, 96-well plates were purchased from Dynatech Laboratories (Chantilly, Va.).

Quantitative neuraminidase assay to determine inhibitor concentration. Neuraminidase activity was determined in an enzymatic assay with purified influenza A/PR/8/34 (H1N1) virus as the source of viral neuraminidase. The virus was grown in embryonated hen eggs and was purified from allantoic fluid with a sucrose gradient (15). The purified virus was harvested, resuspended in 10 mM Tris (pH 7.5) containing 0.1 M NaCl and 60% (vol/vol) glycerol, and stored at 70°C until use.

a  d

Conversion of the prodrug to parent compound by plasma esterase activity. The ethyl ester prodrugs GS 4104 and GS 4109 were incubated at a concentration of 50 µM in the presence or absence of plasma for 30 min at 37°C. The amount of parent compound generated during the incubation period was then determined by the quantitative neuraminidase assay described above, and the extent of conversion observed during the 30-min incubation was taken as a relative measure of the stability of the prodrug in plasma. No attempt was made to further characterize the in vitro conversion of the prodrugs to their respective parent compounds.

Pharmacokinetic studies. Studies with animals were conducted in accordance with guidelines set forth in the Guide for the Care and Use of Laboratory Animals (20a). In rat studies, GS 4071, its ethyl ester prodrug GS 4104, GS 4116, its ethyl ester prodrug GS 4109, and zanamivir were each administered to four Sprague-Dawley rats (age, 8 to 10 weeks) as a single intravenous (i.v.) dose (10 mg/kg of body weight or a single oral dose (10 mg-eq/kg) of compound by gavage. The oral doses are presented as milligram equivalents per kilogram to indicate that the dose of compound given by this route has been corrected to ensure delivery of the same amount (moles) of compound delivered in the i.v. dose. This is important when parent compound is given by the i.v. route and the prodrug, which has a different molecular weight, is given by the oral route. In dog studies, a single 5-mg/kg i.v. dose of GS 4071 was administered to five beagle dogs (average weight, 7.9 kg). After a 1-week washout period, the same animals received a 5-mg-eq/kg oral dose of GS 4104. In other studies, groups of four mice (age, 8 to 10 weeks) or three ferrets (average weight, 1.4 kg) received either a single i.v. dose (10 or 1 mg/kg, respectively) of GS 4071 or a single oral dose (10 or 5 mg-eq/kg, respectively) of GS 4104 by gavage. All compounds were administered as aqueous solutions in 0.9% sodium chloride.

Determination of pharmacokinetic parameters. The area under the plasma concentration-versus-time curve (AUC) for the 24 h following administration of compound (AUC_0-24), the terminal-phase half-life (t_1/2), the maximum concentration of compound in plasma (C_max), the time to C_max (T_max), and the mean residence time (MRT) were calculated by a noncompartmental method with PCNONLIN software (Statistical Consultants, Inc., Lexington, Ky.). The oral bioavailability (F) of the parent compound from prodrug or from oral administration of the parent compound was calculated from the AUC_0-24 of the oral dose divided by the AUC_0-24 of the i.v. dose of the parent compound. For example, the oral bioavailability of GS 4071 from orally administered GS 4104 was calculated as follows: F = [(AUC_p.o.)/(dose_p.o.)]/[(AUC_i.v.)/(dose_i.v.)], where AUC_p.o. is the AUC of GS 4071 after oral (p.o.) administration of GS 4104, dose_p.o. is the dose of GS 4104 in milligram equivalents of GS 4071/per kilogram, AUC_i.v. is the AUC of GS 4071 after i.v. administration, and dose_i.v. is the i.v. dose of GS 4071. Total clearance (CL) from plasma was calculated as dose/AUC_i.v.. The volume of distribution at steady-state (V_SS) of drug administered i.v. was calculated as CL × MRT.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Neuraminidase inhibitor prodrugs are readily hydrolyzed to parent compound in rat plasma. As shown in Fig. 2, GS 4071 is a potent inhibitor of influenza virus neuraminidase activity, with an IC₅₀ and an IC₉₀ of 2 and 10 nM, respectively, in an in vitro neuraminidase assay (12, 20). In contrast, GS 4104, the ethyl ester prodrug of GS 4071, exhibited poor activity in this assay, with an IC₅₀ of approximately 100 nM (Fig. 2). GS 4109 is an ethyl ester prodrug of GS 4116, the guanidino derivative of GS 4071, and also was a poor inhibitor of influenza virus neuraminidase activity in the in vitro enzymatic assay, with no inhibitory activity detected at a concentration of 100 nM; for comparison, the IC₅₀ and the IC₉₀ of GS 4116 in this assay were 0.9 and 5 nM, respectively. The parent compounds and the prodrugs were stable under the assay conditions, with no change in potency observed for either of the compounds after 24 h at room temperature (data not shown).

View larger version (15K): [in this window] [in a new window]   FIG. 2.   GS 4104 is converted to GS 4071 by rat plasma esterase activity. Dose-dependent inhibition of influenza virus neuraminidase activity by GS 4104, the ethyl ester prodrug of GS 4071, after a 30-min incubation at 37°C in buffer () or rat plasma (). A dose-response curve for GS 4071, assayed in buffer, is shown for comparison ().

GS 4104 is orally bioavailable in rats. To determine the oral bioavailabilities of GS 4071, GS 4116, and their respective ethyl ester prodrugs, GS 4104 and GS 4109, the four compounds were administered to rats by the i.v. and oral routes. The oral bioavailability of zanamivir was also determined for comparison. A summary of the values of the pharmacokinetic parameters determined from experiments described in this section is provided in Table 1.

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Concentration-time profiles of influenza virus neuraminidase inhibitors in rat plasma. (A) Concentration of GS 4071 following i.v. () or oral () administration of a 10-mg/kg dose of GS 4071 or oral administration of a 10 mg-eq/kg-dose of the prodrug GS 4104 () (datum points represent means ± standard deviations for four animals). (B) Concentration of GS 4116 following i.v. () or oral () administration of a 10-mg/kg dose of GS 4116 or oral administration of a 10-mg-eq/kg dose of the prodrug GS 4109 () (datum points represent the means ± standard deviations for four animals). (C) Concentration of zanamivir following i.v. () or oral () administration of a single 10-mg/kg dose (datum points represent the means ± standard deviations for three animals). A horizontal line indicating a concentration in plasma of 10 nM, the IC90 of GS 4071 in the influenza virus neuraminidase enzymatic assay (Fig. 2), is shown for reference.

GS 4104 is detected in the plasma of dogs and ferrets. To ensure that the bioavailability of the parent compound from orally administered GS 4104 was not specific to rats, similar experiments were performed with mice, dogs, and ferrets. Mice and ferrets were included because they are used in animal models of influenza virus infection, and dogs were chosen because they are often used as a nonrodent species in preclinical studies. A comparison of the values of the pharmacokinetic parameters determined for the four species are summarized in Table 2.

View larger version (16K): [in this window] [in a new window]   FIG. 4.   Concentration-time profiles of GS 4071 and the prodrug GS 4104 in dog and ferret plasma. (A) Concentration of GS 4071 () in dog plasma following administration of an i.v. dose (5 mg/kg) of GS 4071 or the concentration of GS 4071 () and GS 4104 () in dog plasma following administration of an oral dose (5 mg-eq/kg) of the prodrug GS 4104 (datum points represent the means ± standard deviations for five animals). (B) Concentration of GS 4071 () in ferret plasma following administration of an i.v. dose (1 mg/kg) of GS 4071 or the concentration of GS 4071 () and GS 4104 () in ferret plasma following administration of an oral dose (5 mg-eq/kg) of the prodrug GS 4104 (datum points represent the means ± standard deviations for three animals). A horizontal line indicating a concentration in plasma of 10 nM, the IC90 of GS 4071 in the influenza virus neuraminidase enzymatic assay (Fig. 2), is shown for reference.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The influenza viruses are well understood but poorly controlled human pathogens which are responsible for approximately 30,000 deaths each year in the United States alone (18). The demonstration that zanamivir, a potent and selective inhibitor of the influenza virus neuraminidases in vitro, is effective both in animal models of influenza infection and in challenge studies with humans (10, 26, 27, 32, 35) indicates that the influenza virus neuraminidase is a valid antiviral target. However, due to its poor oral bioavailability and rapid elimination, zanamivir is administered topically to the respiratory tract as a nasal spray or inhalant. We have identified GS 4071, a member of a series of novel carbocyclic transition-state analog inhibitors of the influenza virus neuraminidases, as a lead compound with the potential to maintain potency while being orally bioavailable (12).

Although GS 4071 is more lipophilic than zanamivir, with an amino group replacing the guanidino group of zanamivir and a lipophilic 3-pentyloxy group replacing the polar glycerol group, GS 4071 itself exhibited poor oral bioavailability in rats. However, the parent compound was bioavailable (F = 11 to 73%) in rats, mice, dogs, and ferrets following the oral administration of GS 4104, the ethyl ester prodrug of GS 4071. Ester prodrugs have been used to increase the bioavailability of carboxylic acid-containing compounds (30). Once absorbed, these prodrugs are readily hydrolyzed by a variety of esterases present in the blood and tissues of many species and the parent compound is released (16, 24). Notably, esterification of the carboxylic group of GS 4116, the more polar guanidino analog of GS 4071, did not increase its bioavailability.

The results presented in this report indicate that oral administration of GS 4104 led to higher and more sustained concentrations of the active form of the drug in plasma than did oral administration of GS 4071 itself. This was particularly evident in dogs and ferrets, the species with detectable levels of intact prodrug in their plasma. The observation that the apparent terminal t_1/2 of GS 4071 following oral administration to rats (10.6 h) was significantly longer than that following intravenous dosing (1.6 h) indicates that the elimination of GS 4071 is rate limited by slow oral absorption of the polar parent molecule (flip-flop kinetics). In the case of the prodrug GS 4104, i.v. and oral administration to rats gave similar apparent terminal t_1/2s for the parent (6.2 and 7.0 h, respectively). These data suggest that elimination of the parent compound after administration of prodrug is rate limited by conversion to GS 4071 rather than by absorption. While the stability of the prodrug in plasma would be expected to affect its rate of conversion to GS 4071, the ability of prodrug to aid delivery of the compound to extravascular tissues may also play a role in maintaining the concentrations of GS 4071 in plasma. For example, it has recently been demonstrated that GS 4104 binds selectively to the lung tissue of rats (6), and slow release of prodrug from this or other tissue compartments may also contribute to the longer t_1/2 of the parent compound observed after the administration of prodrug.

The ability of GS 4104, like several other basic lipophilic compounds containing a primary amino group (3, 25, 31), to accumulate in lung tissue may also affect its antiviral activity at the primary site of influenza virus infection and replication. Consistent with this hypothesis, Eisenberg et al. (6) have demonstrated that GS 4071 is present in bronchoalveolar lavage fluid following oral administration of the prodrug to rats. Of particular interest, the peak concentration of GS 4071 in the bronchoalveolar lavage fluid was similar to that in the plasma, but it declined more slowly compared to the rate of decline in plasma, indicating that the antiviral activity in lung tissue may be more persistent than is suggested by the plasma concentration-versus-time profile (6). Because these experiments were carried out with rats, which do not have detectable levels of circulating prodrug in plasma, we would anticipate that the accumulation of prodrug in lung tissue might be even more pronounced in animal species such as dogs or ferrets which have greater amounts of circulating prodrug. In this respect, it is noteworthy that prodrug is detected in human plasma samples following the oral administration of GS 4104 (34).

In summary, we have demonstrated that GS 4104 is an orally bioavailable prodrug of the potent influenza virus neuraminidase inhibitor GS 4071 in each of the four animal species tested. We have also demonstrated that in each of the four animal species, including ferrets, in which the lowest bioavailability of parent compound (11%) from orally administered GS 4104 was found, the concentrations of parent compound in plasma are well above those necessary to inhibit influenza virus neuraminidase activity, even at 12 h postdosing. These results are consistent with the subsequent demonstration that GS 4104 administered orally twice daily is effective in the ferret and mouse models of influenza virus infection (11, 20, 28). On the basis of the demonstrated efficacy of orally administered GS 4104 in animal models of influenza virus infection and preliminary animal toxicity data (20), we conclude that GS 4104 has the potential to be an oral agent for the prophylaxis and treatment of influenza A and B virus infections in humans. This conclusion is supported by the recent demonstrations (34) that the levels of GS 4071 in plasma following oral administration of GS 4104 to humans are similar to those reported here and that early treatment with oral GS 4104 is associated with significant clinical efficacy and antiviral effect in experimental influenza A virus infection in humans (8).