|Home | About | Journals | Submit | Contact Us | Français|
West Nile Virus (WNV) is a potentially deadly mosquito-borne flavivirus which has spread rapidly throughout the world. Currently there is no effective vaccine against flaviviral infections. We previously reported the identification of pyrazole ester derivatives as allosteric inhibitors of WNV NS2B-NS3 proteinase. These compounds degrade rapidly in pH 8 buffer with a half life of 1-2h. We now report the design, synthesis and in vitro evaluation of pyrazole derivatives that are inhibitors of WNV NS2B-NS3 proteinase with greatly improved stability in the assay medium.
West Nile virus (WNV) is a mosquito-borne pathogen of the genus, Flavivirus.1 This genus also contains many other human pathogens, such as Dengue virus (Den), Japanese encephalitis virus (JE), and yellow fever virus (YF). First isolated in 1937 from Uganda, WNV has since been widely distributed around the world.2 Infections in humans are usually asymptomatic or cause a mild flu-like illness for a few days called West Nile fever. However, recent infections of WNV have been associated with much higher fatality rates particularly among the senior population.3 Since the identification of WNV in New York in 1999 the virus has distributed itself widely throughout North America, infecting over 28,000 people (see the Center for Disease Control and Prevention site on the World Wide Web at www.cdc.gov/ncidod/dvbid/westnile/index.htm). Currently there is no effective vaccine against flaviviral infections.2 WNV has a single-stranded RNA genome, which encodes a single polyprotein. This consists of three structural (C, prM, and E) and seven nonstructural (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) proteins. Post translational processing of the polyprotein precursor is required to produce functional viral proteins.4 In particular, the WNV NS2B-NS3 proteinase holds promise as a potential target for therapeutic intervention with small molecule drugs. 5-9 Thus the identification of potent small molecule inhibitors of WNV would be a highly useful starting point for drug discovery and development.
We previously reported the discovery of pyrazole derivatives as allosteric inhibitors of WNV NS2B-NS3 proteinase identified by high-throughput screening (HTS) of a small molecule library through the NIH Molecular Libraries Initiative (MLI).10 The structures and data for compounds were deposited to the PubChem database under AID 577 (http://pubchem.ncbi.nlm.nih.gov/assay/assay.cgi?aid=577&loc=ea_ras) and AID 653 (http://pubchem.ncbi.nlm.nih.gov/assay/assay.cgi?aid=653&loc=ea_ras). The hit compounds exhibited potencies ranging from 0.105 – 1.353 μM (Figure 1). However, these pyrazolyl benzoic acid ester derivatives (1-3) were rapidly hydrolyzed in an aqueous buffer (pH = 8) to the corresponding pyrazol-3-ol in approximately 1-2 hours. Clearly these compounds, while potent, were of limited use as tools to investigate the biochemistry and enzymatic kinetics of WNV NS2B-NS3 proteinase in vitro because of their instability. We hypothesized that it might be possible to design and synthesize analogues of the hits which retain potency as inhibitors of WNV NS2B-NS3 proteinase and, in addition, exhibit improved hydrolytic stability in the assay medium. Herein, we report the synthesis and structure-activity relationship (SAR) of pyrazole WNV NS2B-NS3 proteinase inhibitors with substantially increased stability towards hydrolysis in an aqueous medium.
The initial phase of studies to address this problem focused on the development of modified benzoate ester analogues of the lead structures. Our approach to the design of ester derivatives with the potential for improved stability in buffer was to introduce substituents in the ortho positions of the aryl ring in an attempt to prevent hydrolysis of the ester. Compounds 1 and 2 were used as starting points because they were the most potent in that series. We therefore designed and synthesized a focused library of pyrazole ester analogues. The synthetic chemistry used for the preparation of the pyrazole ester analogues of 1 and 2 is outlined in Scheme 1.11
The reaction of commercially available cyanoacetohydrazide 4 with 4-methoxyphenylsulfonyl chloride in ethanol yielded the sulfonamide 5 which precipitated from solution. Cyclization of sulfonamide 5 to pyrazolone 6 was achieved in an ethanolic solution of KOH, followed by neutralization with AcOH. Lastly, pyrazolone 6 was treated with an arylcarbonyl chloride to obtain the pyrazole esters 7. The analogues in this library were designed to probe both the steric and electronic requirements of the benzoic acid ester moiety and the in vitro data for selected examples are shown in Table 1. Although the introduction of ortho difluoro substituents on the aryl ring was tolerated (e.g. 7a, 7b) with a small loss of potency, more bulky substituents such as dichloro (7n) led to a complete loss of activity. Similarly, the presence of electron donating (7o) or withdrawing (7p) substituents in the ortho position gave inactive analogues, although the ortho methyl derivative (7f) retained its micromolar inhibitory activity. In general the addition of substituents led to a reduction in potency for both the benzoate and the thiophenecarboxylate (7c, 7i) analogues (Table 1). To test the effect of N-alkylation of the pyrazole 5-amino group on inhibitory activity we synthesized the monomethylamino (7q) and dimethylamino (7r) derivatives which were prepared by methylation of compound 1 using (CH3)2SO4. Unfortunately this led to loss of inhibitory activity (IC50 > 100 μM). It should be noted that the products of ester hydrolysis 6 and 8 are inactive (IC50 > 100 μM) as inhibitors of WNV NS2B-NS3 proteinase.10
We also investigated the structural requirements important for potency and stability of 3-substituted pyrazolyl esters related to compound 3. Pyrazole derivative 8 was selected as the key synthon for the preparation of ester analogues. The synthetic chemistry used to prepare the second series of pyrazole ester derivatives is outlined in Scheme 2. The reaction of commercially available chloromethyl methoxybenzene with anhydrous hydrazine in MeOH provided the benzyl hydrazine derivative 9, which reacted with ethylacetoacetate in AcOH under reflux to obtain pyrazole 8 in excellent yield. Lastly, the treatment of 8 with the appropriate arylcarbonyl chloride in the presence of Et3N in CHCl3 furnished the corresponding pyrazole esters 10.12
The in vitro data for the pyrazole ester analogues based on compound 3 are shown in Table 2. The potency of the second set of analogues ranged from 4.03 to 9.43 μM in the in vitro assay13, with compound 10a being the most potent.
The stability of the most potent compounds in each series in pH 8 buffer was determined by analyzing the amount of compound remaining with time using LC/MS detection. To ensure an accurate quantitation of degradants chlorpromazine was used as an internal standard. The corresponding half lives are reported in Table 3. We were gratified to observe a significant improvement in aqueous stability for compounds 7e (t½ = 450 min.) and 10a (t½ = 900 min.) compared with the initial hits10 (Table 3). To help address whether compounds appearing to be inactive under the assay conditions were in fact being rapidly hydrolyzed we also tested the stability of the inactive 2,6-dichloro derivative 7n (IC50 > 100 μM). Interestingly, as shown in Table 3, this compound exhibited significantly improved stability (t½ = 300 min.) compared with the initial hit (1).
Based on the data for the stability experiments in a pH 8 buffer, we selected 7e and 10a for additional stability studies in the in vitro assay buffer with or without the enzyme WNV NS2B-NS3 proteinase (Figure 2). The goal in this study was to mimic the assay conditions used for the in vitro experiments and to gain insight into any differences in stability for the two chemical series. As shown in Figure 2, the results were quite different for each compound. Thus, while 7e was relatively stable in the assay buffer, it degraded rapidly in the presence of the enzyme (Figure 2a). On the other hand, the stability of 10a was approximately the same in the presence or absence of the enzyme (Figure 2b). Taken together, the stability data suggest that the relative stability of the benzoate ester derivatives may be related to electronic rather than steric effects of substituents. These results are in agreement with Charton's studies on the hydrolysis of ortho-substituted benzoate esters.14,15
Although at this juncture we had made significant progress in enhancing the stability of the pyrazole ester derivatives, we next investigated pyrazole analogues containing non-hydrolyzable ester isosteres. In particular, we wanted to determine whether it was possible to replace the ester moiety with alcohol, ketone, alkene or amide functional groups in the pyrazole derivatives while retaining potency as inhibitors of WNV NS2B-NS3 proteinase. The pyrazole analogues containing ester isosteres were prepared according to the procedures outlined in Schemes 3 and and4.4. Thus, commercially available 3,5-dimethylisoxazole was reacted with LDA followed by reaction with the corresponding substituted benzaldehyde to yield compound 11 (Scheme 3). The isoxazole derivative 11 was converted to pyrazole 1216 using Raney Ni and 12 was then oxidized using the Dess-Martin reagent to produce the ketone derivative 13. Acid catalyzed dehydration to afford the alkene derivative 14 was accomplished using pTsOH in hot toluene (Scheme 3).
Treatment of 3-methyl-1-phenyl-1H-pyrazol-5-amine with benzoyl chloride in the presence of pyridine and dioxane furnished the amide derivative 15. N-(1-(4-Methoxyphenylsulfonyl)-3-methyl-1H-pyrazol-5-yl)benzamide (17) was prepared by the condensation of 3-aminocrotononitrile with 4-methoxybenzenesulfonohydrazide to afford pyrazole 16 followed by reaction of benzoyl chloride (Scheme 4). The in vitro data for some of the target compounds are shown in Table 4. All of the alcohol derivatives were inactive up to the highest concentration tested (100 μM) while the two ketone derivatives exhibited IC50 values in the high micromolar range. Encouragingly, however, the alkene derivative 14 was more potent with IC50 = 13.8 μM, while the amide derivatives 15 (IC50 = 16.0 μM) and 17 (IC50 = 9.2 μM) showed activity in a comparable range in the in vitro enzyme assay. Compounds 14, 15 and 17 were also tested for stability in pH 8 buffer (Table 5). Interestingly, while the alkene (14) and amide 15 were highly stable, the amide 17 possessed a relatively short half-life of 1.25 h.
In conclusion, we have described the design and synthesis of 3-substituted pyrazole ester derivatives which are active as allosteric inhibitors of West Nile Virus NS2B-NS3 proteinase. Two compounds, 7a and 10a, while less potent than the original hits (1.96 and 4.03 μM IC50, respectively) are significantly more stable in pH 8 buffer. In addition, we have designed, synthesized and evaluated the in vitro activity of a series of pyrazole derivatives containing ester isosteres. Of these analogues, the alkene 14 (IC50 = 13.8 μM) and amide 15 (IC50 = 16.0 μM) derivatives are highly stable inhibitors of WNV NS2B-NS3 proteinase. These compounds, which interact with an allosteric site on the enzyme,10 are promising leads for additional optimization studies and may find utility in in vitro studies to elucidate the biochemistry and enzyme kinetics of WNV NS2B-NS3 proteinase.
This work was supported by NIH grants U01 AI078048 and U54 HG005033.