2012新型抗流感病毒药物

Design, Synthesis, and in Vitro Biological Evaluation of 1H-1,2,3- Triazole-4-carboxamide Derivatives as New Anti-influenza A Agents Targeting Virus Nucleoprotein Huimin Cheng,†,⊥ Junting Wan,† Meng-I Lin,‡ Yingxue Liu,† Xiaoyun Lu,† Jinsong Liu,† Yong Xu,† Jianxin Chen,§ Zhengchao Tu,† Yih-Shyun E. Cheng,‡ and Ke Ding*,† †Key Laboratory of Regenerative Biology and Institute of Chemical Biology, Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences, 190 Kaiyuan Avenue, Guangzhou 510530, People’s Republic of China ‡Genomics Research Center, Academia Sinica, Taipei, 115, Taiwan, Republic of China §South China Agricultural University, 483 Wushan Road, Guangzhou 510642, People’s Republic of China ⊥Graduate School of Chinese Academy of Sciences, #19 Yuquan Road, Beijing 100039, China *S Supporting Information ABSTRACT: The influenza virus nucleoprotein (NP) is an emerging target for anti- influenza drug development. Nucleozin (1) and its closely related derivatives had been identified as NP inhibitors displaying anti-influenza activity. Utilizing 1 as a lead molecule, we successfully designed and synthesized a series of 1H-1,2,3-triazole-4-carboxamide derivatives as new anti-influenza A agents. One of the most potent compounds, 3b, inhibited the replication of various H3N2 and H1N1 influenza A virus strains with IC50 values ranging from 0.5 to 4.6 μM. Compound 3b also strongly inhibited the replication of H5N1 (RG14), amantidine-resistant A/WSN/33 (H1N1), and oseltamivir-resistant A/WSN/1933 (H1N1, 274Y) virus strains with IC50 values in sub-μM ranges. Further computational studies and mechanism investigation suggested that 3b might directly target influenza virus A nucleoprotein to inhibit its nuclear accumulation. ■ INTRODUCTION Influenza is a seasonally epidemic acute respiratory disease persistently threatening public health caused by infection of ribonucleic acid (RNA) influenza viruses of orthomyxoviridae family. The epidemic and pandemics of influenza have caused serious impact on worldwide morbidity, mortality, and economy.1 Three distinct types of influenza viruses are classified based on their serological subtypes, among which influenza A viruses are the most virulent human pathogens. Influenza A viruses possesses eight segmented RNA genomes. The continuous genetic mutation and reassortment of influenza A viruses may cause the evasion of primary human immunities and lead to the resistance against current therapies. More importantly, the viral genetic shift may eventually cause the breakthrough of interspecies and interhuman transmission barriers which will be disastrous to public health.2−5 Currently, two types of small molecular drugs are available for the treatment of influenza A including M2 ion channel blockers (i.e., amantidine and rimanitidine) and neuraminidase inhibitors (i.e., oseltamivir and zanamivir).6−8 However, the application of M2 ion channel blockers, particularly for amantidine, has been strictly limited because of the rapid emergence of drug resistance and the occurrence of central nervous side effects.9 Oseltamivir resistant viruses have also been reported since 2005,10,11 and almost all the seasonal influenza H1N1 viruses circulating in the United States were resistant to Oseltamivir in the 2008−2009 flu season.12−17 Most recently, cases of reassortment between swine influenza and 2009 influenza A (H1N1) in human were reported,18 which highlights the emergent need for developing new anti- influenza agents with new mechanism of action. The influenza virus nucleoprotein (NP) is encoded by the fifth genome segment and abundantly expressed during the course of infection. Comparing with the viral surface spike proteins like hemagglutinin and neuraminidase, the influenza virus nucleoprotein is highly conserved.20−22 During the viral life cycle, nucleoprotein binds with influenza viral RNA segments and polymerase subunit proteins (PB1, PB2, and PA) to form virus ribonucleoprotein (vRNP) complex, which is transported to host cell nucleus to trigger viral RNA transcription, replication, and virion assembly.19 Owing to its indispensable roles in numerous stages of viral multiplication, influenza virus nucleoprotein has been considered as a novel target for new anti-influenza drug development. Several nucleoprotein inhibitors have been reported to display promising anti-influenza virus activities.23−28 Nucleozin (1)24 is the arguably first small molecular influenza virus nucleoprotein inhibitor reported by Yuen and Su independently.24,25 Studies have demonstrated that 1 and its related derivatives could potently trigger the aggregation of nucleoprotein and inhibit its nuclear accumulation to display Received: October 8, 2011 Published: February 14, 2012 Article pubs.acs.org/jmc © 2012 American Chemical Society 2144 dx.doi.org/10.1021/jm2013503 | J. Med. Chem. 2012, 55, 2144−2153 potent anti-influenza virus activities. Further mutational studies suggested that these inhibitors might directly bind to nucleoprotein and “induce formation of higher-order nucleo- protein oligomers” that are unable to migrate into the nucleus.24−27 However, it remains elusive that how the inhibitors induced the formation of nucleoprotein oligomers.27 None of the current inhibitors has been approved for clinical investigation. It is highly desirable to identify new inhibitors for to further validate nucleoprotein as novel molecular target for anti-influenza A development. Herein, we report the structural design, synthesis, and in vitro biological evaluation of 1H-1,2,3- triazole-4-carboxamide derivatives as new anti-influenza A agents using 1 as the lead compound.27 ■ CHEMISTRY Compounds 2 and 4 were prepared by using ethyl 2-benzoyl-3- oxobutanoate as the same starting material (Scheme 1). The condensation/cyclization of ethyl 2-benzoyl-3-oxobutanoate (6) with hydroxylamine hydrochloride or hydrazine hydrate yielded ethyl 3-methyl-5-phenylisoxazole-4-carboxylate (7) or ethyl 5-methyl-3-phenyl-1H-pyrazole- 4-carboxylate (9), re- spectively. Compound 7 or 9 was hydrolyzed and reacted with 1-(2-chloro-4-nitrophenyl)piperazine to produce the designed (4-(2-chloro-4-nitrophenyl) piperazin-1-yl)(3-methyl-5-phenyl- isoxazol-4-yl)methanone (2) or (4-(2-chloro-4-nitrophenyl)- piperazin-1-yl) (5-methyl-3-phenyl-1H-pyrazol-4-yl) metha- none (4). (4-(2-Chloro-4-nitrophenyl) piperazin-1-yl)(2′- biphenyl)methanone (5) was readily prepared by a direct condensation of 2-phenyl benzoic acid (11) with 1-(2-chloro-4- nitrophenyl)piperazine. Compounds 3 were synthesized by using a traditional “3 + 2” cycloaddition as the key step.29−31 Briefly, the anilines 12 were diazotized and then reacted with NaN3 to produce the azidobenzenes 13. Compounds 13 were directly reacted with Scheme 1. Synthesis of Compounds 2, 4, and 5a aReagents and conditions: (a) NH2OH·HCl, ethanol aq, 60 °C overnight, 76.4%; (b) NaOH, MeOH:THF:H2O = 1:1:3, overnight, 95.7%; (c) (COCl)2, dichloromethane, 2.0 h, then 1-(2-chloro-4-nitrophenyl)piperazine, Et3N, dichloromethane, 3.0 h, 60−80%; (d) N2H4·H2O, ethanol, reflux overnight, 76.4%; (e) BBr3, DCM, −10 °C, 3.0 h, 56.2%. Scheme 2. Synthesis of Compounds 3 and the X-Ray Structure of Intermediate 16aa aReagents and conditions: (a) H2O/ethyl acetate, then HCl,NaNO2; (b) NaN3, 0 °C, 2.0 h,59.0%; (c) toluene, reflux, overnight, 80.9% (14 + 15); (d) LiOH, MeOH:THF:H2O = 1:1:3, overnight, 96%; (e) (COCl)2, dichloromethane, DMF cat., 2.0 h, then 1-(2-chloro-4-nitrophenyl)piperazine, Et3N, dichloromethane, 3.0 h, 58.0%. Journal of Medicinal Chemistry Article dx.doi.org/10.1021/jm2013503 | J. Med. Chem. 2012, 55, 2144−21532145 different alkynoates to obtain the key intermediates ethyl 3- phenyl-3H-1,2,3-triazole-4-carboxylates (14) and the byprod- ucts ethyl 1-phenyl-1H-1,2,3-triazole-4-carboxylates (15). Com- pounds 14 and 15 could be separated by silica gel column chromatography, and the structures of 14 were further determined by X-ray crystallographic analysis of a representa- tive compound 16a (Scheme 2).32 With the key intermediates 14 in hands, compounds 3 were readily obtained by hydrolysis and coupling with substituted piperazines or 1, 4-diazepane with good yields. ■ RESULTS AND DISCUSSION Taking compound 1 as the lead compound, we initially designed isoxazol-4-carboxamide (2a), 1H-1,2,3-triazole-4- Figure 1. Design of new anti-influenza A agents using a scaffold-hopping strategy. Table 1. Inhibitory Activities of the Compounds against Influenza A Replication IC50 (μM) compd n R1 R2 R3 R4 R5 R6 R7 CC50 c (μM) H3N2a H1N1b 1 15.9 1.08 0.32 2a 6.00 4.83 2b 2.62 0.71 4a 8.64 3.38 4b 2.87 0.72 5a 7.61 5.24 5b 3.68 1.57 3a 1 Cl NO2 Me H H H H >100 5.16 3.13 3b 1 Cl NO2 Me OMe H H H >100 1.97 0.68 3c 1 Cl NO2 Me H OMe H H >100 13.77 5.35 3d 1 Cl NO2 Me H H OMe H >100 14.72 5.61 3e 1 Cl NO2 Me Cl H H H >100 5.65 1.20 3f 1 Cl NO2 Me H Cl H H >100 9.38 4.89 3g 1 Cl NO2 Me H H Cl H >100 >50 4.93 3h 1 Cl NO2 Me OH H H H >100 7.84 1.58 3i 1 Cl NO2 Me Me H H H >100 4.92 2.40 3j 1 Cl NO2 Me OEt H H H >100 8.97 3.22 3k 1 Cl NO2 Me OiPr H H H >100 16.16 8.42 3l 1 Cl NO2 Me COMe H H H >100 14.80 13.72 3m 1 Cl NO2 Me OMe H H OMe >100 2.42 1.34 3n 1 Cl NO2 H OMe H H H >100 >50 8.36 3o 1 Cl NO2 Et OMe H H H >100 >50 >50 3p 1 Cl H Me OMe H H H >100 >50 48.12 3q 1 H NO2 Me OMe H H H >100 >50 16.98 3r 1 H H Me OMe H H H >100 >50 >50 3s 1 OMe NO2 Me OMe H H H >100 >50 >50 3t 2 Cl NO2 Me OMe H H H >100 >50 35.45 AMDd >100 2.22 >50 aInfluenza A/HK/8/68 (H3N2) stains. bInfluenza A/WSN/33 (H1N1) stains. cAntiproliferation against MDCK cells. dAMD: amantidine. The data are means of results from at least three independent experiments. Journal of Medicinal Chemistry Article dx.doi.org/10.1021/jm2013503 | J. Med. Chem. 2012, 55, 2144−21532146 carboxamide (3a), 1H-pyrazol-4-carboxamide (4a), and 2- biphenyl carboxamide (5a) as new anti-influenza A agents by using a scaffold-hopping strategy (Figure 1). The anti-influenza A virus activities of the compounds were preliminarily evaluated using Madin−Darby canine kidney (MDCK) cell-based CPE (cytopathic effect protection) assays. Under the screening conditions, 1 inhibited the replication of H3N2 (A/HK/8/68) and H1N1 (A/WSN/33) strains, with IC50 values of 1.01 and 0.32 μM, respectively, which were comparable to the reported data.24,25 The results also suggested that all the four designed compounds potently inhibited the replication of H3N2 (A/ HK/8/68) and H1N1 (A/WSN/33) strains, among which 1H- 1,2,3-triazole- 4-carboxamide (3a) displayed the greatest potency and inhibited the replication of H3N2(A/HK/8/68) and H1N1 (A/WSN/33) strains, with IC50 values of 5.16 and 3.13 μM, respectively (Table 1). Although compound 3a is about 5-fold less potent than the original lead compound 1, it represented a new anti-influenza A agent with a different chemical scaffold. Therefore, further structural optimization was conducted to improve its anti-influenza activity. The results were summarized in Table 1. Similar to the previous observation on the derivatives of compound 1,25,27 a substituted group in phenyl ring A might have great impact on the anti-influenza activity of the 1H-1,2,3- triazole-4-carboxamide compounds. For instance, when a methoxyl group was introduced at R4 position of compound 3a, the anti-influenza activity was obviously improved and the resulting compound 3b displayed IC50 values of 1.96 and 0.68 μM against the replication of H3N2 (A/HK/8/68) and H1N1 (A/WSN/33) strains, respectively, which was about 3 times more potent than the original compound 3a. However, when a methoxyl group was introduced in R5 or R6 position (3b and 3c), the potency was decreased. The R4-Cl− substituted compound 3e was also more potent than the corresponding R5- or R6-substituted compound 3f or 3g, which suggested that R4 might be a feasible position for further optimization. Therefore, a variety of other moieties such as hydroxyl (3h), methyl (3i), ethoxyl (3j), or acetyl (3k) groups were introduced for further investigation. However, it was disappointing that none of these substitutions further improved the potency. The 2′,6′- dimethoxyl compound 3m (R4, R7-disubstituted compound) also displayed good potency against the replication of H3N2 (A/HK/8/68) and H1N1 (A/WSN/33) strains. The impact of R3 was also investigated, and the results indicated that removal of the methyl group (3n) or replacing it with a slightly bigger ethyl group (3o) caused dramatic decrease of the potency. Investigation on the substituted group in phenyl ring B revealed that both the nitro- (R2) and chloro- (R1) were necessary for the anti-influenza activity. The potencies were significantly lost by removal of the nitro or/and chloro groups (3p−3s). It was also demonstrated that replacement of the piperazine linker with 1,4-diazenpane (3t) led to about 50-time decrease of the potency (comparing with 3b). Utilizing the knowledge of the structure−activity relationship on compounds 3, methoxyl or chloro substituted derivatives of compounds 2a, 4a, and 5a were also design and synthesized. The resulting compounds 2b, 4b, and 5b displayed improved anti-influenza activities against H3N2 (A/HK/8/68) and H1N1 (A/WSN/33) strains, with IC50 values comparable to that of 3b. Antiproliferative activities of the compounds against MDCK cells were also evaluated to monitor the potential cytotoxic effects. Under the assay conditions, 1 displayed moderate toxicity against the MDCK cells with IC50 value of 16 μM. However, none of the 1H-1,2,3-triazole-4-carboxamide com- pounds showed obvious cellular growth inhibition against MDCK cells under 100 μM (Table 1), indicating that the compounds beard great selectivity indexes (SI = CC50/IC50). Although 3b was about 2 times less potent than compound 1, it displayed equal potency with amantidine against the replication of influenza A/HK/8/68 (H3N2) stains with great selectivity index. Furthermore, 3b also potently inhibited the replication of A/WSN/33 (H1N1) stains which were resistant to the clinical M2 blocker amantidine (Table 1), suggesting that this compound might be used a new lead compound for novel anti-influenza drug discovery. Therefore, further biological evaluations were performed to validate its anti-influenza activity. Plague counting based viral yield reduction assay is one of the most direct and reliable methods for evaluating the inhibitory effects on viral replication. Therefore, the anti- influenza function of 3b was further validated by using a plaque reduction assay. As shown in Figure 2, compound 3b dose- dependently inhibited the plaque formation of MDCK cells by A/WSN/33 (H1N1) influenza virus infection. The plaque formation was completely inhibited by 3b under a concen- tration of 1.0 μM. Highly consistent with the results from our CPE assay, compound 1 displayed better plague suppressive effect than 3b and induced a complete abrogation of viral plagues at 0.3 μM. The viral yield reduction activities of 1 and 3b were further validated by using microplate counting of infectious particles (McIP) assay,25 and the deduced IC50 values were 0.10 and 0.23 μM, respectively. The anti-influenza activities of 3b were also evaluated against a panel of other influenza A virus strains, and the results were summarized in Table 2. It was shown that 3b also potently inhibited several other H1N1 and H3N2 influenza A virus strains with varied IC50 values in low μM range, except for the H3N2 (Br/10/09) strains (IC50 > 50 μM). It was also noteworthy that 3b potently inhibited the replication of H5N1 (RG14) strains, tamiflu resistant WSN (274Y) A/WSN/1933 (H1N1) strains, and A/Guangdong/1996 Chicken H9N2 influenza virus strains with IC50 values of 0.69, 1.11, and 4.62 μM, respectively. Compound 1 had been reported as a nucleoprotein inhibitor that triggered the aggregation and inhibited the nuclear accumulation of influenza virus nucleoproteins.24,25 Given the structural similarity of the 1H-1,2,3-triazole-4-carboxamide Figure 2. Compound 3b potently inhibited the plaque formation of influenza virus infected MDCK cells. Monolayer MDCK cells were infected with A/WSN/33 (H1N1) at 0.01 MOI (multiplicity of infection) and overlaid with 1% agar containing various concentrations of nucleozin (1) (upper panel) and 3b (bottom panel). At 72 h postinfection, plaques were visualized by stained with crystal violet. The results shown are representative of at least three independent experiments. Journal of Medicinal Chemistry Article dx.doi.org/10.1021/jm2013503 | J. Med. Chem. 2012, 55, 2144−21532147 derivatives to that of 1, we hypothesized that the new compounds might also target influenza A nucleoproteins. Therefore, the effect of 3b on nucleoproteins was inspected under fluorescence microscopy (Figure 3). Not surprisingly, 3b potently inhibited the nuclear accumulation of influenza virus A nucleoprotein and caused the nucleoproteins to be trapped in the cytoplasm and scattered randomly in H1N1 (A/WSN/33) virus infected MDCK cells, indicating nucleoproteins might be the molecular target of the 1H-1,2,3-triazole-4-carboxamide analogues. Most recently, the X-ray crystal structure of nucleoprotein complex with a compound 1 analogue was disclosed. It was demonstrated that each inhibitor bridged two molecules of NP (i.e., NP_A and NP_B) protein to form a hexameric complex.27 The interactions of the inhibitor with the Y289 region in NP_A and the Y52 region in NP_B were critical for binding of the inhibitor to interface of NP_A and NP_B (Figure 4A,B). The importance of the Y289 and Y52 regions in nucleoprotein had also been observed by previous mutational studies.24,25 To understand the potential interaction of 3b with influenza nucleoproteins, a computational study was performed. The results suggested that 3b bound to nucleoproteins with a similar mode to that of the nucleozin analogues (Figure 4). The carbonyl group of 3b formed an essential hydrogen bond with the OH group of the S376 residue in NP_B. The 4-nitro-2- chloro-phenyl moiety might form strong π−π stacking interaction with the Y289 residue of NP_A, while the 2- methoxyl phenyl group and the piperazine moiety could achieve important hydrophobic effect with the Y52 side chain of NP_B. To further validate the proposed interaction, the anti- influenza activity of 3b were determined against A/New Jersey/ 8/76 strains bearing Y289H mutation and recombinant WSN viruses harboring either wild-type NP (52Y) or mutated NP (52H) using 1 as a reference compound. Similar to compound 1, 3b is a potent anti-influenza agent to the WSN virus with wild-type NP (52Y). However, both Y52H mutated isogenic WSN virus and Y289H mutated A/New Jersey/8/76 strains were resistant to 3b (IC50 > 50 μM), highlighting the critical interaction of 3b with these two residues. These results were highly consistent to our computational modeling results and further supported that virus nucleoprotein might be the molecular target of the 1H-1,2,3-triazole-4-carboxamide anti- fluenza agents. ■ CONCLUSION In summary, using scaffold-hopping and bioisosteric replace- ment strategies, we successfully designed and synthesized a series of 1H-1, 2, 3-triazole-4-carboxamide derivatives as new anti-influenza A agents based on the chemical structure of compound 1. One of the most potent compounds 3b potently inhibited the replication of various H3N2 and H1N1 influenza A virus strain with IC50 values ranging from 0.5 to 4.6 μM. Furthermore, 3b also potently inhibited the replication of H5N1 (RG14), amantidin-resistant A/WSN/33 (H1N1), and tamiflu-resistant WSN (274Y) A/WSN/1933 (H1N1) with IC50 values of 0.69, 0.68, and 1.11 μM, respectively. Further computational study and mechanism investigation suggested that 3b might directly target influenza virus A nucleoprotein to inhibit its nuclear accumulation. Our study provided a new series of lead c