Identification and biochemical characterization of

European Journal of Medicinal Chemistry 124 (2016) 981e991

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry journal homepage: http://www.elsevier.com/locate/ejmech

Research paper

Identification and biochemical characterization of DC07090 as a novel potent small molecule inhibitor against human enterovirus 71 3C protease by structure-based virtual screening Guang-Hui Ma a, b, c, 1, Yan Ye d, e, b, 1, Dan Zhang a, f, 1, Xin Xu a, Pei Si a, g, Jian-Long Peng d, Yong-Long Xiao a, Rui-Yuan Cao h, Yu-Ling Yin a, Jing Chen a, Lin-Xiang Zhao f, Yu Zhou a, ***, Wu Zhong h, Hong Liu a, Xiao-Min Luo d, e, **, Li-Li Chen a, *, Xu Shen a a

Key Laboratory of Receptor Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road, Shanghai 201203, China University of Chinese Academy of Sciences, No.19A Yuquan Rd, Beijing 100049, China c School of Life Science and Technology, ShanghaiTech University, 100 Haike Rd, Pudong, Shanghai 201210, China d Drug Discovery and Design Center, State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road, Shanghai 201203, China e State Key Laboratory of Natural and Biomimetic Drugs, Peking University, 38 Xueyuan Rd, Beijing 100191, China f Key Laboratory of Structure-Based Drugs Design & Discovery of Ministry of Education, Shenyang Pharmaceutical University, 103 Culture Rd, Shenyang 110016, China g College of Life and Environmental Sciences, Shanghai Normal University, 100 Guilin Road, Shanghai 200234, China h Beijing Institute of Pharmacology and Toxicology, 27 Taiping Rd., Beijing 100850, China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 July 2016 Received in revised form 21 September 2016 Accepted 10 October 2016 Available online 11 October 2016

Hand, foot and mouth disease (HFMD) is a serious, highly contagious disease. HFMD caused by Enterovirus 71 (EV71), results in severe complications and even death. The pivotal role of EV71 3Cpro in the viral life cycle makes it an attractive target for drug discovery and development to treat HFMD. In this study, we identified novel EV71 3Cpro inhibitors by docking-based virtual screening. Totally 50 compounds were selected to test their inhibitory activity against EV71 3Cpro. The best inhibitor DC07090 exhibited the inhibition potency with an IC50 value of 21.72 ± 0.95 mM without apparent toxicity (CC50 > 200 mM). To explore structure-activity relationship of DC07090, 15 new derivatives were designed, synthesized and evaluated in vitro enzyme assay accordingly. Interestingly, four compounds showed inhibitory activities against EV71 3Cpro and only DC07090 inhibited EV71 replication with an EC50 value of 22.09 ± 1.07 mM. Enzyme inhibition kinetic experiments showed that the compound was a reversible and competitive inhibitor. The Ki value was determined to be 23.29 ± 12.08 mM. Further molecular docking, MD simulation and mutagenesis studies confirmed the binding mode of DC07090 and EV71 3Cpro. Besides, DC07090 could also inhibit coxsackievirus A16 (CVA16) replication with an EC50 value of 27.76 ± 0.88 mM. Therefore, DC07090 represents a new non-peptidyl small molecule inhibitor for further development of antiviral therapy against EV71 or other picornaviruses. © 2016 Elsevier Masson SAS. All rights reserved.

Keywords: HFMD EV71 3Cpro Non-peptidyl inhibitor Virtual screening

1. Introduction Hand, foot and mouth disease (HFMD) is a serious, highly

* Corresponding author. ** Corresponding author. *** Corresponding author. E-mail addresses: [email protected] (Y. (X.-M. Luo), [email protected] (L.-L. Chen). 1 These authors contributed equally to this work.

Zhou),

http://dx.doi.org/10.1016/j.ejmech.2016.10.019 0223-5234/© 2016 Elsevier Masson SAS. All rights reserved.

[email protected]

contagious disease caused by polioviruses, coxsackievirus (CV), echoviruses and Enterovirus (EV) [1e3]. It is characterized by fever, vomiting, vesicular rashes on the hands, feet, buttocks, and oropharyngeal ulcers [4]. HFMD occurs mainly in children under 10 years old, and can also be found in adults occasionally. HFMD are contagious mainly through direct contact with saliva, mucus, fluid from blisters and stool of infected people [5]. Outbreaks of HFMD usually occur in many Asian countries, including China, Japan, Hong Kong (China), Republic of Korea, Malaysia, Singapore, Thailand, Taiwan (China) and Viet Nam [6,7]. Especially, in 2011, Viet Nam

982

G.-H. Ma et al. / European Journal of Medicinal Chemistry 124 (2016) 981e991

experienced a large outbreak of HFMD, causing 113,121 cases and 169 deaths from 63/63 provinces [8]. In China, the annual HFMD incidence has increased from 37.6/100,000 in 2008 to 139.6/ 100,000 in 2014 [9e11]. Therefore, HFMD is a serious threat to human health. As we know, HFMD is most commonly caused by CVA16, leading to a mild self-limiting disease with few complications [12]. However, HFMD caused by Enterovirus 71 (EV71) results in severe complications, including neurological, cardiovascular and respiratory problems, and even death. EV71 was firstly isolated in California, from patients with central nervous system diseases [13,14]. In recent years, HFMD caused by EV71 has continued increasing throughout the world [15e17]. Enterovirus 71 (EV71) is a single-stranded, positive-sense RNA virus, belonging to the Picornaviridae family. The genome of EV71 is about 7.4 kb, and its open reading frame (ORF) includes three consecutive parts, P1, P2, and P3 [18]. P1 region encodes a polyprotein which can be cleaved into four structural proteins VP1eVP4. P2 and P3 regions respectively encode the non-structural proteins 2A-2C and 3A-3D. The key enzyme in the processing of polyproteins is EV71 3C protease (EV71 3Cpro), which is a cysteine protease with the typical chymotrypsin-like fold and is involved in almost all cleavage processes of EV71 polyprotein, with the exception of the cleavages of VP1/2A and 3C/3D by 2A protease [19]. Due to its functional importance in the viral life cycle, EV71 3Cpro is considered to be an attractive target for drug discovery and development to treat HFMD. So far, a variety of inhibitors against the EV71 3Cpro have been reported, including substrate-based peptidomimetics [2,20e22], flavonoids and derivatives [23e25], etc. Although peptidomimetics exhibit submicromolar potency with an EC50 value of 18e20 nM against EV71 [18,21], their further bioavailability and efficacy in the treatment of HFMD have yet to be confirmed. While AG7088 (rupintrivir) as the substrate based peptidomimetic was found to have potent inhibitory activity against EV71 3Cpro (IC50 ¼ 2.3 mM), it failed during Phase II in clinical trials as human rhinovirus 3Cpro inhibitor due to unfavorable bioavailability and limited efficacy [1,26,27]. A natural flavonoid, chrysin (CR) and its derivative, diisopropyl chrysin-7-yl phosphate (CPI), exhibited anti-viral activities with EC50 of 15.89 and 9.06 mM [24], respectively. This kind of flavonoids trends to have apparent side effects in clinical practice [28]. Additionally, an adenosine analog (NITD008) was reported to have potent antiviral activity against EV71 with an EC50 value of 0.67 mM. However, it shows a relatively high cytotoxicity (CC50 ¼ 119.97 mM) [29]. To date, there are no specific antiviral drugs available for the treatment and prevention of EV71 causing HFMD. Therefore, it is significant to develop non-peptidyl small molecule inhibitors against EV71. In this study, we identified novel EV71 3Cpro inhibitors from the commercial database by docking-based virtual screening. There were 50 compounds selected to test their inhibitory activity against EV71 3Cpro. The best inhibitor DC07090 exhibited the inhibition potency with an IC50 value of 21.72 ± 0.95 mM without apparent toxicity (CC50 > 200 mM). To explore structure-activity relationship of DC07090, 15 new derivatives were designed and synthesized accordingly. Four compounds showed inhibitory activities against EV71 3Cpro and only DC07090 inhibited EV71 replication with an EC50 value of 22.09 ± 1.07 mM. Further molecular docking, MD simulation and mutagenesis studies confirmed the binding mode of DC07090 and EV71 3Cpro. DC07090 forms stable hydrogen-bonding interaction with the main chains of S128, G145, G164 and hydrophobic interactions with F25, L125, L127 and F170. Besides, DC07090 could inhibit coxsackievirus A16 (CVA16) replication with an EC50 value of 27.76 ± 0.88 mM. Therefore, DC07090 represents a new non-peptidyl small molecule inhibitor for further development of antiviral therapy against EV71 or other picornaviruses.

2. Results and discussion 2.1. Identification of DC07090 by virtual screening and enzyme inhibition assay To identify novel small molecule inhibitors against EV71 3Cpro, the crystal structure (PDB entry: 3SJO [30]) of EV71 3Cpro was used for conducting the structure-based virtual screening with the Glide docking algorithms. The workflow of the overall virtual screening procedure is shown in Fig. 1A. In the primary screening, the database SPECS containing about 200,000 molecules was searched for potential binders using the HTVS mode of the program GLIDE. Subsequently, the top ranked 5000 molecules were chosen for the second round docking by the XP (extra-precision) mode of Glide. As a result, the top 500 compounds were reserved for visual inspection. Finally, we selected 50 molecules for biological assays according to their docking scores and binding modes. The 50 candidates were tested for their inhibitory activities against EV71 3Cpro. As shown in Fig. 1B, IC50 value of the positive compound AG7088 is 2.62 ± 0.92 mM. DC07090 showed micromolar potency against EV71 3Cpro with an IC50 value of 21.72 ± 0.95 mM, as described in Fig. 1C. 2.2. 1 DC07090 acted as a competitive inhibitor of EV71 3Cpro To confirm the binding affinity of DC07090 and EV71 3Cpro, SPRbased assay and fluorescence-quenching method were performed. As shown in Fig. 2A, DC07090 presents obvious concentrationdependently binding curves. The dissociation equilibrium constant (Kd) (16.5 ± 2.13 mM) was determined by fitting the sensorgrams with a 1:1 binding fitting model. In addition, given the tryptophan residue Trp48 in EV71 3Cpro, an intrinsic fluorescencequenching based assay was performed to further determine the interaction of DC07090 and EV71 3Cpro. As demonstrated in Fig. 2B, EV71 3Cpro displayed the maximal fluorescence at 340 nm, and the addition of DC07090 resulted in fluorescence quenching of EV71 3Cpro in a dose dependent manner, with a Kd value of 13.41 ± 4.3 mM. After confirming the direct binding of DC07090 to EV71 3Cpro, we further explore the inhibitory mechanism of compound DC07090 against EV71 3Cpro. Enzyme kinetics experiments were performed at different inhibitor concentrations (0e20 mM) and different substrate concentrations (40e150 mM). As shown in Fig. 2C, the LineweavereBurk plot of 1/v versus 1/[S] of the lines with different inhibitor concentrations resulted in the same y-axis intercept reflecting DC07090 as a competitive inhibitor toward EV71 3Cpro. Ki value was determined to be 23.29 ± 12.08 mM calculated using the Dixon plots (Fig. 2D). These results suggest that DC07090 potently impairs the catalytic activity of EV71 3Cpro by binding in the enzyme's catalytic sites, and acts as a competitive inhibitor of EV71 3Cpro with a Ki value of 23.29 ± 12.08 mM. Therefore, it was shown that DC07090 directly bound to EV71 3Cpro and acted as a competitive inhibitor of EV71 3Cpro with a Ki value of 23.29 ± 12.08 mM. Although the positive compound AG7088 as a covalent inhibitor has potent activity and lasting effect, the irreversible inhibitor can also react with non-target proteins, and the lasting effect is detrimental [30]. However, relative to AG7088, DC07090 as a competitive reversible inhibitor might have fewer side effects. 2.3. Structural modification and structure-activity relationships (SAR) analysis Based on the above results, we attempted to improve the activity of DC07090 against EV71 3Cpro by optimizing its structure. The designed compounds (1e15) were synthesized according to the procedures shown in Scheme 1. Firstly, treatment of commercially


983

Fig. 1. Virtual screening workflow and in vitro enzyme inhibition assay. (A) The workflow of the overall virtual screening procedure. (B) In vitro enzyme inhibition assay, compound AG7088 was a positive control with an IC50 value of 2.62 ± 0.92 mM. (C) The structure and the dose-dependent effect of compound DC07090 against EV71 3Cpro. IC50 value was determined to be 21.72 ± 0.95 mM.

Fig. 2. DC07090 as a competitive inhibitor directly bound to EV71 3C protease. (A) The binding affinity of DC07090 with EV71 3Cpro was determined by SPR-based assay with a Kd value of 16.5 ± 2.13 mM. (B) The binding affinity of DC07090 with EV71 3Cpro was performed by an intrinsic fluorescence quenching assay with a Kd value of 13.41 ± 4.3 mM (C) DC07090 as the competitive inhibitor was indicated in Double-reciprocal plot. (D) The fitted Ki value of DC07090 (23.29 ± 12.08 mM) was calculated using Dixon plots.

984


R2 NH2

O R3

Cl +

N

O

R2 a

O

n

1a

O

O HO

H N

N

R3

N

b

R3

n 1c

1b R2

H N

OH R1

n

N 1e c

1d

NH2

R3

N

N

H N

R1

N

R2

n

O

1-15

Scheme 1. The synthesis of target compounds. Reagents and conditions: (a) Et3N, toluene, reflux, over night; (b) 2 mol/L NaOH, MeOH, 60 C, 4 h; (c) PPA, 180 C, 0.5 h.

available 2-(chloromethyl) pyridine (1a) and methyl 4aminobenzoate (1b) generated the key amine intermediates 1c in the presence of Et3N with excellent yields. Carboxylic acid 1d was obtained by hydrolysis reaction, and further cyclized with 2-aminoTable 1 Structures and activities of the EV71 3C protease inhibitors.

H N

N

R3

R2

N

R1

N n

O

Compound

n

R1

R2

R3

Inhibition Rate (%)a

IC50 (mM)b

DC07090 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 AG7088

0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 _

H CH3 H CH3 H CH3 CH3 H H CH3 H CH3 H CH3 CH3 H _

H H H H F F F F Cl Cl Cl Cl H H H H _

H H CH3 CH3 H H CH3 CH3 H H CH3 CH3 H H CH3 CH3 _

80.18 19.25 68.07 200 mM) (Fig. 3C). Therefore, this study provides valuable insight into the novel scaffold of potential anti-EV71 virus agents. 2.4. Binding mode of DC07090 with EV71 3Cpro In order to study the binding mode of DC07090 bound to EV71

985

3Cpro, we analyzed the binding pose of DC07090 docked into EV71 3Cpro by molecular docking. As shown in Fig. 4, the top-ranked pose was chosen to be superimposed with the crystal conformation of the reference molecule AG7088. AG7088 occupies five pockets S0 , S1, S2, S3 and S4 of EV71 3Cpro, forming a network of extensive hydrogen bonds and hydrophobic interactions [30]. For the compound DC07090, the oxazole [4,5-b]pyridine moiety is situated in the pocket S10 and forms hydrophobic interaction with F25. The pyridine nitrogen (N2) of oxazole [4,5-b]pyridine moiety is involved in the hydrogen bonding network with the main chain of G145, while the amino nitrogen in the middle (N4) of DC07090 forms hydrogen-bonding interaction with the main chain of G164. In the terminal of DC07090, the pyridine ring occupies the pocket S4 and has hydrophobic interactions with the residues Leu125, Leu127, and Phe170. The pyridine nitrogen (N3) of DC07090 has hydrogen bonding interaction with the main chain of Ser128. The three hydrogen-bonding interactions and hydrophobic interactions, which also exist in the conformation of AG7088 with EV71 3Cpro make the compound DC07090 well fitted into the catalytic pocket and compete against endogenous substrates (e.g. precursor polyproteins) binding. 2.5. Molecular dynamics simulation for EV71 3Cpro in complex with DC07090 To further verify the stability of the binding mode between compound DC07090 and EV71 3Cpro, a 100-ns molecular dynamics (MD) simulation was conducted. The stability under simulation was evaluated by the root-mean-square deviation (RMSD). Fig. 5 showed the time dependences of RMSD of the backbone of EV71 3Cpro and the ligand DC07090 over the simulation. From 2ns to 100 ns, the backbone of the protease was relatively stable with the RMSD value fluctuating around 1.28 Å, and DC07090 maintained a stable binding pose during the simulation. Next, we calculated the occupancy of hydrogen bonds formed by DC07090 and EV71 3Cpro in the simulation. As shown in Table 2, the hydrogen bond between the backbone NH of G164 and the N4 atom of DC07090 is the most stable with a percentage of 59.84%. In addition, the occupancy of the hydrogen bond between backbone NH of G145 and the N2 atom

Fig. 4. The potential binding mode of DC07090 with EV71 3Cpro (PDB code 3SJO): The green stick is AG7088, while the orange stick is DC07090. In the solid surface, the pocket S1 (magenta), S1’ (orange), S2 (blue), S3 (cyan) and S4 (wheat) are labeled. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

986


Fig. 5. The root mean square deviations of the backbone of EV71 3Cpro (red) and the ligand DC07090 (blue) over the 100ns MD simulation at 300 K. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 2 The occupancy of three hydrogen-binding interactions. Donor

Acceptor

Occupancy

GLY164-Main-N GLY145-Main-N SER128-Main-N

DC07090 -N4 DC07090 -N2 DC07090 -N3

59.84% 56.88% 1.94%

of DC07090 is 56.88%, which implies that the hydrogen bond is also relatively stable. However, the hydrogen bond between the backbone NH of S128 and the terminal pyridine N3 atom of DC07090 is weak with the occupancy of 1.94%. Therefore, our MD simulation result is consistent with the above binding mode revealed by the molecular docking. The binding between DC07090 and EV71 3Cpro is relatively stable except for the interactions between the terminal pyridine ring and the protease. 2.6. Determination of DC07090 binding sites Based on the above binding mode speculated by molecular docking and MD simulation, DC07090 forms stable hydrogenbonding interactions with the main chain of S128, G145, G164 and hydrophobic interactions with F25, L125, L127 and F170. To validate the interactions, the site-directed mutagenesis studies were performed. Given the infeasibility to disturb hydrogen bonding interactions in the main chain of S128, G145 and G164, only the hydrophobic residues in the existing interaction can be mutated to influence the hydrophobic interaction between EV71 3Cpro and DC07090. We chose two residues (F25 and L127) respectively, in pocket S10 and S4 to construct mutants. The mutants F25W and L127F can increase the hydrophobic interactions with DC07090, while the mutant F25A is considered to decrease the hydrophobic interaction. In addition, there are two nitrogens on the oxazole [4,5-b] pyridine moiety of DC07090, which could act as the hydrogen acceptors to form new hydrogen bonding interactions with EV71 3Cpro. The mutants A144S and G145S were obtained to add hydrogen bonding interactions. Moreover, the mutant G163L was designed to form new hydrophobic interaction with the benzene ring of DC07090. Apart from the residues (F25, L127, A144, G145 and G163) mentioned above, we also designed the mutant S128A to validate the mutagenesis studies according to the previous paper [20]. First, a series of EV71 3Cpro mutants were generated and expressed according to the method similar to EV71 3Cpro wild type. Next, the

structure changes of these proteases were detected by CD spectrum. As shown in Fig. 6A, the CD spectra showed the negative band at 218 nm and the positive one at 196 nm, suggesting that EV71 3Cpro wild type and mutants are largely b-sheet in solution. Then, we determined the enzymatic activities of EV71 3Cpro mutants. Strikingly, except for the mutant A144S having the same potency of activity as the wild type, the mutant F25A, F25W and S128A had only a fraction of the enzymatic activity, while other mutant proteases are nearly inactive (Fig. 6B). The enzymatic activity of the mutant S128A is consistent with the result published in the paper [20]. In order to confirm the binding mode, we further determined the binding and inhibitory activity of DC07090 with EV71 3Cpro mutant A144S, F25A and F25W. As shown in Fig. 6C, the addition of DC07090 led to remarkable fluorescence quenching of EV71 3Cpro mutant A144S, with a Kd value of 6.37 ± 1.94 mM, suggesting that the binding affinity between them was increased. At the same time, it was also observed that the inhibitory activity of DC07090 against EV71 3Cpro mutant A144S was enhanced, with an IC50 value of 9.73 ± 0.93 mM (Fig. 6D). In addition, we also detected the interactions between DC07090 and the mutant F25A/F25W. As shown in Fig. S2A, the mutant F25A had only 12.5% enzymatic activity compared to that of EV71 3Cpro WT. The enzymatic activity was weaker when different concentrations of DC07090 were added in the reaction system. We then determined their interaction by fluorescence quenching assay. The result showed that the binding affinity between F25A and DC07090 was reduced, with Kd of 25.0 ± 6.07 mM (Fig. S2B). We next detected the inhibition of DC07090 on mutant protein F25W. It was shown that DC07090 inhibited EV71 3Cpro (F25W) with IC50 of 2.28 ± 0.92 mM (Figs. S2C and D). The fluorescence quenching experiment showed that F25W protein had stronger binding with DC07090 with Kd of 12.10 ± 4.56 mM (Fig. S2E). Based on the above binding mode, the binding between DC07090 and the mutant F25A was reduced, because the mutant protein decreased the hydrophobic interaction which was unfavorable to the binding of DC07090. On the contrary, the mutant F25W increased the hydrophobic interaction contributing to the binding of DC07090, so the binding between them was also increased. Therefore, these results were in accordance with the binding mode between DC07090 and EV71 3Cpro. Based on the above results, we analyzed structural modification, SAR and the binding mode of DC07090. In order to improve the inhibitory activity of DC07090 against EV71 3Cpro, 15 new derivatives were designed, synthesized, and evaluated for their biological activities. Among of them, four compounds showed inhibitory activities against EV71 3Cpro and only DC07090 inhibited EV71 replication. The preliminary SARs were obtained based on the results in Table 1. The introduction of methyl group at the R1, R3 position resulted in a decrease of inhibitory activity against EV71 3Cpro. Especially, the presence of methyl group at the R1 position led to the compounds complete loss of activity. The substituent of F and Cl at the R2 position largely decreased the inhibitory activity. To further explain SARs, we analyzed the binding mode of DC07090 in the catalytic domain of EV71 3Cpro (Fig. 4) and molecular dynamics simulations result (Fig. 5). The RMSD results revealed that compound DC07090 stably bound into the catalytic domain of EV71 3Cpro, including three hydrogen-bonding interactions with the main chains of S128, G145, G164 and hydrophobic interactions with F25, L125, L127, F170. The loss of inhibitory activity attributes to the steric hindrance upon introducing methyl group at the R1 position and F, Cl group at the R2 position. The R3 position exposes to the solvent and the substituent of hydrophobic methyl group results in a decrease of activity. Therefore, molecular binding model provides the rational explanations about SARs, which is in good agreement with biological results. Although the 15 structural analogues had no great improvement to DC07090, maybe owing to the limited space


987

Fig. 6. The binding mode of DC07090 with EV71 3Cpro was confirmed by site-directed mutagenesis, CD, enzyme inhibition and binding assay. (A) CD result showed that EV71 3Cpro mutants are largely b-sheet and have no obvious changes in structure. (B) In vitro enzyme inhibition assay, the activity of A144S was similar to EV71 3Cpro WT. The mutant F25A, F25W and S128A only kept fraction of the enzymatic activity, while L127F, G145S and G163L were almost inactive. (C) Fluorescence quenching revealed that the binding affinity between A144S and DC07090 was increased with a Kd value of 6.37 ± 1.94 mM. (D) The inhibitory activity of DC07090 against EV71 3Cpro WT or the mutant A144S was determined. The IC50 values are 21.72 ± 0.95 mM for EV71 3Cpro WT (blue) and 9.73 ± 0.93 mM for A144S (red), respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

of protein hydrophobic pocket when methyl was introduced, we still got some useful indications for further structural modifications based on structural optimization and molecular dynamics simulation. From the superimposed binding modes of DC07090 and AG7088 (Fig. 4), we speculate that the unoccupied cavities S1 and S2 may be the modification direction to improve the activity of DC07090. At the same time, we found that DC07090 could inhibit the replication both EV71 and CVA16, which indicated that the compound could have broader activity against other picornaviruses. Now, molecular design and structural modification are underway in our laboratory. In silico docking predicted that the binding site of DC07090 is near the substrate binding site (Fig. 4). The prediction is consistent with the observed competitive inhibition type of DC07090. Besides, we conducted the site-directed mutagenesis studies to validate the interaction between DC07090 and EV71 3Cpro. In the mutagenesis studies, we observed an interesting phenomenon. After some key amino acid residues of EV71 3Cpro involved in binding of DC07090 were mutated, mutant F25A, F25W and S128A only keep fraction of the enzymatic activity, while other mutant L127F, G163L are nearly inactive. The residue L127 on the tip of flexible b-ribbon of EV71 3Cpro is very important for substrate recognition [31] and restraints the mobility of the b-ribbon, which is harmful to the substrate binding. The sequence G145 is highly conserved in viral 3C proteases, which must remain a certain flexibility to orient the backbone NH groups of Gly145 to form an “oxyanion hole” for stabilization of a tetrahedral transition state [32]. Therefore, the mutant G145S might destroy the formation of the oxyanion hole by the limited motion of the conserved

sequence. The residue G163 is near to the active sites of the oxyanion hole and the mutant G163L might affect the structure of the active sites. These results argue that local perturbations surrounding the active sites strongly destroy proteolytic activity. The residue F25 has hydrophobic interaction with the substrate and the mutant F25A reduces the activity of protease by a decrease of hydrophobic interaction. For the mutant F25W, steric hindrance will prevent the substrate binding, which results in a decrease of the activity of protease. The residue S128 is reported to be important for substrate recognition [31], which is in good agreement with the result that the enzymatic activity suffered from significant loss due to the missing serine side-chain oxygen for the mutant S128A. The mutant A144S does not affect the enzymatic activity significantly, which is accorded with the previous report [33]. Finally, we tested the enzymatic inhibition assay for DC07090 against the mutant A144S, F25A and F25W. DC07090 inhibits the mutant A144S at a potency that is 2.2-fold stronger than that of the wild type, which supports the point that the hydroxyl group on the side chain of Ser144 may form hydrogen bonding interaction with hydrogenacceptors of DC07090. The binding between DC07090 and the mutant F25A was reduced, because the mutant protein decreased the hydrophobic interaction which was unfavorable to the binding of DC07090. On the contrary, the mutant F25W increased the hydrophobic interaction contributing to the binding of DC07090, so the binding between them was also increased. Therefore, the results validate the above binding model of DC07090 with EV71 3Cpro, which provides valuable clues for further optimization of the lead and the discovery of the antiviral agents.

988


3. Conclusion

4.2. Biological assay

In summary, with the combination of virtual screening and in vitro assays, we identified DC07090 as an effective inhibitor of EV71 3Cpro. DC07090 competitively inhibited the activity of the protease with a Ki of 23.29 ± 12.08 mM and EV71 virus infection with an EC50 of 22.09 ± 1.07 mM. To improve the inhibitory activity against EV71 3Cpro and EV71 virus, we preliminarily studied SAR of DC07090. Further molecular docking, MD simulation and mutagenesis studies confirmed the binding mode of DC07090 and EV71 3Cpro. Besides, DC07090 could also inhibit CVA16 replication with an EC50 value of 27.76 ± 0.88 mM. Therefore, DC07090 represents a new non-peptidyl small molecule inhibitor for further development of antiviral therapy against EV71 or other picornaviruses.

4.2.1. Expression and purification of the recombinant wild-type (WT) EV71 3Cpro and mutants The plasmid encoding N-terminal 6 His-tagged EV71 3Cpro was kindly provided by Dr. Cui (Institute of Pathogen Biology, Beijing). EV71 3Cpro mutants were obtained using the QuickChange II site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's instruction. The expression and purification of EV71 3Cpro and mutants were carried out as described previously [20]. Briefly, the recombinant proteases were overexpressed in Escherichia coli (Rosetta; Novagen) under the induction of 0.5 mM isopropyl-b-thiogalactopyranoside (IPTG) at 18 C overnight, and purified using Ni-nitrilotriacetic acid (NTA) resin (Qiagen), followed by desalting column to remove imidazole using the buffer containing 50 mM Tris (pH7.5), 150 mM NaCl, 2 mM DTT, 1 mM EDTA and 10% glycerol. The protein purity was determined by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis. Aliquots of the purified protein were stored at 80 C freezer for enzymatic assay.

4. Experimental 4.1. Computational chemistry 4.1.1. Virtual screening The crystal structure of EV71 3Cpro was retrieved from the protein data bank (PDB entry: 3SJO [30]), and the co-crystallized inhibitor AG7088, a peptide mimic inhibitor with a covalent linkage to Cys147, was deleted. All crystallographic water molecules were also removed from the coordinate set. Glide calculations were performed with Maestro v9.1 (Schrodinger, Inc.) [34,35]. First, hydrogen atoms and charges were added through the “Protein Preparation” panel in Maestro, and a restrained partial minimization was performed until the root-mean-square deviation (rmsd) reached a maximum value of 0.3 Å to relieve steric clashes. Then, the grid-enclosing box was centered on the place where the center of AG7088 locates. To soften the potential for nonpolar parts of the receptor, a scaling factor of 1.0 was set to van der Walls (VDW) radii of those receptor atoms with the partial atomic charge less than 0.25. In the docking process, HTVS (high throughput virtual screening) mode was first used to screen against the whole SPECS database (http://www.specs.net). The top ranked 5000 molecules were filtered for more precisely docking procedure using the XP (extra-precision) mode of Glide. Finally, the top 500 molecules were selected for conformational and structural diversity analysis, yielding a set of 50 compounds for further biological assays. 4.1.2. Molecular dynamics simulation The docked complex of DC07090 and EV71 3Cpro was used as starting point for molecular dynamics (MD) simulation using GROMACS package 4.5.3 [36,37], adopting the AMBER99sb force field parameters [38]. The ACPYPE [39] was used to generate the topology parameters of DC07090. The complex was located in a cubic box with the periodic boundary conditions, which was filled with extended simple-point charge (SPC) water molecules [40]. As the solvated system was already neutralized, no ions were needed to be added. First, 1-ns Energy minimization was performed by the steepest descent method. Next, the system was equilibrated by performing the position-restrained MD simulations (NVT and NPT) at 300 K at time period of 500ps and 2 ns, respectively. Finally, the full system was subjected to MD simulations for 100 ns at the constant temperature (300 K) and pressure (1.0 bar). The long range electrostatic interactions were handled by the particle mesh Ewald (PME) algorithm [41] with the cutoff radius of 14 Å. During the simulations, the trajectory snapshots were stored for structural analysis at every 2 ps. Root mean square deviation (RMSD) and hydrogen bonds were analyzed through GROMACS utilities g_rms and g_hbond. The occupancy of hydrogen-bonding interactions was analyzed by VMD [42].

4.2.2. Enzyme inhibition assay The activity of EV71 3Cpro was performed according to the published method with some modifications [20]. The fluorescence peptide substrate Dabcyl-RTATVQGPSLDFE-Edans was synthesized by Shanghai Saiyi biotechnology Ltd. Briefly, 25 mL reaction mixtures containing the reaction buffer (20 mM Tris-HCl pH 7.5, 50 mM NaCl, 1 mM EDTA, 1 mM DTT, 10% glycerol), 8 mM EV71 3Cpro and the tested compounds dissolved in DMSO were incubated for 30 min at 30 C in 384-well plates. The substrates were added to initiate the reaction. Hydrolysis of the substrate was monitored with SpectraMax® M5 Multi-Mode Microplate Reader at excitation and emission wavelengths of 340 and 500 nm, respectively. The initial velocities (before 600 s) of the reaction was plotted as the activity of EV71 3Cpro. Inhibition of EV71 3Cpro activity of the tested compounds was calculated using the formula (VDMSO-Vtested compounds)/ VDMSO*100%. Values are averaged from three independent experiments. For IC50 determination of the tested compounds, the same reaction was performed as described above with serial dilutions of the inhibitors as indicated, and IC50 value was fitted using Graphpad Prism 5. The inhibition type was determined using direct linear plots at various substrate and inhibitor concentrations. 4.2.3. Surface plasmon resonance (SPR) technology-based binding assay Binding affinity of the compound to EV71 3Cpro was investigated by SPR based Biacore 3000 instrument (GE Healthcare). Firstly, purified EV71 3Cpro in the buffer (20 mM Tris-HCl pH 7.5, 50 mM NaCl, 1 mM EDTA, 1 mM DTT) was immobilized on CM5 sensor chip by primary amine coupling procedure. Before immobilization, the CM5 sensor chip was equilibrated with HBS-EP buffer (10 mM HEPES, 3 mM EDTA, 150 mM NaCl, and 0.005% surfactant P20, pH 7.4). Then different concentrations of the compound diluted in HBS-EP buffer were injected into the chip at a flow rate of 30 mL/ min for 60s, followed by dissociation for 120s. The equilibrium dissociation constant (Kd) was calculated by 1:1 Langmuir binding model based on BIA evaluation software (version 3.1; Biacore). 4.2.4. Fluorescence quenching assay All fluorescence spectra were recorded on F-2500 fluorescence spectrophotometer (Hitachi, Japan) equipped with 1 cm quartz cell. Different concentrations of the compound were mixed with 2 mM EV71 3Cpro for 1 h on ice. Sample then was placed into a quartz cell and kept for 5 min in dark before measurement. The emission spectra were recorded from 300 to 400 nm at 25 C under the


excitation wavelength of 280 nm. The Kd value was fitted according to the published paper [43]. 4.2.5. Circular dichroism (CD) measurements Characterization of wild-type (WT) EV71 3Cpro and mutants were determined by CD measurement. CD spectra were recorded by a JASCO J-810 spectropolarimeter at 25 C. 20 mM proteases were respectively prepared in the solution of 20 mM Tris-HCl, Ph 7.5, 50 mM NaCl, 1 mM EDTA, 1 mM DTT and 10% glycerol. All proteins including EV71 3Cpro WT and mutants were kept the same concentration. Far-UV CD spectra were collected from 190 to 260 nm at a scanning speed of 20 nm/min using 0.1 cm path length cuvette. Experimental data were the average values of at least three measurements and corrected by subtracting the buffer spectra in the same condition in the absence of protease. 4.2.6. Anti-viral assay Human embryonal rhabdomyosarcoma (RD) cells were plated into 96-well plates with concentration of 1 104/mL in DMEM containing 10% FBS, 100 unit/mL penicillin and 100 mg/mL streptomycin at 37 C in a humidified 5% CO2 incubator for 24 h. EV71 virus (50 TCID50) mixed with different concentrations of the tested compounds was added to the cells and incubated at 37 C in a humidified 5% CO2 incubator for 72 h. Similarly, the antiviral activity of the tested compounds against CVA16 virus (100 TCID50) was measured as described above. The protection effect of the tested compounds on RD cells was determined with CellTiter-Glo® Luminescent Cell Viability assay kit (Promega, USA). The concentration of the tested compound to reduce the virus-induced cell death by 50% relative to the virus control was defined as EC50. All assays were performed in triplicate and at least twice. 4.2.7. Cytotoxicity The cytotoxicity was measured in RD cells using the CellTiterGlo® Luminescent Cell Viability assay kit according to the manufacturer's protocol (Promega, USA). Briefly, RD cells were planted in 96-well opaque-walled plates at 1 103 cells per well in 100 mL DMEM medium supplemented with 10% FBS and 100 unit/mL penicillin and 100 mg/mL streptomycin and cultured overnight at 37 C under 5% CO2. The compound with different concentrations (0e200 mM) was added and incubated for 48 h at 37 C and then CellTiter-Glo® reagent was added to each well. The luminescent signals were read with a multimode microplate reader (SpectraMax M5, Molecular Devices, USA). The cell viability in each well was calculated as follows: percent of inhibition ¼ 1-Rlu(treated sample)/Rlu(cell control sample)*100%. The 50% cytotoxic concentration (CC50) was calculated by regression analysis using GraphPad Prism. 4.3. Synthesis 4.3.1. Chemistry The reagents (chemicals) were purchased from a commercial chemical reagent company and used without further purification unless otherwise stated. Analytical thin-layer chromatography (TLC) was done with HSGF 254. All target products were characterized by NMR, LRMS, and HRMS. Chemical shifts were reported in parts per million (ppm, d) downfield from tetramethylsilane. Proton coupling patterns are described as singlet (s), doublet (d), triplet (t), quartet (q), multiplet (m), and broad (br). Low- and high-resolution mass spectra (LRMS and HRMS) were obtained with electric, electrospray, and matrix-assisted laser desorption ionization (EI and ESI) produced by FinniganMAT-95 and LCQ-DECA spectrometers.

989

4.3.2. 4-(Oxazolo[4,5-b]pyridin-2-yl)-N-(pyridin-2-ylmethyl) aniline (DC07090) To a solution of 2-Cl-pyridine (0.60 g, 4.70 mmol) in toluene (10 mL), methyl 4-aminobenzoate (0.71 g, 4.70 mmol) and triethylamine (0.48 g, 4.70 mmol) were added at room temperature. Under argon protection, the resulting solution was refluxed for 24 h. After being cooled to room temperature and diluted with ether, the resulting mixture was extracted with NaHCO3, then the extract was washed with H2O and dried with MgSO4. After evaporation of the solvent, the desired Methyl 4-((pyridin-2-ylmethyl) amino) benzoate (1c, 0.84 g) was obtained as white solid in 74% yield via column chromatography (silica, PE/AcOEt, 2/1, v/v). 1H NMR (400 MHz, CDCl3) d 8.59 (d, 1H), 7.85 (d, 2H), 7.64 (tri, 1H), 7.29 (d, 1H), 7.19 (tri, 1H), 6.61 (d, 2H), 5.49 (s, 1H), 4.48 (s, 2H), 4.59 (s, 2H) 3.81 (s, 3H). The ester from above (0.84 g, 3.47 mmol) was dissolved in MeOH (20 mL) and 2 M NaOH (10.40 mmol, 5.20 mL) was added. The mixture was stirred at 60 C for 4 h and then cooled to room temperature. MeOH was removed under reduced pressure and the residue acidified to pH ¼ 4 with glacial AcOH. The precipitated carboxylic acid was collected by filtration, washed with water and dried to give 4-((Pyridin-2-ylmethyl)amino)benzoic acid (1d, 0.71 g, 89% yield). 1H NMR (400 MHz, DMSO-d6) d 12.01 (S, 1H), 8.54 (d, 1H), 7.75 (tri, 1H), 7.65 (d, 2H), 7.34 (d, 1H), 7.26 (tri, 2H), 6.61 (d, 2H), 4.41 (d, 2H), 4.02 (q, 1H). The acid from above (0.71 g 3.11 mmol) and 2-aminopyridin-3ol (0.34 g 3.11 mmol) were added to a 10 mL round bottom flask, followed by PPA (5 g) was added to the mixture, then the reaction mixture was vigorously stirred for 0.5 h at 180 C. After that, water was added and the residue acidified to pH ¼ 7 with 2 M NaOH, then the resulting mixture was extracted with AcOEt (20 mL x 3), dried with Na2SO4 and purified via column chromatography (silica, PE/ AcOEt, 1/3, v/v). The desired product DC07090 was obtained as yellow solid (0.41 g, 63% yield). M.P. 131e133 C. 1H NMR (400 MHz, CDCl3) d 8.62 (d, J ¼ 4.3 Hz, 1H), 8.49 (dd, J ¼ 4.9, 1.3 Hz, 1H), 8.13 (d, J ¼ 8.8 Hz, 2H), 7.84e7.67 (m, 2H), 7.39 (d, J ¼ 7.8 Hz, 1H), 7.32e7.23 (m, 2H), 7.19 (dd, J ¼ 8.0, 5.0 Hz, 1H), 6.77 (d, J ¼ 8.8 Hz, 2H), 4.59 (s, 2H). LRMS (ESI) [MþH]þ found m/z 303.1. 4.3.3. 4-(5-Methyloxazolo[4,5-b]pyridin-2-yl)-N-(pyridin-2ylmethyl)aniline (1) By the same manner as described for the preparation of DC07090, compound 1 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 56%, yellow solid, mp 136e138 C. 1H NMR (400 MHz, CDCl3) d 8.45 (q, 1H), 8.10 (d, 2H), 7.72 (q, 1H), 7.54 (tri, 1H), 7.15 (d, 1H), 7.11 (d, 1H), 7.06 (d, 1H), 6.74 (d, 2H), 5.67 (s, 1H), 4.48 (s, 2H), 2.57 (s, 3H). LRMS (ESI) [MþH]þ found m/z 317.1. 4.3.4. N-((6-methylpyridin-2-yl)methyl)-4-(oxazolo [4,5-b]pyridin2-yl)aniline (2) By the same manner as described for the preparation of DC07090, compound 2 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 52%. Yellow solid, mp 134e136 C. 1H NMR (400 MHz, CDCl3) d 8.45 (dd, J ¼ 5.0, 1.4 Hz, 1H), 8.10 (d, J ¼ 8.9 Hz, 2H), 7.73 (dd, J ¼ 8.0, 1.4 Hz, 1H), 7.54 (t, J ¼ 7.7 Hz, 1H), 7.12 (ddd, J ¼ 29.8, 15.1, 6.3 Hz, 3H), 6.74 (d, J ¼ 8.9 Hz, 2H), 5.67 (s, 1H), 4.48 (s, 2H), 2.57 (s, 3H). LRMS (ESI) [MþH]þ found m/z 317.1. 4.3.5. 4-(5-Methyloxazolo[4,5-b]pyridin-2-yl)-N-((6methylpyridin-2-yl)methyl)aniline (3) By the same manner as described for the preparation of DC07090, compound 3 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 58%, yellow solid,

990


mp 151e153 C. 1H NMR (400 MHz, CDCl3) d 8.17e8.04 (m, 2H), 7.67e7.53 (m, 2H), 7.18e6.97 (m, 3H), 6.74 (d, J ¼ 8.9 Hz, 2H), 5.58 (s, 1H), 4.51 (s, 2H), 2.64 (s, 3H), 2.60 (s, 3H). LRMS (ESI) [MþH]þ found m/z 331.1. 4.3.6. 2-Fluoro-4-(oxazolo[4,5-b]pyridin-2-yl)-N-(pyridin-2ylmethyl)aniline (4) By the same manner as described for the preparation of DC07090, compound 4 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 52%, yellow solid, mp 136e138 C. 1H NMR (400 MHz, CDCl3) d 8.62 (d, J ¼ 4.2 Hz, 1H), 8.50 (dd, J ¼ 4.9, 1.4 Hz, 1H), 7.95 (ddd, J ¼ 13.9, 10.0, 1.5 Hz, 2H), 7.78 (dd, J ¼ 8.0, 1.4 Hz, 1H), 7.74e7.66 (m, 1H), 7.34 (d, J ¼ 7.9 Hz, 1H), 7.26e7.18 (m, 2H), 6.75 (t, J ¼ 8.4 Hz, 1H), 5.70 (s, 1H), 4.59 (s, 2H). LRMS (ESI) [MþH]þ found m/z 321.0. 4.3.7. 2-Fluoro-4-(5-methyloxazolo[4,5-b]pyridin-2-yl)-N(pyridin-2-ylmethyl)aniline (5) By the same manner as described for the preparation of DC07090, compound 5 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 61%, yellow solid, mp 144e146 C. 1H NMR (400 MHz, CDCl3) d 8.63 (d, J ¼ 4.6 Hz, 1H), 7.91 (dd, J ¼ 17.6, 5.3 Hz, 3H), 7.72 (t, J ¼ 7.8 Hz, 1H), 7.65 (d, J ¼ 8.2 Hz, 1H), 7.36 (d, J ¼ 7.6 Hz, 1H), 7.06 (d, J ¼ 8.1 Hz, 1H), 6.74 (t, J ¼ 8.5 Hz, 1H), 5.66 (s, 1H), 4.61 (s, 3H), 2.66 (s, 5H). LRMS (ESI) [MþH]þ found m/z 335.0. 4.3.8. 2-Fluoro-4-(5-methyloxazolo[4,5-b]pyridin-2-yl)-N-((6methylpyridin-2-yl)methyl)aniline (6) By the same manner as described for the preparation of DC07090, compound 6 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 66%, yellow solid, mp 154e156 C, 1H NMR (400 MHz, CDCl3) d 7.92 (t, J ¼ 10.9 Hz, 2H), 7.70e7.53 (m, 2H), 7.19e7.02 (m, 3H), 6.74 (t, J ¼ 8.5 Hz, 1H), 5.64 (s, 1H), 4.56 (s, 2H), 2.66 (s, 3H), 2.61 (s, 3H). LRMS (ESI) [MþH]þ found m/z 349.0. 4.3.9. 2-Fluoro-N-((6-methylpyridin-2-yl)methyl)-4-(oxazolo[4,5b]pyridin-2-yl)aniline (7) By the same manner as described for the preparation of DC07090, compound 7 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 62%, yellow solid, mp 141e143 C, 1H NMR (400 MHz, CDCl3) d 8.46 (d, J ¼ 5.0 Hz, 1H), 7.94e7.80 (m, 2H), 7.74 (dd, J ¼ 8.1, 1.4 Hz, 1H), 7.55 (t, J ¼ 7.7 Hz, 1H), 7.18 (dd, J ¼ 8.0, 4.9 Hz, 1H), 7.08 (dd, J ¼ 19.2, 7.5 Hz, 2H), 6.71 (t, J ¼ 8.4 Hz, 1H), 5.73 (s, 1H), 4.51 (d, J ¼ 4.0 Hz, 2H), 2.56 (s, 3H). LRMS (ESI) [MþH]þ found m/z 335.0.

mp 162e164 C, 1H NMR (400 MHz, CDCl3) d 8.64 (d, J ¼ 4.6 Hz, 1H), 8.22 (d, J ¼ 1.9 Hz, 1H), 8.05 (dd, J ¼ 8.6, 1.9 Hz, 1H), 7.69 (ddd, J ¼ 21.4, 10.6, 4.9 Hz, 2H), 7.32 (d, J ¼ 7.7 Hz, 1H), 7.07 (d, J ¼ 8.2 Hz, 1H), 6.72 (d, J ¼ 8.6 Hz, 1H), 6.09 (s, 1H), 4.61 (d, J ¼ 4.9 Hz, 2H), 2.66 (s, 3H). LRMS (ESI) [MþH]þ found m/z 351.0. 4.3.12. 2-Chloro-N-((6-methylpyridin-2-yl)methyl)-4-(oxazolo [4,5-b]pyridin-2-yl)aniline (10) By the same manner as described for the preparation of DC07090, compound 10 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 62%, yellow solid, mp 156e158 C, 1H NMR (400 MHz, CDCl3) d 8.50 (d, J ¼ 3.7 Hz, 1H), 8.22 (d, J ¼ 1.9 Hz, 1H), 8.05 (dd, J ¼ 8.6, 1.7 Hz, 1H), 7.77 (dd, J ¼ 8.0, 1.4 Hz, 1H), 7.56 (t, J ¼ 7.7 Hz, 1H), 7.21 (dd, J ¼ 8.0, 5.0 Hz, 1H), 7.09 (t, J ¼ 7.2 Hz, 2H), 6.71 (d, J ¼ 8.6 Hz, 1H), 6.16 (s, 1H), 4.55 (d, J ¼ 5.1 Hz, 2H), 2.59 (s, 3H). LRMS (ESI) [MþH]þ found m/z 351.0. 4.3.13. 2-Chloro-4-(5-methyloxazolo [4,5-b]pyridin-2-yl)-N-((6methylpyridin-2-yl)methyl)aniline (11) By the same manner as described for the preparation of DC07090, compound 11 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 65%, yellow solid, mp 168e170 C, 1H NMR (400 MHz, CDCl3) d 8.21 (d, J ¼ 1.9 Hz, 1H), 8.04 (dd, J ¼ 8.5, 1.8 Hz, 1H), 7.70e7.53 (m, 2H), 7.18e7.10 (m, 2H), 7.06 (d, J ¼ 8.2 Hz, 1H), 6.70 (d, J ¼ 8.6 Hz, 1H), 6.12 (d, J ¼ 4.7 Hz, 1H), 4.59 (s, 2H), 2.65 (s, 3H), 2.62 (s, 3H). LRMS (ESI) [MþH]þ found m/z 365.1. 4.3.14. 4-(Oxazolo [4,5-b]pyridin-2-ylmethyl)-N-(pyridin-2ylmethyl)aniline (12) By the same manner as described for the preparation of DC07090, compound 12 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 53%, brown solid, mp 113e115 C, 1H NMR (400 MHz, CDCl3) d 8.62e8.51 (m, 1H), 7.76 (dd, J ¼ 4.6, 1.5 Hz, 1H), 7.64 (td, J ¼ 7.7, 1.8 Hz, 1H), 7.41e7.23 (m, 2H), 7.18 (dd, J ¼ 6.7, 5.2 Hz, 1H), 7.13e6.99 (m, 3H), 6.65 (t, J ¼ 5.6 Hz, 2H), 4.44 (s, 2H), 3.67 (s, 2H). LRMS (ESI) [MþH]þ found m/z 317.1. 4.3.15. 4-((5-Methyloxazolo[4,5-b]pyridin-2-yl)methyl)-N(pyridin-2-ylmethyl)aniline (13) By the same manner as described for the preparation of DC07090, compound 13 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 57%, brown solid, mp 116e118 C, 1H NMR (400 MHz, CDCl3) d 8.55 (d, 1H), 7.65 (tri, 1H), 7.57 (d, 1H), 7.31 (d, 1H), 7.18 (d, 3H), 7.04 (d, 1H), 6.61 (d, 2H), 4.43 (s, 2H), 4.16 (s, 2H), 2.62 (s, 3H), LRMS (ESI) [MþH]þ found m/z 331.1.

4.3.10. 2-Chloro-4-(oxazolo[4,5-b]pyridin-2-yl)-N-(pyridin-2ylmethyl)aniline (8) By the same manner as described for the preparation of DC07090, compound 8 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 43%, yellow solid, mp 145e147 C, 1H NMR (400 MHz, CDCl3) d 8.64 (d, J ¼ 4.0 Hz, 1H), 8.50 (dd, J ¼ 4.9, 1.4 Hz, 1H), 8.23 (dd, J ¼ 13.7, 1.9 Hz, 1H), 8.15e8.02 (m, 1H), 7.83e7.75 (m, 1H), 7.70 (td, J ¼ 7.7, 1.7 Hz, 1H), 7.32 (t, J ¼ 8.0 Hz, 1H), 7.25e7.13 (m, 2H), 6.72 (t, J ¼ 11.2 Hz, 1H), 6.14 (d, J ¼ 8.5 Hz, 1H), 4.61 (t, J ¼ 6.8 Hz, 2H). LRMS (ESI) [MþH]þ found m/z 303.1. LRMS (ESI) [MþH]þ found m/z 337.0.

4.3.16. 4-((5-Methyloxazolo [4,5-b]pyridin-2-yl)methyl)-N-((3methylpyridin-2-yl)methyl)aniline (14) By the same manner as described for the preparation of DC07090, compound 14 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 45%, brown solid, mp 124e126 C, 1H NMR (400 MHz, CDCl3) d 7.52 (t, J ¼ 7.7 Hz, 1H), 7.18 (d, J ¼ 8.1 Hz, 1H), 7.10 (dd, J ¼ 14.9, 8.0 Hz, 3H), 7.03 (d, J ¼ 7.6 Hz, 1H), 6.87 (d, J ¼ 8.1 Hz, 1H), 6.65 (d, J ¼ 8.2 Hz, 2H), 4.39 (s, 2H), 3.65 (s, 2H), 2.56 (s, 3H), 2.30 (s, 3H). LRMS (ESI) [MþH]þ found m/z 345.1.

4.3.11. 2-Chloro-4-(5-methyloxazolo[4,5-b]pyridin-2-yl)-N(pyridin-2-ylmethyl)aniline (9) By the same manner as described for the preparation of DC07090, compound 9 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 51%, yellow solid,

4.3.17. N-((3-methylpyridin-2-yl)methyl)-4-(oxazolo [4,5-b] pyridin-2-ylmethyl)aniline (15) By the same manner as described for the preparation of DC07090, compound 15 was prepared and purified by silica gel chromatography (silica, PE/AcOEt, 1/3, v/v). Yield: 51%, brown solid,


mp 119e121 C, 1H NMR (400 MHz, CDCl3) d 8.58e8.52 (m, 1H), 7.62 (td, J ¼ 7.7, 1.8 Hz, 1H), 7.57 (d, J ¼ 8.3 Hz, 1H), 7.31 (d, J ¼ 7.8 Hz, 1H), 7.17 (dd, J ¼ 7.5, 5.5 Hz, 3H), 7.04 (d, J ¼ 8.3 Hz, 1H), 6.64e6.57 (m, 2H), 4.43 (s, 2H), 4.16 (s, 2H), 2.62 (s, 3H). LRMS (ESI) [MþH]þ found m/z 331.1. Acknowledgements We gratefully acknowledge financial support from the National Natural Science Foundation of China (grant 81220108025, 81573351) and the State Key Laboratory of Natural and Biomimetic Drugs. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ejmech.2016.10.019. References [1] C. Wu, L. Zhang, P. Li, Q. Cai, X. Peng, K. Yin, X. Chen, H. Ren, S. Zhong, Y. Weng, Y. Guan, S. Chen, J. Wu, J. Li, T. Lin, Fragment-wise design of inhibitors to 3C proteinase from enterovirus 71, Biochim. Biophys. Acta 1860 (2016) 1299e1307. [2] Y. Wang, B. Yang, Y. Zhai, Z. Yin, Y. Sun, Z. Rao, Peptidyl aldehyde NK-1.8k suppresses enterovirus 71 and enterovirus 68 infection by targeting protease 3C, Antimicrob. Agents Chemother. 59 (2015) 2636e2646. [3] J.P. Lott, K. Liu, M.L. Landry, W.A. Nix, M.S. Oberste, J. Bolognia, B. King, Atypical hand-foot-and-mouth disease associated with coxsackievirus A6 infection, J. Am. Acad. Dermatol 69 (2013) 736e741. [4] M.J. Ang, Q.Y. Lau, F.M. Ng, S.W. Then, A. Poulsen, Y.K. Cheong, Z.X. Ngoh, Y.W. Tan, J. Peng, T.H. Keller, J. Hill, J.J. Chu, C.S. Chia, Peptidomimetic ethyl propenoate covalent inhibitors of the enterovirus 71 3C protease: a P2-P4 study, J. Enzyme Inhib. Med. Chem. 31 (2016) 332e339. [5] S. Owatanapanich, R. Wutthanarungsan, W. Jaksupa, U. Thisyakorn, Risk factors for severe enteroviral infections in children, J. Med. Assoc. Thai 99 (2016) 322e330. [6] C.C. Lai, D.S. Jiang, H.M. Wu, H.H. Chen, A dynamic model for the outbreaks of hand, foot, and mouth disease in Taiwan, Epidemiol. Infect. 144 (2016) 1500e1511. [7] P. Wang, W.B. Goggins, E.Y. Chan, Hand, foot and mouth disease in Hong Kong: a time-series analysis on its relationship with weather, PLoS One 11 (2016) e0161006. [8] N.T. Nguyen, H.V. Pham, C.Q. Hoang, T.M. Nguyen, L.T. Nguyen, H.C. Phan, L.T. Phan, L.N. Vu, N.N. Tran Minh, Epidemiological and clinical characteristics of children who died from hand, foot and mouth disease in Vietnam, 2011, BMC Infect. Dis. 14 (2014) 341. [9] J. Liao, Z. Qin, Z. Zuo, S. Yu, J. Zhang, Spatial-temporal mapping of hand foot and mouth disease and the long-term effects associated with climate and socio-economic variables in Sichuan Province, China from 2009 to 2013, Sci. Total Environ. 563e564 (2016) 152e159. [10] F.C. Jiang, F. Yang, L. Chen, J. Jia, Y.L. Han, B. Hao, G.W. Cao, Meteorological factors affect the hand, foot, and mouth disease epidemic in Qingdao, China, 2007-2014, Epidemiol. Infect. 144 (2016) 2354e2362. [11] H.T. Zhou, Y.H. Guo, M.J. Chen, Y.X. Pan, L. Xue, B. Wang, S.H. Tao, N. Yu, Changes in enterovirus serotype constituent ratios altered the clinical features of infected children in Guangdong Province, China, from 2010 to 2013, BMC Infect. Dis. 16 (2016) 399. [12] C.T. Lim, L. Jiang, S. Ma, L. James, L.W. Ang, Basic reproduction number of coxsackievirus type A6 and A16 and enterovirus 71: estimates from outbreaks of hand, foot and mouth disease in Singapore, a tropical city-state, Epidemiol. Infect. 144 (2016) 1028e1034. [13] H.L. Wen, L.Y. Si, X.J. Yuan, S.B. Hao, F. Gao, F.L. Chu, C.X. Sun, Z.Y. Wang, Complete genome sequencing and analysis of six enterovirus 71 strains with different clinical phenotypes, Virol. J. 10 (2013) 115. [14] N.J. Schmidt, E.H. Lennette, H.H. Ho, An apparently new enterovirus isolated from patients with disease of the central nervous system, J. Infect. Dis. 129 (1974) 304e309. [15] J.M. Bible, M. Iturriza-Gomara, B. Megson, D. Brown, P. Pantelidis, P. Earl, J. Bendig, C.Y. Tong, Molecular epidemiology of human enterovirus 71 in the United Kingdom from 1998 to 2006, J. Clin. Microbiol. 46 (2008) 3192e3200. [16] C.Y. Tan, G. Gonfrier, L. Ninove, C. Zandotti, A. Dubot-Peres, X. de Lamballerie, R.N. Charrel, Screening and detection of human enterovirus 71 infection by a real-time RT-PCR assay in Marseille, France, 2009-2011, Clin. Microbiol. Infect. 18 (2012) E77eE80. [17] C. Nanda, R. Singh, S.K. Rana, An outbreak of hand-foot-mouth disease: a report from the hills of northern India, Natl. Med. J. India 28 (2015) 126e128. [18] X. Han, N. Sun, H. Wu, D. Guo, P. Tien, C. Dong, S. Wu, H.B. Zhou, Identification and structure-activity relationships of diarylhydrazides as novel potent and

991

selective human enterovirus inhibitors, J. Med. Chem. 59 (2016) 2139e2150. [19] L. Shang, M. Xu, Z. Yin, Antiviral drug discovery for the treatment of enterovirus 71 infections, Antivir. Res. 97 (2013) 183e194. [20] J. Wang, T. Fan, X. Yao, Z. Wu, L. Guo, X. Lei, J. Wang, M. Wang, Q. Jin, S. Cui, Crystal structures of enterovirus 71 3C protease complexed with rupintrivir reveal the roles of catalytically important residues, J. Virol. 85 (2011) 10021e10030. [21] C.J. Kuo, J.J. Shie, J.M. Fang, G.R. Yen, J.T. Hsu, H.G. Liu, S.N. Tseng, S.C. Chang, C.Y. Lee, S.R. Shih, P.H. Liang, Design, synthesis, and evaluation of 3C protease inhibitors as anti-enterovirus 71 agents, Bioorg Med. Chem. 16 (2008) 7388e7398. [22] Y. Zhai, X. Zhao, Z. Cui, M. Wang, Y. Wang, L. Li, Q. Sun, X. Yang, D. Zeng, Y. Liu, Y. Sun, Z. Lou, L. Shang, Z. Yin, Cyanohydrin as an anchoring group for potent and selective inhibitors of enterovirus 71 3C protease, J. Med. Chem. 58 (2015) 9414e9420. [23] Y.J. Lin, Y.C. Chang, N.W. Hsiao, J.L. Hsieh, C.Y. Wang, S.H. Kung, F.J. Tsai, Y.C. Lan, C.W. Lin, Fisetin and rutin as 3C protease inhibitors of enterovirus A71, J. Virol. Methods 182 (2012) 93e98. [24] J. Wang, T. Zhang, J. Du, S. Cui, F. Yang, Q. Jin, Anti-enterovirus 71 effects of chrysin and its phosphate ester, PLoS One 9 (2014) e89668. [25] Z. Cao, Y. Ding, Z. Ke, L. Cao, N. Li, G. Ding, Z. Wang, W. Xiao, Luteoloside acts as 3C protease inhibitor of enterovirus 71 in vitro, PLoS One 11 (2016) e0148693. [26] D. Zeng, Y. Ma, R. Zhang, Q. Nie, Z. Cui, Y. Wang, L. Shang, Z. Yin, Synthesis and structure-activity relationship of alpha-keto amides as enterovirus 71 3C protease inhibitors, Bioorg Med. Chem. Lett. 26 (2016) 1762e1766. [27] P.S. Dragovich, T.J. Prins, R. Zhou, E.L. Brown, F.C. Maldonado, S.A. Fuhrman, L.S. Zalman, T. Tuntland, C.A. Lee, A.K. Patick, D.A. Matthews, T.F. Hendrickson, M.B. Kosa, B. Liu, M.R. Batugo, J.P. Gleeson, S.K. Sakata, L. Chen, M.C. Guzman, J.W. Meador 3rd, R.A. Ferre, S.T. Worland, Structure-based design, synthesis, and biological evaluation of irreversible human rhinovirus 3C protease inhibitors. 6. Structure-activity studies of orally bioavailable, 2-pyridone-containing peptidomimetics, J. Med. Chem. 45 (2002) 1607e1623. [28] G. Galati, P.J. O'Brien, Potential toxicity of flavonoids and other dietary phenolics: significance for their chemopreventive and anticancer properties, Free Radic. Biol. Med. 37 (2004) 287e303. [29] C.L. Deng, H. Yeo, H.Q. Ye, S.Q. Liu, B.D. Shang, P. Gong, S. Alonso, P.Y. Shi, B. Zhang, Inhibition of enterovirus 71 by adenosine analog NITD008, J. Virol. 88 (2014) 11915e11923. [30] G. Lu, J. Qi, Z. Chen, X. Xu, F. Gao, D. Lin, W. Qian, H. Liu, H. Jiang, J. Yan, G.F. Gao, Enterovirus 71 and coxsackievirus A16 3C proteases: binding to rupintrivir and their substrates and anti-hand, foot, and mouth disease virus drug design, J. Virol. 85 (2011) 10319e10331. [31] S. Cui, J. Wang, T. Fan, B. Qin, L. Guo, X. Lei, J. Wang, M. Wang, Q. Jin, Crystal structure of human enterovirus 71 3C protease, J. Mol. Biol. 408 (2011) 449e461. [32] D.A. Matthews, W.W. Smith, R.A. Ferre, B. Condon, G. Budahazi, W. Sisson, J.E. Villafranca, C.A. Janson, H.E. McElroy, C.L. Gribskov, et al., Structure of human rhinovirus 3C protease reveals a trypsin-like polypeptide fold, RNAbinding site, and means for cleaving precursor polyprotein, Cell 77 (1994) 761e771. [33] D.A. Matthews, P.S. Dragovich, S.E. Webber, S.A. Fuhrman, A.K. Patick, L.S. Zalman, T.F. Hendrickson, R.A. Love, T.J. Prins, J.T. Marakovits, R. Zhou, J. Tikhe, C.E. Ford, J.W. Meador, R.A. Ferre, E.L. Brown, S.L. Binford, M.A. Brothers, D.M. DeLisle, S.T. Worland, Structure-assisted design of mechanism-based irreversible inhibitors of human rhinovirus 3C protease with potent antiviral activity against multiple rhinovirus serotypes, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 11000e11007. [34] R.A. Friesner, J.L. Banks, R.B. Murphy, T.A. Halgren, J.J. Klicic, D.T. Mainz, M.P. Repasky, E.H. Knoll, M. Shelley, J.K. Perry, D.E. Shaw, P. Francis, P.S. Shenkin, Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy, J. Med. Chem. 47 (2004) 1739e1749. [35] T.A. Halgren, R.B. Murphy, R.A. Friesner, H.S. Beard, L.L. Frye, W.T. Pollard, J.L. Banks, Glide: a new approach for rapid, accurate docking and scoring. 2. Enrichment factors in database screening, J. Med. Chem. 47 (2004) 1750e1759. [36] D. Van Der Spoel, E. Lindahl, B. Hess, G. Groenhof, A.E. Mark, H.J. Berendsen, GROMACS: fast, flexible, and free, J. Comput. Chem. 26 (2005) 1701e1718. [37] B. Hess, C. Kutzner, D. van der Spoel, E. Lindahl, GROMACS 4: Algorithms for Highly Efficient, Load-balanced, and Scalable Molecular Simulation. [38] V. Hornak, R. Abel, A. Okur, B. Strockbine, A. Roitberg, C. Simmerling, Comparison of multiple Amber force fields and development of improved protein backbone parameters, Proteins 65 (2006) 712e725. [39] A.W. Sousa da Silva, W.F. Vranken, Acpype - AnteChamber PYthon parser interfacE, BMC Res. Notes 5 (2012) 367. [40] K.L. Meagher, H.A. Carlson, Solvation influences flap collapse in HIV-1 protease, Proteins 58 (2005) 119e125. [41] U. Essmann, L. Perera, M.L. Berkowitz, T. Darden, H. Lee, L.G. Pedersen, A smooth particle mesh Ewald method, J. Chem. Phys. 103 (1995) 8577e8593. [42] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol. Graph. 14 (1996) 33e38. [43] X. Shen, K. Tomoo, S. Uchiyama, Y. Kobayashi, T. Ishida, Structural and thermodynamic behavior of eukaryotic initiation factor 4E in supramolecular formation with 4E-binding protein 1 and mRNA cap analogue, studied by spectroscopic methods, Chem. Pharm. Bull. (Tokyo) 49 (2001) 1299e1303.