3D Modeling of Cytochrome P450 2C19: Implications for Drug Design, Papers of Biology

Download 3D Modeling of Cytochrome P450 2C19: Implications for Drug Design and more Papers Biology in PDF only on Docsity! www.elsevier.com/locate/ybbrc Biochemical and Biophysical Research Communications 355 (2007) 513–5193D structure modeling of cytochrome P450 2C19 and its implication for personalized drug design Jing-Fang Wang a,b, Dong-Qing Wei a,*, Lin Li a,c, Si-Yuan Zheng b, Yi-Xue Li a,b, Kuo-Chen Chou d,* a College of Life Science and Biotechnology, Shanghai Jiaotong University, Shanghai 200240, China b Bioinformation Center, Key Lab of Systems Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China c Department of Biochemistry and Molecular Biology, Peking University, Beijing 100871, China d Gordon Life Science Institute, 13784 Torrey Del Mar Drive, San Diego, CA 92130, USA Received 26 January 2007 Available online 8 February 2007Abstract Cytochrome P450 2C19 (CYP2C19) is a member of the cytochrome P-450 enzyme superfamily and plays an important role in the metabolism of drugs. In order to gain insights for developing personalized drugs, the 3D (dimensional) structure of CYP2C19 has been developed based on the crystal structure of CYP2C9 (PDB code 1R90), and its structure–activity relationship with the ligands of CEC, Fluvoxamine, Lescol, and Ticlopidine investigated through the structure–activity relationship approach. By means of a series of docking studies, the binding pockets of CYP2C19 for the four compounds are explicitly defined that will be very useful for conducting mutagen- esis studies, providing insights into personalization of drug treatments and stimulating novel strategies for finding desired personalized drugs.  2007 Elsevier Inc. All rights reserved. Keywords: CYP2C19; Cytochrome P-450; Structure–activity relationship; CEC; Fluvoxamine; Lescol; Ticlopidine; Personalized drug designCytochrome P450 enzymes are the most important enzymes in phase I metabolism in mammals, and are pri- marily responsible for the metabolism for drugs [1]. These enzymes belong to a superfamily of mono-oxygenases which can be found in all kingdoms of life and show the extraordinary diversity in their reaction chemistry. In mammals, as well as many other cell types, they are found mainly in the membranes of the endoplasmic reticulum (microsomes) within liver cells (hepatocytes) [2], and are capable of utilizing haem iron to oxidize molecules, and often making them more water-soluble for clearance by either adding or unmasking a polar group. Mammalian CYP enzymes can oxidize both xenobiotics and endoge-0006-291X/$ - see front matter  2007 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2007.01.185 * Corresponding authors. E-mail addresses: dqwei@sjtu.edu.cn (D.-Q. Wei), kchou@san.rr.com (K.-C. Chou).nous compounds, and are significant for the detoxification of foreign substances, as well as for the control of the level of endogenous compounds, such as vitamin D metabolism, cholesterol synthesis, hormone synthesis and breakdown. CYP enzymes also relate to vascular autoregulation, espe- cially in the brain, and are vital to the formation of choles- terol, steroids and arachidonic acid metabolites. They can also clear the body of metabolic products such as bilirubin, which arises from the breakdown of haemoglobin. These enzymes are found throughout the body, where they often have specialized roles, although there is a high concentra- tion of CYP proteins in the liver. Some CYP proteins have been detected to exist as fused domains with one or more of their redox partners, which are considered to have the evolutionary advantage of increasing catalytic efficiency [3]. Fusion proteins have been found in prokaryotic and eukaryotic systems, with a wide variety of cytochrome P450/redox partner fusions. For 514 J.-F. Wang et al. / Biochemical and Biophysical Research Communications 355 (2007) 513–519instance, in Bacillus megaterium, soluble cytochrome P450 produces a highly efficient system for the oxygenation of fatty acids by covalently linking to a reductase enzyme. In another bacterial system, it has been found that CYP are attached as an N-terminal domain to a phthalate diox- ygenase reductase module, which has the function to allow the reductase to facilitate the breakdown and metabolism of aromatic phthalate. People can carry different alleles of CYP genes, which have a little variation in their genetic sequence attributable to nucleotide changes or polymorphisms. These polymor- phic variations are capable of causing individual and pop- ulation differences in the tolerance to toxins and drugs. Because CYP proteins are specific in terms of which drugs they clear and which they activate, polymorphic variations in different CYP genes may have different effects. CYP2C19 is a member of the cytochrome P-450 enzyme superfamily [4]. It affects a series clinically important drugs, such as proton pump inhibitors (omeprazole [5], lanzopraz- ole, and rabeprazole), phenytoin, tricyclic antidepressants (imipramine, amitryptiline), propranolol, and benzodiaze- pines (diazepam) [6]. Marked interethnic differences in the polymorphism frequency [4,7] have led to 21 variant alleles, from CYP2C19*1 to CYP2C19*21, being documented. So far no experimental structural data whatsoever for CYP2C19 are available, and hence we have to resort to structural bioinformatics [8] to deal with the problem. In this study, a homology model of CYP2C19 has been devel- oped based on the crystal structure of CYP2C9 (PDB code 1R90) [10]; the latter belongs to the same family as the CYP2C19. Subsequently, the structure–activity relation- ship research was carried out based on the modeled 3D structures of CYP2C19 for the ligands CEC, Fluvoxamine, Lescol, and Ticlopidine, respectively (Table 1). Finally, the drug-resistance of CYP2C19 was analyzed.Modeling the 3D structure of CYP2C19 The crystal structure of CYP2C9 (PDB code 1R90), which was used as a template to build the 3D structure of CYP2C19, was first reported in 2004 [9]. It is known that both CYP2C19 and CYP2C9 belong to the same family. Accordingly, it is rational to use the crystal structureTable 1 The detailed descriptions of the four ligands: CEC, Fluvoxamine, Lescol, and Ticlopidine Name Formula Chemical name CEC C11H9NO3 3-Cyano-7-ethoxycoumarin Fluvoxamine C15H21F3N2O2 2-[(5-Methoxy-1-[4-(trifluoromethyl) phenyl]pentylidene)amino]oxyethanamine Lescol C24H26FNO4 7-[3-(4-Fluorophenyl)-1-(1-methylethyl)- 1H-indol-2-yl]- 3,5-dihydroxy-hept-6- enoic acid Ticlopidine C14H14ClNS 3-[(2-Chlorophenyl)methyl]-7-thia-3- azabicyclo[4.3.0]nona-8,10-dieneCYP2C9 as a template to derive the 3D structure of CYP2C19 by the method of structural bioinformatics [8], as described below. The sequence of CYP2C9 contains 492 amino acids, and the sequence of CYP2C19 contains 490 amino acids. The sequence alignment between the CYP2C19 and CYP2C9 has been performed by the in-house program TIMM (Tian- jin Molecular Modeling). The result of alignment is shown in Fig. 1, where the segments (residues 1–10 and 480–490) for the CYP2C19 and the segment (residues 1–25) for the CYP2C9 are not shown because they are outside of the protease domains. The similarity of the two sequences is about 62%. Based on the atomic coordinates of the crystal structure CYP2C9 (PDB code 1R90) and the sequence alignment (Fig. 1), the 3D structure of CYP2C19 was developed by the segment matching or coordinate reconstruction approach [11–15]. The segment matching method was used to model the 3D structure of the protease domain of cas- pase-8 [16] before the crystal coordinates of caspase-3 were released [17]. To deal with the situation of lacking a proper template, the crystal structure of caspase-1 was first used as a template to model the 3D structure of the catalytic domain of caspase-3, and the structure thus obtained was subsequently utilized as a template to further model the protease domain of caspase-8. After the crystal coordinates of the caspase-3 protease domain and the caspase-8 prote- ase domain were finally released [18], it was found that the computed structures of caspase-3 and caspase-8 were quite close to their corresponding crystal structures. For instance, the root-mean-square-deviation (RMSD) for all the backbone atoms of the caspase-3 protease domain between the crystal and computed structures was 2.7 Å, while the corresponding RMSD for caspase-8 was 3.1 Å with only 1.2 Å for the core structure. Shortly afterwards, a number of other cases were reported in successively applying the segment matching approach, such as model- ing the CARDs (caspase recruitment domains of Apaf-1, Ced-4, and Ced-3 by using the NMR structure of the RAIDD CARD [19] as a template, and modeling the Cdk5-Nck5a* complex [20] as well as the protease domain of caspase-9 [21]. Two years after the computed Cdk5- Nck5a* complex structure was published [21], its crystal structure was determined [22], and it was found that the predicted Cdk5 and its crystal structure are almost the same. Also, it was observed [22] that the buried surface area was 3400 Å2 for the binding of Cdk5 with Nck5a*, while the corresponding area was 3461 Å2 as derived from the computed structure. Meanwhile, stimulated by the computed Cdk5-Nck5a*-ATP structure, the molecular truncation experiments were conducted [23], and a conclu- sion was drawn that the experimental results ‘‘confirm and extend specific aspects of the original predicted computer model’’. The segment match method was also used to derive the 3D structure of b-secretase zymogen [24], leading to a compelling elucidation of why the prodomain of b-se- cretase did not suppress activity like in a strict zymogen, as Table 2 List of interaction energies (kcal/mol) obtained by docking CEC, Fluvoxamine, Lescol, and Ticlopidine to CYP2C19, respectively Ligands E (electronic) E (van der Waals) E (binding) CEC 4.23 14.82 19.05 Fluvoxamine 3.25 14.84 18.09 Lescol 7.14 13.45 20.59 Ticlopidine 5.53 13.63 19.17 J.-F. Wang et al. / Biochemical and Biophysical Research Communications 355 (2007) 513–519 517such as the hinge binding. In some cases, this kind of change might be integral to the function of a protein through allosteric transition and signal transductions (see, e.g., [36,37]) When a new conformation of the ligand is gen- erated, the search for favorable binding configurations, by either simulated annealing [38] or Tabu search [39], is con- ducted within a specified 3D docking box. Both methods try to optimize the purely spatial contacts as well as electro- static interactions. The interaction energy is calculated with electrostatic and van der Waals potential fields. In all ourTable 3 Residues in forming the binding pockets of CYP 2C19 for the four ligands Ligands Pocket CEC Arg139, Glu142, Glu328, Ile331, Gly332, As Fuvoxamine Glu142, Gln146, Ile331, Gly332, Asn334, Ar Lescol Arg139, Glu142, Gln146, Glu147, Ile331, Gl Ticlopidine Glu328, Ile331, Gly332, Arg333, Asn334, Gl Fig. 4. Illustrations showing the lipophilic and hydrophilic surfaces of the Fluvoxamine, (C) Lescol, and (D) Ticlopidine, where the lipophilic and hydroph of the references to color in this figure legend, the reader is referred to the wecomputations, the CHARMM33 force field parameters were utilized [40]. The method utilized here to do the molecular docking [41] is called Monte Carlo simulated annealing, a Metrop- olis algorithm. It was used to dock CEC, Fluvoxamine, Lescol, and Ticlopidine to the computed 3D structure of CYP2C19, respectively, for finding the most favorable binding interaction. The similar molecular docking method had been used for studying the binding interaction of alpha 7 nAChR dimer with GTS-21 [42], and for anti-SARS drug screening [43–44].Results and discussions A close view of the binding interactions of CYP2C19 with CEC, Fluvoxamine, Lescol, and Ticlopidine, respec- tively, is shown in Fig. 3. It can be seen from Fig. 3B that there are six hydrogen bonds formed between Fluvoxamine and CYP2C19, indicating that Fluvoxamine has strongern334, Arg_335, Ser336, Pro337, Phe451, Gln454 g335, Ser336, Pro337, Gln340, Phe451, Gln454, Asn455 y332, Arg333, Asn334, Arg335, Ser336, Pro337, Phe451, Gln454, Asn455 n454, Asn455, Ile488, Pro489, Val490 binding pocket (or cavity) of cytochrome P450 2C19 for (A) CEC, (B) ilic surfaces are colored in green and blue, respectively. (For interpretation b version of this paper.) 518 J.-F. Wang et al. / Biochemical and Biophysical Research Communications 355 (2007) 513–519hydrogen-bonding interaction with CYP2C19 than the other three ligands (panels A, C, and D for Fig. 3). The amino acids involved in forming the hydrogen bonds with Fluvoxamine include Ile331, Asn334, Gln454, and Asn455. Of these four, Ile331 and Asn334 are considered to have more contributions to the binding interaction because more hydrogen bonds are formed by them. Besides, Ile331 is also involved in forming hydrogen bonds with Lescol and Tic- lopidine, and Asn334 involved in forming hydrogen bonds with CEC and Lescol. Listed in Table 2 are the binding energies obtained by docking CEC, Fluvoxamine, Lescol, and Ticlopidine to CYP2C19, respectively. As shown from the table, Lescol has the lowest binding energy. The amino acid residues in forming the four binding pockets are given in Table 3, from which we can see that Ile331, Gly332, Arg333, Asn334, and Gln454 occur in all the four pockets, while Glu142 in three of the four. Shown in the Fig. 4 are the lipophilic and hydrophilic surfaces of the CYP2C19 with the four ligands: (A) CEC, (B) Fluvoxamine, (C) Lescol, and (D) Ticlopi- dine, respectively. It can be seen from the figure (and Table 3 as well) that the active cavities are mainly hydrophilic. Conclusion The lipophilicity of inhibitors is a key factor for design- ing orally active drugs owing to the complemental lipophil- ic and hydrophilic interactions between the bioreceptors and ligands, and the balancing lipophilicity and water sol- ubility. The latter is necessary for its absorption from the intestinal tract. It is hard to understand the hydrophobic effects on the molecular level, but a remarkable change of the docking free energy between ligands and enzymes might be an important relevant factor. The different lipo- philicity in the binding pocket or active cavity of CYP2C19 might be the main reason for causing individually different drug effect. The key amino acid residues of the receptor’s binding pockets found through this study for CEC, Flu- voxamine, Lescol, and Ticlopidine, respectively, will be very useful for conducting mutagenesis studies for finding desired drugs or proper treatments according to the charac- teristics of an individual patient to improve efficacy and reduce the number and adverse drug reactions. Acknowledgments Valuable discussions with Professor Chen Chao of Chi- nese National Engineering Center on Micro-measurement and Professor Li Rulin of North West University are grate- fully acknowledged. This work was supported by the grants from Chinese National Science Foundation under the Con- tract No.10376024, the Tianjin Commission of Education under contract No. 20030001 and the Tianjin Commission of Sciences and Technology under the Contract No. 033801911. Additional supports were from the special fund for intensive computation, Virtual Laboratory for Compu- tational Chemistry of CNIC, and the SupercomputingCenter of CNIC, Chinese Academy of Sciences under the contract No. 043185111-4.References [1] S.P. Tanmay, F.N. Shaun, D. Sanish, M.S. Vishal, A.K. Nilima, J.G. Nithya, Evaluation of the activity of CYP2C19 in Gujrati and Marwadi subjects living in Mumbasi (Bombay), BMC Clin. Pharm. 6 (2006) 1–5. [2] F.J. Gonzale, C. Radek, W. Daniel, Characterization of the common genetic defect in humans deficient in debrisoquine metabolism, J. Nature 331 (1988) 442–444. [3] E. Regina, L.D. Tammy, Z. Arne, M. Romy, Identification and functional characterization of eight CYP3A4 protein variants, Phar- macogenetics 11 (2001) 447–458. [4] M. Ingelman-Sundberg, M. Oscarson, A. K. Daly, S. Garte, D. W. Nebert, Human cytochrome P450 (CYP) alleles a web page for the nomenclature of alleles [http://www.imm.ki.se/CYPalleles/] (accessed 22.11.2005). [5] Z. Desta, X. Zhao, J.G. Shin, D.A. Flockhart, Clinical significance of the cytochrome P450 2C19 genetic polymorphism, Clin. Pharmaco- kinet 41 (2002) 913–958. [6] C.R. Kumana, I.J. Lauder, M. Chan, W. Ko, H.J. Lin, Differences in diazepam pharmacokinetics in Chinese and white Caucasians-relation to body lipid stores, Eur. J. Clin. Pharmacol. 32 (1987) 211–215. [7] A.K. Daly, Pharmacogenetics of the cytochromes P450, Curr. Top Med. Chem. 4 (2004) 1733–1744. [8] K.C. Chou, Review: structural bioinformatics and its impact to biomedical science, Curr. Med. Chem. 11 (2004) 2105–2134. [9] M.R. Wester, J.K. Yano, G.A. Schoch, C. Yang, K.J. Griffin, C.D. Stout, E.F. Johnson, The structure of human cytochrome P450 2C9 complexed with flurbiprofen at 2.0 Å resolution, J. Bio. Chem. 279 (2004) 35630–35637. [10] M.O. Tanira, M.K. Al-Mukhaini, A.T. Al-Hinai, K.A. Al-Balushi, I.S. Ahmed, Frequency of CYP2C9 genotypes among omani patients receiving warfarin and its correlation with warfarin dose, Community Genet. 10 (2007) 32–37. [11] M. Claessens, E.V. Cutsem, I. Lasters, S. Wodak, Modeling the polypeptide backbone with ‘spare parts’ from known protein struc- tures, Protein Eng. 4 (1989) 335–345. [12] T.A. Jones, S. Thirup, Using known substructures in protein model building and crystallography, EMBO J. 5 (1986) 819–822. [13] T.L. Blundell, B.L. Sibanda, M.J.E. Sternberg, J.M. Thornton, Knowledge-based prediction of protein structures and design of novel molecules, Nature (London) 326 (1987) 347–352. [14] K.C. Chou, G. Nemethy, M. Pottle, H.A. Scheraga, Energy of stabilization of the right-handed beta-alpha-beta crossover in pro- teins, J. Mol. Biol. 205 (1989) 241–249. [15] T. Fechteler, U. Dengler, D. Schomburg, Predition of protein three- dimensional structures in insertion and deletion regions: a procedure for searching data bases of representative protein fragments using geometric scoring criteria, J. Mol. Biol. 253 (1995) 114–131. [16] K.C. Chou, D. Jones, R.L. Heinrikson, Prediction of the tertiary structure and substrate binding site of caspase-8, FEBS Lett. 419 (1997) 49–54. [17] J. Rotonda, D.W. Nicholson, K.M. Fazil, M. Gallant, Y. Gareau, M. Labelle, E.P. Peterson, D.M. Rasper, R. Ruel, J.P. Vaillancourt, N.A. Thornberry, J.W. Becker, The three-dimensional structure of apo- pain/CPP32, a key mediator of apoptosis, Nat. Struct. Biol. 3 (1996) 619–625. [18] W. Watt, K.A. Koeplinger, A.M. Mildner, R.L. Heinrikson, A.G. Tomasselli, K.D. Watenpaugh, The atomic resolution structure of human caspase-8, a key activator of apoptosis, Structure 7 (1999) 1135–1143. [19] J.J. Chou, H. Matsou, J. Duan, G. Wagner, Solution structure of the RAIDD CARD and model for CARD/CARD interaction in caspase- 2 and caspase-9 recruitment, Cell 94 (1998) 171–180. J.-F. Wang et al. / Biochemical and Biophysical Research Communications 355 (2007) 513–519 519[20] K.C. Chou, K.D. Watenpaugh, R.L. Heinrikson, A model of the complex between cyclin-dependent kinase 5( Cdk5) and the activation domain of neuronal Cdk5 activator, Biochem. Biophys. Res. Com- mun. 259 (1999) 420–428. [21] K.C. Chou, A.G. Tomasselli, R.L. Heinrikson, Prediction of the tertiary structure of a caspase-9/inhibitor complex, FEBS Lett. 470 (2000) 249–256. [22] C. Tarricone, R. Dhavan, J. Peng, L.B. Arecces, L.H. Tsai, A. Musacchio, Structure and regulation of the Cdk5-p25 (Nck5a) complex, Mol. Cell 8 (2001) 657–669. [23] J. Zhang, C.H. Luan, K.C. Chou, G.V.W. Johnson, Identification of the N-terminal functional domains of Cdk5 by molecular truncation and computer modeling, Proteins Struct. Funct. Genet. 48 (2002) 447–453. [24] K.C. Chou, W.J. Howe, Prediction of the tertiary structure of the beta-secretase zymogen, Biochem. Biophys. Res. Commun. 292 (2002) 702–708. [25] X.P. Shi, E. Chen, K.C. Yin, S. Na, V.M. Garsky, M.T. Lai, Y.M. Li, M. Platchek, R.B. Register, M.K. Sardana, M.J. Tam, J. Thiebeau, T. Wood, J.A. Shafer, S.J. Gardell, The pro domain of beta-secretase does not confer strict zymogen-like properties but does assist proper folding of the protease domain, J. Biol. Chem. 276 (2001) 10336– 10373. [26] S. Benjannet, A. Elagoz, L. Wickham, M. Mamarbachi, J.S. Munzer, A. Basak, C. Lazure, J.A. Cromlish, S. Sisodia, F. Checler, M. Chretien, N.G. Seidah, Post-translational processing of beta-secre- tase-amylodi-converting enzyme and its ectodomain shedding, J. Biol. Chem. 276 (2001) 10879–10887. [27] K.C. Chou, Modeling extracellular domains of GABA-A receptors: subtypes 1, 2, 3 and 5, Biochem. Biophys. Res. Commun. 316 (2004) 636–642. [28] K.C. Chou, Insights from modelling the tertiary structure of BACE2, J. Proteome Res. 3 (2004) 1069–1072. [29] D.Q. Wei, Q.S. Du, H. Sun, K.C. Chou, Insights from modeling the 3D structure of H5N1 influenza virus neuraminidase and its binding interactions with ligands, Biochem. Biophys. Res. Commun. 344 (2006) 1048–1055. [30] K.C. Chou, Molecular therapeutic target for type-2 diabetes, J. Proteome Res. 3 (2004) 1284–1288. [31] K.C. Chou, Modeling the tertiary structure of human cathepsin-E, Biochem. Biophys. Res. Commun. 331 (2005) 56–60.[32] K.C. Chou, Insights from modeling the 3D structure of the extracellular domain of alpha 7 nicotinic acetylcholine receptor, Biochem. Biophys. Res. Commun. 319 (2004) 433–438. [33] K.C. Chou, Insights from modeling three-dimensional structures of the human potassium and sodium channels, J. Proteome Res. 3 (2004) 856–861. [34] I.D. Kuntz, E.C. Meng, B.K. Shoichet, Structure-based strategies for drug design and discovery, Acc. Chem. Res. 27 (1994) 117–123. [35] G.M. Morris, D.S. Goodesll, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, Automated docking using a Lamarckian Genetic Algorithm and empirical binding free energy function, J. Comput. Chem. 19 (1998) 1639–1662. [36] K.C. Chou, Low-frequency resonance and cooperativity of hemoglo- bin, Trends Biochem. Sci. 14 (1989) 212. [37] K.C. Chou, Review: Low-frequency collective motion in biomacro- molecules and its biological functions, Biophys. Chem. 30 (1988) 3– 48. [38] K.C. Chou, L. Carlacci, Simulated annealing approach to the study of protein structures, Protein Eng. 4 (1991) 661–667. [39] F. Glover, Tabu search: a tutorial, Interfaces 20 (1990) 74–94. [40] A.D. MacKerell Jr., D. Bashford, M. Bellott, R.L. Dunbrack Jr., J.D. Evanseck, M.J. Field, S. Fischer, J. Gao, H. Guo, S. Ha, D. Joseph- McCarthy, L. Kuchnir, K. Kuczear, F.T.K. Lau, C. Mattos, S. Michnick, T. Ngo, D.T. Nguyen, B. Prodhom, W.E. Reither III, B. Roux, M. Schlenkrich, J.C. Smith, R. Stoe, J. Straub, M. Watanabe, J. Wiorkiewicz-Kuczera, D. Yin, M. Karplus, All-atom empirical potential for molecular modeling and dynamics studies of proteins, J. Phys. Chem. B 102 (1998) 3586–3616. [41] G.M. Morris, D.S. Goodsell, R. Huey, A.J. Olson, Distributed automated docking of flexible ligands to proteins: parallel applications of AutoDock 2.4, J. Comput. Aided Mol. Des. 10 (1996) 293–304. [42] D.Q. Wei, S. Sirois, Q.S. Du, H.R. Arias, K.C. Chou, Theoretical studies of Alzheimer’s disease drug candidate [(2,4-dimethoxy) benzylidene]-anabaseine dihydrochloride (GTS-21) and its deriva- tives, Biochem. Biophys. Res. Commun. 338 (2005) 1059–1064. [43] D.Q. Wei, R. Zhang, Q.S. Du, W.N. Gao, Y. Li, H. Gao, S.Q. Wang, X. Zhang, A.X. Li, S. Sirois, K.C. Chou, Anti-SARS drug screening by molecular docking, Amino Acids 31 (2006) 73–80. [44] K.C. Chou, D.Q. Wei, Q.S. Du, S. Sirois, W.Z. Zhong, Review: Progress in computational approach to drug development against SARS, Curr. Med. Chem. 13 (2006) 3263–3270.