Computational Biology and Chemistry 34 (2010) 226–231 Contents lists available at ScienceDirect Computational Biology and Chemistry journal homepage: www.elsevier.com/locate/compbiolchem Research article Homology modeling, binding site identification and docking in flavone hydroxylase CYP105P2 in Streptomyces peucetius ATCC 27952 Bashistha Kumar Kanth, Kwangkyoung Liou, Jae Kyung Sohng ∗ Institute of Biomolecule Reconstruction (iBR), Department of Pharmaceutical Engineering, Sun Moon University, #100, Kalsan-ri, Tangjeong-myeon, Asansi, Chungnam 336-708, Republic of Korea article info Article history: Received 21 June 2010 Received in revised form 11 August 2010 Accepted 18 August 2010 Keywords: Cytochrome P450 Homology modeling Molecular dynamics simulation Molecular docking Flavone Ligand-binding site abstract Homology models of cytochrome P450 105P2 (CYP105P2) were constructed using four P450 structures, CYP105A1, CYP105, CYP165B3 and CYP107L1, as templates for the model building. Using Accelrys Discov- ery Studio 2.1 software, the lowest energy CYP105P2 model was then assessed for stereochemical quality and side-chain environment. Further active site optimization of the CYP105P2 model built using these templates was performed by molecular dynamics to generate the final CYP105P2 model. The substrates, flavone, flavanone, quercetin and naringenin, were docked into the model. The model-flavone complex was used to validate the active site architecture, and structurally and functionally important residues were identified by subsequent characterization of the secondary structure. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction Cytochrome P450s (CYPs) are well-known for their monooxy- genase reactions. These are heme-containing enzymes and are involved in phase I metabolism. They comprise a superfamily of enzymes that are mainly responsible for the hydroxylation of a wide variety of hydrophobic compounds (Nelson et al., 1996). They also catalyze reactions such as N-oxidation, N-, O- and S-dealkylation, sulfoxidation, epoxidation, peroxidation, deamination, desulfuration, dehalogenation and N-oxide reduction (Bernhardt, 1996). In addition, they participate in a wide variety of reactions in oxidative, peroxidative and reductive metabolic bio- transformation and metabolize endogenous compounds, such as fatty acids and prostaglandins, and exogenous compounds, such as therapeutic drugs and xenobiotics (Ortiz de Montellano, 2004; Nelson et al., 1996). Streptomyces peucetius ATCC 27952 possesses 23 cytochrome P450s, 6 ferredoxins (FDXs) and 7 ferredoxin reductases (FDRs). The expression of a P450 gene with high catalytic activity in a heterologous host is a challenging task that will lead to efficient bio- transformation and biodegradation. In cytochrome P450 research, a long-sought-after practical goal is to capitalize on the specificity of these enzymes in regio- and stereoselective reactions for the production of chemicals that are difficult to prepare by traditional ∗ Corresponding author. Tel.: +82 41 530 2246; fax: +82 41 544 2919. E-mail address: sohng@sunmoon.ac.kr (J.K. Sohng). organic synthesis methods. However, the prerequisite for rational engineering of enzymes is the knowledge of their structure. Many attempts to crystallize cytochrome P450 have been unsuccessful; thus, a homology model can be created for deeper insights into the structure and function and later for the rational design of mutants to directly influence the specific reaction as desired. In recent years, homology modeling has become a promising tool with which to study cytochrome P450 function. Whereas the number of modeled bacterial cytochrome P450 structures is rather low, a cytochrome P450 homology model for CYP105P2 has been constructed, and the resulting structures have been successfully evaluated. 2. Materials and methods 2.1. Construction of the homology model Homology modeling was used to build the model for CYP105P2. The protein is 399 amino acids long and the GenBank accession number is CAE53708. The Discovery Studio v 2.1 (2008) (DS 2.1) was used for homology model construction (Accelrys, San Diego, USA). Sequence analysis identified homologs for CYP105P2 pro- tein sequences by searching either the NCBI (The National Center for Biotechnology Information) website or PSI-BLAST (compari- son matrix, BLOSUM62; E-threshold, 10) using The ExPASy (2010) server. Four templates were selected for the model building. Pro- tein data bank (PDB) identifications of the templates are the 2ZBY, 2Z36, 1LFK, and 2C6H crystal structures of CYP105A1 (Sugimoto et al., 2008), CYP105 (Yasutake et al., 2007), CYP165B3 (Zerbe et al., 1476-9271/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.compbiolchem.2010.08.002