DOI: 10.1002/jcc.21895 Zinc–Homocysteine Binding in Cobalamin-Dependent Methionine Synthase and its Role in the Substrate Activation: DFT, ONIOM, and QM/MM Molecular Dynamics Studies Safwat Abdel-Azeim, [a] Xin Li, [a] Lung Wa Chung, [a] and Keiji Morokuma [a,b]* Cobalamin-dependent methionine synthase (MetH) is an important metalloenzyme responsible for the biosynthesis of methionine. It catalyzes methyl transfer from N 5 -methyl- tetrahydrofolate to homocysteine (Hcy) by using a zinc ion to activate the Hcy substrate. Density functional theory (B3LYP) calculations on the active-site model in gas phase and in a polarized continuum model were performed to study the Zn coordination changes from the substrate-unbound state to the substrate-bound state. The protein effect on the Zn 2þ coordination exchange was further investigated by ONIOM (B3LYP:AMBER)-ME and EE calculations. The Zn 2þ -coordination exchange is found to be highly unfavorable in the gas phase with a high barrier and endothermicity. In the water solution, the reaction becomes exothermic and the reaction barrier is drastically decreased to about 10.0 kcal/mol. A considerable protein effect on the coordination exchange was also found; the reaction is even more exothermic and occurs without barrier. The enzyme was suggested to constrain the zinc coordination sphere in the reactant state (Hcy-unbound state) more than that in the product state (Hcy-bound state), which promotes ligation of the Hcy substrate. Molecular dynamics simulations using molecular mechanics (MM) and PM3/MM potentials suggest a correlation between the flexibility of the Zn 2þ -binding site and regulation of the enzyme function. Directed in silico mutations of selected residues in the active site were also performed. Our studies support a dissociative mechanism starting with the ZnAO (Asn234) bond breaking followed by the ZnAS (Hcy) bond formation; the proposed associative mechanism for the Zn 2þ - coordination exchange is not supported. V C 2011 Wiley Periodicals, Inc. J Comput Chem 32: 3154–3167, 2011 Keywords: ONIOM methionine synthase cobalamin-dependent homocysteine QM/MM MD Introduction Cobalamin-dependent methionine synthase (MetH) is a large and flexible multidomain metalloenzyme responsible for the biosynthesis of methionine. It catalyzes methyl transfer from N 5 -methyltetrahydrofolate to homocysteine (Hcy) using cobala- min as a methyl carrier (Fig. 1). [1] MetH is a modular protein containing different domains which are implicated for the binding and the activation of different substrates and cofac- tors involved in the catalytic cycle. For example, in Escherichia coli MetH, amino acids 2353 are responsible for binding and activation of Hcy; amino acids 345649 are involved in the binding and the activation of N 5 -methyltetrahydrofolate; amino acids 650896 are responsible for binding of methylcobala- min; and amino acids 8971227 are involved in binding of adenosylmethionine and are also required for reductive activa- tion of the cob(II)alamin state. [2] MetH uses a zinc ion to bind and activate the Hcy substrate. The extended X-ray absorption fine structure spectra of native MetH confirmed that Zn 2þ - coordination changes from ZnS 3 (O/N) to ZnS 4 on the substrate binding. [3,4] Also, zinc is found to be in the þ2 oxidation state, Zn(II), with a d 10 electron configuration, which makes the metal able to adopt a variety of coordination numbers ranging from 3 to 8 with very little energy differences. Metals with a filled d shell are not subjected to geometry-dependent ligand field stabilization effects, but the selectivity and the specificity of its complexes depend on the size and charge of the ligands. [5] Furthermore, binding of Hcy to MetH was accompa- nied by the release of a proton (0.9 H þ equiv of MetH at pH 7.9), consistent with the binding of Hcy as a thiolate and the role of Zn 2þ as a Lewis acid catalyst to lower pK a of Hcy. [6] On the basis of the recent available X-ray crystal structures and biophysical studies on MetH and cobalamin-independent methionine synthase (MetE), Koutmos et al. [7] found an inver- sion of the metal geometry and large displacement from the oxygen–donor ligand (Asn234 in MetH) or (Glu642 in MetE) to Hcy (R1 ! P , see Scheme 1). They also proposed that this [a] S. Abdel-Azeim, X. Li, L. W. Chung, K. Morokuma Fukui Institute for Fundamental Chemistry, Kyoto University, 34-4 Takano Nishihiraki-cho, Sakyo-ku, Kyoto 606-8103, Japan E-mail: morokuma@fukui.kyoto-u.ac.jp [b] K. Morokuma Department of Chemistry and Cherry L. Emerson Center for Scientific Computation, Emory University, Atlanta, Georgia 30322 Contributions: S. Abdel-Azeim, L. W. Chung, and K. Morokuma design the project and writing the manuscript; S. Abdel-Azeim and X. Li perform the calculations and analyze the results. Grant sponsor: Japan Science and Technology Agency [JST; with a Core Research for Evolutional Science and Technology (CREST) grant in the area of high performance computing for multiscale and multiphysics phenomena]. Additional Supporting Information may be found in the online version of this article. Journal of Computational Chemistry V C 2011 Wiley Periodicals, Inc. 3154 ORIGINAL ARTICLES