Catalysis in Glycine N-Methyltransferase: Testing the Electrostatic Stabilization and Compression Hypothesis ² Alejandro Soriano, ‡ Raquel Castillo, Christo Christov, § Juan Andre ´s, and Vicente Moliner* Departament de Cie ` ncies Experimentals, UniVersitat Jaume I, 12071 Castello ´ n, Spain In ˜aki Tun ˜o ´n* Departament de Quı ´mica Fı ´sica, UniVersidad de Valencia, 46100 Burjasot, Valencia, Spain ReceiVed June 30, 2006; ReVised Manuscript ReceiVed September 20, 2006 ABSTRACT: Glycine N-methyltransferase (GNMT) is an S-adenosyl-L-methionine dependent enzyme that catalyzes glycine transformation to sarcosine. Here, we present a hybrid quantum mechanics/molecular mechanics (QM/MM) computational study of the reaction compared to the counterpart process in water. The process takes place through an S N 2 mechanism in both media with a transition state in which the transferring methyl group is placed in between the donor (SAM) and the acceptor (the amine group of glycine). Comparative analysis of structural, electrostatic, and electronic characteristics of the in-solution and enzymatic transition states allows us to get a deeper insight into the origins of the enzyme’s catalytic power. We found that the enzyme is able to stabilize the substrate in its more active basic form by means of a positively charged residue (Arg175) placed in the active site. However, the maximum stabilization is attained for the transition state. In this case, the enzyme is able to form stronger hydrogen bonds with the positively charged amine group. Finally, we show that in agreement with previous computational studies on other methyltransferases, there is no computational evidence for the compression hypothesis, as was formulated by Schowen (Hegazi, M. F., Borchardt, R. T., and Schowen, R. L. (1979) J. Am. Chem. Soc. 101, 4359-4365). Enzymes are able to speed up chemical reactions by several orders of magnitude, making metabolic transforma- tions fast enough to be compatible with life (1). In fact, enzymes are amazing catalysts, not only because of their catalytic power but also because of their specificity and selectivity. The origin of this catalytic efficiency is still a controversial topic (2-5). In general, two different theories, with several variants, have been elaborated to explain enzymatic activity. The first, stresses the effect of the enzyme on the reaction transition state (TS 1 ). According to TS theories, energy barrier reduction is attained by means of TS stabilization relative to the uncatalyzed process, the counterpart reaction in aqueous solution (6-8). This stabi- lization is mainly done by electrostatic interactions that take place in the enzyme but not in solution. The second theory explains the energy barrier reduction on the basis of a reactant state (RS) destabilization (9-13). An explanation based on the enhanced formation of an especially reactive conforma- tion (near attack conformations or NACs) (13) in the enzyme can be interpreted as an example of these types of theories. Recently, some of us proposed that RS destabilization could be related to TS stabilization, considering that an active site complementary to the TS may favor the reactant conforma- tion electronically and geometrically closer to the TS (14). Similarly, the work of Garcı ´a-Viloca et al. (3) shows and discusses that the two concepts are physically equivalent within the limits of transition state theory (TST) and amount only to two deceptively different descriptions of TS stabiliza- tion. The analysis of an increasing number of enzymatic systems is helpful to highlight this fundamental problem, although new challenges, such as the understanding of the coupling between reaction and protein dynamics (15, 16), or the fact that TST may be inadequate to account for some catalytic effects (2), are emerging and requiring a more complex formulation. ² We are indebted to DGI for project BQU2003-04168, BANCAIXA for project P1-1B2002-02, P1-1B2005-13, and P1-1B2005-15, and Generalitat Valenciana for projects GV06-021, GV04B-131, GV06/ 152, GV06/016, ACOMP06/122, and GRUPOS04/08, which supported this research. A.S. is thankful for the Decom Valencia fellowship. The Marie Curie Development Host Fellowship program supported the work of C.C. (contract no. HPMD-CT-2000-00055). The authors are solely responsible for the information communicated, and it does not represent the opinion of The European Community. The European Community is not responsible for any use that might be made of data appearing herein. C.C. also acknowledges grant CTESIN/2004019 from Gener- alitat Valenciana, grant RIG 981486 from NATO, and grant YS-CH- 1202 from the Bulgarian Science Fund. * To whom correspondence should be addressed. Tel: 34-964-728- 084. Fax: 34-964-728-066. E-mail: moliner@uji.es (V.M.); Tel: 34- 963-544-880. Fax: 34-963-544-564. E-mail: tunon@uv.es (I.T.). ‡ Current address: Servei de Informa ´tica, Universidad de Valencia. § Current address: School of Chemistry, University of Bristol. On leave from Institute of Organic Chemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria. 1 Abbreviations: GNMT, glycine N-methyl transferase; QM/MM, quantum mechanics/molecular mechanics; SAM, S-adenosyl-methion- ine; SAH, S-adenosyl-L-homocysteine; TS, transition state; RS, reactant state; TST, transition state theory; PMF, potential of mean force; WHAM, weighted hystogram analysis method; ZW, zwitterionic form of glycine; BA, basic form of glycine; NE, neutral form of glycine; AC, acid form of glycine; KIE, kinetic isotope effect. 14917 Biochemistry 2006, 45, 14917-14925 10.1021/bi061319k CCC: $33.50 © 2006 American Chemical Society Published on Web 11/23/2006