How is a co-methyl intermediate formed in the reaction of cobalamin-dependent methionine synthase? Theoretical evidence for a two-step methyl cation transfer mechanism.

A methyl-Co(cobalamin) species has been characterized to be a crucial intermediate in the last step of the de novo biosynthesis of methionine catalyzed by cobalamin-dependent methionine synthase (MetH). However, exactly how it is formed is still an open question. In the present article, the formation of the methyl-Co(cobalamin) species in MetH has been investigated with B3LYP* hybrid DFT including van der Waals (vdW) interactions (i.e., dispersion) and using a chemical model built on X-ray crystal structures. The methyl cation and radical transfer mechanisms have been examined in various protonation states. The calculations reveal that the CH(3)-Co(III)(cobalamin) formation in MetH proceeds along a stepwise pathway, where the first step is a methyl cation transfer from the protonated methyltetrahydrofolate (CH(3)-THF) substrate to the Co(I)cobalamin. The second step is a binding of His759 to the other side (α-face) of Co. The former methyl transfer is computed to be the rate-limiting step with a barrier of 18 kcal/mol, which is reduced to 13 kcal/mol when dispersion is included. For the first step, the protonation at the methyl-bound nitrogen of CH(3)-THF is very important. The methyl transfer is otherwise unreachable with a very high barrier of ~38 kcal/mol. The deprotonation of the α-face His759-Asp757-Ser810 triad is found to be much less significant but slightly facilitates the CH(3)-Co(III)Cbl formation. There has been a long-standing discrepancy of 10-20 kcal/mol between theory and experiment in previous B3LYP computations of the Co-C bond dissociation energy for the methyl-Co(cobalamin) species. The calculations indicate that the lack of dispersion (~11 kcal/mol) is the main origin of this puzzling problem. With these effects, B3LYP* gives a bond strength of 32 kcal/mol compared to the experimental value of 37 ± 3 kcal/mol. Overall, the present calculations give many examples of dispersion that makes non-negligible contributions to the energetics of enzyme reactions, especially for systems involving at least one large reacting fragment approaching or departing.