Ab Initio Molecular Dynamics Simulation and Energetics of the Ribulose-1,5-biphosphate Carboxylation Reaction Catalyzed by Rubisco: Toward Elucidating the Stereospecific Protonation Mechanism

In the carboxylation reaction catalyzed by ribulose 1,5-bisphosphate (RuBP) carboxylase–oxygenase (Rubisco), which is fundamental to photosynthesis, scission of a C–C bond in the six-carbon gemdiolate intermediate forms a carbanion that must be protonated stereospecifically to form product. It is thought that a conserved lysine side chain (LYS175 in spinach Rubisco), in the immediate vicinity of the carbanion, provides the necessary proton. Here, we endeavor to determine from the electronic-structure calculations whether protonation via this route is energetically possible. The two-dimensional energy surface was mapped to determine the minimum energy path (MEP) using density functional theory (B3LYP) and incorporating basis set superposition and classical (London) dispersion corrections. The potential of mean force (free energy) was then calculated from ab initio molecular dynamics simulations with umbrella sampling in the vicinity of the MEP on the scission–protonation reaction coordinate. MEP calculations were also carried out to evaluate the possibility of an active-site water near the phosphate (P1) of RuBP, with an excess proton positioned at P1, as an alternative facilitator of stereospecific protonation via a classical Grotthuss mechanism. In both cases, the C–C bond scission in the six-carbon intermediate and proton transfer from the donor was found to be concerted and highly asynchronous, without a stable carbanion intermediate. However, the free energy change was unfavorable for direct protonation by the LYS175 side chain. In contrast, the Grotthuss mechanism yielded stable products and an activation energy in good agreement with experiment. It also provides a plausible mechanism for alternative product formed in enzyme mutations at the LYS175 position and is consistent with the observed deuterium isotope effects.This research was undertaken with the assistance of resources and services from the National Computational Infrastructure (NCI), which is supported by the Australian Governmen