Genomes to Life Contractor-Grantee Workshop III
February 6-9, 2005, Washington, D.C.
Genomics:GTL Program Projects
Sandia National Laboratories
27
Modeling RuBisCO’s Gating Mechanism Using Targeted Molecular Dynamics
Paul S. Crozier1 (pscrozi@sandia.gov), Steven J. Plimpton1, Mark D. Rintoul1*, Christian Burisch2, and Jürgen Schlitter2
1Sandia National Laboratories, Albuquerque, NM and 2Ruhr-Universität Bochum, Bochum, Germany
RuBisCO is the enzymatic bottleneck of carbon sequestration in Synechococcus, which is partly due to its catalysis of a competing oxygenase reaction that limits its specificity and efficiency. The binding niche residues are highly conserved across RuBisCO species, yet experimentally-measured specificities and carboxylation rates vary widely. The residues that make up the gate to the binding niche affect the gate’s opening and closing rates, and in turn, RuBisCO specificities and reaction rates. We have performed molecular dynamics (MD) simulations of RuBisCO’s gating mechanism to gain insight into how residue-level changes in RuBisCO’s primary sequence affect enzyme performance.
Traditional MD is currently limited to the sub-microsecond timescale, but hardware and algorithm improvements continue to push the attainable timescales upward. We have recently developed several upgrades for our open-source parallel MD simulation package, LAMMPS (http://www.cs.sandia.gov/~sjplimp/lammps.html), which essentially double the algorithm’s performance on typical biomolecular simulations. Performance enhancements have come through implementation of the rRESPA hierarchical time-stepping method and a high-performance tabulation algorithm for rapid evaluation of CPU-intensive coulombic interatomic forces. The LAMMPS simulation package was officially released as an open-source parallel MD code available for download on the first of September. Since then, it has been downloaded 1,075 times.
In addition to improving the traditional MD capabilities in LAMMPS, we have implemented advanced MD methods that allow simulation of events that occur on much longer timescales. One such method, targeted molecular dynamics (TMD), allows simulation of user-specified transition events, like the RuBisCO gating mechanism, by imposing a dynamic holonomic constraint on the macromolecular complex. TMD yields the free energy profile of the transition event, which is related to the rate of the transition event.
We performed TMD simulations of the gating event of spinach and Synechococcus RuBisCO, each for WT and for mutant D473A. Our simple implicit solvent reduced-model predictions of gating free energy profiles have been encouraging since they have demonstrated the ability to discriminate between RuBisCO structural differences, and are in qualitative agreement with expected trends. For example, our TMD prediction shows a much higher gate opening barrier for Synechococcus than for spinach, which indicates more time in the closed state, more photorespiration, and lower specificity for Synechococcus RuBisCO. This is in qualitative agreement with the experimentally-measured specificities of Synechococcus RuBisCO (47) and spinach RuBisCO (92). Likewise, D473A mutations performed in silico for both RuBisCO species show a much lower free energy barrier for gate opening than do wild type RuBisCOs. Experiments show that D473A mutants are not catalytically competent, which is probably due to the fact that the binding niche gate can not properly close (and rapidly opens), without the D473 – R134 salt bridge.
This project is supported by the U.S.Department of Energy’s Genomics:GTL Program under project “Carbon Sequestration in Synechococcus sp: From Molecular Machines to Hierarchical Modeling” (http://www.genomes2life.org/). Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000.
* Presenting author
