We present a new more general way to combine ab initio quantum mechanical calculations with classical mechanical free energy perturbation approach to calculate the energetics of enzyme-catalyzed reactions and the same reaction in solution. This approach, which enables enzyme and solution reactions to be compared without the use of empirical parameters, is applied to the formation of the tetrahedral intermediate in trypsin, but it should be generally applicable to any enzymatic reaction. Critical to the accurate calculation of the reaction energetics in solution is the estimate of the free energy to assemble the reacting groups, where the approach recently published by Hermans and Wang (J. Am. Chem. Soc. 1997, 119, 2707) was used. A central aspect of this new approach is the use of the RESP protocol to calculate the charge distribution of structures along the reaction pathway, which enables us to circumvent problems in partitioning the charge across a residue that is being divided into QM and MM parts. The classical mechanical free energy calculations are implemented with two different approaches, “Cartesian mapping” and “flexible FEP”. The similarity of the results found by using these two approaches supports the robustness of the calculated free energies. The calculated free energies are in quite good agreement with available experimental data for the activation free energies in the enzyme and aqueous phase reactions.

How do enzymes achieve very large rate enhancements compared to corresponding uncatalyzed reactions in solution? We present a computational approach which combines high-level ab initio quantum mechanical calculations with classical free energy calculations to address this question. Our calculations lead to accurate estimates of DeltaG for both trypsin and catechol O-methyltransferase-catalyzed and reference uncatalyzed reactions and give new insights into the nature of enzyme catalysis. The same methodology applied to steps in the catalytic mechanism of citrate synthase further supports the conclusion that one need not invoke special concepts such as "low-barrier hydrogen bonds" or "pK(a) matching" to explain enzyme catalysis.

No abstract

The human constitutive androstane receptor (CAR, NR1I3) is an important ligand-activated regulator of oxidative and conjugative enzymes and transport proteins. Because of the lack of a crystal structure of the ligand-binding domain (LBD), wide species differences in ligand specificity and the scarcity of well characterized ligands, the factors that determine CAR ligand specificity are not clear. To address this issue, we developed highly defined homology models of human CAR LBD to identify residues lining the ligand-binding pocket and to perform molecular dynamics simulations with known human CAR modulators. The roles of 22 LBD residues for basal activity, ligand selectivity, and interactions with co-regulators were studied using sitedirected mutagenesis, mammalian co-transfection, and yeast two-hybrid assays.

Parameterization and test calculations of a reduced protein model with new energy terms are presented. The new energy terms retain the steric properties and the most significant degrees of freedom of protein side chains in an efficient way using only one to three virtual atoms per amino acid residue. The energy terms are implemented in a force field containing predefined secondary structure elements as constraints, electrostatic interaction terms, and a solvent-accessible surface area term to include the effect of solvation. In the force field the main-chain peptide units are modeled as electric dipoles, which have constant directions in α-helices and β-sheets and variable conformation-dependent directions in loops. Protein secondary structures can be readily modeled using these dipole terms. Parameters of the force field were derived using a large set of experimental protein structures and refined by minimizing RMS errors between the experimental structures and structures generated using molecular dynamics simulations. The final average RMS error was 3.7 Å for the main-chain virtual atoms (C α atoms) and 4.2 Å for all virtual atoms for a test set of 10 proteins with 58-294 amino acid residues. The force field was further tested with a substantially larger test set of 608 proteins yielding somewhat lower accuracy. The fold recognition capabilities of the force field were also evaluated using a set of 27,814 misfolded decoy structures.