Nam Kwangho, Cui Qiang, Gao Jiali, York Darrin M
Department of Chemistry and Supercomputing Institute and the Digital Technology Center, University of Minnesota, Minneapolis, Minnesota 55455-0431, and Department of Chemistry and Theoretical Chemistry Institute, University of Wisconsin, 1101 University Avenue, Madison, Wisconsin 53706.
J Chem Theory Comput. 2007 Mar;3(2):486-504. doi: 10.1021/ct6002466.
A semiempirical AM1/d Hamiltonian is developed to model phosphoryl transfer reactions catalyzed by enzymes and ribozymes for use in linear-scaling calculations and combined quantum mechanical/molecular mechanical simulations. The model, designated AM1/d-PhoT, is parametrized for H, O, and P atoms to reproduce high-level density-functional results from a recently constructed database of quantum calculations for RNA catalysis ( http://theory.chem.umn.edu/Database/QCRNA ), including geometries and relative energies of minima, transition states and reactive intermediates, dipole moments, proton affinities, and other relevant properties. The model is tested in the gas phase and in solution using a QM/MM potential. The results indicate that the method provides significantly higher accuracy than MNDO/d, AM1, and PM3 methods and, for the transphosphorylation reactions, is in close agreement with the density-functional calculations at the B3LYP/6-311++G(3df,2p) level with a reduction in computational cost of 3-4 orders of magnitude. The model is expected to have considerable impact on the application of semiempirical QM/MM methods to transphosphorylation reactions in solution, enzymes, and ribozymes and to ultimately facilitate the design of improved next-generation multiscale quantum models.
开发了一种半经验的AM1/d哈密顿量,用于模拟酶和核酶催化的磷酰基转移反应,以用于线性标度计算和量子力学/分子力学联合模拟。该模型命名为AM1/d-PhoT,对氢、氧和磷原子进行了参数化,以重现最近构建的RNA催化量子计算数据库(http://theory.chem.umn.edu/Database/QCRNA)中的高水平密度泛函结果,包括最小值、过渡态和反应中间体的几何结构和相对能量、偶极矩、质子亲和势及其他相关性质。该模型在气相和溶液中使用QM/MM势进行了测试。结果表明,该方法比MNDO/d、AM1和PM3方法具有显著更高的精度,对于转磷酸化反应,与B3LYP/6-311++G(3df,2p)水平的密度泛函计算结果非常吻合,同时计算成本降低了3-4个数量级。预计该模型将对半经验QM/MM方法在溶液、酶和核酶中转磷酸化反应的应用产生重大影响,并最终有助于设计改进的下一代多尺度量子模型。