%0 Generic %A Schwarzl, Sonja M. %D 2005 %F heidok:6389 %K QM/MM , Reaktionsweg , Ladungsverteilung , chemomechanische KopplungQM/MM , reaction path , charge distribution , chemomechanical coupling %R 10.11588/heidok.00006389 %T Understanding the ATP hydrolysis mechanism in myosin using computer simulation techniques %U https://archiv.ub.uni-heidelberg.de/volltextserver/6389/ %X Molecular motors are proteins that convert energy from nucleoside triphosphate hydrolysis into mechanical work. A prominent example is myosin which drives muscle contraction and a large number of additional cellular transport phenomena in all living organisms. While hydrolyzing ATP, myosin translocates along an actin filament. The catalytic cycle for ATP hydrolysis and the mechanical motor cycle are closely coupled. Although a large number of studies have been devoted to understanding the functioning of myosin since its isolation in the 19th century, the details of the chemical mechanism underlying ATP hydrolysis and its coupling to the necessary conformational changes of myosin are poorly understood. In this thesis, theoretical methods are developed and used to gain a detailed understanding of the mechanism of ATP hydrolysis in myosin and of mechanical events that immediately follow hydrolysis. Three different possible reaction routes are investigated using combined quantum mechanical and molecular mechanical (QM/MM) reaction path simulations. To include solvent screening effects in the calculations, a new approximate method "Non-Uniform Charge Scaling" (NUCS) was developed which scales the partial atomic charges on the molecular mechanical atoms so as to optimally reproduce electrostatic interaction energies between groups of protein atoms and the QM region as determined from an initial continuum solvent analysis with a simple Coulomb potential and scaled charges. NUCS is a generally-applicable method that is particularly useful in cases where an explicit treatment of water molecules is not feasible and interfaces to implicit solvent models are lacking, as is the case for current QM/MM calculations. Path optimizations were done using Hartree-Fock calculations with 3-21G(d) and 6-31G(d,p) basis sets, followed by point energy calls using density-functional theory B3LYP/6-31+G(d,p). Despite the inaccuracies inherent in this method, the present calculations currently represent the most accurate QM/MM theoretical investigation of an enzyme-catalyzed phosphoanhydride hydrolysis reaction. Possible methodological improvements for future investigations are discussed. The three pathways studied are isoenergetic within error and are thus equally likely to be populated. The 6-31G(d,p) basis set proved to be reliable in describing the geometries during the phosphate hydrolysis reactions, whereas the 3-21G(d) basis set was found to be too inaccurate. Although the energies were not sufficiently accurate, a number of structural conclusions on the mechanism of ATP hydrolysis can be drawn and related to experimental findings from isotope exchange and mutation studies. All three paths investigated follow a single-step associative-like mechanism (see movies at http://www.iwr.uni-heidelberg.de/groups/biocomp/fischer) and show very similar heavy-atom positions in the transition states regardless of the positions of the protons. In the product states, the coordination bond between Mg2+ and Ser237 (and thus the switch-1 loop) is broken. This indicates that product release is likely to occur via an exit route that opens by complete opening of the switch-1 loop ("trap door" mechanism). Moreover, the coordination distance between Mg2+ and inorganic phosphate (Pi) is extended. This indicates that after hydrolysis this bond may be completely cleaved as an early event necessary for phosphate exit. Inspired by the simulation results, a Network Hypothesis on the mechanism of ATP hydrolysis in myosin is put forward that combines previous mechanistic proposals and that is consistent with experimental data available from mutational and isotope exchange studies. Moreover, a mechanism is suggested to explain how the catalytic cycle is coupled to the motor activity of myosin. %Z Teile in: S.M. Schwarzl, Understanding the ATP hydrolysis mechanism in myosin using computer simulation techniques, Mensch und Buch Verlag Berlin 2006, ISBN 3-86664-044-7