On the Approximation of Solvent Effects on the Conformation and Dynamics of Cyclosporin A by Stochastic Dynamics Simulation Techniques

The molecular simulation technique of stochastic dynamics (SD) is tested by application to the immunosuppressive drug cyclosporin A (CPA). Two stochastic dynamics simulations are performed, one (SDCCl4 ) with atomic friction coefficients proportional to the viscosity of the nonpolar solvent CCl4, and one (SDH2O) with atomic friction coefficients corresponding to an aqueous solution. The atomic friction coefficients are also taken proportional to an approximate expression for the atomic accessible surface area. The properties of both stochastic dynamics simulations are compared to those of two full molecular dynamics (MD) simulations of cyclosporin A, one in a box with 591 CCl4 molecules, and one in a box with 632 H2O molecules. The properties of cyclosporin A as found in the molecular dynamics simulation in CCl4 are well reproduced by the SDCCl4 simulation. This indicates that the neglect of a mean force reresenting the average solvent effects on the solute is justified in the case of nonpolar solvents. For polar solvents, like water, this mean force may not be neglected. The SDH2O simulation of cyclosporin A clearly fails to reproduce the amount of hydrogen bonding found in the molecular dynamics stimulation of cyclosporin A in water. A comparison with a molecular dynamics simulation of cyclosporin A in vacuo shows that both the SDCCl4 and the SDH2O simulation come closer to the properties of the molecular dynamics simulations in CCl4 and in H2O than a molecular dynamics simulation in vacuo.