The charges of the ligand atoms were determined by fitting to the electrostatic potential computed by QM calculations on the HF/6C31G(d) level of theory according to the RESP scheme, as implemented in AmberTools18 all in line with our previous reports [56,57,58,69,70,74,75,76,82,83]. Open in a separate window Figure 5 The structure of the hydrated MAO Santacruzamate A B with rasagiline (in red) and selegiline (in violet) placed in the active site. kcal mol?1 in SEL (imag = 1390cm?1). The latter suggests a moderately higher reactivity for RAS in the aqueous solution, which would be different from the trend observed within the MAO B active site. Still, it is very likely that a much simpler and highly polar aqueous environment favors hydride abstraction from a system with a more polar secondary amine moiety in its immediate vicinity, as in RAS, than with a more hydrophobic tertiary amine, as in SEL. Nevertheless, the thermodynamic picture of the investigated reaction is in line with the expected situation in the enzyme, as the overall reaction free energy was, by cm?1 for RAS and 1390cm?1 for SEL) and the intrinsic reaction coordinate (IRC) calculations. The reaction free energy was calculated as the difference between the energy of the reactants complex and the transient intermediate point on the products reaction coordinate path in which the hydrided LMFH? moiety remained planar, in line with our previous reports [56,74,75,76]. All QM calculations were performed using the Gaussian 16 program package . The starting points for our EVB simulations were the coordinates of the MAO B enzyme in complex with the bound NYP inhibitor (PDB ID: 1GOS) . The inhibitor was removed, but its position in this structure served as a reference point for the initial manual positioning of RAS and SEL into the active site (Figure 5) using the UCSF Chimera program . The protein model included one subunit of the dimeric MAO B Santacruzamate A enclosed in a simulation sphere, with a 30 ? radius, centered at the reactive N5 atom of the FAD cofactor. Such a setup encompassed the vast majority of the proteineither RAS or SELand 1662 TIP3P water molecules. All protein atoms outside this sphere were kept restrained to their starting positions by applying a 200 kcal mol?1 ??2 harmonic restraint. The simulations were built around the OPLS-AA force field , with the ligand parameters acquired by the ffld_server utility and assisted by the Maestro v. 11.7 graphical interface . The charges of the ligand atoms were determined by fitting to the electrostatic potential computed by QM calculations on the HF/6C31G(d) level of theory according to the RESP scheme, as implemented in AmberTools18 all in line with our previous reports [56,57,58,69,70,74,75,76,82,83]. Open in a separate window Figure 5 The structure of the hydrated MAO B with rasagiline (in red) and selegiline (in violet) placed in the active site. The position of the FAD cofactor is also shown in the stick representation. The system Vegfa was first equilibrated in several distinct steps, by slowly increasing both the temperature (starting at 1 K and ending at 300 K) and the time-step (from 0.1 to 1 1 fs), as well as gradually removing the restraints. An additional equilibration step of 10 ns was carried out at 300 K with minimal position restraints. Such an equilibrated structure was used as the starting point for the subsequent simulations, which employed standard EVB procedure based on the free energy perturbation/umbrella sampling (FEP/US) approach [67,84,85]. In the case of FEP, the force fields which describe the valence states of reactants and the products (Figure 2) must first be established. This Santacruzamate A force field was appropriately tuned to allow for the breaking and formation of bonds, by replacing the harmonic potentials of the CCH and NCH bonds with Morse functions, as well as substituting the 12-6 Lennard-Jones potential with a less restrictive Buckingham-type nonbonding potential on the three reacting Santacruzamate A atoms. The reactants were then converted to the products in a series of mapping steps, using a mapping potential of the type : m = ?1 + (1 ? )?2 where the force field of the reactants (1) was gradually transformed into the force field of the products (2) via the coupling parameter lambda (). In our case, the initial structure was equilibrated at = 0.5 (i.e., a structure in the vicinity of the transition state). Thus, the subsequent FEP procedure was carried out starting at = 0.5 and finishing at either = 0 or = 1, corresponding to reactants or the.