Hybrid quantum/classical molecular dynamics simulations of the proton transfer reactions catalyzed by ketosteroid isomerase: analysis of hydrogen bonding, conformational motions, and electrostatics

Abstract

Hybrid quantum/classical molecular dynamics simulations of the two proton transfer reactions catalyzed by ketosteroid isomerase are presented. The potential energy surfaces for the proton transfer reactions are described with the empirical valence bond method. Nuclear quantum effects of the transferring hydrogen increase the rates by a factor of approximately 8, and dynamical barrier recrossings decrease the rates by a factor of 3-4. For both proton transfer reactions, the donor-acceptor distance decreases substantially at the transition state. The carboxylate group of the Asp38 side chain, which serves as the proton acceptor and donor in the first and second steps, respectively, rotates significantly between the two proton transfer reactions. The hydrogen-bonding interactions within the active site are consistent with the hydrogen bonding of both Asp99 and Tyr14 to the substrate. The simulations suggest that a hydrogen bond between Asp99 and the substrate is present from the beginning of the first proton transfer step, whereas the hydrogen bond between Tyr14 and the substrate is virtually absent in the first part of this step but forms nearly concurrently with the formation of the transition state. Both hydrogen bonds are present throughout the second proton transfer step until partial dissociation of the product. The hydrogen bond between Tyr14 and Tyr55 is present throughout both proton transfer steps. The active site residues are more mobile during the first step than during the second step. The van der Waals interaction energy between the substrate and the enzyme remains virtually constant along the reaction pathway, but the electrostatic interaction energy is significantly stronger for the dienolate intermediate than for the reactant and product. Mobile loop regions distal to the active site exhibit significant structural rearrangements and, in some cases, qualitative changes in the electrostatic potential during the catalytic reaction. These results suggest that relatively small conformational changes of the enzyme active site and substrate strengthen the hydrogen bonds that stabilize the intermediate, thereby facilitating the proton transfer reactions. Moreover, the conformational and electrostatic changes associated with these reactions are not limited to the active site but rather extend throughout the entire enzyme.

Figure 1

Schematic depiction of the proton transfer reactions catalyzed by KSI. In the first step, the proton transfers from the C4 atom of the substrate to the Asp38 residue. In the second step, the proton transfers from the Asp38 residue to the C6 atom of the substrate. The reactant, intermediate, and product states of the overall reaction are labeled.

Figure 2

Model system of the KSI active site used to assist in the parameterization of the EVB potential. In this simple model, the substrate is represented by cyclohexone, Asp38 is represented by ethanoic acid, and Tyr14 and Asp99 are represented by water molecules to include the key hydrogen bonding interactions. The transferring hydrogen is identified with an asterisk.

Figure 3

Free energy profiles calculated for (a) the first proton transfer step and (b) the second proton transfer step catalyzed by KSI. These PMF curves were generated along an energy gap collective reaction coordinate using two-state EVB models with a series of mapping potentials. The QCP free energy profiles, which include the nuclear quantum effects of the transferring hydrogen, are indicated by the dashed lines in the barrier region.

Figure 4

Snapshot of the KSI active site for the Intermediate state shown in Figure 1. The snapshot was obtained from the MD simulation of the first proton transfer step with the mapping potential corresponding to λ = 0.95. The proton has transferred from the substrate to Asp38, and residues Tyr14 and Asp99 are hydrogen bonded to the substrate, which is in the dienolate form. The transferring hydrogen is identified with an asterisk.

Figure 5

Thermally averaged distances within the proton transfer interface calculated along the collective reaction coordinate for the two proton transfer reactions catalyzed by KSI. For the first proton transfer step, the substrate C4 atom is the donor and the Asp38 OD2 atom is the acceptor. For the second proton transfer step, the Asp38 OD2 atom is the donor and the substrate C6 atom is the acceptor. The donor-acceptor distance is depicted for the first step in (a) and the second step in (b). The donor-hydrogen (red) and acceptor-hydrogen (blue) distances are depicted for the first step in (c) and the second step in (d).

Figure 6

Thermally averaged distances and angles within the active site calculated along the collective reaction coordinate for the two proton transfer reactions catalyzed by KSI. The hydrogen bond donor-acceptor distances between the substrate O3 atom and Tyr14 (solid blue), between the substrate O3 atom and Asp99 (dashed red), and between Tyr14 and Tyr55 (dotted black) are depicted for the first step in (a) and the second step in (b). The angles Tyr14-Asp99-SubstrateO3 (solid blue) and Asp99-Tyr14-SubstrateO3 (dashed red) are depicted for the first step in (c) and the second step in (d). These angles are defined in terms of the heavy atoms involved in the hydrogen bonds between the substrate and both Tyr14 and Asp99.

Figure 7

Thermally averaged structures of KSI along the reaction pathway for both proton transfer reactions. The reactant state (black), transition state (red) and product state (green) for the first proton transfer step and the reactant state (brown), transition state (magenta) and product state (blue) for the second proton transfer step are presented. The loop regions exhibiting significant structural changes are labeled. A version of this figure separating the structures for the two proton transfer steps is given in Supporting Information.

Figure 8

RMSF values for the C_α atoms of the protein backbone corresponding to the reactant state (black), transition state (red), and product state (green) for the first proton transfer step and the reactant state (brown), transition state (magenta), and product state (blue) for the second proton transfer step. The peaks indicated with arrows correspond to the loop regions labeled in Figure 7. The red blocks under the data identify the active site residues as defined in the text.

Figure 9

Qualitative illustration of the conformational changes of Asp38 occurring between the first and second proton transfer steps catalyzed by KSI. The thermally averaged structures of the substrate and Asp38 are depicted for (a) the product of the first step and (b) the reactant of the second step. Although both structures correspond to the Intermediate state defined in Figure 1, each structure arises from a mixture of states in the two-state EVB models describing the two proton transfer steps. The transferring hydrogen is identified with an asterisk.

Figure 10

Illustration of the changes in the electrostatic potential on the solvent accessible surface in KSI during the two proton transfer reactions. The electrostatic potential is depicted for the thermally averaged structures of the Reactant in (a) and (d), the Intermediate in (b) and (e), and the Product in (c) and (f) for the overall reaction pathway shown in Figure 1. The active site is identified in (a), (b), and (c). The regions exhibiting significant changes in electrostatic potential on the side of the protein opposite the active site are identified in (d), (e), and (f). The red regions correspond to negative potential and the blue regions correspond to positive potential. Since the substrate is not included in the electrostatic potential calculations, the Intermediate includes an extra positive charge in the active site from the transferred proton.