Transition Path Sampling Study of Engineered Enzymes That Catalyze the Morita-Baylis-Hillman Reaction: Why Is Enzyme Design so Difficult?

Abstract

It is hoped that artificial enzymes designed in laboratories can be efficient alternatives to chemical catalysts that have been used to synthesize organic molecules. However, the design of artificial enzymes is challenging and requires a detailed molecular-level analysis to understand the mechanism they promote in order to design efficient variants. In this study, we computationally investigate the mechanism of proficient Morita-Baylis-Hillman enzymes developed using a combination of computational design and directed evolution. The powerful transition path sampling method coupled with in-depth post-processing analysis has been successfully used to elucidate the different chemical pathways, transition states, protein dynamics, and free energy barriers of reactions catalyzed by such laboratory-optimized enzymes. This research provides an explanation for how different chemical modifications in an enzyme affect its catalytic activity in ways that are not predictable by static design algorithms.

Figure 1.

Chemical scheme for the proposed mechanism depicting the steps involved in the MBH reaction between 2-cyclohexen-1-one (1) and 4-nitrobenzaldehyde (2) leading to the formation of 2-[hydroxy(4-nitrophenyl)methyl]cyclohex-2-en-1-one (3) catalyzed by a nucleophile (N̈uc) and a water molecule. The atoms labeled in red are directly involved in the proton transfer between the water molecule and the intermediate 2 and are given unique names to label them for the computational analysis in this manuscript.

Figure 2.

Proposed mechanism of the MBH reaction catalyzed by the BH32.12 enzyme. The substrate is covalently bound to the His23 residue, and oxyanionic intermediate states are stabilized by the Arg124 residue, which is essential for catalysis. The proton transfers are mediated by a water molecule.

Figure 3.

(a) Time series of the evolution of various proton transfer atomic distances along a typical TPS trajectory for the BH32.12 enzyme with an arginine residue and water-mediated reaction mechanism. (b) Geometry of the substrate along with the Arg124 and His23 residues at the second transition state calculated using committor analysis for a typical TPS trajectory, which indicates the formation of an OH⁻ ion. (c) Reaction free energy as a function of the order parameter χ = d(OH2–H1) – d(H1–O4) + d(O4–C7) + d(C7–C8) + d(C8–H5) – d(H5–OH2) calculated using a window-based modified TPS algorithm, where the error bars obtained from bootstrapping analysis are indicated in blue.

Figure 4.

Predicted reaction scheme for the BH_MeHis1.8 variant containing (a) the ionized form of the Glu26 residue (glutamate) on the left and (b) the glutamine Gln26 residue on the right. For the Gln26 case, the carbonyl group acts as a hydrogen bond acceptor, whereas for the glutamate case, both the OE1 and OE2 atoms with delocalized negative charge act as hydrogen bond acceptors.

Figure 5.

(a) Time series of the evolution of various proton transfer atomic distances along a typical TPS trajectory for the BH_MeHis1.8 enzyme with a glutamate residue and water-mediated reaction mechanism. (b) Geometry of the substrate along with the Glu26 and MeHis23 residues at the transition state calculated using committor analysis for a typical TPS trajectory, which indicates the formation of a H₃O⁺ ion. (c) Reaction free energy as a function of the order parameter ζ = d(C8–H5) – d(OH2–H5) calculated using a window-based modified TPS algorithm, where the error bars obtained from bootstrapping analysis are indicated in blue.

Figure 6.

Reaction mechanism promoted by the BH_MeHis1.8 enzyme with a GluH26 residue protonating the O4 oxyanion and a catalytic water molecule involved in the transfer of proton from the substrate to the glutamate residue.

Figure 7.

(a) Time series of the evolution of various proton transfer atomic distances along a typical TPS trajectory for the BH_MeHis1.8 enzyme with a glutamic acid residue and water-mediated reaction mechanism. (b) Geometry of the substrate along with the GluH26 and MeHis23 residues at the transition state calculated using committor analysis for a typical TPS trajectory, which indicates the formation of a H₃O⁺ ion. (c) Reaction free energy as a function of the order parameter ζ = d(C8–H5) – d(OH2–H5) calculated using a window-based modified TPS algorithm, where the error bars obtained from bootstrapping analysis are indicated in blue.

Figure 8.

Reaction scheme for the direct protonation mechanism between the substrate and the Glu26 residue. The atoms marked in red color are given special labels, which will be used in the analysis.

Figure 9.

(a) Time series of the evolution of various proton transfer atomic distances along a typical TPS trajectory for the BH_MeHis1.8 enzyme with a glutamic acid residue directly responsible for both proton transfers. (b) Geometry of the substrate along with the GluH26 and MeHis23 residues at the transition state calculated using committor analysis for a typical TPS trajectory. (c) Reaction free energy as a function of the order parameter ζ = d(C8–H5) – d(OH2–H5) calculated using a window-based modified TPS algorithm, where the error bars obtained from bootstrapping analysis are indicated in blue.

Figure 10.

(a) Time series of the evolution of various proton transfer atomic distances along a typical TPS trajectory for the BH_MeHis1.8 enzyme with a glutamine residue and water-mediated reaction mechanism. (b) Geometry of the substrate along with the Gln26 and MeHis23 residues at the transition state calculated using committor analysis for a typical TPS trajectory, which indicates the formation of a H₃O⁺ ion. (c) Reaction free energy as a function of the order parameter ζ = d(C8–H5) – d(OH2–H5) calculated using a window-based modified TPS algorithm, where the error bars obtained from bootstrapping analysis are indicated in blue.

Figure 11.

Residues (a) Trp10 in the BH32.12 variant and (b) Leu87 in the BH_MeHis1.8 variant participating in protein dynamics.