Probing the electrostatics of active site microenvironments along the catalytic cycle for Escherichia coli dihydrofolate reductase

Abstract

Electrostatic interactions play an important role in enzyme catalysis by guiding ligand binding and facilitating chemical reactions. These electrostatic interactions are modulated by conformational changes occurring over the catalytic cycle. Herein, the changes in active site electrostatic microenvironments are examined for all enzyme complexes along the catalytic cycle of Escherichia coli dihydrofolate reductase (ecDHFR) by incorporation of thiocyanate probes at two site-specific locations in the active site. The electrostatics and degree of hydration of the microenvironments surrounding the probes are investigated with spectroscopic techniques and mixed quantum mechanical/molecular mechanical (QM/MM) calculations. Changes in the electrostatic microenvironments along the catalytic environment lead to different nitrile (CN) vibrational stretching frequencies and (13)C NMR chemical shifts. These environmental changes arise from protein conformational rearrangements during catalysis. The QM/MM calculations reproduce the experimentally measured vibrational frequency shifts of the thiocyanate probes across the catalyzed hydride transfer step, which spans the closed and occluded conformations of the enzyme. Analysis of the molecular dynamics trajectories provides insight into the conformational changes occurring between these two states and the resulting changes in classical electrostatics and specific hydrogen-bonding interactions. The electric fields along the CN axes of the probes are decomposed into contributions from specific residues, ligands, and solvent molecules that make up the microenvironments around the probes. Moreover, calculation of the electric field along the hydride donor-acceptor axis, along with decomposition of this field into specific contributions, indicates that the cofactor and substrate, as well as the enzyme, impose a substantial electric field that facilitates hydride transfer. Overall, experimental and theoretical data provide evidence for significant electrostatic changes in the active site microenvironments due to conformational motion occurring over the catalytic cycle of ecDHFR.

Figure 1

(A) Superposition of the T46C-CN and L54C-CN ecDHFR mutants in the closed conformation with folate and NADP⁺ bound, where only the thiocyanate residue is shown for the L54C-CN mutant. (B) ecDHFR in the closed (red) and the occluded (blue) conformations exhibited by the (C) five major complexes in its catalytic cycle.

Figure 2

(A) FTIR (SCN) and (B) NMR (S¹³CN) measurements along the catalytic cycle of ecDHFR. The color scales for the FTIR and NMR data have units of cm^–1 and ppm, respectively. The cofactor is orange, and the substrate/product is purple in all complexes (the coloring of the cofactor and substrate/product does not correspond to the color scales that represent the NMR shift or IR frequency). Enzyme complex labels in blue are in the occluded conformation, and those in red are in the closed conformation. The measured values for IR frequency and NMR chemical shift are provided in Table S4.

Figure 3

Configurations from MD simulations for the (A) T46C-CN and (B) L54C-CN ecDHFR mutants. The residues with thicker lines are included in the QM region for the QM/MM calculations of vibrational frequencies.

Figure 4

Calculated and experimentally measured IR spectra of the CN vibrational stretching frequency for the (A) T46C-CN and (B) L54C-CN ecDHFR systems with NADP⁺/FOL bound (closed conformation; black line) and with NADP⁺/THF bound (occluded conformation; red line). The solid lines are the simulated spectra from the QM/MM calculations, and the dotted lines are the measured spectra from the FTIR experiments.

Figure 5

Calculated electric field along the CN bond at the midpoint of this bond for the (A) T46C-CN and (B) L54C-CN ecDHFR systems with NADP⁺/FOL bound (closed conformation; black line) and with NADP⁺/THF bound (occluded conformation; red line).

Figure 6

Contributions to the calculated electric field along the CN bond at the midpoint of this bond for the (A) T46C-CN and (B) L54C-CN ecDHFR systems with NADP⁺/FOL bound and the Met20 loop in the closed conformation. The color for each residue corresponds to the calculated values in Tables 1 and 2, respectively, using the color scale provided, although residues contributing a magnitude >4.0 MV/cm are depicted in the darkest color. The residues that contribute negatively to the field are colored red, the residues that contribute positively to the field are colored blue, and the residues that have no net contribution to the field along the CN bond are colored white.

Figure 7

Depiction of the component of the electric field along the hydride transfer D–A axis calculated from an MD simulation of WT ecDHFR with NADP⁺/FOL bound and the Met20 loop in the closed conformation. The color for each residue corresponds to the calculated electric field values using the color scale provided, although residues contributing a magnitude >4.0 MV/cm are depicted in the darkest color. The three arrows in the red oval represent the total electric field of −48.9 MV/cm (purple), the field of −32.4 MV/cm resulting from the ligands (yellow), and the field of −16.5 MV/cm resulting from the rest of the system (green) projected along the D–A axis.