Calculation of vibrational shifts of nitrile probes in the active site of ketosteroid isomerase upon ligand binding

Abstract

The vibrational Stark effect provides insight into the roles of hydrogen bonding, electrostatics, and conformational motions in enzyme catalysis. In a recent application of this approach to the enzyme ketosteroid isomerase (KSI), thiocyanate probes were introduced in site-specific positions throughout the active site. This paper implements a quantum mechanical/molecular mechanical (QM/MM) approach for calculating the vibrational shifts of nitrile (CN) probes in proteins. This methodology is shown to reproduce the experimentally measured vibrational shifts upon binding of the intermediate analogue equilinen to KSI for two different nitrile probe positions. Analysis of the molecular dynamics simulations provides atomistic insight into the roles that key residues play in determining the electrostatic environment and hydrogen-bonding interactions experienced by the nitrile probe. For the M116C-CN probe, equilinen binding reorients an active-site water molecule that is directly hydrogen-bonded to the nitrile probe, resulting in a more linear C≡N--H angle and increasing the CN frequency upon binding. For the F86C-CN probe, equilinen binding orients the Asp103 residue, decreasing the hydrogen-bonding distance between the Asp103 backbone and the nitrile probe and slightly increasing the CN frequency. This QM/MM methodology is applicable to a wide range of biological systems and has the potential to assist in the elucidation of the fundamental principles underlying enzyme catalysis.

Figure 1

Schematic diagram of the one-dimensional potential energy curve as a function of the CN bond length and the corresponding ground and first excited vibrational state energy levels. The anharmonicity of this curve leads to different average CN bond lengths, where 〈x₁〉 > 〈x₀〉, and therefore different dipole moments, where 〈μ〉 > 0, associated with these two vibrational states. As a result, the CN vibrational frequency, indicated by the red arrow, is sensitive to the local electrostatic field, leading to the vibrational Stark effect. This figure was inspired by and modeled after Figure S1 in Ref. .

Figure 2

Comparison of chemical structures of the intermediate dienolate form of the 5-AND steroid substrate and the intermediate analog EQU.

Figure 3

Representative snapshots from MD simulations of D40N KSI with bound EQU in its anionic form for the (A) M116C-CN and (B) F86C-CN systems. The residues that are included in the QM region for the QM/MM calculations of the vibrational shifts and line shapes are depicted with thicker lines.

Figure 4

Calculated CN frequencies for the gas phase methyl thiocyanate-water dimer treated fully quantum mechanically using the following methods: DFT B3LYP/6-311++G(d,p) (black circles), PM3 with parameters developed for acetonitrile by Corcelli and coworkers (red), and PM3 from the present work obtained with no waters (blue), one water (green), and two waters (violet) included in the QM region during the fitting procedure. The CN frequency is calculated as a function of (A) the CNH angle at the equilibrium N--H distance of 2.1 Å and (B) the N--H distance with a CNH angle of 135°. The analogs of Figure 4B with CNH angles of 90° and 180° are provided in Figure S2 of Supporting Information.

Figure 5

Calculated IR spectra of the CN vibrational stretching frequency for the M116C-CN D40N KSI system with no ligand bound and Tyr57 depronated (solid red line), with EQU bound in its anionic form (solid black line), and with EQU bound in its neutral form and Tyr16 deprotonated (dashed black line).

Figure 6

Distribution of the CNH angle between the nitrile probe and the hydrogen-bonded water molecule as determined from the MD simulations of the M116C-CN D40N KSI system with no ligand bound and Tyr57 depronated (solid red line), with EQU bound in its anionic form (solid black line), and with EQU bound in its neutral form and Tyr16 deprotonated (dashed black line).

Figure 7

Calculated IR spectra of the CN vibrational stretching frequency for the F86C-CN D40N KSI system with no ligand bound and Tyr57 depronated (solid red line), with EQU bound in its anionic form (solid black line), and with EQU bound in its neutral form and Tyr16 deprotonated (dashed black line).

Figure 8

Distribution of the distance between the amide hydrogen of the Asp103 backbone and the nitrogen of the F86C-CN probe as determined from the MD simulations of the F86C-CN D40N KSI system with no ligand bound and Tyr57 depronated (solid red line), with EQU bound in its anionic form (solid black line), and with EQU bound in its neutral form and Tyr16 deprotonated (dashed black line).