Proton transfer function of carbonic anhydrase: Insights from QM/MM simulations

Abstract

Recent QM/MM analyses of proton transfer function of human carbonic anhydrase II (CAII) are briefly reviewed. The topics include a preliminary analysis of nuclear quadrupole coupling constant calculations for the zinc ion and more detailed analyses of microscopic pK(a) of the zinc-bound water and free energy profile for the proton transfer. From a methodological perspective, our results emphasize that performing sufficient sampling is essential to the calculation of all these quantities, which reflects the well solvated nature of CAII active site. From a mechanistic perspective, our analyses highlight the importance of electrostatics in shaping the energetics and kinetics of proton transfer in CAII for its function. We argue that once the pK(a) for the zinc-bound water is modulated to be in the proper range (approximately 7.0), proton transfer through a relatively well solvated cavity towards/from the protein surface (His64) does not require any major acceleration. Therefore, although structural details like the length of the water wire between the donor and acceptor groups still may make a non-negligible contribution, our computational results and the framework of analysis suggest that the significance of such "fine-tuning" is likely secondary to the modulation of pK(a) of the zinc-bound water. We encourage further experimental analysis with mutation of (charged) residues not in the immediate neighborhood of the zinc ion to quantitatively test this electrostatics based framework; in particular, Phi analysis based on these mutations may shed further light into the relative importance of the classical Grotthus mechanism and the "proton hole" pathway that we have proposed recently for CAII.

FIG. 1:

The active site of CAII rendered from the crystal structure (PDB ID: 2CBA). All dotted lines correspond to hydrogen-bonding interactions with distances ≤3.5 Å. Glu117 and Glu106 are in close proximity to His119, and Glu106 also interacts with Thr199 through the presumed hydroxyl proton of Thr199 (not shown for clarity). His64 is resolved to partially occupy both the “in” and “out” rotameric states; whether both states are productive toward proton transfer from the zinc-bound water is one of the mechanistic issues under debate (see text).

FIG. 2:

The contribution to the electrostatic component of free energy of deprotonation for the zinc-bound water from protein atoms (only MM atoms are considered) in the WT and E106Q mutant CAII based on perturbative analysis. Contribution from individual residue is plotted against the distance (Cα) from the zinc ion. Note the striking similarity between the WT and E106Q results, except, apparently, for residue 106.

FIG. 3:

Snapshots and excess coordination number plots (see text) for PMF and MEP simulations of proton transfers in CAII. (a) Transition state from a MEP spanning four bridging water molecules between the zinc-bound water and His64; (b) the excess coordination number plot for MEPs involving four bridging water molecules, which indicate the concerted nature of the transfer pathways; (c) a snapshot from PMF simulations (ζ ~ 0.7 window) with the “out” configuration of His64, which clearly shows a hydroxide being stabilized by water molecules and polar residues (Asn67, Gln92) in the active site; (d) the excess coordination number plot for the “out”-His64 PMF simulations, which highlights a “proton hole” mechanism.

FIG. 4:

Calculated classical potentials of mean force (PMFs) for the proton transfer between the zinc-bound water and His 64 in the “in” and “out” configurations for the wild type CAII. The x-axis is a generalized coordinate (ζ) that characterizes the progress of the proton transfer based on a modified center of excess charge.^,

FIG. 5:

A schematics that illustrates the relative energetics of the Grotthus and “proton hole” transfer mechanisms, which involve step-wise transfer of hydronium and hydroxide ions, respectively. When pK_a differences between key groups are large, the basic energetic profile can be estimated based on the pK_a values of the proton donor/acceptor groups (AH, BH) and water (W) as well as hydronium (WH⁺); the distance between the donor/acceptor is not an important factor for the proton transfer kinetics (assuming that the process is not controlled by the diffusion of the hydronium or hydroxide).

FIG. 6:

Simplified characteristic charge distribution in the transition state for the proton-hole and Grotthus mechanisms in CAII. The former is featured with +2 for the zinc-center, +1 for His64H⁺ and −1 for the hydroxide; the Grotthus mechanism is featured with +1 for the zinc-center, 0 for the neutral His64 and +1 for the hydronium. Accordingly, Φ values for residues in the active site are likely to have different patterns in the two mechanisms (estimated Φ values based on perturbative analysis are shown for several residues for an illustration). Thus a systematic Φ analysis can provide experimental insights into the relative importance of the proton-hole and Grotthuss mechanisms in CAII.