Protein-solvent interface

As an illustration, we briefly discuss the SCC-DFTB/MM simulations of carbonic anhydrase II (CAII), which is a zinc-enzyme that catalyzes the interconversion of CO2 and HCO [86], The rate-limiting step of the catalytic cycle is a proton transfer between a zinc-bound water/hydroxide and the neutral/protonated His64 residue close to the protein/solvent interface. Since this proton transfer spans at least 8-10 A depending on the orientation of the His 64 sidechain ( in vs. out , both observed in the X-ray study [87]), the transfer is believed to be mediated by the water molecules in the active site (see Figure 7-1). To carry out meaningful simulations for the proton transfer in CAII, therefore, it is crucial to be able to describe the water structure in the active site and the sidechain flexibility of His 64 in a satisfactory manner. 

Bizzarri, A. R. and Cannistraro, S., Molecular dynamics of water at the protein-solvent interface. J. Phys. Chem. B 106, 6617-33 (2002). 

The protein-solvent interface was studied in an explicit solvent environment of 3182 water molecules by MD simulations performed on metmyoglobin [31].Both the structure and dynamics of the hydrated surface of myoglobin are similar to that obtained by experimental methods calculating three-dimensional density distributions, temperature factors and occupancy weights of the solvent molecules. On the basis of trajectories they identified multiple solvation layers around the protein surface including more than 500 hydration sites. Properties of theoretically calculated hydration clusters were compared to that obtained from neutron and X-ray data. This study indicates that the simulation unified the hydration picture provided by X-ray and neutron diffraction experiments. 

Although no hexane molecules were found in the protein s interior for the CTWAT and CTMONO systems, hydrophobic contacts were observed between hexane molecules near the protein surface and hydrophobic side chains in all three systems. Hexane molecules on the protein surface tend to reside in the surface "clefts" formed by the hydrophobic side chains extended into the hexane solvent. At the same time, the hydrophilic residues tended to fold back onto the surface of the protein in order to minimize surface contacts. In our CTMONO simulation, we further observed the water molecules clustered around charged hydrophilic residues, while leaving the hydrophobic residues exposed to the soIvent.(Fig. 1) It has been reported that preferential solvation of the hydrophobic regions of the protein surface by the non-polar solvent is due to the thermodynamically unfavorable formation of a complete monolayer of water in a non-polar solvent. Klibanov and co-workers have also shown that hexane does not strip the water layer - nor does it immobilize the water molecules at the protein/solvent interface. Instead, rearrangements of the water molecules on the protein surface is the more favored process. Our simulations clearly support these experimental observations. 

M. Tarek D.J. Tobias (2002). Phys. Rev. Lett., 89, art no. 275501. Single particle and collective dynamics of protein hydration water A molecular djmamics study. A.R. Bizzarri S. Cannistrato (2002). J. Phys. Chem. B, 106, 6617-6633. Molecular dynamics of water at the protein solvent interface. 

It is worth noting that the protein-solvent interface is uniquely defined by the whole protein and is used in all the electronic structure calculations of the capped fragments and conjugate caps. Thus in each cycle of MFCC calculation, all the fragments and caps are interacting with a common external ESP created by the same set of induced charges on the cavity surface. 

The other approach is to map the polarization effect onto the protein-solvent interface in the form of induced charge, and then take the induced charge as a single electron operator in the MFCC Hamiltonian. The latter approach is employed in working with the discrete representation of the electron density by (Ji et al., 2008). Similar to the MFCC-CPCM approach, this procedure must be iterated until convergence is reached. 

Ttichsen, E., Woodward, C. (1985) Hydrogen kinetics of peptide amide protons at the bovine pancreatic trypsin inhibitor protein-solvent interface. Journal of Molecular Biology, 185 (2), 405 19. 

Purkiss A, Skoulakis S, Goodfellow JM (2001) The protein-solvent interface a big splash. Phil Trans R Soc Lond A Math Phys Eng Sci 359 1515-1527... 

---