Recognition of Water Molecules by Protein

It is not necessary to emphasize how important water is for living systems to maintain their life [24-26], No wonder that many scientists in the field of X-ray and neutron diffraction measurement have been trying to determine positions and orientations of water molecules around and inside biomolecules, or protein and DNA [27,28], However, it is not so easy even for modern experimental technology to locate the position of water molecules, partly due to the limited resolution of diffraction measurements in space as well as in time. This is because water molecules at the surface of protein are not necessarily bound firmly to some particular site of biomolecules, but exchange their positions quite frequently. Actually this flexibility and fluctuation of water molecules are essential for living systems to control their life. The diffraction measurement can identify only some water molecules that have long residence time at some particular position of the biomolecules. 

In this study [15,29], we have carried out the 3D-RISM calculation for a hen egg-white lysozyme immersed in water and obtained the 3D-distribution function of oxygen and hydrogen of water molecules around and inside the protein. The native 3D structure of the protein is taken from the protein data bank (PDB). The protein is known to have a cavity composed of the residues from Y53 to 158 and from A82 to S91, in which four water molecules have been determined by means of the X-ray diffraction measurement [30]. In our calculation, those water molecules are not included explicitly. 

It is interesting to compare the hydration structure obtained by the 3D-RISM theory with crystallographic water sites of X-ray structure [30]. The crystallographic water molecules in the cavity are depicted in the right of Fig. 10.3, showing four water sites in the cavity, much as the 3D-RISM theory has detected. Moreover, the water distributions obtained from the theory and experiment are quite similar to each other. Thus the 3D-RISM theory can predict the water-binding sites with great success. 

It should be noted that one peak of the 3D-distribution function does not necessarily correspond to one molecule. If a water molecule transfers back and forth between two sites in the equilibrium state, two peaks correspondingly appear in the 3D-distribution function. In fact, the number of water molecules within the cavity calculated from the 3D-distribution function is 3.6. It is less than the number of water-binding sites and includes decimal fractions. To explain that, we carried out molecular dynamics (MD) simulation using the same parameters and under the same thermodynamic conditions as 

---