| dc.description.abstract | It has been demonstrated earlier in this laboratory that many protein crystals, but not all, undergo reversible structural transformations, as evidenced by abrupt changes in unit cell dimensions, diffraction pattern, and solvent content, when the environmental humidity is systematically varied.
The progress made so far in exploring the variability in protein hydration and its structural consequences using such water?mediated transformations is reviewed in the first chapter of the thesis. In addition to carrying out preliminary investigations on several protein crystals, the crystal structure of the low?humidity form of the well?known tetragonal hen egg?white lysozyme has been analysed. A comparison of this structure with the already known native structure showed that the gentle removal of a small amount of bulk water, which is what water?mediated transformations involve, results in significant changes in the hydration shell, although the hydration shell tends to move as a whole along with the protein molecule. These changes lead to structural perturbations in the protein, which are most pronounced in the regions involved in substrate binding.
Encouraged by the results obtained on tetragonal lysozyme, the X?ray analysis of low?humidity monoclinic lysozyme was taken up. Indeed, monoclinic lysozyme exhibits the most remarkable water?mediated transformation observed so far. The two crystallographically independent molecules in the native crystals become equivalent in the low?humidity form which at 22% has the lowest solvent content among the protein crystals examined to date. The low?humidity monoclinic crystals diffract better than the native ones.
Intensity data collection, at a relative humidity of 88%, up to a resolution of 1.75 Å from low?humidity monoclinic lysozyme is described in the second chapter. Oscillation photography followed by computer?controlled microdensitometry was used for data collection. The data were processed using programmes originally developed by M. G. Rossmann. The final processed data set contained 8523 independent reflections.
Attempts to refine the structure using one or the other of the two crystallographically independent molecules in the native structure as the starting model did not succeed. Subsequently, the structure was solved using rotation function and R?factor calculations employing the molecule in low?humidity tetragonal lysozyme as the search model. These calculations and the subsequent refinement of the structure using Hendrickson–Konnert restrained least squares method, interspersed with model building based on Fourier maps, are described in the third chapter. The final model, containing 997 protein atoms, 148 water molecules and 2 nitrate ions, had an R?factor of 0.175 for 7684 reflections with F > 4?(F) in the 10 to 1.75 Å resolution shell.
The refined structure, discussed in detail in the next chapter, provides the most accurate description to date of the enzyme molecule. The packing coefficient of the protein molecules in the crystal is, at 0.78, as high as that observed in closest?packed crystals of organic molecules. Each molecule is surrounded by, and in contact with, 14 other such molecules. The interstices in this arrangement are filled with the solvent. It turns out that about 90% of the total solvent, mainly water molecules, in the structure could be located. They provide a wealth of information on protein?water interactions, favourable sites of hydration on the protein surface, the hydration of secondary structural features and side chains, intermolecular solvent bridges, and the water structure associated with proteins.
A comparison of the low?humidity structure with the native structure shows that substantial rearrangement of molecules takes place during the transformation. Differences in detail also exist in the molecular structure of the protein. A comparison of the protein molecule and its hydration shell with those in native tetragonal lysozyme, low?humidity tetragonal lysozyme, high?pressure tetragonal lysozyme and triclinic lysozyme is presented in the final chapter. This study leads to the delineation of the relatively rigid, moderately flexible, and highly flexible regions of the molecule. The relatively rigid region forms a contiguous structural unit close to the molecular centroid and encompasses parts of the main ??structure and three ??helices. The hydration shell of the molecule contains 30 invariant water molecules. Many of them are involved in holding different parts of the molecule together or in stabilizing the local structure. Five of the six invariant molecules attached to the substrate binding region form part of a water cluster contiguous with the side chains of the catalytic residues Glu?35 and Asp?52.
While pursuing the detailed study of low?humidity monoclinic lysozyme, the author has also been involved in the high?resolution X?ray analysis of the complexes of tetragonal lysozyme with two closely related indicator dyes bromophenol red (BPR) and bromophenol blue (BPB). This work is presented in an appendix. The dyes BPR and BPB bind to lysozyme and inhibit its activity against bacterial cell wall, but not against the polysaccharide. The binding site of BPR had already been characterized earlier at 5.5 Å. The 2 Å analysis of the BPR complex confirms this site to be outside the cleft close to subsite F. The 2 Å analysis of the BPB complex, however, shows that this dye binds at a site far away from the active site cleft. The pH’s at which the dyes were soaked into the crystal were such that BPR (pH 4.6) was neutral and BPB (pH 5.6) was ionised. The results thus appear to indicate that the neutral and the ionised species of the dyes have different binding sites. These sites are perhaps involved in the interaction of the protein with the peptide component of peptidoglycan.
Most of the work presented in the thesis is not published yet. Only the results described in the fifth chapter have been reported in the following publication:
Rigid and flexible regions in lysozyme and invariant features of the hydration shell, Current Science (1991) 60, 165–170 (with M. Vijayan). | |
