Computational techniques for the study of enzyme active sites.
Andrew C. Wallace · OpenGrey (Institut de l'Information Scientifique et Technique) · 1996
The aim of this thesis, which is based on the coordinate data from known X-ray crystal and nuclear magnetic resonance (NMR) structures, is to further our understanding of the ways in which enzymes catalyse their reactions. Many enzyme structures are solved with inhibitors or substrate analogs bound to their active sites. To enable evaluation of the protein-ligand interactions that occur in these complexes, a computational tool called LIGPLOT has been developed which allows swift pictorial evaluation of the ligand and its interactions with the enzyme. These LIGPLOT diagrams, along with other relevent information has been compiled into an enzyme database that is available over the World Wide Web. The rest of the thesis is devoted to studying structural organisation of the amino acid residues that are directly involved in chemical catalysis. A detailed analysis of the geometry of the Ser-His-Asp catalytic triad found in the serine proteinases and lipase X-ray and NMR structures showed that it is possible to define a 3D consensus template that identifies all catalytic Ser-His-Asp triads with the exclusion of all other interactions. To create 3D consensus templates describing other enzyme active sites we needed a generalised search method. To do this, a computer program called 'TESS' has been developed which is based on the geometric hashing paradigm. Using this program, a database of enzyme active site templates has been created which enables swift evaluation of the function of a new protein structure as it is solved and aid protein design and engineering experiments.