published_or_final_version / Chemistry / Doctoral / Doctor of Philosophy

The kringle is a name given to a sequence of seventy nine residues linked by three disulphide bridges, multiple copies of which have been found in a number of proteins involved in haemostasis. Kringles from plasminogen and prothrombin have been studied using high resolution nuclear magnetic resonance (NMR) spectroscopy. Preliminary experiments demonstrated the kringles to be globular proteins in aqueous solution having well defined tertiary folds. Resonances have been assigned in the NMR spectrum to all but two of the aromatic residues of kringle 4 of human plasminogen and in addition a number of aliphatic residues have also been assigned. The majority of these assignments have been made to specific protons in the protein with the aid of chemically modified kringle 4 derivatives and comparison with other kringle protein spectra. Nuclear Overhauser enhancement studies have demonstrated a number of structurally significant side chain interactions. The lysine binding site of kringle 4 has been explored using two ω-aminocarboxylic acids as ligands. Further structural probes have included copper and lanthanide ions as well as chemically modified kringle 4 species. Combining all the data has allowed a preliminary molecular model to be built of part of kringle 4 incorporating all the residues between the second and third disulphide bridges. Preliminary studies demonstrate the possibility of being able to refine parts of this structure to an accuracy approaching that of X-ray crystallographic studies. A similar number of assignments have been made in the spectrum of bovine prothrombin fragment 2. Comparison of the data on the kringles of both fragment 2 and kringle 4 shows the two proteins to have preserved structural elements, and in particular one of these involves the conserved residues Trp 25, Leu 45 and Trp 61. Furthermore, the two tryptophan residues have been shown to have quite unique spectral charateristics. Brief studies of human prothrombin fragment 2 reveals similar findings. The results discussed indicate that the kringle sequence gives rise to a well defined fold. In kringle 4 of human plasminogen this is seen to provide a highly specific binding site for certain ω-aminocarboxylic acids. The data is all in accord with kringles being autonomous units within the non protease portion of zymogen molecules and which are ideally suited to include the mediation of protein-protein interactions amongst their biological functions.

Molecular docking algorithms are computational methods that predict the binding site and docking pose of specified ligands with a protein target. They have proliferated in recent years, due to the explosion of structural data in biology. Oxdock is an algorithm that uses various techniques to simplify this complex task, the most significant being the use of a multiscale approach to analyse the problem using a simple representation in the early stages. Oxdock is shown to be a very useful tool in computational biology, as exemplified by two cases. The first case is the analysis of the NMDA subclass of neuronal glutamate receptors and the subsequent elucidation of their function. The second is the investigation of the newly discovered plant glutamate receptors and the clarification of their natural ligands. The results in both instances open new areas of research into exciting areas of biology. Despite its effectiveness in solving many problems, Oxdock does fail in a number of circumstances. It is thus important to devise a new and improved method for molecular docking. This is achieved by combining the speed of the multiscale approach with the optimising ability of Evolutionary Programming. This yields an algorithm that is shown to be precise, accurate and specific. The new algorithm, Eve, is then modified to illustrate its potential in both lead optimisation and de novo drug design. These capacities, combined with its ability to predict the location of binding sites and the docking pose of a ligand, highlight the promise of computational methods in solving problems in many areas of biological chemistry.

Thesis (Ph. D.)--University of Oregon, 2007. / Typescript. Includes vita and abstract. Includes bibliographical references (leaves 67-80). Also available for download via the World Wide Web; free to University of Oregon users.

Diss. (sammanfattning) Umeå : Univ., 2003. / Härtill 4 uppsatser.

Thesis (Ph.D.) -- University of Adelaide, Dept. of Medicine, 1999. / Copy of author's previously published article in pocket on back end-paper. Includes bibliography (leaves 135-185).

Thesis (Ph. D.)--Michigan State University. Dept. of Microbiology and Molecular Genetics, 2006. / Title from PDF t.p. (viewed on Nov. 20, 2008) Includes bibliographical references. Also issued in print.

Thesis (Ph. D.)--Michigan State University. Dept. of Chemistry, 2006. / Title from PDF t.p. (viewed on Nov. 20, 2008) Includes bibliographical references. Also issued in print.

Thesis (M.S.)--Cleveland State University, 2008. / Abstract. Title from PDF t.p. (viewed on July 29, 2008). Includes bibliographical references (p. 66-73). Available online via the OhioLINK ETD Center. Also available in print.

Thesis (M.S.)--Cleveland State University, 2008. / Abstract. Title from PDF t.p. (viewed on July 29, 2008). Includes bibliographical references (p. 66-73). Available online via the OhioLINK ETD Center. Also available in print.