|dc.description.abstract||Stereochemistry, defined as being the study of the spatial behaviour of atoms in molecules and complexes, has become increasingly important in the pharmaceutical sector. In particular, the existence and pharmacological effects of pharmaceutical enantiomers has given rise to significant research efforts, aiming not only to study the biological effects of enantiopure products but also establishing methods of obtaining chiral molecules in their enantiopure form.
One of these methods, involving the use of enantioselective hydrolytic enzymes, has received considerable attention. Utilising the catalytic potential of epoxide hydrolase, an enzyme found in a wide variety of living organisms, highly reactive racemic epoxides may be obtained in their enantiopure form together with their corresponding vicinal diols. These two enantiopure products may themselves be biologically active or, in turn, serve as precursors during the synthesis of high value enantiopure pharmaceuticals.
The project described in this thesis had four main objectives. Firstly, to optimise the previously established enantioselectivity exhibited by whole cells of Rhodotorula glutinis towards terminal epoxides using styrene oxide as a model substrate. Successful completion of this goal illustrated that a pH of 7.2, optimal temperature of 15 °C and an initial substrate concentration of 50 mM yielded the most promising reaction. Further studies also showed that this reaction may be run as a salt free process, reducing costs and following a more environmentally friendly approach.
One of the major challenges during the hydrolysis of epoxides is the fact that these substrates tend to be highly insoluble in water. The enzyme, on the other hand, is far more active in its native aqueous medium. For this reason the second objective was to investigate not only the effects of two commonly used water miscible organic solvents (DMSO & DMF), but also the possibility of utilising solubility enhancing cyclodextrins for this reaction. It was found that hydroxypropyl-β-cyclodextrin (HPB) had a far greater solubilsation potential than the two solvents investigated. In addition, HPB was found to have the least negative effects on the reaction and was therefore shown to be a viable alternative to the use of solubility enhancing organic solvents.
The third goal of this project was to investigate the possibility of scaling the reaction up to a bench scale bioreactor, thereby not only investigating the factors that influence such a scale-up, but also to investigate the economic viability of using whole cells in a bioreactor to catalyse the enantioselective hydrolysis of a terminal epoxide. Initial reaction rate, enzymatic stability, reaction enantioselectivity and optimal cell/buffer ratios were all investigated and the optimal conditions reported. Recycling the whole yeast cells by means of micro-filtration was found to be ineffective.
Finally, as downstream processing of the products of a bioreaction contribute significantly to the costs involved, the use of a selective liquid-liquid extraction step was investigated, not only to separate the products from the reaction media, but also to simultaneously separate the residual epoxide and produced diol from one another. It was established that solvents with low log P values were best suited for the simultaneous extraction of both the epoxide and diol from the reaction medium. Solvents with high log P values, however, were useful for the selective extraction of the residual epoxide from the reaction medium.||