Doctor of Philosophy (PhD) / This thesis primarily details the application of high-level ab initio quantum chemistry techniques in order to understand aspects of free-radical mediated protein oxidation. Traditionally, product analysis and electron paramagnetic resonance (EPR) spectroscopy are the primary means for elucidating the chemistry of protein oxidation. However, in experiments involving relatively small proteins reacting with a controlled radical-flux, a vast array of compounds can be produced, which are often difficult to analyse. Quantum chemical techniques on the other hand, can calculate the properties of any particular species directly, without suffering from the problems associated with experiment, such as side-reactions and chain processes. The results presented in this thesis are aimed at elucidating mechanistic details of protein oxidation, which might otherwise be difficult to probe experimentally. Chapter 1 gives an overview of the free-radical hypothesis of disease and ageing. Protein-derived radicals can undergo a variety of reactions, with the particular reaction that occurs depending on numerous aspects. Many types of reactions have been identified through radiolysis experiments of amino acids, and these are detailed in this chapter. In addition, the key reactive species are characterized and their different chemistries explained. Chapter 2 details the theoretical tools used throughout this thesis. Species with unpaired electrons (radicals) present unique problems for quantum chemistry to handle, thus an appropriate choice of theoretical technique is needed. The approach taken in this thesis is to use high-level compound methods, many of which have been directly formulated to give improved results for radical species, to provide benchmark quality results by which other less demanding techniques can be assessed. During the course of this study, it became apparent there was a void in the armoury of tools that could be used for the theoretical chemistry calculations. Chapter 3 details the formulation of a new tool in an attempt to fill this gap. Historically, the formulation of this new procedure came after much of the work in this thesis had been carried out. Thus, for the study of many of the reactions of this thesis the new method has not been used. However, it is most appropriate to place its formulation after summarizing the current status of techniques in common use today. Chapters 4 and 5 detail computations carried out on models of peptides containing backbone carbon- and nitrogen-centered radicals. A number of different theoretical techniques are used in these chapters, ranging from the highly accurate and computationally intensive to the less reliable and less demanding. The highly accurate techniques are used to gauge the accuracy of the other less demanding theoretical techniques so that the latter can be used with confidence in larger systems. Not only is the choice of theoretical technique important but also the judicious choice of model is essential. With this in mind, models are incrementally built until convergence of the particular property of interest is reached. Chapters 6 and 7 detail the calculations of β-scission reactions of alkoxyl radicals, which are a particular class of reaction known to occur on peptide backbones. Alkoxyl radicals are particularly difficult for theory to describe correctly. Therefore, Chapter 6 extensively assesses and then identifies the theoretical methods needed to portray them. Chapter 7 uses the techniques identified in the previous chapter in order to predict how the preference for a particular type of β-scission reaction changes.
| Identifer | oai:union.ndltd.org:ADTP/283275 |
| Date | January 2006 |
| Creators | Wood, Geoffrey Paul Farra |
| Publisher | University of Sydney., School of Cemistry |
| Source Sets | Australiasian Digital Theses Program |
| Language | en_AU |
| Detected Language | English |
| Rights | The author retains copyright of this thesis., http://www.library.usyd.edu.au/copyright.html |
Page generated in 0.0022 seconds