Degree Name

Doctor of Philosophy


School of Biological Sciences - Faculty of Science


The incidence of protein conformational disorders is a problem affecting the aging population and is an increasing cost on the health system. These diseases involve the breakdown of control mechanisms within and outside of cells leading to accumulation of unfolded and misfolded proteins. The control of protein folding, both inside and outside the cell, is important to maintain the integrity of all proteins. Dysfunctions in this control may lead to the development of protein conformational disorders. Intracellularly the mechanisms monitoring the correct translation, modification, assembly and transport of synthesised proteins are very well studied. However, in the harsh extracellular environment, mechanisms to control protein folding need further investigation. Extracellular chaperones have recently been proposed as an important part of extracellular quality control for protein folding. The ubiquitous glycoprotein clusterin has been implicated in this extracellular quality control mechanism. A primary aim of the work described in this thesis was to examine the relationship between clusterin structure and function. This was done using three different approaches; characterisation of mutant clusterin expressed in transfected CHO cells; characterisation of enzymatically deglycosylated clusterin; and characterisation of bacterially expressed recombinant clusterin domains. Initially a panel of mutants was designed to disrupt predicted features of the protein, consisting of five point mutants and five truncation mutants. The clusterin molecules were expressed in CHO cells and were proteolytically processed with two subunits and appeared to be normally glycosylated. The mutant clusterin molecules appeared to be chaperone-active to some degree also, evidenced by their copurification with non- specific media proteins. In a rolling culture proteins in the media would be exposed to sheer stress and one of the actions of clusterin is to bind to exposed hydrophobicity on stressed proteins. A decrease in the amount of contaminating proteins copurified with clusterin when static cultures were used instead of rolling cultures supports this argument. In an effort to understand the importance of the glyscosylation of clusterin, the conjugated sugars were enzymatically removed to determine their effects on clusterin function. The success of the deglycosylation process was first verified using SDSPAGE and mass spectrometry. Structural features of the deglycosylated molecule were tested, with no change seen in the secondary structure of the deglycosylated protein compared to the native protein. A shift to slightly acidic pH produced no change in secondary structure as assessed by circular dichroism spectroscopy. Exposed hydrophobicity was seen to increase with the removal of the sugars but the Kd of the interaction with bisANS was not significantly different. At slightly acidic pH both the wild type and deglycosylated molecules exposed a greater amount of hydrophobicity to solution. Binding to the known clusterin ligand megalin was also tested with a similar binding affinity seen for both molecules, however the removal of sugars increased the rate at which the molecule bound to and dissociated from megalin. Binding of the deglycosylated molecule to known native ligands (IgG and GST) and stressed protein ligands (lysozyme and BSA) was not affected by the absence of sugar moieties. The chaperone activity of the deglycosylated molecule was also not different to the wild type molecule. Following the determination that the sugars appeared to have little effect on the activity of clusterin, a series of domains was cloned into bacterial expression vectors. Two domains (αN: D23-E112 and βC: R306-E449) were purified and their structure and function was examined. In the presence of a polar solvent both domains exhibited alpha helical secondary structure as expected based on amino acid analysis, though in aqueous buffer the domains had very little or no significant structure. The domains alone did not have any effect on the heat stressed precipitation of citrate synthase. Neither domain alone bound to heat-stressed lysozyme, however the αN domain bound to the native protein ligand IgG and this was enhanced at slightly acidic pH, as seen with native clusterin. The αN domain also bound with high affinity to LRP indicating that this domain of clusterin is likely to be involved in binding interactions of the entire molecule. Interaction between the αN and βC domains was also observed and further investigation is required to determine binding and chaperone activity of the interacting domains. Understanding the functional relationship between the structure and function of clusterin is important in understanding the role of clusterin in protein folding control. This knowledge may also be important in the development of novel treatments for the prevention or slowing of protein conformational diseases. The work in this manuscript showed that the deglycosylated form of clusterin is as potent a chaperone as the entire clusterin molecule. This indicated that a bacterially expressed form of clusterin may be used to further investigate the function of the molecule. Recombinant domains of clusterin were also studied to localise the regions of the molecule important in clusterin binding and chaperone function. Further investigation is needed, however the use of bacterially expressed recombinant proteins allows the potential development of recombinant chaperone-active clusterin which may ultimately be used as a therapeutic for protein conformational diseases. This research has gone some way to showing the best methods of characterising the structural relevance to the binding and chaperone activity of clusterin, and to providing a knowledge base for future investigations.

02Whole.pdf (5232 kB)