A detailed investigation into varied aspects of the Maillard reaction: In vitro studies on protein crosslinking and the role of amadoriase enzymes
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
The Maillard reaction represents a complex series of processes, initiated on reaction of a carbonyl moeity with an amine. Amongst the vast array of products (advanced glycation end products, AGEs) are protein crosslinks, which can foml on reaction of a-dicarbonyls with lysine and arginine residues, to form molecular bridges between proteins. It is thought that these crosslinks affect the native function of the protein, due to a structural change in the protein, and they have been implicated in the progression of some age-related diseases. Little is known about the precise amino acid requirements for protein crosslinking. Thus, this thesis sought to determine which amino acids were critical for crosslinking to occur. A lysine containing (arginine-free) protein and an arginine-containing (lysine-free) protein were reacted with three a-dicarbonyl compounds, methyl glyoxal, glyoxal and diacetyl. It was demonstrated, for the first time in vitro, that an arginine residue is not essential for protein crosslinking to occur. Only an N-terminus or a lysine residue is required for crosslinking with the three a-dicarbonyls tested. A model protein, RNase A, which contains both lysine and arginine, was also incubated with the three dicarbonyls. Parallel measurements on crosslinking and activity were performed in order to test the assumption that crosslinking affects the function of the protein. To establish whether the loss funetion was a direct result of the crosslinking process, or a co-incident event, two previously reported crosslinking inhibitors, aminoguanidine and 3,5- dimethylpyrazole-l-carboxamidine, were included. The results demonstrated that although crosslinldng of RNase A could be inhibited by these compounds in the presence of adicarbonyl, the activity of RNase A was not necessarily preserved. Inhibition of crosslinking may not preserve the funetion of the protein, as evidenced by the RNase A incubated with methylglyoxal and 3,5-dimethylpyrazole-l-carboxamidine. This loss in activity but inhibition of crosslinking may be explained by the presence of non-crosslinking AGEs that have formed on the protein, rendering it inactive. These results are significant when considering any approach to intervention of the Maillard reaction. Also examined was a recently-discovered enzyme, amadoriase I, which has been shown to reverse the early stages of the Maillard reaction under some experimental conditions. A time resolved assay was developed that allowed for accurate determination of ldnetic parameters and purification of amadoriasc I. By this method, amadoriase I was shown to have a Km of 11 μM for fructosyl propylamine, a known substrate, and a kcatlKm of 3.25x10 5 both of which differ from previous literature reports. This assay was also used to successfully measure the activity of two mutants, H357N and S370A, which were created in order to determine which amino acid residues within amadoriase I are critical for catalysis. Both mutants were catalytically inactive, demonstrating their importance in amadoriase I. Finally, a second assay was developed in order to determine whether amadoriase I could act on a mildly glycated protein. Amadoriase I is reported not to turnover glycated protein, the physiologically relevant substrate. However, past experiments have involved conditions that favour formation of a heavily glycated protein, which may be inaccessible to amadoriase I. The novel assay method involved direct assessment of activity of the protein substrate, before and after treatment with amadoriase I. The results confirmed previous studies that amadoriase I could not use glycated protein as a substrate, even in the case of a mildly glycated protein. The assay method was also employed to search for a potential amadoriasc enzyme from a Pseudomonad which had been isolated from raw mille. The crude extract, however, did not show any deglycating activity as assessed by this assay and the time-resolved assay. The studies undertaken in this thesis will inform design of therapeutic strategies aimed at inhibiting or undoing the damaging effects of Maillard chemistry in vivo.