Studies on collagen post-translational modifications

• Main Author: Langley, Gareth William
• Other Authors: Schofield, Christopher J. ; Claridge, Timothy D. W. ; Jung, Manfred ; Carr, Andrew J.
• Published: University of Oxford 2017
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.740832

The collagen proteins are the principal component of the extracellular matrix and are the most abundant protein family in humans. Therefore, collagens form the major protein component of bone, tendon and cartilage. During collagen biosynthesis, proline and lysine residues are enzymatically converted to (2S,4R)-hydroxyproline, (2S,3S)-hydroxyproline and (2S,5R)-hydroxylysine by the collagen prolyl-4-hydroxylases (C-P4Hs), prolyl-3-hydroxylases (P3Hs) and procollagen-lysine, 2-oxoglutarate 5-dioxygenases (PLODs) respectively. The C-P4Hs, P3Hs and PLODs are members of the Fe^II and 2-oxoglutarate dependent oxygenase (2OG oxygenase) superfamily. Despite being amongst the first 2OG oxygenases to be identified, the collagen hydroxylases are poorly understood from structural and biochemical perspectives, a consequence of inefficient recombinant production methods. Initial efforts focussed on developing a robust expression system for recombinant C-P4H. An E. coli expression system was pursued to allow for relatively large quantities of protein to be produced, as required for structural studies. A number of novel expression constructs were generated and used in small-scale expression trials. However, attempts to isolate purified C-P4H tetramer have ultimately proved unsuccessful. A prolyl hydroxylase from Paramecium bursaria Chlorella Virus I (vCPH), was pursued as a model collagen prolyl hydroxylase; production of vCPH in E. coli affords large quantities of high purity recombinant protein for subsequent analyses. vCPH catalyses the formation of (2S,4R)-hydroxyproline at proline 3 and proline 5 of the short 8-mer peptide PAPKPAPK and kinetic analyses have revealed that the ascorbate dependence of vCPH-catalysis at these two positions differs significantly, indicating that ascorbate dependence of the collagen prolyl hydroxylases varies dependent upon the residues adjacent to the target residue. A vCPH inhibition assay was developed and a library of potential inhibitors was screened. This screen included compounds that inhibit a different subfamily of 2OG dependent prolyl-4-hydroxylases, the hypoxia inducible factor prolyl hydroxylases (HIF PHDs), which also catalyse the formation of (2S,4R)-hydroxyproline. The studies revealed that inhibitors of the HIF PHDs that are currently in clinical trials also inhibit vCPH. However, analysis of novel HIF PHD inhibitors and inhibitors of the prolyl-3-hydroxylase OGFOD1 indicate that achieving selectivity between these prolyl hydroxylases is possible. The inhibition studies have been complemented with amino acid analyses of collagen samples. No evidence for the existence of hydroxylated residues other than (2S,4R)-hydroxyproline, (2S,3S)-hydroxyproline and (2S,5R)-hydroxylysine was acquired. These amino acid analysis approaches may be taken further to analyse collagen samples derived from tissue to assess hydroxylation levels in diseased tendon. Additionally, metathesis-derived lysine analogues have been synthesised and used to analyse 2OG-dependent histone demethylase (KDM) catalysis. These analogues have revealed new modes of activity with KDM3A and provided insight into the charge and conformational requirements of 2OG-dependent KDM catalysis. Overall, the work presented in this thesis has provided insight into the biochemistry of a collagen prolyl hydroxylase and demonstrated the potential for the development of selective inhibitors of prolyl hydroxylases. Additionally, novel reactions catalysed by histone demethylases have been observed.