Phylogenomic and structure-function relationship studies of proteins involved in EBV associated oncogenesis

Hussain, Mushtaq (2013) Phylogenomic and structure-function relationship studies of proteins involved in EBV associated oncogenesis. PhD thesis, University of Glasgow.

Abstract

This study covers the investigation of evolutionary and structure-function relationship aspects of several cancer related proteins. One part of the study deals with the investigation of a critical protein of Epstein-Barr Virus (EBV) the Nuclear Antigen 1 (EBNA1), and its interactions with different host proteins. One of these host proteins is a member of a large gene family, encoding ubiquitin specific proteases (USP), known as USP7. The second section of the thesis deals with the molecular evolution of the USP gene family. Another set of cellular proteins deregulated during EBV associated oncogenesis are members of the glycoside hydrolase (GH18) family. Their phylogenetic relationships and protein structures were investigated in the third section of this thesis.
EBNA1 is the only EBV protein that consistently expressed in all latent forms of the EBV infections. The protein is involved in the genome maintenance and a substantial body of evidence suggests that it has a role in EBV associated oncogenesis. In this study, full length molecular models of the EBNA1 protein were generated using the programmes, I-TASSER, MOE and Modeller. The best models were selected on the basis of plausibility in structural and thermodynamical parameters and from this models of EBNA1 homologues of primates lymphocryptoviruses (LCVs) were generated. The C-terminal DNA binding and homodimerisation domain was predicted to be structurally similar between different LCV EBNA1 homologues, indicative of functional conservation. The central glycine alanine repeat (GAr) domain was predicted to be primarily composed of α helices, while almost all of the protein interaction region was found to be unstructured, irrespective of the prediction approach used and sequence origin. Predicted USP7 and Casein kinase 2 (CK2) binding sites and GAr were observed in the EBNA1 homologues of Old World primate LCVs, but not in the marmoset homologue suggesting the co-evolution of both these sites. Dimer conformations of the EBNA1 monomer models were constructed using SymmDock, where the C-terminal tail was predicted to wrap around the proline rich loop of another monomer, possibly contributing to dimer stability. This feature could be exploited in therapeutic design, hence an inhibitor peptide was designed and a preliminary evaluation was conducted to explore its ability to inhibit EBNA1 function in cell survival. The peptide array libraries of EBNA1 were used to investigate the binding regions and critical contact points between EBNA1 and partner proteins. Human EBP2 and USP7 proteins were expressed in bacteria and probed on the EBNA1 array. The data confirm the previously known binding region for EBNA1-EBP2 and EBNA1-USP7 interactions. In addition further information was gained regarding the critical contact residues and the potential role of phosphorylation of serine residues of EBNA1 in its binding with EBP2 and USP7.
The human genome encodes nearly 100 USPs which contribute to regulate the turnover of cellular proteins. These homologues are divided into 16 paralogous groups, all sharing a characteristic peptidase C19 domain. Evolutionary relationships between these homologues were explored by datamining and the phylogenetic reconstruction of peptidase C19 domain sequences. The data reveal an ancient relationship between the genes, with expansion occurring throughout the course of evolution, but particularly at the base of the vertebrates, at the time of the two whole genome duplications. A comparison between the phylogenetic architecture and protein interaction networks suggests the parallel emergence of many molecular pathways and the associated USPs.
The GH18 gene family includes chitinases and related non catalytic proteins. Most mammals encode at least three chitinases (CHIT1, CHIA/AMCase and CTBS), as well as several homologues encoding catalytically inactive chitinase-like proteins or chilectins. Phylogenomic analysis shows that the family has undergone extensive expansion, initiating with a duplication event at the root of the vertebrate tree, resulting in the origin of the ancestors of CHIT1 and CHIA. Two further duplications of ancestral CHIA predate the divergence of bony fishes, one leading to a newly identified paralogous group (we have termed CHIO). In tetrapods, additional CHIA duplications predate and postdate the amphibian/mammalian split and relics of some exist as pseudogenes in the human genome. Homology modelling of structurally unresolved GH18 homologues in mouse and human was conducted using Modeller and I-TASSER. All resolved and predicted structures share a TIM barrel (β/α)8 and α+β domain. A central ligand binding cavity was also found in all GH18 homologues. The variation in size and shape of different paralogous proteins, indicate the difference in their ligands specificity and in turn potential functions.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: EBV, EBNA1, Ubiquitin Specific Peptidase, Chitinase, Oncogenesis
• Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Molecular Biosciences > Molecular Biosciences
• Supervisor's Name: Wilson, Dr. Joanna B.
• Date of Award: 2013
• Depositing User: Dr Mushtaq Hussain
• Unique ID: glathesis:2013-5357
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 06 Nov 2014 09:16
• Last Modified: 06 Nov 2017 13:16
• URI: https://theses.gla.ac.uk/id/eprint/5357