Analysis of Protein-Protein Interactions in the Shell of Herpes Simplex Virus Type 1 (HSV-1) Capsids

Bain, Sharon Joanne Macnab (1999) Analysis of Protein-Protein Interactions in the Shell of Herpes Simplex Virus Type 1 (HSV-1) Capsids. PhD thesis, University of Glasgow.

Abstract

Much is known about the structural appearance of the HSV-1 capsid however, the aim of this project was to build on existing knowledge by studying the interactions and properties of individual capsid proteins, principally VP26, but also VP5 and preVP22a. As good antibodies to many capsid proteins were not available, a short amino acid sequence tag, derived from the HCMV pp65 protein was inserted. This approach provided a tool for monitoring VP26, VP5 and VP19c in a series of interaction studies using an anti- pp6S monoclonal antibody. Extensive immunofluorescence studies of plasmid transfection experiments, showed that VP19c had an intrinsic ability to localise to the nucleus. Furthermore, nuclear localisation of VP26 was only observed when VP5 was present together with either VP 19c or preVP22a (see Chapter 3). In an attempt to produce an unrelated tag, a short amino acid sequence derived from the CAT enzyme and recognised by an anti-CAT monoclonal antibody, was mapped and characterised. An oligopeptide of this sequence was synthesised and used in an unsuccessful attempt to raise monospecific antisera in rabbits for use in double labelling experiments. The VP26 ORF was sub-cloned into a bacterial expression vector to allow more detailed studies of the protein to be performed. The pET expression plasmid offered a system where high quantities of protein could be produced, readily purified (using a 6xhis tag and nickel affinity chromatography) and easily manipulated to produce mutant proteins. Although the VP26 tagged protein was induced to high levels, it was extremely insoluble. A successfiil procedure was developed by which VP26 was purified in a denatured form and renatured by dialysis against buffer. This process consistently produced concentrations of purified VP26 protein at 0.5-1.0mg/ml (see Chapter 4). The functionality of the purified renatured VP26 was demonstrated by its ability to assemble correctly on capsids. An experimental protocol was developed for attaching the tagged VP26 to capsids lacking VP26. Cryo-EM analysis of these capsids confirmed that the VP26 protein had attached at the correct location and in the expected orientation. forming a star shaped ring on the tips of the VP5 hexons. This provided proof that the bacterially expressed protein mimicked the properties of the natural VP26 protein and that the extra sequences, contributed by the epitope and 6xhis tags, did not interfere with the binding of VP26 to VP5. An area of extra mass that was detected in the hexons with the tagged VP26 protein attached, was considered to be indicative of the tag sequence. As the tag had been inserted into the N-terminus of the UL35 ORF, this identified the probable location of the N-terminus of VP26 (see Chapter 5). To analyse sequences important for the interaction of VP26 with VP5 an ELISA assay was developed. Initially, binding of purified VP26 to capsids which lacked the endogenous VP26 were examined. Subsequently, the assay was refined by replacing the VP26 minus capsids with purified and soluble VP5, expressed from a baculovirus vector. Using this assay, it was possible to map the points of contact between the two proteins. Ten overlapping oligopeptides, spanning the entire UL35 ORF, were made and tested for their ability to block binding of VP26 to VP5 in an inhibition type ELISA assay. Several of the oligopeptides had an inhibitory effect, most notably those containing sequences between amino acid residues 31 and 70 in the UL35 ORF (see Chapter S). The solubility of the VP26 tagged protein was not high enough to allow X-ray crytallography or NMR studies. Therefore a protein predict computer package was used to try and gain an insight into the secondary structure of the protein. The protein predict model was used to guide the construction of a series of mutant VP26 proteins. Although the prediction has since proved inaccurate, through CD analysis, the mutant proteins produced have been extremely useful. Experiments using VP26 proteins, with three glycine insertions at different locations, showed that insertion in the region between amino acid residues 40 and 60 on the UL3S ORF (a region incorrectly predicted to consist largely of alpha helix) affected the binding of VP26 to capsids. These insertions mapped within the region of VP26 previously shown to be important for binding in the oligopeptide inhibition studies, suggesting that this region contains sequences involved in the interaction with VP5. To account for the presence of VP26 on hexons but not pentons (which are also formed by VP5) it had been suggested that VP26 formed 6-mers prior its binding to capsids. Analysis of the oligomeric status of purified VP26 suggested that the protein was present as monomers and dimers (in a ratio of 3:1) but not 6-mers. The implications of this result, with regard to capsid assembly models, is discussed in Chapter 5.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: Adviser: Frazer J Rixon
• Keywords: Virology
• Date of Award: 1999
• Depositing User: Enlighten Team
• Unique ID: glathesis:1999-76350
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 19 Nov 2019 15:23
• Last Modified: 19 Nov 2019 15:23
• URI: https://theses.gla.ac.uk/id/eprint/76350