Abstract
The CAP protein superfamily comprises: the Cysteine-rich secretory proteins, Antigen 5 and the Pathogenesis Related group 1 proteins. Their diverse biological functions include mammalian reproduction (e.g. CRISP1-4) and ion channel regulatory activity (e.g. Natrin, Stecrisp), plant immune responses (P14a) and the processing of conotoxins (Tex31). CAP proteins have also been identified in venoms (Triflin, Vesv5) and cancer (GLIPR1). The CAP proteins are thought to share common functions based on their highly conserved sequence motifs (namely CAP1-4) and disulfide bonds in the core of most CAP proteins. The spatial arrangement of amino acids within the large cleft of the CAP proteins suggests a possible enzymatic active site. Tex31, a cysteine-rich protein from the venom of Conus textile, was shown to be a substrate specific protease and is the only CAP protein identified with any enzymatic function. Tex31 is a multiple domains protein and its conserved CAP domain has been hypothesized to possess the catalytic activity. Over a hundred of CAP proteins have been identified but only six of them were successfully expressed using heterologous systems. Of these six, just one protein GAPR1 containing the CAP domain was solubly expressed in Escherichia coli and had its crystal structure determined. It has been difficult to express soluble CAP proteins heterologously. Therefore, this thesis aimed to (i) develop the expression protocol for Tex31 using GLIPR1 as the trial protein in Escherichia coli and Saccharomyces cerevisiae systems, (ii) to solubly express the recombinant Tex31 in E. coli or S. cerevisiae using the developed protocol in (i) and to produce enough correctly folded protein for Tex31 X-ray crystallographic analysis and protease activity assay, and (iii) to investigate the function of the CAP domain of Pry3p from S. cerevisiae using confocal microscopy.
The GLIPR1 open reading frame was cloned into the pQE30 (N-terminal 6× His-tag) and pQE60 (C-terminal 6× His-tag) vectors in the E. coli strain M15[pREP4] and into the PDR5 locus of S. cerevisiae. The in frame and correct DNA sequence of GLIPR1 in pQE vectors and PDR5 locus were confirmed by DNA sequencing. The expression of GLIPR1 was not detected using SDS-PAGE gel and western blot analysis. Disruption in the T5 promoter region of pQE30 vector was identified by DNA sequencing but it remained unknown as to why GLIPR1 protein did not express in pQE60/E. coli and S. cerevisiae systems. Nevertheless, the main aim of this section which was to develop the expression protocol for Tex31 had been achieved.
The E. coli expression protocol developed for GLIPR1 in (i) was applied to Tex31. Multiple constructs of Tex31 were cloned into pQE30 and pQE60 systems. Initial expression of Tex31 resulted in the formation of inclusion bodies. Refolding experiments were attempted but the refolded Tex31 was unsuccessful due to Tex31 instability. Similar Tex31 constructs were expressed in E. coli strain DH5α as soluble protein using a maltose-binding protein (MBP) as a fusion partner. MBP-Tex31 chimeras were affinity purified using amylose resin. Further analysis revealed that these chimeras formed large multimeric soluble aggregate complexes partially due to the mis-formation of disulfide bonds of the recombinant Tex31. Additionally, small heat-shock proteins IbpA/B were found in the affinity purified MBP-Tex31 and were deemed to be a marker of misfolded protein. Removal of specific disulfide bonds in recombinant Tex31 constructs plus the addition of the mild detergent n-dodecyl-β-D-maltoside and sonication prior to size-exclusion chromatography eliminated IbpA/B from the sample. Approximately 50% of the chimeric constructs were successfully recovered in the soluble and monodisperse form. Crystallization of MBP-Tex31 mutants using hanging-drop vapour-diffusion gave non-diffracting, 10 – 30 μm diameter crystals within 3 – 6 weeks. None of the recombinant Tex31 constructs tested thus far have shown any detectable enzymatic activity. Nevertheless, the developed method to express the mutant Tex31 fusing with MBP and to purify the chimeras using mild detergent and sonication before size-exclusion chromatography should be considered to express any other CAP proteins.
In attempts to investigate the function of the CAP domain, the behaviour of Pry3p, a CAP protein from S. cerevisiae was studied under the influence of yeast α-mating pheromone. This thesis aimed to detect the influence of mating pheromone on the CAP domain expression and localization of Pry3p. Initially, the S. cerevisiae strain ADΔ MATα was converted to strain ADΔ MATa. Wild-type PRY3 was successfully knocked out from the strain ADΔ MATa. A number of mutant PRY3 constructs containing 6× His-tag, FLAG-tag and monomer red fluorescent protein, were then generated by recombinant PCR and transformed into PRY3 locus. While the latter could be directly detected using confocal microscopy, the former two tags required specific primary antibody conjugated with fluorophores. The correct colonies that contained in frame and precise DNA sequences were identified by PCR and DNA sequencing. However, the confocal observation did not produce any significant signal to locate the Pry3p mutants and western blot analysis did not detect any mutant Pry3p expression level. The extensively modified PRY3 constructs at the PRY3 locus possibly caused cells to develop an endogenous mechanism to down regulate the expression of mPry3p(s) or to degrade them.
This thesis provides a recombinant E. coli expression system which hopefully can be applied to solubly express and purify other CAP proteins. Nevertheless, alternative expression systems such as Pichia pastoris and mammalian expression system should also be considered. The overall biological function of the CAP protein still remains elusive and so different research approaches to study their functions (in addition to enzyme assays) are necessary e.g. protein-protein interaction or cellular active transportation.