Characterisation of Proteins from Grevillea robusta and NMR Studies of the Serine Protease Inhibitor

Abstract

Proteins that recognise the sugar surface structures on cells have an enormous potential to be used as tools in the characterisation of these structures. A group of proteins, called lectins, have been identified that can bind to carbohydrate complexes on the receptors of cells. The crude extract from Grevillea robusta seeds was found to contain lectin-like proteins that were different from most other lectins, as they would specifically target the receptors of white blood cells and not those found on red blood cells. Therefore, the lectin isolated from G.robusta could be used as a tool to identify the specific surface structures on white blood cells. The lectin was isolated using affinity chromatography where a complex (oligosaccharide) matrix was used. Agglutination, binding and sugar inhibition assays confirmed the isolated protein was a lectin. The lectin was found in low amounts (up to 5% of the total protein content) within the seeds of G.robusta. As a result of this low yield, the identification of the lectin by PAGE was difficult because the levels of protein were beyond the detection limit of the commercial staining reagents. The lectin was called the GR2 protein and was characterised as a monocot mannose binding lectin based on its sugar specificity for only mannose. A serine protease inhibitor was isolated from the seeds of G.robusta using two different chromatography methods, reverse phase HPLC (GR1.HPLC) and gel filtration chromatography (GR1.GF). Ion exchange chromatography was used to initially separate the proteins in the crude extract and the fraction containing the GR1 protein was further purified using reverse phase HPLC (GR1.HPLC). N-terminal sequencing results of the GR1.HPLC protein, showed evidence of proteolytic cleavage during the extraction process, which lead to the second purification method being established. Protease inhibitors were added to the buffers prior to being purified by gel filtration chromatography, which resulted in the GR1 protein being isolated from the crude extract without the presence of the contaminating protein. Mass spectroscopy identified the molecular weight of the GR1 protein to be 6669Da and the full amino acid sequence was derived by cDNA techniques. Sequence alignment studies of the GR1 protein showed significant similarities with the Bowman-Birk inhibitor. The positioning of the cysteine residues were conserved throughout the Bowman-Birk superfamily, however these residues were not conserved within the GR1 protein. Competitive inhibition assays on the GR1 protein revealed the protein could inhibit both trypsin and chymotrypsin at similar levels to that seen for the Bowman-Birk inhibitor. Therefore, the GR1 protein was characterised as a member of the Bowman-Birk superfamily of serine protease inhibitors. The three-dimensional structure of the GR1 protein was determined using two-dimensional NMR spectroscopy. Computer programs such as XEASY, DYANA and SYBYL® were used to tabulate the information taken from the 2D experiments, generate structures and minimise these structures respectively. The solution structure of the GR1 protein was found to contain a region of antiparallel β-sheet structure that corresponded to the trypsin binding site and the remainder of the protein consisted of loops and turns that were held together by disulfide bridges (the chymotrypsin-binding region). Structural similarities between the GR1 protein and the Bowman-Birk inhibitor existed only in the trypsin-binding site of the Bowman-Birk inhibitor. The GR1 protein is the first member of the Proteaceae family to be characterised as a Bowman-Birk inhibitor. This thesis outlines the isolation and biochemical characterisation of the two proteins found within Grevillea robusta and also describes the steps involved and results obtained in determining the three-dimensional structure of the GR1 protein.