In eukaryotic cells, it has been well established that large GTPases of the dynamin superfamily are important drivers of membrane curvature and constriction during membrane fusion and fission processes such as clathrin-coated endocytosis (1, 2). More recently, dynamins have been identified in a number of bacteria including Bacillus subtilis, Nostoc punctiforme and Escherichia coli (3-5). However, despite their putative identification in over 1500 bacterial species to date, minimal understanding of their biological function has been established in bacteria (6). One such bacterial dynamin, a previously characterised enterotoxigenic E. coli virulence factor, LeoA has been shown to be involved in outer membrane vesicle formation for the release of heat-lable LT toxin (5, 7). Upstream of LeoA there were two other dynamins identified, LeoC and LeoB, encoding a full-length dynamin only when expressed together. This is a seemingly split ORF, a first for dynamin (5). The closest homologs to these genes are found in the gastric carcinogenic pathogen Helicobacter pylori and we have termed them dlp1a, dlp1b and dlp2 respective to their LeoCBA homologs. Therefore, this thesis endeavoured to investigate for the first time the presence of dynamin-like proteins (DLPs) in H. pylori, an important and surprisingly understudied human pathogen that is known to be dependent on membrane dynamics for virulence (8-10). It is also a pathogen that survives and thrives in one of the harshest environments, the acidic environment of the human stomach, where membrane integrity is essential to its protection.
The investigation into the presence of DLPs in H. pylori undertaken in this thesis was achieved through a diverse series of aims. Firstly, a bioinformatic survey undertaken to determine the conservation of the DLP genes in the H. pylori genome highlighted conservation greater than 90% in H. pylori, unprecedented for any known bacterial dynamins. For the first time, in vivo expression of dlps was shown in H. pylori using detection of the mRNA transcripts as well as the protein products via Western blotting. Interestingly, a single-nucleotide indel was identified in a region of overlap between dlp1a and dlp1b that differs between sub-isolates of H. pylori, leading to either the split or the joined forms of dlp1 (the full-length gene). Furthermore, the analyses of dlp1 and dlp2 mRNA and their protein products were consistent with this genotypic variation and might represent a novel regulation mechanism.
A preliminary investigation into the biochemical features of the DLPs such as localisation, self-interaction and lipid binding capabilities was commenced. In strains containing the fused dlp1, Western blotting and immuno-fluorescence microscopy suggested that Dlp1 is membrane-associated, whereas Dlp2 is approximately equally distributed between the cytosol and membrane. Also, initial studies suggested that Dlp2 may dimerise, interact with liposomes, as well as showing potential for interaction with its operon partner Dlp1a; all fundamental dynamin features. The gene arrangement and presence of a predicted transmembrane helix in Dlp1, similar to the membrane-binding domain of eukaryotic dynamin, suggested that these proteins might act as a hetero-complex involved in bacterial membrane dynamics.
Finally, the previously uncharted function of DLPs in H. pylori was explored using both direct and global approaches. Knockout of the dlp operon in H. pylori caused a high proportion of cells to develop a compromised cell membrane in dye-penetration assays, and minimal capacity to overcome acidic exposure as well as minimal growth in acidic conditions. Utilising a global proteome approach, SWATH-MS, one of the most comprehensive quantitative proteomics study to date of the acid response in wild-type and knock-out strains was achieved. This analysis highlighted the potential role that dynamins may have in acid adaptation, with proteins involved in processes such as motility and cytoplasmic neutralisation detected in higher abundance in the absence of dynamin. Taken together, this suggests that H. pylori, with a compromised membrane in the absence of dynamin, is attempting to overcome the higher influx of acid into the cell by “escaping” to conditions that are more neutral.
Overall, this thesis provides the first evidence that this important human pathogen, H. pylori utilises membrane remodelling events in some capacity driven, by dynamins, to maintain membrane integrity. A feature that is necessary to ensure continual pathogenesis in its niche environment the stomach of over half the world’s population.