Membranes provide a barrier to cells and organelles, and allow the selective transport of molecules between compartments. Membrane fusion is essential for organelle biogenesis as well as trafficking of molecules between cellular compartments. Membrane fusion is also required for the formation of the branched network of tubules that make up the Endoplasmic Reticulum (ER). One protein implicated in ER fusion is Atlastin, a dynamin like GTPase. Mutations in Atlastin-1, among others, cause Hereditary Spastic Paraplegias (HSP), a group of neurological disorders that cause progressive weakness of lower extremities. We have shown that the C-terminal tail of atlastin is necessary for membrane fusion. The requirement of the C-terminal tail can be partially abrogated in an unstable lipid environment. This implies that the C-terminal tail of Atlastin plays a role in perturbing the lipid bilayer to allow membrane fusion. Understanding the molecular details of how Atlastin drives membrane fusion may help elucidate the pathogenesis of HSP. Intracellular fusion at the plasma membrane is SNARE mediated and regulated by Sec1p/Munc18 (SM) proteins. Increased rate of glucose transport into fat and muscles cells by translocation of glucose transporter GLUT4 in response to insulin is a SNARE regulated fusion process. Recent reports have linked Munc18c and Syntaxin4 with obesity and Type 2 diabetes. We characterized the function of Munc18c, an SM protein, in regulating GLUT-4 containing vesicle fusion with the plasma membrane. We have shown that Munc18c directly inhibits membrane fusion by interacting with its cognate SNARE complexes. Characterization of membrane fusion in a minimal system as the in vitro liposome fusion assay offers a powerful tool with which to finely dissect the mechanistic basis of SM protein function.

The biogenesis and maintenance of the entire endomembrane system is dependent upon membrane fusion proteins. Mounting evidence indicates that the integral membrane GTPase Atlastin is a membrane fusion protein involved in the homotypic fusion of the endoplasmic reticulum (ER) membrane suggesting a role in the biogenesis and maintenance of ER structure. I helped show that recombinant Drosophila atlastin is able to promote the fusion of synthetic membranes in vitro and that this fusion is dependent upon atlastin GTPase activity. The structure-function experiments presented here assist in elucidating domains required in the mechanism of atlastin mediated membrane fusion. ER homotypic fusion is dependent upon the self-association of Atlastin subunits in adjacent membranes to bring the bilayers into close molecular contact. Atlastin dimerization occurs in the presence of GTPγS but not GDP and stable dimerization is dependent upon a juxtamembrane middle domain three-helix bundle (3HB). The atlastin GTPase domain and 3HB form a potent soluble domain inhibitor of atlastin homotypic fusion, while the GTPase domain alone shows little inhibition. Designed GTPase domain mutations show that GTP binding and atlastin dimerization is insufficient to support fusion without GTP hydrolysis. Additionally, domain analysis of atlastin reveals that the C-terminal cytoplasmic domain of atlastin is absolutely required for membrane fusion, possibly through a protein-lipid interaction of an amphipathic alpha-helix. Genetic lesions in the human Atlastin-1 gene, SPG3A, result in a form of autosomal dominant hereditary spastic paraplegia (HSP). A better understanding of Atlastin function should lend significant insight into normal ER biogenesis and maintenance, as well as the pathology of human disease.