Mechanistic Insights into Homotypic Fusion of ER Membranes by the Atlastin GTPase

A variety of diseases, including Hereditary Spastic Parapalegia (HSP), are associated with defects in Endoplasmic Reticulum (ER) morphogenesis, highlighting the significance of forming and maintaining proper ER structure in the context of human health. While the overall shape and structure adopted by the ER is mainly influenced by the lipid and protein composition of its membrane, fusion plays an equally important role. Recent studies have implicated a conserved family of proteins called atlastin/Sey1 as the fusion machinery responsible for generating three-way junctions within the peripheral ER; however, the actual mechanism used by atlastin (ATL) to catalyze homotypic membrane fusion remains to be clarified. Structural and biochemical studies performed largely with the soluble domain of ATL, suggested that GTP binding facilitated membrane tethering between ATLs anchored in opposing membranes (pre-fusion), followed by GTP hydrolysis catalyzing a crossover conformational change that pulled opposing membranes together for fusion (post-fusion). Through structure-function analysis, I identified key residues that are required for stabilizing the post-fusion conformation, which assisted in elucidating the energy requirements for achieving the post-fusion conformational state of the ATL soluble domain. Using various nucleotides and analogs, I discovered that the soluble domain of ATL was capable of adopting the post-fusion dimer in the absence of GTP hydrolysis, suggesting that GTP hydrolysis may be required for another discrete step within the ATL fusion cycle, such as disassembly. However, this result appeared to be inconsistent with the requirement for GTP hydrolysis in the ATL-mediated fusion of synthetic liposomes and may perhaps be attributed to the different behaviors exhibited by the soluble domain verses membrane-anchored ATL molecules. Therefore, I extended our initial analysis of the ATL soluble domain to membrane-anchored ATL, specifically focusing on identifying the energetic and conformational requirements for ATL-mediated tethering. My investigation revealed that membrane tethering depended on GTP hydrolysis; but, unlike fusion, it did not depend on crossover.