Atlastin Mediated Endoplasmic Reticulum Network Formation In Hereditary Spastic Paraplegia

The endoplasmic reticulum is an extensive multifunctional membrane bound organelle present in all eukaryotic cells. It houses a wide array of essential processes including protein and lipid synthesis, drug detoxification and regulation of intracellular Ca+2 . This very large organelle is organized into morphologically distinct subdomains, presumably to maximize the efficiency of each of its many functions. Yet the ER is interconnected at hundreds of branchpoints. maintaining both luminal and membrane continuity. Despite its complex structure, the ER undergoes continuous membrane remodeling, which may enable it to adopt to environmental changes. Due to their extreme polarity and the long distances that need to be traversed by cellular constituents, neurons may rely more heavily than other cell types on the proper structure, function and dynamics of organelles such as the ER. In support of this idea, a number of neurological disorders are linked to mutations in genes whose products are proposed to structure the ER. In particular, mutations in the neuronal isoform of atlastin (ATL), a conserved dynamin-related GTPase implicated in homotypic ER membrane fusion and ER network formation, cause a motor neurological disorder called Hereditary Spastic Paraplegia (HSP). Determining the role of ATLs in ER morphology has obvious implications in the context of the neurodegeneration seen in Hereditary Spastic Paraplegia patients. To this end, in my thesis I worked on three projects. One focused on testing the hypothesis that disease mutations cause HSP because they disrupt neuronal ATL-1’s fusion-dependent ER structuring function. Using a cell-based assay for ATL-mediated ER network formation, I showed that neuronal ATL-1 can fully restore a branched ER network in HeLa cells depleted of endogenous ATL, and yet surprisingly, not all the disease mutations disrupt ER morphology. Furthermore, at least two disease variants, including that most commonly identified in patients, displayed wild type levels of activity in all assays, including a biochemical assay for membrane fusion. The second project tested the role of an N-terminal extension of ATL-1 that is highly conserved across vertebrate species. My results indicated that this extension was dispensable for ER structuring at least in non-neuronal cells. Therefore, the significant conservation observed within this region may reflect a regulatory role specific to neurons, an idea that remains to be tested. Lastly, I collaborated with James McNew and his group to investigate the precise role of the cytoplasmic C-terminal tail of ATL. Together we showed that the C-terminal tail is important for both the fusion and ER network formation functions of ATL. And yet in the context of less stable lipid bilayers, the requirement for the C-terminal tail during fusion was alleviated. Altogether, my findings reveal a discrepancy with the hypothesis that disease mutations disrupt ER morphology and highlight a gap in the understanding of the cause of ATL-1 linked SPG3A. The apparent lack of a requirement for a highly conserved N-terminal extension, as well as residues implicated in HSP, is surprising. It suggests that the ER in neurons might rely on a neuron specific factor that binds and regulates the fusion. Alternatively, ATL-1 may mediate an additional (non-ER fusion) function specific to neurons. Overall, my investigation reveals that there is more to be understood in terms of precise role (s) and regulation of ATL a well as the basis of SPG3A pathogenesis.