Torsins are developmentally essential AAA+ proteins, and mutation of human torsinA causes the neurological disease DYT1 dystonia. They localize in the ER membranes, but their cellular function remains unclear. We now show that dTorsin is required in Drosophila adipose tissue, where it suppresses triglyceride levels, promotes cell growth, and elevates membrane lipid content. We also see that human torsinA at the inner nuclear membrane is associated with membrane expansion and elevated cellular lipid content. Furthermore, the key lipid metabolizing enzyme, lipin, is mislocalized in dTorsin-KO cells, and dTorsin increases levels of the lipin substrate, phosphatidate, and reduces the product, diacylglycerol. Finally, genetic suppression of dLipin rescues dTorsin-KO defects, including adipose cell size, animal growth, and survival. These findings identify that torsins are essential regulators of cellular lipid metabolism and implicate disturbed lipid biology in childhood-onset DYT1 dystonia.

Repeat expansions in the C9orf72 gene cause amyotrophic lateral sclerosis and frontotemporal dementia characterized by dipeptide-repeat protein (DPR) inclusions. The toxicity associated with two of these DPRs, poly-GR and poly-PR, has been associated with nucleocytoplasmic transport. To investigate the causal role of poly-GR or poly-PR on active nucleocytoplasmic transport, we measured nuclear import and export in poly-GR or poly-PR expressing Hela cells, neuronal-like SH-SY5Y cells and iPSC-derived motor neurons. Our data strongly indicate that poly-GR and poly-PR do not directly impede active nucleocytoplasmic transport.

Energy metabolism has been repeatedly linked to amyotrophic lateral sclerosis (ALS). Yet, motor neuron (MN) metabolism remains poorly studied and it is unknown if ALS MNs differ metabolically from healthy MNs. To address this question, we first performed a metabolic characterization of induced pluripotent stem cells (iPSCs) versus iPSC-derived MNs and subsequently compared MNs from ALS patients carrying FUS mutations to their CRISPR/Cas9-corrected counterparts. We discovered that human iPSCs undergo a lactate oxidation-fuelled prooxidative metabolic switch when they differentiate into functional MNs. Simultaneously, they rewire metabolic routes to import pyruvate into the TCA cycle in an energy substrate specific way. By comparing patient-derived MNs and their isogenic controls, we show that ALS-causing mutations in FUS did not affect glycolytic or mitochondrial energy metabolism of human MNs in vitro. These data show that metabolic dysfunction is not the underlying cause of the ALS-related phenotypes previously observed in these MNs.