ER-localized LULL1 function is required for the abnormal INM concentration of disease-related delta-E torsinA

DYT1 dystonia is an autosomal dominant neurological disease caused by a single amino acid deletion in the protein torsinA, resulting in expression of a mutant ΔE [delta E] torsinA isoform in DYT1 patients. Research has consistently found ΔE [delta E] torsinA abnormally concentrated in the nuclear envelope (NE) lumen of the endoplasmic reticulum (ER), and this has led to the hypothesis NE accumulation of ΔE [delta E] torsinA may underlie disease pathogenesis.

We first investigated where and how ΔE [delta E] NE accumulation occurs. We found that ΔE [delta E] torsinA accumulates at the inner nuclear membrane (INM) NE subdomain. Furthermore, modulating expression of NE-localized interacting partners alters NE accumulation of torsinA, but surprisingly, increasing expression of ER resident interacting partner LULL1 also leads to torsinA NE concentration. An ER-localized mechanism that promotes the concentration of another ER protein at the NE has never been described, so we tested whether LULL1 is functional in the ER. We used ER-restricted LULL1 fusion proteins and antibody accessibility assays, and our data reveals that there is not an additional subset of LULL1 present in the NE, despite relocalization of ΔE torsinA to the INM.

We broadly tested the relative importance of LULL1 domains. We co-expressed GFP-ΔE [delta E] torsinA with LULL1 truncation mutants and found that the torsinA-interacting luminal LULL1 domain is required to promote ΔE [delta E] torsinA accumulation in the NE. We also find that the cytosolic domain of LULL1 is required for LULL1 function, suggesting that a region other than the torsinA interacting domain contributes to the activity of LULL1. Unexpectedly, we find that dimerization-competent heterologous domains can complement for loss of the required LULL1 cytosolic domain. Our current focus is the functional significance of LULL1 dimerization in vivo.

Much of our work has focused on the disease-related torsinA mutant, but very little is known about its normal function. Development of an in vitro system to investigate the biochemical cycle of torsinA has proven challenging, but we present the foundation for an in vitro system using recombinant torsinA produced in an insect cell based system.