Feline Lindhout

126 4 of proteins. Here, the difference could be explained by incomplete depletion of endogenous SCRN1 during the SCRN1 shRNA silencing period, whereas on the other hand dominant- negative SCRN1-F402A actively recruits and captures endogenous SCRN1 proteins and thereby impairs its function. Nevertheless, we cannot exclude that additional functions of SCRN1-F402A might be at play. Together, our data indicate that SCRN1-VAP interactions are engaged in controlling the VAP-associated phenotypes, and further investigations are required to examine whether additional VAP-interacting proteins could also be involved in modulating SV cycling. VAP-SCRN1 interactions control ER continuity and dynamics The ER network is composed of a well-maintained interconnected structure that undergoes continuous remodeling for proper functioning. Here, we identified the VAP-SCRN1 interactions as critical regulators for maintaining ER structure and dynamics. Considering that VAPs are best known to act as major ER receptors by facilitating tight membrane contact sites between the ER and other organelles, it is tempting to speculate that these membrane contact sites might be engaged in controlling VAP-mediated functions on ER integrity and dynamics (Muallem et al, 2017; Wu et al, 2018). One plausible mechanism could be that these membrane contact sites may act as “anchor points” which locally stabilize the ER network. These VAP-facilitated membrane contact sites are abundantly present throughout the cell and can thus stabilize local spots within the large interconnected and highly dynamic ER network. As such, this hypothesis follows the assumption that proper ER integrity relies on a defined balance between ER dynamics and stability. In line with this, it was previously shown that depleting either ER-forming protein atlastin or ER-stabilizing protein reticulon resulted in ER tubule fragmentation, which could remarkably be rescued by depleting both proteins simultaneously (Wang et al, 2016). Alternatively, VAP-mediated membrane contact sites could control ER remodeling by enabling fast lipid delivery. We showed that ER remodeling in neurons is a fast process and that ER tubule growth events can occur in a few seconds. This fast ER remodeling requires the continuous and rapid rearrangement of lipids. This may be accomplished via fast lipid transfer at membrane contact sites between ER and other organelles, rather than via the relatively slower lipid synthesis and redistribution at the ER membrane. Note worthily, considering that VAPs are implicated in a wide range of functions, various other possible direct or indirect VAP- mediated mechanisms to control ER remodeling might be at play. Maintaining axonal ER structures is important to preserve presynaptic function In recent EM studies, it was reported that ER structures in axons show specific adaptations, characterized by small ER lumen and low density of 1–2 tubules per axon diameter. These adaptations likely allow the axonal ER to extend throughout all axon brancheswhile remaining continuous with the remaining ER network, which is typically more dense and branched. In this study, we found that VAP and SCRN1 depletion affected ER structures throughout the cell, though axonal ER structures seemed most severely disrupted. Interestingly, mutations