Feline Lindhout

124 4 photo-bleached TagRFP-ER co-expressed with either GFP or GFP-SCRN1 showed full recovery within 60 s of the ER patches (Fig 4F–H). This suggests rapid and complete redistribution of luminal ER content consistent with an intact ER structure. In contrast, the recovery of TagRFP-ER at ER patches was markedly reduced by ~45% in neurons expressing either GFP-SCRN-F402A, SCRN1 shRNA, or VAPA/B shRNAs, which implies incomplete redistribution of the luminal ER marker due to discontinuity (Fig 4F,I,J). Together, these data indicate that loss of VAP and SCRN1 results in discontinuous ER structures and impaired ER dynamics in neurons, which is mediated by VAP-SCRN1 interactions. SCRN1-VAP interaction controls SV cycling Next, we aimed to test whether the VAP-SCRN1 interactions at the ER membrane may be engaged in regulating the SV cycle, as we previously observed that VAP or SCRN1 depletion resulted in decreased SV cycling (Fig 1B–D,J,K). To test this, we conducted the Syt uptake assay in neurons depleted from SCRN1 and expressing SCRN1-F402A, thereby abolishing VAP-SCRN1 interactions. Unlike wild-type SCRN1 expression, exogenous mutant SCRN1- F402A was unable to rescue the effect of SCRN1 knockdown on SV cycling (Fig 5A,B). These data imply that the VAP-SCRN1 interactions, which we identified as regulators of ER remodeling, are engaged in modulating SV cycling at presynaptic sites. VAP and SCRN1 are engaged in regulating evoked presynaptic Ca 2+ responses Neurotransmitter release is induced by a local cytoplasmic Ca 2+ influx upon neuronal stimulation. Thus, to further investigate the phenotype of VAP and SCRN1 on SV cycling, we next assessed whether they are engaged in regulating evoked Ca 2+ influx. To test this, we measured Ca 2+ dynamics at single boutons with the genetically encoded Ca 2+ indicator GCaMP6f and triggered trains of action potentials (50 APs, 20 Hz) using electric field stimulation (Fig 5C). Presynaptic boutons were identified as GCaMP6f-positive swellings along axons, which were shown to co-localize with the presynaptic active zone marker RIM1a-mCherry (Fig 5C). Loss of VAP or SCRN1 both resulted in a marked ~ 25% decreased peak amplitude of evoked Ca 2+ transients (Fig 5D–F). These results indicate that both the ER receptor VAP and the VAP-interacting protein SCRN1 are involved in modulating presynaptic Ca 2+ influx and thereby could affect SV cycling. VAP-SCRN1 interactions modulate Ca 2+ homeostasis at presynaptic sites To better understand the effects of VAP and SCRN1 on evoked Ca 2+ influx, we next sought to determine whether these proteins could interfere with ER-mediated Ca 2+ homeostasis. Maintaining basal Ca 2+ levels is one of the key functions of smooth ER, which is the only type of ER that is present in axons. Thus, we hypothesized that the observed structural ER defects with abolished VAP-SCRN1 interactions could perturb ER-mediated Ca 2+ homeostasis and thereby affect Ca 2+ -mediated SV cycling. We compared relative basal Ca 2+ levels at presynaptic sites in neurons expressing Ca 2+ indicators GCaMP6f or R-GECO1, as well as mRFP or GFP to identify transfected cells and presynaptic boutons. Relative