Nonsense-mediated mRNA decay and loss-of-function of the protein underlie the X-linked epilepsy associated with the W356× mutation in synapsin I

Abstract

Synapsins are a family of neuronal phosphoproteins associated with the cytosolic surface of synaptic vesicles. Experimental evidence suggests a role for synapsins in synaptic vesicle clustering and recycling at the presynaptic terminal, as well as in neuronal development and synaptogenesis. Synapsin knock-out (Syn1(-/-) ) mice display an epileptic phenotype and mutations in the SYN1 gene have been identified in individuals affected by epilepsy and/or autism spectrum disorder. We investigated the impact of the c.1067G>A nonsense transition, the first mutation described in a family affected by X-linked syndromic epilepsy, on the expression and functional properties of the synapsin I protein. We found that the presence of a premature termination codon in the human SYN1 transcript renders it susceptible to nonsense-mediated mRNA decay (NMD). Given that the NMD efficiency is highly variable among individuals and cell types, we investigated also the effects of expression of the mutant protein and found that it is expressed at lower levels compared to wild-type synapsin I, forms perinuclear aggregates and is unable to reach presynaptic terminals in mature hippocampal neurons grown in culture. Taken together, these data indicate that in patients carrying the W356× mutation the function of synapsin I is markedly impaired, due to both the strongly decreased translation and the altered function of the NMD-escaped protein, and support the value of Syn1(-/-) mice as an experimental model mimicking the human pathology.

Figure 1. The SYN1 transcript carrying the G1067A nonsense transition is subjected to NMD in HeLa cells.

A. Schematic representation of the SYN1 gene, the Syn I protein (WT and W356× variants) and the minigene construct. The position of the G1067A mutation is indicated by an asterisk in exon 9. B. NMD assay performed in HeLa cells. Top and middle panels: HeLa cells were transfected with siRNAs for luciferase (Luc; lanes 1–2) or UPF1 (lanes 3–4). As a rescue control, cells transfected with the siRNA for UPF1 were co-transfected with a UPF1 cDNA construct carrying silent mutations that render it insensible to knocking-down (UPF1 rescue; lanes 5–6). For these three conditions, cells were co-transfected with either the WT or nonsense (NS) SYN1 minigenes and with a WT HBB coding plasmid (as a transfection control). When UPF1 is present, most of the mutant SYN1 mRNA is degraded. Bottom panel: as a control for NMD efficiency, the same experiment was performed with the WT and NS39 variant of the HBB gene, with the WT+300 HBB construct as a transfection control. C. Quantification of the Northern blot analysis. Values were calculated from signal intensities according to the ratio (C_WT · S_NS)/(C_NS · S_WT), where S_WT and S_NS are the WT and NS samples, respectively, and C_WT and C_NS are the respective transfection controls. Data represent the mean (± SD) percentage of remaining nonsense mRNA of n = 3 independent experiments. Statistical analysis was carried out by one-way ANOVA, followed by the post-hoc Tukey's multiple comparison test (*, p<0.05; **, p<0.01).

Figure 2. The W356× Syn I variant is poorly expressed in transfected HeLa cells.

HeLa cells were transfected with either the complementary (cDNA) or genomic (minigene) DNA coding for either WT or W356× (NS) Syn I. A. Western blotting analysis shows that WT Syn I is efficiently expressed in both cDNA- and minigene-transfected cells (the two bands in the minigene-transfected sample corresponds to the Syn I a and b splicing isoforms). In contrast, the mutant protein is expressed with much lower efficiency in cDNA-transfected cells, and its levels are hardly detectable after transfection with the intron-containing minigene. B. Immunofluorescence images of HeLa cells co-transfected with pEGFP (in green) and either cDNAs or minigenes encoding either WT or mutant Syn I (anti-Syn I staining, in red). Immunostaining for mutant Syn I is appreciable in cells transfected with cDNA, but not with the minigene. Scale bar: 20 µm.

Figure 3. The W356× Syn I variant accumulates in aggregates that are labelled by an autophagosome marker.

HeLa cells were transfected with FLAG-tagged WT or W356× Syn I coding plasmids. A.Representative immunofluorescence images show that the distribution of the WT protein (green) overlaps with the distribution of the F-actin filaments labelled with phalloidin (red), while the mutant form accumulates in perinuclear aggregates. WT Syn I is expressed in the majority of the cells (not shown), while W356× Syn I is only expressed by a small fraction of the cDNA-transfected cells. B-E.W356× Syn I aggregates (red) do not co-localize with organelle markers (green) specific for either early endosomes (EEA1; B), recycling endosomes (TfR; C) or lysosomes (LAMP1; D), while they co-localize with an autophagosome marker (LC3; E). Scale bars: 15 µm.

Figure 4. W356× Syn I aggregates are Triton X-100 insoluble and are not ubiquitinated.

HeLa cells transfected with FLAG-tagged WT or W356× Syn I were lysed in 1% Triton X-100 or 1% SDS buffers after overnight treatment in the presence (+) or absence (−) of the proteasome inhibitor MG132 (1 µM). Total lysates (TL) were subjected to immunoprecipitation (IP) with an anti-FLAG antibody and the IP samples were analyzed by Western blotting with anti-Ubiquitin antibody to reveal protein ubiquitination (upper panels). Neither WT nor W356× Syn I appear ubiquitinated. To check for recovery of the transfected proteins after IP, the same membranes were stained with anti-FLAG antibody (lower panels). White asterisks indicate the IgG heavy chains. Black arrowheads indicate either the WT or W356× Syn I band.

Figure 5. The W356× Syn I variant is not targeted to nerve terminals in either Syn1^−/− or WT hippocampal neurons.

A. Syn1^−/− hippocampal neurons were transduced at 4 DIV with lentiviruses expressing EYFP-labeled either WT or W356× Syn I (green), and fixed at 8 DIV. VAMP2 (red) was used as a marker of presynaptic contacts. B. A similar experiment was performed in WT hippocampal neurons. In both neuronal backgrounds, WT Syn I displays a punctuate pattern overlapping with VAMP2 staining, while W356× Syn I is dispersed in the cytoplasm. The presence of endogenous Syn I in WT neurons is not sufficient to correct the mis-localisation and to redirect the mutant protein to presynaptic terminals. C. Syn1^−/− hippocampal neurons were transduced at 4 DIV and fixed at either 11 or 14 DIV. The prolonged expression of W356× Syn I results in the formation of somatic microaggregates. Scale bars: 20 µm.