Other mutations in the SYNE region exons have been
Other mutations in the SYNE1 3′ region (exons 81–85) have been implicated in an upper motor neuron disease, ARCA1 (Gros-Louis et al., 2007). The disease is characterized by progressive movement, coordination, and balance problems caused by disrupted Purkinje cell function and their impaired signaling to other cerebellar neurons (Gros-Louis et al., 2007). To date it is still unclear how the SYNE1 gene mutations lead to the cerebellar cellular phenotypes contributing to ARCA1. Interestingly, a large SYNE1 transcript has recently been identified in mouse cerebellum, which encodes a nesprin-1 isoform that lacks the KASH domain for nuclear anchorage and instead localizes to glomeruli of cerebellar mossy fibers (Razafsky and Hodzic, 2015). We speculate that the larger transcript size band observed in the cystamine in our Northern blotting analysis might belong to human homologs of the cerebellum specific transcript identified in mice (Razafsky and Hodzic, 2015) that potentially harbor the cerebellar ataxia associated SNPs (Gros-Louis et al., 2007). A second group of mutations in SYNE1 have been implicated in synapse-associated psychiatric disorders, ASD, BD, and major depression, which display considerable comorbidity, and diagnostic boundaries that are often difficult to define (Cross-Disorder Group of the Psychiatric Genomics, 2013). While family studies have repeatedly demonstrated that genetic risk factors are important in the causality of all these diseases, the role of specific genes remain unclear (Berrettini, 2000). The mutations in SYNE1 associated with psychiatric disorders map to the part of the gene that spans the rat Cpg2 homologous locus and the nearby 3′ and 5′ regions. In recent meta-analyses of GWAS data (Green et al., 2013a, Psychiatric, G.C.B.D.W.G., 2011, Xu et al., 2014), the markers in SYNE1 that showed genome-wide statistically significant association with BD tag (via linkage disequilibrium) a missense SNP that alters the protein encoded by CPG2. Several other missense SNPs are in close proximity, and exome sequencing suggests additional polymorphisms in the CPG2 coding region, which could potentially alter CPG2 protein function. The ASD associated mutations fall outside the CPG2 region that we mapped in this study. While they are close enough (exon 9 and exon 60) that we initially considered the possibility that they would be contained within a human CPG2 transcript longer than the rat homolog, our data suggest that other as yet unidentified transcripts are more likely to exist in these regions such as transcripts encoding the N-terminal nesprin isoforms identified in mammalian cell lines with suggested functions in the Golgi complex (Gough et al., 2003). To further investigate the CPG2 locus of SYNE1 as the potential risk region for neuropsychiatric disorders, we used an integrated approach combining several complementary gene-mapping strategies, overcoming their individual limitations. Our data verified several human CPG2 transcripts, confidently determining their 3′ and 5\' ends. We cloned a full-length human CPG2 transcript as well as two shorter forms. The complex transcriptional pattern from the SYNE1/CPG2 locus with several potential splice variants suggests multiple layers of transcriptional regulation. The shorter transcripts could potentially have regulatory functions at the mRNA level through competitive binding to microRNAs or to the translational machinery. Notably, a large number of microRNAs have been identified that interfere with expression of synaptic proteins such as NMDA receptor and calcium/calmodulin-dependent kinase II (CaMKII), and might influence neuronal plasticity (Bredy et al., 2011). Such functions may be impaired in the case of specific SYNE1 mutations/polymorphisms, either by regulation of transcript abundance or by alternative RNA splicing. Alternatively, truncated CPG2-like proteins could regulate CPG2 function through competitive association with cellular binding partners as naturally occurring dominant negatives. While our inability to detect the two CPG2 short forms in human brain extracts suggests the former possibility as more likely, we cannot exclude that this is due to limitations of our antibodies or to limited probe sensitivity.