Here we use RNA sequencing RNA seq chromatin

Here, we use RNA sequencing (RNA-seq), chromatin immunoprecipitation sequencing (ChIP-seq) and rapid amplification of cDNA ends (RACE) to map the human SYNE1 transcriptome, providing a framework for localizing the multiple disease-associated mutations identified in this complex gene in relation to specific protein coding transcripts. Several CPG2 transcripts from the human SYNE1 gene were identified, including ones not previously annotated in public databases. The first full-length human CPG2 cDNA was cloned, and the corresponding transcript shown to be expressed in the human neocortex. Further, we raised specific polyclonal Fmoc-Glu(OtBu)-OPfp sale against human CPG2 and used them to confirm that the protein is expressed in several human brain regions. Using a lenti-viral gene knock down/replacement strategy and a surface receptor internalization assay, we demonstrate that human CPG2 localizes to dendritic spines in rat hippocampal neurons and is functionally equivalent to rat CPG2 in regulating glutamate receptor internalization. The conserved function of CPG2 between rat and human provides a platform for testing the effect of missense SNPs identified by BD patient exome sequencing on neuronal function.
Materials and methods
Results
Discussion In the past decade, GWAS have attempted to identify genetic variants that confer risk for many human diseases, whose inherited components remain unexplained (Manolio et al., 2009). In a few cases risk variants identified by GWAS have paved the way for a molecular understanding of disease causes. Crohn\'s disease, ulcerative colitis, and type I diabetes are all examples where deep sequencing follow-up on GWAS hits have revealed specific disease causing mutations (Lee et al., 2013, Nejentsev et al., 2009). However, experience from these studies and others indicates that when individual gene effects are relatively small, as would be the case for neuropsychiatric diseases such as BD, data from tens of thousands of patients is required in order for GWAS results to be meaningful (Craddock and Sklar, 2009). Only in recent years have patient databases become large and diagnostically detailed enough to confidently identify individual risk genes as well as complex genetic pathways (Cross-Disorder Group of the Psychiatric Genomics, C, 2013, Psychiatric, G.C.B.D.W.G., 2011). An additional difficulty, making many GWAS hits challenging to follow up functionally, is the absence of comprehensive transcriptional, and protein coding maps for genes in the region of the identified risk loci. Without knowing the gene products potentially altered by GWAS identified variations, mechanistic testing of disease processes is impossible. Further, knowledge of the human gene products is essential for evaluating the relevance of studies using animal models to human disease. While annotated human sequence databases, such as the UCSC Genome database, are a good starting point, they suffer from several limitations. Most annotated database sequences are unverified, in particular in relation to their transcript start and stop sites. Due to the way most database sequences are generated from poly-A-primed cDNA templates, transcript ends are frequently a result of early termination of cDNA synthesis and lack the true 5′ terminus. This is particularly true for the long transcripts typical of brain tissue (Zylka et al., 2015). In addition, transcript priming can arise from genomic poly-A stretches, thus denoting false 3′ termini. Further, current databases are by no means comprehensive and it is likely that additional unreported transcripts exist, especially in the case of genes with differential tissue specific expression patterns. Given these concerns, the first step in evaluating the potential significance of various mutations and polymorphisms that fall near GWAS risk loci is mapping of the transcriptional and protein coding region of interest.