Genome wide association studies (GWAS) reveal disease associated genetic variation, but more often than not point to multiple variants with correlated genotypes due to linkage disequilibrium (LD). Many of the variants in LD are correlated with disease not because they have a functional consequence but because they associate with others that do. Whether each independent GWAS signal, i.e. each set of disease SNPs in strong LD includes one or more functional variant driving the association, and whether alleles in LD are related in function is generally unknown. To accurately model the effects of disease associated variants it is important to understand the possible dynamic relationship between functionality and LD. From our and others' published work we have evidence that in some cases LD may be driven by selection for the co-occurrence of functional variant alleles in cis on the same haplotype. We hypothesize that new regulatory alleles undergo selective pressure that depends on the existing regulatory alleles on the same haplotype. In this R21 exploratory research grant application we propose to use the CACNA1C gene, on which we have previously performed extensive work, as a model to look for evidence of cooperative activity of the SZ-associated SNPs. In that work we acquired evidence that interaction between multiple SZ variants may be mediated by a transcription factor (TF) called PKNOX2, a TF highly expressed in the brain and previously associated with substance abuse and SZ presentation. Here we will use human induced pluripotent stem cells (iPSCs) differentiated into neurons by NGN2 induction (iNs) and CRISPR/Cas9 editing of each SZ associated variant and the PKNOX2 gene to study the consequences of modifying each variant and the connecting TF. In particular we propose the following specific aims: (1) We will employ CRISPR/Cas9 genome editing on iPSC lines from healthy controls carrying each of the two major haplotypes. We will delete the region surrounding each SZ-SNP (~50 bp) and measure the effect of each deletion on CACNA1C expression in each background. We expect more than one deletion to have a consequence on the gene's regulation. (2) We will use Chromatin immunoprecipitation followed by sequencing (ChIP-Seq) using a PKNOX2 antibody on opposite haplotype homozygotes. We expect to demonstrate interaction with PKNOX2 with the SNP regions and differential pull down of the two alleles of multiple CACNA1C SZ SNPs. We will use the same data to identify other genomic sequences interacting with PKNOX2, allelic differences where the genotypes differ, and possible association of these regions with SZ in public GWAS data. (3) We will knock out PKNOX2 by genome editing and use the otherwise isogenic iN cells to: (i) Measure the changes of CACNA1C and the transcriptome by RNAseq on iNs', (ii) Measure the consequences to open chromatin by ATAC-seq on iNs' (iii) Show the effects of removing PKNOX2 on chromatin interactions of the CACNA1C promoter with the SZ-associated SNPS by 4Cseq in iNs
Genome wide association studies reveal disease associated genetic variation, but most often point to multiple genetic variants, any one or more of which might be responsible for increasing disease risk. In order to make the most out of this new knowledge on the genetics of disease we need to find out whether and how such variants combine to influence the development and function of the brain. In this exploratory project we propose a series of experiments tailored to gain more insight on these questions using induced pluripotent stem cells, induced neuronal cells and a gene strongly associated with schizophrenia called CACNA1C as a model.
