ChIP-chip from fetal brain tissue with b-Catenin

ABSTRACT: GSK3b signaling has been implicated in the pathophysiology of bipolar affective disorder based on the inhibitory effect of lithium salts on this pathway. This has been supported by a recent genome wide association study. In addition, GSK3b signaling may also play a role in schizophrenia, an idea that is based on the finding that the protein encoded by the schizophrenia candidate gene DISC1 inhibits GSK3b. Gene expression is affected by GSK3b signaling through the activation of the transcription factor b-catenin, which modulates expression through a protein-protein binding interaction with members of the TCF/LEF family. To identify genes targeted by GSK3b/b-catenin, investigators have used expression profiling in lithium treated cells and animals. Gene knockdown and transfection are other experimental methodologies one can employ. However, these approaches are not possible using human brain tissue. Consequently, we used ChIP-chip to identify promoters bound by b-catenin as an indirect method for identifying genes targeted by GSK3b. Fetal brain was used because schizophrenia is viewed as a neurodevelopmental disorder. In addition, fetal brain tissue is not affected by the confounding factors associated with postmortem tissue (postmortem delay, cause of death, brain pH, use of medication and other environmental factors). We carried out two complete promoter array hybridizations on different biological isolates using the 385K two-Array Human Promoter Set developed by Nimblegen/Roche. Significant enrichment for b-catenin immunoprecipitates was determined in each experiment if the log2 ratio of signal obtained for the b-catenin antibody compared with control was below an FDR<0.2 using NimbleScan. NimbleScan reports three different levels of enrichment (peaks) based on the FDR score, as described in methods. Since whole brain tissue was used in these experiments, we chose the least conservative FDR (<0.2) to reduce the type II error (false negative) rate, which could be caused by enrichment at some promoters in a subset of cells being diluted by the absence of b-catenin immunoprecipitates at those loci in other cells. A total of 640 gene promoters showed evidence for b-catenin binding on two arrays. We validated enrichment at 9 gene promoters using real time PCR. The 640 genes are analyzed using SLEP (Sullivan Lab Evidence Project, http://slep.unc.edu). Analysis of the SLEP database was restricted to genome wide association studies (GWAS) and copy number variants (CNVs) in ASD, BD and SZ. In addition, a gene was scored as a potential candidate if it fell within a linkage peak or if multiple positive association studies were found upon a candidate gene analysis. Promoters for 24 genes implicated in these disorders were targeted by b-catenin. Of these, a total of 16 genes are very strong candidates based on unbiased genome-wide surveys (GWAS and CNV screening). The other 8 have been implicated by linkage analysis and candidate gene analysis, which are the least robust genetic methods for targeting specific genes in psychiatric disorders. We carried out a similar analysis on 640 genes that did not show evidence for enrichment in the b-catenin ChIP-chip. Only 5 genes were found, none of which are GWAS or CNV targets. Based on the finding that five candidate genes were found in a group of 640 randomly selected genes, while 24 were identified in the ChIP-chip experiment, the binding of b-catenin to schizophrenia and bipolar disorder candidate gene promoters is highly significant (p = 6.5 x 10-12). The findings suggest that many seemingly disparate candidate genes for psychiatric disorders can be grouped into the GSK3b/b-catenin signaling pathway. The results could also have therapeutic implications in helping to identify bipolar patients who respond best to lithium, as well as identify patients with schizophrenia who might benefit from the drug.

ORGANISM(S): Homo sapiens

PROVIDER: GSE21151 | GEO | 2010/04/01

SECONDARY ACCESSION(S): PRJNA126631

REPOSITORIES: GEO