Pablo V. Gejman - US grants

DESCRIPTION: (Adapted from investigator's abstract) Susceptibility to schizophrenia appears to be transmitted in part through a complex genetic mechanism involving interaction of multiple genes, each of relatively small effect. Detection of such loci through genetic linkage studies is likely to require a very large sample of multiply affected pedigrees. In response to RFA MH-99-005, nine investigators propose a Collaborative Study of Mental Disorders to collect in three years a sample of 517 affected sibling pairs (ASPs) with DSM-IV schizophrenia, to complete a genome scan of these pedigrees for multipoint ASP analysis to detect susceptibility loci, and to share biological materials, genotypes and blinded clinical data with the scientific community through an NIMH-sponsored mechanism. Each site will recruit families in a large geographic area, using an opportunistic ascertainment strategy and efficient assessment procedures to maximize the number of ASPs collected. Subjects suspected of having schizophrenia will be assessed with the Diagnostic Interview for Genetic Studies supplemented by information from the Family interview for Genetic studies and medical records. Diagnoses will be made by consensus best-estimate procedures. Interviewer training and quality assurance monitoring of protocol adherence will be provided for all sites. Blood specimens will be obtained from all individuals with psychotic disorders plus their parents and (when both parents are not available) up to two additional siblings to provide genetic phase information. Permanent cell lines will be created and DNA extracted at the NIMH-sponsored Center for Genetic Studies. All clinical data will be merged regularly into a central study database, and blinded data transmitted to the Center for Genetic Studies. At the end of the four-year project period, biological materials and blinded pedigree and clinical data will be made available to scientific community for genetic studies of schizophrenia and related disorders. In year 4, a genome scan will be undertaken at CIDR (if approved) or at the University of Chicago. Affected subjects and relatives needed for phase information will be genotyped using the latest screening map, currently the Weber Version 9 Linkage Mapping Set, containing 387 microsatellite markers at approximately 10 cM spacing. The primary statistical approach will be multipoint analysis of allele sharing in affected individuals, with DSM-IV schizophrenia defined as affected. Secondary analyses will also be carried out. Power analyses suggest that this study would have excellent power to detect loci associated with lambda sibs of 1.4, and moderate power for a value of 1.3, i.e., loci with relatively small etiologic effects.

DESCRIPTION (provided by applicant): This is a revised application for 3 years of funding for two sites and 5 consortium sites (PA 05-106, "Deep Sequencing and Haplotype Profiling of Mental Disorders"). The goal is to identify and characterize genetic variation that contributes to schizophrenia (SZ) susceptibility, by carrying out a genome-wide association GWA) study followed by resequencing, genotyping and biological experiments. Two samples will be studied: 3,000 SZ and 3,000 control subjects of European ancestry (EA), and 1,200 cases and 1,200 controls of African-American (AA) ancestry. The GWA datasets will include 550,000 SNPs in the EA sample (the revised Affymetrix 500K array and 50K Gene-Focused chip that includes 20K nsSNPs), and the new Affymetrix 1M array in the AA sample (the 500K array and 500K additional SNPs with increased coverage of African variation). The new 500K array also provides genomewide assays of additional copy number variants (CNVs). The Genetic Association Information Network (GAIN) will genotype 1450/1450 EA cases/controls (500K) and the entire AA sample (1M). The Affymetrix consortium site will genotype the remaining 1550/1550 EA cases/controls with the 500K array, and all EA subjects with the 50K chip. Preliminary statistical studies are proposed, to select an optimal data analysis strategy that tests every HapMap SNP using single- and multi-marker tests, evaluates evidence for association on European- and African-ancestry chromosomes (after inferring local ancestry in admixed individuals) and in the combined data, controls for subtle population substructure, and evaluates empirical p-values through permutation. A set of" 15 candidate intervals will be selected based on p-value threshold, Rank Truncated Product analysis, replication experiments, and bioinformatic and biological information. Deep resequencing experiments will detect any significant case-control difference in rare functional mutations, and will discover new rare and common SNPs. Further genotyping of each region will include rare/functional SNPs and additional common SNPs for optimal tagging of common variants. Based on evidence for association and available information about each gene, the associated genomic interval, and the associated variants, biological studies will be undertaken to begin to evaluate the functional effects of these variants and the implications for hypothesis about mechanisms underlying susceptibility to SZ.

DESCRIPTION (provided by applicant): Schizophrenia is a common, severe, highly heritable psychotic disorder for which biological insights and etiological knowledge-based treatments have yet to be achieved. The vast majority of patients suffering from schizophrenia remains ill after the initial episode, suffering from chronic and severely incapacitating symptoms, and are unable to work. Genome-wide association studies (GWAS) have been successful in uncovering individual common susceptibility loci reproducibly associated with schizophrenia. However, identifying the underlying causal variants, risk genes, and etiological gene networks have proven difficult for schizophrenia, like for most other complex disorders. It is likely that many risk variants in these loci are regulatory in nature. Here, we propose to fill the gap in our biological understanding of schizophrenia etiology through an integrative genomics approach based on SNP genotype and RNAseq data from a large collection of lymphoblastoid cell lines (LCLs) derived from the Molecular Genetics of Schizophrenia case-control sample, which includes rich demographic and psychiatric information. Expression signatures will be generated at baseline and after a perturbation with dopamine, predicated on the hypothesis that cell perturbation using this pharmacologically relevant agent will reveal etiologically relevant genes which are undetectable in the unperturbed (baseline) state. We present evidence of two supporting facts: (1) Signals from expression analysis of LCLs and GWAS results converge at the major histocompatibility complex region on chromosome 6. (2) Dopamine stimulation strongly regulates the expression of many genes located in genome-wide significant GWAS loci, and in copy number variants associated with schizophrenia. Our study will capitalize on an ongoing experiment (RC2MH90030) of unstimulated (baseline) genome-wide expression profiles of LCLs from the same sample. We will identify dopamine-responsive transcripts associated with schizophrenia, analyze the underlying regulatory DNA variants (i.e., expression quantitative trait nucleotides, eQTNs), and assess the association of eQTNs with schizophrenia. We will then perform validation testing of LCL findings in neural tissues, and will functionally characterize a set of most important eQTNs. The proposed study is expected to identify new loci influencing schizophrenia risk, reveal the causal genes in already identified GWAS loci, and shed light on the underlying etiological mechanisms by establishing a connection to the mechanisms of action of antipsychotics, spearheading clinical applications in the field for diagnostic classification and treatment.

DESCRIPTION (provided by applicant): Schizophrenia (SZ) is a severe psychiatric disorder that affects 1% of the population worldwide and has a strong genetic influence on susceptibility. Recent genetic investigations of SZ, such as genome-wide association studies (GWAS) and structural genomic studies have made remarkable progress but leave a substantial part of the genetic risk unexplained, suggesting alternative models should be explored. We have designed a targeted sequencing experiment, by selecting coding and regulatory sequence in genomic intervals with high prior evidence for involvement in SZ. Our targets come from (i) genes that reside within SZ- associated copy number variant (CNV) intervals, or (ii) genes that show extreme transcriptional departures in SZ, for a total of ~600kb of sequence. We propose sequencing sample from the Molecular Genetics of SZ (MGS) collection and extracting sequence from the Genomic Psychiatry Cohort (GPC), resulting in a large, combined European ancestry (EA) discovery sample of 3,181 SZ cases and 3,500 matched controls. By limiting our target to a region that is small but likely enriched for SZ associated low frequency variants, we both lower the statistical threshold required for significance, and economically allow for a large sample size, giving us maximal power to identify new associations for SZ. We will then examine top hits for replication in the remaining GPC EA sample of 4,100 SZ cases and screened 5,400 controls. In addition, we propose adding another dimension of information, by functional evaluation of our most promising candidates. To accomplish this goal, in subsequent initial, exploratory work, we will generate and phenotypically characterize induced pluripotent stem cell (iPSC)-differentiated neurons from patients harboring associated mutations and from controls. If successful, our study will identify genes, putative mutations, and a mechanism of action by which those mutations contribute to SZ pathology. In this way we expect to refine our understanding of SZ and advance new, focused hypotheses to be tested. All data and biological materials will be rapidly shared through the designated NIMH repository (www.nimhgenetics.org) and dbGaP (dbgap.ncbi.nlm.nih.gov).