Molecular Genetics of Schizophrenia

Schizophrenia is a severe disorder affecting approximately 1% of the population. Sufferers experience ?psychotic? symptoms, in particular delusions (false beliefs) and hallucinations (false perceptions) usually in the form of hearing voices. They also tend to show disturbed emotions, impairments of thinking and reasoning and bizarre behaviour. Signs can be present from childhood, but usually the disorder is diagnosed in the late teens and early 20?s. This age of onset, combined with the facts that many patients respond poorly or not at all to current treatments, and for those that do respond, relapse is usually frequent, means that the illness typically impacts on the vast majority of an individual?s adult life. This makes schizophrenia a major burden on the patient, their family and wider society. It has been clear for a century that schizophrenia runs in families, and this is now known to be largely due to genes rather than the family environment. Schizophrenia is clearly a brain disease but, despite much research, specific abnormalities that cause the disorder have not been identified. Such knowledge is likely to be required for the development of truly effective treatments. It is our belief that the best hope of identifying the molecular and cellular abnormalities that underlie schizophrenia is to identify specific genes that are involved. This has proven difficult because schizophrenia does not occur as a result of a single genetic mutation, but reflects a large number of different ?risk genes?. It is really the combination of genes inherited at birth, along with as yet unknown environmental factors, that determine someone?s risk. Modern genetic methods are, for the first time, allowing the great majority of variation in a person?s DNA to be identified in a single experiment and will soon allow the whole DNA sequence to be obtained from individuals at realistic costs. Since people with schizophrenia are relatively unlikely to have children, it is likely that one class of genetic risk factor that will be important are new mutations that arise relatively frequently in the population. We have recently obtained evidence that this is the case, and are now proposing to use new technology to find many more of these by obtaining the sequence of all the genes in 700 patients and their parents. Identifying these mutations and studying the function of the genes they affect will help us understand how schizophrenia arises and to identify molecular targets for new treatments.

Schizophrenia (SZ) is a severe psychiatric disorder with a lifetime risk of approximately 1% and around 80% heritability. Recent genomic studies support expectations from population genetics that susceptibility is likely to reflect a large number of risk alleles occupying a spectrum of frequencies and effect sizes. Thus, genome-wide association studies (GWAS) support the existence of many common risk alleles, and although each individual allele confers small effects (odds ratios 1.2), en masse these contribute a significant role to population risk. At the other end of the spectrum, an important role for rare alleles is supported by the demonstration that rare genomic copy number variants (CNVs) contribute to disease risk, and that at least some of these rare alleles confer relatively large effects on risk (OR~10). Given the marked reduction in fecundity associated with schizophrenia, rare alleles of large effect must be maintained in the population by de novo mutation. Our recent work on CNVs in SZ, as well as work on point mutations in SZ and intellectual disability, highlights the value of seeking these de novo events as a highly efficient means of identifying pathogenic mutations. Our recent CNV work also shows that identification of specific disrupted genes (eg NRXN1) and of enrichment of mutations in genes coding for functionally related proteins can provide important pointers to disease pathways.
Initially we will use whole-exome sequencing to detect de novo mutations in SZ in 700 well-characterised parent-proband trios. This will build on current pilot work with the Wellcome Trust Sanger Institute (WTSI) and involve collaboration with, and substantial leveraged funding from, the Broad Institute and Mount Sinai School of Medicine. After a synthesis of those data with those from emerging sequencing programmes and GWAS, we will use targeted sequencing in exceptionally large case-control samples to confirm the pathogenic status of individual mutations and genes, and to explore pleiotropic effects in phenotypes for which the genetic risk overlaps with schizophrenia. As in our CNV work, we will combine computational biology with increasingly refined and experimentally derived systems annotations to identify biological systems of pathogenic relevance. Following our work on CNVs, an initial major focus will be to explore systematically components of the synaptic machinery in collaboration with the G2C group at WTSI, but the whole-exome scale of our study will allow us to extend systems-wide analyses into the remaining exome to identify further novel loci and pathogenic systems.