Genetics of schizophrenia: What do we know?

Focusing on single nucleotide polymorphisms

Genetic research of diseases previously relied on linkage studies, which focus on linking a chromosome region to transmission of a particular trait across multiple familial generations. This approach has identified several genomic regions that may be associated with schizophrenia, but most of these regions contain multiple genes and are not specific to schizophrenia.

Today, many genetic studies examine variations of a single nucleotide in the DNA sequence, ie, a change of 1 letter in a particular location on the DNA chain. Single nucleotide polymorphisms (SNPs)—relatively common DNA variations found in >5% of the population—have been a major focus of psychiatric genetics in the past decade. Technology now allows researchers to simultaneously genotype millions of SNPs across the genome, producing tremendous power to investigate the entire genome in relation to a phenotype (a disease or a trait) in genome-wide association studies (GWAS).³ GWAS do not require an a priori hypothesis regarding which regions or genes may be important, and have yielded many novel genetic variants implicated in schizophrenia.

Susceptibility genes

Genetic researchers initially hoped to find that one or a few genes are responsible for schizophrenia. However, recent research revealed that many genes may be involved in susceptibility to schizophrenia, and that a particular gene may contribute to the risk of not only schizophrenia but also other psychiatric disorders such as bipolar disorder (BD).

Discovery of the DISC1 gene is an example of how our understanding of the complex genetic architecture in psychiatric disorders has evolved. In 2000, a linkage study in a Scottish family cohort found a translocation on chromosome 1, t(1:11), highly correlated with schizophrenia.⁴ Later studies found that this translocation directly disrupts a gene, which researchers named “disrupted in schizophrenia 1.” The protein encoded by DISC1 appears to provide a scaffold to other proteins involved in multiple cellular functions, particularly regulation of brain development and maturation. It is involved in neuronal proliferation, differentiation, and migration via various signaling pathways by interacting with many other proteins.⁵ Disruption of DISC1 results in dysfunction in multiple neurodevelopmental processes, significantly increasing susceptibility not only for schizophrenia but also for BD and depression.

Many common variants of DISC1 slightly alter expression levels of the gene, which may exert subtle but pervasive effects on neural circuitry development. DISC1 knockout mouse models showed close interactions between DISC1 and N-methyl-d-aspartate receptors and dopamine D2 receptors, linking to the glutamate hypothesis of schizophrenia and the common site of action of antipsychotics. Despite advances in understanding the biology of DISC1, large case-control studies have not found a consistent association between DISC1 and schizophrenia.^6,7 It is possible that DISC1 pathology represents one subtype of schizophrenia that is not prevalent among the general population; therefore, large-scale epidemiologic studies could not find evidence to support DISC1’s role in schizophrenia.

DTNBP1 is another schizophrenia susceptibility gene discovered in linkage studies. Originally found in a large Irish cohort, several SNPs of DTNBP1 were significantly associated with schizophrenia.⁸ A meta-analysis of candidate genes identified DTNBP1 as one of 4 genes with the strongest evidence for association with schizophrenia (the other 3 are DRD1, MTHFR, and TPH1).⁹ DTNBP1 is widely expressed in the brain and is present in presynaptic, postsynaptic, and microtubule locations implicated in a number of brain functions, including synaptic transmission and neurite outgrowth in a developing organism. Furthermore, DTNBP1 is associated with cognitive functions in schizophrenia patients¹⁰ as well as in control subjects.¹¹ Cognitive impairment is considered an endophenotype for schizophrenia. Similar to DISC1 and other candidate genes, DTNBP1 has not emerged as a significant hit in later, large-scale GWAS studies.

Since the first schizophrenia GWAS in 2007,¹² >15 GWAS have been published, with increasingly larger samples sizes. GWAS are based on the “common disease/common variant hypothesis” that common disorders such as diabetes, macular degeneration, and schizophrenia are caused by multiple common variants in the genome. Because GWAS can analyze hundreds of thousands of SNPs simultaneously, a stringent criterion (usually P < 5×10^-8) is used to gauge statistical significance to correct for multiple testing. Because most effect sizes associated with genetic markers in psychiatry are fairly small (odds ratios [ORs] are approximately 1.1 to 1.2), large samples are required to detect significant effects. Several international consortia have accumulated large samples. The Psychiatric GWAS Consortium has >17,000 patients with schizophrenia, >11,000 with BD, >16,000 with major depression, and >50,000 healthy controls. This wave of GWAS has implicated several novel genomic regions in schizophrenia pathophysiology, including ZNF804A, the major histocompatibility complex (MHC) region, and MIR137.