Copy number variation contributes substantially to human evolution, normal phenotypic variation, and human disease (Malhotra and Sebat, 2012). To date, thousands of different genomic duplications and deletions, each spanning hundreds to millions of basepairs, have been mapped genome-wide, and collectively account for a significant fraction of human genetic variation. By reorganizing broad swaths of DNA, structural mutations potentially create new genes and regulatory motifs while also disrupting established genes.

Copy number variants (CNVs) tend to occur in regions rich in genes, segmental duplications, and mobile elements(Malhotra and Sebat, 2012). Some arise repeatedly as de novo mutations in genomic ‘hotspots.’ The genomic architecture of hotspots is typically characterized by large repeated segments that increase the risk for errors in replication, a process known as nonallelic homologous recombination. If a CNV is pathogenic, it may persist for only a few generations given negative selection. At the same time, in each generation many de novo CNVs are introduced given the highly repetitive nature of the human genome.

The importance of rare copy number mutations for neuropsychiatric disease is now well-established (Malhotra and Sebat, 2012). Our group was the first to demonstrate that individuals with schizophrenia are significantly more likely than unaffected persons to harbor rare gene-impacting CNVs, with greater effect for patients with illness onset before age 18 years. Genes disrupted by CNVs in patients function disproportionately in signaling and neurodevelopmental processes, including neuregulin and glutamate pathways (Walsh et al, 2008). Subsequent research replicated and extended these findings. Most rare copy number mutations implicated in schizophrenia are unique, and many arose de novo or in recent generations. Others recur at genomic hotspots, including chromosomes 1q21.1, 3q29, 15q11.2, 15q13.3, 16p11.2, 16p12.1, 16p13.11, 17p12, 22q11.2, and the neuropeptide receptor VIPR2 (Malhotra and Sebat, 2012) (Figure 1).

Figure 1
figure 1

CNVs in schizophrenia. The figure depicts the cytogenetic locations of known genomic hotspots (Malhotra and Sebat, 2012) (red), case-specific CNVs (Walsh et al, 2008) (blue), and case-specific CNVs that created chimeric genes (Rippey et al, 2013) (green).

PowerPoint slide

Full size image

One novel mechanism by which CNVs may cause schizophrenia is the creation of chimeric genes (Rippey et al, 2013). We recently found that persons with schizophrenia were more likely than controls to carry genomic duplications or deletions that resulted in a fusion gene. Each of the chimeric genes in patients differed from their parent genes in ways likely to be important to brain function, including differences in neuronal subcellular localization, levels of expression, and/or protein interactions. For example, one patient harbored a 150-kb deletion that resulted in a MAP3K3–DDX42 fusion gene. The MAP3K3–DDX42 chimera produced a novel protein isoform that binds activated MEK5 and likely acts as a dominant-negative inhibitor of ERK5 signaling, a key pathway regulating neuronal differentiation and adult neurogenesis. By interfering with parent gene function, chimeras potentially disrupt critical brain processes important to schizophrenia.

Rare deletions and duplications, including in the same recurrent hotspot CNVs associated with schizophrenia, are also enriched in other neuropsychiatric conditions, including autism, intellectual disability, and idiopathic generalized epilepsy (Malhotra and Sebat, 2012). Many clinical laboratories now have assays that can detect hundreds of potentially pathogenic CNVs. Each disorder is associated with many different structural mutations. Each pathogenic CNV is associated with different phenotypes in different persons, including cognitive deficits in individuals characterized as ‘unaffected’ (Stefansson et al, 2014). Variable phenotypic expression of pathogenic events is likely mediated by a variety of different factors, including dose effects, epistasis, epigenetic mechanisms, and environmental exposures (McClellan and King, 2010).

Genes and loci disrupted by pathogenic CNVs provide a window to study key neurobiological mechanisms important to brain development and pathophysiology. For example, rare de novo CNVs in persons with schizophrenia were found to be enriched for genes that function within N-methyl-D-aspartate receptor postsynaptic signaling complexes (Kirov et al, 2012). Given the extreme genetic heterogeneity, such research is critical towards delineating specific genomically defined neurobiological subtypes within large clinically heterogeneous diagnostic groups, and ultimately towards developing biologically informed targeted treatments.

FUNDING AND DISCLOSURE

This work was supported by NIMH grants R01MH083989 and F30MH085467. The authors declare no conflict of interest.