Genetic Variation And Schizophrenia

REVIEW

The Role of DNA Copy Number Variation in Schizophrenia Gloria W.C. Tam, Richard Redon, Nigel P. Carter, and Seth G.N. Grant Schizophrenia is a major psychiatric disease with strong evidence of genetic risk factors. Recent studies based on genome-wide study of copy number variations (CNVs) have detected novel recurrent submicroscopic copy number changes, including recurrent deletions at 1q21.11, 15q11.3, 15q13.3, and the recurrent CNV at the 2p16.3 neurexin 1 locus. These schizophrenia susceptibility CNV loci demonstrate that schizophrenia is, at least in part, genetic in origin and provide the basis for further investigation of mutations associated with the disease. The studies combined have also established the role of rare and—in sporadic cases— de novo variants in schizophrenia. Furthermore, neuronal-related genes and genetic pathways are starting to emerge from the CNV loci associated with schizophrenia. Here, we review the major findings in the recent literature, which begin to unravel the genetic and biological architecture of this complex human neuropsychiatric disorder. Key Words: Association study, copy number variation, neurodevelopment, schizophrenia, synapse

S

chizophrenia (OMIM181500) is a debilitating psychiatric illness with a prevalence of 1% worldwide. The disease is characterized by positive symptoms such as delusions, hallucinations, and disorganized thinking, together with negative symptoms, including cognitive and social impairment (1). Although family and twin studies have suggested the existence of a strong genetic component in schizophrenia (for review see [2]), the search for genetic susceptibility factors remains a challenge. Early schizophrenia genetic studies were based on linkage analysis, progressing to association studies of biological or positional candidates. Some of the early linkage and association studies generated reproducible loci leading to susceptibility genes or genetic regions with biologically plausible candidates such as neuregulin 1, dysbindin, D-amino acid oxidase activator (DAOA), and regulator of G-protein signaling 4 (for reviews see [2– 4]). These molecular genetic studies were complemented by identification of chromosomal abnormalities in patients (see review by [5]), as well as the latest genome-wide association studies (GWAS) with single nucleotide polymorphisms (SNPs) (6 –11). More recently, copy number variation (CNV) was established as a major source of human genetic variation, which could be linked to complex psychiatric diseases. This review will focus on the role of genomic structural abnormalities in schizophrenia, with particular emphasis on recent copy number variation studies.

Chromosomal Abnormality and Schizophrenia Chromosomal anomalies detected by cytogenetics and karyotyping have long been identified as mutations causing schizophrenia with high penetrance in familial studies (12). Two abnormalities in particular provided examples of rare chromosomal rearrangements in the pathogenesis of schizophrenia. 22q11 Microdeletion and Schizophrenia The 22q11 microdeletion syndrome (22q11DS; also known as velocardiofacial syndrome/DiGeorge syndrome [VCFS/DGS]) is a complex disorder responsible for learning disability, cardiac From The Wellcome Trust Sanger Institute, Hinxton, United Kingdom. Address correspondence to Seth G. N. Grant, M.B., B.S., The Wellcome Trust Sanger Institute, Hinxton, Cambridge CB10 1SA, UK; E-mail: sg3@sanger. ac.uk. Received Jan 30, 2009; revised Jul 21, 2009; accepted Jul 21, 2009.

0006-3223/09/$36.00 doi:10.1016/j.biopsych.2009.07.027

defects, craniofacial abnormalities, and palatal defects (13,14). It is caused by the recurrent deletion of a 3 megabase (Mb) region— or in some cases a smaller 1.5 Mb region—mediated by nonallelic homologous recombination (15–17). The manifestation of the neuropsychiatric phenotypes in the deletion syndrome is highly variable. Among patients with the 22q11 microdeletion syndrome, many have mental retardation with different degrees of cognitive deficits. There is also an unusually high prevalence (⬃30%) of behavioral abnormalities, including schizophrenia, bipolar disorder, autism, and other psychoses (18 –20). The 22q11 deletion is well accepted as a true genetic subtype of schizophrenia. The increased risk of schizophrenia likely results from haploinsufficiency of one or more genes or unmasking of deleterious genetic variant(s) at 22q11, although there has been limited success in delineating the exact gene(s) responsible for the behavioral phenotypes. Studies on point mutation screening and linkage disequilibrium mapping using SNPs have presented some candidate genes/regions (for example [21,22]). Efforts have been made to delineate the overlapping region and establish the minimal “critical regions” of the disorder. However, genotype-phenotype correlation has remained difficult, complicated with the discoveries of rare atypical deletions and the possibility of long-range effects. Even for patients previously thought to have indistinguishable deletion loci (under conventional cytogenetic methods), the size and the genes involved in the deletion are now under controversy. For example, studies using NimbleGen (Roche NimbleGen, Madison, Wisconsin) oligonucleotide array comparative genome hybridization (array CGH) and high-resolution paired-end mapping on patients with the same 22q11 deletion locus revealed differences of up to 14 genes and 200 kilobase (kb) either side of the breakpoint (23). These candidates may play a role in the variability of phenotypes in patients previously considered to have identical deletion breakpoints. Although the gene(s) predisposing to schizophrenia is (are) unknown, functional candidates within the deleted region have been proposed, including catechol-O-methyltransferase (COMT), proline dehydrogenase (PRODH), the micro-RNA processing gene DiGeorge syndrome critical region 8 (DGCR8), the brain expressed putative palmitoyltransferase ZDHHC8, and the guanine nucleotide binding protein beta-subunit-like polypeptide GNB1L. Genetic and clinical investigations into the syndrome have provided important insights into other genomic disorders caused by duplication or deletion of dosage-sensitive genes, with altered genomic segments frequently flanked by segmental duplications (24). BIOL PSYCHIATRY 2009;66:1005–1012 © 2009 Society of Biological Psychiatry

1006 BIOL PSYCHIATRY 2009;66:1005–1012 The Disrupted in Schizophrenia 1 Breakpoint in Schizophrenia Disrupted in schizophrenia 1 (DISC1) is a schizophrenia candidate gene with supporting evidence involving human genetics, molecular and cell biology, and insights from animal models. DISC1 was first identified as a balanced reciprocal translocation t(1;11) (q42;q14), which cosegregated in a large Scottish family with schizophrenia and other psychiatric diseases (25,26). The initial pedigree included 21 individuals with schizophrenia, bipolar disorder, recurrent depression, and conduct disorder, of which 16 were identified with the 1q42 translocation, giving a highly significant linkage signal with a logarithm of odds (LOD) score exceeding 7.0. Two brain-expressed genes were revealed at the breakpoint: Disrupted in Schizophrenia 1 and 2 (DISC1 and DISC2) (25,27). The 1q42 linkage result was replicated in a subsequent study in a Finnish cohort (28), and the involvement of DISC1 in schizophrenia and bipolar disorder was later suggested in a number of association studies (29 –31). These follow-up studies complemented the original rare Scottish translocation by targeting more common SNP alleles, which demonstrated lower odds ratio and suggested locus heterogeneity. Several mouse models have provided evidence of DISC1 involvement in behavior (32–35). DISC1 is involved in neurodevelopmental, cytoskeletal, and cyclic adenosine monophosphate (cAMP) signaling functions (36). The study of DISC1 and its interacting proteins have provided us with important clues into disease pathogenesis (37). In particular, phosphodiesterase 4B (PDE4B), an interacting partner of DISC1, was also revealed as a disease-causing chromosomal translocation (38). Both DISC1 and PDE4B are involved in cAMP signaling, an important pathway in neuronal signaling networks and the synapse (38,39). The DISC1 example illustrates the role of studying schizophrenia families to identify genetic susceptibility factors and the potential for combining protein interactions with genetics, a strategy embraced by systems biology.

Large-Scale CNV Screens in Schizophrenia Patients The detection of chromosomal rearrangements in patients with schizophrenia provided evidence that genome structural variation could alter complex behavioral traits. The first generation of CNV studies were mostly performed using array CGH platforms based on large insert clones, such as bacterial artificial chromosome arrays. Arrays using oligonucleotides as hybridization probes have increased CNV detection resolution dramatically. These techniques include representational oligonucleotide microarray analysis, a platform consisting of oligonucleotide probes hybridized with “representations” of the test and reference genome. Furthermore, arrays originally designed to genotype SNPs in genome-wide linkage association studies are now providing estimations of copy number changes as the associated bioinformatics tools have matured. The technical advancements led to a turning point in schizophrenia genetics in 2008 with the publication of several largescale whole-genome schizophrenia association studies reporting the detection of copy number variants (Table 1). These CNV studies not only uncovered particular schizophrenia susceptibility loci but also revealed key insights into the genetic architecture of the disease. Identification of New Schizophrenia Susceptibility Loci Two large-scale association studies using genome-wide SNP genotyping arrays have recently associated CNVs with schizophrenia (40,41) (Table 1). Together, the two reports revealed three novel schizophrenia-associated recurrent CNV loci at chrowww.sobp.org/journal

G.W.C. Tam et al. mosome 1q21.1, 15q13.3, and 15q11.2, with the former two replicated in both studies (Table 2). The regions at 1q21 and 15q13 were extremely rare events, occurring in ⬃.1% to .3% of cases and tenfold less frequent in control subjects. Estimated odds ratios for these two regions were high (⬎10 and 6 to 18 in the second study by the International Schizophrenia Consortium, herein referred to as ISC). The region at 15q11 was relatively more frequent among the three but still rare (.55%) in cases (and five times less frequent in control subjects), with odds ratio at 2.7. Of the genes affected in the three regions, a number of highly plausible schizophrenia candidates are worth mentioning. At 1q21, the gene connexin-50 (GJA8) encodes a gap junction subunit and was previously reported as associated with schizophrenia in case-control and family studies (42). This CNV region has also been associated with schizophrenia by linkage analysis (41). The 15q11.2 CNV partially overlaps with the Prader-Willi Syndrome locus (43) and maps to the breakpoints identified in some of the Prader-Willi patients (44). Within the ⬃500 kb region, there is the cytoplasmic FMR1 interacting protein 1 isoform (CYFIP1) gene, which encodes a protein that binds both the fragile X mental retardation protein (FMRP) and the translation initiation factor eIF4E (45). The CYFIP1/FMRP interaction regulates messenger RNA (mRNA) translation in neuronal dendrites, demonstrating its role in synaptic plasticity and brain development. It was noted that both Prader-Willi Syndrome patients and fragile X patients frequently have autistic features (46,47), while various psychiatric diseases such as autism were suggested to share genetic overlap with schizophrenia (48). Furthermore, FMRP was shown to regulate mRNA stability for the postsynaptic density protein 95 (PSD-95) (49,50), and as will be noted later, PSD-95 is a neurexin-interacting protein. Finally, at the third locus 15q13.3, the ␣7 nicotinic receptor gene CHRNA7 is a schizophrenia candidate gene (51) and was shown to be targeted by neuregulin1-erbB signaling to axons for surface expression in sensory neurons (52). A noteworthy characteristic of the three recurrent loci was that they were all flanked by segmental duplications or low copy repeats (LCRs), which, analogous to the 22q11 microdeletion, reflects the underlying architecture of the genome (40). Nonallelic homologous recombination (NAHR) could be mediated by these LCRs, generating recurrent deletions and reciprocal duplications (24). Consistent with this, reciprocal duplications have been observed for these recurrent loci (40). The mutational rate for this type of genomic rearrangement hotspot is estimated to be 10⫺6 to 10⫺4, manyfold higher than for SNPs (10⫺8) (53). This may partially explain the recurrent feature of schizophrenia at a similar rate in all populations and the maintenance of schizophrenia in the population even though the disease confers reduced productive fitness (54). Increased Mutation Burden in Schizophrenia Patients A finding of considerable interest from the ISC study was an increased mutational burden as conferred by rare CNVs (defined as CNV ⬍1% frequency in all samples screened) in schizophrenia cases versus control subjects. This excess of rare CNVs in patients had previously been described by Walsh et al. (55) using both representational oligonucleotide microarray analysis and SNP arrays (Table 1). In particular, Walsh et al. (55) reported a threefold increase in the proportion of rare, genic variants (those that delete or duplicate genes) compared with total CNVs in cases (15%) versus control subjects (5%). The effect was magnified when only childhood-onset cases were taken into account (20%), reflecting a possible increased genetic burden of child-

G.W.C. Tam et al.

Table 1. Summary of Recent Genome-Wide CNV Screens in Schizophrenia Patients Year

Platform

Cases

Control Subjects

Main Findings

International Schizophrenia Consortium (40)

2008

Affy SNP 5.0 & 6.0 array

3391

3181

Two recurrent deletions were found significantly associated with SCZ: 1q21.1 and 15q13.3 (⫹ previously detected 22q11) CNV burden is higher in cases than in control subjects both in terms of CNV count or by gene count CNVs detected at genes involved in the synapse, e.g., NRXN1 and APBA2 Three recurrent deletions were found significantly associated with SCZ: 1q21.1, 15q11.2 and 15q13.3 (⫹ previously detected 22q11) Odds ratio estimations for the 3 regions range from 2 to 15 Phase I: 13 rare CNVs in patients; 4 candidate loci involved in neuronal functions (NRXN1, MYT1L, AXTN2, and CTNND2) Phase II: additional 7 CNVs detected at 3 of the 4 candidate loci in patients (1 in control subject) CNV burden is higher in cases than in control subjects (genic CNVs) Genes implicated are overrepresented in categories related to brain development (e.g., ERBB4, SLC3A1, DLG2, and NRXN1) de novo CNV burden is higher in sporadic cases than in control trios (8x increase of de novo CNV in cases) No de novo CNV detected in familial cases

Kirov et al. (58)

2008

Tiling Path BAC Array

Steffanson et al. (41)

2008

Illumina HumanHap chip, Affymetrix SNP6.0, Taqman qPCR

1433 ⫹ 3285

Vrijenhoek et al. (59)

2008

Affy 250K SNP array, Illumina HumanHap550 chip

54 ⫹ 752

Walsh et al. (55)

2008

ROMA, Affy 500K SNP array

150 ⫹ 83 childhood-onset cases

268 ⫹ 154 nontransmitting chr from parents

Xu et al. (56)

2008

Affy SNP 5.0 array

Sporadic cases: 152 affected child (trios) Familial cases: 48 affected child (trios)

159 unaffected trios

93

Database of Genomic Variants ⫹ 372 33250 ⫹ 7951 (de novo CNV Discovery: 2160 trios and 5558 pairs) Previous literature ⫹ 706

BAC, bacterial artificial chromosome; CNV, copy number variation; ROMA, representational oligonucleotide microarray analysis; SCZ, schizophrenia; SNP, single nucleotide polymorphism.

www.sobp.org/journal

BIOL PSYCHIATRY 2009;66:1005–1012 1007

Author

1008 BIOL PSYCHIATRY 2009;66:1005–1012

G.W.C. Tam et al.

Table 2. Three Novel Recurrent Deletions Associated with Schizophrenia

Estimated Chr Coordinates Estimated Size Prevalence SGENE⫹ Case (n ⫽ 4718) Control (n ⫽ 41199) ISC Case (n ⫽ 3391) Control (n ⫽ 3181) Odds Ratio (p Value) SGENE⫹ ISC Affected Genes

1q21.1

15q11.2

15q13.3

chr1:142–146.3 (Mb) ⬃1.3 Mb (small)/⬃3.5 Mb (large)

chr15:20.31–20.78 (Mb) 470 kb

chr15:28–31 (Mb) ⬃3 Mb

.23% .02%

.55% .19%

.17% .02%

.2949% .0314%

NA NA

.2654% .0000%

14.83 (p ⫽ .024) 6.6 (p ⫽ .023) SEC22B, NOTCH2NL, HFE2, TXNIP, POLR3GL, LIX1L, RBM8A, PEX11B, ITGA10, ANKRD35, PIAS3, POLR3C, ZNF364, CD160, PDZK1, NBPF11, FAM108A3, PRKAB2, FMO5, CHD1L, BCL9, ACP6, GJA5, GJA8, GPR89A, NM_207400, NBPF15

2.73 (p ⫽ .007) NA TUBGCP5, CY FIP1, NIPA2, NIPA1, BC044583, WHDC1L1

11.54 (p ⫽ .040) 17.9 (p ⫽ .0023) CHRFAM7A, MTMR10, TRPM1, has-mir-211, KLF13, OTUD7A, CHRNA7

Chr, chromosome; ISC, Report from International Schizophrenia Consortium (40); kb, kilobase; Mb, megabase; NA, not applicable; SGENE⫹, Report from Stefansson et al. (41).

hood psychiatric disorders (55). The extent of the CNV burden was smaller in the ISC study compared with the report from Walsh et al. (55) (from 1.15 to 1.4 times increase). However, the signal was intensified (to 1.45 to 1.67) when only the extremely rare CNV events—the singletons occurring once in all samples screened—were considered (40). Another report addressed the role of de novo copy number variations in schizophrenia (56). By screening 200 trios with an affected child (152 sporadic cases and 48 familial cases) and a similar number of nonaffected control trios using Affymetrix SNP arrays (Affymetrix Inc., Santa Clara, California), the authors revealed an eightfold increased rate of de novo copy number mutations in sporadic cases compared with control subjects. The effect was much smaller in familial cases, showing a distinct difference in the genetic determinations of sporadic versus familial cases (56). These de novo CNVs were previously shown to play important roles in autism (57). Taken together, these studies came to the conclusion that rare variants were collectively significant risk factors for schizophrenia.

Rare Variants Converging into Synaptic Pathways? Although the role of rare CNVs in schizophrenia pathogenesis has been collectively demonstrated, the bulk of the published schizophrenia CNV regions are composed of unique events, with minimal overlap between studies with the exception of a few recurrent loci. It is therefore extremely difficult to pinpoint which CNV(s) or gene(s) is/are schizophrenia risk factors. As a method to decipher the molecular basis of the disease, one could look for convergence of molecular components into biological pathways. One common feature from these rare CNV events is the involvement of proteins related to synaptic development and functions (55,58,59), in particular those that contribute to neuregulin and neurexin signaling in the synapse, as highlighted in red in Figure 1. Of particular recent interest is the gene neurexin 1 (NRXN1), which was identified as a CNV locus in both studies and was later confirmed as one of the very few schizophreniaassociated recurrent CNV (60). Neurexins are presynaptic proteins that induce postsynaptic differentiation in dendrites. There are three neurexin genes—NRXN1, NRXN2, and NRXN3—at

Figure 1. Schematic diagram of the synapse displaying known involvement of synaptic proteins in schizophrenia CNV loci. The proteins identified as schizophrenia CNV loci are indicated in red. AMPAR, AMPA (⫺amino-3-hydroxy-5methyl-4-isoxazole propionic acid) receptor; ERBB4, v-erb-a erythroblastic leukemia viral oncogene; Ca⫹ channel, calcium channel; CASK, calcium/calmodulin-dependent serine protein; CNV, copy number variation; K⫹ channel, potassium channel; KIF17, kinesin family member 17; KIF1B␣, kinesin family member 1B ␣; MALS, lin-7 homolog; Mint, amyloid beta A4 precursor protein-binding; NMDAR, NMDA (N-methylD-aspartate) receptor; PSD-95, postsynaptic density protein 95; PSD-93, postsynaptic density protein 93; SAP-97, synapseassociated protein 97; SAP-102, synapse-associated protein 102.

www.sobp.org/journal

G.W.C. Tam et al. human chromosome loci 2q32, 11q13, and 14q24.3– q31.1, respectively. The 2p16.3 neurexin 1 locus has been previously shown as associated with autism (61– 65). Evidence suggesting its role in schizophrenia first came from Kirov et al. (58) who screened 93 patients using array CGH and identified a segregating NRXN1 CNV in a pair of affected schizophrenia siblings and their asymptomatic mother. Walsh et al. (55) also detected a neurexin CNV in a pair of identical twins diagnosed with childhood-onset schizophrenia. Subsequently, Vrijenhoek et al. (59) identified one CNV in the region by screening an initial cohort of 54 patients by array CGH and revealed another four cases when screening this locus in more patients. Furthermore, the ISC detected a number of neurexin CNVs with different breakpoints in cases (as well as in control subjects) (40). In addition to the above genome-wide studies, Rejuscu et al. (60) recently targeted the three neurexin genes using a candidate CNV approach to look for structural rearrangement in ⬃3000 patients and ⬎30,000 control subjects (60). They identified 61 NRXN1 deletions (one de novo) and 5 duplications throughout the locus. By restricting the analysis to CNVs that disrupt exons, the authors revealed .17% in cases harboring such CNVs compared with .020% in control subjects (odds ratio ⬃10). This recurrent locus distinguishes itself from the three aforementioned large recurrent deletions in two ways. First, the breakpoints of the neurexin CNVs were not flanked by LCRs. There was no common breakpoint (consistent with previous findings), and the CNVs vary in size (18 kb to 420 kb). Long-range phasing analysis on the de novo case provided evidence against NAHR as a rearrangement-generating mechanism. Second, rather than deleting or duplicating whole gene(s) as described previously, CNVs in this locus most likely acted by disrupting the gene NRXN1. It was concluded that NRXN1 deletions (in particular the ones that affect exons) confer risk of schizophrenia (60). Neurexin 1 belongs to a large family of proteins that act as neuronal cell surface receptors. Neurexin is concentrated at the presynaptic membrane, where it functions as a neuroligin receptor (66). Neuroligins were shown to be sufficient to trigger presynaptic differentiation through neurexin (67), and neurexins could in turn trigger postsynaptic differentiation (68). This transsynaptic neurexin-neuroligin complex is important for excitatory glutamatergic and inhibitory gamma-aminobutyric acid (GABA)ergic synapses in the mammalian brain (68), playing a central role in synapse formation and neurotransmission. Moreover, neuroligin directly binds scaffold proteins in the membrane associated guanylate kinase (MAGUK) family (dlg2/ PSD-93/Chapsyn-110, dlg3/SAP102, dlg4/PSD-95/SAP90), and CNVs in PSD-93 were reported in schizophrenia (55), while SAP102 mutations were shown to result in X-linked mental retardation (69). As shown in Figure 1, these proteins are physically associated in macromolecular complexes in the postsynaptic terminal and link to the N-methyl-D-aspartate (NMDA) receptor, forming the MAGUK associated signaling complex (MASC) or NMDA receptor complex (NRC) of postsynaptic proteins (70 –73). This set of proteins is also involved with at least six forms of X-linked mental retardation (XLMR) and half of all genes involved with XLMR are encoded by postsynaptic proteome genes (74). Lists of genetic disorders involved with the NRC/MASC and the synapse proteome can be found at Genes to Cognition Database (http://www.genes2cognition.org/cgi-bin/ DiseaseView) (75). Of direct relevance to the NXRN1 CNV was a schizophrenia study identifying 3 patients carrying genomic deletions of the CNTNAP2 gene, while no CNV was detected in 512 control

BIOL PSYCHIATRY 2009;66:1005–1012 1009 subjects (76). CNTNAP2 encodes contactin-associated protein 2 (Caspr2), a member of the neurexin superfamily (77). Furthermore, the same study that first detected the NXRN1 disruption in schizophrenia revealed in another patient a CNV involving APBA2 (Mint2) (58), a neuronal adapter protein that binds neurexins as part of a CASK-containing protein complex (78). The study by Walsh et al. (55) revealed a number of CNVs disrupting genes involved in synaptic transmission or neurodevelopment. Of particular interest is the neuregulin (NRG) signaling pathway, which has diverse roles from neuronal migration, axon guidance, glial cell development, axon myelination, neurotransmitter receptor expression, and synapse formation (79). These important functions position neuregulin as a key signaling protein for synaptic plasticity and neuronal survival. Genes involved in the NRG pathway and disrupted by CNVs included two interacting partners, ERBB4 and MAGI2 (55). ERBB4 is a type I transmembrane tyrosine kinase receptor for neuregulin-1 (NRG1). It was detected as a deletion in a patient. The neuregulin-erbB complex was suggested to interact with PDZ domains (originally found in PSD-95, Dlg, and ZO-1) containing proteins at neuronal synapses (80,81). The complex interacts with MAGI2 at neuronal synapses, detected as duplicated in another patient (55). In addition, ERBB4 was suggested to be recruited by PSD95 (a PDZ-containing protein) to the neurexin-neuroligin complex for synapse formation (82), presenting a link between the CNVs discussed in this section so far. This evidence has provided a glimpse into the complexity of the neuronal network, suggesting how the apparent genetic heterogeneity revealed by the schizophrenia CNV screens could be reconciled, although caution must be taken when linking rare single events in relation to diseases, as some neuronal genes were also detected in CNVs in control individuals (83). Finally, the emergence of larger CNV datasets, combined with large-scale GWAS approaches to reveal both common and rare diseaseassociated SNPs in schizophrenia, could help confirm some of these candidate genes or genetic regions and strengthen statistical power to evaluate frequency differences.

Understanding the Genetic Model of Schizophrenia One controversy surrounding schizophrenia genetics and many other complex disease traits is the debate between common disease-common variant (CD-CV) (84) versus common disease-rare variant (CD-RV) models (85,86). The CD-CV model proposes that common alleles with small to moderate disease risks may have an additive or multiplicative effect on schizophrenia (87) and have inspired many SNP association studies investigating common polymorphisms in schizophrenia. Recent largescale association studies have successfully identified such risk variants (7,9 –11). The CD-CV model alone, however, is not able to explain the genetic architecture of schizophrenia. Another proposed theory that has received increasing attention in schizophrenia genetics in the past decade is the CD-RV model, which suggests heterogeneity of schizophrenia comes from multiple rare variants (85,88). Recent CNV studies further demonstrated that rare risk loci, individually or collectively, could predispose to schizophrenia (40,41), supporting the CD-RV model. Apart from locus heterogeneity, CNV studies showed that even within the same disease-associated genomic locus there could be allelic heterogeneity, as in the case of the neurexin1 gene (60). Some of these rare alleles could be recurrent in the population (either inherited or sporadic) but would remain at low frequency due to the www.sobp.org/journal

1010 BIOL PSYCHIATRY 2009;66:1005–1012 selective disadvantage they confer (86), while others may be private events found only in single individuals or families. By contrast, current CNV discoveries may be skewed toward the lower end of the minor allelic frequency spectrum (89), possibly due to the bias of CNV platforms and algorithms detecting relatively large mutations. As advances in CNV discovery techniques allow accurate detection of smaller, more frequent copy number polymorphisms, together with increasing data released from association studies using SNPs (9 –11), we are starting to detect common alleles associated with schizophrenia. These may translate into neuronal functions by, for example, altering gene dosage or protein functions. Genetic studies of other complex diseases (e.g., Alzheimer’s disease) have demonstrated that the CD-CV and CD-RV theories need not be mutually exclusive in disease populations (85). Copy number variation studies on schizophrenia combined with SNP association have further suggested that these two models coexist, leading to the genetic heterogeneity in schizophrenia.

Etiological Overlap of Schizophrenia with Other Psychiatric Diseases The well-documented link between the 22q11 deletion syndrome and schizophrenia demonstrated that the same structural variant can be associated with numerous psychiatric phenotypes, from schizophrenia to attention-deficit/hyperactivity disorder, bipolar disorders, and autism spectrum disorders (ASDs) (3,87,90). These and other genetic loci argued for pleiotropic genetic effects among psychiatric disorders. The major recurrent CNVs reported on schizophrenia so far seem to have reinforced this notion of etiological overlap among neuropsychiatric diseases. The 1q21.1 region, for instance, was found to be associated with a range of developmental disorders including autism (91). The 15q13.3 recurrent deletion was also previously associated with mental retardation and seizures (92), as well as with ASDs (93). Moreover, the 15q11.2 recurrent deletion is a region involved in Prader-Willi/Angelman Syndrome with known cognitive and neurological impairment (94), and a subset of patients also exhibited autistic features (95). A recent study using genomewide expression profiling to investigate autistic patients with different genetic etiology, including the 15q duplication, suggested that dsyregulation (increased expression) of gene CYFIP1 could be a molecular link between autistic patients with different genetic background factors (96). Instead of a duplication, this gene is deleted within the schizophrenia CNV loci. Furthermore, recurrent schizophrenia CNV in neurexin1 (55,58,60) have also been detected in autism (97) and mental retardation patients (98), and neurexin1 point mutations have been implicated in autism (61,99). Interestingly, the specificity of the association of neurexin proteins with autism (that only neurexin1 is involved) seems to be recapitulated in schizophrenia. These novel data from CNV research suggested that some genetic risk loci apparently give rise to diverse psychopathology, leading to several diseases falling across diagnostic boundaries. This may be due to a number of reasons, for example, the limitations of traditional diagnostic boundaries used to classify neuropsychiatric disorders into distinct disease entities or the variable expressivity of the genes involved. Given the subtlety and intricacy of the human nervous system, one can imagine a small deviation in gene expression, for example, or the existence of an interacting genetic or environmental factor could change the balance in certain biological pathway(s) and result in different psychiatric phenotypes. Moreover, the lessons learned from the genetics www.sobp.org/journal

G.W.C. Tam et al. and biology of schizophrenia might provide further insight into our understanding of psychiatric disorders as a whole.

Conclusions The identification of CNVs as schizophrenia candidate loci begins to provide insights into the biological basis and genetic model of the disease. Looking forward, CNVs are just the initial step toward understanding of schizophrenia genetics. The identification of SNPs by linkage and association studies has generated a number replicated loci. Recent GWAS have further revealed both common and rare alleles. Multiple risk factors in schizophrenia, including those that are genetic (SNPs, small insertions/deletions, inversions, and translocations), epigenetic, environmental, or stochastic events, are all part of the comprehensive picture of this complex psychiatric illness. In addition, the same genetic locus may harbor multiple allelic variants of different types, further complicating the issue. The genetic architecture of schizophrenia, taking into account the number of variants at a given disease locus, their frequencies, and the penetrance of the genotype combinations (100), will add another layer of complexity to our comprehension of this disease. Improvement of genomic technologies—in particular more accurate, higher-resolution, and higher-throughput CNV detection platforms, as well as advancement in sequencing technologies— will facilitate the search for such schizophrenia risk factors in a near future. New bioinformatic tools for CNV analysis will also be indispensable. Better algorithms for CNV detection and genotyping are needed (89,101,102), together with improved statistical methods to look for disease association (100). Databases for CNV information will also be important in providing baseline comparison and new discoveries. For instance, the Database of Genomic Variants (http:// projects.tcag.ca/variation/) and DatabasE of Chromosomal Imbalance and Phenotype in Humans using Ensembl Resources (https:// decipher.sanger.ac.uk/) are two public repositories collating CNV discoveries in normal and affected individuals, respectively. Databases dedicated to schizophrenia CNVs, such as one analogous to the Autism Chromosome Rearrangement Database (http://projects.tcag.ca/autism/), will also be valuable to the schizophrenia research and clinical community. Finally, the rate of genetic discovery is not being met with equivalent capacity for efficient functional testing of the variants by biological assays. A systematic linking of genetic and molecular biological experimentation will expedite the search for schizophrenia risk factors and provide a rational basis for new drug discovery.

This work was supported by The Wellcome Trust and The Wellcome Trust Genes to Cognition Programme. The authors reported no biomedical financial interests or potential conflicts of interest. 1. Andreasen NC (1995): Symptoms, signs, and diagnosis of schizophrenia. Lancet 346:477– 481. 2. Sullivan PF (2005): The genetics of schizophrenia. PLoS Med 2:e212. 3. Burmeister M, McInnis MG, Zollner S (2008): Psychiatric genetics: Progress amid controversy. Nat Rev Genet 9:527–540. 4. Owen MJ, Craddock N, O’Donovan MC (2005): Schizophrenia: Genes at last? Trends Genet 21:518 –525. 5. Bassett AS, Chow EW, Weksberg R (2000): Chromosomal abnormalities and schizophrenia. Am J Med Genet 97:45–51. 6. Lencz T, Morgan TV, Athanasiou M, Dain B, Reed CR, Kane JM, et al. (2007): Converging evidence for a pseudoautosomal cytokine receptor gene locus in schizophrenia. Mol Psychiatry 12:572–580.

G.W.C. Tam et al. 7. O’Donovan MC, Craddock N, Norton N, Williams H, Peirce T, Moskvina V, et al. (2008): Identification of loci associated with schizophrenia by genome-wide association and follow-up. Nat Genet 40:1053–1055. 8. Sullivan PF, Lin D, Tzeng JY, van den Oord E, Perkins D, Stroup TS, et al. (2008): Genomewide association for schizophrenia in the CATIE study: Results of stage 1. Mol Psychiatry 13:570 –584. 9. Purcell SM, Wray NR, Stone JL, Visscher PM, O’Donovan MC, Sullivan PF, Sklar P (2009): Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature 460:748 –752. 10. Shi J, Levinson DF, Duan J, Sanders AR, Zheng Y, Pe’er I, et al. (2009): Common variants on chromosome 6p22.1 are associated with schizophrenia. Nature 460:753–757. 11. Stefansson H, Ophoff RA, Steinberg S, Andreassen OA, Cichon S, Rujescu D, et al. (2009): Common variants conferring risk of schizophrenia. Nature 460:744 –747. 12. MacIntyre DJ, Blackwood DH, Porteous DJ, Pickard BS, Muir WJ (2003): Chromosomal abnormalities and mental illness. Mol Psychiatry 8:275–287. 13. Karayiorgou M, Morris MA, Morrow B, Shprintzen RJ, Goldberg R, Borrow J, et al. (1995): Schizophrenia susceptibility associated with interstitial deletions of chromosome 22q11. Proc Natl Acad Sci U S A 92: 7612–7616. 14. Murphy KC (2002): Schizophrenia and velo-cardio-facial syndrome. Lancet 359:426 – 430. 15. Edelmann L, Pandita RK, Spiteri E, Funke B, Goldberg R, Palanisamy N, et al. (1999): A common molecular basis for rearrangement disorders on chromosome 22q11. Hum Mol Genet 8:1157–1167. 16. Shaikh TH, Kurahashi H, Emanuel BS (2001): Evolutionarily conserved low copy repeats (LCRs) in 22q11 mediate deletions, duplications, translocations, and genomic instability: An update and literature review. Genet Med 3:6 –13. 17. Sharp AJ, Locke DP, McGrath SD, Cheng Z, Bailey JA, Vallente RU, et al. (2005): Segmental duplications and copy-number variation in the human genome. Am J Hum Genet 77:78 – 88. 18. Fine SE, Weissman A, Gerdes M, Pinto-Martin J, Zackai EH, McDonaldMcGinn DM, et al. (2005): Autism spectrum disorders and symptoms in children with molecularly confirmed 22q11.2 deletion syndrome. J Autism Dev Disord 35:461– 470. 19. Murphy KC, Jones LA, Owen MJ (1999): High rates of schizophrenia in adults with velo-cardiofacial syndrome. Arch Gen Psychiatry 56:940 –945. 20. Murphy KC, Owen MJ (2001): Velo-cardio-facial syndrome: A model for understanding the genetics and pathogenesis of schizophrenia. Br J Psychiatry 179:397– 402. 21. Chen WY, Shi YY, Zheng YL, Zhao XZ, Zhang GJ, Chen SQ, et al. (2004): Case-control study and transmission disequilibrium test provide consistent evidence for association between schizophrenia and genetic variation in the 22q11 gene ZDHHC8. Hum Mol Genet 13:2991–2995. 22. Liu H, Heath SC, Sobin C, Roos JL, Galke BL, Blundell ML, et al. (2002): Genetic variation at the 22q11 PRODH2/DGCR6 locus presents an unusual pattern and increases susceptibility to schizophrenia. Proc Natl Acad Sci U S A 99:3717–3722. 23. Urban AE, Korbel JO, Selzer R, Richmond T, Hacker A, Popescu GV, et al. (2006): High-resolution mapping of DNA copy alterations in human chromosome 22 using high-density tiling oligonucleotide arrays. Proc Natl Acad Sci U S A 103:4534 – 4539. 24. Lupski JR, Stankiewicz PT (2006): Genomic Disorders: The Genomic Basis of Disease. Towata, NJ: Humana Press, 448. 25. Blackwood DH, Fordyce A, Walker MT, St. Clair DM, Porteous DJ, Muir WJ (2001): Schizophrenia and affective disorders— cosegregation with a translocation at chromosome 1q42 that directly disrupts brainexpressed genes: Clinical and P300 findings in a family. Am J Hum Genet 69:428 – 433. 26. St. Clair D, Blackwood D, Muir W, Carothers A, Walker M, Spowart G, et al. (1990): Association within a family of a balanced autosomal translocation with major mental illness. Lancet 336:13–16. 27. Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, Semple CA, et al. (2000): Disruption of two novel genes by a translocation co-segregating with schizophrenia. Hum Mol Genet 9:1415–1423. 28. Ekelund J, Hovatta I, Parker A, Paunio T, Varilo T, Martin R, et al. (2001): Chromosome 1 loci in Finnish schizophrenia families. Hum Mol Genet 10:1611–1617. 29. Hennah W, Thomson P, McQuillin A, Bass N, Loukola A, Anjorin A, et al. (2008): DISC 1 association, heterogeneity and interplay in schizophrenia and bipolar disorder. Mol Psychiatry 14:865– 873.

BIOL PSYCHIATRY 2009;66:1005–1012 1011 30. Hodgkinson CA, Goldman D, Jaeger J, Persaud S, Kane JM, Lipsky RH, et al. (2004): Disrupted in schizophrenia 1 (DISC1): Association with schizophrenia, schizoaffective disorder, and bipolar disorder. Am J Hum Genet 75:862– 872. 31. Kockelkorn TT, Arai M, Matsumoto H, Fukuda N, Yamada K, Minabe Y, et al. (2004): Association study of polymorphisms in the 5= upstream region of human DISC1 gene with schizophrenia. Neurosci Lett 368:41– 45. 32. Clapcote SJ, Lipina TV, Millar JK, Mackie S, Christie S, Ogawa F, et al. (2007): Behavioral phenotypes of Disc1 missense mutations in mice. Neuron 54:387– 402. 33. Koike H, Arguello PA, Kvajo M, Karayiorgou M, Gogos JA (2006): Disc1 is mutated in the 129S6/SvEv strain and modulates working memory in mice. Proc Natl Acad Sci U S A 103:3693–3697. 34. Pletnikov MV, Ayhan Y, Nikolskaia O, Xu Y, Ovanesov MV, Huang H, et al. (2008): Inducible expression of mutant human DISC1 in mice is associated with brain and behavioral abnormalities reminiscent of schizophrenia. Mol Psychiatry 13:173–186, 115. 35. Shen S, Lang B, Nakamoto C, Zhang F, Pu J, Kuan SL, et al. (2008): Schizophrenia-related neural and behavioral phenotypes in transgenic mice expressing truncated Disc1. J Neurosci 28:10893–10904. 36. Chubb JE, Bradshaw NJ, Soares DC, Porteous DJ, Millar JK (2008): The DISC locus in psychiatric illness. Mol Psychiatry 13:36 – 64. 37. Brandon NJ (2007): Dissecting DISC1 function through protein-protein interactions. Biochem Soc Trans 35:1283–1286. 38. Millar JK, Pickard BS, Mackie S, James R, Christie S, Buchanan SR, et al. (2005): DISC1 and PDE4B are interacting genetic factors in schizophrenia that regulate cAMP signaling. Science 310:1187–1191. 39. Bradshaw NJ, Ogawa F, Antolin-Fontes B, Chubb JE, Carlyle BC, Christie S, et al. (2008): DISC1, PDE4B, and NDE1 at the centrosome and synapse. Biochem Biophys Res Commun 377:1091–1096. 40. International Schizophrenia Consortium (2008): Rare chromosomal deletions and duplications increase risk of schizophrenia. Nature 455:237–241. 41. Stefansson H, Rujescu D, Cichon S, Pietilainen OP, Ingason A, Steinberg S, et al. (2008): Large recurrent microdeletions associated with schizophrenia. Nature 455:232–236. 42. Ni X, Valente J, Azevedo MH, Pato MT, Pato CN, Kennedy JL (2007): Connexin 50 gene on human chromosome 1q21 is associated with schizophrenia in matched case control and family-based studies. J Med Genet 44:532–536. 43. Christian SL, Robinson WP, Huang B, Mutirangura A, Line MR, Nakao M, et al. (1995): Molecular characterization of two proximal deletion breakpoint regions in both Prader-Willi and Angelman’s syndrome patients. Am J Hum Genet 57:40 – 48. 44. Murthy SK, Nygren AO, El Shakankiry HM, Schouten JP, Al Khayat AI, Ridha A, et al. (2007): Detection of a novel familial deletion of four genes between BP1 and BP2 of the Prader-Willi/Angelman syndrome critical region by oligo-array CGH in a child with neurological disorder and speech impairment. Cytogenet Genome Res 116:135–140. 45. Napoli I, Mercaldo V, Boyl PP, Eleuteri B, Zalfa F, De Rubeis S, et al. (2008): The fragile X syndrome protein represses activity-dependent translation through CYFIP1, a new 4E-BP. Cell 134:1042–1054. 46. Dimitropoulos A, Schultz RT (2007): Autistic-like symptomatology in PraderWilli syndrome: A review of recent findings. Curr Psychiatry Rep 9:159–164. 47. Rogers SJ, Wehner DE, Hagerman R (2001): The behavioral phenotype in fragile X: Symptoms of autism in very young children with fragile X syndrome, idiopathic autism, and other developmental disorders. J Dev Behav Pediatr 22:409 – 417. 48. Rzhetsky A, Wajngurt D, Park N, Zheng T (2007): Probing genetic overlap among complex human phenotypes. Proc Natl Acad Sci U S A 104: 11694 –11699. 49. Todd PK, Mack KJ, Malter JS (2003): The fragile X mental retardation protein is required for type-I metabotropic glutamate receptor-dependent translation of PSD-95. Proc Natl Acad Sci U S A 100:14374 –14378. 50. Zalfa F, Eleuteri B, Dickson KS, Mercaldo V, De Rubeis S, di Penta A, et al. (2007): A new function for the fragile X mental retardation protein in regulation of PSD-95 mRNA stability. Nat Neurosci 10:578 –587. 51. Freedman R, Coon H, Myles-Worsley M, Orr-Urtreger A, Olincy A, Davis A, et al. (1997): Linkage of a neurophysiological deficit in schizophrenia to a chromosome 15 locus. Proc Natl Acad Sci U S A 94:587–592. 52. Hancock ML, Canetta SE, Role LW, Talmage DA (2008): Presynaptic type III neuregulin1-ErbB signaling targets {alpha}7 nicotinic acetylcholine receptors to axons. J Cell Biol 181:511–521.

www.sobp.org/journal

1012 BIOL PSYCHIATRY 2009;66:1005–1012 53. Lupski JR (2007): Genomic rearrangements and sporadic disease. Nat Genet 39:S43–S47. 54. Bassett AS, Bury A, Hodgkinson KA, Honer WG (1996): Reproductive fitness in familial schizophrenia. Schizophr Res 21:151–160. 55. Walsh T, McClellan JM, McCarthy SE, Addington AM, Pierce SB, Cooper GM, et al. (2008): Rare structural variants disrupt multiple genes in neurodevelopmental pathways in schizophrenia. Science 320:539 –543. 56. Xu B, Roos JL, Levy S, van Rensburg EJ, Gogos JA, Karayiorgou M (2008): Strong association of de novo copy number mutations with sporadic schizophrenia. Nat Genet 40:880 – 885. 57. Sebat J, Lakshmi B, Malhotra D, Troge J, Lese-Martin C, Walsh T, et al. (2007): Strong association of de novo copy number mutations with autism. Science 316:445– 449. 58. Kirov G, Gumus D, Chen W, Norton N, Georgieva L, Sari M, et al. (2008): Comparative genome hybridization suggests a role for NRXN1 and APBA2 in schizophrenia. Hum Mol Genet 17:458 – 465. 59. Vrijenhoek T, Buizer-Voskamp JE, van der Stelt I, Strengman E, Sabatti C, Geurts van Kessel A, et al. (2008): Recurrent CNVs disrupt three candidate genes in schizophrenia patients. Am J Hum Genet 83:504 –510. 60. Rujescu D, Ingason A, Cichon S, Pietilainen OP, Barnes MR, Toulopoulou T, et al. (2009): Disruption of the neurexin 1 gene is associated with schizophrenia. Hum Mol Genet 18:988 –996. 61. Feng J, Schroer R, Yan J, Song W, Yang C, Bockholt A, et al. (2006): High frequency of neurexin 1 beta signal peptide structural variants in patients with autism. Neurosci Lett 409:10 –13. 62. Kim HG, Kishikawa S, Higgins AW, Seong IS, Donovan DJ, Shen Y, et al. (2008): Disruption of neurexin 1 associated with autism spectrum disorder. Am J Hum Genet 82:199 –207. 63. Marshall CR, Noor A, Vincent JB, Lionel AC, Feuk L, Skaug J, et al. (2008): Structural variation of chromosomes in autism spectrum disorder. Am J Hum Genet 82:477– 488. 64. Szatmari P, Paterson AD, Zwaigenbaum L, Roberts W, Brian J, Liu XQ, et al. (2007): Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nat Genet 39:319 –328. 65. Zahir FR, Baross A, Delaney AD, Eydoux P, Fernandes ND, Pugh T, et al. (2008): A patient with vertebral, cognitive and behavioural abnormalities and a de novo deletion of NRXN1alpha. J Med Genet 45:239 –243. 66. Reissner C, Klose M, Fairless R, Missler M (2008): Mutational analysis of the neurexin/neuroligin complex reveals essential and regulatory components. Proc Natl Acad Sci U S A 105:15124 –15129. 67. Dean C, Scholl FG, Choih J, DeMaria S, Berger J, Isacoff E, et al. (2003): Neurexin mediates the assembly of presynaptic terminals. Nat Neurosci 6:708 –716. 68. Craig AM, Kang Y (2007): Neurexin-neuroligin signaling in synapse development. Curr Opin Neurobiol 17:43–52. 69. Tarpey P, Parnau J, Blow M, Woffendin H, Bignell G, Cox C, et al. (2004): Mutations in the DLG3 gene cause nonsyndromic X-linked mental retardation. Am J Hum Genet 75:318 –324. 70. Collins MO, Husi H, Yu L, Brandon JM, Anderson CN, Blackstock WP, et al. (2006): Molecular characterization and comparison of the components and multiprotein complexes in the postsynaptic proteome. J Neurochem 97(suppl 1):16 –23. 71. Husi H, Ward MA, Choudhary JS, Blackstock WP, Grant SG (2000): Proteomic analysis of NMDA receptor-adhesion protein signaling complexes. Nat Neurosci 3:661– 669. 72. Pocklington AJ, Cumiskey M, Armstrong JD, Grant SG (2006): The proteomes of neurotransmitter receptor complexes form modular networks with distributed functionality underlying plasticity and behaviour. Mol Syst Biol 2:2006.0023. 73. Ferna´ndez E, Collins MO, Uren RT, Kopanitsa MV, Komiyama NH, Croning MD, et al. (2009): Targeted tandem affinity purification of PSD-95 recovers core postsynaptic complexes and schizophrenia susceptibility proteins. Mol Syst Biol 5:269. 74. Laumonnier F, Cuthbert PC, Grant SG (2007): The role of neuronal complexes in human X-linked brain diseases. Am J Hum Genet 80:205–220. 75. Croning MD, Marshall MC, McLaren P, Armstrong JD, Grant SG (2009): G2Cdb: The Genes to Cognition database. Nucleic Acids Res 37:D846–D851. 76. Friedman JI, Vrijenhoek T, Markx S, Janssen IM, van der Vliet WA, Faas BH, et al. (2008): CNTNAP2 gene dosage variation is associated with schizophrenia and epilepsy. Mol Psychiatry 13:261–266.

www.sobp.org/journal

G.W.C. Tam et al. 77. Poliak S, Gollan L, Martinez R, Custer A, Einheber S, Salzer JL, et al. (1999): Caspr2, a new member of the neurexin superfamily, is localized at the juxtaparanodes of myelinated axons and associates with K⫹ channels. Neuron 24:1037–1047. 78. Biederer T, Sudhof TC (2000): Mints as adaptors. Direct binding to neurexins and recruitment of munc18. J Biol Chem 275:39803–39806. 79. Mei L, Xiong WC (2008): Neuregulin 1 in neural development, synaptic plasticity and schizophrenia. Nat Rev Neurosci 9:437– 452. 80. Nourry C, Grant SG, Borg JP (2003): PDZ domain proteins: plug and play! Sci STKE 2003(179):RE7. 81. Garcia RA, Vasudevan K, Buonanno A (2000): The neuregulin receptor ErbB-4 interacts with PDZ-containing proteins at neuronal synapses. Proc Natl Acad Sci U S A 97:3596 –3601. 82. Lin W, Sanchez HB, Deerinck T, Morris JK, Ellisman M, Lee KF (2000): Aberrant development of motor axons and neuromuscular synapses in erbB2-deficient mice. Proc Natl Acad Sci U S A 97:1299 –1304. 83. Cantor RM, Geschwind DH (2008): Schizophrenia: Genome, interrupted. Neuron 58:165–167. 84. Lohmueller KE, Pearce CL, Pike M, Lander ES, Hirschhorn JN (2003): Metaanalysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nat Genet 33:177–182. 85. McClellan JM, Susser E, King MC (2007): Schizophrenia: A common disease caused by multiple rare alleles. Br J Psychiatry 190:194 –199. 86. Pritchard JK (2001): Are rare variants responsible for susceptibility to complex diseases? Am J Hum Genet 69:124 –137. 87. Cook EH Jr, Scherer SW (2008): Copy-number variations associated with neuropsychiatric conditions. Nature 455:919 –923. 88. Craddock N, O’Donovan MC, Owen MJ (2007): Phenotypic and genetic complexity of psychosis. Invited commentary on Schizophrenia: A common disease caused by multiple rare alleles. Br J Psychiatry 190:200 –203. 89. Ionita-Laza I, Perry GH, Raby BA, Klanderman B, Lee C, Laird NM, et al. (2008): On the analysis of copy-number variations in genome-wide association studies: A translation of the family-based association test. Genet Epidemiol 32:273–284. 90. Antshel KM, Aneja A, Strunge L, Peebles J, Fremont WP, Stallone K, et al. (2007): Autistic spectrum disorders in velo-cardio facial syndrome (22q11.2 deletion). J Autism Dev Disord 37:1776 –1786. 91. Mefford HC, Sharp AJ, Baker C, Itsara A, Jiang Z, Buysse K, et al. (2008): Recurrent rearrangements of chromosome 1q21.1 and variable pediatric phenotypes. N Engl J Med 359:1685–1699. 92. Sharp AJ, Mefford HC, Li K, Baker C, Skinner C, Stevenson RE, et al. (2008): A recurrent 15q13.3 microdeletion syndrome associated with mental retardation and seizures. Nat Genet 40:322–328. 93. Veltman MW, Craig EE, Bolton PF (2005): Autism spectrum disorders in Prader-Willi and Angelman syndromes: A systematic review. Psychiatr Genet 15:243–254. 94. Cassidy SB, Dykens E, Williams CA (2000): Prader-Willi and Angelman syndromes: Sister imprinted disorders. Am J Med Genet 97:136 –146. 95. Hogart A, Wu D, Lasalle JM, Schanen NC (2008): The comorbidity of autism with the genomic disorders of chromosome 15q11.2-q13 [published online ahead of print September 18]. Neurobiol Dis. doi:10.1016/ j.nbd.2008.08.011. 96. Nishimura Y, Martin CL, Vazquez-Lopez A, Spence SJ, Alvarez-Retuerto AI, Sigman M, et al. (2007): Genome-wide expression profiling of lymphoblastoid cell lines distinguishes different forms of autism and reveals shared pathways. Hum Mol Genet 16:1682–1698. 97. Autism Genome Project Consortium (2007): Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nat Genet 39:319 –328. 98. Friedman JM, Baross A, Delaney AD, Ally A, Arbour L, Armstrong L, et al. (2006): Oligonucleotide microarray analysis of genomic imbalance in children with mental retardation. Am J Hum Genet 79:500 –513. 99. Yan J, Noltner K, Feng J, Li W, Schroer R, Skinner C, et al. (2008): Neurexin 1alpha structural variants associated with autism. Neurosci Lett 438:368 –370. 100. Pritchard JK, Cox NJ (2002): The allelic architecture of human disease genes: Common disease-common variantѧor not? Hum Mol Genet 11:2417–2423. 101. Barnes C, Plagnol V, Fitzgerald T, Redon R, Marchini J, Clayton D, et al. (2008): A robust statistical method for case-control association testing with copy number variation. Nat Genet 40:1245–1252. 102. McCarroll SA, Altshuler DM (2007): Copy-number variation and association studies of human disease. Nat Genet 39:S37–S42.