Copy Number Variants for Schizophrenia and Related Psychotic Disorders in Oceanic Palau: Risk and Transmission in Extended Pedigrees
Background
We report on copy number variants (CNVs) found in Palauan subjects ascertained for schizophrenia and related psychotic disorders in extended pedigrees in Palau. We compare CNVs found in this Oceanic population to those seen in other samples, typically of European ancestry. Assessing CNVs in Palauan extended pedigrees yields insight into the evolution of risk CNVs, such as how they arise, are transmitted, and are lost from populations by stochastic or selective processes, none of which is easily measured from case-control samples.
Methods
DNA samples from 197 subjects affected with schizophrenia and related psychotic disorders, 185 of their relatives, and 159 controls were successfully characterized for CNVs using Affymetrix Genomewide Human SNP Array 5.0.
Results
CNVs thought to be associated with risk for schizophrenia and related disorders also occur in affected individuals in Palau, specifically 15q11.2 and 1q21.1 deletions, partial duplication of IL1RAPL1 (Xp21.3), and chromosome X duplications (Klinefleter’s syndrome). Partial duplication within A2BP1 appears to convey an 8-fold increased risk in males (95% CI, 0.8–84.4) but not females (OR=0.4, 95% CI, 0.03–4.9). Affected-only linkage analysis using this variant yields a LOD score of 3.5.
Conclusions
This study reveals CNVs that confer risk to schizophrenia and related psychotic disorders in Palau, most of which have been previously observed in samples of European ancestry. Only a few of these CNVs show evidence that they have existed for many generations, consistent with risk variants diminishing reproductive success.
INTRODUCTION
Schizophrenia, like so many psychiatric disorders, shows a puzzling pattern of complex inheritance that has complicated efforts to discover susceptibility genes. Recently several inherited and de novo copy number variants (CNVs) have been shown to be a replicable source of risk for schizophrenia and related psychotic disorders. These CNVs are rare and account for only a small portion of risk. Still, by implicating genes and biological pathways, they could yield a deeper understanding of the genetic etiology of psychotic disorders.
One risk CNV, deletion of the 22q11.2 – the velocardiofacial syndrome or VCFS region (OMIM #192430) – has long been known (1). The VCFS region covers many genes in a 1.5 to 3 Mb region. Those with a VCFS deletion appear to have at least a twenty fold increased lifetime risk of a psychotic disorder, principally schizophrenia. A recently-identified and confirmed region is a roughly 1.5 Mb region of 1q21.1 (2–5). Typically risk is associated with deletion of the region and the many genes therein. Another is duplication/deletion of 15q11.2, between breakpoints one and two of the Prader-Willi/Angelman syndrome region (2, 4). Other genomic regions associated with schizophrenia, autism, and other psychiatric phenotypes are 3q29, 16p11.2, and 17q12 (6–14). Somewhat less support is available for several other genomic regions, including 15q13.3, 16p13.1, and 17p12 (3, 4, 16–18), but they too are likely risk loci. None of these CNV risk regions appear specific to schizophrenia or even psychosis. In fact myriad phenotypes are associated with deletions and sometimes duplications in these regions, including intellectual disability, autism, and attention deficit hyperactivity disorder (ADHD) (6, 8, 10, 12, 13, 18,19).
We use data from the Palauan population to determine if previously-identified risk CNVs also affect risk in Palau (20–24). The nation of Palau is an archipelago of islands in the South Pacific. Geographically isolated, it is situated about 550 miles east of the Philippines. Its nearest neighbor is Yap, approximately 400 miles northeast. Evidence suggests the population of Palau was founded by East Asians and developed over approximately 2,000 years in relative isolation until two centuries ago. Because most studies of psychosis involve samples of European ancestry, analysis of CNVs in a sample from Palau allows us to seek similarities and differences across ancestries. In addition, nearly complete ascertainment of Palauan subjects affected by schizophrenia, most of whom fall in extended pedigrees, could shed some light on the evolution of risk CNVs, specifically the process of how these CNVs arise in the population, whether they are transmitted through generations, and their loss from the population by stochastic or selective processes. Case-control analyses tell us little about the evolutionary processes affecting this kind of variation. Data from extended pedigrees in Oceanic Palau could fill some of this gap.
METHODS AND MATERIALS
Subjects
Through an ongoing study of schizophrenia in Palau (20–24), with pedigree information for almost 1,000 subjects tracing beyond maternal and paternal grandparents, nearly complete ascertainment of individuals affected with schizophrenia and related psychotic disorders and their first-degree relatives has been achieved. Because individuals in this small isolated population are often if not always related at some level, selecting control subjects present a challenge. We selected individuals as controls if they had no history of psychiatric treatment and were no closer than third-degree relatives of cases. For this study, blood samples from 573 subjects were genotyped including all affected subjects, a sample of relatives of an affected (ROA), and control subjects. There were 208 (36.3%) affected individuals, with 125 having a diagnosis of schizophrenia (60%), 36 schizoaffective (17%), 20 bipolar disorder (10%), and 27 psychotic disorders not otherwise specified (13%); 365 (63.7%) were unaffected by any of these disorders, 191 ROA and 174 controls (Figure S1 in Supplement 1).
Potentially affected individuals, identified by medical records and/or referral by a family member, were interviewed by an experienced clinician (WB, MMW) working with a Palauan mental health professional (JT) using a modified version of the Schedule for Affective Disorders and Schizophrenia-Lifetime Version or SADS-L (25). Subjects were interviewed in their language of preference; if it was Belauan, JT translated. The SADS-L interviews were followed by a detailed review of hospital psychiatric medical records, which document in English inpatient and outpatient visits and thus describe symptoms and treatment over the course of illness, often providing in-depth longitudinal diagnostic data. The SADS-L interview data and medical records were used to reach a DSM-IV based Best Estimate Final Diagnosis (26). Research protocols and procedures were approved by institutional review boards (IRBs) at each of the sites in the US and the Republic of Palau. All subjects provided written informed consent to participate after receiving a full explanation of the study in both English and Palauan.
Copy Number Variation analyses
CNVs were called for 573 Palauan subjects using the genotype image intensities from Affymetrix Genomewide Human SNP Array 5.0 and two different calling algorithms: Birdseye of the Birdsuite package (27) and PennCNV (28). Genotype-based gender was supplied for calling CNVs on X. We evaluated nominal gender based on the level of heterozygosity on chromosome X after excluding SNPs in the pseudo-autosomal region (Figure S2 in Supplement 1). There were 5 (0.87%) discrepancies between the genotype-based gender and the pedigree information based on the assumption that all males would have <1% heterozygosity on the X-linked portion of chromosome X. Where a sample swap was suspected, an attempt to redraw a DNA sample from the subjects was made to resolve the discrepancy. CNVs were called only when covered by 20 or more SNPs or monomorphic markers (henceforth collectively called ‘probes’). Rare CNVs were defined as CNVs called by both Birdsuite and PennCNV that occurred in ≤ 1% of the Palauan chromosomes. Based on the distribution of rare CNVs per sample, we declared a subject to have excessive number of rare CNVs if more than 10 were called from that subject’s DNA; samples with excessive number were excluded (Figure S3 in Supplement 1). Of the 541 remaining samples [267 males (M), 274 females (F)], 197 (36.4%, 128 M: 69 F) were affected, 185 (34.2%, 62 M: 123 F) were ROA, and 159 (29.4%, 77 M: 82 F) were controls. Within this sample, the average count of rare CNVs per sample is 0.68; the average size of a rare CNV is .67 Mb; and an average rare CNV is called using 68 probes.
Risk CNVs were identified using the following criteria: 1) associated with increased risk in this sample; or 2) occur in genes of relevance to schizophrenia and related disorders, typically identified in studies of samples of European ancestry. We consider these risk CNVs to be validated if they are called by both Birdsuite and PennCNV algorithms and if they segregate in families. If a risk CNV is called by at least one algorithm in a subject and the subject is an obligate carrier or a first-degree relative of a subject carrying the CNV, we consider that CNV to be validated also. The association of these CNVs with affection status is assessed by the odds ratio (OR). When there is a zero cell (undefined or infinity OR), we used the Yates correction, which adds 0.5 to all of the cells of a contingency table. Given the relatively small sample size of this study, we do not expect the association of these CNVs with diagnosis to be statistically significant.
We examined the segregation of risk CNVs in families by determining the relationships and haplotype sharing among CNV carriers. Relationships in this Oceanic population are complex and sometimes unknown, although most relationships fifth degree or less can be inferred by analysis of numerous independent genotypes, and most are known from pedigrees. We examined the genetic-based estimates of relationships by using tag SNPs, which were chosen as SNPs with no Mendelian errors, no missingness, and minor allele frequency of > 0.05 in controls. We then sub-selected SNPs that are 25 kb apart, yielding 59,627 SNPs and used Hclust (29, 30) to select 6,860 SNPs that had a pairwise correlation of r ≤ 0.2. Pairwise identity-by-descent (IBD) relationships for all pairs of samples were estimated using maximum likelihood. Genetic relationships were computed as p(IBD2)+ 0.5 p(IBD1) where p(IBD1) and p(IBD2) are the probabilities that two individuals share one and two alleles, respectively, IBD. Allele sharing among subjects was also examined to identify samples of non-Palauan ancestry. None were identified as having a distinct ancestry, as might be predicted by recent European admixture, although two individuals were born in Yap.
To complement estimated relationships, we also evaluated haplotype sharing in the CNV region amongst all CNV carriers (31). Extended haplotype-sharing, teamed with sharing in risk CNV, is clear evidence that the CNV is inherited IBD from a common ancestor. One hundred SNPs around the CNV, 50 SNPs from each side, were phased using Beagle 3.0.1 (32). If fewer than 50 SNPs were available, all available SNPs were used.
Subjects
Through an ongoing study of schizophrenia in Palau (20–24), with pedigree information for almost 1,000 subjects tracing beyond maternal and paternal grandparents, nearly complete ascertainment of individuals affected with schizophrenia and related psychotic disorders and their first-degree relatives has been achieved. Because individuals in this small isolated population are often if not always related at some level, selecting control subjects present a challenge. We selected individuals as controls if they had no history of psychiatric treatment and were no closer than third-degree relatives of cases. For this study, blood samples from 573 subjects were genotyped including all affected subjects, a sample of relatives of an affected (ROA), and control subjects. There were 208 (36.3%) affected individuals, with 125 having a diagnosis of schizophrenia (60%), 36 schizoaffective (17%), 20 bipolar disorder (10%), and 27 psychotic disorders not otherwise specified (13%); 365 (63.7%) were unaffected by any of these disorders, 191 ROA and 174 controls (Figure S1 in Supplement 1).
Potentially affected individuals, identified by medical records and/or referral by a family member, were interviewed by an experienced clinician (WB, MMW) working with a Palauan mental health professional (JT) using a modified version of the Schedule for Affective Disorders and Schizophrenia-Lifetime Version or SADS-L (25). Subjects were interviewed in their language of preference; if it was Belauan, JT translated. The SADS-L interviews were followed by a detailed review of hospital psychiatric medical records, which document in English inpatient and outpatient visits and thus describe symptoms and treatment over the course of illness, often providing in-depth longitudinal diagnostic data. The SADS-L interview data and medical records were used to reach a DSM-IV based Best Estimate Final Diagnosis (26). Research protocols and procedures were approved by institutional review boards (IRBs) at each of the sites in the US and the Republic of Palau. All subjects provided written informed consent to participate after receiving a full explanation of the study in both English and Palauan.
Copy Number Variation analyses
CNVs were called for 573 Palauan subjects using the genotype image intensities from Affymetrix Genomewide Human SNP Array 5.0 and two different calling algorithms: Birdseye of the Birdsuite package (27) and PennCNV (28). Genotype-based gender was supplied for calling CNVs on X. We evaluated nominal gender based on the level of heterozygosity on chromosome X after excluding SNPs in the pseudo-autosomal region (Figure S2 in Supplement 1). There were 5 (0.87%) discrepancies between the genotype-based gender and the pedigree information based on the assumption that all males would have <1% heterozygosity on the X-linked portion of chromosome X. Where a sample swap was suspected, an attempt to redraw a DNA sample from the subjects was made to resolve the discrepancy. CNVs were called only when covered by 20 or more SNPs or monomorphic markers (henceforth collectively called ‘probes’). Rare CNVs were defined as CNVs called by both Birdsuite and PennCNV that occurred in ≤ 1% of the Palauan chromosomes. Based on the distribution of rare CNVs per sample, we declared a subject to have excessive number of rare CNVs if more than 10 were called from that subject’s DNA; samples with excessive number were excluded (Figure S3 in Supplement 1). Of the 541 remaining samples [267 males (M), 274 females (F)], 197 (36.4%, 128 M: 69 F) were affected, 185 (34.2%, 62 M: 123 F) were ROA, and 159 (29.4%, 77 M: 82 F) were controls. Within this sample, the average count of rare CNVs per sample is 0.68; the average size of a rare CNV is .67 Mb; and an average rare CNV is called using 68 probes.
Risk CNVs were identified using the following criteria: 1) associated with increased risk in this sample; or 2) occur in genes of relevance to schizophrenia and related disorders, typically identified in studies of samples of European ancestry. We consider these risk CNVs to be validated if they are called by both Birdsuite and PennCNV algorithms and if they segregate in families. If a risk CNV is called by at least one algorithm in a subject and the subject is an obligate carrier or a first-degree relative of a subject carrying the CNV, we consider that CNV to be validated also. The association of these CNVs with affection status is assessed by the odds ratio (OR). When there is a zero cell (undefined or infinity OR), we used the Yates correction, which adds 0.5 to all of the cells of a contingency table. Given the relatively small sample size of this study, we do not expect the association of these CNVs with diagnosis to be statistically significant.
We examined the segregation of risk CNVs in families by determining the relationships and haplotype sharing among CNV carriers. Relationships in this Oceanic population are complex and sometimes unknown, although most relationships fifth degree or less can be inferred by analysis of numerous independent genotypes, and most are known from pedigrees. We examined the genetic-based estimates of relationships by using tag SNPs, which were chosen as SNPs with no Mendelian errors, no missingness, and minor allele frequency of > 0.05 in controls. We then sub-selected SNPs that are 25 kb apart, yielding 59,627 SNPs and used Hclust (29, 30) to select 6,860 SNPs that had a pairwise correlation of r ≤ 0.2. Pairwise identity-by-descent (IBD) relationships for all pairs of samples were estimated using maximum likelihood. Genetic relationships were computed as p(IBD2)+ 0.5 p(IBD1) where p(IBD1) and p(IBD2) are the probabilities that two individuals share one and two alleles, respectively, IBD. Allele sharing among subjects was also examined to identify samples of non-Palauan ancestry. None were identified as having a distinct ancestry, as might be predicted by recent European admixture, although two individuals were born in Yap.
To complement estimated relationships, we also evaluated haplotype sharing in the CNV region amongst all CNV carriers (31). Extended haplotype-sharing, teamed with sharing in risk CNV, is clear evidence that the CNV is inherited IBD from a common ancestor. One hundred SNPs around the CNV, 50 SNPs from each side, were phased using Beagle 3.0.1 (32). If fewer than 50 SNPs were available, all available SNPs were used.
RESULTS
CNV analysis
We detected a substantial set of CNVs in the Palau sample, some of which confer risk to schizophrenia and related disorders, namely, CNVs in the 1q21.1, 15q11.2, 16p13.2, and Xp21.3 regions (Table 1). A 2.8Mb deletion on chromosome 1q21.1 was observed in one male subject affected with schizophrenia. Deletions of 1q21.1 spanning 1.35–2.19Mb have been previously reported in schizophrenia (2, 3, 5), autism, mental retardation, and other behavioral abnormalities (OMIM #612474).
Table 1
List of CNVs in Palau that appear to confer risk to schizophrenia and related psychotic disorders
Chr band | Gene | Start | End | Dup / Del | Intronic/ Exonic | Affected: ROA: Controls | OR (95% CI) | ||||
---|---|---|---|---|---|---|---|---|---|---|---|
Total | Females | Males | Total | Females | Males | ||||||
1q21.13 | Several Genes4 | 143,560,909 | 146,395,590 | Del | Exonic | 1:0:0 | 0:0:0 | 1:0:0 | 2.5 (0.2–30.3) | --- | 1.8 (0.15–22.4) |
15q11.2 | TUBGCP5, CYFIP1, NIPA2, NIPA1 | 19,537,067 | 20,852,161 | Del | Exonic | 2:3:0 | 1:2:0 | 1:1:0 | 4.1 (0.36–45.6) | 3.6 (0.29–44.4) | 1.8 (0.15–22.4) |
16p13.2 | A2BP1 | 6,581,119 | 6,645,572 | Dup | Intronic | 6:6:1 | 0:4:1 | 6:2:0 | 5.0 (0.8–31.6) | 0.4 (0.03–4.9) | 8.2 (0.8–84.4) |
Xp21.3 | IL1RAPL1 | 29,163,088 | 29,455,307 | Dup | Exonic | 4:6:2 | 2:5:1 | 2:1:1 | 1.6 (0.3–7.7) | 2.4 (0.3–18.8) | 1.2 (0.15–9.3) |
47, XXY | Kleinfelter Syndrome | 230,897 | 154,582,819 | Dup | Both | 2:0:1 | 0:0:0 | 2:0:1 | 1.2 (0.2–9.2) | --- | 1.2 (0.2–9.2) |
A deletion in the 15q11.2 (2, 4, 5, 33) region, between breakpoints 1 and 2 of the Prader-Willi/Angelman syndrome region (34) and covering the genes TUBGCP5, CYFIP1, NIPA2, and NIPA1, occurred in two full-sibs (one affected and one unaffected), their paternal affected half-sib, and their unaffected father, as well as an unrelated unaffected female (Tables S1 & S2 in Supplement 1). All relationships were confirmed using genetic-based estimates of relationships. The deletion seems to affect risk (OR=4.1; 95% CI, 0.36–45.6) and is clearly incompletely penetrant. As expected, all deletion carriers related through their carrier father share an extended haplotype identical-by-descent (Table S3 in Supplement 1). However, the unrelated female with the 15q11.2 deletion does not carry that haplotype, consistent with this CNV arising at least twice in this population.
An intronic duplication at 16p13.2 within A2BP1 (32–40) appears to convey risk in males (OR=8.2, 95% CI, 0.8–84.4) but not females (OR=0.4, 95% CI, 0.03–4.9). There were 6 males carrying the duplication, all affected, whereas none of the females carrying the duplication were affected (Table S4 in Supplement 1). Pedigree relationships and genetic-based estimates of relationships show that this CNV is inherited (Figure 1, Table 2). All females with the CNV are either mothers or full siblings to an affected individual, except for one subject who is an adolescent and is not yet through the age of risk for developing schizophrenia and related disorders. Haplotype sharing also demonstrates that this CNV is inherited IBD (Table 3). To evaluate how the IBD-sharing translated into linkage, we performed an affected-only linkage analysis with the set of carriers and their close relatives. We assumed the CNV was rare and inherited IBD. Some individuals did not have clear pedigree relationships, but were genetically inferred to be related. We constructed a simple pedigree consistent with the genetically-inferred relationships that also fit with the known pedigrees (Figure S4 in Supplement 1). We then computed a LOD score using this pedigree and an affected-only model in which the probability of being affected was extremely small (0.0001) and the probability of being affected given the subject carries the CNV is 0.49. For this model and pedigree, a LOD score of 3.5 results.

Nominal pedigree relationships between subjects carrying a duplication in A2BP1. Subjects in red are affected; triangle symbol represents unknown gender. Subjects with OC are obligate carriers who were not genotyped to confirm their carrier status. The pedigree section linking 2034, 22554, 22532, and 60038 was inferred from genetic-based estimates of the relationships. Genetic-based estimates uncover additional distant relationships between CNV carriers (Table 2).
Table 2
Estimated relationship matrix for subjects carrying a duplication in A2BP1
2058 | 17615 | 19800 | 20248 | 22415 | 22554 | 20244 | 22352 | 22563 | 60038 | 22353 | 2034 | 22532 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2058 | 1 | 0.015 | 0.01 | 0.006 | 0.007 | 0.092 | 0.005 | 0.018 | 0.005 | 0.5 | 0.003 | 0.038 | 0.118 |
17615 | 1 | 0 | 0 | 0 | 0.004 | 0 | 0.509 | 0 | 0.033 | 0 | 0.004 | 0.021 | |
19800 | 1 | 0.144 | 0 | 0.003 | 0.498 | 0 | 0.008 | 0.011 | 0.145 | 0 | 0 | ||
20248 | 1 | 0 | 0 | 0.228 | 0.0002 | 0 | 0.006 | 0.417 | 0 | 0 | |||
22415 | 1 | 0.001 | 0 | 0.005 | 0.506 | 0 | 0 | 0 | 0.014 | ||||
22554 | 1 | 0.003 | 0 | 0.002 | 0.178 | 0.001 | 0.24 | 0.476 | |||||
20244 | 1 | 0 | 0 | 0.005 | 0.31 | 0 | 0 | ||||||
22352 | 1 | 0.008 | 0 | 0 | 0.004 | 0 | |||||||
22563 | 1 | 0 | 0 | 0.001 | 0.0001 | ||||||||
60038 | 1 | 0.002 | 0.075 | 0.279 | |||||||||
22353 | 1 | 0 | 0.010 | ||||||||||
2034 | 1 | 0.262 | |||||||||||
22532 | 1 |
Table 3
Haplotype sharing surrounding the A2BP1 CNV
Subject | Haplotype |
---|---|
22353 | *****//221212121212212\\***** |
20248 | *****//221212121212212\\***** |
20244 | *****//221212121212212\\***** |
19800 | *****//221212121212212\\***** |
22532 | *****//221212121212212\\***** |
22554 | *****//222212121212212\\***** |
22352 | *****//222212121212212\\***** |
17615 | *++**//221212121212212\\***** |
60038 | *++**//221212121212212\\***** |
2058 | *++**//221212121212212\\***** |
2034 | *++**//222212121212212\\***** |
22563 | ^^^^^//221212121212212\\+++++ |
22415 | ^^^^^//221212121212212\\+++++ |
Haplotypes are ordered in terms of similarity; CNV region is the region between // and \\; Symbols represent similarity on a stretch of 10 SNPs except in the CNV region where the actual haplotype is presented; Variation in the CNV region is presented in red and bold.
At Xp21.3 (41–47), we find a partial duplication of the gene encoding interleukin 1 receptor accessory protein-like 1, IL1RAPL1 (OR=1.6, 95% CI, 0.3–7.7). While sex-specific effects of CNVs on chromosome X is the expectation, it is curious that females rather than males with the IL1RAPL1 duplication appear to be at higher risk (Table 1). Notably, however, 5 out of the 6 unaffected female carriers were first-degree relative of an affected. The IL1RAPL1 CNV is inherited based on identified and genetic relationships (Figure S5 & Tables S5 & S6 in Supplement 1) and IBD on the basis of haplotype sharing (Table S7 in Supplement 1). Aside from parent-offspring pairs, most other relationships are distant. Intriguingly on the proximal side of the duplication we find clusters of SNPs with evidence that they are heterozygous in males, suggesting that after the duplication event there was a recombination that disrupted a portion of the duplication. This region (between rs5943613 and rs4893560) has been identified as a recombination hotspot (http://mathgen.stats.ox.ac.uk/Recombination.html).
Three male subjects, 2 affected and 1 unaffected, were identified to have Klinefelter’s syndrome (KS; 47, XXY). None of these subjects were related and none of their parents were genotyped. However, they are most likely to be de novo events as KS is associated with male hypogonadism and infertility (48). Notably, XXY was only called by PennCNV, which is not completely surprising because most algorithms have diminished accuracy of calling CNVs on chromosome X. However, XXY was confirmed based on the rate of heterozygosity on the X-linked SNPs, which was similar to that of females and by previous unreported results for genotyping of Short Tandem Repeats (STRs) by the Center for Inherited Disease Research (CIDR) (20–23). CIDR results also reveal Y markers consistent with an intact Y chromosome, thereby confirming KS.
We also detected several CNVs for which the risk ratio of affected to unaffected individuals is not large but they involved genes in neuronal-developmental pathways that have been previously reported to affect the risk for schizophrenia or related disorders, including autism. We report these CNVs (Table S8 in Supplement 1) and all other rare CNVS (<1%) called by Birdsuite and PennCNV (Table S9 in Supplement 1). GRM7 is among the CNVs that have been previously reported in schizophrenia (49). We find a large duplication in one affected subject on 17p12, a region that falls between Potocki-Lupski syndrome (PLS, OMIM#610883) and Charcot-Marie-Tooth Type 1A (CMT1A, OMIM#118220). We also find another large duplication on 19p13.3 in one subject affected with schizophrenia and another with mild mental retardation, a duplication that was previously reported in schizophrenia (50).
CNV analysis
We detected a substantial set of CNVs in the Palau sample, some of which confer risk to schizophrenia and related disorders, namely, CNVs in the 1q21.1, 15q11.2, 16p13.2, and Xp21.3 regions (Table 1). A 2.8Mb deletion on chromosome 1q21.1 was observed in one male subject affected with schizophrenia. Deletions of 1q21.1 spanning 1.35–2.19Mb have been previously reported in schizophrenia (2, 3, 5), autism, mental retardation, and other behavioral abnormalities (OMIM #612474).
Table 1
List of CNVs in Palau that appear to confer risk to schizophrenia and related psychotic disorders
Chr band | Gene | Start | End | Dup / Del | Intronic/ Exonic | Affected: ROA: Controls | OR (95% CI) | ||||
---|---|---|---|---|---|---|---|---|---|---|---|
Total | Females | Males | Total | Females | Males | ||||||
1q21.13 | Several Genes4 | 143,560,909 | 146,395,590 | Del | Exonic | 1:0:0 | 0:0:0 | 1:0:0 | 2.5 (0.2–30.3) | --- | 1.8 (0.15–22.4) |
15q11.2 | TUBGCP5, CYFIP1, NIPA2, NIPA1 | 19,537,067 | 20,852,161 | Del | Exonic | 2:3:0 | 1:2:0 | 1:1:0 | 4.1 (0.36–45.6) | 3.6 (0.29–44.4) | 1.8 (0.15–22.4) |
16p13.2 | A2BP1 | 6,581,119 | 6,645,572 | Dup | Intronic | 6:6:1 | 0:4:1 | 6:2:0 | 5.0 (0.8–31.6) | 0.4 (0.03–4.9) | 8.2 (0.8–84.4) |
Xp21.3 | IL1RAPL1 | 29,163,088 | 29,455,307 | Dup | Exonic | 4:6:2 | 2:5:1 | 2:1:1 | 1.6 (0.3–7.7) | 2.4 (0.3–18.8) | 1.2 (0.15–9.3) |
47, XXY | Kleinfelter Syndrome | 230,897 | 154,582,819 | Dup | Both | 2:0:1 | 0:0:0 | 2:0:1 | 1.2 (0.2–9.2) | --- | 1.2 (0.2–9.2) |
A deletion in the 15q11.2 (2, 4, 5, 33) region, between breakpoints 1 and 2 of the Prader-Willi/Angelman syndrome region (34) and covering the genes TUBGCP5, CYFIP1, NIPA2, and NIPA1, occurred in two full-sibs (one affected and one unaffected), their paternal affected half-sib, and their unaffected father, as well as an unrelated unaffected female (Tables S1 & S2 in Supplement 1). All relationships were confirmed using genetic-based estimates of relationships. The deletion seems to affect risk (OR=4.1; 95% CI, 0.36–45.6) and is clearly incompletely penetrant. As expected, all deletion carriers related through their carrier father share an extended haplotype identical-by-descent (Table S3 in Supplement 1). However, the unrelated female with the 15q11.2 deletion does not carry that haplotype, consistent with this CNV arising at least twice in this population.
An intronic duplication at 16p13.2 within A2BP1 (32–40) appears to convey risk in males (OR=8.2, 95% CI, 0.8–84.4) but not females (OR=0.4, 95% CI, 0.03–4.9). There were 6 males carrying the duplication, all affected, whereas none of the females carrying the duplication were affected (Table S4 in Supplement 1). Pedigree relationships and genetic-based estimates of relationships show that this CNV is inherited (Figure 1, Table 2). All females with the CNV are either mothers or full siblings to an affected individual, except for one subject who is an adolescent and is not yet through the age of risk for developing schizophrenia and related disorders. Haplotype sharing also demonstrates that this CNV is inherited IBD (Table 3). To evaluate how the IBD-sharing translated into linkage, we performed an affected-only linkage analysis with the set of carriers and their close relatives. We assumed the CNV was rare and inherited IBD. Some individuals did not have clear pedigree relationships, but were genetically inferred to be related. We constructed a simple pedigree consistent with the genetically-inferred relationships that also fit with the known pedigrees (Figure S4 in Supplement 1). We then computed a LOD score using this pedigree and an affected-only model in which the probability of being affected was extremely small (0.0001) and the probability of being affected given the subject carries the CNV is 0.49. For this model and pedigree, a LOD score of 3.5 results.

Nominal pedigree relationships between subjects carrying a duplication in A2BP1. Subjects in red are affected; triangle symbol represents unknown gender. Subjects with OC are obligate carriers who were not genotyped to confirm their carrier status. The pedigree section linking 2034, 22554, 22532, and 60038 was inferred from genetic-based estimates of the relationships. Genetic-based estimates uncover additional distant relationships between CNV carriers (Table 2).
Table 2
Estimated relationship matrix for subjects carrying a duplication in A2BP1
2058 | 17615 | 19800 | 20248 | 22415 | 22554 | 20244 | 22352 | 22563 | 60038 | 22353 | 2034 | 22532 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2058 | 1 | 0.015 | 0.01 | 0.006 | 0.007 | 0.092 | 0.005 | 0.018 | 0.005 | 0.5 | 0.003 | 0.038 | 0.118 |
17615 | 1 | 0 | 0 | 0 | 0.004 | 0 | 0.509 | 0 | 0.033 | 0 | 0.004 | 0.021 | |
19800 | 1 | 0.144 | 0 | 0.003 | 0.498 | 0 | 0.008 | 0.011 | 0.145 | 0 | 0 | ||
20248 | 1 | 0 | 0 | 0.228 | 0.0002 | 0 | 0.006 | 0.417 | 0 | 0 | |||
22415 | 1 | 0.001 | 0 | 0.005 | 0.506 | 0 | 0 | 0 | 0.014 | ||||
22554 | 1 | 0.003 | 0 | 0.002 | 0.178 | 0.001 | 0.24 | 0.476 | |||||
20244 | 1 | 0 | 0 | 0.005 | 0.31 | 0 | 0 | ||||||
22352 | 1 | 0.008 | 0 | 0 | 0.004 | 0 | |||||||
22563 | 1 | 0 | 0 | 0.001 | 0.0001 | ||||||||
60038 | 1 | 0.002 | 0.075 | 0.279 | |||||||||
22353 | 1 | 0 | 0.010 | ||||||||||
2034 | 1 | 0.262 | |||||||||||
22532 | 1 |
Table 3
Haplotype sharing surrounding the A2BP1 CNV
Subject | Haplotype |
---|---|
22353 | *****//221212121212212\\***** |
20248 | *****//221212121212212\\***** |
20244 | *****//221212121212212\\***** |
19800 | *****//221212121212212\\***** |
22532 | *****//221212121212212\\***** |
22554 | *****//222212121212212\\***** |
22352 | *****//222212121212212\\***** |
17615 | *++**//221212121212212\\***** |
60038 | *++**//221212121212212\\***** |
2058 | *++**//221212121212212\\***** |
2034 | *++**//222212121212212\\***** |
22563 | ^^^^^//221212121212212\\+++++ |
22415 | ^^^^^//221212121212212\\+++++ |
Haplotypes are ordered in terms of similarity; CNV region is the region between // and \\; Symbols represent similarity on a stretch of 10 SNPs except in the CNV region where the actual haplotype is presented; Variation in the CNV region is presented in red and bold.
At Xp21.3 (41–47), we find a partial duplication of the gene encoding interleukin 1 receptor accessory protein-like 1, IL1RAPL1 (OR=1.6, 95% CI, 0.3–7.7). While sex-specific effects of CNVs on chromosome X is the expectation, it is curious that females rather than males with the IL1RAPL1 duplication appear to be at higher risk (Table 1). Notably, however, 5 out of the 6 unaffected female carriers were first-degree relative of an affected. The IL1RAPL1 CNV is inherited based on identified and genetic relationships (Figure S5 & Tables S5 & S6 in Supplement 1) and IBD on the basis of haplotype sharing (Table S7 in Supplement 1). Aside from parent-offspring pairs, most other relationships are distant. Intriguingly on the proximal side of the duplication we find clusters of SNPs with evidence that they are heterozygous in males, suggesting that after the duplication event there was a recombination that disrupted a portion of the duplication. This region (between rs5943613 and rs4893560) has been identified as a recombination hotspot (http://mathgen.stats.ox.ac.uk/Recombination.html).
Three male subjects, 2 affected and 1 unaffected, were identified to have Klinefelter’s syndrome (KS; 47, XXY). None of these subjects were related and none of their parents were genotyped. However, they are most likely to be de novo events as KS is associated with male hypogonadism and infertility (48). Notably, XXY was only called by PennCNV, which is not completely surprising because most algorithms have diminished accuracy of calling CNVs on chromosome X. However, XXY was confirmed based on the rate of heterozygosity on the X-linked SNPs, which was similar to that of females and by previous unreported results for genotyping of Short Tandem Repeats (STRs) by the Center for Inherited Disease Research (CIDR) (20–23). CIDR results also reveal Y markers consistent with an intact Y chromosome, thereby confirming KS.
We also detected several CNVs for which the risk ratio of affected to unaffected individuals is not large but they involved genes in neuronal-developmental pathways that have been previously reported to affect the risk for schizophrenia or related disorders, including autism. We report these CNVs (Table S8 in Supplement 1) and all other rare CNVS (<1%) called by Birdsuite and PennCNV (Table S9 in Supplement 1). GRM7 is among the CNVs that have been previously reported in schizophrenia (49). We find a large duplication in one affected subject on 17p12, a region that falls between Potocki-Lupski syndrome (PLS, OMIM#610883) and Charcot-Marie-Tooth Type 1A (CMT1A, OMIM#118220). We also find another large duplication on 19p13.3 in one subject affected with schizophrenia and another with mild mental retardation, a duplication that was previously reported in schizophrenia (50).
DISCUSSION
In this population, we identify several CNVs that are associated with increased risk of schizophrenia and related psychotic disorders, similar to those detected in populations of European ancestry. We report CNVs in the 1q21.1 and 15q11.2, which are confirmed regions of risk for schizophrenia (2–5). We also add to the existing evidence of the importance of CNVs in 16p13.2 and Xp21.3 involving A2BP1 and IL1RAPL1 genes, respectively, in the risk for schizophrenia and related disorders. Most of these CNVs (except for 1q21.1) are inherited and have incomplete penetrance, the latter consistent with studies reporting these deletions and duplications in unaffected individuals (51, 52).
This study has limitations. Our sample size of affected and unaffected subjects lacks power to detect statistically significant associations between CNVs and case-control status. The complex relationships amongst individuals in this Oceanic population allows the evaluation of the transmission of CNVs, which a strength of this study; however, it poses a limitation in the choice of controls. We include controls who are no less than third degree relatives to cases because relationships among case and control subjects are inevitable. Indeed, relationships amongst controls are also certain. Another limitation is that controls and relatives of an unaffected were not administered the SADS-Lifetime. While the absence of psychiatric disorders has been confirmed through medical records without a full psychiatric interview, it does not preclude milder and subsyndromal diagnoses for which medical care may not have been sought.
A duplication affecting A2BP1 is associated with an 8-fold increased risk in males but not in females. A2BP1 codes for a brain-expressed RNA binding or splicing factor involved in neurological function. A2BP1 binds with ataxin-2 (53), which is the product of the gene causing spinocerebellar ataxia 2, a neurological disease that can express psychosis as one of its symptoms (54). Reduced mRNA expression in lymphocytes was found in an autistic patient with a deletion of the first exon of A2BP1 (35). Duplications and deletions disrupting A2BP1 are reported in schizophrenia (36), attention deficit hyperactivity disorder (37), epilepsy, and mental retardation (38). A2BP1 is also reported as a candidate gene for bipolar disorder (39, 40). This duplication has been present in Palau for many generations because all carriers show evidence of haplotype-sharing and genetic relationship to other carriers, with some of the related carriers separated by many generations.
The sex difference we found for duplications in A2BP1 is intriguing given the well-known sex differences in the age at onset and clinical features of schizophrenia (55, 56). Also notable in this regard are potential regulatory pathways whereby sex hormones could regulate A2BP1. A2BP1 has a predicted binding site for the microRNA has-mir-21 (57) and other evidence suggests has-mir-21 is regulated by androgen and estrogen receptors (58, 59). This raises the possibility that A2BP1 signaling may mediate sex effects in the expression of psychosis. However, given our small sample size, further studies are needed to confirm the sex difference in A2BP1 duplications.
IL1RAPL1 encodes for a protein that is a member of the Interleukin 1 receptor family and is highly expressed in postnatal brain structures involved in hippocampal memory. IL1RAPL1 plays a role in the regulation of neurite outgrowth and exocytosis and in the down-regulation of calcium channels. The IL1RAPL1 consists of 10 coding exons. A loss of function results in a truncated protein that is unable to control neurite out-growth in hippocampal neurons (41). Males with deletions and mutations of IL1RAPL1 exhibit a range of neurological disorders such as X-linked mental retardation (MR), autism, and cognitive impairments (OMIM #300206; #300143) (41–47) whereas females are not necessarily affected. Duplicated genes on chromosome X often show sexually dimorphic expression due to skewed X inactivation; in females, a duplicated or disrupted gene is often silenced by the X inactivation mechanism. Thus, females with an X-linked abnormality can develop normally. Contrary to expectation for the Palauan sample, females with IL1RAPL1 duplications are affected more than would be expected by chance. Selective or incomplete inactivation of the X chromosome could explain the diversity of the phenotype in females, consistent with several studies reporting some X-linked genes escaping X-inactivation in the Xp21.3-p22.1 boundary (60, 61). Because the CNV duplicates two exons of IL1RAPL1, it is likely that it alters the conformation of the resulting protein, thereby functioning as a mutation. However, it is also possible that the duplicated exons are not transcribed or translated and therefore does not alter gene function.
We identified three subjects with Klinfelter’s Syndrome in this sample. This represents an increased prevalence of KS in Palau (3 per 268 males) as compared to the expected prevalence of 1 per 500 males or higher (62). The risk for KS is known to increase with increased maternal age in maternally-derived cases and recent studies show increased risk with increased paternal age (63–65). Subjects with KS present with a variety of clinical features including developmental delays and behavioral problems (66). An increased prevalence of psychotic disorders and increased levels of schizophrenia-spectrum pathology in KS patients have been previously reported (67, 68).
Deletion of a 300Kb region in 15q11.2, between breakpoints 1 and 2 of the Prader-Willi/Angelman syndrome region (33) has previously been reported as a risk region for schizophrenia in samples of European ancestry (2, 4, 5). The deletion is found in one nuclear family and an unrelated individual and therefore has arisen at least twice in Palau. Yet the deletion appears rare because it is not found in our study’s controls. Like results from populations of European ancestry, our results suggest deletions convey modest risk for schizophrenia.
We identified one affected male with a deletion on 1q21.1. Deletions on 1q21.1 have been consistently implicated for schizophrenia (2–5), as well as developmental delay, intellectual disability, behavioral abnormalities, autism, nonspecific dysmorphic features, and congenital abnormalities (OMIM #612474). Because his parents were deceased, inherited or a de novo status could not be established. Still, because this CNV is not found in any other individual, is not present in his affected sibling, and because of the high penetrance of this deletion, it is very likely a de novo mutation, consistent with observations in populations of European ancestry (2–5).
This study reveals several CNVs that confer risk to schizophrenia and related psychotic disorders in Palau. These CNVs have been previously reported in samples of European ancestry. These findings suggest that rare de novo and inherited CNVs account for a proportion of the risk to these disorders regardless of ancestry. However, most of these CNVs have incomplete penetrance. A few of these CNVs show evidence that they have existed in Palau for many generations, while others show no such evidence. It is interesting to note that the CNVs showing some evidence of extended longevity in the population are also those that show (A2BP1) or should show (IL1RAPL1) sex-dependent risk. In theory this would expose the variant to natural selection, as a result of reduced reproductive success typically associated with schizophrenia, in a sex-differential manner.
Supplementary Material
01
01
ACKNOWLEDGMENTS
This study was supported by NIH grants MH080375, MH080373, MH080299, and MH077930. We are grateful to the people of Palau for their participation in this study. For STR and Y marker results, genotyping services were provided by the Center for Inherited Disease Research (CIDR). CIDR is fully funded through a federal contract from the National Institutes of Health to The Johns Hopkins University, contract number HHSN268200782096C.
Abstract
Background
We report on copy number variants (CNVs) found in Palauan subjects ascertained for schizophrenia and related psychotic disorders in extended pedigrees in Palau. We compare CNVs found in this Oceanic population to those seen in other samples, typically of European ancestry. Assessing CNVs in Palauan extended pedigrees yields insight into the evolution of risk CNVs, such as how they arise, are transmitted, and are lost from populations by stochastic or selective processes, none of which is easily measured from case-control samples.
Methods
DNA samples from 197 subjects affected with schizophrenia and related psychotic disorders, 185 of their relatives, and 159 controls were successfully characterized for CNVs using Affymetrix Genomewide Human SNP Array 5.0.
Results
CNVs thought to be associated with risk for schizophrenia and related disorders also occur in affected individuals in Palau, specifically 15q11.2 and 1q21.1 deletions, partial duplication of IL1RAPL1 (Xp21.3), and chromosome X duplications (Klinefleter’s syndrome). Partial duplication within A2BP1 appears to convey an 8-fold increased risk in males (95% CI, 0.8–84.4) but not females (OR=0.4, 95% CI, 0.03–4.9). Affected-only linkage analysis using this variant yields a LOD score of 3.5.
Conclusions
This study reveals CNVs that confer risk to schizophrenia and related psychotic disorders in Palau, most of which have been previously observed in samples of European ancestry. Only a few of these CNVs show evidence that they have existed for many generations, consistent with risk variants diminishing reproductive success.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
FINANCIAL DISCLOSURE
In the past year, Dr. Faraone received consulting fees and was on Advisory Boards for Shire Development and received research support from Shire and the National Institutes of Health (NIH). In previous years, he received consulting fees or was on Advisory Boards or participated in continuing medical education programs sponsored by: Shire, McNeil, Janssen, Novartis, Pfizer and Eli Lilly. In previous years he received research support from Eli Lilly, Shire, Pfizer and the NIH. Dr. Faraone receives royalties from a book published by Guilford Press: Straight Talk about Your Child’s Mental Health.
All other authors report no biomedical financial interests or potential conflicts of interest.
REFERENCES
References
- 1. Basset AS, Marshall CR, Lionel AC, Chow EW, Scherer SWCopy number variations and risk for schizophrenia in 22q11.2 deletion syndrome. Hum Mol Genet. 2008;17:4045–4053.[Google Scholar]
- 2. Stefansson H, Rujescu D, Cichon S, Pietilainen OP, Ingason A, Steinberg S, et al Large recurrent microdeletions associated with schizophrenia. Nature. 2008;455:232–236.[Google Scholar]
- 3. International schizophrenia consortiumRare chromosomal deletions and duplications increase risk of schizophrenia. Nature. 2008;455:237–241.[Google Scholar]
- 4. Kirov G, Grozeva D, Norton N, Ivanov D, Mantripragada KK, Holmans P, et al Support for the involvement of a large copy number variants in the pathogenesis of schizophrenia. Hum Mol Genet. 2009;18:1497–1503.[Google Scholar]
- 5. Need AC, Ge D, Weale ME, Maia J, Feng S, Heinzen EL, et al A genome-wide investigation of SNPs and CNVs in schizophrenia. PLoS Genet. 2009;5(2):e1000373.[Google Scholar]
- 6. Weiss LA, Shen Y, Korn JM, Arking DE, Miller DT, Fossdal R, et al Association between microdeletion and microduplication at 16p11.2 and autism. N Engl J Med. 2008;358:667–675.[PubMed][Google Scholar]
- 7. Mulle JG, Dodd AF, McGrath JA, Wolyniec PS, Mitchell AA, Shetty AC, et al Microdeletions of 3q29 confer high risk for schizophrenia. Am J Hum Genet. 2010;87(2):229–236.[Google Scholar]
- 8. Quintero-Rivera F, Sharifi-Hannauer P, Martinez-Agosto JAAutistic and psychiatric findings associated with the 3q29 microdeletion syndrome: case report and review. Am J Med Genet A. 2010;152A(10):2459–2467.[PubMed][Google Scholar]
- 9. Levinson DF, Duan J, Oh S, Wang K, Sanders AR, Shi J, et al Copy number variants in schizophrenia: confirmation of five previous findings and new evidence for 3q29 microdeletions and VIPR2 duplications. Am J Psychiatry. 2011;168:302–316.[Google Scholar]
- 10. Kumar RA, KaraMohamed S, Sudi J, Conrad DF, Brune C, Badner JA, et al Recurrent 16p11.2 microdeletions in autism. Hum Mol Genet. 2008;17:628–638.[PubMed][Google Scholar]
- 11. McCarthy SE, Makarov V, Kirov G, Addington AM, McClellan J, Yoon S, et al Microduplications of 16p11.2 are associated with schizophrenia. Nature Genet. 2009;41(11):1223–1227.[Google Scholar]
- 12. Shinawi M, Liu P, Kang SH, Shen J, Belmont JW, Scott DA, et al Recurrent reciprocal 16p11.2 rearrangements associated with global developmental delay, behavioural problems, dysmorphism, epilepsy, and abnormal head size. J Med Genet. 2010;47(5):332–341.[Google Scholar]
- 13. Moreno-De-Luca D, SGENE Consortium. Mulle JG, Kaminsky EB, Sanders SJ, et al. Simons Simplex Collection Genetics Consortium. Deletion 17q12 is a recurrent copy number variant that confers high risk of autism and schizophrenia. Am J Hum Genet. 2010;87(5):618–630.
- 14. Magri C, Sacchetti E, Traversa M, Valsecchi P, Gardella R, Bonvicini C, et al New copy number variations in schizophrenia. PLoS One. 2010;5(10):e13422.[Google Scholar]
- 15. Coon H, Myles-Worsley M, Tiobech J, Hoff M, Rosenthal J, Bennett P, et al Evidence for a chromosome 2p13-14 schizophrenia susceptibility locus in families from Palau, Micronesia. Mol Psychiatry. 1998;3:521–527.[PubMed][Google Scholar]
- 16. Kirov GThe role of copy number variation in schizophrenia. Expert Rev Neurother. 2010;10:25–32.[PubMed][Google Scholar]
- 17. Ingason A, Rujescu D, Cichon S, Sigurdsson E, Sigmundsson T, Pietilainen OP, et al Copy number variations of chromosome 16p13.1 region associated with schizophrenia. Mol Psychiatry. 2011;16(1):17–25.[Google Scholar]
- 18. Williams NM, Zaharieva I, Martin A, Langley K, Mantripragada K, Fossdal R, et al Rare chromosomal deletions and duplications in attention-deficit hyperactivity disorder: a genome-wide analysis. Lancet. 2010;376(9750):1401–1408.[Google Scholar]
- 19. Sanders SJ, Ercan-Sencicek AG, Hus V, Luo R, Murtha MT, Moreno-De-Luca D, et al Multiple Recurrent De Novo CNVs, Including Duplications of the 7q11.23 Williams Syndrome Region, Are Strongly Associated with Autism. Neuron. 2011;70(5):863–885.[Google Scholar]
- 20. Devlin B, Roeder K, Otto C, Tiobech S, Byerley WGenome-wide distribution of linkage disequilibrium in the population of Palau and its implications for gene flow in Remote Oceania. Hum Genet. 2001;108(6):521–528.[PubMed][Google Scholar]
- 21. Devlin B, Bacanu SA, Roeder K, Reimherr F, Wender P, Galke B, et al Genome-wide multipoint linkage analyses of multiplex schizophrenia pedigrees from the oceanic nation of Palau. Mol Psychiatry. 2002;7(7):689–694.[PubMed][Google Scholar]
- 22. Klei L, Bacanu SA, Myles-Worsley M, Galke B, Xie W, Tiobech J, et al Linkage analysis of a completely ascertained sample of familial schizophrenics and bipolars from Palau, Micronesia. Hum Genet. 2005;117(4):349–356.[PubMed][Google Scholar]
- 23. Devlin B, Klei L, Myles-Worsley M, Tiobech J, Otto C, Byerley W, Roeder KGenetic liability to schizophrenia in Oceanic Palau: a search in the affected and maternal generation. Hum Genet. 2007;121(6):675–684.[PubMed][Google Scholar]
- 24. Myles-Worsley M, Tiobech J, Blailes F, Middleton FA, Vinogradov S, Byerley W, et al Familial transmission of schizophrenia in Palau: A 20-year genetic epidemiological study in three generations. Am J Med Genet B Neuropsychiatr Genet. 2011;156B(3):247–254.[PubMed][Google Scholar]
- 25. Endicott J, Spitzer RLA diagnostic interview: the schedule for affective disorders and schizophrenia. Arch Gen Psychiatry. 1978;35(7):837–844.[PubMed][Google Scholar]
- 26. Spitzer RL, Endicott J, Robins EResearch diagnostic criteria: rationale and reliability. Arch Gen Psychiatry. 1978;35(6):773–782.[PubMed][Google Scholar]
- 27. Korn JM, Kuruvilla FG, McCarroll SA, Wysoker A, Nemesh J, Cawley S, et al Integrated genotype calling and association analysis of SNPs, common copy number polymorphisms and rare CNVs. Nature Genet. 2008;40:1253–1260.[Google Scholar]
- 28. Wang K, Li M, Hadley D, Liu R, Glessner J, Grant SF, et al PennCNV: an integrated hidden Markov model designed for high-resolution copy number variation detection in whole-genome SNP genotyping data. Genome Res. 2007;17:1665–1674.[Google Scholar]
- 29. Rinaldo A, Bacanu S, Devlin B, Sonpar V, Wasserman L, Roeder KCharacterization of multilocus linkage disequilibrium. Genet Epidemiol. 2005;28:193–206.[PubMed][Google Scholar]
- 30. Roeder K, Bacanu S, Sonpar V, Zhang X, Devlin BAnalysis of single locus tests to detect disease/gene associations. Genet Epidemiol. 2005;28:207–219.[PubMed][Google Scholar]
- 31. Gusev A, Lowe JK, Stoffel M, Daly MJ, Altshuler D, Breslow JL, et al Whole population, genome-wide mapping of hidden relatedness. Genome Res. 2009;19(2):318–326.[Google Scholar]
- 32. Browning SR, Browning BLRapid and accurate haplotype phasing and missing data inference for whole genome association studies using localized haplotype clustering. Am J Hum Genet. 2007;81:1084–1097.[Google Scholar]
- 33. Vassos E, Collier DA, Holden S, Patch C, Rujescu D, St Clair D, Lewis CMPenetrance for copy number variants associated with schizophrenia. Hum Mol Genet. 2010;19(17):3477–3481.[PubMed][Google Scholar]
- 34. Butler MG, Fischer W, Kibiryeva N, Bittel DCArray comparative genomic hybridization (aCGH) analysis in Prader-Willi syndrome. Am J Hum Genet. 2008;146:854–860.[Google Scholar]
- 35. Martin CL, Duvall JA, Ilkin Y, Simons JS, Arreaza MG, Wilkes K, et al Cytogenetic and molecular characterization of A2BP1/FOX1 as a candidate gene for autism. Am J Med Genet B Neuropsychiatr Genet. 2007;144B(7):869–876.[PubMed][Google Scholar]
- 36. Xu B, Roos JL, Levy S, van Rensburg EJ, Gogos JA, Karayiorgou MStrong association of de novo copy number mutations with sporadic schizophrenia. Nature Genet. 2008;40:880–885.[PubMed][Google Scholar]
- 37. Elia J, Gai X, Xie HM, Perin JC, Geiger E, Glessner JT, et al Rare structural variants found in attention-deficit hyperactivity disorder are preferentially associated with neurodevelopmental genes. Mol Psychiatry. 2010;15(6):637–646.[Google Scholar]
- 38. Bhalla K, Phillips HA, Crawford J, McKenzie OL, Mulley JC, Eyre H, et al The de novo chromosome 16 translocations of two patients with abnormal phenotypes (mental retardation and epilepsy) disrupt the A2BP1 gene. J Hum Genet. 2004;49(6):308–311.[PubMed][Google Scholar]
- 39. Hamshere ML, Green EK, Jones IR, Jones L, Moskvina V, Kirov G, et al Genetic utility of broadly defined bipolar schizoaffective disorder as a diagnostic concept. Br J Psychiatry. 2009;195(1):23–29.[Google Scholar]
- 40. Le-Niculescu H, Patel SD, Bhat M, Kuczenski R, Faraone SV, Tsuang MT, et al Convergent functional genomics of genome-wide association data for bipolar disorder: comprehensive identification of candidate genes, pathways and mechanisms. Am J Med Genet B: Neuropsychiatr Genet. 2009;150B(2):155–181.[PubMed][Google Scholar]
- 41. Piton A, Michaud JL, Peng H, Aradhya S, Gauthier J, Mottron L, et al Mutations in the calcium-related gene IL1RAPL1 are associated with autism. Hum Mol Genet. 2008;17:3674–3965.[PubMed][Google Scholar]
- 42. Honda S, Hayashi S, Imoto I, Toyama J, Okazawa H, Nakagawa E, et al Copy-number variations on the X chromosome in Japanese patients with mental retardation detected by array-based comparative genomic hybridization analysis. J Hum Genet. 2010;55(9):590–599.[PubMed][Google Scholar]
- 43. Handley MT, Lian LY, Haynes LP, Burgorne RDStructural and functional deficits in a neuronal calcium sensor-1 mutant identified in a case of autistic spectrum disorder. PLoS One. 2010;5(5):e10534.[Google Scholar]
- 44. Piton A, Gauthier J, Hamdan FF, Lafreniere RG, Yang Y, Henrion E, et al Systematic resequencing of X-chromosome synaptic genes in autism spectrum disorder and schizophrenia. Mol Psychiatr. 2010 e-pub ahead of print 18 May 2010: mp.2010. [Google Scholar]
- 45. Marshall CR, Noor A, Vincent JB, Lionel AC, Feuk L, Skaug J, et al Structural variation of chromosomes in autism spectrum disorder. Am J Hum Genet. 2008;82:477–488.[Google Scholar]
- 46. Nawara M, Klapecki J, Borg K, Jurek M, Moreno S, Tryfon J, et al Novel mutation of IL1RAPL1 gene in a nonspecific X-linked mental retardation (MRX) family. Am J Hum Genet A. 2008;146A:3167–3172.[PubMed][Google Scholar]
- 47. Gao X, Xi G, Niu Y, Zhang S, Fu R, Zheng Z, et al A study on the correlation between IL1RAPL1 and human cognitive ability. Neurosci Lett. 2008;438(2):163–167.[PubMed][Google Scholar]
- 48. Bojesen A, Gravholt CHKlinefelter syndrome in clinical practice. Nat Clin Pract Urol. 2007;4(4):192–204.[PubMed][Google Scholar]
- 49. Walsh T, McClellan JM, McCarthy SE, Addington AM, Pierce SB, Cooper GM, et al Rare structural variants disrupt multiple genes in neurodevelopmental pathways in schizophrenia. Science. 2008;320:539–543.[PubMed][Google Scholar]
- 50. Bassett AS, Costain G, Fung WL, Russell KJ, Pierce L, Kapadia R, et al Clinically detectable copy number variations in a Canadian catchment population of schizophrenia. J Psychiatr Res. 2010;44(15):1005–1009.[Google Scholar]
- 51. Shaikh TH, Gai X, Perin JC, Glessner JT, Xie H, Murphy K, et al High-resolution mapping and analysis of copy number variations in the human genome: a data resource for clinical and research applications. Genome Res. 2009;19(9):1682–1690.[Google Scholar]
- 52. Itsara A, Cooper GM, Baker C, Girirajan S, Li J, Absher D, et al Population analysis of large copy number variants and hotspots of human genetic disease. Am J Hum Genet. 2009;84(2):148–161.[Google Scholar]
- 53. Shibata H, Huynh DP, Pulst SMA novel protein with RNA-binding motifs interacts with ataxin-2. Hum Mol Genet. 2009;9(9):1303–1313.[PubMed][Google Scholar]
- 54. Rottnek M, Riggio S, Byne W, Sano M, Margolis RL, Walker RHSchizophrenia in a patient with spinocerebellar ataxia 2: coincidence of two disorders or a neurodegenerative disease presenting with psychosis? Am J Psychiatry. 2008;165(8):964–967.[PubMed][Google Scholar]
- 55. Goldstein JM, Seidman LJ, Makris N, Ahern T, O'Brien LM, Caviness VS, et al Hypothalamic abnormalities in schizophrenia: sex effects and genetic vulnerability. Biol Psychiatry. 2007;61(8):935–945.[PubMed][Google Scholar]
- 56. Goldstein JM, Seidman LJ, O'Brien LM, Horton NJ, Kennedy DN, Makris N, et al Impact of normal sexual dimorphisms on sex differences in structural brain abnormalities in schizophrenia assessed by magnetic resonance imaging. Arch Gen Psychiatry. 2002;59(2):154–164.[PubMed][Google Scholar]
- 57. John B, Enright AJ, Aravin A, Tuschl T, Sander C, Marks DSHuman MicroRNA targets. PLoS Biol. 2004;2(11):e363.[Google Scholar]
- 58. Ribas J, Ni X, Haffner M, Wentzel EA, Salmasi AH, Chowdhury WH, et al miR-21: an androgen receptor-regulated microRNA that promotes hormone-dependent and hormone-independent prostate cancer growth. Cancer Res. 2009;69(18):7165–7169.[Google Scholar]
- 59. Maillot G, Lacroix-Triki M, Pierredon S, Gratadou L, Schmidt S, Benes V, et al Widespread estrogen-dependent repression of micrornas involved in breast tumor cell growth. Cancer Res. 2009;69(21):8332–8340.[PubMed][Google Scholar]
- 60. Page DC, Disteche CM, Simpson EM, de la Chapelle A, Andersson M, Alitalo T, et al Chromosomal localization of ZFX--a human gene that escapes X inactivation--and its murine homologs. Genomics. 1990;7(1):37–46.[PubMed][Google Scholar]
- 61. Davies KE, Mandel JL, Weissenbach J, Fellous MReport of the committee on the genetic constitution of the X and Y chromosomes. Cytogenet Cell Genet. 1987;46(1–4):277–315.[PubMed][Google Scholar]
- 62. Morris JK, Alberman E, Scott C, Jacobs PIs the prevalence of Klinefelter syndrome increasing? Eur J Hum Genet. 2008;16(2):163–170.[PubMed][Google Scholar]
- 63. Jacobs PA, Hassold TJ, Whittington E, Butler G, Collyer S, Keston M, Lee MKlinefelter's syndrome: an analysis of the origin of the additional sex chromosome using molecular probes. Ann Hum Genet. 1988;52(2):93–103.[PubMed][Google Scholar]
- 64. Lorda-Sanchez I, Binkert F, Maechler M, Robinson WP, Schinzel AAReduced recombination and paternal age effect in Klinefelter syndrome. Hum Genet. 1992;89(5):524–530.[PubMed][Google Scholar]
- 65. De Souza E, Morris JK EUROCAT Working GroupCase-control analysis of paternal age and trisomic anomalies. Arch Dis Child. 2010;95(11):893–897.[PubMed][Google Scholar]
- 66. Visootsak J, Graham JM., Jr Klinefelter syndrome and other sex chromosomal aneuploidies. Orphanet J Rare Dis. 2006;1:42.
- 67. Boks MP, de Vette MH, Sommer IE, van Rijn S, Giltay JC, Swaab H, Kahn RSPsychiatric morbidity and x-chromosomal origin in a Klinefelter sample. Schizophr Res. 2007;93:1–3.[PubMed][Google Scholar]
- 68. van Rijn S, Aleman A, Swaab H, Kahn RKlinefelter's syndrome (karyotype 47, XXY) and schizophrenia-spectrum pathology. Br J Psychiatry. 2006;189:459–460.[PubMed][Google Scholar]