Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on May 25, 2005
Human Molecular Genetics 2005 14(14):1947-1954; doi:10.1093/hmg/ddi199

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 14/14/1947    most recent ddi199v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (70) Request Permissions Google Scholar Articles by Bray, N. J. Articles by O'Donovan, M. C. Search for Related Content PubMed PubMed Citation Articles by Bray, N. J. Articles by O'Donovan, M. C. Social Bookmarking          What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oupjournals.org

Haplotypes at the dystrobrevin binding protein 1 (DTNBP1) gene locus mediate risk for schizophrenia through reduced DTNBP1 expression

Nicholas J. Bray¹, Anna Preece¹, Nigel M. Williams¹, Valentina Moskvina^1,2, Paul R. Buckland¹, Michael J. Owen¹ and Michael C. O'Donovan^1,*

¹Department of Psychological Medicine and ²Biostatistics Bioinformatics Unit, School of Medicine, Cardiff University, Heath Park, Cardiff CF14 4XN, UK

^* To whom correspondence should be addressed. Tel: +44 2920743242; Fax: +44 2920746554; Email: odonovanmc{at}cardiff.ac.uk

Received April 4, 2005; Accepted May 17, 2005

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
The DTNBP1 gene, encoding dysbindin, is now generally considered to be a susceptibility gene for schizophrenia. However, the confidence with which this hypothesis can be held has to be tempered by the poor reproducibility between studies in terms of the exact nature of the associated haplotypes, by the failure so far to identify any specific susceptibility variants and by the absence of any demonstrated function associated with any of the risk haplotypes. In the present study, we show that a defined schizophrenia risk haplotype tags one or more cis-acting variants that results in a relative reduction in DTNBP1 mRNA expression in human cerebral cortex. Subsidiary analyses suggest that risk haplotypes identified in other sample groups of white European ancestry also index lower DTNBP1 expression, whereas putative ‘protective’ haplotypes index high DTNBP1 expression. Our data indicate that variation in the DTNBP1 gene confers susceptibility to schizophrenia through reduced expression, and that this, therefore, represents a primary aetiological mechanism in the disorder.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
The identification of susceptibility genes for complex psychiatric disorders holds great promise in shedding light on the primary pathophysiological mechanisms. For schizophrenia, a common and debilitating psychiatric condition of high heritability (1), several genes have recently emerged as strong candidates for conferring disease susceptibility (2). Of these, the gene with arguably the strongest evidence for genetic association with schizophrenia so far is dystrobrevin binding protein 1 (DTNBP1), encoding dysbindin.

DTNBP1 spans 140 kb of chromosome 6p22.3, one of the most consistent linkage regions for schizophrenia (3). Following an initial report of genetic association between DTNBP1 and schizophrenia in Irish pedigrees showing linkage to chromosome 6p (4,5), significant haplotypic associations have been found by several other research group in independent populations of schizophrenics (6–12). These have included samples that do not show significant linkage to the 6p region. However, the associated haplotypes have differed between studies, suggesting possible allelic heterogeneity or, alternatively, population differences in the linkage disequilibrium (LD) structure across the gene.

DTNBP1 does not contain common coding sequence changes that might account for any of the reported genetic associations with schizophrenia (10). The functional variants remain unknown, and the mechanism by which they might confer susceptibility to the disorder has not been clearly demonstrated (13).

However, a previous study from our group (14) provided support to the possibility that unidentified sequence variants in regulatory regions of DTNBP1 might promote susceptibility to schizophrenia by altering expression of the gene. In this earlier study, DTNBP1 in human cerebral cortex was found to show large differences in allelic expression, suggesting variable cis-acting regulatory influences. However, the small sample precluded any investigation into how these might relate to haplotypes showing genetic association with schizophrenia (10).

Two recent post-mortem investigations suggest that altered DTNBP1 expression is relevant to schizophrenia pathophysiology. The first (15) found reduced dysbindin expression in pre-synaptic glutamatergic terminals of the hippocampus in two sets of schizophrenia cases when compared with controls. Importantly, reduced dysbindin expression was seen at both the mRNA and protein levels. A second group (16) found that DTNBP1 mRNA levels were also reduced in the dorsolateral pre-frontal cortex of patients with schizophrenia when compared with controls. This latter study also reported preliminary evidence for an effect of genotype on DTNBP1 expression. Interestingly, significant differences were observed in total DTNBP1 expression between carriers of alternate alleles of a 3'-UTR polymorphism (denoted as P3230), which we had previously shown to be associated with altered allelic expression of DTNBP1 (14). However, there was no reported investigation into how these expression differences might relate to haplotypes implicated in schizophrenia susceptibility.

Given the complex pattern of associations in the literature, and the absence of demonstration of the mechanisms by which risk haplotypes in DTNBP1 lead to altered dysbindin function, it is difficult to confidently mimic the appropriate alteration in DTNBP1 function in model organisms and in other analytic systems, or to extrapolate from those systems to pathophysiology. Indeed, it is possible that reduced total DTNBP1 expression observed in schizophrenia represents not a primary aetiological mechanism per se, but a compensatory response to an enhancement of dysbindin function. It therefore remains important to close the circle between the observations of low dysbindin expression and the genetic findings.

Assays of allele-specific expression in individual heterozyotes offer great power to indirectly detect the influence of cis-acting regulatory variants on gene expression (17,18). Under this approach, an expressed single nucleotide polymorphism (SNP) is used as a copy-specific tag to allow measurement of the relative abundance of mRNA transcribed from each chromosome in individual heterozygotes. In the absence of sequence polymorphism or epigenetic modification affecting expression, each gene copy is expected to be equally expressed. Conversely, differences in the expression of each gene copy within an individual suggest that the subject is heterozygous for a cis-acting functional locus, or that the assayed gene is subject to epigenetic influences.

Because assays of allelic expression expose only the variance in gene expression resulting from cis-acting effects, they are particularly powerful for examining the influence of specific genotypes or haplotypes on expression of the target gene (19,20). For example, we have shown that, despite the small magnitude of the effects, haplotypes at the gene encoding catecholamine-methyl-transferase (COMT) are associated with altered COMT mRNA expression (19), a finding that has been independently replicated (21). The ability to examine small effects in modest sample sizes contrasts with assays based on whole gene expression, in which putative relationships between gene expression and genotype are weakened by the trans-acting influences of other genes on expression of the target gene, and by artefacts like imperfect matching, variable tissue quality or systematic differences in relevant environmental exposures (e.g. drugs) between samples of any given genotype.

In order to assess whether a particular haplotype is associated with altered gene expression, the relative expression of the mRNAs from each chromosome is compared between individuals who are heterozygous for that haplotype and those who carry no copies or are homozygous for the haplotype. If the particular haplotype is associated with altered gene expression, individuals who are heterozygous will display greater relative differences in the expression of each gene copy than those who are not. In the event of an observed difference, in order to assign a direction of effect, it is then necessary to determine the likely phase of the assayed marker allele with respect to the haplotype. This is achieved by constructing the diplotype probabilities in each individual, conditional upon the frequencies of each haplotype that could possibly be present, given the observed genotypes (20).

In the present study, we have extended our analyses of DTNBP1 allelic expression in human cerebral cortex (14) using a considerably enlarged sample group. We have also re-analysed our genetic association data (10) to identify the optimal risk haplotype with respect to phase at the polymorphism informative for the allelic expression assay, and genotyped brain samples of similar ethnicity for the additional SNPs that defined the risk haplotype identified by us, and haplotypes reported in other samples of white European ancestry. We hypothesized that if reduced DTNBP1 expression observed in schizophrenia is a primary aetiological factor, the defined risk haplotype would also be associated with reduced allelic expression of DTNBP1.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
We conducted our allelic expression analyses using expressed SNP rs1047631, which is located within the 3'-UTR of the majority of predicted DTNBP1 transcripts (10). The frequency of the minor G-allele of rs1047631 was 0.13 in the Caucasian brain sample. Of the 149 individuals in that sample, 31 were heterozygous for this SNP and therefore informative for allelic expression analysis. Consistent with our previous finding (14), and with measures of total DTNBP1 expression in brain (16), the common A-allele of SNP rs1047631 showed a highly statistically significant (P<0.0001) reduction in the expression relative to the G-allele (Fig. 1). The mean A/G ratio was 0.86, corresponding to a mean reduction of 14% in the expression of A-allele (using the G-allele as an arbitrary reference point) or a mean increase of 19% in the expression of G-allele (using the A-allele as an arbitrary reference point). Relative under-expression of the A-allele remained significant when the nine individuals assayed in the previous study (14) were excluded (P=0.0005). Repeat assays showed good reproducibility of individual cDNA ratios, with an average coefficient of variation (SD/mean) of 0.05. However, as in our previous study, considerable variability in cDNA ratios was observed between individuals, ranging from 0.64 (a relative decrease of 36% in the expression of A-allele) to a ratio of 1.07 (a relative increase of 7% in the expression of A-allele).

View larger version (8K): [in this window] [in a new window]   Figure 1. Comparison between corrected genomic and cDNA allele ratios in heterozygotes for SNP rs1047631. Data are represented as a ratio of A/G alleles. The data for genomic DNA are the averages of two measurements for each individual sample. The data for cDNA are the averages of four measurements for each individual sample. Significant under-representation of the A-allele is observed in cDNA (P<0.0001).

 
The spread of the data observed in the assay of SNP rs1047631 could potentially reflect multiple cis-acting haplotypes influencing expression, interactive effects between a single cis-acting variant and other trans-acting or environmental effects or (if the cis-acting influences operate in only a proportion of the sampled cells) sampling variance due to differences in the cellular composition between samples. To determine whether the variance was due to sampling, we assayed SNP rs1047631 in an additional 15 samples for which tissue was available from two brain regions. Relative expression of the A-allele was again decreased on average in both brain areas, with a mean decrease of 12% in BA10 tissue (mean ratio: 0.88, range: 0.77–1.11, P=0.0002) and a mean decrease of 11% in BA20 tissue (mean ratio: 0.89, range: 0.70–1.06, P=0.003). Within-subject comparisons between brain regions indicated no significant differences (paired t-test, P=0.61). Moreover, consistent with the spread of data reflecting inter-individual differences in cis-acting influences (rather than random effects), individual cDNA ratios were significantly (P<0.05) correlated between brain regions.

To investigate whether cis-acting influences on DTNBP1 expression are relevant to schizophrenia aetiology, we undertook a post hoc re-analysis of our case–control genetic association data (10) to identify the risk haplotype that includes phase information with respect to rs1047631 and that maximally differentiates cases and controls (Table 1). This was achieved by a 3-marker haplotype that comprises the T-allele of rs2619538, the A-allele of rs3213207 and the A-allele of rs1047631. The frequency of this T–A–A haplotype in schizophrenia cases was 45.6% compared with a frequency of 40.4% in controls (²=6.4, P=0.01).

View this table: [in this window] [in a new window]   Table 1. Determination of the risk haplotype based upon assayed SNP rs1047631 and the three markers that best defined an associated haplotype by global P-value in the study of Williams et al. (10)

 
To assess the effect of the specific T–A–A risk haplotype on allelic expression of DTNBP1, 149 individuals in the Caucasian brain sample were additionally genotyped for SNPs rs2619538 and rs3213207. There were no significant differences in the frequencies of haplotypes, including the specific T–A–A haplotype, between the brain samples and the controls from our case–control analysis (maximum ²=1.3, P=0.25). Diplotype probabilities were calculated for each individual based on the observed haplotype frequencies in the sample. Thirty of the 31 assayed heterozygotes at rs1047631 were successfully genotyped for the additional SNPs. Of these, 15 were predicted to be heterozygous for the specific T–A–A risk haplotype, with a mean diplotype probability of 0.97 (range=0.73–1). To a high level of confidence, the predicted diplotypes for this group of individuals were T–A–A/A–A–G and T–A–A/A–G–G. The mean reduction in the expression of A-allele on chromosomes carrying the T–A–A risk haplotype, relative to the G-allele on the other two haplotypes was 21% (mean cDNA ratio: 0.79, range: 0.64–0.98). The other 15 assayed individuals carried no copies of the risk haplotype. In this group, the mean decrease in the expression of A-allele, relative to the G-allele was only 8% (mean cDNA ratio: 0.92, range: 0.73–1.07). These results are shown in Figure 2. The relative expression of the A-allele was significantly lower when it was carried on the risk haplotype compared with when it was carried on the non-risk haplotypes (P=0.002), with a mean relative reduction of 14% for the risk haplotype (95% CI: 21–6%).

View larger version (9K): [in this window] [in a new window]   Figure 2. Allele ratios at SNP rs1047631, stratified by heterozygosity for the defined schizophrenia risk haplotype. Data are represented as a ratio of A/G alleles. The risk haplotype includes the A-allele of rs1047631. cDNA samples from individuals who are heterozygous for the risk haplotype show lower expression of the A-allele than those from individuals who carry no copies of the risk haplotype (P=0.002).

 
In order to determine the relationship between marker rs1047631, used for allelic expression analysis, and haplotypes showing association with schizophrenia in other studies, we additionally genotyped the Caucasian brain sample for markers P1325 (rs1011313), P1320 (rs760761), P1578 (rs1018381) and P1757 (rs2005976). These SNPs were selected because they constitute core markers used in a number of studies that have found association with schizophrenia (5–7,9,11). Although these studies used samples of principally Caucasian origin, allelic heterogeneity and/or differences in LD structure between the assayed SNP and the actual functional variant(s) might still exist between these and the current sample. Therefore, because genotype/association data for the marker used in our allelic expression analysis are unavailable for these other samples, we must emphasize the more tentative nature of our conclusions with respect to these datasets.

The risk (over-transmitted) 6-marker haplotype of Schwab et al. (6) is defined by the (common) C-allele at P1325 and the (common) C-allele at P1320. This haplotype occurred in the present sample at a frequency of 0.76, approximating the frequency observed in the study of Schwab et al. The vast majority (92%) of chromosomes in our sample that are predicted to carry this risk haplotype also carry the low expression A-allele of rs1047631. Given the strong association between the risk haplotype of Schwab et al. and the low expression allele in the present sample (OR=7, ²=37, P<0.0001), it is likely that it also indexes lower DTNBP1 expression in the sample used by Schwab et al. Moreover, a haplotype defined by alleles G and T at P1635 and P1320, respectively, was markedly under-transmitted in that study (a 9% frequency in transmitted versus a 17% frequency in non-transmitted haplotypes) (S.G. Schwab, personal communication) suggesting the possibility of a protective haplotype. In our sample, this haplotype carries the high expression G-allele of rs1047631 on 88% occasions, and is strongly associated with it (OR=105, ²=145, P<0.0001). These findings are thus consistent with the hypothesis that low expression of DTNBP1 confers susceptibility to schizophrenia, whereas high DTNBP1 expression confers a protective effect.

In the study of Kirov et al. (9), which was based on a Bulgarian sample, the common allele of marker P1635 was significantly over-transmitted to schizophrenia probands. This allele occurred in the brain sample at a frequency of 0.89, approximating that observed in the study of Kirov et al. In the present sample, this allele is very strongly associated with the low expression A-allele of rs1047631 (OR=105, ²=145, P<0.0001), with which it is in phase on 93% of occasions. Moreover, the G-allele of P1635, under-transmitted in the study of Kirov et al., was in phase with the high expression G-allele of rs1047631 on 88% of occasions. Interestingly, in our sample, this under-transmitted allele defines the putative protective haplotype of Schwab et al. perfectly, suggesting congruence between the two genetic studies.

Haplotypic association with schizophrenia has also been reported in the ‘white’ sample of Funke et al. (11) and the familial Swedish sample of van den Bogaert et al. (7). In both cases, the risk haplotypes can be perfectly defined by the T-allele of P1578. Moreover, this particular allele was itself significantly associated with schizophrenia in both these and a third sample group, of Hispanic ethnicity (11). In the present brain sample, the frequency of the T-allele of P1578 was 9%, similar to the frequency observed in the studies of Funke et al. and van den Bogaert et al. Although showing no overlap with the risk haplotype of Schwab et al., this putative risk allele was, in the present sample, exclusively in phase with the lower expressed A-allele of rs1047631 (OR=21, Fisher's exact P<0.0001).

According to van den Oord et al. (22), the risk haplotype of Straub et al. (5) can be tagged in the Irish high-density family sample using the common and rare alleles of markers P1578 and P1757, respectively. In contrast to the other reported haplotypes showing association with schizophrenia, in our brain sample, this haplotype was generally (85% of occasions) in phase with the higher expressed G-allele of rs1047631. However, based upon the number of individuals genotyped by van den Oord et al. and a reported frequency of around 0.058, we calculate that the frequency of the risk haplotype in that sample is significantly lower than in our sample of brains (11%) or in our case–control sample (10%) (P-value for each comparison <0.001). Although the estimate of significance is not exact, because not all the individuals in the study of van den Oord et al. were unrelated, the data nonetheless suggest that important differences might exist between these samples.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
In the present study, we have confirmed variable cis-acting influences affecting allelic expression of DTNBP1 in human brain using a considerably enlarged sample group. Furthermore, having identified an optimal schizophrenia risk haplotype with respect to phase with the SNP used for allelic expression analysis, we were able to show that this haplotype is associated with significantly reduced DTNBP1 expression. This is as predicted by the hypothesis, and strongly suggests that reduced DTNBP1 expression is a primary aetiological factor in the pathophysiology of schizophrenia. As a subsidiary analysis, we also genotyped our brain sample for markers used in several other genetic studies that have found association with schizophrenia using samples of principally Caucasian origin. Consistent with an association between reduced expression of DTNBP1 and risk for schizophrenia, the various risk and protective haplotypes were, in the present sample, predominantly in phase with the lower and higher expressed alleles, respectively.

Our finding of a reduction in the relative expression of the A-allele of SNP rs1047631 is consistent with the findings of Weickert et al. (16), based on measures of total DTNBP1 expression. In that study, carriers of the alternate G-allele showed a significant increase in total DTNBP1 mRNA levels in the pre-frontal cortex. The samples used in the study of Weickert et al. were derived predominantly from African-Americans. In the present study, lower expression of the A-allele was also observed in a large sample of Caucasians, as well as in an additional sample that included individuals of African-American and Asian ethnicity, suggesting the presence of a shared cis-acting influence operating across populations. However, consistent with additional cis-acting influences on DTNBP1 expression, allele ratios were found to vary considerably between individuals.

To investigate whether cis-acting influences on DTNBP1 expression are relevant to schizophrenia, we performed a post hoc re-analysis of genetic association data derived from a case–control sample of similar white European ancestry (10). This sample had been previously genotyped for SNP rs1047631, enabling determination of an optimal risk haplotype with respect to phase at the assayed SNP. In our previous case–control analysis (10), haplotypes constructed from many different permutations of markers yielded evidence for association with schizophrenia, a finding reported in other studies (6,9). In our earlier report (10), the smallest global P-value was yielded by haplotypes constructed from markers rs2619538, rs3213207 and rs2619539. In the present analysis, we also required phase information at rs1047631. It is difficult to choose between different competing marker definitions of the risk haplotype; so, blind to the expression data, we chose to examine the effect of the haplotype that maximized the difference in frequencies between cases and controls (Table 1). As predicted by the hypothesis, the mean expression of the A-allele of rs1047631 was significantly lower when it was on a chromosome carrying this risk haplotype. However, as can be seen from the observed allele ratios (Fig. 2), this haplotype cannot in itself account for all the cis-acting variation in DTNBP1 expression, because a proportion of samples that do not carry the risk haplotype also showed allelic distortion.

A matter of current concern in psychiatric genetics is how to unite the complex patterns of association revealed between schizophrenia and DTNBP1 in the different genetic studies so far. Any attempt to do so may be well confounded by allelic heterogeneity, as well as potential differences in LD structure between sample groups. However, as a subsidiary analysis, we genotyped the brain sample for core markers showing genetic association with schizophrenia in other studies where samples of similar ethnicity were used. Risk haplotypes defined in the majority of these studies were, in the present sample, strongly associated with the low-expression allele of the assayed SNP. Moreover, a putative protective haplotype identified in the samples of Schwab et al. (6) and Kirov et al. (9) showed strong association with the high expression G-allele of the assayed SNP, again consistent with the hypothesis that low DTNBP1 expression confers susceptibility to schizophrenia, whereas high DTNBP1 expression confers a protective effect. Our observation that the disease-associated haplotypes from several different studies have similar functional associations in our sample of brains suggests that the heterogeneity apparent in those association studies may not be grounded in population-specific patterns of LD between functional variants and the markers used (at least in white European founded populations). However, in the absence of genotype data from these other samples confirming the relationships between their associated haplotypes and our marker of allelic expression, these additional analyses, although suggestive, must be considered tentative. We also note that there is an exception to this pattern. Thus, the low frequency risk haplotype that drives the association described by Straub et al. (5), and subsequently refined by van den Oord et al. (22), was, in our brain sample, generally in phase with the higher expression G-allele of rs1047631. However, as mentioned earlier, this haplotype was observed in the present sample at a substantially greater frequency than that was seen in the study of van den Oord et al., suggesting potentially important differences between the two samples. Because the original sample of Straub et al. consisted of families with multiple affected members that demonstrated linkage to 6p, their haplotype could, for example, tag a risk allele that is rare in other samples. However, until the actual risk variant in that sample is identified, any explanation must remain speculative.

We do not assume that any of the SNPs used in the present study have direct effects on DTNBP1 expression, rather than being in LD with the actual functional variants. It is also important to note that, in common with previous studies of DTNBP1 expression (15,16), the present investigation used measures that assayed multiple DTNBP1 transcripts simultaneously, and, therefore, our findings relate to net effects on those transcripts. Thus, although the present findings implicate mechanisms promoting a general down-regulation of DTNBP1 expression in schizophrenia, effects of risk haplotypes on certain transcripts may have gone undetected. It is also possible that, in addition to potential effects on transcription and mRNA stability, our findings reflect allelic differences in the splicing of DTNBP1 transcripts. The effect of risk haplotypes on individual DTNBP1 transcripts warrants further investigation at the protein, as well as mRNA, level.

Besides these considerations, our finding that a specific haplotype showing genetic association with schizophrenia is also associated with reduced expression of DTNBP1 strongly suggests that the reductions in dysbindin protein and mRNA expression observed in schizophrenia by other groups (15,16) are not simply compensatory mechanisms or epiphenomena, but represent a primary aetiological mechanism. The present findings therefore suggest that all mechanisms that down-regulate DTNBP1 expression (i.e. both cis- and trans-acting) will impact on risk for the disorder.

The cellular functions of dysbindin are yet to be fully elucidated. Originally characterized as a dystobrevin-binding protein (23), dysbindin has been found to also form part of the BLOC-1 complex (24). DTNBP1 shows widespread expression in the human brain (16), and a role in glutamatergic synapses has been discussed (5,15,25). A recent study (12) found that, in neuronal culture, over-expression of dysbindin induced expression of SNAP25 and synapsin-I, promoted phosphorylation of Akt and increased glutamate release. Conversely, knockdown of dysbindin expression resulted in lower pre-synaptic protein expression and a decrease in glutamate release (12). Our findings therefore appear consistent with hypotheses of deficient glutamatergic transmission in the pathophysiology of schizophrenia (26,27).

In conclusion, we have clearly demonstrated the presence of variable cis-acting influences on DTNBP1 expression in human brain and shown that a haplotype that is associated with schizophrenia is also associated with reduced expression of DTNBP1. Subsidiary analyses suggest that risk haplotypes identified in other sample groups also index lower DTNBP1 expression. Our findings provide strong support to the hypothesis that reduced dysbindin expression represents a primary aetiological mechanism in schizophrenia.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Brain samples
Principal experiments were performed using post-mortem brain tissue (either frontal, temporal or parietal cortex) from 149 unrelated, anonymized Caucasians (86 males, 63 females; mean age=58, SD=19). Of these, 86 had received no psychiatric or neurological diagnosis at the time of death, 22 had a diagnosis of Alzheimer's disease, 12 a diagnosis of schizophrenia, 14 a diagnosis of bipolar disorder and 15 a diagnosis of major depression. These samples were obtained from three sources (The MRC London Neurodegenerative Diseases Brain Bank, UK; The Stanley Medical Research Institute Brain Bank, Bethesda, USA and The Karolinska Institute, Stockholm, Sweden). Within-subject comparisons of allelic expression between brain regions were made using a second sample, drawn from an additional 60 unrelated anonymized individuals of mixed ethnicity and mixed neuropsychiatric diagnoses from the USA (The Mount Sinai School of Medicine, Department of Psychiatry, Alzheimer's disease and Schizophrenia Brain Bank). Tissue was available from cortical regions BA10 (frontopolar) and BA20 (inferior temporal) for each of these 60 individuals. For all samples, genomic DNA was extracted using standard phenol–chloroform procedures, and total RNA extracted using the RNAwiz^TM isolation reagent (Ambion). Total RNA was treated with DNase prior to reverse transcription using random decamers and the RETROscript^TM kit (Ambion).

Re-analysis of schizophrenia case–control data
In order to identify the optimal risk haplotype with respect to phase at the polymorphism informative for the allelic expression assay, we re-analysed genotype data derived from the Cardiff case–control sample described in the study of Williams et al. (10). This sample comprises 708 DSM-IV schizophrenia cases and 711 matched controls. Determination of the risk haplotype was based upon the expressed SNP rs1047631 and the three markers that defined an associated haplotype in our previous study (10) showing the smallest global P-value. By sequential testing, we first identified which of these three markers, when combined with rs1047631, maximized the difference in frequency between cases and controls. This procedure was then repeated by sequentially adding each of the remaining markers to the optimal 2-marker haplotype to identify the optimal 3-marker haplotype. Addition of genotype information from the remaining marker reduced the difference between cases and controls. Note that this should not be taken to mean that this haplotype is the most strongly associated haplotype, merely that it is the haplotype that, when combined with phase information from the expressed SNP, maximally distinguishes between cases and controls.

Genotyping
All SNPs (and their aliases) used in the present study are shown in Table 2. PCR was carried out by standard procedures using ‘Hot Star’ Taq polymerase (Qiagen). Genotyping was performed by primer extension using the SNaPshot Multiplex Kit (Applied Biosystems). It is important to note that, in keeping with previous studies (10), all alleles are described in relation to the positive strand in the 5'–3' direction by which DTNBP1 is transcribed (rather than chromosomal direction).

View this table: [in this window] [in a new window]   Table 2. DTNBP1 SNPs used in the present study and their aliases.

 
Allelic expression assay
Genomic DNA from all subjects was initially genotyped in order to identify heterozygotes for the expressed marker polymorphism rs1047631. The cDNA samples from heterozygous subjects were then assayed twice, each time as two separate RT reactions, alongside the corresponding genomic DNA samples. Samples were amplified using primers based on single exonic sequence, capable of amplifying either cDNA or genomic DNA: 5'-GTGGTGAGGACAGCGACTCT-3' and 5'-GCTGTTCTTTAAGTTTCTCACACA-3'. RNA samples did not yield detectable levels of product in the absence of an RT step. The same analytic conditions were used for genomic DNA and cDNA so that we could use, for each assay, the average of the ratios observed from genomic DNA (representing a perfect 1:1 ratio of the two alleles) to correct allelic ratios obtained from cDNA for any inequalities in allelic representation specific to that assay (28). Allele ratios for genomic DNA and cDNA are, therefore, reported as the mean of the ratios for that sample after correction by the average genomic ratio for the corresponding assays.

Allelic representation was measured by primer extension and SNaPshot chemistry (Applied Biosystems), as described previously (20), using the extension primer: 5'-TTCTCACACATTATTGGCAATTA-3'. Peak heights of allele-specific extended primers were determined using Genotyper version 2.5 software (Applied Biosystems). The ratio of cDNA peak heights, corrected using the average genomic ratio, was used to calculate relative expression of the two alleles in each individual sample.

Statistical analysis
Differences in allelic expression were tested by comparing genomic ratios with cDNA ratios from the same heterozygous samples. Group comparisons were analysed by Mann–Whitney or, where indicated, by paired t-test. Tests of association were performed using the ² test. All significant tests are two-tailed.

Predicted haplotype frequencies were calculated using EH plus (29), and these formed the basis for calculation of individual diplotype probabilities. We calculate the probability that an individual carries a specific diplotype by first reconstructing all possible combinations of diplotypes for an individual, given the observed genotypes at each locus. We then use the expected distribution of diplotype frequencies, given the observed haplotypes frequencies within the specific sample to identify the most probable diplotype for that individual. The probability of the diplotype within an individual is then the frequency of that diplotype divided by the sum of the frequencies for all possible diplotypes (20).

	ACKNOWLEDGEMENTS

 
We are grateful to the MRC London Neurodegenerative Diseases Brain Bank (UK), the Department of Clinical Neuroscience at the Karolinska Institute (Stockholm, Sweden) and The Mount Sinai School of Medicine Department of Psychiatry Alzheimer's disease and Schizophrenia Brain Bank (New York, USA) for donating brain tissue. Post-mortem brain tissue was also donated by The Stanley Medical Research Institute's brain collection, courtesy of Drs Michael B. Knable, E. Fuller Torrey, Maree J. Webster, Serge Weis and Robert H. Yolken. This work was funded by the Medical Research Council (UK).

Conflict of Interest statement. None declared.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Cardno, A.G., Marshall, E.J., Coid, B., Macdonald, A.M., Ribchester, T.R., Davies, N.J., Venturi, P., Jones, L.A., Lewis, S.W., Sham, P.C. et al. (1999) Heritability estimates for psychotic disorders: the Maudsley twin psychosis series. Arch. Gen. Psychiatry, 56, 162–168.[Abstract/Free Full Text]

2. O'Donovan, M.C., Williams, N.M. and Owen, M.J. (2003) Recent advances in the genetics of schizophrenia. Hum. Mol. Genet., 2, R125–R133.

3. Lewis, C.M., Levinson, D.F., Wise, L.H., DeLisi, L.E., Straub, R.E., Hovatta, I., Williams, N.M., Schwab, S.G., Pulver, A.E., Faraone, S.V. et al. (2003) Genome scan meta-analysis of schizophrenia and bipolar disorder, part II: Schizophrenia. Am. J. Hum. Genet., 73, 34–48.[CrossRef][Web of Science][Medline]

4. Straub, R.E., MacLean, C.J., O'Neill, F.A., Burke, J., Murphy, B., Duke, F., Shinkwin, R., Webb, B.T., Zhang, J., Walsh, D. et al. (1995) A potential vulnerability locus for schizophrenia on chromosome 6p24–22: evidence for genetic heterogeneity. Nat. Genet., 11, 287–293.[CrossRef][Web of Science][Medline]

5. Straub, R.E., Jiang, Y., MacLean, C.J., Ma, Y., Webb, B.T., Myakishev, M.V., Harris-Kerr, C., Wormley, B., Sadek, H., Kadambi, B. et al. (2002) Genetic variation in the 6p22.3 gene DTNBP1, the human ortholog of the mouse dysbindin gene, is associated with schizophrenia. Am. J. Hum. Genet., 71, 337–348.[CrossRef][Web of Science][Medline]

6. Schwab, S.G., Knapp, M., Mondabon, S., Hallmayer, J., Borrmann-Hassenbach, M., Albus, M., Lerer, B., Rietschel, M., Trixler, M., Maier, W. and Wildenauer, D.B. (2003) Support for association of schizophrenia with genetic variation in the 6p22.3 gene, dysbindin, in sib-pair families with linkage and in an additional sample of triad families. Am. J. Hum. Genet., 72, 185–190.[CrossRef][Web of Science][Medline]

7. van den Bogaert, A., Schumacher, J., Schulze, T.G., Otte, A.C., Ohlraun, S., Kovalenko, S., Becker, T., Freudenberg, J., Jonsson, E.G., Mattila-Evenden, M. et al. (2003) The DTNBP1 (dysbindin) gene contributes to schizophrenia, depending on family history of the disease. Am. J. Hum. Genet., 73, 1438–1443.[CrossRef][Web of Science][Medline]

8. Tang, J.X., Zhou, J., Fan, J.B., Li, X.W., Shi, Y.Y., Gu, N.F., Feng, G.Y., Xing, Y.L., Shi, J.G. and He, L. (2003) Family-based association study of DTNBP1 in 6p22.3 and schizophrenia. Mol. Psychiatry, 8, 717–718.[CrossRef][Web of Science][Medline]

9. Kirov, G., Ivanov, D., Williams, N.M., Preece, A., Nikolov, I., Milev, R., Koleva, S., Dimitrova, A., Toncheva, D., O'Donovan, M.C. and Owen, M.J. (2004) Strong evidence for association between the dystrobrevin binding protein 1 gene (DTNBP1) and schizophrenia in 488 parent–offspring trios from Bulgaria. Biol. Psychiatry, 15, 971–975.

10. Williams, N.M., Preece, A., Morris, D.W., Spurlock, G., Bray, N.J., Stephens, M., Norton, N., Williams, H., Clement, M., Dwyer, S. et al. (2004) Identification in 2 independent samples of a novel schizophrenia risk haplotype of the dystrobrevin binding protein gene (DTNBP1). Arch. Gen. Psychiatry, 61, 336–344.[Abstract/Free Full Text]

11. Funke, B., Finn, C.T., Plocik, A.M., Lake, S., DeRosse, P., Kane, J.M., Kucherlapati, R. and Malhotra, A.K. (2004) Association of the DTNBP1 locus with schizophrenia in a US population. Am. J. Hum. Genet., 75, 891–898.[CrossRef][Web of Science][Medline]

12. Numakawa, T., Yagasaki, Y., Ishimoto, T., Okada, T., Suzuki, T., Iwata, N., Ozaki, N., Taguchi, T., Tatsumi, M., Kamijima, K. et al. (2004) Evidence of novel neuronal functions of dysbindin, a susceptibility gene for schizophrenia. Hum. Mol. Genet., 13, 2699–2708.[Abstract/Free Full Text]

13. Kendler, K.S. (2004) Schizophrenia genetics and dysbindin: a corner turned? Am. J. Psychiatry, 161, 1533–1536.[Free Full Text]

14. Bray, N.J., Buckland, P.R., Owen, M.J. and O'Donovan, M.C. (2003) Cis-acting variation in the expression of a high proportion of genes in human brain. Hum. Genet., 113, 149–153.[Web of Science][Medline]

15. Talbot, K., Eidem, W.L., Tinsley, C.L., Benson, M.A., Thompson, E.W., Smith, R.J., Hahn, C.G., Siegel, S.J., Trojanowski, J.Q., Gur, R.E. et al. (2004) Dysbindin-1 is reduced in intrinsic, glutamatergic terminals of the hippocampal formation in schizophrenia. J. Clin. Invest., 113, 1353–1363.[CrossRef][Web of Science][Medline]

16. Weickert, C.S., Straub, R.E., McClintock, B.W., Matsumoto, M., Hashimoto, R., Hyde, T.M., Herman, M.M., Weinberger, D.R. and Kleinman, J.E. (2004) Human dysbindin (>DTNBP1) gene expression in normal brain and in schizophrenic prefrontal cortex and midbrain. Arch. Gen. Psychiatry, 61, 544–555.[Abstract/Free Full Text]

17. Singer-Sam, J., Chapman, V., LeBon, J.M. and Riggs, A.D. (1992) Parental imprinting studied by allele-specific primer extension after PCR: paternal X chromosome-linked genes are transcribed prior to preferential paternal X chromosome inactivation. Proc. Natl Acad. Sci. USA, 89, 10469–10473.[Abstract/Free Full Text]

18. Yan, H., Yuan, W., Velculescu, V.E., Vogelstein, B. and Kinzler, K.W. (2002) Allelic variation in human gene expression. Science, 297, 1143.[Free Full Text]

19. Bray, N.J., Buckland, P.R., Williams, N.M., Williams, H.J., Norton, N., Owen, M.J. and O'Donovan, M.C. (2003) A haplotype implicated in schizophrenia susceptibility is associated with reduced COMT expression in human brain. Am. J. Hum. Genet., 73, 152–161.[CrossRef][Web of Science][Medline]

20. Bray, N.J., Jehu, L., Moskvina, V., Buxbaum, J.D., Dracheva, S., Haroutunian, V., Williams, J., Buckland, P.R., Owen, M.J. and O'Donovan, M.C. (2004) Allelic expression of APOE in human brain: effects of epsilon status and promoter haplotypes. Hum. Mol. Genet., 13, 2885–2892.[Abstract/Free Full Text]

21. Zhu, G., Lipsky, R.H., Xu, K., Ali, S., Hyde, T., Kleinman, J., Akhtar, L.A., Mash, D.C. and Goldman, D. (2004) Differential expression of human COMT alleles in brain and lymphoblasts detected by RT-coupled 5' nuclease assay. Psychopharmacology, 177, 178–184.[CrossRef][Medline]

22. van den Oord, E.J., Sullivan, P.F., Jiang, Y., Walsh, D., O'Neill, F.A., Kendler, K.S. and Riley, B.P. (2003) Identification of a high-risk haplotype for the dystrobrevin binding protein 1 (DTNBP1) gene in the Irish study of high-density schizophrenia families. Mol. Psychiatry, 8, 499–510.[CrossRef][Web of Science][Medline]

23. Benson, M.A., Newey, S.E., Martin-Rendon, E., Hawkes, R. and Blake, D.J. (2001) Dysbindin, a novel coiled-coil-containing protein that interacts with the dystrobrevins in muscle and brain. J. Biol. Chem., 276, 24232–24241.[Abstract/Free Full Text]

24. Li, W., Zhang, Q., Oiso, N., Novak, E.K., Gautam, R., O'Brien, E.P., Tinsley, C.L., Blake, D.J., Spritz, R.A., Copeland, N.G. et al. (2003) Hermansky–Pudlak syndrome type 7 (HPS-7) results from mutant dysbindin, a member of the biogenesis of lysosome-related organelles complex 1 (BLOC-1). Nat. Genet., 35, 84–89.[CrossRef][Web of Science][Medline]

25. Harrison, P.J. and Owen, M.J. (2003) Genes for schizophrenia? Recent findings and their pathophysiological implications. Lancet, 361, 417–419.[CrossRef][Web of Science][Medline]

26. Kim, J.S., Kornhuber, H.H., Schmid-Burgk, W. and Holzmuller, B. (1980) Low cerebrospinal fluid glutamate in schizophrenic patients and a new hypothesis on schizophrenia. Neurosci. Lett., 20, 379–382.[CrossRef][Web of Science][Medline]

27. Tamminga, C.A. (1998) Schizophrenia and glutamatergic transmission. Crit. Rev. Neurobiol., 12, 21–36.[Web of Science][Medline]

28. Hoogendoorn, B., Norton, N., Kirov, G., Williams, N., Hamshere, M.L., Spurlock, G., Austin, J., Stephens, M.K., Buckland, P.R., Owen, M.J. and O'Donovan, M.C. (2000) Cheap, accurate and rapid allele frequency estimation of single nucleotide polymorphisms by primer extension and DHPLC in DNA pools. Hum. Genet., 107, 488–493.[CrossRef][Web of Science][Medline]

29. Zhao, J.H., Curtis, D. and Sham, P.C. (2000) Model-free analysis and permutation tests for allelic associations. Hum. Hered., 50, 133–139.[Web of Science][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				P. V. Ryder and V. Faundez Schizophrenia: The "BLOC" May Be in the Endosomes Sci. Signal., October 20, 2009; 2(93): pe66 - pe66. [Abstract] [Full Text] [PDF]

				N. J. Bray Gene Expression in the Etiology of Schizophrenia Schizophr Bull, May 1, 2008; 34(3): 412 - 418. [Abstract] [Full Text] [PDF]

				N. J. Bray, P. A. Holmans, M. B. van den Bree, L. Jones, L. A. Elliston, G. Hughes, A. L. Richards, N. M. Williams, N. Craddock, M. J. Owen, et al. Cis- and trans- loci influence expression of the schizophrenia susceptibility gene DTNBP1 Hum. Mol. Genet., April 15, 2008; 17(8): 1169 - 1174. [Abstract] [Full Text] [PDF]

				J. M. McClellan, E. Susser, and M.-C. King Schizophrenia: a common disease caused by multiple rare alleles The British Journal of Psychiatry, March 1, 2007; 190(3): 194 - 199. [Abstract] [Full Text] [PDF]

				S. R. G. Setty, D. Tenza, S. T. Truschel, E. Chou, E. V. Sviderskaya, A. C. Theos, M. L. Lamoreux, S. M. Di Pietro, M. Starcevic, D. C. Bennett, et al. BLOC-1 Is Required for Cargo-specific Sorting from Vacuolar Early Endosomes toward Lysosome-related Organelles Mol. Biol. Cell, March 1, 2007; 18(3): 768 - 780. [Abstract] [Full Text] [PDF]

				K. Newell-Litwa, E. Seong, M. Burmeister, and V. Faundez Neuronal and non-neuronal functions of the AP-3 sorting machinery J. Cell Sci., February 15, 2007; 120(4): 531 - 541. [Abstract] [Full Text] [PDF]

				T. M. Caffrey, C. Joachim, S. Paracchini, M. M. Esiri, and R. Wade-Martins Haplotype-specific expression of exon 10 at the human MAPT locus Hum. Mol. Genet., December 15, 2006; 15(24): 3529 - 3537. [Abstract] [Full Text] [PDF]

				K. Talbot, D.-S. Cho, W.-Y. Ong, M. A. Benson, L.-Y. Han, H. A. Kazi, J. Kamins, C.-G. Hahn, D. J. Blake, and S. E. Arnold Dysbindin-1 is a synaptic and microtubular protein that binds brain snapin Hum. Mol. Genet., October 15, 2006; 15(20): 3041 - 3054. [Abstract] [Full Text] [PDF]

				D. Li, D. A. Collier, and L. He Meta-analysis shows strong positive association of the neuregulin 1 (NRG1) gene with schizophrenia Hum. Mol. Genet., June 15, 2006; 15(12): 1995 - 2002. [Abstract] [Full Text] [PDF]

				K. E. Burdick, T. Lencz, B. Funke, C. T. Finn, P. R. Szeszko, J. M. Kane, R. Kucherlapati, and A. K. Malhotra Genetic variation in DTNBP1 influences general cognitive ability Hum. Mol. Genet., May 15, 2006; 15(10): 1563 - 1568. [Abstract] [Full Text] [PDF]

				N. DOI and C. USUI Are High-Risk Haplotypes in DTNBP1 and NRG1 Resistance Genes for Schizophrenia? Am J Psychiatry, May 1, 2006; 163(5): 940 - 941. [Full Text] [PDF]

				N. M Williams, M. C O'Donovan, and M. J Owen Is the Dysbindin Gene (DTNBP1) a Susceptibility Gene for Schizophrenia? Schizophr Bull, October 1, 2005; 31(4): 800 - 805. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 14/14/1947    most recent ddi199v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (70) Request Permissions Google Scholar Articles by Bray, N. J. Articles by O'Donovan, M. C. Search for Related Content PubMed PubMed Citation Articles by Bray, N. J. Articles by O'Donovan, M. C. Social Bookmarking          What's this?

Online ISSN 1460-2083 - Print ISSN 0964-6906

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics