Markers at the pericentriolar material 1 gene (PCM1) have shown genetic association with schizophrenia in both a University College London (UCL) and a USA-based case–control sample. In this paper we report a statistically significant replication of the PCM1 association in a large Scottish case–control sample from Aberdeen. Resequencing of the genomic DNA from research volunteers who had inherited haplotypes associated with schizophrenia showed a threonine to isoleucine missense mutation in exon 24 which was likely to change the structure and function of PCM1 (rs370429). This mutation was found only as a heterozygote in 98 schizophrenic research subjects and controls out of 2246 case and control research subjects. Among the 98 carriers of rs370429, 67 were affected with schizophrenia. The same alleles and haplotypes were associated with schizophrenia in both the London and Aberdeen samples. Another potential aetiological base pair change in PCM1 was rs445422, which altered a splice site signal. A further mutation, rs208747, was shown by electrophoretic mobility shift assays to create or destroy a promoter transcription factor site. Five further non-synonymous changes in exons were also found. Genotyping of the new variants discovered in the UCL case–control sample strengthened the evidence for allelic and haplotypic association (P=0.02–0.0002). Given the number and identity of the haplotypes associated with schizophrenia, further aetiological base pair changes must exist within and around the PCM1 gene. PCM1 protein has been shown to interact directly with the disrupted-in-schizophrenia 1 (DISC1) protein, Bardet-Biedl syndrome 4, and Huntingtin-associated protein 1, and is important in neuronal cell growth. In a separate study we found that clozapine but not haloperidol downregulated PCM1 expression in the mouse brain. We hypothesize that mutant PCM1 may be responsible for causing a subtype of schizophrenia through abnormal cell division and abnormal regeneration in dividing cells in the central nervous system. This is supported by our previous finding of orbitofrontal volumetric deficits in PCM1-associated schizophrenia patients as opposed to temporal pole deficits in non-PCM1-associated schizophrenia patients. Caution needs to be exercised in interpreting the actual biological effects of the mutations we have found without further cell biology. However, the DNA changes we have found deserve widespread genotyping in multiple case–control populations.
The schizophrenias have a lifetime prevalence of approximately 0.85% in the general population. Twin, adoption and family studies point to a strong genetic component with estimates of heritability varying between 66 and 93% and with little or no evidence for family environment having an effect on aetiology.1, 2, 3 Three genetic linkage analyses of the 8p22-21 region in independent schizophrenia family samples have confirmed linkage to schizophrenia with LODs above 3.00.4, 5, 6 Another linkage study provided supportive evidence with a LOD of between 2.00 and 3.00.7
We reasoned that because linkage to schizophrenia on chromosome 8p21–22 could be found in several populations, that it would be a good disease locus to attempt to localize down to the gene level by fine mapping. Our intention from the outset was to find linkage disequilibrium (LD) to a disease locus on 8p21–22 and to sequence any gene implicated in order to find aetiological base pair changes. We knew that the marker D8S261, which is within PCM1, showed both linkage and transmission disequilibrium with schizophrenia in our family linkage sample and that this marker was also associated with schizophrenia in a USA sample of schizophrenic trios but not in a sample of 200 cases and 200 controls from Edinburgh.8 Later we genotyped the University College London (UCL) case–control sample with other markers within PCM1 and found a total of five markers which were associated on chromosome 8p21.3.8 The associated markers were rs445422 (P=0.016), rs13276297 (P=0.019), D8S261 (P=0.0096) and rs370429 (P>0.01). Two other markers showed a trend toward association and these were D8S2616 (P>0.07) and rs3214087 (P=0.07). The marker D8S2615 had also shown association with schizophrenia in an earlier sub-sample, but this marker later became non significant in the whole UCL sample of 450 cases and 450 controls.8 Two and three-marker haplotypic associations with schizophrenia in the London sample were also positive as confirmed with permutation tests (P=0.002 and 0.003, respectively).8 Having found what we thought was good evidence of LD between markers and disease susceptibility alleles at PCM1 we carried out an investigation to find out whether research subjects with schizophrenia had brain volumetric changes associated with PCM1 alleles and haplotypes. It was found that PCM1-associated schizophrenic subjects showed a significant reduction in the volume of orbitofrontal cortex grey matter in comparison to non-PCM1-associated schizophrenia patients who showed grey matter volume reduction in the temporal poles, hippocampus and inferior temporal cortex.8 We have also investigated expression of PCM1 mRNA in the mouse brain as part of a microarray gene expression study comparing clozapine and haloperidol. We found that clozapine significantly downregulated PCM1 gene expression whereas haloperidol did not. This effect was confirmed by quantitative in situ hybridization.9 We argued that this sort of action of drug which is closer to the genetic aetiology of a subtype of schizophrenia, might be one reason why clozapine is the most effective antipsychotic drug.
The PCM1 protein was originally identified by virtue of its distinct cell-cycle-dependent association with the centrosome complex and microtubules.10 The protein appears to associate with the centrosome complex during the G1, S and part of the G2 phase in the cell cycle. Dissociation occurs during mitosis when PCM1 is dispersed throughout the cell.10 Immunolabelling studies performed by Kubo et al.11 found that PCM1 was present in centriolar satellites and in electron dense granules between 70 and 100 nm in diameter. These were originally thought to be scattered only around the centrosomes, but further studies proved that PCM1 was also found throughout the cytoplasm.12 PCM1 has four known transcripts, the longest of which has 39 exons. The open reading frame of PCM1 encodes a protein of 2024 amino acids. The protein contains coil-coiled regions between areas of low complexity as well as an adenosine triphosphate (ATP)/GTPase domain, a nuclear localization domain and a eukaryotic molybdopterin domain. The eukaryotic molybdopterin binding domain is currently found in only five other human genes, xanthine dehydrogenase, sulfite oxidase (mitochondrial precursor), aldehyde oxidase, erythropoietin receptor precursor and the ATP-binding cassette, sub-family A, member 2. PCM1 mRNA expression in the mouse brain has been found to be highest in the hippocampal formation (http://www.brain-map.org/welcome.do). In the human it is expressed above the median level of central nervous system (CNS) expression in most parts of the brain (Probe set 202174_s_at in Human GeneAtlas GNF1H, MAS5; http://symatlas.gnf.org/SymAtlas/).
PCM1 was shown to be essential for cell division because PCM1 antibodies cause cell-cycle arrest when microinjected into fertilized murine eggs.13 Targeting of centrin, pericentrin and ninein was also dramatically reduced after PCM1 depletion using siRNA, overexpression of PCM1 deletion mutants and PCM1 antibody microinjection. As a result of this depletion, the radial organization of the microtubules was found to be disrupted, but did not appear to effect microtubule nucleation.14 Kamiya et al. found that PCM1 forms a complex with disrupted-in-schizophrenia 1 (DISC1) and Bardet-Biedl syndrome 4 (BBS4) through discrete binding domains in each protein.15 DISC1 and BBS4 are required for targeting PCM1 and other cargo proteins, such as ninein, to the centrosome. In the developing cerebral cortex, in utero knockdown of PCM1 mRNA leads to neuronal migration defects, an effect which is also seen after knockdown of either DISC1 or BBS4.15 With simultaneous knockdown of both PCM1 and DISC1 neuronal migration defects become even more prominent.15 Mutations in PCM1 have not yet been associated with any disease. However, PCM1 has been secondarily implicated in what has been called a ‘centrosomal defect’ in BBS, where it is known that the genetically abnormal BBS protein interacts with PCM1.16, 17 BBS related and other proteins known to interact with PCM1 are shown in Figure 3. Below we describe the results of sequencing of specific individuals who had inherited the PCM1 marker alleles and haplotypes associated with schizophrenia. We also present further genetic association findings in the UCL sample and in an independent Scottish case–control sample from Aberdeen.
Materials and methods
The UCL schizophrenia case and control samples were recruited from London and South England and consist of 450 volunteers with schizophrenia and 450 controls. All subjects were included only if both parents were of English, Irish, Welsh or Scottish descent and if three out of four grandparents were of the same descent. One grandparent was allowed to be of Caucasian European origin but not of Jewish or non-EU ancestry, based on the EU countries before the 2004 enlargement. UK National Health Service multicentre and local research ethics approval was obtained and all subjects signed an approved consent form after reading an information sheet. All 450 schizophrenic cases were selected for having prior International Classification of Diseases 10 (ICD10) diagnosis of schizophrenia made by National Health Service (NHS) psychiatrists. The research subjects were then given interviews with the Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L) schedule18 and further data were collected from NHS medical and nursing case notes and all other available sources. Therefore, all cases were selected on the basis of having a primary clinical diagnosis of schizophrenia made by a psychiatrist at interview according to ICD10 criteria and then at the probable level of schizophrenia with Research Diagnostic Criteria (RDC) made at interview by a second research psychiatrist. Research subjects with schizophrenia associated with brain damage were excluded. The ‘supernormal’ control subjects were also interviewed with the initial clinical screening questions of the SADS-L and selected on the basis of not having a family history of schizophrenia, alcoholism or bipolar disorder and for having no past or present personal history of any RDC-defined mental disorder. Genomic DNA was extracted from frozen whole blood samples using a standard cell lysis, proteinase K digestion, phenol/chloroform, ethanol precipitation method. All DNA samples were quantified with PicoGreen (Invitrogen, Paisley, UK) by fluorimetry. A total of 32 cases of schizophrenia were selected for resequencing, based on the criteria of whether an individual carried any of the associated alleles in at least two out of the three microsatellite markers D8S2615, D8S2616 and D8S261 which at that time were all significantly associated with schizophrenia in the first sample genotyped. Where novel nucleotide changes were detected in the schizophrenic samples, 32 randomly selected individuals from the control sample were then sequenced.
The Aberdeen sample consisted of 858 cases of schizophrenia and 591 controls. The cases were recruited through Scottish psychiatric hospitals and met DSM-III-R or DSM-IV criteria of schizophrenia using an operational criteria checklist (OPCRIT). Diagnosis of schizophrenia was confirmed through agreement by two independent senior psychiatrists based on structured clinical interviews for DSM-III-R/DSM-IV (severe combined immunodeficient) and inspection of psychiatric case notes. All controls were volunteers recruited through general practices from the same region of Scotland and were ethnically matched. The control samples were screened for absence of psychiatric illness. Informed consent was obtained from all patients and control individuals. A proportion of the Aberdeen sample was previously employed to test for genetic association between Neuregulin 1 (NRG1) and schizophrenia.19
Sequencing and genotyping
Primer sequences for PCR amplification of the microsatellite markers D8S261, D8S2616 and D8S2615 and the single nucleotide polymorphisms (SNPs) were obtained from ENSEMBL (http://www.ensembl.org/Homo_sapiens/index.html) the University of California Santa Cruz (UCSC) Genome Bioinformatics site (http://genome.ucsc.edu/), or the National Center for Biotechnology Information SNP Database (http://www.ncbi.nlm.nih.gov/projects/SNP/). To sequence the nucleotides encoding the PCM1 gene it was necessary to design primers flanking each of the exons using the intron/exon gene structure found on the Ensembl database (www.ensembl.org). A total of 39 exons and a minimum of 60 base pairs either side of the splice site junctions were sequenced from PCR products amplified from genomic DNA. In addition to this, the putative promoter region (1500 bp 5′ of PCM1 exon 1) was also sequenced with overlapping primers. Primers were designed using PRIMER 3 software.20 Where possible, two exons were amplified together. The primers that were used in this investigation are listed in Table 1. Exon 39 was amplified with two sets of primers because the 3′-untranslated region homopolymeric region could not be resequenced without obtaining shorter amplification products. The PCR product was then sequenced using the Sequitherm EXCEL II DNA Sequencing Kit-LC (Epicentre, Madison, WI, USA). Sequenced PCR products were run on polyacrylamide gels made with Sequagel XR (National Diagnostics, Hessle, UK) using the Li-Cor Global IR2 System, which allows simultaneous bidirectional sequencing from one reaction.21
SNPs selected for screening were genotyped by KBiosciences (Hertfordshire, UK) which employs either the Amplifluor, KASPar or Taqman SNP genotyping methods. Internal controls were used to check for accuracy with approximately 17% of samples duplicated in order to detect error and confirm the reproducibility of genotypes. Because many of the SNPs were discovered by sequencing the selected schizophrenic DNA samples it was possible to validate some of the SNP genotyping using sequence data.
Analyses were performed using SCANGROUP, LDPAIRS and RUNGC, which are part of the GENECOUNTING support package.22, 23 Before association analysis, the program SCANGROUP was used to check the data for errors.22, 23 SCANGROUP takes data from each 96-well DNA microtitre plate in turn and compares this with the rest of the plates to check if there are any anomalies in haplotype frequencies between one plate compared to all of the other plates. Once these data were found to be error free, the genotypes were then analysed to confirm Hardy–Weinberg equilibrium (HWE). Markers with a lack of HWE in the control group (P<0.05) were rejected and genotyping was repeated. Single marker allelic association analyses for SNPs were then performed using standard χ2-tests. The genotypes were then analysed for marker-to-marker LD using LDPAIRS which computes D′ and Cramer's V-tests of LD and maximum likelihood estimates of haplotype frequencies from phase-unknown case–control data.22, 23 Cramer's V is equivalent to r, we therefore transformed these figures by squaring to obtain r2 values. Multi-marker haplotype tests of association with schizophrenia were performed using RUNGC with the significance of any overall haplotype association computed using a permutation test.23, 24 To compare individual haplotype associations between the UCL and Aberdeen samples, haplotype analysis was performed using Haploview25 with permutation testing to generate empirical significance values. LD blocks were defined using a solid spine of D′.
Bioinformatic sequence analyses
The putative promoter region, consisting of approximately 1500 bp 5′ of exon 1 was analysed for predicted transcription factor binding sites using TESS26 to assess the potential relevance of any SNPs found within this region. Protein modelling was performed by the program 3D-PSSM.27, 28, 29 This is a program that uses a method for protein fold recognition using 1D and 3D sequence profiles as well as taking into account the secondary structure and solvation potential data (http://www.sbg.bio.ic.ac.uk/~3dpssm/).
Preparation of nuclear extracts from HeLa cells
HeLa cells were rinsed with cold phosphate-buffered saline and then harvested by scraping in cytoplasmic lysis buffer (10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) pH 7.6, 1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol (DTT), 1 mM vanadate, 1 tablet ‘complete’ protease inhibitor (Roche Diagnostics Ltd., Burgess Hill, UK)). The cells were then lysed with Nonidet P-40 (Sigma-Aldrich, Poole, Dorset, UK) and the nuclear cell pellets collected by centrifugation. The nuclear pellets were then resuspended in nuclear lysis buffer (20 mM HEPES pH 7.6, 0.2 mM EDTA, 0.1 mM EGTA, 25% glycerol, 0.42 M NaCl, 1 mM DTT, 1 mM vanadate and 1/2 tablet ‘complete’ protease inhibitor; Roche Diagnostics Ltd.). The nuclei were disrupted by repeated freeze/thawing in dry ice with ethanol and a water bath at 37 °C. Cellular debris was then removed by centrifugation and the nuclear extract stored in aliquots at −80 °C.
Electrophoretic mobility shift assays
Equimolar amounts of complementary single-stranded oligonucleotides were annealed to form duplexes by gradual cooling from 94 °C to room temperature in sodium chloride-Tris-EDTA (STE) buffer (10 mM Tris pH 8.0, 50 mM, NaCl, 1 mM EDTA). The double-stranded probes were 5′ end labelled with γ-32P-ATP (GE Healthcare, Bucks, UK) using T4 polynucleotide Kinase (Promega, Southampton, UK). The probes were then incubated in a 20 μl reaction with 5 μg nuclear extract, 1 μl of poly(dI•dC) (1 mg ml−1), 10 μl of 2 × Parker buffer (16% Ficoll, 40 mM HEPES, 100 mM KCl, 2 mM EDTA, 1 mM DTT) and various amounts of unlabelled competitor probe for 30 min at room temperature.
Controls for the experiments included incubating the labelled probes without nuclear extract, probes with nuclear extract without competitor and finally probes with nuclear extract and various amounts of specific and homologous and heterologous competitors. Protein/probe complexes were resolved on non-denaturing 4% polyacrylamide gels.
Aetiological base pair change discovery and analysis
A total of 37 SNPs were found by sequencing DNA from PCM1-associated subjects with schizophrenia and these are given in Table 2. Sequence analysis of the promoter region which included 1500 bp 5′ of exon 1 revealed nine SNPs. On the basis of differences in the frequencies of the SNPs in 32 cases as compared to 32 normal control sample, 14 possibly aetiological SNPs were identified. Genotype data had already been obtained for nine of these SNPs, rs916550, rs412750, rs454755, rs13276297, rs208753, rs3780103, rs6991775, rs3214087 and rs370429.8 Therefore five additional SNPs were genotyped for this study rs208747, rs7814692, rs445422, PCM1_11305_67 and PCM1_51659_67 in the UCL case and control sample (Tables 2 and 4).
TESS analysis of the promoter region indicated that the SNP rs208747 appeared to alter either a T allele multiprotein bridging factor-1 (MBF1) or TATA binding protein (TBP) transcription factor recognition site to create a myocyte enhancer factor-2 (MEF2), interferon-stimulated gene factor-1 (ISGF1) or CCAAT enhancer binding protein alpha (C/EBPα) factor binding site with allele A. The rs208747 allele A was more common in the cases selected for sequencing based on being associated with schizophrenia than in random controls (21/64 alleles vs 0/64 alleles). Several other changes in potential transcription factor binding sites detected by TESS were also found. SNP rs7814692 appeared to change a binding site for the Sp1 transcription factor and had a frequency of 35/64 ‘A’ alleles in the schizophrenia patients compared to 19/64 ‘A’ alleles in the controls. SNPs rs916550 and rs445422 and rs3214087 were in close proximity to splice site junctions. SNPs causing changes to the amino-acid sequence were rs412750, rs2285302, rs208753, rs4440668, rs7009117, rs172298, rs370429 and PCM1_51659_67. SNPs which caused synonymous substitutions within the coding sequence were rs454755, PCM1_11305_67, rs3780103 and rs6991775, details of which can be found in Table 2. SNPs detected by sequencing were genotyped in the entire case–control sample after differences in minor allele frequencies were found in PCM1-associated schizophrenia patients compared to random controls.
Further genetic analyses in the UCL sample
One new SNP, rs208747 that was predicted to alter either an MBF1 or TBP transcription factor binding site to MEF2, ISGF1 or C/EBPα showed allelic association in the UCL case–control sample (χ2=5.47, P=0.019; Tables 2 and 4). Pairwise LD was calculated between all pairs of markers genotyped in the UCL case–control sample at the PCM1 locus using LDPAIRS (Table 3). The markers typed in this study were in high LD with each other according to the D′ statistic. The r2 values confirmed that extensive LD is present over the region for all markers.
Single marker allelic associations with the new SNPs and other markers are given in Table 4 for the UCL and Aberdeen samples. Haplotypic associations in the UCL sub-sample are listed in Table 5 and are an update on those previously found to be associated with schizophrenia. A sliding window of two locus haplotypes showed that haplotypes across the whole length of the gene were significantly associated with schizophrenia. One three-marker haplotype (PCM1F1: rs7814692 G, rs454755T, rs3780103 G) included markers which did not individually display allelic association with schizophrenia. However examination of the LD in the region of the three SNPs showed very high r2 values between each SNP. Therefore, it is likely that an unidentified mutation or aetiological base pair change exists in this haplotype.
Replication studies in the Aberdeen schizophrenia case–control sample
Eight SNPs that had been genotyped in the UCL sample were successfully genotyped in the Aberdeen sample as given in Table 4. Of the four SNPs that showed association in the UCL sample one (rs445422) showed a significant association in the Aberdeen sample (one sided P=0.0185; Table 4). For two of the remaining SNPs (rs208747 and rs370429) the allele elevated in frequency in the Aberdeen schizophrenia cohort was consistent with that observed in the UCL schizophrenia sample. In a combined analysis of data from the two samples three markers: rs208747, rs445422 and rs370429 showed significant two-tailed evidence for association P=0.015, 0.002 and 0.006, respectively. In the combined sample the marker rs445422 gave an odds ratio of 1.985 (95% CI 1.27–3.10). SNPs rs208747, rs445422 and rs370429 all show high levels of LD with one another: rs208747, rs445422 D′=0.988, r2=0.908; rs208747, rs370429 D′=0.917, r2=0.821; and rs445422, rs370429 D′=0.978, r2=0.914 (Figure 1).
In Gurling et al.8 we reported three sets of global haplotype analysis that yielded global empirical significant evidence for association (P=0.002–0.003). We tested the Aberdeen case–control sample for evidence that the same haplotypes were also associated in this sample and then carried out a combined analysis of the two samples together (Table 6). Haplotype PCM1A2 contained the same alleles significantly increased in cases as compared to controls in both the London (two-sided empirical P=0.029) and Aberdeen samples (one-sided asymptotic P=0.027). Combined analysis of the two data sets for haplotype PCM1A2 gave a permuted two-tailed empirical significance of P=0.007. The previously reported haplotype PCM1B1 which was associated with schizophrenia in the UCL sample showed a trend towards association in the Aberdeen sample (χ2=2.020, one-sided asymptotic P=0.078) and the related haplotype PCM1C1 was significantly associated in both samples (Aberdeen χ2=2.76, one-sided asymptotic P=0.049). Both of these haplotypes were also elevated when the two data sets were combined (PCM1B1 empirical P=0.033; PCM1C1 empirical P=0.020).
Haplotype analysis of all SNPs genotyped across the UCL and Aberdeen case–control samples gave further support for haplotypic association. An eight-locus haplotype comprising alleles rs208747 allele A, rs445422 allele T, rs454755 allele T, rs13276297 allele C, rs3780103 allele G, rs6991775 allele A, rs3214087 allele C, rs370429 allele A, was elevated in frequency in both the UCL (empirical P=0.016) and Aberdeen (empirical P=0.031) schizophrenia samples (UCL schizophrenia 3.1%, UCL controls 1.4%; Aberdeen schizophrenia 2.3%, Aberdeen controls 1%). In the combined analysis this haplotype gave permuted evidence for association (P=0.008). These data are summarized in Supplementary Table 7 online at www.ucl.ac.uk/~rejuhxg.
Consistent permuted haplotypic association was observed after defining haplotype blocks by a solid spine of D′ in the UCL sample. This block definition was then applied to the Aberdeen and combined data sets. Two haplotype blocks were identified using this method (Figure 1a–c). The data from this haplotype block analysis are shown in Supplementary Table 8 at www.ucl.ac.uk/~rejuhxg. Block 1 consisted of SNPs rs208747 and rs445422; block two consisted of SNPs rs454755, rs13276297, rs3780103, rs6991775, rs3214087 and rs370429. The block 2 six-marker haplotype consisting of rs454755 allele T, rs13276297 allele C, rs3780103 allele G, rs6991775 allele A, rs3214087 allele C and rs370429 allele A, showed permuted evidence for association in the UCL sample (P=0.046), the Aberdeen sample (P=0.049) and in the combined sample (P=0.002 (see Supplementary Table 8 online at www.ucl.ac.uk/~rejuhxg). This haplotype is highly related to the eight-marker haplotype described above.
Incubation of HeLa cell nuclear extracts with probes for the alternate alleles of rs208747 indicate that both sequences specifically interact with a nuclear protein, most likely to be a transcription factor. The competition experiments indicate that binding of the rs208747 ‘A’ probe with the likely transcription factor takes place with greater affinity than the binding of the rs208747 ‘T’ probe (compare the visible band in lanes 7, 11 and 12 with the absence of a band in lane 8 in Figure 2).
It is possible that any or all of the three aetiological base pair changes we have identified in PCM1 that are associated with schizophrenia could be pathogenic. SNP rs208747 that changes a transcription factor binding site, rs445442 that alters a splice site and SNP rs370429 that causes a threonine to isoleucine amino-acid change are present in the same haplotype and they may have a synergistic negative effect on the expression and function of the gene. We present biochemical evidence that rs208747 may have an effect on transcription factor binding, however further in vitro experimental data on the effect of these base pair changes on gene expression are needed. Humanizing transgenic mice with combinations of the three mutations would be a useful strategy to identify which change, if any, is the most pathological. It is now known that PCM1 is involved in maintaining centrosome integrity and the regulation of the microtubule cytoskeleton. Kamiya et al.15 and others have shown that perturbation of PCM1 expression through in utero RNAi experiments can lead to altered neuronal cell organization in the mouse brain. We hypothesize that this mechanism is a chronogenetic effect leading to the development of schizophrenia which is pathogenetic when CNS cells particularly in the prefrontal cortex and elsewhere are dividing in adolescence. The base pair change, SNP rs445422 which is 5 bp 5′ of exon 3, could alter the exon 3 splice acceptor site. This is a type of mutation known to cause frontotemporal dementia with Parkinson's disease and phenylketonuria31 and this demands to be investigated in relation to PCM1.
SNP rs13276297 was also found to be associated with schizophrenia and was localized within intron 6 of PCM1. It is approximately 59 bp 3′ from the microsatellite marker D8S2616 which showed a trend towards association with schizophrenia (χ2=10.1, empirical P=0.066, asymptotic P=0.002). It is possible that this base pair change has an effect on mRNA stability or processing. Another marker associated with schizophrenia in our sample and a USA trio sample is the microsatellite D8S261.8 This marker has also shown linkage to schizophrenia in several studies.32, 33 Speculation exists as to whether microsatellites themselves can effect gene expression as has been found in bacteria.34 Alternatively a silencing element or an enhancer could be affected by microsatellite length, as illustrated by the marker NACP-Rep1 found 10 kb upstream of the α-synuclein (SNCA) gene. This marker is thought to act as a negative modulator and is postulated to influence differing risks of sporadic Parkinson's disease.35 Microsatellite variation affecting transcription factor binding capabilities has been reported in several investigations.36, 37 There have been reports of intronic microsatellites directly effecting gene expression, such as in 97% of Freidrich's ataxia patients, and the GT microsatellite repeat from the human cardiac Na+Ca2+ exchanger gene.38, 39 Microsatellites are found to be non-randomly distributed within the genomes of plants, fungi and animals, located more frequently near transcribed gene-rich areas and could provide another basis for quantitative genetic variation.40, 41, 42, 43
Three synonymous exonic SNPs, PCM1_11305_67, rs3780103 and rs6991775 in exons 14, 16 and 17 were observed. Synonymous SNPs have been found to affect mRNA stability and degradation.44, 45 However, none of these SNPs were individually associated with schizophrenia. Four non-synonymous SNPs were found rs412750, rs208753, rs370429 and PCM1_51659_67 in exons 5, 12, 28 and 33. Of these only rs370429 was found to be associated with schizophrenia. rs370429 encodes a threonine to isoleucine substitution. In silico protein analysis was carried out to investigate whether this amino-acid substitution significantly changed the conformation of PCM1 at either a secondary, tertiary or quaternary level. The threonine to isoleucine polymorphism was found to be within a helical region of the protein. There are two cysteine residues on either side of the amino-acid change which in turn creates a phosphorylation site. This causes the formation of a disulphide bridge that could potentially change the conformation of the protein in combination with the amino-acid change. Using comparative proteomics we found that this region is conserved in mouse, but not in chicken, rat or in Fugu.
Protein interaction databases as well as the work of Kamiya et al.15 have listed experiments that show PCM1 interacting with several proteins including Huntingtin-associated protein-1 (HAP1), DISC1, pericentrin (PCNT2) and two of the recently characterized proteins involved in BBS; tetratricopeptide repeat domain 8 (TTC8) and BBS4. These interactions are shown in Figure 3.17, 46, 47, 48 In addition to the PCNT2 interaction with PCM1, the protein also interacts with DISC1 and calmodulin 1 (CALM1). DISC1 has been implicated in bipolar disorder and schizophrenia and CALM1 interacts with many proteins including the stable tubule only polypeptide (STOP) proteins mentioned below.49, 50 As described in the ‘Introduction’ Kamiya et al.15 have shown that PCM1, DISC1 and BBS4 can all disrupt neuronal organization in the mouse when their expression is downregulated. HAP1 is a cytoskeletal protein that binds to PCM1 between residues 1279 and 1799.46 It has been suggested that HAP1 is involved in neuronal vesicle trafficking within the cell because it is expressed in neurons and has been found to be strongly expressed in terminal neuronal vesicle rich cell fractions.51 Other studies have also concluded that both Huntingtin and HAP1 have a function in vesicular trafficking.52, 53, 54 Of note is the fact that psychotic symptoms sometimes leading to a diagnosis of schizophrenia are found in Huntington's disease patients.55, 56 BBS4 and TTC8 are two members of the BBS family of proteins. TTC8 localizes to the centrosomes and basal bodies and interacts with PCM1.48 Silencing or truncation of BBS4 disrupts the cellular localization of PCM1 and its associated proteins. This causes microtubular destabilization, resulting in an arrest of mitosis/meiosis and eventual apoptosis.16 BBS is a heterogeneous disorder; symptoms include polydactyly, cognitive impairment, obesity, retinal dystrophy, diabetes mellitus and decreased olfaction. As with Huntington's disease, BBS sufferers appear to be at a higher risk of developing schizophrenia.48
Several genes which have a microtubule-associated function have been implicated in schizophrenia either through genetic association or gene expression studies, these include DISC1,57, 58, 59, 60, 61 fasciculation and elongation protein-ζ-1 (FEZ1), nuclear-distribution gene E homologue-like 1 (NUDEL),62, 63, 64 reelin (RELN),65, 66 v-akt murine thymoma viral oncogene homologue 1 (AKT1) and glucose synthase kinase-3β (GSK3β).67, 68 Further evidence for the effect of microtubule function on behaviour has come from experiments that involve destabilizing of neuronal cold-stable microtubules that were found to cause behavioural deficits in mice. These deficits were reversed by using neuroleptics.50 Amongst the genes affected was STOP which encodes calmodulin-binding and calmodulin-regulated microtubular-associated proteins (MAPs). STOP null mice exhibit no gross anatomical changes, however reduced synaptic vesicle pools and impaired synaptic plasticity were observed in STOP knockouts. Behavioural changes included reduced olfaction, social withdrawal, abnormal freezing and burrowing, anxiety-like activity, and perturbed interaction with physical environment. Interestingly, long-term administration of neuroleptics appeared to reduce these symptoms. Specifically a mixture of chlorpromazine and haloperidol administered from 6 days pre-partum appeared to be the optimal medication for these mice. This finding in combination with the fact that HAP1 has been implicated in synaptic trafficking is an alternative to the hypothesis that abnormalities in neuronal migration in schizophrenia might be caused by PCM1 genetic variation and mutations.
Allelic and haplotypic analyses in our sample have implicated all parts of the PCM1 gene. We found very strong LD between several PCM1 markers and good evidence of association with several haplotypes, not all of which carried the more obvious putative aetiological base pair changes, thus it would appear that a number of disease-related DNA mutations or SNPs are present in several haplotype backgrounds. This is reflected in a consistent 2- to 3-fold change in frequencies for most alleles or haplotypes associated with schizophrenia in cases compared to controls. It is obvious that only a small percentage of the schizophrenic subjects are PCM1 associated in our sample and even this subgroup cannot have been accounted for in terms of the number of putative aetiological base pair changes found so far. Further sequencing of PCM1-associated cases in an enlarged sample should reveal new aetiological base pair changes.
The evidence of replication in the Aberdeen sample is encouraging. However, only one marker and one haplotype achieved formal two-tailed levels of significance levels of association in both the UCL and Aberdeen samples. A number of other individual markers and haplotypes show allele or haplotype frequency changes that are in the same direction in the two samples. In conclusion there is accumulating evidence for the involvement of MAPs in the aetiology of schizophrenia. Genes involved in microtubular processes such as DISC1, NUDEL, RELN and AKT1/GSK3β have all been either associated with schizophrenia through genetic studies or have been implicated in expression studies. Possible disease pathways may involve defects in neuronal migration, or through a mechanism in which synaptic vesicular trafficking becomes abnormal. Further investigation into PCM1 is warranted to discover and characterize more SNPs and mutations in exonic regions and locus control regions. At present because we have found that the threonine to isoleucine amino-acid change is in LD with the splice site and transcription factor site mutations and there could be mutation synergy between all three changes in producing a pathological effect. The gene expression data provide evidence for the downregulation of PCM1 with administration of clozapine which was confirmed with in situ hybridization in mouse experiments. It is possible that neuroleptics work by having a function in stabilizing structural protein changes in the CNS. This could imply we are expecting a gain of function mutation of PCM1 in the disease state. Further research such as ‘humanizing’ a mouse transgenic model with any, all or combinations of the possible aetiological base pair changes we have found could help resolve which genetic change is pathological.
Fulker DW . A biometrical genetic approach to intelligence and schizophrenia. Soc Biol 1973; 20: 266–275.
McGue M, Gottesman II, Rao DC . The transmission of schizophrenia under a multifactorial threshold model. Am J Hum Genet 1983; 35: 1161–1178.
Sullivan PF, Kendler KS, Neale MC . Schizophrenia as a complex trait: evidence from a meta-analysis of twin studies. Arch Gen Psychiatry 2003; 60: 1187–1192.
Blouin JL, Dombroski BA, Nath SK, Lasseter VK, Wolyniec PS, Nestadt G et al. Schizophrenia susceptibility loci on chromosomes 13q32 and 8p21. Nat Genet 1998; 20: 70–73.
Brzustowicz LM, Honer WG, Chow EW, Little D, Hogan J, Hodgkinson K et al. Linkage of familial schizophrenia to chromosome 13q32. Am J Hum Genet 1999; 65: 1096–1103.
Gurling HM, Kalsi G, Brynjolfson J, Sigmundsson T, Sherrington R, Mankoo BS et al. Genomewide genetic linkage analysis confirms the presence of susceptibility loci for schizophrenia, on chromosomes 1q32.2, 5q33.2, and 8p21–22 and provides support for linkage to schizophrenia, on chromosomes 11q23.3-24 and 20q12.1-11.23. Am J Hum Genet 2001; 68: 661–673.
Kendler KS, MacLean CJ, O'Neill FA, Burke J, Murphy B, Duke F et al. Evidence for a schizophrenia vulnerability locus on chromosome 8p in the Irish Study of High-Density Schizophrenia Families. Am J Psychiatry 1996; 153: 1534–1540.
Gurling HM, Critchley H, Datta SR, McQuillin A, Blaveri E, Thirumalai S et al. Genetic association and brain morphology studies and the chromosome 8p22 pericentriolar material 1 (PCM1) gene in susceptibility to schizophrenia. Arch Gen Psychiatry 2006; 63: 844–854.
Rizig MA, McQuillin A, Robinson M, Hunt SP, Ng A, Harrison A et al. A gene expression and systems pathway analysis of the effects of clozapine compared to haloperidol in the mouse brain implicates susceptibility genes for schizophrenia. (Unpublished data).
Balczon R, Bao L, Zimmer WE . PCM-1, A 228-kD centrosome autoantigen with a distinct cell cycle distribution. J Cell Biol 1994; 124: 783–793.
Kubo A, Sasaki H, Yuba-Kubo A, Tsukita S, Shiina N . Centriolar satellites: molecular characterization, ATP-dependent movement toward centrioles and possible involvement in ciliogenesis. J Cell Biol 1999; 147: 969–980.
Kubo A, Tsukita S . Non-membranous granular organelle consisting of PCM-1: subcellular distribution and cell-cycle-dependent assembly/disassembly. J Cell Sci 2003; 116 (Part 5): 919–928.
Balczon R, Simerly C, Takahashi D, Schatten G . Arrest of cell cycle progression during first interphase in murine zygotes microinjected with anti-PCM-1 antibodies. Cell Motil Cytoskeleton 2002; 52: 183–192.
Dammermann A, Merdes A . Assembly of centrosomal proteins and microtubule organization depends on PCM-1. J Cell Biol 2002; 159: 255–266.
Kamiya AT, Kubo PL, Engelhard K-I, Ishizuka C, Kubo K, Tsukita A et al. A PCM1 is recruited to the centrosome by the cooperative action of DISC1 and BBS4 and is a candidate for psychiatric illness. Arch Gen Psychiatry 2008; 65: 996–1006.
Kim JC, Badano JL, Sibold S, Esmail MA, Hill J, Hoskins BE et al. The Bardet-Biedl protein BBS4 targets cargo to the pericentriolar region and is required for microtubule anchoring and cell cycle progression. Nat Genet 2004; 36: 462–470.
Badano JL, Teslovich TM, Katsanis N . The centrosome in human genetic disease. Nat Rev Genet 2005; 6: 194–205.
Spitzer R, Endicott J . The Schedule for Affective Disorders and Schizophrenia, Lifetime Version, 3rd edn. New York State Psychiatric Institute: New York, 1977.
Stefansson H, Sarginson J, Kong A, Yates P, Steinthorsdottir V, Gudfinnsson E et al. Association of neuregulin 1 with schizophrenia confirmed in a Scottish population. Am J Hum Genet 2003; 72: 83–87.
Rozen S, SkaletskyIn HJ . Primer3 on the WWW for General Users and for Biologist Programmers. Humana Press: Totowa, NJ, 2000; 365–386.
Middendorf LR, Bruce JC, Bruce RC, Eckles RD, Grone DL, Roemer SC et al. Continuous, on-line DNA sequencing using a versatile infrared laser scanner/electrophoresis apparatus. Electrophoresis 1992; 13: 487–494.
Curtis D, Knight J, Sham PC . Program report: GENECOUNTING support programs. Ann Hum Genet 2006; 70 (Part 2): 277–279.
Zhao JH, Lissarrague S, Essioux L, Sham PC . GENECOUNTING: haplotype analysis with missing genotypes. Bioinformatics 2002; 18: 1694–1695.
Zhao JH, Curtis D, Sham PC . Model-free analysis and permutation tests for allelic associations. Hum Hered 2000; 50: 133–139.
Barrett JC, Fry B, Maller J, Daly MJ . Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 21: 263–265.
Schug J . Using TESS to predict transcription factor binding sites in DNA sequence. In: Baxevanis AD (ed). Current Protocols in Bioinformatics. John Wiley and Sons, London, 2003.
Kelley LA, Maccallum R, Sternberg MJE . Recognition of remote protein homologies using three-dimensional information to generate a position specific scoring matrix in the program 3D-PSSM. In: Istrail I, Pevzner P, Waterman M (eds). RECOMB 99, Proceedings of the Third Annual Conference on Computational Molecular Biology. The Association for Computing Machinery: New York, 1999, pp 218–225.
Kelley LA, MacCallum RM, Sternberg MJE . Enhanced genome annotation using structural profiles in the program 3D-PSSM. J Mol Biol 2000; 299: 499–520.
Fischer D, Barret C, Bryson K, Elofsson A, Godzik A, Jones D et al. CAFASP-1: critical assessment of fully automated structure prediction methods. Proteins Struct Funct Genet 1999; Suppl 3: 209–217.
Gossett LA, Kelvin DJ, Sternberg EA, Olson EN . A new myocyte-specific enhancer-binding factor that recognizes a conserved element associated with multiple muscle-specific genes. Mol Cell Biol 1989; 9: 5022–5033.
Hutton M, Lendon CL, Rizzu P, Baker M, Froelich S, Houlden H et al. Association of missense and 5[prime]-splice-site mutations in tau with the inherited dementia FTDP-17. Nature 1998; 393: 702–705.
Levinson DF, Mowry BJ, Sharpe L, Endicott J . Penetrance of schizophrenia-related disorders in multiplex families after correction for ascertainment. Genet Epidemiol 1996; 13: 11–21.
Pulver AE, Lasseter VK, Kasch L, Wolyniec P, Nestadt G, Blouin JL et al. Schizophrenia: a genome scan targets chromosomes 3p and 8p as potential sites of susceptibility genes. Am J Med Genet 1995; 60: 252–260.
Li YC, Korol AB, Fahima T, Nevo E . Microsatellites within genes: structure, function, and evolution. Mol Biol Evol 2004; 21: 991–1007.
Chiba-Falek O, Kowalak JA, Smulson ME, Nussbaum RL . Regulation of alpha-synuclein expression by poly (ADP ribose) polymerase-1 (PARP-1) binding to the NACP-Rep1 polymorphic site upstream of the SNCA gene. Am J Hum Genet 2005; 76: 478–492.
Martin P, Makepeace K, Hill SA, Hood DW, Moxon ER . Microsatellite instability regulates transcription factor binding and gene expression. PNAS 2005; 102: 3800–3804.
Iglesias AR, Kindlund E, Tammi M, Wadelius C . Some microsatellites may act as novel polymorphic cis-regulatory elements through transcription factor binding. Gene 2004; 341: 149–165.
Gabellini N . A polymorphic GT repeat from the human cardiac Na+Ca2+ exchanger intron 2 activates splicing. Eur J Biochem 2001; 268: 1076–1083.
Becker E, Richardson DR . Frataxin: its role in iron metabolism and the pathogenesis of Friedreich's ataxia. Int J Biochem Cell Biol 2001; 33: 1–10.
Bachtrog D, Weiss S, Zangerl B, Brem G, Schlotterer C . Distribution of dinucleotide microsatellites in the Drosophila melanogaster genome. Mol Biol Evol 1999; 16: 602–610.
Khashnobish A, Hamann A, Osiewacz HD . Modulation of gene expression by (CA) n microsatellites in the filamentous ascomycete Podospora anserina. Appl Microbiol Biotechnol 1999; 52: 191–195.
La Rota M, Kantety R, Yu J-K, Sorrells M . Nonrandom distribution and frequencies of genomic and EST-derived microsatellite markers in rice, wheat, and barley. BMC Genomics 2005; 6: 23.
Morgante M, Hanafey M, Powell W . Microsatellites are preferentially associated with nonrepetitive DNA in plant genomes. Nat Genet 2002; 30: 194–200.
Duan J, Antezana MA . Mammalian mutation pressure, synonymous codon choice, and mRNA degradation. J Mol Evol 2003; 57: 694–701.
Duan J, Wainwright MS, Comeron JM, Saitou N, Sanders AR, Gelernter J et al. Synonymous mutations in the human dopamine receptor D2 (DRD2) affect mRNA stability and synthesis of the receptor. Hum Mol Genet 2003; 12: 205–216.
Engelender S, Sharp AH, Colomer V, Tokito MK, Lanahan A, Worley P et al. Huntingtin-associated protein 1 (HAP1) interacts with the p150Glued subunit of dynactin. Hum Mol Genet 1997; 6: 2205–2212.
Li Q, Hansen D, Killilea A, Joshi HC, Palazzo RE, Balczon R . Kendrin/pericentrin-B, a centrosome protein with homology to pericentrin that complexes with PCM-1. J Cell Sci 2001; 114 (Part 4): 797–809.
Ansley SJ, Badano JL, Blacque OE, Hill J, Hoskins BE, Leitch CC et al. Basal body dysfunction is a likely cause of pleiotropic Bardet-Biedl syndrome. Nature 2003; 425: 628–633.
Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, Semple CA et al. Disruption of two novel genes by a translocation co-segregating with schizophrenia. Hum Mol Genet 2000; 9: 1415–1423.
Andrieux A, Salin PA, Vernet M, Kujala P, Baratier J, Gory-Faure S et al. The suppression of brain cold-stable microtubules in mice induces synaptic defects associated with neuroleptic-sensitive behavioral disorders. Genes Dev 2002; 16: 2350–2364.
Li X-J, Sharp AH, Li S-H, Dawson TM, Snyder SH, Ross CA . Huntingtin-associated protein (HAP1): discrete neuronal localizations in the brain resemble those of neuronal nitric oxide synthase. PNAS 1996; 93: 4839–4844.
DiFiglia M, Sapp E, Chase K, Schwarz C, Meloni A, Young C et al. Huntingtin is a cytoplasmic protein associated with vesicles in human and rat brain neurons. Neuron 1995; 14: 1075–1081.
Trottier Y, Devys D, Imbert G, Saudou F, An I, Lutz Y et al. Cellular localization of the Huntington's disease protein and discrimination of the normal and mutated form. Nat Genet 1995; 10: 104–110.
Sharp AH, Loev SJ, Schilling G, Li SH, Li XJ, Bao J et al. Widespread expression of Huntington's disease gene (IT15) protein product. Neuron 1995; 14: 1065–1074.
Shiwach R . Psychopathology in Huntington's disease patients. Acta –Psychiatr Scand 1994; 90: 241–246 *LHM: This title is currently taken by UCL: Boldero & Neurology ISSN: 0001-0690X.
Tsuang D, DiGiacomo L, Lipe H, Bird TD . Familial aggregation of schizophrenia-like symptoms in Huntington's disease. Am J Med Genet 1998; 81: 323–327 issn: 0148–7299.
Blackwood DH, Fordyce A, Walker MT, St Clair DM, Porteous DJ, Muir WJ . Schizophrenia and affective disorders—cosegregation with a translocation at chromosome 1q42 that directly disrupts brain-expressed genes: clinical and P300 findings in a family. Am J Hum Genet 2001; 69: 428–433.
Hodgkinson CA, Goldman D, Jaeger J, Persaud S, Kane JM, Lipsky RH et al. Disrupted in schizophrenia 1 (DISC1): association with schizophrenia, schizoaffective disorder, and bipolar disorder. Am J Hum Genet 2004; 75: 862–872.
St Clair D, Blackwood D, Muir W, Carothers A, Walker M, Spowart G et al. Association within a family of a balanced autosomal translocation with major mental illness. Lancet 1990; 336: 13–16.
Hennah W, Varilo T, Kestila M, Paunio T, Arajarvi R, Haukka J et al. Haplotype transmission analysis provides evidence of association for DISC1 to schizophrenia and suggests sex-dependent effects. Hum Mol Genet 2003; 12: 3151–3159.
Millar JK, James R, Brandon NJ, Thomson PA . DISC1 and DISC2: discovering and dissecting molecular mechanisms underlying psychiatric illness. Ann Med 2004; 36: 367–378.
Ozeki Y, Tomoda T, Kleiderlein J, Kamiya A, Bord L, Fujii K et al. Disrupted-in-schizophrenia-1 (DISC-1): mutant truncation prevents binding to NudE-like (NUDEL) and inhibits neurite outgrowth. Proc Natl Acad Sci USA 2003; 100: 289–294.
Miyoshi K, Honda A, Baba K, Taniguchi M, Oono K, Fujita T et al. Disrupted-in-schizophrenia 1, a candidate gene for schizophrenia, participates in neurite outgrowth. Mol Psychiatry 2003; 8: 685–694.
Brandon NJ, Handford EJ, Schurov I, Rain JC, Pelling M, Duran-Jimeniz B et al. Disrupted in schizophrenia 1 and Nudel form a neurodevelopmentally regulated protein complex: implications for schizophrenia and other major neurological disorders. Mol Cell Neurosci 2004; 25: 42–55.
Fatemi SH, Kroll JL, Stary JM . Altered levels of Reelin and its isoforms in schizophrenia and mood disorders. Neuroreport 2001; 12: 3209–3215.
Impagnatiello F, Guidotti AR, Pesold C, Dwivedi Y, Caruncho H, Pisu MG et al. A decrease of reelin expression as a putative vulnerability factor in schizophrenia. PNAS 1998; 95: 15718–15723.
Ikeda M, Iwata N, Suzuki T, Kitajima T, Yamanouchi Y, Kinoshita Y et al. Association of AKT1 with schizophrenia confirmed in a Japanese population. Biol Psychiatry 2004; 56: 698–700.
Bhat RV, Budd SL . GSK3b signalling: casting a wide net in Alzheimer's disease. Neurosignals 2002; 11: 251–261.
This work was funded by the Neuroscience Research Charitable Trust.
Supplementary Information accompanies the paper on the Molecular Psychiatry website
About this article
Cite this article
Datta, S., McQuillin, A., Rizig, M. et al. A threonine to isoleucine missense mutation in the pericentriolar material 1 gene is strongly associated with schizophrenia. Mol Psychiatry 15, 615–628 (2010) doi:10.1038/mp.2008.128
EMBO reports (2019)
Human Mutation (2019)
Schizophrenia Research (2018)
Autosomal genetic variation is associated with DNA methylation in regions variably escaping X-chromosome inactivation
Nature Communications (2018)