Molecular Psychiatry (2006) 11, 847–857 & 2006 Nature Publishing Group All rights reserved 1359-4184/06 $30.00 www.nature.com/mp ORIGINAL ARTICLE Cytogenetic and genetic evidence supports a role for the kainate-type , GRIK4,in schizophrenia and BS Pickard1, MP Malloy1, A Christoforou1, PA Thomson1, KL Evans1, SW Morris1, M Hampson1,3, DJ Porteous1, DHR Blackwood2 and WJ Muir2 1Medical Genetics Section, School of Clinical and Molecular Medicine, Molecular Medicine Centre, University of Edinburgh, Edinburgh, UK and 2Psychiatry, Kennedy Tower, Royal Edinburgh Hospital, University of Edinburgh, Edinburgh, UK

In the search for the biological causes of schizophrenia and bipolar disorder, glutamate neurotransmission has emerged as one of a number of candidate processes and pathways where underlying gene deficits may be present. The analysis of chromosomal rearrangements in individuals diagnosed with neuropsychiatric disorders is an established route to candidate gene identification in both Mendelian and complex disorders. Here we describe a set of disrupted by, or proximal to, chromosomal breakpoints (2p12, 2q31.3, 2q21.2, 11q23.3 and 11q24.2) in a patient where chronic schizophrenia coexists with mild learning disability (US: mental retardation). Of these disrupted genes, the most promising candidate is a member of the kainate-type ionotropic glutamate receptor family, GRIK4 (KA1). A subsequent systematic case–control association study on GRIK4 assessed its contribution to psychiatric illness in the karyotypically normal population. This identified two discrete regions of disease risk within the GRIK4 locus: three single single nucleotide polymorphism (SNP) markers with a correspond- ing underlying haplotype associated with susceptibility to schizophrenia (P = 0.0005, odds ratio (OR) of 1.453, 95% CI 1.182–1.787) and two single SNP markers and a haplotype associated with a protective effect against bipolar disorder (P = 0.0002, OR of 0.624, 95% CI 0.485–0.802). After permutation analysis to correct for multiple testing, schizophrenia and bipolar disorder haplotypes remained significant (P = 0.0430, s.e. 0.0064 and P = 0.0190, s.e. 0.0043, respectively). We propose that these convergent cytogenetic and genetic findings provide molecular evidence for common aetiologies for different psychiatric conditions and further support the ‘glutamate hypothesis’ of psychotic illness. Molecular Psychiatry (2006) 11, 847–857. doi:10.1038/sj.mp.4001867; published online 4 July 2006 Keywords: association study; chromosome abnormality; glutamate receptor; haplotype; psy- chiatry

Introduction phrenia and bipolar disorder has come from linkage and case–control association studies on karyotypi- Both de novo and familial chromosome abnormalities cally normal populations8–12 and from our recent have been described in patients with psychiatric or demonstration of its activity-dependent interaction neurological conditions.1,2 A small proportion of with the protein encoded by an independently these have been studied at the molecular level.3 We disrupted gene, PDE4B.13 Through the analysis of have previously described a large Scottish family in further cytogenetic abnormalities in schizophrenic which there is a highly significant co-segregation patients, we and others have identified NPAS3, (LOD > 7)4 between major mental illness and the GRIA3 and DIBD1 as potential candidate genes.14–17 presence of a balanced translocation between chro- Here we describe a patient with chronic schizo- mosomes 1 and 11.5 The DISC1 gene is directly phrenia and mild learning disability (US: mental disrupted by the chromosome 1 breakpoint6,7 and retardation). This co-morbid state has been previously supportive evidence for its involvement with schizo- shown to be highly familial, associated with changes in brain structure on imaging typical of schizophrenia Correspondence: Dr BS Pickard, Medical Genetics Section, alone, and also associated with increased prevalences School of Clinical and Molecular Medicine, Molecular Medicine of chromosomal abnormalities.18,19 The patient Centre, University of Edinburgh, Edinburgh EH4 2XU, UK. possesses a complex chromosomal rearrangement E-mail: [email protected] 3 involving several chromosomes. We hypothesized Current address: Department of Advanced Biotechnologies, that the patient’s diagnosis was as a direct result of Janssen Pharmaceutica NV, Beerse B-2340, Belgium. Received 3 February 2006; revised 4 May 2006; accepted 5 June gene deficits caused by the cytogenetic abnorma- 2006; published online 4 July 2006 lity. Therefore, we have positioned the constituent GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 848 chromosomal breakpoints and identified a number of The patient was not dysmorphic and was of average potential candidate genes in their vicinity. One of stature. During childhood, repeated mastoid infec- these, GRIK4, is directly disrupted at a breakpoint and tions and corrective surgery led to conduction deaf- encodes a member of the ionotropic glutamate ness in one ear and partial conduction deafness in the receptor family – an established focus for candidate other. Hearing-aid correction was excellent and meant disease gene hunting because of the hypothesis that that her communication skills were fully intact. defects in glutamate neurotransmission may underlie Although there is an older literature suggesting a link schizophrenia. between some cases of schizophrenia (especially late- As the most promising candidate, we wished to onset paraphrenia) and chronic deafness, more recent determine the relevance of GRIK4 to psychiatric analysis suggests that there is little evidence of any illness beyond this unique patient. Therefore, a large true association.22 During her 60s, the patient devel- case–control association study was undertaken, oped sero-negative rheumatoid arthritis, principally which demonstrated that single nucleotide poly- affecting her distal joints. morphism (SNP) markers and haplotypes within the GRIK4 locus are strongly associated with both Cell culture schizophrenia and bipolar disorder. The patient was karyotyped using conventional Giemsa, G-band, staining. Lymphocytes were ex- Materials and methods tracted from 7 ml of patient venous blood (for storage and generation of Epstein–Barr virus-transformed cell Clinical findings lines) using density gradient separation (Histopaque- A female patient was identified as part of a survey of 1077, Sigma, Gillingham, UK). The resulting lympho- patients with mild mental retardation (UK term – blastoid cell line failed to thrive, perhaps owing to the learning disability) and coexisting schizophrenia.18 patient age and medication, such that limited cell There was no evidence for any previous karyotype material was produced. However, metaphase-arrested investigation before this ascertainment. Full, in- chromosomes for cytogenetic analysis were success- formed consent for the current study was obtained fully prepared from 0.8 ml of patient blood cultured from the subject. for 71 h in medium containing phytohaemagglutinin In her mid 60s when examined, the patient gave a (Peripheral Blood Medium, Sigma). These short-term near 30-year history of severe chronic schizophrenia cultures were treated with colcemid for 1 h followed requiring repeated admissions to a hospital specializ- by a conventional cell membrane lysis and methanol/ ing in the care of people with mental retardation and acetic acid fixing procedure. Fixed chromosomes concurrent psychiatric or behavioural disorders. In- were dropped onto microscope slides and stored for itially in mainstream schooling, from the age of 8 she 1 week before use in fluorescence in situ hybridiza- received education in an establishment for those with tion (FISH) experiments. special educational needs – she had a mild degree of intellectual impairment (IQ 65–70). Her early psy- Selection of YAC/BAC and cosmid clones for FISH chiatric history is unclear and, although psychosis probe synthesis was only formally documented from her early 40s, Clones were initially selected from the Whitehead/ there is evidence of earlier onset. The schizophrenic MIT map of the relevant chromosome in the cyto- illness showed the same features during all episodes genetic intervals to which the G-band defined break- with clear and persistent auditory hallucinations and points were mapped. Yeast artificial chromosomes persecutory delusions. Treatment was initially with (YACs) were obtained from the HGMP Resource standard typical anti-psychotic medications and, Centre, Babraham Bioincubator, Babraham, Cam- latterly, with atypical anti-psychotics owing to the bridge, UK (http://www.geneservice.co.uk). Clone onset of persistent extrapyramidal side effects. Be- DNA was prepared by standard methods and PCR tween episodes resolution was good and she regained amplified using primers designed against consensus full insight. The diagnosis of chronic schizophrenia sequence elements within the archetypal Alu repeat.23 was confirmed using SADS-L structured inter- This ‘Alu-PCR’ gives a representative spread of non- view,20,21 to generate DSM-IV and ICD-10 criteria, by repetitive sequence over the full length of the YAC a psychiatrist experienced in both general psychiatry and generates a better FISH probe than native YAC and the psychiatry of mental retardation (WM). DNA. Alu-PCR was performed using the Expand Long Consensus diagnosis was reached on review by two Template PCR kit (Roche, Lewes, UK). Cycling psychiatrists (WM and DB). IQ scores were generated conditions were as follows: 941C – 45 s, 551C – 30 s, from WAIS-R and their stability shown by similar 681C – 8 min, 35 cycles; 681C À10 min final extension. levels detected by psychological examination at Bacterial artificial chromosome (BAC) clones corre- different times throughout her life. No other family sponding to positive YAC regions were arranged into member agreed to participate in the study or to be contigs by consulting the Washington University karyotyped. Clear family history information is not FPC (http://www.agcol.arizona.edu/fpc/human/) and available. However, as far as can be ascertained, there UCSC Human Genome Browser (http://genome. were no first-degree relatives known with schizo- cse.ucsc.edu/cgi-bin/hgGateway) databases. BAC phrenia or mental retardation. clones were supplied by BACPAC Resources,

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 849 Oakland, CA, USA (http://www.chori.org/bacpac/). the transcribed gene) were defined using Haploview’s Clone selection was biased to gene-containing BACs. ‘solid spine of LD’ method with D0 values of greater Once a breakpoint-spanning BAC was identified, the than 0.8. Tagging SNPs were then selected both by precise position of the breakpoint in relation to their ability to represent haplotype diversity within candidate gene exons was determined by FISH probes these blocks down to the 10% haplotype frequency generated from cosmids isolated from gridded chro- level and also their predicted ability to form reliable mosome-specific libraries (HGMP Resource Centre, assays on the Illumina Inc., San Diego, CA, USA see below for primers used to amplify GRIK4 exon genotyping platform. Twenty-seven such SNPs were regions for use as radioactive probes in cosmid library chosen over a span of 344 kb to cover the GRIK4 gene screens) or precisely positioned, repeat element-free, (326 kb). These SNPs were typed at Illumina Inc., San long-range PCR products (Expand long range PCR kit, Diego, CA, USA using their proprietorial bead-array Roche; see below for primer sequences). Cycling technology on 368 bipolar disorder cases and 386 conditions were as follows: 941C – 45 s, 521C – 30 s, schizophrenia cases comprised of patients of hospi- 681C – 11 min, 35 cycles; 681C – 15 min final tals in southeast and south central Scotland, and 458 extension. controls of matching geographical distribution from the Scottish National Blood Transfusion Service: a FISH protocol cohort described in more detail elsewhere.25 Analyses Probe template DNA (pooled YAC Alu-PCR products, by our own and other groups on the Scottish BAC clone DNA, cosmid clone DNA or long-range population suggest that it is generally stable and PCR products) was labelled by nick translation and homogeneous, with no obvious evidence for sub- hybridized to patient metaphase spreads using stan- structure. dard FISH methods. Slides were counterstained with 4,6-diamidino-2-phenylindole in Vectashield anti- Statistical analysis fade solution (Vector Laboratories, Peterborough, Formatted genotype data were analysed using Coca- UK). A Zeiss Axioskop2 fluorescence microscope phase26 (http://portal.litbio.org/Registered/Option/ with a chroma number 81000 multi-spectral filter unphased.html) in order to derive P-values for both set was used in conjunction with CoolSnap HQ single marker associations and also two-, three- and camera (Photometrics). Images were captured using four-marker global and individual haplotype tests – a SmartCapture2 (Digital Scientific, UK). FISH signals surrogate for linkage disequilibrium (LD) blocks in observed on derived chromosomes dictated the our population. selection of further clones required to ‘walk’ towards, Cocaphase uses the EM algorithm to estimate the and eventually cross, the breakpoint. haplotype frequencies of unphased genotype data and standard unconditional logistic regression analysis, PCR primers applying the likelihood ratio test under a log-linear Long-range PCR for FISH probe templates: model to compare haplotype frequencies between cases and controls. In order to avoid misleading 50 ITGA4a: TGCTCTTCAAAACCAGTCAGAA, results caused by rare haplotypes, all haplotypes with 50 ITGA4b: CCCCTTGGGATTATCCTATCTC, a frequency less than or equal to 1% in both cases and 30 ITGA4a: CTCATTTGACGGAGATTTTTC, controls were declared as rare and clumped together 30 ITGA4b: GTGATGGTGAGGAATGTTCTGA, for the test of the null hypothesis using the command line option ‘-rare 0.01’. P-values for both global and GRIK4 exon region-specific PCR: screening of individual tests of haplotype frequencies were deter- cosmid libraries: mined. The global test P-value assesses the signifi- cance of the overall difference in the distribution of Ex1 a: GCAGAGTTATGTCATGCCCA, haplotype frequencies between cases and controls. Ex1 b: CCTGTGCAGCACTCTCATGT, The P-value from the individual test represents the Ex2/3 a: TTGAACCCAAGAGAACAGGG, significance of the difference in frequency of an Ex2/3 b: TCCCCTTCTCCTTCCAGTTT individual haplotype between cases and controls. Cycling conditions were as follows: 941C – 2 min In order to account for the multiple SNPs and initial denaturation; 941C – 1 min, 521C – 1 min, 721C haplotypes tested, permutation analysis (1000 per- – 75 s, 33 cycles; 721C – 15 min final extension. mutations) was performed also using Cocaphase. Cocaphase randomly reassigns the diagnosis labels Association study (case vs control) of the individuals. All single markers SNP marker selection was achieved by analysing or haplotypes of a specified window size are then the public International HAPMAP Project (http:// tested and the most significant P-value from each www.hapmap.org/index.html.en) genotyping data permutation is stored. Based on the distribution of (release #7) from the CEU population (Utah residents these stored P-values, an ‘experimentwise’ signifi- with ancestry from northern and western Europe) cance level27 is provided for the most significant P- in the Haploview application (v.2.5) (http://www. value observed for the particular sliding-window size broad.mit.edu/mpg/haploview/index.php).24 LD blocks (i.e. one to four). Odds ratios (ORs) were calculated across the GRIK4 locus (including regions 50 and 30 of online (http://www.hutchon.net/confidor.htm).

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 850 Results coupled with a modifed translocation event between chromosomes 2 and 11 (Figure 1). In addition, a Characterization of loci affected by the karyotype segment of 11q adjacent to the translocation has been rearrangement transferred onto chromosome 8. A representative The chromosomal rearrangement was discovered on image from a BAC FISH probe crossing the breakpoint fine analysis to be more complex than the description on chromosomes 11 is shown in Figure 2. FISH in the original study18 (see legend to Figure 1). It is analysis positioned five breakpoints. Table 1 details likely that the cytogenetic abnormality originates the YAC and BAC probes that define the breakpoint from a pericentric inversion of chromosome 2 positions and Figure 3 shows the gene content in their

Figure 1 Representation of complex chromosomal rearran- Figure 2 FISH hybridization. The red signals (biotinylated gement in the translocation patient. The pericentric probe) correspond to whole chromosome 8 paint. The green chromosome 2 inversion is coupled with a reciprocal signals (digoxigeninylated probe) indicate hybridization of translocation with chromosome 11. The chromosome 11 a probe derived from CEPH YAC 936_d_9. This YAC probe region between the 11q23.3 and 11q24.3 breakpoints has hybridizes to normal 11q23.3, the derived chromosome 11 inserted on chromosome 8q13. The full karyotype of and the derived chromosome 8. This split signal indicates this rearrangement is (46, XX, ins(8;11)(q13;q23.3q24.2) that the YAC spans the breakpoint in the region of the inv(2)(p12q32.1) t(2;11)(q21.3;q24.2) der(2)(2qter- GRIK4 gene. A BAC RP11-89p5-derived probe and cosmid- 2q32.1::2p12-2q21.3::11q24.2-11qter) der(11)(11pter- derived probes (data not shown) gave FISH results indicat- 11q23.3::2q21.3-2q32.1::2p12-2pter) der(8)(8pter- ing that the breakpoint lies within the body of the GRIK4 8q13::11q23.3-11q24.2::8q13-8qter). gene (Figures 3 and 4).

Table 1 FISH probes used in the definition of the breakpoints in the translocation patient

Patient Cytogenetic position Breakpoint YAC clones Breakpoint BAC clones (acc. no.)

1 2p12 915_f_7 RP11-9o10 (AC104775)–RP11-113b11 (AQ341814/AQ341812) 1 2q31.3 941_h_12 RP11-358m9 (AC020595) 1 2q21.2 766_c_12 RP11-250h22 (AC011996) 1 11q23.3 936_d_9 RP11-89p5 (AZ517840/AQ284967) 1 11q24.2 749_d_2 RP11-687M24 (AP001007)–RP11-100p11 (AP003061)

Abbreviations: BAC, bacterial artificial chromosome; YAC, yeast artificial chromosome. CEPH library YACs spanning the breakpoints are listed. Also detailed are the BAC clones (and accession numbers) from the RPCI-11 BAC library that span or flank (indicated by en-dashes) the breakpoints. Breakpoints at 8q13 were not characterized in detail in this study.

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 851

Figure 3 Genomic arrangement of the genes within 500 kb of each chromosomal breakpoint in the translocation patient. To the right of the figure is a block listing the cytogenetic bands at which breakpoints were localized. The horizontal lines to the left of the block represent 1 Mb stretches of genomic DNA centred on the position of the breakpoints. The resolution accuracy of each breakpoint position, as determined by FISH, is indicated by the length of the black bars superimposed on the chromosome stretches. Rectangular boxes indicate the location, genomic extent and transcriptional orientation of each gene: those situated above the chromosome line are transcribed from left to right, and those below from right to left. Shading within the rectangles indicates a direct gene disruption. vicinity. The relatively recent annotation of sequences this gene is probably incomplete; further 50 exons from the Human Genome Project has meant that our extending to a CpG island are present in several ESTs understanding of the precise relationship between (e.g. BC045626). The putative complete gene se- breakpoint and transcriptional units has changed quence is disrupted by the breakpoint resulting in over time: the data presented here are a combination the loss of one functional allele in this patient. of the May 2004 freeze of the UCSC Human PKNOX2/PREP2 encodes a member of the MEINOX- Genome Browser (http://genome.cse.ucsc.edu/cgi-bin/ TALE class of homeodomain-containing proteins that hgGateway) and in-house annotation through the heterodimerizes with PBX1A to form a functional collation of expressed sequence tags (ESTs) and transcriptional regulator. Little is known about the incomplete cDNA sequences from multiple species. function of this gene product other than that it is We were not able to characterize in detail the region of expressed widely with the highest levels observed in 8q13 where the segment of chromosome 11q23.3– the brain, pancreas and lung.31 The integrin A4 cell q24.2 has inserted owing to a limited supply of adhesion molecule (ITGA4, acc. L12002 and X16983) patient chromosome material. The breakpoint at 2p12 and the candidate retinitis pigmentosa gene, RP26/ lies within a gene-poor region (3 Mb in size) and was CERKL (acc. NM_201548), are both located at the not pursued beyond the resolution of YAC clones. 2q31.3 breakpoint. The breakpoint-spanning BAC The breakpoint at 2q21.2 disrupts a potentially RP11-358m9 contains both genes but long-range aberrant fusion transcript, which appears to be a PCR-based FISH probes corresponding to the 50 and non-genic LINE1 transcript intergenically spliced to a 30 ends of ITGA4 (see Materials and methods for genuine, CpG island-associated gene, NAP5 (NCK- primer sequences) indicated that the breakpoint must Associated Protein). The functional NAP5 gene has lie within a 50 kb window downstream of its coding been described in its full-length form only in the sequence. This region consists entirely of RP26/ mouse (NM_172484), but the cognate human ortho- CERKL gene sequence, indicating that this gene was logue is not disrupted, lying centromeric to the also directly disrupted. A recessive nonsense muta- breakpoint. tion in the RP26/CERKL gene has been described in The 11q23.3 breakpoint lies within the high-affinity two unrelated Spanish families with retinitis pigmen- kainate ionotropic glutamate receptor, GRIK4, locus tosa,32 indicating that disruption and the resulting (acc. S67803 and NM_014619, alternative nomencla- haploinsufficiency of this gene is unlikely to explain ture KA1/EAA1). The function of this gene and the the clinical phenotype of the patient. The direct gene receptor it encodes are described in disruptions of the PKNOX2/PREP2 and RP26/CERKL more detail in the Discussion section. FISH with genes are not discussed further in this paper. cosmid-derived probes (Figure 4) positioned the breakpoint between exons 2 and 3 of GRIK4. The Association expected outcome of the position of the breakpoint is Of the several genes disrupted or potentially deregu- the truncation of all putative transcript forms such lated in this subject, GRIK4 is the outstanding that a functional receptor cannot be encoded from the candidate to explain the observed symptomatology. derived chromosome 11. PKNOX2/PREP2, encoding a This is owing to its neuronal expression pattern, a transcription factor, is located at the 11q24.2 break- previous report of a disrupted glutamate receptor point.28–30 The reference sequence (NM_022062) for in a psychiatric condition16 and the hypothesized

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 852

Figure 4 The GRIK4 gene: annotation pertaining to the cytogenetics and association studies. Exons are represented by numbered vertical lines. The breakpoint position is marked by a longer vertical line. The large rectangle at the top signifies the breakpoint-spanning BAC clone as determined by FISH. Labelled arrows indicate the location of cosmid clones used in the positioning of the breakpoint. The direction of the arrows represents the direction towards the breakpoint as interpreted from FISH images using probes derived from these clones. The 27 SNPs typed in the association study are shown together with the extended haplotype regions associated with schizophrenia and bipolar disorder. Filled triangles indicate significant single marker association with disease. Core (filled rectangles) and extended (bracketed lines) haplotypes associated with both schizophrenia and bipolar disorder are shown beneath the individual markers. The three rows at the bottom of the figure show LD blocks calculated from HapMap genotyping data processed in the Haploview application set at three defined stringency levels of multiallelic D0 before adjacent blocks were joined. A scale bar is shown.

involvement of aberrant glutamate neurotransmission of combined diagnoses, resulted in less significant in susceptibility to psychiatric illness (see Discussion). P-values at both regions when compared to the tests In order to assess the contribution of the GRIK4 of the individual diagnoses, suggesting that different gene to psychiatric illness in the karyotypically regions within the GRIK4 locus confer disease- normal population, we undertook a case–control specific risk/protective effects for the two disorders. association study employing 27 SNP markers selected We examined the associations in further detail by across the gene for their ability to ‘tag’ common assessing the contribution of individual haplotypes haplotype variation (see Materials and methods). to the global associations across the entire gene Analysis of genotyping data from schizophrenia (Table 3). An individual haplotype consisting of three cases, bipolar disorder cases and controls was carried SNPs (CTT, SNPs 15–17) was responsible for the out using Cocaphase. At the level of individual most significant association with schizophrenia markers (Table 2), two principal regions of the gene (P = 0.0005, est. freq. cases: 0.38, est. freq. controls: demonstrated allele frequency changes in the 0.29, OR of 1.453 (95% CI 1.182–1.787)). Although case groups: SNPs 15, 16 and 17 showed signi- this three-SNP haplotype just failed to remain ficant association with schizophrenia (P = 0.001, significant after permutation correction, the compo- 0.004 and 0.015, respectively) and SNPs 23 nent two-SNP haplotype (TT, SNPs 16 and 17; and 26 were associated with bipolar disorder P = 0.0006) did remain significant (P = 0.0430, s.e. (P = 0.003 and 0.020, respectively). Of these, SNP 15 0.0064; Table 3) after such correction. An underlying remained significantly associated with schizophrenia extended haplotype consisting of nine SNPs (P = 0.030, s.e. 0.005; Table 2) after permutation (TCCTTGTGA, SNPs 13–21) appeared upon align- analysis to counter multiple testing errors (see ment of the local individual haplotypes: this haplo- Materials and methods). These positionally bi-partite type conferred susceptibility to schizophrenia. For findings (Table 2) were further supported by a sliding bipolar disorder, a two-SNP haplotype (GC, SNPs 26 window global analysis of haplotypes of lengths and 27) showed the greatest significance (P = 0.0002, two to four, such that distinct clusters of significant est. freq. cases: 0.16, est. freq. controls: 0.23, OR P-values were observed in both schizophrenia (best of 0.624 (95% CI 0.485–0.802)). This haplotype global haplotype, SNPs 15 and 16; P = 0.005) and remained significant after permutation correction bipolar disorder (best global haplotype, SNPs 26 and (P = 0.0190, s.e. 0.0043; Table 3). Again, overlapping 27; P = 0.002). The latter global haplotype approached individual haplotype results in this region suggested significance after permutation analysis (P = 0.049, s.e. that this was a component of an underlying seven- 0.007; Table 2). Additional significant global P-values SNP haplotype (ATCATGC, SNPs 21–27), in this case were observed outside of the two regions, but without conferring a protective effect against bipolar disorder. clear clustering between overlapping windows. Table In addition, it has been shown previously that a 2 also shows that tests of the case group, consisting genuinely significant individual haplotype can exist

Molecular Psychiatry Table 2 Single SNP marker and sliding window global haplotype (2–4 SNP markers)

Marker SNP Genome Allele Allele Freq. Freq. Global P-values: all cases vs Global P-values: SCZ cases vs Global P-values: BPD cases vs identity location 1 2 1 2 controls controls controls

Single 2-SNP 3-SNP 4-SNP Single 2-SNP 3-SNP 4-SNP Single 2-SNP 3-SNP 4-SNP

1 rs4245040 120058281 C T 0.68 0.32 0.250 0.089 0.288 0.221 0.390 0.364 0.704 0.484 0.269 0.050 0.187 0.158 2 rs7928347 120073545 C T 0.80 0.20 0.382 0.583 0.430 0.306 0.279 0.735 0.612 0.187 0.693 0.559 0.486 0.601 3 rs2000869 120082523 A G 0.66 0.34 0.677 0.448 0.391 0.483 0.984 0.538 0.466 0.365 0.484 0.421 0.498 0.246 4 rs7479743 120091693 C G 0.57 0.43 0.621 0.779 0.718 0.712 0.842 0.870 0.597 0.757 0.286 0.538 0.239 0.094 5 rs6589832 120097132 A G 0.52 0.48 0.833 0.913 0.800 0.831 0.600 0.474 0.837 0.833 0.365 0.480 0.038 0.125 6 rs7925270 120100950 C T 0.55 0.45 0.695 0.950 0.876 0.860 0.859 0.952 0.886 0.984 0.387 0.768 0.795 0.544 7 rs3133855 120122503 C T 0.66 0.34 0.624 0.971 0.885 0.824 0.811 0.985 0.999 0.965 0.549 0.957 0.485 0.594 8 rs3133226 120129688 C T 0.45 0.55 0.904 0.492 0.564 0.597 0.806 0.980 0.840 0.767 0.965 0.179 0.326 0.503 9 rs2187495 120148960 C T 0.76 0.24 0.357 0.495 0.508 0.774 0.864 0.819 0.713 0.827 0.157 0.312 0.459 0.723 10 rs3132780 120157811 A G 0.83 0.17 0.514 0.779 0.543 0.171 0.464 0.619 0.696 0.905 0.708 0.954 0.540 0.022 11 rs948028 120182099 A C 0.85 0.15 0.687 0.383 0.158 0.005 0.504 0.598 0.938 0.043 0.991 0.309 0.011 0.009 12 rs2852217 120184878 C T 0.26 0.74 0.175 0.232 0.039 0.041 0.405 0.825 0.030 0.006 0.136 0.034 0.035 0.120 13 rs879602 120194224 C T 0.13 0.87 0.936 0.949 0.449 0.176 0.677 0.702 0.067 0.028 0.557 0.885 0.892 0.652 14 rs1954787 120201015 C T 0.54 0.46 0.515 0.088 0.287 0.320 0.472 0.008 0.024 0.068 0.701 0.665 0.927 0.562 15 rs4935752 120226771 A C 0.68 0.32 0.022 0.103 0.160 0.128 0.001 0.005 0.015 0.016 0.584 0.706 0.354 0.434 16 rs6589846 120237799 G T 0.43 0.57 0.141 0.115 0.340 0.335 0.004 0.007 0.059 0.055 0.704 0.349 0.506 0.578 disorder Pickard bipolar BS and schizophrenia in GRIK4 17 rs4430518 120261085 G T 0.28 0.72 0.244 0.769 0.855 0.758 0.015 0.140 0.135 0.367 0.653 0.918 0.993 0.904 18 rs433110 120268321 A G 0.43 0.57 0.362 0.606 0.570 0.707 0.084 0.149 0.419 0.305 0.835 0.988 0.688 0.917

19 rs7111184 120308715 G T 0.33 0.67 0.736 0.277 0.173 0.271 0.821 0.363 0.045 0.118 0.725 0.351 0.486 0.489 al et 20 rs2156635 120309320 C G 0.58 0.42 0.041 0.027 0.129 0.117 0.081 0.004 0.040 0.091 0.081 0.269 0.446 0.311 21 rs949056 120317011 A G 0.71 0.29 0.476 0.767 0.364 0.910 0.870 0.972 0.931 0.997 0.162 0.317 0.054 0.382 22 rs949054 120333540 C T 0.20 0.80 0.509 0.093 0.589 0.566 0.894 0.769 0.990 0.945 0.315 0.007 0.124 0.222 23 rs602104 120347035 C T 0.22 0.78 0.028 0.274 0.314 0.340 0.425 0.908 0.781 0.695 0.003 0.056 0.137 0.136 24 rs2276319 120375473 A G 0.61 0.39 0.675 0.472 0.186 0.113 0.940 0.643 0.496 0.781 0.434 0.402 0.039 0.015 25 rs513548 120385469 C T 0.22 0.78 0.786 0.093 0.054 0.680 0.288 0.410 0.375 0.030 0.010 26 rs2282586 120396062 A G 0.37 0.63 0.058 0.048 0.358 0.708 0.020 0.002 27 rs1944522 120402574 C T 0.54 0.46 0.405 0.842 0.216

Best global P-value (permutation corrected) 0.403 0.420 0.550 0.130 0.030 0.072 0.250 0.168 0.066 0.049 0.207 0.240 Associated standard error 0.016 0.016 0.016 0.011 0.005 0.008 0.014 0.012 0.008 0.007 0.013 0.014

Abbreviations: BPD, bipolar disorder; SCZ, schizophrenia; SNP, single nucleotide polymorphism. P-values from the case–control association study (those less than 0.05 are boxed and are in bold). Details of the identity, location, allele forms and control group allele frequencies of the 27 SNPs used in the study are shown on the left-hand side of the table. P-values for the single-marker and global sliding window SNP analyses are shown for the three case–control groupings: all cases vs controls, SCZ cases vs controls and BPD cases vs controls. Permutation corrected P-values (with standard error) are shown at the bottom for the best uncorrected P-value in the column above. oeua Psychiatry Molecular 853 GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 854 Table 3 Key individual haplotypes associated with SCZ (top) and BPD (bottom)

Schizophrenia vs. Controls SNP markers 1 2 3456789101112131415161718192021222324252627p-valueEffectCorrectionS.E. CC 0.0094 susc. CT 0.0009 susc. TT 0.0006 susc. 0.0430 0.0064 GA 0.0022 susc. TCC 0.0090 susc. CCT 0.0055 susc. CTT 0.0005 susc. 0.0589 0.0074 TTG 0.0020 susc. TTCC 0.0094 susc. TCCT 0.0069 susc. CCTT 0.0050 susc. CTTG 0.0009 susc. 0.1818 0.0122 GATT 0.0048 susc. Assembled haplotype TCCTTGTGA

Bipolar Disorder vs. Controls SNP markers 1 2 3456789101112131415161718192021222324252627p-valueEffectCorrectionS.E. TC 0.0028 prot. CA 0.0064 prot. TG 0.0037 prot. GC0.0002 prot. 0.0190 0.0043 ATC 0.0043 prot. TCA 0.0071 prot. ATG 0.0040 prot. TGC0.0021 prot. ATCA 0.0098 prot. CATG 0.0080 prot. ATGC0.0035 prot. GCT 0.0018 prot. 0.2258 0.0132 CGCT 0.0080 prot. GCTC 0.0018 prot. 0.3187 0.0147 Assembled haplotypes CGCTC A TCA TGC

Abbreviations: BPD, bipolar disorder; SCZ, schizophrenia; SNP, single nucleotide polymorphism. The P-value of each haplotype is shown to the immediate right of the tables (values less than 0.001 are boxed). The effect on disease status (prot. – protective, susc. – susceptibility) is also shown for each haplotype. Permutation corrected P-values and standard errors are shown next to the best uncorrected two-, three- and four-marker individual haplotypes for both SCZ and BPD. Larger compound haplotypes assembled from inspection of individual haplotypes are shown underneath each section in bold (the SCZ susceptibility haplotype includes a T at SNP 19 deduced from a number of individual haplotypes not shown but with P-values between 0.05 and 0.01).

in the absence of a significant global test.33 We rule out other mechanisms. We have not yet examined identified several consistent individual haplotypes these genes at the molecular or genotyping level and between markers 4 and 8, which were significantly so these alternative explanations remain uninvesti- associated with a protective effect against bipolar gated in this patient. disorder despite this region not appearing in the GRIK4 lies at the edge of a schizophrenia linkage global analysis. region described in a recent publication39 with one microsatellite marker, D11S925, located within an intron at the 30 end of the gene. In contrast, a recent Discussion meta-analysis of several bipolar disorder linkage Five chromosomal breakpoints have been positioned studies failed to observe any significant findings on at the gene resolution level in a patient with chromosome 11.40 However, a conventional linkage schizophrenia and mental retardation. We have study approach would be unlikely to discover loci described the disruption of the gene, GRIK4 (KA1/ with protective effects. Recently, an association study EAA1), encoding the high-affinity subunit of the was carried out on GRIK4 in Japanese schizophrenia kainate-type ionotropic glutamate receptor.34,35 In this patients.41 The study failed to identify any statisti- patient, we believe that haploinsufficiency of this cally significant single SNP or haplotype associations gene is most likely responsible for the psychiatric despite good marker coverage. Explanations for this component of the patient’s diagnosis although the might be the modest size of the study (100 cases, 100 existence of several other disrupted genes in this controls) and the different population studied. patient, the potential for position effects on genes at a Beyond the bona fide gene mutation in the distance from the breakpoints36,37 and the possibility translocation patient, the most compelling evidence of cryptic deletions associated with the chromosomal we describe here in support of a role for GRIK4 in rearrangement38 mean that we cannot definitively schizophrenia and bipolar disorder comes from our

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 855 findings from a comprehensive case–control associa- been identified on the basis of their pharmacological tion study in Scottish population. Interestingly, two profiles and sequence homologies: NMDA receptors, physically separated haplotype regions within the AMPA receptors, kainate receptors and delta recep- gene showed independent and significant association tors. Intriguingly, with the inclusion of this study, with schizophrenia and bipolar disorder, respectively. examples from each class have now been associated This finding mirrors the repeated observation of with the pathophysiology of psychiatric illness. overlapping linkage ‘hotspots’ for these two condi- NMDAR1/Grin1 knockdown mice display compo- tions.42 The OR values detailed above for the schizo- nents of schizophrenia-like behaviour,53 the AMPA phrenia and bipolar disorder core individual receptor GRIA3 has been shown to be disrupted in a haplotypes reflect their high frequency in the popula- patient diagnosed with bipolar disorder co-morbid tion (0.38 and 0.23, respectively) and are of a with mental retardation16 and a breakpoint has been magnitude entirely consistent with previously de- discovered adjacent to a delta-type receptor (Pickard scribed complex disease-susceptibility variants.43 et al., manuscript in preparation). Although the corrected P-values are only moderately GRIK4 is expressed in the amygdala, hippocampal significant, we believe they should be taken in the formation (CA3 pyramidal and dentate granule cells) context of the prior probability conferred on the and entorhinal cortex,54,55 a pattern matching brain GRIK4 gene through the cytogenetic study. Replicat- regions implicated in the psychoses.56 In addition, the ing these findings in additional European sample sets potential role of long-term potentiation (LTP; the should be an important target for future genetic molecular model now generally accepted to underlie studies. observed changes in activity-dependent synaptic The schizophrenia risk region lies primarily over plasticity57–59) in psychiatric illness has also been exons encoding the N-terminal extracellular domain, widely hypothesized. This is as a result of the well- whereas the bipolar disorder protective region lies documented deficits in particular cognitive tasks over exons encoding the cytoplasmic C-terminus of involving learning and memory in individuals with the protein, suggesting that the, as yet undiscovered, schizophrenia, bipolar disorder and unipolar major underlying causative mutations for these two regions depression.60–62 Presynaptic changes in kainate re- could act through different aspects of GRIK4 function ceptor-dependent plasticity have been described at or regulation. Further analysis (data not shown) failed hipocampal mossy fibre synapses,63,64 suggesting that to reveal any effect of the protective haplotype on the expected reduction in GRIK4 protein levels in the diagnostic/phenotypic subclasses of the bipolar dis- karyotypically abnormal patient might be expected to order cohort (e.g. BP1 vs BP2, hallucinations vs modify channel properties, and their delusions). This suggests that the haplotype confers contribution to LTP, by altering subunit stoichiometry. a protective effect on all forms of bipolar disorder. The findings described in this report provide To our knowledge, no previous example has been supportive evidence for a glutamatergic contribution documented where a single gene contains distinct to psychosis. The susceptibility and protective hap- and independent risk and protective haplotypes for lotypes described above present future opportunities two disorders. However, there are examples of for replication studies and for re-sequencing coding complex genetic disorders where all the individual and non-coding portions of GRIK4, bounded by the features of the GRIK4 association findings have been relevant LD blocks, to identify causative mutations. described. Independent CARD15/NOD2 gene muta- The findings also suggest that the study of rare tions can give rise to Blau syndrome44 or Crohn’s chromosomal rearrangements can direct research disease45,46 – distinct disorders with a common towards pertinent biological pathways and processes inflammatory basis. There are also examples where that may be targets for further genetic investigation both risk and protective alleles have been identified and potential therapeutic strategies. in genes such as PPARG in insulin resistance disorders.47 Moreover, in this example, like GRIK4 Acknowledgments and DISC1, the PPARG mutations exist in both rare, severe familial forms and common, low-penetrance We thank Judy Fantes, Maura Walker, Margaret van population-level forms. Beck, Paul Perry, Veronica van Heyningen and With regard to GRIK4 function, abnormalities in Western General Hospital Cytogenetics service for glutamatergic neurotransmission have previously expert assistance and advice. We also wish to been proposed as a mechanism for psychiatric illness. acknowledge the HGMP Resource Centre for the The ‘Glutamate Hypothesis’ of schizophrenia was provision of YAC, Cosmid and IMAGE clones and devised to explain the psychotic symptoms exhibited gridded DNA libraries. This work was supported by a by individuals following administration of ionotropic collaborative research agreement with Merck, Sharp glutamate receptor antagonists such as phencyclidine and Dohme (The Neuroscience Research Centre, (PCP; ‘Angel Dust’) and ketamine.48,49 Several studies Terlings Park, UK), SHERT Grant RG45/01, Wellcome also point to changes, predominantly decreases, in Trust ‘Genes to Cognition’ grant and CSO Grant K/ glutamate receptor subunit expression in the post- MRS/50/C2789. Part of the patient sample collection mortem brains of patients with schizophrenia.50–52 was supported by a grant from the State Hospital for Four classes of ionotropic glutamate receptors have Scotland at Carstairs.

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 856 The authors declare that there are no conflicts of 19 Sanderson TL, Best JJ, Doody GA, Owens DG, Johnstone EC. interest. Neuroanatomy of comorbid schizophrenia and learning disability: a controlled study. Lancet 1999; 354: 1867–1871. 20 Wing JK, Ustun TB, Sartorius N (eds) Diagnosis and Clinical References Measurement in Psychiatry: A Reference Manual for Scan/PSE-10. Cambridge University Press: Cambridge, 1998. 1 MacIntyre DJ, Blackwood DH, Porteous DJ, Pickard BS, Muir WJ. 21 Endicott J, Spitzer RL. A diagnostic interview: the schedule for Chromosomal abnormalities and mental illness. Mol Psychiatry affective disorders and schizophrenia. Arch Gen Psychiat 1978; 2003; 8: 275–287. 35: 837–844. 2 Nielsen J, Wohlert M. Chromosome abnormalities found among 22 Carvill S. Sensory impairments, intellectual disability and 34,910 newborn children: results from a 13-year incidence study psychiatry. J Intellect Disabil Res 2001; 45: 467–483. in Arhus, Denmark. Hum Genet 1991; 87: 81–83. 23 Breen M, Arveiler B, Murray I, Gosden JR, Porteous DJ. YAC 3 Pickard BS, Millar JK, Porteous DJ, Muir WJ, Blackwood DH. mapping by FISH using Alu-PCR-generated probes. Genomics Cytogenetics and gene discovery in psychiatric disorders. Phar- 1992; 13: 726–730. macogenomics J 2005; 5: 81–88. 24 Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and 4 Blackwood DH, Fordyce A, Walker MT, St Clair DM, Porteous DJ, visualization of LD and haplotype maps. Bioinformatics 2005; 21: Muir WJ. Schizophrenia and affective disorders – cosegregation 263–265. with a translocation at chromosome 1q42 that directly disrupts 25 Thomson PA, Wray NR, Millar JK, Evans KL, Le Hellard S, Condie brain-expressed genes: clinical and P300 findings in a family. A et al. Association between the TRAX/DISC locus and both Am J Hum Genet 2001; 69: 428–433. bipolar disorder and schizophrenia in the Scottish population. 5 St Clair D, Blackwood D, Muir W, Carothers A, Walker M, Spowart Mol Psychiatry 2005; 10: 657–668. G et al. Association within a family of a balanced autosomal 26 Dudbridge F. Pedigree disequilibrium tests for multilocus haplo- translocation with major mental illness. Lancet 1990; 336: 13–16. types. Genet Epidemiol 2003; 25: 115–121. 6 Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, 27 Churchill GA, Doerge RW. Empirical threshold values for Semple CA et al. Disruption of two novel genes by a translocation quantitative trait mapping. Genetics 1994; 138: 963–971. co-segregating with schizophrenia. Hum Mol Genet 2000; 9: 28 Imoto I, Sonoda I, Yuki Y, Inazawa J. Identification and 1415–1423. characterization of human PKNOX2, a novel homeobox-contain- 7 Millar JK, Christie S, Anderson S, Lawson D, Hsiao-Wei Loh D, ing gene. Biochem Biophys Res Commun 2001; 287: 270–276. Devon RS et al. Genomic structure and localisation within a 29 Haller K, Rambaldi I, Kovacs EN, Daniels E, Featherstone M. linkage hotspot of Disrupted In Schizophrenia 1, a gene disrupted Prep2: cloning and expression of a new prep family member. by a translocation segregating with schizophrenia. Mol Psychiatry Dev Dyn 2002; 225: 358–364. 2001; 6: 173–178. 30 Fognani C, Kilstrup-Nielsen C, Berthelsen J, Ferretti E, Zappavigna 8 Ekelund J, Hovatta I, Parker A, Paunio T, Varilo T, Martin R et al. V, Blasi F. Characterization of PREP2, a paralog of PREP1, Chromosome 1 loci in Finnish schizophrenia families. Hum Mol which defines a novel sub-family of the MEINOX TALE homeo- Genet 2001; 10: 1611–1617. domain transcription factors. Nucleic Acids Res 2002; 30: 9 Hennah W, Varilo T, Kestila M, Paunio T, Arajarvi R, Haukka J 2043–2051. et al. Haplotype transmission analysis provides evidence of 31 Villaescusa JC, Verrotti AC, Ferretti E, Farookhi R, Blasi F. association for DISC1 to schizophrenia and suggests sex-depen- Expression of Hox cofactor genes during mouse ovarian follicular dent effects. Hum Mol Genet 2003; 12: 3151–3159. development and oocyte maturation. Gene 2004; 330: 1–7. 10 Gejman PV, Martinez M, Cao Q, Friedman E, Berrettini WH, 32 Tuson M, Marfany G, Gonzalez-Duarte R. Mutation of CERKL, Goldin LR et al. Linkage analysis of fifty-seven microsatellite loci a novel human ceramide kinase gene, causes autosomal to bipolar disorder. Neuropsychopharmacology 1993; 9: 31–40. recessive retinitis pigmentosa RP26. Am J Hum Genet 2004; 74: 11 Hwu HG, Liu CM, Fann CS, Ou-Yang WC, Lee SF. Linkage of 128–138. schizophrenia with chromosome 1q loci in Taiwanese families. 33 Fallin D, Cohen A, Essioux L, Chumakov I, Blumenfeld M, Cohen Mol Psychiatry 2003; 8: 445–452. D et al. Genetic analysis of case/control data using estimated 12 Curtis D, Kalsi G, Brynjolfsson J, McInnis M, O’Neill J, Smyth C et haplotype frequencies: application to APOE locus variation and al. Genome scan of pedigrees multiply affected with bipolar Alzheimer’s disease. Genome Res 2001; 11: 143–151. disorder provides further support for the presence of a suscept- 34 Werner P, Voigt M, Keinanen K, Wisden W, Seeburg PH. Cloning of ibility locus on chromosome 12q23–q24, and suggests the a putative high-affinity kainate receptor expressed predominantly presence of additional loci on 1p and 1q. Psychiatr Genet 2003; in hippocampal CA3 cells. Nature 1991; 351: 742–744. 13: 77–84. 35 Kamboj RK, Schoepp DD, Nutt S, Shekter L, Korczak B, True 13 Millar JK, Pickard BS, Mackie S, James R, Christie S, Buchanan SR RA et al. Molecular cloning, expression, and pharmacological et al. DISC1 and PDE4B are interacting genetic factors in characterization of humEAA1, a human kainate receptor subunit. schizophrenia that regulate cAMP signalling. Science 2005; 310: J Neurochem 1994; 62: 1–9. 1187–1191. 36 Velagaleti GV, Bien-Willner GA, Northup JK, Lockhart LH, 14 Kamnasaran D, Muir WJ, Ferguson-Smith MA, Cox DW. Disrup- Hawkins JC, Jalal SM et al. Position effects due to chromosome tion of the neuronal PAS3 gene in a family affected with breakpoints that map approximately 900 Kb upstream and schizophrenia. J Med Genet 2003; 40: 325–332. approximately 1.3 Mb downstream of SOX9 in two patients with 15 Pickard BS, Malloy MP, Porteous DJ, Blackwood DH, Muir WJ. campomelic dysplasia. Am J Hum Genet 2005; 76: 652–662. Disruption of a brain transcription factor, NPAS3, is associated 37 Kleinjan DA, van Heyningen V. Long-range control of gene with schizophrenia and learning disability. Am J Med Genet B expression: emerging mechanisms and disruption in disease. Am Neuropsychiatr Genet 2005; 136: 26–32. J Hum Genet 2005; 76: 8–32. 16 Gecz J, Barnett S, Liu J, Hollway G, Donnelly A, Eyre H et al. 38 Gribble SM, Prigmore E, Burford DC, Porter KM, Ng BL, Douglas EJ Characterization of the human glutamate receptor subunit 3 gene et al. The complex nature of constitutional de novo apparently (GRIA3), a candidate for bipolar disorder and nonspecific X-linked balanced translocations in patients presenting with abnormal mental retardation. Genomics 1999; 62: 356–368. phenotypes. J Med Genet 2005; 42: 8–16. 17 Baysal BE, Willett-Brozick JE, Badner JA, Corona W, Ferrell RE, 39 Gurling HM, Kalsi G, Brynjolfson J, Sigmundsson T, Sherrington Nimgaonkar VL et al. A mannosyltransferase gene at 11q23 is R, Mankoo BS et al. Genomewide genetic linkage analysis disrupted by a translocation breakpoint that co-segregates with confirms the presence of susceptibility loci for schizophrenia, on bipolar affective disorder in a small family. Neurogenetics 2002; chromosomes 1q32.2, 5q33.2, and 8p21–22 and provides support 4: 43–53. for linkage to schizophrenia, on chromosomes 11q23.3–24 and 18 Doody GA, Johnstone EC, Sanderson TL, Owens DG, Muir WJ. 20q12.1–11.23. Am J Hum Genet 2001; 68: 661–673. ‘Pfropfschizophrenie’ revisited: schizophrenia in people with 40 McQueen MB, Devlin B, Faraone SV, Nimgaonkar VL, Sklar P, mild learning disability. Br J Psychiat 1998; 173: 145–153. Smoller JW et al. Combined analysis from eleven linkage studies

Molecular Psychiatry GRIK4 in schizophrenia and bipolar disorder BS Pickard et al 857 of bipolar disorder provides strong evidence of susceptibility Loci ‘neuroleptic-free’ schizophrenics: evidence on reversible up-regu- on chromosomes 6q and 8q. Am J Hum Genet 2005; 77: 582–595. lation by typical neuroleptics. J Neurochem 1998; 71: 2454–2464. 41 Shibata H, Aramaki T, Sakai M, Ninomiya H, Tashiro N, Iwata N 53 Mohn AR, Gainetdinov RR, Caron MG, Koller BH. Mice with et al. Association study of polymorphisms in the GluR7, KA1 and reduced NMDA receptor expression display behaviors related to KA2 kainate receptor genes (GRIK3, GRIK4, GRIK5) with schizo- schizophrenia. Cell 1999; 98: 427–436. phrenia. Psychiat Res 2006; 141: 39–51. 54 Chittajallu R, Braithwaite SP, Clarke VR, Henley JM. Kainate 42 Kohn Y, Lerer B. Excitement and confusion on chromosome 6q: receptors: subunits, synaptic localization and function. Trends the challenges of neuropsychiatric genetics in microcosm. Pharmacol Sci 1999; 20: 26–35. Mol Psychiatry 2005; 10: 1062–1073. 55 Lerma J, Paternain AV, Rodriguez-Moreno A, Lopez-Garcia JC. 43 Wang WY, Barratt BJ, Clayton DG, Todd JA. Genome-wide Molecular physiology of kainate receptors. Physiol Rev 2001; 81: association studies: theoretical and practical concerns. Nat Rev 971–998. Genet 2005; 6: 109–118. 56 Antonova E, Sharma T, Morris R, Kumari V. The relationship 44 Miceli-Richard C, Lesage S, Rybojad M, Prieur AM, Manouvrier- between brain structure and neurocognition in schizophrenia: a Hanu S, Hafner R et al. CARD15 mutations in Blau syndrome. Nat selective review. Schizophr Res 2004; 70: 117–145. Genet 2001; 29: 19–20. 57 Miller S, Mayford M. Cellular and molecular mechanisms of 45 Hugot JP, Chamaillard M, Zouali H, Lesage S, Cezard JP, Belaiche J memory: the LTP connection. Curr Opin Genet Dev 1999; 9: et al. Association of NOD2 leucine-rich repeat variants with 333–337. susceptibility to Crohn’s disease. Nature 2001; 411: 599–603. 58 Martin SJ, Grimwood PD, Morris RG. Synaptic plasticity and 46 Ogura Y, Bonen DK, Inohara N, Nicolae DL, Chen FF, Ramos R et memory: an evaluation of the hypothesis. Annu Rev Neurosci al. A frameshift mutation in NOD2 associated with susceptibility 2000; 23: 649–711. to Crohn’s disease. Nature 2001; 411: 603–606. 59 Abel T, Lattal KM. Molecular mechanisms of memory acquisition, 47 Meirhaeghe A, Amouyel P. Impact of genetic variation of consolidation and retrieval. Curr Opin Neurobiol 2001; 11: 180–187. PPARgamma in humans. Mol Genet Metab 2004; 83: 93–102. 60 Kuperberg G, Heckers S. Schizophrenia and cognitive function. 48 Goff DC, Wine L. Glutamate in schizophrenia: clinical and Curr Opin Neurobiol 2000; 10: 205–210. research implications. Schizophr Res 1997; 27: 157–168. 61 Elvevag B, Goldberg TE. Cognitive impairment in schizophrenia is 49 Javitt DC. Glutamate as a therapeutic target in psychiatric the core of the disorder. Crit Rev Neurobiol 2000; 14: 1–21. disorders. Mol Psychiatry 2004; 9: 984–997. 62 Blackwood DH, Glabus MF, Dunan J, O’Carroll RE, Muir WJ, 50 Ibrahim HM, Hogg AJ, Healy Jr DJ, Haroutunian V, Davis KL, Ebmeier KP. Altered cerebral perfusion measured by SPECT in Meador-Woodruff JH. Ionotropic glutamate receptor binding and relatives of patients with schizophrenia. Correlations with subunit mRNA expression in thalamic nuclei in schizophrenia. memory and P300. Br J Psychiat 1999; 175: 357–366. Am J Psychiat 2000; 157: 1811–1823. 63 Contractor A, Swanson G, Heinemann SF. Kainate receptors are 51 Meador-Woodruff JH, Davis KL, Haroutunian V. Abnormal kainate involved in short- and long-term plasticity at mossy fiber synapses receptor expression in prefrontal cortex in schizophrenia. Neuro- in the hippocampus. Neuron 2001; 29: 209–216. psychopharmacology 2001; 24: 545–552. 64 Lauri SE, Bortolotto ZA, Bleakman D, Ornstein PL, Lodge D, Isaac 52 Sokolov BP. Expression of NMDAR1, GluR1, GluR7, and KA1 JT et al. A critical role of a facilitatory presynaptic kainate receptor glutamate receptor mRNAs is decreased in frontal cortex of in mossy fiber LTP. Neuron 2001; 32: 697–709.

Molecular Psychiatry