Analysis of single nucleotide polymorphisms in genes in the chromosome 12Q24.31 region points to P2RX7 as a susceptibility gene to bipolar affective disorderкод для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 141B:374 –382 (2006) Analysis of Single Nucleotide Polymorphisms in Genes in the Chromosome 12Q24.31 Region Points to P2RX7 as a Susceptibility Gene to Bipolar Affective Disorder Nicholas Barden,1* Mario Harvey,1 Bernard Gagné,1 Eric Shink,1 Monique Tremblay,1 Catherine Raymond,1 Michel Labbé,1 André Villeneuve,1 Denis Rochette,2 Lise Bordeleau,1 Herbert Stadler,3 Florian Holsboer,4 and Bertram Müller-Myhsok4 1 Neuroscience, CHUL Research Centre and Université Laval, Quebec Canada Complexe Hospitalier de la Sagamie, Chicoutimi, Quebec, Canada 3 Affectis Pharmaceuticals AG, Munich, Germany 4 Max-Planck Institute of Psychiatry, Munich, Germany 2 Previous results from our genetic analyses using pedigrees from a French Canadian population suggested that the interval delimited by markers on chromosome 12, D12S86 and D12S378, was the most probable genomic region to contain a susceptibility gene for affective disorders. Association studies with microsatellite markers using a case/control sample from the same population (n ¼ 427) revealed significant allelic associations between the bipolar phenotype and marker NBG6. Since this marker is located in intron 9 of the P2RX7 gene, we analyzed the surrounding genomic region for the presence of polymorphisms in regulatory, coding and intron/exon junction sequences. Twenty four (24) SNPs were genotyped in a case/control sample and 12 SNPs in all pedigrees used for linkage analysis. Allelic, genotypic or family-based association studies suggest the presence of two susceptibility loci, the P2RX7 and CaMKK2 genes. The strongest association was observed in bipolar families at the non-synonymous SNP P2RX7-E13A (rs2230912, P-value ¼ 0.000708), which results from an over-transmission of the mutant G-allele to affected offspring. This Gln460Arg polymorphism occurs at an amino acid that is conserved between humans and rodents and is located in the C-terminal domain of the P2X7 receptor, known to be essential for normal P2RX7 function. ß 2006 Wiley-Liss, Inc. KEY WORDS: bipolar disorder; association analysis; linkage disequilibrium; purinergic receptor Please cite this article as follows: Barden N, Harvey M, Gagné B, Shink E, Tremblay M, Raymond C, Labbé M, Villeneuve A, Rochette D, Bordeleau L, Stadler H, Holsboer F, Müller-Myhsok B. 2006. Analysis of Single This article contains supplementary material, which may be viewed at the American Journal of Medical Genetics website at http://www.interscience.wiley.com/jpages/1552-4841/suppmat/ index.html. *Correspondence to: Dr. Nicholas Barden, Neuroscience, CHUQ pavillon CHUL, 2705 Blvd. Laurier, Québec, QC, G1V 4G2, Canada. E-mail: firstname.lastname@example.org Received 22 November 2005; Accepted 8 February 2006 DOI 10.1002/ajmg.b.30303 ß 2006 Wiley-Liss, Inc. Nucleotide Polymorphisms in Genes in the Chromosome 12Q24.31 Region Points to P2RX7 as a Susceptibility Gene to Bipolar Affective Disorder. Am J Med Genet Part B 141B:374–382. INTRODUCTION Mood disorders, including bipolar (BP) and major depressive disorders (MDD) are the most common psychiatric disorders with a combined lifetime prevalence of around 15% [Blazer et al., 1994]. Patients cycle through episodes of depression and euthymia (MDD) or mania (BP) demonstrating profound changes in affect and mood, cognition, neurovegetative function, and psychomotor activity. Mood disorders are complex traits involving both environmental and genetic factors. Although several chromosomal regions, including 4p16, 12q23-24, 13q32-33, 18p11, 18q21-23, 21q22, and Xq26, show genetic linkage with either BP, MDD, or both, only a few candidate genes have been pointed to [Hattori et al., 2003; Sjoholt et al., 2004; Hoefgen et al., 2005]. Candidate gene approaches, focusing on neurotransmitter systems (e.g., norepinephrine, serotonin, and excitatory amino acids), stress hormone regulation (hypothalamic-pituitary-adrenocorticalaxis, HPA-axis) or factors involved in the postulated failure of adult neurogenesis in depression have had limited success [Zhang et al., 2005]. Our previous genetic studies in families from the Saguenay-Lac-St.-Jean (SLSJ) region of Quebec led to the identification of a susceptibility locus for both BP and MDD in the region of chromosome 12q24 with a parametric LOD score value of 3.35 under a recessive model and an MLS score value of 5.05 [Morissette et al., 1999; Shink et al., 2005a]. These results have received support from other studies in the 12q2324 region that suggested linkage with mood disorders including both BP and MDD [Abkevich et al., 2003; Curtis et al., 2003]. One marker, NBG6, gave a MLOD value equal to 3.35 with associated P-value less than 0.0001 [Shink et al., 2005b]. This marker has been located within intron 9 of the P2RX7 gene and for this reason we investigated genes within 100 kb up- and down-stream of this marker. In this study, we report association studies on three consecutive genes P2RX7, P2RX4, and CAMKK2 genes in a case/control sample from the SLSJ region. All of these genes share common features including expression in brain and a role in Ca2þ-dependent signaling pathways. We screened for mutations and genotyped SNPs with minor allele frequency greater than 0.01 in case/control and pedigree samples. Taken together, results of allelic and genotypic association as well as haplotype analysis and family-based association point to P2RX7 as a susceptibility gene for bipolar disorder. P2RX7 and Bipolar Disorder METHODS Ascertainment and Diagnosis Pedigrees. We ascertained 485 individuals from 41 families of the SLSJ region of Quebec. All Individuals were interviewed using a French translation of the Structured Clinical Interview for DSM-IIIR/IV [Spitzer et al., 1987] and a best estimate diagnosis established by panel including at least two psychiatrists. Diagnoses among the whole genotyped individuals were distributed as follows: 105 bipolar I disorder or schizoaffective disorder, bipolar type (BPI), 42 bipolar II disorder (BPII), 54 recurrent major depression (MDD), and 57 single episode major depression. Case/control sample. Diagnostic and ascertainment procedures were as previously described [Morissette et al., 1999; Shink et al., 2005b]. The SLSJ case sample was composed of 213 unrelated individuals with attachments to the founding population of SLSJ and included BPI (n ¼ 182, mean age at onset 28 11 (mean SD), 60% female) and BP II (n ¼ 31, mean age at onset 27 11 (mean SD), 55% female) diagnosed subjects. The control sample contained 214 individuals also coming from the SLSJ region, most of whom had previously participated in non-psychiatric genetic projects and provided informed written consent to participate in other genetic studies. No detailed genealogic information on this sample is available and, while only individuals drawn from clearly different pedigrees were sampled, we cannot rule out more distant degrees of relationship. All protocols were approved by and conducted in strict adherence to the ‘Comité d’éthique de la recherche clinique du CHUQ’. No screen to exclude control subjects with a history of psychiatric illness was done since there is no real gain in power to screen controls for association studies of diseases with lifetime risk estimated at 1% for BP [Owen et al., 1997]. Genders were distributed similarly in both samples according to the Fisher’s exact test (P-value ¼ 0.08) and no attempt was made to match the controls and cases by age. Using a critical level of 0.05, power calculations done with PAWE [Gordon et al., 2002, 2003] showed that our case/control sample had a power of 74% to detect allelic association assuming an odds ratio (OR) of 1.7 and a diallelic polymorphism with minor allele frequency of 10% in the general population. Power was 64% for genotypic association under the same conditions. Genomic DNA preparation. Blood samples from each individual were collected in 10-ml K3 EDTA Vacutainer tube (Becton-Dickinson) and genomic DNA was isolated using a Puregene DNA Isolation kit (Gentra Systems). DNA was solubilized in 500 ml of DNA Hydration Solution and the final concentration was adjusted to 300–400 mg/ml by spectroscopy at 260 nm. Mutation analyses. Single nucleotide polymorphisms and other variations were searched in coding sequences and exon–intron boundaries of genes using direct sequencing. The starting sample was composed of 16 unrelated BP affected individuals selected to maximize the number of different haplotypes and according to their link with families that gave positive genetic linkages on chromosome 12q24. Genotyping of SNPs. SNPs were genotyped by resequencing of the amplification products. The sequencing traces for each individual are automatically typed for the corresponding SNP using a home-developed program, GENO.pl and genotyping results compiled in a 4D database. Deviation from Hardy– Weinberg equilibrium (HWE) was assessed for each SNP in both case and control groups with the exact test from Genepop software package (http://wbiomed.curtin.edu.au/genepop). Statistics. Testing for allelic and genotypic association with the BPI or BPII phenotype, as well as the computation of odds ratios and 95% confidence intervals, was tested with the Fisher’s exact test from the SISA webserver (http://home.- 375 clara.net/sisa/). Haplotypes, either within block or inter-block, were estimated with the expectation-maximization (EM) algorithm [Excoffier and Slatkin, 1995] implanted in the cocaphase module of UNPHASE Version 2.40 [Dudbridge, 2003]. Since the EM algorithm has limited precision to estimate haplotype frequencies <1%, such haplotypes were excluded using the–droprare option. P-values for individual haplotypes were calculated with the–individual option. The family-based association tests were done using the FBAT software version 1.5.5 [Rabinowitz and Laird, 2000], in which alleles transmitted to affected offspring are compared with the expected distribution of alleles among offspring. The affected status is defined as BP I and II, all other individuals are considered as unaffected or unknown. We tested for the null hypothesis of no association and no linkage, and used a biallelic test. Since we may not estimate disease prevalence we used the–o option with the FBAT command, thus estimating an offset value to minimize the variance. RESULTS DNA from 16 unrelated BPI (n ¼ 12) and BPII (n ¼ 4) individuals, who were selected among families and trios from the SLSJ region to maximize the number of different potential haplotypes, was sequenced from both strands over the genes P2RX7, P2RX4, and CAMKK2. We identified 126 variations, of which 39 were found in untranslated sequences, 34 in coding sequences, and 53 in intronic regions (see supplementary information on-line, Table S1). The P2RX7-associated SNPs accounted for 56% (71) of total identified SNPs in this genomic region (Fig. 1). Thirteen of these polymorphisms lead to nonsynonymous mutations (nsSNP) causing amino acid changes and one described a deletion of seven amino acids (del488-494). Selection of SNPs for genetic analysis was mostly based on the putative SNP functionality. We typed 24 polymorphisms identified with minor allele frequency higher or equal to 5% distributed in regulatory sequences (10), coding sequences (13), and exon/intron junctions (1) of these three adjacent genes (Table I). The case/control sample was composed of 213 BP patients and 214 controls that gave 71% power to detect significant allelic association, at P-value ¼ 0.05, for polymorphisms of minor allele frequency of 0.05 and showing a relative risk of 2. Accordingly, our analysis strategy was to determine positive associations under the P-value 0.05 without correction for multiple testing and further confirm the positive results in an independent sample. The deviation from Hardy–Weinberg equilibrium was calculated in both case and control groups. We observed disequilibrium with one polymorphism, P2RX7E11B, which gave a heterozygote deficiency in the control group and an excess among cases. The hypothesis of allelic or genotypic association with BP was evaluated with a Fisher’s exact test. Since we may not anticipate the correct inheritance model, the genotypic analysis was done under additive, dominant and recessive models. Only three markers, P2RX7I07E, P2RX7-E11B, and P2RX4-UTR3A gave significant allelic and/or genotypic association (Table IIA). The pairwise LD measures between P2RX7-I07E and P2RX7-E11B revealed that both markers are closely linked together, however, considering the HW disequilibrium at P2RX7-E11B locus we should exclude these positive results. The P2RX4-UTR3A polymorphism is about 75 bp downstream of the P2RX4 gene and is unlikely to interfere with gene function, but it might be linked to other functional polymorphisms. We tried to reduce phenotypic heterogeneity by excluding the 31 BPII patients, and observed only slight decrease in significance (not shown) probably caused by the loss of power. Considering both D0 > 0.33 and r2 > 0.1 as a threshold for meaningful LD [Kruglyak, 1999; Nakajima et al., 2002], P2RX4-UTR3A is found in LD with six distant SNPs in the case group, P2RX7- 376 Barden et al. TABLE I. Description of Single Nucleotide Polymorphisms Analyzed SNP ID P2RX7-UTR5F P2RX7-UTR5G P2RX7-UTR5A P2RX7-UTR5B P2RX7-UTR5E P2RX7-E02A* P2RX7-I04B P2RX7-E05A* P2RX7-I07E P2RX7-E08A* P2RX7-E11B* P2RX7-E11C* P2RX7-E13A* P2RX7-E13B* P2RX7-E13C* P2RX7-UTR3A P2RX7-UTR3B P2RX4-UTR5A P2RX4-UTR5B P2RX4-E07A P2RX4-UTR3A* CAMKK2-E09A* CAMKK2-E01B* CAMKK2-E01A* dbSNPa Allele(modification) Minor allele frequency Positionb rs523977 rs520396 rs494986 rs2393799 rs684201 ss35031323 rs208293 rs208294 rs504677 rs7958311 rs1718119 rs6489795 rs2230912 rs3751144 rs3751143 ss35031375 rs1653625 ss35031381 ss35031382 rs11065500 rs11065501 rs4980999 rs3817190 ss35031413 T-C T-G C-A C-T G-A T-C(Val76Ala) C-T T-C(Tyr155His) C-T G-A(Arg270His) G-A(Ala348Thr) C-G(Thr357Ser) A-G(Gln460Arg) C-T(Pro474Pro) A-C(Glu496Ala) C-A A-C C-A A-G A-G(Ser242Gly) C-G C-T(Arg363Cys) T-A(Thr85Ser) C-T(Ser10Asn) 0.19 0.10 0.05 0.23 0.05 0.06 0.25 0.49 0.35 0.25 0.37 0.10 0.18 0.10 0.23 0.46 0.07 0.16 0.31 0.15 0.31 0.15 0.38 0.08 12138634 12139014 12139445 12139521 12139852 12162198 12169689 12169762 12174698 12174864 12184612 12184640 12191705 12191748 12191813 12192393 12192394 12216833 12216944 12236155 12241479 12260605 12281586 12281810 All SNPs were genotyped in the case/control sample, whereas only those with asterisk (*) were genotyped in pedigrees. a Available in dbSNP build 125. b Position relative to the contig NT_009775.15 from the NCBI. UT5G (r2 ¼ 0.131, D0 ¼ 0.764, distance ¼ 102.5 kb), P2RX7E02A (r2 ¼ 0.112, D0 ¼ 0.857, distance ¼ 79.3 kb), P2RX7-E08A (r2 ¼ 0.168, D0 ¼ 0.483, distance ¼ 66.7 kb), P2RX7-E13C (r2 ¼ 0.132, D0 ¼ 0.952, distance ¼ 49.7 kb), P2RX4-UTR5B (r2 ¼ 0.106, D0 ¼ 0.688, distance ¼ 24.7 kb), and CAMKK2E01A (r2 ¼ 0.168, D0 ¼ 0.945, distance ¼ 40.4 kb). This irregular pattern of LD has been noted in many studies on small genomic regions with densely selected markers [Taillon-Miller et al., 2000; Abecasis et al., 2001; Nakajima et al., 2002], and is to be expected when recombination becomes rare relative to other factors influencing LD, such as mutation rate, genetic drift, and gene conversion [Ardlie et al., 2001]. Consequently, considering P2RX4-UTR3A as a true positively associated locus, the causal disease mutation(s) might be found anywhere within the three adjacent genes. To support the single marker association study and define more accurately a susceptibility locus, we then carried out haplotype analysis. The LD measures, especially the very low r2 values, reflect large variation in marker allele frequencies, and consequently suggest important haplotype diversity (Fig. 2). Considering this, we firstly segmented the region in haplotype blocks to minimize estimated haplotype and increase power to detect positive associations (Fig. 2). We defined four blocks, and estimated haplotypes using the EM algorithm implemented in cocaphase program. We identified global significant P-values for blocks 2 and 4 with P-value ¼ 0.0410 and 0.0390, respectively (Table IIB). The haplotype block 2 defines the last six exons of the P2RX7 gene while block 4 encloses the associated SNP P2RX4-UTR3A, hence supporting the prior single marker analysis. Although useful for statistical analysis, it is however inappropriate to arbitrarily segment a continuous chromosomal region. To circumvent this without loss of power, we treated the selected region as a continuum by performing haplotype analysis using a sliding window method (in cocaphase program) with a small window of seven SNPs (Table IIC). We considered only estimated haplotypes showing frequencies greater than or equal to 0.03 in either group, thus decreasing the degree of freedom and giving greater power to detect effects in common haplotypes. This method showed two significantly associated haplotypes, which are close to those aforementioned. The 7-SNP positively associated haplotype window bordered by P2RX7-I04B and P2RX7-E13A overlaps blocks 1 and 2 and is characterized by a haplotype strictly observed in the case group (P-value ¼ 0.0006) bearing the mutated allele for the non-synonymous SNP P2RX7-E08A (Table IIC). The other 7-SNP positively associated window bordered by markers P2RX7-UTR3A and CAMKK2-E01B comprises blocks 3 and 4 and the associated marker P2RX4UTR3A (Table IIC). This haplotype analysis supports the single marker result at the locus P2RX4-UTR3A and raises the possibility of a susceptibility locus in the P2RX7 gene. However, according to the LD structure, we may not answer the question as to whether these loci are linked between each other, linked with other loci or independently involved. The selection of this region for a case/control SNP-based association studies was based on linkage results in pedigrees from the SLSJ region. We therefore genotyped in the 41 pedigrees that were used for linkage studies [Shink et al., 2005a] all 11 non-synonymous SNPs (with minor allele frequency >0.01) and the associated SNP P2RX4-UTR3A that are found in the region delineated by the positive haplotype results. We used the family-based approach for association (FBAT), which uses data from all family members and avoids complications due to population admixture often found in case/ control studies. Allelic association was first tested under the null hypothesis of no association and no linkage, and we generated data under three inheritance models: additive, dominant, and recessive, as well as using either or both bipolar I and bipolar II as affected status (see supplementary Table S2). Considering that BP disorder is a complex trait and having no estimation of disease prevalence, we used the–o option giving an offset value in order to minimize the variance of the statistic. Table IIIA summarizes the results by showing significant associations (P-value <0.05). We observed two positively associated polymorphisms, P2RX7-E13A and P2RX7 and Bipolar Disorder 377 TABLE IIA. Allelic and Genotypic Associations SAMPLE SNP SAMPLE Fisher’s (P-value) OR (CI 95%) Genotype Allele N A P2XR7-I07E C T 272 156 278 140 0.387 0.88 (0.66–1.17) P2XR7-E11B G A 271 155 269 157 0.9343 1.02 (0.77–1.35) P2XR4-UTR3A C G 313 115 279 147 0.018 1.43 (1.07–1.92) C/C C/T T/T G/G G/A A/A C/C C/G G/G N A 92 88 34 93 85 35 114 72 17 87 104 18 78 113 22 91 96 25 Fisher’s (P-value) OR (CI 95%) Model 0.0262 0.50 (0.27–0.92) Recessive 0.0172 1.34 (0.90–1.98) Additive 0.008 1.70 (1.16–2.51) Dominant SNP-based association study. Positive results from the SNP-based association study conducted in the case/control sample from the SLSJ region (BP ¼ 213, controls ¼ 214). TABLE IIB. Haplotype Analysis Within Blocks Frequency Block Marker Allele Case Control Chi2 (P-value) LRS (global P-value) 0.0410 2 P2RX7-I07E P2RX7-E08A P2RX7-E11B P2RX7-E11C P2RX7-E13A P2RX7-E13B P2RX7-E13C P2RX7-UTR3A C-A-G-C-A-C-A-C C-G-G-C-A-C-C-C T-G-A-C-G-C-A-A T-G-A-C-A-C-A-A C-G-G-G-A-T-A-A C-G-G-C-A-C-A-C C-G-A-C-A-C-A-A C-A-G-C-G-C-A-C T-G-G-C-A-C-A-C 0.26 0.21 0.18 0.13 0.09 0.04 0.04 0.01 0 0.22 0.21 0.16 0.18 0.11 0.05 0.02 0 0.02 0.159 0.979 0.551 0.095 0.346 0.249 0.101 0.012 0.133 4 P2RX4-UTR3A CAMKK2-E09A CAMKK2-E01B CAMKK2-E01A C-C-T-C C-C-A-C G-C-T-C C-T-T-C G-C-A-T G-C-A-C 0.26 0.26 0.19 0.13 0.09 0.07 0.34 0.22 0.13 0.17 0.06 0.07 0.012 0.314 0.065 0.179 0.235 0.411 0.0390 SNP-based association study. Positive results from the SNP-based association study conducted in the case/control sample from the SLSJ region (BP ¼ 213, controls ¼ 214). TABLE IIC. Haplotype Analysis With a Sliding Window Strategy Frequency 7-SNP window (markers) Allele Case Control Chi2 P-value C-C-C-A-G-C-A 0.16 0.15 0.523 C-C-C-G-G-C-A C-T-C-A-G-C-A C-T-C-G-G-C-A C-T-C-G-G-G-A C-T-T-G-A-C-G T-C-C-A-G-C-A T-C-T-G-A-C-A T-C-T-G-A-C-G 0.06 0.08 0.19 0.07 0.13 0.03 0.11 0.05 0.06 0.08 0.19 0.09 0.14 0 0.16 0.02 0.864 0.911 0.959 0.391 0.824 0.0006 0.060 0.048 A-A-G-G-C-T-T 0.11 0.13 0.362 A-C-A-A-C-C-A A-C-A-A-C-C-T A-C-A-A-C-T-T A-C-A-A-G-C-A A-C-A-A-G-C-T A-C-G-A-C-C-A A-C-G-A-C-C-T C-C-A-A-C-C-T 0.14 0.18 0.02 0.13 0.19 0.09 0.05 0.03 0.16 0.2 0.03 0.10 0.13 0.06 0.07 0.07 0.685 0.487 0.544 0.106 0.073 0.163 0.327 0.012 LRS (global P-value) 0.0178 P2RX7-I04B— P2RX7-E13A 0.0249 P2RX7-UTR3B— CAMKK2-E01B SNP-based association study. Positive results from the SNP-based association study conducted in the case/control sample from the SLSJ region (BP ¼ 213, controls ¼ 214). 378 Barden et al. Fig. 1. Gene structure of P2RX7, P2RX4, and CAMKK2. a: The P2RX7 gene codes for 13 exons. All coding, untranslated sequences and 1,800 pb upstream of the start codon were analyzed. The P2RX4 gene codes for 12 exons. The CAMKK2 gene has two protein variants from differential usage of the last coding exon, resulting in distinct 30 -untranslated sequences. b: The P2RX7 gene product has four major domains: a short intracellular N-terminal domain, an extracellular loop (ECL) separated by two trans-membrane domains (TM), and a long intracellular C-terminal domain. Eight SNPs were identified in the extracellular loop, one in the second hydrophobic domain (TM2) and eight in the intracellular C-terminus. Residues Gly150, Glu186, and Arg276 are mostly conserved between P2X family members and Leu191, Thr357, Gln460, Glu496, and Arg578 among P2RX7 ortholog genes. The P2RX4 gene product shares the same structural organization as P2RX7. CAMKK2 encodes for two isoforms of 587 and 533 amino acids and is composed of a kinase domain (KD), ATP binding site (ABS) and two overlapping domains, the calmodulin binding domain and the autoinhibitory domain (CaMBD/AID). The CAMKK2 gene extends over 60 kb, and is divided into 20 exons. Fig. 2. Linkage disequilibrium (LD) measures and haplotype block structure across the 167 kb P2RX7-adjacent region using the Haploview program [Barrett et al., 2004]. Haplotype blocks are defined by successive SNPs given a mean jD0 j > 0.80. P2RX7 and Bipolar Disorder 379 TABLE IIIA. Allelic Analysis Allele Frequency Model Z P-value Offset P2RX7-E13A G A 0.146 0.854 Additive 3.387 0.000708 0.311360 CAMKK2-E01B A T 0.419 0.581 Additive 2.421 0.01548 0.275688 Marker Family-based association study. Description of positive results (P-value <0.05) from the family-based association study. TABLE IIIB. Genotypic Analysis Genotype Frequency Z P-value P2RX7E-13A A/A A/G G/G 0.722 0.265 0.014 2.028 0.688 3.157 0.04253 0.49170 0.00156 CAMKK2-E01B A/A A/T T/T 0.127 0.583 0.290 1.973 0.254 1.549 0.04846 0.79963 0.12131 Marker Family-based association study. Description of positive results (P-value <0.05) from the family-based association study. CAMKK2-E01B, with the additive model. Both are nonsynonymous SNPs, and describe an over-transmission of the mutant allele (minor allele) to affected offspring. If the phenotype was restricted to BPI only, results remained significant but decreased slightly (not shown) probably due to loss of power. It is noteworthy that the marker P2RX4-UTR3A is not overtransmitted. When we tested for the null hypothesis in large pedigrees that show linkage on chromosome 12, the locus P2RX7-E13A remained positively associated but with lesser significance (P-value ¼ 0.008, not shown). This indicates that the strong positive association (nominal P-value ¼ 0.000708, P-value of 0.0252 Bonferroni corrected for a total of 36 SNPs tested both in the case-control and the FBAT analysis) is not a reflection of a pedigree-specific effect but has a broader incidence in the SLSJ population. The P2RX7-E13A locus defines a missense mutation leading to amino acid change Gln460Arg (Table IIIA). This residue of the C-terminal domain is located in an SH3-like domain and is conserved between humans and rodents. In contrast, the other associated polymorphism in the CAMKK2 gene, Thr85Ser, is not conserved, and is not a phosphorylation site. The analysis of genotype distribution gave support to allelic results (Table IIIB). We noted a significant over-transmission of the homozygote mutant/mutant-genotype to affected children. This is also in agreement with the previous linkage findings where positive LOD scores on 12q were observed under a recessive model (Morissette et al., 1999; Shink et al., 2005a]. This conclusion is strengthened by TDT analysis in the SLSJ families used for the linkage finding and which gave a significant result for P2RX7-E13A, with the G-allele, which is positively associated in the case-control studies, being transmitted preferentially in 71% of all informative meioses (P ¼ 0.04). Using only one sib sampled at random from each family we estimated the relative risk in these families to be 2.5. We also performed haplotype analysis in order to reveal any other loci that might be hidden by single marker analysis, but did not observe any significantly associated haplotypes except for those bearing the P2RX7-E13A locus (data not shown). DISCUSSION Several groups have pointed to the long arm of chromosome 12 as a susceptibility locus for bipolar disorders. However, extensive genetic studies on 12q did not demonstrate a consensus region but suggested three major and overlapping susceptibility loci extending over 35 Mb (see Figure 3). Our linkage analyses revealed two loci on chromosome 12q. The major locus is located at 12q24.31 where non-parametric analysis, based on a broad affection status (bipolar, schizoaffective disorders, and recurrent major depression), gave a maximum MLS score of 5.05 at the marker D12S378. Both linkage and association studies have already highlighted this narrow critical region [Dawson et al., 1995; Degn et al., 2001; Curtis et al., 2003]. While linkage studies from Danish pedigrees also demonstrated positive linkages more telomeric on 12q24 with a maximum LOD score at D12S1639 [Ewald et al., 2002], other studies pointed to regions more centromeric on 12q23-24 [Craddock et al., 1994; Ekholm et al., 2003]. A recent study on two pedigrees that cosegregated mood disorders and Darier’s disease delimited another critical region of interest at 12q23-q24.1 between D12S1127 and D12S1646 [Green et al., 2005]. A more conservative analysis of these data described a full region extending over 26 cM. Accordingly, the long arm of chromosome 12 may include more than one gene involved in the susceptibility to mood disorders. Bipolar disorder is described as a complex genetic trait with influenced by several genes with common natural variants. Consequently, it is quite likely that more than one gene associated to bipolar disorders will be found on chromosome 12q. Recent candidate gene analyses on 12q have pointed to three genes with positive association to bipolar disorder [Glaser et al., 2005a,b; LyonsWarren et al., 2005]. Considerable distances between each of these genes strongly supports the assumption of several BPassociated genes or more speculatively, a mood-regulating gene cluster. In this report, we present the analysis of three genes in the vicinity of the associated marker, NBG6. These three genes, P2RX7, P2RX4, and CAMKK2, are potentially implicated in Ca2þ signaling and neurotransmitter release. Changes in intracellular concentration of Ca2þ have been observed in patients suffering from mood disorders [Wasserman et al., 2004]. The mutation screening among BP affected individuals covered 26 kb, mostly in coding and regulatory sequences. This genomic region showed a mutation rate of 5 103 (71/9789), which is greater than the mean genome mutation rate of 8 104 [Reich et al., 2002]. Interestingly, the P2RX7 gene 380 Barden et al. Fig. 3. Positive genetic linkage findings in bipolar disorders on 12q23-24. The relevant studies are shown, together with the location of the maximum LOD score in support of linkage. At left, the positively associated candidate genes are listed. Their approximate genomic position is relative to the human genome build 35. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] revealed several non-synonymous polymorphisms (16), but after confirmation in a larger sample (50 affected individuals) many of these had a minor allele frequency lower than 0.01. It is possible that our extensive mutation screening within this genomic region revealed variations usually missed by other more conservative mutation screens. Considering that this region is associated with bipolar disorder (results presented here), the observation of such mutation variability [Splendore et al., 2000] could underline the possible contribution of many relatively rare polymorphisms in alteration of protein function rather than a few common variants [Wright et al., 2003] a fact that decreases the power to detect associated SNPs in small case/control samples. While the functional relevance of all these mutations is unknown, the high variability is intriguing, suggesting that P2RX7 is not essential for survival as indicated by P2RX7 knockout transgenic mice [Solle et al., 2001]. Regardless, our case/control association studies revealed positive association with polymorphisms in P2RX4. These polymorphisms are putatively non-functional and most likely in LD with another functional SNP in P2RX7, P2RX4, or CAMKK2. Fewer susceptibility alleles should be found among family members compared to unrelated case/control samples and a family-based approach with large pedigrees would give more power to detect specific alleles. In fact, this strategy allowed us to detect a strong association between BP disorder and the functional polymorphism P2RX7-E13A, which leads to change of amino acid Gln460 to Arginine. Retrospectively, we could ask why P2RX7E13A was not associated in the case/control sample. Firstly, single marker analysis gave an odds ratio of 1.26 at this locus (frequency in cases ¼ 0.200, frequency in controls ¼ 0.165, see supplementary Table S3), and our sample was thus underpowered to detect an association. However, haplotype analysis pointed to P2RX7-E13A since the haplotype unique to the case group has the P2RX7-E13A mutant-G allele, which is the only variation from the most frequent haplotype. The family-based association study also pointed to a second associated polymorphism in the CAMKK2 gene. This mutation causes the change Thr85Ser. Thr85 is not conserved between mammals and is not a predicted phosphorylation site. This locus is in meaningful LD (r2 ¼ 0.170, D0 ¼ 0.695) with the strongly associated marker P2RX7-E13A, and could explain the positive result. However, another genotyped marker with a similar frequency, P2RX7-E11B, is also in useful LD with P2RX7-E13A (r2 ¼ 0.358, D0 ¼ 0.847) but has not been associated to BP. Therefore, we cannot exclude the CAMKK2 gene as an additional susceptibility gene, as hinted by microarray gene expression analysis with mouse models that showed both methamphetamine and valproate influence the expression of CAMKK2 [Ogden et al., 2004]. ATP-gated P2X receptors are cation-selective ion channels with high calcium permeability that open on binding of extracellular ATP [Khakh, 2001; North, 2002]. The P2RX7E13A susceptibility mutation described here is located in the intracellular C-terminal domain, which is known to be essential for protein functions such as large pore formation, intracellular signaling, membrane blebbing and receptor trafficking [Kim et al., 2001; Wilson et al., 2002]. Two mutations, E496A (rs3751143) and I568N (rs1653624) in this region have been reported as critical for P2RX7 function [Gu et al., 2001; Wiley et al., 2003]. Chimeric protein experiments between human and rodent P2RX7 have demonstrated that amino acids 347–595 are responsible for the functional differences between these native receptors [Rassendren et al., 1997] indicating that non-synonymous polymorphisms in the C-terminal domain (conserved or non-conserved) are likely to be dysfunctional. Little is known about the functional effect of the Q460R mutation. However, the Q460 residue is conserved between humans and rodents and is part of an SH3-like domain [Denlinger et al., 2001]. This residue could thus be involved in P2RX7 dimerization or in other protein–protein interactions involving SH3 domain-containing proteins. P2RX7 and Bipolar Disorder Although conserved missense mutations are not always disease-predisposing alleles [Saunders et al., 2006], the likelihood that these variations would be more frequently observed associated to diseases is suggested by extrapolation from monogenic diseases [Botstein and Risch, 2003] and also by some complex disorders [Bonifati et al., 2003]. Expression of P2X7 receptor is mostly restricted to immunerelated cells, such as monocyte/macrophage, NK-cells, T- and Bcells [Collo et al., 1997; Gu et al., 2000; Chakfe et al., 2002]. In brain, both microglia and astrocytes are stimulated, by ATPinduced P2X7 receptor activation, to produce cytokines, chemokines and growth factors. These effects of P2RX7 are thought to promote inflammatory response by activating and recruiting immune cells. On the other hand, the activation of astrocytes to produce growth and trophic factors enhances neuronal survival and promotes neurogenesis [Walter et al., 2004]. There is also evidence for the presence of P2X7 receptors in presynaptic terminals, however this is controversial since the P2RX7-KO mice showed the same anti-P2X7 labeling characteristics as the control animals [Sim et al., 2004]. While there is some support for a neurodegenerative role of P2RX7 [Le Feuvre et al., 2002], recent studies demonstrated its neuroprotective effect [Wang et al., 2003; Suzuki et al., 2004; Walter et al., 2004]. Consequently, the role of P2RX7 during an immune response would not be limited to the activation of immune associatedcells, but would extend to the modulation of the response intensity, thus preventing extended or uncontrolled inflammatory response. The P2X7 receptor has already been associated to polygenic inflammatory diseases such as systemic lupus erythematosus [Nath et al., 2004; Elliott et al., 2005]. The concept of an inter-relationship between the psychological state and immune status can be traced back several years and it is now well accepted that the nervous, endocrine, and immune systems are so closely linked that they should be regarded as a single network. Recently it has been suggested that immune activation could be present in depressed patients. Several works reported increased concentrations of proinflammatory cytokines and their receptors in depressed individuals [Song et al., 1994; Maes, 1995; Kim et al., 2002]. Consequently, harmful or sustained inflammatory responses could be associated to depressive behavior, and it is likely to propose a role for the inflammatory mediator P2X7 receptor. In conclusion, the association studies presented here, especially the family-based study, clearly demonstrate the presence of a BP susceptibility locus close to the already associated marker NBG6. Although the strongest result has been observed with the P2RX7 gene, we may not exclude the CAMMK2 gene as a second putative candidate gene. Further genetic work will probably answer whether these loci are linked, interact together or are totally independent. The presence of several non-synonymous SNPs in the P2RX7 gene suggests a role for different mutant haplotypes and recent molecular studies on P2RX7 haplotypes lead to the assumption that many mutations would trigger a similar phenotype, that is a dysfunctional P2X7 receptor. ACKNOWLEDGMENTS We acknowledge the participation of individuals and family members in this work. REFERENCES Abecasis GR, Noguchi E, Heinzmann A, Traherne JA, Bhattacharyya S, Leaves NI, Anderson GG, Zhang Y, Lench NJ, Carey A, Cardon LR, Moffatt MF, Cookson WO. 2001. Extent and distribution of linkage disequilibrium in three genomic regions. Am J Hum Genet 68:191–197. Abkevich V, Camp NJ, Hensel CH, Neff CD, Russell DL, Hughes DC, Plenk AM, Lowry MR, Richards RL, Carter C, Frech GC, Stone S, Rowe K, 381 Chau CA, Cortado K, Hunt A, Luce K, O’Neil G, Poarch J, Potter J, Poulsen GH, Saxton H, Bernat-Sestak M, Thompson V, Gutin A, Skolnick MH, Shattuck D, Cannon-Albright L. 2003. Predisposition locus for major depression at chromosome 12q22-12q23.2. Am J Hum Genet 73:1271–1281. Ardlie K, Liu-Cordero SN, Eberle MA, Daly M, Barrett J, Winchester E, Lander ES, Kruglyak L. 2001. Lower-than-expected linkage disequilibrium between tightly linked markers in humans suggests a role for gene conversion. Am J Hum Genet 69(3):582–589. Barrett JC, Fry B, Maller J, Daly MJ. 2004. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 21:263–265. Blazer DG, Kessler RC, McGonagle KA, Swartz MS. 1994. The prevalence and distribution of major depression in a national community sample: The National Comorbidity Survey. Am J Psychiatry 151:979–986. Bonifati V, Rizzu P, van Baren MJ, Schaap O, Breedveld GJ, Krieger E, Dekker MC, Squitieri F, Ibanez P, Joosse M, et al. 2003. Mutations in the DJ-1 gene associated with autosomal recessive early-onset parkinsonism. Science 299:256–259. Botstein D, Risch N. 2003. Discovering genotypes underlying human phenotypes: Past successes for mendelian disease, future approaches for complex disease. Nat Genet 33(Suppl):228–237. Chakfe Y, Seguin R, Antel JP, Morissette C, Malo D, Henderson D, Seguela P. 2002. ADP and AMP induce interleukin-1beta release from microglial cells through activation of ATP-primed P2X7 receptor channels. J Neurosci 22:3061–3069. Collo G, Neidhart S, Kawashima E, Kosco-Vilbois M, North RA, Buell G. 1997. Tissue distribution of the P2X7 receptor. Neuropharmacology 36:1277–1283. Craddock N, Owen M, Burge S, Kurian B, Thomas P, McGuffin P. 1994. Familial cosegregation of major affective disorder and Darier’s disease (keratosis follicularis). Br J Psychiatry 164:355–358. Curtis D, Kalsi G, Brynjolfsson J, McInnis M, O’Neill J, Smyth C, Moloney E, Murphy P, McQuillin A, Perursson J, Gurling H. 2003. Genome scan of pedigrees multiply affected with bipolar disorder provides further support for the presence of a susceptibility locus on chromosome 12q23-q24, and suggests the presence of additional loci on 1p and 1q. Psychiatry Genet 13:77–84. Dawson E, Parfitt E, Roberts Q, Daniels J, Lim L, Sham P, Nothen M, Propping P, Lanczik M, Maier W, et al. 1995. Linkage studies of bipolar disorder in the region of the Darier’s disease gene on chromosome 12q2324.1. Am J Med Genet 60:94–102. Degn B, Lundorf MD, Wang A, Vang M, Mors O, Kruse TA, Ewald H. 2001. Further evidence for a bipolar risk gene on chromosome 12q24 suggested by investigation of haplotype sharing and allelic association in patients from the Faroe Islands. Mol Psychiatry 6:450–455. Denlinger LC, Fisette PL, Sommer JA, Watters JJ, Prabhu U, Dubyak GR, Proctor RA, Bertics PJ. 2001. Cutting edge: The nucleotide receptor P2X7 contains multiple protein- and lipid-interaction motifs including a potential binding site for bacterial lipopolysaccharide. J Immunol 167:1871–1876. Dudbridge F. 2003. Pedigree disequilibrium tests for multilocus haplotypes. Genet Epidemiol 25:115–121. Ekholm JM, Kieseppa T, Hiekkalinna T, Partonen T, Paunio T, Perola M, Ekelund J, Lonnqvist J, Pekkarinen-Ijas P, Peltonen L. 2003. Evidence of susceptibility loci on 4q32 and 16p12 for bipolar disorder. Hum Mol Genet 12:1907–1915. Elliott JI, McVey JH, Higgins CF. 2005. The P2X7 receptor is a candidate product of murine and human lupus susceptibility loci: A hypothesis and comparison of murine allelic products. Arthritis Res Ther 7:R468– R475. Ewald H, Flint T, Kruse TA, Mors O. 2002. A genome-wide scan shows significant linkage between bipolar disorder and chromosome 12q24.3 and suggestive linkage to chromosomes 1p22-21, 4p16, 6q14-22, 10q26 and 16p13.3. Mol Psychiatry 7:734–744. Excoffier L. Slatkin M. 1995. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol 12:921–927. Glaser B, Kirov G, Green E, Craddock N, Owen MJ. 2005a. Linkage disequilibrium mapping of bipolar affective disorder at 12q23-q24 provides evidence for association at CUX2 and FLJ32356. Am J Med Genet Part B 132B:38–45. Glaser B, Kirov G, Bray NJ, Green E, O’Donovan MC, Craddock N, Owen MJ. 2005b. Identification of a potential bipolar risk haplotype in the gene encoding the winged-helix transcription factor RFX4. Mol Psychiatry 10:920–927. 382 Barden et al. Gordon D, Finch SJ, Nothnagel M, Ott J. 2002. Power and sample size calculations for case-control genetic association tests when errors present: Application to single nucleotide polymorphisms. Human Heredity 54:22–33. Gordon D, Levenstien MA, Finch SJ, Ott J. 2003. Errors and linkage disequilibrium interact multiplicatively when computing sample sizes for genetic case-control association studies. Pac Symp Biocomput 490– 501. Green E, Elvidge G, Jacobsen N, Glaser B, Jones I, O’Donovan MC, Kirov G, Owen MJ, Craddock N. 2005. Localization of bipolar susceptibility locus by molecular genetic analysis of the chromosome 12q23-q24 region in two pedigrees with bipolar disorder and Darier’s disease. Am J Psychiatry 162:35–42. Gu BJ, Zhang WY, Bendall LJ, Chessell IP, Buell GN, Wiley JS. 2000. Expression of P2X(7) purinoceptors on human lymphocytes and monocytes: Evidence for nonfunctional P2X(7) receptors. Am J Physiol Cell Physiol 279:C1189–C1197. Gu BJ, Zhang W, Worthington RA, Sluyter R, Dao-Ung P, Petrou S, Barden JA, Wiley JS. 2001. A Glu-496 to Ala polymorphism leads to loss of function of the human P2X7 receptor. J Biol Chem 276:11135–11142. Hattori E, Liu C, Badner JA, Bonner TI, Christian SL, Maheshwari M, Detera-Wadleigh SD, Gibbs RA, Gershon ES. 2003. Polymorphisms at the G72/G30 gene locus, on 13q33, are associated with bipolar disorder in two independent pedigree series. Am J Hum Genet 72:1131–1140. Hoefgen B, Schulze TG, Ohlraun S, von Widdern O, Hofels S, Gross M, Heidmann V, Kovalenko S, Eckermann A, Kolsch H, Metten M, Zobel A, Becker T, Nothen MM, propping P, Heun R, Maier W, Rietschel M. 2005. The power of sample size and homogenous sampling: Association between the 5-HTTLPR serotonin transporter polymorphism and major depressive disorder. Biol Psychiatry 57:247–251. Khakh B. 2001. Molecular physiology of P2X receptors and ATP signalling at synapses. Nat Rev Neurosci 2:165–174. Kim M, Jiang LH, Wilson HL, North RA, Surprenant A. 2001. Proteomic and functional evidence for a P2X7 receptor signalling complex. EMBO J 20:6347–6358. Kim YK, Suh IB, Kim H, Han CS, Lim CS, Choi SH, Licinio J. 2002. The plasma levels of interleukin-12 in schizophrenia, major depression, and bipolar mania: Effects of psychotropic drugs. Mol Psychiatry 7:1107–1114. Kruglyak L. 1999. Prospects for whole-genome linkage disequilibrium mapping of common disease genes. Nat Genet 22:139–144. Le Feuvre R, Brough D, Rothwell N. 2002. Extracellular ATP and P2X7 receptors in neurodegeneration. Eur J Pharmacol 447:261–269. Lyons-Warren A, Chang JJ, Balkissoon R, Kamiya A, Garant M, Nurnberger J, Scheftner W, Reich T, McMahon F, Kelsoe J, Gershon E, Coryell W, Byerley W, Berrettini W, Depaulo R, McInnis M, Sawa A. 2005. Evidence of association between bipolar disorder and Citron on chromosome 12q24. Mol Psychiatry 10:807–809. Maes M. 1995. Evidence for an immune response in major depression: A review and hypothesis. Prog Neuropsychopharmacol Biol Psychiatry 19:11–38. Morissette J, Villeneuve A, Bordeleau L, Rochette D, Laberge C, Gagne B, Laprise C, Bouchard F, Plante M, Gobeil L, Shink E, Weissenbach J, Barden N. 1999. Genome-wide search for linkage of bipolar affective disorders in a very large pedigree derived from a homogeneous population in Quebec points to a locus of major effect on chromosome 12q23-q24. Am J Med Genet 88:567–587. Nakajima T, Jorde LB, Ishigami T, Umemura S, Emi M, Lalouel JM, Inoue I. 2002. Nucleotide diversity and haplotype structure of the human angiotensinogen gene in two populations. Am J Hum Genet 70:108–123. Nath SK, Quintero-Del-Rio AI, Kilpatrick J, Feo L, Ballesteros M, Harley JB. 2004. Linkage at 12q24 with systemic lupus erythematosus (SLE) is established and confirmed in Hispanic and European American families. Am J Hum Genet 74:73–82. North RA. 2002. Molecular physiology of P2X receptors. Physiol Rev 82:1013–1067. Ogden CA, Rich ME, Schork NJ, Paulus MP, Geyer MA, Lohr JB, Kuczenski R, Niculescu AB. 2004. Candidate genes, pathways and mechanisms for bipolar (manic-depressive) and related disorders: An expanded convergent functional genomics approach. Mol Psychiatry 9:1007–1029. Owen MJ, Holmans P, McGuffin P. 1997. Association studies in psychiatric genetics. Mol Psychiatry 2:270–273. Rabinowitz D, Laird N. 2000. A unified approach to adjusting association tests for population admixture with arbitrary pedigree structure and arbitrary missing marker information. Hum Hered 50:211–223. Rassendren F, Buell GN, Virginio C, Collo G, North RA, Surprenant A. 1997. The permeabilizing ATP receptor, P2X7. Cloning and expression of a human cDNA. J Biol Chem 272:5482–5486. Reich DE, Schaffner SF, Daly MJ, McVean G, Mullikin JC, Higgins JM, Richter DJ, Lander ES, Altshuler D. 2002. Human genome sequence variation and the influence of gene history, mutation and recombination. Nat Genet 32:135–142. Saunders RE, Goodship TH, Zipfel PF, Perkins SJ. 2006. An interactive web database of factor H-associated hemolytic uremic syndrome mutations: Insights into the structural consequences of disease-associated mutations. Hum Mutat 27:21–30. Shink E, Morissette J, Sherrington R, Barden N. 2005a. A genome-wide scan points to a susceptibility locus for bipolar disorder on chromosome 12. Mol Psychiatry 10:538–544. Shink E, Harvey M, Tremblay M, Gagne B, Belleau P, Raymond C, Labbé M, Dubé M-P, Lafrenière RG, Barden N. 2005b. Analysis of microsatellite markers and single nucleotide polymorphisms in candidate genes for susceptibility to bipolar affective disorder in the chromosome 12Q24.31 region. Am J Med Genet Part B 135B:50–58. Sim JA, Young MT, Sung HY, North RA, Surprenant A. 2004. Reanalysis of P2X7 receptor expression in rodent brain. J Neurosci 24:6307–6314. Sjoholt G, Ebstein RP, Lie RT, Berle JO, Mallet J, Deleuze JF, Levinson DF, Laurent C, Mujahed M, Bannoura I, Murad I, Molven A, Steen VM. 2004. Examination of IMPA1 and IMPA2 genes in manic-depressive patients: Association between IMPA2 promoter polymorphisms and bipolar disorder. Mol Psychiatry 9:621–629. Solle M, Labasi J, Perregaux DG, Stam E, Petrushova N, Koller BH, Griffiths RJ, Gabel CA. 2001. Altered cytokine production in mice lacking P2X(7) receptors. J Biol Chem 276:125–132. Song C, Dinan T, Leonard BE. 1994. Changes in immunoglobulin, complement and acute phase protein levels in the depressed patients and normal controls. J Affect Disord 30:283–288. Spitzer RL, Williams JBW, Gibbon M. 1987. Structured clinical interview for DSM-III-R. Biometrics Research. New York: New York State Psychiatric Institute. Splendore A, Silva EO, Alonso LG, Richieri-Costa A, Alonso N, Rosa A, Carakushanky G, Cavalcanti DP, Brunoni D, Passos-Bueno MR. 2000. High mutation detection rate in TCOF1 among Treacher Collins syndrome patients reveals clustering of mutations and 16 novel pathogenic changes. Hum Mutat 16:315–322. Suzuki T, Hide I, Ido K, Kohsaka S, Inoue K, Nakata Y. 2004. Production and release of neuroprotective tumor necrosis factor by P2X7 receptoractivated microglia. J Neurosci 24:1–7. Taillon-Miller P, Bauer-Sardina I, Saccone L, Putzel J, Laitinen T, Cao A, Kere J, Pilia G, Rice JP, Kwok PY. 2000. Juxtaposed regions of extensive and minimal linkage disequilibrium in human Xq25 and Xq28. Nat Genet 25:324–328. Walter L, Dinh T, Stella N. 2004. ATP induces a rapid and pronounced increase in 2-arachidonoylglycerol production by astrocytes, a response limited by monoacylglycerol lipase. J Neurosci 24:8068–8074. Wang CM, Chang YY, Sun SH. 2003. Activation of P2X7 purinoceptorstimulated TGF-beta 1 mRNA expression involves PKC/MAPK signalling pathway in a rat brain-derived type-2 astrocyte cell line, RBA-2. Cell Signal 15:1129–1137. Wasserman MJ, Corson TW, Sibony D, Cooke RG, Parikh SV, Pennefather PS, Li PP, Warsh JJ. 2004. Chronic lithium treatment attenuates intracellular calcium mobilization. Neuropsychopharmacology 29:759– 769. Wiley JS. Dao-Ung LP, Li C, Shemon AN, Gu BJ, Smart ML, Fuller SJ, Barden JA, Petrou S, Sluyter R. 2003. An Ile-568 to Asn polymorphism prevents normal trafficking and function of the human P2X7 receptor. J Biol Chem 278:17108–17113. Wilson H, Wilson SA, Surprenant A, North RA. 2002. Epithelial membrane proteins induce membrane blebbing and interact with the P2X7 receptor C terminus. J Biol Chem 277:34017–34023. Wright A, Charlesworth B, Rudan I, Carothers A, Campbell H. 2003. A polygenic basis for late-onset disease. Trends Genet 19:97–106. Zhang X, Gainetdinov RR, Beaulieu JM, Sotnikova TD, Burch LH, Williams RB, Schwartz DA, Krishnan KR, Caron MG. 2005. Loss-of-function mutation in tryptophan hydroxylase-2 identified in unipolar major depression. Neuron 45:11–16.