- Email: [email protected]
An association study on polymorphisms in the PEA15, ENTPD4, and GAS2L1 genes and schizophrenia
Psychiatry Research 185 (2011) 9–15
Contents lists available at ScienceDirect
Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s
An association study on polymorphisms in the PEA15, ENTPD4, and GAS2L1 genes and schizophrenia Atsushi Saito a,⁎, Yuta Fujikura-Ouchi b, Chihiro Ito c, Hiroo Matsuoka c, Kazutaka Shimoda d, Kazufumi Akiyama a a
Department of Biological Psychiatry and Neuroscience, Dokkyo Medical University School of Medicine, 880 Kitakobayashi, Mibu, Japan Kodama Hospital, Ishinomaki, Japan Department of Psychiatry, Tohoku University Graduate School of Medicine, Sendai, Japan d Department of Psychiatry, Dokkyo Medical University School of Medicine, 880 Kitakobayashi, Mibu, Japan b c
a r t i c l e
i n f o
Article history: Received 28 June 2008 Received in revised form 27 August 2009 Accepted 29 September 2009 Keywords: Schizophrenia Serial analysis of gene expression (SAGE) Phosphoprotein enriched in astrocyte 15 (PEA15) Ectonucleoside triphosphate diphosphohydrolase 4 (ENTPD4) Growth arrest-specific 2 like 1 (GAS2L1)
a b s t r a c t Our previous study examined a number of methamphetamine (METH)/phencyclidine (PCP)-reactive tags in rat brain, using a serial analysis of gene expression. Among human homologous genes, which matched METH/PCP-reactive tags, three human genes were identified: phosphoprotein enriched in astrocyte 15 (PEA15), ectonucleoside triphosphate diphosphohydrolase 4 (ENTPD4), and growth arrest-specific 2 like 1 (GAS2L1), which are localized in the chromosome 1q21.1, 8p21.3, and 22q12.2, respectively. We postulated that these genes are plausible candidate genes that play a role in pathogenesis for schizophrenia. Using tagging single-nucleotide polymorphisms (SNPs), we performed a case-control comparison for three SNPs in the PEA15 gene, and six SNPs in the GAS2L1 gene in a sample set of subjects (240 schizophrenia patients and 286 control subjects). Twelve SNPs in the ENTPD4 gene were analyzed in a subset of subjects (94 schizophrenia patients and 94 control subjects). No single SNP displayed a significant difference regarding the allelic frequency or genotypic distribution between the affected cases and controls for any of the genes examined. There was neither a significant difference in the frequency of three marker haplotype in the PEA15 gene or of six marker haplotype in the GAS2L1 gene between the cases and controls. The present study fails to provide evidence for the contribution of PEA15, ENTPD4, and GAS2L1 genes to the etiology of schizophrenia in the Japanese population. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Schizophrenia is a common mental disorder affecting 0.5–1% of the general population at some point in their lives. The aggregation of schizophrenia in families, and the results of many twin and adoption studies support the hypothesis of an inherited risk and predisposition to the disease. We previously reported that allelic frequencies of singlenucleotide polymorphism (SNP) in the 5'-upstream of gene encoding neuroplastin (NPTN), which belongs to the immunoglobulin superfamily, are significantly associated with schizophrenia (Saito et al., 2007). The NPTN gene was deduced from one of a large number of transcripts which underwent up-regulation after acute administration of either methamphetamine (METH) or phencyclidine (PCP) during the serial analysis of gene expression (SAGE)-based screening strategy (Ouchi et al., 2005). From our previous study (Saito et al., 2007), it has been suggested that some susceptibility genes for schizophrenia may be captured from tags which are regulated by METH/PCP, psychostimu-
⁎ Corresponding author. Tel.: +81 282 87 2478; fax: +81 282 86 2538. E-mail address: [email protected] (A. Saito). 0165-1781/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2009.09.018
lants that produce schizophrenia-like symptoms. The present study extends this strategy (Ouchi et al., 2005; Saito et al., 2007) to assess whether sequences of METH/PCP-regulated tags in the SAGE-based screening would predict human homologous genes which are located on chromosomal loci strongly implicated in schizophrenia susceptibility. With regard to this, METH/PCP-reactive tags reported in the supplementary data of the previous study (Ouchi et al., 2005) were matched to those human genes using UniGene (http://www.ncbi.nlm. nih.gov/sites/entrez?db=unigene). As a result, three rodent transcripts, which were down-regulated by either METH or PCP (supplementary data to Ouchi et al., 2005), provided three human homologous genes which correspond to chromosomal regions suggested by a linkage analysis, to contain a susceptibility gene for schizophrenia: phosphoprotein enriched in astrocyte 15 (PEA15), ectonucleoside triphosphate diphosphohydrolase 4 (ENTPD4), and growth arrest-specific 2 like 1 (GAS2L1). PEA15 is an abundant phosphoprotein expressed in brain astrocytes and neurons, and plays a pivotal role in the chain of intra-cellular signal transductions pathways (Condorelli et al., 1999; Renault et al., 2003; Whitehurst et al., 2004). In particular, PEA15 has been shown to inhibit formation of the death-inducing signaling complex, and thus exert an
10
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
antiapoptotic action (Condorelli et al., 1999; Renault et al., 2003; Whitehurst et al., 2004). In the brain, PEA-15 is enriched in progenitor cells which are localized in the subventricular zone, indicating that it may be involved in adult neurogenesis (Sharif et al., 2004). In light of the view that schizophrenia may be associated with impairment of adult neurogenesis (Reif et al., 2006; Toro and Deakin, 2007), the PEA15 gene remains as interesting susceptibility loci in schizophrenia. In addition, overexpression of the PEA15 gene is reported to be correlated with reduced insulin sensitivity in patients with type 2 diabetes (Condorelli et al., 1998). The human PEA15 gene is encoded by a gene of approximately 10 kb, located on chromosome 1q21.1 and consists of 4 exons (OMIM ⁎603434). ENTPD4, which is also known as lysosomal apyrase-like protein of 70 kDa, is one of the enzyme families capable of cleaving nucleotide tri- and diphosphates and participates in salvaging nucleotides from the lysosomal/autophagic vacuole lumen (Biederbick et al., 1999). Sorting mechanisms of lysosome-related organelles have been recently implicated in schizophrenia (Morris et al., 2008; NewellLitwa et al., 2007). The human ENTPD4 gene is encoded by a gene of approximately 30 kb, located on chromosome 8p21.3, and consists of 14 exons. GAS2L1 is a member of a family of genes that are highly expressed in 3T3 fibroblasts during growth arrest (Brancolini et al., 1992). The sequences of the GAS2L1 proteins contain both a putative calponin homology domain (CH) and GAS2-related (GAR) domain (Goriounov et al., 2003). The CH and GAR domains are associated with actin and microtubule, respectively, indicating that the product of GAS2L1 is a component of the microfilament system (Goriounov et al., 2003). Although the function of GAS2L1 remains unknown, it might be involved in the execution of apoptotic process (Lee et al., 1999). Given that schizophrenia might be associated with altered apoptotic process during the early manifestations of the disease (Gottfried et al., 2007), GAS2L1 may be an interesting susceptibility locus in schizophrenia. The human GAS2L1 gene is encoded by a gene of approximately 5.8 kb, located on chromosome 22q12.2 and consists of 5 exons. Therefore, on the basis of both the SAGE-based screening strategy and the positional candidate loci, we assume that these genes may confer susceptibility to schizophrenia. We performed the association analysis of these genes with schizophrenia in a Japanese population using selected genetic variants, such as tagging SNPs, which are proxies for SNPs in local linkage disequilibrium (LD) structures (Gu et al., 2008).
Fig. 1. Genomic structure and linkage disequilibrium (LD) of PEA15 gene. (a) Genomic structure of PEA15 gene and single-nucleotide polymorphisms (SNPs) used in association analyses. PEA15 gene spans over 10 kb, and is composed of 4 exons shown as boxes. The numbers either above or below the vertical arrows represent SNP numbers. SNP labeled with (H) represents a tagging SNP. Two pairs of primers (a pair of F4 and R4, and another of F6 and R6) are indicated by single horizontal lines with arrowheads. (b) LD evaluation of PEA15 gene. The number in each cell represents D′ value after the decimal point between SNPs. D′ values of 1.0 are not shown. A block is defined by a sold spine of LD.
for the screening scan was analyzed for precise determination of the pair-wise LD (Figs. 1b, 2b, and 3b). LD was based on 95% confidence bounds on D, according to Gabriel et al. (2002). Comparison of a categorical variable (gender) was performed using a χ2-test. DNA was extracted from blood using a DNA isolation kit. The SNPs were individually genotyped using either polymerase chain reaction (PCR)-based restriction fragment length polymorphism (PCR-RFLP), direct DNA sequencing of PCR products, or TaqMan® assay (Applied Biosystems, Foster City, CA, USA). Specific primer pairs for PCR-RFLP and direct DNA sequencing of PCR products are shown in Table 1. Three SNPs (rs680083 and rs4656252 for PEA15, and rs6006090 for GAS2L1) and eight SNPs (rs12549049, rs4872136, rs2272641, rs2272642 and rs2272643 for ENTPD4, and rs1894414, rs2301585, and rs2301586 for GAS2L1) were genotyped with PCR-RFLP and with direct DNA sequencing, respectively. Ten SNPs (rs2795070 for PEA15, rs34309579,
2. Methods The total number of subjects consisted of 240 schizophrenia patients (133 males and 107 females; age, 54.9 ± 12.8 years, mean ± S.D.) and 286 control subjects (143 males and 143 females; age: 41.6 ± 12.5 years). Subjects in both groups were from the Tohoku and Kanto regions of Japan. Review of their medical charts reporting the onset and characteristic clinical course clearly showed that the patients had suffered from schizophrenia in their lives prior to the present study. At the beginning of the present study, the diagnosis of schizophrenia was finally confirmed by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) criteria (American Psychiatric Association, 2000) for all cases. The control subjects were unrelated healthy volunteers and did not suffer from any neuropsychiatric disorder. The objective of the present study was clearly explained, and written informed consent was obtained from all subjects in accordance with the Declaration of Helsinki (http://www.wma.net). Since several non-tagging SNPs adopted in the present study have not been validated for their allele frequencies in the Japanese ethnic group, we performed a screening scan in a subset of subjects that were included in the total subjects. This group composed of 94 schizophrenia patients (47 males and 47 females; age, 47.0 ± 8.8 years, mean ± S. D.) and 94 control subjects (47 males and 47 females; age: 42.0 ± 11. 3 years). The study was formally approved by the Institutional Review Board of the Ethical Committees of Tohoku University Graduate School of Medicine and of Dokkyo Medical University School of Medicine. We first consulted the International HapMap project database on dbSNP (http://www. ncbi.nlm.nih.gov/SNP/) and chose 16 tagging SNPs (two for PEA15, 10 for ENTPD4, and four for GAS2L1, shown in Figs. 1a, 2a, and 3a, respectively). We added five non-tagging SNPs which were located around the 5-near region of the respective genes in order to extend the sequenced region and to utilize successfully designed TaqMan probes (SNP browser™). Putative LD blocks of the three genes were estimated according to HAPLOVIEW (http://www.broad.mit.edu/mpg/haploview/), and the subset of subjects
Fig. 2. Genomic structure and linkage disequilibrium (LD) of ENTPD4 gene. (a) Genomic structure of ENTPD4 gene and single-nucleotide polymorphisms (SNPs) used in association analyses. ENTPD4 gene spans over 30 kb, and is composed of 14 exons shown as boxes. The numbers either above or below the vertical arrows represent SNP numbers. SNP labeled with (H) represents a tagging SNP. Two pairs of primers (a pair of F1 and R1, and another of F2 and R2) are indicated by single horizontal lines with arrowheads. (b) LD evaluation of ENTPD4 gene. The number in each cell represents D′ value after the decimal point between SNPs. D′ values of 1.0 are not shown. A block is defined by a sold spine of LD.
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
11
schizophrenia patients. The permutation P-values from the individual test represent the difference in frequency of an individual haplotype between the control and schizophrenia patients. All tests were two-tailed, and statistical significance was P b 0.05.
3. Results
Fig. 3. Genomic structure and linkage disequilibrium (LD) of GAS2L1 gene. (a) Genomic structure of GAS2L1 gene and single-nucleotide polymorphisms (SNPs) used in association analyses. GAS2L1 gene spans over 5.8 kb, and is composed of 5 exons shown as boxes. The numbers either above or below the vertical arrows represent SNP numbers. SNP labeled with (H) represents a tagging SNP. Two pairs of primers (a pair of F1 and R1, and another of F3 and R3) are indicated by single horizontal lines with arrowheads. (b) LD evaluation of GAS2L1 gene. The number in each cell represents D′ value after the decimal point between SNPs. A block is defined by a sold spine of LD.
rs6557673, rs11777391, rs2293936, rs2272640, rs2280832, and rs1061293 for ENTPD4, and rs174761 and rs174762 for GAS2L1) were genotyped with TaqMan® assay. PCR amplification was performed in a total volume of 5 μl reaction mixture containing 1 × PCR Buffer, 2.5 mM MgCl2, 0.5 μM of each primer, 0.25 mM of each dNTP, 5 ng of the genomic DNA, and 0.25 U of Taq DNA polymerase. LA Taq polymerase (Takara, Japan) was used for rs680083, rs4656252, rs1894414, rs2301585, rs2301586, and rs6006090. Taq DNA polymerase (Sigma, St. Louis, MO, USA) was used for the rest of SNPs. The PCR amplification was conducted under the following conditions: an initial denaturation phase at 95 °C for 2 min, followed by 40 cycles of 95 °C for 15–30 s, 58– 72 °C (details shown in Table 1) for 15–30 s, and 72 °C for 30 s–1 min, followed by a final extension phase of 72 °C for 5 min. Among the SNPs genotyped with direct DNA sequencing, the PCR products containing rs12549049 and rs4872136 were amplified using nested PCR. In PCR-RFLP, each 5 μl sample of PCR product was completely digested with 5 U of restriction enzyme, ScrFI for rs680083, BseYI for rs4656252, and AvaII for rs6006090. The resulting digested product was separated on 2–3% agarose gel, which was subsequently stained with ethidium bromide. Identification of the three SNPs, which were genotyped with PCR-RFLP, was confirmed by direct DNA sequencing in 16 randomly selected samples (eight controls and eight cases in the screening sample set). In direct DNA sequencing of PCR reactions, the PCR products were purified and sequenced with a BigDye Terminator Cycle Sequencing Reagent (Applied Biosystems, Foster City, CA, USA) on an ABI PRISM 3100 DNA Sequencer (Applied Biosystems, Foster City, CA, USA). Among SNPs which are eligible for TaqMan® assay, rs2795070, rs174761 and rs174762 were analyzed using the Takara Cycler Dice ™ Real Time System, and the rest of the seven SNPs were outsourced to be analyzed at the HuBit Genomics (Tokyo). All alleles and genotypes were expressed as those in a coding strand of the genes. Deviations of each SNP from the Hardy–Weinberg equilibrium (HWE) were analyzed by χ2 goodness-of-fit tests with a degree of freedom being one. Statistical differences in genotypic distribution and allelic frequencies between the schizophrenia patients and control subjects were evaluated by Fischer's exact tests. An odds ratio (OR) and its 95% confidence intervals (CIs) were determined based on a contingency table for allelic frequencies of each SNP. SNPAyze 5.1. standard program (Dynacom, Japan) was used to obtain the permutation P-values for constructed haplotype and genetic associations at polymorphism loci. The global test permutation P-values assess the overall difference in the distribution of haplotype frequencies between the control and
There were no significant differences in gender between the schizophrenia and control subjects. The average age was significantly higher in the schizophrenia patients than the control subjects (P b 0.0001, Student's t test). The direct DNA sequence-based genotyping of three SNPs (rs680083, rs4656252, and rs6006090) in 16 randomly selected samples was identical to those shown by PCR-RFLP (data not shown). Accordingly, we speculate that it was unlikely that a genotyping error had occurred. The genotypic distributions of all SNPs detected in the sample set did not deviate from the HWE. The LD block structures, which encompass the three genes, are shown in Figs. 1b, 2b, and 3b. Based on the screening scan sample subset, all SNPs investigated for PEA15 and GAS2L1 were placed on the entire genomic region with a single strong LD block. The SNPs in the ENTPD4 gene showed a strong, but not single, LD block. In all cases of the total sample set, we genotyped nine SNPs (three for PEA15 and six for GAS2L1), and constructed a haplotypic combination using these selected SNPs. Table 2 summarizes results of the allelic frequency and genotypic distribution for the PEA15 gene and GAS2L1 gene for the total sample set of subjects (240 schizophrenia patients and 286 control subjects). We circumscribed genotyping of 12 SNPs in the ENTPD4 gene within the screening scan sample subset (94 schizophrenia patients and 94 control subjects) due to high P values (N0.5) at this stage (Table 2). No single marker showed significant association with schizophrenia with regard to the allelic frequency or genotypic distribution for any of the genes examined. The results of the haplotype analyses are summarized in Table 3. Haplotype which is constructed of three SNPs (rs2795070(T/C)rs680083(A/G)-rs4656252(G/A)) in the strong LD block of the PEA15 gene (Fig. 1b) failed to show an overall difference in frequency between the affected cases and controls (global P value = 0.1097). When individual haplotypes were examined, C-G-A, a minor combination of the same haplotype, was underrepresented in the cases only at a marginally significant level (P = 0.0367). A second haplotype defined by six SNPs (rs174761(C/T)-rs1894414(A/C)rs2301585(G/C)-rs2301586(T/A)-rs6006090(A/G)-rs174762 (C/T)) in the GAS2L1 gene did not show a significant association with schizophrenia regarding the global P value or individual haplotypic p value. 4. Discussion The three genes (PEA15, ENTPD4, and GAS2L1) which were analyzed in the present study are localized in 1q21.1, 8p21.3, and 22q12.2, all of which have already been shown as positional candidate regions associated with schizophrenia in previous linkage studies (Brzustowicz et al., 2000; Gurling et al., 2001; Kendler et al., 2000; Takahashi et al., 2003). Individual genes mapped to this region can be considered to be candidates for schizophrenia susceptibility genes. Several candidate genes on these loci have indeed been reported to be associated with
Table 1 Forward primer
Primer sequence
Reverse primer
Primer sequence
Ta (°C)
PEA15 F4 PEA15 F6 ENTPD4 F1 ENTPD4 F2 GAS2L1 F1 GAS2L1 F3
5′-ATTGCCCCTATCTTGTGC-3′ 5′-CATGAGTTCCTACCAGACC-3′ 5′-CAAGAATGAAATCCTGGCTCTAC-3′ 5′-GTTTGTATCCTGACTCTCC-3′ 5′-TGAGCATTCAGTGAGCACAAG-3′ 5′-TGGACCAGGGTCACTCAAGG-3′
PEA15 R4 PEA15 R6 ENTPD4 R1 ENTPD4 R2 GAS2L1 R1 GAS2L1 R3
5′-TCTGTGTTCCTATTGCTGG-3′ 5′-GTTACCATAGCAACCTGC-3′ 5′-CACCAGTGGGCAGCATC-3′ 5′-TGCTGTGACTATTCAACC-3′ 5′-CGAGCCGCAGCTCTATG-3′ 5′-ATGGGGTACAGCTTCCACACG-3′
58 60 62 62 58 72
DMSO(+) means that adding 1/10 volume of dimethyl sulfoxide in each PCR reaction. Ta (°C) means annealing temperature.
DMSO
+ +
Method RFLP RFLP sequence sequence sequence RFLP
12
SNP number
IDs on dbSNPs
PEA15(1q21.1) SNP1
rs2795070
SNP2
SNP3
ENTPD4(8p21.3) SNP1
SNP2
SNP3
SNP4
SNP5
SNP6
SNP7
Groups
n
Schizophrenics Controls
n = 240 n = 286
Schizophrenics Controls
n = 237 n = 286
Schizophrenics Controls
n = 238 n = 286
Schizophrenics Controls
n = 91 n = 92
Schizophrenics Controls
n = 94 n = 89
Schizophrenics Controls
n = 93 n = 89
Schizophrenics Controls
n = 88 n = 94
Schizophrenics Controls
n = 88 n = 94
Schizophrenics Controls
n = 92 n = 94
Schizophrenics Controls
n = 91 n = 91
rs680083
rs4656252
rs34309579
rs12549049
rs4872136
rs6557673
rs11777391
rs2293936
rs2272640
Genotype count (frequency)
P-value
T/T 75(0.31) 73(0.26) A/A 75(0.32) 73(0.25) G/G 224(0.94) 256(0.89)
C/T 105(0.44) 150(0.52) A/G 105(0.44) 148(0.52) A/G 14(0.06) 29(0.10)
C/C 60(0.25) 63(0.22) G/G 57(0.24) 65(0.23) A/A 0(0) 1(0.01)
C/C 10(0.11) 15(0.16) C/C 62(0.66) 65(0.73) T/T 33(0.35) 32(0.36) C/C 18(0.20) 16(0.17) G/G 38(0.43) 34(0.36) A/A 81(0.88) 81(0.86) A/A 38(0.42) 33(0.36)
C/T 41(0.45) 38(0.41) C/T 28(0.30) 20(0.22) A/T 43(0.46) 40(0.45) C/T 35(0.40) 44(0.47) G/T 37(0.42) 44(0.47) A/C 10(0.11) 12(0.13) A/G 39(0.43) 44(0.48)
T/T 40(0.44) 39(0.43) T/T 4(0.04) 4(0.05) A/A 18(0.19) 17(0.19) T/T 35(0.40) 34(0.36) T/T 13(0.15) 16(0.17) C/C 1(0.01) 1(0.01) G/G 14(0.15) 14(0.16)
0.1356 (χ2 = 4.0495) 0.1759 (χ2 = 3.4768) 0.0934 (χ2 = 4.0025)
0.5971 (χ2 = 1.1212) 0.5209 (χ2 = 1.2685) 1 (χ2 = 0.0158) 0.6355 (χ2 = 0.9607) 0.6217 (χ2 = 0.9407) 0.9101 (χ2 = 0.1603) 0.7478 (χ2 = 0.6533)
Allele count (frequency) T 255(0.53) 296(0.52) A 255(0.54) 294(0.51) G 462(0.97) 541(0.95)
C 225(0.47) 276(0.48) G 218(0.46) 278(0.49) A 14(0.03) 31(0.05)
C 61(0.34) 68(0.37) C 152(0.81) 150(0.86) T 109(0.58) 104(0.58) C 71(0.40) 76(0.40) G 113(0.64) 112(0.60) A 172(0.93) 174(0.92) A 115(0.63) 110(0.60)
T 121(0.66) 116(0.63) T 36(0.19) 28(0.14) A 79(0.42) 74(0.42) T 105(0.60) 112(0.60) T 63(0.36) 76(0.40) C 12(0.07) 14(0.07) G 67(0.37) 72(0.40)
P-value
Odds ratio (95%CI)
0.6649 (χ2 = 0.1983)
0.9463 (0.7422−1.2066)
0.4552 (χ2 = 0.6048)
0.9077 (0.711–1.1587)
0.0652 (χ2 = 3.8832)
0.5288 (0.278–1.0062)
0.5128 (χ2 = 0.4744)
0.86 (0.5598−1.3212)
0.4115 (χ2 = 0.7406)
0.7881 (0.4579−1.3565)
1 (χ2 = 0.0076)
1.0186 (0.6722−1.5435)
1 (χ2 = 0.0003)
0.9965 (0.6554−1.5151)
0.3888 (χ2 = 0.8256)
0.1271 (0.7965−1.8599)
0.8396 (χ2 = 0.1224)
1.1533 (0.5185−2.5649)
0.6662 (χ2 = 0.291)
1.1235 (0.7359−1.7152)
Method TaqMan
RFLP
RFLP
TaqMan
Sequence
Sequence
TaqMan
TaqMan
TaqMan
TaqMan
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
Table 2 Allelic frequencies and genotypic distribution of 21 SNPs in thePEA15, ENTPD4, and GAS2L1 genes.
SNP8
SNP9
SNP10
SNP11
SNP12
GAS2L1(22q12.2) SNP1
SNP2
SNP4
SNP5
SNP6
Schizophrenics Controls
n = 91 n = 92
Schizophrenics Controls
n = 94 n = 94
Schizophrenics Controls
n = 94 n = 94
Schizophrenics Controls
n = 94 n = 94
Schizophrenics Controls
n = 94 n = 92
Schizophrenics Controls
n = 240 n = 286
Schizophrenics Controls
n = 239 n = 283
Schizophrenics Controls
n = 239 n = 282
Schizophrenics Controls
n = 239 n = 282
Schizophrenics Controls
n = 238 n = 286
Schizophrenics Controls
n = 240 n = 286
rs2272641
rs2272642
rs2272643
rs1061293
rs174761
rs1894414
rs2301585
rs2301586
rs6006090
rs174762
A/A 7(0.08) 9(0.10) A/A 41(0.44) 45(0.48) C/C 1(0.01) 1(0.01) A/A 1(0.01) 1(0.01) A/A 1(0.01) 1(0.01)
A/G 44(0.48) 38(0.41) A/G 46(0.49) 40(0.42) C/T 10(0.11) 12(0.13) A/G 10(0.11) 12(0.13) A/G 11(0.12) 12(0.13)
G/G 40(0.44) 45(0.49) G/G 7(0.07) 9(0.10) T/T 83(0.88) 81(0.86) G/G 83(0.88) 81(0.86) G/G 82(0.87) 79(0.86)
C/C 121(0.50) 156(0.55) A/A 217(0.92) 259(0.91) G/G 123(0.52) 156(0.55) T/T 106(0.44) 140(0.50) A/A 118(0.50) 155(0.54) C/C 123(0.51) 159(0.56)
C/T 97(0.41) 103(0.36) A/C 20(0.07) 23(0.08) C/G 96(0.40) 100(0.36) A/T 104(0.44) 108(0.38) A/G 98(0.41) 104(0.36) C/T 95(0.40) 100(0.35)
T/T 22(0.09) 27(0.09) C/C 2(0.01) 1(0.01) C/C 20(0.08) 26(0.09) A/A 29(0.12) 34(0.12) G/G 22(0.09) 27(0.10) T/T 22(0.09) 27(0.09)
Fischer's exact test was used to compare allelic frequencies and genotypic distribution between the patients and controls.
0.6654 (χ2 = 0.9777) 0.6881 (χ2 = 0.8547) 0.91 (χ2 = 0.2062) 0.91 (χ2 = 0.2062) 0.9127 (χ2 = 0.0779)
0.5767 (χ2 = 1.0982) 0.8074 (χ2 = 0.5436) 0.5407 (χ2 = 1.2269) 0.4476 (χ2 = 1.6337) 0.5131 (χ2 = 1.3172) 0.4901 (χ2 = 0.538)
A 58(0.32) 56(0.30) A 128(0.68) 130(0.69) C 12(0.06) 14(0.07) A 12(0.06) 14(0.07) A 13(0.07) 14(0.08)
G 124(0.68) 128(0.70) G 60(0.32) 58(0.31) T 176(0.94) 174(0.93) G 176(0.94) 174(0.93) G 175(0.93) 170(0.92)
C 339(0.71) 415(0.73) A 454(0.95) 541(0.96) G 342(0.72) 412(0.69) T 255(0.72) 388(0.69) A 334(0.70) 414(0.72) C 341(0.71) 418(0.73)
T 141(0.29) 157(0.27) C 24(0.05) 25(0.04) C 136(0.28) 152(0.31) A 101(0.29) 176(0.31) G 142(0.30) 158(0.28) T 139(0.29) 154(0.27)
TaqMan 0.8216 (χ2 = 0.0877)
0.9353 (0.6009−1.456)
0.9115 (χ2 = 0.0494)
0.9518 (0.6156−1.4715)
0.8394 (χ2 = 0.1653)
1.1801 (0.5308−2.6237)
0.8394 (χ2 = 0.1653)
1.1801 (0.5308−2.6237)
0.8435 (χ2 = 0.0665)
1.1086 (0.5062−2.4278)
0.493 (χ2 = 0.4776)
1.0994 (0.8402−1.4386)
0.6621 (χ2 = 0.2113)
1.144 (0.6445−2.0306)
0.6266 (χ2 = 0.2917)
0.9278 (0.7067−1.218)
0.388 (χ2 = 0.8514)
0.8848 (0.6822−1.1475)
0.4506 (χ2 = 0.6208)
1.114 (0.0.8516–1.4573)
0.538 (χ2 = 0.9335)
1.1064 (0.8444–1.4497)
Sequence
Sequence
Sequence
TaqMan
TaqMan
Sequence
Sequence
Sequence
RFLP
TaqMan
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
SNP3
rs2280832
13
14
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
Table 3 PEA15 haplotype in Japanese case-control subjects Gene
Haplotype
Schizophrenics
Frequency
Controls
PEA15
T-A-G 248 0.537 294 C-G-G 201 0.435 243 C-G-A 13 0.028 31 Haplotype pairs mean rs2795070(T/C)-rs680083(A/G)-rs4656252(G/A) in PEA15. Global P-value = 0.1097. Permutation P-values were calculated by SNPAyze 5.1. standard program (Dynacom, Japan).
Frequency
Permutation P-value
0.517 0.428 0.055
0.5406 0.8275 0.0367
GAS2L1 haplotype in Japanese case-control subjects Gene
Haplotype
Schizophrenics
Frequency
Controls
GAS2L1
Frequency
C-A-G-T-A-C 307 0.670 379 0.694 T-A-C-A-G-T 128 0.280 143 0.262 C-C-G-A-A-C 23 0.050 24 0.044 Haplotype pairs mean rs174761(C/T)-rs1894414(A/C)-rs2301585(G/C)-rs2301586(T/A)-rs6006090(A/G)-rs174762(C/T) in GAS2L1. Global P-value = 0.7100. Permutation P-values were calculated by SNPAyze 5.1. standard program (Dynacom, Japan).
schizophrenia in case-control and family-based studies (Liu et al., 2007; Nakata et al., 2003; Ni et al., 2007; Porton et al., 2004). In the present study, we focused on analyses of PEA15 and GAS2L1 genes due to the following reasons: (1) all SNPs investigated for PEA15 and GAS2L1 were placed on a single strong LD block, and (2) these two genes may be involved in key signaling pathways for neurogenesis and apoptosis (Lee et al., 1999; Renault et al., 2003; Sharif et al., 2004; Whitehurst et al., 2004). No significant differences in genotypic and allelic frequencies of the selected SNPs nor frequencies of multi-marker haplotype were observed between the affected cases and controls. Combining these findings together, the present study suggests that PEA15 and GAS2L1 do not play an important role in susceptibility to schizophrenia in the Japanese population. We left ENTPD4 out of further analysis beyond the screening scan sample subset, since high P values for allelic frequencies suggest that this gene is also unlikely to be related to schizophrenia. In the present study, tagging SNPs which have been increasingly utilized in genetic studies of schizophrenia were applied to minimize the size of the core set of SNPs by the reducing redundancy among the SNPs, since no previous affected case-control comparison data regarding SNPs in the PEA15, ENTPD4, and GAS2L1 genes existed. The failure to demonstrate association with the three genes is unlikely to be due to population stratification, since both groups of individuals were drawn from the ethnically homogenous population of the Tohoku and Kanto areas of Japan. The finding that most tagging SNPs employed hereby based on the HapMap data are associated with strong LD, lends support to the theory of favorable transferability across different ethnicities (Gu et al., 2008). Several caveats should be considered when interpreting our results. The first concerns the ability of tagging SNPs to capture rare SNPs. The rare SNPs with minor allele frequency of b5% are generally excluded from a HapMap scheme (Gu et al., 2008). Thus, it remains possible that an as-yet unidentified causative mutation might be involved in pathogenesis of schizophrenia. In addition, most of the tagging SNPs investigated in the present study exist in introns, 5-near regions and 3-near regions with unknown function. Given that SNPs which are not found in the LD may be associated with schizophrenia, further analyses on the basis of detailed SNP coverage of these genes are necessary. The second concerns significant differences in the average age between the controls and schizophrenics in the present study. However, this should not be regarded as being a shortage, since SNP frequencies are generally unlikely to change with age (Lohoff et al., 2005). Finally, the study reporting the plausible involvement of the overexpression of the PEA15 gene in type 2 diabetes mellitus (Condorelli et al., 1998) is worthy of attention. Although this has been shown to occur in fibroblasts of patients with diabetes mellitus (Condorelli et al., 1998), Wolford et al. (2000) reported that none of
Permutation P-value 0.4369 0.5398 0.6266
the three SNPs in the PEA15 gene are associated with type 2 diabetes mellitus in Pima Indians, a population with the highest documented prevalence of this disease worldwide. Intriguingly, the two SNPs they (Wolford et al., 2000) examined appear to correspond to the two SNPs (rs2795070 and rs680083) analyzed in the current study. Despite the general understanding that the risk of developing diabetes mellitus is accelerated by the administration of antipsychotics, recent studies on medication-naïve patients with schizophrenia do not support the hypothesis that schizophrenia is associated with an underlying abnormality predicting the occurrence of diabetes mellitus (PerezIglesias et al., 2009; Sengupta et al., 2008). Combining these findings together, it is likely that the PEA15 gene variation is involved in neither schizophrenia nor diabetes mellitus. Nonetheless, the possibility that PEA15 may be involved somehow in glucose metabolism, or other alterations of glucose homeostasis in the context of antipsychotic-induced liability, in developing diabetes mellitus is a topic that in the future might be worthwhile to study. In conclusion, the present results suggest that PEA15, ENTPD4, and GAS2L1 genes do not confer susceptibility to schizophrenia in the Japanese population. However, further research on these genes, which are located on schizophrenia susceptibility loci, using different SNPs and a larger sample set will be required.
Acknowledgments The present study was supported by Dokkyo Medical University, Young Investigator Award, and the Kobayashi Magobe Memorial Medical Foundation.
References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. American Psychiatric Association, Washington DC. Biederbick, A., Rose, S., Elsässer, H.-P., 1999. A human intracellular apyrase-like protein, LALP70, localizes to lysosomal/autophagic vacuoles. Journal of Cell Science 112, 2473–2484. Brancolini, C., Bottega, S., Schneider, C., 1992. Gas2, a growth arrest-specific protein, is a component of the microfilament network system. The Journal of Cell Biology 117, 1251–1261. Brzustowicz, L.M., Hodgkinson, K.A., Chow, E.W.C., Honer, W.G., Bassett, A.S., 2000. Location of a major susceptibility locus for familial schizophrenia on chromosome 1q21–q22. Science 288, 678–682. Condorelli, G., Vigliotta, G., Cafieri, A., Trencia, A., Andalò, P., Oriente, F., Miele, C., Caruso, M., Formisano, P., Beguinot, F., 1999. PED/PEA-15: an anti-apoptotic molecule that regulates FAS/TNFR1-induced apoptosis. Oncogene 18, 4409–4415. Condorelli, G., Vigliotta, G., Iavarone, C., Caruso, M., Tocchetti, C.G., Andreozzi, F., Cafieri, A., Tecce, M.F., Formisano, P., Beguinot, L., Beguinot, F., 1998. PED/PEA-15 gene controls glucose transport and is overexpressed in type 2 diabetes mellitus. The EMBO Journal 17, 3858–3866. Gabriel, S.B., Schaffner, S.F., Nguyen, H., Moore, J.M., Roy, J., Blumenstiel, B., Higgins, J., DeFelice, M., Lochner, A., Faggart, M., Liu-Cordero, S.N., Rotimi, C., Adeyemo, A., Cooper, R., Ward, R., Lander, E.S., Daly, M.J., Altshuler, D., 2002. The structure of haplotype blocks in the human genome. Science 296, 2225–2229.
A. Saito et al. / Psychiatry Research 185 (2011) 9–15 Goriounov, D., Leung, C.L., Liem, R.K.H., 2003. Protein products of human Gas2-related genes on chromosomes 17 and 22 (hGAR17 and hGAR22) associate with both microfilaments and microtubules. Journal of Cell Science 116, 1045–1058. Gottfried, Y., Rotem, A., Klein, E., Larisch, S., 2007. The pro-apoptotic ARTS/Sept4 protein is significantly reduced in post-mortem brains from schizophrenic patients. Schizophrenia Research 96, 257–266. Gu, C.C., Yu, K., Rao, D.C., 2008. Characterization of LD structures and the utility of HapMap in genetic association studies. Advances in Genetics 60, 407–435. Gurling, H.M.D., Kalsi, G., Brynjolfson, J., Sigmundsson, T., Sherrington, R., Mankoo, B.S., Read, T., Murphy, P., Blaveri, E., McQuillin, A., Petursson, H., Curtis, D., 2001. Genomewide genetic linkage analysis confirms the presence of susceptibility loci for schizophrenia, on chromosomes 1q32.2, 5q33.2, and 8p21–22 and provides support for linkage to schizophrenia, on chromosomes 11q23.3–24 and 20q12.1– 11.23. American Journal of Human Genetics 68, 661–673. Kendler, K.S., Myers, J.M., O'Neill, F.A., Martin, R., Murphy, B., MacLean, C.J., Walsh, D., Straub, R.E., 2000. Clinical features of schizophrenia and linkage to chromosomes 5q, 6p, 8p, and 10p in the Irish study of high-density schizophrenia families. The Amercan Journal of Psychiatry 157, 402–408. Lee, K.K.H., Tang, M.K., Yew, D.T.W., Chow, P.H., Yee, S.P., Schneider, C., Brancolini, C., 1999. Gas2 is a multifunctional gene involved in the regulation of apoptosis and chondrogenesis in the developing mouse limb. Developmental Biology 207, 14–25. Liu, Y.L., Fann, C.S.J., Liu, C.M., Chang, C.C., Yang, W.C., Hung, S.I., Yu, S.L., Hwang, T.J., Hsieh, M.H., Liu, C.C., Tsuang, M.M., Wu, J.Y., Jou, Y.S., Faraone, S.V., Tsuang, M.T., Chen, W.J., Hwu, H.-G., 2007. More evidence supports the association of PPP3CC with schizophrenia. Molecular Psychiatry 12, 966–974. Lohoff, F.W., Ferraro, T.N., McNabb, L., Schwebel, C., Dahl, J.P., Doyle, G.A., Buono, R.J., Berrettini, W.H., 2005. No association between common variations in the neuronal nicotinic acetylcholine receptor alpha2 subunit gene (CHRNA2) and bipolar I disorder. Psychiatry Research 135, 171–177. Morris, D.W., Murphy, K., Kenny, N., Purcell, S.M., McGhee, K.A., Schwaiger, S., Nangle, J.-M., Donohue, G., Clarke, S., Scully, P., Quinn, J., Meagher, D., Baldwin, P., Crumlish, N., O'Callaghan, E., Waddington, J.L., Gill, M., Corvin, A.P., 2008. Dysbindin (DTNBP1) and the biogenesis of lysosome-related organelles complex 1 (BLOC-1): main and epistatic gene effects are potential contributors to schizophrenia susceptibility. Biological Psychiatry 63, 24–31. Nakata, K., Ujike, H., Sakai, A., Takaki, M., Imamura, T., Tanaka, Y., Kuroda, S., 2003. The human dihydropyrimidinase-related protein 2 gene on chromosome 8p21 is associated with paranoid-type schizophrenia. Biological Psychiatry 53, 571–576. Newell-Litwa, K., Seong, E., Burmeister, M., Faundez, V., 2007. Neuronal and non-neuronal functions of the AP-3 sorting machinery. Journal of Cell Science 120, 531–541. Ni, X., Valente, J., Azevedo, M.H., Pato, M.T., Pato, C.N., Kennedy, J.L., 2007. Connexin 50 gene on human chromosome 1q21 is associated with schizophrenia in matched case-control and family-based studies. Journal of Medical Genetics 44, 532–536.
15
Ouchi, Y., Kubota, Y., Kuramasu, A., Watanabe, T., Ito, C., 2005. Gene expression profiling in whole cerebral cortices of phencyclidine- or methamphetamine-treated rats. Molecular Brain Research 140, 142–149. Perez-Iglesias, R., Mata, I., Relayo-Teran, J.M., Amado, J.A., Garcia-Unzueta, M.T., Berja, A., Martinez-Garcia, O., Vazquez-Barquero, J.L., Crespo-Facorro, B., 2009. Glucose and lipid disturbances after 1 year of antipsychotic treatment in a drug-naïve population. Schizophrenia Research 107, 115–121. Porton, B., Ferreira, A., DeLisi, L.E., Kao, H.-T., 2004. A rare polymorphism affects a mitogen-activated protein kinase site in synapsin III: possible relationship to schizophrenia. Biological Psychiatry 55, 118–125. Reif, A., Fritzen, S., Finger, M., Strobel, A., Lauer, M., Schmitt, A., Lesch, K.-P., 2006. Neural stem cell proliferation is decreased in schizophrenia, but not in depression. Molecular Psychiatry 11, 514–522. Renault, F., Formstecher, E., Callebaut, I., Junier, M.-P., Chneiweiss, H., 2003. The multifunctional protein PEA-15 is involved in the control of apoptosis and cell cycle in astrocytes. Biochemical Pharmacology 66, 1581–1588. Saito, A., Fujikura-Ouchi, Y., Kuramasu, A., Shimoda, K., Akiyama, K., Matsuoka, H., Ito, C., 2007. Association study of putative promoter polymorphisms in the neuroplastin gene and schizophrenia. Neuroscience Letters 411, 168–173. Sengupta, S., Parrilla-Escobar, M.A., Klink, R., Fathalli, F., Ying, K.N., Stip, E., Baptista, T., Malla, A., Joober, R., 2008. Are metabolic indices different between drug-naïve first-episode psychosis patients and healthy controls? Schizophrenia Research 102, 329–336. Sharif, A., Renault, F., Beuvon, F., Castellanos, R., Canton, B., Barbeito, L., Junier, M.P., Chneiweiss, H., 2004. The expression of PEA-15 (phosphoprotein enriched in astrocytes of 15KDa) defines subpopulations of astrocytes and neurons throughout the adult mouse brain. Neuroscience 126, 263–275. Takahashi, S., Cui, Y.-h., Kojima, T., Han, Y.-h., Zhou, R.-l., Kamioka, M., Yu, S.-y., Matsuura, M., Matsushima, E., Wilcox, M., Arinami, T., Shen, Y.-c., Faraone, S.V., Tsuang, M.T., 2003. Family-based association study of markers on chromosome 22 in schizophrenia using African-American, European-American, and Chinese families. American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 120B, 11–17. Toro, C.T., Deakin, J.F.W., 2007. Adult neurogenesis and schizophrenia: A window on abnormal early brain development? Schizophrenia Research 90, 1–14. Whitehurst, A.W., Robinson, F.L., Moore, M.S., Cobb, M.H., 2004. The death effector domain protein PEA-15 prevents nuclear entry of ERK2 by inhibiting required interactions. The Journal of Biological Chemistry 279, 12840–12847. Wolford, J.K., Bogardus, C., Ossowski, V., Prochazka, M., 2000. Molecular characterization of the human PEA15 gene on 1q21–q22 and association with type 2 diabetes mellitus in Pima Indians. Gene 241, 143–148.
Contents lists available at ScienceDirect
Psychiatry Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s
An association study on polymorphisms in the PEA15, ENTPD4, and GAS2L1 genes and schizophrenia Atsushi Saito a,⁎, Yuta Fujikura-Ouchi b, Chihiro Ito c, Hiroo Matsuoka c, Kazutaka Shimoda d, Kazufumi Akiyama a a
Department of Biological Psychiatry and Neuroscience, Dokkyo Medical University School of Medicine, 880 Kitakobayashi, Mibu, Japan Kodama Hospital, Ishinomaki, Japan Department of Psychiatry, Tohoku University Graduate School of Medicine, Sendai, Japan d Department of Psychiatry, Dokkyo Medical University School of Medicine, 880 Kitakobayashi, Mibu, Japan b c
a r t i c l e
i n f o
Article history: Received 28 June 2008 Received in revised form 27 August 2009 Accepted 29 September 2009 Keywords: Schizophrenia Serial analysis of gene expression (SAGE) Phosphoprotein enriched in astrocyte 15 (PEA15) Ectonucleoside triphosphate diphosphohydrolase 4 (ENTPD4) Growth arrest-specific 2 like 1 (GAS2L1)
a b s t r a c t Our previous study examined a number of methamphetamine (METH)/phencyclidine (PCP)-reactive tags in rat brain, using a serial analysis of gene expression. Among human homologous genes, which matched METH/PCP-reactive tags, three human genes were identified: phosphoprotein enriched in astrocyte 15 (PEA15), ectonucleoside triphosphate diphosphohydrolase 4 (ENTPD4), and growth arrest-specific 2 like 1 (GAS2L1), which are localized in the chromosome 1q21.1, 8p21.3, and 22q12.2, respectively. We postulated that these genes are plausible candidate genes that play a role in pathogenesis for schizophrenia. Using tagging single-nucleotide polymorphisms (SNPs), we performed a case-control comparison for three SNPs in the PEA15 gene, and six SNPs in the GAS2L1 gene in a sample set of subjects (240 schizophrenia patients and 286 control subjects). Twelve SNPs in the ENTPD4 gene were analyzed in a subset of subjects (94 schizophrenia patients and 94 control subjects). No single SNP displayed a significant difference regarding the allelic frequency or genotypic distribution between the affected cases and controls for any of the genes examined. There was neither a significant difference in the frequency of three marker haplotype in the PEA15 gene or of six marker haplotype in the GAS2L1 gene between the cases and controls. The present study fails to provide evidence for the contribution of PEA15, ENTPD4, and GAS2L1 genes to the etiology of schizophrenia in the Japanese population. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Schizophrenia is a common mental disorder affecting 0.5–1% of the general population at some point in their lives. The aggregation of schizophrenia in families, and the results of many twin and adoption studies support the hypothesis of an inherited risk and predisposition to the disease. We previously reported that allelic frequencies of singlenucleotide polymorphism (SNP) in the 5'-upstream of gene encoding neuroplastin (NPTN), which belongs to the immunoglobulin superfamily, are significantly associated with schizophrenia (Saito et al., 2007). The NPTN gene was deduced from one of a large number of transcripts which underwent up-regulation after acute administration of either methamphetamine (METH) or phencyclidine (PCP) during the serial analysis of gene expression (SAGE)-based screening strategy (Ouchi et al., 2005). From our previous study (Saito et al., 2007), it has been suggested that some susceptibility genes for schizophrenia may be captured from tags which are regulated by METH/PCP, psychostimu-
⁎ Corresponding author. Tel.: +81 282 87 2478; fax: +81 282 86 2538. E-mail address: [email protected] (A. Saito). 0165-1781/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2009.09.018
lants that produce schizophrenia-like symptoms. The present study extends this strategy (Ouchi et al., 2005; Saito et al., 2007) to assess whether sequences of METH/PCP-regulated tags in the SAGE-based screening would predict human homologous genes which are located on chromosomal loci strongly implicated in schizophrenia susceptibility. With regard to this, METH/PCP-reactive tags reported in the supplementary data of the previous study (Ouchi et al., 2005) were matched to those human genes using UniGene (http://www.ncbi.nlm. nih.gov/sites/entrez?db=unigene). As a result, three rodent transcripts, which were down-regulated by either METH or PCP (supplementary data to Ouchi et al., 2005), provided three human homologous genes which correspond to chromosomal regions suggested by a linkage analysis, to contain a susceptibility gene for schizophrenia: phosphoprotein enriched in astrocyte 15 (PEA15), ectonucleoside triphosphate diphosphohydrolase 4 (ENTPD4), and growth arrest-specific 2 like 1 (GAS2L1). PEA15 is an abundant phosphoprotein expressed in brain astrocytes and neurons, and plays a pivotal role in the chain of intra-cellular signal transductions pathways (Condorelli et al., 1999; Renault et al., 2003; Whitehurst et al., 2004). In particular, PEA15 has been shown to inhibit formation of the death-inducing signaling complex, and thus exert an
10
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
antiapoptotic action (Condorelli et al., 1999; Renault et al., 2003; Whitehurst et al., 2004). In the brain, PEA-15 is enriched in progenitor cells which are localized in the subventricular zone, indicating that it may be involved in adult neurogenesis (Sharif et al., 2004). In light of the view that schizophrenia may be associated with impairment of adult neurogenesis (Reif et al., 2006; Toro and Deakin, 2007), the PEA15 gene remains as interesting susceptibility loci in schizophrenia. In addition, overexpression of the PEA15 gene is reported to be correlated with reduced insulin sensitivity in patients with type 2 diabetes (Condorelli et al., 1998). The human PEA15 gene is encoded by a gene of approximately 10 kb, located on chromosome 1q21.1 and consists of 4 exons (OMIM ⁎603434). ENTPD4, which is also known as lysosomal apyrase-like protein of 70 kDa, is one of the enzyme families capable of cleaving nucleotide tri- and diphosphates and participates in salvaging nucleotides from the lysosomal/autophagic vacuole lumen (Biederbick et al., 1999). Sorting mechanisms of lysosome-related organelles have been recently implicated in schizophrenia (Morris et al., 2008; NewellLitwa et al., 2007). The human ENTPD4 gene is encoded by a gene of approximately 30 kb, located on chromosome 8p21.3, and consists of 14 exons. GAS2L1 is a member of a family of genes that are highly expressed in 3T3 fibroblasts during growth arrest (Brancolini et al., 1992). The sequences of the GAS2L1 proteins contain both a putative calponin homology domain (CH) and GAS2-related (GAR) domain (Goriounov et al., 2003). The CH and GAR domains are associated with actin and microtubule, respectively, indicating that the product of GAS2L1 is a component of the microfilament system (Goriounov et al., 2003). Although the function of GAS2L1 remains unknown, it might be involved in the execution of apoptotic process (Lee et al., 1999). Given that schizophrenia might be associated with altered apoptotic process during the early manifestations of the disease (Gottfried et al., 2007), GAS2L1 may be an interesting susceptibility locus in schizophrenia. The human GAS2L1 gene is encoded by a gene of approximately 5.8 kb, located on chromosome 22q12.2 and consists of 5 exons. Therefore, on the basis of both the SAGE-based screening strategy and the positional candidate loci, we assume that these genes may confer susceptibility to schizophrenia. We performed the association analysis of these genes with schizophrenia in a Japanese population using selected genetic variants, such as tagging SNPs, which are proxies for SNPs in local linkage disequilibrium (LD) structures (Gu et al., 2008).
Fig. 1. Genomic structure and linkage disequilibrium (LD) of PEA15 gene. (a) Genomic structure of PEA15 gene and single-nucleotide polymorphisms (SNPs) used in association analyses. PEA15 gene spans over 10 kb, and is composed of 4 exons shown as boxes. The numbers either above or below the vertical arrows represent SNP numbers. SNP labeled with (H) represents a tagging SNP. Two pairs of primers (a pair of F4 and R4, and another of F6 and R6) are indicated by single horizontal lines with arrowheads. (b) LD evaluation of PEA15 gene. The number in each cell represents D′ value after the decimal point between SNPs. D′ values of 1.0 are not shown. A block is defined by a sold spine of LD.
for the screening scan was analyzed for precise determination of the pair-wise LD (Figs. 1b, 2b, and 3b). LD was based on 95% confidence bounds on D, according to Gabriel et al. (2002). Comparison of a categorical variable (gender) was performed using a χ2-test. DNA was extracted from blood using a DNA isolation kit. The SNPs were individually genotyped using either polymerase chain reaction (PCR)-based restriction fragment length polymorphism (PCR-RFLP), direct DNA sequencing of PCR products, or TaqMan® assay (Applied Biosystems, Foster City, CA, USA). Specific primer pairs for PCR-RFLP and direct DNA sequencing of PCR products are shown in Table 1. Three SNPs (rs680083 and rs4656252 for PEA15, and rs6006090 for GAS2L1) and eight SNPs (rs12549049, rs4872136, rs2272641, rs2272642 and rs2272643 for ENTPD4, and rs1894414, rs2301585, and rs2301586 for GAS2L1) were genotyped with PCR-RFLP and with direct DNA sequencing, respectively. Ten SNPs (rs2795070 for PEA15, rs34309579,
2. Methods The total number of subjects consisted of 240 schizophrenia patients (133 males and 107 females; age, 54.9 ± 12.8 years, mean ± S.D.) and 286 control subjects (143 males and 143 females; age: 41.6 ± 12.5 years). Subjects in both groups were from the Tohoku and Kanto regions of Japan. Review of their medical charts reporting the onset and characteristic clinical course clearly showed that the patients had suffered from schizophrenia in their lives prior to the present study. At the beginning of the present study, the diagnosis of schizophrenia was finally confirmed by the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) criteria (American Psychiatric Association, 2000) for all cases. The control subjects were unrelated healthy volunteers and did not suffer from any neuropsychiatric disorder. The objective of the present study was clearly explained, and written informed consent was obtained from all subjects in accordance with the Declaration of Helsinki (http://www.wma.net). Since several non-tagging SNPs adopted in the present study have not been validated for their allele frequencies in the Japanese ethnic group, we performed a screening scan in a subset of subjects that were included in the total subjects. This group composed of 94 schizophrenia patients (47 males and 47 females; age, 47.0 ± 8.8 years, mean ± S. D.) and 94 control subjects (47 males and 47 females; age: 42.0 ± 11. 3 years). The study was formally approved by the Institutional Review Board of the Ethical Committees of Tohoku University Graduate School of Medicine and of Dokkyo Medical University School of Medicine. We first consulted the International HapMap project database on dbSNP (http://www. ncbi.nlm.nih.gov/SNP/) and chose 16 tagging SNPs (two for PEA15, 10 for ENTPD4, and four for GAS2L1, shown in Figs. 1a, 2a, and 3a, respectively). We added five non-tagging SNPs which were located around the 5-near region of the respective genes in order to extend the sequenced region and to utilize successfully designed TaqMan probes (SNP browser™). Putative LD blocks of the three genes were estimated according to HAPLOVIEW (http://www.broad.mit.edu/mpg/haploview/), and the subset of subjects
Fig. 2. Genomic structure and linkage disequilibrium (LD) of ENTPD4 gene. (a) Genomic structure of ENTPD4 gene and single-nucleotide polymorphisms (SNPs) used in association analyses. ENTPD4 gene spans over 30 kb, and is composed of 14 exons shown as boxes. The numbers either above or below the vertical arrows represent SNP numbers. SNP labeled with (H) represents a tagging SNP. Two pairs of primers (a pair of F1 and R1, and another of F2 and R2) are indicated by single horizontal lines with arrowheads. (b) LD evaluation of ENTPD4 gene. The number in each cell represents D′ value after the decimal point between SNPs. D′ values of 1.0 are not shown. A block is defined by a sold spine of LD.
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
11
schizophrenia patients. The permutation P-values from the individual test represent the difference in frequency of an individual haplotype between the control and schizophrenia patients. All tests were two-tailed, and statistical significance was P b 0.05.
3. Results
Fig. 3. Genomic structure and linkage disequilibrium (LD) of GAS2L1 gene. (a) Genomic structure of GAS2L1 gene and single-nucleotide polymorphisms (SNPs) used in association analyses. GAS2L1 gene spans over 5.8 kb, and is composed of 5 exons shown as boxes. The numbers either above or below the vertical arrows represent SNP numbers. SNP labeled with (H) represents a tagging SNP. Two pairs of primers (a pair of F1 and R1, and another of F3 and R3) are indicated by single horizontal lines with arrowheads. (b) LD evaluation of GAS2L1 gene. The number in each cell represents D′ value after the decimal point between SNPs. A block is defined by a sold spine of LD.
rs6557673, rs11777391, rs2293936, rs2272640, rs2280832, and rs1061293 for ENTPD4, and rs174761 and rs174762 for GAS2L1) were genotyped with TaqMan® assay. PCR amplification was performed in a total volume of 5 μl reaction mixture containing 1 × PCR Buffer, 2.5 mM MgCl2, 0.5 μM of each primer, 0.25 mM of each dNTP, 5 ng of the genomic DNA, and 0.25 U of Taq DNA polymerase. LA Taq polymerase (Takara, Japan) was used for rs680083, rs4656252, rs1894414, rs2301585, rs2301586, and rs6006090. Taq DNA polymerase (Sigma, St. Louis, MO, USA) was used for the rest of SNPs. The PCR amplification was conducted under the following conditions: an initial denaturation phase at 95 °C for 2 min, followed by 40 cycles of 95 °C for 15–30 s, 58– 72 °C (details shown in Table 1) for 15–30 s, and 72 °C for 30 s–1 min, followed by a final extension phase of 72 °C for 5 min. Among the SNPs genotyped with direct DNA sequencing, the PCR products containing rs12549049 and rs4872136 were amplified using nested PCR. In PCR-RFLP, each 5 μl sample of PCR product was completely digested with 5 U of restriction enzyme, ScrFI for rs680083, BseYI for rs4656252, and AvaII for rs6006090. The resulting digested product was separated on 2–3% agarose gel, which was subsequently stained with ethidium bromide. Identification of the three SNPs, which were genotyped with PCR-RFLP, was confirmed by direct DNA sequencing in 16 randomly selected samples (eight controls and eight cases in the screening sample set). In direct DNA sequencing of PCR reactions, the PCR products were purified and sequenced with a BigDye Terminator Cycle Sequencing Reagent (Applied Biosystems, Foster City, CA, USA) on an ABI PRISM 3100 DNA Sequencer (Applied Biosystems, Foster City, CA, USA). Among SNPs which are eligible for TaqMan® assay, rs2795070, rs174761 and rs174762 were analyzed using the Takara Cycler Dice ™ Real Time System, and the rest of the seven SNPs were outsourced to be analyzed at the HuBit Genomics (Tokyo). All alleles and genotypes were expressed as those in a coding strand of the genes. Deviations of each SNP from the Hardy–Weinberg equilibrium (HWE) were analyzed by χ2 goodness-of-fit tests with a degree of freedom being one. Statistical differences in genotypic distribution and allelic frequencies between the schizophrenia patients and control subjects were evaluated by Fischer's exact tests. An odds ratio (OR) and its 95% confidence intervals (CIs) were determined based on a contingency table for allelic frequencies of each SNP. SNPAyze 5.1. standard program (Dynacom, Japan) was used to obtain the permutation P-values for constructed haplotype and genetic associations at polymorphism loci. The global test permutation P-values assess the overall difference in the distribution of haplotype frequencies between the control and
There were no significant differences in gender between the schizophrenia and control subjects. The average age was significantly higher in the schizophrenia patients than the control subjects (P b 0.0001, Student's t test). The direct DNA sequence-based genotyping of three SNPs (rs680083, rs4656252, and rs6006090) in 16 randomly selected samples was identical to those shown by PCR-RFLP (data not shown). Accordingly, we speculate that it was unlikely that a genotyping error had occurred. The genotypic distributions of all SNPs detected in the sample set did not deviate from the HWE. The LD block structures, which encompass the three genes, are shown in Figs. 1b, 2b, and 3b. Based on the screening scan sample subset, all SNPs investigated for PEA15 and GAS2L1 were placed on the entire genomic region with a single strong LD block. The SNPs in the ENTPD4 gene showed a strong, but not single, LD block. In all cases of the total sample set, we genotyped nine SNPs (three for PEA15 and six for GAS2L1), and constructed a haplotypic combination using these selected SNPs. Table 2 summarizes results of the allelic frequency and genotypic distribution for the PEA15 gene and GAS2L1 gene for the total sample set of subjects (240 schizophrenia patients and 286 control subjects). We circumscribed genotyping of 12 SNPs in the ENTPD4 gene within the screening scan sample subset (94 schizophrenia patients and 94 control subjects) due to high P values (N0.5) at this stage (Table 2). No single marker showed significant association with schizophrenia with regard to the allelic frequency or genotypic distribution for any of the genes examined. The results of the haplotype analyses are summarized in Table 3. Haplotype which is constructed of three SNPs (rs2795070(T/C)rs680083(A/G)-rs4656252(G/A)) in the strong LD block of the PEA15 gene (Fig. 1b) failed to show an overall difference in frequency between the affected cases and controls (global P value = 0.1097). When individual haplotypes were examined, C-G-A, a minor combination of the same haplotype, was underrepresented in the cases only at a marginally significant level (P = 0.0367). A second haplotype defined by six SNPs (rs174761(C/T)-rs1894414(A/C)rs2301585(G/C)-rs2301586(T/A)-rs6006090(A/G)-rs174762 (C/T)) in the GAS2L1 gene did not show a significant association with schizophrenia regarding the global P value or individual haplotypic p value. 4. Discussion The three genes (PEA15, ENTPD4, and GAS2L1) which were analyzed in the present study are localized in 1q21.1, 8p21.3, and 22q12.2, all of which have already been shown as positional candidate regions associated with schizophrenia in previous linkage studies (Brzustowicz et al., 2000; Gurling et al., 2001; Kendler et al., 2000; Takahashi et al., 2003). Individual genes mapped to this region can be considered to be candidates for schizophrenia susceptibility genes. Several candidate genes on these loci have indeed been reported to be associated with
Table 1 Forward primer
Primer sequence
Reverse primer
Primer sequence
Ta (°C)
PEA15 F4 PEA15 F6 ENTPD4 F1 ENTPD4 F2 GAS2L1 F1 GAS2L1 F3
5′-ATTGCCCCTATCTTGTGC-3′ 5′-CATGAGTTCCTACCAGACC-3′ 5′-CAAGAATGAAATCCTGGCTCTAC-3′ 5′-GTTTGTATCCTGACTCTCC-3′ 5′-TGAGCATTCAGTGAGCACAAG-3′ 5′-TGGACCAGGGTCACTCAAGG-3′
PEA15 R4 PEA15 R6 ENTPD4 R1 ENTPD4 R2 GAS2L1 R1 GAS2L1 R3
5′-TCTGTGTTCCTATTGCTGG-3′ 5′-GTTACCATAGCAACCTGC-3′ 5′-CACCAGTGGGCAGCATC-3′ 5′-TGCTGTGACTATTCAACC-3′ 5′-CGAGCCGCAGCTCTATG-3′ 5′-ATGGGGTACAGCTTCCACACG-3′
58 60 62 62 58 72
DMSO(+) means that adding 1/10 volume of dimethyl sulfoxide in each PCR reaction. Ta (°C) means annealing temperature.
DMSO
+ +
Method RFLP RFLP sequence sequence sequence RFLP
12
SNP number
IDs on dbSNPs
PEA15(1q21.1) SNP1
rs2795070
SNP2
SNP3
ENTPD4(8p21.3) SNP1
SNP2
SNP3
SNP4
SNP5
SNP6
SNP7
Groups
n
Schizophrenics Controls
n = 240 n = 286
Schizophrenics Controls
n = 237 n = 286
Schizophrenics Controls
n = 238 n = 286
Schizophrenics Controls
n = 91 n = 92
Schizophrenics Controls
n = 94 n = 89
Schizophrenics Controls
n = 93 n = 89
Schizophrenics Controls
n = 88 n = 94
Schizophrenics Controls
n = 88 n = 94
Schizophrenics Controls
n = 92 n = 94
Schizophrenics Controls
n = 91 n = 91
rs680083
rs4656252
rs34309579
rs12549049
rs4872136
rs6557673
rs11777391
rs2293936
rs2272640
Genotype count (frequency)
P-value
T/T 75(0.31) 73(0.26) A/A 75(0.32) 73(0.25) G/G 224(0.94) 256(0.89)
C/T 105(0.44) 150(0.52) A/G 105(0.44) 148(0.52) A/G 14(0.06) 29(0.10)
C/C 60(0.25) 63(0.22) G/G 57(0.24) 65(0.23) A/A 0(0) 1(0.01)
C/C 10(0.11) 15(0.16) C/C 62(0.66) 65(0.73) T/T 33(0.35) 32(0.36) C/C 18(0.20) 16(0.17) G/G 38(0.43) 34(0.36) A/A 81(0.88) 81(0.86) A/A 38(0.42) 33(0.36)
C/T 41(0.45) 38(0.41) C/T 28(0.30) 20(0.22) A/T 43(0.46) 40(0.45) C/T 35(0.40) 44(0.47) G/T 37(0.42) 44(0.47) A/C 10(0.11) 12(0.13) A/G 39(0.43) 44(0.48)
T/T 40(0.44) 39(0.43) T/T 4(0.04) 4(0.05) A/A 18(0.19) 17(0.19) T/T 35(0.40) 34(0.36) T/T 13(0.15) 16(0.17) C/C 1(0.01) 1(0.01) G/G 14(0.15) 14(0.16)
0.1356 (χ2 = 4.0495) 0.1759 (χ2 = 3.4768) 0.0934 (χ2 = 4.0025)
0.5971 (χ2 = 1.1212) 0.5209 (χ2 = 1.2685) 1 (χ2 = 0.0158) 0.6355 (χ2 = 0.9607) 0.6217 (χ2 = 0.9407) 0.9101 (χ2 = 0.1603) 0.7478 (χ2 = 0.6533)
Allele count (frequency) T 255(0.53) 296(0.52) A 255(0.54) 294(0.51) G 462(0.97) 541(0.95)
C 225(0.47) 276(0.48) G 218(0.46) 278(0.49) A 14(0.03) 31(0.05)
C 61(0.34) 68(0.37) C 152(0.81) 150(0.86) T 109(0.58) 104(0.58) C 71(0.40) 76(0.40) G 113(0.64) 112(0.60) A 172(0.93) 174(0.92) A 115(0.63) 110(0.60)
T 121(0.66) 116(0.63) T 36(0.19) 28(0.14) A 79(0.42) 74(0.42) T 105(0.60) 112(0.60) T 63(0.36) 76(0.40) C 12(0.07) 14(0.07) G 67(0.37) 72(0.40)
P-value
Odds ratio (95%CI)
0.6649 (χ2 = 0.1983)
0.9463 (0.7422−1.2066)
0.4552 (χ2 = 0.6048)
0.9077 (0.711–1.1587)
0.0652 (χ2 = 3.8832)
0.5288 (0.278–1.0062)
0.5128 (χ2 = 0.4744)
0.86 (0.5598−1.3212)
0.4115 (χ2 = 0.7406)
0.7881 (0.4579−1.3565)
1 (χ2 = 0.0076)
1.0186 (0.6722−1.5435)
1 (χ2 = 0.0003)
0.9965 (0.6554−1.5151)
0.3888 (χ2 = 0.8256)
0.1271 (0.7965−1.8599)
0.8396 (χ2 = 0.1224)
1.1533 (0.5185−2.5649)
0.6662 (χ2 = 0.291)
1.1235 (0.7359−1.7152)
Method TaqMan
RFLP
RFLP
TaqMan
Sequence
Sequence
TaqMan
TaqMan
TaqMan
TaqMan
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
Table 2 Allelic frequencies and genotypic distribution of 21 SNPs in thePEA15, ENTPD4, and GAS2L1 genes.
SNP8
SNP9
SNP10
SNP11
SNP12
GAS2L1(22q12.2) SNP1
SNP2
SNP4
SNP5
SNP6
Schizophrenics Controls
n = 91 n = 92
Schizophrenics Controls
n = 94 n = 94
Schizophrenics Controls
n = 94 n = 94
Schizophrenics Controls
n = 94 n = 94
Schizophrenics Controls
n = 94 n = 92
Schizophrenics Controls
n = 240 n = 286
Schizophrenics Controls
n = 239 n = 283
Schizophrenics Controls
n = 239 n = 282
Schizophrenics Controls
n = 239 n = 282
Schizophrenics Controls
n = 238 n = 286
Schizophrenics Controls
n = 240 n = 286
rs2272641
rs2272642
rs2272643
rs1061293
rs174761
rs1894414
rs2301585
rs2301586
rs6006090
rs174762
A/A 7(0.08) 9(0.10) A/A 41(0.44) 45(0.48) C/C 1(0.01) 1(0.01) A/A 1(0.01) 1(0.01) A/A 1(0.01) 1(0.01)
A/G 44(0.48) 38(0.41) A/G 46(0.49) 40(0.42) C/T 10(0.11) 12(0.13) A/G 10(0.11) 12(0.13) A/G 11(0.12) 12(0.13)
G/G 40(0.44) 45(0.49) G/G 7(0.07) 9(0.10) T/T 83(0.88) 81(0.86) G/G 83(0.88) 81(0.86) G/G 82(0.87) 79(0.86)
C/C 121(0.50) 156(0.55) A/A 217(0.92) 259(0.91) G/G 123(0.52) 156(0.55) T/T 106(0.44) 140(0.50) A/A 118(0.50) 155(0.54) C/C 123(0.51) 159(0.56)
C/T 97(0.41) 103(0.36) A/C 20(0.07) 23(0.08) C/G 96(0.40) 100(0.36) A/T 104(0.44) 108(0.38) A/G 98(0.41) 104(0.36) C/T 95(0.40) 100(0.35)
T/T 22(0.09) 27(0.09) C/C 2(0.01) 1(0.01) C/C 20(0.08) 26(0.09) A/A 29(0.12) 34(0.12) G/G 22(0.09) 27(0.10) T/T 22(0.09) 27(0.09)
Fischer's exact test was used to compare allelic frequencies and genotypic distribution between the patients and controls.
0.6654 (χ2 = 0.9777) 0.6881 (χ2 = 0.8547) 0.91 (χ2 = 0.2062) 0.91 (χ2 = 0.2062) 0.9127 (χ2 = 0.0779)
0.5767 (χ2 = 1.0982) 0.8074 (χ2 = 0.5436) 0.5407 (χ2 = 1.2269) 0.4476 (χ2 = 1.6337) 0.5131 (χ2 = 1.3172) 0.4901 (χ2 = 0.538)
A 58(0.32) 56(0.30) A 128(0.68) 130(0.69) C 12(0.06) 14(0.07) A 12(0.06) 14(0.07) A 13(0.07) 14(0.08)
G 124(0.68) 128(0.70) G 60(0.32) 58(0.31) T 176(0.94) 174(0.93) G 176(0.94) 174(0.93) G 175(0.93) 170(0.92)
C 339(0.71) 415(0.73) A 454(0.95) 541(0.96) G 342(0.72) 412(0.69) T 255(0.72) 388(0.69) A 334(0.70) 414(0.72) C 341(0.71) 418(0.73)
T 141(0.29) 157(0.27) C 24(0.05) 25(0.04) C 136(0.28) 152(0.31) A 101(0.29) 176(0.31) G 142(0.30) 158(0.28) T 139(0.29) 154(0.27)
TaqMan 0.8216 (χ2 = 0.0877)
0.9353 (0.6009−1.456)
0.9115 (χ2 = 0.0494)
0.9518 (0.6156−1.4715)
0.8394 (χ2 = 0.1653)
1.1801 (0.5308−2.6237)
0.8394 (χ2 = 0.1653)
1.1801 (0.5308−2.6237)
0.8435 (χ2 = 0.0665)
1.1086 (0.5062−2.4278)
0.493 (χ2 = 0.4776)
1.0994 (0.8402−1.4386)
0.6621 (χ2 = 0.2113)
1.144 (0.6445−2.0306)
0.6266 (χ2 = 0.2917)
0.9278 (0.7067−1.218)
0.388 (χ2 = 0.8514)
0.8848 (0.6822−1.1475)
0.4506 (χ2 = 0.6208)
1.114 (0.0.8516–1.4573)
0.538 (χ2 = 0.9335)
1.1064 (0.8444–1.4497)
Sequence
Sequence
Sequence
TaqMan
TaqMan
Sequence
Sequence
Sequence
RFLP
TaqMan
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
SNP3
rs2280832
13
14
A. Saito et al. / Psychiatry Research 185 (2011) 9–15
Table 3 PEA15 haplotype in Japanese case-control subjects Gene
Haplotype
Schizophrenics
Frequency
Controls
PEA15
T-A-G 248 0.537 294 C-G-G 201 0.435 243 C-G-A 13 0.028 31 Haplotype pairs mean rs2795070(T/C)-rs680083(A/G)-rs4656252(G/A) in PEA15. Global P-value = 0.1097. Permutation P-values were calculated by SNPAyze 5.1. standard program (Dynacom, Japan).
Frequency
Permutation P-value
0.517 0.428 0.055
0.5406 0.8275 0.0367
GAS2L1 haplotype in Japanese case-control subjects Gene
Haplotype
Schizophrenics
Frequency
Controls
GAS2L1
Frequency
C-A-G-T-A-C 307 0.670 379 0.694 T-A-C-A-G-T 128 0.280 143 0.262 C-C-G-A-A-C 23 0.050 24 0.044 Haplotype pairs mean rs174761(C/T)-rs1894414(A/C)-rs2301585(G/C)-rs2301586(T/A)-rs6006090(A/G)-rs174762(C/T) in GAS2L1. Global P-value = 0.7100. Permutation P-values were calculated by SNPAyze 5.1. standard program (Dynacom, Japan).
schizophrenia in case-control and family-based studies (Liu et al., 2007; Nakata et al., 2003; Ni et al., 2007; Porton et al., 2004). In the present study, we focused on analyses of PEA15 and GAS2L1 genes due to the following reasons: (1) all SNPs investigated for PEA15 and GAS2L1 were placed on a single strong LD block, and (2) these two genes may be involved in key signaling pathways for neurogenesis and apoptosis (Lee et al., 1999; Renault et al., 2003; Sharif et al., 2004; Whitehurst et al., 2004). No significant differences in genotypic and allelic frequencies of the selected SNPs nor frequencies of multi-marker haplotype were observed between the affected cases and controls. Combining these findings together, the present study suggests that PEA15 and GAS2L1 do not play an important role in susceptibility to schizophrenia in the Japanese population. We left ENTPD4 out of further analysis beyond the screening scan sample subset, since high P values for allelic frequencies suggest that this gene is also unlikely to be related to schizophrenia. In the present study, tagging SNPs which have been increasingly utilized in genetic studies of schizophrenia were applied to minimize the size of the core set of SNPs by the reducing redundancy among the SNPs, since no previous affected case-control comparison data regarding SNPs in the PEA15, ENTPD4, and GAS2L1 genes existed. The failure to demonstrate association with the three genes is unlikely to be due to population stratification, since both groups of individuals were drawn from the ethnically homogenous population of the Tohoku and Kanto areas of Japan. The finding that most tagging SNPs employed hereby based on the HapMap data are associated with strong LD, lends support to the theory of favorable transferability across different ethnicities (Gu et al., 2008). Several caveats should be considered when interpreting our results. The first concerns the ability of tagging SNPs to capture rare SNPs. The rare SNPs with minor allele frequency of b5% are generally excluded from a HapMap scheme (Gu et al., 2008). Thus, it remains possible that an as-yet unidentified causative mutation might be involved in pathogenesis of schizophrenia. In addition, most of the tagging SNPs investigated in the present study exist in introns, 5-near regions and 3-near regions with unknown function. Given that SNPs which are not found in the LD may be associated with schizophrenia, further analyses on the basis of detailed SNP coverage of these genes are necessary. The second concerns significant differences in the average age between the controls and schizophrenics in the present study. However, this should not be regarded as being a shortage, since SNP frequencies are generally unlikely to change with age (Lohoff et al., 2005). Finally, the study reporting the plausible involvement of the overexpression of the PEA15 gene in type 2 diabetes mellitus (Condorelli et al., 1998) is worthy of attention. Although this has been shown to occur in fibroblasts of patients with diabetes mellitus (Condorelli et al., 1998), Wolford et al. (2000) reported that none of
Permutation P-value 0.4369 0.5398 0.6266
the three SNPs in the PEA15 gene are associated with type 2 diabetes mellitus in Pima Indians, a population with the highest documented prevalence of this disease worldwide. Intriguingly, the two SNPs they (Wolford et al., 2000) examined appear to correspond to the two SNPs (rs2795070 and rs680083) analyzed in the current study. Despite the general understanding that the risk of developing diabetes mellitus is accelerated by the administration of antipsychotics, recent studies on medication-naïve patients with schizophrenia do not support the hypothesis that schizophrenia is associated with an underlying abnormality predicting the occurrence of diabetes mellitus (PerezIglesias et al., 2009; Sengupta et al., 2008). Combining these findings together, it is likely that the PEA15 gene variation is involved in neither schizophrenia nor diabetes mellitus. Nonetheless, the possibility that PEA15 may be involved somehow in glucose metabolism, or other alterations of glucose homeostasis in the context of antipsychotic-induced liability, in developing diabetes mellitus is a topic that in the future might be worthwhile to study. In conclusion, the present results suggest that PEA15, ENTPD4, and GAS2L1 genes do not confer susceptibility to schizophrenia in the Japanese population. However, further research on these genes, which are located on schizophrenia susceptibility loci, using different SNPs and a larger sample set will be required.
Acknowledgments The present study was supported by Dokkyo Medical University, Young Investigator Award, and the Kobayashi Magobe Memorial Medical Foundation.
References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. American Psychiatric Association, Washington DC. Biederbick, A., Rose, S., Elsässer, H.-P., 1999. A human intracellular apyrase-like protein, LALP70, localizes to lysosomal/autophagic vacuoles. Journal of Cell Science 112, 2473–2484. Brancolini, C., Bottega, S., Schneider, C., 1992. Gas2, a growth arrest-specific protein, is a component of the microfilament network system. The Journal of Cell Biology 117, 1251–1261. Brzustowicz, L.M., Hodgkinson, K.A., Chow, E.W.C., Honer, W.G., Bassett, A.S., 2000. Location of a major susceptibility locus for familial schizophrenia on chromosome 1q21–q22. Science 288, 678–682. Condorelli, G., Vigliotta, G., Cafieri, A., Trencia, A., Andalò, P., Oriente, F., Miele, C., Caruso, M., Formisano, P., Beguinot, F., 1999. PED/PEA-15: an anti-apoptotic molecule that regulates FAS/TNFR1-induced apoptosis. Oncogene 18, 4409–4415. Condorelli, G., Vigliotta, G., Iavarone, C., Caruso, M., Tocchetti, C.G., Andreozzi, F., Cafieri, A., Tecce, M.F., Formisano, P., Beguinot, L., Beguinot, F., 1998. PED/PEA-15 gene controls glucose transport and is overexpressed in type 2 diabetes mellitus. The EMBO Journal 17, 3858–3866. Gabriel, S.B., Schaffner, S.F., Nguyen, H., Moore, J.M., Roy, J., Blumenstiel, B., Higgins, J., DeFelice, M., Lochner, A., Faggart, M., Liu-Cordero, S.N., Rotimi, C., Adeyemo, A., Cooper, R., Ward, R., Lander, E.S., Daly, M.J., Altshuler, D., 2002. The structure of haplotype blocks in the human genome. Science 296, 2225–2229.
A. Saito et al. / Psychiatry Research 185 (2011) 9–15 Goriounov, D., Leung, C.L., Liem, R.K.H., 2003. Protein products of human Gas2-related genes on chromosomes 17 and 22 (hGAR17 and hGAR22) associate with both microfilaments and microtubules. Journal of Cell Science 116, 1045–1058. Gottfried, Y., Rotem, A., Klein, E., Larisch, S., 2007. The pro-apoptotic ARTS/Sept4 protein is significantly reduced in post-mortem brains from schizophrenic patients. Schizophrenia Research 96, 257–266. Gu, C.C., Yu, K., Rao, D.C., 2008. Characterization of LD structures and the utility of HapMap in genetic association studies. Advances in Genetics 60, 407–435. Gurling, H.M.D., Kalsi, G., Brynjolfson, J., Sigmundsson, T., Sherrington, R., Mankoo, B.S., Read, T., Murphy, P., Blaveri, E., McQuillin, A., Petursson, H., Curtis, D., 2001. Genomewide genetic linkage analysis confirms the presence of susceptibility loci for schizophrenia, on chromosomes 1q32.2, 5q33.2, and 8p21–22 and provides support for linkage to schizophrenia, on chromosomes 11q23.3–24 and 20q12.1– 11.23. American Journal of Human Genetics 68, 661–673. Kendler, K.S., Myers, J.M., O'Neill, F.A., Martin, R., Murphy, B., MacLean, C.J., Walsh, D., Straub, R.E., 2000. Clinical features of schizophrenia and linkage to chromosomes 5q, 6p, 8p, and 10p in the Irish study of high-density schizophrenia families. The Amercan Journal of Psychiatry 157, 402–408. Lee, K.K.H., Tang, M.K., Yew, D.T.W., Chow, P.H., Yee, S.P., Schneider, C., Brancolini, C., 1999. Gas2 is a multifunctional gene involved in the regulation of apoptosis and chondrogenesis in the developing mouse limb. Developmental Biology 207, 14–25. Liu, Y.L., Fann, C.S.J., Liu, C.M., Chang, C.C., Yang, W.C., Hung, S.I., Yu, S.L., Hwang, T.J., Hsieh, M.H., Liu, C.C., Tsuang, M.M., Wu, J.Y., Jou, Y.S., Faraone, S.V., Tsuang, M.T., Chen, W.J., Hwu, H.-G., 2007. More evidence supports the association of PPP3CC with schizophrenia. Molecular Psychiatry 12, 966–974. Lohoff, F.W., Ferraro, T.N., McNabb, L., Schwebel, C., Dahl, J.P., Doyle, G.A., Buono, R.J., Berrettini, W.H., 2005. No association between common variations in the neuronal nicotinic acetylcholine receptor alpha2 subunit gene (CHRNA2) and bipolar I disorder. Psychiatry Research 135, 171–177. Morris, D.W., Murphy, K., Kenny, N., Purcell, S.M., McGhee, K.A., Schwaiger, S., Nangle, J.-M., Donohue, G., Clarke, S., Scully, P., Quinn, J., Meagher, D., Baldwin, P., Crumlish, N., O'Callaghan, E., Waddington, J.L., Gill, M., Corvin, A.P., 2008. Dysbindin (DTNBP1) and the biogenesis of lysosome-related organelles complex 1 (BLOC-1): main and epistatic gene effects are potential contributors to schizophrenia susceptibility. Biological Psychiatry 63, 24–31. Nakata, K., Ujike, H., Sakai, A., Takaki, M., Imamura, T., Tanaka, Y., Kuroda, S., 2003. The human dihydropyrimidinase-related protein 2 gene on chromosome 8p21 is associated with paranoid-type schizophrenia. Biological Psychiatry 53, 571–576. Newell-Litwa, K., Seong, E., Burmeister, M., Faundez, V., 2007. Neuronal and non-neuronal functions of the AP-3 sorting machinery. Journal of Cell Science 120, 531–541. Ni, X., Valente, J., Azevedo, M.H., Pato, M.T., Pato, C.N., Kennedy, J.L., 2007. Connexin 50 gene on human chromosome 1q21 is associated with schizophrenia in matched case-control and family-based studies. Journal of Medical Genetics 44, 532–536.
15
Ouchi, Y., Kubota, Y., Kuramasu, A., Watanabe, T., Ito, C., 2005. Gene expression profiling in whole cerebral cortices of phencyclidine- or methamphetamine-treated rats. Molecular Brain Research 140, 142–149. Perez-Iglesias, R., Mata, I., Relayo-Teran, J.M., Amado, J.A., Garcia-Unzueta, M.T., Berja, A., Martinez-Garcia, O., Vazquez-Barquero, J.L., Crespo-Facorro, B., 2009. Glucose and lipid disturbances after 1 year of antipsychotic treatment in a drug-naïve population. Schizophrenia Research 107, 115–121. Porton, B., Ferreira, A., DeLisi, L.E., Kao, H.-T., 2004. A rare polymorphism affects a mitogen-activated protein kinase site in synapsin III: possible relationship to schizophrenia. Biological Psychiatry 55, 118–125. Reif, A., Fritzen, S., Finger, M., Strobel, A., Lauer, M., Schmitt, A., Lesch, K.-P., 2006. Neural stem cell proliferation is decreased in schizophrenia, but not in depression. Molecular Psychiatry 11, 514–522. Renault, F., Formstecher, E., Callebaut, I., Junier, M.-P., Chneiweiss, H., 2003. The multifunctional protein PEA-15 is involved in the control of apoptosis and cell cycle in astrocytes. Biochemical Pharmacology 66, 1581–1588. Saito, A., Fujikura-Ouchi, Y., Kuramasu, A., Shimoda, K., Akiyama, K., Matsuoka, H., Ito, C., 2007. Association study of putative promoter polymorphisms in the neuroplastin gene and schizophrenia. Neuroscience Letters 411, 168–173. Sengupta, S., Parrilla-Escobar, M.A., Klink, R., Fathalli, F., Ying, K.N., Stip, E., Baptista, T., Malla, A., Joober, R., 2008. Are metabolic indices different between drug-naïve first-episode psychosis patients and healthy controls? Schizophrenia Research 102, 329–336. Sharif, A., Renault, F., Beuvon, F., Castellanos, R., Canton, B., Barbeito, L., Junier, M.P., Chneiweiss, H., 2004. The expression of PEA-15 (phosphoprotein enriched in astrocytes of 15KDa) defines subpopulations of astrocytes and neurons throughout the adult mouse brain. Neuroscience 126, 263–275. Takahashi, S., Cui, Y.-h., Kojima, T., Han, Y.-h., Zhou, R.-l., Kamioka, M., Yu, S.-y., Matsuura, M., Matsushima, E., Wilcox, M., Arinami, T., Shen, Y.-c., Faraone, S.V., Tsuang, M.T., 2003. Family-based association study of markers on chromosome 22 in schizophrenia using African-American, European-American, and Chinese families. American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 120B, 11–17. Toro, C.T., Deakin, J.F.W., 2007. Adult neurogenesis and schizophrenia: A window on abnormal early brain development? Schizophrenia Research 90, 1–14. Whitehurst, A.W., Robinson, F.L., Moore, M.S., Cobb, M.H., 2004. The death effector domain protein PEA-15 prevents nuclear entry of ERK2 by inhibiting required interactions. The Journal of Biological Chemistry 279, 12840–12847. Wolford, J.K., Bogardus, C., Ossowski, V., Prochazka, M., 2000. Molecular characterization of the human PEA15 gene on 1q21–q22 and association with type 2 diabetes mellitus in Pima Indians. Gene 241, 143–148.