Skip Navigation


Human Molecular Genetics Advance Access originally published online on August 30, 2007
Human Molecular Genetics 2007 16(23):2854-2869; doi:10.1093/hmg/ddm244
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/23/2854    most recent
ddm244v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Miyashita, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miyashita, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Genetic association of CTNNA3 with late-onset Alzheimer's disease in females

Akinori Miyashita1, Hiroyuki Arai2, Takashi Asada3, Masaki Imagawa4, Etsuro Matsubara5, Mikio Shoji6, Susumu Higuchi7, Katsuya Urakami8, Akiyoshi Kakita9, Hitoshi Takahashi9, Shinichi Toyabe10,{dagger}, Kohei Akazawa10, Ichiro Kanazawa11, Yasuo Ihara12,{ddagger}, Ryozo Kuwano1,* and The Japanese Genetic Study Consortium for Alzeheimer's Disease

1 Bioresource Science Branch, Center for Bioresources, Brain Research Institute, Niigata University, Niigata 951-8585, Japan, 2 Department of Geriatrics and Gerontology, Center for Asian Traditional Medicine, Tohoku University Graduate School of Medicine, Sendai 980-8574, Japan, 3 Department of Psychiatry, Institute of Clinical Medicine, University of Tsukuba, Tsukuba 305-8575, Japan, 4 Imagawa Clinic, Fukushima-ku, Osaka 553-0003, Japan, 5 Department of Alzheimer's Disease Research, National Institute for Longevity Sciences, National Center for Geriatrics and Gerontology, Obu 474-8522, Japan, 6 Department of Neurology, Neuroscience and Biophysiological Science, Hirosaki University, School of Medicine, Hirosaki 036-8562, Japan, 7 Division of Clinical Research, Kurihama Alcoholism Center, National Hospital Organization, Yokosuka 239-0841, Japan, 8 Department of Biological Regulation, Section of Environment and Health Science, Faculty of Medicine, Tottori University, Yonago 683-8503, Japan, 9 Departments of Pathology and Pathological Neuroscience, Brain Research Institute, Niigata University, Niigata 951-8585, Japan, 10 Department of Medical Informatics, Niigata University, Niigata 951-8520, Japan, 11 National Center for Neurology and Psychiatry, Kodaira 187-8502, Japan and 12 Department of Neuropathology, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan

* To whom correspondence should be addressed at: 1-757 Asahimachi, Chuo-ku, Niigata 951-8585, Japan. Tel: +81 252272343; Fax: +81 252270793; Email: ryosun{at}bri.niigata-u.ac.jp

Received July 18, 2007; Accepted August 22, 2007


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Alzheimer's disease (AD), the most common form of dementia in the elderly, was found to exhibit a trend toward a higher risk in females than in males through epidemiological studies. Therefore, we hypothesized that gender-related genetic risks could exist. To reveal the ones for late-onset AD (LOAD), we extended our previous genetic work on chromosome 10q (genomic region, 60–107 Mb), and single nucleotide polymorphism (SNP)-based genetic association analyses were performed on the same chromosomal region, where the existence of genetic risk factors for plasma Aß42 elevation in LOAD was implied on a linkage analysis. Two-step screening of 1140 SNPs was carried out using a total of 1408 subjects with the APOE-{varepsilon}3*3 genotype: we first genotyped an exploratory sample set (LOAD, 363; control, 337), and then genotyped some associated SNPs in a validation sample set (LOAD, 336; control, 372). Seven SNPs, spanning about 38 kb, in intron 9 of CTNNA3 were found to show multiple-hit association with LOAD in females, and exhibited more significant association on Mantel–Haenszel test (allelic P-valuesMH-F = 0.000005945–0.0007658). Multiple logistic regression analysis of a total of 2762 subjects (LOAD, 1313; controls, 1449) demonstrated that one of the seven SNPs directly interacted with the female gender, but not with the male gender. Furthermore, we found that this SNP exhibited no interaction with the APOE-{varepsilon}4 allele. Our data suggest that CTNNA3 may affect LOAD through a female-specific mechanism independent of the APOE-{varepsilon}4 allele.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Alzheimer's disease (AD) is a neurodegenerative disorder clinically characterized by progressive cognitive deterioration and is the most common form of dementia in the elderly. Its neuropathological features are amyloid plaques [extracellular deposition of amyloid ß-protein (Aß)] and neurofibrillary tangles (intracellular aggregation of highly phosphorylated microtubule-associated protein tau), which finally lead to synaptic loss and/or neuronal death.

Recent epidemiological studies on AD revealed gender-related differences in its prevalence (1–3) and incidence (4–6). Compared with males, females are more likely to develop AD, although results contradicting this gender difference have been reported (7–9). In blood mononuclear cells in AD, there are substantial gender differences in gene expression (10). The plasma level of amyloid beta-protein 42 (Aß42), a major constituent of senile plaques, is significantly increased in females with mild cognitive impairment, a transitional state between normal aging and mild dementia (11). In transgenic animal models of AD, gender-dependent accumulation and deposition of Aß42 and Aß40 have been observed (12–15). Moreover, there has been increasing research on gender-related genetic risk factors in AD: ACT (16), MPO (17,18), ACE (19), ESR2 (20), DSC1 (21) and ABCA1 (22). Therefore, based on these findings, we hypothesized that gender-related genetic risk factors that modify Aß metabolism in late-onset AD (LOAD), which accounts for 95–99% of AD, could exist.

We have paid a great deal of attention to chromosome 10q, especially because the existence of genetic risk factors for plasma Aß42 elevation in it was implied on linkage analysis of LOAD families (23). Furthermore, through other genetic approaches, including genome-wide linkage screening of affected sib pairs (24) and candidate gene-based analysis of multiplex AD families (25), chromosome 10q was strongly suggested to be the most prominent one for LOAD. Therefore, regarding a genomic region on chromosome 10q (60–107 Mb), we previously performed large-scale single nucleotide polymorphism (SNP)-based screening of a Japanese population to identify additional genetic risk factors to APOE (19q13.2), which is universally recognized as a major risk gene for the development of LOAD (OMIM + 107741 [OMIM] ). Consequently, we found that DNMBP, which is involved in synaptic vesicle recycling, was associated with LOAD with the APOE-{varepsilon}3*3 genotype or lacking the APOE-{varepsilon}4 allele in several sample sets (26).

Interestingly, replicated evidence for a parent-of-origin effect of chromosome 10q was recently reported for LOAD (27,28), which suggests that gender-related genes such as imprinting genes could be responsible for the disease development. Here, in order to determine whether or not gender-related loci associated with LOAD are present, our previous genetic work on chromosome 10q (26) was extended. Two sample sets for screening, Exploratory and Validation, comprising only APOE-{varepsilon}3*3 subjects were prepared, which were used for a case–control association study after being stratified as to gender. We first genotyped the Exploratory set, and then genotyped some significantly associated SNPs in the Validation set. Through this stepwise screening, among the 1140 SNPs subjected to the exploratory screening, we finally found seven SNPs located in intron 9 of CTNNA3 that showed reproducible association with LOAD in females. These replicated SNPs were further examined by means of genotyping of all the subjects with all APOE genotypes ({varepsilon}2*2, {varepsilon}2*3, {varepsilon}2*4, {varepsilon}3*3, {varepsilon}3*4 and {varepsilon}4*4), i.e. 1526 LOAD patients (female, 1103; male, 423) and 1666 controls (female, 998; male, 668), some of them exhibiting significance only in a female sub-sample set. In terms of biological functions, CTNNA3 (29,30), encoding {alpha}-T catenin, is thought to be a promising candidate for LOAD because it is a binding partner of ß-catenin, which interacts with PSEN1 (31), and because it was recently shown to be associated with the level of plasma Aß42 in a set of families with LOAD (32). Multiple logistic regression analysis in a total of 2762 subjects (LOAD, 1313; controls, 1449) revealed that one (SNP rs713250) of the seven associated SNPs exhibits a significant interaction with the female gender, but not with the male gender and the APOE-{varepsilon}4 allele. Our data suggest that CTNNA3 could affect LOAD through a female-specific mechanism independent of the APOE-{varepsilon}4 allele.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Allelic association
To determine whether gender-related loci associated with LOAD on chromosome 10q (60–107 Mb) exist or not, we stratified the Exploratory sample set (Table 1) by gender, resulting in female and male subsets. An allelic contingency table (2 x 2)-based {chi}2 test was performed using already-obtained genotype data (26) for 1140 SNPs for the Exploratory set. Calculation of allelic P-values and odds ratios (ORs) with 95% confidence interval (CI) was carried out to examine the genetic association of these SNPs. In a Japanese population, these SNPs were actually polymorphic and showed a P-value >0.05 in exact tests of Hardy–Weinberg equilibrium (HWE) in both cases and controls of the Exploratory set (details given under Materials and Methods). The results of {chi}2 tests for the gender-stratified sets are presented in Fig. 1. In the female group (LOAD, 249; controls, 223), 106 of the 1140 SNPs had significant allelic P-values <0.05, and 34 of these 106 showed more significant values (allelic P-values <0.01). In the male group (LOAD, 114; controls, 114), 53 of the 1140 SNPs showed allelic P-values <0.05, and 7 of these 53 showed more significant association with allelic P-values <0.01.


Figure 1
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Allelic P-values of 1140 SNPs for the Exploratory set comprising female (A) (LOAD, 249; control, 223) or male (B) (LOAD, 114; control, 114) APOE-{varepsilon}3*3 subjects. Dotted and dashed lines indicate allelic P-values at the 0.05 and 0.01 levels, respectively. The significantly associated locus focused on in this study is indicated by the thick line, which is labeled ‘CTNNA3’. The genomic position conformed to NCBI build 35.1.

 


View this table:
[in this window]
[in a new window]

  		Table 1. Subject information

 
A total of 41 SNPs (34 and 7 SNPs in female and male Exploratory sets, respectively) showing allelic P-values <0.01 were further analyzed by means of {chi}2 tests to determine whether or not these SNPs actually exhibit reproducible allelic association using another sample set, Validation, sub-grouped as to gender (Table 1). In the male Validation set (LOAD, 94; controls, 159), three of the above-mentioned seven SNPs showed reproducible association (allelic P-values = 0.0342–0.046). Among these three SNPs, only SNP rs1000280 exhibited a significant value on Mantel–Haenszel test (allelic P-valueMH-M = 0.0009112). This SNP is located in the intergenic region between LOXL4 (100.00–100.02 Mb) and C10orf33 (100.13–100.16 Mb); therefore, we did not investigate this SNP further. In the female Validation set (LOAD, 242; controls, 213), 16 of the above-mentioned 34 SNPs exhibited allelic association with P-values <0.05. These SNPs exhibited significance on Mantel–Haenszel test of the two female sets (allelic P-valuesMH-F = 0.000005945–0.0008809). These allelic P-valuesMH-F remained at significant levels even after Bonferroni's correction for 34 tests (allelic P-valuesMH-F(B) = 0.0002021–0.02995). Of the 16 SNPs, 9 (rs911541, rs3740066, rs11190302, rs35715207, rs3758394, rs3740058, rs3740057, rs11190315 and rs6584331) are located in a locus between ENTPD7 and DNMBP recently reported by our group (26). The remaining seven, rs7909676, rs2394287, rs4459178, rs10997307, rs12258078, rs10822890 and rs713250, spanning about 38 kb, are encompassed by intron 9 of CTNNA3, which consists of 18 exons (Fig. 2A and C). The allelic P-values of these seven SNPs in the two sample sets, Exploratory and Validation, are presented in Table 2, and marker information on them is summarized in Table 3. The genotypic and allelic distributions are presented in the Supplementary Material, Table S1.


Figure 2
View larger version (47K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Genomic position and LD block structure of an associated locus within CTNNA3. (A) Genomic region including CTNNA3 from web site Database of Genomic Variation (http://projects.tcag.ca/variation/). Boxes filled in green and black, also used in Fig. 2B, represent the associated genomic regions identified here and in studies on Caucasians (32,35,37), respectively. Each exon of CTNNA3 and LRRTM3 is numbered; CNV, copy number variation. (B) Overview of the LD pattern between 67.4 and 68.2 Mb in Japanese. HapMap genotype data (1768 SNPs) on 45 unrelated Japanese in Tokyo (JPT) were used as calculation of LD measures, D'. (C) Physical positions of the seven replicated SNPs. Vertical lines indicate the SNPs used in this study: significantly associated SNPs are indicated by the long labeled vertical lines. Horizontal arrows within open boxes indicate the transcription orientations of individual genes. The mapping position of each SNP is according to dbSNP build 125 on NCBI build 35.1.

 


View this table:
[in this window]
[in a new window]

  		Table 2. Statistics for seven reproducible SNPs found on two-step screening involving APOE-{varepsilon}3*3

 


View this table:
[in this window]
[in a new window]

  		Table 3. Summary of seven associated SNPs within intron 9 of CTNNA3

 
To examine the gender-specific effects of the seven CTNNA3 SNPs on LOAD, we additionally performed joint analysis regarding gender (Table 2). For this analysis, female and male allelic contingency tables were combined for the Exploratory and Validation sets, respectively (Supplementary Material, Table S1). {chi}2 tests based on the combined 2 x 2 allelic contingency tables and calculation of the ORs with 95% CI were carried out. In the Exploratory set comprising both genders, none of these seven SNPs showed more significant association (allelic P-values = 0.00005431–0.0235) in comparison with the Exploratory set only including females (allelic P-values = 0.00004614–0.008). The ORs exhibited a tendency to decrease; for example, for SNP rs10822890, from 1.72 to 1.55. A similar trend for both the allelic P-values and ORs of these seven SNPs was observed on Mantel–Haenszel test.

The reproducible seven SNPs on CTNNA3 were further examined by means of stratified analysis, based on the carrier status of the APOE-{varepsilon}4 allele, with the {chi}2 test (Table 4). The genotypic and allelic distributions are presented in the Supplementary Material, Table S2. We used the overall sample set, All, including all subjects (LOAD, 1526; controls, 1666) with all APOE genotypes (2*2, 2*3, 2*4, 3*3, 3*4 and 4*4), and two sub-sample sets, Negative-{varepsilon}4 and Positive-{varepsilon}4, which were stratified as to the presence (2*4, 3*4 and 4*4) or absence (2*2, 2*3 and 3*3) of the APOE-{varepsilon}4 allele (Table 1). As shown in Table 4, in the All set, five (rs7909676, rs2394287, rs4459178, rs10822890 and rs713250) of the seven SNPs were statistically significant in females (allelic P-values = 0.0009719–0.00126). In the Negative-{varepsilon}4 set, all seven SNPs exhibited more significant association with LOAD in females (allelic P-values = 0.00001019–0.002555). No evidence was found of association with any of the seven SNPs in males in any sample set.


View this table:
[in this window]
[in a new window]

  		Table 4. Allelic association of seven associated SNPs, encompassed by intron 9 of CTNNA3, in the overall sample set, All, and two sub-sample sets, Negative-{varepsilon}4 and Positive-{varepsilon}4, stratified as to the presence or absence of the APOE-{varepsilon}4 allele

 
For joint analysis concerning gender, female and male contingency tables (2 x 2) with the allelic distributions were combined for the All, Negative-{varepsilon}4 and Positive-{varepsilon}4 sample sets, respectively (Supplementary Material, Table S2). Allelic P-values and ORs (95% CI) derived from the combined contingency tables were used to evaluate the gender-specific effects on LOAD (Table 4). This analysis revealed that in the All set including both genders, the ORs of significant SNPs (rs7909676, rs10822890 and rs713250) tended to be lower, compared with those in the female All set; for example, from 1.23 to 1.11 for SNP rs713250. A similar decreasing tendency for ORs of significant SNPs (rs7909676, rs2394287, rs4459178, rs10997307, rs12258078, rs10822890 and rs713250) in the Negative-{varepsilon}4 set including both genders was also observed in comparison with those in the female Negative-{varepsilon}4 set; for example, from 1.42 to 1.24 for SNP rs10822890.

Multiple logistic regression analysis, involving APOE-{varepsilon}4, gender, age, the seven replicated SNPs on CTNNA3 and their interactions as independent variables, was performed to assess the potential effects of these variables on the association with LOAD, using 2762 subjects [LOAD, 1313 (female, 949; male, 364); controls, 1449 (female, 877; male, 572)] (Table 5). In this analysis, the subjects used were not sub-grouped as to gender and/or carrier status of the APOE-{varepsilon}4 allele. Initially, we carried out multiple logistic regression analysis with a forward stepwise method without interaction terms to elucidate which variables explained an association with LOAD independently. Model 1 in Table 5 shows significant risk factors selected by this analysis. Expectedly, the APOE-{varepsilon}4 allele, gender and age, which are well-known risk factors for LOAD, had significant effects on the LOAD risk. Among the seven associated SNPs, SNP rs713250 was chosen as representative and selectively entered in this model [for genotype CC: OR (95% CI), 1.36 (1.08–1.71); P-value = 0.009]. Following this primary analysis, we further assessed second-order interaction terms created by the four significant risk factors including the SNP rs713250 (Model 2 in Table 5). Six interactions were tested by means of a forward stepwise method in addition to APOE-{varepsilon}4, gender, age and the SNP rs713250. It was demonstrated that the SNP rs713250 exhibited significant interaction with the female gender in a dose-dependent manner as to the allele C [TC_female, OR (95% CI) = 1.68 (1.12–2.54); CC_female, OR (95% CI) = 2.57 (1.59–4.17)].


View this table:
[in this window]
[in a new window]

  		Table 5. Multiple logistic regression analysis

 
Linkage disequilibrium and case–control haplotype analyses
To reveal genetic relationship between each significant SNP on CTNNA3, linkage disequilibrium (LD) and haplotype estimation analyses were performed. For these analyses, we used four sample sets (All as the overall sample set, and Negative-{varepsilon}4, Positive-{varepsilon}4 and {varepsilon}3*3 as sub-sample sets) after being sub-grouped as to gender (Table 1). From the Japanese HapMap genotype data (JPT), these SNPs were found to be encompassed by a highly structured LD block extending about 80 kb from 68.10 to 68.18 Mb (Fig. 2B). They were in strong LD: the robust LD block structures did not differ between females and males or between LOAD and controls in any sample set (Supplementary Material, Fig. S1). Four haplotypes were estimated in each LD block consisting of the seven SNPs: three major haplotypes (frequency >0.1), [H1]C-A-T-T-T-A-T, [H2]A-G-C-C-G-G-C and [H3]A-G-C-T-T-G-C, and one minor haplotype, [H4]C-A-T-T-T-A-C (Table 6). H1 exhibited the highest frequency (range 0.4363–0.5356) and H4 the lowest (range 0.0084–0.031). Haplotypes H1, H2 and H3 were always estimated with the expectation-maximization (EM) algorithm in the four sample sets examined. Haplotype H4 was not inferred in either Negative-{varepsilon}4 or {varepsilon}3*3 consisting of male subjects.


View this table:
[in this window]
[in a new window]

  		Table 6. Case–control haplotype analysis

 
Because multiple SNPs may increase the risk of LOAD in combination, we carried out a case–control haplotype analysis (Table 6). In the All set, haplotypes H1 (permutation P-value = 0.0029) and H3 (permutation P-value = 0.0043) exhibited significant association in females. In both the Negative-{varepsilon}4 and {varepsilon}3*3 sets, haplotypes H1, H2 and H3 exhibited significance in females (permutation P-value H1<H2<H3). In the All, Negative-{varepsilon}4 and {varepsilon}3*3 sets, the frequency of haplotype H1 was decreased in LOAD, suggesting it is a protective haplotype for LOAD. On the other hand, haplotypes H2 and H3 were increased in LOAD, implying that they are risk haplotypes for LOAD. In males, each haplotype showed no significant difference in any sample set.

Of the four sample sets of females, three showed significant association in global tests: All (global permutation P-value = 0.0006), Negative-{varepsilon}4 (global permutation P-value = 0.0008), and {varepsilon}3*3 (global permutation P-value = 0.001). We did not detect significance in any haplotype in the female sub-sample set Positive-{varepsilon}4 (global permutation P-value = 0.3323).

Relationship between the Aß40/42 ratio and genetic variation on CTNNA3
The levels of plasma Aß40 and Aß42 and their ratio (Aß40/42) were compared between LOAD patients (N = 456) and control subjects (N = 147) within different gender groups (Fig. 3A–C). The Mann–Whitney U-test was adopted as a non-parametric method for this analysis. In both the female and male groups, the Aß40 levels (Fig. 3A) and A ß40/42 ratio (Fig. 3C) were significantly higher in LOAD in comparison with those in controls. The Aß42 levels were significantly lower in LOAD compared with those in controls (Fig. 3B).


Figure 3
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Comparison of the plasma levels of Aß40 and Aß42, and the Aß40/42 ratio. The differences in the relative amounts of Aß40 (A) and Aß42 (B), and the Aß40/42 ratio (C) were compared between LOAD patients and controls by means of Mann–Whitney's U-test within different gender groups. (D, E, F) Correlation between the Aß40/42 ratio, an associated SNP on CTNNA3, and the diagnosis (LOAD or control). Using log-transformed Aß40/42 ratio values, two-way ANOVA tests were performed after Bartlett's test for the homogeneity of variances and the KS normality test. The results for SNP rs713250 are presented here as being representative of the seven associated SNPs identified in this study. The horizontal line inside each box denotes the median value. The box extends from the 25th and 75th percentiles. The error bars extend down to the lowest value and up to the highest. Genotypes CC, CT and TT represent major-allele homozygotes, heterozygotes and minor-allele homozygotes, respectively.

 
To determine whether or not the difference in the Aß40/42 ratio between LOAD and the controls is due to the SNPs identified here, two-way ANOVA was performed across diagnosis (LOAD and control) and three genotypic groups (major homozygotes, heterozygotes and minor homozygotes) within different gender and their combined groups (Fig. 3D–F). SNP rs713250 was used as a representative of the seven associated SNPs because it showed the most significant association with LOAD on Mantel–Haenszel test (allelic P-valueMH-F = 0.000005945), as shown in Table 2. The log-transformed Aß40/42 ratio values [log2(Aß40/42 ratio + 1)] were used in this analysis. Before two-way ANOVA, the Kolmogorov–Smirnov (KS) normality test and Bartlett's test for equal variances were performed for the each dataset as to gender. Almost every sub-group examined passed the KS normality test. Both the female–male (Fig. 3D) and female (Fig. 3E) groups passed the Bartlett's test, but not the male group (Fig. 3F, P = 0.01178). Through two-way ANOVA, a significant effect of diagnosis was observed for every group (P-values <0.0001). However, we did not detect any genotype-dependent effect of this SNP on the Aß40/42 ratio, and no interaction between the SNP, Aß40/42 ratio and diagnosis.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
In this study, we extended our previous work on chromosome 10q (26), and thoroughly reanalyzed the genotype data for 1140 SNPs in order to discover gender-related genetic loci for LOAD. In a single SNP-based case–control study, we found seven SNPs on CTNNA3 showing genetic association with LOAD in females with the APOE-{varepsilon}3*3 genotype or without the APOE-{varepsilon}4 allele. Furthermore, multiple logistic regression analysis revealed that one (SNP rs713250) of these seven SNPs directly interacted with the female gender, but not with the male gender, and did not show any interaction with the APOE-{varepsilon}4 allele at all. These are the first findings constituting evidence that CTNNA3 may affect the development of sporadic LOAD through a novel female-specific mechanism independent of the APOE-{varepsilon}4 allele. We consider the genetic association identified here to reflect one single signal. The reasons are: (1) the seven significant SNPs span only ~38 kb and are clustered in intron 9 of CTNNA3 (Fig. 2A and C), which suggests a multiple-hit genomic region of SNPs associated with LOAD; (2) solid linkage disequilibrium was observed between all of these seven SNPs (D' > 0.9) (Supplementary Material, Fig. S1); and (3) the associated region was encompassed by a tight structured LD block extending ~80 kb (Fig. 2B).

Janssens et al. (29,30) cloned full-length CTNNA3 cDNA as a novel member of the {alpha}-catenin gene family and determined its genomic structure. CTNNA3 contains 18 exons and spans ~1.78 Mb (67.35–69.13 Mb), being the longest of all genes located on chromosome 10. The chromosomal location of CTNNA3 is 10q21 (30), which includes the suggestive linkage region between microsatellite markers D10S1227 (57.20 Mb) and D10S1211 (66.39 Mb) in LOAD (24). Ertekin-Taner et al. (23) found a linkage with a maximum LOD score of 3.93 at 81 cM close to D10S1225 (64.43 Mb) using the plasma Aß42 level as a surrogate trait in a set of LOAD families, and the same chromosomal region was identified by Myers et al. (24) by means of genome-wide screening of sibling pairs with LOAD. To date, there have been six papers on the genetic association of CTNNA3 with LOAD (32–37). In the first report (32), it was demonstrated that two SNPs located in intron 13 of CTNNA3 are associated with familial LOAD with high levels of plasma Aß42, which was used as an intermediate phenotype related to AD. These intronic SNPs, spanning 423 bp, are rs12357560 and rs7070570: the former lies 1174 bp upstream, and the latter 1597 bp downstream from exon 14, respectively. They are in strong LD: D' = 1 in all four populations, CEU, CHB, JPT and YRI, used in the HapMap project (38). A genotype-dependent correlation between SNP rs7070570 and the plasma Aß42 level has also been detected: the major homozygote (TT) is associated with the highest level of Aß42, the heterozygote (TC) with an intermediate level and the minor homozygote (CC) with the lowest level (32). Martin et al. (34). found that SNP rs7074454 located in intron 13 of CTNNA3, lying 355 bp upstream from SNP rs7070570, was significantly associated with both familial and sporadic cases of LOAD. Non-synonymous SNP rs4548513 (AGC->AAC, Ser596Asn) located in exon 13 of CTNNA3, lying 175 721 bp upstream from SNP rs7070570, has been shown to be associated with familial AD (37). All of these four SNPs, rs7070570, rs12357560, rs7074454 and rs4548513, lie in a genomic region extending from exons 13 to 14 (Fig. 2A), which has been shown to be located within a large LD block spanning around 310 kb (67.43–67.74 Mb) in CEU subjects (37) (Supplementary Material, Fig. S2). They have a tendency to exhibit selective association with familial rather than sporadic LOAD (32,35,37). Therefore, it is likely that the large LD block region contributes to a specific form of familial LOAD in Caucasians. We also assessed these four SNPs and SNPs neighboring them in our Japanese sporadic LOAD subjects, however, none of these SNPs exhibited significant association (data not shown). In the genomic region including the four SNPs, different LD block structures were observed in Japanese and CEPH subjects (Fig. 2B and Supplementary Material, Fig. S2). As one of the reasons why reproducible association could not be detected for these four SNPs, we mainly consider that an ethnic difference may exist.

High-level gene expression of CTNNA3 is detected predominantly in heart and testis, and low-level expression in several tissues including brain (29). Coimmunoprecipitation analysis revealed that CTNNA3 binds directly to ß-catenin in both a human cell line transfected with CTNNA3 cDNA, and heart and testis tissue extracts of mouse (30). ß-Catenin forms a complex with presenilin 1 (PSEN1) (31,39,40), mutations of which cause familial cases of early-onset AD (EOAD) [Alzheimer Disease & Frontotemporal Dementia Mutation Database (AD&FTDMDB), http://www.molgen.ua.ac.be/ADMutations/]. The expression level of ß-catenin is reduced in the brains of EOAD patients with PSEN1 mutations (31). Intracellular trafficking of ß-catenin is affected in human cells bearing PSEN1 mutations (41), resulting in sustained loss of Wnt/ß-catenin signal transduction, which is probably followed by the onset and development of AD (42,43). Although, at present, there is no direct evidence suggesting that CTNNA3 interacts with PSEN1, it is assumed that their genetic polymorphisms or combinations in CTNNA3 may have a negative influence on the Wnt/ß-catenin signaling pathway, leading to potential involvement in the pathogenesis of AD. In this study, it was clarified that seven intronic SNPs on CTNNA3 were significantly and reproducibly associated with sporadic female cases of LOAD without the APOE-{varepsilon}4 allele. Intronic variants are considered to have the potential to directly affect gene-expression levels in some cases (44); therefore, we performed quantitative real-time RT–PCR analysis of CTNNA3 using the postmortem brains of 19 neuropathologically-confirmed LOAD cases and 22 control ones. Two-way ANOVA revealed that there was no statistically significant interaction between the CTNNA3 expression level, the associated SNPs identified here and the diagnosis (data not shown). Additionally, although a genotype-dependent transition effect on the plasma Aß42 level was observed for intronic SNP rs7070570 by Ertekin-Taner et al. (32), it was found that none of these SNPs influence the plasma levels of Aß peptides (Fig. 3D–F).

However, interestingly, by means of a search of a public genome database, the Database of Genomic Variants (http://projects.tcag.ca/variation/), we discovered that there is copy number variation (CNV) (45) in the genomic region comprising the seven associated SNPs on CTNNA3: variation ID 3807 at Locus 2128, which was detected in a Japanese subject (ID, NA18973) (Fig. 2A). CNV, i.e. deletion, insertion and duplication with >1 kb in length of the genomic sequence (46), rather than SNP could cause phenotypic diversity and complex diseases in humans by altering the gene dose or disrupting the coding or regulatory sequences of genes, and may account for the LOAD susceptibility. Regarding our LOAD subjects, we did not examine the presence or absence of CNV within CTNNA3. Therefore, in a further study, it is very important to determine whether or not CNV in CTNNA3 is associated with LOAD.

Recently, in LOAD families, notable evidence was obtained suggesting a maternal parent-of-origin effect on chromosome 10q between microsatellite markers D10S1233 (44.05 Mb) and D10S1225 (64.43 Mb) with a non-parametric LOD score >1.0: the highest LOD score of 3.73 was seen for microsatellite marker D10S1221 (57.20 Mb) (27,28). Moreover, it was found that CTNNA3 is subject to genomic imprinting with cell-type specificity in placental tissues: biallelic and monoallelic (maternal-allele) expression is observed in extravillus and villus trophoblasts, respectively (47). Mouse Ctnna3 (Clone ID 4933408A16 on FANTOM2), orthologous to human CTNNA3, has been deposited as a maternal imprinting gene on chromosome 10 in the Expression-based Imprint Candidate Organizer DataBase (48; EICO DB, http://fantom2.gsc.riken.jp/EICODB/imprinting/), provided by RIKEN (Japan). These findings led us to examine whether or not CTNNA3 shows allele-specific expression caused by a molecular mechanism such as genomic imprinting in the brain. We conducted real-time RT–PCR analysis with allele-specific amplification using postmortem human brains heterozygous for non-synonymous SNP rs4548513 in exon 13 [LOAD, 7 (female:male = 3:4); control, 8 (female:male = 3:5)]. Unexpectedly, biallelic expression was detected in brain tissues, and there was no significant difference between LOAD patients and control subjects in the expression level of CTNNA3 (data not shown). Since as in placental tissues, as described above, it is possible that cell-type dependent imprinting for CTNNA3 may occur in the brain, further expression analysis should be carefully carried out using homogeneous populations of specific cells from brain tissues. Now genome-wide prediction and the discovery of imprinted genes have progressed (49,50), and 600 (2.5%) of 23 788 annotated autosomal genes have been found to be potentially imprinted in the mouse genome by computational estimation: 384 (64%) of these candidate-imprinted genes show maternal-allele expression (50). It is expected that failure of imprinted gene expression in the human brain may lead to cognition and behavior defects such as Alzheimer's disease, schizophrenia, the bipolar affective disorder and epilepsy (51–53). Therefore, it is important and interesting to actively examine imprinted genes present in the genetic linkage region of LOAD.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 
Subjects
The Japanese Genetic Study Consortium for AD (JGSCAD) was organized in 2000, and blood samples were collected to survey risk genes for LOAD by means of a genome-wide association study. All individuals included in this study were Japanese. Probable AD cases met the criteria of the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders. Control subjects who had no signs of dementia and lived in an unassisted manner in the local community were also recruited. Age at onset (AAO) is here defined as the age at which the family and/or individuals first noted cognitive problems during work or in daily activities. The Mini-Mental State Examination (MMSE), and Clinical Dementia Rating and/or the Function Assessment Staging were used for the evaluation of cognitive impairment: MMSE was used for almost every subject.

The basic demographics of the LOAD patients and non-demented control subjects are presented in Table 1. A total of 3192 subjects comprising 1526 LOAD patients [female, 1103 (72.3%); male, 423 (27.7%)] and 1666 controls [female, 998 (59.9%); male, 668 (40.1%)], which is referred to as overall sample set All in this study, were used to discover gender-related loci associated with LOAD on chromosome 10q: information on these subjects was also presented in our recent paper, Kuwano et al. (26). The mean AAO ± standard deviation (SD) in the 1526 LOAD patients was 73.5 ± 6.6 (range 60–93). The mean age at examination (AAE) ± SD of the control subjects was 73.1 ± 7.8 (range 60–96). There was no significant difference between AAO in LOAD patients and AAE in control subjects with the unpaired Student's t-test (P-value = 0.1239). The mean MMSE score in the 1526 LOAD patients was 16.5 (SD 7.0), which was significantly lower (P-value with unpaired Student's t-test <0.0001) than that in the 1666 controls (mean ± SD 28.0 ± 1.8). The numbers (frequency) of APOE-{varepsilon}2*2, {varepsilon}2*3, {varepsilon}2*4, {varepsilon}3*3, {varepsilon}3*4 and {varepsilon}4*4 in the 1526 LOAD subjects were 1 (0.07%), 49 (3.21%), 17 (1.11%), 699 (45.81%), 613 (40.17%) and 147 (9.63%), and those in the 1666 control subjects were 3 (0.18%), 132 (7.92%), 15 (0.90%), 1243 (74.61%), 256 (15.37%) and 17 (1.02%). The allelic distribution of APOE was significantly different between LOAD patients ({varepsilon}2, 68; {varepsilon}3, 2060; {varepsilon}4, 924) and control subjects ({varepsilon}2, 153; {varepsilon}3, 2874; {varepsilon}4, 305), as expected (P-value with {chi}2 test using a 2 x 3 contingency table, <0.0001).

The present study was approved by the Institutional Review Board of Niigata University and by all participating institutes. Informed consent was obtained from all controls and appropriate proxies for patients, and all samples were anonymously analyzed for genotyping.

SNPs and genotyping
SNP information was obtained from five open databases: NCBI dbSNP (Build 125, http://www.ncbi.nlm.nih.gov/SNP/), UCSC Genome Bioinformatics (http://genome.ucsc.edu/), International HapMap Project (Rel#20/phaseII on NCBI Build 35.1 assembly and dbSNP Build 125, http://www.hapmap.org/index.html), Ensemble Human (Version 37 on NCBI Build 35.1, http://www.ensembl.org/Homo_sapiens/) and Celera myScience (Version R27 g on NCBI Build 35.1, http://myscience.appliedbiosystems.com/). We selected 1322 SNPs in the region from 60 to 107 Mb on chromosome 10q; mean intermarker distance ± SD, 34.9 ± 87.4 kb; 95% CI, 30.2–39.6 kb. The information on all SNPs, including rs or Celera IDs and genomic positions on NCBI build 35.1, used here was presented in detail elsewhere (26). These SNPs consisted of 29 missense mutations, 27 silent mutations, 6 SNPs in the 5'-UTR, 29 SNPs in the 3'-UTR, 921 SNPs in introns, 282 SNPs in intergenic regions and 28 SNPs in four loci shared by two different genes (CTNNA3/LRRTM3, CDH23/C10orf54, C10orf55/PLAU and PGAM1/EXOSC1). Among the 1322 SNPs, 28 SNPs could not be genotyped. To examine deviation from HWE of 1294 SNPs, exact tests (details given under Statistical analysis) were performed with both 363 LOAD patients and 337 control subjects (carrying APOE-{varepsilon}3*3 in the exploratory sample set, as shown in Table 1). We used 1140 SNPs that were shown to be actually polymorphic in the Japanese population and showed P-values >0.05 with the exact tests; mean intermarker distance ± SD, 40.5 ± 96.7 kb; 95% CI, 34.9–46.1 kb.

Genomic DNA was extracted from peripheral blood with a QIAamp DNA Blood Maxi Kit (Qiagen, Duseldorf, Germany) and examined fluorometrically with a PicoGreen dsDNA quantification kit (Molecular Probes, California, USA). SNP genotyping of individual samples was performed with an ABI PRISM 7900HT instrument using TaqMan technology, and TaqMan SNP Genotyping Assays were purchased from Applied Biosystems (California, USA).

Case–control study
To discover gender-related genetic loci on chromosome 10q (60–107 Mb on NCBI build 35.1), allelic association was assessed by means of the {chi}2 test based on a 2 x 2 contingency table in comparison with allele frequencies in LOAD patients and control subjects within different gender groups. For screening, two independent sample sets, Exploratory and Validation, comprising case–control subjects with APOE-{varepsilon}3*3 were first used after being stratified as to gender (Table 1). Sample set Exploratory comprising 363 LOAD patients and 337 control subjects was genotyped (26), and SNPs showing significant association (allelic P-value <0.01) were then subjected to further examination using another sample set, Validation, comprising 336 LOAD patients and 372 control subjects. Multistage, including two-stage, genotyping designs for large-scale association surveys have been proved to be practically as well as theoretically effective for identifying common genetic variants that predispose to human disease (54–58). Therefore, we considered that replication in both the Exploratory and Validation sample sets implicates an association of particular SNPs with LOAD.

Subsequently, for stratified analysis we increased the number of subjects and constructed an overall sample set, All. Furthermore, to construct three sub-sample sets, overall sample set All was stratified as to the APOE carrier status: Negative-{varepsilon}4, APOE-{varepsilon} 2*2, 2*3 and 3*3; {varepsilon}3*3, APOE-{varepsilon} 3*3; Positive-{varepsilon}4, APOE-{varepsilon} 2*4, 3*4 and 4*4 (Table 1). The sample numbers for LOAD patients and controls in All, Negative-{varepsilon}4, {varepsilon}3*3 and Positive-{varepsilon}4 were 1526 and 1666, 749 and 1378, 699 and 1243, and 777 and 288, respectively. These four sample sets were used for the {chi}2 test after being sub-grouped as to gender.

Case–control haplotype analysis with significant SNPs was also performed using the following sample sets: All, Negative-{varepsilon}4, {varepsilon}3*3 and Positive-{varepsilon}4. These four sample sets were used after being stratified as to gender.

Aß40 and Aß42 quantification
For Aß40 and Aß42 quantification, 603 subjects consisting of 456 LOAD patients (female, 332; male, 124) and 147 control subjects (female, 95; male, 52) were used. They are included in the All set. The sandwich enzyme-linked immunosorbent assay (59–61) was used to specifically quantify whole plasma Aß species. The standardization, sensitivity and specificity of the method were described in a previous paper (61). Briefly, microplates (Immunoplate I; Nunc, Rockilde, Denmark) were pre-coated with monoclonal BNT77 (IgA isotype specific for Aß11–16) and then sequentially incubated for 24 h at 4°C (100 µl of whole plasma/well), followed by 24 h incubation at 4°C with horseradish-peroxidase-conjugated BA27 (anti-Aß1–40, specific for Aß40) or BC05 (anti-Aß35–43, specific for Aß42). Color was developed with 3,3',5,5'-tetramethylbenzidine and evaluated at 450 nm with a microplate reader (Molecular Devices, CA). Synthetic Aß40 and Aß42 (Sigma, St Louis, MO) of known concentration (estimated from the amino acid composition) were used as standards. The plates were normalized as to each other by inclusion of three standard plasma samples on all plates.

Statistical analysis
Allele frequencies were calculated by allele counting. To evaluate deviation from the HWE of each SNP marker, we carried out an exact test (62) based on the probability of occurrence of genotypic contingency tables with fixed total numbers of alleles within each sample set (LOAD patients and controls included in two screening sets, Exploratory and Validation). For single SNP case–control analysis, the allelic distributions in LOAD patients and controls were compared by means of {chi}2 tests via standard 2 x 2 contingency tables. Evidence of replication, rather than multiple testing corrections, was used to evaluate the significance of associated SNPs. To comprehensively assess the reproducible SNPs, we conducted a Mantel–Haenszel test, where Exploratory and Validation samples in our case–control study were considered as the strata (63), and computed pooled ORs with 95% CI and P-values from Mantel–Haenszel statistics (Statcel 2; OMS, Tokyo, Japan). Estimation of haplotypes and their frequencies was carried out for LOAD patients and controls separately by the maximum-likelihood method from unphased diploid genotype data using an EM algorithm (64) with the following parameters: iteration counter, 5000; conversion criterion, 0.000001. To assess the differences in haplotype distribution between LOAD patients and controls, a permutation test (65) was performed. In this test, all permutation P-values were empirically computed using 10 000 iterations of random sampling with fixed total numbers of both LOAD and control subjects. OR (95% CI), as an estimate of the relative risk of disease, of each marker or haplotype was calculated from a 2 x 2 contingency table. For all statistical methods mentioned above, except the Mantel–Haenszel test, we used SNPAlyze software versions 3.2.3 or 6.0.1 (DYNACOM, Chiba, Japan; http://www.dynacom.co.jp/). For calculation of LD measures (D') and LD block definition by Gabriel et al.'s method (66), we used Haploview version 3.32 (67, http://www.broad.mit.edu/mpg/haploview/index.php).

Using SPSS version 13.0 software (SPSS, Chicago, USA), multiple logistic regression analysis (Table 5) was performed to reveal the effects of the APOE-{varepsilon}4 [non-carrier of the {varepsilon}4 allele ({varepsilon}2*2, {varepsilon}2*3 and {varepsilon}3*3)/carrier of the {varepsilon}4 allele ({varepsilon}2*4, {varepsilon}3*4 and {varepsilon}4*4)], gender (male/female), age and significant SNPs identified here (major-allele homozygote/heterozygote/minor-allele homozygote) on the risk for LOAD as well as their second-order interaction terms. The strength of association between these variables and disease status (control/LOAD) was evaluated with ORs with 95% CI, based on Wald statistics. We examined the four variables by means of a two-step multiple logistic regression analysis according to Akazawa et al. (68). In order to examine which variables explain an association with LOAD independently, we initially carried out stepwise logistic regression analysis (forward selection method) without interaction terms. A significance level of 0.05 was used to enter a variable in the model. Through this analysis, the following multiple logistic regression model was fitted (Model 1 in Table 5): log(P/(1–P)) = {alpha} + ß1X1 + ß2X2 + ß3X3 + ß4X4, where P denotes the probability of having LOAD, {alpha} is the intercept, ßi represents the estimated parameters and Xj the independent variables (X1, APOE-{varepsilon}4; X2, gender; X3, age; X4, SNP). We next analyzed the four variables including their second-order interaction terms (SNP_gender, SNP_APOE-{varepsilon}4, SNP_age, gender_APOE-{varepsilon}4, gender_age and age_APOE-{varepsilon}4) by means of a forward stepwise regression method with a significance level of 0.05 for the inclusion of a variable in the model. As a result, the following model was fitted (Model 2 in Table 5): log(P/(1–P)) = {alpha} + ß1X1 + ß2X2 + ß3X3 + ß4X4 + ß5X5 + ß6X6 + ß7X7, where P denotes the probability of having LOAD, {alpha} is the intercept, ßi represents the estimated parameters and Xj the independent variables (X1, APOE-{varepsilon}4; X2, gender; X3, age; X4, SNP; X5, SNP_gender; X6, gender_APOE-{varepsilon}4; X7, age_APOE-{varepsilon}4). Subjects with undetermined SNP genotype data were omitted for multiple logistic regression analysis.

The Mann–Whitney U-test was applied to compare differences in the levels of Aß40 and Aß42, and their ratio (Aß40/42) between LOAD patients and controls (Prism 4.0b; GraphPad Software, CA, USA). After Bartlett's test for the homogeneity of variances (Statcel 2) and the KS normality test (Prism 4.0b), the effects of three SNP genotypes (minor-allele homozygotes, heterozygotes and major-allele homozygotes) in three sub-groups stratified as to gender (female–male mixture, female or male) were examined as to levels of the plasma Aß40/42 ratio using two-way ANOVA (Prism 4.0b). To create more normally distributed datasets, the Aß40/42 ratio was subjected to log transformation [log2(Aß40/42 ratio + 1)] before the two-way ANOVA.

The statistical significance was set at P<0.05.


   SUPPLEMENTARY MATERIAL
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
FUNDING
 REFERENCES
 
Supplementary Material is available at HMG Online.


   FUNDING
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
REFERENCES
 
This study was supported by KAKENHI (Grant-in-Aid for Scientific Research) on Priority Areas, Comprehensive Genomics (R.K.), and a Grant-in-Aid for Scientific Research on Priority Areas (C), Advanced Brain Science Project (Y.I.), from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and a Grant for the Promotion of Niigata University Research Projects, Japan (A.M.).


   ACKNOWLEDGEMENTS
 
We wish to thank the patients with AD and their families, and the control individuals for their participation in this study, without whom this genetic work would have been impossible. We also thank N. Takei, K. Horigome, M. Hirose, N. Yahata, T. Tsukie, K. Takadono, A. Kitamura, Y. Satoh, A. Hirokawa, T. Hoshino, S. Yanagihara and M. Saitoh for their technical assistance, and N. J. Halewood for critical reading of the manuscript.

The members of JGSCAD who participated in the collection of blood samples from AD patients and controls were as follows: All authors on the title page; and Akihiko Nunomura, MD, and Shigeru Chiba, MD, Department of Psychiatry and Neurology, Asahikawa Medical College, Asahikawa; Takeshi Kawarabayashi, MD, Department of Neurology, Neuroscience and Biophysiological Science, Hirosaki University, School of Medicine, Hirosaki; Satoshi Takahashi, MD, Department of Neurology, Iwate Medical University, Morioka; Naoki Tomita, MD, Department of Geriatric and Complementary Medicine, Tohoku University Graduate School of Medicine, Sendai; Jyunzo Ito, MD, Alpine Kawasaki, Kawasaki, Miyagi; Haruo Hanyu, MD, Department of Geriatric Medicine, Tokyo Medical University, Tokyo; Shin Kitamura, MD, Second Department of Internal Medicine, Nippon Medical School, Tokyo; Hitoshi Shinotoh, MD, Asahi Hospital for Neurological Disease, Chiba; Hiroyuki Iwamoto, MD, Department of Neurology, Hatsuishi Hospital, Kashiwa; Masahiko Takahashi, MD, Department of Old Age Psychiatry and Memory Clinic, Tokyo Metropolitan Geriatric Medical; Yasuo Harigaya, MD, Department of Neurology, Maebashi Red Cross Hospital, Gunma; Masaki Ikeda, MD, and Masakuni Amari, MD, Department of Neurology, Gunma University Graduate School of Medicine, Maebashi; Takeo Takahashi, MD, Ina Neurological Hospital, Ina; Ryoichi Nakano, MD, and Masatoyo Nishizawa, MD, Department of Neurology, Brain Research Institute, Niigata University, Niigata; Takeshi Ikeuchi, MD, and Osamu Onodera, MD, Department of Molecular Neuroscience, Bioresource Science Branch, Center for Bioresources, Brain Research Institute, Niigata University, Niigata; Masaichi Suga, MD, Higashi Niigata Hospital, Niigata; Makoto Hasegawa, MD, Niigata Shin-ai Hospital, Niigata; Yasuhiro Kawase, MD, Kawase Neurology Clinic, Sanjo; Kenichi Honda, MD, Honda Hospital, Uonuma; Toshiro Kumanishi, MD, and Yukiyosi Takeuchi, MD, Niigata Longevity Research Institute, Shibata; Atsushi Ishikawa, MD, Department of Neurology, Brain Disease Center, Agano Hospital, Agano; Masahiro Morita, MD, Department of Psychiatry, Mishima Hospital, Mishima; Fumihito Yoshii, MD, Department of Neurology, Tokai University School of Medicine, Isehara; Hiroyasu Akatsu, MD, and Kenji Kosaka, MD, Choju Medical Institute, Fukushimura Hospital, Toyohashi; Masahito Yamada, MD, and Tsuyoshi Hamaguchi, MD, Department of Neurology and Neurobiology of Aging, Kanazawa University Graduate School of Medical Science, Kanazawa; Satoshi Masuzugawa, MD, Department of Neurology, Mie Prefectural Shima Hospital, Shima; Takeo Takao, MD, and Nobuko Ota, Kurashiki Heisei Hospital, Kurashiki; Ken Sasaki, MD, Yoshikatsu Fujisawa, MD, and Kenji Nakata, MD, Kinoko Espoir Hospital, Kasaoka; Ken Watanabe, MD, Watanabe Hospital, Tottori; Yosuke Wakutani, MD, and Kenji Nakashima, MD, Department of Neurology, Institute of Neurological Science, Tottori University, Yonago; Toshiyuki Hayabara, MD, Iwaki Hospital, Kagawa; Terumi Ooya, Town Office, Oonan, Shimane; Mitsuo Takahashi, MD, Department of Clinical Pharmacology, Fukuoka University, Fukuoka; Tatsuo Yamada, MD, Fifth Department of Internal Medicine, Fukuoka University, Fukuoka; Taihei Miyakawa, MD, Labour Welfare Corporation Kumamoto Rosai Hospital, Yatsushiro; Eiichiro Uyama, MD, Department of Neurology, Graduate School of Medical Science, Kumamoto University, Kumamoto; Takefumi Yuzuriha, MD, Department of Psychiatry, National Hospital Organization Hizen Psychiatric Center, Sefuri; and Ryuji Nakagawa, MD, Shizushi Yoshimoto, MD, and Kayoko Serikawa, MD, Ureshino-Onsen Hospital, Saga.

Conflict of interest statement. None declared.


   FOOTNOTES
 
{dagger} Present address: Risk Management Office, Niigata University Medical and Dental Hospital, Niigata 951-8520, Japan Back

{ddagger} Present address: Planning Office, Faculty of Life and Medical Sciences, Doshisha University, Kyoto 619-0225, Japan Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 SUPPLEMENTARY MATERIAL
 FUNDING
 REFERENCES
 

  1. Bachman D.L., Wolf P.A., Linn R., Knoefel J.E., Cobb J., Belanger A., D'Agostino R.B., White L.R. Prevalence of dementia and probable senile dementia of the Alzheimer type in the Framingham Study. Neurology (1992) 42:115–119.[Abstract/Free Full Text]

  2. Hy L.X., Keller D.M. Prevalence of AD among whites: a summary by levels of severity. Neurology (2000) 55:198–204.[Abstract/Free Full Text]

  3. Nakamura S., Shigeta M., Iwamoto M., Tsuno N., Niina R., Homma A., Kawamuro Y. Prevalence and predominance of Alzheimer type dementia in rural Japan. Psychogeriatrics (2003) 3:97–103.[CrossRef]

  4. Andersen K., Launer L.J., Dewey M.E., Letenneur L., Ott A., Copeland J.R., Dartigues J.F., Kragh-Sorensen P., Baldereschi M., Brayne C., et al. Gender differences in the incidence of AD and vascular dementia: The EURODEM Studies. EURODEM Incidence Research Group. Neurology (1999) 53:1992–1997.[Abstract/Free Full Text]

  5. Fratiglioni L., Launer L.J., Andersen K., Breteler M.M., Copeland J.R., Dartigues J.F., Lobo A., Martinez-Lage J., Soininen H., Hofman A., and Neurologic Diseases in the Elderly Research Group. Incidence of dementia and major subtypes in Europe: a collaborative study of population-based cohorts. Neurology (2000) 54(Suppl. 5):S10–S15.[Web of Science][Medline]

  6. Fujishima M., Kiyohara Y. Incidence and risk factors of dementia in a defined elderly Japanese population: the Hisayama study. Ann. NY Acad. Sci. (2002) 977:1–8.[CrossRef][Web of Science][Medline]

  7. Kukull W.A., Higdon R., Bowen J.D., McCormick W.C., Teri L., Schellenberg G.D., van Belle G., Jolley L., Larson E.B. Dementia and Alzheimer disease incidence: a prospective cohort study. Arch. Neurol. (2002) 59:1737–1746.[Abstract/Free Full Text]

  8. Edland S.D., Rocca W.A., Petersen R.C., Cha R.H., Kokmen E. Dementia and Alzheimer disease incidence rates do not vary by sex in Rochester, Minn. Arch. Neurol. (2002) 59:1589–1593.[Abstract/Free Full Text]

  9. Gatz M., Fiske A., Reynolds C.A., Wetherell J.L., Johansson B., Pedersen N.L. Sex differences in genetic risk for dementia. Behav. Genet. (2003) 33:95–105.[CrossRef][Web of Science][Medline]

  10. Maes O.C., Xu S., Yu B., Chertkow H.M., Wang E., Schipper H.M. Transcriptional profiling of Alzheimer blood mononuclear cells by microarray. Neurobiol. Aging (2006) doi:10.1016/j.neurobiolaging.2006.08.004.

  11. Assini A., Cammarata S., Vitali A., Colucci M., Giliberto L., Borghi R., Inglese M.L., Volpe S., Ratto S., Dagna-Bricarelli F., et al. Plasma levels of amyloid beta-protein 42 are increased in women with mild cognitive impairment. Neurology (2004) 63:828–831.[Abstract/Free Full Text]

  12. Wang J., Tanila H., Puolivali J., Kadish I., van Groen T. Gender differences in the amount and deposition of amyloid beta in APPswe and PS1 double transgenic mice. Neurobiol. Dis. (2003) 14:318–327.[CrossRef][Web of Science][Medline]

  13. van Groen T., Kiliaan A.J., Kadish I. Deposition of mouse amyloid beta in human APP/PS1 double and single AD model transgenic mice. Neurobiol. Dis. (2006) 23:653–662.[CrossRef][Web of Science][Medline]

  14. Maynard C.J., Cappai R., Volitakis I., Cherny R.A., Masters C.L., Li Q.X., Bush A.I. Gender and genetic background effects on brain metal levels in APP transgenic and normal mice: implications for Alzheimer beta-amyloid pathology. J. Inorg. Biochem. (2006) 100:952–962.[CrossRef][Web of Science][Medline]

  15. Schafer S., Wirths O., Multhaup G., Bayer T.A. Gender dependent APP processing in a transgenic mouse model of Alzheimer's disease. J. Neural Transm. (2007) 114:387–394.[CrossRef][Web of Science][Medline]

  16. Kamboh M.I., Sanghera D.K., Aston C.E., Bunker C.H., Hamman R.F., Ferrell R.E., DeKosky S.T. Gender-specific nonrandom association between the alpha 1-antichymotrypsin and apolipoprotein E polymorphisms in the general population and its implication for the risk of Alzheimer's disease. Genet. Epidemiol. (1997) 14:169–180.[CrossRef][Web of Science][Medline]

  17. Reynolds W.F., Rhees J., Maciejewski D., Paladino T., Sieburg H., Maki R.A., Masliah E. Myeloperoxidase polymorphism is associated with gender specific risk for Alzheimer's disease. Exp. Neurol. (1999) 155:31–41.[CrossRef][Web of Science][Medline]

  18. Crawford F.C., Freeman M.J., Schinka J.A., Morris M.D., Abdullah L.I., Richards D., Sevush S., Duara R., Mullan M.J. Association between Alzheimer's disease and a functional polymorphism in the Myeloperoxidase gene. Exp. Neurol. (2001) 167:456–459.[CrossRef][Web of Science][Medline]

  19. Crawford F., Abdullah L., Schinka J., Suo Z., Gold M., Duara R., Mullan M. Gender-specific association of the angiotensin converting enzyme gene with Alzheimer's disease. Neurosci. Lett. (2000) 280:215–219.[CrossRef][Web of Science][Medline]

  20. Pirskanen M., Hiltunen M., Mannermaa A., Helisalmi S., Lehtovirta M., Hanninen T., Soininen H. Estrogen receptor beta gene variants are associated with increased risk of Alzheimer's disease in women. Eur. J. Hum. Genet. (2005) 13:1000–1006.[CrossRef][Web of Science][Medline]

  21. Ramos M.C., Tenorio R., Martinez-Garcia A., Sastre I., Vilella-Cuadrada E., Frank A., Rosich-Estrago M., Valdivieso F., Bullido M.J. Association of DSC1, a gene modulated by adrenergic stimulation, with Alzheimer's disease. Neurosci. Lett. (2006) 408:203–208.[CrossRef][Web of Science][Medline]

  22. Sundar P.D., Feingold E., Minster R.L., Dekosky S.T., Kamboh M.I. Gender-specific association of ATP-binding cassette transporter 1 (ABCA1) polymorphisms with the risk of late-onset Alzheimer's disease. Neurobiol. Aging (2006) doi:10.1016/j.neurobiolaging.2006.04.005.

  23. Ertekin-Taner N., Graff-Radford N., Younkin L.H., Eckman C., Baker M., Adamson J., Ronald J., Blangero J., Hutton M., Younkin S.G. Linkage of plasma Abeta42 to a quantitative locus on chromosome 10 in late-onset Alzheimer's disease pedigrees. Science (2000) 290:2303–2304.[Abstract/Free Full Text]

  24. Myers A., Holmans P., Marshall H., Kwon J., Meyer D., Ramic D., Shears S., Booth J., DeVrieze F.W., Crook R., et al. Susceptibility locus for Alzheimer's disease on chromosome 10. Science (2000) 290:2304–2305.[Abstract/Free Full Text]

  25. Bertram L., Blacker D., Mullin K., Keeney D., Jones J., Basu S., Yhu S., McInnis M.G., Go R.C., Vekrellis K., et al. Evidence for genetic linkage of Alzheimer's disease to chromosome 10q. Science (2000) 290:2302–2303.[Abstract/Free Full Text]

  26. Kuwano R., Miyashita A., Arai H., Asada T., Imagawa M., Shoji M., Higuchi S., Urakami K., Kakita A., Takahashi H., et al. Dynamin-binding protein gene on chromosome 10q is associated with late-onset Alzheimer's disease. Hum. Mol. Genet. (2006) 15:2170–2182.[Abstract/Free Full Text]

  27. Bassett S.S., Avramopoulos D., Fallin D. Evidence for parent of origin effect in late-onset Alzheimer disease. Am. J. Med. Genet. (2002) 114:679–686.[CrossRef][Web of Science][Medline]

  28. Bassett S.S., Avramopoulos D., Perry R.T., Wiener H., Watson B. Jr., Go R.C., Fallin M.D. Further evidence of a maternal parent-of-origin effect on chromosome 10 in late-onset Alzheimer's disease. Am. J. Med. Genet. B Neuropsychiatr. Genet (2006) 141:537–540.[Medline]

  29. Janssens B., Goossens S., Staes K., Gilbert B., van Hengel J., Colpaert C., Bruyneel E., Mareel M., van Roy F. AlphaT-catenin: a novel tissue-specific beta-catenin-binding protein mediating strong cell-cell adhesion. J. Cell. Sci. (2001) 114:3177–3188.[Abstract/Free Full Text]

  30. Janssens B., Mohapatra B., Vatta M., Goossens S., Vanpoucke G., Kools P., Montoye T., van Hengel J., Bowles N.E., van Roy F., et al. Assessment of the CTNNA3 gene encoding human alpha T-catenin regarding its involvement in dilated cardiomyopathy. Hum. Genet. (2003) 112:227–236.[Web of Science][Medline]

  31. Zhang Z., Hartmann H., Do V.M., Abramowski D., Sturchler-Pierrat C., Staufenbiel M., Sommer B., van de Wetering M., Clevers H., Saftig P., et al. Destabilization of beta-catenin by mutations in presenilin-1 potentiates neuronal apoptosis. Nature (1998) 395:698–702.[CrossRef][Medline]

  32. Ertekin-Taner N., Ronald J., Asahara H., Younkin L., Hella M., Jain S., Gnida E., Younkin S., Fadale D., Ohyagi Y., et al. Fine mapping of the alpha-T catenin gene to a quantitative trait locus on chromosome 10 in late-onset Alzheimer's disease pedigrees. Hum. Mol. Genet. (2003) 12:3133–3143.[Abstract/Free Full Text]

  33. Busby V., Goossens S., Nowotny P., Hamilton G., Smemo S., Harold D., Turic D., Jesus L., Myers A., Womick M., et al. Alpha-T-catenin is expressed in human brain and interacts with the Wnt signaling pathway but is not responsible for linkage to chromosome 10 in Alzheimer's disease. Neuromolecular Med. (2004) 5:133–146.[CrossRef][Web of Science][Medline]

  34. Blomqvist M.E., Andreasen N., Bogdanovic N., Blennow K., Brookes A.J., Prince J.A. Genetic variation in CTNNA3 encoding alpha-3 catenin and Alzheimer's disease. Neurosci. Lett. (2004) 358:220–222.[CrossRef][Web of Science][Medline]

  35. Martin E.R., Bronson P.G., Li Y.J., Wall N., Chung R.H., Schmechel D.E., Small G., Xu P.T., Bartlett J., Schnetz-Boutaud N., et al. Interaction between the alpha-T catenin gene (VR22) and APOE in Alzheimer's disease. J. Med. Genet. (2005) 42:787–792.[Abstract/Free Full Text]

  36. Cellini E., Bagnoli S., Tedde A., Nacmias B., Piacentini S., Sorbi S. Insulin degrading enzyme and alpha-3 catenin polymorphisms in Italian patients with Alzheimer disease. Alzheimer Dis. Assoc. Disord. (2005) 19:246–247.[CrossRef][Web of Science][Medline]

  37. Bertram L., Mullin K., Parkinson M., Hsiao M., Moscarillo T.J., Wagner S.L., Becker K.D., Velicelebi G., Blacker D., Tanzi R.E. Is alpha-T catenin (VR22) an Alzheimer's disease risk gene? J. Med. Genet. (2007) 44:e63.[Abstract/Free Full Text]

  38. International HapMap Consortium. A haplotype map of the human genome. Nature (2005) 437:1299–1320.[CrossRef][Medline]

  39. Murayama M., Tanaka S., Palacino J., Murayama O., Honda T., Sun X., Yasutake K., Nihonmatsu N., Wolozin B., Takashima A. Direct association of presenilin-1 with beta-catenin. FEBS Lett. (1998) 433:73–77.[CrossRef][Web of Science][Medline]

  40. Yu G., Chen F., Levesque G., Nishimura M., Zhang D.M., Levesque L., Rogaeva E., Xu D., Liang Y., Duthie M., et al. The presenilin 1 protein is a component of a high molecular weight intracellular complex that contains beta-catenin. J. Biol. Chem. (1998) 273:16470–16475.[Abstract/Free Full Text]

  41. Nishimura M., Yu G., Levesque G., Zhang D.M., Ruel L., Chen F., Milman P., Holmes E., Liang Y., Kawarai T., et al. Presenilin mutations associated with Alzheimer disease cause defective intracellular trafficking of beta-catenin, a component of the presenilin protein complex. Nat. Med. (1999) 5:164–169.[CrossRef][Web of Science][Medline]

  42. De Ferrari G.V., Inestrosa N.C. Wnt signaling function in Alzheimer's disease. Brain Res. Rev. (2000) 33:1–12.[CrossRef][Medline]

  43. De Ferrari G.V., Moon R.T. The ups and downs of Wnt signaling in prevalent neurological disorders. Oncogene (2006) 25:7545–7553.[CrossRef][Web of Science][Medline]

  44. Tokuhiro S., Yamada R., Chang X., Suzuki A., Kochi Y., Sawada T., Suzuki M., Nagasaki M., Ohtsuki M., Ono M., et al. An intronic SNP in a RUNX1 binding site of SLC22A4, encoding an organic cation transporter, is associated with rheumatoid arthritis. Nat. Genet. (2003) 35:341–348.[CrossRef][Web of Science][Medline]

  45. Redon R., Ishikawa S., Fitch K.R., Feuk L., Perry G.H., Andrews T.D., Fiegler H., Shapero M.H., Carson A.R., Chen W., et al. Global variation in copy number in the human genome. Nature (2006) 444:444–454.[CrossRef][Medline]

  46. Feuk L., Marshall C.R., Wintle R.F., Scherer S.W. Structural variants: changing the landscape of chromosomes and design of disease studies. Hum. Mol. Genet. (2006) 15(Spec. No. 1):R57–R66.[Abstract/Free Full Text]

  47. van Dijk M., Mulders J., Konst A., Janssens B., van Roy F., Blankenstein M., Oudejans C. Differential downregulation of alphaT-catenin expression in placenta: trophoblast cell type-dependent imprinting of the CTNNA3 gene. Gene Expr. Patterns (2004) 5:61–65.[CrossRef][Medline]

  48. Nikaido I., Saito C., Wakamoto A., Tomaru Y., Arakawa T., Hayashizaki Y., Okazaki Y. EICO (Expression-based Imprint Candidate Organizer): finding disease-related imprinted genes. Nucleic Acids Res. (2004) 32:D548–D551.[Abstract/Free Full Text]

  49. Nikaido I., Saito C., Mizuno Y., Meguro M., Bono H., Kadomura M., Kono T., Morris G.A., Lyons P.A., Oshimura M., et al. Discovery of imprinted transcripts in the mouse transcriptome using large-scale expression profiling. Genome Res. (2003) 13:1402–1409.[Abstract/Free Full Text]

  50. Luedi P.P., Hartemink A.J., Jirtle R.L. Genome-wide prediction of imprinted murine genes. Genome Res. (2005) 15:875–884.[Abstract/Free Full Text]

  51. Isles A.R., Wilkinson L.S. Imprinted genes, cognition and behaviour. Trends Cogn. Sci. (2000) 4:309–318.[CrossRef][Web of Science][Medline]

  52. Morison I.M., Ramsay J.P., Spencer H.G. A census of mammalian imprinting. Trends Genet. (2005) 21:457–465.[CrossRef][Web of Science][Medline]

  53. Davies W., Isles A.R., Wilkinson L.S. Imprinted gene expression in the brain. Neurosci. Biobehav. Rev. (2005) 29:421–430.[CrossRef][Web of Science][Medline]

  54. Ozaki K., Ohnishi Y., Iida A., Sekine A., Yamada R., Tsunoda T., Sato H., Sato H., Hori M., Nakamura Y., et al. Functional SNPs in the lymphotoxin-alpha gene that are associated with susceptibility to myocardial infarction. Nat. Genet. (2002) 32:650–654.[CrossRef][Web of Science][Medline]

  55. Li Y., Nowotny P., Holmans P., Smemo S., Kauwe J.S., Hinrichs A.L., Tacey K., Doil L., van Luchene R., Garcia V., et al. Association of late-onset Alzheimer's disease with genetic variation in multiple members of the GAPD gene family. Proc. Natl. Acad. Sci. USA (2004) 101:15688–15693.[Abstract/Free Full Text]

  56. Grupe A., Li Y., Rowland C., Nowotny P., Hinrichs A.L., Smemo S., Kauwe J.S., Maxwell T.J., Cherny S., Doil L., et al. A scan of chromosome 10 identifies a novel locus showing strong association with late-onset Alzheimer disease. Am. J. Hum. Genet. (2006) 78:78–88.[CrossRef][Web of Science][Medline]

  57. Li Y., Grupe A., Rowland C., Nowotny P., Kauwe J.S., Smemo S., Hinrichs A., Tacey K., Toombs T.A., Kwok S., et al. DAPK1 variants are associated with Alzheimer's disease and allele-specific expression. Hum. Mol. Genet. (2006) 15:2560–2568.[Abstract/Free Full Text]

  58. Sato Y., Suganami H., Hamada C., Yoshimura I., Yoshida T., Yoshimura K. Designing a multistage, SNP-based, genome screen for common diseases. J. Hum. Genet. (2004) 49:669–676.[CrossRef][Web of Science][Medline]

  59. Suzuki N., Cheung T.T., Cai X.D., Odaka A., Otovos L., Eckman C., Golde T.E., Younkin S.G. An increased percentage of long amyloid ß protein secreted by familial amyloid ß protein precursor (ßAPP717) mutants. Science (1994) 264:1336–1340.[Abstract/Free Full Text]

  60. Asami-Odaka A., Ishibashi Y., Kikuchi T., Kitada C., Suzuki N. Long amyloid ß-protein secreted from wild-type human neuroblastoma IMR-32 cells. Biochemistry (1995) 32:10272–10278.

  61. Matsubara E., Ghiso J., Frangione B., Amari M., Tomidokoro Y., Ikeda Y., Harigaya Y., Okamoto K., Shoji M. Lipoprotein-free amyloidogenic peptides in plasma are elevated in patients with sporadic Alzheimer's disease and Down's syndrome. Ann. Neurol. (1999) 45:537–541.[CrossRef][Web of Science][Medline]

  62. Guo S.W., Thompson E.A. Performing the exact test of Hardy–Weinberg proportion for multiple alleles. Biometrics (1992) 48:361–372.[CrossRef][Web of Science][Medline]

  63. Mantel N., Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J. Natl. Cancer Inst. (1959) 22:719–748.[Web of Science][Medline]

  64. Excoffier L., Slatkin M. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol. Biol. Evol. (1995) 12:921–927.[Abstract]

  65. Fallin D., Cohen A., Essioux L., Chumakov I., Blumenfeld M., Cohen D., Schork N.J. Genetic analysis of case/control data using estimated haplotype frequencies: application to APOE locus variation and Alzheimer's disease. Genome Res. (2001) 11:143–151.[Abstract/Free Full Text]

  66. Gabriel S.B., Schaffner S.F., Nguyen H., Moore J.M., Roy J., Blumenstiel B., Higgins J., DeFelice M., Lochner A., Faggart M., et al. The structure of haplotype blocks in the human genome. Science (2002) 296:2225–2229.[Abstract/Free Full Text]

  67. Barrett J.C., Fry B., Maller J., Daly M.J. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics (2005) 21:263–265.[Abstract/Free Full Text]

  68. Akazawa K., Nakamura T., Palesch Y. Power of logrank test and Cox regression model in clinical trials with heterogeneous samples. Stat. Med. (1997) 16:583–597.[CrossRef][Web of Science][Medline]


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


This article has been cited by other articles:


Home page
 IOVSHome page
N. Kondo, S. Honda, K. Ishibashi, Y. Tsukahara, and A. Negi
Elastin Gene Polymorphisms in Neovascular Age-Related Macular Degeneration and Polypoidal Choroidal Vasculopathy
Invest. Ophthalmol. Vis. Sci., March 1, 2008; 49(3): 1101 - 1105.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
16/23/2854    most recent
ddm244v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Miyashita, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miyashita, A.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?