| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Experimental Therapeutics (H.H., S.I., A.S., T.A.), Translational Research Center, Kyoto University Hospital, Kyoto University School of Medicine, Kyoto 606-8507, Japan; Department of Clinical Laboratory Science (H.H., Y.I.), School of Allied Health Science, Osaka University, Osaka 565-0871, Japan; Department of Pathology (S.S.), International Medical Center of Japan, Tokyo 162-8655, Japan; and Department of Biochemistry (D.W.B.), Wake Forest University School of Medicine, Winston-Salem, North Carolina 27157
Address all correspondence and requests for reprints to: Takashi Akamizu, Translational Research Center, Kyoto University Hospital, 54 Shogoin-Kawaharacho, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: akamizu{at}kuhp.kyoto-u.ac.jp.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
TSHR is a major thyroid-specific autoantigen, and autoantibodies against TSHR cause GD and contribute to primary myxedema. Anti-TSHR antibody is often, although usually weakly, expressed in HT. The human TSHR gene is located on chromosome 14q31 (5) and is split into 10 exons (6). Point mutations have been identified in the coding region of the TSHR gene in patients with GD (7), and these mutations may affect autoimmune responses to TSHR. A mutation in the first position of codon 52 (C52
A52), resulting in the substitution of threonine for proline in the extracellular domain of the TSHR protein, has been shown to be significantly associated with AITD in Caucasian females (8). However, others could not confirm this (9). The D727E single-nucleotide polymorphism (SNP) was associated with GD in a Russian (10) but not a Caucasian (11) or Singapore population (12). Ho et al. (12) reported that a SNP in intron 1 of the TSHR gene was associated with GD in the same study, and Chistiakov et al. (13) demonstrated that a haplotype consisting of the TSHR D727E SNP and the SNP of DIO2 gene encoding type II iodothyronine deiodinase showed significant preferential transmission from parents to probands. Previously we investigated the genetic contribution of the TSHR gene to AITD susceptibility in Japanese using two microsatellite markers located near or in intron 2 of the TSHR gene (14, 15). We found significant associations of these microsatellite markers with AITD. An obvious next step is to evaluate this evidence for association in greater detail using SNPs spanning the TSHR gene in an effort to more clearly define the exact location of AITD susceptibility variants. In the present study, we first performed a case-control study of AITD using 12 SNPs from the Japanese SNP (JSNP) and the Single Nucleotide Polymorphism Database (dbSNP) databases spaced approximately 1050 kb apart and spanning the TSHR gene. Then we found significant associations in SNPs 16 and 18 with GD, and furthermore, using an additional 10 SNPs, we analyzed a region in which significant associations with GD were observed. In addition, we examined the linkage disequilibrium structure of the region to apply haplotype case-control analysis to the data.
| Subjects and Methods |
|---|
|
|
|---|
Four hundred one unrelated Japanese individuals with AITD (female to male ratio, 3.5) were evaluated. In cases in which more than one individual from a family was collected, data from only a single individual from the family were included in the genetic analysis. Patients were classed as having either GD (n = 250; female to male ratio, 2.7) or HT (n = 151, female to male ratio, 6.1). These patients were diagnosed and assayed for antibodies as described previously (2, 14, 15). As an unaffected control population, we also collected an expanded set of controls, consisting of 238 (female to male ratio, 1.8) unrelated Japanese individuals. They are healthy and have no antibodies against thyroid-specific antigens including thyroid peroxidase, thyroglobulin, and TSHR. Informed consent was obtained from all the subjects used in this study, and the present study was approved by the Ethics Committee of Kyoto University Graduate School of Medicine.
SNP typing
Twenty-two intronic SNPs in the TSHR gene were chosen from a database of common gene variation in the Japanese population (JSNP) (16) and dbSNP (Fig. 1
and Table 1
). D727E has been previously reported (7, 17). These 22 SNPs were typed by SNaPshot (Applied Biosystems, Foster City, CA) and direct sequencing analyses. All regions that included SNPs were amplified by PCR using the GeneAmp System 9600 (Applied Biosystems) with standard protocols. After amplification 1 µl of PCR product (mixture of two to five SNPs) was incubated at 37 C for 60 min with 2.5 U exonuclease (New England BioLabs, Beverly, MA), 2.5 U calf intestinal alkaline phosphatase (New England BioLabs) and 2 µl distilled H2O to inactivate or dissolve unincorporated primers and deoxynucleotide triphosphates. Single nucleotide primer extension reactions were carried out using a SNaPshot Multiplex ready reaction mix (Applied Biosystems). SNaPshot typing primers were designed to be annealed to the target DNA 3' to the SNP. Poly-T was added to the 5' end of typing primers longer than 25 mer. Before loading onto the genetic analyzer, 1 µl of products was mixed with 10 µl Hi-Di Formamide (Applied Biosystems) and 0.2 µl GeneScan-120LIZ size standard (Applied Biosystems), and heated to 95 C for 3 min and placed on ice. The products were electrophoresed with ABI PRISM genetic analyzer (Applied Biosystems), analyzed by GeneScan software (version 3.7; Applied Biosystems) and SNPs typed by Genotyper software (version 3.7; Applied Biosystems).
|
|
Statistical analysis
Case-control analysis and Hardy-Weinberg equilibrium (HWE) test of SNP were performed using SNPALyze software (version 3.1; Dynacom, Mobara-shi, Japan). HWE tests were carried out for all loci among subjects and controls separately. Tests in subjects and controls did not show any significant deviation from HWE for any of the SNPs.
Linkage disequilibrium (LD) between SNPs was evaluated by the D' and r2 of pair-wise LD using SNPALyze software (version 3.1; Dynacom). Pair-wise D' values are computed as: D' = D/Dmax, Dmax = min (f1+f+2, f+1f2+) when D is greater than 0, and Dmax = min (f1+f+1, f+2f2+) when D is less than 0, where fs are sample estimates of SNP frequencies. D = pA1B1 pA1 pB1, where pAi and pBi stand for the frequency of allele Ai and Bi (i = 1, 2 at locus A and B) and pAiBj stand for the frequency of the AiBj haplotype. D-abs is the absolute value of D and normalized to take values between 0 and 1, regardless of the allele frequencies (18). r2 is the squared correlation in allelic state between the two loci as they occur in haplotypes and is quantified as r2 = (f11f22 f12f21)2/[(f11 + f12)(f21 + f22)(f11 + f21)(f12 + f22)], where fs are sample estimates of haplotype frequencies (19).
In the haplotype analysis, the frequency was calculated by phase estimation using the expectation maximization algorithm and examined by the permutation method using SNPALyze software (version 3.1; Dynacom) (20, 21). LD block was identified with the criterion of D' is greater than 0.8 or r2 is greater than 0.5 (22, 23).
| Results |
|---|
|
|
|---|
Allele frequencies of the 22 SNPs were not significantly different between males and females. Table 2
shows frequencies of these alleles in patients and controls and the results of the case-control association analysis of alleles of these SNPs. With JST022302 and JST140345, we found significant differences between allele frequencies in subjects and controls, which were reflected in an increased frequency of the major alleles in AITD and GD (P <0.01), compared with frequencies in controls. The evidence for association appears to be driven by associations with GD. When allele frequencies were compared between GD subjects and controls, nine of the 10 SNPs in the interval JST022302 and JST140345 showed evidence of association. In contrast, there was little or no evidence of association between these SNPs and HT alone. In addition, several other SNPs located proximal to JST022302 were significantly associated with GD, reflecting an increased frequency of the major allele in controls. Similar results were obtained when allele frequencies reported in the JSNP database were used to compare instead of control frequencies generated from our control group (data not shown). Results of genotypic analysis were broadly similar to those observed in the allelic association analysis: consistent evidence of association with SNPs between JST022302 and JST140345 in GD subjects. In addition, we performed sequencing of exons 7 and 8 on DNA from AITD (n = 339) and normal (n = 181) subjects in an effort to identify additional SNPs, but we did not find any new SNPs in these exons (data not shown).
|
Table 3
shows pair-wise LD values between SNPs for both subjects and controls expressed as D'. When evaluated by D', we identified evidence for two extended LD blocks: JST002606-JST140349, which encompasses intron 5 through intron 8 (approximately 16.2 kb), and JST140345-JST022306, which covers intron 8 through 3'-untranslated region (approximately 16.5 kb). When evaluated by r2, however, the extended block JST002606-JST140349 is divided into three restricted blocks: JST002606-JST002609-JST056274 (block 1), JST022300-rs (reference SNP ID) 11845715 (block 2), and JST022302-JST022303-JST022304-rs12885526-JST140351-TT/CA-JST140350-JST140349 (block 3), ranging in size from 2.7 to 6.2 kb.
|
In r2 block 3, the haplotype distribution was calculated for the eight SNP haplotypes of these SNPs, and case-control haplotype association analysis was performed in patients and controls as shown in Table 4
. Although there were eight SNPs in this block, only two haplotypes were common. One haplotype consisted of all major alleles of the eight SNPs [AATG(CT)6(TT)AG], and the other consisted of all minor alleles [GCAA(CT)8(CA)GT]. The haplotype of all major alleles was significantly increased in AITD (P = 0.0058, permutation P = 0.01) and GD (P = 0.0029, permutation P = 0.005), and the haplotype of all minor alleles was significantly decreased in AITD (P = 0.0020, permutation P = 0.0010) and GD (P = 0.00026, permutation P < 0.0001). Finally, genotype data showed evidence of recessive effect of the associated SNPs or the associated haplotype.
|
| Discussion |
|---|
|
|
|---|
0.05 differences between subjects and controls corresponding to relative risks of 1.5 for AITD, 1.6 for GD, and 1.7 for HT based on polymorphisms with minor allele frequency of 0.35. On the contrary, JST034658 showed significant association with HT but not GD. Most SNPs that showed LD with JST022302 showed significant association with GD but not HT. It is a possibility that the lack of association with HT may be due to a minor effect of the gene in HT, compared with GD. These findings support our previous report using microsatellite markers that the TSHR gene is a susceptibility gene for AITD/GD/HT (14, 15). Allele 5 (294 bp) of (AT)n microsatellite marker in the intron 2 (15) tended to be the major haplotype in block 3, suggesting LD between the (AT)n microsatellite marker and SNPs in intron 7/8. Therefore, the previous association of the microsatellites with AITD was the reasonable starting point of this study. Particularly, the present study suggested higher association between GD and TSHR gene, which is a major autoantigen of GD. The single SNP analysis was followed by haplotype analysis. For this purpose, we calculated intermarker LD for the 22 SNPs. Based on D' and r2, the TSHR gene can be divided into several blocks. A haplotype block defined by high r2 (block3, JST022302-JST022303-JST022304-rs12885526-JST140351-TT/CA-JST140350-JST140349) contained most of the SNPs that show association with GD. This LD block covers the exon 7-exon 8 part of the gene. Only two common haplotypes were observed, and haplotype association analysis in this block showed that a single common haplotype is associated with GD susceptibility.
An important observation from this study is that changes in the coding region of TSHR do not seem to be driving the evidence of association. Noncoding SNPs are the source of evidence of association. What biological role that these noncoding SNPs may play is unclear; similar observations have been reported by other investigators (24, 25, 26, 27, 28). It is unclear whether the associated SNPs include functional variants or functional variants within the LD block remain to be discovered, but sequencing of the exons suggests that any remaining sequence polymorphisms lie in the intron. In addition, it is noteworthy that the common risk and common protective haplotypes are completely mismatching (29). An obvious next step is a comprehensive sequencing of this region and functional studies to clarify whether the TSHR gene polymorphisms in intron 7 contribute to autoantibody production or other immunological derangement in these disorders.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online March 1, 2005
Abbreviations: AITD, Autoimmune thyroid disease; GD, Graves disease; HT, Hashimotos thyroiditis; HWE, Hardy-Weinberg equilibrium; JSNP, Japanese SNP; LD, linkage disequilibrium; SNP, single-nucleotide polymorphism; TSHR, TSH receptor.
Received November 1, 2004.
Accepted February 18, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. J. Brand, J. C. Barrett, M. J. Simmonds, P. R. Newby, C. J. McCabe, C. K. Bruce, B. Kysela, J. D. Carr-Smith, T. Brix, P. J. Hunt, et al. Association of the thyroid stimulating hormone receptor gene (TSHR) with Graves' disease Hum. Mol. Genet., May 1, 2009; 18(9): 1704 - 1713. [Abstract] [Full Text] [PDF] |
||||
![]() |
H.-D. Song, J. Liang, J.-Y. Shi, S.-X. Zhao, Z. Liu, J.-J. Zhao, Y.-D. Peng, G.-Q. Gao, J. Tao, C.-M. Pan, et al. Functional SNPs in the SCGB3A2 promoter are associated with susceptibility to Graves' disease Hum. Mol. Genet., March 15, 2009; 18(6): 1156 - 1170. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Huber, F. Menconi, S. Corathers, E. M. Jacobson, and Y. Tomer Joint Genetic Susceptibility to Type 1 Diabetes and Autoimmune Thyroiditis: from Epidemiology to Mechanisms Endocr. Rev., October 1, 2008; 29(6): 697 - 725. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Sutherland, J. Davies, C. J. Owen, S. Vaikkakara, C. Walker, T. D. Cheetham, R. A. James, P. Perros, P. T. Donaldson, H. J. Cordell, et al. Genomic Polymorphism at the Interferon-Induced Helicase (IFIH1) Locus Contributes to Graves' Disease Susceptibility J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3338 - 3341. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. J. Owen, H. Kelly, J. A. Eden, M. E. Merriman, S. H. S. Pearce, and T. R. Merriman Analysis of the Fc Receptor-Like-3 (FCRL3) Locus in Caucasians with Autoimmune Disorders Suggests a Complex Pattern of Disease Association J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 1106 - 1111. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. P Peeters, W. M van der Deure, and T. J Visser Genetic variation in thyroid hormone pathway genes; polymorphisms in the TSH receptor and the iodothyronine deiodinases. Eur. J. Endocrinol., November 1, 2006; 155(5): 655 - 662. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. A. Aliesky, P. N. Pichurin, C.-R. Chen, R. W. Williams, B. Rapoport, and S. M. McLachlan Probing the Genetic Basis for Thyrotropin Receptor Antibodies and Hyperthyroidism in Immunized CXB Recombinant Inbred Mice Endocrinology, June 1, 2006; 147(6): 2789 - 2800. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |