| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Oxagen Ltd. (J.C.T., D.Z., I.M.), Abingdon, Oxon OX14 4RY, United Kingdom; Oxford Genetics Knowledge Park, Wellcome Trust Center for Human Genetics (J.C.T.), Headington, Oxford OX3 7BN, United Kingdom; Department of Medicine, Division of Medical Sciences, University of Birmingham (S.C.G., J.A.F.), Birmingham B15 2TT, United Kingdom; Department of Endocrinology, Christchurch Hospital (P.J.H.), Christchurch 8001, New Zealand; Department of Endocrinology and Metabolism, Odense University Hospital (T.H.B., L.H.), DK-5000 Odense C, Denmark; Department of Medicine, Addenbrooke’s Hospital (K.C.), Cambridge CB2 2QQ, United Kingdom; University Department of Medicine and Therapeutics, Western Infirmary (J.M.C.), Glasgow, Scotland G11 6NT, United Kingdom; Kolling Institute and Department of Endocrinology, Royal North Shore Hospital, University of Sydney (B.G.R.), Sydney, NSW 2065, Australia; Department of Endocrinology, University Hospital Academisch Medisch Centrum, University of Amsterdam (W.M.W.), 1100 DE Amsterdam, The Netherlands; Department of Endocrinology, Oxford Center for Diabetes, Endocrinology, and Metabolism, Churchill Hospital (J.A.H.W.), Oxford OX3 7LJ, United Kingdom; and University of Sheffield Clinical Sciences Center, Northern General Hospital (A.P.W.), Sheffield S5 7AU, United Kingdom
Address all correspondence and requests for reprints to: Dr. J. C. Taylor, Oxford Genetics Knowledge Park, Wellcome Trust Center for Human Genetics, Roosevelt Drive, Headington, Oxford OX3 7BN, United Kingdom. E-mail: jenny.taylor{at}well.ox.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of this study was to identify regions of genetic linkage to AITD that could potentially harbor genetic susceptibility factors for these conditions.
Design: The study design was a genome-wide screen performed on affected relative pairs with AITD.
Setting: Patients were recruited through hospital endocrinology clinics.
Participants: Some 1119 Caucasian relative pairs affected with AITD (Graves’ disease or autoimmune hypothyroidism) were recruited into the study.
Intervention: Blood samples were obtained from each participant for DNA analysis, and clinical questionnaires were completed.
Main Outcome Measure: The study aimed to identify regions of genetic linkage to AITD.
Results: Three regions of suggestive linkage were obtained on chromosomes 18p11 (maximum LOD score, 2.5), 2q36 (maximum LOD score, 2.2), and 11p15 (maximum LOD score, 2.0). No linkage to human leukocyte antigen was found.
Conclusions: The absence of significant evidence of linkage at any one locus in such a large dataset argues that genetic susceptibility to AITD reflects a number of loci, each with a modest effect. Linkage analysis may be limited in defining such loci, and large-scale association studies may prove to be more useful in identifying genetic susceptibility factors for AITD.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The importance of genetic factors in determining susceptibility to AITD has been demonstrated by animal models of AITD (6), the increased risk of thyroid disease in siblings of affected individuals (7), and the higher concordance rate in monozygotic compared with dizygotic twins (8, 9). The most recent modeling analysis of twin data suggests that up to 79% of the liability to the development of GD is attributable to genetic factors (10). The remainder is presumed to arise from environmental determinants. Concordance rates for AIH have also been estimated to be significantly higher in monozygotic compared with dizygotic twins, suggesting that genetic factors also contribute to the development of AIH (11).
Identification of the susceptibility genes in AITD has, in general, relied upon the investigation of candidate genes in population-based association studies or linkage analysis in families. Such studies have collectively provided overwhelming evidence that loci in the human leukocyte antigen (HLA) region (chromosome 6p21) are important (8, 12, 13), providing relative risks of 1.9–4.0 for GD and 1.9 for AIH (13, 14). Polymorphisms in the region of the cytotoxic T lymphocyte antigen 4 (CTLA-4) gene on chromosome 2q33 have also been associated with GD and AIH, and linkage confirmed in family studies (15, 16, 17). Primary susceptibility has now been mapped to CTLA-4 itself with an odds ratio of 1.5 [confidence interval (CI), 1.31–1.75; P = 2.72 x 10–8] for GD and 1.45 [95% CI, 1.17–1.80; P = 0.0005] for AIH (17). Most recently, the gene encoding lymphoid protein tyrosine phosphatase (PTPN22) has been identified as a general susceptibility locus for a number of autoimmune diseases, including GD (18, 19), for which an odds ratio of 1.43 [95% CI, 1.17–1.76; P = 6.24 x 10–4] has been previously reported (18). These studies suggest that the HLA region and CTLA-4 and LYP genes make only a modest contribution to the susceptibility to AITD; therefore, other loci remain to be identified.
Genome scans have been used to provide hypothesis-free information about genetic susceptibility in a number of complex diseases, including autoimmune diseases. Several genome scans, each of modest size, have been reported for AITD; however, the putative loci have not generally been replicated (20, 21, 22, 23). In an attempt to address this lack of consistency, which may relate to the size and genetic heterogeneity of the sample sets, we have conducted a genome scan on 1119 Caucasian relative pairs affected with AITD.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Families were recruited through the eight participating centers in the United Kingdom, continental Europe, and Australasia. The study received ethical approval from each institution’s local review ethics committee and additionally from the regional multicenter ethics committees where appropriate. Written informed consent was obtained from every participant. Individuals affected with either GD or AIH were accepted into the study. To minimize genetic heterogeneity, only subjects with at least a three-generation history of white Caucasian origin were recruited. Families were recruited if they had either a sibling pair or at least two affected members (including avuncular and grandparent/grandchild pairs, but excluding parent/offspring relationships) affected with AITD. Unaffected individuals, including parents, were recruited where possible to maximize the genetic information obtainable.
GD was defined by the presence of documented biochemical hyperthyroidism in combination with 1) a diffuse goiter on a scan, 2) positive autoantibodies to TSH receptor (TSHR), TG, or TPO, 3) Graves’ ophthalmopathy, or 4) confirmation of a lymphocytic infiltrate in thyroid histology. The diagnostic criteria for AIH were defined as documented biochemical hypothyroidism and 1) positive autoantibodies to TG or TPO, 2) histological confirmation of a lymphocytic infiltrate in a fine needle aspirate, or 3) presence of a goiter on clinical examination.
Serum TSH concentration and thyroid antibody (TG and TPO) status were tested for all participants at the time of entry into the study. This permitted confirmation of the autoimmune nature of the thyroid disease in a considerable number of participants whose autoantibody levels had not been measured in the course of their treatment. Any subjects recruited as apparently unaffected with AITD, but found subsequently to have positive autoantibodies and a suppressed serum TSH concentration (with raised thyroid hormone levels) or an elevated TSH level, were reclassified as being affected with either GD or AIH, respectively, and the individuals were notified accordingly.
Family studies have repeatedly demonstrated a higher prevalence of thyroid autoantibodies among first degree relatives of patients with AITD than in the general population alone (24). The presence of thyroid autoantibodies may be associated with an increased risk of developing AITD (25), suggesting that the development of autoantibodies might be an early event in the pathogenesis of AITD. To evaluate whether distinct genetic loci influence the development of autoantibodies, an additional autoantibody (AB) phenotypic trait was defined as the subset of individuals positive for TG, TPO, or TSHR autoantibodies. This cohort included both GD and AIH patients who had positive autoantibodies either at the time of diagnosis or at follow-up and, in addition, any biochemically euthyroid patients with positive autoantibodies.
Laboratory methods
Blood samples were collected from all participants for DNA and serum. DNA was isolated using a Puregene kit (Gentra Systems, Minneapolis, MN). Serum samples from all individuals were tested for the presence of TPO autoantibodies by ELISA (26), and positivity was defined as values more than 3 SD above the mean from a sample of 150 healthy blood donors. The serum TSH concentration was measured using the Bayer Diagnostics ACS 180 immunoassay analyzer (Newbury, UK).
A genome-wide screen was carried out using 389 markers spaced at 10-cM intervals from the LMS2 marker set (Applied Biosystems, Foster City, CA). Additional markers were included to replace those excluded by error-checking procedures. The markers were amplified using standard ABI protocols, pooled in sets of 15–20 markers, and products were separated by electrophoresis in polyacrylamide gels using the ABI 3700 automated sequencers. Products were sized using the GENESCAN program (Applied Biosystems), and genotypes were assigned semiautomatically using GENOTYPER.
Statistical analysis
Our power calculations used a 
s = 5 based on the most conservative estimate of this parameter (reviewed in Ref.8) and assumed that five genes accounted for the genetic effects. This would give 99% power to detect any one locus and 87% power to detect any specific locus, with a locus-specific s of 1.37 in a dataset of this size.
To minimize genotyping errors, extensive checking procedures were employed. The genetic relationships between family members were verified using a proprietary version of the software GRR (27). This program identified 13 pairs of monozygotic twins who were subsequently removed from the dataset and a number of half-sibling pairs who were originally considered to be full siblings. PedCheck (28) and Merlin (29) were used to screen for parent-child allele incompatibilities and possible genotyping errors. Tests for marker heterozygosity and double recombination were also carried out.
For the GD analysis, only family members with GD were defined as affected. Similarly, the AIH analysis used relative pairs that were purely AIH and not mixed with GD. The AITD-affected status was derived from those families with members affected with GD, AIH, or both. The AB subgroup included any individuals who had at any time tested positive for autoantibodies to TPO, TG, or TSHR regardless of their clinical status.
Merlin was used for multipoint nonparametric linkage analysis of the affected relative pairs. Affected relative pair analysis has been reported to be more powerful than affected sibling pair analysis and was therefore the method of choice for this study (30). Simulation studies were conducted in Merlin to assess the genome-wide significance of the findings. One thousand simulations were run using a dataset that comprised the majority of families used in the experimental genome scan. The simulations were therefore based on the same ascertainment criteria and mode of statistical analysis as the genome scan reported in this study.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 558 families containing 1119 relative pairs affected with AITD were collected for this study. Affected relative pairs included full siblings, half siblings, grandparent/grandchild pairs, and avuncular pairs, but excluded parent-offspring relationships. Of the 558 AITD families, 238 contained only individuals with GD, 206 were purely AIH, and the remaining 114 were mixed GD and AIH. The AB cohort numbered 349 families. A breakdown of the cohort composition by phenotype is shown in Table 1
. DNA from parents was collected wherever possible to ensure that a family was genetically informative using identity by descent analysis. Where parents were not available, unaffected siblings were collected.
|
The results of the nonparametric multipoint analysis of affected relative pairs for each trait are summarized in Fig. 1. Empirical threshold LOD values for suggestive and significant linkages were defined from simulation studies. Genotypes for the four traits (GD, AIH, AITD, and AB) were each simulated 1000 times for the majority of the families in the cohort. The mean LOD score from these simulations was used to derive a threshold value of suggestive linkage for each trait. Suggestive linkages were 1.79 for GD, 1.80 for AIH, 1.74 for AB, and 1.74 for AITD. LOD scores for linkages significant at the 5%, 1%, and 0.1% levels were also obtained. The number of linkages that would be expected purely by chance was also simulated and found to be one locus for each trait investigated (GD, AIH, AB, or AITD). The simulation data are summarized in Table 2.
|
|
, and LOD plots for these chromosomes are shown in Fig. 2. For the GD subset of families, three regions showed evidence suggestive of linkage (LOD, >1.79) when the threshold defined by the simulation data were applied. The highest maximum LOD score (MLS) obtained for any of the four phenotypes was for marker D18S53 on chromosome 18p11 in GD (MLS 2.5). Additional suggestive linkages were found on chromosomes 2q36 (MLS 2.2) and 11p15 (MLS 2.0). Nominal evidence of linkage to GD (LOD, >1.0) was obtained for another nine regions, of which two, on chromosomes 5q23 (MLS 1.6) and Xp11 (MLS 1.6), showed linkage to GD with MLS greater than 1.5. Simulation studies predicted that only one GD locus would be expected by chance, suggesting that two of the three GD loci with suggestive evidence for linkage may represent true positives.
|
|
None of the markers showed linkage to the AITD cohort as a whole. Consistent with this, no single marker showed suggestive linkage to both GD and AIH phenotypes. Although separate markers on chromosome 5 showed linkage to GD and AIH, these markers are located at 5q23 and 5p14, respectively, and are therefore likely to represent distinct susceptibility loci.
The highest MLS obtained for the AB subset of patients (those with positive autoantibodies) was 1.4 on chromosome 9p21. This was derived principally from the GD subset of families who showed nominal linkage to this region (MLS 1.34), because linkage to AIH was not observed at this locus.
No linkage on chromosome 6 in the region of the HLA locus was observed for any of the four phenotypic traits (GD, AIH, AITD, or AB) despite the presence of markers localized in the class I and II regions of the major histocompatibility complex.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We found evidence suggestive of linkage to three regions for GD (chromosomes 2q36, 11p15, and 18p11). Elevated LODs were obtained for two additional regions (on chromosomes 5q23 and Xp11) for GD and four regions for AIH (on chromosomes 1p36, 14p11, 17q21, and 19q13). None of the linkages obtained for GD or AIH was the same, and none of these showed more than nominal linkage to the combined AITD cohort.
Our strongest region of linkage, to marker D18S53 on chromosome 18p11 in the GD families, has not been previously reported for GD, although nominal evidence of linkage was observed in a study of Old Order Amish families with positive TPO or TG antibodies (20). This region contains a number of genes involved with immune system function, including another member of the PTP family (PTPN2), which is involved with T cell proliferation and IL signaling; a member of the TNF superfamily (TNFSF5IP1); and a cell death activator (CIDE-A).
Evidence for the suggestive linkage to 11p15 overlaps with the type 1 diabetes (T1D) locus IDDM2 (MIM 125852) (31). Linkage of this region has also been reported to multiple sclerosis (32) and systemic lupus erythematosus (33). Although in T1D, this linkage has been largely attributed to the variable number of tandem repeats located in the 5'-upstream region of INS (34), recent analysis in GD and multiple sclerosis patients suggests that the linkages to these diseases are likely to be the result of an alternative locus (32). The CD81 antigen, which plays a role in early T cell development, and the IGF-II precursor (IGF2) are two such candidate loci.
Nominal linkage (defined here as MLS >1.5) was found in an additional seven regions, on chromosomes 1p36, 5q23, 9p21, 14p11, 17q21, 19q13, and Xp11. Linkage to 1p36, observed in this study for AIH, has previously been reported for a number of autoimmune diseases, including systemic lupus erythematosus (33, 35), multiple sclerosis (36), ankylosing spondylitis (37), rheumatoid arthritis (38), and inflammatory bowel disease (39). In addition, a proximal region at 1p31-p32 has been linked to vitiligo and AIH (40).
Previous linkage has been reported to 5q33 in a number of autoimmune conditions, including AITD. The region Xp11 showing linkage to GD in this study is the same region previously reported to be linked to GD. A Toll-like receptor precursor (TLR7) is a candidate gene located in this region. Toll-like receptors recognize microbial pathogens and trigger innate immune responses.
Although there is convincing evidence that HLA, CTLA-4, and PTPN22 are genetic susceptibility factors for AITD, we have been unable to detect linkage to these loci despite the fact that the analysis included markers within these gene regions. This suggests that the contributions of these loci are insufficient to be detected by linkage. Consistent with this, the odds ratio for HLA has previously been estimated from several association studies to be between 1.9 and 4.0. This is in marked contrast to T1D (41) and rheumatoid arthritis (42), in which the HLA region is the major locus, conferring 40–50% of the genetic susceptibility.
Our data do not support other previously reported linkages, including those on chromosomes 8q24 (22, 23), 12q22 (23), 13q32 (23), 14q31 (43), 18q21 (44), 20q11 (45, 46), and Xq21 (23). This lack of replication may arise from the differences in size or ethnic origin between the cohorts or the statistical methods used. This study reports a nonparametric linkage analysis of 1119 Caucasian AITD-affected relative pairs. Sakai et al. (22) also used a nonparametric linkage approach, but on a cohort of 123 Japanese AITD-affected sibling pairs. In contrast Tomer et al. (23) used a parametric linkage approach on a cohort numbering a maximum of 102 multiplex families (although most linkages reported were on smaller GD or HT subsets) of diverse ethnic origin (including families from North America, Israel, the United Kingdom, and Italy). Underpowered studies and genetic heterogeneity are well recognized to be major factors confounding replication in linkage studies and are likely to be the explanation for the differences between the linkages reported in this study and those cited in previous AITD genome scans.
Although the current linkage study in AITD is considerably larger than those in any previous studies, we have still been unable to detect a major locus for AITD. This suggests that the genetic susceptibility to AITD comprises a number of loci, each with modest effects, and that alternative strategies are required for their detection.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
This work was supported by Oxagen Ltd. L.H. and T.H.B. were supported by the Agnes and Knut Moerk Foundation.
First Published Online November 8, 2005
Abbreviations: AB, Individual with positive thyroid autoantibodies; AIH, autoimmune hypothyroidism; AITD, autoimmune thyroid disease(s); CI, confidence interval; CTLA-4, cytotoxic T lymphocyte antigen 4 gene; GD, Graves’ disease; HLA, human leukocyte antigen; HT, Hashimoto’s thyroiditis; LYP, lymphoid protein tyrosine phosphatase; MLS, maximum LOD score; PTPN22, gene encoding lymphoid protein tyrosine phosphatase; T1D, type 1 diabetes; TG, thyroglobulin; TPO, thyroid peroxidase; TSHR, TSH receptor.
Received April 1, 2005.
Accepted October 27, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. S. Douglas, T. H. Brix, C. J. Hwang, L. Hegedus, and T. J. Smith Divergent Frequencies of IGF-I Receptor-Expressing Blood Lymphocytes in Monozygotic Twin Pairs Discordant for Graves' Disease: Evidence for a Phenotypic Signature Ascribable to Nongenetic Factors J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1797 - 1802. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Anderson Update in Endocrine Autoimmunity J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 3663 - 3670. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kondrashova, H. Viskari, A.-M. Haapala, T. Seiskari, P. Kulmala, J. Ilonen, M. Knip, and H. Hyoty Serological Evidence of Thyroid Autoimmunity among Schoolchildren in Two Different Socioeconomic Environments J. Clin. Endocrinol. Metab., March 1, 2008; 93(3): 729 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J Owen, J. A Eden, C. E Jennings, V. Wilson, T. D Cheetham, and S. H S Pearce Genetic association studies of the FOXP3 gene in Graves' disease and autoimmune Addison's disease in the United Kingdom population. J. Mol. Endocrinol., August 1, 2006; 37(1): 97 - 104. [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 |
