The common variants of E-selectin gene in Graves’ disease

Article metrics


Adhesion molecules are involved in cell invasion in autoimmune thyroid disease. It was also reported that patients with untreated Graves’ disease (GD) had high serum level of soluble form of E-selectin (sE-selectin), the concentration of which correlated with the activity of the disease. The aim of the present study was to elucidate whether the common variants in E-selectin gene (SELE) were associated with the development of GD. Six tagSNPs within SELE were studied in 297 patients with GD and 208 healthy subjects in Chinese population. Our data showed that common SELE variants were associated with GD (P=0.012–0.036). Haplotype analysis of the single nucleotide polymorphisms revealed an association of a haplotype ATAACC with GD (P=0.005). Furthermore, quantitative trait analysis showed a significant association of SELE haplotype with sE-selectin levels (P=0.0438). This study therefore could provide us to a certain degree the insight that common SELE variants may be associated with susceptibility to GD in Chinese population, though the limitation of sample size and multiple test problems exists.


Graves’ disease (GD) is a common organ-specific autoimmune disease, which is, to a significant extent, determined by genetic factors.1, 2 The search for genes that predispose such disease is complicated by their polygenic nature. Until recently, only the major histocompatibility complex (MHC)3, 4 and cytotoxic T-lymphocyte antigen-4 (CTLA-4)5, 6, 7 have been consistently found associated with GD. With regard to the adhesion molecule, Kretowski A et al.8 had reported that intercellular adhesion molecule 1 (ICAM-1) gene polymorphisms were associated with GD. In our previous study, we had demonstrated the association between L-selectin gene (SELL) polymorphism and GD.9 E-selectin, similar to L-selectin, is one of the three members of the selectin family and has been shown to mediate the recruitment of circulating leukocytes by physically supporting adhesive interactions, and participating in cell signalling and rolling, which in turn leads to firm adhesion through the activation and binding of β2-integrin.10, 11, 12, 13 Furthermore, it was well documented that patients with untreated GD had high serum levels of a soluble form of E-selectin (sE-selectin), and the concentrations of this adhesion molecule correlated with the activity of the disease, probably reflecting an ongoing immune process.14, 15 Accordingly, E-selectin gene (SELE) may be a potential candidate gene contributing to the development of GD or influencing the clinical course of the disease.

Despite a plausible role of E-selectin in the development of GD, it has not, to our knowledge, been previously examined whether common variation in SELE is involved in GD aetiology. We aimed to answer this question by selecting haplotype-tagging single nucleotide polymorphisms (tagSNPs) that could predict at least 90% of the entire common variation in this gene. We therefore assessed the association between these SNPs or their haplotypes and GD risk, in a population-based study of Chinese patients.

In total, 297 patients with GD (72 men and 225 women; aged 8–81 years; mean age, 37.0±13.4 years) being treated at Ruijin Hospital in Shanghai were included in this study. GD was diagnosed on the basis of clinical manifestations, biochemical criteria of thyrotoxicosis (TSH <0.05 mIU l−1 and increased free T3 and/or free T4) and the presence of TSH receptor antibodies.16 Graves’ ophthalmopathy was diagnosed by experienced ophthalmologists and was classified using the NOSPECS classification.17 For statistical analysis, patients with classes 2–6 were considered as having Graves’ ophthalmopathy. Two hundred and eight healthy Chinese (45 men and 163 women; aged 20–71 years; mean age 37.5±8.9 years) without family history of GD or other autoimmune diseases served as the control group. Informed consent was obtained from the GD patients and controls, and the study was approved by the Ethics Committee of the Ruijin Hospital, School of Medicine, Shanghai JiaoTong University.

For the selection of SNPs, Haploview software ( was used to conduct linkage disequilibrium and haplotype block analyses, using the Hapmap phase II genotype data for chromosomal region 1: 166 420 439–166 437 836 (CHB database, Hapmap release 21a (January 2007)). The amplicon of interest is a 17.4-kb region, with SELE and approximately 3 kb upstream and 3 kb downstream of the gene in it. The selection of tagSNPs was performed by running the tagger program implemented in Haploview.18 The criteria for r2 was set at >0.8. This meant that any marker that was not eventually chosen as a tagging marker was already strongly correlated with at least one of the tagging markers with r2>0.8. One SNP (rs5368) was selected as tagSNP because it was located in an exon and caused an amino-acid substitution. Another five tagSNPs were chosen from HapMap SNPs with minor allele frequencies >10% (rs10800469, rs3917406, rs3917412, rs2179172 and rs3917419). Using the six SNPs (r2>0.8), we were able to capture 91% (21/23 SNPs) of the Hapmap phase II CHB common variation in SELE. All six selected tagSNPs were genotyped in 505 GD cases and control subjects by PCR-RFLP (primers and conditions are available on request).

Results and discussion

SELE consists of 14 exons that span approximately 13 kb of genomic DNA. All SNPs were in Hardy–Weinberg equilibrium (Table 1).

Table 1 SELE variations and associations with GD

Genotype data for the six SELE SNPs successfully typed in the patients with GD and controls were examined by single variant analysis (Table 1). The result showed an association of the missense SNP (rs5368) with GD in Chinese patients (allelic P=0.025, odds ratio, 1.37 (95% CI: 1.04–1.81)). The most strongly associated SNP was rs3917412 (allelic P=0.014, odds ratio, 1.43 (95% CI: 1.07–1.91)). The six SNPs were further submitted to haplotype analysis (Table 2). Rare haplotypes (frequencies <0.05) were removed. The most common haplotype (GCGACT) served as reference in our analysis. The frequency of haplotype ATAACC was significantly higher in patients than in controls (0.253 vs 0.191, P=0.005, odds ratio, 1.81 (95% CI: 1.20–2.72)). The frequency of haplotype GCGCCC was more prevalent in GD patients compared with controls, but it did not reach significant difference (0.175 vs 0.148, P=0.067, odds ratio,1.49 (95%CI: 0.97–2.29)). With regard to the quantitative trait, Table 3 shows the mean effect of SELE haplotype on serum sE-selectin level (log-transformed) with reference to the most common haplotype. With statistical analysis by likelihood ratio test for global haplotype effect, a significant association of SELE haplotype with circulating sE-selectin level was observed (P=0.0438) (Table 3).

Table 2 Haplotypes of SELE and associations with GD
Table 3 Effects of the main haplotypes of SELE on serum sE-selectin levelsa by comparison with the most frequent haplotype (N=204)

In addition, available clinical phenotypes of patients with GD were analysed for possible association with the different alleles or genotypes of these SELE polymorphisms. However, no correlation was obtained between genotype at any SNP and clinical phenotype, including the severity of ophthalmopathy indicated by the NOSPECS rating (0–1 vs 2 or greater) and age onset of GD age at biochemical diagnosis of thyroid dysfunction (<40 vs greater than 40).

This study represents the first attempt to investigate the association between the polymorphisms of SELE and the susceptibility to GD. The result revealed that variations of SELE, especially the haplotype ATAACC, were associated with susceptibility to GD.

On the basis of the current study, it is not possible to ascertain the functional genetic variant(s) within this haplotype block. It may be one of the SNPs used to identify the SELE haplotype, or it may be another genetic variant in linkage disequilibrium with the haplotype. The current study examined an a priori hypothesis that haplotypes within the SELE were associated with GD. This was based on the previous studies demonstrating that serum sE-selectin was elevated in patients with GD.14, 15 Taking the results of the current study within the context of these previous reports, it is most likely that the functional variant(s) responsible for this association is within SELE.

The promoter region SNP rs10800469 initially seemed to be well suited for a plausible functional role. The possible mechanism is that rs10800469 is located in the promoter region of the SELE, and it would influence the transcription rate of SELE, which in turn affected serum E-selectin protein level. It is equally likely that it is merely in close LD with one or more SNPs that affect SELE expression or protein function. No polymorphism in the promoter of the SELE, however, has been reported to influence gene expression, and we also could not identify a promoter SNP that was significantly associated with E-selectin level (unpublished data). Nevertheless, this does not preclude subtle effects in the regulation of SELE expression but were not detected in our study.

The coding SNP rs5368, which causes an amino-acid substitution from Histidine to Tyrosine at position 468 (H468Y), exhibited an association with GD in our study and was the only nonsynonymous SNP. This SNP would most likely affect some aspect of the post-translational function of the E-selectin protein. However, Takei et al.'s20 functional experiment did not show the evidence of H468Y having a role in the modulation of E-selectin activity. E-selectin was shown to support the rolling of leukocytes on activated endothelial cells, and much attention has been directed to the function of the extracellular portions of these molecules. Though the S128R (rs5361) polymorphism found in its epidermal growth factor-like domain was reported to alter leukocyte–endothelial interactions,21, 22 the substitution of the 468th amino acid in short consensus repeats did not seem to be critical for its leukocyte adhesion capacity. PolyPhen ( was used to investigate the possible impact of this nonsynonymous base-pair change on the structure and function. The nonsynonymous change of rs5368, however, was predicted to be benign. Thus, the pathological mechanism that underlies the predisposition to the development of GD by variants of the H468Y polymorphism remains unknown.

In haplotype analysis, significant associations of SELE haplotypes were observed with both circulating sE-selectin level and the susceptibility to GD. Because haplotype ATAACC is the only one carrying the A allele of rs3917412, it points towards the potential functional role of the polymorphism. This SNP is located at intron χ in a non-expressed region of the gene. Although the probability of exerting biological effects on gene expression is higher for polymorphisms located in exonic region, it is possible that polymorphisms located in non-expressed region could interfere in gene expression, for example polymorphism located in enhancer region consisting of coded or uncoded sequences that physically act on the regulation of gene expression. Alternatively, this haplotype might be only a marker for a functional mutation located in unexplored regions of the gene.

We also investigated the effects of the main haplotypes of SELE on serum sE-selectin level by comparison with the most frequent haplotype in 204 control subjects. A significant association of SELE haplotype with circulating sE-selectin levels was observed (P=0.0438). The result therefore further suggests that the SELE haplotype correlates with the serum sE-selectin level, and confers susceptibility to GD.

The selectins represent a family of three vascular cell adhesion molecules that appear to modulate the migration of leukocytes from blood into extravascular tissue, and share 60–70% identity between the amino-acid sequences of their lectin domains.23, 24 Recently, we identified the association of GD and specific SNPs within SELL by means of a case–control study.9 In addition, the common variants in selectin gene cluster on chromosome 1q24–25, including SELE and SELL, are in nearly complete linkage disequilibrium (data from The two selectin genes might therefore contribute to the susceptibility to GD through either a mutual interaction or a common pathway, though we did not observe any direct evidence in this study. Moreover, SELE and SELL within the selectin gene cluster seem to be prone to contribute susceptibility to a single autoimmune disease together, as shown in the findings in IgA nephropathy25 and coeliac disease.26 In the light of the evidence in the association of SELL with GD,9 the association of SELE with GD could be a reasonable deduction.

One limitation of the present study that needed to be addressed is the multiple comparison problem resulting from the number of SNPs genotyped. We performed the permutation for the multiple test correction with the Haploview software. Indeed, almost all of the single-marker associations became insignificant, but a marginal P-value of 0.065 was observed for the SNP rs3917412 after controlling for multiple comparisons by permutation test. Bonferroni correction was used for the haplotype association test. The strongest haplotype association was still significant after a conservative Bonferroni adjustment for six multiple comparisons (six SNPs). On the other hand, although the tagSNPs were used for this study, some markers were in certain linkage disequilibrium and therefore not independent of each other (that is D′=1 between rs2179172 and 3917419), which may overcorrect by Bonferroni method, resulting in a reduction in power.27

Therefore, this study could provide us the suggestive insight that common SELE variants may be associated with susceptibility to GD in Chinese population, and further genetic investigation in SELE may cast light on the aetiology of this autoimmune disease.


  1. 1

    Gough SC . The genetics of Graves’ disease. Endocrinol Metab Clin North Am 2000; 29: 255–266.

  2. 2

    Simmonds MJ, Gough SC . Unravelling the genetic complexity of autoimmune thyroid disease: HLA, CTLA-4 and beyond. Clin Exp Immunol 2004; 136: 1–10.

  3. 3

    Schleusener H, Schernthaner G, Mayr WR, Kotulla P, Bogner U, Finke R et al. HLA-DR3 and HLA-DR5 associated thyrotoxicosis—two different types of toxic diffuse goiter. J Clin Endocrinol Metab 1983; 56: 781–785.

  4. 4

    Weetman AP, Zhang L, Webb S, Shine B . Analysis of HLA-DQB and HLA-DPB alleles in GD by oligonucleotide probing of enzymatically amplified DNA. Clin Endocrinol (Oxf) 1990; 33: 65–71.

  5. 5

    Vaidya B, Imrie H, Perros P, Dickinson J, McCarthy MI, Kendall-Taylor P et al. Cytotoxic T lymphocyte antigen-4 (CTLA-4) gene polymorphism confers susceptibility to thyroid associated orbitopathy. Lancet 1999; 354: 743–744.

  6. 6

    Kouki T, Sawai Y, Gardine CA, Fisfalen ME, Alegre ML, DeGroot LJ . CTLA-4 gene polymorphism at position 49 in exon1 reduces the inhibitory function of CTLA-4 and contributes to the pathogenesis of Graves’ disease. J Immunol 2000; 165: 6606–6611.

  7. 7

    Bednarczuk T, Hiromatsu Y, Fukutani T, Jazdzewski K, Miskiewicz P, Osikowska M et al. Association of cytotoxic T-lymphocyte-associated antigen-4 (CTLA-4) gene polymorphism and non-genetic factors with Graves’ ophthalmopathy in European and Japanese populations. Eur J Endocrinol 2003; 148: 13–18.

  8. 8

    Kretowski A, Wawrusiewicz N, Mironczuk K, Mysliwiec J, Kretowska M, Kinalska I . Intercellular adhesion molecule 1 gene polymorphisms in Graves’ disease. J Clin Endocrinol Metab 2003; 88: 4945–4949.

  9. 9

    Chen HY, Cui B, Wang S, Zhao ZF, Sun H, Zhao YJ et al. L-selectin gene polymorphisms in Graves’ disease. Clin Endocrinol (Oxf) 2007; 67: 145–151.

  10. 10

    Roldan V, Marin F, Lip GY, Blann AD . Soluble E-selectin in cardiovascular disease and its risk factors. A review of the literature. Thromb Haemost 2003; 90: 1007–1020.

  11. 11

    Abbassi O, Kishimoto TK, McIntire LV, Anderson DC, Smith CW . E-selectin supports neutrophil rolling in vitro under conditions of flow. J Clin Invest 1993; 92: 2719–2730.

  12. 12

    Lawrence MB, Springer TA . Neutrophils roll on E-selectin. J Immunol 1993; 151: 6338–6346.

  13. 13

    Simon SI, Hu Y, Vestweber D, Smith CW . Neutrophil tethering on E-selectin activates beta 2 integrin binding to ICAM-1 through a mitogen-activated protein kinase signal transduction pathway. J Immunol 2000; 164: 4348–4358.

  14. 14

    Hara H, Sugita E, Sato R, Ban Y . Plasma selectin levels in patients with Graves’ disease. Endocr J 1996; 43: 709–713.

  15. 15

    Wenisch C, Myskiw D, Gessl A, Graninger W . Circulating selectins, intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 in hyperthyroidism. J Clin Endocrinol Metab 1995; 80: 2122–2126.

  16. 16

    Meller J, Jauho A, Hufner M, Gratz S, Becker W . Disseminated thyroid autonomy or Graves’ disease reevaluation by second generation TSH receptor antibody assay. Thyroid 2000; 10: 1073–1079.

  17. 17

    Werner SC . Modification of the classification of the eye changes of Graves’ disease: recommendation of the ad hoc committee of the American Thyroid Association. J Clin Endocrinol Metab 1977; 44: 203–204.

  18. 18

    Barrett JC, Fry B, Maller J, Daly MJ . Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 21: 263–265.

  19. 19

    Solé X, Guinó E, Valls J, Iniesta R, Moreno V . SNPStats: a web tool for the analysis of association studies. Bioinformatics 2006; 22: 1928–1929.

  20. 20

    Takei T, Hiraoka M, Nitta K, Uchida K, Deushi M, Yu T et al. Functional impact of IgA nephropathy-associated selectin gene haplotype on leukocyte–endothelial interaction. Immunogenetics 2006; 58: 355–361.

  21. 21

    Yoshida M, Takano Y, Sasaoka T, Izumi T, Kimura A . E-selectin polymorphism associated with myocardial infarction causes enhanced leukocyte–endothelial interactions under flow conditions. Arterioscler Thromb Vasc Biol 2003; 23: 783–788.

  22. 22

    Rao RM, Clarke JL, Ortlepp S, Robinson MK, Landis RC, Haskard DO . The S128R polymorphism of E-selectin mediates neuraminidase-resistant tethering of myeloid cells under shear flow. Eur J Immunol 2002; 32: 251–260.

  23. 23

    Bevilacqua MP, Nelson RM . Selectins. J Clin Invest 1993; 91: 379–387.

  24. 24

    Bevilacqua M, Butcher E, Furie B, Furie B, Gallatin M, Gimbrone M et al. Selectins: a family of adhesion receptors. Cell 1991; 67: 233.

  25. 25

    Takei T, Iida A, Nitta K, Tanaka T, Ohnishi Y, Yamada R et al. Association between single-nucleotide polymorphisms in selectin genes and immunoglobulin A nephropathy. Am J Hum Genet 2002; 70: 781–786.

  26. 26

    Kaur G, Rapthap CC, Kumar S, Bhatnagar S, Bhan MK, Mehra NK . Polymorphism in L-selectin, E-selectin and ICAM-1 genes in Asian Indian pediatric patients with celiac disease. Hum Immunol 2006; 67: 634–638.

  27. 27

    Nyholt DR . A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet 2004; 74: 765–769.

Download references


The present study would not have been possible without the participation of the patients and healthy volunteers. The study is supported by the grants from the E-Institute of Shanghai Universities (no. E03007) and Shanghai Leading Academic Discipline Project (Project no. Y0204).

Author information

Correspondence to G Ning.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Chen, H., Cui, B., Wang, S. et al. The common variants of E-selectin gene in Graves’ disease. Genes Immun 9, 182–186 (2008) doi:10.1038/sj.gene.6364452

Download citation


  • E-selectin
  • genetic association
  • polymorphism
  • haplotype
  • Graves’ disease

Further reading