Association of vitamin D receptor gene 3′-variants with Hashimoto's thyroiditis in the Croatian population

• Published in: International journal of immunogenetics Vol. 35; no. 2; pp. 125 - 131
• Main Authors: Štefanić, M., Papić, S., Suver, M., Glavaš-Obrovac, L., Karner, I.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01.04.2008
• Subjects: 3' Untranslated Regions - genetics; Adult; Aged; Aged, 80 and over; Allelic Imbalance - genetics; Croatia; Female; Genetic Predisposition to Disease - genetics; Genotype; Hashimoto Disease - genetics; Humans; Male; Middle Aged; Polymorphism, Restriction Fragment Length; Receptors, Calcitriol - genetics; RNA, Messenger - genetics; Croatia
• ISSN: 1744-3121; 1744-313X
• DOI: 10.1111/j.1744-313X.2008.00748.x

Summary Hashimoto's thyroiditis (HT) is the most frequent autoimmune thyroid disease with strong genetic background. Vitamin D receptor (VDR) endocrine system affects immunosuppressive, regulatory and tolerogenic decisions required for induction and maintenance of peripheral immune tolerance. With respect to the biological function of the VDR and functionally plausible gene‐expression data, we sought to test whether particular 3′‐restriction fragment length polymorphisms (RFLP) and haplotypes previously directly or indirectly associated with VDR mRNA 3′‐allelic imbalance phenotype and differences in total VDR mRNA expression are implicated in HT susceptibility. Thus, 145 Croatian HT patients and 145 age‐, sex‐ and ethnically matched euthyroid controls were genotyped for VDR rs1544410 (BsmI), rs7975232 (ApaI) and rs731236 (TaqI) polymorphisms by polymerase chain reaction‐RFLP method. Covariate‐adjusted single‐locus and haplotype–phenotype regression analyses were performed. Permutation corrections (Pc) and Akaike Information Criteria were used for model comparisons. The best‐fit [global Pc = 7.2 × 10−4]BsmI‐TaqI BT haplotype was found significantly more often in subjects without HT [12.2% vs. 3.7%; odds ratio (OR, 95% confidence intervals) = 0.28 (0.14–0.56), Pc = 8 × 10−4], whereas the bT haplotype was significantly more frequent in individuals with HT [45.7% vs. 61.8%; OR = 1.91 (1.37–2.65), Pc = 4 × 10−4]. Two extended BsmI‐ApaI‐TaqI RFLP haplotypes, the common baT [35.7 vs. 47.3%, OR = 1.63 (1.17–2.27), Pc = 0.012] and rare BaT variants [6.5 vs. 1.2%, OR = 0.17 (0.06–0.55), Pc = 1.2 × 10−3] were associated with HT, representing predisposing and protective haplotypes, respectively. In single‐RFLP association analyses, only rs1544410 polymorphism was associated with HT phenotype (allelic Pc = 0.0078) and appeared to function under the recessive model, with decreased risk of HT among the BB homozygotes [OR = 0.39 (0.21–0.7), Pc = 0.0052] when compared to the reference b+‐genotypes. These data suggest that common haplotypic variants within the VDR gene 3′‐region previously linked to VDR mRNA expression and allelic imbalance could be associated with HT in the general population, and thus, may be involved in the pathogenesis of HT.