Vitamin D Receptor Gene Polymorphisms Influence T1D Susceptibility among Pakistanis

Background The vitamin D receptor (VDR) gene regulates insulin secretion from the pancreas and acts as a mediator of the immune response through vitamin D. Polymorphism in VDR causes alterations in the functioning of vitamin D, leading to type 1 diabetes (T1D) predisposition. The aim of the present study was to determine VDR gene polymorphism in association with T1D in Pakistanis. Methods The association was evaluated by selecting rs2228570 (FokΙ), rs7975232 (ApaΙ), and rs731236 (TaqΙ) polymorphic sites in 102 patients and 100 controls. Genotypes were identified by DNA sequencing and PCR-RFLP. Results The allelic and genotypic frequencies of FokΙ and ApaI were significantly associated with T1D (p < 0.001) development. At the FokΙ site, tryptophan was replaced with arginine due to polymorphism. A novel SNP (GeneBank acc number KT280406) was identified through the sequencing of intron 8, 62 bp downstream from the ApaI polymorphic site, and significantly associated with T1D development. The TaqΙ did not depict any association with T1D at the allelic or genotypic level (p > 0.05). CCGC, CCGG, CCTC, and CCTG haplotypes were significantly associated with disease development (p < 0.05). However, CTGG haplotype was protective towards T1D (p < 0.01). Conclusion VDR polymorphisms were identified as susceptible regions for T1D development in the Pakistani population.


Introduction
T1D is a polygenic disease with several protective and susceptible alleles interacting with each other [1]. Genetics plays a key role in the onset of T1D [2,3] and shows a significant clustering in a family; as in siblings, the average risk prevalence is 6% as compared to the general population with a 0.6% risk [4].
Vitamin D displays vital immunomodulatory properties that aid in preventing diabetes development in T1D animal models [5]. It activates human macrophages, antigenpresenting cell maturation, and inhibits dendritic cell differentiation as well as affects cytokine production by interacting with most immune cells [6,7]. Vitamin D reduced the MHC class I molecules and Fas expression (transmembrane cell surface receptor mediator) leading to the suppression of pancreatic β-cell apoptosis [8,9]. In addition, it increases A20 protein expression, which has an antiapoptotic function against pancreatic β-cells, and increases insulin production [9].
Vitamin D is reported to be metabolically active and activates the nuclear vitamin D receptors (VDR) to exert its genomic effect. The VDR belongs to the super family of ligand-activated transcription factors located on the 12q12-q14 chromosome and encoded by the VDR gene in humans [10]. It consists of two promoter regions, six untranslated exons (exons 1a-1f) that are spliced alternatively, and eight protein-coding exons (exons 2-9) [11]. Basically, the VDR gene contains four polymorphism sites which were identified with restriction fragment length polymorphism (RFLPs) TaqI within exon 9, ApaΙ and BsmI within intron 8, FokI within exon 2, and mononucleotide polymorphism in the 3′ untranslated region [12]. In our previous study, two SNPs: rs1544410 on VDR and rs2476601 on PTPN22 were screened and were found significantly associated with T1D in the Pakistani population [13]. The genomic action of vitamin D is initiated by its binding to the VDR, which leads to the transcription of genes regulated by it. This transcriptional regulation occurs to a very complicated mechanism [14].
The Pakistani population is heterogeneous and becoming more complex due to cast-specific marriages. This phenomenon produces narrow genetic pool and transfers the genetic mutations more often to the next generations. Therefore, the current study was designed to evaluate the association of polymorphism in the VDR gene with T1D.

Subjects.
The study was ethically approved by the Board of Advance Research, G.C. University, Lahore, Pakistan. This case-control study was carried on T1D patients recruited from the Diabetic Center of Shalamar Hospital, Lahore (public sector hospital). Written consent was obtained from patients/guardians of the studied subjects. All patients that participated were already clinically diagnosed with T1D by a physician according to WHO criteria such as hyperglycemia, insulin requirement from diagnosis, recurrent infections, high levels of glycosuria, increased urine volume and thirst, unexplained weight loss, and in severe cases coma and drowsiness [15]. A total of 102 T1D cases and 100 controls were included in the study. The clinical characteristics (gender, age, the age of diagnosis, and positive family history as well as physical activity including exercise and playing outdoor games) that were recorded after interviewing the patients who participated in the study were presented in Table 1. All control subjects were healthy and had a negative family history of T1D.
2.2. DNA Isolation and SNP Selection. Blood samples (3 ml) from each subject were collected in EDTA-coated tubes, and DNA was extracted by the modified organic extraction method [16]. The extracted DNA was stored at −20°C (Haier) for further genetic analysis. DNA quantification was carried out by the nanodrop (Thermo 2000). Three polymorphic sites FokI, ApaI, and TaqI were selected by using the HapMap database (http://hapmap.ncbi.nlm.nih.gov/) and SNP Browser software 4.0 (Applied Biosystems).
2.4. VDR Genotyping. VDR genotyping for FokI, ApaI, and TaqI was performed by DNA sequencing and restriction fragment length polymorphism (RFLP), whereas for KT280406, all samples were sequenced with forward primer. For sequencing, the reaction mixture of 10 μl was prepared to contain 3 μl of purified PCR product, 1 μl of forward primer, 1 μl of big dye mixture, 1 μl of 5X reaction buffer, and 4 μl of DEPC water. The products were amplified by following PCR conditions: initial denaturation at 95°C for 2 mins followed by repeated 35 cycles of denaturation at 95°C for 30 s, annealing at 50°C for 15 s, and extension at 60°C for 4 mins, and at end final extension at 60°C for 5 mins. By using isopropanol, amplified PCR products were precipitated. Product pellets were dissolved in 12 μl formamide and were incubated for 5 minutes at 95°C and chilled quickly. The samples were loaded to ABI PRISM genetic analyzer 3130 XL to sequence the fragment of interest.  Figure 4).

Statistical
Analysis. All data of the controls passed the Hardy-Weinberg equilibrium (p > 0 05). The chi-square test was used to determine allelic and genotypic frequencies.  Linkage disequilibrium and haplotypes were calculated to study their association with T1D by SHEsis (http:// analysis.bio-x.cn/SHEsisMain.htm). The change in amino acid sequences was determined by aligning sequences in Mega 6 software.

Results
The allelic and genotype frequencies of the VDR gene at FokI, ApaI, and TaqI polymorphic sites were assessed in 102 T1D patients and 100 controls and were presented in Table 2. In patients and controls, FokI, ApaI, TaqI, and KT280406 were validated (p > 0 05) by the Hardy-Weinberg equilibrium (HWE). In the genetic analysis, "C" was identified as a risk allele (p = 0 001) from FokI (rs2228570), while the risk genotype was CC (p = 0 0015) making the population at risk of T1D development. Allele "T" from ApaI (rs7975232) was found to be significant with T1D (p = 0 036), but at the genotype level "TT" and "TG," significant association was detected between T1D and the studied SNP (p = 0 014). No significant difference was observed for TaqI among patients and controls for allelic and genotypic frequencies (p > 0 05). A novel SNP was identified as a result of the sequencing shown in Figure 2 about 62 bp downstream from the ApaI polymorphic site and submitted to NCBI (GeneBank Acc number KT280406); the detected mutation was G into C. This novel SNP was only identified in 4.18% of patients while it was not found in the controls.
Protein alignment showed that tryptophan changes into arginine in FokΙ with a change in allele whereas no change in amino acid occurred in TaqI sites.
Haplotype analysis depicted both the protected and susceptible haplotypes for T1D. For the four genetic polymorphisms, haplotypes CCGC, CCGG, CCTC, and CCTG were significantly associated with T1D (p < 0 01). However, the frequency of CTGG and CTTG was higher in controls (0.933, 0.067, resp.) as compared to patients (0.051, 0.000, resp.); so they were protective against T1D development (p < 0 01), as shown in Table 3. A correction with eight gives only significance for the CTTG haplotype which was protective against T1D development.

Discussions
The results of the present study revealed the existence of an association between VDR-FokI, VDR-ApaI polymorphisms, and T1D in the Pakistani population. The association of the VDR gene polymorphism at four polymorphic sites (FokΙ, ApaΙ, KT280406, and TaqΙ) with T1D was examined in the present study. In the current study, ApaI does violate HWE (p < 0 05). This might be because Pakistan is a multicultural country with specific traditions which were followed for generations. Among those traditions, one of the most common traditions is intercaste marriages and absence of random mating, ultimately reducing genetic recombination. These also might be the reason for the HWE violation.
The current study reported that FokΙ polymorphism significantly (p = 0 001) increased the chances of disease development in the Pakistani population. However, the frequency of the FokΙ "C" allele was higher in cases in Hungary [18]. The findings on Dalmatian and Japanese populations were in line with the results of our study that the FokΙ restriction site was significantly associated with T1D [19,20]. In contrast with the current results, the findings on two subpopulations of Spain and Iran reported no association of FokΙ polymorphism with T1D. Moreover, the frequency of the CT allele was reported to be higher in cases as compared to controls in the Iranian Population [21,22].
No significant genotypic association was observed between ApaΙ polymorphism and T1D in the current study; however (p < 0 05), in the Taiwanese population, a significant association was reported between the ApaΙ polymorphism and the onset of T1D [23]. In line with the present observation, studies on Greece, Finland, and Norway reported that no association exists between ApaΙ polymorphism with the development of T1D [24][25][26].
The present study demonstrated that the C allele of TaqΙ was frequently present in both the case and control, and no association exists between TaqΙ polymorphism and T1D development in the Pakistani population. In contrast to our study, it was reported that the T allele was more common in 79.5% of healthy subjects as compared to diseased, and a significant association exists between TaqΙ polymorphism and T1D [27]. Genetic heterogeneity was reported to be influenced by the TaqΙ polymorphism in Romanian T1D individuals and significantly associated with T1D [20,28].
The effect of VDR gene polymorphism on its functioning has not been completely understood, and it was considered that different SNP interactions lead to the alteration in gene functioning [25]. The current study revealed that the frequency of haplotypes CTGG and CTTG of TaqI-FokI-ApaI-KT280406 was significant and protects against disease development. CCGG, CCTC, and CCTG were significantly associated (p < 0 01) and involved in disease development. In line with current observation, CCG was reported as a significantly associated haplotype in the German population and acts as a protective marker whereas CTG, CTT, and CCT were not associated with the haplotype of T1D [29].

Conclusions
Our case-control study demonstrated that VDR polymorphism in the FokI, ApaI, and KT280406 regions is susceptible to T1D development in the Pakistani population. Further studies should be carried out to determine the role of genetics in diabetes onset so that the susceptible communities can be targeted.

Conflicts of Interest
The authors declare that they have no competing interests.