Background

Vitamin D deficiency as a common health problem is a global problem, thought to be related to lack of sunlight exposure, and usually accompanied by reduced dietary intake [1]. The Vitamin-D receptor (VDR) was studied as a genetic factor of spine pathologies and plays a part in normal bone mineralization and remodeling. It is an endocrine member belongs to the nuclear receptor superfamily for steroid hormones. Its gene polymorphisms are thought to contribute to osteoarthritis, osteoporosis and degenerative disc disease. Also researchers found that VDR regulates vitamin D levels and calcium metabolism in the body and these are known to be associated with endocrine dysfunctions, insulin resistance [2, 3]. Vitamin D has been reported to influence glucose regulation via effects on insulin secretion and action [4]. Evidence is accumulating to suggest that altered vitamin D and Ca homoeostasis may play a role in the development of metabolic disturbances in insulin resistance related diseases [5,6,7]. More and more studies found that the vitamin D was useful for insulin resistance diseases [8,9,10].

T2DM, MetS, and IFG are common metabolic disorders which are observed with increasing prevalence, and which are caused by a complex interplay between genetic and environmental factors, and these metabolic disorders are all characterized by insulin resistance [11,12,13]. PCOS is by far the most common cause of anovulatory infertility and has been reported to be associated with insulin resistance (IR), hyperinsulinemia, dyslipidemia, and central obesity, which are all risk factors for the MetS, T2DM, and cardiovascular disease. Several studies have assessed vitamin D receptor gene polymorphism in relationship with these diseases; however, results remain inconsistent.

Vitamin D condition depends mainly on the sunlight and skin. It is both an environmental and biological determinant of health. Skin pigmentation may predispose subpopulations to vitamin D deficiency [14]. Some studies demonstrate that vitamin D deficiency is much higher in dark-pigmented population and Asians due to a reduced ability to produce vitamin D in their skin [15, 16]. Wondering whether there was any correlation or diverseness among these different population and their living latitude, in this research we also performed sub-group studies by skin pigmentation and latitude. Our study was conducted to elucidate whether VDR Gene polymorphisms could predict insulin resistance on a large scale.

Methods

Search strategy and selection criteria

Two investigators (Fei-fei Han, Ya-li Lv) independently searched PubMed and Embase (from 1980 until December 16th, 2016) database using the terms ((Gene polymorphism or gene variation)) AND (((((((diabetes mellitus) OR Diabetes) OR insulin resistance) OR metabolic syndrome) OR polycystic ovarian syndrome)) AND (vitamin D receptor OR VDR)).

Furthermore, we reviewed citations in the retrieved articles to search for additional relevant studies. Articles included in meta-analysis were in English or Chinese, with human subjects, published in primary literature and with no obvious overlap of subjects with other studies. The retrieved literatures were then read in their entirety to assess their appropriateness for the inclusion in this meta-analysis. Conference abstracts, case reports, editorials, review articles, and letters were excluded. We defined strict criteria for inclusion of studies. Studies were included if the exposure of interest was the VDR genotype.

Data extraction

Two independent authors extracted data and reached a consensus on the author, year of publication, ethnicity, number of patients and controls and disease types.

Statistical analysis

All statistical analyses were performed using Review Manager (Review Manager 5.0 software) and Stata/MP 11.0. Cochran’s w2 test and the inconsistency index (I2) were used to evaluate heterogeneity across the included studies. Random-effects model was applied in all the analysis. OR and their corresponding 95% confidence intervals (CI) were estimated. Z-test was performed to determine the statistical significance of pooled OR, and was considered significant when P < 0.05. We assessed potential publication bias by using a funnel plot and Egger’s test. Sensitivity analysis was performed by sequential removal (statistics of study remove) of individual studies (we did not show these results) [17].

Results

Eligible studies for meta-analysis

This study is focusing on VDR ApaI rs7975232 (G > T) variant, BsmI rs1544410 (A > G) variant, Taq1rs731236 (T > C) variant and FokIrs2228570 (C > T) variant and Insulin resistance related diseases susceptibility including (T2DM, MetS and PCOS). Characteristics of studies investigating the association of the variants with Insulin resistance related diseases susceptibility are presented in Table 1. The research of the VDR variant identified 54 articles. However, 26 studies were excluded for no case–control or no data. Finally, 28 studies were included in the current meta-analysis (Fig. 1).

Table 1 Characteristics of studies on VDR ApaI rs7975232 (G > T) variant and Insulin resistance related diseases susceptibility
Full size table
Fig. 1
figure 1

Flow diagram for study selection in meta-analysis

Full size image

Of these, 14 case–control studies examined the association of the ApaI rs7975232 (G > T) variant [3, 18,19,20,21,22,23,24,25,26,27,28,29,30] (Table 1), 22 studies in 20 case–control papers examined the association of the BsmI rs1544410 (A > G) variant [18, 22, 23, 27,28,29,30,31,32,33,34,35,36,37,38,39] (Table 2), 19 studies in 18 case–control studies examined the association of the Taq1rs731236 (T > C) variant[3, 18,19,20,21,22,23,24,25,26,27,28, 32, 33, 35, 38,39,40] (Table 3) and 18 studies in 16 case–control studies in15 papers examined the association of FokIrs2228570 (C > T)variant [3, 18, 23,24,25, 27, 30,31,32, 36, 41,42,43,44,45] (Table 4) with Insulin resistance related diseases susceptibility.

Table 2 Characteristics of studies on VDR BsmI rs1544410 (A > G) variant and Insulin resistance related diseases susceptibility
Full size table
Table 3 Characteristics of studies on VDR Taq1rs731236 (T/C) variant and Insulin resistance related diseases susceptibility
Full size table
Table 4 Characteristics of studies on VDR FokIrs2228570 (C > T) variant and Insulin resistance related diseases susceptibility
Full size table

Association between VDR ApaI rs7975232 (G > T) variant and insulin resistance related diseases susceptibility

Fourteen studies (3212 cases and 3360 controls) examining the association between the VDR ApaI rs7975232 (G > T) variant and Insulin resistance related diseases susceptibility were included. Sub-group analysis (nine studies about T2DM and five studies about PCOS) was performed. All the original data were combined by means of the Random effect model. We found no association of the VDR ApaI rs7975232 (G > T) variant with Insulin resistance related diseases (OR, 1.08; 95% CI, 0.91–1.28; P = 0.37) in the recessive genetic model (G/G vs.G/T or T/T), dominant genetic model in the (G/G or G/T vs.T/T) (OR, 1.04; 95% CI, 0.89–1.21; P = 0.62) and G allele vs. T allele analysis (OR, 1.04; 95% CI, 0.95–1.1; P = 0.36). sub-group analysis indicated that there was no association between VDR ApaI rs7975232 (G > T)variant and T2DM, PCOS patients (Table 5). sub-group analysis by skin pigmentation and living latitude showed that ApaI rs7975232 (G > T) variant was associated with insulin resistance related diseases in Asians (GG/GT + TT) (OR, 1.62; 95% CI, 1.03–2.53; P = 0.04) and population who lived in middle latitude district (30–60°) (GG/GT + TT) (OR, 1.22; 95% CI, 1.04–1.43; P = 0.02). No publication bias was detected by either the funnel plot or Egger’s tests (P > 0.05, each comparison).

Table 5 Summary of meta-analysis
Full size table

Association between VDR BsmI rs1544410 (A > G) variant and insulin resistance related diseases susceptibility

Twenty-two studies (4294 cases and 4157 controls) in 17 papers examining the association between the VDR BsmI rs1544410 (A > G) variant and Insulin resistance related diseases susceptibility were included. Sub-group analysis (14 studies about T2DM, four studies about PCOS and four studies about Mets) was performed. All the original data were combined by means of the Random effect model. We found no association of the VDR BsmI rs1544410 (A > G)variant with Insulin resistance related diseases (OR, 0.95; 95% CI, 0.78–1.16; P = 0.64) in the recessive genetic model (A/A vs.A/G or G/G), dominant genetic model in th e (A/A or A/G vs. G/G) (OR, 1.06; 95% CI, 0.86–1.31; P = 0.59) and A allele vs. G allele analysis (OR, 0.97; 95% CI, 0.83–1.13; P = 0.67). sub-group analysis indicated that there was no association between BsmI rs1544410 (A > G) variant and T2DM, PCOS patients. However, significant association was found in MetS sub-group analysis G allele vs. A allele analysis (OR, 1.41; 95% CI, 1.07–1.85; P = 0.01) (Table 5). sub-group analysis by skin pigmentation and living latitude showed that VDR BsmI rs1544410 (A > G) variant was associated with insulin resistance related diseases in Caucasian (dark-pigmented) (AA + GA/GG) (OR, 1.50; 95% CI, 1.16–1.93; P = 0.002), (A allele) (OR, 1.23; 95% CI, 1.07–1.42; P = 0.004). No publication bias was detected by either the funnel plot or Egger’s tests (P > 0.05, each comparison).

Association between VDR TaqI rs731236 (T/C) variant and insulin resistance related diseases susceptibility

Nineteen studies (3533 cases and 4024 controls) examining the association between the VDR Taq1rs731236 (T/C) variant and Insulin resistance related diseases susceptibility were included. Sub-group analysis (13 studies about T2DM, six studies about PCOS) was performed. All the original data were combined by means of the Random effect model. We found no association of the VDR TaqI rs731236 (T/C) variant with Insulin resistance related diseases (OR, 1.00; 95% CI, 0.82–1.21; P = 0.96) in the recessive genetic model (T/T vs.T/C or C/C), dominant genetic model in the (T/T or T/C vs. C/C) (OR, 0.88; 95% CI, 0.73–1.06; P = 0.17), T allele (OR, 0.89; 95% CI, 0.75–1.06; P = 0.18). Sub-group analysis indicated significant association between VDR Taq1rs731236 C allele and PCOS in C allele analysis (OR1.47; CI 1.03–2.09; P = 0.03) (Table 5). sub-group analysis by skin pigmentation and living latitude showed that VDR TaqI rs731236 (T/C) variant was associated with insulin resistance related diseases in Caucasian (dark-pigmented) (C allele) (OR, 1.24; 95% CI, 1.05–1.47; P = 0.01). No publication bias was detected by either the funnel plot or Egger’s tests (P > 0.05, each comparison).

Association between VDR FokI rs2228570 (C > T) variant and insulin resistance related diseases susceptibility

Eighteen studies (4851 cases and 6174 controls) from 17 papers examining the association between the VDR FokIrs2228570 (C > T) variant and Insulin resistance related diseases susceptibility were included. Sub-group analysis (nine studies about T2DM, five studies about PCOS, three studies about MetS and one study about IFG) was performed. All the original data were combined by means of the Random effect model. We found no association of the VDR FokIrs2228570 (C > T)variant with Insulin resistance related diseases (OR, 1.00; 95% CI, 0.68–1.47; P = 0.99) in the recessive genetic model (C/C vs.C/T or T/T), dominant genetic model in the ((C/C or C/T vs. T/T) (OR, 0.86; 95% CI, 0.67–1.09; P = 0.21) and C allele vs. T allele analysis (OR, 0.96; 95% CI, 0.84–1.10; P = 0.53). sub-group analysis indicated that there was no association between FokIrs2228570 (C > T) variant and T2DM, PCOS and MetS patients (Table 5). sub-group analysis by skin pigmentation and living latitude showed that there were no association between VDR TaqI rs731236 (T/C) variant and insulin resistance related diseases in ethics with different skin pigment and in different latitudes. No publication bias was detected by either the funnel plot or Egger’s tests (P > 0.05, each comparison).

Discussion

VDR, which is considered as a pleiotropic gene, is a transcription factor that mediates the action of vitamin D3 by controlling the expression of hormone sensitive genes such as Calmodulin-Dependent Kinase (CaMKs), and CaMKs stimulates VDR-Mediated transcription by phosphorylation levels of VDR [46]. Recent research found that deletion of macrophage VDR promotes insulin resistance and monocyte cholesterol transport to accelerate atherosclerosis in mice [47] which suggested that VDR dysfunction might result in insulin resistance. The association between VDR polymorphisms and insulin resistance related diseases including T2DM, PCOS and Mets has been extensively researched, but the results obtained so far are conflictive, and the role of VDR polymorphisms remains unclear. The reasons for this disparity may be small sample sizes, low statistical power, differences in ethnicities, extensive geographic variations, and interactions with other genetic or environmental factors. Therefore, in order to overcome the limitations of individual studies, we performed a meta-analysis. Meta-analysis increases statistical power and resolution by pooling the results of independent analyses. In this meta-analysis, we combined data from published case–control studies to evaluate the genetic associations of TaqI, BsmI, ApaI and FokI polymorphisms with these insulin resistance diseases.

To the best of our knowledge, this is the first meta-analysis which takes into account the interaction of individual VDR polymorphisms with in insulin resistance diseases. This meta-analysis, which included a total of 28 articles, examined the associations among four studied polymorphisms in the VDR ApaI variant, VDR BsmI variant, VDR Taq1 variant and VDR FokI variant and insulin resistance related diseases. The results indicated that VDR ApaI variant, VDR BsmI variant and VDR FokI variant were not conspicuous risk factors for insulin resistance related diseases. The result provided no evidence of the association between VDR variant and Insulin resistance related diseases. Yet the results were different when the researches were grouping by skin pigment and living latitude. Sub-group analysis suggested that the association between insulin resistance related diseases and VDR ApaI, BsmI, FokI variant was obvious in dark-pigmented Caucasian population and Asians.

However, to make conclusive estimates, many factors should be considered. In complex diseases such as T2DM, complex interactions between genetic and environmental factors have differential effects on disease susceptibility. Further characterization of VDR, in addition to traditional and related risk factors may facilitate early identification of patients at high risk for T2DM, and then elucidate new approaches for prevention and treatment. However, several limitations of the meta-analysis should be addressed. First, lack of the original data of the reviewed studies limited our further evaluation of potential interactions, because the interactions between and even different polymorphic loci of the same gene may influence the risk. Second, our results were based on unadjusted published estimates, and hence, we were unable to adjust them by possible confounders, for example Vitamin D level, and diet did not take into consider. Third, the number of articles and cases taking in this research is relatively small. In order to provide a more precise estimation on the basis of adjustment for confounders, more well-designed studies should be taking into account. Additionally, current evidence from prospective studies on the association between vitamin D gene polymorphism and risk of insulin resistance related diseases was limited by the use of vitamin D gene polymorphism or a single measurement of 25(OH)D concentrations. A single baseline measure of dietary vitamin D may not be able to take into account the within-individual variations of vitamin D levels across seasons or geographical location, as evident in sub-group analysis. Studies are, therefore, needed with geographical location and dietary vitamin D levels to adjust for its variability while quantifying the associations.

Conclusion

In summary, this meta-analysis provided evidence of the association between VDR BsmI variant and MetS and supporting that VDR BsmI variant G allele might be a susceptibility marker of MetS. TaqI variant was associated with PCOS for C allele and supporting that VDR TaqI variant C allele might be a susceptibility marker of PCOS. No significant association was found in the rest gene polymorphisms and these diseases related with insulin resistance diseases. The relationship of VDR gene polymorphism was more important with PCOS and MetS than T2DM. However, sub-group analysis showed VDR ApaI variant was associated with insulin resistance related diseases in Asians, VDR BsmI and VDR TaqI variant was associated with insulin resistance related diseases in Caucasian (dark-pigmented).The results suggested that the association between insulin resistance related diseases and VDR ApaI, BsmI, FokI variant was more obvious in dark-pigmented Caucasians and Asians but not in Caucasian with white skin.