Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population

XU Ze ZHANG Dong Dong LIU Ya Ping ZHANG Yu Jing XUE Yuan GAO Jiao Jiao ZHAO Tong HUANG Hao Yue LI Wen Jie LI Xing

XU Ze, ZHANG Dong Dong, LIU Ya Ping, ZHANG Yu Jing, XUE Yuan, GAO Jiao Jiao, ZHAO Tong, HUANG Hao Yue, LI Wen Jie, LI Xing. Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population[J]. Biomedical and Environmental Sciences, 2022, 35(11): 1074-1078. doi: 10.3967/bes2022.136
Citation: XU Ze, ZHANG Dong Dong, LIU Ya Ping, ZHANG Yu Jing, XUE Yuan, GAO Jiao Jiao, ZHAO Tong, HUANG Hao Yue, LI Wen Jie, LI Xing. Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population[J]. Biomedical and Environmental Sciences, 2022, 35(11): 1074-1078. doi: 10.3967/bes2022.136

doi: 10.3967/bes2022.136

Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population

Funds: The study was funded by the National Natural Science Foundation of China [Grant no. 81872626, 82003454]; Chinese Nutrition Society - Zhendong National Physical Fitness and Health Research Fund [Grant no. CNS-ZD2019066]; and Key R&D and promotion projects in Henan Province [212102310219 and 202102310120]
More Information
    Author Bio:

    XU Ze, female, born in 1995, Master, majoring in nutrition and food hygiene

    Corresponding author: LI Xing, E-mail: lixing530@zzu.edu.cn, Tel: 86-371-67781305.
  • Figure  1.  Multiplicative interactions between rs739837 (G > C) and BMI (A) or WC (B) and the associations of these interactions with FPG. The blue trend line represents the correlations between BMI or WC and FPG levels in individuals with risk alleles. The red trend line represents the correlation between BMI or WC and FPG levels in individuals without risk alleles. The point where the two lines cross indicates that for individuals whose BMI or WC is higher than the level value at this point, risk allele carriers have a greater effect on systolic or diastolic blood pressure than non-risk allele carriers. BMI, body mass index; FPG, fasting plasma glucose; WC, waist circumference.

    S1.   Basic characteristics of study participants

    CharacteristicsT2D (n = 397)Control (n = 794)χ2/ZP
    Male156 (39.3%)312 (39.3%)01.000
    Age (years)60 (52.5−67.5)60.5 (51.00−67.00)−0.020.982
    Smoking295 (25.7%)576 (27.5%)0.420.518
    Alcohol drinking50 (12.6%)98 (12.3%)0.020.901
    Physical activity15.40< 0.001
     Low193 (39.2%)311(48.7%)
     Moderate76 (17.0%)135 (19.2%)
     High127 (43.8%)347 (32.1%)
    Family history of T2D71 (17.9%)56 (7.1%)32.59< 0.001
    BMI (kg/m2)26.33 (23.98−28.70)24.80 (22.30−27.20)−6.28< 0.001
    WC (cm)91.25 (84.13−97.50)86.00 (78.00−93.50)−7.76< 0.001
    FPG (mmHg)8.46 (6.84−10.94)4.98 (4.52−5.42)−23.77< 0.001
    FINS (mIU/L)13.25 (9.55−18.97)11.04 (8.65−13.77)−6.68< 0.001
    HOMA-β54.61 (29.81−96.93)153.57 (110.26−217.01)−19.14< 0.001
    HOMA-IR5.02 (3.29−7.80)2.42 (1.80−3.23)−17.76< 0.001
    TC (mmol/L)4.61 (4.06−5.29)4.51 (3.94−5.16)−2.210.027
    TG (mmol/L)1.81 (1.19−2.80)1.36 (0.90−2.04)−7.12< 0.001
    HDL-C (mmol/L)1.13 (0.96−1.34)1.18 (0.99−1.39)−2.660.008
    LDL-C (mmol/L)2.63 (2.19−3.17)2.53 (2.02−3.09)−2.470.014
    25(OH)D3 (ng/mL)19.1 (15.02−27.21)19.36 (15.5−24.5)−0.520.601
      Note. Data are given as number (%) or median (interquartile range). Age refers to the age at which the participants were enrolled in the study. Physical activity was divided into three categories (low, moderate, and high) according to the International Physical Activity Questionnaire. BMI, body mass index; WC, waist circumference; FPG, fasting plasma glucose; FINS, fasting insulin; HOMA-β, Homeostasis Model Assessment-β [HOMA-β (%) = 20 × FINS (mIU/L) / (FPG (mmol/L) − 3.5)]; HOMA-IR, Homeostasis Model Assessment Insulin Resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L) / 22.5]; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol. P-value > alpha (0.05) means not statistically significant.
    下载: 导出CSV

    Table  1.   Genotypic distribution of VDR polymorphisms and the association of VDR polymorphisms with T2D risk

    SNPT2DControlχ2POR (95% CI)aPHWE
    rs21894804.1550.1250.76
     GG182 (46.08%)327 (42.69%)1
     GA177 (44.26%)339 (44.26%)0.90 (0.69–1.18)0.459
     AA36 (9.11%)100 (13.05%)0.60 (0.38–0.94)0.026
     AA/GA+GG0.63 (0.41–0.97)0.036
     AA+GA/GG0.84 (0.65–1.09)0.180
    rs7398373.7730.1520.77
     GG202 (50.88%)432 (56.18%)1
     GC161 (40.56%)288 (37.45%)1.26 (0.96–1.66)0.095
     CC34 (8.56%)49 (6.37%)1.42 (0.86–2.32)0.170
     CC/GC+GG1.28 (0.79–2.08)0.308
     CC+GC/GG1.32 (1.03–1.70)0.028
      Note. aAdjusted for age, smoking, alcohol drinking, exercise, level of 25(OH)D, and family history of T2D. CI, confidence interval; HWE, Hardy–Weinberg equilibrium; SNP, single nucleotide polymorphism; OR, odds ratio; T2D, type 2 diabetes; VDR, vitamin D receptor. P-value > alpha (0.05) means not significant.
    下载: 导出CSV

    S2.   Association of BMI and WC levels with the risk of T2D

    ContinuousModel 1Model 2
    BMI
     Continuous
     T1 (< 23.7 kg/m2)11
     T2 (23.7−26.8 kg/m2)1.68 (1.22−2.31) *1.35 (0.97−1.89)
     T3 (≥ 26.8 kg/m2)2.52 (1.84−3.43) *1.88 (1.35−2.63) *
     P-trend< 0.0010.003
    WC
     Continuous
     T1 (< 82.5 cm)11
     T2 (82.5−92.5 cm)2.15 (1.56−2.97) *1.87 (1.34−2.61) *
     T3 (≥ 92.5 cm)2.94 (2.13−4.04) *2.22 (1.58−3.14) *
     P-trend< 0.001< 0.001
      Note. Model 1: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D; Model 2: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D, SBP, DBP, TC, TG, HDL-C, LDL-C. BMI, body mass index; WC, waist circumference *P < 0.05. P-value > alpha (0.05) means not statistically significant.
    下载: 导出CSV

    S3.   Associations (β, 95% CI) of BMI and WC with blood glucose related parameters

    OutcomesBMIWC
    Model 1Model 2Model 1Model 2
    FPG0.076 (0.032−0.119)*0.010 (−0.035−0.055)0.040 (0.025−0.054)*0.017 (0.002−0.032)*
    FINS0.592 (0.479−0.706)*0.493 (0.367−0.619)*0.233 (0.194−0.271)*0.202 (0.159−0.245)*
    HOMA-IR0.249 (0.198−0.299)*0.186 (0.132−0.241)*0.100 (0.084−0.117)*0.080 (0.062−0.098)*
    HOMA-β3.183 (−11.166−17.531)2.837 (−13.628−19.301)0.671 (−4.246−5.587)0.690 (−4.946−6.326)
      Note. Model 1: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D; Model 2: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D, SBP, DBP, TC, TG, HDL-C, LDL-C. BMI, body mass index; WC, waist circumference; FPG, fasting plasma glucose; FINS, fasting insulin; HOMA-IR, Homeostasis Model Assessment Insulin Resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L) / 22.5]; HOMA-β, Homeostasis Model Assessment-β [HOMA-β (%) = 20 × FINS (mIU/L) / (FPG (mmol/L) − 3.5)]; *P < 0.05. P-value > alpha (0.05) means not statistically significant.
    下载: 导出CSV

    Table  2.   Interaction analysis of VDR polymorphisms and obesity on T2D

    SNPOverweight/obesityPinteractionAbdominal obesityPinteraction
    NoYesNoYes
    rs21894800.4220.626
     AA11.51 (0.67–3.41)13.99 (1.24–7.24)
     GA+GG1.14 (0.56–2.31)2.47 (1.23–4.91)1.79 (0.81–3.95)4.07 (1.88–8.83)
    rs7398370.0290.826
     GG12.86 (1.93–4.26)11.26 (0.80–1.98)
     CC+GC1.96 (1.24–3.10)3.06 (2.04–4.61)2.27 (1.54–3.33)3.00 (2.02–4.46)
      Note. SNP, single nucleotide polymorphism; T2D, type 2 diabetes; VDR , vitamin D receptor. P-value > alpha (0.05) means not significant.
    下载: 导出CSV

    S4.   Associations of VDR variants (rs2189480 and rs739837) genotypes with cardiovascular metabolic indexes

    Cardiovascular metabolic indexesrs2189480rs739837
    GA+GGAAP*GGGC+CCP*
    TG (mmol/L)1.65 ± 1.041.67 ± 1.260.9571.70 ± 1.151.60 ± 1.180.215
    TC (mmol/L)4.62 ± 0.954.57 ± 0.980.2774.63 ± 0.984.54 ± 0.960.325
    HDL-C (mmol/L)1.17 ± 0.311.17 ± 0.310.9621.19 ± 0.311.15 ± 0.300.189
    LDL-C (mmol/L)2.73 ± 0.772.68 ± 0.800.3332.70 ± 0.802.69 ± 0.760.936
    SBP (mmHg)128.14 ± 20.19128.80 ± 19.230.818127.45 ± 18.94130.00 ± 20.370.030
    DBP (mmHg)78.17 ± 10.4878.98 ± 10.950.40678.10 ± 10.8579.47 ± 10.630.084
    FPG (mmol/L)4.96 ± 0.764.99 ± 0.710.5524.98 ± 0.734.97 ± 0.730.825
    FINS (mIU/L)11.98 ± 5.7012.17 ± 6.190.79511.28 ± 4.3212.60 ± 6.510.001
    HOMA-IR2.67 ± 1.372.73 ± 1.510.7612.23 ± 0.322.21 ± 0.260.238
    HOMA-β2.22 ± 0.282.22 ± 0.310.9892.72 ± 1.502.70 ± 1.370.757
      Note. *Adjusted for age, smoking, alcohol drinking, exercise. TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SBP, systolic blood pressure; DBP, diastolic blood pressure; FPG, fasting plasma glucose; FINS, Fasting insulin; HOMA-β, Homeostasis Model Assessment-β [HOMA-β (%) = 20 × FINS (mIU/L) / (FPG (mmol/L) − 3.5)]; HOMA-IR, Homeostasis Model Assessment Insulin Resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L) / 22.5]; P-value > alpha (0.05) means not statistically significant.
    下载: 导出CSV
  • [1] Sun H, Saeedi P, Karuranga S, et al. IDF diabetes atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract, 2022; 183, 109119. doi:  10.1016/j.diabres.2021.109119
    [2] Yang L, Ma JF, Zhang XL, et al. Protective role of the vitamin D receptor. Cell Immunol, 2012; 279, 160−6. doi:  10.1016/j.cellimm.2012.10.002
    [3] American Diabetes Association. Nutrition recommendations and interventions for diabetes: a position statement of the American Diabetes Association. Diabetes Care, 2008; 31 Suppl 1, S61−78.
    [4] Han JC, Du J, Zhang YJ, et al. Vitamin D receptor polymorphisms may contribute to asthma risk. J Asthma, 2016; 53, 790−800. doi:  10.3109/02770903.2016.1158267
    [5] Saccone D, Asani F, Bornman L. Regulation of the vitamin D receptor gene by environment, genetics and epigenetics. Gene, 2015; 561, 171−80. doi:  10.1016/j.gene.2015.02.024
    [6] Zeitz U, Weber K, Soegiarto DW, et al. Impaired insulin secretory capacity in mice lacking a functional vitamin D receptor. FASEB J, 2003; 17, 509−11.
    [7] Ogunkolade BW, Boucher BJ, Prahl JM, et al. Vitamin D receptor (VDR) mRNA and VDR protein levels in relation to vitamin D status, insulin secretory capacity, and VDR genotype in Bangladeshi Asians. Diabetes, 2002; 51, 2294−300. doi:  10.2337/diabetes.51.7.2294
    [8] Kamei Y, Kawada T, Kazuki R, et al. Vitamin D receptor gene expression is up-regulated by 1, 25-dihydroxyvitamin D3 in 3T3-L1 preadipocytes. Biochem Biophys Res Commun, 1993; 193, 948−55. doi:  10.1006/bbrc.1993.1717
    [9] Vasilopoulos Y, Sarafidou T, Kotsa K, et al. VDR TaqI is associated with obesity in the Greek population. Gene, 2013; 512, 237−9. doi:  10.1016/j.gene.2012.10.044
    [10] Burton GR, Guan Y, Nagarajan R, et al. Microarray analysis of gene expression during early adipocyte differentiation. Gene, 2002; 293, 21−31. doi:  10.1016/S0378-1119(02)00726-6
  • [1] LIU Jing, LI Sheng Bing, LUO Xie, YUAN Lei, LAI Ye Rui, ZHANG Li Li.  Type 2 Diabetes and Hyperlipidemia Caused by AKT2 Gene Combined with PLIN1 Gene Mutation: A Case Report . Biomedical and Environmental Sciences, 2023, 36(4): 371-375. doi: 10.3967/bes2023.043
    [2] ZHANG Yu Jing, LI Wen Jie, ZHANG Dong Dong, LIU Ya Ping, XU Ze, GAO Jiao Jiao, GU Chen Xi, LI Xing.  Association of VDR Gene Variants with Hyperglycemia in Henan Rural Population . Biomedical and Environmental Sciences, 2023, 36(3): 284-288. doi: 10.3967/bes2023.031
    [3] LI Na, ZHANG Li Hong, GAO Yu Min, LI Hai Ling, LI Wu Yun Ta Na, PANG Hui, ZHAO Yan Ping, ZHENG Hai Fang, DONG Jia Hui, ZHAO Ling Yan.  Adipocytokine Gene SNPs and Type 2 Diabetes in Mongolians: A Gene-Gene and Gene-Environment Interaction Study . Biomedical and Environmental Sciences, 2023, 36(12): 1183-1188. doi: 10.3967/bes2023.155
    [4] HE Ya Ling, LIU Xiao Tian, TU Run Qi, PAN Ming Ming, NIU Miao Miao, CHEN Gong Bo, HOU Jian, MAO Zhen Xing, HUO Wen Qian, LI Shan Shan, GUO Yu Ming, WANG Chong Jian.  Association of Residential Greenness with the Prevalence of Metabolic Syndrome in a Rural Chinese Population: the Henan Rural Cohort Study . Biomedical and Environmental Sciences, 2022, 35(1): 89-94. doi: 10.3967/bes2022.013
    [5] YIN Yi Hong, HU Wei, ZHU Zheng Yu, HAN Min, XIE Zhao Hong, BI Jian Zhong.  The SOST (Sclerostin) Gene +219 Site Methylation is the Risk Factor of Acute Cerebral Ischemia in the Han Chinese Population: A Case-control Study . Biomedical and Environmental Sciences, 2020, 33(9): 708-712. doi: 10.3967/bes2020.092
    [6] SUN Xiao Ya, HE Zhi Qiang, WANG Li Qun, WANG Zhi Zhong, BAI Fei Hu, YOU Yan Jie.  Interaction Effects of Gene Polymorphisms and Obesity on Nonalcoholic Fatty Liver Disease . Biomedical and Environmental Sciences, 2019, 32(10): 793-796. doi: 10.3967/bes2019.100
    [7] WANG Bo Shen, XU Kai, ZHANG Hong, PU Yue Pu, YIN Li Hong, HAN Lei, CHEN Lin, WANG Ning, ZHU Bao Li, ZHANG Juan.  Association between NFE2L2 Gene Polymorphisms and Noise-induced Hearing Loss in a Chinese Population . Biomedical and Environmental Sciences, 2019, 32(6): 465-470. doi: 10.3967/bes2019.063
    [8] WANG Yan, HAN Han, WANG Jun, SHEN Fang, YU Fei, WANG Ling, YU Song Cheng, ZHANG Dong Dong, SUN Hua Lei, XUE Yuan, BA Yue, WANG Chong Jian, LI Wen Jie.  Polymorphisms in CYP2R1 Gene Associated with Serum Vitamin D Levels and Status in a Chinese Rural Population . Biomedical and Environmental Sciences, 2019, 32(7): 550-553. doi: 10.3967/bes2019.072
    [9] HAN Han, ZHAO Ming Xu, WANG Yan, WANG Jun, REN Bing Nan, GE Hui Na, WANG Teng, SUN Bei Bei, BA Yue, LI Wen Jie.  Significant Polymorphisms of Vitamin D Receptor Gene (rs2189480 and rs3847987) Related to the Risk of Type 2 Diabetes in Henan Rural Area . Biomedical and Environmental Sciences, 2019, 32(1): 58-62. doi: 10.3967/bes2019.009
    [10] LI Xiu Ting, YU Peng Fei, GAO Yan, GUO Wen Hui, WANG Jun, LIU Xin, GU Ai Hua, JI Gui Xiang, DONG Qiu, WANG Bo Shen, CAO Ying, ZHU Bao Li, XIAO Hang.  Association between Plasma Metal Levels and Diabetes Risk:a Case-control Study in China . Biomedical and Environmental Sciences, 2017, 30(7): 482-491. doi: 10.3967/bes2017.064
    [11] YOU Hai Fei, ZHAO Jing Zhi, ZHAI Yu Jia, YIN Lei, PANG Chao, LUO Xin Ping, ZHANG Ming, WANG Jin Jin, LI Lin Lin, WANG Yan, WANG Qian, WANG Bing Yuan, REN Yong Cheng, HU Dong Sheng.  Association between Low-density Lipoprotein Receptor-related Protein 5 Polymorphisms and Type 2 Diabetes Mellitus in Han Chinese:a Case-control Study . Biomedical and Environmental Sciences, 2015, 28(7): 510-517. doi: 10.3967/bes2015.073
    [12] WANG Sheng Li, YU Lu Gang, LIU Ren Ping, ZHU Wan Zhan, GAO Wei Min, XUE Li Ping, JIANG Xu.  Gene-Gene Interaction of GJB2, SOD2, and CAT on Occupational Noise-induced Hearing Loss in Chinese Han Population . Biomedical and Environmental Sciences, 2014, 27(12): 965-968. doi: 10.3967/bes2014.131
    [13] LI Jing Yun, WANG Sheng, HE Li Hua, WU Shan Shan, YANG Lei, YU Shan Fa, LI Li Ping, WANG Jian Xin, HUANG Yan Di.  Risk Factors of Low Back Pain among the Chinese Occupational Population:A Case-control Study . Biomedical and Environmental Sciences, 2012, 25(4): 421-429. doi: 10.3967/0895-3988.2012.04.007
    [14] HE Yu Na, FESKENS EJ, LI Yan Ping, ZHANG Jian, FU Ping, MA Guan Sheng, YANG Xiao Guang.  Association between High Fat-low Carbohydrate Diet Score and Newly Diagnosed Type 2 Diabetes in Chinese Population . Biomedical and Environmental Sciences, 2012, 25(4): 373-382. doi: 10.3967/0895-3988.2012.04.001
    [15] YU HongJie, FU ChaoWei, WANG JianMing, XUE HengChuan, XU Biao.  Interaction between XRCC1 Polymorphisms and Intake of Long-Term Stored Rice in the Risk of Esophageal Squamous Cell Carcinoma: A Case-Control Study . Biomedical and Environmental Sciences, 2011, 24(3): 268-274. doi: 10.3967/0895-3988.2011.03.010
    [16] WANG Qiang, BAI Xue Tao, XU Dong Qun, LI Hong, XU Chun Yu, FANG Jian Long.  Association of Polymorphisms of STAT6 and SO2 with Chinese Childhood Asthma: a Case-control Study . Biomedical and Environmental Sciences, 2011, 24(6): 670-677. doi: 10.3967/0895-3988.2011.06.012
    [17] CONG-RONG WANG, CHENG HU, RONG ZHANG, QI-CHEN FANG, XIAO-JING MA, WEI-PING JIA, KUN-SAN XIANG.  Association of A Common Haplotype of Hepatocyte Nuclear Factor 1α With Type 2 Diabetes in Chinese Population . Biomedical and Environmental Sciences, 2007, 20(1): 41-46.
    [18] MIN YANG, CHANG-CHUN QIU, QUN XU, HONG-DING XIANG.  Association of Angiotensin Converting Enzyme Gene I/D Polymorphism With Type 2 Diabetes Mellitus . Biomedical and Environmental Sciences, 2006, 19(4): 323-327.
    [19] Xiao-Xin He, Shui-Gao Jin.  Interaction and Its Solution in Individual Matching Case-control Study . Biomedical and Environmental Sciences, 2003, 16(1): 40-46.
    [20] SUN Hong-xia, ZHANG KUI-XING, DU WEI-NAN, SHI JIN-XIU, JIANG ZHENG-WEN, SUN HAO, ZUO JIN, HUANG WEI, CHEN ZHU, SHEN YAN, YAO Zhi-Jian, QIANG Bo-Qin, FANG FU-DE.  Single Nucleotide Polymorphisms in CAPN10 Gene of Chinese People and Its Correlation With Type 2 Diabetes Mellitus in Han People of Northern China . Biomedical and Environmental Sciences, 2002, 15(1): 75-82.
  • 22051Supplementary Materials.pdf
  • 加载中
图(1) / 表ll (6)
计量
  • 文章访问数:  482
  • HTML全文浏览量:  206
  • PDF下载量:  61
  • 被引次数: 0
出版历程
  • 收稿日期:  2022-02-10
  • 录用日期:  2022-09-20
  • 刊出日期:  2022-11-25

Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population

doi: 10.3967/bes2022.136
    基金项目:  The study was funded by the National Natural Science Foundation of China [Grant no. 81872626, 82003454]; Chinese Nutrition Society - Zhendong National Physical Fitness and Health Research Fund [Grant no. CNS-ZD2019066]; and Key R&D and promotion projects in Henan Province [212102310219 and 202102310120]
    作者简介:

    XU Ze, female, born in 1995, Master, majoring in nutrition and food hygiene

    通讯作者: LI Xing, E-mail: lixing530@zzu.edu.cn, Tel: 86-371-67781305.

English Abstract

XU Ze, ZHANG Dong Dong, LIU Ya Ping, ZHANG Yu Jing, XUE Yuan, GAO Jiao Jiao, ZHAO Tong, HUANG Hao Yue, LI Wen Jie, LI Xing. Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population[J]. Biomedical and Environmental Sciences, 2022, 35(11): 1074-1078. doi: 10.3967/bes2022.136
Citation: XU Ze, ZHANG Dong Dong, LIU Ya Ping, ZHANG Yu Jing, XUE Yuan, GAO Jiao Jiao, ZHAO Tong, HUANG Hao Yue, LI Wen Jie, LI Xing. Association of VDR Polymorphisms and Gene–obesity Interaction with Type 2 Diabetes: A Case–control Study among Chinese Rural Population[J]. Biomedical and Environmental Sciences, 2022, 35(11): 1074-1078. doi: 10.3967/bes2022.136
  • In recent years, the prevalence of type 2 diabetes (T2D) has increased at an alarming rate, and the epidemic of T2D mellitus and its complications poses a major global health threat. Globally, the International Diabetes Federation estimated that 10.5% of adults aged 20–79 years (536 million adults) had diabetes mellitus in 2021. If these trends continue, approximately 783 million people will have diabetes by 2045[1]. T2D is a complex metabolic disease caused by a complex interplay of genetic and environmental risk factors. Obesity, a sedentary lifestyle, and energy-dense diets are the best-known environmental factors associated with T2D. The prevalence of overweight and obesity is high in Chinese, and the number of people with obesity accounts for approximately 46% of the adult population.

    Accumulating evidence has demonstrated that low circulating vitamin D levels are associated with an increased risk of developing T2D across diverse populations. The vitamin D receptor (VDR) is a member of nuclear receptor superfamily 1, and it is expressed in many cells and tissues, such as pancreatic β-cells, liver, and adipose tissue. VDR mediates the biological effects of vitamin D, and 1,25-dihydroxyvitamin D depends on VDR to exert its biological effects. Most studies on the association between VDR polymorphism and T2D have focused on six single nucleotide polymorphisms (SNPs): Fok1, Bsm1, Taq1, Apa1, Tru91, and 3’UTR [2]. However, the association of VDR (rs2189480 and rs739837) variants with T2D remains unclear, and the interaction between the loci and obesity on T2D susceptibility has not been explored. Therefore, this case–control study aimed to investigate the relationship between VDR SNPs and T2D and further explore the combined effects of obesity and VDR gene polymorphisms on T2D, which may help in predicting the risk of T2D and identifying individuals at high risk. Furthermore, the study may provide novel strategies for T2D prevention and individualized anti-diabetes treatment.

    From July to August 2013 and from July to August 2015, we recruited 2,378 Han Chinese from Wuzhi, Xin’an, and Houzhai County in Henan Province. All participants completed a standard questionnaire (general characteristics, lifestyle, medical history, and family history of disease), body measurements, and fasting blood collection. T2D was defined as fasting serum glucose ≥ 7.0 mmol/L and/or taking hypoglycemic medication and/or having a self-reported T2D history [3]. Individuals with a self-reported history of gestational diabetes or other types of diabetes, coronary heart disease, chronic kidney diseases, cancer, or failed DNA extraction were excluded from the study. We performed a 1:2 ratio case–control match by sex and age (within 3 years); finally, a total of 397 T2D and 794 non-T2D cases were included in our study.

    All participants provided informed consent before the examination, and the study was approved by the Medical Ethics Committee of Zhengzhou University.

    For each participant, height, weight, waist circumference (WC), and hip circumference were measured twice to the nearest 0.5 kg and 0.1 cm, and the body mass index (BMI) was calculated in kg/m2. All participants were asked to assemble on time after an overnight fast (> 10 h) for blood sample collection. Oral antidiabetic drugs, insulin, and alcohol should have been avoided for at least 12 h before blood collection. Blood samples were used for the DNA isolation and measurements of glucose, lipid profile, and serum 25-hydroxyvitamin D [25(OH)D] concentrations. 25(OH)D levels were measured by human enzyme-linked immunosorbent assay kits (Sangon Biotech, Shanghai, China). The levels of fasting plasma glucose (FPG), triglyceride (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and fasting insulin (FINS) were measured by an automatic biochemical analyzer (KHB, Shanghai, China). The Homeostasis Model Assessment Insulin Resistance (HOMA-IR) was applied as a measure of the degree of insulin resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L)/22.5]. Genomic DNA was isolated from the peripheral blood following the standard procedures of the DNA extraction kit (Bioteke, Beijing, China). The TaqMan probe method and real-time polymerase chain reaction technology were used to identify the rs2189480 and rs739837 genotypes.

    Non-normally distributed variables were shown as median values with interquartile range, and categorical variables were described with frequencies and percentages. The distribution of participants’ characteristics between the case and control groups was compared using the nonparametric Wilcoxon rank-sum test or Chi-square test. Multivariate logistic regression was used to examine the relationship between VDR SNPs and T2D, and odds ratio (OR) and corresponding 95% confidence interval (CI) were calculated after adjustment for smoking, alcohol consumption, physical activity, 25(OH)D, and diabetes family history, assuming codominant, recessive, and dominant models of inheritance. Multivariable linear regression was performed to explore the association of BMI and WC with blood glucose-related parameters, including FPG, hemoglobin A1c, HOMA-IR, and HOMA-β. Multiple logistic regression analysis was used to examine the interaction between VDR SNPs and obesity on T2D risk according to the genotype of each SNP (under the recessive or dominant genotype model) and obesity exposure status. Multiplicative interactions were evaluated using a multiplicative model based on generalized linear regression, with FPG as the dependent variable. Covariance analysis was used to explore the influence of VDR variations on cardiovascular metabolic indicators. Unless otherwise stated, statistical analyses were performed using SAS v.9.1 (SAS Institute, Cary, NC, USA), and the significance level was set at P < 0.05.

    Results of the comparisons of the basic characteristics of the two groups are presented in Supplementary Table S1 (available in www.besjournal.com). A total of 1,191 individuals (397 T2D cases/794 controls) were included in this case–control study, and male and female participants accounted for 39.3% and 61.6%, respectively. Compared with the control group, patients with T2D had higher levels of FPG, FINS, HOMA-IR, TC, TG, and LDL-C but lower levels of HOMA-β and HDL-C (P < 0.05). As expected, patients with T2D had significantly higher BMI and WC than controls, respectively (P < 0.05).

    Table S1.  Basic characteristics of study participants

    CharacteristicsT2D (n = 397)Control (n = 794)χ2/ZP
    Male156 (39.3%)312 (39.3%)01.000
    Age (years)60 (52.5−67.5)60.5 (51.00−67.00)−0.020.982
    Smoking295 (25.7%)576 (27.5%)0.420.518
    Alcohol drinking50 (12.6%)98 (12.3%)0.020.901
    Physical activity15.40< 0.001
     Low193 (39.2%)311(48.7%)
     Moderate76 (17.0%)135 (19.2%)
     High127 (43.8%)347 (32.1%)
    Family history of T2D71 (17.9%)56 (7.1%)32.59< 0.001
    BMI (kg/m2)26.33 (23.98−28.70)24.80 (22.30−27.20)−6.28< 0.001
    WC (cm)91.25 (84.13−97.50)86.00 (78.00−93.50)−7.76< 0.001
    FPG (mmHg)8.46 (6.84−10.94)4.98 (4.52−5.42)−23.77< 0.001
    FINS (mIU/L)13.25 (9.55−18.97)11.04 (8.65−13.77)−6.68< 0.001
    HOMA-β54.61 (29.81−96.93)153.57 (110.26−217.01)−19.14< 0.001
    HOMA-IR5.02 (3.29−7.80)2.42 (1.80−3.23)−17.76< 0.001
    TC (mmol/L)4.61 (4.06−5.29)4.51 (3.94−5.16)−2.210.027
    TG (mmol/L)1.81 (1.19−2.80)1.36 (0.90−2.04)−7.12< 0.001
    HDL-C (mmol/L)1.13 (0.96−1.34)1.18 (0.99−1.39)−2.660.008
    LDL-C (mmol/L)2.63 (2.19−3.17)2.53 (2.02−3.09)−2.470.014
    25(OH)D3 (ng/mL)19.1 (15.02−27.21)19.36 (15.5−24.5)−0.520.601
      Note. Data are given as number (%) or median (interquartile range). Age refers to the age at which the participants were enrolled in the study. Physical activity was divided into three categories (low, moderate, and high) according to the International Physical Activity Questionnaire. BMI, body mass index; WC, waist circumference; FPG, fasting plasma glucose; FINS, fasting insulin; HOMA-β, Homeostasis Model Assessment-β [HOMA-β (%) = 20 × FINS (mIU/L) / (FPG (mmol/L) − 3.5)]; HOMA-IR, Homeostasis Model Assessment Insulin Resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L) / 22.5]; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol. P-value > alpha (0.05) means not statistically significant.

    At present, the results of studies on the association between VDR polymorphism and chronic diseases are still contradictory. Our study explored the association between two loci (rs2189480 and rs739837) and T2D risk through multiple logistic regression analysis (Table 1). The genotypic distributions of the two SNPs did not deviate from the Hardy–Weinberg equilibrium (P > 0.05). Multiple logistic regression analysis showed that participants with the “AA” genotype of rs2189480 were significantly less likely to develop T2D than those with the “GA+GG” genotype (P = 0.036, OR = 1.30, 95% CI 1.02–1.66). The rs2189480 locus is one of the VDR SNPs located in the intron and consists of a G-A change. Studies have shown that rs2189480 affects regulatory T cell function and T cells help regulate inflammatory activity [4]. Therefore, the rs2189480 mutation may have a protective effect on T2D by regulating the inflammatory response. Studies have reported the effect of rs739837 variants on diabetes, but no unanimous conclusions have been drawn, and the role of the rs739837 locus is still unclear. Our study results showed that carrying “CC+GC” of rs739837 rather than “GG” was significantly associated with an increased hazard of developing T2D (P = 0.028, OR = 1.32, 95% CI 1.03–1.70). The rs739837 is located in the 3’UTR of the VDR gene, which contains many regulatory elements that influence its post-transcriptional regulation. A study showed that variations in rs739837 may influence the stability of VDR mRNA and its gene transcription and further increase the risk of diabetes [5].

    Table 1.  Genotypic distribution of VDR polymorphisms and the association of VDR polymorphisms with T2D risk

    SNPT2DControlχ2POR (95% CI)aPHWE
    rs21894804.1550.1250.76
     GG182 (46.08%)327 (42.69%)1
     GA177 (44.26%)339 (44.26%)0.90 (0.69–1.18)0.459
     AA36 (9.11%)100 (13.05%)0.60 (0.38–0.94)0.026
     AA/GA+GG0.63 (0.41–0.97)0.036
     AA+GA/GG0.84 (0.65–1.09)0.180
    rs7398373.7730.1520.77
     GG202 (50.88%)432 (56.18%)1
     GC161 (40.56%)288 (37.45%)1.26 (0.96–1.66)0.095
     CC34 (8.56%)49 (6.37%)1.42 (0.86–2.32)0.170
     CC/GC+GG1.28 (0.79–2.08)0.308
     CC+GC/GG1.32 (1.03–1.70)0.028
      Note. aAdjusted for age, smoking, alcohol drinking, exercise, level of 25(OH)D, and family history of T2D. CI, confidence interval; HWE, Hardy–Weinberg equilibrium; SNP, single nucleotide polymorphism; OR, odds ratio; T2D, type 2 diabetes; VDR, vitamin D receptor. P-value > alpha (0.05) means not significant.

    Obesity is believed to be a promoter of T2D. Obesity is strongly associated with the development of insulin resistance, which in turn plays a key role in the pathogenesis of obesity-associated cardiometabolic complications, including metabolic syndrome components, T2D, and cardiovascular diseases. Our study investigated the association between BMI/WC and T2D in the rural Chinese population (Supplementary Table S2, available in www.besjournal.com) and the influence of BMI/WC and blood glucose-related parameters (Supplementary Table S3, available in www.besjournal.com). The results showed that higher BMI or WC was significantly associated with an increased risk of T2D in Chinese rural populations. Moreover, after multivariable adjustment, we found that for each increase in BMI of 1 kg/m2, the level of FINS and HOMA-IR increased by 0.49 mIU/L and 0.19 (P < 0.05), respectively. For each increase in WC of 1 cm, the level of FPG, FINS, and HOMA-IR increased by 0.017 mmHg, 0.49 mIU/L, and 0.19 (P < 0.05), respectively.

    Table S2.  Association of BMI and WC levels with the risk of T2D

    ContinuousModel 1Model 2
    BMI
     Continuous
     T1 (< 23.7 kg/m2)11
     T2 (23.7−26.8 kg/m2)1.68 (1.22−2.31) *1.35 (0.97−1.89)
     T3 (≥ 26.8 kg/m2)2.52 (1.84−3.43) *1.88 (1.35−2.63) *
     P-trend< 0.0010.003
    WC
     Continuous
     T1 (< 82.5 cm)11
     T2 (82.5−92.5 cm)2.15 (1.56−2.97) *1.87 (1.34−2.61) *
     T3 (≥ 92.5 cm)2.94 (2.13−4.04) *2.22 (1.58−3.14) *
     P-trend< 0.001< 0.001
      Note. Model 1: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D; Model 2: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D, SBP, DBP, TC, TG, HDL-C, LDL-C. BMI, body mass index; WC, waist circumference *P < 0.05. P-value > alpha (0.05) means not statistically significant.

    Table S3.  Associations (β, 95% CI) of BMI and WC with blood glucose related parameters

    OutcomesBMIWC
    Model 1Model 2Model 1Model 2
    FPG0.076 (0.032−0.119)*0.010 (−0.035−0.055)0.040 (0.025−0.054)*0.017 (0.002−0.032)*
    FINS0.592 (0.479−0.706)*0.493 (0.367−0.619)*0.233 (0.194−0.271)*0.202 (0.159−0.245)*
    HOMA-IR0.249 (0.198−0.299)*0.186 (0.132−0.241)*0.100 (0.084−0.117)*0.080 (0.062−0.098)*
    HOMA-β3.183 (−11.166−17.531)2.837 (−13.628−19.301)0.671 (−4.246−5.587)0.690 (−4.946−6.326)
      Note. Model 1: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D; Model 2: Adjusted for age, sex, smoking, alcohol drinking, exercise, family history of T2D, SBP, DBP, TC, TG, HDL-C, LDL-C. BMI, body mass index; WC, waist circumference; FPG, fasting plasma glucose; FINS, fasting insulin; HOMA-IR, Homeostasis Model Assessment Insulin Resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L) / 22.5]; HOMA-β, Homeostasis Model Assessment-β [HOMA-β (%) = 20 × FINS (mIU/L) / (FPG (mmol/L) − 3.5)]; *P < 0.05. P-value > alpha (0.05) means not statistically significant.

    To assess the interaction between obesity status and SNPs of the VDR gene on T2D, the participants were divided into four subgroups by genetic models of SNPs and obesity status (Table 2). An interaction between VDR variants and obesity on T2D risk has been found (Pinteraction = 0.029). Compared with non-obese individuals carrying rs739837 “GG” genotype, those with overweight/obesity carrying the “CC” or “GC” genotype had a higher risk of developing T2D (OR = 3.06, 95% CI 2.04–4.61). Further, the results of the generalized linear model showed a significant interaction between rs739837 and BMI (P = 0.041). At higher BMI (> 26.96 kg/m2), rs739837 “C” allele carriers had higher FPG levels and showed a stronger effect than individuals with the “GG” genotype (Figure 1). VDR is expressed in pancreatic β cells, and alleles of VDR regulate β-cell function. Human and animal studies have shown that disruption of the VDR signaling pathway can affect insulin secretion and decrease insulin mRNA levels [6, 7]. The ability of fat cells to store fat is gradually reduced with obesity, and fat is stored in places such as visceral tissue, which adversely affects insulin signaling. The VDR gene mutation and obesity may have an interactive effect on T2D risk by affecting insulin secretion. Meanwhile, our results found that subjects carrying "GC" or "CC "of rs739837 had higher fasting insulin levels than "GG" carriers, suggesting that mutations in the VDR gene may affect the insulin signaling pathway and lead to insulin resistance in the subjects (Supplementary Table S4, available in www.besjournal.com). On the contrary, studies have reported that VDR is expressed in adipose tissue and dynamically upregulated during adipocyte differentiation [8]. The T allele of TaqI loci was associated with a higher risk of obesity in Greek individuals (contributing to an elevated BMI of 3 kg/m2 per risk allele) [9]. VDR plays a key role in fat formation by affecting vitamin D levels. Meanwhile, the expression of VDR mRNA also changes during adipocyte differentiation [10]. Therefore, we speculate that the interaction between VDR gene variation and adipocyte differentiation and metabolism may contribute to the risk of T2D. However, the specific mechanism of interaction needs further study.

    Figure 1.  Multiplicative interactions between rs739837 (G > C) and BMI (A) or WC (B) and the associations of these interactions with FPG. The blue trend line represents the correlations between BMI or WC and FPG levels in individuals with risk alleles. The red trend line represents the correlation between BMI or WC and FPG levels in individuals without risk alleles. The point where the two lines cross indicates that for individuals whose BMI or WC is higher than the level value at this point, risk allele carriers have a greater effect on systolic or diastolic blood pressure than non-risk allele carriers. BMI, body mass index; FPG, fasting plasma glucose; WC, waist circumference.

    Table 2.  Interaction analysis of VDR polymorphisms and obesity on T2D

    SNPOverweight/obesityPinteractionAbdominal obesityPinteraction
    NoYesNoYes
    rs21894800.4220.626
     AA11.51 (0.67–3.41)13.99 (1.24–7.24)
     GA+GG1.14 (0.56–2.31)2.47 (1.23–4.91)1.79 (0.81–3.95)4.07 (1.88–8.83)
    rs7398370.0290.826
     GG12.86 (1.93–4.26)11.26 (0.80–1.98)
     CC+GC1.96 (1.24–3.10)3.06 (2.04–4.61)2.27 (1.54–3.33)3.00 (2.02–4.46)
      Note. SNP, single nucleotide polymorphism; T2D, type 2 diabetes; VDR , vitamin D receptor. P-value > alpha (0.05) means not significant.

    This study explores the interaction between VDR gene (rs2189480 and rs739837) variation and obesity on T2D susceptibility in the Chinese rural population. Multiplicative interactions between rs739837 variants and BMI or WC were evaluated using a multiplicative model based on generalized linear regression, with FPG as the dependent variable. This study also employed a 1:2 case–control design, including serum samples from 1191 subjects covering a wide age span, which may be suitable for the exploration of T2D etiology. However, this study has several limitations. First, this was a single study limited to Chinese Han populations, which may limit the generalizability of the findings to other ethnic groups. Second, only one gene was discussed in the study; other regulatory genes may contribute to the pathogenesis of T2D. Furthermore, the study was limited by the lack of data on more informative measures of adiposity in addition to BMI and WC, such as visceral fat index or percentage of body fat. Therefore, further studies are needed to verify the observed interactions in other populations and further investigate the mechanisms.

    In conclusion, our study found that the “CC+GC” genotypes of rs739837 in the VDR gene were associated with an increased risk for T2D development. Interactions between VDR and obesity were associated with the susceptibility for T2D. The observations from this study constitute evidence that people who carry particular variants of the VDR-rs739837, which improved BMI level control, may be beneficial in inhibiting T2D progression.

  • Table S4.  Associations of VDR variants (rs2189480 and rs739837) genotypes with cardiovascular metabolic indexes

    Cardiovascular metabolic indexesrs2189480rs739837
    GA+GGAAP*GGGC+CCP*
    TG (mmol/L)1.65 ± 1.041.67 ± 1.260.9571.70 ± 1.151.60 ± 1.180.215
    TC (mmol/L)4.62 ± 0.954.57 ± 0.980.2774.63 ± 0.984.54 ± 0.960.325
    HDL-C (mmol/L)1.17 ± 0.311.17 ± 0.310.9621.19 ± 0.311.15 ± 0.300.189
    LDL-C (mmol/L)2.73 ± 0.772.68 ± 0.800.3332.70 ± 0.802.69 ± 0.760.936
    SBP (mmHg)128.14 ± 20.19128.80 ± 19.230.818127.45 ± 18.94130.00 ± 20.370.030
    DBP (mmHg)78.17 ± 10.4878.98 ± 10.950.40678.10 ± 10.8579.47 ± 10.630.084
    FPG (mmol/L)4.96 ± 0.764.99 ± 0.710.5524.98 ± 0.734.97 ± 0.730.825
    FINS (mIU/L)11.98 ± 5.7012.17 ± 6.190.79511.28 ± 4.3212.60 ± 6.510.001
    HOMA-IR2.67 ± 1.372.73 ± 1.510.7612.23 ± 0.322.21 ± 0.260.238
    HOMA-β2.22 ± 0.282.22 ± 0.310.9892.72 ± 1.502.70 ± 1.370.757
      Note. *Adjusted for age, smoking, alcohol drinking, exercise. TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; SBP, systolic blood pressure; DBP, diastolic blood pressure; FPG, fasting plasma glucose; FINS, Fasting insulin; HOMA-β, Homeostasis Model Assessment-β [HOMA-β (%) = 20 × FINS (mIU/L) / (FPG (mmol/L) − 3.5)]; HOMA-IR, Homeostasis Model Assessment Insulin Resistance [HOMA-IR = FPG (mmol/L) × FINS (mIU/L) / 22.5]; P-value > alpha (0.05) means not statistically significant.
参考文献 (10)
补充材料:
22051Supplementary Materials.pdf

目录

    /

    返回文章
    返回