The Associations of Novel Vitamin D3 Metabolic Gene CYP27A1 Polymorphism, Adiponectin/Leptin Ratio, and Metabolic Syndrome in Middle-Aged Taiwanese Males

Metabolic syndrome (MetS) confers increased risks of cardiovascular disease (CVD). Both vitamin D3 and adipocytokines (especially adiponectin and leptin) have a great impact on CVD and MetS. In vitamin D3 metabolism, the vitamin D3 25-hydroxylase (CYP27A1) and 25-hydroxyvitamin D3 1-alpha-hydroxylase (CYP27B1) are two key enzymes. This study aimed to examine the influence of vitamin D3 CYP27 single nucleotide polymorphisms (SNPs) on adipocytokines and MetS. Cross-sectional data and DNA samples were collected from male volunteers (n = 649, age: 55.7 ± 4.7 years). Two tagging SNPs, CYP27A1 rs4674344 and CYP27B1 rs10877012, were selected from the HapMap project. MetS was significantly associated with the CYP27A1 rs4674344 SNP (P = 0.04) and the ratio of adiponectin/leptin (A/L ratio) was most correlated to the CYP27A1 rs4674344 SNP, appearing to be significantly lower in T-carriers than in AA subjects (3.7 ± 4.0 versus 5.1 ± 6.0, P = 0.001) and significantly negatively associated after adjustment. For each MetS component associated with the CYP27A1 rs4674344 SNP, the A/L ratios were significantly negative in preclinical stage (condition not meeting the individual criteria), except the blood pressure. In conclusion, CYP27A1 rs4674344 SNP, A/L ratio, and MetS are significantly associated and T-carriers might have a higher risk of developing MetS due to low A/L ratios in the preclinical stage.


Introduction
Metabolic syndrome (MetS), a collection of cardiometabolic risk factors that includes obesity, insulin resistance, hypertension, and dyslipidemia, confers increased risks of coronary Subjects with AA genotype in CYP27A1 rs4674344 (n = 447) (2) SNPs had a minor allele frequency (MAF) ≥ 0.10; (3) the other unselected SNPs could be captured by one of the tag SNPs with a LD of ≥ 0.80.
Vitamin D 3 and adipocytokine interact with each other and influence the development of MetS. However, in the view of genetics, little about these two vitamin D 3 metabolic genes with adipocytokines and MetS is studied. Therefore, our purpose was to explore the potential associations of vitamin D 3 metabolism genes, CYP27A1 and CYP27B1, with adiponectin and leptin and, subsequently, with MetS.

Study Population and Protocol.
In this cross-sectional study, 694 Taiwanese males (age: 44-77 years) living in Kaohsiung City were recruited from a free health screening program held by our institution [1,32]. Each of them participated with informed written consent. Ethical approval following the Declaration of Helsinki was authorized by the Institutional Research Ethics Committee of Kaohsiung Medical University Hospital.
The men who had a previous diagnosis of hypertension, diabetes mellitus (DM), or hyperlipidemia controlled by regular medication were included in this study, but those who were diagnosed with labile hypertension, labile DM, and advanced liver and/or renal disease were excluded. Entire records of medical, surgical, and psychosexual history, as well as the results of physical examinations including measurements of body weight, height, and blood pressure for each participant, were collected ( Figure 1).
International Journal of Endocrinology 3 The subjects were classified as alcohol drinkers, cigarette smokers, or betel nut chewers if they had regularly consumed any alcoholic beverage ≥1 time per week, smoked ≥10 cigarettes per week, or chewed ≥7 betel quids per week for at least the last 6 months [33,34]. Hypertension was defined as a systolic blood pressure of ≥140 mmHg or diastolic blood pressure of ≥90 mmHg. Hyperlipidemia was defined as a total cholesterol level of ≥200 mg/dL or triglyceride level of ≥200 mg/dL [1,35]. DM was diagnosed when the fasting blood glucose (FBG) was ≥126 mg/dL.

Selection, Identification, and Genotyping of the Single
Nucleotide Polymorphism. Tag single nucleotide polymorphisms (SNPs) were selected by searching data on Han Chinese in the HapMap project (http://www.hapmap.org/) using Haploview software [37]. Afterward the tag SNPs were identified by the following criteria: (1) SNPs located in the CYP27A1 and CYP27B1 genes or within the flanking 5 kb regions; (2) a minor allele frequency of ≥0.10; and (3) other unselected SNPs that could be captured by one of the tag SNPs with a linkage disequilibrium of 2 ≥ 0.80. Selection of SNPs had been completed in March 2006, and only 1 tag SNP had been identified for each of the CYP27A1 and CYP27B1 genes, namely, rs4675344 on the intron 1 region and rs10877012 on the promoter region ( Figure 1). DNA was extracted from the peripheral whole blood using a Puregene DNA Isolation Kit (Gentra Systems Inc., Minneapolis, MN). The two SNPs were determined using a TaqMan 5 allelic discrimination assay performed on an assays-on-demand SNP genotyping kit (Applied Biosystems, Foster City, CA). SNP amplification assays including 10 ng of sample DNA in 25 L of reaction solution contained in 12.5 L of 2x TaqMan Universal PCR Mix (Applied Biosystems) and 1.25 L of predeveloped assay reagent from the SNP genotyping product (Applied Biosystems) contained in two primers and two MGB-TaqMan probes were performed according to the manufacturer's instructions. Reaction conditions consisted of preincubation at 50 ∘ C for 2 min and 95 ∘ C for 10 min, followed by 40 cycles at 92 ∘ C for 15 s and 60 ∘ C for 1 min. Amplifications were performed in an ABI Prism 7500 Sequence Detection System (Applied Biosystems). Our genotyping results were also validated in accordance with the identical results from the 25 patients and 25 controls (about 10% of our participants) who were randomly selected for regenotyping by direct sequencing.

Statistical
Analysis. The demographic and laboratory data are presented as mean ± standard deviation (SD). For the controls, the Hardy-Weinberg equilibrium was checked using the chi-square test. The categorical variables and continuous variables were examined by the contingency tables (Pearson chi-square test) and the Student's -test, respectively. Linkage disequilibrium between the polymorphisms was assessed via Haploview software [37]. The ANCOVA test was used to evaluate the associations between the CYP27A1 rs4675344 and adiponectin/leptin ratio (A/L ratio) controlling for other covariates. Multiple linear regression analysis was used to examine the associations between the A/L ratio and MetS score for the CYP27A1 rs4675344. Since the established cardiovascular risk factors including hypertension and DM are the components of MetS, we were only adjusted for sex, age, coronary artery disease (CAD), stroke, alcohol consumption, and smoking status. We also studied the effects of the individual MetS component. All statistical analyses were performed by SPSS 18.0 software (Chicago, IL, USA) with significance levels of = 0.05.

Patient Characteristics.
The baseline characteristics and biochemical data of our subjects with respect to CYP27A1 rs4675344 were summarized in Table 1. A total of 649 patients had adequate qualities for analysis, of whom 447 (68.9%), 179 (27.6%), and 23 (3.5%) patients had the AA, AT, and TT genotypes, respectively. Due to the low minor allele frequency, the patients with TT and patients with AT were united into the T-carriers for analysis. Compared with the patients with the AA genotype, the T-carriers had significantly higher WC (87.1 ± 6.9 versus 85.7 ± 7.2 cm, = 0.003), positive history of dyslipidemia, total and HDL cholesterol  (Table 2). However, no significant associations were found in the CYP27B1 rs10877012 SNP.

Regression Analysis of the CYP27A1 rs4675344 SNP against Other Study Parameters for A/L Ratio.
Four models were established for our regression analysis, namely, Model 1 that was adjusted for baseline characteristics including age, BMI, WC, hip circumference, systolic blood pressure, and diastolic blood pressure; Model 2 that was then adjusted for exsmoker, current smoker, ex-betel nut chewer, current betel nut chewer, exdrinker, and current drinker; Model 3 that was further adjusted for comorbidities including DM, hypertension, dyslipidemia, CAD, and stroke; and Model 4 that was farther adjusted for biochemistry data including fasting glucose, total cholesterol, low density lipoprotein (LDL), HDL, and triglycerides (TG). Negative associations between the CYP27A1 rs4675344 SNP and the A/L ratio were found consistently from Model 1 to Model 4. In addition, obesity and the parameters related to MetS, including BMI and WC, were also negatively related to the A/L ratio. In Model 4, LDL and fasting glucose were also significantly, negatively associated with the A/L ratio (Table 3).

Associations between the Components of the MetS and CYP27A1 rs4674344 for the A/L Ratio.
The A/L ratio showed greatest association with the CYP27A1 rs4674344 SNP, and the MetS also had significant association, too. Therefore, we further examined the influence of MetS and its components on the associations between the CYP27A1 rs4674344 SNP and A/L ratio. Greater association between A/L ratio and the CYP27A1 rs4674344 SNP was noted in the low MetS score group when compared to the high MetS score group ( Figure 2). Significant A/L ratio was only found when most of the MetS components were in the preclinical stage (not meeting individual criteria). However, there was no significant association with the BP component after adjusting for age, BMI, smoking, alcohol drinking, betel nut chewing, CAD, total cholesterol, LDL, HDL, and the other 4 components for the CYP27A1 rs4674344 SNP (Table 4, Figure 3).

Discussion
This study selected the SNPs by searching Han Chinese data from the HapMap project (http://www.hapmap.org/) using Haploview software [37]. These SNPs were located in the CYP27A1 genes and CYP27B1 genes or within the flanking 5 kb regions with a minor allele frequency ≥0. 10. Accordingly, the other unselected SNPs could be captured by one of the tag SNPs with a linkage disequilibrium of 2 ≥ 0.80. As a result, only 1 tag SNP was obtained for each of the CYP27A1 and CYP27B1 genes, namely, the rs4675344 on intron 1 region and the rs10877012 on the promoter region ( Figure 1). The MetS and its HDL component are significantly associated with the CYP27A1 rs4674344 SNP. CYP27A1 mRNA expression is also significantly associated with MetS in one in vivo study [38]. The HDL cholesterol has been reported to have a strong correlation with the adiponectin and to affect the adipocyte metabolism and adiponectin expression [39,40]. And it is known that mutations in CYP27A1 are associated with cerebrotendinous xanthomatosis, a rare lipid storage disease with the deposition of a form of cholesterol (cholestanol) in the brain and other tissues [41][42][43][44]. Our study provides further novel information of the impact of CYP27A1 rs4674344 SNP on both of them. However, like previous document, no CYP27 SNPs in our study were associated with 1,25(OH)(2)D levels [45].
A/L ratio is the best surrogate for inflammatory burden and insulin resistance [46,47]. Adiponectin acts as an insulin-sensitizing adipocytokine and an anti-inflammatory factor especially with regard to atherosclerosis [48]. Leptin is known to be a satiety factor regulating body weight by suppressing appetite and stimulating energy expenditure. Leptin is also a proinflammatory adipocytokine, which induces the development of T helper 1 cells that contribute to the progress of inflammation [49][50][51]. For in-depth analysis in our study, A/L ratio was not only significantly higher in the CYP27A1 rs4674344 SNP T-carriers than in those with the AA genotype (3.7 ± 4.0 versus 5.1 ± 6.0, = 0.001), but was also most correlated with this SNP. Against other study parameters for the A/L ratio, the CYP27A1 rs4675344 SNP consistently had significantly negative association with the A/L ratio from Model 1 to Model 4 after adjusting for baseline characteristics, life habits, comorbidities, and biochemistry profiles. In addition, obesity and MetS-related parameters including BMI, WC, and hip circumference were also negatively associated with the A/L ratio. In Model 4, LDL and fasting glucose were significantly negatively associated with the A/L ratio, too (Table 3). These results all indicated that the CYP27A1 rs4675344 SNP, A/L ratio, and MetS are closely associated with each other.
Another set of novel in-depth information provided by our study related to the CYP27A1 rs4674344 SNP was that the A/L ratio was only significantly different in the preclinical stage of most individual MetS criteria, except for the BP component, but not in the clinical stage of MetS. The SNPs of genes are imprinted in the germ-cell stage and one previous investigation had also reported that the changes of A/L ratio and development of the MetS partly accorded with one's own genetics [52]. Therefore, our novel findings indicated that those T-carriers with the CYP27A1 rs4674344 SNP tend to have a higher risk of developing MetS, owing to an adipocytokine imbalance. However, the CYP27A1 rs4674344 SNP did not seem to play a role in the development of BP component.
Based on our findings and the previous evidence, the CYP27A1 rs4674344 might participate in lipid and adipocytokine homeostasis and in the crosstalk of HDL with adiponectin. However, unequal number of subjects with the polymorphism of CYP27A1 rs4674344 was the main limitation in this study, which might lead to bias in the results. A large-scale cohort study with equal numbers of subjects may be dedicated to investigate the refined causality, 6 International Journal of Endocrinology    including the SNP of CYP27A1 rs4674344 for the major cardiovascular events. In vitro studies may also be needed to clarify the comprehensive role of the CYP27A1 rs4674344 SNP.

Conclusions
This study reported the novel finding that vitamin D 3 metabolism gene CYP27A1 rs4674344 was closely associated International Journal of Endocrinology 9 with A/L ratio and MetS. The A/L ratio alone had significant association with the CYP27A1 rs4675344 SNP. In addition, for the CYP27A1 rs4674344 SNP, the A/L ratio was only significantly different in the preclinical stage of the most individual MetS criteria except the BP component. This finding indicated that the T-carriers were at a higher risk of developing MetS.