Lower Vitamin D Metabolites Levels Were Associated With Increased Coronary Artery Diseases in Type 2 Diabetes Patients in India

The purpose of the present study was to measure six vitamin D metabolites and to find the association between vitamin D deficiency and coronary artery diseases in diabetes (T2DM_CAD). Four groups [control (n = 50), type 2 diabetes (T2DM, n = 71), coronary artery diseases (CAD, n = 28), T2DM_CAD (n = 38)] of total 187 subjects were included in the study. Six vitamin D metabolites (D2, D3, 25(OH)D2, 25(OH)D3, 1,25(OH)2D2, 1,25(OH)2D3), total 25(OH)D and total 1,25(OH)2D were measured by UPLC/APCI/HRMS method in these subjects. Although all the vitamin D metabolites were significantly decreased in T2DM_CAD as compared to both control and T2DM subjects (p < 0.05), only two metabolites i.e., 25(OH)D3 and total 25(OH)D were significantly (p < 0.05) decreased in the T2DM subjects as compared with the control subjects (p < 0.05). Vitamin D3, 1,25(OH)2D2, 25(OH)D, and 1,25(OH)2D levels were significantly decreased in T2DM_CAD subjects as compared with CAD subjects (p < 0.05). Further, multiple logistic regression analysis revealed that total 25(OH)D and total 1,25(OH)2D can be used to predict T2DM (OR 0.82.95% CI 0.68–0.99; p = 0.0208) and T2DM with CAD (OR 0.460, 95% CI 0.242–0.874; p = 0.0177), respectively. Our data concludes that lower concentration of 1,25(OH)2D is associated with type 2 diabetes coexisting with coronary artery diseases in South Indian subjects.

cells. Vitamin D plays pivotal role in the bone and mineral metabolism. Vitamin D deficiency is a common health problem worldwide and is the cause for osteoporosis and osteomalacia, rickets and other bone related disorders. In the last decade, researchers observed that lower vitamin D levels were associated with metabolic diseases like type 1 diabetes, obesity, insulin resistance, cardiovascular diseases, and cancer [6][7][8] . Many studies reported that Indian subjects are more prone to vitamin D deficiency despite the availability of abundant sunshine throughout the year in many parts of India [9][10][11][12][13][14][15][16][17] . Although, there are some studies from Indian population but none of these studies looked into the association of vitamin D metabolites with coronary artery diseases coexisting with type 2 diabetes. For last few decades more than 50 metabolites of vitamin D have been described 18 . Among them very few have been quantified in blood.
Currently, general awareness of association of vitamin D insufficiency and increased risk of several diseases has been increased. However, it is not clear which vitamin D metabolite should be quantified and what is the association of each metabolite with increased risk of diseases. Among all vitamin D metabolites, only 25(OH)D and 1,25(OH) 2 D have received great attention. Hence in the present study six vitamin D metabolites have been quantified by UPLC/APCI/HRMS method. Therefore, the purpose of the present study was to measure six vitamin D metabolites, total 25(OH)D and total 1,25(OH) 2 D levels in control, T2DM, CAD and T2DM with CAD and to find the association between vitamin D deficiency and coronary artery diseases in diabetes.

Results
Four groups (Control, T2DM, CAD and T2DM_CAD) of total 187 subjects were included in this present study. The serum level of total 25(OH)D was well classified in the previous literature 19,20 . Anthropometric and clinical characteristics of all subjects based on 25(OH)D categories were presented in Table 1. There were no significant differences in age, body mass index (BMI), systolic and diastolic blood pressure (BP), creatinine, estimated glomerular filtration rate (eGFR), uric acid, creatinine kinase myoglobin isoenzyme (CK-MB), and apolipoprotein B (Apo-B). Glycated hemoglobin (HbA1c) levels were increased significantly (p < 0.05) in patients having 20-30 ng/ml and < 20 ng/ml at 25(OH)D levels as compared with subjects having > 30 ng/ml of 25(OH)D. Similarly, fasting blood sugar (FBS) levels were also increased in patient groups having 25(OH)D levels 20-30 ng/ ml and < 20 ng/ml as compared with patients having > 30 ng/ml. Among all T2DM subjects, 43 subjects had history of hypertension. Among CAD subjects, 23 subjects were diagnosed with acute myocardial infarction (MI) and five were with unstable angina. Among T2DM_CAD subjects, 31 subjects were diagnosed with acute myocardial infraction and 7 were diagnosed with unstable angina.
Distribution of vitamin D 2 and D 3 metabolites in control, T2DM, CAD and T2DM_CAD groups. Distribution of total 25(OH)D levels in all subjects was mentioned in Fig. 2. The median total 25(OH) D concentrations in the total subjects was 32.3 (range 7.9-99.0). According to the previous literature, serum total 25(OH)D concentrations were well categorized as sufficient > 30 ng/ml, insufficient 30-20 ng/ml and deficient < 20 ng/ml 20 . However, no such categories were established for the other metabolites. In present study, cut off points was set based on the median of the individual metabolites in total subjects. Cutoff values for all three 25(OH)D 3 , Vitamin D 2 and total 25(OH)D metabolites were categorized as > 30, 30-20, and < 20 ng/ml. Other vitamin D metabolites like vitamin D 3 (> 20 and < 20 ng/ml), 1,25(OH) 2 D 3 (> 6 and < 6 ng/ml), 25(OH)D 2 (> 4 and < 4 ng /ml), 1,25(OH) 2 D 2 (> 4 and < 4 ng/ml) and 1,25(OH) 2 D (> 11 and < 11 ng/ml) were categorized as stated to find distribution of patients in different groups. In the present study, higher percentage of subjects having optimum or higher level of vitamin D 3 (> 20 ng/ ml), 25(OH)D 3 (> 30 ng/ml) and 1,25(OH) 2 D 3 (> 6 ng/ml) metabolites were observed in control group. However, the percentage of diabetes along with CAD subjects were increased when we considered lower concentration of vitamin D3 (< 20 ng/ml), 25(OH)D 3 (< 20 ng/ml) and 1,25(OH) 2 D 3 (< 6 ng/ml) metabolites. Lower percentage of T2DM, CAD and T2DM with CAD subjects were observed when we considered higher concentration of vitamin D 3 metabolites. However, the percentage has been increased when we consider lower concentration of vitamin D 3 metabolites (Fig. 3A,B and C). Similar changes were also observed for all three vitamin D 2 metabolites (Vitamin D 2 , 25(OH)D 2 , 1, 25(OH) 2 D 2 ) (Fig. 3D,E and F).

Discussion
In the present study, both vitamin D 2 and vitamin D 3 metabolites levels were measured in Indian subjects associated with type 2 diabetes and coronary artery diseases. As per the previous literature, in Indian subjects prevalence of vitamin D deficiency is 70-100%, despite living in the 8.4° to 37.6° north latitude and sufficient sunlight exposure throughout the year in most of the places [9][10][11][12][13][14][15][16][17] . Unlike previous studies, our study found that 83% of control subjects were vitamin D sufficient i.e., more than 30 ng/ml 25(OH)D levels and only 13% subjects were insufficient i.e., below < 20 ng/ml. Maximum percentage of healthy subjects have normal range (30-40 ng/ml) of 25(OH)D i.e., within optimum levels. Our results differ from the other Indian studies where maximum number of people were vitamin D deficient [7][8][9][10][11][12][13][14][15][16][17] . This difference might be due to the use of different kind of detection methods. Previously, radio immune assay (RIA) method was used to find out vitamin D levels from Indian subjects. However, in the present study UPLC/APCI/HRMS method has been used to measure vitamin D metabolites in Indian subjects. Type 2 diabetes and coronary artery diseases are two pathological conditions closely related to each other 21,22 . Severity of the disease complications generally increases when diabetes coexists with coronary artery diseases. Different type of pathological changes like oxidative stress, advanced end glycation products, hyperglycemia and inflammation are involved in cardiovascular complications in diabetes 23 . Researchers reported that vitamin D deficiency is associated with insulin resistance in type 2 diabetic patients 6,24 and coronary artery disease 25 27 . Several other researchers reported that low levels of vitamin D(25(OH)D 3 ) were associated with asymptomatic CAD in high-risk type 2 diabetic patients with elevated urinary albumin excretion rate 28 . However, there is no study yet to find the association of vitamin D deficiency with coronary artery diseases in type 2 diabetes. This is the first cross sectional study to compare six vitamin D metabolites and total 25(OH)D, total 1,25(OH) 2 D in control, T2DM, CAD and T2DM with CAD patients from south part of India.
The present study shows that 83% and 73% of control subjects have sufficient total 25(OH)D (> 30 ng/ml) and total 1,25(OH) 2 D (> 11 ng/ml) levels, respectively. However, the percentage of diabetes and diabetes with CAD subjects were increased when we considered insufficient 25(OH)D and 1,25(OH)D 2 levels. All the vitamin D metabolites were significantly lowered in T2DM with CAD subjects when compared with control and T2DM  subjects. However, only 25(OH)D 3 and total 25(OH)D levels were significantly decreased in the T2DM subjects as compared with control subjects. The study also indicated that T2DM patients with total 1,25(OH) 2 D deficiency are more prone to coronary artery diseases. It was observed that deficiency of vitamin D metabolites were associated with coronary artery diseases among diabetic patients. However, this deficiency is much less for T2DM or CAD alone. Significant negative correlation was observed between total 25(OH)D and HbA1c or FBS but not with BMI. Similar to our study, other studies also showed that total 25(OH)D levels were negatively correlated with HbA1C and blood glucose 29 but not with the BMI 30 . Interestingly, our data found a significant correlation between vitamin D metabolites and the prevalence of CAD among T2DM patients.
To further confirm the causal relationship between lower levels of plasma vitamin D metabolites and T2DM or T2DM with CAD, multivariate logistic regression was performed. The analysis of our results revealed that while lower serum 25(OH)D predicts type 2 diabetes, lower serum 1,25(OH) 2 D predicts coronary artery diseases in type 2 diabetes. Although, we have not able to investigate the mechanism behind the role of vitamin D on diabetes and CAD, several previous literatures shed light regarding the molecular effect of Vitamin D. 1,25(OH)2D stimulates the expression of insulin receptors, which in turn will affect insulin sensitivity 31 . 1,25(OH)2D may also enhance insulin sensitivity by activating peroxisome proliferator-activated receptor delta (PPAR-δ ), which is a transcription factor regulates the metabolism of fatty acids in skeletal muscle and adipose tissue 32 . Vitamin D insufficiency has been associated with increased fat infiltration in skeletal muscle and contributes to decrease in insulin sensitivity 33 . Similar to diabetes, lower concentration of 1,25(OH)2D is associated with athreosclerosis and coronary artery disease through affecting vitamin D receptor signaling. Impairment of VDR signaling accelerates foam cell formation in diabetics 26 .
Unlike previous studies, the strength of the present study is to measure six vitamin D metabolites by novel and specific mass spectrometry assay in four different groups like control, T2DM, CAD, and T2DM with CAD.

Limitations
Several limitations of our study should be noted. (1) It was a single-center one-point time study from South part of India. (2) We have small sample size in each group. (3) Although we adjusted for potential confounders, we cannot eliminate the possibility of residual confounding factors influencing our results. (4) We have not measured parathyroid hormone and fibroblast growth factor (FGF-23) levels in blood. (5) This study is an observational study and does not address causality.

Conclusion
In conclusion, vitamin D metabolites were decreased in T2DM with CAD subjects as compared with control, T2DM, and CAD subjects. Vitamin D metabolites like 25(OH)D can be used to predict T2DM while 1,25(OH) 2 D can be used to predict CAD in T2DM. Further studies are required to show that vitamin D supplementation may be beneficial to reduce coronary artery diseases among diabetic patients.
Inclusion and Exclusion Criteria. Group 1 Control subjects had no prior history of T2DM, hypertension, coronary artery diseases or any other cardiovascular diseases, and were not taking medication for any chronic medical condition. Fasting blood glucose, HbA1c and blood chemistry were normal. Group 2 (T2DM) included subjects with HbA1c levels ≥ 6.5% as per American Diabetes Association (ADA) guidelines with proven history of T2DM but no other complications. Group 3 (CAD) included subjects with narrowing or blockage of one or more epicardial coronary artery with greater than 25% stenosis shown in coronary angiography and diagnosed by cardiologists 3 . This group had no prior history of T2DM. Group 4 (T2DM_CAD) included subjects with coronary artery disease as defined for group 3 but patient had HbA1c levels ≥ 6.5% and prior history of T2DM 3 . Exclusion criteria defined for this study were clinical or laboratory evidence of liver failure, renal failure (plasma creatinine levels > 1.5 mg/dl), type 1 diabetes, cancer, thyroid disease and pregnancy; and subjects on vitamin D treatment therapy were also excluded in this study. The study conforms to the principles outlined in the Declaration of Helsinki and was approved by the Mediciti Ethics Committee (Institutional Ethics Committee), Hyderabad. All patients were given detailed information of the study. Informed consent was obtained from all the subjects.

Sample preparation and analytical method for estimation of vitamin D metabolites. Serum
vitamin D metabolites concentrations were determined by liquid chromatography-Orbitrap mass spectrometry (UPLC/APCI/HRMS). Stock solutions of vitamin D 3 and vitamin D 2 and its respective metabolites (25(OH) D 3 , 1,25(OH) 2 D 3 , 25(OH)D 2 and 1,25(OH) 2 D 2 ) were prepared in ethanol; Calibration curve (5-200 ng/ml) and quality control sample was prepared from stock solution using methanol; Similarly the dihydrotachysterol 50 ng/ ml was used as an internal standard (IS). 100 μ l of serum (test sample) was added with 10 ul of IS solution followed by 1 ml of hexane:heptane:acetone in the ratio of 45:40:15. After vertex it properly, the solution was incubated in a shaker for 10 min. The supernatant was separated after centrifugation at 6000 rpm at 4 °C. Supernatant was collected and evaporated using scan vac speed vacuum concentrator. After re-dissolving the concentrate sample in 100 μ l of methanol, 10 ul of dissolved sample was used for analysis.
Separation of metabolites was achieved using waters Xselect CSH phenyl hexyl column (150 mm*4.6 mm ID; particle size 3.5 um) with gradient mode. Both mobile phase i.e., mobile phase A (10 mM of ammonium formate in methanol) and mobile phase B (ACN: Acetone: IPA (5:4:1)) were set as follows: (T min /% proportion of solvent (B): 0 /5, 0-5 /95, 5-6 /5, and 6-8 /5. The flow rate of the mobile phase was 1.00 ml/min while the column temperature and the injection volume was set as 25 °C and 10 μ L, respectively. Mass spectrometric detection was carried out by Orbitrap mass analyzer (Exactive Thermo Scientific, Germany) equipped with an atmospheric pressure chemical ionization (APCI) source. The data acquisition was under the control of Xcalibur software. The typical operating source conditions for MS scan in positive ion APCI mode were optimized as follows sheath gas flow rate 65; Aux  Table S1).

Statistical analysis.
Normally distributed variables were summarized as mean and standard deviation, while non-normally distributed variables were expressed as median and interquartile ranges. One-way analysis of variance (ANOVA) was used to determine whether there are any significant differences between the means of three independent (unrelated) groups of 25(OH)D. Anthropometric and clinical characteristics were presented by 25(OH)D categories (> 30, 20-30 and < 20 ng/ml). Kruskal-Wallis test was used for comparing non-normally distributed variables with post-hoc Dunn's multiple comparison test. One-way ANOVA with Bonferroni test was used for normally distributed variable. Kruskal-Wallis test was used to compare vitamin D metabolites levels in control, CAD and T2DM_CAD groups and control, T2DM and T2DM_CAD groups. Spearman rank order correlation was used to find correlation among 25(OH) D, HbA1c, FBS and BMI. Total 25(OH)D [sum of 25(OH)D+ 25(OH)D 3 ] and total 1,25(OH) 2 D [sum of 1,25(OH) 2 D 2 + 1,25(OH) 2 D 3 ] were evaluated for overall comparison. Multivariate logistic regression analysis was also performed to test the association of the serum 25(OH)D level with the presence of T2DM (i.e. the dependent variable). Two forced-entry multivariate regression models were performed (i.e. unadjusted and adjusted). The dependent variable was involved as a compound end point: patients with T2DM or CAD or both complications (coded as 1) and patients with neither T2DM nor CAD (coded as 0 i.e. control). Univariate analysis was used to identify covariates as potential confounders on the basis of their significance (p < 0.05). ORs with 95% CIs were used to report the results. Distribution data were presented with Origin 6.0. Box and whisker plots were made with GraphPad Prism version 5.01. Statistical analysis was performed with SAS ® V.9.4 software.