Is Vitamin D Deficiency Related to Accumulation of Advanced Glycation End Products, Markers of Inflammation, and Oxidative Stress in Diabetic Subjects?

Figure 4

Multivariate analysis data from OPLS-DA model comparing 25(OH)D deficient diabetic patients (25(OH)D <20 ng/mL) with those presenting sufficient levels (25(OH)D >30 ng/mL). (a) Score scatter plot of 25(OH)D deficient diabetic patients (DM-D, red circles) and those presenting sufficient 25(OH)D levels (DM-S, yellow circles). Scores are orthogonal (=completely independent from each other), representing new variables summarizing the input of all determined variables (herein gender, presence or absence of comorbidities, age, SAF, and biochemical variables) so that one score vector corresponds to one subject, having its own score vector. Observations situated far outside Hotelling’s T2 tolerance ellipse are outliers. Model reveals separation of 25(OH)D deficient and sufficient diabetic subjects (separation in direction of -axis). Separation in direction of -axis represents within group variability. (b) Loading scatter plot of 25(OH)D deficient diabetic subjects and those presenting sufficient 25(OH)D levels. Dummy variables (blue circles) characterize the respective 2 groups categorized according to 25(OH)D levels, deficient group at left, and sufficient one at right side of the plot. Vitamin D₃ (25D3) adjacent to dummy variable representing 25(OH)D sufficient group represents the most significant component with discriminatory power determining the separation between the groups; being situated in the vicinity of vitamin D sufficient group presenting dummy variable it indicates that it is higher in this group. 25(OH)D deficient subjects also tend to present higher CML, total protein and sVAP-1 levels (positioned in vicinity of respective dummy variable), and lower AGE-associated fluorescence of plasma, AOPPs and grip strength (far opposite, right to respective dummy). Variables positioned near to intersect and on -axis are similar in 25(OH)D deficient and sufficient groups and thus do not contribute to between-group separation. (c) Plot of variables of importance contributing to between-group separation among 25(OH)D deficient controls and those presenting sufficient 25(OH)D levels. Plot of variables importance for the projection (VIP) summarizes the importance of the variables both to explain and to correlate with dummy variables (in (a), and (b)). VIP values >1 indicate “important” variables, <0.5 “unimportant” variables, in the “grey interval” (0.5-to-1) the importance depends on the sample size. This plot confirms the OPLS-DA loadings scatter plot (b), showing that the variables adjacent to the origin in the former plot do not contribute to between-group separation significantly. Abbreviations used in Figures 3(b) and 3(c): 25D3: 25(OH) vitamin D₃; TP: plasma total protein concentration; sVAP: soluble vascular receptor adhesion protein-1; CML: N^ε-carboxymethyllysine; AGEfl: advanced glycation end products associated fluorescence of plasma; sRAGE: soluble receptor for advanced glycation end products; F: female; M: male; BMI: body mass index; HbA1c: glycated hemoglobin A1c; DM1: type 1 diabetes mellitus; DM2: type 2 diabetes mellitus; eGFR: estimated glomerular filtration rate; CHD: coronary heart disease; N: no, absent; Y: yes, present; GFR: estimated glomerular filtration rate; grips: grip strength; SAF: skin autofluorescence; AOPPs: advanced oxidation protein products; hsCRP: high sensitive C-reactive protein; HT: hypertension; PAD: peripheral artery disease.

(a)

(b)

(c)