Pediatric obesity and vitamin D deficiency: a proteomic approach identifies multimeric adiponectin as a key link between these conditions

Abstract

Key circulating molecules that link vitamin D (VD) to pediatric obesity and its co-morbidities remain unclear. Using a proteomic approach, our objective was to identify key molecules in obese children dichotomized according to 25OH-vitamin D (25OHD) levels. A total of 42 obese children (M/F = 18/24) were divided according to their 25OHD3 levels into 25OHD3 deficient (VDD; n = 18; 25OHD<15 ng/ml) or normal subjects (NVD; n = 24; >30 ng/ml). Plasma proteomic analyses by two dimensional (2D)-electrophoresis were performed at baseline in all subjects. VDD subjects underwent a 12mo treatment with 3000 IU vitamin D3 once a week to confirm the proteomic analyses. The proteomic analyses identified 53 "spots" that differed between VDD and NVD (p<0.05), amongst which adiponectin was identified. Adiponectin was selected for confirmational studies due to its tight association with obesity and diabetes mellitus. Western Immunoblot (WIB) analyses of 2D-gels demonstrated a downregulation of adiponectin in VDD subjects, which was confirmed in the plasma from VDD with respect to NVD subjects (p<0.035) and increased following 12mo vitamin D3 supplementation in VDD subjects (p<0.02). High molecular weight (HMW) adiponectin, a surrogate indicator of insulin sensitivity, was significantly lower in VDD subjects (p<0.02) and improved with vitamin D3 supplementation (p<0.042). A direct effect in vitro of 1α,25-(OH)2D3 on adipocyte adiponectin synthesis was demonstrated, with adiponectin and its multimeric forms upregulated, even at low pharmacological doses (10(-9) M) of 1α,25-(OH)2D3. This upregulation was paralleled by the adiponectin interactive protein, DsbA-L, suggesting that the VD regulation of adiponectin involves post-transciptional events. Using a proteomic approach, multimeric adiponectin has been identified as a key plasma protein that links VDD to pediatric obesity.

Figure 1. Proteomic evaluation predicts adiponectin isoforms as being differentially expressed between VDD and NVD.

A 2D-electrophoretic analysis was performed in duplicate for the 42 subjects using IPG3-10, with proteins detected by Sypro Ruby staining. Spot/s predicted to be adiponectin are indicated by the PDQuest identification number (SSP3050). Supportive evidence for the prediction was given by performing a WIB of 2D-electrohoretic gels using anti-adiponectin antibody. Representative gels are shown.

Figure 2. WIB of proteomic analyses and plasma samples confirms that total adiponectin is decreased in VD deficient pediatric obese subjects (VDD) when compared to pediatric obese with normal VD levels (NVD).

A. A WIB of 2D-electrophoretic analyses was performed in VDD (n = 10) and NVD (n = 10) subjects using anti-adiponectin antibody. B. A WIB analysis under reduced conditions of total adiponectin in the plasma of representative VDD (<15 ng/ml) and NVD (>30 ng/ml) subjects.

Figure 3. The multimeric forms of adiponectin are reduced, in particular the HMW form, in VD deficient pediatric obese subjects.

A WIB analysis under non-reduced conditions and a quantitative densitometric analysis of the multimeric forms of adiponectin (HMW, MMW, LMW) in the plasma of representative VDD (<15 ng/ml; n = 18) and NVD (>30 ng/ml; n = 24) subjects. Densitometric results were normalized to plasma protein concentrations.

Figure 4. Total and the HMW and MMW multimeric forms of adiponectin increase in VD deficient pediatric obese subjects following cholecalciferol supplementation for 12mo.

A. A WIB analysis under reduced conditions and a quantitative densitometric analysis of total adiponectin in the plasma of representative VDD (<15 ng/ml) and NVD (>30 ng/ml) subjects. B. A WIB analysis under non-reduced conditions and a quantitative densitometric analysis of the multimeric forms of adiponectin in the plasma of VDD (<15 ng/ml; n = 18) and NVD (>30 ng/ml; n = 24) subjects.

Figure 5. Total adiponectin secretion increases in 3T3-L1 adipocytes treated with 1α,25-(OH)2D3.

3T3-L1 adipocytes, generated using a standard differentiation protocol, at day 10 were treated with increasing concentrations of 1α,25-(OH)2D3 (10⁻⁹ to 10⁻⁷ M) in SFM or SFM with vehicle for 48 h. The CM at 7, 24 and 48 h from the same treatment was analyzed by WIB under reduced condition and analyzed densitometrically for the synthesis of adiponectin. Results were normalized to α-tubulin and are presented as fold-increase with respect to the 7 h SFM sample (n = 4).

Figure 6. The secretion of the adiponectin multimeric forms, in particular the LMW form, in parallel with the ER-chaperon DsbA-L, increases in 3T3-L1 adipocytes treated with 1α,25-(OH)2D3.

3T3-L1 adipocytes, generated using a standard differentiation protocol, at day 10 were treated with increasing concentrations of 1α,25-(OH)2D3 (10-9 to 10-7M) in SFM or SFM with vehicle for 48 h. The CM at 7, 24 and 48 h from the same treatment and CL at 48 h, was analyzed by WIB under non-reduced or reduced conditions for the synthesis of adiponectin multimeric forms or DsbA-L with α-tubulin, respectively.