Insulin and metformin regulate circulating and adipose tissue chemerin

Abstract

Objective: To assess chemerin levels and regulation in sera and adipose tissue from women with polycystic ovary syndrome (PCOS) and matched control subjects.

Research design and methods: Real-time RT-PCR and Western blotting were used to assess mRNA and protein expression of chemerin. Serum chemerin was measured by enzyme-linked immunosorbent assay. We investigated the in vivo effects of insulin on serum chemerin levels via a prolonged insulin-glucose infusion. Ex vivo effects of insulin, metformin, and steroid hormones on adipose tissue chemerin protein production and secretion into conditioned media were assessed by Western blotting and enzyme-linked immunosorbent assay, respectively.

Results: Serum chemerin, subcutaneous, and omental adipose tissue chemerin were significantly higher in women with PCOS (n = 14; P < 0.05, P < 0.01). Hyperinsulinemic induction in human subjects significantly increased serum chemerin levels (n = 6; P < 0.05, P < 0.01). In adipose tissue explants, insulin significantly increased (n = 6; P < 0.05, P < 0.01) whereas metformin significantly decreased (n = 6; P < 0.05, P < 0.01) chemerin protein production and secretion into conditioned media, respectively. After 6 months of metformin treatment, there was a significant decrease in serum chemerin (n = 21; P < 0.01). Importantly, changes in homeostasis model assessment-insulin resistance were predictive of changes in serum chemerin (P = 0.046).

Conclusions: Serum and adipose tissue chemerin levels are increased in women with PCOS and are upregulated by insulin. Metformin treatment decreases serum chemerin in these women.

FIG. 1.

A: Chemerin mRNA expression relative to β-actin was significantly increased in human subcutaneous (Sc) and omental (Om) adipose tissue depots when comparing PCOS women (n = 14) with normal control subjects (n = 14), using real-time RT-PCR. Data are expressed as percent difference of median of human subcutaneous adipose tissue of normal control subjects. Each experiment was carried out in three replicates. Group comparison by Kruskal-Wallis ANOVA and post hoc Dunn's test; *P < 0.05, **P < 0.01. B: Densitometric analysis of chemerin immune complexes having normalized to β-actin revealed that protein levels of chemerin were significantly increased in human subcutaneous and omental adipose tissue depots, respectively, when comparing all women with PCOS with all normal control subjects. Data are expressed as percent difference of median of normal control subjects. Each experiment was carried out in three replicates. Group comparison by Kruskal-Wallis ANOVA (post hoc analysis: Dunn's test). *P < 0.05, **P < 0.01; PSL, phosphostimulated light units.

FIG. 2.

Mean concentrations of chemerin in nanogram per milliliter in all subjects, before and after insulin infusion. Insulin infusion resulted in elevation of fasting insulinemia from 78.1 ± 12.0 pmol/l to 294.6 ± 31.0 pmol/l. Insulin levels remained elevated until the end of the prolonged insulin-glucose infusion (366.0 ± 37.0 pmol/l). Data are means ± SD. Group comparison by Student's t test. *P < 0.05, **P < 0.01.

FIG. 3.

A: Dose-dependent effects of insulin (10⁻¹¹ M, 10⁻⁹ M, 10⁻⁷ M) in the presence of 5 mmol/l d-glucose on chemerin net protein production in control human omental adipose tissue explants at 24 h were assessed by Western blotting. Western blot analysis of protein extracts from omental adipose tissue demonstrate that the antibody against chemerin and the antibody against β-actin recognized bands with apparent molecular weights of 18 kDa and 45 kDa, respectively (Fig. 3A, inserts). Densitometric analysis of chemerin immune complexes having normalized to β-actin, respectively, revealed that protein levels of chemerin were significantly increased by insulin (10⁻⁹ M, 10⁻⁷ M) in control human omental adipose tissue explants. Data are expressed as percent difference of median of basal. Each experiment was carried out with six different samples from six different subjects in three replicates. Group comparison by Friedman's ANOVA and post hoc Dunn's test. *P < 0.05, **P < 0.01. B: Dose-dependent effects of insulin (10⁻¹¹ M, 10⁻⁹ M, 10⁻⁷ M) in the presence of 5 mmol/l d-glucose on chemerin secretion into conditioned media from control human omental adipose tissue after 24 h were measured by ELISA. Chemerin secretion was significantly increased (by 10⁻⁹ M, 10⁻⁷ M) from control human omental adipose tissue explants. Data are expressed as percent difference of median of basal. Each experiment was carried out with six different samples from six different subjects in three replicates. Group comparison by Friedman's ANOVA and post hoc Dunn's test. *P < 0.05, **P < 0.01.

FIG. 4.

A: Dose-dependent effects of metformin (0.01, 0.1, and 2.00 μg/ml) in the presence of 5 mmol/l d-glucose on chemerin net protein production in control human omental adipose tissue explants at 24 h were assessed by Western blotting. Western blot analysis of protein extracts from omental adipose tissue demonstrate that the antibody against chemerin and the antibody against β-actin recognized bands with apparent molecular weights of 18 kDa and 45 kDa, respectively (Fig. 4A, inserts). Densitometric analysis of chemerin immune complexes having normalized to β-actin, respectively, revealed that protein levels of chemerin were significantly decreased by metformin (0.01, 0.1, and 2.00 μg/ml) in control human omental adipose tissue explants. Data are expressed as percent difference of median of basal. Each experiment was carried out with six different samples from six different subjects in three replicates. Group comparison by Friedman's ANOVA and post hoc Dunn's test. *P < 0.05, **P < 0.01. B: Dose-dependent effects of insulin (0.01, 0.1, and 2.00 μg/ml) in the presence of 5 mmol/l d-glucose on chemerin secretion into conditioned media from control human omental adipose tissue after 24 h were measured by ELISA. Chemerin secretion was significantly decreased (by 0.01, 0.1, and 2.00 μg/ml) from control human omental adipose tissue explants. Data are expressed as percent difference of median of basal. Each experiment was carried out with six different samples from six different subjects in three replicates. Group comparison by Friedman's ANOVA and post hoc Dunn's test. *P < 0.05, **P < 0.01.