Weight loss improves the adipogenic capacity of human preadipocytes and modulates their secretory profile

Abstract

Calorie restriction-induced weight loss is accompanied by profound changes in adipose tissue characteristics. To determine the effect of weight loss on differentiation of preadipocytes and secretory capacity of in vitro differentiated adipocytes, we established cultures of these cells from paired subcutaneous adipose tissue biopsies obtained before and at the end of weight-reducing dietary intervention (DI) in 23 obese women. Based on lipid accumulation and the expression of differentiation markers, in vitro adipogenesis increased after weight loss and it was accompanied by enhanced expression of genes involved in de novo lipogenesis. This effect of weight loss was not driven by changes of peroxisome proliferator-activated receptor γ sensitivity to rosiglitazone. Weight loss also enhanced the expression of adiponectin and leptin while reducing that of monocyte chemoattractant protein 1 and interleukin-8 by cultured adipocytes. Thus, the weight-reducing (DI) increased adipogenic capacity of preadipocytes and shifted their secretion toward lower inflammatory profile. Reprogramming of preadipocytes could represent an adaptation to weight loss leading to partial restoration of preobese adipose tissue traits and thus contribute to the improvement of metabolic status. However, enhanced adipogenesis could also contribute to the unwanted weight regain after initial weight loss.

Figures

FIG. 1.

Weight loss improves in vitro adipogenesis. Cells were differentiated for 12 days, and then accumulation of lipids, protein content, or gene expression was analyzed. A: Effect of weight loss on lipid accumulation. Representative images of adipocytes from one donor before and after DI stained with ORO and quantification of neutral lipid accumulation expressed as percent of stock ORO (n = 22). B: Effect of weight loss on gene expression. mRNA expression (arbitrary units [AU]) in adipocytes normalized to GUSB expression (n = 15). C: Effect of weight loss on protein content. Ratio between total protein content in adipocytes vs. preadipocytes (n = 22). Data are means ± SE; *P < 0.05, **P < 0.01, ***P < 0.001. D: Linear regression between ORO accumulation and protein content in adipocytes (A) vs. preadipocytes (P) at baseline. E: Linear regression between relative Δprotein (adipocytes vs. protein) and ΔmRNA expression of aP2 and PPARγ.

FIG. 2.

Weight loss does not enhance the sensitivity to PPARγ ligand rosiglitazone (Rosi) but is connected with the suppression of RUNX2 expression. A and B: Cells were differentiated for 3 days in the presence or absence of 1 mmol/L rosiglitazone. Cells kept in serum-free medium supplemented with transferrin and insulin that did not undergo adipogenesis were used as controls. A: Relative mRNA levels of selected genes were detected by quantitative RT-PCR (n = 6). □, Baseline; ■, DI. Data are means ± SE; *P < 0.05. B: The ratio between expression of selected genes in cells differentiated in the presence and absence of rosiglitazone was calculated in cells derived before and after the DI. C: Cells were differentiated for 6 days in the presence or absence of 1 mmol/L rosiglitazone (n = 4). After staining with BODIPY, the 45 microscopy images encompassing an average of 2,700 cells were analyzed and numbers of BODIPY+ cells were counted. Chart represents the percentage of BODIPY+ cells within analyzed populations (each on average from 2,700 cells). Data are means ± SE; *P < 0.05. AU, arbitrary units; pread, preadipocytes.

FIG. 3.

Weight loss alters the expression and secretion of cytokines in adipocytes differentiated in vitro. Cells were differentiated for 12 days, and conditioned media and cells were collected after 24 h of incubation in freshly added media for analysis of cytokine expression and secretion. A: Effect of weight loss on cytokine gene expression. Quantitative RT-PCR analysis of mRNA for selected cytokines, normalized to GUSB expression (n = 15). B: Effect of weight loss on cytokine secretion. Fold change over the basal values for MCP1, IL-6, and adiponectin in conditioned media measured by ELISA, normalized to protein content (n = 22). C: Effect of weight loss on adiponectin isoform secretion. Quantification of adiponectin isoforms by native polyacrylamide gel electrophoresis and Western blot analysis (n = 21). Data are means ± SE; *P < 0.05, **P < 0.01, ***P < 0.001. HMW, high molecular weight; LMW, low molecular weight, MMW, medium molecular weight.