Dexamethasone Inhibits White Adipose Tissue Browning

Abstract

White adipose tissue (WAT) regulates energy balance through energy storage, adipokines secretion and the thermogenesis process. Beige adipocytes are responsible for WAT thermogenesis. They are generated by adipogenesis or transdifferentiation during cold or β3-adrenergic agonist stimulus through a process called browning. Browning has gained significant interest for to its preventive effect on obesity. Glucocorticoids (GCs) have several functions in WAT biology; however, their role in beige adipocyte generation and WAT browning is not fully understood. The aim of our study was to determine the effect of dexamethasone (DXM) on WAT thermogenesis. For this purpose, rats were treated with DXM at room temperature (RT) or cold conditions to determine different thermogenic markers. Furthermore, the effects of DXM on the adipogenic potential of beige precursors and on mature beige adipocytes were evaluated in vitro. Our results showed that DXM decreased UCP-1 mRNA and protein levels, mainly after cold exposure. In vitro studies showed that DXM decreased the expression of a beige precursor marker (Ebf2), affecting their ability to differentiate into beige adipocytes, and inhibited the thermogenic response of mature beige adipocytes (Ucp-1, Dio2 and Pgc1α gene expressions and mitochondrial respiration). Overall, our data strongly suggest that DXM can inhibit the thermogenic program of both retroperitoneal and inguinal WAT depots, an effect that could be exerted, at least partially, by inhibiting de novo cell generation and the thermogenic response in beige adipocytes.

Keywords: beige adipocytes; glucocorticoids; thermogenesis.

Figure 1

In vivo effects of DXM treatment. (A) Experimental design for in vivo DXM effects in male rats housed at RT (22 °C) or cold (4 °C). (B) Body weight change and (C) mean of caloric intake during DXM treatment. Percentage masses of (D) RPAT and (H) IAT. Representative images from (E) RPAT and (I) IAT depots. Adipocyte size (diameter) from (F) RPAT and (J) IAT adipocytes. Representative images from haematoxylin-eosin stained sections from (G) RPAT and (K) IAT. Magnification ×400. Scale bar: 50 µm. n = 3–4 samples per group. (L) Percentage mass and (M) representative images from BAT, respectively. In all cases, two-way ANOVA was performed for factors (Cold or DXM) and interaction analysis (Cold × DXM) followed by Tukey’s post-test. Results for factors and interaction are shown in each graph. * p < 0.05, *** p < 0.001 and **** p < 0.0001 vs. CTR-RT. ⁺⁺ p < 0.01 vs. CTR-C. n = 10–12 rats per group. Data shown are mean ± SEM.

Figure 2

Effect of DXM on cold-induced browning in RPAT and IAT. Ucp-1 mRNA levels from (A) RPAT and (C) IAT. UCP1 protein levels from (B) RPAT and (D) IAT (representative immunoblots and densitometry analysis). β-ACTIN was used as loading control. Dio2 and Arβ3 mRNA levels from RPAT (E,F) and from IAT (G,H), respectively. Two-way ANOVA was performed for factors (Cold or DXM) and interaction analysis (Cold × DXM) followed by Tukey’s post-test. *** p < 0.001 and **** p < 0.0001 vs. CTR-RT. ⁺⁺⁺ p < 0.001 and ⁺⁺⁺⁺ p < 0.0001 vs. CTR-C. Results for factors and interaction are shown in each graph. Western blot data were analysed using Student t-test. ** p < 0.01 vs. CTR-RT. For qPCR quantification, n = 5 samples per group were analysed in duplicate. For Western blot analysis, n = 3–4 samples per group. Data shown are mean ± SEM.

Figure 3

Mitochondrial content in RPAT and IAT from DXM-treated rats. mRNA levels of Pgc1α in (A) RPAT and (C) IAT. DNA mitochondrial content from (B) RPAT and (D) IAT. Two-way ANOVA was performed for factors (Cold or DXM) and interaction analysis (Cold × DXM) followed by Tukey’s post-test. Results for factors and interaction are shown in each graph. Mitochondrial content was analysed using Student t-test. * p < 0.05 and **** p < 0.0001 vs. CTR-RT. ⁺ p < 0.05 and ⁺⁺⁺ p < 0.001 vs. CTR-C. For qPCR quantification, n = 5 samples per group were analysed in duplicate. For mitochondrial content, n = 4 samples per group were analysed in duplicate. Data shown are mean ± SEM.

Figure 4

Gene expression of beige precursor markers in SFV cells treated with DXM. mRNA levels of Pdgfrα, Ebf2 and Zpf423 in SVF cells from (A–C) RPAT and (D–F) IAT. Two-way ANOVA was performed for factors (Cold or DXM) and interaction analysis (Cold × DXM) followed by Tukey’s post-test. Results for factors and interaction are shown in each graph. ** p < 0.01, *** p < 0.001 and **** p < 0.0001 vs. CTR-RT. ⁺⁺⁺ p < 0.001 and ⁺⁺⁺⁺ p < 0.0001 vs. CTR-C. n = 5 samples per group were analysed in duplicate. Data shown are mean ± SEM.

Figure 5

DXM effect on beige precursor expression in cultured APCs. (A) Experimental design for in vitro DXM effects on APCs. Briefly, cells were incubated in presence or absence of DXM 0.25 μM (DXM and CTR, respectively) for 48 h. Then, DXM was removed and cells were harvested immediately (D0). BDM: Beige differentiation mix; BMM: Beige Maintenance Medium. mRNA levels of Pdgfrα, Ebf2 and Zpf423 were quantified in cultured APCs from (B–D) RPAT and (E–G) IAT. n = 5 independent experiments were analysed in duplicate. Data was analysed using Student t-test. * p < 0.05 and ** p < 0.01 vs. CTR. (H) Additionally, cells were induced to differentiate to beige adipocytes and processed at differentiation day 8 (D8), after stimulation or not with Fsk in the last 4 h (CTR-Fsk and DXM-Fsk, CTR and DXM, respectively). mRNA levels of Ucp-1, Pgc1α and Dio2 from (I–K) RPAT cells and (L–N) IAT cells. Two-way ANOVA was performed for factors (Fsk or DXM) and interaction analysis (Fsk × DXM) followed by Tukey’s post-test. Results for factors and interaction are shown in each graph. * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001 vs. CTR. ⁺ p < 0.05 vs. CTR-Fsk. ^#### p < 0.0001 vs. DXM. n = 5–6 independent experiments were analysed in duplicate. Data shown are mean ± SEM.

Figure 6

DXM effect on thermogenic program in differentiated beige adipocytes. (A) Experimental design for in vitro DXM effects on beige mature adipocytes. Briefly, cells were incubated or not with 0.25 µM DXM the last 48 h of culture and stimulated or not with Fsk the last 4 h (CTR-Fsk and DXM-Fsk, CTR and DXM, respectively). mRNA levels of Ucp-1, Pgc1α and Dio2 from (B–D) RPAT cells and (E–G) IAT cells. Two-way ANOVA was performed for factors (Fsk or DXM) and interaction analysis (Fsk × DXM) followed by Tukey’s post-test. Results for factors and interaction are shown in each graph. **** p < 0.0001 vs. CTR. ⁺ p < 0.05 and ⁺⁺⁺⁺ p < 0.0001 vs. CTR-Fsk. ^# p < 0.05 and ^## p < 0.01 vs. DXM. n = 6 independent experiments were analysed in duplicate. Data shown are mean ± SEM.

Figure 7

DXM effect on mitochondrial respiration in 3T3-L1 beige adipocytes. (A) Experimental design for in vitro DXM effects on 3T3-L1 beige adipocytes. Briefly, cells were incubated or not with 0.25 µM DXM the last 48 h of culture (CTR and DXM, respectively). (B) mRNA level of Ucp-1. (C) Oxygen consumption rate (OCR) profile of CTR and DXM 3T3-L1 beige adipocytes. (D) Basal respiration, (E) proton leak, (F) maximal respiration and (G) ATP production parameters from CTR and DXM 3T3-L1 adipocytes. Student t-test was used for statistical analysis. * p < 0.05 vs. CTR. n = 3 independent experiments in duplicate. Data shown are mean ± SEM.