Glucocorticoid modulation of insulin signaling in human subcutaneous adipose tissue

Abstract

Context: Glucocorticoid (GC) excess is characterized by central obesity, insulin resistance, and in some cases, type 2 diabetes. However, the impact of GC upon insulin signaling in human adipose tissue has not been fully explored.

Objective: We have examined the effect of GC upon insulin signaling in both human sc primary preadipocyte cultures and a novel human immortalized sc adipocyte cell line (Chub-S7) and contrasted this with observations in primary cultures of human skeletal muscle.

Design and setting: This is an in vitro study characterizing the impact of GC upon insulin signaling in human tissues.

Patients: Biopsy specimens were from healthy volunteers who gave their full and informed written consent.

Interventions: Combinations of treatments, including GC, RU38486, and wortmannin, were used.

Main outcome measures: Insulin signaling cascade gene and protein expression and insulin-stimulated glucose uptake were determined.

Results: In human adipocytes, pretreatment with GC induced a dose-dependent [1.0 (control); 1.2 +/- 0.1 (50 nm); 2.2 +/- 0.2 (250 nm), P < 0.01 vs. control; 3.4 +/- 0.2 (1000 nm), P < 0.001 vs. control] and time-dependent [1.0 (1 h); 3.2 +/- 2.0 (6 h); 9.1 +/- 5.9 (24 h), P < 0.05 vs. 1 h; 4.5 +/- 2.2 (48 h)] increase in insulin-stimulated protein kinase B/akt phosphorylation. In addition, whereas insulin receptor substrate (IRS)-1 protein expression did not change, IRS-1 tyrosine phosphorylation increased. Furthermore, GC induced IRS-2 mRNA expression (2.8-fold; P < 0.05) and increased insulin-stimulated glucose uptake [1.0 (control) 1.8 +/- 0.1 (insulin) vs. 2.8 +/- 0.2 (insulin + GC); P < 0.05]. In contrast, in primary cultures of human muscle, GC decreased insulin-stimulated glucose uptake [1.0 (control) 1.9 +/- 0.2 (insulin) vs. GC 1.3 +/- 0.1 (insulin + GC); P < 0.05].

Conclusions: We have demonstrated tissue-specific regulation of insulin signaling by GC. Within sc adipose tissue, GCs augment insulin signaling, yet in muscle GCs cause insulin resistance. We propose that enhanced insulin action in adipose tissue increases adipocyte differentiation, thereby contributing to GC-induced obesity.

Figure 1

Chub-S7 cells and primary cultures of human sc preadipocytes, differentiate in chemical defined media (14 days) and acquire a mature adipocyte phenotype with clearly visible lipid droplets.

Figure 2

Insulin stimulation (15 minutes) causes a dose dependent increase in PKB/akt phosphorylation in differentiated chub-S7 cells (A). Dexamethasone pre-treatment (1μM, 48hrs), does not change IRS1 (B) or PKB/akt protein expression (C), but enhances insulin stimulated phosphorylation (B and C). (C=control, D=dexamethasone, representative Western blots are shown with the quantification relative to β-actin of n=3-5 experiments shown below, * p<0.05 vs. control)

Figure 3

Dexamethasone (D) pre-treatment (1μM, 48h) increases insulin stimulated (0.5ng/mL, 15 minutes) PKB/akt phosphorylation in both undifferentiated (A) and differentiated chub-S7 cells (B) and in primary cultures of differentiated subcutaneous human pre-adipocytes (C). In chub-S7 cells, these effects are blocked by co-incubation with the GR antagonist, RU38486 (R, 10μM) or the PI3 kinase inhibitor, wortmannin (W, 100nM)). Data are presented as the mean±se of n=3-5 experiments and quantified relative to β-actin. Representative western blots and light microscopic images are shown inserted (* p<0.05 vs. control (C), † p<0.05 vs. dexamethasone).

Figure 4

Cortisol induces a dose (A) and time (B) dependent increase in insulin stimulated PKB/akt phosphorylation (white bars) without alteration in total PKB/akt expression (black bars) in human differentiated chub-S7 cells. Data are presented as the mean±se of n=5 experiments performed in triplicate. Representative western blots are shown inserted (*p<0.05, ** p<0.01, † p<0.001 vs. control).

Figure 5

Both cortisol and cortisone (1μM, 24hrs) pre-treatment enhance insulin stimulated PKB/akt phosphorylation. The effect of cortisone is completely blocked by the 11β-HSD1 inhibitor, glycerrhetinic acid (GE) which prevents the activation of cortisone to cortisol. Data are presented as the mean±se of n=5 experiments performed in triplicate. Representative western blots are shown inserted (*p<0.05, ** p<0.01 vs. control).

Figure 6

Insulin stimulated 2-deoxy-D-[³H] glucose uptake in undifferentiated and differentiated chub-S7 cells and skeletal myocytes. Data are expressed as the fold increase in glucose uptake with basal unstimulated uptake set as 1 (white bars). In all cell systems, insulin treatment significantly increased glucose uptake (grey bars). Dexamethasone (1μM, 48h) pre-treatment increased 2-deoxy-D-[³H] glucose uptake in undifferentiated and differentiated chub-S7 cells whilst decreasing uptake in human skeletal muscle cells (black bars). Data are presented as the mean±se of n=3-5 experiments performed in triplicate (*p<0.05 vs. control).