Skeletal Muscle Insulin Resistance and Absence of Inflammation Characterize Insulin-Resistant Grade I Obese Women

Abstract

Context: Obesity is associated with insulin-resistance (IR), the key feature of type 2 diabetes. Although chronic low-grade inflammation has been identified as a central effector of IR development, it has never been investigated simultaneously at systemic level and locally in skeletal muscle and adipose tissue in obese humans characterized for their insulin sensitivity.

Objectives: We compared metabolic parameters and inflammation at systemic and tissue levels in normal-weight and obese subjects with different insulin sensitivity to better understand the mechanisms involved in IR development.

Methods: 30 post-menopausal women were classified as normal-weight insulin-sensitive (controls, CT) and obese (grade I) insulin-sensitive (OIS) or insulin-resistant (OIR) according to their body mass index and homeostasis model assessment of IR index. They underwent a hyperinsulinemic-euglycemic clamp, blood sampling, skeletal muscle and subcutaneous adipose tissue biopsies, an activity questionnaire and a self-administrated dietary recall. We analyzed insulin sensitivity, inflammation and IR-related parameters at the systemic level. In tissues, insulin response was assessed by P-Akt/Akt expression and inflammation by macrophage infiltration as well as cytokines and IκBα expression.

Results: Systemic levels of lipids, adipokines, inflammatory cytokines, and lipopolysaccharides were equivalent between OIS and OIR subjects. In subcutaneous adipose tissue, the number of anti-inflammatory macrophages was higher in OIR than in CT and OIS and was associated with higher IL-6 level. Insulin induced Akt phosphorylation to the same extent in CT, OIS and OIR. In skeletal muscle, we could not detect any inflammation even though IκBα expression was lower in OIR compared to CT. However, while P-Akt/Akt level increased following insulin stimulation in CT and OIS, it remained unchanged in OIR.

Conclusion: Our results show that systemic IR occurs without any change in systemic and tissues inflammation. We identified a muscle defect in insulin response as an early mechanism of IR development in grade I obese post-menopausal women.

Fig 1. Dietary intake.

Calorie intake (Panel A) and diet composition (Panel B) were determined by self-administered questionnaire and analysis using GENI software. Data are expressed as mean ± SEM. Calorie, * P_{OIS vs. CT} = 0.0007; $ P_{OIR vs. CT} = 0.0003. Proteins, * P_{OIS vs. CT} = 0.0007; $ P_{OIR vs. CT} = 0.009. Fats, $ P_{OIR vs. CT} = 0.0003. Carbohydrates, * P_{OIS vs. CT} = 0.04.

Fig 2. Inflammation and fibrosis in SAT.

Panel A: Total macrophages (CD68⁺), pro-inflammatory (CD68⁺/CD86⁺), and anti-inflammatory (CD68⁺/CD206⁺) macrophages were counted after staining in SAT. For CD68, and CD86 or CD206 markers, four and two non-consecutive entire sections per subject were analyzed for 7 CT, 5 OIS and 6 OIR subjects. Data are expressed as mean ± SEM. $: P_{OIR vs. CT} = 0.008. Panel B: Adipocyte average size (μm²) in SAT; 10–15 sections per subject were respectively analyzed for 4 subjects per group. Data are expressed as mean ± SEM. Panel C: Relative quantification of Col5A, Col6A, TNFα, IL-1β, IL-6 and MCP-1 mRNA expression in SAT (n_CT = 10, n_OIS = 11, n_OIR = 9). Data are expressed as mean ± SEM and relatively to CT values, which were set at 1.0. $ P_{OIR vs. CT} = 0.004. Panel D: Western blot analysis of IκBα expression in SAT (n_CT = 9, n_OIS = 8, n_OIR = 8). The graph represents IκBα protein quantification after correction by α/β tubulin protein levels, used as an indicator of protein loading. Data are expressed as mean ± SEM and relatively to CT values, which were set at 1.0.

Fig 3. Insulin response in SAT.

Panel A: Western blot analysis of P-Akt/Akt ratio in SAT with and without incubation with insulin (n_CT = 9, n_OIS = 10, n_OIR = 9). The graph represents P-Akt protein quantification after correction by total Akt protein levels, used as an indicator of protein loading. Data are expressed as mean ± SEM and relatively to CT values, which were set at 1.0. § P_CT±insulin = 0.008; P_OIS±insulin = 0.002; P_OIR±insulin = 0.004. Panel B: Fold induction of P-Akt/Akt level in SAT after insulin stimulation. Data are expressed as mean ± SEM. Panels C and D: Correlations (Spearman analysis) between insulin-stimulated P-Akt/Akt fold induction in SAT and HOMA_IR (C) or GIR (D).

Fig 4. Inflammation and fibrosis in skeletal muscle.

Panel A: Total macrophages (CD68⁺), and pro-inflammatory (CD68⁺/CD86⁺) and anti-inflammatory (CD68⁺/CD206⁺) macrophages were counted after staining in skeletal muscle. For CD68, and CD86 or CD206 markers, four and two non-consecutive entire sections per subject were respectively analyzed for 6 CT, 6 OIS and 5 OIR subjects. Data are expressed as mean ± SEM. Panel B: Relative quantification of Col5A, Col6A, TNFα, IL-1β, IL-6 and MCP-1, mRNA expression in skeletal muscle (n_CT = 10, n_OIS = 11, n_OIR = 9). Data are expressed as mean ± SEM and relatively to CT values, which were set at 1.0. Panel C: Western blot analysis of IκBα expression in skeletal muscle (n_CT = 10, n_OIS = 9, n_OIR = 9). The graph represents IκBα protein quantification after correction by α/β tubulin protein levels, used as an indicator of protein loading. Data are expressed as mean ± SEM and relatively to CT values, which were set at 1.0. $ P_{OIR vs. CT} = 0.04.

Fig 5. Insulin response in skeletal muscle.

Panel A: Western blot analysis of P-Akt/Akt ratio in skeletal muscle with and without incubation with insulin (n_CT = 10, n_OIS = 11, n_OIR = 8). The graph represents P-Akt protein quantification after correction by total Akt protein levels, used as an indicator of protein loading. Data are expressed as mean ± SEM and relatively to CT values, which were set at 1.0. § P_CT±insulin = 0.002; P_OIS±insulin = 0.002. P_OIR±insulin = 0.129. Panel B: Fold induction of P-Akt/Akt in skeletal muscle after insulin stimulation. Data are mean ± SEM. $ P_{OIR vs. CT} = 0.008 and # P_{OIR vs. OIS} = 0.0005. Panels C and D: Correlations (Spearman analysis) between insulin-stimulated P-Akt/Akt fold induction in skeletal muscle and HOMA_IR (C) or GIR (D).