Diet-induced muscle insulin resistance is associated with extracellular matrix remodeling and interaction with integrin alpha2beta1 in mice - PubMed (original) (raw)

Diet-induced muscle insulin resistance is associated with extracellular matrix remodeling and interaction with integrin alpha2beta1 in mice

Li Kang et al. Diabetes. 2011 Feb.

Abstract

Objective: The hypothesis that high-fat (HF) feeding causes skeletal muscle extracellular matrix (ECM) remodeling in C57BL/6J mice and that this remodeling contributes to diet-induced muscle insulin resistance (IR) through the collagen receptor integrin α(2)β(1) was tested.

Research design and methods: The association between IR and ECM remodeling was studied in mice fed chow or HF diet. Specific genetic and pharmacological murine models were used to study effects of HF feeding on ECM in the absence of IR. The role of ECM-integrin interaction in IR was studied using hyperinsulinemic-euglycemic clamps on integrin α(2)β(1)-null (itga2(-/-)), integrin α(1)β(1)-null (itga1(-/-)), and wild-type littermate mice fed chow or HF. Integrin α(2)β(1) and integrin α(1)β(1) signaling pathways have opposing actions.

Results: HF-fed mice had IR and increased muscle collagen (Col) III and ColIV protein; the former was associated with increased transcript, whereas the latter was associated with reduced matrix metalloproteinase 9 activity. Rescue of muscle IR by genetic muscle-specific mitochondria-targeted catalase overexpression or by the phosphodiesterase 5a inhibitor, sildenafil, reversed HF feeding effects on ECM remodeling and increased muscle vascularity. Collagen remained elevated in HF-fed itga2(-/-) mice. Nevertheless, muscle insulin action and vascularity were increased. Muscle IR in HF-fed itga1(-/-) mice was unchanged. Insulin sensitivity in chow-fed itga1(-/-) and itga2(-/-) mice was not different from wild-type littermates.

Conclusions: ECM collagen expansion is tightly associated with muscle IR. Studies with itga2(-/-) mice provide mechanistic insight for this association by showing that the link between muscle IR and increased collagen can be uncoupled by the absence of collagen-integrin α(2)β(1) interaction.

PubMed Disclaimer

Figures

FIG. 1.

FIG. 1.

HF feeding to C57BL/6J mice increased muscle collagen expression. A: Immunohistochemical detection of ColIII and ColIV proteins in gastrocnemius of chow-fed and HF-fed C57BL/6J mice. The magnification of images was 20×. B: mRNA levels of ColIα1, ColIα2, ColIIIα1, and ColIVα1. C: Gelatin zymogram of protein extracts from gastrocnemius of mice after either chow or HF feeding for 20 weeks. The white bands indicate the presence of type IV collagenase activity, MMP-9, pro-MMP2, and MMP2. Data are represented as means ± SE; n = 4 ∼7. *P < 0.05 chow vs. HF. (A high-quality color representation of this figure is available in the online issue.)

FIG. 2.

FIG. 2.

Rescue of muscle inflammation and collagen expansion by catalase overexpression and chronic treatment with sildenafil. A_–_D: Muscle inflammation was assessed by measuring the mRNA levels of F4/80 and TNF-α using the quantitative real-time PCR. E: Immunohistochemical detection of ColIII and ColIV proteins in gastrocnemius of mice. The magnification of images was 20×. Values represent means ± SE of integrated intensity of staining for ColIII and ColIV. Data are normalized to WT HF or vehicle HF, and n is equal to 4 for the WT and _mcat_Tg mice, 6 ∼8 for the vehicle- and sildenafil-treated mice. #P < 0.05 compared with WT chow. *P < 0.05 compared with WT HF or vehicle HF. (A high-quality color representation of this figure is available in the online issue.)

FIG. 3.

FIG. 3.

Catalase overexpression and chronic treatment with sildenafil rescued muscle ECM adaptation and improved muscle vascularization. A and B: mRNA levels of ColIα2 and ColIIIα1 in gastrocnemius of WT and _mcat_Tg mice. C: Gelatin zymogram of protein extracts from gastrocnemius of mice. MMP2 was not detectable in muscle under current experimental conditions. Arbitrary density of the MMP9 bands was listed. D: Immunohistochemical detection of vascular markers, CD31 and VWF. The magnification of images was 20×. Representative images are presented. Values represent means ± SE of numbers of CD31-positive structures or of areas occupied by VWF-positive structures. Data are normalized to WT HF or vehicle HF, and n is equal to 4 for the WT and _mcat_Tg mice, 6 ∼8 for the vehicle- and sildenafil-treated mice. #P < 0.05 compared with WT chow. *P < 0.05 compared with WT HF or vehicle HF. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 4.

FIG. 4.

Blood glucose, glucose infusion rates, and muscle glucose uptake (_R_g) in the chow- and HF-fed integrin α1- and integrin α2-null mice during the hyperinsulinemic-euglycemic clamp. A, D, G, and J: Blood glucose was maintained at 150 ∼160 mg/dL in all groups of mice. Time course is presented to illustrate the quality of the clamps. B, E, H, and K: To maintain this euglycemia, 50% glucose was infused into mice throughout the clamp. C, F, I, and L: Nonmetabolizable glucose analog [14C]2-deoxyglucose was administered as an intravenous bolus to determine the glucose uptake (_R_g) in muscles. Values represent means ± SE, n = 3 ∼7 for each group. *P < 0.05 compared with itga1+/+ HF or _itga2+/+_HF. SVL, superficial vastus literalis.

FIG. 5.

FIG. 5.

Insulin sensitivity in liver and adipose tissue of HF-fed integrin α1- and integrin α2-null mice during the hyperinsulinemic-euglycemic clamp. A and D: Endogenous glucose production (endo_R_a) was calculated by subtracting the glucose infusion rate from total _R_a. B and E: Glucose disappearance rate (_R_d) was determined using non-steady-state equations. C and F: Nonmetabolizable glucose analog [14C]2-deoxyglucose was administered as an intravenous bolus to determine the glucose uptake (_R_g) in adipose tissue. Values represent means ± SE; n = 5 ∼6 for each group. For A, B, D, and E, *P < 0.05 compared with itga1+/+ HF at basal or itga2+/+ HF at basal. For C, *P < 0.05 compared with itga1+/+ HF.

FIG. 6.

FIG. 6.

Insulin signaling, collagen expression, and vascularization in gastrocnemius muscle of HF-fed itga2+/+ and itga2−/− mice. A and B: Tissue homogenates extracted from gastrocnemius were applied to 4–12% SDS-PAGE gel. Western blotting was then performed using the antiphospho-Akt or antiphospho-IRS-1. Values are expressed as means ± SE of integrated intensity, and representative bands are presented; n = 5 ∼6. *P < 0.05 compared with itga2+/+ HF. C: Paraffin-embedded tissue sections were stained with anti-ColIII, anti-ColIV, or anti-VWF antibodies. Frozen tissue sections were stained with anti-CD31 antibody. The magnification of images was 20×. Data are expressed as means ± SE of integrated intensity of staining for ColIII and ColIV, or of numbers of CD31-positive structures, or of areas occupied with VWF-positive structures in each section. Representative images are presented; N is equal to 6 for the itga2+/+ HF group and 3 ∼6 for the _itga2_−/− HF group. *P < 0.05 _itga2_−/− HF vs. itga2+/+ HF. (A high-quality color representation of this figure is available in the online issue.)

FIG. 7.

FIG. 7.

Proposed model for how ECM remodeling is linked to HF diet-induced muscle insulin resistance. HF diet potentially leads to insulin resistance by several mechanisms. This scheme illustrates our hypothesis that multiple mechanisms are involved in this pathway, including inflammation, ECM expansion, integrin α2β1 activation, and vascularization in muscle. Interruptions of the pathway at several steps rescue insulin resistance.

Similar articles

Cited by

References

    1. Aumailley M, Gayraud B. Structure and biological activity of the extracellular matrix. J Mol Med 1998;76:253–265 - PubMed
    1. Khoshnoodi J, Pedchenko V, Hudson BG. Mammalian collagen IV. Microsc Res Tech 2008;71:357–370 - PMC - PubMed
    1. Yurchenco PD, Patton BL. Developmental and pathogenic mechanisms of basement membrane assembly. Curr Pharm Des 2009;15:1277–1294 - PMC - PubMed
    1. Richardson DK, Kashyap S, Bajaj M, et al. Lipid infusion decreases the expression of nuclear encoded mitochondrial genes and increases the expression of extracellular matrix genes in human skeletal muscle. J Biol Chem 2005;280:10290–10297 - PubMed
    1. Berria R, Wang L, Richardson DK, et al. Increased collagen content in insulin-resistant skeletal muscle. Am J Physiol Endocrinol Metab 2006;290:E560–E565 - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources