Slow-twitch fiber proportion in skeletal muscle correlates with insulin responsiveness

Abstract

Context: The metabolic syndrome, characterized by central obesity with dyslipidemia, hypertension, and hyperglycemia, identifies people at high risk for type 2 diabetes.

Objective: Our objective was to determine how the insulin resistance of the metabolic syndrome is related to muscle fiber composition.

Design: Thirty-nine sedentary men and women (including 22 with the metabolic syndrome) had insulin responsiveness quantified using euglycemic clamps and underwent biopsies of the vastus lateralis muscle. Expression of insulin receptors, insulin receptor substrate-1, glucose transporter 4, and ATP synthase were quantified with immunoblots and immunohistochemistry.

Participants and setting: Participants were nondiabetic, metabolic syndrome volunteers and sedentary control subjects studied at an outpatient clinic.

Main outcome measures: Insulin responsiveness during an insulin clamp and the fiber composition of a muscle biopsy specimen were evaluated.

Results: There were fewer type I fibers and more mixed (type IIa) fibers in metabolic syndrome subjects. Insulin responsiveness and maximal oxygen uptake correlated with the proportion of type I fibers. Insulin receptor, insulin receptor substrate-1, and glucose transporter 4 expression were not different in whole muscle but all were significantly less in the type I fibers of metabolic syndrome subjects when adjusted for fiber proportion and fiber size. Fat oxidation and muscle mitochondrial expression were not different in the metabolic syndrome subjects.

Conclusion: Lower proportion of type I fibers in metabolic syndrome muscle correlated with the severity of insulin resistance. Even though whole muscle content was normal, key elements of insulin action were consistently less in type I muscle fibers, suggesting their distribution was important in mediating insulin effects.

Figures

Figure 1.

Metabolic syndrome subjects' fiber composition of vastus lateralis muscle is different from controls. A, Bright-field image of a transverse section of vastus lateralis muscle from a subject with the metabolic syndrome. Antibodies against slow-twitch myosin heavy chain and fast-twitch myosin heavy chain are added sequentially as described by Behan and coworkers (18). Type I fibers are darkest, type IIx fibers are pink, and type IIa fibers are intermediate because they contain both myosin types. Scale bar, 100 μm. B, Resulting fiber composition data for sedentary controls and metabolic syndrome subjects. The fiber composition data from subjects with the metabolic syndrome demonstrated that there were fewer type I fibers (38% vs 48%) and more type IIa fibers (27% vs 13%) than in muscle from sedentary controls. **, Significant difference from the controls at P < .01.

Figure 2.

Muscle type I fiber content predicts insulin responsiveness and aerobic fitness. A, Graph of the SSGIR plotted against the type I fiber percentage in the muscle biopsy specimens from 17 controls and 22 subjects with the metabolic syndrome. The open symbols represent the controls, and the filled symbols represent the subjects with the metabolic syndrome. R is the correlation coefficient and p is the probability of the null hypothesis. B, VO2max plotted vs the percentage of type I fibers in the muscle biopsies from the same controls and metabolic syndrome subjects displayed in A.

Figure 3.

Muscle content of insulin receptor, IRS-1, and GLUT4. Shown here are sample immunoblots for the 3 proteins in muscle homogenates. In the blots shown, the first and third bands were from control subject muscle. The second and fourth bands were from metabolic syndrome muscle. None of expression of these proteins in metabolic syndrome muscle was significantly different from that of the control subjects. Each subject's muscle sample was run in a minimum of 3 separate analyses.

Figure 4.

Muscle fiber type-specific expression of insulin receptor, IRS-1, GLUT4, and ATP synthase. Shown here are confocal microscopy images from 4 separate biopsy specimens labeled with antibodies against slow-twitch myosin heavy chain (stained blue) and antibodies specific to the human insulin receptor β-subunit (green), IRS-1 (green), GLUT4 (red), or ATP synthase (blue). Panels A–C are from a single portion of a slide viewed on the confocal microscope. Panels D–F, similarly, are one confocal view of another slide, as are panels G–I and J–L. A–C, Insulin receptor signal (A), slow-twitch myosin (B), and composite view of both antibodies (C). D–F, Antihuman IRS-1 (D), slow-twitch myosin (E), and composite image with IRS-1 and slow-twitch myosin signal (F). G–I, Human GLUT4 signal (G), slow-twitch myosin (H), and composite image with GLUT4 and the slow-twitch myosin heavy chain label (I). J–L, Signal from the ATP synthase-specific antibody (J), image of the same section probed with rabbit antifast twitch myosin heavy chain (K), and composite image of those of J and K (L). Note that in each of the 4 subjects displayed here, the type IIa fibers can be identified in B, E, H, and K as the intermediate intensity stained fibers. Scale bar in J, 100 μm. All panels are at the same magnification.

Figure 5.

Metabolic syndrome muscle type I fibers contain a lower proportion of insulin receptor, IRS-1, GLUT4, and ATP synthase. The 4 panels on the left display the signal intensity in each of the 3 types of fibers and compare controls and metabolic syndrome muscle. These data were determined based on digital image analysis of immunohistochemical studies, examples of which are shown in Figure 4. For each individual subject, an estimate of the total amount of each of these 3 proteins associated with the specific fiber types was calculated from the signal intensity, the percentage of each fiber type, and the fiber size of each type. The means and SEM of these adjusted data are shown in the panels on the right. *P value < .05 by t test. For each of these proteins, less was present in the type I muscle fibers of the metabolic syndrome subjects. More of IRS-1, GLUT4, and ATP synthase was expressed in the IIa fibers. GLUT4 was significantly less in both type I and type IIx fibers of metabolic syndrome subjects. The mitochondrial marker data shown in D and H represent only the most recently studied 18 subjects (7 controls and 11 metabolic syndrome subjects), whereas the other panels represent the entire subject groups. Unexpectedly, the IIx fibers expressed more mitochondrial marker than the type I fibers in both groups. The difference between type I and IIx ATP synthase signal intensity reached statistical significance in the metabolic syndrome muscle only (P = .039, paired t test, vs P = .101 for control muscle). As with the insulin receptor, IRS-1, and GLUT4 data, the total ATP synthase expression in each fiber type adjusted for fiber number and size was significantly different in the metabolic syndrome muscle.