Insulin signalling, exercise and cellular integrity.

Although the effects of exercise on insulin sensitivity are generally positive, eccentric exercise presents a paradox because it induces a transient state of insulin resistance that persists for up to 48 h after the exercise bout. Excessive eccentric contractions, such as prolonged downhill running, or marathon running, causes muscle damage and disruption of the integrity of the cell. Down-regulation of insulin receptor tyrosine phosphorylation and subsequent steps in the insulin signalling pathway, including insulin receptor substrate-1 (IRS-1)-associated phosphoinositide 3-kinase (PI3K), Akt kinase serine phosphorylation and activity and glucose transporter (GLUT-4) protein content, are evident in skeletal muscle after eccentric exercise. Furthermore, increased tumour necrosis factor alpha (TNF-alpha) secretion from monocytes is associated with the decrease in PI3K activity after this type of exercise. Recent studies have shown that TNF-alpha can increase IRS-1 serine/threonine phosphorylation, which impairs IRS-1 docking to the insulin receptor, and this inhibits insulin signalling. Thus a unifying hypothesis to explain insulin resistance after eccentric exercise may include inflammation arising from the disruption of muscle-cell integrity, leading to an acute-phase response that includes TNF-alpha, with the latter inhibiting insulin signalling and subsequent metabolic events. In contrast, exercise training increases insulin signalling and GLUT-4 expression, decreases TNF-alpha expression in skeletal muscle, and is associated with enhanced insulin sensitivity. These observations highlight the complexity of the cellular and molecular adaptations to exercise. Understanding these adaptations is essential in order to establish a sound theoretical basis for recommending exercise as a therapeutic intervention for insulin resistance and type 2 diabetes.

Human aging is associated with altered TNF-alpha production during hyperglycemia and hyperinsulinemia.

Changes in tumor necrosis factor-alpha (TNF-alpha) may provide a mechanism to explain impaired glucose metabolism with advancing age. Hyperglycemic clamps (180 min, 10 mM) were performed on seven older [67 +/- 2 yr; body mass index (BMI) 24.7 +/- 1.0 kg/m(2)] and seven younger (22 +/- 1 yr; BMI 21.8 +/- 1.3 kg/m(2)) healthy sedentary males with normal glucose tolerance. TNF-alpha production at basal and at the end of 180 min of hyperglycemia and hyperinsulinemia was measured ex vivo from lipopolysaccharide-stimulated (1 ng/ml) peripheral blood mononuclear cells. Plasma glucose, insulin, and C-peptide levels were similar in both groups at basal and during the last 30 min of the hyperglycemic clamp. Glucose infusion rates were lower (P < 0.004) in the older group compared with the young, indicating decreased insulin action among the older subjects. Basal TNF-alpha secretion was similar in older and younger subjects. TNF-alpha was suppressed (P < 0.02) in the younger group (230 +/- 46 vs. 126 +/- 49 pg/ml; basal vs. clamp) but not in the older group (153 +/- 37 vs. 182 +/- 42 pg/ml), with significant group differences in response (P < 0.05). A significant correlation was observed between the level of suppression in TNF-alpha production and insulin action (Kendall's rank, tau = 0.40, P < 0.05). Furthermore, the TNF-alpha response during the clamp was related to fat mass (r = 0.88, P < 0.001) and abdominal fat (r = 0.81, P < 0.003). In conclusion, these findings suggest a possible mechanism by which TNF-alpha may modulate glucose metabolism in younger people. Aging and modest increases in adiposity prevent the "normal" suppression of TNF-alpha production after a sustained postprandial-like hyperglycemic-hyperinsulinemic stimulus, which may contribute in part to the decline in insulin sensitivity in older men.

Insulin and exercise differentially regulate PI3-kinase and glycogen synthase in human skeletal muscle.

The purpose of this study was to determine the separate and combined effects of exercise and insulin on the activation of phosphatidylinositol 3-kinase (PI3-kinase) and glycogen synthase in human skeletal muscle in vivo. Seven healthy men performed three trials in random order. The trials included 1) ingestion of 2 g/kg body wt carbohydrate in a 10% solution (CHO); 2) 75 min of semirecumbent cycling exercise at 75% of peak O(2) consumption; followed by 5 x 1-min maximal sprints (Ex); and 3) Ex, immediately followed by ingestion of the carbohydrate solution (ExCHO). Plasma glucose and insulin were increased (P < 0.05) at 15 and 30 (Post-15 and Post-30) min after the trial during CHO and ExCHO, although insulin was lower for ExCHO. Hyperinsulinemia during recovery in CHO and ExCHO led to an increase (P < 0.001) in PI3-kinase activity at Post-30 compared with basal, although the increase was lower (P < 0. 004) for ExCHO. Furthermore, PI3-kinase activity was suppressed (P < 0.02) immediately after exercise (Post-0) during Ex and ExCHO. Area under the insulin response curve for all trials was positively associated with PI3-kinase activity (r = 0.66, P < 0.001). Glycogen synthase activity did not increase during CHO but was increased (P < 0.05) at Post-0 and Post-30 during Ex and ExCHO. Ingestion of the drink increased (P < 0.05) carbohydrate oxidation during CHO and ExCHO, although the increase after ExCHO was lower (P < 0.05) than CHO. Carbohydrate oxidation was directly correlated with PI3-kinase activity for all trials (r = 0.63, P < 0.001). In conclusion, under resting conditions, ingestion of a carbohydrate solution led to activation of the PI3-kinase pathway and oxidation of the carbohydrate. However, when carbohydrate was ingested after intense exercise, the PI3-kinase response was attenuated and glycogen synthase activity was augmented, thus facilitating nonoxidative metabolism or storage of the carbohydrate. Activation of glycogen synthase was independent of PI3-kinase.

Muscle damage impairs insulin stimulation of IRS-1, PI 3-kinase, and Akt-kinase in human skeletal muscle.

Physiological stress associated with muscle damage results in systemic insulin resistance. However, the mechanisms responsible for the insulin resistance are not known; therefore, the present study was conducted to elucidate the molecular mechanisms associated with insulin resistance after muscle damage. Muscle biopsies were obtained before (base) and at 1 h during a hyperinsulinemic-euglycemic clamp (40 mU x kg(-1) x min(-1)) in eight young (age 24+/-1 yr) healthy sedentary (maximal O(2) consumption, 49.7+/-2.4 ml x kg(-1) x min(-1)) males before and 24 h after eccentric exercise (ECC)-induced muscle damage. To determine the role of cytokines in ECC-induced insulin resistance, venous blood samples were obtained before (control) and 24 h after ECC to evaluate ex vivo endotoxin-induced mononuclear cell secretion of tumor necrosis factor (TNF)-alpha, interleukin (IL)-6, and IL-1beta. Glucose disposal was 19% lower after ECC (P<0.05). Insulin-stimulated insulin receptor substrate (IRS)-1 tyrosine phosphorylation was 45% lower after ECC (P<0.05). Insulin-stimulated phosphatidylinositol (PI) 3-kinase, Akt (protein kinase B) serine phosphorylation, and Akt activity were reduced 34, 65, and 20%, respectively, after ECC (P < 0.05). TNF-alpha, but not IL-6 or IL-1beta production, increased 2.4-fold 24 h after ECC (P<0.05). TNF-alpha production was positively correlated with reduced insulin action on PI 3-kinase (r = 0.77, P = 0.04). In summary, the physiological stress associated with muscle damage impairs insulin stimulation of IRS-1, PI 3-kinase, and Akt-kinase, presumably leading to decreased insulin-mediated glucose uptake. Although more research is needed on the potential role for TNF-alpha inhibition of insulin action, elevated TNF-alpha production after muscle damage may impair insulin signal transduction.

Regular exercise enhances insulin activation of IRS-1-associated PI3-kinase in human skeletal muscle.

Insulin action in skeletal muscle is enhanced by regular exercise. Whether insulin signaling in human skeletal muscle is affected by habitual exercise is not well understood. Phosphatidylinositol 3-kinase (PI3-kinase) activation is an important step in the insulin-signaling pathway and appears to regulate glucose metabolism via GLUT-4 translocation in skeletal muscle. To examine the effects of regular exercise on PI3-kinase activation, 2-h hyperinsulinemic (40 mU. m(-2). min(-1))-euglycemic (5.0 mM) clamps were performed on eight healthy exercise-trained [24 +/- 1 yr, 71.8 +/- 2.0 kg, maximal O(2) uptake (VO(2 max)) of 56.1 +/- 2.5 ml. kg(-1). min(-1)] and eight healthy sedentary men and women (24 +/- 1 yr, 64.7 +/- 4.4 kg, VO(2 max) of 44.4 +/- 2.7 ml. kg(-1). min(-1)). A [6, 6-(2)H]glucose tracer was used to measure hepatic glucose output. A muscle biopsy was obtained from the vastus lateralis muscle at basal and at 2 h of hyperinsulinemia to measure insulin receptor substrate-1(IRS-1)-associated PI3-kinase activation. Insulin concentrations during hyperinsulinemia were similar for both groups (293 +/- 22 and 311 +/- 22 pM for trained and sedentary, respectively). Insulin-mediated glucose disposal rates (GDR) were greater (P < 0.05) in the exercise-trained compared with the sedentary control group (9.22 +/- 0.95 vs. 6.36 +/- 0.57 mg. kg fat-free mass(-1). min(-1)). Insulin-stimulated PI3-kinase activation was also greater (P < 0.004) in the trained compared with the sedentary group (3.8 +/- 0.5- vs. 1.8 +/- 0.2-fold increase from basal). Endurance capacity (VO(2 max)) was positively correlated with PI3-kinase activation (r = 0.53, P < 0.04). There was no correlation between PI3-kinase and muscle morphology. However, increases in GDR were positively related to PI3-kinase activation (r = 0.60, P < 0.02). We conclude that regular exercise leads to greater insulin-stimulated IRS-1-associated PI3-kinase activation in human skeletal muscle, thus facilitating enhanced insulin-mediated glucose uptake.

TNF-alpha impairs insulin signaling and insulin stimulation of glucose uptake in C2C12 muscle cells.

Physiological stressors such as sepsis and tissue damage initiate an acute immune response and cause transient systemic insulin resistance. This study was conducted to determine whether tumor necrosis factor-alpha (TNF-alpha), a cytokine produced by immune cells during skeletal muscle damage, decreases insulin responsiveness at the cellular level. To examine the molecular mechanisms associated with TNF-alpha and insulin action, we measured insulin receptor substrate (IRS)-1- and IRS-2-mediated phosphatidylinositol 3-kinase (PI 3-kinase) activation, IRS-1-PI 3-kinase binding, IRS-1 tyrosine phosphorylation, and the phosphorylation of two mitogen-activated protein kinases (MAPK, known as p42(MAPK) and p44(MAPK)) in cultured C2C12 myotubes. Furthermore, we determined the effects of TNF-alpha on insulin-stimulated 2-deoxyglucose (2-DG) uptake. We observed that TNF-alpha impaired insulin stimulation of IRS-1- and IRS-2-mediated PI 3-kinase activation by 54 and 55% (P < 0.05), respectively. In addition, TNF-alpha decreased insulin-stimulated IRS-1 tyrosine phosphorylation by 40% (P < 0.05). Furthermore, TNF-alpha repressed insulin-induced p42(MAPK) and p44(MAPK) tyrosine phosphorylation by 81% (P < 0.01). TNF-alpha impairment of insulin signaling activation was accompanied by a decrease (P < 0.05) in 2-DG uptake in the muscle cells (60 +/- 4 vs. 44 +/- 6 pmol. min-1. mg-1). These data suggest that increases in TNF-alpha may cause insulin resistance in skeletal muscle by inhibiting IRS-1- and IRS-2-mediated PI 3-kinase activation as well as p42(MAPK) and p44(MAPK) tyrosine phosphorylation, leading to impaired insulin-stimulated glucose uptake.

Expression of vascular permeability factor/vascular endothelial growth factor by melanoma cells increases tumor growth, angiogenesis, and experimental metastasis.

Vascular permeability factor (VPF)/vascular endothelial growth factor (VEGF) is an angiogenic cytokine expressed by many human and animal tumors. Hypoxia often up-regulates VPF/VEGF expression further. To better define the role of VPF/VEGF in tumor biology, we screened tumorigenic lines for those expressing minimal constitutive and hypoxia-inducible VPF/VEGF. Human melanoma SK-MEL-2 cells best fit these criteria and formed small, poorly vascularized tumors in immunodeficient mice. We transfected SK-MEL-2 cells stably with sense or antisense mouse VPF/VEGF cDNA or with vector alone. Cells transfected with sense VPF/VEGF (V+) expressed and secreted large amounts of mouse VPF/VEGF and formed well-vascularized tumors with hyperpermeable blood vessels and minimal necrosis in nude/SCID mice. Antisense-transfected VPF/VEGF (V-) cells expressed reduced constitutive VPF/VEGF and no detectable mouse VPF/VEGF, and formed small, minimally vascularized tumors exhibiting extensive necrosis. Vector-alone transfectants (N1 cells) behaved like parental cells. V+ cells formed numerous lung tumor colonies in SCID mice, approximately 50-fold more than N1 cells, whereas V- cells formed few or none. These experiments demonstrate that VPF/VEGF promotes melanoma growth by stimulating angiogenesis and that constitutive VPF/VEGF expression dramatically promotes tumor colonization in the lung.