Genome-wide analysis of mitochondrial DNA copy number reveals loci implicated in nucleotide metabolism, platelet activation, and megakaryocyte proliferation.

Mitochondrial DNA copy number (mtDNA-CN) measured from blood specimens is a minimally invasive marker of mitochondrial function that exhibits both inter-individual and intercellular variation. To identify genes involved in regulating mitochondrial function, we performed a genome-wide association study (GWAS) in 465,809 White individuals from the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) consortium and the UK Biobank (UKB). We identified 133 SNPs with statistically significant, independent effects associated with mtDNA-CN across 100 loci. A combination of fine-mapping, variant annotation, and co-localization analyses was used to prioritize genes within each of the 133 independent sites. Putative causal genes were enriched for known mitochondrial DNA depletion syndromes (p = 3.09 × 10-15) and the gene ontology (GO) terms for mtDNA metabolism (p = 1.43 × 10-8) and mtDNA replication (p = 1.2 × 10-7). A clustering approach leveraged pleiotropy between mtDNA-CN associated SNPs and 41 mtDNA-CN associated phenotypes to identify functional domains, revealing three distinct groups, including platelet activation, megakaryocyte proliferation, and mtDNA metabolism. Finally, using mitochondrial SNPs, we establish causal relationships between mitochondrial function and a variety of blood cell-related traits, kidney function, liver function and overall (p = 0.044) and non-cancer mortality (p = 6.56 × 10-4).

The effect of leptin on Lep expression is tissue-specific and nutritionally regulated.

Leptin, the product of the Obese (Lep) gene, orchestrates behavioral and metabolic responses to nutrient intake. Here, we demonstrate tissue-specific autoregulation of Lep. Moderate increases in circulating leptin considerably decreased Lep expression in adipose tissue and induced lep expression in skeletal muscle, a tissue that normally does not express this gene. Changes in nutrient availability resulted in rapid alterations in Lep autoregulation. These findings demonstrate negative feedback regulation of Lep in fat, and indicate that leptin secretion can function as a vehicle of 'cross-talk' between adipose tissue and skeletal muscle, leading to tissue-specific modulation of the 'leptin signal'.

In vivo glucosamine infusion induces insulin resistance in normoglycemic but not in hyperglycemic conscious rats.

To test the hypothesis that increased flux through the hexosamine biosynthetic pathway can induce insulin resistance in skeletal muscle in vivo, we monitored glucose uptake, glycolysis, and glycogen synthesis during insulin clamp studies in 6-h fasted conscious rats in the presence of a sustained (7-h) increase in glucosamine (GlcN) availability. Euglycemic (approximately 7 mM) insulin (approximately 2,500 pM) clamps with saline or GlcN infusions were performed in control (CON; plasma glucose [PG] = 7.4 +/- 0.2 mM), diabetic (D; PG = 19.7 +/- 1.1), and phlorizin-treated (3-wk) diabetic rats (D + PHL; PG = 7.6 +/- 0.9). 7-h euglycemic hyperinsulinemia with saline did not significantly decrease Rd (360-420 min = 39.2 +/- 3.6 vs. 60-120 min = 42.2 +/- 3.7 mg/kg.min; P = NS). GlcN infusion raised plasma GlcN concentrations to approximately 1.2 mM and increased muscle and liver UDP-GlcNAc levels by 4-5-fold in all groups. GlcN markedly decreased Rd in CON (360-420 min = 30.4 +/- 1.3 vs. 60-120 min = 44.1 +/- 3.5 mg/kg.min; P < 0.01) and D + PHL (360-420 min = 29.4 +/- 2.5 vs. 60-120 min = 43.8 +/- 2.9 mg/kg.min; P < 0.01), but not in D (5-7 h = 21.5 +/- 0.8 vs. 0-2 h = 24.3 +/- 1.1 mg/kg.min; P = NS). Thus, increased GlcN availability induces severe skeletal muscle insulin resistance in normoglycemic but not in chronically hyperglycemic rats. The lack of additive effects of GlcN and chronic hyperglycemia (experimental diabetes) provides support for the hypothesis that increased flux through the GlcN pathway in skeletal muscle may play an important role in glucose-induced insulin resistance in vivo.

Leptin selectively decreases visceral adiposity and enhances insulin action.

Intraabdominal adiposity and insulin resistance are risk factors for diabetes mellitus, dyslipidemia, arteriosclerosis, and mortality. Leptin, a fat-derived protein encoded by the ob gene, has been postulated to be a sensor of energy storage in adipose tissue capable of mediating a feedback signal to sites involved in the regulation of energy homeostasis. Here, we provide evidence for specific effects of leptin on fat distribution and in vivo insulin action. Leptin (LEP) or vehicle (CON) was administered by osmotic minipumps for 8 d to pair-fed adult rats. During the 8 d of the study, body weight and total fat mass decreased similarly in LEP and in CON. However, while moderate calorie restriction (CON) resulted in similar decreases in whole body (by 20%) and visceral (by 21%) fat, leptin administration led to a specific and marked decrease (by 62%) in visceral adiposity. During physiologic hyperinsulinemia (insulin clamp), leptin markedly enhanced insulin action on both inhibition of hepatic glucose production and stimulation of glucose uptake. Finally, leptin exerted complex effects on the hepatic gene expression of key metabolic enzymes and on the intrahepatic partitioning of metabolic fluxes, which are likely to represent a defense against excessive storage of energy in adipose depots. These studies demonstrate novel actions of circulating leptin in the regulation of fat distribution, insulin action, and hepatic gene expression and suggest that it may play a role in the pathophysiology of abdominal obesity and insulin resistance.

Role of the glucosamine pathway in fat-induced insulin resistance.

To examine whether the hexosamine biosynthetic pathway might play a role in fat-induced insulin resistance, we monitored the effects of prolonged elevations in FFA availability both on skeletal muscle levels of UDP-N-acetyl-hexosamines and on peripheral glucose disposal during 7-h euglycemic-hyperinsulinemic (approximately 500 microU/ml) clamp studies. When the insulin-induced decrease in the plasma FFA levels (to approximately 0.3 mM) was prevented by infusion of a lipid emulsion in 15 conscious rats (plasma FFA approximately 1.4 mM), glucose uptake (5-7 h = 32.5+/-1.7 vs 0-2 h = 45.2+/-2.8 mg/kg per min; P < 0.01) and glycogen synthesis (P < 0.01) were markedly decreased. During lipid infusion, muscle UDP-N-acetyl-glucosamine (UDP-GlcNAc) increased by twofold (to 53.4+/-1.1 at 3 h and to 55.5+/-1.1 nmol/gram at 7 h vs 20.4+/-1.7 at 0 h, P < 0.01) while glucose-6-phosphate (Glc-6-P) levels were increased at 3 h (475+/-49 nmol/gram) and decreased at 7 h (133+/-7 vs 337+/-28 nmol/gram at 0 h, P < 0.01). To discern whether such an increase in the skeletal muscle UDP-GlcNAc concentration could account for the development of insulin resistance, we generated similar increases in muscle UDP-GlcNAc using three alternate experimental approaches. Euglycemic clamps were performed after prolonged hyperglycemia (18 mM, n = 10), or increased availability of either glucosamine (3 micromol/kg per min; n = 10) or uridine (30 micromol/kg per min; n = 4). These conditions all resulted in very similar increases in the skeletal muscle UDP-GlcNAc (to approximately 55 nmol/gram) and markedly impaired glucose uptake and glycogen synthesis. Thus, fat-induced insulin resistance is associated with: (a) decreased skeletal muscle Glc-6-P levels indicating defective transport/phosphorylation of glucose; (b) marked accumulation of the endproducts of the hexosamine biosynthetic pathway preceding the onset of insulin resistance. Most important, the same degree of insulin resistance can be reproduced in the absence of increased FFA availability by a similar increase in skeletal muscle UDP-N-acetyl-hexosamines. In conclusion, our results support the hypothesis that increased FFA availability induces skeletal muscle insulin resistance by increasing the flux of fructose-6-phosphate into the hexosamine pathway.

Mechanism by which hyperglycemia inhibits hepatic glucose production in conscious rats. Implications for the pathophysiology of fasting hyperglycemia in diabetes.

To examine the relationship between the plasma glucose concentration (PG) and the pathways of hepatic glucose production (HGP), five groups of conscious rats were studied after a 6-h fast: (a) control rats (PG = 8.0 +/- 0.2 mM); (b) control rats (PG = 7.9 +/- 0.2 mM) with somatostatin and insulin replaced at the basal level; (c) control rats (PG = 18.1 +/- 0.2 mM) with somatostatin, insulin replaced at the basal level, and glucose infused to acutely raise plasma glucose by 10 mM; (d) control rats (PG = 18.0 +/- 0.2 mM) with somatostatin and glucose infusions to acutely reproduce the metabolic conditions of diabetic rats, i.e., hyperglycemia and moderate hypoinsulinemia; (e) diabetic rats (PG = 18.4 +/- 2.3 mM). All rats received an infusion of [3-3H]glucose and [U-14C]lactate. The ratio between hepatic [14C]UDP-glucose sp act (SA) and 2X [14C]-phosphoenolpyruvate (PEP) SA (the former reflecting glucose-6-phosphate SA) measured the portion of total glucose output derived from PEP-gluconeogenesis. In control rats, HGP was decreased by 58% in hyperglycemic compared to euglycemic conditions (4.5 +/- 0.3 vs. 10.6 +/- 0.2 mg/kg.min; P < 0.01). When evaluated under identical glycemic conditions, HGP was significantly increased in diabetic rats (18.9 +/- 1.4 vs. 6.2 +/- 0.4 mg/kg.min; P < 0.01). In control rats, hyperglycemia increased glucose cycling (by 2.5-fold) and the contribution of gluconeogenesis to HGP (91% vs. 45%), while decreasing that of glycogenolysis (9% vs. 55%). Under identical plasma glucose and insulin concentrations, glucose cycling in diabetic rats was decreased (by 21%) and the percent contribution of gluconeogenesis to HGP (73%) was similar to that of controls (84%). These data indicate that: (a) hyperglycemia causes a marked inhibition of HGP mainly through the suppression of glycogenolysis and the increase in glucokinase flux, with no apparent changes in the fluxes through gluconeogenesis and glucose-6-phosphatase; under similar hyperglycemic hypoinsulinemic conditions: (b) HGP is markedly increased in diabetic rats; however, (c) the contribution of glycogenolysis and gluconeogenesis to HGP is similar to control animals.

Caloric restriction reverses hepatic insulin resistance in aging rats by decreasing visceral fat.

Hyperinsulinemia and increased visceral/abdominal fat (VF) are common features of human aging. To examine the relationships among VF, peripheral, and hepatic insulin sensitivity, we studied 4- and 18-mo-old male Sprague-Dawley rats (n = 42) fed ad libitum (4 AL and 18 AL) or moderately calorie restricted (18 CR) up to 18 mo of age. Total fat mass (FM) and VF were decreased in 18 CR to approximately one-third of that of 18 AL (P < 0.001), while lean body mass (LBM) was unchanged. Most important, 18 CR had more FM (65+/-6 vs. 45+/-6 g) but less VF (7.8+/-0.6 vs. 12.3+/-3.3 g) compared with 4 AL (P < 0.01 for both). Thus, the effects of variable VF on HIS could be assessed, independent of FM and age. Marked hepatic insulin resistance ensued with aging (18 AL) and CR restored hepatic insulin sensitivity to the levels of young rats, while peripheral insulin sensitivity remained unchanged (by insulin clamp of 18 mU/kg/min). In fact, the rates of insulin infusion required to maintain basal hepatic glucose production in the presence of pancreatic clamp were 0.75+/-0.10, 1.41+/-0.13, and 0.51+/-0.12 mU/kg . min, in 4 AL, 18 AL, and 18 CR, respectively (P < 0.01 between all groups), and in 18 CR rats infused with insulin at similar rates as in the 18 AL (1.4 mU/kg/min) hepatic glucose production was decreased by 32% (P < 0. 005). Furthermore, when 18 CR rats were fed AL for 14 d, VF rapidly and selectively increased and severe hepatic insulin resistance was induced. We propose that in this animal model the age-associated decrease in hepatic (rather than peripheral) insulin action is the major determinant of fasting hyperinsulinemia and that increased visceral adiposity plays the major role in inducing hepatic insulin resistance. Thus, interventions designed to prevent the accumulation of VF are likely to represent an effective mean to improve carbohydrate metabolism in aging.

Peripheral but not hepatic insulin resistance in mice with one disrupted allele of the glucose transporter type 4 (GLUT4) gene.

Glucose transporter type 4 (GLUT4) is insulin responsive and is expressed in striated muscle and adipose tissue. To investigate the impact of a partial deficiency in the level of GLUT4 on in vivo insulin action, we examined glucose disposal and hepatic glucose production (HGP) during hyperinsulinemic clamp studies in 4-5-mo-old conscious mice with one disrupted GLUT4 allele [GLUT4 (+/-)], compared with wild-type control mice [WT (+/+)]. GLUT4 (+/-) mice were studied before the onset of hyperglycemia and had normal plasma glucose levels and a 50% increase in the fasting (6 h) plasma insulin concentrations. GLUT4 protein in muscle was approximately 45% less in GLUT4 (+/-) than in WT (+/+). Euglycemic hyperinsulinemic clamp studies were performed in combination with [3-3H]glucose to measure the rate of appearance of glucose and HGP, with [U-14C]-2-deoxyglucose to estimate muscle glucose transport in vivo, and with [U-14C]lactate to assess hepatic glucose fluxes. During the clamp studies, the rates of glucose infusion, glucose disappearance, glycolysis, glycogen synthesis, and muscle glucose uptake were approximately 55% decreased in GLUT4 (+/-), compared with WT (+/+) mice. The decreased rate of in vivo glycogen synthesis was due to decreased stimulation of glucose transport since insulin's activation of muscle glycogen synthase was similar in GLUT4 (+/-) and in WT (+/+) mice. By contrast, the ability of hyperinsulinemia to inhibit HGP was unaffected in GLUT4 (+/-). The normal regulation of hepatic glucose metabolism in GLUT4 (+/-) mice was further supported by the similar intrahepatic distribution of liver glucose fluxes through glucose cycling, gluconeogenesis, and glycogenolysis. We conclude that the disruption of one allele of the GLUT4 gene leads to severe peripheral but not hepatic insulin resistance. Thus, varying levels of GLUT4 protein in striated muscle and adipose tissue can markedly alter whole body glucose disposal. These differences most likely account for the interindividual variations in peripheral insulin action.

Regulation of gene expression in PC12 cells via an activator of dual second messengers: pituitary adenylate cyclase activating polypeptide.

In this study we demonstrate that the activator protein-1 (AP-1) DNA motif, initially considered to be unresponsive to cyclic AMP (cAMP), does function as a cAMP-response element in PC12 cells. A luciferase reporter gene driven by the collagenase promoter that contains the AP-1 motif is responsive to cAMP as well as phorbol esters when transfected in PC12 cells. We have recently shown that pituitary adenylate cyclase activating peptide (PACAP) has neurotrophic properties and activates both adenylylcyclase and the inositol lipid cascade in PC12 cells. Consistent with these actions, we demonstrate that PACAP is an effective activator of luciferase reporter genes whose promoters bear the AP-1 motif, as well as the related DNA element that binds the protein CREB. Both the cAMP and inositol lipid pathways appear to play a role in the activation of these motifs by PACAP. Mutation of the AP-1 motif and its juxtaposition to a heterologous promoter proves that the AP-1 motif is a locus for response to cAMP and PACAP. The luciferase reporter genes bearing the AP-1 motif are not cAMP responsive in HeLa tk- cells, indicating that the mode of second-messenger responsiveness is cell-type specific.

Overfeeding rapidly induces leptin and insulin resistance.

In common forms of obesity, hyperphagia, hyperinsulinemia, and hyperleptinemia coexist. Here, we demonstrate rapid induction of insulin and leptin resistance by short-term overfeeding. After 3 and 7 days on the assigned diet regimen, rats were tested for their biological responses to acute elevations in plasma insulin and leptin concentrations. Severe resistance to the metabolic effects of both leptin and insulin ensued after just 3 days of overfeeding. During the insulin clamp studies, glucose production was decreased by approximately 70% in control rats and 28-53% in overfed rats. Similarly, leptin infusion doubled the contribution of gluconeogenesis to glucose output in control rats but failed to modify gluconeogenesis in overfed animals. These findings demonstrate a paradoxical and rapid collapse of the leptin system in response to nutrient excess. This partial failure is tightly coupled with the onset of insulin resistance.

Glucosamine-induced inhibition of liver glucokinase impairs the ability of hyperglycemia to suppress endogenous glucose production.

Although the kinetic characteristics of hepatic glucokinase (GK) suggest its potential role as the hepatic "glucose sensor," its impact on the regulation of in vivo hepatic glucose production (HGP) is still controversial. Since decreased GK activity has been linked to experimental and human diabetes, we examined whether a moderate and transient inhibition of GK activity diminishes the ability of hyperglycemia to suppress HGP. We first determined the concentration of the competitive inhibitor, glucosamine (GlcN), which decreases hepatic GK activity by approximately 60% in vitro. GlcN was then infused into conscious rats to achieve a similar inhibition of the in vivo GK activity (plasma GlcN levels = approximately 2 mmol/l; rats infused with saline served as control, n = 20). To maintain equal plasma insulin and glucagon concentrations throughout the studies, somatostatin and insulin (basal replacement) were infused for 4 h. [3-(3H)]-glucose and [U-(14C)]-lactate were infused to measure HGP, gluconeogenesis, and glucose cycling (GC) during 2 h of euglycemia (glucose approximately 8 mmol/l) followed by 2 h of hyperglycemia (glucose approximately 18 mmol/l). Our results support the notion that hepatic GK activity is indeed decreased by GlcN in vivo. In fact, in response to hyperglycemia the "direct" pathway of hepatic glucose-6-phosphate (G-6-P) formation was approximately 40% lower with GlcN compared with saline infusion (37 +/- 3 vs. 63 +/- 3%; P < 0.001). Furthermore, while hyperglycemia stimulated GC by approximately 2.5-fold during saline infusion (from 3.0 +/- 0.6 to 7.7 +/- 1.4 mg.kg-1.min-1, P < 0.001, euglycemia vs. hyperglycemia), this increase was blunted in the presence of GlcN (4.6 +/- 0.6 mg.kg-1.min-1, P = NS). Finally, in the presence of GlcN, the hepatic concentration of G-6-P was decreased by approximately 40% compared with saline (234 +/- 38 and 390 +/- 24 nmol/g, P < 0.01). During the euglycemic studies, HGP was similar (12.6 +/- 0.6 and 11.3 +/- 0.2 mg .kg-1.min-1 with GlcN or saline infusion, respectively). However, while hyperglycemia per se suppressed HGP by approximately 65%, HGP was inhibited by approximately 38% and it was approximately twofold higher than in the saline-infused rats (7.8 +/- 0.8 and 4.0 +/- 0.3 mg.kg-1.min-1, P < 0.01) in the presence of GlcN-induced inhibition of hepatic GK. This increase in HGP was largely accounted for by the decreased inhibition of hepatic net glycogenolysis by hyperglycemia (3.3 +/- 0.8 and 1.1 +/- 0.3 mg.kg-1.min-1 with GlcN or saline infusion, respectively, P < 0.01). We conclude that intact GK activity is required for the normal suppression of HGP by hyperglycemia and its impairment may contribute to increased HGP in experimental and human diabetes.

Induction of hepatic glucose-6-phosphatase gene expression by lipid infusion.

The distal enzymatic step in the process of glucose output is catalyzed by the glucose-6-phosphatase (Glc-6-Pase) complex. The recently cloned catalytic unit of this complex has been shown to be regulated by insulin, dexamethasone, cAMP, and glucose. Using a combination of intralipid and/or nicotinic acid infusions and a pancreatic clamp technique, we maintained plasma free fatty acids (FFAs) at three different levels (0.26 +/- 0.07, 0.56 +/- 0.09, and 1.59 +/- 0.12 mmol/l) in the presence of well-controlled hormonal and metabolic conditions. An increase in the plasma FFA concentration within the physiological range caused a rapid, greater than threefold increase in the mRNA and protein levels of the catalytic subunit of Glc-6-Pase in the liver. These data indicate that the in vivo gene expression of Glc-6-Pase in the liver is regulated by circulating lipids independent of insulin and thus that prolonged hyperlipidemia may contribute to the increased production of glucose via increased expression of this protein.

Increased hexosamine availability similarly impairs the action of insulin and IGF-1 on glucose disposal.

Prolonged glucosamine (GlcN) infusion increases the skeletal muscle hexosamine concentration and induces peripheral insulin resistance in conscious rats. IGF-1 and insulin share common steps in signal transduction, and the action of IGF-1 on carbohydrate metabolism is preserved in certain insulin-resistant states. In our study, we attempted to delineate whether increased GlcN availability also impairs the effects of IGF-1 on glucose uptake (Rd), glycolysis, and glycogen synthesis. We performed euglycemic IGF-1 (5 and 15 microg x kg(-1) x min(-1)) and insulin (3 and 18 mU mg x kg(-1) x min(-1)) clamp studies at 0-2 h and 5-7 h in conscious rats (n = 44) during saline or GlcN infusions. GlcN infusion raised plasma GlcN levels to approximately 2.0 mmol/l and skeletal muscle uridinediphospho-n-acetylglucosamine to 80-150 nmol/g (approximately three- to fivefold over basal). During physiological hyperinsulinemia (3 mU x kg(-1) x min(-1), plasma insulin approximately 50 microU/ml), GlcN infusion caused comparable decreases in Rd (15.7 +/- 1.0 [5-7 h] vs. 21.7 +/- 2.3 [0-2 h] mg x kg(-1) x min(-1); P < 0.01) and glycogen synthesis (5.4 +/- 0.5 [5-7 h] vs. 10.4 +/- 1.9 [0-2 h] mg x kg(-1) x min(-1); P < 0.005). Furthermore, GlcN markedly decreased Rd by 7.8 +/- 1.2 mg x kg(-1) x min(-1) (18.7 +/- 0.7 [5-7 h] vs. 26.5 +/- 1.3 [0-2 h] mg x kg(-1) x min(-1); P < 0.001 vs. control) during IGF-1 (5 microg x kg(-1) x min(-1)) clamp studies. This decline was associated with a 26% decrease in the steady-state concentration of skeletal muscle Glc-6-P (286 +/- 45 vs. 386 +/- 36 nmol/g; P < 0.01) and was primarily caused by impaired glycogen synthesis (6.7 +/- 0.5 [5-7 h] vs. 13.9 +/- 0.9 [0-2 h] mg x kg(-1) x min(-1); P < 0.005). The effects of GlcN infusion on glucose disposal (percentage decrease in Rd) were correlated (r2 = 0.803; P < 0.01) with the skeletal muscle concentration of UDP-GlcNAc. To investigate whether IGF-1 can overcome GlcN-induced insulin resistance, GlcN and insulin (18 mU x kg(-1) x min(-1)) were infused for 7 h during euglycemic clamps, and IGF-1 (15 microg x kg(-1) x min(-1)) was superimposed during the final 2 h. GlcN infusion induced severe impairment of insulin action on Rd (39.4 +/- 3.2 [4-5 h] vs. 49.8 +/- 3.6 [1-2 h] mg x kg(-1) x min(-1); P < 0.05), which the addition of IGF-1 failed to improve (35.9 +/- 2.3 [6-7 h] vs. 39.4 +/- 3.2 [4-5 h] mg x kg(-1) x min(-1); P > 0.1). In summary, GlcN induced severe resistance to the actions of both insulin and IGF-1 on glucose uptake and glycogen synthesis, and IGF-1 was unable to overcome GlcN-induced insulin resistance. Thus, it is likely that GlcN causes peripheral insulin resistance acting at a site common to both IGF-1 and insulin signaling pathways.

Surgical removal of visceral fat reverses hepatic insulin resistance.

We directly examined whether visceral fat (VF) modulates hepatic insulin action by randomizing moderately obese (body wt approximately 400 g) Sprague-Dawley rats to either surgical removal of epididymal and perinephric fat pads (VF-; n = 9) or a sham operation (VF+; n = 11). Three weeks later, total VF was fourfold increased (8.5 +/- 1.2 vs. 2.1 +/- 0.3 g, P < 0.001) in the VF+ compared with the VF- group, but whole-body fat mass (determined using 3H2O) was not significantly different. The rates of insulin infusion required to maintain plasma glucose levels and basal hepatic glucose production in the presence of hepatic-pancreatic clamp were markedly decreased in VF- compared with VF+ rats (0.57 +/- 0.02 vs. 1.22 +/- 0.19 mU x kg(-1) x min(-1), P < 0.001). Similarly, plasma insulin levels were more than twofold higher in the VF+ group (P < 0.001). The heightened hepatic insulin sensitivity is supported by the decrease in gene expression of both glucose-6-phosphatase and PEPCK and by physiological hyperinsulinemia in VF- but not VF+ rats. The improvement in hepatic insulin sensitivity in VF- rats was also supported by a approximately 70% decrease in the plasma levels of insulin-like growth factor binding protein-1, a marker of insulin's transcription regulation in the liver. The removal of VF pads also resulted in marked decreases in the gene expression of tumor necrosis factor-alpha (by 72%) and leptin (by 60%) in subcutaneous fat. We conclude that visceral fat is a potent modulator of insulin action on hepatic glucose production and gene expression.

Physiological increase in plasma leptin markedly inhibits insulin secretion in vivo.

The demonstration of leptin receptors on the pancreatic beta-cells suggests the possibility of direct actions of leptin on insulin secretion. In vitro studies on islets or perfused pancreas and beta-cell lines produced inconsistent results. We performed an in vivo study to distinctly examine whether leptin has an effect on glucose-stimulated insulin secretion. Young chronically catheterized Sprague-Dawley rats (n = 28) were subjected to a 4-h hyperglycemic clamp study (approximately 11 mmol/l). At minute 120 to 240, rats were assigned to receive either saline or leptin (0.1, 0.5, and 5 microg x kg(-1) x min) infusion. Leptin decreased plasma insulin levels abruptly, and an approximately twofold decrease in plasma insulin levels compared with saline control was sustained over the 2 h of the study (14.8 +/- 5.8 vs. 34.8 +/- 2.6 ng/ml with leptin and saline infusion, respectively, P < 0.001). Moreover, a dose-dependent decrease in plasma insulin levels was noted (r = -0.731, P < 0.01). Since milrinone, an inhibitor of cAMP phosphodiesterase (PDE) 3, did not reverse the effect of leptin on glucose-induced insulin secretion, its action may be independent of PDE3. These findings suggest that acute physiological increase in plasma leptin levels acutely and significantly inhibits glucose-stimulated insulin secretion in vivo. The site of leptin effects on insulin secretion remains to be determined.

Dose-dependent effects of glyburide on insulin secretion and glucose uptake in humans.

OBJECTIVE: To examine the relationship between plasma glyburide concentrations (0, 50, 100, 200, 400, and 800 nM) and the insulin response and glucose metabolism during euglycemic (4.6 +/- 0.1 mM) and hyperglycemic (11.6 +/- 0.2 mM) conditions. RESEARCH DESIGN AND METHODS: Nine healthy subjects participated in the study. Steady-state plasma glyburide concentrations were achieved by primed continuous intravenous infusion of glyburide. RESULTS: During both euglycemia and hyperglycemia, glyburide enhanced insulin secretion and glucose disposal only to drug levels of 100-200 nM corresponding to an oral dose less than or equal to 10 mg. CONCLUSIONS: The data suggest that glyburide (and probably other sulfonylureas), operates within a narrow range of plasma concentrations (50-200 nM), which can be achieved with very low doses of the drug. It remains to be shown whether the threshold of maximal effect also in clinical practice is achieved with lower sulfonylurea doses than that currently used.

Mechanism of metformin action in obese and lean noninsulin-dependent diabetic subjects.

The effect of metformin on glucose metabolism was examined in eight obese (percent ideal body weight, 151 +/- 9%) and six lean (percent ideal body weight, 104 +/- 4%) noninsulin-dependent diabetic (NIDD) subjects before and after 3 months of metformin treatment (2.5 g/day). Fasting plasma glucose (11.5-8.8 mmol/L), hemoglobin-A1c (9.8-7.7%), oral glucose tolerance test response (20.0-17.0 mmol/L; peak glucose), total cholesterol (5.67-4.71 mmol/L), and triglycerides (2.77-1.52 mmol/L) uniformly decreased (P less than 0.05-0.001) after metformin treatment; fasting plasma lactate increased slightly from baseline (1.4 to 1.7 mmol/L; P = NS). Body weight decreased by 5 kg in obese NIDD subjects, but remained constant in lean NIDD. Basal hepatic glucose production declined in all diabetics from 83 to 61 mg/m2.min (P less than 0.01), and the decrease correlated (r = 0.80; P less than 0.01) closely with the fall in fasting glucose concentration. Fasting insulin (115 to 79 pmol/L) declined (P less than 0.05) after metformin. During a 6.9 mmol/L hyperglycemic clamp, glucose uptake increased in every NIDD subject (113 +/- 15 to 141 +/- 12 mg/m2.min; P less than 0.001) without a change in the plasma insulin response. During a euglycemic insulin clamp, total glucose uptake rose in obese NIDD subjects (121 +/- 10 to 146 +/- 9 mmol/m2.min; P less than 0.05), but decreased slightly in lean NIDD (121 +/- 10 to 146 +/- 0.5; P = NS). Hepatic glucose production was suppressed by more than 80-90% in all insulin clamp studies before and after metformin treatment. In conclusion, metformin lowers the fasting plasma glucose and insulin concentrations, improves oral glucose tolerance, and decreases plasma lipid levels independent of changes in body weight. The improvement in fasting glucose results from a reduction in basal hepatic glucose production. Metformin per se does not enhance tissue sensitivity to insulin in NIDD subjects. The improvement in glucose metabolism under hyperglycemic, but not euglycemic, conditions suggests that metformin augments glucose-mediated glucose uptake. Metformin has no stimulatory effect on insulin secretion.

Lack of suppression of insulin secretion by hyperinsulinemia in a patient with an insulinoma.

The regulation of insulin secretion in patients with insulinoma is known to be abnormal. For example, physiological and pharmacological stimuli often fail to stimulate insulin in such patients. Recently, insulin has been found to inhibit its own secretion in normal subjects. To determine if insulin has this effect in patients with insulinoma, we infused insulin at rates of 1 and 10 mU/kg X min in such a patient and in eight normal subjects. Euglycemia was maintained by the euglycemic glucose clamp technique, and endogenous insulin secretion was estimated by measuring plasma C-peptide levels. In the normal subjects, plasma C-peptide declined from 1.60 +/- 0.22 (+/- SEM) to 1.16 +/- 0.17 and 0.82 +/- 0.11 ng/ml during the low and high dose insulin infusions, respectively, indicating 27% (P less than 0.01) and 48% (P less than 0.001) decreases in endogenous insulin secretion at moderately elevated and extremely elevated insulin levels, respectively. In the insulinoma patient, plasma C-peptide was 2.6 ng/ml basally, did not change during the low dose insulin infusion, and rose to 3.4 ng/ml during the high dose insulin infusion. We conclude that the feedback regulation of insulin secretion by insulin that occurs in normal subjects is absent in insulinoma patients. This finding could have pathophysiological and possibly diagnostic significance.

Insulin effects on glucose and potassium metabolism in vivo: evidence for selective insulin resistance in humans.

UNLABELLED: The effects of insulin on in vivo glucose use and potassium uptake in healthy humans are well documented. However, the interrelationship between these two processes is not fully defined. In order to characterize it, we have used the euglycemic clamp technique on six normal volunteers, two patients with acanthosis nigricans and insulin resistance (AN), and one patient with idiopathic nonazotemic hyperkalemia (HK). In the basal state, all patients had normal fasting blood sugar, the AN patients had fasting hyperinsulinemia (600% of controls), and the HK patient had an elevated plasma potassium level of 5.1 mmol/L (n = 4.2 +/- 0.2 mmol/L). During low dose (1 mU/kg.min), and high dose (10 mU/kg.min) insulin infusions, normals used glucose at a rate of 220 +/- 10 and 470 +/- 20 mg/M2.min, respectively. The HK patient had a normal glucose use at both infusion rates, but the AN patients had a 20% decrease of glucose use compared to normals at the two infusion rates. In normal patients, plasma potassium fell by 0.7 and 1.4 mmol/L at the end of the two infusion periods, respectively. AN patients had a similar fall in potassium, but the HK patient displayed no change in plasma potassium levels during a low dose insulin infusion, and only a 0.6 mmol/L drop during the high dose insulin infusion. These results indicate that: 1) patients with AN are resistant to insulin action on glucose use, 2) AN patients have a normal response to insulin on potassium uptake, 3) HK is a patient with normal response to insulin on glucose use, and 4) this patient is resistant to insulin action on potassium uptake. IN CONCLUSION: 1) we have demonstrated the independence of insulin action on glucose and potassium uptake in vivo, 2) we documented the existence of selective insulin resistance in the above patients, 3) we speculate, that in patients with a normal response to insulin on one parameter of its actions, and subnormal response on another parameter, a postreceptor defect rather than a receptor abnormality must exist.

Caloric restriction, body fat and ageing in experimental models.

Caloric restriction in animal models delays many age-related pathological conditions. Ageing rats have characteristically increased body weight, fat mass and a specific body fat distribution. This report will focus on the potential cause-effect relationship between increased fat mass and accelerated ageing. In humans, increased fat mass (obesity), and in particular increases in abdominal obesity as a result of deposition of visceral fat, are associated with the metabolic syndrome of ageing. This syndrome is associated with hyperinsulinaemia, dyslipidaemia, type 2 diabetes mellitus, atherosclerosis, hypercoagulability and hypertension. Fat tissue, however, plays a major role by secreting multiple metabolically active factors, which are potentially responsible for the development of insulin resistance. This article will review various experimental models (in animals) used to prevent insulin resistance of ageing by decreasing fat mass, and in particular, decreasing visceral fat. We suggest that this decrease in fat mass and its beneficial repercussions observed in ageing animal models may apply also to human ageing and its related pathology.