Nonalcoholic steatohepatitis: association of insulin resistance and mitochondrial abnormalities.

BACKGROUND AND AIMS: The pathogenesis of nonalcoholic steatohepatitis (NASH) is unknown. We tested the hypothesis that NASH is associated with 2 defects: (1) peripheral insulin resistance, which increases lipolysis, delivery of free fatty acids (FFA) to the liver, and hepatic fatty acid beta oxidation, thereby creating oxidative stress; and (2) an abnormality within the hepatocytes that might render them more susceptible to injury from oxidative stress. METHODS: The hypothesis was tested by evaluation of (1) insulin resistance by a 2-step hyperinsulinemic (10 and 40 mU. m(-2). min(-1)) euglycemic clamp; (2) insulin effects on lipolysis by enrichment of [U-(13)C]glycerol; (3) frequency and severity of structural defects in hepatocyte mitochondria in vivo; (4) fatty acid beta oxidation from serum [beta-OH butyrate], release of water-soluble radioactivity from (3)H-palmitate by cultured fibroblasts and urinary dicarboxylic acid excretion; and (5) hepatic lipid peroxidation by immunohistochemical staining for 3-nitrotyrosine (3-NT). Subjects with NASH (n = 6-10 for different studies) were compared with those with fatty liver (n = 6) or normal controls (n = 6). RESULTS: NASH and fatty liver were both associated with insulin resistance, with mean glucose infusion rates (normal/fatty liver/NASH) of step 1, 4.5/1.6/0.9; step 2, 9.5/7.7/4.5 (P < 0.03 for both steps). Although baseline rates of glycerol appearance were higher in those with NASH than in those with fatty liver (means, 14.6 vs. 21.6 micromol. kg(-1). min(-1); P < 0.05), neither group significantly suppressed glycerol appearance at insulin infusion rates of 10 mU. m(-2). min(-1). NASH was associated with loss of mitochondrial cristae and paracrystalline inclusions in 9 of 10 subjects, compared with 0 of 6 subjects with fatty liver. However, no evidence of a generalized defect in fatty acid beta oxidation was noted in any group. Also, mean [beta-OH butyrate] was highest in those with NASH (means, 90 vs. 110 vs. 160 micromol/L; P < 0.04). Increased staining for 3-NT was present in fatty liver, and even greater staining was seen in NASH. CONCLUSIONS: These data indicate that peripheral insulin resistance, increased fatty acid beta oxidation, and hepatic oxidative stress are present in both fatty liver and NASH, but NASH alone is associated with mitochondrial structural defects.

Effects of fructose and troglitazone on phospholipid fatty acid composition in rat skeletal muscle.

Skeletal muscle phospholipid fatty acid (PLFA) composition is associated with insulin sensitivity in animal models and in man. However, it is not clear whether changes in insulin sensitivity cause a change in PLFA composition or vice versa. The present studies have examined the effects of agents known to increase or decrease insulin sensitivity on PLFA composition of the major phospholipids, phosphatidylcholine (PC) and phosphatidylethanolamine (PE), in soleus and extensor digitorum longus muscle. Four groups of Sprague-Dawley rats--control, 0.2% troglitazone (Tgz), 60% fructose fed, and fructose + Tgz--were treated for 3 wk. Fructose feeding was associated with a decrease in muscle membrane polyunsaturated fatty acids (PUFA) and n-3 fatty acids in both PC and PE. Administration of Tgz alone resulted in an increase in liver (3.75 +/- 0.93 to 6.93 +/- 1.00 micromol/min/mg tissue, P < 0.05) and soleus muscle (0.34 +/- 0.03 to 0.67 +/- 0.09 micromol/min/mg, P < 0.01) elongase activity, which would be expected to increase membrane PUFA. However, Tgz decreased PLFA associated with greater insulin sensitivity (e.g., PUFA and n-3 fatty acids) and increased PLFA associated with decreased insulin sensitivity (16:0 and n-6 fatty acids) in both PC and PE. Co-administration of fructose and Tgz did not reverse the decrease in PUFA observed with fructose alone. We conclude that the improvement in insulin sensitivity reported with Tgz is associated with an apparently paradoxical effect to decrease PUFA and n-3 PLFA composition in rat skeletal muscle. These studies suggest that Tgz-mediated increases in insulin sensitivity do not result in improved PLFA composition.

Glucose-6-phosphatase flux in vitro is increased in type 2 diabetes.

Despite the effects of hyperinsulinemia and hyperglycemia, 2 factors known to inhibit endogenous glucose production (EGP) in nondiabetic subjects, increased EGP is a consistent feature of type 2 diabetes. Recent studies have suggested that increased glucose-6-phosphatase (G6Pase) and/or decreased glucokinase (GK) may explain the increase in EGP. However, no studies to date have clearly established this relationship in type 2 diabetes. The present studies were designed to determine rates of EGP and the activities of G6Pase and GK in obese patients scheduled for gastric bypass surgery. The study group consisted of 14 obese nondiabetic subjects and 13 patients with type 2 diabetes (BMI 53.7 +/- 2.4 vs. 50.1 +/- 1.6 kg/m2). Rates of EGP were determined after an overnight fast with a 4-h infusion of [6,6]-D-glucose, and they were significantly higher in the type 2 diabetic patients (85.9 +/- 10.0 vs. 137.8 +/- 14.4 mg x m(-2) x min(-1), P < 0.001) despite greater plasma glucose (5.1 +/- 0.1 vs. 12.0 +/- 1.1 mmol/l) and similar insulin concentrations (130.8 +/- 19.8 vs. 112.8 +/- 16.2 pmol/l, NS). Moreover, resistance to insulin-induced suppression of EGP was observed in the patients with type 2 diabetes when insulin concentrations were increased from approximately 120 to 180 pmol/l. Hepatic G6Pase activity determined from freshly isolated microsomes was significantly increased in the type 2 diabetic patients compared with the obese control subjects (0.16 +/- 0.02 vs. 0.09 +/- 0.01 micromol x min(-1) x mg(-1) protein, P < 0.02), whereas levels of GK were decreased (1.20 +/- 0.16 vs. 2.01 +/- 0.01 micromol x min(-1) x mg(-1) protein, P < 0.01). Net flux through G6Pase was significantly increased in type 2 diabetic patients (P < 0.01). We conclude that increased EGP is mediated in part by increased G6Pase flux in type 2 diabetes.

Changes in phosphatidylcholine fatty acid composition are associated with altered skeletal muscle insulin responsiveness in normal man.

The fatty acid composition of skeletal muscle cell membrane phospholipids (PLs) is known to influence insulin responsiveness in man. We have recently shown that the fatty acid composition of phosphatidylcholine (PC), and not phosphatidylethanolamine (PE), from skeletal muscle membranes is of particular importance in this relationship. Efforts to alter the PL fatty acid composition in animal models have demonstrated induction of insulin resistance. However, it has been more difficult to determine if changes in insulin sensitivity are associated with changes in the skeletal muscle membrane fatty acid composition of PL in man. Using nicotinic acid (NA), an agent known to induce insulin resistance in man, 9 normal subjects were studied before and after treatment for 1 month. Skeletal muscle membrane fatty acid composition of PC and PE from biopsies of vastus lateralis was correlated with insulin responsiveness using a 3-step hyperinsulinemic-euglycemic clamp. Treatment with NA was associated with a 25% increase in the half-maximal insulin concentration ([ED50] 52.0 +/- 7.5 to 64.6 +/- 9.0 microU/mL, P < .05), consistent with decreased peripheral insulin sensitivity. Significant changes in the fatty acid composition of PC, but not PE, were also observed after NA administration. An increase in the percentage of 16:0 (21% +/- 0.3% to 21.7% +/- 0.4%, P < .05) and decreases in 18:0 (6.2% +/- 0.5% to 5.1% +/- 0.4%, P = .01), long-chain n-3 fatty acids (1.7% +/- 0.2% to 1.4% +/- 0.1%, P < .01), and total polyunsaturated fatty acids ([PUFAs] 8.7% +/- 0.8% to 8.0% +/- 0.8%, P < .05) are consistent with a decrease in fatty acid length and unsaturation in PC following NA administration. The change in ED50 was significantly correlated with the change in PUFAs (r = -.65, P < .05). These studies suggest that the induction of insulin resistance with NA is associated with changes in the fatty acid composition of PC in man.

Skeletal muscle phosphatidylcholine fatty acids and insulin sensitivity in normal humans.

The fatty acid composition of skeletal muscle membrane phospholipids (PL) is known to influence insulin responsiveness in humans. However, the contribution of the major PL of the outer (phosphatidylcholine, PC) and inner (phosphatidylethanolamine, PE) layers of the sarcolemma to insulin sensitivity is not known. Fatty acid composition of PC and PE from biopsies of vastus lateralis from 27 normal men and women were correlated with insulin sensitivity determined by the hyperinsulinemic euglycemic clamp technique at insulin infusion rates of 0.4, 1.0, and 10.0 mU . kg-1 . min-1. Significant variation in the half-maximal insulin concentration (ED50) was observed in the normal volunteers (range 24.0-146.0 microU/ml), which correlated directly with fasting plasma insulin (r = 0.75, P < 0.0001). ED50 was inversely correlated with the degree of membrane unsaturation (C20-C22 polyunsaturated fatty acids; r = 0. 58, P < 0.01) and directly correlated with fatty acid elongation (ratio of 16:0 to 18:0, r = 0.45, P < 0.05) in PC. However, no relationship between fatty acid composition and insulin sensitivity was observed in PE (NS). These studies suggest that the fatty acid composition of PC may be of particular importance in the relationship between fatty acids and insulin sensitivity in normal humans.

Phospholipid fatty acid composition in type I and type II rat muscle.

The fatty acid composition of the membrane phospholipids phosphatidylcholine (PC) and phosphatidylethanolamine in insulin-sensitive Type I (soleus) and insulin-resistant Type II (EDL) muscle is not known. In the present studies, soleus and EDL muscles were removed from 250-300 g Sprague-Dawley rats, and the fatty acid composition of total and individual phospholipid (PL) species was quantitated. As expected, triglyceride content was increased twofold in soleus muscle. No quantitative differences in the individual PL species or cholesterol content were found between the two muscles. However, a striking difference in PL fatty acid composition was observed in the PC fraction. An increase in 16:0 with decreases in 18:0, 18:1, 22:5n-3, and 22:6n-3 (P < 0.001 for each) was observed in the PC fraction of EDL compared to that from soleus, consistent with reduced elongation of PC fatty acids. Inhibition of fatty acid oxidation with the carnitine palmitoyl transferase-1 inhibitor, etomoxir, did not alter the fatty acid pattern in either muscle. We conclude that an alteration in PL fatty acid composition consistent with reduced elongation of both saturated and unsaturated fatty acids is observed in Type II muscle. The restriction of these alterations to the PC fraction has important implications.

Dehydroepiandrosterone and body fat.

Dehydroepiandrosterone sulfate (DHEA-S) is the most abundant circulating adrenal steroid in man, yet its physiologic role and that of its parent compound DHEA are unknown. Age-related decreases in DHEA in association with increases in obesity, insulin resistance, and atherosclerosis are well known. Recent investigations in lower mammals (which do not secrete DHEA) have suggested that DHEA (or its metabolites) may function as an antiobesity agent in these models of obesity independent of food intake. Proposed mechanisms for the decrease in fat mass and lower weight gain when DHEA is given orally include increases in futile cycling and peroxisomal beta-oxidation and decreases in de novo lipogenesis. Alterations in the availability of reducing equivalents for lipid synthesis do not appear to explain this decrease. Changes in pancreatic insulin secretion or insulin sensitivity may also be responsible for some of these effects. Studies in humans have failed to demonstrate a beneficial effect of DHEA on body composition or energy expenditure at either pharmacologic or physiologic replacement doses for 1-3 months. Administration of DHEA to men or women has also not been shown to alter insulin sensitivity as measured by the minimal model or the euglycemic clamp technique. The effect of DHEA on peroxisomal beta-oxidation and de novo lipogenesis is not known. We conclude that a significant role for DHEA in the pharmacologic treatment of human obesity is unlikely.

Loss of hepatic autoregulation after carbohydrate overfeeding in normal man.

To determine the effect of increased glycogen stores on hepatic carbohydrate metabolism, 15 nondiabetic volunteers were studied before and after 4 d of progressive overfeeding. Glucose production and gluconeogenesis were assessed with [2-3H] glucose and [6-14C] glucose (Study I, n = 6) or [3-3H] glucose and [U-14C]-alanine (Study II, n = 9) and substrate oxidation was determined by indirect calorimetry. Overfeeding was associated with significant (P < 0.01) increases in plasma glucose (4.97 +/- 0.10 to 5.09 +/- 0.11 mmol/liter), insulin (18.8 +/- 1.5 to 46.6 +/- 10.0 pmol/liter) and carbohydrate oxidation (4.7 +/- 1.4 to 18.0 +/- 1.5 mumol.kg-1.min-1) and a decrease in lipid oxidation (1.2 +/- 0.2 to 0.3 +/- 0.1 mumol.kg-1.min-1). Hepatic glucose output (HGO) increased in Study I (10.2 +/- 0.5 to 13.1 +/- 0.9 mumol.kg-1.min-1, P < 0.01) and Study II (11.17 +/- 0.67 to 13.33 +/- 0.83 mumol.kg-1.min-1, P < 0.01), and gluconeogenesis decreased (57.6 +/- 6.4 to 33.4 +/- 4.9 mumol/min, P < 0.01), indicating an increase in glycogenolysis. The increase in glycogenolysis was only partly compensated by an increase in glucose cycle activity (2.2 +/- 0.2 to 3.4 +/- 0.4 mumol.kg-1.min-1, P < 0.01) and the fall in gluconeogenesis, thus resulting in increased HGO. The suppression of gluconeogenesis despite increased lactate and alanine (glycerol was decreased) was associated with decreased free fatty acid (FFA) oxidation and negligible FFA enhanced gluconeogenesis. These studies suggest that increased liver glycogen stores alone can overwhelm normal intrahepatic mechanisms regulating carbohydrate metabolism resulting in increased HGO in nondiabetic man.

Amylin/insulin secretory ratios in morbidly obese man: inverse relationship with glucose disappearance rate.

Amylin/insulin secretory ratios were determined in nine morbidly obese subjects consenting to portal venous catheterization at the time of gastric bypass surgery. By subtracting recirculating insulin and amylin concentrations (arterial values) from portal venous hormone concentrations, instantaneous amylin/insulin secretory ratios could be determined before and after iv glucose administration. Baseline portal venous amylin levels were 32% higher than peripheral concentrations (7.3 +/- 0.8 vs. 5.6 +/- 0.6 pmol/L). Portal venous amylin and insulin concentrations peak 90 s after the initiation of a 2-min glucose infusion. When instantaneously secreted amylin and insulin were compared at each of the eight time points, a highly significant correlation was observed in seven of the nine subjects. However, large interindividual variations in amylin/insulin secretory ratios were observed, with molar ratios from 0.2-1.6%. The amylin/insulin secretory ratios calculated at the time of surgery varied inversely (r = -0.89; P < 0.001) with glucose disappearance rates obtained 5-7 months later after 19- to 29-kg weight loss. These data corroborate those obtained from animal studies and indicate that amylin and insulin are cosecreted in man. Despite evidence for cosecretion of amylin and insulin, the large intersubject variation in amylin/insulin secretory ratios and its inverse correlation with glucose disappearance rates suggest a constitutional factor that may either play a role in the pathogenesis of carbohydrate intolerance or result from it.

Suppression of gluconeogenesis after a 3-day fast does not deplete liver glycogen in patients with NIDDM.

To determine the effect of inhibition of gluconeogenesis on liver glycogen stores in patients with non-insulin-dependent diabetes mellitus (NIDDM) after a 3-day fast, 10% ethanol (EtOH) was administered intravenously to nine obese patients with NIDDM and six obese nondiabetic subjects. Rates of glucose appearance (3-[3H]glucose) and [U-14C]alanine incorporation into glucose (alanine gluconeogenesis [Ala-GNG]) were determined before and during EtOH administration, and residual glycogen stores were assessed by the incremental glucose response to glucagon (glucoseAUC). Hepatic glucose output (HGO) was closely correlated with plasma glucose levels (r = 0.71, P < 0.001) after the 3-day fast and was significantly greater in the diabetic compared with the nondiabetic subjects (13.8 +/- 1.4 vs. 7.6 +/- 0.6 mumol.kg-1 FFM.min-1, P < 0.01). During the 120-min EtOH infusion, Ala-GNG fell by more than 50% in both groups and did not increase after intravenous glucagon administration. HGO fell modestly in both the diabetic and nondiabetic subjects during the first 30 min of EtOH infusion and stabilized thereafter. In contrast to Ala-GNG, HGO increased significantly after intravenous glucagon administration in both the diabetic and nondiabetic subjects, but the increase was significantly greater in the patients with NIDDM (P < 0.01). The glucose area under the curve in response to glucagon (glucoseAUC) was lower in the presence of EtOH than in its absence (14.9 +/- 7 vs. 68 +/- 15.6 mM/min, P < 0.01) in the obese nondiabetic subjects, which suggests a decrease in liver glycogen stores.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of corticosteroids on urinary ammonium excretion in humans.

This study was designed to examine the selective effects of glucocorticoid and mineralocorticoid classes of steroid hormones on urinary ammonium excretion in humans. In 22 10-day studies, normal male volunteers received either 9 alpha-fludrohydrocortisone or hydrocortisone, alone or with the receptor antagonist spironolactone or mifepristone. The small but significant increase in ammonium excretion noted with the administration of 9 alpha-fludrohydrocortisone was associated with a significant decrease in serum potassium. In contrast, a significantly larger increase in ammonium excretion was noted with hydrocortisone, without concomitant electrolyte changes. Spironolactone did not alter the effect on ammonium excretion by either corticosteroid, whereas mifepristone markedly blunted the hydrocortisone-induced increase in urinary ammonium excretion. It was concluded that glucocorticoids increase urinary ammonium excretion in humans and that this effect occurs through binding to the Type II (glucocorticoid) receptor rather than by cross-occupancy of the Type I (mineralocorticoid) receptor.

Insulin sensitivity of splanchnic and peripheral adipose tissue in vivo in morbidly obese man.

Epidemiologic studies demonstrate an association between increased waist to hip ratio ([WHR] android obesity, central obesity) and diabetes mellitus in man. To study the relative insulin sensitivity of splanchnic versus peripheral adipose tissue, portal vein catheterization via the collapsed umbilical vein was performed in 14 morbidly obese subjects at the time of surgery. Catheters were also placed in a peripheral artery and antecubital vein such that simultaneous arterio-venous (A-V) differences (glycerol, free fatty acids [FFA], and lactate) could be determined. After two baseline samples obtained 3 minutes apart, 25 g intravenous (i.v.) glucose (14 subjects) was administered over a 2-minute period, with samples being obtained every 5 minutes for 30 additional minutes. Arterial plasma glycerol levels decreased from 173.9 +/- 17.4 mumol/L at baseline to 89.1 +/- 7.6 mumol/L at 30 minutes (P < .01). Peripheral and splanchnic A-V glycerol differences were similar at baseline, but within 10 minutes after glucose administration the difference across the splanchnic area decreased by 52% and remained significantly less than that across the periphery (P < .01). Despite a 49% decrease in arterial plasma glycerol level, plasma FFA level decreased only 18.3% over the 30-minute period (942 +/- 74.8 to 770.0 +/- 76 mumol/L, NS). These studies in morbidly obese man (glycerol data) indicate a greater insulin sensitivity of splanchnic adipose tissue than of peripheral adipose tissue. Thus hypertrophy of fat in the splanchnic area might be an expected consequence of the hyperinsulinemia associated with insulin-resistant states.

Skeletal muscle lipids and glycogen mask substrate competition (Randle cycle).

The glucose-free fatty acid (FFA) cycle (Randle) was examined in soleus muscle, a red muscle with a high lipid oxidation rate, and extensor digitorum longus (EDL) muscle, a white muscle with a low lipid oxidation rate, using a carnitine palmethyltransferase (CPT-I) inhibitor as a probe. Exogenous palmitate by itself had little if any effect on glycolysis or glycogen accumulation in the two muscle types. The CPT-I inhibitor markedly decreased glycogen accumulation in both muscles (from fed rats), but increased glycolysis (lactate formation) and glucose oxidation to carbon dioxide only in the red muscle. When the muscles were made more dependent on FFA oxidation by prior fasting or exercise, the CPT-I stimulatory effect on glycolysis and glucose oxidation in white muscle was unmasked. In conclusion, the competition between lipid and carbohydrate utilization (Randle cycle) is easily demonstrated in both red and white muscle using a CPT-I inhibitor as a probe. The difficulties encountered in showing this competition in other studies using exogenous FFA may be explained by a combination of factors, including (1) low tissue lipid oxidation rates, (2) competition between exogenous and endogenous lipids such that provision of exogenous lipids fails to increase overall lipid oxidation, and (3) preferential utilization of exogenous glucose with glycogen sparing in the presence of FFA.

Evidence for dissociation of gluconeogenesis stimulated by non-esterified fatty acids and changes in fructose 2,6-bisphosphate in cultured rat hepatocytes.

In order to examine the role of fructose 2,6-bisphosphate (Fru-2,6-P2) in non-esterified-fatty-acid-stimulated gluconeogenesis, Fru-2,6-P2 levels were measured in cultured rat hepatocytes under conditions mimicking the fasted state. After addition of either 1.5 mM-palmitate or 10 nM-glucagon, [U-14C]lactate incorporation into glucose increased 2-fold, but only glucagon suppressed Fru-2,6-P2. Prevention of palmitate oxidation with a carnitine palmitoyltransferase-I inhibitor (2-bromopalmitate) diminished glucose production and Fru-2,6-P2 levels. Addition of exogenous glucose to the media increased Fru-2,6-P2 in a dose-related manner, which was further augmented by addition of palmitate. When Fru-2,6-P2 levels were examined in cells cultured under conditions mimicking the fed state (significantly higher basal Fru-2,6-P2 levels and lower glucose production), palmitate oxidation was associated with a significant fall in Fru-2,6-P2. In conclusion, the present studies have demonstrated a dissociation between fatty-acid-stimulated gluconeogenesis and changes in Fru-2,6-P2 in cultured rat hepatocytes. Further experiments suggest that the accumulation of intracellular hexose 6-phosphate as a result of fatty-acid-stimulated gluconeogenesis masks a putative inhibitory effect of fatty acids on Fru-2,6-P2 concentrations.

Dehydroepiandrosterone: the "missing link" between hyperinsulinemia and atherosclerosis?

A well-established epidemiologic association exists between hyperinsulinemia and macrovascular disease. However, the mechanism or mechanisms by which hyperinsulinemia promotes atherogenesis is unknown. Recent evidence indicates that the adrenal steroid dehydroepiandrosterone (DHEA) exerts multiple antiatherogenic effects and also suggests that hyperinsulinemia may reduce serum DHEA and DHEA-sulfate levels by decreasing production and enhancing metabolic clearance. We advance the hypothesis that hyperinsulinemia promotes macrovascular disease in part by reducing serum DHEA and DHEA-sulfate levels and illustrate how this may be the case in two clinical conditions characterized by hyperinsulinemic insulin resistance: aging and obesity.

The effect of hyperosmolarity on muscle glycogen accumulation.

Soleus (red) and extensor digitorum longus (white) muscles from Sprague Dawley rats were incubated with 6-14C-labelled glucose in normal and in hyperosmotic media. Hyperosmolarity decreased 6-14C-glucose incorporation into muscle glycogen in a dose dependent manner and increased glycolysis and glucose oxidation. Increased glycogenolysis rather than decreased glycogenesis was responsible for the reduction in labelled glycogen accumulation.

Evidence for increased liver glycogen in patients with noninsulin-dependent diabetes mellitus after a 3-day fast.

In order to assess hepatic glycogen stores in patients with noninsulin dependent diabetes mellitus (NIDDM) after a 3-day fast, the incremental glucose response to 1.0 mg iv glucagon (glucose area under the curve, glucoseAUC) was assessed in 19 obese diabetic subjects after an overnight (14 h) fast and again after a 3-day (64 h) fast. Results were compared to those of lean (n = 6) and obese (n = 15) nondiabetic subjects. During the fast, plasma glucose fell significantly in the lean (4.9 +/- 0.2 to 3.9 +/- 0.2 mmol/L), obese (5.1 +/- 0.1 to 4.2 +/- 0.2 mmol/L), and diabetic (14.7 +/- 0.7 to 10.3 +/- 1.0 mmol/L) subjects. However, in contrast to the fall in glucoseAUC observed in the lean (92.4 +/- 15.4 to 39.9 +/- 8.1 mmol min-1 L-1, P less than 0.02) and obese (64.4 +/- 11.1 to 48.4 +/- 9.4 mmol min-1 L-1) subjects, the glucoseAUC increased in diabetic subjects from 81.6 +/- 8.6 to 103.9 +/- 8.8 mmol min-1 L-1 during the fast, and was significantly greater than that of either the lean (P less than 0.001) or obese (P less than 0.001) nondiabetic subjects after the 64-h fast. Evidence that the glucose response to glucagon after a 64-h fast represents glycogenolysis and not gluconeogenesis was provided by studies in 10 additional subjects (5 obese nondiabetic subjects and 5 patients with NIDDM). Overall hepatic glucose output calculated from glucose kinetic data [( 3-3H]glucose) increased in diabetic and nondiabetic subjects during the first 30 min after glucagon administration and fell progressively thereafter. However, no increase in alanine gluconeogenesis (14C-alanine incorporation into glucose) was observed after glucagon administration in either subject group. The paradoxical accumulation of glycogen in the patients with NIDDM during the fast occurred despite basal rates of hepatic glucose output on the third day of the fast which were greater than those of obese nondiabetic subjects (9.0 +/- 1.2 vs. 5.6 +/- 0.5 mumol kg-1 min-1, P less than 0.05). A glycogen sparing action of increased gluconeogenesis is proposed as the explanation for the preservation of liver glycogen in patients with NIDDM.

Insulin inhibits adrenal 17,20-lyase activity in man.

Experimentally induced hyperinsulinemia reduces serum adrenal androgen levels in man, but does not alter cortisol secretion. To determine whether insulin might selectively inhibit adrenal androgen production by suppressing 17,20-lyase activity, ACTH-stimulated androgen secretion was assessed in 10 normal men after an insulin infusion (hyperinsulinemic-euglycemic clamp) or a control saline infusion. For the insulin clamp study, each man received a 2-U (14.4-nmol) insulin bolus dose, followed by a 2.0-mU/kg.min (14.4-pmol/kg.min) insulin infusion for 5 h. An average insulin level of 746 +/- 35 (+/- SE) pmol/L was achieved; serum glucose was maintained at 4.96 +/- 0.03 mmol/L. At the end of the insulin infusion, an ACTH stimulation test was performed, and serum steroid levels were determined 30 and 60 min later. Subjects returned 1-3 weeks later for control studies, during which 0.45% saline was infused at rates matched exactly to the rates of the dextrose and insulin infusions during the insulin clamp studies, and an ACTH stimulation test was performed after 5 h of saline infusion. After the insulin infusion, stimulation by ACTH resulted in a significant rise in the serum molar ratio of 17 alpha-hydroxyprogesterone to androstenedione (from 0.914 +/- 0.110 at zero time to 1.388 +/- 0.278 60 min after ACTH; P less than 0.05), whereas no change occurred in the ACTH-stimulated ratio of these steroids after the saline infusion (1.067 +/- 0.109 at zero time to 1.060 +/- 0.109 60 min after ACTH; P = NS). The insulin-induced change in this steroid ratio was due to a relative increase in precursor (17 alpha-hydroxyprogesterone) and decrease in product (androstenedione) responsiveness to ACTH. Similarly, insulin treatment resulted in a greater than 100% rise in the difference from baseline in the serum molar ratio of 17 alpha-hydroxypregnenolone to dehydroepiandrosterone 30-60 min after ACTH (P less than 0.004), whereas no change in this difference was observed after the saline infusion (P = 0.71). Again, the insulin-induced change in this steroid ratio was due to a relative increase in precursor (17 alpha-hydroxypregnenolone) and decrease in product (dehydroepiandrosterone) responsiveness to ACTH. Of note, insulin treatment altered neither cortisol responsiveness to ACTH nor 17 alpha-hydroxylase activity, as indicated by similar ACTH-stimulated responses in the serum molar ratio of progesterone to 17 alpha-hydroxyprogesterone after the insulin and saline infusions (P = 0.71). Hence, the results of this study indicate that the acute elevation of serum insulin levels into the high physiological range selectively inhibits adrenal 17,20-lyase activity in man.

In vivo evidence for hepatic autoregulation during FFA-stimulated gluconeogenesis in normal humans.

To examine the effect of increased gluconeogenesis [by increasing free fatty acids (FFA)] on hepatic glucose output (HGO) and on the first substrate (glucose) cycle, a primed continuous infusion of [2-3H]- and [6-14C]glucose was infused to isotopic steady state in 12 normal male volunteers after an overnight fast. Blood samples for the determination of glucose specific activity were obtained before and after an infusion of saline (n = 6) or 10% Intralipid and heparin (90 mU.kg-1.min-1, n = 6). Plasma FFA (593.3 +/- 74.5 to 971.1 +/- 127.1 mumol/l, P = 0.007) and glycerol (68.0 +/- 5.9 vs. 222.4 +/- 32.0 mumol/l, P = 0.002) increased during the lipid infusion, and beta-hydroxybutyrate levels rose from 0.24 +/- 0.12 to 0.50 +/- 0.17 mmol/l (P = 0.01). No change in plasma glucose, insulin, or glucagon levels was observed during the study, and levels of the gluconeogenic substrates alanine and lactate were also unchanged. Baseline rates of glucose cycling (rate of appearance of [2-3H]glucose minus rate of appearance of [6-14C]glucose) were similar in the two groups [1.44 +/- 0.33 vs. 1.33 +/- 0.44 mumol.kg-1.min-1, not significant (NS)] and did not change during either saline or lipid infusion, respectively. However, Cori cycle activity (the conversion of [6-14C]- to [1-14C]glucose) increased significantly from 0.59 +/- 0.19 to 1.28 +/- 0.19 mumol.kg-1.min-1 (P = 0.002) after FFA and glycerol levels had been increased, in marked contrast to the saline control (0.51 +/- 0.18 to 0.39 +/- 0.18 mumol.kg-1.min-1, NS).(ABSTRACT TRUNCATED AT 250 WORDS)