Effects of saxagliptin on ß-cell stimulation and insulin secretion in patients with type 2 diabetes.

AIM: To study the effect of dipeptidyl peptidase-4 (DPP-4) inhibition with saxagliptin on ß-cell function as reflected by the stimulated insulin secretion rate after an enteral glucose load in patients with type 2 diabetes. METHODS: Patients in this randomized, parallel-group, double-blind, placebo-controlled study were drug-naïve, aged 43-69 years, with baseline haemoglobin A1c (HbA1c) 5.9-8.1%. Twenty patients received saxagliptin 5 mg once daily; 16 received placebo. Patients were assessed at baseline and week 12 by intravenous hyperglycaemic clamp (0-180 min, fasting state), and intravenous-oral hyperglycaemic clamp (180-480 min, postprandial state) following oral ingestion of 75 g glucose. Primary and secondary endpoints were percent changes from baseline in insulin secretion during postprandial and fasting states, respectively. Insulin secretion was calculated by C-peptide deconvolution. RESULTS: After 12 weeks, saxagliptin significantly increased insulin secretion percent change from baseline during the postprandial state by an 18.5% adjusted difference versus placebo (p = 0.04), an improvement associated with increased peak plasma concentrations of intact glucagon-like peptide-1 and glucose-dependent insulinotropic polypeptide. In the fasting state, saxagliptin significantly increased insulin secretion by a 27.9% adjusted difference versus placebo (p = 0.02). Saxagliptin also improved glucagon area under the curve in the postprandial state (adjusted difference -21.8% vs. placebo, p = 0.03). CONCLUSIONS: DPP-4 inhibition with saxagliptin improves pancreatic ß-cell function in postprandial and fasting states, and decreases postprandial glucagon concentration. Given the magnitude of enhancement of the insulin response in the fasting state, further study into the effect of DPP-4 inhibition on the ß-cell is warranted.

Inhibition of LDL oxidation in vitro but not ex vivo by troglitazone.

Diabetic subjects are at increased risk for developing coronary artery disease, in part because of increased oxidation of LDL, which promotes atherogenesis. Troglitazone, a new antidiabetic drug of the thiazolidinedione class, acts as an insulin sensitizer and improves hyperglycemia. Structurally, it contains a tocopherol moiety similar to vitamin E and has been shown to have antioxidant properties in vitro. Therefore, we evaluated whether troglitazone inhibited LDL oxidation both in vitro and in type 2 diabetic subjects ex vivo. Troglitazone inhibited oxidation of LDL induced by Cu2+ or 2'2'-azobis-2-amidinopropane hydrochloride (AAPH) with 50% inhibition at 1 micromol/l and 100% inhibition at 5-10 micromol/l troglitazone. The inhibition of LDL oxidation by troglitazone also was time dependent. In addition, troglitazone inhibited oxidation of 125I-labeled LDL and its subsequent uptake and degradation by macrophages. To determine whether troglitazone was incorporated into LDL particles or acted in the aqueous milieu, troglitazone was incubated overnight at 37 degrees C with LDL or plasma before LDL re-isolation. After re-isolation, LDL that was incubated with troglitazone was no longer protected from oxidation, compared with probucol-treated LDL, which remained protected. Further, [14C]troglitazone did not get incorporated into LDL. This suggests that troglitazone exerts its antioxidant effect in the aqueous milieu of LDL. Consistent with this was the observation that the lag phases of copper-induced conjugated diene formation, a measure of the susceptibility in vivo, was similar for subjects taking troglitazone (76 +/- 5 min, n = 9) to subjects not taking the drug (77 +/- 3 min, n = 11; NS). Thus, troglitazone may be of value as an aqueous-phase antioxidant in addition to its effect on glucose homeostasis.

Acute and chronic effects of insulin on leptin production in humans: Studies in vivo and in vitro.

This study was undertaken to investigate the changes in obesity (OB) gene expression and production of leptin in response to insulin in vitro and in vivo under euglycemic and hyperglycemic conditions in humans. Three protocols were used: 1) euglycemic clamp with insulin infusion rates at 40, 120, 300, and 1,200 mU / m / min carried out for up to 5 h performed in 16 normal lean individuals, 30 obese individuals, and 31 patients with NIDDM; 2) 64-to 72-h hyperglycemic (glucose 12.6 mmol/l) clamp performed on 5 lean individuals; 3) long-term (96-h) primary culture of isolated abdominal adipocytes in the presence and absence of 100 nmol/l insulin. Short-term hyperinsulinemia in the range of 80 to > 10,000 microU/ml had no effect on circulating levels of leptin. During the prolonged hyperglycemic clamp, a rise in leptin was observed during the last 24 h of the study (P < 0.001). In the presence of insulin in vitro, OB gene expression increased at 72 h (P < 0.01), followed by an increase in leptin released to the medium (P < 0.001). In summary, insulin does not stimulate leptin production acutely; however, a long-term effect of insulin on leptin production could be demonstrated both in vivo and in vitro. These data suggest that insulin regulates OB gene expression and leptin production indirectly, probably through its trophic effect on adipocytes.

Insulin aspart (B28 asp-insulin): a fast-acting analog of human insulin: absorption kinetics and action profile compared with regular human insulin in healthy nondiabetic subjects.

OBJECTIVE: To study the pharmacokinetic and pharmacodynamic profile of insulin aspart (a new fast-acting human insulin analog) after subcutaneous administration in the deltoid, abdominal, and thigh sites and to compare this profile with regular human insulin (Novolin; Novo Nordisk A/S, Copenhagen). RESEARCH DESIGN AND METHODS: A total of 20 healthy subjects were studied in a single-center six-period double-blind randomized crossover trial with 6 study days and a washout period of 1 week between each single daily dose of the trial drug. Subjects were randomized to receive a single dose of 0.2 U/kg of insulin aspart or regular insulin on each of the 6 study days in three different sites (the deltoid, the abdomen, and the thigh) during a 10-h euglycemic clamp (two drugs and three injection sites). Pharmacokinetic and pharmacodynamic measurements were derived from blood sample measurements of glucose, insulin, and C-peptide during these clamps. RESULTS: The pharmacodynamic data from the euglycemic clamp study showed that, regardless of injection site, the maximal glucose infusion rate (GIR Cmax) was greater and occurred at an earlier time (GIR Tmax) after administration of insulin aspart than regular insulin (GIR Cmax: abdomen 813 vs. 708, deltoid 861 vs. 736, and thigh 857 vs. 720 g/min, P < 0.05 for all; GIR Tmax: abdomen 94 vs. 173, deltoid 111 vs. 192, and thigh 145 vs. 193 g/min, P < 0.05 for all). Pharmacokinetic parameters were also consistent with faster absorption and higher peak insulin concentrations after insulin aspart administration. From all sites, the peak insulin concentration (Cmax) was higher and occurred earlier (Tmax) after administration of insulin aspart than of regular insulin (Cmax: abdomen 501 vs. 260, deltoid 506 vs. 252, thigh 422 vs. 220 pmol/l, P < 0.001 for all sites; Tmax: abdomen 52 vs. 109, deltoid 54 vs. 98, and thigh 60 vs. 107 min, P < 0.01 for all sites). The absorption and glucose-lowering action of insulin aspart did not differ between sites (similar GIR Cmax, Tmax, and area under the curve parameters). However, the duration of the glucose-lowering effect was up to 34 min shorter (P < 0.01) for the abdomen injections than for the deltoid or thigh injections (lower time of 50% glucose disposal). In addition, the amount of glucose infused was significantly lower by 10-14% in the abdomen than in other sites. CONCLUSIONS: Subcutaneous administration of insulin aspart causes a more rapid and intense maximal effect compared with regular insulin during euglycemic clamp studies in nondiabetic subjects. Abdominal administration of insulin aspart has a shorter duration of glucose-lowering effect compared with administration in the deltoid or thigh.

Pharmacodynamic and pharmacokinetic properties of a premixed 85/15 human insulin preparation.

BACKGROUND: Many patients with diabetes use mixtures of fast-acting (regular human) insulin and intermediate-acting (neutral protamine Hagedorn [NPH]) insulin to control their blood glucose levels. Premixed insulin is available in a 70%/30% mixture and a 50%/50% mixture of NPH/regular human insulin. For some patients, however, a premixed formulation containing > or =30% regular human insulin can provide too much fast-acting insulin, potentially causing an increased risk for hypoglycemia in the early hours after injection. OBJECTIVE: The pharmacokinetic and pharmacodynamic properties of a premixed formulation of 85% NPH insulin and 15% regular human insulin (85/15) were compared with those of a premixed 70%/30% NPH/regular human insulin preparation and 100% NPH insulin. METHODS: A 12-hour euglycemic clamp approach was used to assess glucose-lowering effects and serum insulin levels in 36 healthy male volunteers in a single-dose (0.5 U/kg), randomized, double-blind, 3-period, crossover study. RESULTS: From 0 to 8 hours after injection, the glucose-lowering effects and serum insulin levels for the 85/15 premixed insulin preparation were significantly greater than those for NPH insulin (P < or = 0.05) but significantly less than those for the 70/30 premixed insulin preparation. The mean (+/- SEM) maximum glucose infusion rate (GIRmax) was 8+/-0.6 mg/(min x kg) for the 85/15 preparation, 7+/-0.6 mg/(min x kg) for NPH, and 9+/-0.6 mg/(min x kg) for the 70/30 preparation, with time to peak GIR (tmax(GIR)) occurring at 313, 360, and 272 minutes, respectively. Time to peak insulin levels did not differ significantly for the 3 preparations, but maximum serum insulin concentration (Cmax(ins)) was significantly different between the groups (70/30 premix: 54+/-2.2 microU/mL; 85/15 premix: 44+/-2.4 microU/mL; NPH: 35+/-1.7 microU/mL). Glucodynamic effect and serum insulin levels did not differ significantly among preparations during the interval from 8 to 12 hours after injection. Mean serum C-peptide levels ranged from -0.6 to 1.0 ng/mL for each preparation during the 12-hour period after injection. CONCLUSIONS: The 85/15 premixed insulin preparation demonstrated clinical pharmacokinetic and pharmacodynamic properties that were intermediate between, and significantly different from, those of NPH insulin and the 70/30 premixed insulin preparation.

Role of glycemic control and protein restriction in clinical management of diabetic kidney disease.

OBJECTIVE: To assess the role of control of blood glucose levels and restriction of dietary protein in the management of diabetic nephropathy. METHODS: We summarize the results of pertinent published studies of glycemic control and modification of protein intake to provide information about strategies that potentially could benefit patients with diabetes and renal dysfunction. RESULTS: Considerable evidence is available to support the contention that improved glycemic control may have beneficial effects on the development and progression of diabetic renal disease. Maximal benefits of improved glycemia occur when instituted before the onset of macroalbuminuria. Once overt diabetic nephropathy is established, improved glycemic control may not be beneficial. Current evidence indicates that a glycosylated hemoglobin level of less than 8.1% should be the glycemic goal. At this level, the risk of developing micro-albuminuria is substantially reduced, and the risk of hypoglycemia is minimized. Most studies have been conducted in type I diabetes, and the results have been extrapolated to type II diabetes. Whether improved glycemic control will be equally beneficial in the nephropathy of type II diabetes has yet to be determined. Although some scientific evidence supports dietary protein restriction in patients with diabetic nephropathy, the extent of restriction needed for optimal benefits and minimal side effects remains to be determined. On the basis of current information, patients with both types of diabetes who have evidence of nephropathy should have protein limited to the recommended dietary allowance for adults (0.8 g/kg of body weight per day or approximately 10% of total daily caloric intake), and protein should be derived primarily from vegetable and lean animal sources. CONCLUSION: End-stage renal disease is not inevitable in patients with diabetic nephropathy. Normalization of glucose levels and modification of protein intake can favorably influence the course of diabetes-related kidney disease.

Strategies for preventing type II diabetes. What can be done to stem the epidemic?

Non-insulin-dependent (type II) diabetes is a major target for prevention because of its staggering emotional and financial burdens on society. The theory behind preventive efforts is that individuals at high risk often show signs of impaired glucose tolerance before diabetes fully develops. Once identified, these patients may respond to several pharmacologic and nonpharmacologic interventions that lower blood glucose levels and improve cardiovascular fitness with few adverse effects. However, the long-term efficacy of these approaches is uncertain. Information from the National Institutes of Health Diabetes Prevention Program should help clarify the impact of preventive measures and their cost-effectiveness.

Effect of acute exercise on citrate synthase activity in untrained and trained human skeletal muscle.

Maximal citrate synthase activity (CS) is routinely used as a marker of aerobic capacity and mitochondrial density in skeletal muscle. However, reported CS has been notoriously variable, even with similar experimental protocols and sampling from the same muscles. Exercise training has resulted in increases in CS ranging from 0 to 100%. Previously, it has been reported that acute exercise may significantly affect CS. To investigate the hypothesis that the large variation in CS that occurs with training is influenced by alterations during the exercise itself, we studied CS in human vastus lateralis both in the rested and acutely exercised state while trained and untrained (n = 6). Tissues obtained from four biopsies (untrained rested, untrained acutely exercised, trained rested, and trained acutely exercised) were analyzed spectrophotometrically for maximal CS. Exercise training measured in a rested state resulted in an 18.2% increase in CS (12.3 +/- 0.3 to 14.5 +/- 0.3 micromol x min(-1) x g tissue(-1), P < or = 0.05). However, even greater increases were recorded 1 h after acute exercise: 49.4% in the untrained state (12.3 +/- 0.3 to 18.3 +/- 0.5 micromol x min(-1) x g tissue(-1), P < or = 0.05) and 50.8% in the trained state (14.5 +/- 0.3 to 21.8 +/- 0.4 micromol x min(-1) x g tissue(-1), P < or = 0.05). Ultrastructural analysis, by electron microscopy, supported an effect of acute exercise with the finding of numerous swollen mitochondria 1 h after exercise that may result in greater access to the CS itself in the CS assay. In conclusion, although unexplained, the increased CS with acute exercise can clearly confound training responses and artificially elevate CS values. Therefore, the timing of muscle sampling relative to the last exercise session is critical when measuring CS and offers an explanation for the large variation in CS previously reported.

Exercise adaptation attenuates VEGF gene expression in human skeletal muscle.

Angiogenesis is a component of the multifactoral adaptation to exercise training, and vascular endothelial growth factor (VEGF) is involved in extracellular matrix changes and endothelial cell proliferation. However, there is limited evidence supporting the role of VEGF in the exercise training response. Thus we studied mRNA levels of VEGF, using quantitative Northern analysis, in untrained and trained human skeletal muscle at rest and after a single bout of exercise. Single leg knee-extension provided the acute exercise stimulus and the training modality. Four biopsies were collected from the vastus lateralis muscle at rest in the untrained and trained conditions before and after exercise. Training resulted in a 35% increase in muscle oxygen consumption and an 18% increase in number of capillaries per muscle fiber. At rest, VEGF/18S mRNA levels were similar before (0.38 +/- 0.04) and after (1.2 +/- 0.4) training. When muscle was untrained, acute exercise greatly elevated VEGF/18S mRNA levels (16.9 +/- 6.7). The VEGF/18S mRNA response to acute exercise in the trained state was markedly attenuated (5.4 +/- 1.3). These data support the concept that VEGF is involved in exercise-induced skeletal muscle angiogenesis and appears to be subject to a negative feedback mechanism as exercise adaptations occur.

Human VEGF gene expression in skeletal muscle: effect of acute normoxic and hypoxic exercise.

Vascular endothelial growth factor (VEGF) is involved in extracellular matrix changes and endothelial cell proliferation, both of which are precursors to new capillary growth. Angiogenesis is a vital adaptation to exercise training, and the exercise-induced reduction in intracellular PO2 has been proposed as a stimulus for this process. Thus we studied muscle cell PO2 [myoglobin PO2 (MbPO2)] during exercise in normoxia and in hypoxia (12% O2) and studied the mRNA levels of VEGF in six untrained subjects after a single bout of exercise by quantitative Northern analysis. Single-leg knee extension provided the acute exercise stimulus: a maximal test followed by 30 min at 50% of the peak work rate achieved in this graded test. Because peak work rate was not affected by hypoxia, the absolute and relative work rates were identical in hypoxia and normoxia. Three pericutaneous needle biopsies were collected from the vastus lateralis muscle, one at rest and then the others at 1 h after exercise in normoxia or hypoxia. At rest (control), VEGF mRNA levels were very low (0.38 +/- 0.04 VEGF/18S). After exercise in normoxia or hypoxia, VEGF mRNA levels were much greater (16.9 +/- 6.7 or 7.1 +/- 1.8 VEGF/18S, respectively). In contrast, there was no measurable basic fibroblast growth factor mRNA response to exercise at this 1-h postexercise time point. Magnetic resonance spectroscopy of myoglobin confirmed a reduction in MbPO2 in hypoxia (3.8 +/- 0.3 mmHg) compared with normoxia (7.2 +/- 0.6 mmHg) but failed to reveal a relationship between MbPO2 during exercise and VEGF expression. This VEGF mRNA increase in response to acute exercise supports the concept that VEGF is involved in exercise-induced skeletal muscle angiogenesis but questions the importance of a reduced cellular PO2 as a stimulus for this response.