M. Dubois

Thiazolidinediones (TZDs) form a new class of oral antidiabetic agents. They improve insulin sensitivity and reduce glycemia, lipidemia and insulinemia in patients with type 2 diabetes. Their mechanism is original, since they activate the nuclear receptor Peroxisome Proliferator-Activated Receptor gamma (PPARgamma), altering the expression of genes involved in glucose and lipid homeostasis. Stimulating PPARgamma improves insulin sensitivity via several mechanisms: 1) it raises the expression of GLUT4 glucose transporter; 2) it regulates release of adipocyte-derived signaling factors that affect insulin sensitivity in muscle, and 3) it contributes to a turn-over in adipose tissue, inducing the production of smaller, more insulin sensitive adipocytes. TZDs also affect free fatty acids (FFA) lipotoxicity on islets, improving pancreatic B-cell function. In addition, triglycerides and FFA levels are lowered by TZDs. Two TZDs, rosiglitazone and pioglitazone, have recently obtained the European commercial licence, but their use is restricted to the association with metformin or sulfonylureas. At the moment, they are indicated in type 2 diabetes but could be of interest in a broader array of diseases related to insulin resistance. As for side effects, rosiglitazone and pioglitazone may cause increased plasma volume, edema and dose-related weight gain. TZDs offer an attractive option in the treatment of type 2 diabetes, though it may be too soon to determine if they prevent vascular complications, as do other oral antidiabetic agents. An important issue for the future will be to assess the influence of weight gain in the long time.

Glucagon immunoreactivity (IRG) was measured in portal plasma from control and pancreatectomized rats and in arterial plasma from eviscerated rats with a functional liver. Portal IRG was 0.41 ± 0.02 ng/ml in control rats and 0.22 ± 0.01 in pancreatectomized rats. After evisceration, values of 0.08 ± 0.01 ng/ml were found (unextracted plasma, antiserum 30 K). Acid-ethanol plasma extracts demonstrated lower values, but a similar stepwise decrease was observed after pancreatectomy, then gastrectomy. Rat gastric extracts contained a low concentration of IRG (approximately 1/1200 the C-terminal IRG concentration of the corresponding pancreas). No IRG-positive cells were detected by immunofluorescence in the gastric mucosa. In the pancreatectomized rats, portal IRG remained stable for 75 min in the absence of further manipulation. From IRG concentrations and hepatic blood flow estimation in both control and pancreatectomized rats, the contribution of the stomach to portal IRG in the basal state could be estimated as 20% of the total. Gastric IRG release was increased by acute hypoglycemia (peak value 0.75 ± 0.18 ng/ml; N = 10; P < 0.01) and by 2-deoxyglucose infusion (0.45 ± 0.15 ng/ml; N = 4; P < 0.05). Administration of glucose + insulin induced a decrease in portal IRG (0.13 ± 0.01 ng/ml; N = 4; P < 0.001). Vagal stimulation and arginine infusion induced a rise in portal IRG: 0.84 ± 0.27 ng/ml (N = 10; P < 0.05) and 0.31 ± 0.03 ng/ml (N = 9; P < 0.01), respectively, while portal insulin remained low or undetectable (0-18 μu/ml). A rise in blood glucose accompanied the increase of plasma IRG. A concomitant insulin-induced hypoglycemia (36 ± 5 mg/dl) strongly potentiated the effects of both arginine infusion (1.46 ± 0.47 ng/ml; N = 6; P / 0.005) and vagal stimulation (1.39 ± 0.47 ng/ml; N = 4; P < 0.005).Higher IRG values were observed after pancreatectomy in alloxan-diabetic rats: 0.36 ± 0.02 ng/ml (N = 3; P < 0.001). We conclude therefore that: (a) the rat stomach contributes to the release of IRG in blood but to a limited extent, (b) the factors controlling this release appear very similar to those controlling pancreatic A cells; and (c) gastric IRG may be hyperglycemic in the rat.