Characterization of the molecular mode of action of the sulfonylurea, glimepiride, at adipocytes.

The possibility of an insulin-independent blood glucose decreasing activity of sulfonylureas was re-evaluated. Single dose studies in dogs with different sulfonylureas revealed a ranking in the ratio of plasma insulin release/blood glucose decrease with glimepiride exhibiting the lowest and glibenclamide the highest ratio. This ranking suggests that sulfonylureas have extrapancreatic activity and that this is most pronounced for glimepiride. Further evidence for this was derived from single dose studies in rabbits, euglycemic hyperinsulinemic clamp studies in rats and subchronic studies in manifestly diabetic KK-AY mice. Extrapancreatic activity of sulfonylureas as deduced from the ranking in vivo between glimepiride and glibenclamide directly on peripheral tissues would imply a similar ranking between the two drugs in glucose utilizing processes in isolated muscle and fat cells. Indeed, glimepiride exhibits a higher potency compared to glibenclamide with respect to stimulation of glucose transport, glucose transporter isoform 4 (GLUT4) translocation and lipid and glycogen synthesis in normal and insulin-resistant adipocytes and in muscle cells, as well as of the potential underlying signalling processes examined at the molecular level. The molecular basis for the sulfonylurea-induced increase of glucose transport and non-oxidative glucose metabolism may rely on the dephosphorylation of key metabolic proteins/enzymes, like GLUT4 as demonstrated in isolated rat adipocytes. Activation of certain serine/threonine-specific protein phosphatases by insulin has been postulated to be mediated by the mitogen-activated protein kinase (MAPK) pathway and phosphatidylinositol (P1)-3'-kinase. However, there was no evidence that these pathways are involved in the regulation of protein phosphatase activity by sulfonylureas. Binding and photoaffinity studies showed that glimepiride associates in a time- and concentration dependent non-saturable manner with detergent-insoluble complexes of the plasma membrane which may correspond to caveolae. This association seems to be based on the interaction of glimepiride with glycosyl-phosphatidylinositol (GPI) lipids and membrane protein anchors. These were found to be enriched in detergent-insoluble complexes together with a GPI-specific phospholipase (PLC), the caveolae-specific coast protein, caveolin, and acylated tyrosine kinases of the src family. Sulfonylureas were found to stimulate the GPI-PLC and tyrosine phosphorylation of caveolin. This is presumably caused by direct interaction of the sulfonylurea into caveolar glycolipids and stimulation of a caveolar src tyrosine kinase, respectively. In accordance with the higher potency of glimepiride in vivo and in glucose transport/metabolism in vitro, the EC50 values for GPI-PLC activation and caveolin phosphorylation were lower for glimepiride than those for glibenclamide. The stimulation of protein tyrosine phosphorylation by sulfonylureas via this pathway not involving the insulin signaling cascade may be coupled to activation of specific protein phosphatases regulating glucose transport and metabolism. The concentrations required in vitro were higher than the reported therapeutic plasma concentrations. However, provided that the observed time-dependent accumulation of glimepiride in caveolae of peripheral cells were of functional relevance for stimulation of glucose transport/metabolism and would also occur in vivo, due to the longer exposure times even at lower drug concentrations the insulin-independent blood glucose decreasing activity of sulfonylureas might become effective in vivo.