Selectivity of repaglinide and glibenclamide for the pancreatic over the cardiovascular K channels

Aims/hypothesis: Sulfonylureas and glinides close beta cell ATP-sensitive K (K) channels to increase insulin release; the concomitant closure of cardiovascular K channels, however, leads to complications in patients with cardiac ischaemia. The insulinotrope repaglinide is successful in therapy, but has been reported to inhibit the recombinant K channels of beta cells, cardiocytes and non-vascular smooth muscle cells with similar potencies, suggesting that the (patho-)physiological role of the cardiovascular K channels may be overstated. We therefore re-examined repaglinide's potency at and affinity for the recombinant pancreatic, myocardial and vascular K channels in comparison with glibenclamide. Methods: K channel subunits (i.e. inwardly rectifying K channels [Kir6.x] and sulfonylurea receptors [SURx]) were expressed in intact human embryonic kidney cells and assayed in whole-cell patch-clamp and [H]glibenclamide binding experiments at 37°C. Results: Repaglinide and glibenclamide, respectively, were ≥30 and ≥1,000 times more potent in closing the pancreatic than the cardiovascular channels and they did not lead to complete inhibition of the myocardial channel. Binding assays showed that the selectivity of glibenclamide was essentially based on high affinity for the pancreatic SUR, whereas binding of repaglinide to the SUR subtypes was rather non-selective. After coexpression with Kir6.x to form the assembled channels, however, the affinity of the pancreatic channel for repaglinide was increased 130-fold, an effect much larger than with the cardiovascular channels. This selective effect of coexpression depended on the piperidino substituent in repaglinide. Conclusions/interpretation: Repaglinide and glibenclamide show higher potency and efficacy in inhibiting the pancreatic than the cardiovascular K channels, thus supporting their clinical use. © Springer-Verlag 2006.