Characterization of the G protein coupling of a glucagon receptor to the KATP channel in insulin-secreting cells

The G-protein-mediated coupling of a glucagon receptor to ATP-dependent K channels—K_ATP—has been studied in insulin-secreting cells using the patch clamp technique. In excised outside-out patches, K_ATP channel activity was inhibited by low concentrations of glucagon (IC₅₀ = 2.4 nm); the inhibitory effect vanished at concentrations greater than 50 nm. In cell-attached patches, inhibition by bath-applied glucagon was seen most often, although stimulation was observed in a few cases. A dual action of the hormone is proposed to resolve these apparently divergent results. In excised inside-out patches, K_ATP channel activity was inhibited by addition of subunits purified from either erythrocyte or retina (IC₅₀ = 50 pm and 1 nm, respectively). Subsequent exposure of the patch to _i or _o reversed this effect. In excised inside-out patches, increasing Mg²⁺ in the bath stimulated the channel activity between 0 and 0.5 mm, but blocked it at higher concentrations (IC₅₀ = 2.55 mm). In most cases (70%), GTP had a stimulatory effect at concentrations up to 100 m. However, in three cases, similar GTP levels had clear inhibitory effects. In excised inside-out patches, cholera toxin (CTX) caused channel inhibition. Although the effect could not be reversed by removal of the toxin, the activity was restored by subsequent addition of purified _i or _o . These results are compatible with a model whereby channel inhibition by activated G _S -coupled receptors occurs, at least in part, via association of the subunits of G _S with _i / _o subunits and deactivation of the _i / _o -dependent stimulatory pathway. On the basis of this hypothesis, a model is developed to describe the effects of G proteins on the K_ATP channel, as well as to account for the concentration-dependent stimulation and inhibition of K_ATP channel by Mg²⁺. An interpretation of the ability of glucagon to potentiate, but not initiate, insulin release is also given in terms of this model and the effects of ATP on K_ATP channels.This work was supported by grant DCB-89 19368 from the National Science Foundation and a research grant (W-P 880513) from the American Diabetes Association to B.R.The authors would like to thank Dr. A.E. Boyd, III for supplying the RINm5F and HIT cells, Drs. J. Codina and L. Birnbaumer for supplying the G protein and subunits from erythrocyte, Dr. R.A. Cerione for supplying the G protein subunit from retina, and Mrs. Satoko Hagiwara for preparing and maintaining the cell cultures.