The mu-opioid receptor (MOR) contains four highly conserved cytoplasmic tyrosine residues that may serve to regulate receptor activity. For Xenopus laevis oocytes coexpressing the rat MOR and the heteromultimeric potassium channel, K(IR)3.1/3.2, pretreatment with insulin produced both a 40% suppression in the basal channel conductance and potentiation of response to the mu-opioid agonist [D-Ala(2),methyl-Phe(4),Gly(5)-ol]enkephalin (DAMGO) to 155% of matched, untreated control cells. Insulin-induced potentiation of the DAMGO response was concentration-dependent and reversed after 1 h. Insulin pretreatment increased the maximal effect of DAMGO, but did not change its EC(50) value. Potentiation of the DAMGO response did not result from a recruitment of MOR to the cell surface, as measured by specific binding of the opioid peptide antagonist [(3)H]d-Phe((3)H)-Cys-Tyr-d-Trp-Arg-Thr-Pen-Thr-NH(2) (cyclic) to whole-oocytes, but instead the potentiation was probably caused by an increase in intrinsic efficacy of G protein coupling. The involvement of tyrosine residues on the putative intracellular loops of the MOR was demonstrated with four point-mutated receptors, replacing tyrosine with phenylalanine to create MOR(Y96F), MOR(Y106F), MOR(Y166F), and MOR(Y336F). None of these mutations significantly altered the EC(50) value for DAMGO compared with wild-type MOR, and insulin pretreatment still potentiated the effect of 1 microM DAMGO in oocytes containing either MOR(Y96F) or MOR(Y336F) to 137 +/- 10 and 124 +/- 8%, respectively. However, insulin did not significantly potentiate the DAMGO response with oocytes containing either MOR(Y106F) or MOR(Y166F), suggesting that these two sites were responsible for the insulin-induced opioid potentiation. The tyrosine-kinase inhibitors genistein (100 microM) or K-252a (20 microM) did not block the insulin-induced potentiation of the DAMGO response, but coincubation of insulin with either the MAP kinase inhibitor PD98,059 (20 microM) or phosphatase inhibitor orthovanadate (30 microM) completely blocked the potentiation. The results suggest the hypothesis that the potentiation was caused by dephosphorylation of the two tyrosines in MOR. To test this hypothesis, we measured the recovery rates after insulin treatment. As predicted, tyrosine kinase inhibition by K-252a significantly slowed the reversal and phosphatase inhibition by orthovanadate significantly accelerated the recovery. These findings support a rapid modulatory role for insulin on opioid signal transduction, possibly through the dephosphorylation of the MOR at tyrosines 106 and 166 by an insulin-activated MAP kinase/protein tyrosine phosphatase cascade. We conclude that tyrosine phosphorylation of the mu-opioid receptor regulates receptor-G protein coupling efficacy.