The three known beta-adrenergic receptors (beta1-, beta2- and beta3-AR) are members of the G protein-coupled receptor (GPCR) superfamily. In response to epinephrine and norepinephrine, all three can couple to the stimulatory G protein (Gs), activating adenylyl cyclase to produce cyclic AMP. The latter activates protein kinase A (PKA) which by phosphorylation of various proteins, produces distinct physiological responses. This signaling pathway is highly regulated. We have focused on regulation of the receptors and previously showed that beta1-AR undergoes less agonist stimulated phosphorylation, ubiquitination, desensitization, internalization and down-regulation than does beta2-AR. Using chimeras in which the C-tails are exchanged, we demonstrated that the C-tail is the major determinant of subtype differences in these regulatory processes.? ? The mechanism(s) of intracellular sorting of receptors between recycling and degradation pathways is a major interest in GPCR research. It is proposed that motifs on the receptor are recognized by specific sorting proteins and that phosphorylation and ubiquitination act as motifs or modulate motif expression. As the ubiquitination of beta2-AR is dependent on its C-tail, we replaced the three C-tail lysines with arginines (K348/372/375R) to generate the mutant 3K/R. Agonist-mediated ubiquitination, degradation and down-regulation of 3K/R are reduced whereas internalization and recycling are similar compared to WT-beta2-AR. Intracellular trafficking of 3K/R is similar to that of beta1-AR, and unlike WT, 3K/R does not accumulate in lysosomes. Agonist-stimulated phosphorylation of both is similar but upon agonist removal, 3K/R undergoes dephosphorylation 3-4 times more rapidly than does WT. As the increased rate is observed in cell-free and soluble assays, it appears to be intrinsic to the mutation. In this regard, total protein phosphatase activity was similar in control and agonist-treated cells expressing WT or 3K/R. The activity was partially inhibited by okadaic acid and inhibitor-2, potent inhibitors of protein phosphatases PP2A and PP1, respectively. We confirmed that both are present in the cells by Western blotting. In addition, we observed that both co-immunoprecipitated with beta2-AR and in similar amounts from control and agonist-treated cells expressing WT or 3K/R. It appears that beta2-AR exists in a complex with PP2A and PP1 that is agonist-independent and similar for the WT and 3K/R receptors. When we compared the rates of internalization and dephosphorylation, we found that WT is being internalized more rapidly than it is being dephosphorylated whereas it is the opposite for 3K/R. Taken together with our finding that the receptor is associated with PP2A and PP1, our data suggest that dephosphorylation can occur before endocytosis of the receptor. It has generally been accepted that dephosphorylation occurs in acidic endosomes. Finally, we established that the increased rate of dephosphorylation of 3K/R was functionally relevant. In okadaic acid-treated cells, lysosomal trafficking and degradation of 3K/R are increased almost to the levels of WT.? ? Thus phosphorylation and ubiquitination appear to be involved in the intracellular sorting of beta2-AR between recycling to the plasma membrane and lysosomal degradation. Phosphorylation may be a sufficient signal for degradation whereas ubiquitination may enhance the signal as well as protect the receptor from dephosphorylation. Although we favor this hypothesis, we cannot rule out the possibility that the mutations alter the structure of the receptor making the C-tail more susceptible to phosphatases. We believe that it is unlikely as WT and 3K/R have similar properties of ligand binding, adenylyl cyclase activation and desensitization, and phosphorylation and internalization. As all are dependent on the C-tail, any structural change should be reflected in these properties. Furthermore, we found the K348R mutant is similar to WT and the K372/375R mutant is similar to 3K/R. Thus, the C-tail region (GYSSNGNTGEQSGYHVEQEKENKLL) which contains the three serines that are phosphorylated by G protein-coupled receptor kinases (GRKs) and two lysines that are ubiquitinated appears to be involved in beta2-AR trafficking.? ? The regulation of beta-ARs by PKA has been extensively investigated and shown to occur at multiple levels. A rapid desensitization of beta1-AR and beta2-AR is observed in cells where PKA but not GRKs are activated, and is blocked by treating the cells with PKA inhibitors. A similar desensitization of the two subtypes occurs in membranes incubated with ATP and the catalytic subunit of PKA. The desensitization is due phosphorylation of the receptors by PKA. More prolonged activation of PKA leads to down-regulation of both beta1-AR and beta2-AR as measured by loss of receptor binding activity and protein. PKA-mediated down-regulation is highly cell-type-specific and due to a reduction of beta-AR mRNA levels by either message destabilization or transcriptional repression. Activation of PKA, however, does not result in internalization of beta2-AR as shown by others or beta1-AR as shown by us. Neither endogenous beta1-AR in human SK-N-MC cells nor heterologous human beta1-AR in Chinese hamster fibroblasts (CHW) undergo endocytosis when exposed to a permeable cAMP derivative. It was recently reported that beta1-AR is internalized via caveolae by a PKA-dependent process. The cells (HEK 293), the species of beta1-AR (mouse) and the methods used were different from ours, which may explain the conflicting results. To clarify the situation, we examined the effect of activating PKA on B1AR internalization in CHW expressing either human or rat beta1-AR; in HEK 293 cells expressing the G389 and R389 variants of human beta1-AR or mouse beta1-AR; and in rat C6 glioma cells expressing endogenous beta1-AR. The cells were exposed to a high and low concentration of agonist or to a permeable cAMP derivative in the absence and presence of a PKA inhibitor (H89). In all of the cells exposed to the high agonist concentration, beta1-AR internalization occurred and H89 had no effect. In the presence of the low agonist concentration, less than 10% of beta1-AR were internalized and H89 did not inhibit this internalization. The cAMP derivative had no effect on internalization. We then compared the effect of agonist concentration on intracellular cAMP levels, receptor occupancy by the agonist and receptor internalization. Whereas the EC50 values for agonist occupancy and stimulation of internalization were similar, the value for agonist-stimulated cAMP accumulation was an order of magnitude lower. Thus, under conditions of high intracellular cAMP and low receptor occupancy, there is little beta1-AR internalization. The correlation between receptor occupancy and internalization is consistent with GRK-catalyzed phosphorylation of the agonist-occupied receptors followed by arrestin binding and promotion of internalization via clathrin-coated pits. In this regard, we demonstrated that beta1-AR endocytosis is blocked by hypertonic sucrose which disrupts clathrin baskets but not by nystatin which disrupts caveolae. We conclude that our results do not support a role for PKA in the endocytosis of beta1-AR
