Our long-term research goal is to identify and elucidate mechanisms for regulation of the receptor-coupled adenylyl cyclase signaling pathway. We are using the beta-adrenergic receptor (BAR) as a model and focusing on differences in regulation of the three BAR subtypes, B1AR, B2AR and B3AR. They are members of the G protein-coupled receptor (GPCR) superfamily and activate adenylyl cyclase via the stimulatory G protein (Gs) in response to epinephrine and norepinephrine. In addition to differences in tissue distribution, we and others have shown they differ in their ability to be regulated (B2AR> B1AR>>B3AR). This may provide a physiological basis for the existence of three subtypes. Previously, we showed that: (i) B1AR undergoes less agonist-mediated internalization than B2AR; (ii) over-expressing GPCR kinases (GRKs) or arrestins enhances internalization of B1AR to that of B2AR; (iii) B1AR is endocytosed by the clathrin-coated pit pathway similar to that of B2AR; (iv) and during or soon after endocytosis, the two subtypes are sorted to different endosomal compartments. We are currently investigating the mechanisms and consequences of this subtype sorting. Following agonist-mediated endocytosis, most GPCRs either are recycled back to the plasma membrane for reuse or are targeted to lysosomes for proteolysis and down-regulation. The balance between these two processes depends not only on the GPCR but also on other factors which together with the underlying mechanisms are under intense investigation. We have found that the difference in trafficking of the two subtypes affects both their recycling and down-regulation. We used baby hamster kidney (BHK) and human embryo kidney (HEK 293) cells expressing similar levels of either B1AR or B2AR. To investigated recycling, cells were treated with agonist for 15-30 min to maximize internalization without any down-regulation, washed free of agonist and allowed to recycle BARs for different times. In BHK cells, both subtypes recycled but B1AR recycling was difficult to quantify due to its limited internalization (<20% vs >45% for B2AR). Co-expression of arrestin-2 increased internalization to 55-65% for both subtypes and the rates and extent of recycling were very similar. As HEK 293 cells have more endogenous arrestins than BHK, internalization was greater (~35% for B1AR and ~70% for B2AR) and co-expression of arrestin-2 was not required. In HEK 293 cells, internalized B1AR recycled at the same rate as B2AR. Receptor recycling is inhibited by the ionophore monensin which blocks endosomal acidification. In monensin-treated BHK and HEK 293 cells, we found inhibition of B2AR but not B1AR recycling. These results are consistent with B2AR and B1AR being sorted to monensin-sensitive and -insensitive recycling endosomes, respectively. We showed previously that upon prolonged agonist stimulation, B2AR but not B1AR is down-regulated in both BHK and HEK 293 cells. We had speculated that due to the subtype sorting differences, B1AR may be in endosomes that do not converge with the degradative pathway. Since then we have found that: although internalization is necessary for down-regulation, increasing B1AR endocytosis by over-expressing arrestin-2 did not lead to B1AR down-regulation. In the absence of agonist, both subtypes turned over at similar rates. Thus, B1AR is capable of being degraded in these cells. We then explored the role of structural motifs of the receptors in determining their fate. Key areas of differences reside in the 3rd intracellular loop and the C-terminus. B1AR has a proline-rich, 24-residue sequence in its 3rd intracellular loop, and although its deletion increases B1AR internalization, we found it had little effect on B1AR down-regulation. Human B1AR also has a polymorphism at residue 389 that affects function. The Gly-389 variant was the first to be cloned and thus used by most laboratories. The Arg-389 variant, however, couples more efficiently to Gs. We constructed an Arg-389 B1AR, and confirmed that it is more efficient than the Gly-389 B1AR when expressed in BHK or HEK 293 cells. The EC50 for agonist-stimulated cAMP formation was equal to that of B2AR and 10-fold lower than that of Gly-389 B1AR. Despite the increase in function, the Arg-389 variant was as resistant to down-regulation as the Gly-389 B1AR. We then directed our attention to the C-tails and expressed chimeras in BHK cells: B1AR through the 7th transmembrane region joined to B2AR C-tail, B1(1-7)/B2(ct), and the analogous B2(1-7)/B1(ct). We confirmed that B1(1-7)/B2(ct) internalized more than B1AR and B2(1-7)/B1(ct) less than B2AR. We then established that the chimeras down-regulated in the same order: B2AR>B1(1-7)/B2(ct)>B2(1-7)/B1(ct)>B1AR. Thus, the BAR C-tails are important determinants for both internalization and down-regulation. Undoubtably other structural features are involved in BAR regulation and trafficking as well as specific sorting proteins. We have shown that human B1AR expressed both endogenously in SK-N-MC neurotumor cells and stably in Chinese hamster (CHW) fibroblasts undergo agonist-mediated down-regulation. One explanation for these cell-specific differences is the presence or absence of sorting protein(s) that target B1AR to the lysosomal degradative pathway. Such a protein recently has been described for the delta opioid receptor. We also continue to investigate targeting of BAR subtypes to caveolae, the cholesterol- and sphingolipid-enriched microdomains of the plasma membrane. Last year we reported that in rat C6 glioma cells, endogenous B2AR and B1AR respectively are located in caveolae and distributed more randomly. We recently found that endogenous B1AR in human SK-N-MC cells was targeted to caveolae. With both cell lines, we identify the receptors by Western blotting. In total cell extracts as well as the subcellular fractions, hB1AR appeared as a doublet of 52 and 65 kDa. In contrast, rB1AR appeared as multiple species ranging from 37 to 100 kDa in cell extracts and non-caveolae fractions whereas caveolae-enriched fractions contained major 50 kDa and minor 65 kDa species. As the B1ARs have 91% homology at the protein level, the reason for this anomaly is not clear. We assayed the subcellular fractions for binding activity and recovered most of it in the caveolae-enriched fractions from both cell lines. Thus, distribution of hB1AR protein and activity corresponded whereas that of rB1AR did not. We considered the possibilities of either cell-specific or receptor species differences. To address both, we analyzed CHW cells expressing similar levels of either h- or rB1AR and found the same pattern as with the endogenous receptors. To show that the caveolae-associated activity represents cell surface B1ARs located in caveolae and not in internalized lipid rafts, we incubated the cells with a radioligand that only binds to surface BAR before we homogenize them and isolated the subcellular fractions. With both r- and hB1ARs, most of the bound radioligand was recovered in the caveolae-enriched fractions. One explanation for the presence of multiple forms of the rat B1AR, their heterogenous subcellular distribution and the dissociation of binding and immunoreactivity is that the additional protein species detected in the non-caveolae fractions represent inactive receptors. Possibly they have undergone ubiquitination and are targeted for proteosome-mediated degradation. Proteins usually are conjugated with multiple molecules of ubiquitin resulting in a spectrum of molecular masses. We are testing this possibility by treating the cells with proteosome inhibitors. Our findings have important implications for this research area as most investigators are only using Western blotting to determine whether BARs as well as other GPCRs are targeted to caveolae.
