GPCRs activate MAPKs by multiple converging mechanisms 2021. To test the ability of the fusion protein β₂AR-Gα_s to induce MAPK signalling, the agonist-mediated activation of extra cellular signal-regulated kinases (ERK1/2), also known as p42/44 MAPK, was analysed by western blotting. HEK293 cells stably expressing the β₂AR-Gα_s (HEK293/β₂AR-Gα_s) or wild type β₂-adrenergic receptor (HEK293/β₂AR) were treated for increasing times with the agonist isoproterenol (1 μM), then cell extracts were subjected to SDS-PAGE and blots were probed with a monoclonal antibody specific for the phosphorylated ERK1/2 proteins. As shown in figure 1, the isoproterenol stimulation of HEK293/β₂AR-Gα_s resulted in both ERK-2 (major band at 42 kDa) and ERK-1 (minor band at 44 kDa) phosphorylation, whose activities were enhanced following 10 min of agonist treatment and then rapidly declined without other activation peaks within 4 hours. A similar time-course of ERK activation was observed in the cells stably expressing the wild type receptor.

In agreement with previous findings, this result confirms that the fusion of the G protein α-subunit to β₂AR does not alter the functional properties of the receptor, which retains its ability to activate cellular signalling.

The exposure of cell surface β₂ARs to agonist results in a time-dependent internalization of receptors. Internalization properties of the β₂AR-Gα_s fusion protein were investigated using both qualitative and quantitative experimental strategies. Immunofluorescence experiments were performed on HEK293 cells stably expressing the Flag-tagged chimeric (HEK293/β₂AR-Gα_s) or wild type (HEK293/β₂AR) receptors following visualization with an epifluorescence microscope. Figure 2A shows a typical experiment in which the cells were treated with 1 μM isoproterenol for the indicated times, fixed in formaldehyde and subjected to indirect immunofluorescence using an anti-Flag antibody, as described in the Methods section. As shown in figure 2A, the wild type receptor rapidly disappeared from the plasma membrane within 5 minutes of agonist treatment. The same amount of time was not sufficient to induce a significant internalization of the chimeric receptor, as indicated by the high immunoreactivity to Flag-antibody retained on the plasma membrane of cells expressing the fusion protein. An appreciable decrease in the surface levels of the fusion protein started at 20 minutes, but the extensive loss of plasma membrane β₂AR-Gα_s was only observed after 60 minutes of agonist stimulation. At this time, indeed, the permeabilization of samples with a non-anionic detergent, performed before the immuno-staining procedure, allowed for the observation of an intracellular redistribution of the wild type as well as the chimeric receptor (fig. 2B). To quantify the internalization rates of the chimeric and wild type proteins, we performed a fluorimetric assay. HEK293 cells stably expressing β₂AR-Gα_s or native receptor were seeded onto multiwell plates and treated with 1 μM isoproterenol for the indicated times (fig. 2C). After fixation, the surface receptors were immuno-stained with an anti-Flag antibody and labelled with Alexa Fluor 488-conjugated anti mouse IgG. The fluorescence measured by a multi-label counter clearly indicates that the chimeric protein has a lower rate of internalization than the wild type receptor. The computed t_1/2 was 12 and 33 min for β₂AR and β₂AR-Gα_s, respectively.

To investigate whether the fusion of β₂AR to G protein α-subunit also affects the recycling processes of the receptor, we used the same technical approach used for the internalization studies. First, HEK293/β₂AR or HEK293/β₂AR-Gα_s cells were exposed to 1 μM isoproterenol for 5 hours to induce the maximal agonist-mediated internalization, as shown in figure 2C. Then we added the antagonist propranolol and incubated the cells for an additional hour, to allow the cell surface recovery of the receptors. Finally, an immunofluorescence assay was performed to visualize the recycling of epitope-tagged receptors via fluorescence microscopy (fig. 3A). Flag-epitopes immuno-staining revealed that, after 5 hours of incubation with isoproterenol, both receptors disappeared from the cell surface (fig. 3A panels b,b'). However, following an additional hour of incubation with the antagonist propranolol, native β₂ AR was almost completely returned to the cell surface, in sharp contrast with the β₂AR-Gα_s protein, which showed only a slight plasma membrane redistribution (fig. 3A panels c,c'). The inability of the chimeric receptor to recycle back to the plasma membrane was also confirmed by a fluorimetric assay performed to estimate the cell surface recovery of receptors upon antagonist incubation (fig. 3B). HEK293/β₂AR or HEK293/β₂AR-Gα_s cells were cultured on 96 multiwell plates and exposed to isoproterenol (1 μM) for 5 hours. The agonist-mediated recruitment of receptors to intracellular compartments was stopped by the addition of propranolol (1 μM) for increasing times, after which an immunofluorescence assay was performed to visualize the surface receptors (fig. 3B). At each time point, the cell surface recovery of either epitope-tagged β₂ AR or β₂AR-Gα_s was evaluated by the measurement of cell surface fluorescence by a multi-label counter. As shown in figure 3B, upon addition of the antagonist, the wild type receptor rapidly relocalized to the plasma membrane, in a time-dependent fashion. Approximately 50% of internalized β₂AR returned to the plasma membrane within 15 minutes, thus showing a fast component of recycling. Over this time, the further exposure of cells to antagonist favoured a nearly complete recycling of β₂AR by a slow component. As expected, the difference in the recycling rates between fused and non-fused receptors was dramatic. In fact, the fluorimetric measurement performed on cells expressing the β₂AR-Gα_s resulted in a slow as well as modest relocalization of receptors to the plasma membrane, and no significant recycling of the fusion protein was observed until after 60 minutes of antagonist treatment. Moreover, to exclude the membrane exposure of newly synthesized receptors from analysis, we also performed the same experiments in the presence of cycloheximide (CHX), a protein synthesis inhibitor. Under this condition, we observed that the recovery of the wild type receptor occurred by both recycling (fast component) and resynthesis (slow component). In contrast, the recovery of the fused receptor occurred only by new synthesis.

In order to investigate the failure of the chimeric receptor to recycle back to the plasma membrane in more detail, we compared the agonist-mediated sub-cellular localization of β₂AR and β₂AR-Gα_s receptors with that of endocytosed transferrin, a well-established marker of early and rapid recycling endosomes 142223. The extent of co-localization between endocytosed transferrin and either native or chimeric receptors was examined by dual-label confocal microscopy (fig. 4A). As indicated by the yellow spots in the merged panel of figure 4A, most of the wild type β₂AR co-localized with the tracer transferrin, consistent with the typical rapid recycling of the β₂AR, whereas β₂AR-Gα trafficked through transferrin-positive endosomal compartments, with an apparent lower grade compared to the wild type receptor. The fraction of vesicles showing transferrin-receptor co-localization was approximately 43% and 37% for the wild type and chimeric receptors, respectively (fig. 4B). This slight difference could be explained by the different kinetics of internalization displayed by the two receptors (fig. 2). As the above experiment demonstrated, 30 minutes corresponds to the agonist-exposure time at which β₂ AR, but not β₂AR-Gα_s receptor, shows maximal agonist-mediated internalization. Thus, the resistance to agonist-induced endocytosis showed by the fusion protein could also result in a different kinetic rate of transfer into and out of specific vesicles, without, however, changing its endosomal targeting. Taken together, these data support the hypothesis that the failure of β₂AR-Gα_s to undergo recycling is not linked to it sorting into endosomal compartments distinct from those that efficiently translate their cargo to the plasma membrane.

The experiments reported above do not exclude the possibility that, following agonist-mediated internalization, β₂AR-Gα_s and wild type β₂AR may traffic through similar endosomal compartments. To better understand this, we carried out experiments in which the endocytic trafficking of β₂-adrenergic receptors was analyzed in a co-expression system using a different tagged version of the native receptor.

HEK293/β₂AR-Gα_s cells were transiently transfected with an amino-terminus HA-tagged β₂AR (HA-β₂AR), were incubated with or without 1 μM isoproterenol for 5 hours to induce maximal internalization without affecting the recycling of the native receptor, were fixed, and were then subjected to confocal microscopy. In the absence of agonist stimulation, the immunoreactivity of Flag-β₂AR-Gα_s, as well as that HA-β₂AR, was typically localized at the plasma membrane. In response to agonist, each receptor redistributed from the cell surface to a population of cytoplasmic vesicles that resulted in complete overlapping in the merged colour images (fig. 5). These results, in addition to the transferrin co-localization experiments, confirm that the fusion of the G protein α-subunit to the wild type β₂AR does not target the chimeric receptor to endosomal compartments distinct from those observed for the wild type protein. Next, using the same experimental strategy, we also examined whether or not the inability of the chimeric receptor to recycle back was affected by the presence of the wild type receptor.

Analysis by epifluorescence microscopy of HEK293/β₂AR-Gα_s cells transiently expressing the wild type β₂AR, and treated for five hours with 1 μM isoproterenol, confirmed that both receptors were recruited into the same endocytic vesicles (fig. 6). However, when the antagonist propranolol was added to stop receptor internalization, we observed an almost complete recovery of the wild type receptor on the cell surface, whereas the β₂AR-Gα_s was entirely retained in the intracellular vesicles. Moreover, an intracellular pool of both receptors continued to co-localize, probably in correspondence of compartments that drive them to a common degradative pathway.

Materials came from the following sources: cell culture media, fetal bovine serum, G418 and Lipofectamine were from Invitrogen. Isoproterenol, propranolol, anti-FLAG monoclonal antibody (M1 clone), anti-HA polyclonal antibody and Hoechst 33258 were from Sigma. Phospho-p44/42 MAPK rabbit monoclonal antibody was from Cell Signalling Technology. Alexa-Fluor 488-conjugated transferrin, and antibodies Alexa-Fluor 488 and Alexa-Fluor 594 goat anti-mouse IgG, were from Molecular Probes. Nitrocellulose transfer membrane was from Schleicher&Schuell. Enhanced chemiluminescent substrate (ECL) for detection of HRP was from PIERCE.

The construction of full-length cDNA encoding the Flag-tagged β2AR-Gαs fusion protein was described previously 19.

The amino terminal Flag tagged wild receptor was obtained in a similar way. Briefly, the cleavable prolactin signal peptide tethered to the Flag epitope (DYKDDDDK) was added immediately before the Gly² in β₂ AR by PCR-based strategies and cloned in a pcDNA3 vector (Invitrogen).

The cDNA encoding for 3HA-β₂AR was obtained by the annealing of two synthetic oligonucleotides containing the sequences of three sequential modules of the influenza hemagglutinin HA epitope (AYPYDVPDYA), and was cloned into pcDNA3 vector. Then, by PCR, we obtained the cDNA encoding for the β₂AR deprived of the Met initiator codon, which was subcloned into the vector, in frame with the 3 × HA epitope.

Human embryonic kidney (HEK293) cells stably expressing the wild type β₂AR or chimeric protein β₂AR-Gα_s were generated as described previously 19. The total number of expressed receptors was measured in membranes prepared from the transfected cells using radioligand binding assays, as described previously 19. The levels of receptor expression were 15 and 10.8 pmol/mg in cells expressing wild type β₂AR and β₂AR-Gα_s, respectively.

Cells were grown in Dulbecco's modified Eagle's medium (DMEM; GIBCO) supplemented with 10% (v/v) fetal bovine serum (FBS; GIBCO), 100 units/ml penicillin, 100 μg/ml streptomycin sulphate, and 200 μg/ml G418 (GIBCO) in a humidified atmosphere of 5% CO₂ at 37°C.

For transient transfections, cells cultured in 35-mm tissue culture dishes were transfected with 0.3 μg of pcDNA3/HA-β₂AR and 0.7 μg of empty vector using Lipofectamine (Invitrogen), according to the manufacturer's instructions. The cells were allowed to express the transfected gene for 48 hrs before harvesting.

Cells were grown on 35-mm tissue culture dishes and following the various treatments were fixed with 4% buffered formaldehyde for 20 minutes. If necessary, fixed cells were permeabilized with 0.2% Nonidet P-40 (NP-40) in phosphate-buffered saline (PBS) for 10 min to assure the accessibility of intracellular and intravesicular antigens. For the detection of epitope-tagged receptors, anti-Flag mouse monoclonal antibody was added in blocking buffer (1% BSA) for 50 min at room temperature (r.t.) followed by Alexa-Fluor 594 conjugated anti-mouse IgG (Molecular Probes). In dual staining experiments, co-localization of Flag-tagged receptors and HA-tagged receptors in a single cell line, was performed by incubating cells with anti-Flag and anti-HA rabbit polyclonal antibody (Sigma).

Stained specimens were examined by conventional epifluorescence microscope (Olympus BX51; Tokyo, Japan) or confocal microscope.

Fluorescently labelled preparations were also observed by a confocal fluorescent imaging system using the confocal laser scanning microscope LEICA TCS 4D (Leach Instruments, Heidelberg, Germany) supplemented with an Argon/Kripton laser and equipped with 40 × 1.00–0.5 and 100 × 1.3–0.6 oil immersion lenses. The excitation/emission wavelengths employed were 488 nm/510 nm, and 568 nm/590 nm for specific Alexa-Fluor labelling. Confocal sections were acquired at intervals of 0.5 μm from the middle to the bottom toward the cells, and a 3D reconstruction image of the fluorescent signal was obtained. In double staining experiments, confocal sections for both fluorescent signals were taken simultaneously, the 3D reconstruction images were recorded, and merged images of the two signals were obtained using confocal microscope software.

Cells, cultured on 96 multiwell plates (Nunc) until they reached confluence, were stimulated with 1 μM isoproterenol (Sigma) at 37°C for different times to drive agonist-induced internalization. At the end of the time-course, cells were immediately fixed with 4% buffered formaldehyde for 20 min, incubated in blocking buffer for 30 min and immuno-stained for 50 min with anti-FLAG mouse monoclonal antibody (Sigma), followed by Alexa-Fluor 488 conjugated anti-mouse IgG (Molecular Probes). Nuclei were stained with Hoechst 33258 (1 μg/ml). Fluorescence was measured by a multi-label counter (Victor 2; Wallac), setting excitation and emission filters as follows: λ_ex = 485/λ_em = 535, and λ_ex = 355/λ_em = 460 for Alexa-Fluor 488 and Hoechst, respectively. Each point was performed in triplicate and the receptor internalization was monitored in relationship to the decrease of fluorescence from the plasma membrane.

For recycling studies, cells cultured on 96 multiwell plates were incubated with 1 μM isoproterenol for 5 hrs to induce maximal receptor internalization. Reversal of internalization was initiated by adding the antagonist, 1 μM Propranolol (Sigma), for different times. In other experiments, 50 μg/ml cycloheximide (CHX; Sigma) was added during the last 2 hrs of agonist treatment. At the end of the time-course, the cells were fixed and subjected to immunofluorence followed by fluorimetry analysis as above.

Receptor recycling from the endocytic pathway was estimated by assaying the recovery of immunoreactive epitopes at the cell surface that were accessible by monoclonal antibody.

In order to estimate the degree of co-localization of the receptor with endocytosed transferrin, cells were incubated for 30 min at 37°C in culture medium containing 100 μM Alexa-Fluor 488-conjugated transferrin (Molecular Probes), and then 1 μM isoproterenol was added for an additional 30 min.

Next, cells were washed twice in pre-warmed PBS to remove non-internalized transferrin and were subjected to immunofluorescence as above. Briefly, fixed cells were permeabilized and immuno-stained for detection of Flag-tagged receptors. Dual-label fluorescence microscopy was performed as described above.

Cells were first sensitized by incubating with the agonist isoproterenol (1 μM) for different times and were then dissolved in Laemmli's sample buffer. Samples were electrophoresed through a 12% SDS-polyacrylamide gel (SDS-PAGE) and transferred onto nitrocellulose membranes (Schleicher&Schuell). Blots were incubated in blocking buffer (5% w/v non fat dry-milk, 0,1% Tween-20 in Tris buffered saline) for 1 hrs at r.t., followed by incubation overnight at 4°C with Phospho-p44/42 MAP kinases rabbit monoclonal antibody (Cell Signalling Technology), which detects endogenous levels of p42 and p44 MAP kinases (ERK1/2) when phosphorylated at Thr202 and Tyr204, respectively. Immunoreactive bands probed with horseradish peroxidase-conjugated anti rabbit IgG antibody were visualized by chemiluminescence (ECL; PIERCE), according to the manufacturer's instructions.

Samples were normalized for protein content by reprobing blots with an alpha-tubulin monoclonal antibody (Sigma).