INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

Numerous hormones and neurotransmitters exert their effects through G-protein-coupled receptors (GPCRs)¹ (1-4). The ₂-adrenoreceptor (₂AR), a prototypical GPCR, interacts with the G-protein G_s, causes GDP/GTP exchange at its -subunit (G_s) and, thereby, leads to activation of adenylyl cyclase (AC). Recently, the ternary complex model of GPCR activation has been extended to explain the finding that GPCRs can activate G-proteins, even in the absence of agonist, and that certain receptor ligands, namely inverse agonists, can suppress the G-protein activation mediated by agonist-free GPCRs (5-14). The agonist-independent activity of a GPCR is referred to as constitutive activity. The extended ternary complex model (two-state model) assumes that agonists stabilize GPCRs in the active (R*) state, while inverse agonists stabilize the inactive (R) state. Although constitutive GPCR activity can be most easily observed when receptors are overexpressed (10-12) or mutated (7, 8, 13), it also occurs at physiological receptor expression levels (5, 6, 9, 14). Hallmarks of constitutive GPCR activity are increased potency and efficacy of partial agonists, increased efficacy of inverse agonists, and elevated basal G-protein activity (5-14). These properties of constitutive activity are generally associated with GPCR function, and little is known about the ability of different G-proteins to influence the efficacy and potency of ligands.

G_s exists as a short (G_sS) and a long (G_sL) splice variant. Compared with G_sS, G_sL contains additional 15 amino acids inserted at position 72 of the polypeptide chain, and there is an exchange of glutamate for aspartate at position 71 (15, 16) (Fig. 1A). Based on the -carbon model of the -subunit of the retinal G-protein tansducin (17), the sequence within which the 15-amino acid insert is localized in G_sL serves as a linker between the Ras-like domain and the -helical domain (Fig. 1B). The guanine nucleotide-binding site is embedded between these two domains. Thus, a change in this linker sequence might be expected to influence the binding kinetics of guanine nucleotides. In fact, purified G_sL releases GDP more than twice as fast as G_sS (18).

The results of a previous study indicate that G_sS may be more effective than G_sL in activating AC (19), but with regard to ₂AR coupling, studies have not revealed significant differences between G_sS and G_sL (18, 20, 21). Studying differences in the interaction of structurally very similar G-proteins with a given GPCR is technically difficult, because functional interactions between receptors and G-proteins are strongly influenced by their relative expression levels (22). Specifically, defined receptor/G-protein stoichiometries have to be achieved to be able to detect subtle differences in GPCR/G-protein coupling.

To facilitate the examination of receptor/G-protein interactions we constructed fusion protein DNAs in which the C terminus of the ₂AR was linked to the N terminus of G_sS (₂ARG_sS) or G_sL (₂ARG_sL) (Fig. 1A) and expressed the fusion proteins in Sf9 cells. Fusion proteins have a fixed ratio of receptor to -subunit (23, 24). Thus, ambiguities in data analysis because of varying stoichiometry of the signaling partners can be eliminated. Using the fusion protein approach, we observed that the ₂AR coupled to G_sL has properties of constitutively active GPCR.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Materials-- Rat G_sL DNA was kindly provided by Dr. R. R. Reed (Johns Hopkins University, Baltimore, MD) (25). For generation of recombinant baculoviruses encoding for rat G_sL, its DNA sequence was transferred into the baculovirus transfer vector pVL 1392 (11). [-³²P]GTP (6000 Ci/mmol) and [-³²P]ATP (3000 Ci/mmol) were from NEN Life Science Products. [³H]Dihydroalprenolol ([³H]DHA) (85-90 Ci/mmol) was from Amersham Corp. Anti-G_s antibody was from Calbiochem. Guanosine 5'-phosphorothioate (GMPS) was from U. S. Biochemical Corp. All other nucleotides were from Boehringer Mannheim (Mannheim, Germany). Sources of other materials have been described elsewhere (11, 13, 26).

Construction of ₂ARG_sL and ₂ARG_sS DNAs-- ₂ARG_sL DNA was generated by a two-step PCR protocol using Pfu polymerase. A set of fusion primers (sense and antisense), encoding 18 base pairs from the C terminus of the ₂AR, 18 base pairs encoding a hexahistidine tag, and 21 base pairs from the N terminus of G_sL, were synthesized. In PCR 1A, the sequence between a primer 5' of the EcoRV site of the human ₂AR and the antisense fusion primer was amplified using ₂AR DNA in pGEM-3Z as template. In this vector, referred to as pGEM-3Z-SF-₂AR-6His, the ₂AR is tagged at the N terminus with the cleavable influenza-hemagglutinin signal sequence followed by the Flag epitope (IBI, New Haven, CT), and the C terminus of the ₂AR is tagged with a hexahistidine tail (Fig. 1A) (26). In PCR 1B, the sequence between the sense fusion primer and the antisense primer with an extra SalI site 3' of the stop codon of G_sL was amplified using rat G_sL DNA in pGEM-3Z as template. In PCR 2, the products of PCRs 1A and 1B were annealed and the sense primer 5' of the EcoRV site in the ₂AR sequence and the antisense primer 3' of the stop codon of G_sL were used. In this way, a fragment encoding the C terminus of the ₂AR, a hexahistidine tag, and G_sL was obtained. This fragment was digested with EcoRV and SalI and cloned into pGEM-3Z-SF-₂AR-6His digested with EcoRV and SalI to obtain the full-length fusion protein DNA sequence (pGEM-3Z-SF-₂AR-6His-G_sL). For generation of ₂ARG_sS DNA, a set of deletion primers (sense and antisense) and an antisense primer 3' of the EcoRI site of G_sL were synthesized. In PCR 3A, the sequence between a primer 5' of the EcoRV site in the ₂AR and the antisense deletion primer was amplified using pGEM-3Z-SF-₂AR-6His-G_sL as template. In PCR 3B, the sequence between the sense deletion primer and the antisense primer 3' of the EcoRI site of G_sL was amplified using the same template as in PCR 3A. In PCR 4, the products of PCRs 3A and 3B were annealed, and the sense primer 5' of the EcoRV site in the ₂AR and the antisense primer 3' of the EcoRI site of G_sL were used. In this way, a DNA fragment encoding the C terminus of the ₂AR, a hexahistidine tail and the N-terminal portion of G_sS, missing the sequence for amino acids 72-86 in G_sL and encoding the Glu-71  Asp substitution (15), was created (Fig. 1A). This fragment was digested with EcoRV and EcoRI and cloned into pGEM-3Z-SF-₂AR-6His-G_sL digested with EcoRV and EcoRI. PCR-generated DNA sequences were confirmed by enzymatic sequencing. Fusion protein DNAs were cut out from pGEM-3Z vectors with HindIII and SalI and cloned into the baculovirus transfer vector pVL 1392 (11). Recombination of viruses was confirmed by reverse transcriptase PCR.

Cell Culture-- Recombinant baculoviruses were generated and amplified as described (11). Sf9 cells were seeded at 3.0 × 10⁶ cells/ml and infected with 1:50 or 1:500 dilutions of high titer virus stocks. Cells were cultured for 24-48 h to obtain various expression levels of fusion proteins and ₂AR. For co-expression studies, Sf9 cells were infected with a 1:10,000 dilution of a high titer ₂AR baculovirus stock and a 1:50 dilution of a high titer G_sL baculovirus stock to achieve a receptor to G-protein stoichiometry of ~1:100. Cells were cultured for 48 h. Membranes were prepared according to Gether et al. (11).

[³H]DHA Binding-- For determination of K_d and B_max values, Sf9 membranes (5 µg of protein) were suspended in 500 µl of buffer containing 75 mM Tris/HCl, pH 7.4, 12.5 mM MgCl₂, and 1 mM EDTA, supplemented with 0.1-10 nM [³H]DHA and 0.2% (w/v) bovine serum albumin. Nonspecific binding was assessed in the presence of 10 µM ()-alprenolol (ALP). Incubations were performed for 90 min at 25 °C and shaking at 200 rpm. Competition binding experiments were carried out with 15-30 µg of membrane protein with 1 nM [³H]DHA in the presence of unlabeled ligands at various concentrations without or with guanosine 5'-O-(3-thiotriphosphate) (GTPS) (10 µM). In some experiments, tubes contained 1 nM [³H]DHA, 1 µM salbutamol (SAL), and various nucleotides at increasing concentrations.

GTPase Activity-- Assay tubes (100 µl) contained 10 µg of membrane protein, 0.1 mM [-³²P]GTP (0.1-0.5 mCi/tube), 1.0 mM MgCl₂, 0.1 mM EDTA, 0.1 mM ATP, 1 mM adenylyl imidodiphosphate, 5 mM creatine phosphate, 40 µg of creatine kinase, 0.2% (w/v) bovine serum albumin in 50 mM Tris/HCl, pH 7.4, and ligands at various concentrations. Reactions were conducted for 20 min at 25 °C and were terminated by the addition of 900 µl of a slurry consisting of 5% (w/v) activated charcoal and 50 mM NaH₂PO₄, pH 2.0. Reaction mixtures were centrifuged for 15 min at room temperature and 15,000 × g. Seven-hundred µl of the supernatant fluid of reaction mixtures were removed and [³²P]P_i was determined by liquid scintillation counting.

AC Activity-- Assay tubes (50 µl) contained 15 µg of membrane protein, 1 mM GTP, 40 µM [-³²P]ATP (2.5 µCi/tube), 2.7 mM mono(cyclohexyl)ammonium phosphoenolpyruvate, 0.125 IU of pyruvate kinase, 1 IU of myokinase, 0.1 mM cAMP, 5 mM MgCl₂, 0.4 mM EDTA, 30 mM Tris/HCl, pH 7.4, and ligands at various concentrations. Reactions were conducted for 20 min at 37 °C. Separation of [³²P]cAMP from [-³²P]ATP was performed as described (27).

Western Blot Analysis-- Solubilized Sf9 membrane proteins (5-10 µg of protein/lane) were separated by SDS-PAGE (8% (w/v) acrylamide). Proteins were visualized using either M1 antibody or anti-G_s antibody and the ECL Western blotting system (Amersham). G_sL expression in Sf9 membranes was quantitated by immunoblotting with anti-G_s antibody using defined amounts of ₂ARG_s fusion protein as standard.

Miscellaneous-- Protein was determined using the Bio-Rad DC protein assay kit (Bio-Rad). Data were analyzed by nonlinear regression, using the program Prism (GraphPad, Prism, San Diego, CA). Statistical comparisons between ₂ARG_sS and ₂ARG_sL were done with the Wilcoxon test. Data are given as means ± S.D. of three to seven independent experiments performed in duplicate or triplicate.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

Expression of ₂ARG_sS and ₂ARG_sL in Sf9 Membranes-- Expression of fusion proteins in Sf9 membranes was confirmed by SDS-PAGE using the M1 monoclonal antibody to detect the N-terminal Flag epitope of the ₂AR (Fig. 1A). The nonfused ₂AR expressed in Sf9 cells runs as a broad glycosylated 52-kDa protein in SDS-PAGE (11, 26). The apparent molecular masses of G_sS and G_sL are 45 and 52 kDa, respectively (16). Accordingly, the apparent molecular masses of ₂ARG_sS and ₂ARG_sL were expected to be 97 and 104 kDa, respectively. The data obtained are in agreement with this expectation (Fig. 1C). Immunoblots with an anti-G_s antibody confirmed the presence of G_s in the fusion proteins and the difference in apparent molecular mass between ₂ARG_sS and ₂ARG_sL. In membranes from uninfected cells, no immunoreactive bands in the 97-104-kDa region were detected with the M1 and anti-G_s antibody (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Structure of 2ARGs fusion proteins and three-dimensional model of transducin-. A, schematic structure of fusion proteins. The DNA of the human 2AR, tagged with the Flag epitope at the N terminus and a hexahistidine tag at its C terminus, was fused to the DNA of GsS or GsL. The differences in amino acid sequence between GsS and GsL are given in the single letter code. B, three-dimensional -carbon model of the -subunit of transducin. Blue, -helical domain; gray, Ras-like domain; red, linker 1 (where insert is located in GsL); orange, linker 2; green, 5-helix; yellow, GDP (van der Waals representation). The membrane would be in a horizontal plane below the molecule. The N terminus of transducin- is at the lower right. C, immunological characterization of fusion proteins. Sf9 membranes were separated by SDS-PAGE, transferred to nitrocellulose, and probed with M1 antibody or anti-Gs antibody as described under "Experimental Procedures." Numbers on the left indicate molecular masses of marker proteins. Shown are autoluminograms of gels containing 8% (w/v) acrylamide.

Ligand Binding Properties of ₂ARG_sS and ₂ARG_sL, Comparison with the Nonfused ₂AR-- The K_d values of [³H]DHA for ₂ARG_sS and ₂ARG_sL were very similar (Table I). In competition experiments, we studied the effects of ()-isoproterenol (()-ISO), (+)-isoproterenol ((+)-ISO), SAL, dobutamine (DOB), ()-ephedrine (EPH), dichloroisoproterenol (DCI) and ICI 118,551 (ICI) on [³H]DHA binding. ()-ISO binds to bARs with higher affinity than (+)-ISO, but both stereoisomers are full agonists (28-30). SAL, DOB, EPH, and DCI are partial ₂AR agonists (7, 10, 11), and ICI is an inverse agonist (8, 11-13). At both fusion proteins, full and strong partial agonists (()-ISO, (+)-ISO, SAL, and DOB) showed a high- and low-affinity binding component (Table I). The high-affinity agonist binding was abolished by GTPS. For agonists with lower intrinsic activity (EPH and DCI), high- and low-affinity binding sites were not discriminated by curve fitting analysis, but GTPS still reduced the affinity of these ligands to ₂ARG_s fusion proteins. There were no significant differences in the low- and high-affinity K_i values of the agonists studied between ₂ARG_sS and ₂ARG_sL. There was a trend toward higher fractions of high-affinity agonist-binding sites for full agonists and strong partial agonists at ₂ARG_sS compared with ₂ARG_sL, but this was significant only for (+)-ISO.

View this table: [in this window] [in a new window]   Table I Binding properties of 2AR ligands at 2ARGsS and 2ARGsL [3H]DHA binding was determined as described under "Experimental Procedures" in membranes expressing 2ARGsS or 2ARGsL at 3.5-6.5 pmol/µg. Competition binding isotherms were analyzed by nonlinear regression for best fit to single-site or two-site binding. Kh and Kl designate the dissociation constants for the high- and low-affinity state of the 2AR, respectively. %Rh indicates the percentage of high-affinity binding sites. When competition isotherms were best fit to a single-site model, the respective Ki values are listed under Kl. Ki (+GTPS) indicates the Ki values obtained in the presence of 10 µM GTPS. Data shown represent the means ± S.D. of five to seven independent experiments performed in duplicate or triplicate. Ki values are expressed in nanomolar.

Regulation of GTPase Activity in ₂AR_s Fusion Proteins, Comparison with a Co-expression System Consisting of ₂AR and G_s-- Activation of the GTPase of G_s by agonist-occupied ARs can be studied with great sensitivity in reconstituted systems (32, 33), but in most plasma membrane systems, the GTPase stimulation induced by ARs is small relative to the high background GTPase activity of other cellular G-proteins with higher GTP turnover than G_s and the presence of low-affinity nucleotidases (34, 35). In S49 lymphoma cell membranes, a prototypical system for studying ₂AR/G_s interaction (23, 36), the ₂AR and G_s are expressed at levels of ~0.2 and ~20 pmol/µg, respectively, i.e. there is an ~100-fold molar excess of G-protein compared with receptor (37). We co-expressed the ₂AR at a level of 1.4 pmol/µg with G_sL at a level of ~100 pmol/µg in Sf9 membranes, achieving a similar receptor/G-protein ratio as in S49 lymphoma cells, and studied the regulation of GTPase activity by ()-ISO and ICI. However, despite the relatively high expression of ₂AR and G_sL at a stoichiometry similar to that in the mammalian cell line, we detected only marginal activation of GTPase by agonist in Sf9 membranes and failed to see inhibition by inverse agonist (Fig. 2A). Similar results were obtained when the expression level of ₂AR was increased to 11.8 pmol/µg (data not shown). In marked contrast, ()-ISO increased GTP hydrolysis in membranes expressing ₂ARG_s (5.0 pmol/µg) by up to 245% above basal, and ICI reduced GTP hydrolysis by up to 50% (Fig. 2B). These findings demonstrate that fusion of the ₂AR to G_s greatly facilitates detection of ligand-regulated GTP hydrolysis in Sf9 membranes.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Comparison of the ligand regulation of GTPase activity in Sf9 membranes expressing the 2AR and GsL as separate proteins and in membranes expressing the 2ARGsL fusion protein. GTP hydrolysis was determined with 100 nM [-32P]GTP as substrate as described under "Experimental Procedures." Reaction mixtures contained ()-ISO and ICI at the indicated concentrations and membranes expressing the 2AR (1.4 pmol/µg) plus GsL (~100 pmol/µg) (A) or membranes expressing 2ARGsL (5.0 pmol/µg) (B). Data shown are the means ± S.D. of three to five independent experiments performed in duplicate.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Effects of ()-ISO and ICI on GTPase and AC activity in Sf9 membranes expressing 2ARGsS and 2ARGsL. GTPase activity (A and B) and AC activity (C and D) in 2ARGsS membranes (A and C) and 2ARGsL membranes (B and D) were determined as described under "Experimental Procedures" in the presence of ()-ISO or ICI at various concentrations. For GTPase studies, the expression level of fusion proteins was 4.5-5.0 pmol/µg, and for AC studies, 2.3-2.6 pmol/µg. GTP hydrolysis was determined with 100 nM [-32P]GTP as substrate. AC activity was determined in the presence of 1 µM GTP. Data shown are the means ± S.D. of three to five independent experiments performed in duplicate. The dotted lines are extrapolations of basal GTPase and AC activities to illustrate the relative contributions of ()-ISO and ICI at the ligand-regulated enzyme activities.

The Effects of G_sS and G_sL on the Efficacy of Agonists and Inverse Agonists at the ₂ARG_s Fusion Proteins-- The precise determination of the intrinsic activities of partial agonists constitutes a major problem in the functional characterization of GPCRs, because the intrinsic activity of a given ligand may depend on numerous variables, i.e. receptor and G-protein expression level and the availability of effector molecules such as AC (11, 22, 38-40). In most studies, the intrinsic activities of partial ₂AR agonists were characterized by measuring AC activity (7, 10, 11, 41). The AC assay takes advantage of the signal amplification at the G_s level, but it is difficult to control for the impact of G_s and AC availability on intrinsic activities of ligands. We reasoned that with the GTPase activity of ₂ARG_s fusion proteins as parameter, determination of the intrinsic activities of partial agonists should be less ambiguous because of the fixed stoichiometry of the signaling components. Moreover, signal amplification by AC is not required, thereby reducing the number of variables that can influence the determination of intrinsic activity. To validate this assumption, we studied the potencies and intrinsic activities of a series of partial ₂AR agonists at the GTPase of ₂ARG_sL with expression levels ranging from 0.6 to 7.6 pmol/µg. Within this broad range of expression, we did not observe significant differences in the potency and intrinsic activity of partial ₂AR agonists (data not shown).

View this table: [in this window] [in a new window]   Table II Potencies of full and partial 2AR agonists at the GTPase of 2ARGsS and 2ARGsL For determination of the potency of ligands, GTP hydrolysis was measured as described under "Experimental Procedures" in membranes expressing 2ARGsS and 2ARGsL at 4.5-5.0 pmol/µg. Reaction mixtures contained ligands at 0.1 nM to 1 mM as appropriate to obtain saturated concentration-response curves. EC50 values were calculated by nonlinear regression. Data shown represent the means ± S.D. of five to seven independent experiments performed in duplicate or triplicate. Potencies are expressed in nanomolar.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Intrinsic activities of full and partial 2AR agonists at the GTPase of 2ARGsS and 2ARGsL. For determination of the intrinsic activities of ligands, GTP hydrolysis was measured as described under "Experimental Procedures" in membranes expressing 2ARGsS and 2ARGsL at 4.5-5.0 pmol/µg. Reaction mixtures contained ligands at 0.1 nM to 1 mM as appropriate to obtain saturated concentration-response curves. The intrinsic activities were derived from plateau values of concentration-response curves generated by nonlinear regression analysis. A, comparison of the intrinsic activities of various ligands at 2ARGsS and 2ARGsL. *, p < 0.05. Data shown are the means ± S.D. of five to seven independent experiments performed in duplicate. B, replot of the data shown in A. The intrinsic activities of ligands at 2ARGsL were plotted against their intrinsic activities at 2ARGsS. Data points were best fitted by a hyperbolic function as assessed by nonlinear regression. The straight line represents the theoretical curve that would have fitted data best if the intrinsic activities of ligands at 2ARGsS and 2ARGsL had been the same. For comparison, the data for the neutral antagonist (propranolol) (PRO) are included in the panel.

Regulation of High-affinity Agonist Binding at ₂ARG_sS and ₂ARG_sL by Guanine Nucleotides-- High-affinity agonist binding to GPCRs depends on their interaction with G-protein -subunits, presumably in the nucleotide-free state (1, 42). Occupation of the guanine nucleotide-binding site of -subunits disrupts high-affinity agonist binding (1, 7). To determine the guanine nucleotide binding affinities of G_sS and G_sL in ₂ARG_s fusion proteins, we examined binding of a fixed concentration of the antagonist [³H]DHA in the presence of a subsaturating concentration of the strong partial agonist SAL in membranes expressing ₂ARG_sS and ₂ARG_s. Various nucleotides at increasing concentrations were added to the binding assays. Guanine nucleotide binding to G_s reduces the affinity of the ₂AR for agonist and, thereby, increases [³H]DHA binding (Fig. 5). In this way, the affinity of G-proteins for nucleotides can be measured. It should be noted that our binding experiments were performed in the absence of a nucleotide-regenerating system, excluding the possibility that effects caused by nucleoside 5'-monophosphates and -diphosphates are due to transphosphorylation. GTP was similarly potent at inhibiting high-affinity agonist binding at ₂ARG_sS and ₂ARG_sL (EC₅₀, 59 ± 15 and 49 ± 20 nM, respectively). In contrast, GDP was far more potent at ₂ARG_sS (EC₅₀, 83 ± 23 nM) than at ₂ARG_sL (EC₅₀, 1.8 ± 0.2 µM). Like GDP, its nucleotidase-resistant phosphorothioate analog, guanosine 5'-O-(2-thiodiphosphate) (GDPS), inhibited agonist binding at ₂ARG_sS more potently than at ₂ARG_sL (EC₅₀, 490 ± 150 nM and 2.2 ± 0.3 µM, respectively). These data show that in the ₂ARG_s fusion proteins, G_sS has a higher affinity for guanosine 5'-diphosphates than G_sL and are in agreement with data obtained with purified G_sL and G_sS (18).

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Effects of guanine nucleotides on high-affinity agonist binding in Sf9 membranes expressing 2ARGsS and 2ARGsL. Binding experiments were carried out as described under "Experimental Procedures" with 2ARGsS membranes (A) or 2ARGsL membranes (B). Reactions mixtures additionally contained 1 nM [3H]DHA, 1 µM SAL, and guanine nucleotides at increasing concentrations. Data shown are the means ± S.D. of two independent experiments performed in triplicate.

Regulation of AC Activity in Sf9 Membranes Expressing ₂ARG_sS and ₂ARG_sL by Agonist and Inverse Agonist-- The analysis of AC activity in Sf9 membranes expressing ₂ARG_sS and ₂ARG_sL must take into consideration the fact that Sf9 cells express endogenous G_s-like G-proteins (10, 11, 30, 31). This is of particular relevance because for AC studies, we expressed fusion proteins at relatively low levels to avoid AC availability becoming the limiting factor (38). However, the AC activity in membranes from uninfected Sf9 cells in the presence of 10 µM GTPS was ~5.5-fold lower than in Sf9 membranes expressing ₂ARG_sL at 2.6 pmol/µg (0.089 ± 0.015 nmol/µg/20 min versus 0.491 ± 0.054 nmol/µg/20 min). These data show that even under maximal stimulation of AC, the contribution of endogenous G_s-like G-proteins in Sf9 cells to total AC activity is small.

AC Regulation in Sf9 Membranes Expressing ₂ARG_sS and ₂ARG_sL in the Absence of Exogenous Guanine Nucleotides-- In the presence of GTP, agonists at G_s-coupled GPCRs cause AC activation (1, 2). However, in the absence of added guanine nucleotides, agonists at G_s-coupled receptors can reduce AC activity (43, 44). The most likely explanation for these observations is that agonists induce release of prebound guanine nucleotide from G_s, generating guanine nucleotide-free G_s and, thereby, reducing AC activity. Indeed, ()-ISO reduced the basal AC activity in membranes expressing ₂ARG_sS by about 30% and with an IC₅₀ of 40 ± 12 nM (Fig. 6). In contrast, ()-ISO had no significant deactivating effect on AC activity in membranes expressing ₂ARG_sL. These results suggest that the nucleotide-binding pocket of G_sL in ₂ARG_sL is already nucleotide-free.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Deactivation of AC by ()-ISO in the absence of exogenous nucleotides. AC activity in 2ARGsS and 2ARGsL membranes was determined as described under "Experimental Procedures" in the absence of exogenous guanine nucleotides and in the presence of ISO at different concentrations. The expression level of fusion proteins was 2.3-2.6 pmol/µg. Data shown are the means ± S.D. of three to four independent experiments performed in duplicate.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

The ₂AR Fused to G_sL Has Properties of a Constitutively Active Receptor-- Previous studies have shown that there are subtle differences in the GDP affinities of purified G_sS and G_sL (18) and that G_sS may activate AC more efficiently than G_sL (19). However, studies aiming to reveal differences between G_sS and G_sL in their coupling to the ₂AR have remained inconclusive because of the difficulties to ensure exactly defined receptor/G-protein stoichiometry (18, 20, 21). This is important because functional interactions between GPCRs and G-proteins are strongly influenced by their relative expression levels (11, 22, 40). To circumvent this problem, we constructed fusion proteins in which the C terminus of the ₂AR was linked to the N terminus of G_sS or G_sL (Fig. 1A), thereby guaranteeing a defined stoichiometry of receptor to G-protein and increasing the efficiency of receptor/G-protein coupling. Using this approach we observed that the efficacy and potency of partial agonists acting on the ₂AR were significantly higher when the receptor was fused to G_sL than when the receptor was fused to G_sS (Fig. 4 and Table II). Moreover, the basal GTPase and AC activities in membranes expressing ₂ARG_sL were more sensitive to inverse agonists than the corresponding activities in membranes expressing ₂ARG_sS (Fig. 3). These functional properties of the ₂AR fused to G_sL are similar to those of the ₂AR_CAM (7, 8, 13).

Physiological Considerations-- Although constitutive activation of GPCRs is easily observed with high receptor expression levels (10-12), this is not a prerequisite. There are several examples in the literature documenting constitutive GPCR activity at physiological or near-physiological expression levels (5, 6, 9, 14). These data raise the possibility that constitutive activity of GPCRs is of relevance in vivo and that the R* state can be more readily stabilized or detected by specific G-protein -subunits. In agreement with such a concept is the finding that increases in expression of specific G-proteins can increase high-affinity agonist binding and can promote constitutive receptor activation (40, 46).

Conclusion-- The 15-amino acid insert by which G_sL differs from G_sS (Fig. 1B) lowers the GDP affinity of the G-protein. Using fusion proteins of the ₂AR with G_s splice variants, which ensure precise receptor/G-protein stoichiometry, we could show that the subtle differences in GDP affinity between G_sS and G_sL have important consequences for the interaction with the ₂AR, i.e. G_sL confers to the ₂AR some properties of a constitutively active protein. Future studies will have to examine the effects of partial and inverse agonists of the ₂AR in tissues and cells expressing G_sS and G_sL at different levels and to further define the physiological and pharmacological implications of the differences that we have discovered for the interaction of the ₂AR with the two splice variants of G_s. Because the overall properties of the ₂AR and G_s and their interaction were not changed as a result of fusion, this approach may be applied to a broad variety of receptors and G-proteins to uncover subtle differences in the interaction of closely related G-protein -subunits with GPCRs.