Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Many intercellular signal molecules exert their biological effects via GPCRs. GPCRs interact with heterotrimeric G proteins that, in turn, regulate the activity of effector systems (Gilman, 1987; Birnbaumer et al., 1990). Upon binding of agonist, GPCRs undergo a conformational change, enabling them to promote the dissociation of GDP from G. Subsequently, the agonist-occupied GPCR forms a ternary complex with the nucleotide-free G protein that exhibits high agonist affinity. The GPCR then catalyzes GTP binding to G, followed by the disruption of the ternary complex and dissociation of the G protein into the -complex and G_-GTP. Both G_-GTP and the -complex regulate the activity of effector systems. The GTPase of G deactivates the G protein. The extended ternary complex model assumes that GPCRs isomerize from an inactive (R) state to an active (R*) state and that agonists stabilize the R* state (Lefkowitz et al., 1993; Gether and Kobilka, 1998). R to R* isomerization can also occur independently of agonist and is referred to as constitutive activity. Inverse agonists stabilize the R state and decrease basal G protein activity. In GPCR/G fusion proteins, the C terminus of GPCR is tethered to the N terminus of G. Fusion proteins ensure a defined 1:1 stoichiometry and efficient coupling of the signaling partners and allow for the analysis of GPCR/G protein coupling under exactly defined experimental conditions (Seifert et al., 1999a; Milligan, 2000).

The ₂AR is a prototypical G_s-coupled GPCR (Gilman, 1987; Birnbaumer et al., 1990). G_s exists as two splice variants, G_sL and G_sS (Graziano et al., 1989; Seifert et al., 1998b). The ₂AR fused to G_sL exhibits higher apparent constitutive activity than the ₂AR fused to G_sS (Seifert et al., 1998b; Wenzel-Seifert and Seifert, 2000; Seifert, 2001). Specifically, the efficacies and potencies of partial agonists and the efficacies of inverse agonists are higher at the ₂AR-G_sL fusion protein than at the ₂AR-G_sS fusion protein. We explained those differences in apparent constitutive activity by a model in which the agonist-free ₂AR and the ₂AR occupied by partial agonists promote GDP dissociation from G_sL more efficiently than from G_sS because G_sL possesses a lower GDP affinity than G_sS (Graziano et al., 1989; Seifert et al., 1998b; Seifert, 2001). However, it is unknown whether this model also applies to G_s-coupled GPCRs other than the ₂AR.

To answer this question, we studied H₂R-G_s fusion proteins. We chose the H₂R for several reasons. First, like the ₂AR, the H₂R is a very well studied G_s-coupled GPCR (Hill et al., 1997). Second, the H₂R exhibits constitutive activity and inverse agonists for the H₂R have been identified (Smit et al., 1996; Alewijnse et al., 1998). Third, several partial agonists for the human H₂R have been described, namely, the HIS-related BET (4) (Burde et al., 1989), and several arpromidine-derived guanidines (6, 7, 10, and 11) (Burde et al., 1990) (Fig. 1). Fourth, the overall mobility and arrangement of the coupling partners in H₂R-G_s- and ₂AR-G_s fusion proteins should be similar because the C termini of both GPCRs, serving as a tether between GPCR core and G have a similar length (Gantz et al., 1991; Wenzel-Seifert et al., 1998).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Structures of H2R agonists. Compounds 1 to 4 represent HIS and HIS-related compounds. Compounds 5 to 13 represent guanidines.

In our present study we analyzed the effects of HIS (1), three HIS-related agonists (2-4), and nine guanidines (5-13) (Fig. 1) as well as of the antagonists/inverse agonists CIM (14), RAN (14), ZOL (15), TIO (16) and FAM (17) on H₂R-G_sL- and H₂R-G_sS fusion proteins expressed in Sf9 insect cells. Here we report that in contrast to ₂AR-G_sL and ₂AR-G_sS, H₂R-G_sL, and H₂R-G_sS exhibit similar apparent constitutive activity.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. The cDNA for the human H₂R was kindly provided by Dr. I. Gantz (University of Michigan Medical School and Ann Arbor Veterans Affairs Medical Center, Ann Arbor, MI) (Gantz et al., 1991). IMP was prepared as described (Durant et al., 1978). Guanidines 6-11 were synthesized as described (Buschauer, 1989). Guanidines 12 and 13 were prepared by analogy to the procedures described for guanidines 6-11 (Buschauer, 1989). The anti-FLAG Ig (M1 monoclonal antibody) was from Sigma (St. Louis, MO). The anti-G_s Ig (C-terminal) was from Calbiochem (La Jolla, CA). [-³²P]GTP (6000 Ci/mmol), [³⁵S]GTPS (1100 Ci/mmol), and [³H]TIO (90 Ci/mmol) were from PerkinElmer Life Sciences (Boston, MA). All unlabeled nucleotides were from Roche Molecular Biochemicals (Indianapolis, IN). HIS, BET, CIM, RAN, and FAM were from Sigma. Amthamine, TIO, and ZOL were from Tocris Cookson (Ballwin, MO). Dimaprit was from Sigma/RBI (Natick, MA). All restriction enzymes and T4 DNA ligase were from New England Biolabs (Beverly, MA). Cloned Pfu DNA polymerase was from Stratagene (La Jolla, CA).

Construction of FLAG Epitope- and Hexahistidine-Tagged cDNA for H₂R-G_sS and H₂R-G_sL. A DNA sequence encoding the cleavable signal peptide from influenza hemagglutinin (S) followed by the FLAG epitope (F), which is recognized by the M1 antibody, was placed 5' of the start codon of the H₂R to enhance GPCR expression and allow immunological detection. We also added a hexahistidine tag to the C terminus of H₂R to allow future purification and to provide additional protection against proteolysis (Seifert et al., 1998a). The GPCR modifications were generated by sequential overlap-extension PCRs. In PCR 1A, the DNA sequence of the N-terminal portion of the H₂R was amplified using CMVneo-H₂R as template. The sense primer annealed with the first 18 base pairs of the 5' end of the H₂R and included the last 18 base pairs of the SF in its 5' extension. The antisense primer encoded the sequence GAGCTGTTGATATCCGGTGCGGAAGTCTCTG to generate a silent mutation yielding a new EcoRV site. In PCR 1B, the DNA sequence of the C-terminal portion of the H₂R was amplified using CMVneo-H₂R as template. The sense primer encoded the sequence TTCCGCACCGGATATCAACAGCTCTTCTGCTGC to generate the new EcoRV site. The antisense primer encoded the five C-terminal amino acids of the H₂R, a hexahistidine tag, the stop codon and an XbaI site. In PCR 2, the products of PCRs 1A and 1B annealed in the region encoding the newly created EcoRV site, and the sense primer of PCR 1A and the antisense primer of PCR 1B were used. In this way, a fragment encoding the signal sequence, the FLAG epitope, H₂R cDNA with a new EcoRV site, and a hexahistidine tag followed by an XbaI site was obtained. This fragment was digested with NcoI and XbaI and cloned into pGEM-3Z-SF-human formyl peptide receptor-6His digested with NcoI and XbaI. In PCR 3A, the C-terminal portion of the H₂R was amplified using pGEM-3Z-SF-hH₂R as template, a sense primer annealing 5' of the newly created EcoRV site and an antisense primer annealing with the hexahistidine tag. In PCRs 3B₁ and 3B₂, the sequences of G_sL and G_sS were amplified, using pGEM-3Z-SF-₂AR-G_sL and pGEM-3Z-SF-₂AR-G_ss, respectively, as template, a sense primer annealing with the hexahistidine tag, and an antisense primer annealing with the five C-terminal amino acids of G_s, the stop codon, and an XbaI site. In PCRs 4A and 4B, the products of PCRs 3A and 3B₁ and PCRs 3A and 3B₂, respectively, annealed in the hexahistidine region, and the sense primer of PCR 3A and the antisense primer of PCR 3B were used. In this way, fragments encoding the C-terminal portion of the H₂R, a hexahistidine tag, G_sL and G_sS, respectively, a stop codon, and an XbaI site were created. These fragments were digested with EcoRV and XbaI and cloned into pGEM-3Z-SF-H₂R digested with EcoRV and XbaI. In this way, the full-length cDNAs for H₂R-G_sL and H₂R-G_sS were created. pGEM-3Z-SF-H₂R-G_sL and pGEM-3Z-SF-H₂R-G_sS were digested with NcoI and XbaI to recover the fusion protein cDNAs and cloned into the baculovirus transfer vector pVL 1392-SF-₂AR-G_i2 digested with NcoI and XbaI. PCR-generated DNA sequences were confirmed by restriction enzyme analysis and enzymatic sequencing.

Generation of Recombinant Baculoviruses, Cell Culture, and Membrane Preparation. Recombinant baculoviruses encoding H₂R-G_s fusion proteins were generated in Sf9 cells by using the BaculoGOLD transfection kit (Pharmingen, San Diego, CA) according to the manufacturer's instructions. After initial transfection, high-titer virus stocks were generated by two sequential virus amplifications. Sf9 cells were cultured in 250-ml disposable Erlenmeyer flasks at 28°C under rotation at 125 rpm in SF 900 II medium (Invitrogen, Carlsbad, CA) supplemented with 5% (v/v) fetal calf serum (BioWhittaker, Walkersville, MD) and 0.1 mg/ml gentamicin (BioWhittaker). Cells were maintained at a density of 1.0 to 6.0 × 10⁶ cells/ml. For infection, cells were sedimented by centrifugation and suspended in fresh medium. Cells were seeded at 3.0 × 10⁶ cells/ml and infected with 1:100 dilutions of high-titer baculovirus stocks encoding H₂R-G_s fusion proteins. Cells were cultured for 48 h before membrane preparation. Sf9 membranes were prepared as described (Seifert et al., 1998a), by using 1 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride, 10 µg/ml benzamidine, and 10 µg/ml leupeptin as protease inhibitors. Membranes were suspended in binding buffer (12.5 mM MgCl₂, 1 mM EDTA, and 75 mM Tris/HCl, pH 7.4) and stored at 80°C until use.

[³H]TIO Binding Assay. Membranes were thawed and sedimented by a 15-min centrifugation at 4°C and 15,000g to remove residual endogenous guanine nucleotides as much as possible. Membranes were resuspended in binding buffer. Each tube (total volume 250 µl) contained 200 to 250 µg of protein. Tubes contained 1 to 20 nM [³H]TIO plus unlabeled TIO to obtain final ligand concentrations of up to 300 nM. Nonspecific binding was determined in the presence of [³H]TIO at various concentrations plus 100 µM unlabeled TIO. Incubations were conducted for 90 min at 25°C and shaking at 250 rpm. Bound [³H]TIO was separated from free [³H]TIO by filtration through GF/C filters, followed by three washes with 2 ml of binding buffer (4°C). Filter-bound radioactivity was determined by liquid scintillation counting. The experimental conditions chosen ensured that not more than 5% of the total amount of [³H]TIO added to binding tubes was bound to filters.

[³⁵S]GTPS Binding Assay. Membranes were thawed, sedimented, and suspended in binding buffer. For time course studies, Sf9 membranes expressing H₂R-G_s fusion proteins were suspended in 1500 µl of binding buffer supplemented with 1 nM [³⁵S]GTPS plus 9 nM unlabeled GTPS, 1 µM GDP, and distilled water (basal), HIS (100 µM), or RAN (10 µM). Aliquots of 200 µl containing 15 to 25 µg of protein were withdrawn at different time points. For saturation binding experiments, reaction mixtures (total volume 500 µl) contained Sf9 membranes expressing H₂R-G_s fusion proteins (15 µg of protein/tube) in binding buffer supplemented with 0.05% (w/v) bovine serum albumin, 1 µM GDP, and 0.1 to 2 nM [³⁵S]GTPS plus unlabeled GTPS to reach the final ligand concentrations indicated on the abscissa of Fig. 5. Reaction mixtures additionally contained distilled water (basal), HIS (100 µM), or RAN (10 µM). Incubations were conducted for 90 min at 25°C and shaking at 250 rpm. Bound [³⁵S]GTPS was separated from free [³⁵S]GTPS by filtration through GF/C filters, followed by three washes with 2 ml of binding buffer (4°C). Filter-bound radioactivity was determined by liquid scintillation counting. The experimental conditions chosen ensured that no more than 10% of the total amount of [³⁵S]GTPS added was bound to filters.

Steady-State GPase Activity Assay. Membranes were thawed, sedimented, and resuspended in 10 mM Tris/HCl, pH 7.4. Assay tubes contained Sf9 membranes expressing H₂R-G_s fusion proteins (10 µg of protein/tube), 1.0 mM MgCl₂, 0.1 mM EDTA, 0.1 mM ATP, 100 nM GTP, 1 mM adenylyl imidodiphosphate, 5 mM creatine phosphate, 40 µg of creatine kinase, and 0.2% (w/v) bovine serum albumin in 50 mM Tris/HCl, pH 7.4, and H₂R ligands at various concentrations. Reaction mixtures (80 µl) were incubated for 3 min at 25°C before the addition of 20 µl of [-³²P]GTP (0.2-0.5 µCi/tube). All stock and work dilutions of [-³²P]GTP were prepared in 20 mM Tris/HCl, pH 7.4. Reactions were conducted for 20 min at 25°C. Reactions were terminated by the addition of 900 µl of a suspension consisting of 5% (w/v) activated charcoal and 50 mM NaH₂PO₄, pH 2.0. Charcoal-quenched reaction mixtures were centrifuged for 15 min at room temperature at 15,000g. The supernatant fluid (700 µl) of reaction mixtures was removed, and ³²P_i was determined by liquid scintillation counting. Enzyme activities were corrected for spontaneous degradation of [-³²P]GTP. Spontaneous [-³²P]GTP degradation was determined in tubes containing all of the above-described components plus a very high concentration of unlabeled GTP (1 mM) that, by competition with [-³²P]GTP, prevents [-³²P]GTP hydrolysis by enzymatic activities present in Sf9 membranes. Spontaneous [-³²P]GTP degradation was <1% of the total amount of radioactivity added. The experimental conditions chosen ensured that not more than 10% of the total amount of [-³²P]GTP added was converted to ³²P_i.

SDS-PAGE and Immunoblot Analysis. Membrane proteins were separated on SDS polyacrylamide gels containing 10% (w/v) acrylamide. Proteins were then transferred onto Immobilon P transfer membranes (Millipore Corporation, Bedford, MA). Membranes were reacted with M1 antibody or anti-G_s Ig (1:1000 each). Immunoreactive bands were visualized by sheep anti-mouse IgG (M1 antibody) and donkey anti-rabbit IgG (anti-G_s Ig), respectively, coupled to peroxidase, using o-dianisidine and H₂O₂ as substrates.

Miscellaneous. Protein concentrations were determined using the Bio-Rad DC protein assay kit (Bio-Rad, Hercules, CA). All analyses of experimental data were performed with the Prism III program (GraphPad Software, San Diego, CA).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Immunological Detection of H₂R-G_sL and H₂R-G_sS in Sf9 Cell Membranes. Nonfused H₂R expressed in Sf9 cells migrates as ~33-kDa band in SDS-PAGE (Fukushima et al., 1997). The molecular masses of G_sL and G_sS are ~45 and 52 kDa, respectively (Graziano et al., 1989). Thus, the molecular masses of H₂R-G_sS and H₂R-G_sL were expected to be ~78 and 85 kDa, respectively. In fact, the anti-FLAG Ig and anti G_s Ig detected proteins of the expected masses in immunoblots (Fig. 2). The double bands represent differently glycosylated forms of the fusion proteins (Liu et al., 2001). It should be noted that neither with anti-FLAG Ig nor with anti-G_s Ig, immunoreactive bands below the fusion proteins were detected, indicating that H₂R-G_sL and H₂R-G_sS were not proteolytically degraded.

View larger version (40K): [in this window] [in a new window]   Fig. 2.   Analysis of the expression of H2R-Gs fusion proteins in Sf9 cell membranes. Sf9 membranes (50 µg of protein/lane) expressing H2R-GsL at 4.9 pmol/mg and H2R-GsS at 2.5 pmol/mg as assessed by ligand-regulated GTPS binding were separated by SDS-PAGE with a gel that contained 10% (w/v) acrylamide. Fusion proteins were probed with the anti-FLAG Ig (M1 antibody) (A) or anti-Gs Ig (B). Numbers on the left of the immunoblot indicate molecular masses of marker proteins. Shown are the horseradish peroxidase-reacted Immobilon P membranes of representative gels. Similar results were obtained with three other membrane preparations of H2R-GsL and H2R-GsS.

[³H]TIO Saturation Binding to H₂R-G_sL and H₂R-G_sS Expressed in Sf9 Cell Membranes. Native H₂R binds [³H]TIO with a K_d of ~17 nM (Gajtkowski et al., 1983). However, the use of [³H]TIO in native organs is severely limited by the fact that nonspecific binding with saturating [³H]TIO concentrations amounts to ~85 to 90% of total [³H]TIO binding. In Sf9 membranes, only ~55 to 65% nonspecific [³H]TIO binding occurred with saturating radioligand concentrations. Therefore, a more precise determination of the kinetics of specific [³H]TIO binding was possible (Fig. 3). H₂R-G_s expressed in Sf9 membranes bound [³H]TIO according to monophasic saturation curves: H₂R-G_sL with a K_d of 31.8 ± 4.4 nM and a B_max of 0.52 ± 0.04 pmol/mg (membrane preparation 171; data not shown); H₂R-G_sS with a K_d of 32.0 ± 4.6 nM and a B_max of 0.43 ± 0.02 pmol/mg (membrane preparation 166; Fig. 3). For GTPS binding and GTPase experiments, we used membranes expressing H₂R-G_s fusion proteins with B_max values of [³H]TIO binding between 0.31 and 0.64 pmol/mg.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   [3H]TIO saturation binding in Sf9 membranes expressing H2R-GsS. Sf9 membranes expressing H2R-GsS were incubated in the presence of [3H]TIO at the concentrations indicated on the abscissa as described under Experimental Procedures. Nonspecific binding is the [3H]TIO binding not competed for by 100 µM unlabeled TIO. Specific binding is the difference between total [3H]TIO binding and nonspecific [3H]TIO binding for a given [3H]TIO concentration. A, data were analyzed by nonlinear regression and are the means ± S.D. of three experiments performed in duplicate. The Kd value of specific [3H]TIO binding was 32.0 ± 4.6 nM, and the Bmax value was 0.43 ± 0.02 nM. B, specific [3H]TIO binding data were plotted according to Scatchard. The Kd value in the analysis according to Scatchard was 32.7 nM, and the Bmax value was 0.43 ± 0.02 pmol/mg. Similar results were obtained with five different membrane preparations expressing H2R-GsS or H2R-GsL.

[³⁵S]GTPS Binding Studies. Unlike GTP, GTPS is not hydrolyzed by G proteins (Gilman, 1987), and G proteins bind GTPS with ~100-fold higher affinity than GTP (Seifert et al., 1999b; Wenzel-Seifert et al., 1999; Wenzel-Seifert and Seifert, 2000). Thus, the kinetics of GPCR-mediated GDP/GTP exchange can be readily studied in the [³⁵S]GTPS binding assay. Figure 4 shows the time course of GTPS binding in membranes expressing H₂R-G_sL and H₂R-G_sS. In membranes expressing H₂R-G_sL, the t_1/2 values of GTPS binding under basal conditions and in the presence of HIS and RAN were 16.5 ± 2.1, 6.7 ± 0.5, and 17.1 ± 1.6 min, respectively. The corresponding values for H₂R-G_sS were 34.5 ± 6.4, 12.7 ± 1.5, and 36.9 ± 6.8 min, respectively. These data show that at both fusion proteins, HIS decreases t_1/2 of GTPS binding, indicative for accelerated GDP/GTPS exchange. RAN, which has been reported to be an inverse agonist at the H₂R (Smit et al., 1996; Alewijnse et al., 1998), had only small inhibitory effects on basal GTPS binding, both in membranes expressing H₂R-G_sS and H₂R-G_sL. Under all conditions studied, the t_1/2 values were ~2-fold lower for H₂R-G_sL than for H₂R-G_sS. These data are consistent with the fact that the intrinsic GDP dissociation rate of G_sL is higher than the GDP dissociation rate of G_sS (Graziano et al., 1989).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Time course of [35S]GTPS binding in Sf9 membranes expressing H2R-GsL and H2R-GsS. [35S]GTPS binding in Sf9 membranes was determined as described under Experimental Procedures. Reaction mixtures contained membranes expressing H2R-GsL (A) or H2R-GsS (B) and distilled water (basal), HIS (100 µM), or RAN (10 µM). Binding reactions were conducted for the periods indicated on the abscissa. Data were best fitted to monophasic saturation curves (F test). Data shown are the means ± S.D. of three experiments performed in triplicate.

sL

sL

sL

sS

sS

sS

sL

View larger version (20K): [in this window] [in a new window]   Fig. 5.   [35S]GTPS saturation binding in Sf9 membranes expressing H2R-GsL and H2R-GsS. [35S]GTPS binding in Sf9 membranes was performed as described under Experimental Procedures. Reaction mixtures contained membranes expressing H2R-GsL (A) or H2R-GsS (B) and distilled water (basal), HIS (100 µM), or RAN (10 µM). For each GTPS concentration, the basal GTPS binding was subtracted from GTPS binding observed in the presence of HIS to obtain the HIS-stimulated GTPS binding. In addition, GTPS binding observed in the presence of RAN was subtracted from basal GTPS binding to obtain RAN-inhibited GTPS binding. The dashed lines are extrapolations of basal GTPS binding. Data were best fitted to monophasic saturation curves (F test). Data shown are the means ± S.D. of three experiments performed in triplicate.

s

s

Analysis of Effects of H₂R Ligands in Steady-State GTPase Assay. We determined the potencies and efficacies of 13 H₂R agonists and the efficacies of five potential inverse agonists at H₂R-G_sL and H₂R-G_sS in the steady-state GTPase assay (Table 1). Using this assay, we previously showed that partial agonists and inverse agonists possess higher efficacies at ₂AR-G_sL than at ₂AR-G_sS (Seifert et al., 1998b). Additionally, partial agonists possess higher potencies at ₂AR-G_sL than at ₂AR-G_sS. Ligand potencies and efficacies in the GTPase assay are independent of the fusion protein expression level, allowing for the comparison of various membrane preparations with different expression levels (Seifert et al., 1998a, 1999a). Similar observations were made for H₂R-G_s fusion proteins (data not shown). The most likely explanation for the expression level independence of agonist efficacies at fusion proteins is the defined receptor/G stoichiometry and the apparent independence of fusion protein function of other factors such as -complexes (Seifert et al., 1999a). At ₂AR-G_s fusion proteins, the efficacies of agonists were referred to the efficacy of the full agonist ()-isoproterenol (Seifert et al., 1998b). At H₂R-G_s fusion proteins, the efficacies of agonists were referred to the efficacy of the full agonist HIS. This approach ensured comparison of ₂AR-G_s- and H₂R-G_s fusion proteins with each other as exactly as possible.

View this table: [in this window] [in a new window]   TABLE 1 Ligand potencies and efficacies at H2R-GsL and H2R-GsS expressed in Sf9 cell membranes Potencies and efficacies of ligands at H2R-GsS and H2R-GsL were determined in the GTPase assay as described under Experimental Procedures. Reaction mixtures contained Sf9 membranes expressing fusion proteins and agonists at concentrations from 1 nM-100 µM as appropriate to generate saturated concentration-response curves. Curves were analyzed by nonlinear regression. To calculate agonist efficacies, the maximum stimulatory effect of HIS was set 1.00, and the stimulatory effects of other agonists were referred to this value. To determine the inverse agonist efficacies, the effects of ligands at a fixed concentration (10 µM each) on basal GTPase activity were assessed and referred to the maximum stimulatory effect of HIS (= 1.00). Basal and maximum HIS-stimulated GTPase activities ranged between ~1 and 2 and ~4 to 8 pmol/mg/min, respectively. Data shown are the means ± S.D. of four to eight experiments performed in duplicate. Efficacies and potencies, respectively, of ligands at H2R-GsL were compared to the corresponding parameters at H2R-GsS by using the t test.

sL

sS

sS

sL

sL

sS

sL

sS

sL

sS

sL

sS

sL

sS

sS

sL

sL

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Relations between agonist potencies and agonist efficacies at H2R-GsL and H2R-GsS. pD2 values were derived from the EC50 values shown in Table 1, and agonist efficacies were directly taken from Table 1. Solid lines represent the actual correlations obtained. Dotted lines represent the 95% confidence intervals of the correlations. The straight dashed lines represent the theoretical functions that would have been obtained if pD2 values and efficacies, respectively, had been identical at the two fusion proteins. The theoretical functions have a slope of 1.00. A, correlation of pD2 values at H2R-GsS versus H2R-GsL. Slope, 0.86 ± 0.08 (95% confidence interval, 0.68-1.05); r2 = 0.91; slope significantly different from zero with p = < 0.0001. B, correlation of efficacies at H2R-GsS versus H2R-GsL. Slope, 1.18 ± 0.04 (95% confidence interval, 1.09-1.27); r2 = 0.98; slope significantly different from zero with p = < 0.0001.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The aim of the present study was to answer the question whether the H₂R, like the ₂AR, exhibits higher apparent constitutive activity when fused to G_sL than when fused to G_sS. The differences in apparent constitutive activity of the ₂AR fused to G_sL and G_sS are explained by the lower GDP affinity of G_sL relative G_sS, i.e., the agonist-free ₂AR and the ₂AR occupied by partial agonists promote GDP dissociation from G_sL more efficiently than from G_sS (Seifert et al., 1998b; Wenzel-Seifert and Seifert, 2000; Seifert, 2001). Experimentally, this results in increased potency and efficacy of partial agonists and increased efficacy of inverse agonists at ₂AR-G_sL compared with ₂AR-G_sS. Consistent with the known differences in GDP affinity of G_sL and G_sS (Graziano et al., 1989; Seifert et al., 1998b), basal and HIS-stimulated GDP/GTPS exchange proceeded faster at H₂R-G_sL than at H₂R-G_sS (Fig. 4).

Although the fusion protein approach certainly has advantages for the question addressed in this study, namely, the defined 1:1 stoichiometry of the coupling partners and the high sensitivity both in the GTPS binding and GTPase assay, one has also to keep in mind certain limitations when extrapolating the results obtained with fusion proteins to nonfused systems. Most importantly, fusion proteins are artificial proteins. Although the fusion does not alter the properties of GPCRs and G fundamentally (Seifert et al., 1999a), it cannot be excluded that there are subtle differences in GPCR/G interaction between fused and nonfused systems that could well have an impact on the coupling of GPCRs to very closely related G subunits. This notion is supported by the fact that differences in the interaction of the ₂AR with G_s splice variants are readily detected in fusion proteins (Seifert et al., 1998b; Wenzel-Seifert and Seifert, 2000; Seifert, 2001), but not in nonfused systems (Graziano et al., 1989; Jones et al., 1990; O'Donnell et al., 1991). Even more caution than for the comparison of the coupling of a given GPCR to various fused G subunits has to be exerted when comparing the coupling of various GPCRs to a given G. Specifically, the intracellular G_s-coupling domains of the ₂AR and H₂R are quite different from each other (Dohlman et al., 1991; Kobilka, 1992; Wang et al., 2000), and in a constrained system such as fusion proteins, such differences may accentuate differences in GPCR/G coupling compared with nonfused systems. Therefore, it will be necessary to repeat the experiments reported in this study with nonfused proteins. However, studies with nonfused proteins have drawbacks as well. Specifically, it is very difficult to obtain exactly defined GPCR/G-protein stoichiometries in nonfused systems (Jones et al., 1990; Brys et al., 2000; Seifert and Wenzel-Seifert, 2001). Ultimately, it will be necessary to compare the crystal structures of GPCR-G fusion proteins versus GPCR coexpressed with G to assess the validity of the fusion protein approach.

With the above-discussed limitations in mind, we compared ₂AR-G_s fusion proteins with H₂R-G_s fusion proteins. Based on the data obtained with ₂AR-G_s fusion proteins (Seifert et al., 1998b; Wenzel-Seifert and Seifert, 2000; Seifert, 2001) and the time course of GTPS binding to H₂R-G_s fusion proteins (Fig. 4) we expected H₂R-G_sL to exhibit a higher constitutive activity than H₂R-G_sS. Surprisingly, however, when examining inverse agonists and partial agonists, we failed to uncover consistent differences in apparent constitutive activity between H₂R-G_sL and H₂R-G_sS. Particularly, in the GTPS binding assay, we did not find higher inverse agonistic efficacies of RAN at H₂R-G_sL than at H₂R-G_sS, regardless of whether extended time course studies (Fig. 4) or GTPS saturation binding studies (Fig. 5) were conducted. In contrast, inverse agonists show very large inhibitory effects on basal GTPS binding to ₂AR-G_sL, particularly at late time points of the binding reaction (Wenzel-Seifert and Seifert, 2000). In addition, inverse agonists at the ₂AR decrease the apparent GTPS affinity of G_sL (Wenzel-Seifert and Seifert, 2000), but no such observation was made for H₂R-G_sL (Fig. 5A). For ₂AR-G_sL and ₂AR-G_sS, we observed differences in efficacy of inverse agonists both in the GTPS binding assay and the GTPase assay (Seifert et al., 1998b; Wenzel-Seifert and Seifert, 2000), but for H₂R-G_s fusion proteins, we observed differences between H₂R-G_sL and H₂R-G_sS only for RAN and FAM in the GTPase assay (Figs. 4 and 5; Table 1). An explanation for the differences in the effects of RAN in the GTPS binding and GTPase assay could be that not only GDP dissociation but also other steps of the G protein cycle such as the actual GTP hydrolytic step are GPCR ligand-regulated. Specifically, the RAN-occupied H₂R may be similarly efficient at inhibiting GDP dissociation from fused G_sS and G_sL, but more efficient at reducing GTP hydrolysis at H₂R-G_sL than at H₂R-G_sS. Evidence that GPCRs regulate multiple steps of the G protein cycle, including nucleotide hydrolysis was already presented in previous studies (Brandt and Ross, 1986; Hilf et al., 1992; Seifert et al., 1999b, 2001).

When analyzing the effects of agonists in the GTPase assay, we found increased potencies of agonists at H₂R-G_sL relative to H₂R-G_sS for less than 50% of the compounds studied (Table 1). In contrast, with respect to ₂AR-G_s fusion proteins, the potencies of all agonists studied were higher at ₂AR-G_sL than at ₂AR-G_sS (Seifert et al., 1998b). Moreover, when only considering the agonists for which significant differences between H₂R-G_sL and H₂R-G_sS were observed, we noticed that for an agonist of a given efficacy, the differences in potency were larger for ₂AR-G_s fusion proteins than for H₂R-G_s fusion proteins. Specifically, the full agonist ()-isoproterenol is ~2.5-fold more potent at ₂AR-G_sL than at ₂AR-G_sS (Seifert et al., 1998b), whereas the full agonist HIS is only ~1.5-fold more potent at H₂R-G_sL than at H₂R-G_sS (Table 1). Similarly, the strong partial agonist salbutamol is ~2.6-fold more potent at ₂AR-G_sL than at ₂AR-G_sS, whereas the strong partial agonist IMP is only ~1.5-fold more potent at H₂R-G_sL than at H₂R-G_sS (Table 1). The correlation of the potencies of all agonists studied at H₂R-G_sS versus H₂R-G_sL revealed a linear correlation that was very close to the theoretical function that would have been obtained if the potencies of agonists at the two fusion proteins had been identical (Fig. 6A).

Similar observations as for agonist potencies at H₂R-G_s and ₂AR-G_s were made for agonist efficacies. Specifically, the efficacies of all partial agonists studied in the GTPase assay were significantly higher at ₂AR-G_sL than at ₂AR-G_sS, regardless of whether strong or weak partial agonists were considered (Seifert et al., 1998b). The increases in efficacy were particularly large for partial agonists with moderate efficacy, e.g., ()-ephedrine that has an efficacy of 0.31 at ₂AR-G_sS and an efficacy of 0.63 at ₂AR-G_sL. As a result, there is a strong hyperbolic relation between the efficacies of agonists at ₂AR-G_sS versus ₂AR-G_sL (Seifert et al., 1998b). In marked contrast, for H₂R-G_s fusion proteins, we observed significant increases in efficacy of ligands at H₂R-G_sL versus H₂R-G_sS for less than 50% of the agonists studied, and even those differences were small (Table 1). The most important finding with respect to efficacies was that guanidine D281 (13), which is in a similar efficacy range as ()-ephedrine at the ₂AR, exhibited identical efficacy at H₂R-G_sL and H₂R-G_sS. As the result of the small and inconsistent increases in efficacies of agonists at H₂R-G_sL relative to H₂R-G_sS, we obtained a highly significant linear correlation of the efficacies of agonists at the two fusion proteins that was only slightly shifted toward higher efficacies at H₂R-G_sL (Fig. 6B). This linear correlation is in marked contrast to the hyperbolic relation of agonist efficacies observed for ₂AR-G_s fusion proteins (Seifert et al., 1998b). Taken together, all these data show that the H₂R fused to G_sL and G_sS has similar apparent constitutive activity.

Our present study shows that caution must be exerted when considering the ₂AR as a prototypical GPCR with respect to the mechanisms regulating constitutive activity. Based on the data obtained with the ₂AR, one would have expected that a related biogenic amine GPCR, the H₂R, shows similar coupling to G_s proteins as the ₂AR. However, the apparent constitutive activity of the ₂AR is sensitive to differences in GDP affinity of G_s splice variants, whereas the apparent constitutive activity of the H₂R is insensitive to differences in GDP affinity. Thus, the GDP affinity of G proteins does not generally determine the apparent constitutive activity of GPCRs. Accordingly, intrinsic GPCR properties dominate the apparent constitutive activity of the H₂R but not the apparent constitutive activity of the ₂AR. Insensitivity of apparent constitutive activity of GPCRs to the GDP-affinity of G is not without precedence. Particularly, the various G_i isoforms differ from each other in GDP-affinity as well, but the apparent constitutive activity of the formyl peptide receptor coupled to the various G_i isoforms is very similar (Wenzel-Seifert et al., 1999). If the differential impact of the GDP affinity of G_s splice variants on the apparent constitutive activity of various GPCRs observed in fusion proteins can also be observed in coexpression systems, such data would point to GPCR-specific roles of G_s splice variants in signal transduction in vivo.

The molecular basis for the GPCR-specific interaction with G_sL splice variants is presumably due to differences in the second and third intracellular loops of the H₂R and ₂AR. These loops are crucial for G_s protein coupling (Dohlman et al., 1991; Kobilka, 1992) and are different in the H₂R and ₂AR (Wang et al., 2000). Further support for this hypothesis comes from the fact that chimeras of the ₂AR in which the second and intracellular loops were replaced by the corresponding loops of the H₂R differ considerably from the wild-type ₂AR in their efficacies at activating adenylyl cyclase (Wang et al., 2000). Overall, coupling of different GPCRs to G_s splice variants appears to be much more complex than was previously appreciated.