Similarities and differences in the coupling of human β1- and β2-adrenoceptors to Gsα splice variants

https://doi.org/10.1016/S0006-2952(02)00924-3Get rights and content

Abstract

The human β1-adrenoceptor (β1AR) and β2-adrenoceptor (β2AR) couple to Gs-proteins to activate adenylyl cyclase (AC). There are differences in desensitization between the β2AR and the originally cloned Gly389-β1AR, but with respect to ternary complex formation, constitutive activity, and AC activation the picture is unclear. To learn more about the similarities and differences between the β1AR and β2AR, we analyzed coupling of the Gly389-β1AR to the Gsα splice variants GsαL and GsαS using β1AR-Gsα fusion proteins expressed in Sf9 cells and compared the data with previously published data on β2AR-Gsα fusion proteins (Seifert et al., J Biol Chem 1998;273:5109–16). Fusion ensures defined receptor/G-protein stoichiometry and efficient coupling. The agonist (−)-isoproterenol stabilized the ternary complex at β1AR-GsαS, β1AR-GsαL, β2AR-GsαS, and β2AR-GsαL with similar efficiency. β1AR-GsαL but not β1AR-GsαS showed the hallmarks of constitutive activity as assessed by increased potencies and efficacies of partial agonists and AC activation by the agonist-free receptor. Similar differences were observed previously for β2AR-GsαS and β2AR-GsαL. β1AR-GsαS and β2AR-GsαS were similarly efficient at activating AC, but β1AR-GsαL was ∼4-fold more efficient at activating AC than β2AR-GsαL. Our data show that (i) the β1AR and β2AR are similarly efficient at stabilizing the ternary complex with Gsα splice variants, (ii) GsαL confers constitutive activity to the β1AR and β2AR, and (iii) the β1AR coupled to GsαL is more efficient at activating AC than the β2AR coupled to GsαL. These data help us understand some of the discrepancies regarding similarities and differences between the β1AR and β2AR.

Introduction

Many hormones and neurotransmitters exert their effects via GPCRs [1], [2], [3], [4]. Upon binding of an agonist, GPCRs isomerize from an inactive to an active state, enabling GPCRs to promote GDP dissociation from G-proteins. Not only agonist-occupied but also agonist-free GPCRs can activate G-proteins. Inverse agonists reduce this agonist-independent (constitutive) activity. Agonist-occupied GPCRs form a ternary complex with nucleotide-free G-protein. The ternary complex is characterized by high agonist affinity. GPCRs then promote binding of GTP to Gα, resulting in ternary complex disruption and dissociation of G-protein into Gα-GTP and the βγ-complex. Both Gα-GTP and βγ regulate the activity of effectors. The GTPase (EC 3.6.1.) of Gα accomplishes G-protein deactivation by hydrolyzing GTP. Gα-GDP and βγ re-associate, completing the G-protein cycle.

The β1AR and β2AR are prototypical GPCRs that couple to Gs-proteins to activate AC (EC 4.6.1.1) [5]. The human β1AR exists as two polymorphisms. The originally cloned Gly389-β1AR has an allele frequency of 26% and is less efficient than the Arg389-β1AR (allele frequency 74%) at stabilizing the ternary complex and activating AC [6]. Since the Arg389-β1AR has been identified only recently, most, if not all, published studies aimed at comparing the β1AR and β2AR have been conducted with Gly389-β1AR. For this reason, we analyzed Gly389-β1AR in our present study, although future studies will need to directly compare Arg389-β1AR with β2AR, too.

There is agreement in the literature that the β1AR is less sensitive to desensitization and internalization than the β2AR [7], [8], [9], [10]. However, with respect to other parameters, the picture is unclear. Freissmuth and coworkers [11] observed a higher percentage of high-affinity binding sites with the β2AR reconstituted with Gs than with the β1AR reconstituted with Gs using an Escherichia coli expression system. In contrast, Green and coworkers [12] found a higher percentage of high-affinity binding sites with the β1AR than with the β2AR using CHW cells as the expression system. In the former study, the Kh values for ISO were similar at the β1AR and β2AR, whereas in the latter study the Kh value for the β1AR was ∼5-fold higher than the Kh value for the β2AR. In other studies, it was difficult to discern distinct high-affinity binding sites with the β1AR [6], [13]. When the constitutive activity of the β1AR and β2AR is considered, the situation is unclear, too. The β1AR and β2AR expressed in HEK293 cells exhibit similarly low extents of constitutive activity [14]. In contrast, the β2AR expressed in cardiac myocytes from β1AR/β2AR double knockout mice exhibits significant constitutive activity, whereas the β1AR is devoid of any constitutive activity in this expression system [15]. However, when expressed in COS-7 cells, the β1AR clearly exhibits constitutive activity, although to a lesser extent than the β2AR [16]. Finally, the data concerning AC activation are controversial. In some studies, the agonist-occupied β2AR was found to be more efficient at activating AC than the β1AR [9], [17], whereas in other studies the agonist-occupied β1AR and β2AR were reported to be similarly efficient at activating AC [7], [8], [13], [14].

The G-protein Gs exists as two splice variants, GsαL and GsαS, respectively. GsαL possesses a lower GDP affinity than GsαS[18], [19]. As a result of these biochemical differences between Gsα splice variants, the agonist-free β2AR and the β2AR bound to partial agonists are more efficient at promoting GDP/GTP exchange at GsαL than at GsαS, i.e. GsαL confers the hallmarks of constitutive activity to the β2AR [19], [20]. In addition, the maximum AC activity induced by the GsαL-coupled β2AR is lower than the AC activity induced by the GsαS-coupled β2AR.

Based on these data and the fact that the expression of GsαS and GsαL varies substantially in different tissues [21], we asked the question whether the analysis of β1AR-coupling to Gsα splice variants would help us to understand some of the discrepancies in the literature regarding differences/similarities between the β1AR and β2AR. To address this question, we studied β1AR-Gsα fusion proteins expressed in Sf9 insect cells and compared their properties with the properties of the previously published β2AR-Gsα fusion proteins [19], [20], [22]. In GPCR-Gα fusion proteins, the GPCR C-terminus is linked to the Gα N-terminus [23], [24], [25]. The fusion guarantees a defined 1:1 stoichiometry of the signaling partners, ensures close proximity of the partners, promotes efficient coupling, and allows for the analysis of the coupling of a given GPCR to various G-proteins under defined experimental conditions [19], [22]. The expression level of fusion proteins and, thereby, the expression of a specific Gα, can be precisely determined by GPCR antagonist saturation binding. This information can be used to determine the steady-state GTP turnover of the fusion protein [24], [25], the functional integrity of fusion proteins by assessing ligand-regulated GTPγS saturation binding [22], [26], [27], and the specific efficacies of a given GPCR-Gsα fusion protein at activating AC [27]. Additionally, ternary complex formation, GTP turnover, as well as the efficacies and potencies of partial agonists and inverse agonists in the GTPase and GTPγS binding assays are independent of the expression level of GPCR-Gα fusion proteins [19], [22], [27], [28]. These properties together with the low background signaling of exogenously expressed GPCRs to endogenous insect Gs-like G-proteins and the very stable AC activity in Sf9 cells render these cells a very sensitive system for the analysis of βARs [19], [28].

Section snippets

Materials

The cDNA for the human β1AR (Gly389 polymorphism) in pUC18 was provided by Dr. M. Bouvier (Department of Biochemistry, University of Montreal). [35S]GTPγS (1100 Ci/mmol), [γ-32P]GTP (6000 Ci/mmol), and [α-32P]ATP (3000 Ci/mmol) were from Perkin Elmer. [3H]DHA (85–90 Ci/mmol) was from Amersham Pharmacia Biotech. Unlabeled GTP, GTPγS, GDP, and ATP [high quality, catalogue No. 519 979; <0.01% (w/w) GTP contamination as assessed by HPLC analysis] were obtained from Roche Diagnostics. ICI 118,551 (ICI)

Analysis of the expression of β1AR-GsαL and β1AR-GsαS in Sf9 membranes by antagonist saturation binding, immunoblotting, and GTPγS saturation binding

Membranes from Sf9 cells infected with high-titer baculovirus stocks encoding β1AR-GsαL and β1AR-GsαS were prepared, and the expression levels of constructs were determined by [3H]DHA saturation binding. With a 1:100 dilution of virus stocks, the Bmax value (expression levels) of β1AR-Gsα fusion proteins was 4.5±1.7 pmol/mg (mean±SD, N=6). With a 1:1000 dilution of virus stocks, the expression level of β1AR-Gsα fusion proteins was 1.1±0.4 pmol/mg (mean±SD, N=4). β1AR-GsαL bound [3H]DHA with a Kd

Acknowledgements

We should like to thank the reviewers of this paper for their thoughtful comments. This work was supported by a grant of the Army Research Office (DAAD 19-00-1-0069), the J. R. & Inez Jay Biomedical Research Award of The University of Kansas, and a New Faculty Award of The University of Kansas to R.S.

References (42)

  • K Wenzel-Seifert et al.

    Quantitative analysis of formyl peptide receptor coupling to Giα1, Giα2, and Giα3

    J. Biol. Chem.

    (1999)
  • S.R.J Hoare

    G-protein-coupled receptors: what limits high-affinity agonist binding?

    Trends Pharmacol. Sci.

    (2000)
  • R Seifert et al.

    G-protein-coupled receptors: what limits high-affinity agonist binding? Reply

    Trends Pharmacol. Sci.

    (2000)
  • D.F Matesic et al.

    Pharmacological and biochemical characterization of complexes of muscarinic acetylcholine receptor and guanine nucleotide-binding protein

    J. Biol. Chem.

    (1989)
  • R.V Rebois et al.

    Does subunit dissociation necessarily accompany the activation of all heterotrimeric G proteins?

    Cell Signal.

    (1997)
  • P Chidiac

    Rethinking receptor-G protein-effector interactions

    Biochem. Pharmacol.

    (1998)
  • A.G Gilman

    G proteins: transducers of receptor-generated signals

    Annu. Rev. Biochem.

    (1987)
  • T Gudermann et al.

    Functional and structural complexity of signal transduction via G-protein-coupled receptors

    Annu. Rev. Neurosci.

    (1997)
  • B.K Kobilka

    Adrenergic receptors as models for G protein-coupled receptors

    Annu. Rev. Neurosci.

    (1992)
  • T Suzuki et al.

    Distinct regulation of β1- and β2-adrenergic receptors in Chinese hamster fibroblasts

    Mol. Pharmacol.

    (1992)
  • G Rousseau et al.

    Distinct receptor domains determine subtype-specific coupling and desensitization phenotypes for human β1- and β2-adrenergic receptors

    Mol. Pharmacol.

    (1996)
  • Cited by (0)

    1

    Present address: Neurodegeneration Group, The John P. Robarts Research Institute, London, Ont., Canada.

    View full text