Airway smooth muscle prostaglandin-EP1 receptors directly modulate β2–adrenergic receptors within a unique heterodimeric complex

Multiple and paradoxical effects of airway smooth muscle (ASM) 7-transmembrane–spanning receptors activated during asthma, or by treatment with bronchodilators such as β₂–adrenergic receptor (β₂AR) agonists, indicate extensive receptor crosstalk. We examined the signaling of the prostanoid-EP₁ receptor, since its endogenous agonist prostaglandin E₂ is abundant in the airway, but its functional implications are poorly defined. Activation of EP₁ failed to elicit ASM contraction in mouse trachea via this G_αq-coupled receptor. However, EP₁ activation markedly reduced the bronchodilatory function of β₂AR agonist, but not forskolin, indicating an early pathway interaction. Activation of EP₁ reduced β₂AR-stimulated cAMP in ASM but did not promote or augment β₂AR phosphorylation or alter β₂AR trafficking. Bioluminescence resonant energy transfer showed EP₁ and β₂AR formed heterodimers, which were further modified by EP₁ agonist. In cell membrane [³⁵S]GTPγS binding studies, the presence of the EP₁ component of the dimer uncoupled β₂AR from G_αs, an effect accentuated by EP₁ agonist activation. Thus alone, EP₁ does not appear to have a significant direct effect on airway tone but acts as a modulator of the β₂AR, altering G_αs coupling via steric interactions imposed by the EP₁:β₂AR heterodimeric signaling complex and ultimately affecting β₂AR-mediated bronchial relaxation. This mechanism may contribute to β-agonist resistance found in asthma.

The 7-transmembrane–spanning (7-TM–spanning) receptors represent the largest signaling family in the genome. We estimate that the lung expresses 25–50 7-TM receptors in airway epithelial cells, airway smooth muscle (ASM), pulmonary vasculature, alveolar walls, and resident immune cells (1). In regard to asthma, several 7-TM receptors play established roles in bronchoconstriction (e.g., M₃-muscarinic receptor) and bronchodilation (e.g., β₂–adrenergic receptor [β₂AR]). Despite identification of the endogenous ligands and receptor localization, there are a number of 7-TM receptors expressed in the airway whose functions are unknown, or appear to function paradoxically, based on recombinantly expressed receptors in model cell systems. This lack of understanding of receptor function has impeded our ability to ascertain the role of these ligands (some of which are markedly increased in asthma) in the relaxation/contraction of ASM; thus the mechanistic basis of bronchial hyperreactivity and bronchoconstriction in asthma remains only partially understood. In many cases the basis for incomplete mechanistic information can be attributed to the nature of recombinant expression systems, which may not take into account multiple receptor subtypes, receptor/receptor interactions, crosstalk via second messengers, or subcellular localization of receptors and specific effectors.

These issues are particularly obvious when one considers the biologic actions of prostaglandin E₂ (PGE₂), which, via at least 4 receptor subtypes (EP₁–EP₄), can evoke multiple signals. By activation of the G_s-coupled EP₂ receptor, PGE₂ can evoke ASM relaxation (2). However, blockade of PGE₂ production leads to partial reversal of β₂AR desensitization in the context of inflammation (3), and thus PGE₂ may promote bronchoconstriction via desensitization of β₂AR function or by activation of the G_q-coupled EP₁ receptor. PGE₂ is produced by ASM, airway epithelial cells, and inflammatory cells and is the most abundant prostanoid in epithelial lining fluid (4–6). However, EP₁-mediated signal transduction and its effect on airway responsiveness remains unclear. We show here a unique function of the EP₁ receptor in ASM, which to our knowledge has not been previously described for 7-TM receptors. The EP₁ receptor itself, despite its coupling to inositol 1,4,5-triphosphate (IP₃) and intracellular calcium ([Ca²⁺]_i), failed to exert the expected bronchoconstrictive response. Instead, EP₁ receptor activation modulated β₂AR function via formation of a heterodimeric complex that directly decreased β₂AR coupling to G_αs in an EP₁ receptor, agonist-dependent, fashion.

EP1 receptor activation in isolated and intact ASM. The biologic actions of PGE₂ are potentially mediated by multiple receptor subtypes that differentially couple to G_q, G_i, and G_s. Although a relaxant action of PGE₂, mediated via the EP₂ isoform in ASM, has been documented (7), the extent to which PGE₂ couples to the other subtypes in ASM, particularly the potentially probronchoconstrictive G_q-coupled EP₁ receptor, is not established. EP₁-specific RT-PCR was carried out on murine ASM cells in culture (Figure 1A). A single product of the expected molecular size was obtained, and Southern blots with an EP₁-specific oligonucleotide hybridized to the product. Sequencing further confirmed the product as the murine EP₁ receptor (GenBank accession no. NM_013641). With regard to signal transduction, the EP₁-specific agonist 17-phenyl trinor-PGE₂ (17-PTP) was found to stimulate inositol phosphate production (Figure 1B) and [Ca²⁺]_i transients (Figure 1C). Of note, maximal activation of the EP₁ receptor by 17-PTP stimulated [Ca²⁺]_i to a greater extent, and with greater potency, than the M₃-muscarinic agonist carbachol (Figure 1D). These findings confirmed EP₁ receptor expression in ASM cells and that its activation by agonists resulted in IP₃ and [Ca²⁺]_i turnover as assessed in whole cells. Because these effectors mediate 7-TM receptor contraction of ASM, the data suggested that the EP₁ receptor would likely promote airway contraction.

Figure 1

EP₁ receptor expression and signal transduction in murine ASM cells. (A) EP₁ gene expression was detected in cultured murine ASM cells by RT-PCR. The 501-bp product (left) corresponding to sequence from the EP₁ receptor open reading frame was only detected in reactions that included reverse transcriptase (+RT) and hybridized by Southern blots to an EP₁-specific probe (right). (B) The EP₁-selective agonist 17-PTP (1 μM) stimulated inositol phosphate (IP) production in murine ASM ~70% above basal levels (n = 4). Thrombin-stimulated inositol phosphate production was used for a positive control. *P < 0.05 versus basal. (C) Both 17-PTP and acetylcholine-stimulated [Ca²⁺]_i transients in ASM cells. Shown is a representative experiment using the doses shown in D. (D) Maximal agonist-stimulated [Ca²⁺]_i transients in ASM cells exposed to the indicated doses of agonists. Data are mean ± SEM of 5 experiments. Cch, carbachol; RFU, relative fluorescence units.

Excised mouse tracheal ring preparations were utilized to ascertain the effects of EP₁ activation on active force generation in the intact airway. However, despite its ability to stimulate IP₃ and [Ca²⁺]_i turnover in ASM, 17-PTP caused no detectable contraction of tracheal rings even at concentrations up to 10 μM (Figure 2A). Since 17-PTP did not directly effect ASM tone, we next investigated whether EP₁ receptor activation could modify the response to another contractile agonist acting through G_αq. For these studies, tracheal rings were pretreated with 1 μM 17-PTP for 15 minutes prior to contraction with various concentrations of acetylcholine (Figure 2A). The maximal acetylcholine response in the absence (26 ± 5.3 mN) or presence (26 ± 4.8 mN) of EP₁ coactivation was not different, nor was the EC₅₀ for the response (91 ± 11 versus 82 ± 15 μM, respectively). We next considered whether EP₁ receptor activation alters ASM relaxation. The efficacy and potency of the βAR agonist isoproterenol to relax acetylcholine-constricted ASM in the presence and absence of EP₁ receptor activation was ascertained. Tracheal rings were first treated with vehicle or 1 μM 17-PTP for 15 minutes, contracted continuously with 10 μM acetylcholine, and exposed to increasing concentrations of isoproterenol. In the presence of vehicle, isoproterenol caused a dose-dependent relaxation with an EC₅₀ of 23.4 ± 4.8 nM and a maximal reduction in force of 36.6% ± 2.9% (Figure 2B). However, following pretreatment with 17-PTP, a marked reduction in βAR-mediated relaxation was observed, with isoproterenol evoking only 17.1% ± 2.4% relaxation (n = 6; P < 0.001; Figure 2B). Of note, the EC₅₀ for isoproterenol was also increased under these conditions to 58.2 ± 6.1 nM (P < 0.001). As shown in Figure 2C, this same βAR desensitization effect of 17-PTP was observed when tracheal rings were contracted by 60 mM KCl. The endogenous EP₁ receptor agonist is PGE₂, which activates other EP receptors and thus under physiologic conditions could evoke an effect that dominates airway responses over the EP₁-specific phenotype we observed. We therefore studied the effect of pretreatment with 10 nM PGE₂ (which evokes only ~15% relaxation) using the same protocol as for 17-PTP. As shown in Figure 2D, PGE₂ also evoked a loss of isoproterenol-mediated relaxation, amounting to an approximate 25% desensitization. This lower amount of desensitization (compared with 17-PTP pretreatment) may be due to the combined relaxant effect of PGE₂ and isoproterenol.

Figure 2

Effect of EP₁ receptor activation on ASM contraction and relaxation. (A) Concentrations of 17-PTP up to 10 μM caused no significant contraction of murine tracheal rings and did not enhance acetylcholine-mediated (Ach) contraction. Data are mean ± SEM of 4 experiments. (B) Pretreatment of tracheal rings with 1 μM 17-PTP for 15 minutes reduced both the potency (EC₅₀ increased) and efficacy of isoproterenol-mediated (Iso) relaxation of tracheal rings contracted with 10 μM acetylcholine. Data are mean ± SEM of 6 experiments. ^†P = 0.001 versus vehicle-treated rings. (C) The inhibitory effect of 17-PTP on isoproterenol-mediated relaxation was also present when the tracheal rings were contracted with 60 mM KCl. Data are mean ± SEM of 8 experiments. *P < 0.05 versus vehicle-treated rings. (D) Pretreatment of tracheal rings with 10 nM PGE₂ evokes desensitization of isoproterenol-mediated relaxation. Data are mean ± SEM of 5 experiments. ^#P < 0.02 versus vehicle-treated rings.

Given that mouse trachea has a mixed population of β₁AR and β₂AR subtypes (8), the 17-PTP effect was also assessed using the relatively β₂AR subtype-specific agonist formoterol. Under these conditions, 17-PTP evoked a 55% ± 5% desensitization of formoterol-mediated relaxation, indicating a 17-PTP effect on β₂AR (the predominant βAR subtype of human tracheal ASM). To determine whether the inhibitory effect of 17-PTP extended to other ASM relaxants, we performed analogous experiments with forskolin, which directly activates adenylyl cyclase, mimicking a G_s-coupled receptor. However, unlike the response with isoproterenol, pretreatment with 17-PTP had no effect on the maximal response or the EC₅₀ (Figure 3A). Given that the EP₁ receptor couples to G_αq, we considered that second messengers generated from phospholipase C (PLC) activation may be involved in the apparent crosstalk between β₂AR and EP₁ receptors. However, pretreatment with the PLC inhibitors U73122 (5 μM; Figure 3B) and Et-18-OCH₃ (10 μM; data not shown) failed to alter the 17-PTP effect. β₂ARs have also been shown in some cells to couple to G_αi (9), which may act to oppose G_αs-mediated signaling. Treatment of tracheal rings with pertussis toxin, however, had no effect on the EP₁-mediated β₂AR desensitization (Figure 3C), indicating that enhanced G_i coupling was an unlikely mechanism. Finally, the β₂AR desensitization phenotype evoked by EP₁ activation was not altered by pretreatment with 100 μM of the NO inhibitor l-NAME (51.2% ± 6.8% desensitization) or 10 μM of the cyclooxygenase inhibitor indomethacin (57% ± 5% desensitization).

Figure 3

Characterization of EP₁ receptor–promoted desensitization of β₂ AR-mediated ASM relaxation. (A) No effect of 17-PTP pretreatment was observed on forskolin-mediated ASM relaxation, indicating that the site of EP₁ modulation is likely proximal to adenylyl cyclase (n = 4). (B) Inhibition of phospholipase C (PLC) with U73122 failed to alter the 17-PTP–evoked desensitization of isoproterenol-mediated ASM relaxation. *P < 0.05 versus vehicle-treated rings (n = 4). (C) Ablation of receptor-G_αi coupling with pertussis toxin (PTX) had no effect on the 17-PTP–evoked desensitization of isoproterenol-mediated ASM relaxation. ^‡P < 0.01 versus rings not treated with 17-PTP (n = 4).

Desensitization of β₂ AR by EP₁ receptor activation. The inhibitory effect of 17-PTP on isoproterenol-mediated relaxation observed in the tracheal ring studies indicated functional interplay between the EP₁ and β₂AR signal transduction pathways. Additionally, the absence of a significant EP₁ effect on forskolin-mediated relaxation and no change in phenotype observed with PLC inhibition suggested that this interaction could potentially occur at the level of the β₂AR itself. The effect of 17-PTP on isoproterenol-stimulated cAMP production was determined in isolated ASM cells. In the absence of 17-PTP, 1 μM isoproterenol stimulated cAMP to 88.0% ± 16.3% of the maximum level obtained with 10 μM forskolin. However, when cells were pretreated with 1 μM 17-PTP for 15 minutes prior to isoproterenol stimulation, cAMP production was reduced to 37.0% ± 3.9% (n = 4; P = 0.02) of the forskolin response (Figure 4A). Taken together with the tracheal ring data, these results indicated an interplay between the EP₁ receptor and β₂AR, with activation of the former resulting in decreased function of the latter. Two processes that we considered potential mechanisms to account for uncoupling of β₂AR signal transduction within the time frame of minutes included (a) enhanced loss of cell surface receptors due to internalization and (b) enhanced isoproterenol-stimulated β₂AR phosphorylation due to EP₁ activation or phosphorylation of the β₂AR by EP₁ activation via PKC. To ascertain whether EP₁ activation altered the extent or rate of agonist-mediated β₂AR internalization, we performed radioligand binding on intact ASM cells with the hydrophilic radioligand [³H]-CGP-12177. Since this ligand does not enter intact cells (10), a decrease in binding occurs when surface receptors are internalized. Thus, as expected for agonist-promoted internalization of the β₂AR, exposure of ASM cells to 1 μM isoproterenol caused a time-dependent decrease in [³H]-CGP-12177 binding sites of 39.1% ± 5.7% after 30 minutes (Figure 4B). Pretreatment of cells with 1 μM 17-PTP for 15 minutes had no effect on isoproterenol-stimulated internalization (37.4% ± 3.6%; n = 5). We next examined the potential for 17-PTP to promote phosphorylation of the β₂AR. For the phosphorylation experiments, human embryonic kidney (HEK) cells transiently cotransfected with plasmids so as to express the EP₁ receptor and FLAG-tagged β₂AR were loaded with ³²P-orthophosphate. Cells were treated with vehicle, isoproterenol, 17-PTP, phorbol-12-myristate-13-acetate (a positive control for PKC-mediated event), or isoproterenol plus 17-PTP. β₂AR was purified by FLAG immunoprecipitation. As shown in Figure 4C, isoproterenol evoked robust phosphorylation of the β₂AR, while no detectable phosphorylation over background was noted with 17-PTP exposure alone. Importantly, the extent of isoproterenol–promoted β₂AR phosphorylation was not altered by 17-PTP (48 ± 4.5 relative units versus 42 ± 4.2 relative units in the presence and absence of 17-PTP, respectively; n = 5; P = NS).

Figure 4

Effect of EP₁ receptor activation on β₂ AR agonist–stimulated cAMP production, receptor sequestration and receptor phosphorylation. (A) Pretreatment of ASM cells with 17-PTP (1 μM) for 15 minutes reduced isoproterenol-stimulated cAMP production. Data are mean ± SEM of 4 experiments. ^##P = 0.02 versus vehicle-treated cells. (B) Isoproterenol caused a time-dependent internalization of β₂AR in murine ASM cells over 30 minutes as assessed by whole-cell binding using the hydrophilic antagonist [³H]-CGP-12177; pretreatment with 17-PTP (1 μM for 15 min) had no effect on internalization. Data are mean ± SEM of 4 experiments. (C) EP₁ receptor activation failed to phosphorylate β₂AR or to alter isoproterenol-mediated β₂AR phosphorylation in HEK 293 cells transfected to express β₂AR and EP₁ receptors. Shown is a representative autoradiogram of 4 experiments. PMA, phorbol-12-myristate-13-acetate. NTF, nontransfected.

β₂ ARs and EP₁ receptors form heterodimers. Since these well-defined short-term mechanisms of receptor regulation and crosstalk were not at play during EP₁-induced desensitization of β₂AR, we considered an alternative hypothesis, which was that the 2 receptors formed heterodimers. We also considered whether agonist activation of the EP₁ receptor directly affected the receptor/receptor interaction within the complex, altering β₂AR coupling to G_αs. As a screen for such a complex in ASM cells, fluorescence microscopy was carried out using β₂AR- and EP₁-specific primary antibodies. Secondary antibodies provided for red (β₂AR) and green (EP₁) signals. Merged images revealed yellow signals consistent with areas of colocalization of these 2 receptors (Figure 5A). To rigorously assess the potential for heterodimer formation, bioluminescence resonant energy transfer (BRET) and coimmunoprecipitation studies were undertaken in transfected HEK 293 cells. BRET was carried out in living cells expressing 2 receptors having carboxyterminally fused Renilla luciferase (Rluc) or yellow fluorescent protein (YFP). As described in detail elsewhere (11–13), this approach takes advantage of a signal produced by the energy donor moiety (Rluc) acting on the energy acceptor (YFP) when the two are in very close proximity. As shown in Figure 5B, studies carried out in HEK 293 cells transfected to express β₂AR-Rluc and EP₁-YFP (see Methods) indeed revealed a BRET signal, as indicated by the perturbation of the emission spectrum at 530 nM. To verify that this represented oligomer formation, rather than a “pseudo-BRET” signal (11) due to random collisions of overexpressed receptors, studies were carried out with multiple levels of expression (altering the β₂AR-Rluc/EP₁-YFP ratio). As expected under conditions of dimer formation, the BRET signal saturated, with a calculated maximal ratio (BRET_max) of 0.49 ± 0.04 and a BRET₅₀ of 2.97 ± 0.56 (Figure 5C). In contrast, in experiments utilizing cotransfections of β₂AR-Rluc and a YFP fusion with the γ-aminobutyric acid receptor subtype bR2 (GABAb R2–YFP) a small signal was observed, as was the case with cotransfections of C-C chemokine receptor 5–Rluc (CCR5-Rluc) and β₂AR-YFP (Figure 5D). These low-level signals and a lack of saturation (data not shown) were consistent with prior observations that neither the GABAb R2 nor CCR5 receptors form heterodimers with the β₂AR (13, 14). We next examined whether activation of the EP₁ receptor or the β₂AR altered the BRET signal. As shown in Figure 5E, treatment of cotransfected cells with 17-PTP or isoproterenol for 15 minutes prior to the addition of coelenterazine h increased the BRET signal in both cases. Given that most 7-TM dimers appear to be formed intracellularly and then expressed at the cell surface (11, 12, 15, 16), and given the short-term agonist exposure used in these experiments, we do not interpret these results as representing an increased number of dimers per se. Rather, the data are consistent with a conformational change that occurs (potentially in either receptor) when one receptor is bound by its agonist, which causes the Rluc and YFP moieties to move closer together. The magnitude of the agonist-promoted increase in BRET is not trivial, in that the BRET signal is proportional to the distance between the donor and acceptor by a factor of 10⁶ (11, 12, 15).

Figure 5

The EP₁ receptor and β₂ AR form a heterodimer as assessed by BRET and coimmunoprecipitation. (A) Immunofluorescence microscopy revealed potential colocalization of EP₁ and β₂AR in ASM cells. Magnification, ×200. (B) The emission spectra of HEK 293 cells transfected with the β₂AR-Rluc fusion alone showed a single peak with a maximum at ~475 nm following addition of coelenterazine h (open symbols). However, a second peak was observed at ~530 nm when cells were cotransfected with the EP₁-YFP fusion construct (closed symbols). Shown is a representative experiment of 4. (C) The BRET ratios derived from HEK 293 cells transfected with a fixed amount of the donor β₂AR-Rluc reached saturation with increasing expression of the acceptor EP₁-YFP. Data shown are compiled from 12 experiments. (D) GABAb R2 and C-C chemokine receptor 5 (CCR5) failed to form heterodimers with β₂AR as indicated by low pseudo-BRET signals. (E) The BRET ratios of HEK 293 cells cotransfected with the β₂AR-Rluc and EP₁-YFP were increased by treatment with either the EP₁ agonist 17-PTP or the βAR agonist isoproterenol. *P < 0.05 versus untreated; n = 4. (F) Coimmunoprecipitation of FLAG-EP₁:β₂AR receptor complexes. HEK 293 cells were transfected with FLAG-β₂AR, EP₁ receptor, or both; membranes were solubilized; and proteins were immunoprecipitated with anti-FLAG antibody. Western blots of whole-cell lysates and immunoprecipitates were carried out with polyclonal EP₁ receptor antisera. Shown is a representative blot from 5 experiments.

To further corroborate dimer formation, coimmunoprecipitation studies were carried out in HEK 293 cells transfected to express FLAG-β₂AR and the EP₁ receptor both alone and together. Immunoprecipitation was carried out with anti-FLAG antibody, and the precipitated proteins were fractionated on SDS-PAGE followed by Western blots using a polyclonal EP₁ antisera. As shown in Figure 5F, the EP₁ receptor migrated at approximately 45 kDa in Western blots of cell lysates using the EP₁ antisera. Upon immunoprecipitation with FLAG antibody, protein from cells transfected to express EP₁ or β₂AR showed no immunoreactivity with the EP₁ antisera. However, when the 2 receptors were cotransfected, a band of the appropriate molecular size was observed. Of note, the high–molecular weight bands, which could represent dimers, are often not resistant to SDS and are not typically visualized (14). These coimmunoprecipitation studies are thus consistent with the BRET results, indicating the presence of the EP₁:β₂AR heterodimer.

EP1:β2AR heterodimer formation uncouples β2AR from Gαs. Taken together, the above results indicated that the EP₁ receptor and the β₂AR exist as an oligomeric complex, with activation of the EP₁ receptor markedly desensitizing β₂AR signaling and physiologic function, and that such desensitization is not due to downstream second messenger feedback. This suggested that agonist activation of the EP₁ receptor within the dimer results in a steric effect on the β₂AR, altering its conformation in such a way as to directly affect β₂AR-G_αs interaction. To examine this, we studied agonist-promoted [³⁵S]GTPγS binding in cell membranes. This provided for direct quantitation of receptor–G protein interaction by measuring GTP turnover and, since it is performed with isolated membranes, was not affected by intracellular factors. In these studies, [³⁵S]GTPγS binding was carried out by incubating membranes from cells transfected to express G_αs, β₂AR, and the EP₁ receptor with vehicle, isoproterenol alone, or isoproterenol with 17-PTP. In other studies, G_αs and β₂AR were expressed without the EP₁ receptor so as to ascertain β₂AR coupling to G_αs in the absence of the EP₁:β₂AR dimer. Expression levels of β₂AR in the β₂AR and EP₁ (3774 ± 815 fmol/mg) and β₂AR transfections (3367 ± 185 fmol/mg) were comparable. Net isoproterenol-stimulated [³⁵S]GTPγS binding is shown in Figure 6A. Incubation with 17-PTP resulted in a significant decrease in isoproterenol-stimulated binding, amounting to an approximately 60% reduction (n = 4; P = 0.02). Interestingly, the presence of the EP₁ receptor, even in the absence of its activation by agonist, decreased β₂AR coupling to G_αs. As shown in Figure 6A, cells expressing β₂AR alone had the highest isoproterenol-stimulated [³⁵S]GTPγS binding, and as expected, 17-PTP had no effect in these cells. Additional studies were also carried out using ASM cell membranes (Figure 6B). Analogous to what was observed in the transfected cell studies, [³⁵S]GTPγS turnover was decreased when membranes were incubated with isoproterenol and 17-PTP compared with isoproterenol alone. These data then confirmed the concept of a direct interaction between the 2 receptors in the cell type of interest when the receptors were expressed at physiologic levels.

Figure 6

EP₁ receptor activation uncouples the β₂ AR from G_s . β₂AR coupling to G_s was assessed by measuring isoproterenol-stimulated [³⁵S]GTPγS binding in membranes from COS-7 cells transfected with G_αs and the β₂AR, or with G_αs and β₂AR plus EP₁ (A) or mouse primary ASM cells (B). Reactions were performed with partially purified cell membranes. The [³⁵S]GTPγS bound to G_αs was recovered by immunoprecipitation with G_αs antibody as described in Methods. (A) The presence of 17-PTP in the reaction significantly reduced [³⁵S]GTPγS binding stimulated by isoproterenol in the transfected COS-7 cells, while the β₂AR expressed without the EP₁ had the greatest stimulation by isoproterenol. No effect of 17-PTP was observed on cells transfected to only express β₂AR. **P = 0.05 versus β₂AR only; ^##P = 0.02 versus 17-PTP–untreated cells coexpressing β₂AR and EP₁ receptor. Shown are mean ± SEM from 6 experiments. (B) The presence of 17-PTP in the reaction decreased isoproterenol-stimulated [³⁵S]GTPγS binding in ASM cell membranes. *P < 0.05 versus isoproterenol in the absence of 17-PTP. Shown are mean ± SEM from 6 experiments. In both types of experiments, basal (non-agonist) binding was ~65% of isoproterenol-stimulated binding.

PGE₂ is a cyclooxygenase-derived product of arachidonic acid metabolism that plays a major role in regulating such diverse physiological processes as smooth muscle tone, inflammation, and pain (reviewed in ref. 17). The biological effects of PGE₂ are often multifaceted and can appear contradictory to traditional signaling paradigms. As introduced earlier, some of these properties of PGE₂ are due to the fact that it is an endogenous agonist for 4 receptor subtypes coupled to several signal transduction cascades. The EP₁ receptor is linked to IP₃ and [Ca²⁺]_i turnover through G_q-mediated signal transduction (18), whereas the EP₂ and EP₄ receptors stimulate adenylyl cyclase and cAMP production through G_s (19–21). Splice variants of the EP₃ receptor give rise to multiple isoforms that apparently can couple to G_i, G_s, or G_q (22, 23). Since one or more of these receptors may be expressed by a cell or tissue, the integrated response to PGE₂ often represents the net effect of additive or opposing signal transduction events. Moreover, temporal and spatial changes in patterns of EP receptor expression in response to various stimuli (e.g., injury and inflammation) provide an additional level of dynamic regulation and complexity to PGE₂-regulated processes in the airways. For example, recent studies of the PGE₂-mediated fibroproliferative response due to intratracheal bleomycin have shown an approximately 50% decrease in fibroblast EP₂ receptor mRNA without significant change in the other EP subtypes (24), whereas FITC-mediated fibrosis was associated with decreases in both EP₂ and EP₄ transcript expression (24).

Because PGE₂ is produced endogenously within the lung and activates pathways that may both enhance and inhibit ASM contraction, it may have particular relevance in asthma. PGE₂ is produced by ASM cells, bronchial and alveolar epithelial cells, fibroblasts, and inflammatory cells in the lung, and increased levels have been reported in some studies with human asthma (25, 26). Furthermore, an initial enzyme for prostanoid receptor synthesis (COX-2), as well as PGE₂ precursors, are increased in asthmatic airways (27, 28). Following its discovery, PGE₂ was found to relax ASM and thus has generally been considered to be a bronchoprotective prostanoid. The bronchodilatory effect of PGE₂ appears to be due to activation of the EP₂ receptor subtype, since PGE₂-mediated bronchodilation is present in EP₁-, EP₃-, and EP₄-null mice but absent in EP₂-null mice (29). Interestingly, in nonasthmatic humans, PGE₂ exerts a net bronchodilation; in asthmatics, though, the response is highly variable, sometimes resulting in significant bronchoconstriction (30). The complex nature of PGE₂ action in the asthmatic milieu is further illustrated in that it may serve as an intermediary through which inflammatory cytokines can modulate other receptor systems that govern airway tone. In particular, PGE₂ appears to mediate the inhibitory effects of the inflammatory cytokines TGFβ and interleukin-1 on β₂AR signal transduction in cultured ASM (31–34). These inflammatory cytokines stimulate COX-2 and production of PGE₂ by ASM, and it has been proposed that the resulting increase in cAMP production following EP₂ receptor activation leads to desensitization of the β₂AR (3). However, the mechanism(s) by which this inhibition occurs has not been completely elucidated, nor has the role of other EP receptor subtypes in mediating this effect been explored.

We define here a unique role for the EP₁ receptor expressed on ASM cells using both molecular studies in transfected cells and ASM cells and physiologic studies in trachea. We showed that the EP₁ receptor existed as a heterodimer (or higher-order oligomer) with the β₂AR, and the signaling of this complex was manifested by an altered β₂AR conformation and decreased functional interaction with G_αs. EP₁ activation by agonist further enhanced β₂AR desensitization, as shown in the [³⁵S]GTPγS studies. This effect was also readily observed in ASM cells utilizing whole-cell cAMP measurements and membrane [³⁵S]GTPγS binding and manifested at the physiologic level with intact tracheal rings. Interestingly, the extent of β₂AR desensitization evoked by the EP₁ agonist, as assessed using these distinctly different methods, amounted to approximately 50% in all cases. In addition, the EP₁ receptor, while capable of stimulating intracellular IP₃ and [Ca²⁺]_i (as assessed by global whole-cell assays), did not ultimately cause significant ASM contraction. Thus the dimer may serve not only to form a specialized signaling unit, but could also place the complex in a subcellular compartment that isolates it from activation of the downstream components of contraction. Indeed, β₂ARs have been shown to be localized to caveolae, which are specialized flask-shaped invaginations of the cell membrane that promote association of the receptor with certain signal transduction elements (35). However, we note that EP₁ receptor–mediated contraction has been observed in guinea pig trachea (36), suggesting that the extent of EP₁-mediated tracheal contraction may depend on the species.

We have thus considered the physiologic role of the EP₁ receptor in ASM as primarily one of modulating β₂AR function, with no observable direct effect on contraction. To our knowledge, this “monitor-like” function has not been previously described for a 7-TM receptor with its heterodimeric partner. However, there are somewhat analogous scenarios with other receptors. For the GABAb R1 and R2 subtypes, it appears that the R1:R2 heterodimer represents a distinct signaling unit, in that the R1 subtype binds the agonist while the R2 subtype couples to G protein activation (37). Furthermore, GABAb R1 cell surface expression requires heterodimer formation with GABAb R2 in the endoplasmic reticulum (38). For the α_1DAR, cell surface expression appears to be dependent on oligomeric association with the α_1BAR (39). Once so expressed, each subtype can signal independently. Similarly, the β₂AR can form dimers with the α_1DAR and promote its expression (40). Other examples of homo-and heterodimerization function have been recently reviewed (12, 15, 16). We are unaware of a previous report of the unique function of a 7-TM receptor that we describe here.

In the current report, we utilized the physiologic readout of airway contraction and relaxation as the relevant functional signal and found that activation of the EP₁ receptor had no direct physiologic action. However, we found that EP₁ activation resulted in substantial loss of β₂AR-mediated ASM relaxation. We demonstrated that this interaction of signal transduction networks occurred at the level of the receptors, rather than downstream effectors, and was not due to posttranslational modification of the β₂AR by phosphorylation or other potential distal-pathway crosstalk mechanisms (41). Rather, formation of the EP₁:β₂AR heterodimer was a distinct signaling complex, where the EP₁ receptor modulates β₂AR coupling to G_αs. Using multiple approaches, the EP₁-agonist dependency of the β₂AR modulation was demonstrated. However, we cannot exclude the notion that the mere presence of the heterodimer acts to modulate the function of one of its members, the β₂AR. Indeed, the [³⁵S]GTPγS binding studies showed that β₂AR coupling to G_αs was decreased when coexpressed with the EP₁ receptor. However, the spontaneous switching of the EP₁ receptor to the active conformation may mimic the agonist-occupied form; thus it is not unexpected that coexpression, particularly at relatively high levels, would have an agonist-occupied effect. Regardless, in the native state both receptors were expressed on ASM, and we showed that EP₁ receptor activation decreased β₂AR function in isolated ASM cells and in intact tissue.

In summary, we define a unique modulatory function of the EP₁ receptor, which acts primarily to regulate β₂AR function. In asthma, then, local concentrations of elevated PGE₂ could act to quench β₂AR function, resulting in decreased responsiveness to β-agonist. The mechanism of this receptor interaction appears to be due to the formation of a heterodimer; thus the EP₁:β₂AR dimer should be considered the functional signaling complex in ASM, with a unique phenotype based on steric interactions that lead to attenuated β₂AR-G_αs coupling.

This work was supported by NIH grants HL071609 (to D.W. McGraw) and HL045967 (to S.B. Liggett). The authors thank Michel Bouvier for providing constructs, Cheryl Theiss for assistance with cell culture, and Esther Getz Moses for manuscript preparation.

Address correspondence to: Stephen B. Liggett, University of Maryland School of Medicine, 20 Penn Street, HSF-2, Room S-112, Baltimore, Maryland 21201, USA. Phone: (410) 706-6256; Fax: (410) 706-6262; E-mail: sligg001@umaryland.edu .

Nonstandard abbreviations used: AR, adrenergic receptor; ASM, airway smooth muscle; BRET, bioluminescence resonant energy transfer; [Ca²⁺]_i, intracellular calcium; GABAb R2, γ-aminobutyric acid receptor subtype bR2; HEK, human embryonic kidney; IP₃, inositol 1,4,5-triphosphate; PGE₂, prostaglandin E₂; 17-PTP, 17-phenyl trinor-PGE₂; Rluc, Renilla luciferase; 7-TM, 7-transmembrane; YFP, yellow fluorescenprotein.

Conflict of interest: The authors have declared that no conflict of interest exists.

Citation for this article: J. Clin. Invest. 116:1400–1409 (2006). doi:10.1172/JCI25840.

See the related Commentary beginning on page 1210.

1. Green, S.A., Liggett, S.B. 1996. G protein coupled receptor signalling in the lung. In The genetics of asthma.S. Liggett and D. Meyers, editors. Marcel Dekker Inc.New York, New York, USA.67-90.
2. Sweatman, W.J., Collier, H.O. 1968. Effects of prostaglandins on human bronchial muscle. Nature. 217:69.
   View this article via: PubMed CrossRef Google Scholar
3. Pang, L., Holland, E., Knox, A.J. 1998. Role of cyclo-oxygenase-2 induction in interleukin-1beta induced attenuation of cultured human airway smooth muscle cell cyclic AMP generation in response to isoprenaline. Br. J. Pharmacol. 125:1320-1328.
   View this article via: PubMed CrossRef Google Scholar
4. Delamere, F., et al. 1994. Production of PGE2 by bovine cultured airway smooth muscle cells and its inhibition by cyclo-oxygenase inhibitors. Br. J. Pharmacol. 111:983-988.
   View this article via: PubMed Google Scholar
5. Churchill, L., et al. 1989. Cyclooxygenase metabolism of endogenous arachidonic acid by cultured human tracheal epithelial cells. Am. Rev. Respir. Dis. 140:449-459.
   View this article via: PubMed Google Scholar
6. Ozaki, T., Rennard, S.I., Crystal, R.G. 1987. Cyclooxygenase metabolites are compartmentalized in the human lower respiratory tract. J. Appl. Physiol. 62:219-222.
   View this article via: PubMed Google Scholar
7. Sheller, J.R., Mitchell, D., Meyrick, B., Oates, J., Breyer, R. 2000. EP(2) receptor mediates bronchodilation by PGE(2) in mice. J. Appl. Physiol. 88:2214-2218.
   View this article via: PubMed Google Scholar
8. Henry, P.J., Rigby, P.J., Goldie, R.G. 1990. Distribution of beta 1- and beta 2-adrenoceptors in mouse trachea and lung: a quantitative autoradiographic study. Br. J. Pharmacol. 99:136-144.
   View this article via: PubMed Google Scholar
9. Daaka, Y., Luttrell, L.M., Lefkowitz, R.J. 1997. Switching of the coupling of the β₂ -adrenergic receptor to different G proteins by protein kinase A. Nature. 390:88-91.
   View this article via: PubMed CrossRef Google Scholar
10. De Blasi, A., Lipartiti, M., Motulsky, H.J., Insel, P.A., Fratelli, M. 1985. Agonist-induced redistribution of beta-adrenergic receptors on intact human mononuclear leukocytes: redistributed receptors are nonfunctional. J. Clin. Endocrinol. Metab. 61:1081-1088.
    View this article via: PubMed Google Scholar
11. Angers, S., Salahpour, A., Bouvier, M. 2002. Dimerization: an emerging concept for G protein-coupled receptor ontogeny and function. Annu. Rev. Pharmacol. Toxicol. 42:409-435.
    View this article via: PubMed CrossRef Google Scholar
12. Bulenger, S., Marullo, S., Bouvier, M. 2005. Emerging role of homo- and heterodimerization in G-protein-coupled receptor biosynthesis and maturation. Trends Pharmacol. Sci. 26:131-137.
    View this article via: PubMed CrossRef Google Scholar
13. Breit, A., Lagace, M., Bouvier, M. 2004. Hetero-oligomerization between beta2- and beta3-adrenergic receptors generates a beta-adrenergic signaling unit with distinct functional properties. J. Biol. Chem. 279:28756-28765.
    View this article via: PubMed CrossRef Google Scholar
14. Angers, S., et al. 2000. Detection of beta 2-adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET). Proc. Natl. Acad. Sci. U. S. A. 97:3684-3689.
    View this article via: PubMed CrossRef Google Scholar
15. Salahpour, A., Angers, S., Bouvier, M. 2000. Functional significance of oligomerization of G-protein-coupled receptors. Trends Endocrinol. Metab. 11:163-168.
    View this article via: PubMed CrossRef Google Scholar
16. Bouvier, M. 2001. Oligomerization of G-protein-coupled transmitter receptors. Nat. Rev. Neurosci. 2:274-286.
    View this article via: PubMed CrossRef Google Scholar
17. Narumiya, S., Sugimoto, Y., Ushikubi, F. 1999. Prostanoid receptors: structures, properties, and functions. Physiol. Rev. 79:1193-1226.
    View this article via: PubMed Google Scholar
18. Funk, C.D., et al. 1993. Cloning and expression of a cDNA for the human prostaglandin E receptor EP1 subtype. J. Biol. Chem. 268:26767-26772.
    View this article via: PubMed Google Scholar
19. Katsuyama, M., et al. 1995. The mouse prostaglandin E receptor EP2 subtype: cloning, expression, and northern blot analysis. FEBS Lett. 372:151-156.
    View this article via: PubMed CrossRef Google Scholar
20. Breyer, R.M., et al. 1996. Cloning and expression of the rabbit prostaglandin EP4 receptor. Am. J. Physiol. 270:F485-F493.
    View this article via: PubMed Google Scholar
21. Regan, J.W., et al. 1994. Cloning of a novel human prostaglandin receptor with characteristics of the pharmacologically defined EP2 subtype. Mol. Pharmacol. 46:213-220.
    View this article via: PubMed Google Scholar
22. Namba, T., et al. 1993. Alternative splicing of C-terminal tail of prostaglandin E receptor subtype EP3 determines G-protein specificity. Nature. 365:166-170.
    View this article via: PubMed CrossRef Google Scholar
23. Negishi, M., et al. 1995. Signal transductions of three isoforms of mouse prostaglandin E receptor EP3 subtype. Adv. Prostaglandin Thromboxane Leukot. Res. 23:255-257.
    View this article via: PubMed Google Scholar
24. Moore, B.B., et al. 2005. Bleomycin-induced E prostanoid receptor changes alter fibroblast responses to prostaglandin E2. J. Immunol. 174:5644-5649.
    View this article via: PubMed Google Scholar
25. Krawiec, M.E., et al. 2001. Persistent wheezing in very young children is associated with lower respiratory inflammation. Am. J. Respir. Crit. Care Med. 163:1338-1343.
    View this article via: PubMed Google Scholar
26. Profita, M., et al. 2003. Increased prostaglandin E2 concentrations and cyclooxygenase-2 expression in asthmatic subjects with sputum eosinophilia. J. Allergy Clin. Immunol. 112:709-716.
    View this article via: PubMed CrossRef Google Scholar
27. Redington, A.E., et al. 2001. Increased expression of inducible nitric oxide synthase and cyclo-oxygenase-2 in the airway epithelium of asthmatic subjects and regulation by corticosteroid treatment. Thorax. 56:351-357.
    View this article via: PubMed CrossRef Google Scholar
28. Taha, R., et al. 2000. Prostaglandin H synthase 2 expression in airway cells from patients with asthma and chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 161:636-640.
    View this article via: PubMed Google Scholar
29. Tilley, S.L., et al. 2003. Receptors and pathways mediating the effects of prostaglandin E2 on airway tone. Am. J. Physiol. Lung Cell. Mol. Physiol. 284:L599-L606.
    View this article via: PubMed Google Scholar
30. Mathe, A.A., Hedqvist, P. 1975. Effect of prostaglandins F2 alpha and E2 on airway conductance in healthy subjects and asthmatic patients. Am. Rev. Respir. Dis. 111:313-320.
    View this article via: PubMed Google Scholar
31. Vigano, T., et al. 1997. Cyclooxygenase-2 and synthesis of PGE2 in human bronchial smooth-muscle cells. Am. J. Respir. Crit. Care Med. 155:864-868.
    View this article via: PubMed Google Scholar
32. Belvisi, M.G., et al. 1997. Induction of cyclo-oxygenase-2 by cytokines in human cultured airway smooth muscle cells: novel inflammatory role of this cell type. Br. J. Pharmacol. 120:910-916.
    View this article via: PubMed CrossRef Google Scholar
33. Fong, C.Y., Pang, L., Holland, E., Knox, A.J. 2000. TGF-beta1 stimulates IL-8 release, COX-2 expression, and PGE(2) release in human airway smooth muscle cells. Am. J. Physiol. Lung Cell. Mol. Physiol. 279:L201-L207.
    View this article via: PubMed Google Scholar
34. Pang, L., Knox, A.J. 1997. Effect of interleukin-1 beta, tumour necrosis factor-alpha and interferon-gamma on the induction of cyclo-oxygenase-2 in cultured human airway smooth muscle cells. Br. J. Pharmacol. 121:579-587.
    View this article via: PubMed CrossRef Google Scholar
35. Steinberg, S.F. 2004. beta(2)-Adrenergic receptor signaling complexes in cardiomyocyte caveolae/lipid rafts. J. Mol. Cell. Cardiol. 37:407-415.
    View this article via: PubMed CrossRef Google Scholar
36. Ndukwu, I.M., White, S.R., Leff, A.R., Mitchell, R.W. 1997. EP1 receptor blockade attenuates both spontaneous tone and PGE2-elicited contraction in guinea pig trachealis. Am. J. Physiol. 273:L626-L633.
    View this article via: PubMed Google Scholar
37. Galvez, T., et al. 2001. Allosteric interactions between GB1 and GB2 subunits are required for optimal GABA(B) receptor function. EMBO J. 20:2152-2159.
    View this article via: PubMed CrossRef Google Scholar
38. Kuner, R., et al. 1999. Role of heteromer formation in GABAB receptor function. Science. 283:74-77.
    View this article via: PubMed CrossRef Google Scholar
39. Hague, C., Uberti, M.A., Chen, Z., Hall, R.A., Minneman, K.P. 2004. Cell surface expression of alpha1D-adrenergic receptors is controlled by heterodimerization with alpha1B-adrenergic receptors. J. Biol. Chem. 279:15541-15549.
    View this article via: PubMed CrossRef Google Scholar
40. Uberti, M.A., Hague, C., Oller, H., Minneman, K.P., Hall, R.A. 2005. Heterodimerization with beta2-adrenergic receptors promotes surface expression and functional activity of alpha1D-adrenergic receptors. J. Pharmacol. Exp. Ther. 313:16-23.
    View this article via: PubMed CrossRef Google Scholar
41. McGraw, D.W., Almoosa, K.F., Paul, R.J., Kobilka, B.K., Liggett, S.B. 2003. Antithetic regulation by β-adrenergic receptors of Gq-receptor signaling via phospholipase-C underlies the airway β-agonist paradox. J. Clin. Invest. 112:619-626.
    View this article via: JCI CrossRef PubMed Google Scholar
42. McGraw, D.W., et al. 1999. Transgenic overexpression of β₂ -adrenergic receptors in airway smooth muscle alters myocyte function and ablates bronchial hyperreactivity. J. Biol. Chem. 274:32241-32247.
    View this article via: PubMed CrossRef Google Scholar
43. Small, K.M., et al. 2004. Polymorphisms of the cardiac presynaptic α2Cadrenergic receptors: diverse intragenic variability with haplotype-specific functional effects. Proc. Natl. Acad. Sci. U. S. A. 101:13020-13025.
    View this article via: PubMed CrossRef Google Scholar
44. Mialet-Perez, J., Green, S.A., Miller, W.E., Liggett, S.B. 2004. A primate-dominant third glycosylation site of the β2-adrenergic receptor routes receptors to degradation during agonist regulation. J. Biol. Chem. 279:38603-38607.
    View this article via: PubMed CrossRef Google Scholar
45. Green, S., Liggett, S.B. 1994. A proline-rich region of the third intracellular loop imparts phenotypic β₁ - versus β₂ -adrenergic receptor coupling and sequestration. J. Biol. Chem. 269:26215-26219.
    View this article via: PubMed Google Scholar
46. Jewell-Motz, E.A., Small, K.M., Liggett, S.B. 2000. α_2A /α_2C -adrenergic receptor third loop chimera show that agonist interaction with receptor-subtype backbone establishes G protein-coupled receptor kinase phosphorylation. J. Biol. Chem. 275:28989-28993.
    View this article via: PubMed CrossRef Google Scholar
47. Small, K.M., Forbes, S.L., Brown, K.M., Liggett, S.B. 2000. An Asn to Lys polymorphism in the third intracellular loop of the human α_2A -adrenergic receptor imparts enhanced agonist-promoted G_i coupling. J. Biol. Chem. 275:38518-38523.
    View this article via: PubMed CrossRef Google Scholar
48. Milligan, G. 2003. Principles: extending the utility of [35S]GTP gamma S binding assays. Trends Pharmacol. Sci. 24:87-90.
    View this article via: PubMed CrossRef Google Scholar