Crosstalk between G_i and G_q/G_s pathways in airway smooth muscle regulates bronchial contractility and relaxation

Receptor-mediated airway smooth muscle (ASM) contraction via G_αq, and relaxation via G_αs, underlie the bronchospastic features of asthma and its treatment. Asthma models show increased ASM G_αi expression, considered the basis for the proasthmatic phenotypes of enhanced bronchial hyperreactivity to contraction mediated by M₃-muscarinic receptors and diminished relaxation mediated by β₂-adrenergic receptors (β₂ARs). A causal effect between G_i expression and phenotype has not been established, nor have mechanisms whereby G_i modulates G_q/G_s signaling. To delineate isolated effects of altered G_i, transgenic mice were generated overexpressing G_αi2 or a G_αi2 peptide inhibitor in ASM. Unexpectedly, G_αi2 overexpression decreased contractility to methacholine, while G_αi2 inhibition enhanced contraction. These opposite phenotypes resulted from different crosstalk loci within the G_q signaling network: decreased phospholipase C and increased PKCα, respectively. G_αi2 overexpression decreased β₂AR-mediated airway relaxation, while G_αi2 inhibition increased this response, consistent with physiologically relevant coupling of this receptor to both G_s and G_i. IL-13 transgenic mice (a model of asthma), which developed increased ASM G_αi, displayed marked increases in airway hyperresponsiveness when G_αi function was inhibited. Increased G_αi in asthma is therefore a double-edged sword: a compensatory event mitigating against bronchial hyperreactivity, but a mechanism that evokes β-agonist resistance. By selective intervention within these multipronged signaling modules, advantageous G_s/G_q activities could provide new asthma therapies.

Airway smooth muscle (ASM) contraction and relaxation are primarily regulated by G protein–coupled receptors, the former mediated by receptors signaling to G_q and the latter by those that couple to G_s (1, 2). Many inflammatory cascades in asthma evoke bronchoconstriction by promoting local increases of G_q receptor agonists such as acetylcholine, cysteinyl leukotrienes, prostaglandins, and histamine, which activate their cognate receptors expressed on ASM. There appear to be fewer G_s-coupled receptors that act via endogenous agonists to counteract bronchoconstriction, but the β₂-adrenergic receptor (β₂AR) of ASM is the target of pharmacologically administered β-agonists and is typically highly effective in relaxing constricted airways. The molecular events and critical transduction elements for these 2 classes of receptors are well recognized. Agonist binding to receptors such as the M₃-muscarinic receptor promote disassociation of heterotrimeric G_q into G_α and G_βγ subunits, with the α subunit activating phospholipase C (PLC; which promotes inositol-3 phosphate and diacylglycerol production) and the latter activating PKC. Receptors such as the β₂AR act via G_αs to stimulate the effector adenylyl cyclase, resulting in cAMP production and activation of PKA. Substantial interest has revolved around how these pathways might be modified in asthma, because there may be nodal points which are critical for the pathogenesis of bronchospasm, or may be particularly amenable for pharmacologic intervention. A number of studies have shown, somewhat surprisingly, that a third major class of G proteins, G_i (inhibitory guanine nucleotide binding protein α subunits 2 and 3; G_αi2 and G_αi3, respectively), is increased in ASM in animal models of asthma. Early studies by Grunstein and colleagues (3) showed that rabbit tracheal smooth muscle that was passively sensitized with serum from atopic asthmatics had enhanced ex vivo contraction to acetylcholine and decreased relaxation to the β-agonist isoproterenol. These events were temporally related to increased G_αi protein expression, and these characteristics were attenuated by pertussis toxin (PTX), which inactivates G_αi.

Other studies have shown that cytokine exposure (4, 5) and rhinovirus infection (6) give this same physiologic phenotype, which is accompanied by a several-fold increase in ASM G_αi expression. It remains unclear how the 2 proasthmatic phenotypes of hyperresponsiveness to bronchoconstriction and resistance to bronchodilatation, which are mediated by G_q- and G_s-coupled receptors, respectively, could be influenced by the cellular expression levels of G_i. Recent studies in recombinant cells have shown, however, that β₂AR, once phosphorylated by PKA, has the capacity to couple to G_i (7). Given that G_αi inhibits adenylyl cyclase, this dual coupling may serve to attenuate the cAMP response. However, the magnitude of the physiologic effect in the airway of this coupling “switch” is not known, nor is it clear whether modest changes in G_i could amplify or modify in some other way this bifurcated coupling. A mechanism by which G_i expression alters receptor-G_q signaling is not readily apparent; however, previously unrecognized crosstalk between signaling pathways in ASM presumed to be unrelated has now been documented (8). Of particular interest in regard to the aforementioned models is the difficulty in ascribing a causal effect of increased G_i to the phenotypes in light of the complex regulatory events within many diverse pathways that are undoubtedly underway in asthma. Similarly, it is unclear whether the increase in G_i is participating in the pathologic process or is a compensatory event that may be a factor mitigating against one or both of the asthmatic phenotypes. To ascertain the role of isolated increased or decreased G_i function on bronchoconstrictive and dilating signaling of G protein–coupled receptors in the airway, transgenic mice were generated overexpressing G_αi2 (G_αi2-OE mice) and expressing a G_αi2 inhibitory peptide (G_αi2-IP mice) in ASM. We found that increased G_i did not cause enhanced contraction, but rather decreased the contractile response, while G_i inhibition evoked enhanced contraction. However, although these phenotypes are physiologically antithetic, their mechanistic basis was crosstalk with 2 different elements of the G_q signaling pathway. In the IL-13 transgenic mouse model of asthma and airway hyperresponsiveness, where we found increased G_i in ASM, inhibition of G_i resulted in even greater hyperresponsiveness. Thus the increase in G_i appears to be a compensatory event aimed at attenuating bronchial hyperresponsiveness. We also found that β₂AR-mediated relaxation of ASM was substantially regulated by G_i expression, which we propose is due to direct coupling of this receptor to G_i.

G_αi2 is increased in a genetic model of asthma, and PTX alters airway responsiveness. To assess whether the findings by Grunstein and colleagues of increased G_αi in rabbit models of asthma (3) are true for another model, we examined expression of G_αi2 in tracheae from the inducible IL-13–overexpressing mouse (9). In these mice IL-13 is targeted to the airway by the Clara cell secretory protein (CCSP) promoter, with induction by 3 weeks of doxycycline administration; they develop inflammation, bronchial hyperresponsiveness to methacholine, fibrosis, and airspace enlargement (9, 10). Figure 1 shows that tracheal expression of G_αi2 assessed by Western blot increased approximately 8-fold upon IL-13 induction in these transgenic mice. These data pointed toward increased G_i as a common feature of asthmatic-like airways from passive sensitization with asthmatic serum (3), cytokines (4, 5), and rhinovirus exposure (6) as well as increased IL-13 production. However, these prior studies do not indicate whether isolated changes in G_i levels influence airway contraction or relaxation responses, nor do they implicate the increase in G_i in these models as a mechanism that underlies altered airway signaling or address whether the increase represents a compensatory event. We thus proceeded with screening studies using in vivo PTX, which ablates G_i coupling by ADP-ribosylation of the α subunit. Wild-type mice were treated with i.p. PTX or with vehicle, and 18 hours later airway resistance was measured in the intact, ventilated mouse using previously described methods (8, 11). Unexpectedly, PTX treatment increased the contractile response to methacholine (mediated by the G_q-coupled M₃-muscarinic receptor) as shown by both an increased maximal response and a left-shifted dose-response curve (Figure 2A). To assess βAR-mediated relaxation, mice were pretreated with aerosolized isoproterenol prior to methacholine challenge. With PTX treatment, β₂AR signaling appeared to be enhanced, in that there was little contraction to methacholine under these conditions (Figure 2B). While these findings potentially implicate G_i function as an important element in airway contraction and relaxation, the use of systemic PTX is not specific for ASM and does not provide information as to the effects of increased G_αi expression, as is found in the aforementioned asthma models. These limitations led to the generation of transgenic mice G_αi2-OE (incorporating the full-length human G_αi2 cDNA) and G_αi2-IP (a “minigene” peptide consisting of the C-terminal portion of G_αi2) targeted to ASM.

Figure 1

Induction of IL-13 in mouse airways increases G_αi2 expression. Three-week-old bitransgenic mice generated to have doxycycline-inducible expression of IL-13 targeted to the airways by the CC10 promoter were fed normal food (–Dox) or doxycycline-impregnated food (+Dox) for 3 weeks. Tracheae were homogenized, and membrane preparations were subjected to Western blots performed with G_αi2 and GAPDH antibodies. Induction of IL-13 increased G_αi2 expression by approximately 8-fold.

Figure 2

PTX alters airway contraction and relaxation responses. FVB/N NTG mice were treated with vehicle (veh) or i.p. PTX for 18 hours. Airway resistance and the response to inhaled methacholine (mch) was determined in intact ventilated mice (see Methods). Studies were performed without (A) or with (B) pretreatment with inhaled isoproterenol (Iso). Results are from 5 independent experiments. ^#P < 0.01 versus vehicle.

Transgenic expression of G_αi2 and a G_αi2 competitive inhibitor. Transgenic mice were generated using smooth muscle actin promoter 8 (SMP8); the injected constructs consisted of the full-length cDNA for human G_αi2 or a cDNA encoding the carboxy terminus of human G_αi2 (58 amino acids), which also included a 5' HA-tag (see Methods). Mice were screened for the transgene by PCRs using genomic DNA derived from digested tails. Transgenic mice had a normal body weight and habitus, and their viability was the same as nontransgenic littermate (NTG) mice. To ascertain whether the transgenes were expressed in the cell type of interest, we prepared RNA from cultures of primary ASM cells that were established from tracheal explants of heterozygous F2 and F3 mice. We detected mRNA for the G_αi2 transgene by RT-PCR using human-specific G_αi2 primers and for the G_αi2-IP transgene using primers based on sequence from the 5' HA-tag and human G_αi2. As shown in Figure 3A, the human G_αi2 transcript was identified only in ASM cells derived from G_αi2-OE mice. Immunoblots using G_αi2 antisera further showed overexpression of G_αi2 protein by approximately 10-fold in the ASM cells derived from the full-gene overexpressors (Figure 3B; note that the G_αi2-OE protein was underloaded to obtain a reasonable visual comparison with the NTG signal). Although we were not able to reproducibly detect protein expression of the G_αi2-IP minigene, RT-PCR revealed the chimeric HA–G_αi-IP transcript in ASM cells only in the G_αi-IP–derived cell lines, not the NTG or G_αi2-OE lines. Formalin-fixed sections of the lungs revealed no microscopic evidence of increased ASM mass, basement membrane thickening, inflammation, fibrosis, or parenchymal distortion in either transgenic line. Figure 3C shows representative sections that include both airway and parenchyma. Mean systolic blood pressure measured noninvasively by tail cuff readings over a period of 5 days did not differ among NTG, G_αi2-OE, and G_αi2-IP mice (125 ± 1 mmHg, 125 ± 3 mmHg, and 126 ± 5 mmHg, respectively; n = 10; data not shown).

Figure 3

G_αi2-OE and G_αi2-IP in ASM cells. (A) RT-PCRs from total RNA derived from primary ASM cell cultures from NTG (lanes 1, 2, 7, 8, 13, and 14), G_αi2-IP (lanes 3, 4, 9, 10, 15, and 16), and G_αi2-OE (lanes 5, 6, 11, 12, 17, and 18) mice were performed using transgene-specific primers. h-, human. (B) Overexpression of G_αi2 protein in ASM lysates from G_αi2-OE cells. Note that the protein from G_αi2-OE cells was underloaded by 10-fold compared with NTG cells, as shown by the GAPDH signals. (C) Histopathologic specimens from the lungs of the NTG and 2 representative transgenic lines. Shown are hematoxylin and eosin–stained (upper panels) and trichrome-stained (lower panels) representative sections. Original magnification, ×20.

Effects of altering G_αi by transgenesis on airway contractile and relaxation responses. Baseline airway resistance did not differ among G_αi2-OE, G_αi2-IP, and NTG mice (0.53 ± 0.061 cm H₂O/ml/s, 0.58 ± 0.036 cm H₂O/ml/s, and 0.57 ± 0.041 cm H₂O/ml/s, respectively; n = 6–9; data not shown). However, G_αi2-OE mice exhibited a depressed constrictive response (P < 0.005) to methacholine, primarily due to a 50% decrease in the maximal response compared to the maximal response observed in NTG mice (Figure 4A). Consistent with an antithetic relationship of G_i expression and function to physiologic effect (contraction), G_αi2-IP mice had a greater response to methacholine (P < 0.005), with the maximal response being 50% greater than that of NTG mice (Figure 4A). The contractile responses to methacholine observed in the G_αi2-IP mice were consistent with those observed with PTX treatment (Figure 2A). The effects of G_i expression on β₂AR-mediated relaxation responses were assessed by pretreatment of mice with aerosolized isoproterenol, followed by methacholine challenge. As shown in Figure 4B, in NTG mice this caused the expected approximately 40% decrease in maximal bronchoconstriction due to methacholine. For the G_αi2-IP mice, a markedly enhanced β₂AR function was observed, with very little bronchoconstriction by methacholine in these mice when pretreated with isoproterenol. This effect of G_αi2-IP amounted to an approximate 80% decrease in resistance compared with methacholine treatment alone in these mice (P < 0.01), and indeed resistance after isoproterenol was also less than in NTG mice under the same conditions (P < 0.05). In the G_αi2-OE mice there was an apparent small decrease in contraction in the presence of isoproterenol, but this did not reach statistical significance, suggesting impaired β₂AR function. Thus, like the bronchoconstrictive response (where G_αi2 upregulation and downregulation had opposite effects), G_αi2 inhibition enhanced relaxation via β₂AR, while G_αi2 overexpression decreased it.

Figure 4

G_αi2-OE or G_αi2-IP in mouse ASM alters responsiveness to M₃-muscarinic and β₂AR agonists. (A) Intact, ventilated mice were administered the indicated doses of inhaled methacholine and airway resistance measured. The G_αi2-IP mice had a greater response to methacholine than NTG mice, while G_αi2-OE mice were less responsive than NTG. ^†P < 0.005 versus NTG, ANOVA. (B) Mice were pretreated with vehicle or inhaled isoproterenol and then underwent the methacholine challenge. NTG mice showed the expected decrease in maximal contraction while the G_αi2-OE mice showed no statistical difference in airway resistance with isoproterenol. The G_αi2-IP mice had an enhanced response to isoproterenol. Results are from 6–8 independent experiments. *P < 0.05; ^#P < 0.01.

We considered that the β₂AR-mediated relaxation phenotype of G_αi2-OE mice could be a result of altered expression of ASM β₂AR or altered function of another receptor acting via G_i. However, β₂AR expression, as determined by quantitative ¹²⁵I-cyanopindolol binding, did not differ among NTG, G_αi-OE, and G_αi-IP cells (15 ± 2.1 fmol/mg, 19 ± 3.8 fmol/mg, and 20 ± 2.9 fmol/mg, respectively; n = 4; data not shown). We thus considered the potential for enhanced function of a classic receptor-G_i–coupled pathway (such as the M₂-muscarinic pathway) that might antagonize G_s-coupled relaxation at adenylyl cyclase. To address this potential mechanism, excised tracheal rings were studied ex vivo to measure relaxation responses during KCl contraction. Rings were treated with vehicle or the M₂-muscarinic receptor agonist pilocarpine (0.1 μM) and then contracted continuously with 60 mM KCl. Relaxation was then ascertained in response to increasing concentrations of isoproterenol. As expected, with NTG rings, pilocarpine modestly blunted isoproterenol-mediated relaxation from 51% ± 5.0% to 62% ± 5.0% of maximal contraction (a 20% decrease; Figure 5). However, there was no accentuation of this effect in the G_αi2-OE rings, with pilocarpine reducing isoproterenol-mediated relaxation by an amount comparable to that observed with NTG rings: from 67% ± 3.3% to 81% ± 4.0% (a 22% decrease; Figure 5).

Figure 5

M₂-muscarinic receptor activation does not alter the G_i-effect on β₂AR-mediated smooth muscle relaxation. Shown are experiments where isolated tracheal rings were pretreated with 0.1 μM pilocarpine (pilo) or vehicle, contracted with 60 mM KCl, and then exposed to increasing concentrations of isoproterenol. The brackets clarify the effects of pilocarpine compared with vehicle, which were not significantly different, in the NTG and G_αi2-OE groups (n = 4 experiments).

G_αi2 expression regulates multiple downstream signaling elements. Potential crosstalk between G_αi2 expression and function and relevant downstream mediators of contraction and relaxation was assessed by Western blot analyses from primary ASM cells derived from NTG, G_αi2-IP, and G_αi2-OE mice (Figure 6). Of particular interest — because of the counterintuitive effects on contraction — was expression of elements involved in airway constriction: G_αq, PLC, and PKC isoforms. G_αq levels were not altered in either transgenic. However, its effector, PLCβ3, was downregulated by more than 50% (P < 0.005) in the G_αi2-OE cells but was not different in the G_αi2-IP cells compared with NTG cells (Figure 6B). These results are consistent with our physiologic studies showing that the G_αi2-OE mice were hyporesponsive to methacholine. However, this does not explain the hyperresponsiveness observed in G_αi2-IP mice, which indicates that while the phenotypes of the G_αi2-OE and G_αi2-IP mice are oppositional, the mechanisms by which these responses are attained are different and potentially multifactorial. PKCα was upregulated by approximately 50% (P < 0.01) in G_αi2-IP cells but unchanged in G_αi2-OE cells (Figure 6B). Given that PKCα has been demonstrated to sensitize smooth muscle to contraction (12–15), this upregulation in the G_αi2-IP cells is consistent with the hyperresponsive contractile phenotype. Less is known about the other PKC isoforms in regards to G protein–coupled receptor–mediated smooth muscle contraction. We noted that expression of PKCβ1 was not altered, while PKCε and PKCδ were downregulated by approximately 20% in cells from both lines of mice. In terms of the β₂AR response, we found no pattern of regulation of downstream signaling elements that was consistent with the 2 phenotypes. G_αs was modestly decreased (about 30%) in both G_αi2-OE and G_αi2-IP cells, which does not explain the discordant signaling, and G protein–coupled receptor kinase 2 (GRK2) was decreased by approximately 40% in G_αi2-OE cells. However, this decrease would be expected to enhance β₂AR function, instead of the decreased function we observed with G_αi2-OE. Our findings are consistent with β₂AR coupling to G_i with attenuation of β₂AR stimulation of adenylyl cyclase. Indeed, the primary signal that has been used to assess β₂AR-G_i coupling is p44/p42 activation (7), which was markedly enhanced (about 3-fold; P < 0.001) in the G_αi2-OE cells (Figure 6B).

Figure 6

Downstream modulation of signal transduction elements by altered G_αi in ASM cells. (A) Western blots from 4 independent ASM cell lysates for each line. Symbols at right of blots denote significant increase (up arrows), significant decrease (down arrows), or no change (dashes) in protein expression from G_αi2-OE and G_αi2-IP derived cells (left and right symbols, respectively) compared with NTG cells. (B) Quantification of Western blot data, presented as mean ± SEM. *P < 0.05; ^#P < 0.01; ^†P < 0.005; ^‡P < 0.001 versus NTG.

Increased G_αi2 in a mouse asthma model is a beneficial compensatory event attenuating airway hyperresponsiveness. The physiologic consequences of transgenic overexpression of G_αi2 revealed a decrease, rather than the expected increase, in airway constriction to methacholine. In addition, the molecular events that we defined as a consequence of increased G_αi2 were consistent with this phenotype. This suggested that the increase in G_αi that has been observed in several animal models of asthma is a compensatory event, acting to attenuate bronchial hyperresponsiveness. To test this, we used the inducible IL-13 mice, which displayed a significant increase in G_αi2 upon IL-13 induction (Figure 1). These mice are known to be hyperresponsive, which is a key aspect of their phenotype (10), but the question remains as to whether inactivating G_αi would further increase airway contractility in these mice, thus revealing the partial protective effect against hyperresponsiveness provided by increased G_αi2. The IL-13 mice were given normal chow or chow with doxycycline for 3 weeks, and then some mice were treated with PTX i.p. and studied 18 hours later. As shown in Figure 7, IL-13 induction increased methacholine-promoted airway constriction. Treatment with PTX further enhanced contraction, amounting to a 2-fold increase compared with doxycycline-only mice. These data are consistent with the aforementioned concept that the increase in G_αi that occurs in this animal model of asthma (and others as well) acts in a compensatory manner to limit airway hyperresponsiveness. The β-agonist isoproterenol afforded protection against bronchoconstriction in the noninduced IL-13 mice. Interestingly, β-agonist provided no protection in the IL-13–induced mouse, where extensive airway constriction due to methacholine was observed, consistent with depressed βAR function. However, there was a nonsignificant trend toward less constriction in the presence of isoproterenol with PTX treatment, which supports improved βAR function.

Figure 7

The increase in G_αi in IL-13 mice limits airway hyperresponsiveness. Shown are experiments where the IL-13 doxycycline-inducible mice were studied in the absence or presence of induction with doxycycline. The latter were treated with i.p. PTX or vehicle. Airway resistance was measured using the ventilated mouse model, at a single concentration of aerosolized methacholine (10 mg/ml). After resistance returned to baseline, mice were treated with the same concentration of methacholine with isoproterenol. PTX enhanced methacholine-promoted airway constriction (compare black versus shaded bars under methacholine). These results are consistent with the increased G_αi found in these mice after IL-13 induction acting to attenuate hyperresponsiveness. Results are from independent experiments with 2–3 mice in each group. *P < 0.05.

The aim of these studies was to ascertain the relevance of altered G_αi2 expression on airway contraction and relaxation. As introduced earlier, a several-fold increase in G_αi2 has been found in a number of models of asthma, and it has been proposed that both bronchial hyperresponsiveness to contraction and impaired β₂AR-mediated relaxation can be attributed to this increase (3–6). Undoubtedly, there are expression changes in many signaling elements in these asthma models, and thus a causal relationship between G_αi expression and the above phenotypes cannot be established, nor can the specific effects of altered G_αi expression, in isolation, be ascertained. In the current work we show that increased G_αi2 resulted in decreased contractile responsiveness, which contradicts the aforementioned proposed paradigm. Furthermore, as a complementary and independent approach, we examined the effect of lowering G_αi function via a minigene. This resulted in increased contraction, which is the opposite phenotype to G_αi2-OE, and thus supports the notion that up- or downregulation of G_i levels has opposing effects on contraction. The phenotype of the PTX-treated mice is additional confirmation of the G_αi2-IP results, and also indicates that the effect of the minigene was not because of its interaction with receptor-G_q signaling, since PTX ADP-ribosylates G_αi. Thus, the increase in G_αi observed in animal models of asthma and/or bronchial hyperresponsiveness is clearly not the cause of contractile hyperresponsiveness: it actually decreased the methacholine response. Indeed, it appears that an increase in G_αi is a beneficial compensatory event, acting to limit the extent of hyperresponsiveness. To confirm this notion, the IL-13–inducible mice, which have airway hyperresponsiveness and increased G_αi2, were treated with PTX and displayed further enhancement of bronchoconstriction due to methacholine.

The effects of altering G_i expression and function on methacholine-promoted bronchoconstriction are not readily explained by alterations in muscarinic receptor–G protein interactions (i.e., at the receptor–G protein interface). The M₃-muscarinic receptor couples to G_q, and there is no evidence of promiscuous coupling of this receptor to G_i. Nor can the phenotype be attributed to altered M₂-muscarinic receptor function. Although the M₃ subtype is clearly the major muscarinic receptor that mediates airway contraction to muscarinic agonists, we recognize that the M₂ subtype also contributes (16). This subtype, which is G_i coupled, appears to promote contraction via decreasing cAMP and eliciting nonselective cation conductance (17). However, our results with the G_αi2-OE mice cannot be explained by enhanced M₂-receptor function, because in these mice we found a decreased bronchoconstrictive response to methacholine, and when M₂-G_i coupling would be expected to be diminished with the G_αi2-IP minigene mice, we observed enhanced contraction. Rather, the mechanisms for the phenotypes appear to be due to crosstalk between G_i and downstream mediators of contraction. While the phenotypes of the G_αi2-OE and G_αi2-IP mice are opposites, the mechanism for these phenotypes does not appear to be caused by regulation of a single transduction element. Instead, 2 elements that are closely linked in receptor-G_q signaling, PLCβ3 and PKCα, are each independently regulated. Increased G_αi resulted in downregulation of PLCβ3, the effector for G_q-coupled receptor activation. In contrast, PLCβ3 expression was unchanged in G_αi2-IP ASM, but PKCα, whose activation is a consequence of PLC activation, was upregulated. In a positive feedback loop, PKCα sensitizes smooth muscle to receptor-mediated contraction via G_q receptors. This effect has been particularly well documented in vascular smooth muscle, where PKC increases the myofilament force sensitivity to intracellular Ca²⁺ concentration (12–15). Interestingly, increased G_αi has been reported in many animal models of cardiac hypertrophy and heart failure (18–20) as well as in human heart failure (21). Taken together with the various animal models of asthma, an increase in G_αi may be a generalized response to cellular hypertrophy, which in asthma contributes to the increased ASM mass (22). Although G_αi2 is not increased in hearts overexpressing β₂AR, coupling of β₂AR to G_i appears to provide protection against β₂AR-mediated cardiomyopathy (23).

The β₂AR-mediated relaxation phenotype is consistent with the recent recognition that this receptor, a classic G_s-coupled receptor, can also couple to G_i (7). What has remained unknown is the physiologic relevance of β₂AR-G_i coupling and whether changes in G_i expression can modulate a physiologic effect (as opposed to G_i being in such excess that its expression is not rate limiting). It has not been entirely clear whether β₂AR-G_i coupling serves to substantially modulate G_s-mediated bronchodilatation or whether it is more relevant to non–cAMP-dependent signaling, such as p44/p42 MAP kinase activation. We show here that inhibition of β₂AR-G_i coupling via the minigene markedly enhanced β₂AR-mediated ASM relaxation. This is consistent with the concept that even with normal levels of G_i, β₂AR actively couples to G_i, which has important physiologic consequences. The results with the G_αi2-OE mice revealed that increased G_i expression was associated with increased β₂AR-G_i signaling, which competed with stimulation of β₂AR-mediated relaxation via G_s. This indicates that increasing expression levels by pathologic processes, or by pharmacologic means, will have an impact on the airway relaxation response. Several downstream signaling elements potentially related to β₂AR signaling were regulated by overexpression of G_αi2 or expression of the G_αi2-IP minigene. However, these changes were inversely related to the observed gain, or loss, of β₂AR function. For example, GRK2 (which phosphorylates and desensitizes β₂AR) was downregulated in G_αi2-OE smooth muscle. However, if anything this would cause an increase in β₂AR function, rather than the decreased function that was observed in these mice. Similarly, p44/p42 MAP kinase phosphorylates GRK2 and impairs its function (24). The former kinase was markedly activated in the G_αi2-OE smooth muscle cells, but β₂AR function was impaired, as opposed to enhanced, in these mice. G_αs was decreased approximately 30% in both G_αi2-OE and G_αi2-IP cells, and thus this change is unlikely to represent a major factor given the opposite β₂AR phenotypes in the 2 mice. Similarly, PKCε and PKCδ were downregulated in both mice.

The current work shows a complex, multifactorial interplay between G_i expression and contraction and relaxation signaling via G protein–coupled receptors in the airway (Figure 8). From the standpoint of the pathogenesis of bronchospasm, the increase in G_αi observed in asthmatic-like airways (3–6) appears to serve a protective effect: the increase acts to attenuate bronchial hyperresponsiveness from G_q-mediated receptor signaling. Given that virtually all bronchospastic mediators act via G_q-coupled receptors (e.g., M₃-muscarinic, LT1-cysteinyl leukotriene, and H1-histamine), it would appear that not only is the increase in G_i beneficial in asthma, attempts to adjust these levels at or below normal will increase bronchial hyperreactivity and exacerbate asthma. On the other hand, the increase in G_αi impaired β-agonist bronchodilatation. This appears to be due to an enhancement of β₂AR-G_αi coupling, which opposes β₂AR-G_s functional stimulation of adenylyl cyclase. Furthermore, decreasing G_i function acts to increase β₂AR-mediated bronchodilation, as would be expected based on this mechanism. Therefore, in the absence of β-agonist treatment the increase in G_αi in asthma acts to attenuate bronchospasm, but also limits the effectiveness of β-agonists in relieving bronchoconstriction. This multipronged set of regulatory events may indicate that pharmacologic or genetic methods to modulate G_i could have limited clinical utility. However, there may be opportunities to exploit these relationships by manipulating other downstream effects of altered G_i expression. For example, if decreasing G_i expression and function could be accomplished while also inhibiting PKCα, then both the relaxation and the constrictive pathways would be affected in a manner that would be expected to be clinically beneficial for hyperresponsiveness and β-agonist efficacy (Figure 8).

Figure 8

Multiple roles of G_i in regulating contraction and relaxation signaling in ASM. Increased (green arrows) or decreased (red arrows) expression or function of given parameters are indicated. Dotted lines denote pathways and critical mechanisms of action from increased G_αi; dashed lines denote pathways and critical mechanisms of action from decreased G_αi. AC, adenylyl cyclase; M₃R, M₃-muscarinic receptor.

The authors thank Judy Cain, Cheryl Theiss, and Mary Rose Schwarb for technical assistance and Esther Moses for manuscript preparation. The authors were supported by NIH grants HL045967, HL071609, and HL065899 (to S.B. Liggett), HL072068 (to D.W. McGraw), and AI45898 and HL076383 (to M.E. Rothenberg).

Address correspondence to: Stephen B. Liggett, 20 Penn Street, HSF-II, Room S-114, Baltimore, Maryland 21201-1075, USA. Phone: (410) 706-6256; Fax: (410) 706-6262; E-mail: [email protected].

Nonstandard abbreviations used: AR, adrenergic receptor; ASM, airway smooth muscle; G_αi2-IP, G_αi2 inhibitory peptide expression; G_αi2-OE, G_αi2 overexpression; GRK2, G protein–coupled receptor kinase 2; NTG, nontransgenic littermate; PLC, phospholipase C; PTX, pertussis toxin; SMP8, smooth muscle actin promoter 8.

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

Reference information: J. Clin. Invest.117:1391–1398 (2007). doi:10.1172/JCI30489

1. Green, S.A., and 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. Billington, C.K., Penn, R.B. 2003. Signaling and regulation of G protein-coupled receptors in airway smooth muscle. Respir. Res. 4:2-24.
   View this article via: CrossRef PubMed Google Scholar
3. Hakonarson, H., Herrick, D.J., Grunstein, M.M. 1995. Mechanism of impaired beta-adrenoceptor responsiveness in atopic sensitized airway smooth muscle. Am. J. Physiol. 269:L645-L652.
   View this article via: PubMed Google Scholar
4. Hakonarson, H., Herrick, D.J., Serrano, P.G., Grunstein, M.M. 1996. Mechanism of cytokine-induced modulation of β-adrenoceptor responsiveness in airway smooth muscle. J. Clin. Invest. 97:2593-2600.
   View this article via: JCI PubMed Google Scholar
5. Grunstein, M.M., Hakonarson, H., Maskeri, N., Chuang, S. 2000. Autocrine cytokine signaling mediates effects of rhinovirus on airway responsiveness. Am. J. Physiol. Lung Cell. Mol. Physiol. 278:L1146-L1153.
   View this article via: PubMed Google Scholar
6. Hakonarson, H., et al. 1998. Mechanism of rhinovirus-induced changes in airway smooth muscle responsiveness. J. Clin. Invest. 102:1732-1741.
   View this article via: JCI PubMed Google Scholar
7. 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: CrossRef PubMed Google Scholar
8. McGraw, D.W., Almoosa, K.F., Paul, R.J., Kobilka, B.K., Liggett, S.B. 2003. Antithetic regulation by β-adrenergic receptors of G_q receptor signaling via phospholipase C underlies the airway β-agonist paradox. J. Clin. Invest. 112:619-626.
   View this article via: JCI PubMed Google Scholar
9. Fulkerson, P.C., Fischetti, C.A., Hassman, L.M., Nikolaidis, N.M., Rothenberg, M.E. 2006. Persistent effects induced by IL-13 in the lung. Am. J. Respir. Cell Mol. Biol. 35:337-346.
   View this article via: CrossRef PubMed Google Scholar
10. Zhu, Z., et al. 1999. Pulmonary expression of interleukin-13 causes inflammation, mucus hypersecretion, subepithelial fibrosis, physiologic abnormalities, and eotaxin production. J. Clin. Invest. 103:779-788.
    View this article via: JCI PubMed Google Scholar
11. McGraw, D.W., et al. 2006. Transcriptional response to persistent β2-adrenergic receptor signaling reveals regulation of phospholamban which alters airway contractility. Physiol. Genomics. 27:171-177.
    View this article via: CrossRef PubMed Google Scholar
12. Gokina, N.I., Knot, H.J., Nelson, M.T., Osol, G. 1999. Increased Ca²⁺ sensitivity as a key mechanism of PKC-induced constriction in pressurized cerebral arteries. Am. J. Physiol. 277:H1178-H1188.
    View this article via: PubMed Google Scholar
13. Hill, M.A., Falcone, J.C., Meininger, G.A. 1990. Evidence for protein kinase C involvement in arteriolar myogenic reactivity. Am. J. Physiol. 259:H1586-H1594.
    View this article via: PubMed Google Scholar
14. Lee, M.W., Severson, D.L. 1994. Signal trans­duction in vascular smooth muscle: diacylglycerol second messengers and PKC action. Am. J. Physiol. 267:C659-C678.
    View this article via: PubMed Google Scholar
15. Salamanca, D.A., Khalil, R.A. 2005. Protein kinase C isoforms as specific targets for modulation of vascular smooth muscle function in hypertension. Biochem. Pharmacol. 70:1537-1547.
    View this article via: CrossRef PubMed Google Scholar
16. Struckmann, N., et al. 2003. Role of muscarinic receptor subtypes in the constriction of peripheral airways: studies on receptor-deficient mice. Mol. Pharmacol. 64:1444-1451.
    View this article via: CrossRef PubMed Google Scholar
17. Kotlikoff, M.I., Kume, H., Tomasic, M. 1992. Muscarinic regulation of membrane ion channels in airway smooth muscle cells. Biochem. Pharmacol. 43:5-10.
    View this article via: CrossRef PubMed Google Scholar
18. Dorn (2nd), G.W., Tepe, N.M., Wu, G., Yatani, A., Liggett, S.B. 2000. Mechanisms of impaired β-adrenergic receptor signaling in G_αq-mediated cardiac hypertrophy and ventricular dysfunction. Mol. Pharmacol. 57:278-287.
    View this article via: PubMed Google Scholar
19. Perez, J.M., et al. 2003. β₁-adrenergic receptor polymorphisms confer differential function and predisposition to heart failure. Nat. Med. 9:1300-1305.
    View this article via: CrossRef PubMed Google Scholar
20. Akhter, S.A., et al. 1997. Restoration of β-adrenergic signaling in failing cardiac ventricular myocytes via adenoviral-mediated gene transfer. Proc. Natl. Acad. Sci. U. S. A. 94:12100-12105.
    View this article via: CrossRef PubMed Google Scholar
21. Bohm, M., et al. 1990. Increase of Gi alpha in human hearts with dilated but not ischemic cardiomyopathy. Circulation. 82:1249-1265.
    View this article via: PubMed Google Scholar
22. Ebina, M., Takahashi, T., Chiba, T., Motomiya, M. 1993. Cellular hypertrophy and hyperplasia of airway smooth muscles underlying bronchial asthma. A 3-D morphometric study. Am. Rev. Respir. Dis. 148:720-726.
    View this article via: PubMed Google Scholar
23. Foerster, K., et al. 2003. Cardioprotection specific for the G protein Gi2 in chronic adrenergic signaling through beta 2-adrenoceptors. Proc. Natl. Acad. Sci. U. S. A. 100:14475-14480.
    View this article via: CrossRef PubMed Google Scholar
24. Pitcher, J.A., et al. 1999. Feedback inhibition of G protein-coupled receptor kinase 2 (GRK2) activity by extracellular signal-regulated kinases. J. Biol. Chem. 274:34531-34534.
    View this article via: CrossRef PubMed Google Scholar
25. 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: CrossRef PubMed Google Scholar
26. McGraw, D.W., et al. 2006. Airway smooth muscle prostaglandin-EP₁ receptors directly modulate β₂-adrenergic receptors within a unique heterodimeric complex. J. Clin. Invest. 116:1400-1409.
    View this article via: JCI PubMed Google Scholar
27. Small, K.M., et al. 2006. Alpha2A- and alpha2C-adrenergic receptors form homo- and heterodimers: the heterodimeric state impairs agonist-promoted GRK phosphorylation and beta-arrestin recruitment. Biochemistry. 45:4760-4767.
    View this article via: PubMed CrossRef Google Scholar