Involvement of Allosteric Effect and KCa Channels in Crosstalk between β2-Adrenergic and Muscarinic M2 Receptors in Airway Smooth Muscle

To advance the development of bronchodilators for asthma and chronic obstructive pulmonary disease (COPD), this study was designed to investigate the mechanism of functional antagonism between β2-adrenergic and muscarinic M2 receptors, focusing on allosteric effects and G proteins/ion channels coupling. Muscarinic receptor antagonists (tiotropium, glycopyrronium, atropine) synergistically enhanced the relaxant effects of β2-adrenergic receptor agonists (procaterol, salbutamol, formoterol) in guinea pig trachealis. This crosstalk was inhibited by iberitoxin, a large-conductance Ca2+-activated K+ (KCa) channel inhibitor, whereas it was increased by verapamil, a L-type voltage-dependent Ca2+ (VDC) channel inhibitor; additionally, it was enhanced after tissues were incubated with pertussis or cholera toxin. This synergism converges in the G proteins (Gi, Gs)/KCa channel/VDC channel linkages. Muscarinic receptor antagonists competitively suppressed, whereas, β2-adrenergic receptor agonists noncompetitively suppressed muscarinic contraction. In concentration-inhibition curves for β2-adrenergic receptor agonists with muscarinic receptor antagonists, EC50 was markedly decreased, and maximal inhibition was markedly increased. Hence, muscarinic receptor antagonists do not bind to allosteric sites on muscarinic receptors. β2-Adrenergic receptor agonists bind to allosteric sites on these receptors; their intrinsic efficacy is attenuated by allosteric modulation (partial agonism). Muscarinic receptor antagonists enhance affinity and efficacy of β2-adrenergic action via allosteric sites in β2-adrenergic receptors (synergism). In conclusion, KCa channels and allosterism may be novel targets of bronchodilator therapy for diseases such as asthma and COPD.


Role of Orthosteric Sites in the Effects of Muscarinic Receptor Antagonists
When atropine, tiotropium and glycopyrronium (0.0003-10 µM) were cumulatively applied to the tissues pre-contracted with MCh (1 and 10 µM), these agents inhibited MCh-induced contraction in a concentration-dependent manner, and complete inhibition was observed under these experimental conditions (Table 1). When the concentration of MCh was increased to 10 µM, these agents also caused a concentration-dependent inhibition, and complete relaxation was also observed at 0.3 µM of each agent ( Figure 3A). Moreover, the complete relaxation was not attenuated even when the concentration of these agents were increased to more than 0.3 µM ( Figure 3A). The concentration-inhibition curves for these agents against MCh (1 and 10 µM), the values of EC 50 and the maximal effect of these curves for these agents under each experimental condition are summarized in Table 1. On the other hand, when MCh was cumulatively applied to the tissues, the value of EC 50 was 0.6 ± 0.3 µM [95% CI: 0.26-0.68] (n = 6), and maximal response was observed at 100 µM ( Figure 3B). In the concentration-response curves for MCh in the presence of atropine, tiotropium and glycopyrronium at each 10 nM, values of EC 50 increased to 2.8 ± 0.9 [95% CI: 1.86-3.75] (n = 6), 3.9 ± 1.2 [95% CI: 2.64-5.16] (n = 6), 2.3 ± 0.7 µM [95% CI: 1.57-3.04] (n = 6), respectively (not significant), and the maximal contraction was not diminished under this experimental condition ( Figure 3B).

Role of Orthosteric Sites in the Effects of Muscarinic Receptor Antagonists
When atropine, tiotropium and glycopyrronium (0.0003-10 μM) were cumulatively applied to the tissues pre-contracted with MCh (1 and 10 μM), these agents inhibited MCh-induced contraction in a concentration-dependent manner, and complete inhibition was observed under these experimental conditions (Table 1). When the concentration of MCh was increased to 10 μM, these agents also caused a concentration-dependent inhibition, and complete relaxation was also observed at 0.3 μM of each agent ( Figure 3A). Moreover, the complete relaxation was not attenuated even when the concentration of these agents were increased to more than 0.3 μM ( Figure 3A). The concentrationinhibition curves for these agents against MCh (1 and 10 μM), the values of EC50 and the maximal effect of these curves for these agents under each experimental condition are summarized in Table 1. On the other hand, when MCh was cumulatively applied to the tissues, the value of EC50 was 0.6 ± 0.3 μM [95% CI: 0.26-0.68] (n = 6), and maximal response was observed at 100 μM ( Figure 3B). In the concentration-response curves for MCh in the presence of atropine, tiotropium and glycopyrronium at each 10 nM, values of EC50 increased to 2.8 ± 0.9 [95% CI: 1.86-3.75] (n = 6), 3.9 ± 1.2 [95% CI: 2.64-5.16] (n = 6), 2.3 ± 0.7 μM [95% CI: 1.57-3.04] (n = 6), respectively (not significant), and the maximal contraction was not diminished under this experimental condition ( Figure 3B).

Role of Allosteric Sites in the Effects of β 2 -Adrenergic Receptor Agonists
When procaterol and salbutamol were cumulatively applied to the tissues pre-contracted with MCh (1 and 10 µM), the maximal inhibition was observed for each experimental condition; the inhibitory effects of these agents were attenuated in a concentration-dependent manner at higher concentrations than the concentrations that produce the maximal effects ( Figure 4A,B). Similar results were observed in the cumulative application of formoterol to MCh (10 µM)-induced contraction ( Figure 5D). Procaterol caused almost complete relaxation (100% inhibition) of MCh (1 µM)-induced contraction, whereas, complete inhibition did not occur in other experimental conditions ( Figure 4A,B and Figure 5D). On the other hand, when isoproterenol was cumulatively applied to the tissues pre-contracted with MCh (1 and 10 µM), it caused complete inhibition against MCh-induced conditions; moreover, the inhibitory effects of isoproterenol were not attenuated at the higher concentrations that produce the maximal effects, which is different from procaterol, salbutamol and formoterol ( Figures 4C and 5D). The concentration-inhibition curves for these agents against MCh (1 and 10 µM)-induced contraction, the values of EC 50 and the maximal inhibition are summarized in Table 2.

Role of Allosteric Sites in the Effects of β2-Adrenergic Receptor Agonists
When procaterol and salbutamol were cumulatively applied to the tissues pre-contracted with MCh (1 and 10 μM), the maximal inhibition was observed for each experimental condition; the inhibitory effects of these agents were attenuated in a concentration-dependent manner at higher concentrations than the concentrations that produce the maximal effects ( Figure 4A,B). Similar results were observed in the cumulative application of formoterol to MCh (10 μM)-induced contraction ( Figure 5D). Procaterol caused almost complete relaxation (100% inhibition) of MCh (1 μM)-induced contraction, whereas, complete inhibition did not occur in other experimental conditions ( Figures  4A,B and 5D). On the other hand, when isoproterenol was cumulatively applied to the tissues precontracted with MCh (1 and 10 μM), it caused complete inhibition against MCh-induced conditions; moreover, the inhibitory effects of isoproterenol were not attenuated at the higher concentrations that produce the maximal effects, which is different from procaterol, salbutamol and formoterol ( Figures  4C and 5D). The concentration-inhibition curves for these agents against MCh (1 and 10 μM)-induced contraction, the values of EC50 and the maximal inhibition are summarized in Table 2.

Role of Allosteric Effects in the Synergistic Effects of β2-Adrenergic Receptor Agonists with Muscarinic Receptor Antagonists
When tiotropium, atropine, and gltopyrronium (each 1 nM) were exposed to the tissues precontracted with MCh (10 μM) for 120 min, the maximal inhibition of tiotropium was 8.

Possible Mechanisms of cAMP-Independent Processes in the Synergistic Effect
Forskolin and db-cAMP, which are independent of β 2 -adrenergic receptors, caused a concentration-dependent suppression of MCh (10 µM)-induced contraction, and complete relaxation (100% inhibition) was observed for each experimental condition. The maximal effects of these agents were not attenuated at higher concentrations than those concentrations than produce the maximal effects in the absence and presence of tiotropium, unlike β 2 -adrenerigic receptor agonists. In the concentration-inhibition curves for forskolin and db-cAMP, EC 50 was not significantly reduced in the present of tiotropium (1 nM) ( Figure 6A Figure 6B), respectively.

Possible Mechanisms of cAMP-Independent Processes in the Synergistic Effect
Forskolin and db-cAMP, which are independent of β2-adrenergic receptors, caused a concentrationdependent suppression of MCh (10 μM)-induced contraction, and complete relaxation (100% inhibition) was observed for each experimental condition. The maximal effects of these agents were not attenuated at higher concentrations than those concentrations than produce the maximal effects in the absence and presence of tiotropium, unlike β2-adrenerigic receptor agonists. In the concentrationinhibition curves for forskolin and db-cAMP, EC50 was not significantly reduced in the present of tiotropium (1 nM) ( Figure 6A Figure 5A). *: p < 0.05.

Possible Involvement of Interaction between Muscarinic Receptor Antagonists and Allosteric Sites on β2-Adrenergic Receptors
When the concentration of MCh was decreased from 10 μM to 3 μM, MCh -induced contraction was attenuated by approximate 10% (Figure 3B). The effect of MCh (3 μM) was roughly equivalent to that of MCh (10 μM) with tiotropium (1 nM). In concentration-inhibition curves for procaterol  Figure 5A). *: p < 0.05.

Discussion
This study used physiological methods to demonstrate that G proteins/K Ca channel/VDC channel linkages are an essential process in the synergistic effects induced by crosstalk between β 2 -adrenergic and muscarinic M 2 receptors. Furthermore, allosteric effects play a functionally important role in the alteration of the response to β 2 -adrenergic receptor agonists and muscarinic receptor antagonists via crosstalk between these two GPCRs, i.e., (1) reduced responsiveness to β 2 -adrenergic receptor agonists (partial agonism); and (2) enhanced responsiveness to the combination of β 2 -adrenergic receptor agonists with muscarinic receptor antagonists (synergism).
The combination of β 2 -adrenergic receptor agonists (procaterol, salbutamol) with a muscarinic receptor antagonist (tiotropium) synergistically enhanced the relaxant effects against muscarinic contraction (Figure 1), consistent with previous reports using other β 2 -adrenergic receptor agonists (indacaterol, formoterol) and muscarinic receptor antagonists (glycopyrronium, aclidinium) [4,11,24,33]. Moreover, the combination of atropine with procaterol and the combination of glycopyrronium with formoterol produced synergistic effects ( Figure 5C,D). Therefore, this phenomenon is probably universal, not specific to an agent related to these two receptors.
In airway smooth muscle tone, Ca 2+ signaling contributes to the intracellular processes in the downstream of these two GPCRs [6,8,34,35]. Suppression of K Ca channel activity enhances muscarinic contraction with an increase in Ca 2+ influx pass through VDC channels [17]. Ca 2+ dynamics due to K Ca channels [4,11] and Ca 2+ sensitization due to PKC contributes to the synergistic effects between β 2 -adreneric receptor agonists and muscarinic receptor antagonists [24]. In this study, the effect of VDC channels was analyzed and whether the Ca 2+ dynamics due to K Ca channels is involved in this synergism was investigated. K Ca channel activity is associated with the functional antagonism between β 2 -adrenergic and muscarinic receptors [1,2,4]. K Ca channels are activated not only by protein kinase A (PKA) but also by G s , which is coupled to β 2 -adrenergic receptors [36,37]. In contrast, these channels are inhibited by G i , which is coupled to muscarinic M 2 receptors [1,4,38]. The synergistic effects of β 2 -adrenergic receptor agonists with muscarinic receptor antagonists were markedly augmented when G s was activated by cholera toxin and G i is inhibited by pertussis toxin (Figure 2A,D). These synergistic effects may be caused by β 2 -adreneric receptor agonists/G s protein/K Ca channels' stimulatory coupling and muscarinic M 2 receptors/G i protein/K Ca channels' inhibitory coupling. Moreover, the synergistic effects were inhibited by IbTX, an inhibitor of K Ca channels, in a concentration-dependent manner (Figure 2A,B), indicating that activation of K Ca channels is involved in this phenomenon. In contrast, the synergistic effects were augmented by verapamil, an inhibitor of VDC channels, in a concentration-dependent manner, indicating that inactivation of VDC channels is involved in this phenomenon (Figure 2A,C). IbTX and verapamil were antagonistic to each other in the regulation of synergistic effects ( Figure 2B,C), indicating that coupling between K Ca and VDC channels is involved in this phenomenon. IbTX inhibited a reduction in intracellular Ca 2+ concentration induced by procaterol with atropine, in contrast, verapamil reversed this effect of IbTX ( Figure 2E,F), indicating that suppression of K Ca channels increases the concentration of intracellular Ca 2+ via activation of VDC channels (Ca 2+ dynamics due to K Ca channels/VDC channels coupling). Activation of K Ca channels causes membrane hyperpolarization, which deactivates VDC channels. This channel coupling enhances relaxation of airway smooth muscle [2,17]. Therefore, Ca 2+ dynamics via these G proteins/K Ca channel/VDC channel processes may contribute to the synergistic action (Figure 2). The functional antagonism to β 2 -aderenergic receptors also may be mediated by muscarinic M 2 receptors [39][40][41].
Allosteric affinity and efficacy modulation to muscarinic receptors and β 2 -adenergic receptors was examined using the concentration-inhibition curves for muscarinic receptor antagonists and β 2 -adrenergic receptor agonists against muscarinic contraction (Figures 3 and 4) [4,42,43], and the intrinsic efficacy of muscarinic receptor antagonists and β 2 -adrenergic receptor agonists was determined under these experimental conditions (Tables 1 and 2). Allosteric modulators alter association or dissociation rates of orthosteric ligands (affinity modulation) via acting allosteric sites. Allosteric effects operate intracellular responses and alter the signaling capacity (intrinsic efficacy) of orthosteric ligands (efficacy modulation) [30][31][32]. When agents act on their specific GPCRs at orthosteric sites without allosteric sites, the maximal response to agents as orthosteric ligands is not attenuated (Figure 7A,B) because an allosteric modulation does not occur (full agonism). In the functional antagonism (crosstalk) between β 2 -adrenergic and muscarinic receptors, when agents act upon their specific GPCRs at both orthosteric and allosteric sites, partial agonism of β 2 -adrenergic receptors is generated because response to agents as orthosteric ligands is altered by allosteric efficacy modulation ( Figure 7A,C) [4,43].
To determine the involvement of orthosteric and allosteric sites in the relaxant action of muscarinic receptor antagonists against muscarinic contraction, affinity and efficacy modulation of these antagonists were examined using concentration-inhibition curves for these antagonists on a muscarinic receptor agonist. Muscarinic receptor antagonists, including atropine, tiotropium, and glycopyrroium, caused the maximal effect (complete inhibition) of muscarinic contraction, and the maximal effect was not attenuated even when higher concentrations of these antagonists were applied ( Figure 3A, Table 1). Moreover, in the concentration-response curves for MCh with these antagonists, the values of EC 50 were markedly increased without reducing the maximal response ( Figure 3B). Since muscarinic receptor antagonists competitively inhibit the effects of a ligand on the receptor, these agents do not alter the signal capacity of orthosteric ligands via binding to allosteric sites [42,43]. Hence, muscarinic receptor antagonists do not cause efficacy modulation against orthosteric effects in muscarinic action, and allosteric sites on muscarinic receptors are not associated with muscarinic antagonism induced by these agents ( Figure 7B) [42,43].
To determine the involvement of orthosteric and allosteric sites in the relaxant action of β 2 -adreneric receptor agonists against muscarinic contraction, affinity and efficacy modulation of these agonists were examined using concentration-inhibition curves for these agonists on a muscarinic receptor agonist. Isoproterenol completely antagonized muscarinic contraction, and the complete inhibition induced by isoproterenol was not attenuated ( Figure 4C) at higher concentrations than those that produce maximal relaxation. Isoproterenol acts on orthosteric sites in β 2 -adrenergic receptors and does not act on allosteric sites in these receptors ( Figure 7A,C). Hence, isoproterenol behaves as a full agonist. In contrast, procaterol, salbutamol and formoterol incompletely antagonize muscarinic contraction ( Figure 4A,B and Figure 5D), similar to indacaterol and salmeterol [4,11,44,45]. A lack of complete inhibition by β 2 -adrenergic receptor agonists indicates a decrease in the signal capacity induced by efficacy modulation (an inhibition in response to orthosteric site via allosterism). At higher concentrations than those that produce the maximal relaxation, procaterol, salbutamol and formoterol caused a concentration-dependent contraction, unlike isoproterenol, suggesting that the primary effects (signal capacity) of these agents via orthosteric sites are reduced by efficacy modulation via binding to allosteric sites ( Figure 7A,C). Therefore, in the β 2 -adrenergic receptor agonists, except isoproterenol, intrinsic efficacy is attenuated by stimulating allosteric sites, indicating that these agonists behave as allosteric modulators against β 2 -adrenergic receptors (partial agonists) ( Figure 7A,B, Table 2) [4,11,44,45]. To determine the involvement of orthosteric and allosteric sites in the relaxant action of β2adreneric receptor agonists against muscarinic contraction, affinity and efficacy modulation of these agonists were examined using concentration-inhibition curves for these agonists on a muscarinic receptor agonist. Isoproterenol completely antagonized muscarinic contraction, and the complete inhibition induced by isoproterenol was not attenuated ( Figure 4C) at higher concentrations than those that produce maximal relaxation. Isoproterenol acts on orthosteric sites in β2-adrenergic receptors and does not act on allosteric sites in these receptors ( Figure 7A,C). Hence, isoproterenol behaves as a full agonist. In contrast, procaterol, salbutamol and formoterol incompletely antagonize muscarinic contraction (Figures 4A,B and 5D), similar to indacaterol and salmeterol [4,11,44,45]. A lack of complete inhibition by β2-adrenergic receptor agonists indicates a decrease in the signal capacity induced by efficacy modulation (an inhibition in response to orthosteric site via allosterism). At higher concentrations than those that produce the maximal relaxation, procaterol, salbutamol and formoterol caused a concentration-dependent contraction, unlike isoproterenol, suggesting that the primary effects (signal capacity) of these agents via orthosteric sites are reduced by efficacy Figure 7. Involvement of allosteric effects (affinity and efficacy modulation) in response to β 2 -adrenergic receptor agonists and muscarinic receptor antagonists against muscarinic contraction. (A) A schema of affinity and efficacy modulation via allosteric sites shown in concentration-response curves for β 2 -adrenergic receptor agonists against muscarinic contraction; (B) Muscarinic receptor antagonists inhibit muscarinic action via acting on orthosteric sites on muscarinic receptors, independent of allosteric sites; (C) Full β 2 -adrenergic receptor agonists do not act on allosteric sites. In contrast, partial β 2 -adrenergic receptor agonists act not only on orthosteric sites, but also on allosteric sites on these receptors, and these agonists reduce the signal capacity (intrinsic efficacy) of an orthosteric ligand via efficacy modulation induced by operating upon allosteric sites; (D) Muscarinic receptor antagonists act on allosteric sites of β 2 -adrenergic receptors, and as a result affinity and efficacy of To determine the involvement of orthosteric and allosteric sites in this synergistic effect between β 2 -adreneric receptor agonists and muscarinic receptor antagonists, affinity and efficacy modulation of β 2 -adreneric receptor agonists with muscarinic receptor antagonists were examined using concentration-inhibition curves for these agonists on a muscarinic receptor agonist. Allosteric modulation occurs in the interaction mediated by ligands related to GPCRs [30][31][32]. When agents act not only on their specific GPCGs at orthosteric sites as othosteric ligands, but also on other GPCRs at allosteric sites, synergism may be generated via affinity and efficacy modulation due to allosterism [4,15]. The combination of β 2 -adrenergic receptor agonists with muscarinic receptor antagonists may synergistically inhibit muscarinic contraction via allosterism (affinity and efficacy modulation). In the concentration-inhibition curves for procaterol, salbutamol and formoterol with muscarinic receptor antagonists (tiotropium, atropine, glycopyrronuium), the values of EC 50 were markedly decreased; the maximal effects of these β 2 -adrenergic receptor agonists were markedly augmented in each experimental condition ( Figure 5). Muscarinic receptor antagonists may enhance affinity and efficacy of β 2 -adrenergic receptor agonists, leading to synergistic effects on muscarinic contraction ( Figure 7A,D). However, this synergistic effect may be due to inhibition of muscarinic contraction induced by muscarinic receptor antagonists. To determine whether this phenomenon is involved in the synergistic effect, affinity and efficacy modulation of β 2 -adrenergic receptor agonists were examined using concentration-inhibition curves for these agents without muscarinic receptor antagonists under the condition of a lower concentration of MCh. MCh (10 µM)-induced contraction with 1 nM of atropine, tiotropium and glycopyrronium was roughly equivalent to MCh (3 µM)-induced contraction ( Figure 3A,B). The concentration-inhibition curve for procaterol against MCh (3 µM)-induced contraction was no different from the curve for procaterol with tiotropium (1 nM) against MCh (10 µM)-induced contraction ( Figure 6C). Therefore, this synergistic effect is created independently of the effects of muscarinic receptor antagonists on their receptors. Muscarinic receptor antagonists may operate not only upon orthosteric sites on muscarinic receptors, but also upon allosteric sites on β 2 -adrenergic receptors, and these antagonists enhance both affinity and efficacy to β 2 -adrenergic receptors; as a result, this synergism may be generated via crosstalk between the two GPCRs ( Figure 7D).
To determine whether the downstream of β 2 -adrenergic receptor is involved in the synergistic effect, similar experiments were carried out using cAMP-related agents independent β 2 -adrenergic receptors. Forskolin and N6-dibutylyl cyclic AMP (db-cAMP), which are unrelated to β 2 -adrenergic receptors, did not cause synergistic effects in the presence of tiotropium ( Figure 6A,B), which is different toβ 2 -adrenergic receptor agonists with muscarinic receptor antagonists ( Figure 5). In human bronchial smooth muscle, intracellular concentration of cAMP is associated with the synergism between β 2 -adrenergic receptor agonists and muscarinic receptor antagonists [46]. However, adenylyl cyclase/cAMP/PKA processes may not directly contribute to the synergistic effects between these two agents, because forskolin and db-cAMP caused a modest effect on this synergism ( Figure 6A,B). This result is similar to the phenomenon that G s is more potent in K Ca channel activation than PKA; K Ca channels are markedly activated by G s independent of PKA [37]. The β 2 -adrenergic receptor/G s pathway, and muscarinic M 2 receptor/G i pathway may be essential to the synergism between these two agents.

Tissue Preparation and Tension Records
The methods were similar to those described previously [2,47,48]. Male Hartley guinea pigs (250-350 g) were killed by injection of an overdose of anesthetics (150 mg/kg pentobarbital i.p.), and tracheas were excised from the animals. The tracheal ring with three segments of cartilage was removed and placed horizontally in a 2-mL organ bath for isometric tension recording. Tracheal smooth muscle strips were incubated with 15 µM Fura-2/AM for about 2 h at room temperature (22-24 • C). The non-cytotoxic detergent Pluronic F-127 (0.01% w/v) was added to increase the solubility of Fura-2/AM. The intensities of excitation fluorescence at 340 nm and 380 nm were measured after background subtraction by exposure of the mucosal side to excitation light. The ratio of F 340 to F 380 (F 340 /F 380 ) was used as a relative indicator of intracellular Ca 2+ concentration [6][7][8]. Strips were perfused with the normal bathing solution at a constant flow rate of 3.0 mL/min throughout the experiments. The normal bathing solution was composed of NaCl (137 mM), KHCO 3 (5.9 mM), CaCl 2 (2.4 mM), MgCl 2 (1.2 mM), and glucose (11.8 mM), bubbled with a mixture of 99% O 2 and 1% CO 2 (pH 7.4). After equilibrating the preparation in normal bathing solution, the experiments were started, and the temperature of the organ bath was maintained at 37 • C. The resting tone was abolished by the addition to indomethacin (1 µM) throughout the experiments. All animal procedures in this study were approved by the Animal Care and Use Committee, Kindai University Faculty of Medicine (identification code: KAME-22-007, 24 February 2011).

Analysis of Synergistic Effect
The synergistic effects of the combination of procaterol or salbutamol with tiotropium were evaluated using the Bliss independence (BI) theory. This model assumes that two or more agents act independently, with different modes and sites of action (Greco et al., 1995;Goldoni and Johansson, 2007), and is expressed by the following equation: E(x, y) = E(x) + E(y) − E(x) × E(y) where E is the fractional effect, and x and y are the doses of two compounds in a combination experiment. If the combined experimental value is higher than the expected value, the interaction is synergistic. If it is lower, the interaction is antagonistic [24,49,50].

Experimental Protocols
Concentration-response curves were plotted by the cumulative addition of agents at 10 min for each concentration. To examine the effect of inhibited (uncoupled to the muscarinic M 2 receptor) G i , the inhibitory G protein of adenylyl cyclase, via adenosine diphosphate (ADP) ribosylation of the α-subunit of G i protein, the strips of tracheal smooth muscle were incubated with 1 µg/mL petussis toxin (PTX) for 6 h and then the PTX was washed out. To examine the effects of activated G s , the stimulatory G protein of adenylyl cyclase coupled to β 2 -adrenergic receptors, via ADP ribosylation of the α-subunit of G s protein, the tissues were incubated with 2 µg/mL cholera toxin (CTX) for 6 h and then the CTX was washed out. The involvement of K Ca channels and VDC channels was examined by application of iberiotoxin (IbTX), a potent selective inhibitor of K Ca channels, and verapamil, a selective inhibitor of VDC channels. To examine allosteric effects (alteration of intrinsic efficacy of an orthosteric ligand: efficacy modulation) in the relaxant action of β 2 -adrenergic receptor agonists on muscarinic contraction were cumulatively applied to the tissues pre-contracted by methacholine (MCh, 1 and 10 µM). Moreover, to examine allosteric effects (affinity and efficacy modulation) in the relaxant action of the combination of β 2 -adrenergic receptor agonists with muscarinic receptors antagonists against muscarinic contraction, procaterol and salbutamol were cumulatively applied to the tissues pre-contracted by MCh (10 µM) in the absence and presence of muscarinic receptor antagonists (1 nM). The concentration-inhibition curves for β 2 -adrenergic receptor agonists were analyzed for each experimental condition. Time-matched control tissues were treated similarly to the test tissues, but exposed continuously to the normal bathing solution (sham incubation) instead of agents. Procaterol, salbutamol, and formoterol were used as β 2 -adrenergic receptor agonists. Atropine, tiotropium, and glycopyrronium were used as muscarinic receptor antagonists. Forskolin (a direct activator of adenylyl cyclase) and N6-dibutylyl cyclic AMP (db-cAMP, an analog of cAMP) were used as cAMP-related agents independent of β 2 -adrenergic receptors.

Statistical Analysis
All data are expressed as the mean ± standard deviation (SD) with 95% confidence interval (CI). The response to an agent is described as a percentage of the maximal response. The effect of relaxant agents on MCh-induced contraction is expressed as percent inhibition by taking the control response to MCh (1 or 10 µM) for each experimental condition as 100%. The response to MCh was described as the percent contraction by taking the maximal contraction under the control condition as 100%. Values of the concentration of agents that produced a 50% response (EC 50 ) of contraction were determined from the linear portion of each concentration-response curve. Statistical significance was assessed by unpaired Student's t-test, one-way analysis of variance and the Dunnett test. A probability below 0.05 (p < 0.05) was considered to be significant.

Conclusions
Combining of β 2 -adrenergic receptor agonists with muscarinic receptor antagonists causes a synergistic inhibition against muscarinic contraction of airway smooth muscle. The G proteins/K Ca channel/VDC channel linkage is a key molecule for this synergism, and allosteric GPCRs modulation induced by crosstalk between these two receptors plays a fundamental role in this phenomenon ( Figure 8). Observations obtained from guinea pig trachea may have limited clinical relevance. However, muscarinic and β 2 -adrenergic action is not so different, not only between guinea pig trachea [4,11,24] and human bronchus [46], but also between guinea pig [47] and human trachea [48]. K Ca channels are densely distributed on the airway smooth muscle in various mammals including humans; the electrical characteristics of these channels are no different between these species [51,52]. These results provide evidence that inhalation of a combined β 2 -adrenergic receptor agonist and a muscarinic antagonist causes greater bronchodilation than monotherapy in asthma and COPD [25,[27][28][29]. The K Ca channels/VDC channels coupling may also be responsible for β 2 -adrenergic desensitization [53] and airway remodeling [54]. Hence, elucidation of the mechanisms of this synergism (crosstalk) may contribute to the development of novel bronchodilators for asthma and COPD [4,11,55].

Statistical Analysis
All data are expressed as the mean ± standard deviation (SD) with 95% confidence interval (CI). The response to an agent is described as a percentage of the maximal response. The effect of relaxant agents on MCh-induced contraction is expressed as percent inhibition by taking the control response to MCh (1 or 10 μM) for each experimental condition as 100%. The response to MCh was described as the percent contraction by taking the maximal contraction under the control condition as 100%. Values of the concentration of agents that produced a 50% response (EC50) of contraction were determined from the linear portion of each concentration-response curve. Statistical significance was assessed by unpaired Student's t-test, one-way analysis of variance and the Dunnett test. A probability below 0.05 (p < 0.05) was considered to be significant.

Conclusions
Combining of β2-adrenergic receptor agonists with muscarinic receptor antagonists causes a synergistic inhibition against muscarinic contraction of airway smooth muscle. The G proteins/KCa channel/VDC channel linkage is a key molecule for this synergism, and allosteric GPCRs modulation induced by crosstalk between these two receptors plays a fundamental role in this phenomenon ( Figure 8). Observations obtained from guinea pig trachea may have limited clinical relevance. However, muscarinic and β2-adrenergic action is not so different, not only between guinea pig trachea [4,11,24] and human bronchus [46], but also between guinea pig [47] and human trachea [48]. KCa channels are densely distributed on the airway smooth muscle in various mammals including humans; the electrical characteristics of these channels are no different between these species [51,52]. These results provide evidence that inhalation of a combined β2-adrenergic receptor agonist and a muscarinic antagonist causes greater bronchodilation than monotherapy in asthma and COPD [25,[27][28][29]. The KCa channels/VDC channels coupling may also be responsible for β2-adrenergic desensitization [53] and airway remodeling [54]. Hence, elucidation of the mechanisms of this synergism (crosstalk) may contribute to the development of novel bronchodilators for asthma and COPD [4,11,55]. Possible clinical relevance of the synergistically relaxant effects between β2-adrenergic receptor agonists and muscarinic receptor antagonists. This synergism is caused by crosstalk based on allosteric GPCR modulation via G proteins (Gi, Gs)/KCa channel/VDC channel processes. Since this synergism causes greater bronchodilation, combining two agents may be beneficial to therapy for asthma and COPD. GPCR: G protein-coupled receptor, Gi: the inhibitory G protein of adenylyl cyclase coupled to M2 receptors, Gs: the stimulatory G protein of adenylyl cyclase coupled to β2-adrenergic receptors, KCa: large-conductance Ca 2+ -activated K + channels, VDC: L-type voltage-dependent Ca 2+ channels. Arrows: activation, dotted arrows: inhibition. Figure 8. Possible clinical relevance of the synergistically relaxant effects between β 2 -adrenergic receptor agonists and muscarinic receptor antagonists. This synergism is caused by crosstalk based on allosteric GPCR modulation via G proteins (G i , G s )/K Ca channel/VDC channel processes. Since this synergism causes greater bronchodilation, combining two agents may be beneficial to therapy for asthma and COPD. GPCR: G protein-coupled receptor, G i : the inhibitory G protein of adenylyl cyclase coupled to M 2 receptors, G s : the stimulatory G protein of adenylyl cyclase coupled to β 2 -adrenergic receptors, K Ca : large-conductance Ca 2+ -activated K + channels, VDC: L-type voltage-dependent Ca 2+ channels. Arrows: activation, dotted arrows: inhibition.