Discussion
The results of this study demonstrate that both glycopyrronium and indacaterol have the ability to induce potent, significant and long-lasting relaxation of both medium and small human isolated bronchi pre-contracted with acetylcholine. The co-administration of glycopyrronium and indacaterol produces a synergistic inhibition of ASM tone via modulating the cAMP-dependent pathway, especially when these drugs are administered at low concentrations. Intriguingly, when glycopyrronium and indacateriol were administered at low concentrations in our experimental setting, their ratio was consistent with that of the currently approved fixed dose combinations (FDCs), namely 15.6/27.5 μg in United States and 50/110 μg in European Union [
34,
35].
Overall, the functional data described in this study are consistent with those concerning the pharmacological characterisation of the interaction between aclidinium bromide and formoterol fumarate in human isolated bronchi [
6]. Nevertheless, the glycopyrronium/indacaterol combination produced a greater synergistic interaction in both medium bronchi and bronchioles when compared with that induced by the aclidinium bromide/formoterol fumarate combination. In addition, we must highlight that while formoterol fumarate administered alone did not completely relax small airways, indacaterol was able to abolish the contractile tone of PCLS preparations [
6]. As expected, the duration of the synergistic effect of glycopyrronium plus indacaterol was markedly longer than that elicited by combining aclidinium bromide with formoterol fumarate. In fact, a significant synergism between glycopyrronium and indacaterol was detectable for at least 9 h. On the contrary, combining aclidinium bromide with formoterol fumarate induced a greater post-administration bronchorelaxant peak, though the synergism was significant for only 6 h after the administration of the two drugs [
6]. In any case, the head-to-head comparison of glycopyrronium/indacaterol vs. aclidinium/formoterol combinations indicated that the overall extent of synergistic interaction and onset of action were similar for both these LABA/LAMA combinations, at least undergoing the experimental conditions set up in our
ex vivo model of human isolated bronchi. In fact, we must consider that a recent synthesis of the currently available clinical data suggested for a rank of effectiveness among the approved doses of LAMA/LABA FDCs, with glycopyrronium/indacaterol (15.6/27.5 mg and 50/110 mg) eliciting greater FEV
1 increase than aclidinium/formoterol (400/12 mg) in COPD patients, when compared with the respective monocomponents [
36]. However, this discrepancy may be related with the fact that glycopyrronium/indacaterol and aclidinium/formoterol combinations, administered at the currently approved doses, may be not delivered into the lung at isoeffective concentrations [
36].
When one agent interacts with its specific G protein-coupled receptor (GPCR), the effect of another agent on its GPCR may change, leading to a possible pharmacological interaction [
37,
38]. Since our study has provided evidences for a synergistic cross-talk between an anti-muscarinic agent and a β
2-adrenoceptor agonist undergoing cholinergic stimulation, we have investigated whether the inhibition of muscarinic GPCRs may transmit the signal to β
2-adrenoceptor GPCR also in course of histamine-induced bronchoconstriction. The bronchial contractile tone induced by histamine is mediated by both direct activation of histaminergic receptors expressed on human ASM, and facilitator effect of the acetylcholine release from parasympathetic nerve terminals [
38‐
40]. The findings of our study represent a significant step forward the study of Aizawa et al. [
40], by providing the evidence that the increase in isometric tension elicited by histamine is manly mediated by the resease of acetylcholine via the direct action of histamine on the vagus efferent nerve terminals, independently by neural conduction. In fact, neither TeTX nor epithelium altered the histamine-induced increase of acetylcholine release. Furthermore, neither glycopyrronium nor indacaterol were able to reduce the release of acetylcholine at control levels, even when these drugs were administered in combination. Therefore, the lack of synergism between glycopyrronium and indacaterol on the human ASM contractility histamine-mediated may be explained by the lack of a direct influence of either these drugs on the bronchial histaminergic pathway. Our results confirm that both anti-muscarinic agents and β
2-adrenoceptor agonists are less potent and effective in reducing the bronchial tone elicited by histamine when compared with their impact on the cholinergic tone [
19,
33,
41], and that no cross-talk exists between muscarinic and to β
2-adrenoceptor GPCRs in course of histaminergic stimulation.
Nevertheless, it has been recently reported that the bronchoprotection by a LABA may be synergistically enhanced by a LAMA, at least
in vivo in guinea-pigs [
42,
43]. Indeed, the results of these studies are in evident contrast with our findings, and probably any discrepancy may be related with the considerable differences between our methodological approach and that performed by Smit and colleagues [
42,
43]. However, we must highlight that the data we have presented here may be of great interest, because the responses of autonomic nervous system and ASM are specific for species and for tissues [
40], and results obtained from human isolated tissues have certainly a greater translational potential when compared with those obtained from animal models [
44].
In this study, we have also attempted to elucidate the mechanism(s) underlying the bronchorelaxant interaction between LABAs and LAMAs.
We assumed that the activation of β
2-adrenoceptors by indacaterol would have no effect on the release of acetylcholine from parasympathetic nerves, as previously documented by using human isolated trachea stimulated with the β
2-adrenoceptor agonist isoprenaline [
45]. On the contrary, our data showed that indacaterol was able to reduce the release of acetylcholine, a phenomenon that was dependent by the bronchial epithelium and reverted by blocking the IbTX-sensitive KCa
++ channels.
Since we have confirmed the neuronal origin of acetylcholine by inhibiting its release using TeTX, we can assume that in human bronchi indacaterol has a protective role against the neuronal release of acetylcholine. There is experimental documentation in laboratory animals that β
2-adrenoceptor stimulation may elicit a paradoxical facilitation of acetylcholine release from isolated trachea via the activation of a cAMP/cAMP-dependent protein kinase cascade [
45]. Our data indicate that this is not the case in human bronchial tissue.
Surprisingly, contrary to our assumption that an anti-muscarinic agent would facilitate neurogenic transmission due to inhibition of the pre-synaptic muscarinic M
2 autoreceptor, glycopyrronium inhibited the parasympathetic release of acetylcholine [
9]. So far, it is well known that anti-muscarinic agents are not specifically selective for the post-synaptic muscarinic M
3 receptor [
46]. In fact, LAMAs may bind to human muscarinic M
1-M
5 receptors in a concentration-dependent manner, although they dissociate more slowly from the muscarinic M
3 receptor than they do from the others [
47]. In particular, glycopyrronium showed no selectivity in its binding to the muscarinic M
1-M
3 receptors [
48]. However, a 3-5–fold higher affinity was observed for the muscarinic M
3 receptor compared to the muscarinic M
1 and M
2 receptors, and the Schild plot analysis demonstrated that glycopyrronium has a higher affinity for muscarinic M
1 and M
3 receptors compared to the muscarinic M
2 autoreceptor [
48]. Thus, at the level of post-ganglionic parasympathetic neurons, glycopyrronium prevalently inhibits the muscarinic M
1 receptor expressed on the body cell compared with the inhibition elicited on the muscarinic M
2 autoreceptor localised on the post-ganglionic fibre [
49]. Since the muscarinic M
1 receptor is facilitatory to nicotinic receptors and is involved in setting the efficacy of ganglionic transmission [
49,
50], the overall inhibitory effect of glycopyrronium leads to reduced parasympathetic transmission.
Analogous to indacaterol, modulation of the acetylcholine release by glycopyrronium was related with the integrity of the bronchial epithelium and the functionality of IbTX-sensitive KCa++ channels. Unfortunately, we did not detect any synergistic effect on the inhibitory release of parasympathetic acetylcholine when low concentrations of indacaterol and glycopyrronium were administered in combination.
The autonomic control of human ASM tone is primarily mediated by the release of acetylcholine from parasympathetic fibers [
51]. However, a non-neuronal cholinergic system exists at the level human airways [
52]. The synthesis, recycling, storage and release of non-neuronal acetylcholine is mediated by several mechanisms such as choline acetyltransferase (ChAT), ChAT-like enzymes, carnitine acetyltransferase (CarAT), high-affinity choline transporter (CHT1), vesicular acetylcholine transporter (VAChT) and OCT [
53,
54]. Thus, in addition to the parasympathetic release of acetylcholine, this transmitter may have a crucial local auto-/paracrine role in regulating several aspects on the innate mucosal defense mechanisms, including mucociliary clearance, regulation of macrophage function and modulation of sensory nerves [
55].
Since our findings showed that removal of the bronchial epithelium may influence the release of acetylcholine from human bronchi, we further investigated the role of indacaterol and glycopyrronium on the bronchial non-neuronal cholinergic system [
52,
55,
56]. Interestingly, although indacaterol and glycopyrronium alone did not modify the release of acetylcholine from primary human bronchial epithelial cells, a combination of these drugs inhibited the epithelial release of acetylcholine with the same extent to the inhibitory effect induced by the OCT inhibitor quinine.
Taken together, these evidences allow us ruling out a direct influence of synaptic postganglionic nerve endings in the synergistic interaction between glycopyrronium and indacaterol. Nevertheless, epithelium may be at least partially responsible for the synergistic effect of glycopyrronium/indacaterol, since the drug combination was effective in reducing endogenous and non-neurogenic release of acetylcholine from bronchial epithelial cells compared with either drug administered alone. This latter evidence may explain, to some extent, the relevant synergistic interaction between glycopyrronium and indacaterol in PCLS preparations, since at the level of human bronchioles the density of vagal innervation is insignificant or even absent, thus suggesting a role of the non-neuronal cholinergic system [
57,
58]. In fact, in peripheral airways, the muscarinic M
3 receptor is expressed and may be activated by acetylcholine released from epithelial cells that may express ChAT in response to inflammatory stimuli [
59].
In any case, the function of muscarinic receptors localized on human bronchial epithelium is still speculative. The activation of muscarinic M
1 may induce proliferation of isolated tracheal epithelial cells, and bronchial epithelial muscarinic M
3 receptor seems to mediate the release of diffusible factors, modulating contractility of underlying ASM [
60]. Furthermore, the role of muscarinic M
2 receptor remains unclear in bronchial epithelium [
61]. Therefore, although the nature of these factors remains unknown, we cannot exclude that muscarinic antagonists may reverse the G
i/KCa
++ channel inhibitory linkage induced by the activation of muscarinic M
2 receptor at the level of airway epithelial cells, concurring with the similar effect elicited by β
2-adrenoceptor agonists through intracellular cAMP elevation and reducing the release of acetylcholine [
37].
Since we have demonstrated that the synergism between a LABA and a LAMA cannot be adequately explained by the modulation of acetylcholine release, we investigated further mechanisms that might directly engage the ASM. Our data suggest that cAMP elevation induced in ASM by combining low doses of glycopyrronium plus indacaterol seems to be the main cause that explains the synergistic interaction between these bronchodilators. In fact, a noteworthy enhancement of cAMP levels was detected in human isolated bronchi treated with the glycopyrronium/indacaterol combination compared with isolated airways treated with the monocomponents. As expected, the concentrations of cAMP were not modulated by glycopyrronium, whereas indacaterol enhanced the cAMP levels by approximately three fold, independent by the presence of epithelium. On the other hand, the glycopyrronium/indacaterol combination elicited a noteworthy increase in cAMP concentrations up to approximately seven fold.
Furthermore, our results suggest that the activity of IbTX-sensitive KCa
++ channels is crucial for generating this significant cAMP elevation. Stimulation of the β
2-adrenoceptor activates the KCa
++ channels through both cAMP-dependent and -independent mechanisms, leading to hyperpolarisation of cell membrane and, consequently, to ASM relaxation [
62,
63]. To the best of our knowledge, we have demonstrated for the first time in human ASM that not only cAMP may influence the activity of KCa
++ channels, but also that the IbTX-sensitive KCa
++ channels themselves may modulate cAMP increase following the concomitant activation of the β
2-adrenoceptor and inhibition of muscarinic receptors.
Several pathways have been proposed to explain the intracellular cross-talk between β
2-adrenoceptors and muscarinic receptors at the level of the human ASM [
9]. In our study, the role of IbTX-sensitive KCa
++ channels seems to be predominant, since these channels may induce a direct reduction of ASM tone and, equally important, regulate the intracellular concentrations of cAMP [
64,
65]. Our data suggest that the functionality of IbTX-sensitive KCa
++ channels is crucial for allowing the synergistic interaction between a LABA and a LAMA, leading to a sustained and intense bronchorelaxant effect via cAMP elevation.
Since the bronchial epithelium expresses both β
2-adrenoceptors and muscarinic receptors [
59], we also investigated whether a LABA/LAMA combination might have a role in the synthesis of cAMP in primary human bronchial epithelial cells as well. Intriguingly, glycopyrronium and indacaterol administered alone at low concentrations did not modify the basal level of cAMP, whereas the combination of these drugs increased the cAMP levels in bronchial epithelial cells. This finding suggests that the bronchial epithelium may also contribute to the modulation of cAMP levels in human bronchi following the stimulation of β
2-adrenoceptors and inhibition of muscarinic receptors, thus supporting the synergistic interaction elicited by the glycopyrronium/indacaterol combination.
The findings of this study prove that the synergistic interaction between glycopyrronium and indacaterol is both directly and indirectly mediated downstream by the stimulation of the cAMP-dependent pathway. In fact the block of muscarinic receptors by glycopyrronium induced synergism in the presence of cAMP stimulant agents, such as indaceterol or forskolin. On the other hand, modulating the same intracellular pathway, by combining indacaterol with forskolin, produced only an additive relaxant response.
Our effort to characterize the interaction between different bronchorelaxant agents has demonstrated that synergism may be elicited only when the pharmacological interventions are focused on specific different pathways that, unexpectedly, converge in a noteworthy downstream modulation of intracellular messengers specific for only one of the involved pathways, namely the cAMP increase induced by the activation of β2-adrenoceptors. In effect, no synergistic interaction was produced by the direct activation of AC by forskolin and the concomitant stimulation of β2-adrenoceptors, even in the presence of an anti-muscarinic agent.
Finally, but not less important, these findings fully corroborate the main assumption of the BI criterion, that if two or more agents act independently of one another, neither one interferes with the other, but each contributes to a common result leading to additive effect. In fact, although β
2-adrenoceptors agonists and anti-muscarinic agents interact with different and independent smooth muscle cell membrane receptors, they orchestrate downstream intracellular signalling pathways that interfere each other eliciting synergistic bronchorelaxant interaction [
9]. Therefore, the addition of a second bronchodilator agent should be assessed considering the interference on the signalling cross-talk between different intracellular pathways.