Abstract
Airway hyperresponsiveness (AHR) is a hallmark clinical symptom of asthma. At least two components of AHR have been identified: 1) baseline AHR, which is persistent and presumably caused by airway remodelling due to chronic recurrent airway inflammation; and 2) acute and variable AHR, which is associated with an episodic increase in airway inflammation due to environmental factors such as allergen exposure.
Despite intensive research, the mechanisms underlying acute and chronic AHR are poorly understood. Owing to the complex variety of interactive processes that may be involved, in vitro model systems and animal models are indispensable to the unravelling of these mechanisms at the cellular and molecular level.
The present paper focuses on a number of translational studies addressing the emerging central role of the airway smooth muscle cell, as a multicompetent cell involved in acute airway constriction as well as structural changes in the airways, in the pathophysiology of airway hyperresponsiveness.
- Airway hyperresponsiveness
- airway inflammation
- airway pharmacology
- airway remodelling
- airway smooth muscle
- animal models of asthma
SERIES “CELL AND ANIMAL STUDIES IN RESPIRATORY MEDICINE”
Edited by R. Farré and A.T. Dinh-Xuan
Number 3 in this Series
Allergic asthma is a chronic inflammatory disease of the airways. Characteristic features of this disease are allergen-induced early and late bronchial obstructive reactions, airway inflammation, structural changes to the airway wall associated with progressive decline in lung function, and airway hyperresponsiveness (AHR) 1. AHR is defined by an exaggerated obstructive response of the airways to a variety of pharmacological, chemical and physical stimuli, including histamine, methacholine, AMP, sulphur dioxide, fog and cold air 2, 3. AHR is a risk factor for the development of asthmatic symptoms in children and adults, is associated with the severity of respiratory symptoms and decline in lung function, and determines the need for therapy 3.
Despite intensive research, the mechanisms of AHR are only partially understood. The complex variety of interactive processes that appear to be involved in the pathophysiology of AHR urges detailed in-depth investigations aimed at unravelling the underlying mechanisms at the cellular and molecular level in relation to their functional significance and hence at identifying potential targets for drug therapy. Since there are obvious ethical and experimental limitations, these mechanisms cannot be simply investigated in human subjects; therefore, in vitro model systems and animal models are indispensable. The present review focuses on a number of translational paradigms using these model systems, particularly addressing the role of the airway smooth muscle cell, not only as the key determinant of airway narrowing in asthma but also as an emerging effector of airway inflammation and remodelling 4–11.
VARIABLE AND CHRONIC AHR
It has been recognised that there are at least two components of AHR, the mechanisms of which may be different 12. The first component is baseline and relatively persistent AHR, which is present in the majority of patients with chronic asthma. Superimposed on this, there is a component of variable AHR, which can be induced by episodic exposure to environmental factors, such as allergens or respiratory tract infections 13, 14. The variable component of AHR presumably reflects current airway inflammation associated with asthma activity and severity 15, 16, whereas the underlying chronic component of AHR probably relates to structural alterations in the airways collectively called airway remodelling 12.
Airway remodelling is characterised by persistent structural changes to the airway wall, including epithelial denudation, goblet cell metaplasia, subepithelial fibrosis, increased airway smooth muscle mass, angiogenesis and alterations to extracellular matrix components 17–19. It is generally believed that both airway remodelling and chronic AHR may be induced by chronic or prolonged airway inflammation 12. However, the causal relationship between these factors in patients with asthma has recently been challenged. Thus indices of airway remodelling may already be evident in childhood asthma, with no obvious relationships to asthma symptoms and airway inflammation 20–23. This has led to the concept that inflammation and airway remodelling may occur as parallel rather than sequential events. This concept might fit with the hypothesis that an intrinsic inability of appropriate repair of epithelial injury in response to environmental agents in genetically susceptible individuals activates the epithelial–mesenchymal trophic unit, leading to the secretion of a variety of growth factors, mediators and cytokines by the activated airway epithelium, which drives the airway remodelling and promotes persistent airway inflammation 24. In addition, the causal relationship between AHR and airway remodelling is uncertain 25. Thus, although increased airway smooth muscle mass and function as well as structural changes to the airways may well explain an increase in airway responsiveness 4, 6, 26–29, airway remodelling could also be protective in limiting excessive airway narrowing by means of increased stiffening of the airways 28, 30, 31.
The relationship between variable AHR and inflammation of the airways seems rather well established. For allergic asthma, this is indicated by numerous studies demonstrating a direct association between allergen-induced acute changes in airway responsiveness and type-2 T-helper cell (Th)-driven eosinophilic airway inflammation, as well as its sensitivity to anti-inflammatory therapy 12. Nevertheless, there is some evidence that acute AHR and eosinophilic airway inflammation may dissociate. Thus treatment with anti-interleukin (IL)-5 and anti-immunoglobulin (Ig) E did not affect allergen-induced AHR despite reduced blood and airway eosinophil numbers 32. However, the results of this study should be interpreted with caution 33. Changes in airway geometry caused by both mucosal and adventitial swelling of the airways due to oedema could theoretically play a role in allergen-induced AHR, amplifying the degree of luminal narrowing for a given degree of airway smooth muscle shortening and mechanically uncoupling the airway smooth muscle from the parenchyma, respectively 34, 35. However, even though mucosal thickening has smaller effects on airway resistance than on airway responsiveness 35, allergen-induced AHR persists after airway calibre has fully returned to baseline 36. Moreover, AHR develops after challenge with low subclinical doses of allergen without significant change in pulmonary function 37, 38. Therefore, inflammation-induced alterations in the control of airway smooth muscle function are likely to be of major importance. These changes could involve changes in the neurohumoral control of airway smooth muscle tone as well as changes in the airway smooth muscle cell itself, altering its responsiveness to external stimuli 39–43. However, clinical evidence for such mechanisms is scarce and this area obviously requires further exploration.
IN VITRO MODEL SYSTEMS AND ANIMAL MODELS
In vitro model systems utilising human or animal (particularly bovine and guinea pig) primary airway smooth muscle cells and airway smooth muscle tissue preparations are widely used in studies on AHR, and have been particularly useful in unravelling molecular mechanisms of contractile, proliferative and synthetic cell function, as well as their pharmacological modulation 6, 8, 40, 42, 44–48. Airway smooth muscle tissue from asthmatic subjects permitting contraction studies or cell culture is scarce, and the limited number of studies performed thus far have delivered inconsistent results regarding the responsiveness of these preparations to contractile and relaxant agonists 42, 49, 50. Interestingly, cultures of airway smooth muscle cells from endobronchial biopsy specimens from asthmatic patients have been described recently, demonstrating intrinsic differences between asthmatic and nonasthmatic cells, with increased proliferative and synthetic capacities of the asthmatic cells 50–53.
Small animal models of asthma, using mice, rats and guinea pigs, have proven to be extremely useful for the investigation of potential mechanisms of airway pathophysiology in the intact organism in vivo, as well as in isolated organs and cells ex vivo 54–58. Likewise, animal models have been indispensable for the identification of a vast number of potential drug targets, as well as for efficacy and safety testing of new drugs. Nevertheless, there are appreciable interspecific differences in airway physiology and pathophysiology between the various animal models of asthma, as well as between animal models and human asthmatics, that should be taken into account when studying particular phenotypes of the disease.
Currently, the most widely used experimental animal for modelling allergic responses in the airways is the mouse, particularly because of the availability of trangenic and gene-targeted animals, as well as the variety of commercially available mouse-specific immunological tools for phenotypic and functional analysis of cells and mediators. Mouse models of allergic asthma have proved to be very useful in the investigation of mechanisms of allergic inflammation and the underlying immunological response that are believed to be important in a variety of processes in allergic asthma, including development of AHR 55, 58. Mice are easily sensitised to a number of antigens, including ovalbumin and human allergens, such as house dust mite and Aspergillus fumigatus. Intraperitoneal sensitisation and subsequent inhalational challenge with these antigens results in a clearly defined Th2 response in the airways, characterised by the development of antigen-specific IgE, eosinophilia and AHR, which may, however, vary considerably between strains 58. Similarly, repeated antigen challenge induces airway remodelling in this species, which, depending on strain and/or sensitisation and challenge protocols, may or may not be associated with chronic AHR 59–62. In this regard, the house dust mite model developed by Johnson et al. 59 is of particular interest. In contrast to ovalbumin models, this model permitted sensitisation via the natural route without the development of immunological tolerance, whereas airway remodelling and AHR induced by repeated allergen challenge persisted after cessation of allergen exposure and resolution of airway inflammation.
However, murine models appear to be less well suited to the investigation of the mechanisms of acute hyperresponsiveness in relation to early and late asthmatic responses. Thus AHR and early asthmatic reactions are usually observed only after repeated allergen challenge, and a late asthmatic reaction is rarely observed 58, 63. Difficulties in measuring these physiological responses could be related to the anatomical structure of the mouse lung, which is characterised by relatively large airways and a paucity of airway smooth muscle and mucous glands in the airways 58, 64. In addition, because of its small size, lung function measurement is a significant challenge in the mouse, a situation which has widely favoured the use of noninvasive barometric plethysmography in conscious mice for the measurement of enhanced pause (Penh) 65. Penh is an empirical variable derived from respiratory variations in box pressure with no direct linkage to established mechanical parameters, which has recently been seriously questioned as a valid measure of lung function 66, 67. Another drawback of measuring physiological as well as pharmacological responses in mice is the unresponsiveness of airway smooth muscle to various bronchoconstrictors implicated in the pathophysiology of asthma, including histamine, cysteinyl leukotrienes, neurokinins, bradykinin and prostanoids 54. Moreover, as in rats, the primary mediator of allergen-induced bronchoconstriction is serotonin, not histamine, and inhibitory nonadrenergic–noncholinergic (iNANC) nerves are absent in these species 54. However, in contrast to mice, rat models of allergic asthma, particularly of the Brown Norway strain, develop, usually modest, IgE-mediated early and late asthmatic reactions upon allergen challenge, which are associated with Th2-directed eosinophilic airway inflammation and hyperresponsiveness that is usually measured 18–24 h after challenge. Moreover, repeated allergen challenge causes airway remodelling, including airway smooth muscle hyperplasia and subepithelial fibrosis 56.
From both a physiological and pharmacological point of view, sensitised guinea pigs may be preferable as an animal model for investigating mechanisms of early and late asthmatic reactions and AHR in asthma 54, 57, 58, 68–71. For example, measurements in conscious and unrestrained animals demonstrate allergen-induced early and late asthmatic reactions, eosinophilic airway inflammation and AHR following both the early and the late reaction with a striking similarity to that seen in human allergic asthma, in both a qualitative and a quantitative sense 57, 70. In addition, compared with rodents such as mice and rats, airway smooth muscle responsiveness and autonomic reflexes in guinea pigs more closely resemble those of human airways 54, 69. Furthermore, various features of airway remodelling have been observed after repeated allergen challenge 72–74, which have been associated with AHR ex vivo 73. From a technical point of view, disadvantages of the guinea pig as an experimental animal may be the nonavailability of genetically modified animals and the relative scarcity of immunological tools for this species, although the latter circumstance is rapidly improving.
ACUTE MODULATION OF AHR
Nitric oxide and arginase
Nitric oxide (NO) is a ubiquitous molecule in mammalian species, serving as a signalling molecule that is involved in the control of almost every cellular and organ function in the body, including the airways 75, 76. In the respiratory tract, isoforms of constitutive NO synthase (NOS; cNOS) are mainly expressed in iNANC neurons (neuronal NOS (nNOS)), the endothelium (endothelial NOS (eNOS)) and the epithelium (nNOS and eNOS), which are primarily involved in the regulation of airway and vascular smooth muscle tone 76–78. Moreover, eNOS-derived NO has been shown to inhibit airway inflammation by suppressing the activation of nuclear factor-κB, thereby inhibiting the expression of inducible NOS (iNOS), as well as the production of inflammatory cytokines 79–82. In inflamed asthmatic airways, there is marked upregulation of iNOS expression, particularly in epithelial and inflammatory cells, including macrophages, eosinophils and neutrophils 76, 83, 84, which is associated with greatly increased production of NO and increased concentrations of NO in exhaled air 76, 85. In experimental asthma, iNOS is induced in the airways during the allergen-induced late asthmatic reaction, leading similarly to increased levels of NO in exhaled air 86, 87. High concentrations of iNOS-derived NO have been considered to be detrimental in the airways since they contribute to increased vascular permeability, mucus hypersecretion, inflammatory cell infiltration, epithelial cell damage and perpetuation of the Th2-mediated inflammatory response in the airways 76. Most if not all of the deleterious effects induced by iNOS-derived NO may proceed via formation of peroxynitrite (ONOO-), a highly reactive oxidant synthesised by the rapid reaction of NO with superoxide anion (O2-), generated in the inflamed airways 88, 89. Significant correlations between exhaled NO level, airway eosinophilia and AHR have been observed in asthmatics, whereas all of these parameters are reduced after glucocorticosteroid treatment 89–92. Based on these observations, the NO concentration in exhaled air has been adopted as a sensitive marker of airway inflammation 93. Therefore, it is not surprising that much of the literature regarding NO in asthma has long been focused on iNOS and is heavily biased towards a harmful pro-inflammatory role of NO. However, given the variety of enzymatic and cellular sources of NO in the airways, the different cellular targets and physiological effects of NO, as well as the influence of the local microenvironment on NO homeostasis, the relevance of increased exhaled NO levels to discrete pathophysiological processes in the airway wall may be difficult to determine. Indeed, various studies in animal models and patients have now indicated that AHR and inflammation of the airways is caused by failure of both cNOS- and iNOS-derived NO to exert bronchodilatory and anti-inflammatory effects rather than by an excess of NO 94.
In a guinea pig model of acute allergic asthma, in the absence of iNOS expression, which is induced during the late asthmatic reaction or by chronic allergen exposure in this model, it was demonstrated that the exhaled NO level is transiently increased immediately after allergen challenge, followed by a fall to below control levels 95. The decrease in NO production could well contribute to the early asthmatic reaction, as well as to the ensuing AHR. Indeed, both in vivo and ex vivo studies using the same species indicate that a deficiency of both epithelial and neuronal cNOS-derived NO underlies the development of AHR following the allergen-induced early asthmatic reaction (fig. 1⇓) 77, 96, 97. A deficiency of cNOS activity and endogenous bronchodilating NO contributing to AHR was also demonstrated after repeated allergen challenge of sensitised guinea pigs 98, 99. Importantly, it was demonstrated that a reduction in cNOS-derived NO may also contribute to AHR in patients with severe asthma 100, and may similarly be induced by allergen exposure 101. Interestingly, decreased cNOS (presumably iNANC)-derived NO could also contribute to reduced bronchoprotection by means of deep inspiration in asthmatic patients, which has been recognised as an important factor contributing to AHR 102–105.
Various mechanisms have been implicated in allergen-induced NO deficiency. Reduced expression of eNOS or nNOS has been observed after repeated allergen challenge in guinea pigs and mild asthmatic patients, respectively 101, 106. In addition, it has been demonstrated that reduced bioavailability of l-arginine, the substrate for NOS, may underlie the deficiency in NO and subsequent AHR 96, 107. Animal studies have indicated that polycationic proteins, including eosinophil-derived major basic protein (MBP), inhibit cellular uptake of l-arginine by cationic amino acid y+ transporters 108, which may contribute to the deficiency in cNOS-derived NO and AHR after the early asthmatic reaction 109, 110. A second mechanism that might be crucial in the reduced bioavailability of NO in the airways is increased utilisation of l-arginine by arginase, which hydrolyses l-arginine to l-ornithine and urea (fig. 1⇑) 94, 111, 112. Arginase I and II are both expressed constitutively in the airways, particularly in epithelial cells, (myo)fibroblasts and alveolar macrophages 113–115. In guinea pigs, it has been discovered that arginase activity is functionally involved in basal airway responsiveness by limiting cNOS-derived NO production 116. Using airway preparations from the same animal model, it was demonstrated that increased arginase activity may be involved in allergen-induced AHR and reduced iNANC relaxation after the early asthmatic reaction 96, 111. Moreover, reduced l-arginine availability to iNOS, induced by increased arginase activity as well as reduced transport of the amino acid, may lead to synthesis of both NO and O2- by this enzyme 117, effectively causing the production of ONOO-, which may contribute to the AHR after the late asthmatic reaction 118, 119.
In various mouse and rat models, it has been confirmed that allergen challenge causes a considerable increase in the expression and activity of particularly arginase type I, most probably via Th2 cytokines involved in the asthmatic airway inflammation 120–127. Notably, by microarray analysis of gene expression in Balb/c mice sensitised to ovalbumin or Aspergillus fumigatus, it was shown that, among the 291 common genes that were induced by these allergens, enzymes involved in l-arginine metabolism, particularly arginase I and II, belonged to the most predominantly overexpressed genes 127.
Recent studies in asthmatic patients have also indicated increased expression of arginase I in the airways, particularly in epithelial and inflammatory cells 127. In addition, in asthmatics experiencing an exacerbation, a striking reduction in plasma l-arginine levels has been measured, which was associated with an increase in serum arginase activity 128. Moreover, increased immunoreactivity against arginase I, which could possibly be induced by nicotine, was recently observed in both the epithelium and the smooth muscle layer of smoking asthmatics 129. Interestingly, single nucleotide polymorphisms of arginase I and II have recently been found to be associated with atopy and risk of childhood asthma, respectively 130.
Increased metabolism of l-arginine by arginase in the airways may not only compromise NO homeostasis, leading to AHR as a result of reduced bronchodilation and enhanced inflammation, but may also contribute to airway remodelling in chronic disease, through NO-independent pathways mediated by increased production of l-ornithine. Thus l-ornithine is a precursor of the arginase downstream products l-proline and polyamines, which could promote collagen production and growth of mesenchymal cells, such as fibroblasts and smooth muscle cells, respectively 94, 131, 132. In support of a potential role of arginase in airway fibrosis in asthma, IL-4 and -13 increase arginase I and II expression and arginase activity in cultured rat fibroblasts 114. In addition, increased levels of polyamines have been observed in mouse lung following allergen challenge 127 and in the serum of asthmatic patients 133, respectively. A role for decreased NO synthesis in airway remodelling can also be envisaged, since NO inhibits proliferation of airway smooth muscle cells 134. Moreover, reduced NO synthesis could contribute to airway remodelling by increasing the activity of ornithine decarboxylase, which converts l-ornithine to polyamines 135. The role of arginase in asthmatic airway remodelling remains to be established however.
Cholinergic mechanisms
The parasympathetic nervous system represents a major bronchoconstrictory pathway. During normal breathing, preganglionic nerves innervating the parasympathetic ganglia in the airways evoke action potentials with relatively high frequencies, in the range 10–20 Hz 136. As a result, basal airway smooth muscle tone in vivo is mediated to a significant extent by both ganglionic and post-ganglionic cholinergic nerve activity, of which acetylcholine is the major neurotransmitter. The fidelity with which pre-ganglionic impulses are translated into post-ganglionic action potentials is relatively low, implying a filtering function of these ganglia. Various acute inflammatory mediators, including histamine, prostaglandin (PG) D2 and bradykinin, are able to reduce this filtering function, and, consequently, to enhance ganglionic cholinergic transmission 137. The same is true for tachykinins (substance P and neurokinin A) released by nonmyelinated sensory C-fibres in the airways 138. The release of acetylcholine from parasympathetic nerve endings is regulated by a variety of prejunctional receptors that may inhibit or facilitate neurotransmitter outflow. Autoinhibitory muscarinic M2 receptors, activated by acetylcholine itself, represent an important negative feedback, limiting further release, particularly at higher frequencies 139. In animal models of allergic and viral airway inflammation and asthma, dysfunction of these M2 autoreceptors has been found to contribute to exaggerated acetylcholine release from vagal nerve endings, increased cholinergic reflex activity in response to inhaled stimuli and AHR 140–142. In allergic guinea pigs, the magnitude of the early asthmatic reaction has been reported to correlate significantly with the extent of M2 autoreceptor dysfunction 142. M2 autoreceptors have also been found to be dysfunctional in some but not all asthmatics 143–145. In addition, asthmatics with active viral infections show greater bronchodilatory responses to inhaled anticholinergics, suggesting increased vagal tone 146. Most of the M2 autoreceptor dysfunction is caused by activated eosinophils that migrate to cholinergic nerves and release MBP, which acts as an allosteric M2 antagonist 147–149. Taking into consideration the fact that prejunctional M2 receptor function is more prominent in the larger airways 150, it is no surprise that M2 autoreceptor dysfunction is more prominent in asthma than in chronic obstructive pulmonary disease (COPD). Indeed, in patients with stable COPD, M2 autoreceptors appear to function normally 151, although this does not exclude dysfunction during acute exacerbations. Viral infections, which may play a role in exacerbations of both asthma and COPD, induce dysfunction through neuraminidases that cleave portions of the M2 receptor, and through as yet incompletely characterised mechanisms involving macrophages, CD8+ lymphocytes and possibly interferon-γ 152.
In addition to M2 autoreceptors, a variety of heteroreceptors modulating acetylcholine release have been identified in cholinergic nerve terminals. Catecholamines may inhibit or facilitate acetylcholine overflow through prejunctional α2- and β2-adrenoceptors, respectively 153–155. Neurokinins such as substance P may enhance cholinergic transmission through facilitatory neurokinin-1 and/or -2 receptors 156, 157. Interestingly, substance P may also induce MBP release from eosinophils, causing M2 autoreceptor dysfunction, which could act synergistically to direct facilitation 158. Allergic-inflammation-derived prostanoids, including PGD2, PGF2α and thromboxane A2, as well as histamine, can also augment acetylcholine release through prejunctional receptors 138. Taken together, the above observations indicate that parasympathetic acetylcholine release is governed by various regulatory systems, the set point of which is subject to environmental modulation. During periods of airway inflammation, these modulations often result in enhanced cholinergic transmission (fig. 1⇑). Thus AHR following the early asthmatic reaction is reversed by anticholinergics 141, and the bronchoconstrictor response of inhaled inflammatory mediators (e.g. histamine and thromboxane A2) is, to a large extent, mediated by cholinergic pathways 141, 159.
Acetylcholine released by post-ganglionic parasympathetic nerves may choose to interact with any one of five muscarinic receptor subtypes, M1–M5. Most organs and tissues express more than one subtype, and this is also true for many individual cells. In mammalian airways, including those of humans, M1, M2 and M3 receptors are the most important. M1 receptors are present in type II alveolar cells, presumably mediating/contributing to surfactant production 160, 161, as well as in parasympathetic airway ganglia, where they facilitate ganglionic transmission 138. Thus vagal bronchoconstriction, induced by sulphur dioxide inhalation, has been found to be especially sensitive to inhaled pirenzepine, an M1-selective antagonist 162. M2 and M3 receptors represent the major populations, however, in both intra- and extrapulmonary airways. Postjunctional receptor populations in airway smooth muscle are a mixture of M2 and M3 receptors, the Gi-protein (Gi)-coupled M2 subtype being predominant, particularly in the larger airways. Contraction, however, is primarily mediated by Gq-protein (Gq)-coupled M3 receptors, even in those smooth muscle preparations in which the ratio of M2:M3 is 90:10, the M2 receptor population playing, at most, a minor supporting role 161. This was confirmed in airway smooth muscle preparations from M2 receptor knockout mice, in which carbachol was barely less potent than in preparations from wild-type mice 163.
The principal signalling of Gq-coupled receptors is activation of phospholipase C, mediating hydrolysis of phosphatidylinositol 4,5-bisphosphate into inositol 1,4,5-triphosphate (IP3) and 1,2-diacylglycerol (DAG). IP3 mobilises Ca2+ ions from intracellular stores, generating a rapid and transient rise in cytosolic free Ca2+ concentration. DAG triggers the translocation and activation of protein kinase C, which is able to phosphorylate a variety of protein substrates 164. Cross-talk between Gi-coupled M2 receptors and Gs-protein (Gs)-coupled β-adrenoceptors (having opposing effects on cyclic AMP accumulation) has no major effects on modulation of muscarinic-agonist-induced contraction or β-agonist-induced relaxation, at least under physiological conditions 165. In contrast, Gq-coupled M3 receptors may have a major influence on β-adrenoceptor function, even in noninflamed airways. This is due to DAG-induced activation of protein kinase C, which may: 1) phosphorylate the β2-adrenoceptor, as well as Gs, causing receptor uncoupling and desensitisation 165; and 2) phosphorylate and activate β-adrenoceptor kinases, which are members of the G-protein-coupled receptor kinase family, amplifying homologous β-agonist induced desensitisation 166, 167. These processes may explain the well-known attenuation of β-agonist efficacy during episodes of severe bronchoconstriction, for example during exacerbations.
In contrast to the enhanced release of acetylcholine due to neuronal mechanisms associated with inflammation, as discussed above, no evidence for upregulation of postjunctional M3 and M2 receptors has been found in hyperresponsive airways of patients with asthma and COPD. However, increased expression and enhanced function of signalling molecules involved in muscarinic-agonist-induced smooth muscle contraction (and mucus secretion) have been identified. Thus several pro-inflammatory cytokines, including IL-1β and tumour necrosis factor-α, increase Gαq and Gαi expression in airway smooth muscle, which could account for the increased Ca2+ responses to muscarinic agonists and contraction 168, 169. Coupling of the M3 receptor to CD38 enhances the production of cyclic ADP ribose and the release of Ca2+ ions through ryanodine-sensitive stores in the sarcoplasmic reticulum 170. IL-1β, tumour necrosis factor-α, IL-13 and interferon-γ have also been reported to enhance Ca2+ responses to muscarinic agonists through this mechanism, via increasing CD38 expression 171–174. Conversely, CD38-deficient mice show reduced AHR towards methacholine, both in vivo and ex vivo, following IL-13 challenge 175. Collectively, these studies indicate that airway inflammation results in increased cholinergic bronchoconstrictor responses of the asthmatic airway, caused by enhanced neuronal release of acetylcholine and enhanced airway smooth muscle expression of signalling molecules central to muscarinic receptor function.
Rho kinase
Contraction of airway smooth muscle is primarily regulated by Ca2+-dependent mechanisms, initiated by a (rapid) rise in intracellular Ca2+ concentration, followed by the formation of Ca2+ complexes, which, in turn, activate myosin light chain (MLC) kinase, finally resulting in phosphorylation of the 20-kDa regulatory MLC (MLC20). However, Ca2+-independent mechanisms, characterised by augmented smooth muscle shortening at a fixed Ca2+ concentration are also important. This phenomenon is referred to as Ca-2+- sensitisation 176. Key regulatory factors of the Ca2+ sensitivity of airway smooth muscle are Rho kinase and the small monomeric G-protein RhoA, its main activator. The RhoA/Rho kinase cascade can be stimulated by a variety of receptors, including those coupled to G12/13, Gi and Gq. Activated Rho kinase phosphorylates, amongst others, MLC phosphatase (MLCP), which causes dephosphorylation of MLC20; as a consequence, MLCP is inactivated, resulting in enhancement of MLC20 phosphorylation and augmented contraction 41. Based on this property, and on the pronounced activation of the RhoA/Rho kinase pathway by inflammatory mediators, a role for this pathway in AHR in asthma was recently postulated 41.
The extent to which Rho kinase activation contributes to airway smooth muscle contraction is both agonist- and receptor-dependent. In bovine tracheal smooth muscle preparations, full and partial muscarinic agonists, acting through M3 receptors, are differentially dependent upon Rho kinase for their contractile effects 177. In the lower concentration ranges, all agonists were sensitive to the Rho kinase inhibitor (+)-(R)-trans-4-(1-aminoethyl)-N-(4-pyridyl) cyclohexane carboxamide (Y-27632); in contrast, Y-27632 only reduced maximal contractions induced by partial agonists (such as pilocarpine), whereas the maximal contractions induced by the full agonist methacholine were unaffected. Indeed, an inverse relationship between Ca2+ mobilisation, as well as Ca2+ influx, and the Rho kinase dependency of the contraction induced by these agonists was found 177. Intriguingly, in both human bronchial 178 and guinea pig tracheal 179, 180 smooth muscle preparations, it was found that growth factors and insulin, which act through receptor tyrosine kinases, induce moderate contractions that are totally Rho-kinase-dependent.
In animal models of inflammatory airways disease, evidence indicating a primed role for Rho A/Rho kinase in enhancing obstructive airway responses is accumulating. Thus the functional contribution of Rho kinase in acetylcholine-induced rat bronchial smooth muscle contraction is increased after repeated allergen challenge 181, and parallelled by increased translocation of RhoA to the membrane 182. Furthermore, active allergic sensitisation of guinea pigs without subsequent allergen exposure is already sufficient to increase RhoA expression and enhance its contribution to contraction ex vivo and airway responsiveness in vivo 183. Remarkably, similar observations were made using passively sensitised guinea pig airway preparations 184, and recent data indicate that the same is true for cigarette-smoke-induced hyperresponsiveness of rat bronchial smooth muscle 185. In vivo experiments using a guinea pig model of asthma have further indicated that increased Rho kinase activity contributes to the allergen-induced AHR following both the early and late asthmatic reaction, which is effectively reversed by inhalation of Y-27632 186. These data indicate that enhanced Rho/Rho kinase signalling is a feature of AHR in animal models of asthma; its role in human subjects remains to be determined however.
AIRWAY SMOOTH MUSCLE REMODELLING AND AHR
Structural and phenotypic alterations of airway smooth muscle in asthma
In addition to its central role in limiting airflow and regulating variable AHR, it is increasingly evident that chronic structural and phenotypic alterations of the airway smooth muscle superimpose upon the variable response to exaggerate airway smooth muscle contraction (fig. 1⇑) 4. The smooth muscle mass that encircles the airways and regulates the luminal diameter is considerably thicker in asthmatics 187. Detailed analyses of the airway smooth muscle bundle obtained from asthmatics have indicated that increases in both airway smooth muscle cell number (hyperplasia) and cell size (hypertrophy) contribute to this response, although the extent to which these processes determine the increase in muscle mass may vary among patients 188–190. Based on differences in smooth muscle structural characteristics, Ebina et al. 189 even proposed different asthma phenotypes, one characterised primarily by smooth muscle hyperplasia in the central bronchi and another by smooth muscle hypertrophy throughout the bronchial tree. The exact contribution of airway smooth muscle thickening to AHR is not yet completely clear. However, mathematical modelling studies indicate that airway smooth muscle thickening is probably of major importance, perhaps even the primary cause of exaggerated airway narrowing, in the remodelled airway 27, 29. Importantly, these changes in airway structure worsen with duration of disease, which could contribute to a chronic increase in severity of airway narrowing 191.
Airway smooth muscle proliferation and hypertrophy have been extensively studied in cell culture and animal models of asthma. These studies have revealed that numerous growth factors, contractile agonists, cytokines, proteases and matrix proteins contribute to these responses 7, 192. Moreover, recent studies indicate that asthmatic airway smooth muscle cells are characterised by an intrinsic functional change, facilitating cell proliferation in culture 52. This intrinsic change is partially explained by changes in extracellular matrix protein deposition by asthmatic airway smooth muscle 9. Thus, when normal human airway smooth muscle cells are cultured on an extracellular matrix laid down by asthmatic airway smooth muscle cells, the normal cells also show an increased proliferation rate 9. Abnormal extracellular matrix production, including increases in fibronectin, are also responsible for the enhanced synthetic capacity of airway smooth muscle cells in the production of eotaxin 51. A recent study indicated that, in addition to this abnormality, asthmatic airway smooth muscle expresses increased numbers of mitochondria and exhibits increased mitochondrial activity 53. Increased mitochondrial biogenesis by asthmatic airway smooth muscle is accompanied by increased expression of mitochondrial transcription factor A. Interestingly, the increase in mitochondrial activity and biogenesis is coupled to an increased proliferation rate of these cells 53. Collectively, these intrinsic biochemical differences suggest that an intrinsic functional change in the asthmatic airway smooth muscle makes the muscle more responsive to mitogens 52, 53, 193, 194. However, the pathogenic cause of this intrinsic functional change remains unknown.
Asthmatic airway smooth muscle may also produce enhanced contractile responses because of changes in its phenotype that are associated with increased expression of contractile proteins and contraction-regulatory proteins. Isolated asthmatic airway smooth muscle cells have been reported to contract more profoundly and more rapidly in vitro 49. Furthermore, passive sensitisation of human bronchi with atopic serum increases maximal contractility and agonist sensitivity in vitro 195. Interestingly, this effect is associated with serum IgE 196, suggesting a relationship between allergic sensitisation and increased contractile responsiveness. In a canine model of allergic sensitisation, similar effects have been revealed following active sensitisation 197, 198. This increase in contractility is accompanied by increases in MLC kinase expression, in both the canine model 199 and sensitised human airway smooth muscle 188, 200.
Studies using animal models of allergic asthma have demonstrated additional changes in gene expression following sensitisation and/or repeated allergen exposure that may explain increases in the contractile properties of the muscle; the importance of these changes remains to be confirmed for human asthmatic subjects however. RhoA expression (as discussed above) is increased in repeatedly allergen-challenged rats and allergen-sensitised guinea pigs, which may contribute to increased agonist-induced Ca2+ sensitisation in the muscle 41, 181, 183. A similar role has been identified for the protein CD38, which regulates cyclic ADP ribose production and subsequent activation of ryanodine receptors on intracellular Ca2+stores 170, 171. The reported increase in mitochondrial biogenesis by asthmatic airway smooth muscle may also contribute to this response, since mitochondria play a known role in airway smooth muscle Ca2+ homeostasis 53.
Expression of contractile proteins, including smooth muscle myosin heavy chain (SM-MHC), is also increased after repeated allergen challenge in guinea pigs 73. SM-MHC exists in several isoforms caused by alternative splicing of a single gene. The so-called (+)insert isoform (also called SM-B), characterised by a seven amino acid insert close to the N-terminus, is characterised by increased cross-bridge cycling activity 201. Interestingly, a recent study indicated that the expression of the (+)insert of SM-MHC is increased in the hyperresponsive Fisher rat strain, suggesting that this mechanism could also contribute to AHR 202. These cell and animal studies collectively suggest that changes in the expression of myosin and in proteins that regulate the dynamics of agonist-induced myosin phosphorylation may contribute to allergen-induced AHR. The importance of these findings to the human situation has only partially been revealed and future studies are required in this area.
Pathophysiological mechanisms
The next relevant question concerns which endogenous mediators regulate these responses and whether or not they can be used as pharmacological targets. From in vitro studies, it is well established that a variety of growth factors, G-protein-coupled receptor agonists, cytokines, proteases and matrix proteins can contribute to the aforementioned responses 7. Relatively few data are available from animal models, however, although recent research indicates that cysteinyl leukotrienes, acetylcholine and transforming growth factor-β (TGF-β) act as important endogenous mediators of airway smooth muscle remodelling in vivo, suggesting that targeting these may be of therapeutic value. The focus in this section is, therefore, on these mediators, especially since drugs targeting cysteinyl leukotrienes and acetylcholine are already clinically available.
Cysteinyl leukotrienes probably play an important regulatory role in airway smooth muscle remodelling. In 1993, Wang et al. 203 demonstrated that blockade of the cysteinyl leukotriene receptor 1 was sufficient to significantly reduce the increase in airway smooth muscle mass observed following repeated allergen exposure in Brown Norway rats. Similar observations were subsequently made in a murine model, showing that the increase in airway smooth muscle mass and increases in Penh following allergen exposure were reduced by treatment of the mice with montelukast 204. The observation that montelukast administration is able to completely reverse remodelling of the airway smooth muscle bundle, with the experimental treatment starting only after repeated allergen challenges were completed, is spectacular 205. These data indicate that cysteinyl leukotrienes may play a significant role in the onset and the maintenance of allergen-induced airway smooth muscle thickening.
Cysteinyl leukotrienes probably mediate this response via multiple mechanisms. Cell culture studies indicate that cysteinyl leukotrienes augment the proliferative response of peptide growth factors such as epidermal growth factor (EGF) 206. Whether or not cysteinyl leukotrienes act as airway smooth muscle mitogens per se remains the subject of debate 206, 207. The mechanisms responsible for this response have been partially elucidated and appear to involve reactive oxygen species generation and p42/p44 mitogen-activated protein kinase activation, key players in the mitogenic response of airway smooth muscle 207. These direct effects on the muscle are, however, not the only explanation for the observed effects. Cysteinyl leukotrienes regulate airway inflammation in response to allergen challenge and induce growth factor release from airway structural cells. Thus montelukast inhibits airway eosinophilia and cytokine/chemokine expression in response to allergen challenges in a murine model of asthma 204, 205. In addition, it was recently established that cysteinyl leukotrienes induce the release from airway epithelial cells of TGF-β 208, a key growth factor in many of the structural and phenotypic abnormalities of asthmatic airway smooth muscle, as described below. Therefore, it is likely that these indirect responses too are responsible for the reduction in smooth muscle mass by montelukast.
Acetylcholine, the primary parasympathetic neurotransmitter in the airways, is traditionally associated with bronchoconstriction and mucus secretion. Recent findings are changing this traditional view since acetylcholine production in the airways appears not to be restricted to the parasympathetic nervous system; it is also released from non-neuronal sources, such as the bronchial epithelium and several inflammatory cells 164, 209, 210. Interestingly, acetylcholine (either neuronal or non-neuronal) also appears to regulate remodelling in a guinea pig model of chronic asthma 72, 73. Tiotropium treatment of these guinea pigs significantly reduced airway smooth muscle thickening, and had similar protective effects on allergen-induced increases in pulmonary SM-MHC expression and tracheal contractility 73. This indicates that acetylcholine regulates multiple structural and phenotypic changes in airway smooth muscle caused by allergen exposure.
As observed for cysteinyl leukotrienes, muscarinic receptors on airway smooth muscle also appear to regulate the mitogenic response of airway smooth muscle. The mitogenic effects of platelet-derived growth factor and EGF, for instance, are enhanced, although muscarinic receptors do not themselves mediate mitogenic responses 211, 212. The augmentation is dependent upon M3 receptors, which augment the intracellular signalling of growth factors by cooperatively regulating phosphorylation of p70S6 kinase and glycogen synthase kinase-3, both resulting in increased cell cycle progression 212, 213. Preliminary evidence also indicates that muscarinic receptors regulate the expression of SM-MHC by airway smooth muscle 214. Nonetheless, it is not unlikely that indirect effects of acetylcholine also contribute to this response. As mentioned earlier, (non-neuronal) acetylcholine regulates aspects of airway inflammation, including eosinophilia in response to allergen exposure and neutrophilia in response to diesel particle inhalation 72, 215. Indirect inhibition of airway smooth muscle structural and phenotypic alterations by anticholinergics is therefore also a possibility.
TGF-β, although not a direct target of drugs marketed for asthma treatment at the moment, appears to be among the most relevant growth factors mediating chronic structural and phenotypic changes in airway smooth muscle. Antibodies directed against TGF-β, administered intraperitoneally to mice during an antigen challenge protocol, significantly reduced the increase in airway smooth muscle mass without inducing drastic alterations in inflammatory cell recruitment or pulmonary cytokine expression 216. Likewise, in a Brown Norway rat model of asthma, the TGF-β receptor 1 kinase inhibitor SD-208 reduced allergen-induced AHR, airway smooth muscle cell proliferation in situ and muscle thickness 217. These studies indicate that TGF-β plays a prominent role in allergen-induced airway smooth muscle structural changes, and in the associated AHR.
The prominent role of TGF-β is further revealed by in vitro studies indicating that TGF-β plays an important role in most of the structural and phenotypic changes that are observed in asthma. TGF-β induces airway smooth muscle proliferation, which, interestingly, is partially dependent upon reactive oxygen species generation, as the result of enhanced transcriptional regulation of reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase 4, a catalytic homologue of NADPH oxidase 218. TGF-β also regulates transcription and translation of contractile proteins and induces airway smooth muscle cell hypertrophy, with the involvement of phosphatidylinositol-3’-kinase and downstream signalling 219. This indicates that drugs targeting TGF-β-dependent signalling could achieve a reduction in several smooth muscle abnormalities seen in asthma, although this still needs to be assessed in human subjects.
CONCLUSIONS
In vitro model systems and animal models of allergic asthma have been successfully applied to the investigation of the mechanisms of acute and chronic airway hyperresponsiveness. In particular, the rapidly growing interest in the role of the airway smooth muscle cell, as a multicompetent cell that may be involved in both functional and structural changes in the airways, has produced interesting novel concepts of pathophysiological mechanisms and has disclosed new directions for future drug treatment. The majority of these mechanisms are linked to acute and chronic inflammatory processes in the airway wall, associated with the release of mediators, growth factors and neurotransmitters that may alter airway smooth muscle function by changes in neural and non-neural control, receptor-mediated signalling pathways, and proliferation and maturation of the muscle cells. Most of these concepts are awaiting translation to the asthmatic patient. Moreover, future studies should also be directed towards critical questions addressing the causal role of inflammation, the functional significance of airway remodelling and the role of genetics in airway hyperresponsiveness.
Support statement
R. Gosens is the recipient of a Veni grant (No. 916.86.036) from the Netherlands Organisation for Scientific Research (The Hague, The Netherlands).
Statement of interest
None declared.
Footnotes
-
Previous articles in this series: No. 1: Oeckler RA, Hubmayr RD. Ventilator-associated lung injury: a search for better therapeutic targets. Eur Respir J 2007; 30: 1216–1226. No. 2: Chung KF, Adcock IM. Multifaceted mechanisms in COPD: inflammation, immunity, and tissue repair and destruction. Eur Respir J 2008; 31: 1334–1356.
- Received February 14, 2008.
- Accepted February 18, 2008.
- © ERS Journals Ltd
References
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵