- Email: [email protected]
Eosinophil and airway nerve interactions
Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13 www.elsevier.com/locate/ypupt
Eosinophil and airway nerve interactions P.J. Kinghamb, Richard W. Costelloa,*, W. Graham McLeanb a
Department of Medicine, Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin 9, Ireland b Department of Pharmacology, University of Liverpool, Liverpool, UK Received 21 March 2002; accepted 2 April 2002
Abstract In vivo, eosinophils localise to airway nerves in patients with asthma as well as in animal models of hyperreactivity. In both, in vivo and in vitro studies, we have shown that this localisation changes both cholinergic nerve and eosinophil function. In particular, it leads to an increase in acetylcholine release due to loss of function of a neuronal autoreceptor, the M2 muscarinic receptor. This loss of M2 receptor function occurs because eosinophils become activated and degranulate as a result of interactions that occur via specific adhesion molecules expressed on nerves that are recognised by counter ligands on eosinophils. q 2003 Elsevier Science Ltd. All rights reserved. Keywords: Asthma; Hyperreactivity neuro-immunology
1. Eosinophils: to have and to hold Asthma is characterised by the presence of a variety of inflammatory cells, including lymphocytes and eosinophils [1]. How these cells are linked to the clinical features of asthma has been the subject of intense investigation over the last twenty years. Histological as well as functional studies suggest that eosinophils, in particular, play a pivotal role in the pathogenesis of asthma. For example, increased numbers of eosinophils are found in bronchoalveolar lavage fluid, induced sputum and bronchial wall biopsies of patients with asthma [2]. Histological examination of sections of airway from patients who have died during an asthma attack show increased numbers of activated, degranulating eosinophils along with the presence of released toxic mediators [3,4]. Similarly, resolution of asthma symptoms is accompanied by a reduction in the number of eosinophils in the airways [5 – 8]. Eosinophils have a wide range of functions. For example, they release toxic cationic proteins such as eosinophil major basic protein (MBP), eosinophil peroxidase, eosinophilderived neurotoxin and eosinophil cationic protein. Eosinophils also generate lipid molecules such as platelet-activating factor, prostaglandin E2 and 5-hydroxyeicosa-6,8,11,14tetraenoic acid (5HETE) [1]. Studies in cell culture and in * Corresponding author. Tel.: þ 353-1-8093762; fax: þ 353-1-8093765. E-mail address: [email protected] (R.W. Costello).
animal models have shown that many of these compounds are capable of mimicking the features of asthma. Based on these observations it has been concluded, not unreasonably, that eosinophils are key effector cells in asthma. Eosinophils can also release compounds that activate other inflammatory cells; they are a source of both TH1 and TH2 cytokines, of TNF-a, and of several chemokines. Additionally, eosinophils can function as antigen-presenting cells and they may play a role in tissue repair, for example by secretion of gelatinase-B and transforming growth factor-b (TGF-b). Thus, the possibility remains that eosinophils are playing a role in the repair of damaged tissue. Alternatively, they are innocent bystander cells recruited to the airways as a consequence of activation of CD4 TH2-cells and they play no specific role in the condition.
2. Location Two recent studies, using two different monoclonal antibodies to disrupt eosinophil trafficking to the airways, have demonstrated no significant effect on the late phase response to antigen in sensitive asthmatics although, in both cases, sputum eosinophilia was reduced [9,10]. These studies have raised questions concerning the pathogenic role of eosinophils in asthma. However, it is important to note that eosinophils in the tissues behave differently from those in the airways. In the airways eosinophils interact with
1094-5539/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. PII: S 1 0 9 4 - 5 5 3 9 ( 0 2 ) 0 0 0 9 3 - 7
10
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13
the surrounding tissues through specific surfaces receptors and these interactions lead to significant functional changes. For example, recent studies have shown that eosinophils can avoid both apoptosis and the requirement for otherwise necessary growth factors such as interleukin (IL)-5 by adhesion molecule-dependent autocrine and paracrine generation of growth factors [11,12]. Thus, any investigation into the role of eosinophils in asthma requires a knowledge of the nature of the interactions that occur between eosinophils and airway cells. Included among those cells are the nerves that innervate the airways. Recent studies from our laboratory have focused on the nature of the interactions that occur between eosinophils and airway cholinergic nerves.
3. Parasympathetic nerves in the airways The parasympathetic nerves provide the dominant innervation of airway smooth muscle. Long preganglionic vagal fibres project from nuclei in the brain stem into the thorax. There they relay onto ganglia located on the posterior aspect of the trachea and major bronchi. Short postganglionic fibres arise from these ganglia to innervate the airway smooth muscle, mucous glands and bronchial and pulmonary arterioles. Stimulation of these parasympathetic nerves releases acetycholine which binds to M3 muscarinic receptors on these structures, leading to bronchoconstriction, mucous production and dilation of the bronchial vessels. These nerves provide a protective mechanism for the airways against potentially harmful inhaled toxins, as bronchoconstriction induces turbulent, as opposed to laminar, airflow. This change in airflow causes particles to be deposited in the larger airways where they are trapped by simultaneously generated mucous. Thus, the vagus nerves provide an important respiratory defence mechanism. However, when nerve activity is excessive it may be harmful by causing excessive bronchoconstriction and mucous production, which can also obstruct airflow. Therefore, a system exists to control this activity. In addition to its effect on M3 receptors acetylcholine released from parasympathetic nerves also acts on neuronal M2 muscarinic receptors. Stimulation of these neuronal M2 muscarinic receptors limits further acetylcholine release [13 –15]. It follows that loss of function of these M2 receptors leads to an increase in vagally induced bronchoconstriction and mucous production. We and others have demonstrated in animal models and in some patients with asthma that there is indeed loss of function of these receptors [16 – 22]. For example, in antigen-sensitised animals exposed to further antigen there is an increase in the magnitude of vagally induced bronchoconstriction and hyperreactivity to compounds such as histamine [23]. The mechanism for this loss of M2 muscarinic receptor function appears to be due to the action of MBP. Ligand-binding studies have demonstrated that
eosinophil MBP is an allosteric antagonist at M2 but not at M3 muscarinic receptors [24]. In vitro, anionic compounds such as heparin displace MBP from M2 receptors, suggesting that the cationic charge of MBP is responsible for the inhibition of these receptors. Conversely, in vivo, administration of heparin to antigen-challenged guinea pigs restores M2 receptor function [25]. Furthermore, inhibiting eosinophil recruitment to the airways with an antibody to the eosinophil adhesion molecule VLA-4, depleting eosinophils with an antibody to the eosinophil growth factor IL5, or specifically inhibiting MBP all preserve M2 receptor function [18,20,26]. These data indicate that significant interactions occur between eosinophils and in particular, eosinophil MBP, and cholinergic nerves. MBP is a highly cationic protein that does not readily diffuse once released from eosinophil stores. Thus, we first had to investigate whether or not eosinophils are capable of delivering MBP to cholinergic nerves. In histological studies performed on sections of airway from antigenchallenged guinea pigs, rats and patients who died during an asthma attack, eosinophils were seen to localise to airway cholinergic nerves [22]. In addition, there was frequent evidence of eosinophil degranulation and release of MBP close to these nerves, as shown in Fig. 1. We then undertook further work to investigate the fine details of the interactions of eosinophils with airway nerves, to determine the mechanism of localisation of eosinophils to nerves and to assess the consequences for eosinophil and nerve function once these cells co-localise.
4. Eosinophils localisation to airway nerves via specific adhesion molecules One mechanism whereby eosinophils may localise to airway nerves is that they recognise complementary neuronal adhesion molecules. Eosinophils have an extensive repertoire of adhesion molecules that mediate both initial light ligation and more firm adhesion. For eosinophils this later adhesion is mediated through VLA-4 and CD11/18 recognising VCAM-1 and ICAM-1, respectively. Both VCAM-1 and ICAM-1 are expressed by primary cultures of guinea pig parasympathetic nerves and by the human cholinergic nerve cell line IMR32. In the primary cultures, ICAM-1 expression is constitutive while VCAM-1 requires cytokine pre-treatment in primary culture, while in the cell line both are constitutively expressed [27]. We have used a chromium radiolabelling technique and functional measurements in both systems to demonstrate that inhibiting either VCAM-1 or ICAM-1 can completely prevent adhesion. Similarly, pre-treatment of the nerve cells with heparin inhibits adhesion. The fact that one or other adhesion molecule is sufficient for adhesion suggested that there might be a common process linking the two adhesion molecules. It is known that ligation of VLA-4 leads to a process of inside-out signalling through the protein kinase C
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13
11
Fig. 1. Eosinophils (shown with arrows) localise to airway nerves in humans with asthma (A) and to airway cholinergic nerves in antigen challenged guinea pigs (B –D). Eosinophils were identified using an antibody to MBP in (A), which also showed that extracellular MBP is also seen in seen association with nerves, as demonstrated with the dashed arrow.
(PKC) pathway. Inhibiting the PKC pathway in eosinophils with the compound, rottlerin, completely prevents adhesion to IMR32 cholinergic nerve cells. These data suggest that eosinophil adhesion occurs through the VLA-4/VCAM-1 pathway and that this in turn promotes eosinophil adhesion through CD11/18 interactions with ICAM-1. These data are consistent with the observation that in antigen-challenged guinea pigs preventing VLA-4/VCAM-1 interactions reduces eosinophil recruitment to airway nerves and the consequent loss of M2 muscarinic receptor function. As primary cultures of parasympathetic nerves are known to require cytokine stimulation before expressing VCAM-1, one mechanism of localisation of eosinophils to nerves is that antigen challenge, which is associated with the production of pro-inflammatory cytokines, leads to increased neural VCAM-1 expression and this in turn leads to eosinophil adhesion. Of course, the finding that eosinophils adhere to nerves does not in itself explain how eosinophils become activated and release MBP onto neuronal M2 muscarinic receptors. We therefore investigated whether or not localisation or adhesion of eosinophils to nerves leads directly to eosinophil activation. Co-culture of freshly isolated human eosinophils with IMR32 cholinergic nerve cells leads to the release of eosinophil peroxidase and leukotriene C4 (LtC4). This process of eosinophil activation is dependent on the adhesion of eosinophils to nerves; it is inhibited by paraformaldehyde fixation of the nerves (which does not in itself prevent adhesion), indicating that eosinophil activation is mediated by a factor that requires live nerve cells [28]. Eosinophils that adhere to nerves also show
increased responses to the secretogogue fMLP, suggesting that adhesion leads to eosinophil priming. Priming occurs even when eosinophils adhere to paraformaldehyde-fixed nerve cells. Thus, adhesion to IMR32 nerve cells leads to both autocrine eosinophil priming and a neurally mediated eosinophil activation. This neurally mediated activation and degranulation was associated with adhesion between CD11/18 on the eosinophils and ICAM-1 on the nerves. It has recently been shown in endothelial cells that ICAM-1 is linked to activation of NADPH and the generation of reactive oxidant species [29]. We investigated the possibility that eosinophil adhesion to ICAM-1 may lead to the activation of NADPH, which in turn may release factors from the nerves that then cause eosinophil degranulation. In these studies we demonstrated that adhesion of eosinophils to ICAM-1 induces the generation of reactive oxidant species within IMR32 cholinergic nerves through the activity of NADPH oxidase. Inhibition of NADPH oxidase and the formation of reactive oxidant species prevent eosinophil degranulation. Thus, adhesion triggers a cascade of events within nerves that lead to the generation of reactive oxidant species and promote eosinophil degranulation. Other events that occur within the nerves subsequent to adhesion and activation of reactive oxidant species include the generation of the nuclear transcription factor, NFkB. This may be particularly relevant as NFkB plays a role in neurodegenerative diseases and in some neural tissue is anti-apoptotic. These observations are outlined in Fig. 2. Since our initial observations concerning eosinophil and nerve interactions were indirect, we have recently looked
12
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13
M2 muscarinic receptors. The functional significance of these observations on possible neural plasticity and vagal hyperreactivity will require further in vivo studies.
Acknowledgements The authors acknowledge the support of the Wellcome Trust.
References
Fig. 2. Eosinophil and nerve interactions in vivo lead to changes in both eosinophil and nerve function. Adhesion via specific adhesion molecules leads to the generation of intraneuronal free radicals that are released from nerves and lead to eosinophil activation. This eosinophil activation leads to the loss of function of neuronal M2 muscarinic receptors. In the nerves, eosinophil adhesion nerves leads to the activation of the nuclear transcription factors NFkB and AP-1 and the subsequent activation of several genes including the M2 muscarinic receptor and alterations in nerve growth.
directly at the functional consequences of eosinophil adhesion on cholinergic nerve function. IMR32 cells release acetylcholine and express M2 muscarinic receptors when differentiated with dibutyryl cyclic AMP, which is why they are a useful model of airway parasympathetic nerves. In these studies we incubated eosinophils with IMR32 cells and demonstrated that this leads to a dose-dependent increase in acetylcholine release. This increase in acetylcholine release is due to M2 muscarinic receptor dysfunction since acute desensitisation of the receptor with a high concentration of a muscarinic receptor agonist prevents the effect. Thus, eosinophils can have a direct effect on cholinergic nerve function exemplified by an increase in acetylcholine release. In summary, histological studies have demonstrated that in antigen-challenged guinea pigs and rats, as well as in sections of airways from patients who died during an asthma attack, eosinophils localise to airway nerves. This process is associated with activation of eosinophils as shown by the presence of increased amounts of extracellular MBP. Functional studies in animal models have demonstrated that inhibiting eosinophils and in particular MBP prevents antigen-induced hyperreactivity. Using cell culture systems we have shown that eosinophils adhere to nerves via specific adhesion molecules and this leads to the generation of reactive oxidant species within nerves and the activation of NFkB. The reactive oxidant species generated in the nerves also lead to degranulation of eosinophils and loss of function
[1] Busse WW, Lemanske Jr RF. Asthma. N Engl J Med 2001;344: 350 –62. [2] Bousquet J, Chanez P, Lacoste JY, Barneon G, Ghavanian N, Enander I, Venge P, Ahlstedt S, Simony-Lafontaine J, Godard P, et al. Eosinophilic inflammation in asthma. N Engl J Med 1990;323: 1033–9. [3] Filley WV, Holley KE, Kephart GM, Gleich GJ. Identification by immunofluorescence of eosinophil granule major basic protein in lung tissues of patients with bronchial asthma. Lancet 1982;2:11–16. [4] Gleich GJ. Mechanisms of eosinophil-associated inflammation. J Allergy Clin Immunol 2000;105:651–63. [5] Wang JH, Trigg CJ, Devalia JL, Jordan S, Davies RJ. Effect of inhaled beclomethasone dipropionate on expression of proinflammatory cytokines and activated eosinophils in the bronchial epithelium of patients with mild asthma. J Allergy Clin Immunol 1994;94:1025 –34. [6] Trigg CJ, Manolitsas ND, Wang J, Calderon MA, McAulay A, Jordan SE, Herdman MJ, Jhalli N, Duddle JM, Hamilton SA, et al. Placebocontrolled immunopathologic study of four months of inhale corticosteroids in asthma. Am J Respir Crit Care Med 1994;150: 17– 22. [7] Woolley KL, Adelroth E, Woolley MJ, Ellis R, Jordana M, O’Byrne PM. Effects of allergen challenge on eosinophils, eosinophil cationic protein, and granulocyte-macrophage colony-stimulating factor in mild asthma. Am J Respir Crit Care Med 1995;151:1915–24. [8] Khan LN, Kon OM, Macfarlane AJ, Meng Q, Ying S, Barnes NC, Kay AB. Attenuation of the allergen-induced late asthmatic reaction by cyclosporin A is associated with inhibition of bronchial eosinophils, interleukin-5, granulocyte macrophage colony-stimulating factor, and eotaxin. Am J Respir Crit Care Med 2000;162:1377–82. [9] Bryan SA, O’Connor BJ, Matti S, Leckie MJ, Kanabar V, Khan J, Warrington SJ, Renzetti L, Rames A, Bock JA, Boyce MJ, Hansel TT, Holgate ST, Barnes PJ. Effects of recombinant human interleukin-12 on eosinophils, airway hyper-responsiveness, and the late asthmatic response. Lancet 2000;356:2149. [10] Leckie MJ, ten Brinke A, Khan J, Diamant Z, O’Connor BJ, Walls CM, Mathur AK, Cowley HC, Chung KF, Djukanovic R, Hansel TT, Holgate ST, Sterk PJ, Barnes PJ. Effects of an interleukin-5 blocking monoclonal antibody on eosinophils, airway hyper-responsiveness, and the late asthmatic response. Lancet 2000;356:2144–8. [11] Bartemes KR, McKinney S, Gleich GJ, Kita H. Endogenous plateletactivating factor is critically involved in effector functions of eosinophils stimulated with IL-5 or IgG. J Immunol 1999;162: 2982–9. [12] Kobayashi H, Gleich GJ, Butterfield JH, Kita H. Human eosinophils produce neurotrophins and secrete nerve growth factor on immunologic stimuli. Blood 2002;99:2214–20. [13] Blaber LC, Fryer AD, Maclagan J. Neuronal muscarinic receptors attenuate vagally-induced contraction of feline bronchial smooth muscle. Br J Pharmacol 1985;86:723–8. [14] Blaber LC, Fryer AD. The response of cat airways to histamine in vivo and in vitro. Br J Pharmacol 1985;84:309–16.
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13 [15] Fryer AD, Maclagan J. Muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guinea-pig. Br J Pharmacol 1984;83: 973–8. [16] Fryer AD, Jacoby DB. Parainfluenza virus infection damages inhibitory M2 muscarinic receptors on pulmonary parasympathetic nerves in the guinea-pig. Br J Pharmacol 1991;102:267 –71. [17] Fryer AD, Wills-Karp M. Dysfunction of M2-muscarinic receptors in pulmonary parasympathetic nerves after antigen-challenge. J Appl Physiol 1991;71:2255–61. [18] Elbon CL, Jacoby DB, Fryer AD. Pretreatment with an antibody to interleukin-5 prevents loss of pulmonary M2 muscarinic receptor function in antigen-challenged guinea pigs. Am J Respir Cell Mol Biol 1995;12:320–8. [19] Fryer AD, Elbon CL, Km AL, Xiao HQ, Levey AI, Jacoby DB. Cultures of airway parasympathetic nerves express functional M2 muscarinic receptors. Am J Respir Cell Mol Biol 1996;15: 716 – 25. [20] Evans CM, Fryer AD, Jacoby DB, Gleich GJ, Costello RW. Pretreatment with antibody to esosinophil major basic protein prevents hyperresponsiveness by protecting neuronal M2 muscarinic receptors in antigen-challenged guinea pigs. J Clin Invest 1997;100: 2254– 62. [21] Belmonte KE, Jacoby DB, Fryer AD. Increased function of inhibitory neuronal M2 muscarinic receptors in diabetic rat lungs. Br J Pharmacol 1997;121:1287– 94. [22] Costello RW, Schofield BH, Kephart GM, Gleich GJ, Jacoby DB, Fryer AD. Localization of eosinophils to airway nerves and effect on
[23]
[24]
[25]
[26]
[27]
[28]
[29]
13
neuronal M2 muscarinic receptor function. Am J Physiol 1997;273: L93– L103. Costello RW, Evans CM, Yost BL, Belmonte KE, Gleich GJ, Jacoby DB, Fryer AD. Antigen-induced hyperreactivity to histamine; role of the vagus nerves eosinophils. Am J Physiol 1999;276:L709–14. Jacoby DB, Gleich GJ, Fryer AD. Human eosinophil major basic protein is an endogenous allosteric antagonist at the inhibitory muscarinic M2 receptor. J Clin Invest 1993;91:1314–8. Fryer AD, Jacoby DB. Function of pulmonary M2 muscarinic receptors in antigen-challenged guinea pigs is restored by heparin poly-L-glutamate. J Clin Invest 1992;90:2292–8. Fryer AD, Costello RW, Yost BL, Lobb RR, Tedder TF, Steeber DA, Bochner BS. Antibody to VLA-4, but not to L-selection, protects neuronal M2 muscarinic receptors in antigen-challenged guinea pig airways. J Clin Invest 1997;99:2036 –44. Sawatzky D, Court E, Kingham PJ, Jacoby DB, Krumanwal B, Fryer AD, McLean WG, Costello RW. Eosinophil adhesion to cholinergic nerves via ICAM-1 and VCAM-1. Am J Physiol Lung Cell Mol Physiol 2002;282:L1279–88. Kingham P, Walsh MT, Sawatzky D, McLean WG, Costello RW. Adhesion dependent interactions between eosinophils and cholinergic nerves. Am J Physiol Lung Cell Mol Physiol 2002;282:L1229 –38. Wang Q, Doerschuk CM. The p38 mitogen-activated protein kinase mediates cytoskeletal remodeling in pulmonary microvascular endothelial cells upon intracellular adhesion molecule-1 ligation. J Immunol 2001;166:6877–84.
Eosinophil and airway nerve interactions P.J. Kinghamb, Richard W. Costelloa,*, W. Graham McLeanb a
Department of Medicine, Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin 9, Ireland b Department of Pharmacology, University of Liverpool, Liverpool, UK Received 21 March 2002; accepted 2 April 2002
Abstract In vivo, eosinophils localise to airway nerves in patients with asthma as well as in animal models of hyperreactivity. In both, in vivo and in vitro studies, we have shown that this localisation changes both cholinergic nerve and eosinophil function. In particular, it leads to an increase in acetylcholine release due to loss of function of a neuronal autoreceptor, the M2 muscarinic receptor. This loss of M2 receptor function occurs because eosinophils become activated and degranulate as a result of interactions that occur via specific adhesion molecules expressed on nerves that are recognised by counter ligands on eosinophils. q 2003 Elsevier Science Ltd. All rights reserved. Keywords: Asthma; Hyperreactivity neuro-immunology
1. Eosinophils: to have and to hold Asthma is characterised by the presence of a variety of inflammatory cells, including lymphocytes and eosinophils [1]. How these cells are linked to the clinical features of asthma has been the subject of intense investigation over the last twenty years. Histological as well as functional studies suggest that eosinophils, in particular, play a pivotal role in the pathogenesis of asthma. For example, increased numbers of eosinophils are found in bronchoalveolar lavage fluid, induced sputum and bronchial wall biopsies of patients with asthma [2]. Histological examination of sections of airway from patients who have died during an asthma attack show increased numbers of activated, degranulating eosinophils along with the presence of released toxic mediators [3,4]. Similarly, resolution of asthma symptoms is accompanied by a reduction in the number of eosinophils in the airways [5 – 8]. Eosinophils have a wide range of functions. For example, they release toxic cationic proteins such as eosinophil major basic protein (MBP), eosinophil peroxidase, eosinophilderived neurotoxin and eosinophil cationic protein. Eosinophils also generate lipid molecules such as platelet-activating factor, prostaglandin E2 and 5-hydroxyeicosa-6,8,11,14tetraenoic acid (5HETE) [1]. Studies in cell culture and in * Corresponding author. Tel.: þ 353-1-8093762; fax: þ 353-1-8093765. E-mail address: [email protected] (R.W. Costello).
animal models have shown that many of these compounds are capable of mimicking the features of asthma. Based on these observations it has been concluded, not unreasonably, that eosinophils are key effector cells in asthma. Eosinophils can also release compounds that activate other inflammatory cells; they are a source of both TH1 and TH2 cytokines, of TNF-a, and of several chemokines. Additionally, eosinophils can function as antigen-presenting cells and they may play a role in tissue repair, for example by secretion of gelatinase-B and transforming growth factor-b (TGF-b). Thus, the possibility remains that eosinophils are playing a role in the repair of damaged tissue. Alternatively, they are innocent bystander cells recruited to the airways as a consequence of activation of CD4 TH2-cells and they play no specific role in the condition.
2. Location Two recent studies, using two different monoclonal antibodies to disrupt eosinophil trafficking to the airways, have demonstrated no significant effect on the late phase response to antigen in sensitive asthmatics although, in both cases, sputum eosinophilia was reduced [9,10]. These studies have raised questions concerning the pathogenic role of eosinophils in asthma. However, it is important to note that eosinophils in the tissues behave differently from those in the airways. In the airways eosinophils interact with
1094-5539/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. PII: S 1 0 9 4 - 5 5 3 9 ( 0 2 ) 0 0 0 9 3 - 7
10
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13
the surrounding tissues through specific surfaces receptors and these interactions lead to significant functional changes. For example, recent studies have shown that eosinophils can avoid both apoptosis and the requirement for otherwise necessary growth factors such as interleukin (IL)-5 by adhesion molecule-dependent autocrine and paracrine generation of growth factors [11,12]. Thus, any investigation into the role of eosinophils in asthma requires a knowledge of the nature of the interactions that occur between eosinophils and airway cells. Included among those cells are the nerves that innervate the airways. Recent studies from our laboratory have focused on the nature of the interactions that occur between eosinophils and airway cholinergic nerves.
3. Parasympathetic nerves in the airways The parasympathetic nerves provide the dominant innervation of airway smooth muscle. Long preganglionic vagal fibres project from nuclei in the brain stem into the thorax. There they relay onto ganglia located on the posterior aspect of the trachea and major bronchi. Short postganglionic fibres arise from these ganglia to innervate the airway smooth muscle, mucous glands and bronchial and pulmonary arterioles. Stimulation of these parasympathetic nerves releases acetycholine which binds to M3 muscarinic receptors on these structures, leading to bronchoconstriction, mucous production and dilation of the bronchial vessels. These nerves provide a protective mechanism for the airways against potentially harmful inhaled toxins, as bronchoconstriction induces turbulent, as opposed to laminar, airflow. This change in airflow causes particles to be deposited in the larger airways where they are trapped by simultaneously generated mucous. Thus, the vagus nerves provide an important respiratory defence mechanism. However, when nerve activity is excessive it may be harmful by causing excessive bronchoconstriction and mucous production, which can also obstruct airflow. Therefore, a system exists to control this activity. In addition to its effect on M3 receptors acetylcholine released from parasympathetic nerves also acts on neuronal M2 muscarinic receptors. Stimulation of these neuronal M2 muscarinic receptors limits further acetylcholine release [13 –15]. It follows that loss of function of these M2 receptors leads to an increase in vagally induced bronchoconstriction and mucous production. We and others have demonstrated in animal models and in some patients with asthma that there is indeed loss of function of these receptors [16 – 22]. For example, in antigen-sensitised animals exposed to further antigen there is an increase in the magnitude of vagally induced bronchoconstriction and hyperreactivity to compounds such as histamine [23]. The mechanism for this loss of M2 muscarinic receptor function appears to be due to the action of MBP. Ligand-binding studies have demonstrated that
eosinophil MBP is an allosteric antagonist at M2 but not at M3 muscarinic receptors [24]. In vitro, anionic compounds such as heparin displace MBP from M2 receptors, suggesting that the cationic charge of MBP is responsible for the inhibition of these receptors. Conversely, in vivo, administration of heparin to antigen-challenged guinea pigs restores M2 receptor function [25]. Furthermore, inhibiting eosinophil recruitment to the airways with an antibody to the eosinophil adhesion molecule VLA-4, depleting eosinophils with an antibody to the eosinophil growth factor IL5, or specifically inhibiting MBP all preserve M2 receptor function [18,20,26]. These data indicate that significant interactions occur between eosinophils and in particular, eosinophil MBP, and cholinergic nerves. MBP is a highly cationic protein that does not readily diffuse once released from eosinophil stores. Thus, we first had to investigate whether or not eosinophils are capable of delivering MBP to cholinergic nerves. In histological studies performed on sections of airway from antigenchallenged guinea pigs, rats and patients who died during an asthma attack, eosinophils were seen to localise to airway cholinergic nerves [22]. In addition, there was frequent evidence of eosinophil degranulation and release of MBP close to these nerves, as shown in Fig. 1. We then undertook further work to investigate the fine details of the interactions of eosinophils with airway nerves, to determine the mechanism of localisation of eosinophils to nerves and to assess the consequences for eosinophil and nerve function once these cells co-localise.
4. Eosinophils localisation to airway nerves via specific adhesion molecules One mechanism whereby eosinophils may localise to airway nerves is that they recognise complementary neuronal adhesion molecules. Eosinophils have an extensive repertoire of adhesion molecules that mediate both initial light ligation and more firm adhesion. For eosinophils this later adhesion is mediated through VLA-4 and CD11/18 recognising VCAM-1 and ICAM-1, respectively. Both VCAM-1 and ICAM-1 are expressed by primary cultures of guinea pig parasympathetic nerves and by the human cholinergic nerve cell line IMR32. In the primary cultures, ICAM-1 expression is constitutive while VCAM-1 requires cytokine pre-treatment in primary culture, while in the cell line both are constitutively expressed [27]. We have used a chromium radiolabelling technique and functional measurements in both systems to demonstrate that inhibiting either VCAM-1 or ICAM-1 can completely prevent adhesion. Similarly, pre-treatment of the nerve cells with heparin inhibits adhesion. The fact that one or other adhesion molecule is sufficient for adhesion suggested that there might be a common process linking the two adhesion molecules. It is known that ligation of VLA-4 leads to a process of inside-out signalling through the protein kinase C
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13
11
Fig. 1. Eosinophils (shown with arrows) localise to airway nerves in humans with asthma (A) and to airway cholinergic nerves in antigen challenged guinea pigs (B –D). Eosinophils were identified using an antibody to MBP in (A), which also showed that extracellular MBP is also seen in seen association with nerves, as demonstrated with the dashed arrow.
(PKC) pathway. Inhibiting the PKC pathway in eosinophils with the compound, rottlerin, completely prevents adhesion to IMR32 cholinergic nerve cells. These data suggest that eosinophil adhesion occurs through the VLA-4/VCAM-1 pathway and that this in turn promotes eosinophil adhesion through CD11/18 interactions with ICAM-1. These data are consistent with the observation that in antigen-challenged guinea pigs preventing VLA-4/VCAM-1 interactions reduces eosinophil recruitment to airway nerves and the consequent loss of M2 muscarinic receptor function. As primary cultures of parasympathetic nerves are known to require cytokine stimulation before expressing VCAM-1, one mechanism of localisation of eosinophils to nerves is that antigen challenge, which is associated with the production of pro-inflammatory cytokines, leads to increased neural VCAM-1 expression and this in turn leads to eosinophil adhesion. Of course, the finding that eosinophils adhere to nerves does not in itself explain how eosinophils become activated and release MBP onto neuronal M2 muscarinic receptors. We therefore investigated whether or not localisation or adhesion of eosinophils to nerves leads directly to eosinophil activation. Co-culture of freshly isolated human eosinophils with IMR32 cholinergic nerve cells leads to the release of eosinophil peroxidase and leukotriene C4 (LtC4). This process of eosinophil activation is dependent on the adhesion of eosinophils to nerves; it is inhibited by paraformaldehyde fixation of the nerves (which does not in itself prevent adhesion), indicating that eosinophil activation is mediated by a factor that requires live nerve cells [28]. Eosinophils that adhere to nerves also show
increased responses to the secretogogue fMLP, suggesting that adhesion leads to eosinophil priming. Priming occurs even when eosinophils adhere to paraformaldehyde-fixed nerve cells. Thus, adhesion to IMR32 nerve cells leads to both autocrine eosinophil priming and a neurally mediated eosinophil activation. This neurally mediated activation and degranulation was associated with adhesion between CD11/18 on the eosinophils and ICAM-1 on the nerves. It has recently been shown in endothelial cells that ICAM-1 is linked to activation of NADPH and the generation of reactive oxidant species [29]. We investigated the possibility that eosinophil adhesion to ICAM-1 may lead to the activation of NADPH, which in turn may release factors from the nerves that then cause eosinophil degranulation. In these studies we demonstrated that adhesion of eosinophils to ICAM-1 induces the generation of reactive oxidant species within IMR32 cholinergic nerves through the activity of NADPH oxidase. Inhibition of NADPH oxidase and the formation of reactive oxidant species prevent eosinophil degranulation. Thus, adhesion triggers a cascade of events within nerves that lead to the generation of reactive oxidant species and promote eosinophil degranulation. Other events that occur within the nerves subsequent to adhesion and activation of reactive oxidant species include the generation of the nuclear transcription factor, NFkB. This may be particularly relevant as NFkB plays a role in neurodegenerative diseases and in some neural tissue is anti-apoptotic. These observations are outlined in Fig. 2. Since our initial observations concerning eosinophil and nerve interactions were indirect, we have recently looked
12
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13
M2 muscarinic receptors. The functional significance of these observations on possible neural plasticity and vagal hyperreactivity will require further in vivo studies.
Acknowledgements The authors acknowledge the support of the Wellcome Trust.
References
Fig. 2. Eosinophil and nerve interactions in vivo lead to changes in both eosinophil and nerve function. Adhesion via specific adhesion molecules leads to the generation of intraneuronal free radicals that are released from nerves and lead to eosinophil activation. This eosinophil activation leads to the loss of function of neuronal M2 muscarinic receptors. In the nerves, eosinophil adhesion nerves leads to the activation of the nuclear transcription factors NFkB and AP-1 and the subsequent activation of several genes including the M2 muscarinic receptor and alterations in nerve growth.
directly at the functional consequences of eosinophil adhesion on cholinergic nerve function. IMR32 cells release acetylcholine and express M2 muscarinic receptors when differentiated with dibutyryl cyclic AMP, which is why they are a useful model of airway parasympathetic nerves. In these studies we incubated eosinophils with IMR32 cells and demonstrated that this leads to a dose-dependent increase in acetylcholine release. This increase in acetylcholine release is due to M2 muscarinic receptor dysfunction since acute desensitisation of the receptor with a high concentration of a muscarinic receptor agonist prevents the effect. Thus, eosinophils can have a direct effect on cholinergic nerve function exemplified by an increase in acetylcholine release. In summary, histological studies have demonstrated that in antigen-challenged guinea pigs and rats, as well as in sections of airways from patients who died during an asthma attack, eosinophils localise to airway nerves. This process is associated with activation of eosinophils as shown by the presence of increased amounts of extracellular MBP. Functional studies in animal models have demonstrated that inhibiting eosinophils and in particular MBP prevents antigen-induced hyperreactivity. Using cell culture systems we have shown that eosinophils adhere to nerves via specific adhesion molecules and this leads to the generation of reactive oxidant species within nerves and the activation of NFkB. The reactive oxidant species generated in the nerves also lead to degranulation of eosinophils and loss of function
[1] Busse WW, Lemanske Jr RF. Asthma. N Engl J Med 2001;344: 350 –62. [2] Bousquet J, Chanez P, Lacoste JY, Barneon G, Ghavanian N, Enander I, Venge P, Ahlstedt S, Simony-Lafontaine J, Godard P, et al. Eosinophilic inflammation in asthma. N Engl J Med 1990;323: 1033–9. [3] Filley WV, Holley KE, Kephart GM, Gleich GJ. Identification by immunofluorescence of eosinophil granule major basic protein in lung tissues of patients with bronchial asthma. Lancet 1982;2:11–16. [4] Gleich GJ. Mechanisms of eosinophil-associated inflammation. J Allergy Clin Immunol 2000;105:651–63. [5] Wang JH, Trigg CJ, Devalia JL, Jordan S, Davies RJ. Effect of inhaled beclomethasone dipropionate on expression of proinflammatory cytokines and activated eosinophils in the bronchial epithelium of patients with mild asthma. J Allergy Clin Immunol 1994;94:1025 –34. [6] Trigg CJ, Manolitsas ND, Wang J, Calderon MA, McAulay A, Jordan SE, Herdman MJ, Jhalli N, Duddle JM, Hamilton SA, et al. Placebocontrolled immunopathologic study of four months of inhale corticosteroids in asthma. Am J Respir Crit Care Med 1994;150: 17– 22. [7] Woolley KL, Adelroth E, Woolley MJ, Ellis R, Jordana M, O’Byrne PM. Effects of allergen challenge on eosinophils, eosinophil cationic protein, and granulocyte-macrophage colony-stimulating factor in mild asthma. Am J Respir Crit Care Med 1995;151:1915–24. [8] Khan LN, Kon OM, Macfarlane AJ, Meng Q, Ying S, Barnes NC, Kay AB. Attenuation of the allergen-induced late asthmatic reaction by cyclosporin A is associated with inhibition of bronchial eosinophils, interleukin-5, granulocyte macrophage colony-stimulating factor, and eotaxin. Am J Respir Crit Care Med 2000;162:1377–82. [9] Bryan SA, O’Connor BJ, Matti S, Leckie MJ, Kanabar V, Khan J, Warrington SJ, Renzetti L, Rames A, Bock JA, Boyce MJ, Hansel TT, Holgate ST, Barnes PJ. Effects of recombinant human interleukin-12 on eosinophils, airway hyper-responsiveness, and the late asthmatic response. Lancet 2000;356:2149. [10] Leckie MJ, ten Brinke A, Khan J, Diamant Z, O’Connor BJ, Walls CM, Mathur AK, Cowley HC, Chung KF, Djukanovic R, Hansel TT, Holgate ST, Sterk PJ, Barnes PJ. Effects of an interleukin-5 blocking monoclonal antibody on eosinophils, airway hyper-responsiveness, and the late asthmatic response. Lancet 2000;356:2144–8. [11] Bartemes KR, McKinney S, Gleich GJ, Kita H. Endogenous plateletactivating factor is critically involved in effector functions of eosinophils stimulated with IL-5 or IgG. J Immunol 1999;162: 2982–9. [12] Kobayashi H, Gleich GJ, Butterfield JH, Kita H. Human eosinophils produce neurotrophins and secrete nerve growth factor on immunologic stimuli. Blood 2002;99:2214–20. [13] Blaber LC, Fryer AD, Maclagan J. Neuronal muscarinic receptors attenuate vagally-induced contraction of feline bronchial smooth muscle. Br J Pharmacol 1985;86:723–8. [14] Blaber LC, Fryer AD. The response of cat airways to histamine in vivo and in vitro. Br J Pharmacol 1985;84:309–16.
P.J. Kingham et al. / Pulmonary Pharmacology & Therapeutics 16 (2003) 9–13 [15] Fryer AD, Maclagan J. Muscarinic inhibitory receptors in pulmonary parasympathetic nerves in the guinea-pig. Br J Pharmacol 1984;83: 973–8. [16] Fryer AD, Jacoby DB. Parainfluenza virus infection damages inhibitory M2 muscarinic receptors on pulmonary parasympathetic nerves in the guinea-pig. Br J Pharmacol 1991;102:267 –71. [17] Fryer AD, Wills-Karp M. Dysfunction of M2-muscarinic receptors in pulmonary parasympathetic nerves after antigen-challenge. J Appl Physiol 1991;71:2255–61. [18] Elbon CL, Jacoby DB, Fryer AD. Pretreatment with an antibody to interleukin-5 prevents loss of pulmonary M2 muscarinic receptor function in antigen-challenged guinea pigs. Am J Respir Cell Mol Biol 1995;12:320–8. [19] Fryer AD, Elbon CL, Km AL, Xiao HQ, Levey AI, Jacoby DB. Cultures of airway parasympathetic nerves express functional M2 muscarinic receptors. Am J Respir Cell Mol Biol 1996;15: 716 – 25. [20] Evans CM, Fryer AD, Jacoby DB, Gleich GJ, Costello RW. Pretreatment with antibody to esosinophil major basic protein prevents hyperresponsiveness by protecting neuronal M2 muscarinic receptors in antigen-challenged guinea pigs. J Clin Invest 1997;100: 2254– 62. [21] Belmonte KE, Jacoby DB, Fryer AD. Increased function of inhibitory neuronal M2 muscarinic receptors in diabetic rat lungs. Br J Pharmacol 1997;121:1287– 94. [22] Costello RW, Schofield BH, Kephart GM, Gleich GJ, Jacoby DB, Fryer AD. Localization of eosinophils to airway nerves and effect on
[23]
[24]
[25]
[26]
[27]
[28]
[29]
13
neuronal M2 muscarinic receptor function. Am J Physiol 1997;273: L93– L103. Costello RW, Evans CM, Yost BL, Belmonte KE, Gleich GJ, Jacoby DB, Fryer AD. Antigen-induced hyperreactivity to histamine; role of the vagus nerves eosinophils. Am J Physiol 1999;276:L709–14. Jacoby DB, Gleich GJ, Fryer AD. Human eosinophil major basic protein is an endogenous allosteric antagonist at the inhibitory muscarinic M2 receptor. J Clin Invest 1993;91:1314–8. Fryer AD, Jacoby DB. Function of pulmonary M2 muscarinic receptors in antigen-challenged guinea pigs is restored by heparin poly-L-glutamate. J Clin Invest 1992;90:2292–8. Fryer AD, Costello RW, Yost BL, Lobb RR, Tedder TF, Steeber DA, Bochner BS. Antibody to VLA-4, but not to L-selection, protects neuronal M2 muscarinic receptors in antigen-challenged guinea pig airways. J Clin Invest 1997;99:2036 –44. Sawatzky D, Court E, Kingham PJ, Jacoby DB, Krumanwal B, Fryer AD, McLean WG, Costello RW. Eosinophil adhesion to cholinergic nerves via ICAM-1 and VCAM-1. Am J Physiol Lung Cell Mol Physiol 2002;282:L1279–88. Kingham P, Walsh MT, Sawatzky D, McLean WG, Costello RW. Adhesion dependent interactions between eosinophils and cholinergic nerves. Am J Physiol Lung Cell Mol Physiol 2002;282:L1229 –38. Wang Q, Doerschuk CM. The p38 mitogen-activated protein kinase mediates cytoskeletal remodeling in pulmonary microvascular endothelial cells upon intracellular adhesion molecule-1 ligation. J Immunol 2001;166:6877–84.