Abstract

Airway remodelling and emphysema are major structural abnormalities in chronic obstructive pulmonary disease (COPD). In addition, pulmonary vascular remodelling may occur and contribute to pulmonary hypertension, a comorbidity of COPD. Increased cholinergic activity in COPD contributes to airflow limitation and, possibly, to inflammation and airway remodelling.

This study aimed to investigate the role of acetylcholine in pulmonary inflammation and remodelling using an animal model of COPD. To this aim, guinea pigs were instilled intranasally with lipopolysaccharide (LPS) twice weekly for 12 weeks and were treated, by inhalation, with the long-acting muscarinic receptor antagonist tiotropium.

Repeated LPS exposure induced airway and parenchymal neutrophilia, and increased goblet cell numbers, lung hydroxyproline content, airway wall collagen and airspace size. Furthermore, LPS increased the number of muscularised microvessels in the adventitia of cartilaginous airways. Tiotropium abrogated the LPS-induced increase in neutrophils, goblet cells, collagen deposition and muscularised microvessels, but had no effect on emphysema.

In conclusion, tiotropium inhibits remodelling of the airways as well as pulmonary inflammation in a guinea pig model of COPD, suggesting that endogenous acetylcholine plays a major role in the pathogenesis of this disease.

Chronic obstructive pulmonary disease (COPD) is an inflammatory disease, characterised by a progressive decline in lung function and airflow limitation that is not fully reversible. Structural features of COPD, which contribute to the airflow limitation, include emphysema and airway remodelling, characterised by mucous cell hyperplasia and airway fibrosis [1]. In addition, pulmonary vascular remodelling has been observed, even in patients with mild COPD [2, 3]. The remodelling of pulmonary vessels in COPD is characterised by smooth muscle proliferation, which contributes to thickening of the vessel wall of arteries [4, 5]. Furthermore, there is evidence of muscularisation of small vessels, which do not have a smooth muscle layer under healthy conditions [5, 6]. Although the mechanisms of tissue remodelling in COPD are largely unclear, chronic pulmonary inflammation, characterised by infiltration of neutrophils, macrophages, CD4- and CD8-positive T-lymphocytes, and B-cells, is presumably of major importance [7].

Anticholinergics are indicated as bronchodilator therapy in COPD [8]. However, recent reports indicate that anticholinergics may have effects beyond bronchodilation. Thus, the recent UPLIFT study (Understanding Potential Long-term Impacts on Function with Tiotropium) [911] has shown that the use of the long-acting anticholinergic, tiotropium bromide, is associated with a reduction of the number of exacerbations, and of respiratory and cardiac morbidity and mortality of COPD patients [9, 10]. In addition, pre-specified subgroup analyses of the UPLIFT study have indicated protective effects of tiotropium on post-bronchodilator forced expiratory volume in 1 s (FEV1) decline in Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage II COPD patients [11].

The airways are primarily innervated by parasympathetic cholinergic neurons, which regulate airway smooth muscle (ASM) contraction and mucus secretion [1214]. Bronchomotor tone is increased in COPD, probably due to increased neuronal release of acetylcholine (ACh), which is the major reversible component of airflow obstruction in this disease [8, 12]. Accumulating evidence suggests that in addition to neuronal ACh, non-neuronal ACh may also play a role in the pathogenesis of COPD [12]. Thus, the ACh-synthesising enzyme choline acetyltransferase (ChAT), as well as muscarinic receptors, are expressed in both structural and inflammatory cells in the lung [12]. Remarkably, increased ChAT expression was found in lung fibroblasts from smokers and COPD patients [15]. Muscarinic M3 receptor stimulation has been shown to increase the release of neutrophil chemotactic activity by alveolar macrophages [16], induce interleukin (IL)-8 release by bronchial epithelial cells [17] and monocytes [18], and augment cigarette smoke-induced IL-8 release by ASM cells [19]. In addition, muscarinic receptor agonists were shown to stimulate or potentiate proliferation of lung fibroblasts [20] and ASM cells [21], as well as lung fibroblast collagen synthesis [22]. Collectively, these data suggest a pro-inflammatory and pro-remodelling modulatory role for ACh in the lung. Evidence for such a role has also been found in a guinea pig model of chronic allergic asthma. In this model, it was demonstrated that tiotropium reduces airway eosinophilia, as well as ASM remodelling and goblet cell hyperplasia, upon repeated allergen exposure [23, 24]. In addition, tiotropium has recently been shown to inhibit airway inflammation and remodelling in a mouse model of gastro-oesophageal reflux disease [25].

Currently, the role of ACh in the development and progression of COPD is unknown. In the present study, we addressed this question using a guinea pig model of lipopolysaccharide (LPS)-induced COPD. LPS is a component of the outer wall of Gram-negative bacteria, and a contaminant of organic dusts, environmental pollution and tobacco smoke [26], and has been associated with the development of COPD [27]. In addition, LPS may play an important role in bacterial infection-induced exacerbations of COPD, which contribute to the progression of the disease [28]. Accordingly, in various animal models, inhalation of LPS was shown to induce pathological features of COPD, such as neutrophilia, goblet cell hyperplasia, airway fibrosis and emphysema [2932]. In the present study, we present evidence that ACh plays a major role in the development of pulmonary inflammation as well as in airway remodelling.

METHODS

Animals

Outbred, male, specified pathogen-free Dunkin Hartley guinea pigs (Harlan, Heathfield, UK) weighing 350–400 g were used. All protocols described in this study were approved by the University of Groningen Committee for Animal Experimentation, Groningen, the Netherlands.

LPS instillation

Conscious guinea pigs were held in an upright position, while 200 μL LPS (5 mg·mL−1 in sterile saline) was slowly instilled intranasally. After the intranasally instilled solution was aspirated, the animals were kept in the upright position for an additional 2 min to allow sufficient spreading of the fluid throughout the airways. Control animals were instilled with 200 μL sterile saline.

Experimental protocol

Guinea pigs were challenged by intranasal instillation with either LPS or saline twice weekly, for 12 consecutive weeks. 30 min prior to each instillation, animals received a nebulised dose of tiotropium bromide (Boehringer Ingelheim, Ingelheim, Germany) in saline (0.1 mM for 3 min) or saline (3 min), using a DeVilbiss nebulizer (type 646; DeVilbiss Healthcare, Somerset, PA, USA) as described previously [33]. This dose of tiotropium has previously been shown to induce protection against inhaled methacholine that was measurable for ∼96 h [34]. 24 h after the last instillation, the guinea pigs were sacrificed by experimental concussion followed by rapid exsanguination. The lungs were immediately resected and kept on ice for further processing.

Tissue analysis

Transverse frozen cross-sections (4 μm) of the middle right lung lobe were used for histological and immunohistochemical analyses. To identify smooth muscle and goblet cells, sections were stained for smooth muscle myosin heavy chain (sm-MHC) or MUC5AC, respectively, as described previously [24]. Neutrophils were identified by staining sections for TNAP (tissue nonspecific alkaline phosphatase activity) [35]. For immunohistochemical stainings, primary antibodies (all from Neomarkers, Fremont, CA, USA) were visualised using horseradish peroxidase-linked secondary antibodies (Sigma, St Louis, MO, USA), diaminobenzidine (0.3 mg·mL−1) and ammonium nickel sulphate (25 mg·mL−1), adapted from Adams [36]. Sections were counterstained with haematoxylin. Airways within sections were digitally photographed and classified as cartilaginous or noncartilaginous. All immunohistochemical measurements were carried out digitally using quantification software (ImageJ; National Institutes for Health, Bethesda, MD, USA; http://rsb.info.nih.gov/ij/). For this purpose, the digital photographs were analysed at a magnification of 40–400×. For each animal, two lung sections were prepared per staining, in which a total of 2–6 airways of each classification were analysed. Airway neutrophils were counted in the adventitia and sub-mucosa, and expressed as the number of positively stained cells per millimetre basement membrane length. For parenchymal neutrophil counts, positively stained cells were counted in five random microscopic fields using an eyepiece graticule and expressed as a percentage of total cell counts. MUC5AC-positive cells were counted in the epithelial layer and expressed as number of cells per millimetre basement membrane length. Microvessels that stained positively for sm-MHC were counted in the adventitia of cartilaginous airways and were expressed as number of vessels per square millimetre adventitia.

The upper right lung lobe was removed, inflated and fixed with formalin at a constant pressure of 25 cmH2O for 24 h, and embedded in paraffin. 4-μm thick paraffin sections were cut. For evaluation of emphysema, paraffin sections were stained with haematoxylin and eosin. The mean linear intercept (MLI) was determined as a measure of alveolar airspace size, using 20–25 photomicroscopic images (magnification 200×) per animal, as described previously [37]. For evaluation of airway wall collagen, paraffin sections were stained with Sirius Red and counterstained with haematoxylin. Noncartilaginous airways were digitally photographed (magnification 100–200×) and the individual colour images were split into the red, green and blue channels. Using ImageJ, the black and white images from the green channel were converted to binary images using the threshold function. The positively stained area in the airway wall, from the adventitial border to the basement membrane, was determined in 2–6 airways of each animal. The airway wall collagen area was normalised to the square of the basement membrane length. For evaluation of pulmonary vascular remodelling, paraffin sections were stained with Weigert’s elastin (resorcin/fuchsin) and Van Gieson stain. The measurements were performed according to van Suylen et al. [38]. Total vessel area of pulmonary arteries was defined as the area within the lamina elastica externa, and lumen area was defined as the area within the lamina elastica interna. Medial area is the area between the lamina elastica interna and the lamina elastica externa. The medial area was normalised to the lumen area. Pulmonary arteries with an external diameter of 30–100 μm were analysed. The wall area of pulmonary arterioles was defined as the area between the lamina elastica externa and the lumen, and was normalised to the lumen area. Only vessels with the longest/shortest diameter ratio <2 were analysed.

Hydroxyproline assay

Lungs were analysed for hydroxyproline as an estimate of collagen content [39], as described previously [40]. Lung homogenates were prepared by pulverising tissue under liquid nitrogen, followed by sonification in PBS. Subsequently, homogenates were incubated with 5% v/v trichloroacetic acid on ice for 20 min. Samples were centrifuged and the pellet was resuspended in 10 mL of 12 N hydrochloric acid and heated overnight at 110°C. The samples were reconstituted in 2 mL water by incubating for 72 h at room temperature, applying intermittent vortexing. 5-μL samples were incubated with 100 μL chloramine T (1.4% chloramine T in 0.5 M sodium acetate/10% isopropanol) in a 96-well plate, for 30 min at room temperature. Next, 100 μL Ehrlich's solution (1.0 M 4-dimethylaminobenzaldehyde in 70% isopropanol/30% perchloric acid) was added and samples were incubated at 65°C for 30 min. Samples were cooled to room temperature and the amount of hydroxyproline was quantified by colorimetric measurement (550 nm; Biorad 680 plate reader; Biorad, Hercules, CA, USA) of a pyrrole derivative of hydroxyproline, which forms a chromophore with Erlich's reagent. Concentrations were calculated using a standard curve. Data are expressed as milligrammes of hydroxyproline per lung.

Statistical analysis

Data are presented as mean±sem. Unless otherwise specified, statistical differences between means were calculated using one-way ANOVA, followed by Boniferroni or Newman–Keuls multiple comparison test, as appropriate. Differences were considered significant at p<0.05.

RESULTS

Inflammation

The neutrophil is a major inflammatory cell involved in the pathogenesis of COPD. Repeated LPS instillation induced significant increases in the numbers of neutrophils in both cartilaginous (2.0-fold) and noncartilaginous (1.9-fold) airways as well as in the parenchyma (2.1-fold) (fig. 1). Tiotropium treatment fully inhibited the LPS-induced neutrophilia in these compartments. Neutrophil numbers were not affected by tiotropium in the airways or the parenchyma of saline-challenged animals. These data indicate that tiotropium has a profound anti-inflammatory activity in chronically LPS-instilled guinea pigs.

Figure 1–

Effects of repeated lipopolysaccharide (LPS) challenge and tiotropium treatment on neutrophil numbers in the a) cartilaginous airways, b) noncartilaginous airways, and c) parenchyma of guinea pig lung. Guinea pigs were instilled intranasally with LPS (1 mg per 200 μL) or sterile saline (200 μL) twice weekly for 12 weeks. 30 min prior to each instillation, animals inhaled a nebulised dose of tiotropium in saline (0.1 mM for 3 min) or saline (3 min). Tissue was collected 24 h after the last instillation. Data represent mean±sem of a and b) 8–10 or c) 5–6 experiments; 2–6 airways of each classification or five microscopic fields in the parenchyma were analysed for each animal. BM: basement membrane. *: p<0.05; **: p<0.01.

MUC5AC expression

In order to investigate the effects of LPS and tiotropium on mucus-producing goblet cells, sections were stained with a MUC5AC antibody. Repeated LPS instillation induced a significant 3.1-fold increase in the number of MUC5AC-positive cells in the epithelium of cartilaginous airways of the guinea pigs (fig. 2). Tiotropium treatment fully inhibited the LPS-induced increase, whereas it had no effect in saline-challenged animals.

Figure 2–

Effects of repeated lipopolysaccharide (LPS) challenge and tiotropium treatment on MUC5AC-positive goblet cell number in guinea pig intrapulmonary cartilaginous airways. Guinea pigs were instilled intranasally with LPS (1 mg per 200 μL) or sterile saline (200 μL) twice weekly for 12 weeks. 30 min prior to each instillation, animals inhaled a nebulised dose of tiotropium in saline (0.1 mM, 3 min) or saline (3 min). Tissue was collected 24 h after the last instillation. Data represent mean±sem of seven experiments; 2–5 airways were analysed for each animal. BM: basement membrane. **: p<0.01.

Airway fibrosis

To evaluate fibrotic changes, lungs were analysed for hydroxyproline as an estimate of collagen content. Repeated LPS instillation induced a significant 1.3-fold increase in total lung hydroxyproline content (fig. 3). To confirm that collagen deposition was indeed increased in the airway compartment, Sirius Red staining was evaluated in the airway wall of noncartilaginous airways. LPS induced a 1.7-fold increase in airway wall collagen content. Tiotropium fully inhibited the increase in hydroxyproline and airway wall collagen deposition induced by repeated LPS instillation, whereas it had no effect in saline-challenged animals.

Figure 3–

Effects of repeated lipopolysaccharide (LPS) challenge and tiotropium treatment on guinea pig a) whole lung hydroxyproline content and b) collagen content in noncartilaginous airways. Guinea pigs were instilled intranasally with LPS (1 mg per 200 μL) or sterile saline (200 μL) twice weekly for 12 weeks. 30 min prior to each instillation, animals inhaled a nebulised dose of tiotropium in saline (0.1 mM, 3 min) or saline (3 min). Tissue was collected 24 h after the last instillation. b) Airway wall collagen was determined as Sirius Red-positive area and normalised to the square of the basement membrane length (BM2). Data represent mean±sem of a) 10–12 or b) 5–6 experiments; b) 2–6 airways were analysed for each animal. *: p<0.05; **: p<0.01; ***: p<0.001.

Emphysema

In order to evaluate the alveolar airspace size, MLI was determined in paraffin-embedded lung sections. A 7.3% increase in MLI was observed after 12 weeks of twice-weekly LPS instillations (fig. 4). Tiotropium had no effect on the airspace size in either LPS- or saline-instilled animals.

Figure 4–

Effects of repeated lipopolysaccharide (LPS) challenge and tiotropium treatment on alveolar airspace size (mean linear intercept) in guinea pig lung. Guinea pigs were instilled intranasally with LPS (1 mg per 200 μL) or sterile saline (200 μL) twice weekly for 12 weeks. 30 min prior to each instillation, animals inhaled a nebulised dose of tiotropium in saline (0.1 mM, 3 min) or saline (3 min). Tissue was collected 24 h after the last instillation. Data represent mean±sem of 4–5 experiments. *: p<0.05; ns: not significant.

Vascular remodelling

To evaluate pulmonary vascular remodelling, pulmonary artery medial area and pulmonary arteriole wall area were determined in formalin-fixed, paraffin-embedded guinea pig lung sections stained with Weigert's elastin and Van Gieson stain. Neither repeated LPS instillation nor tiotropium treatment had an effect on the medial area of pulmonary arteries or wall area of pulmonary arterioles (fig. 5). In addition, there was no evidence of intimal proliferation in the pulmonary vessels of either classification. However, repeated LPS instillation increased the number of muscularised (sm-MHC-positive) microvessels in the adventitia of cartilaginous airways (2.4-fold) (fig. 6). This increase was fully inhibited by tiotropium.

Figure 5–

Effects of repeated lipopolysaccharide (LPS) challenge and tiotropium treatment on a) the medial area of pulmonary arteries and b) the wall area of pulmonary arterioles. Guinea pigs were instilled intranasally with LPS (1 mg per 200 μL) or sterile saline (200 μL) twice weekly for 12 weeks. 30 min prior to each instillation, animals inhaled a nebulised dose of tiotropium in saline (0.1 mM for 3 min) or saline (3 min). Tissue was collected 24 h after the last instillation. Data represent mean±sem of 6–8 experiments.

Figure 6–

Effects of repeated lipopolysaccharide (LPS) challenge and tiotropium treatment on the number of muscularised microvessels in the airway adventitia of cartilaginous airways in guinea pig lung. Guinea pigs were instilled intranasally with LPS (1 mg per 200 μL) or sterile saline (200 μL) twice weekly for 12 weeks. 30 min prior to each instillation, animals inhaled a nebulised dose of tiotropium in saline (0.1 mM for 3 min) or saline (3 min). Tissue was collected 24 h after the last instillation. Data represent mean±sem of 7–9 experiments. sm-MHC: smooth muscle myosin heavy chain. **: p<0.01.

DISCUSSION

In this study, we demonstrate for the first time that tiotropium inhalation inhibits neutrophilia, MUC5AC expression and airway fibrosis in an animal model of COPD induced by repeated LPS exposure. In addition, we showed that repeated LPS instillation induced remodelling of the adventitial airway vasculature in this model, which was inhibited by tiotropium. Collectively, these data suggest that endogenous ACh, acting through muscarinic receptors, plays a major role in pulmonary inflammation as well as in airway remodelling in COPD.

Neutrophil numbers are increased in sputum and bronchoalveolar lavage fluid of COPD patients [41]. In addition, correlation between COPD severity and neutrophil numbers in the large airways and the percentage of neutrophil-containing small airways has been reported [41]. ACh may contribute to neutrophilia as it was shown to stimulate the release of neutrophil chemotactic activity from isolated alveolar macrophages [16] and from isolated sputum cells of COPD patients [42]. Furthermore, activation of muscarinic receptors expressed in airway structural cells, including bronchial epithelial [17] and ASM cells [19], may also contribute to neutrophil sequestration in the lungs by inducing or augmenting IL-8 release by these cells. In addition, the high capacity of neutrophils to synthesise ACh [43] implies that neutrophilia may result in increased non-neuronal ACh release in the lung. In our study, repeated LPS exposure increased neutrophil numbers in the airways and in the parenchyma. Because tiotropium inhalation fully inhibited this LPS-induced neutrophilic inflammation, a major role for ACh is inferred.

Mucus hypersecretion is a characteristic feature of COPD that contributes significantly to airflow obstruction. MUC5AC expression can be upregulated by a variety of stimuli, including cigarette smoke, LPS and neutrophil elastase, and is increased in the airway epithelium of COPD patients [44]. The present study indicates that endogenous ACh also plays a crucial role in LPS-induced MUC5AC expression. Interestingly, as early as in 1973, goblet cell hyperplasia has been observed following repeated administration of muscarinic agonists in experimental animals [45].

Airway fibrosis contributes to small airway thickening and airflow limitation in COPD [46]. The role of ACh in airway fibrosis in COPD is currently unknown. However, in vitro studies have indicated that ACh can induce proliferation of lung fibroblasts [20], and collagen synthesis by these cells [22]. In addition, increased ChAT expression has recently been found in lung fibroblasts from healthy smokers and from COPD patients [15], suggesting that non-neuronal ACh may modulate fibroblast function in an autocrine fashion under these conditions. Our results demonstrated that tiotropium inhibits LPS-induced collagen accumulation in the lung and airway wall, indicating that ACh may be a key regulator of airway fibrosis. In addition to a possible direct effect of ACh on fibroblasts in this process, its effect on neutrophils could also play a role. Thus, increased neutrophil elastase activity, as observed in COPD [41], has also been associated with pulmonary fibrotic diseases, such as idiopathic pulmonary fibrosis [47]. The potential role of neutrophil elastase in fibrosis is further supported by studies in mice, showing that neutrophil elastase inhibition [48] or gene deletion [49] inhibits bleomycin-induced pulmonary fibrosis in these animals.

LPS induced a significant increase in alveolar airspace size, which is indicative of emphysema. The observed increase in MLI after 12 weeks of twice-weekly LPS instillation is of a similar magnitude to that described previously for mice following 16 weeks exposure to cigarette smoke [37]. In our study, no evidence for the involvement of ACh in the development of LPS-induced emphysema was found. Thus, tiotropium had no significant effect on airspace size in either LPS- or saline-instilled animals. This could imply that neutrophilic inflammation is not essential for the development of LPS-induced emphysema, which corresponds with previous observations in patients with COPD [41]. The mechanisms leading to the development of LPS-induced emphysema are not entirely clear, but presumably the release of elastolytic enzymes from neutrophils plays a role. Although chronic LPS-induced parenchymal neutrophilia in our model was inhibited by tiotropium, this does not necessarily mean that neutrophils have not contributed to alveolar destruction. A recent study in mice has shown that the acute LPS-induced bronchoalveolar lavage neutrophilia (at 4 h post-LPS exposure) is not affected by tiotropium pre-treatment [50]. This finding indicates that tiotropium does not inhibit the acute LPS-induced signalling, which leads to the release of neutrophil-attracting chemokines, resulting in acute neutrophilia. Since a brief exposure to neutrophil elastase may already lead to emphysema, the acute tiotropium-resistant neutrophilic response after each LPS instillation could be sufficient to initiate the development of emphysema. In addition, other inflammatory cell types, such as macrophages, CD8-positive T-lymphocytes and B-cells, may also contribute [37, 41].

Pulmonary vascular remodelling may occur in patients with COPD and contribute to pulmonary hypertension in these patients [51]. Pulmonary vascular remodelling in COPD is characterised by thickening of the vessel wall as well as muscularisation of microvessels that do not have a smooth muscle layer under healthy conditions [4, 6]. The precise mechanisms underlying the pulmonary vascular remodelling in COPD are not known, but may involve hypoxia, inflammation and cigarette smoke constituents, leading to endothelial dysfunction and release of growth factors [52]. The increase in vessel wall thickness is largely due to smooth muscle cell proliferation in the intima [4] as well as thickening of the media [6]. Interestingly, thickening of pulmonary arteries and increased number of small vessels positive for smooth muscle actin have been found in guinea pigs after 3 or 6 months exposure to cigarette smoke [53]. In contrast, in the present study, no effect of LPS on pulmonary artery or arteriole thickness was found. Similarly, LPS did not affect ASM mass under the same conditions (data not shown). However, LPS did increase the number of muscularised microvessels in the adventitia of cartilaginous airways. The strong inhibitory effect of tiotropium on this increase indicates a major role for endogenous ACh in this process. Although the mechanisms underlying these changes are presently unknown, airway inflammation might play a role [54]. This is supported by the observation that the neutrophil infiltration and its inhibition by tiotropium is particularly observed in the adventitial compartment of the airways (data not shown).

Recently, results of the UPLIFT trial have demonstrated that tiotropium treatment of patients with COPD reduces the number of exacerbations, the incidence of respiratory and cardiac adverse events as well as mortality [9, 10]. A pre-specified subgroup analysis also showed that in GOLD stage II COPD, tiotropium treatment reduces the rate of decline of post-bronchodilator FEV1 [11]. Although the mechanisms underlying these effects remain to be established, data from our animal model suggest that anti-inflammatory and anti-remodelling properties of the drug could be involved.

In conclusion, our study has demonstrated that inhaled tiotropium inhibits pulmonary inflammation and airway remodelling in a guinea pig model of COPD, indicating that endogenous ACh may play a major role in the pathogenesis of this disease.

Footnotes

  • A press release for this article is available from www.erj.ersjournals.com/site/misc/presspack.xhtml

  • Support Statement

    This study was financially supported by Boehringer Ingelheim Pharma GmbH & Co. KG, and the Graduate School of Behavioural and Cognitive Neurosciences, University of Groningen, Groningen, the Netherlands.

  • Statement of Interest

    Statements of interest for R. Gosens, J. Zaagsma, and H. Meurs, and for the study itself can be found at www.erj.ersjournals.com/site/misc/statements.xhtml

  • Received September 15, 2010.
  • Accepted February 9, 2011.

REFERENCES

View Abstract