Published ahead of print on March 1, 2007, doi:10.1164/rccm.200608-1113OC
© 2007 American Thoracic Society doi: 10.1164/rccm.200608-1113OC
Accumulation of Dendritic Cells and Increased CCL20 Levels in the Airways of Patients with Chronic Obstructive Pulmonary Disease1 Department of Respiratory Diseases, Ghent University Hospital, Ghent; 2 Department of Respiratory Diseases, Leuven University Hospital, Leuven; and 3 Department of Thoracic and Vascular Surgery, Ghent University Hospital, Ghent, Belgium Correspondence and requests for reprints should be addressed to Ingel K. Demedts, Department of Respiratory Diseases, Ghent University Hospital 7K12IE, De Pintelaan 185, B-9000 Ghent, Belgium. E-mail M.DemedtsIngelK{at}UGent.be
Rationale: Chronic obstructive pulmonary disease (COPD) is characterized by chronic airway inflammation. It is unclear if dendritic cells (DC) participate in this inflammatory process. Objectives: To evaluate the presence of DC in small airways of patients with COPD. Methods: We evaluated DC infiltration in small airways by immunohistochemistry in patients with COPD (stage I-IV), never-smokers, and smokers without COPD. Chemokine ligand 20 (CCL20, the most potent chemokine in attracting DC) was determined in total lung by RT-PCR and in induced sputum by enzyme-linked immunsorbent assay. Chemokine receptor 6 (CCR6, the receptor for CCL20) expression on human pulmonary DC was evaluated by RT-PCR and flow cytometry. Measurements and Main Results: There is a significant increase in DC number in the epithelium (p = 0.007) and adventitia (p = 0.009) of small airways of patients with COPD compared with never-smokers and smokers without COPD. DC number in epithelium and adventitia increases along with disease severity. CCL20 mRNA expression in total lung and CCL20 protein levels in induced sputum are significantly higher in patients with COPD compared with never-smokers (p = 0.034 for CCL20 mRNA and p = 0.0008 for CCL20 protein) and smokers without COPD (p = 0.016 for CCL20 mRNA and p = 0.001 for CCL20 protein). DC isolated from human lung express CCR6 both at mRNA and at protein level. Conclusions: This is the first description of airway infiltration by DC in COPD. Moreover, interaction between CCL20 and CCR6 provides a possible mechanism for accumulation of DC in the lungs in COPD.
Key Words: langerin • CCL20 • CCR6
Chronic obstructive pulmonary disease (COPD) is an important health problem: it is a major cause of chronic morbidity (1) and is currently listed as the fifth leading cause of death worldwide (2). Moreover, the prevalence and mortality are expected to increase in the coming years (3). In most cases, the disease is caused by cigarette smoking, which induces chronic airway inflammation and development of emphysema, leading to irreversible airflow limitation and an accelerated decline in lung function (4). Several inflammatory cells, both of the innate and the adaptive immune system, participate in the inflammatory response in COPD (5, 6): neutrophils, macrophages, and cytotoxic CD8-positive T-lymphocytes infiltrate the lungs of patients with COPD, and contribute to the development of small airway obstruction (7, 8). Importantly, airway inflammation in patients with COPD persists despite smoking cessation (9), suggesting a memory adaptive immune response. Until now, participation of dendritic cells (DC) in this process of chronic airway inflammation has not been demonstrated. DC are inflammatory cells with a central role in the orchestration of immune responses (10), crucial in linking innate to adaptive immune responses. Both myeloid DC (originating from myeloid precursors) and plasmacytoid DC (named after their resemblance to plasma cells) have been described in human lung, each with divergent roles in pulmonary immunity (11). While the role of DC in the pathogenesis of asthma has been studied extensively (12), investigations on the involvement of DC in the development of COPD in smokers are very scarce. Recent data from a mouse model of COPD describe a strong increase of DC in the lung in response to cigarette smoke, suggesting that DC participate in the disease process (13). However, no data from studies on human subjects are available on the contribution of DC to the pathogenesis of COPD. Some authors describe an increase of DC in the airways of smokers compared with nonsmokers (14, 15), but none of these studies included patients with COPD. By lack of direct studies on DC infiltration in COPD lungs, indirect information could be obtained from the presence of chemotactic signals for DC in the airways. However, while several chemokines for neutrophils and T cells have been described in the lungs of patients with COPD (16), no studies on DC attracting chemokines have been performed.
The aim of this study was to investigate the role of human pulmonary DC in the pathogenesis of COPD. DC infiltration in peripheral airways was compared between never-smokers, smokers without airway obstruction, and patients with COPD (Global Initiative for Chronic Obstructive Lung Disease [GOLD] stage I-IV). Patients with COPD were divided in three different groups: GOLD stage I, GOLD stage II, and a combined group of GOLD stage III and IV. The presence of CCL20 (also known as MIP3
Human Lung Tissue Lung tissue was obtained from patients who underwent lobectomy or pneumectomy for various reasons (mostly lung cancer). Lung tissue from patients with end-stage patients was obtained from explant lungs from patients selected for lung transplantation. In total, resection specimens from 73 patients (14 never-smokers, 15 smokers without COPD, and 44 patients with COPD) were processed. Tissue distant from the primary pathologic lung tissue was collected by a pathologist. Written informed consent was obtained from all subjects according to protocols approved by the Medical Ethical Committee of the Ghent University Hospital and University Hospital Gasthuisberg, Leuven. More details on the processing of the lung tissue are provided in the online data supplement.
Immunohistochemistry
CCL20 and CCR6 mRNA Expression in Total Lung
Sputum Induction and Sputum Processing Sputum induction and processing was performed as previously described (17). Details are provided in the online supplement. CCL20 in induced sputum supernatant was determined by enzyme-linked immunosorbent assay (ELISA; R&D Systems, Abingdon, UK).
Flow Cytometry
Statistical Analysis
Accumulation of DC in Human Lung Accumulation of DC (identified as Langerin-expressing cells) in small airways (Figure 1) was evaluated in 55 subjects: 10 never-smokers, 9 smokers without COPD, 10 patients with COPD GOLD stage I, 16 patients with COPD GOLD stage II, 5 patients with COPD GOLD stage III, and patients with 5 COPD GOLD stage IV. Subject characteristics are shown in Table 1. FEV1 and FEV1/FVC ratio are significantly lower in patients with COPD when compared with control subjects. There is no difference in age and smoking history between patients with COPD and smokers without COPD.
The number of DC in the epithelium and in the adventitia of small airways is significantly higher in patients with COPD compared with never-smokers and smokers without COPD (p = 0.004 for DC accumulation in the epithelium and p = 0.008 for DC accumulation in the adventitia). Moreover, there is a significant increase in DC accumulation in the epithelium and in the adventitia of patients with COPD at more severe stages of the disease compared with mild and moderate stages of the disease (Figures 2A, 2B, and 2D). Mean DC number (± SEM) in the epithelium is 9.1 ± 2.5 x 10–3/µm basement membrane and 387.8 ± 99.0 x 10–6/µm2 epithelium in never-smokers, 8.5 ± 1.3 and 328.9 ± 88.3 in smokers without COPD, 18.8 ± 3.7 and 687.3 ± 150.4 in patients with COPD GOLD stage I, 16.0 ± 2.4 and 623.5 ± 93.3 in patients with COPD GOLD II, and 29.9 ± 8.3 and 1,275.3 ± 352.1 in patients with COPD GOLD III–IV (Figures 2A and 2B).
Mean DC number (± SEM) in the lamina propria is 132.0 ± 82.3 x 10–6/µm2 lamina propria in never-smokers, 85.5 ± 44.3 in smokers without COPD, 137.6 ± 37.8 in patients with COPD GOLD stage I, 145.2 ± 28.8 in patients with COPD GOLD II, and 416.6 ± 198.5 in patients with COPD GOLD III–IV (Figure 2C). There is no significant difference between groups (p = 0.23). Mean DC number (± SEM) in the adventitia is 64.6 ± 29.9 x 10–6/µm2 adventitia in never-smokers, 56.3 ± 14.5 in smokers without COPD, 155.9 ± 71.4 in patients with COPD GOLD stage I, 149.6 ± 30.4 in patients with COPD GOLD II, and 779.7 ± 307.7 in patients with COPD GOLD III–IV (Figure 2D). There is a significant inverse correlation between FEV1% and DC number/length of the basal membrane (r = –0.41, p = 0.003), DC number/area of the epithelium (r = –0.34, p = 0.014), and DC number/area of the adventitia (r = –0.047, p = 0.0005). There is no significant correlation between FEV1% and DC number/area of the lamina propria (r = –0.25, p = 0.078). There is no significant difference in DC numbers in any of the epithelial layers in small airways of patients with COPD who have quit smoking compared with actively smoking patients with COPD. This is the case when patients with COPD from all stages are analyzed together, as well as when each group of patients with COPD is analyzed separately. Importantly, in the COPD GOLD IV group, all of the patients have quit smoking. Taken together, these data show that the accumulation of DC in small airways of patients with COPD persists despite smoking cessation.
CCL20 and CCR6 Expression in Human Lung and Induced Sputum RT-PCR was performed on total lung tissue from 11 never-smokers, 11 smokers without COPD, and 21 patients with COPD. Subject characteristics are shown in Table 2. Expression of CCL20 mRNA transcripts is significantly higher (p = 0.034 and p = 0.016, respectively) in COPD lung tissue compared with lung tissue from never-smokers and smokers without COPD (Figure 3). There is a significant inverse correlation between FEV1% and CCL20 mRNA expression in total lung (r = –0.45, p = 0.004).
CCL20 protein levels were determined by ELISA in induced sputum from 15 never-smokers, 13 smokers without COPD, and 19 patients with COPD. Subject characteristics are shown in Table 3. Mean (± SEM) levels of CCL20 protein levels in sputum from patients with COPD are significantly higher compared with never-smokers and smokers without COPD (2,229.6 ± 471.0 pg/ml in patients with COPD versus 688.7 ± 135.3 in never-smokers and 725.9 ± 110.3 in smokers without COPD, p = 0.00046). There is no significant difference in CCL20 levels between never-smokers and smokers without COPD (Figure 4). There is a significant inverse correlation between FEV1% and CCL20 protein levels in induced sputum (r = –0.40, p = 0.006).
There was no significant difference in CCL20 protein in induced sputum and in CCL20 mRNA expression in total lung in patients with COPD who have quit smoking when compared with actively smoking patients with COPD. There was no significant difference in CCR6 mRNA expression in total lung between never-smokers, smokers without COPD, and patients with COPD (median [± SD] CCR6/GAPDH mRNA ratio 0.63 [± 0.07], 0.66 [± 0.13], and 0.73 [± 0.14], respectively).
Human Pulmonary DC Express CCR6 Myeloid DC in human lung express significant amounts of CCR6 mRNA transcripts, while plasmacytoid pulmonary DC express significantly less CCR6 mRNA (p = 0.004, Figure 5A). Flow cytometric analysis of CCR6 expression on human pulmonary DC confirms CCR6 expression on myeloid DC, in contrast to plasmacytoid DC, where no CCR6 expression could be detected (Figure 5B).
We demonstrate for the first time that there are increased numbers of DC in the airways of patients with COPD compared with never-smokers and smokers without COPD. Moreover, the number of DC infiltrating the small airways of patients with COPD continues to increase along with disease severity. These findings suggest that DC participate in the process of chronic airway inflammation that takes place in COPD. Several methodological aspects of this work contribute to the strengths of this study. First, DC accumulation was investigated in the whole range of severity grades of COPD (GOLD stage I–IV), which provides important additional information (increased DC accumulation in more severe stages of COPD). Second, increased CCL20 expression in patients with COPD was demonstrated both at RNA and protein level and at different compartments of the human lung. Moreover, while there might be a selection bias for investigations on lung resections (selection of patients that are eligible for thoracic surgery), this is not the case for studies on induced sputum, which is performed on patients with COPD in the general population. Our hypothesis that DC are involved in the development of COPD in smokers is based on several findings. First, the intrinsic properties of DC suggest that they might be involved in the pathophysiologic changes as seen in COPD. Indeed, the principal task of DC is to screen the environment for the presence of "danger signals" at epithelial surfaces (10). The airways and lungs contain a rich network of DC, localized near the epithelial surface (23). Consequently, DC are ideally localized to initiate an inflammatory reaction in response to inhaled cigarette smoke. Moreover, DC are traditionally described as key cells in linking innate and adaptive immune responses (24, 25), which are both taking part in the chronic inflammation in COPD (6). Importantly, we recently described that human pulmonary DC are indeed able to induce both innate and adaptive immune responses in the human lung (11). In addition, data from mouse models for COPD support the involvement of DC in the pathogenesis of COPD. D'hulst and coworkers demonstrated a 10-fold increase in DC numbers in bronchoalveolar lavage (BAL) fluid from mice in response to cigarette smoke (13). Moreover, Zeid and colleagues describe a massive increase in DC infiltration in mice lungs after passive exposure to tobacco smoke (26). Complementary to these findings in animal models, several groups describe an effect of smoking on DC recruitment into the human lung (14, 15). However, none of these studies included smokers with COPD. Moreover, there were conflicting results regarding the effect of smoking on DC numbers in the lung. Soler and coworkers studied DC in human lung from 10 smokers and 8 nonsmokers by immunohistochemistry (14). Importantly, they did not find any difference in DC numbers in the bronchiolar epithelium in nonsmokers when compared with smokers. These results are in line with the findings reported here, since there was no significant difference in Langerin+ DC in the airways of smokers without obstructive lung disease compared with never-smokers. However, Soler and colleagues did find a significant increase in CD1a+ DC in alveolar parenchymal tissue in smokers compared with non-smokers. The study of DC in the alveoli was beyond the scope of the current study, since in preliminary experiments, hardly any Langerin+ DC were found in the alveoli. We have previously described the absence of CD1a+ DC in the lung parenchyma, whereas CD1a+ DC were found almost exclusively in association with the airway epithelium (18). The same results were obtained by van Haarst and coworkers (27), who did find CD1a+ DC in bronchial epithelium but describe that there are only very few CD1a+ DC in the alveolar spaces and alveolar walls. While the presence of DC in human lung had been described as early as 20 years ago (28), no detailed analysis of DC infiltration in the airways of patients with COPD has been reported until now. By using immunohistochemical staining for Langerin expression, we demonstrate here an impressive increase of DC in the airways of patients with COPD compared with smokers without COPD (up to fivefold increase in the epithelium and even more in the adventitia). This suggests that there is an increase in DC-attracting chemokines in patients with COPD, while this is probably not the case in smokers without the disease. CCL20 has been described as the most potent chemokine known in attracting DC precursors into sites of inflammation through interaction with CCR6 (19, 20). In this study, we demonstrate for the first time high CCL20 levels in the lungs of patients with COPD compared with never-smokers and smokers without COPD. Moreover, freshly isolated human pulmonary DC in our experiments appear to express CCR6, which confirms previous reports by other groups (21, 22). Taken together, these observations provide a possible mechanism for the increased influx of DC into the airways in COPD. Interestingly, CCR6 is not only expressed on DC, but also on activated neutrophils (29), B cells (30), and memory T cells (31). However, the importance of CCL20/CCR6 interaction for the recruitment of these inflammatory cells into sites of inflammation is much less clear than it is for DC. Future studies are needed to evaluate the role of CCR6 on activated neutrophils, B cells, and memory T cells in the pathogenesis of COPD.
It remains to be determined by which mechanisms DC might contribute to the development of the pathologic changes in COPD. However, several findings suggest that DC might interact with most of the pathophysiologic mechanisms underlying COPD (chronic inflammation, proteolytic activity, and oxidative stress). First, chronic airway inflammation in response to cigarette smoke is the hallmark of COPD. It has been described that pulmonary DC are activated by cigarette smoke and consequentially up-regulate the expression of co-stimulatory molecules such as CD40 and CD86 (13). This could result in the loss of tolerogenic properties of DC and to the initiation of sustained airway inflammation as seen in COPD. Second, data from animal models (32) and recent findings in patients with COPD (17, 33) underscore the importance of proteolysis in general (34), and of MMP-12 in particular, in the pathogenesis of COPD. Alveolar macrophages appear to be the principal source of MMP-12 (35) and increased numbers of CD68+ macrophages have been described in the lungs of patients with COPD (5). However, it has been described that there is a strong up-regulation of MMP-12 mRNA transcription in in vitro cultured human DC (36), and our group recently demonstrated that DC isolated from murine lungs are an important source of MMP-12 (37). Thus, while alveolar macrophages are the most important source of MMP-12 in the lung, pulmonary DC might also contribute to the increase in MMP-12 activity in COPD. Interestingly, MMP-12 activity in the lung generates elastin fragments that are strongly chemotactic for DC precursors (38). Third, oxidative stress induces phenotypical and functional maturation of DC (39) and up-regulates the release of proinflammatory cytokines (IL-8, TNF- As mentioned above, several methodological aspects increase the strength of this study. However, there are some limitations to our study that deserve some further consideration. First, while we describe increased DC numbers as well as high CCL20 levels in COPD, it is not possible to prove a causal role for DC in the pathogenesis of COPD with this approach. For this purpose, in vivo animal experiments are needed to study the effect of overexpression or knockout of DC function on cigarette smoke–induced inflammation. Such animal models are also needed to unravel the mechanism responsible for the accumulation of DC in the airways. Indeed, the increased DC number in small airways can be the consequence of an increased influx into the airways, local proliferation, prolonged survival, or a decreased efflux of DC out of the airways. The elevated CCL20 levels suggest an increased influx of DC into the airways in response to CCL20. Recently, our group investigated the role of the CCR6/CCL20 axis in a mouse model of COPD. We demonstrated the accumulation of DC in the lungs of wild-type mice in response to cigarette smoke exposure. This DC accumulation was significantly attenuated in mice lacking the CCR6 receptor (41). These findings suggest that DC are indeed recruited into the airways in response to cigarette smoke by a CCR6/CCL20-dependent mechanism. However, other mechanisms cannot be excluded. Several hypotheses could explain the increase in DC in the epithelium and the adventitia, but not in the lamina propria. DC or DC precursors are possibly recruited from blood vessels in the adventitia, which could explain the increase of DC in the adventitia. However, the migration of DC from the adventitia toward the epithelium, through the extracellular matrix, should normally result in increased DC numbers in the lamina propria as well. Another possibility would be that DC move from the adventitia to the epithelium through small connecting blood vessels that connect the blood vessels in the adventitia to blood vessels just underneath the epithelium. Second, there could be some influence of pulmonary infections on DC recruitment into the lung. Recently, it has been reported that bacterial colonization is associated with increased airway inflammation in COPD (42). Consequently, one could argue that the increase in DC infiltration in COPD could partly be attributed to bacterial colonization of the airways. While no bacterial cultures were performed in our study, patients with clinical signs of infection are unlikely to be scheduled for curative surgery. Moreover, resection specimens with macroscopic and/or microscopic signs of infection were excluded from analysis, and patients with recent respiratory infections were excluded for sputum induction. Taken together, these measures minimize the risk of infection as a possible confounder. Third, the presence of cancer could have influenced the DC infiltration in small airways, leading to an enhancement of inflammatory cell recruitment. However, primary bronchus carcinoma was the main reason for surgery in all of the groups, minimizing the risk of confounding between groups. Moreover, recent data suggest a suppression of DC accumulation in lung cancer, rather than an enhanced recruitment (43). Most important, the tissue sample taken for analysis was obtained at a distance from the primary pathologic lung tissue. Finally, we did not investigate the cellular source of CCL20 in human lung. However, it has been demonstrated that CCL20 is secreted by human airway epithelial cells in response to ambient particulate matter (44) and by Langerhans cells (45). Again, this provides a plausible explanation for the influx of DC in COPD airways: CCL20, released by airway epithelial cells in response to cigarette smoke, could attract DC into the epithelium. This process could further be enhanced by the release of CCL20 by the DC themselves, providing an amplifying mechanism that would also explain the continuing inflammation that is seen in COPD even after smoking cessation (46). In conclusion, we demonstrated in this study for the first time an increase in DC numbers in the airways of patients with COPD, which could be explained by interaction of CCL20 with CCR6 on pulmonary dendritic cells. Moreover, the number of DC in epithelium and adventitia of small airways in patients with COPD increases significantly with disease severity. Future studies are needed to clarify the exact role of DC in the pathogenesis of COPD and to evaluate if pulmonary DC or the CCL20/CCR6 axis would be an interesting target for new therapeutic strategies for COPD.
The authors thank Prof. M. Decramer (Department of Respiratory Diseases, Leuven University Hospital, Leuven, Belgium) for providing lung tissue from patients with end-stage COPD. They also thank A. Neesen, I. De Borle, M. Mouton, K. De Saedeleer, P. Degryze, G. Barbier, M. Vancoillie, Professor M. Praet, and Dr. L. Van Walleghem for their technical contribution to this work.
The first two authors contributed equally to this work. Supported by the Fund for Scientific Research in Flanders (FWO Vlaanderen, Research Projects G.0011.03 and G.0343.01N) and by Project grant 01251504 from the Concerted Research Initiative of the Ghent University. I.D. is a doctoral research fellow of the Fund for Scientific Research in Flanders (FWO Vlaanderen). This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200608-1113OC on March 1, 2007 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form August 11, 2006; accepted in final form February 26, 2007
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
, a chemokine that attracts DC toward sites of inflammation) in human lung was evaluated in COPD and control groups. Finally, expression of CCR6 (the receptor for CCL20) on human lung DC was investigated. 
