| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Section of Pulmonary and Critical Care Medicine, and Department of Pathology, Yale University School of Medicine, New Haven; and Pathology and Laboratory Medicine Service, VA-CT Health Care System, West Haven, Connecticut
Correspondence and requests for reprints should be addressed to Jack A. Elias, M.D., Section of Pulmonary and Critical Care Medicine, Yale University School of Medicine, 300 Cedar Street, S441 TAC, New Haven, CT 06519. E-mail: jack.elias{at}yale.edu
ABSTRACT
Alveolar destruction is a cardinal feature of emphysema but is not traditionally believed to contribute to the pathogenesis of "classical" asthma. However, the relationship between chronic obstructive pulmonary disease (COPD) and asthma is controversial and the variety of mechanisms that can mediate the alveolar destruction in emphysema have not been adequately defined. To address these issues, we used overexpression transgenic approaches to define the effects of Th1/Tc1 and Th2/Tc2 cytokines in the mature murine lung and compared findings in these transgenic systems to the effects of similar interventions after cigarette smoke (CS) exposure. In these experiments, the Th1/Tc1 and Th2/Tc2 cytokines IFN-
and interleukin (IL)-13, respectively, both caused emphysema. The IFN- response was associated with neutrophilia but was not associated with mucus metaplasia or a major fibrotic response. In this setting, IFN- was a potent stimulator of matrix metalloproteinases (MMPs), cathepsins, and CXC and other chemokines while inhibiting secretory leukocyte proteinase inhibitor (SLPI). Interestingly, IFN- induced its destructive effects via at least two mechanisms, a CCR5/cathepsin-dependent and apoptosis-mediated pathway and an MMP-12–dependent/apoptosis-independent pathway. CS-induced inflammation, apoptosis, and emphysema were also induced by IFN-– and CCR5-dependent mechanisms. In contrast, IL-13–induced emphysema was associated with eosinophilia, mucus metaplasia, and pulmonary fibrosis. In this setting, IL-13 stimulated MMPs, cathepsins, and a variety of CC chemokines while inhibiting 
1-antitrypsin. A cathepsin-dependent apoptosis pathway also contributed to this remodeling response. Interestingly, abnormalities in vascular endothelial growth factor (VEGF) were also appreciated with VEGF165 excess producing an asthmalike pulmonary response and IFN- abrogating this response while inducing emphysematous alveolar destruction. These findings provide molecular support for both points of view in the British/Dutch hypothesis controversy regarding the relationship between asthma and COPD. They also highlight the complexity of the pathways that can induce alveolar destruction and suggest that there is a continuum, based on VEGF, between asthma and COPD.
Key Words: asthma • chronic obstructive pulmonary disease • gamma interferon • IL-13 • vascular endothelial growth factor
Chronic obstructive pulmonary disease (COPD) is a diagnosis that is given to patients with mixtures of chronic bronchitis and emphysema. Chronic bronchitis is defined historically in patients with daily cough and sputum production for at least 3 mo for 2 consecutive years without any other explanation. In contrast, emphysema is defined pathologically when there is abnormal enlargement of the airspaces distal to the terminal bronchiole accompanied by alveolar wall destruction. In Western cultures, COPD is predominantly associated with cigarette smoking. However, only approximately 15% of smokers get COPD, with the others showing a remarkable resistance to this toxic cigarette manifestation (1–4). Seminal studies have demonstrated that an individual's lung function peaks somewhere between 18 and 25 yr of age and slowly declines thereafter. Cigarette smoke (CS) exposure increases the rate of loss of lung function in susceptible individuals, but does not alter the rate of decline in cigarette-resistant individuals. As a result, patients with COPD manifest an accelerated loss of lung function over decades, which can lead to disability and death.
Despite its obvious medical and economic impact, our knowledge of COPD is limited in many ways. These limitations can be appreciated in the controversy that surrounds the relationship between COPD and asthma. Advocates of the "British hypothesis" have suggested that these are distinct diseases mediated via distinct pathogenetic mechanisms. In contrast, advocates of the "Dutch hypothesis" suggest that there are patients who have manifestations of both diseases and that common pathogenetic mechanisms may underlie disease pathogenesis in select patients with COPD and asthma (5, 6). This COPD/asthma controversy is heightened by recent reports demonstrating that a physiologic picture of "pseudoemphysema" can be seen in patients with asthma, highlighting additional levels of similarity between these disorders (7–9). Importantly, we are also limited in our knowledge of disease pathogenesis because we do not know how many ways emphysematous alveolar destruction can be generated and whether different mechanisms are operative in different settings.
SCHOOLS OF THOUGHT IN EMPHYSEMA PATHOGENESIS
A number of theories have been promulgated in attempts to explain emphysema pathogenesis. Since the 1960s, the protease–antiprotease hypothesis has dominated thinking in this area. This theory proposes that the normal lung is protected by an antiprotease shield which protects it from day-to-day exposures that can increase protease production. It also proposes that emphysema occurs when there is an increase in proteases or a decrease in antiproteases (1–4). In this theory, the protease excess is believed to cause matrix injury. With repeated injuries, the damage accumulates and is proposed to lead to alveolar septal rupture. Protease-mediated alveolar epithelial cell injury is not part of this concept.
More recently, pathologic and bronchoalveolar lavage investigations have led to a renewed appreciation of the presence and the potential role of inflammation in the pathogenesis of emphysema. Prominent findings in these studies have been the presence of CD8+ cells in tissues from patients with COPD that produce significant amounts of IFN- and IFN- target genes, such as interferon-inducible protein-10 (IP-10)/CCL-10 (10–13). Studies from our laboratory and others have demonstrated that inflammation can cause protease/antiprotease alterations in the lung. Thus, there is a general consensus that inflammation is likely a major cause of protease/antiprotease imbalance in COPD.
Most recently, a number of investigators have proposed that apoptosis of structural cells plays an important role in the pathogenesis of emphysema. This contention is based on experimental models in which interventions that induce apoptosis lead to emphysema and the appearance of increased numbers of TdT-mediated dUTP nick-end labeling (TUNEL)-staining cells in tissues from patients with emphysema (14–19). However, the relationships between inflammation, protease–antiprotease imbalance, and apoptosis have not been satisfactorily defined.
TRANSGENIC MODELING OF Th1/Tc1 RESPONSES IN THE LUNG
To further understand the mechanism(s) by which Th1/Tc1 responses contribute to the pathogenesis of pulmonary emphysema, we used inducible overexpression transgenic (Tg) systems previously described by our laboratory (20) to selectively overexpress IFN- in the murine lung. This inducible Tg system allowed us to differentiate transgene-induced alveolar abnormalities that are caused by abnormal alveolar development and abnormalities that are induced in otherwise normal mature lungs. This ensured that we were studying the processes that destroy mature lungs (as in human CS-induced emphysema) and not the developmental processes that cause bronchopulmonary dysplasia. In these experimental systems, IFN- caused impressive alveolar destruction and pulmonary emphysema (20). IFN- also caused a patchy mononuclear cell and neutrophil-rich tissue inflammatory response and bronchoalveolar lavage (BAL) inflammation with increased macrophage and neutrophil recovery. Mucus metaplasia was not a prominent finding in these animals and tissue fibrosis was not detected on trichrome stains. However, a modest increase in collagen content could be appreciated in biochemical collagen assays. Interestingly, when the IFN- transgene was activated for significant intervals a barrel chest–like deformity was noted which was characterized by a significant increase in anteroposterior diameter and chest wall indentations at the sites of diaphragmatic insertion. This abnormality is remarkably similar to the barrel chest deformity and "Harrison's grooves" that are seen in patients with severe pulmonary emphysema.
Protease–Antiprotease Alterations
In keeping with the concept that proteases and antiproteases play important roles in the pathogenesis of emphysema, mRNA and protein evaluations were subsequently undertaken to define the protease and antiprotease responses in these Tg animals. These studies demonstrated that IFN- is a prominent stimulator of cathepsins B, D, H and S, and matrix metalloproteinase (MMP)-12, and a powerful inhibitor of the antiprotease secretory leukocyte proteinase inhibitor (SLPI) (20). This is a prominent example of the cause-and-effect relationships between inflammation and protease/antiprotease abnormalities.
Apoptosis
Studies were next undertaken to determine if apoptosis was an ongoing event in lungs from IFN- Tg animals. This was initially done using TUNEL evaluations. These studies demonstrated significant increases in TUNEL staining in lungs from mice in which the IFN- transgene was activated. Double immunohistochemical evaluations subsequently demonstrated that the majority of these TUNEL-positive cells were type II alveolar epithelial cells. To define the cell death response that was causing the TUNEL staining, total lung cells and alveolar type II cells were isolated and subjected to combined propidium iodide and annexin V staining. Subsequent fluorescence-activated cell sorting analysis demonstrated that the majority of these cells were undergoing pure apoptosis. A small number were undergoing apoptosis and necrosis and others were undergoing pure necrosis. When viewed in combination, these studies demonstrate that the overexpression of IFN- in the murine lung causes impressive levels of epithelial cell apoptosis.
Role of Apoptosis in Emphysema
One can envision a scenario in which apoptosis contributes to the pathogenesis of pulmonary emphysema. Alternatively, the apoptosis can be the result of the response that caused the emphysema but might not contribute to the genesis of the emphysema in its own right. To differentiate among these possibilities, we compared the emphysema that was induced by Tg IFN- in mice in which apoptosis proceeded normally and mice in which chemical and genetic approaches were used to selectively block apoptosis. The former was accomplished with the pan caspase inhibitor (Z-vad-fmk), which blocks apoptosis by inhibiting terminal effector caspases, such as caspase 3 (21). The latter was accomplished by breeding IFN- Tg mice with mice with null mutations of caspase 3 and comparing the emphysematous response in mice with wild-type and null caspase-3 loci. In these experiments, the chemical inhibition and the genetic ablation of caspase 3 both caused a significant (50–60%) decrease in the IFN-–induced emphysema (21). This demonstrates that epithelial cell apoptosis is a critical event in the pathogenesis of emphysema in this Tg model.
Protease-mediated Apoptosis
The studies noted above demonstrate that IFN- induces prominent alterations in protease–antiprotease balance and epithelial cell apoptosis. Studies were thus undertaken to determine if there was a relationship between these abnormalities. This was done by breeding the IFN- Tg mice with mice with null mutations of cathepsin S. We then compared the phenotype induced by IFN- in mice with wild-type and null mutant cathepsin-S loci. These comparisons demonstrated that the ablation of cathepsin S significantly diminished the ability of IFN- to induce epithelial cell apoptosis and emphysema. These studies demonstrated that IFN- induces epithelial cell apoptosis via a cathepsin-S–dependent pathway (21). They also demonstrate that this novel protease-mediated epithelial apoptosis response plays a critical role in the pathogenesis of IFN-–induced pulmonary emphysema.
Role of MMP-12
As noted above, interventions that block apoptosis caused significant decreases in pulmonary emphysema. In all cases, however, the emphysematous responses were not completely abrogated. This suggests that other pathways that contribute to the pathogenesis of IFN-–induced emphysema via different mechanisms are simultaneously induced. In an attempt to gain insights into these other pathways, IFN- Tg mice were bred with mice with null mutations of MMP-12. The phenotypes induced by IFN- in mice with wild-type and null MMP-12 loci were then compared. Interestingly, a null mutation of MMP-12 caused a significant decrease in the ability of IFN- to induce pulmonary emphysema (
50%). In contrast to our findings with cathepsin S, null mutations of MMP-12 did not alter the ability of IFN- to induce epithelial cell apoptosis as assessed with TUNEL evaluations. These studies demonstrate that IFN- induces pulmonary emphysema via an MMP-12–dependent pathway that does not involve epithelial apoptosis.
When viewed in combination with the studies noted above, these studies demonstrate that Th1/Tc1 cytokines, such as IFN-, induce pulmonary emphysema via apoptosis-dependent and apoptosis-independent pathways. They also demonstrate that cathepsin S and MMP-12 play critical roles in the apoptosis-dependent and apoptosis-independent pathways, respectively (Figure 1).
|
Inflammation and alveolar remodeling are juxtaposed in lungs from IFN- Tg mice. Studies were thus undertaken to define the mechanisms that IFN- uses to induce tissue inflammation, and the relationships between inflammation and remodeling were explored. The former was accomplished by characterizing the chemokine responses induced by IFN- in the murine lung. In the latter, we defined the effects of chemokine receptor alterations on IFN-–induced tissue remodeling. These studies demonstrated that IFN- is a powerful stimulator of a wide variety of CXC and, to a lesser extent, CC chemokines. Prominent induction of macrophage inflammatory protein (MIP)-1/CCL-3, MIP-1ß/CCL-4 and RANTES (regulated upon activation, T-cell expressed and secreted)/CC5 were readily appreciated. Monocyte chemoattractant protein (MCP)-1/CCL-2, MCP-2/CCL-8, MCP-5/CCL-12, MIP-2/CXCL2/3, keratinocyte cytokine (KC)/CXCL-1, epithelial neutrophil activating (ENA) 78/CXCL-5, monokine induced by gamma interferon (Mig)/CXCL-9, IP-10/CXCL-10, interferon-inducible T-cell alpha chemoattractant (I-TAC)/CXCL-11, stromal cell-derived factor (SDF)-1/CXCL-12, C10/CCL-6, macrophage-derived chemokine (MDC)/CCL-22, and thymus expressed chemokine (TECK)/CCL25 were also induced. To determine if MIP-1/CCL-3, MIP-1ß/CCL-4, and/or RANTES/CCL5 played an important role in these responses, we compared the effects of IFN- in mice with wild-type and null mutant CCR-5 loci, the receptor for these moieties. In the absence of CCR-5, the ability of IFN- to induce tissue inflammation, apoptosis, and pulmonary emphysema was significantly decreased. This was associated with a significant decrease in IFN- induction of selected chemokines (MIP-1/CCL-3, MIP-1ß/CCL-4, MIP-2/CXCL-2/3, RANTES/CCL-5, MCP-1/CCL-2, KC/CXCL-1, and IP-10/CXCL-10) and MMP-9 and increased the expression of SLPI. The decrease in DNA injury and apoptosis was associated with decrease in the induction and activation of the mitochondrial (intrinsic) and death receptor (extrinsic) apoptosis pathways and caspases 3, 8, and 9. Thus, pulmonary inflammatory responses that are associated with exaggerated IFN- production have the ability to induce a wide variety of CC chemokines, including MIP-1/CCL-3, MIP-1ß/CCL-4, and RANTES/CCL-5. These studies also demonstrate that CCR-5 signaling plays a critical role in the pathogenesis of IFN-–induced inflammation and remodeling and the chemokine, protease, antiprotease, and apoptotic responses that underlie these pathologies.
CS-INDUCED EMPHYSEMA
Because the majority of emphysema in the Western world is associated with cigarette smoking, studies were undertaken to determine if similar pathogenic mechanisms were operative in IFN- Tg mice and CS-exposed animals. Six months of CS exposure (two research cigarettes/d, 5 d/wk) caused modest increases in alveolar size in wild-type C57BL/6 mice. This response was associated with a modest increase in BAL macrophage and neutrophil recovery and positive TUNEL staining. Interestingly, these inflammatory, emphysematous, and TUNEL responses were significantly diminished in IFN- knockout mice. In accord with our findings with IFN-, CCR-5 null mice also had significantly decreased inflammatory, apoptotic, and emphysematous responses when compared with control animals. These studies demonstrate that CS induces emphysema in the murine lung via an IFN-–dependent mechanism and that signaling via CCR5 plays an important role in this response. The demonstration that IFN- and CCR-5 play important roles in the pathogenesis CS-induced and IFN-–induced emphysema further substantiates the use of this Tg modeling system and highlights the complementary nature of the two approaches.
Tg MODELING OF Th2/Tc2 RESPONSES IN THE LUNG
To determine if Th2 cytokines also have the ability to contribute to the pathogenesis of pulmonary emphysema, similar approaches were used to inducibly overexpress murine IL-13 in a lung-specific fashion (22). These studies demonstrate that IL-13 causes an eosinophil and mononuclear cell–rich inflammatory response with alveolar enlargement, mucus metaplasia and airway, and, to a lesser degree, parenchymal fibrosis. IL-13 was also a potent stimulator of protease/antiprotease abnormalities augmenting the production of a variety of MMPs (MMP-9, MMP-12) and cathepsins (B, H, D, and S) while simultaneously inhibiting 1-antitrypsin (22). IL-13 also stimulated epithelial cell apoptosis. As noted with IFN-, treatment with Z-vad-fmk or a null mutation of cathepsin S each diminished IL-13–induced apoptosis, emphysema, and inflammation. This highlights the importance of protease-mediated epithelial apoptosis in the pathogenesis of IL-13–induced pulmonary emphysema. These are likely to be important findings because exaggerated levels of IL-13 have been detected in COPD tissues (as reported at this meeting; personal communication, J. Hogg, Vancouver, BC, Canada).
CATHEPSIN S IN SMOKERS' LUNGS
To evaluate the applicability of our murine findings to human COPD, we next compared the expression of cathepsin S in the lung tissue from current smokers, former smokers, and never smokers. Using immunohistochemical quantification, we found that the median expression of cathepsin S was significantly different in current smokers, former smokers, and never smokers, with the highest levels of expression seen in current smokers (median scores of 2.0, 1.0, and 0.5, respectively; p = 0.010; Figure 2). Enhanced cathepsin-S staining was readily appreciated in alveolar macrophages and airway epithelial cells with lesser expression in alveolar epithelium from former smokers (Figure 2). Lower levels of cathepsin S were seen intermittently in alveolar macrophages from nonsmokers. Thus, in accord with our murine findings, cathepsin S is expressed in an exaggerated fashion in the lungs of human smokers.
|
As noted above, the relationship between COPD and asthma is controversial. It is clear, however, that asthma and COPD can be successfully differentiated in a significant percentage of patients. The mechanisms that are responsible for these differences, however, have not been adequately assessed. Recent studies have suggested that vascular endothelial growth factor (VEGF) may contribute to these differences, because decreased levels of VEGF have been noted in COPD and VEGF blockade has been shown to cause emphysema in murine modeling systems (16, 23). In contrast, exaggerated levels of VEGF have been found in tissues and fluids from patients with asthma where VEGF levels correlate with disease activity and inversely with airway caliber and airway hyperresponsiveness (24–26). As a result of these findings, we hypothesized that VEGF excess contributes to an asthmalike phenotype and VEGF deficiency to the development of pulmonary emphysema. To begin to address this hypothesis, VEGF165-overexpressing Tg mice were generated using the methodology noted above. Studies of these animals revealed a remarkable asthmalike phenotype that included increased numbers of leaky superficial blood vessels, mononuclear cell– and eosinophil-rich inflammation, mucus metaplasia, subepithelial fibrosis, myocyte hypertrophy and hyperplasia, and airway hyperresponsiveness. When IFN- and VEGF165 Tg mice were mated to generate mice that produce both cytokines after transgene activation, preliminary studies demonstrated that the tissue effects of VEGF are abrogated by the simultaneous presence of IFN-. These observations suggest that IFN- may induce apoptosis and/or emphysema, in part by blocking the effects of the VEGF that is found in normal lungs. This also allows for the hypothesis that there is a VEGF-based continuum between COPD and asthma, with Th2/Tc2 responses augmenting VEGF and enhancing asthmatic responses and Th1/Tc1 responses diminishing VEGF tissue effects leading and/or predisposing to apoptosis and emphysema (Figure 3). It is also tempting to place the pseudoemphysema that is seen in patients with asthma on this continuum as a midpoint between the polarized responses (Figure 3).
|
Our studies highlight a number of levels of heterogeneity in the pathogenesis of pulmonary emphysema. First, there are a variety of inflammatory mediators, both Th1/Tc1 and Th2/Tc2, that can induce alveolar destruction in different pathologic contexts. IFN-–induced emphysema was associated with BAL neutrophilia and was not associated with mucus metaplasia or significant tissue fibrosis. In contrast, IL-13 induced alveolar destruction in association with an "asthmalike" response characterized by eosinophilia, mucus metaplasia, and airway and parenchymal fibrosis. Second, the different pathways can interact in a synergistic manner with one another to enhance the development of pulmonary emphysema (Figure 4). Third, these studies demonstrate that a single mediator can destroy alveoli via multiple mechanisms. This is nicely illustrated with IFN-, which can activate cathepsin-dependent/apoptosis-dependent and MMP-dependent/apoptosis-independent pathways of alveolar degradation. Last, the responses that are induced are modified by genetic factors and by the presence or absence of other cytokines, like VEGF, in the local microenvironment. The latter may be particularly important, with elevated levels of VEGF leading to asthmalike responses and IFN- inhibition of VEGF effector pathway activation, contributing to its emphysema-generating capacity (Figure 4). When viewed in combination with the British/Dutch hypothesis controversy, it is tempting to speculate that the Th1/Tc1 pathway predominates in patients whose COPD can be easily differentiated from asthma ("British hypothesis patients"); that the Th2/Tc2 pathway dominates in "Dutch hypothesis" patients with features of asthma and emphysema, and that this pathway can contribute to the pathogenesis of asthmatic pseudoemphysema (7–9). It is also tempting to speculate that both pathways can be activated in a subset of patients with rapidly progressive loss of lung function (e.g., smokers with asthma). Last, this heterogeneity suggests that optimal prevention of pulmonary emphysema may require multidrug interventions that simultaneously regulate more than one pathway of alveolar destruction.
|
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 March 17, 2006; accepted in final form April 14, 2006)
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  S. Ying, B. O'Connor, J. Ratoff, Q. Meng, C. Fang, D. Cousins, G. Zhang, S. Gu, Z. Gao, B. Shamji, et al. Expression and Cellular Provenance of Thymic Stromal Lymphopoietin and Chemokines in Patients with Severe Asthma and Chronic Obstructive Pulmonary Disease J. Immunol., August 15, 2008; 181(4): 2790 - 2798. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Petrache, T. R. Medler, A. T. Richter, K. Kamocki, U. Chukwueke, L. Zhen, Y. Gu, J. Adamowicz, K. S. Schweitzer, W. C. Hubbard, et al. Superoxide dismutase protects against apoptosis and alveolar enlargement induced by ceramide Am J Physiol Lung Cell Mol Physiol, July 1, 2008; 295(1): L44 - L53. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. F. Chung and I. M. Adcock Multifaceted mechanisms in COPD: inflammation, immunity, and tissue repair and destruction Eur. Respir. J., June 1, 2008; 31(6): 1334 - 1356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Tang, M. Mura, R. Seth, and M. Liu Acute lung injury and cell death: how many ways can cells die? Am J Physiol Lung Cell Mol Physiol, April 1, 2008; 294(4): L632 - L641. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Daley, C. Emson, C. Guignabert, R. de Waal Malefyt, J. Louten, V. P. Kurup, C. Hogaboam, L. Taraseviciene-Stewart, N. F. Voelkel, M. Rabinovitch, et al. Pulmonary arterial remodeling induced by a Th2 immune response J. Exp. Med., February 18, 2008; 205(2): 361 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Leff Erratum: Mechanistic Heterogeneity in Chronic Obstructive Pulmonary Disease Proceedings of the ATS, August 15, 2007; 4(5): 473 - 473. [Full Text] [PDF]
|
|
|
|
 |
|
 H. Fehrenbach Animal models of pulmonary emphysema: a stereologist's perspective Eur. Respir. Rev., December 1, 2006; 15(101): 136 - 147. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
