| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Pathology, University of British Columbia, Vancouver, British Columbia, Canada
Correspondence and requests for reprints should be addressed to Andrew Churg, M.D., Department of Pathology, University of British Columbia, 2211 Wesbrook Mall, Vancouver, BC, V6T 2B5 Canada. E-mail: achurg{at}interchange.ubc.ca
ABSTRACT
Animal models of cigarette smoke–induced chronic obstructive pulmonary disease (COPD) provide potentially useful ways to test drug therapies, either by direct administration of the treatment of interest, or by use of genetically modified animals that mimic the actions of the drug of interest. Evaluation of the potential effects of a drug in animal models requires a long-term (generally 6-mo) smoke exposure to produce/prevent lesions because acute models do not completely predict chronic events. There are now more than 30 chronic studies in the literature which, in aggregate, show that antiproteolytic therapies, antiinflammatory therapies, and antioxidant therapies substantially or completely prevent emphysema, small airway remodeling, and pulmonary hypertension in laboratory animals. However, the few corresponding trials in humans (anti–TNF-
therapy, PDE4 inhibitors) have produced only minor improvements or failed to prevent disease progression. New data from our laboratory indicates that, at least for murine emphysema, the development of disease goes through different phases, with early repair and late failure to repair smoke-induced damage. These observations suggest that the potential effects of drug treatment in humans may vary depending on the stage of the disease and that treatment may be more effective in relatively early disease. An additional complicating factor is that interventions that ameliorate emphysema may or may not prevent small airway remodeling and/or pulmonary hypertension, suggesting that different therapeutic approaches may be needed for the various different anatomic lesions of COPD.
Key Words: cigarette smoke • emphysema • small airway remodeling • pulmonary hypertension • animal models
With the recognition that chronic obstructive pulmonary disease (COPD) is now among the leading causes of death worldwide, and that smoking cessation programs are generally ineffective, has come an increasing search for pharmacologic interventions. This process has been complicated by a lack of detailed understanding of the pathogenesis of cigarette smoke–induced COPD. Despite a variety of newer theories, the protease–antiprotease hypothesis, which states that smoke evokes an inflammatory infiltrate in the lower respiratory tract, and that these inflammatory cells release proteases that overwhelm the local antiproteolytic defenses, leading to matrix breakdown and emphysema, is generally accepted as the mechanism behind the alveolar destruction of emphysema, and this theory has been the basis of most laboratory animal studies looking at COPD interventions.
During the last 10 years there have been approximately 30 chronic smoke inhalation studies in laboratory animals that have attempted in one way or other to prevent or ameliorate experimental COPD (reviewed in Reference 1). In this article, we briefly comment on these data, consider the progress and limitations of such studies, and present data from our laboratory that suggest a possible different approach to intervention therapy in COPD.
An abstract reporting a portion of these results has been published (2).
ANATOMIC LESIONS IN COPD AND CHOICES OF ANIMAL MODELS OF CIGARETTE SMOKE–INDUCED COPD
Although "COPD" is often viewed as equivalent to "emphysema," this idea is not correct. There are in fact four anatomic lesions in COPD: emphysema, small airway remodeling (SAR), vascular remodeling associated with pulmonary hypertension (PHT), and excessive large airway mucus secretion (chronic bronchitis). Any one or any combination of these morphologic changes can be seen in a given patient. Ideally an animal model would reproduce all of these changes. In practice, the excess mucus production of chronic bronchitis thus far cannot be produced in laboratory animals using cigarette smoke, probably because laboratory animals largely lack bronchial glands, and smoke-induced mucus cell metaplasia of the airways is generally not marked (1, 3). It is likely that the lack of mucus production also prevents creation of the infection-prone milieu that is necessary for development of acute exacerbations, and in fact acute exacerbations are not seen in smoking animal models. Emphysema, SAR, and PHT can be produced with chronic smoke exposure, but the changes induced are all relatively mild and in some reports minimal (1), and it is unclear whether interventions that ameliorate mild disease would be able to affect the typically more severe lesions seen in symptomatic humans.
There are significant differences in the response in different species. The anatomic structure of rodent and guinea pigs lungs is different from that of humans, both in terms of airway branching patterns and in cell type distributions; this difference may influence where the particulate phase of cigarette smoke is deposited in the airways and the response to the smoke, and hence the development of SAR (reviewed in Reference 3). SAR tends to be relatively minimal in mice, possibly because the large component of Clara cells in the bronchiolar epithelium provides strong protection against smoke oxidants (4). More severe SAR is seen in guinea pigs, in which the bronchiolar epithelium is more like that of the largely ciliated epithelium of humans.
Lastly, there are numerous experimental manipulations that appear to be effective in acute (1- to 3-d or even 2-wk) models. Unfortunately, experience has shown that the majority of chronic interventions provide less than 100% protection against emphysema, even if they completely suppress the inflammatory response in an acute model, and some effective acute interventions (for example, smoke exposure to mice lacking TLR4) provide no long-term protection at all (1, 5). Thus, testing of an intervention requires a smoke exposure long enough to produce anatomic abnormalities; in practice this usually means a 6-month exposure. What the discrepancy between acute and chronic studies also implies is that the mechanisms behind smoke-induced lesions in laboratory animals (and presumably in humans) change over time.
INTERVENTIONS AGAINST EMPHYSEMA
Because the protease–antiprotease hypothesis is generally accepted as the driving force behind the matrix destruction seen in emphysema, most interventions have attempted to prevent proteolytic attack on the alveolar wall. In a general sense these can be characterized as inhibition of serine proteases, inhibition of metalloproteases, and inhibition of the inflammatory influx. All of these interventions are broadly effective in chronic studies, although the degree of effectiveness varies considerably. Interference with serine proteases provides about 50 to 70% protection, while interference with metalloproteases provides 0 to 100% protection, depending on the specific metalloprotease inhibited/deleted. Reported antiinflammatory interventions produce 25 to 100% protection (reviewed in Reference 1). One optimistic note is that these interventions have been effective in both mice and guinea pigs, and in the few rat models examined, which suggests that they are probably generalizable to humans.
However, there is a major discrepancy between the animal data and human trials. Two different laboratories have reported very substantial (70 to 100%) protection against emphysema in mice lacking TNF- receptors (6, 7), but two different trials of anti–TNF- therapy in humans have failed to show significant improvements (8). Similarly, anti-PDE4 therapy is 100% protective in mice (9), but has shown minimal benefits in humans. The one possible exception is the use of statins, which have been reported to be completely protective against emphysema in rats (10), and show some promise in humans (11).
The disappointing results of anti–TNF- therapy have led some to question whether animal models are of any use (8). To some extent this problem may arise because human studies generally use pulmonary function tests as the measure of interest, whereas animal studies typically measure airspace size, and the correspondence between the two is uncertain; however, where physiologic measurements have been made in smoke-exposed animals (for example, in Reference 12), the intervention protects against physiologic changes as well as anatomic changes. Probably the more important issue is the fact that that almost all animal studies start the intervention from Day 1 of smoke exposure (i.e., they are directed at primary prevention), whereas human interventions ("therapy") typically occur very late in the course of the disease, in fact at a stage where there is major irreversible damage to the lung.
This idea has been supported by recent observations from our laboratory using laser capture microdissection and real-time RT-PCR to examine the ability of the lung parenchyma (minus airways and vessels) to up-regulate genes for the production of matrix elements (collagens and elastin); enzymes (lysyl oxidase, prolyl-4-hydroxylase) responsible for cross-linking collagens and elastin to the final insoluble product; and the TGF-β regulatory pathway, the presumed driver of fibrogenesis and elastogenesis (10). What we observed in the parenchyma was that, for the first month of smoke exposure, there was efficient repair, at least as judged by gene expression, whereas with longer exposures most of these genes were turned off to the point that, at 6 months (when emphysema was present), they were either at control levels or down-regulated below control levels (Figure 1); that is, later in the disease there was a failure to repair. It should be noted that this sequence of events does not apply to all genes; for example, procollagen 11 gene expression remained elevated in the parenchyma through 6 months, but, of interest, there is evidence that collagen 3 is more important than collagen 1 in terms of protection against emphysema (14).
|
INTERVENTIONS AGAINST SMOKE-INDUCED SMALL AIRWAY REMODELING
The importance of SAR as a cause of airflow limitation has become increasingly apparent (15). Most investigators have assumed that, if emphysema is driven by an inflammatory cell–mediated proteolytic attack, SAR must occur via similar mechanisms. However, careful reading of the literature suggests that emphysema and SAR are independent processes. While some interventions such as knockout of TNF- or IL-1 receptors (16), and, interestingly, use of a selective MMP-9/-12 inhibitor (12), protect against both, CCR6 or CCR5 knockout substantially ameliorates emphysema but does not protect against SAR in mice (17, 18); conversely, STAT4 knockout protects against SAR but not emphysema (A. Churg, unpublished observations).
Using laser capture microdissection, we have shown that smoke rapidly up-regulates gene expression of fibrogenic mediators in the small airways in mice. However, as opposed to the early repair/late failure to repair process seen in the parenchyma, in the airways there is a continuing repair reaction that persists through 6 months of smoke exposure in mice (13) (Figure 1). Thus, both the short-term and the long-term response in the small airways is the laying down of matrix, leading eventually to the narrowed, fibrotic, and distorted airway characteristic of SAR. By the same token, this observation implies that an intervention intended to prevent or, particularly, to repair the matrix destruction of emphysema, might have deleterious effects in the small airways if it led to even more fibrosis.
PULMONARY HYPERTENSION
Pulmonary hypertension is a significant cause of morbidity and mortality in patients with COPD. Its presence is a significant risk factor of hospitalization, and decreased life expectancy, especially when the pressure is high and out of proportion to the degree of airflow obstruction (19, 20). Mild to moderate pulmonary hypertension is a common complication of COPD, occurring in up to 90% of patients with severe airflow obstruction, with an estimated minimal prevalence in patients with at least one hospitalization for COPD of between 10 and 30% (20).
Although it is widely believed that pulmonary hypertension in COPD is secondary to emphysematous loss of vascular bed and/or chronic hypoxia, review of the literature indicates that this is not true (reviewed in Reference 21). Rather, data from our laboratory suggest that smoke exerts direct and rapid effects on vascular expression of vasoconstrictive, vasoproliferative, and vasodilatory mediators with resulting endothelial dysfunction and vascular remodeling. These processes are demonstrable within hours to 2 weeks of initial smoke exposure in guinea pigs, and pulmonary hypertension is present between 2 and 4 weeks (J. L. Wright, unpublished observations)—that is, at a time long before there is morphologically evident emphysema or SAR. Drug therapy of emphysema or SAR may not be effective on the vessels; for example the MMP-9/-12 inhibitor that prevented emphysema and SAR in guinea pigs did not prevent vascular remodeling and pulmonary hypertension (12).
CONCLUSIONS
The experimental literature strongly supports the idea that emphysema, SAR, and pulmonary hypertension are independent process, all driven by direct effects of cigarette smoke on specific compartments of the lung, but not necessarily occurring by the same mechanisms. Further, the time course of development of these lesions is different, and, in the case of emphysema, complex. This complexity may in part explain the apparent failure of drug interventions in humans that appear to be effective in animals, since human interventions typically have been tried late in the disease compared with early in the process in animal models. Whether the different phases (repair versus switching off of genes needed for repair) seen in laboratory animal models also occur in humans is not known, but this observation does raise the question of whether interventions directed toward patients with very early disease may prove more successful.
FOOTNOTES
Supported by grant 42539 from the Canadian Institutes of Health Research.
Conflict of Interest Statement: A.C. received grant support from AstraZeneca R&D $100,001 or more and the Canadian Institutes of Health Research–GlaxoSmithKline $50,001 to $100,000. J.L.W. received grant support from AstraZeneca R&D $100,001 or more and the Canadian Institutes of Health Research–GlaxoSmithKline $50,001 to $100,000.
(Received in original form March 17, 2009; accepted in final form June 8, 2009)
REFERENCES
drives 70-% of cigarette smoke-induced emphysema in the mouse. Am J Respir Crit Care Med 2004;170:492–498.
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
