INTRODUCTION

TOP ABSTRACT INTRODUCTION AIRWAY REPAIR EMPHYSEMA REFERENCES

Chronic obstructive pulmonary disease (COPD) most commonly results from long-standing inhalation of toxins and irritants, e.g., cigarette smoke, into the lung. These toxins can directly injure lung structures. They can also lead to chronic inflammation in the airways and the alveolar structures of the lung, which can further injure lung structures (1). Repair processes are initiated as part of the inflammatory response in many tissues including the lung. If these repair responses can restore normal tissue architecture, function can be preserved. Efforts at repair, however, may result in disruption of normal tissue. In COPD, both in the airways and in the alveolar structures, tissue dysfunction likely results from altered structure due to incompletely effective repair responses (2). Advances in the understanding of the mechanisms underlying these processes offers the opportunity to target these processes and thus favorably affect the clinical outcome in COPD.

	AIRWAY REPAIR

TOP ABSTRACT INTRODUCTION AIRWAY REPAIR EMPHYSEMA REFERENCES

The airways are exposed to a variety of inhaled toxins and pathogens. Appropriately, the airways have considerable capacity both to initiate inflammatory responses and to participate in repair (Figure 1). After mechanical injury of the airway, for example, there is exudation of plasma proteins onto the airway surface (3). These proteins can polymerize into a provisional matrix that forms a basis for airway repair. This process can be exceedingly rapid (4). After mechanical injury in an animal model, for example, within 15 min of injury, epithelial cells on the edge of the wound have flattened and have begun to migrate in order to cover the wound (4). Within hours, an epithelial defect can be covered such that epithelial integrity is restored (4). Within 24 h, a wave of cellular proliferation is initiated, leading to the accumulation of numerous dedifferentiated cells. These cells eventually assume a columnar shape and may acquire a normal differentiated phenotype including expression of secretory granules and cilia.

View larger version (27K): [in this window] [in a new window]   View larger version (19K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   View larger version (34K): [in this window] [in a new window]   View larger version (27K): [in this window] [in a new window]   Figure 1.   Stages of airway repair. (A) Normal epithelium. (B) After a wound has been incurred, portions of the basement membrane become stripped of cells. A provisional matrix derived from plasma proteins including fibrin and fibronectin accumulates rapidly. (C ) Epithelial cells participate in the wound repair process in a number of ways: (1) basal cells remaining in the wound can release cytokines, which can function as autocrine or paracrine factors driving the inflammatory response; (2) epithelial cells can respond to these paracrine factors, leading to the production of additional mediators (e.g., TGF- produced by airway epithelial cells can drive epithelial cell production of fibronectin); (3) mediators derived from epithelial cells and potentially other sources induce dedifferentiation of epithelial cells present at the wound margins; (4) epithelial cells migrate across the provisional matrix, restoring epithelial integrity; and (5) mediators produced by epithelial cells can also lead to activation of mesenchymal cells. (D) After recruitment, epithelial cells undergo accumulation through proliferation. This results in accumulation of a dedifferentiated epithelium. Mesenchymal cells may accumulate concurrently. (E  ) Over the course of days to weeks, the newly recruited epithelial cells can differentiate and have the potential to restore epithelial architecture.

The cells that are responsible for normal airway repair are incompletely defined. By using antibodies specific for cytokeratin 14, a marker of basal cells, Shimizu and colleagues (5) demonstrated that nearly every cell present in the early days after airway injury expresses the basal cell marker. When columnar cells first accumulate in the repairing wound, they initially express this marker. Cytokeratin 14 expression is gradually lost over the next 2 wk, during which the columnar cells increasingly express markers associated with normal epithelial columnar cells. Over this interval, cytokeratin 14 expression becomes limited to cells with a recognizable basal cell morphology. Although this sequential expression of markers suggests that basal cells may be the precursors of the cells present in an airway wound and is consistent with reconstitution studies (6), it is not definitive. It is possible that columnar cells at the edge of the wound dedifferentiate, express cytokeratin 14, and are responsible for the repair response (6). Consistent with this alternative schema are observations that columnar cells can dedifferentiate and can completely restore epithelial structures in model systems.

The processes that regulate epithelial cell recruitment proliferation and differentiation are not fully characterized. It is likely that epithelial cells can respond to a number of chemotactic factors. In this regard, fibronectin, a multifunctional glycoprotein involved in tissue remodeling, is a chemoattractant for airway epithelial cells (7) and can be released in increased amounts by epithelial cells in response to a variety of cytokines (8). Included among these is transforming growth factor  (TGF-) (9), an inflammatory mediator that not only modulates inflammatory responses, but is believed to have an important role in regulating repair. Fibronectin may also be derived from plasma exudation and is incorporated into the provisional matrix that forms after injury (3). Fibronectin has several splice variants and posttranslational modifications. Interestingly, the cell-derived form of fibronectin appears to be a much more potent chemoattractant than is the plasma- derived form (10). This suggests that different forms of the same molecule may play different roles in the repair process.

Increased levels of fibronectin have been reported in the airway and bronchoalveolar lavage fluid of patients with COPD (11). Increased expression of TGF- has also been reported in the airways of patients with chronic bronchitis and asthma, although this observation remains controversial (12, 13). Taken together, however, these data suggest that mediators of the inflammatory response acting locally within the airway can serve to modulate the recruitment of epithelial cells to cover a defect that results from injury.

The ability of epithelial cells to migrate to cover such a defect may also be modulated by components in the inflammatory milieu. In this regard, the chemotactic responsiveness of epithelial cells can be modulated both by inflammatory cytokines (14) and by the composition of the matrix over which the epithelial cells must migrate (15). The cells that initially migrate to cover a wound assume a highly flattened phenotype (4). Interestingly, TGF- causes precisely this type of phenotypic change in cultured epithelial cells (8). TGF- also increases epithelial cell adhesiveness and decreases migratory activity, suggesting it may play a role in directing epithelial cells newly arrived at a site of injury to alter their structure and function (16).

The regulation of epithelial cell proliferation after injury remains to be determined. There are, however, a number of potential sources of growth factors that can regulate epithelial cell proliferation. Both mesenchymal fibroblasts (17) and epithelial cells themselves can release factors that modulate epithelial cell proliferation. In addition, both soluble and matrix components of the extracellular milieu can modulate proliferation (8). Finally, the mononuclear inflammatory cells chronically present in the airways are capable of releasing many growth regulators. In in vitro studies, alveolar macrophages have been demonstrated to drive the proliferation of cultured airway epithelial cells (18). It is likely, therefore, that the inflammatory milieu that characterizes the airway in COPD alters the reparative capacity of the epithelium in response to injury.

Migrating epithelial cells, such as those responding to an injury, express different altered sets of cell surface receptors (19). Bacteria can exploit a number of mechanisms in order to adhere to epithelial surfaces, including fibronectin, which is present in increased amounts on migrating epithelial cells (20). As a result, bacteria will preferentially adhere to migrating epithelial cells. This may account for the tendency of the injured airway to become colonized with bacteria and thus account for the increased incidence of lower respiratory tract infections that follow acute airway injuries.

While the airway has a considerable capacity for repair, repair processes may not effectively restore normal epithelial architecture. For example, instillation of elastase into the airways of hamsters leads to acute airway injury. This process is followed by the accumulation in the airways of these animals of an increased number of secretory cells, which can persist for an extended period of time (21). This resembles the goblet cell metaplasia that characterizes the airways of smokers with chronic bronchitis and patients with asthma. Goblet cell metaplasia, however, is reversible under some conditions (22). The factors responsible for regulating the relative distribution of populations of epithelial cells within the airway remain to be defined.

Repair processes after injury also can lead to alterations in the subepithelial structures. In this regard, epithelial cells have been demonstrated to release factors that can drive fibroblast recruitment (10, 23), proliferation (24), matrix production (25), and matrix remodeling (26). Interestingly, some of the same mediators involved in epithelial repair may also drive mesenchymal cells. Fibronectin, for example, is a potent chemoattractant for both (7, 10). TGF- not only alters the epithelial cell phenotype but stimulates fibroblast matrix production and remodeling (27, 28). Other factors derived from epithelial cells, including insulin-like growth factor I (IGF-I) (29), may also regulate mesenchymal cell function, and some, such as prostaglandin E (PGE), may have inhibitory effects (8). Finally, factors regulating mesenchymal cell function may be derived from the inflammatory cells present in the airway wall. Whether injury results in an accumulation of abnormal fibrotic tissue will likely depend on a complex balance of factors derived from several sources.

Fibrotic peribronchial tissue, like other types of fibrous scar tissue, contracts. In this regard, epithelial cells, at least in part through release of TGF-, can drive fibroblast-mediated contraction of collagenous matrices in vitro (26). It has been suggested that airflow limitation in many patients with COPD is related to narrowing of the small airways (1, 2). Consistent with this concept, the presence of fibrosis correlates with loss of function. This raises several possibilities for therapeutic intervention. Specifically, blocking the accumulation of fibroblasts in peribronchiolar areas and their subsequent contraction and narrowing of the airways may offer an option to alter the relentless loss of lung function that most often characterizes COPD.

In vitro studies suggest that such processes are amenable to therapeutic approach. -Agonists, for example, can attenuate fibroblast-mediated contraction of collagenous matrices in vitro (30). Glucocorticoids, on the other hand, augment contraction, an effect mediated through inhibition of PGE production, which functions as an autocrine/paracrine inhibitor of contraction (31). How such in vitro effects will be related to in vivo responses remains to be defined. However, the altered repair response that characterizes the airway in COPD and likely leads to compromised function may be a target for therapeutic intervention.

	EMPHYSEMA

TOP ABSTRACT INTRODUCTION AIRWAY REPAIR EMPHYSEMA REFERENCES

Inflammation and repair processes also contribute to the development of emphysema (1). In this regard, elastin degradation within alveolar structures is believed to play an important role leading to the development of emphysema. Consistent with this, the congenital deficiency of the neutrophil elastase inhibitor -1 PI is associated with the development of emphysema and instillation of neutrophil elastase can lead to emphysema in animal models (32). Other elastases may also play a role in this process (1, 33). In particular, the macrophage- derived metalloelastase MMP-12 has been suggested to be of particular importance in cigarette smoke-induced emphysema in mice (34). MMP-12 knockout mice exposed to cigarette smoke do not get emphysema and do not have the macrophage accumulations that characterize both control mice and humans exposed to cigarette smoke. In this regard, MMP-12 may function to generate elastin fragments in the lung, which in turn may be responsible for macrophage recruitment (35).

In contrast to the normal lung, where elastin appears to be an exceedingly stable molecule with relatively little turnover (1), emphysema appears to be associated with increased elastin turnover. Patients with COPD have increased excretion of urinary desmosine, an elastin-specific cross-link, suggesting increased elastin degradation (36), and subjects with accelerated decline in lung function excrete larger amounts of desmosine (37). Similarly, after elastase exposure in animal models, elastin gene expression is upregulated (38). Tissue elastin concentrations, which initially drop after acute elastase exposure, can be restored to near normal (32). This indicates that the alveolar structures of the lung can also initiate repair responses after structural injury.

Several lines of evidence suggest that inadequate repair may contribute to the development of emphysema. Starvation, for example, which generally inhibits anabolic responses, can both lead to emphysema (39) and exacerbate the development of elastase-induced emphysema in animals (40). Consistent with this, individuals with a reduced body mass index have been noted to have increased mortality from COPD (41). Among individuals with similar degrees of severe airflow limitation, underweight individuals have a significantly reduced diffusion capacity for carbon monoxide, consistent with worse emphysema (42). In addition, inhibition of elastin synthesis by -amino propionitrile, a lathryogen that inhibits elastin cross-link formation, leads to worse emphysema in elastase-exposed animals (43). Interestingly, cigarette smoke can inhibit this enzyme, suggesting that smoke may cause emphysema not only by initiating inflammatory responses, but also by inhibiting repair (44). Also consistent with a role for cigarette smoke contributing to emphysema through inhibiting repair are direct inhibitory effects of cigarette smoke on lung mesenchymal cells. Specifically, cigarette smoke can inhibit fibroblast recruitment, proliferation, matrix production, and extracellular matrix remodeling (45, 46). Cigarette smoke can also impair the ability of epithelial cells to participate in a repair response (47).

The possibility that altered repair contributes to the development of emphysema also raises the possibility that therapeutic interventions may be able to alter the disease process. In this regard, the effects of nutrition and body weight noted above on COPD mortality and emphysema severity are of interest. Schols and colleagues (41) have noted reduced mortality in patients with COPD who were able to increase their body weight. Whether this was due to altered progression of emphysema remains to be determined. Animal studies, however, also support the concept that alveolar repair can be modulated therapeutically.

As noted above, alveolar injury (e.g., in the elastase model of emphysema) leads to initiation of repair responses including induction of elastin gene expression. Under most circumstances studied to date, however, this process does not effectively restore tissue integrity. In contrast, studies suggest that alveolarization, at least in some circumstances, can be stimulated by exogenous mediators (48, 49). Specifically, retinoic acid can induce the formation of alveoli in neonatal rats (48) and can reverse the inhibition of alveolarization caused by glucocorticoids (48). Retinoic acid can also induce the formation of new alveoli when given after the development of emphysema in elastase-exposed adult rats (49). These results raise the exciting possibility that, consequent to injury, repair responses that may be ineffective in normal adult animals may be manipulated by exogenous agents. Specifically, it may be possible to induce a recapitulation of normal development and thus to restore normal functioning.