Current Concepts in Mechanisms of Emphysema

Abstract

Over the last 10 years, there has been a remarkable degree of progress in our understanding of the pathophysiological mechanisms involved in the genesis of emphysema. This review attempts to summarize these data.

Introduction

Pulmonary emphysema represents one of the most common sources of morbidity and mortality in the developed world (Mannino, 2004). Although our ability to recognize emphysema, first grossly, then microscopically, and now through radiological algorithms has improved markedly, our understanding of the pathobiological processes has not increased until the last few years. Only with this understanding will we be able to formulate appropriate therapies. The following represents an attempt to discuss three of the basic processes that are involved in the development of emphysema. These are not mutually exclusive; rather, there is extensive interaction among all of the processes. Some, such as apoptosis, are speculative, while others, for example, the role of inflammatory cells, are more fully understood.

Emphysema as a Consequence of Proteolytic/Antiproteolytic Imbalance

There is longstanding and incontrovertible evidence of an increase in neutrophils and macrophages in the lungs of cigarette smokers (reviewed in Shapiro, 2000; MacNee, 2005), and it appears that, in smokers who develop COPD, this increase is persistent even after smoking cessation (Retamales et al., 2001), resulting in a continued increase in proteolytic enzymes even after the putative inducing agent has been removed (Shapiro, 2001).

The traditional inflammatory cell implicated in emphysema is the neutrophil and the protease, neutrophil elastase, although neutrophils also secrete cathepsin G and matrix metalloproteinase (MMP) 9. Neutrophil elastase is inhibited by alpha-1-antitrypsin (AAT), and mice treated with exogenous AAT have significantly reduced cigarette smoke-induced airspace enlargement (Churg et al., 2003b). Similar degrees of protection have been obtained by knock-out of the neutrophil elastase gene (Shapiro et al., 2003). These findings confirm the importance of the neutrophil in emphysema.

More recently attention has been focused on the role of the macrophage and macrophage-derived proteases. Macrophages secrete several cathepsins, as well as MMP 2, 9, and 12, and these MMPs are inhibited by the tissue inhibitors of metalloproteinases (TIMP). There is considerable interaction between the metalloproteinases, with MMP 1 activating MMP 2, MMP 2 activating MMMP 1 and MMP 12, and MMP 13 activating MMP 9 (McCawley et al., 2001). Which MMPs are actually important in emphysema is a controversial issue. Increased tissue levels MMP 1 and 9 have been found in the lungs of humans with emphysema (Ohnishi et al., 1998; Iami et al., 2001).

When mice are genetically modified so that they secrete human MMP 1, they develop emphysema (D’Armiento et al., 1992; Foronjy et al., 2003); by contrast, when mice were modified to knockout MMP 12, they are totally protected against smoke induced emphysema (Hautamaki et al., 1997). MMP-12 appears to be central to the induction of emphysema in a variety of (nonsmoke) mouse models (reviewed in (Churg et al., 2005)). However, mice lacking MMP 9 are not protected against interleukin -13 induced emphysema (Lanone et al., 2002). Thus both collagenases and elastases appear to be important in emphysema, but the different metalloproteinases may play different roles in animals compared to humans.

What has not been recognized until recently is that there is a complex inter-relationship between neutrophils and macrophages. Both MMP 12 and neutrophils are required for matrix breakdown in mice exposed to cigarette smoke (Churg et al., 2002b), and the process requires TNFα (Churg et al., 2002a, 2003a). In the proposed mechanism of interaction (Churg et al., 2005). MMP 12 liberates active TNFα from the macrophage surface. This then activates the endothelium resulting in neutrophil migration and secretion of neutrophil elastases. Moreover, MMP 12 is able to inactivate AAT while neutrophil elastase destroys TIMP-1, thus potentiating the action of each protease (Shapiro et al., 2003).

While the above mechanism demonstrates a central role for TNFα in neutrophil recruitment and elastase secretion, it must be stressed that TNFα is only part of the process. TNFα transgenic mice develop only a limited degree emphysema (Vuillemenot et al., 2004), and mice modified to lack the type I and II TNFα receptors only show approximately 70% protection from smoke induced emphysema (Churg et al., 2004). Other inflammatory mediators may be important; for example, IL 13 and IFNγ both appear to act through activation of metalloproteinases (reviewed in Mahadeva et al., 2002; Tuder et al., 2003a; Wright et al., 2003). Phosphodiesterases (PDE) degrade cyclic nucleotides, and PDE4 specifically degrades 3′,5′ adenosine monophosphate, the substance that acts as a second messenger to activate inflammatory cells. Inhibition of inflammation by a phosphodiesterase inhibitor was able to completely abolish cigarette smoke induced emphysema (Martorana et al., 2005), although it is not clear what other processes were inhibited in these experiments

Finally, oxidants probably play a role in the development of emphysema. Cigarette smoke is a highly concentrated source of both reactive oxygen and reactive nitrogen speices; as well, inflammatory cells produce oxidant (reviewed in MacNee, 2005). Macrophages from cigarette smokers appear to release increased amounts of oxidants (Schaberg et al., 1992), and lipid peroxidation products are elevated in the lungs of patients with COPD (Rahman et al., 2002).

Oxidants are believed to inactivate AAT, thus potentiating the effect of serine proteases, and are able to activate pro-MMPs, thus increasing the metalloproteinase burden. Oxidants can inhibit elastin cross-linking, leading to inadequate repair (reviewed in Owen, 2005). As shown in Figure 1, and as discussed in the following sections, oxidative stress is part of several processes, leading to emphysematous lung destruction.

Emphysema as a Result of Disruption of the Lung’s

Homeostatic Maintenance and Repair System

There are several excellent reviews of this subject (Tuder et al., 2003a, 2003b;Voelkel et al., 2006). Figure 2 is a cartoon that demonstrates a very simplified homeostatic system, and illustrates that in the normal lung VEGF is central to this system. As cells undergo apoptosis, largely induced by oxidative stress, they express surface lipid ligands that target them to receptors on phagocytes and result in their being engulfed by phagocytes in a process that has been termed “efferocytosis.” VEGF enhances this phagocytic process (Voelkel et al., 2006), which in turn, increases VEGF secretion (Golpon et al., 2005) and survival signaling, namely increased expression of Bcl-2 which decreases apoptosis, and up-regulation of MnSOD, which reduces oxidative stress (Bratton et al., 2005). Finally, there is some early evidence that suggests that VEGF enhances cell proliferation (Bratton et al., 2005), and thus maintains lung homeostasis.

Disruption of this system could, therefore, lead to disease, with a complex interaction between increasing apoptosis, increased serine and metalloproteinase secretion, leading to airspace enlargement and emphysema (Figure 1). Interruption of the VEGF system by VEGFR2 blockade (Kasahara et al., 2000), or by antibodies against VEGFR2 (Taraseviciene-Stewart et al., 2005), or by disruption of the VEGF-VEGFR2 complex by cigarette smoke (Marwick et al., 2006) all are associated with increased apoptosis.

There appears to be an integral relationship between induction of apoptosis and oxidative stress. Apoptosis as a consequence of VEGFR2 inhibition can be inhibited with a superoxide dismutase mimetic (Tuder et al., 2003c), and alternatively, apoptosis can be induced by application of an oxidative stress (Tuder et al., 2003c). Cigarette smoke exposure is associated with increased apoptosis in rats (Kuo et al., 2005), and there is increased oxidative stress in smoke exposed mice (Aoshiba et al., 2003a). Apoptosis generally involves activation of the caspase cascade (see (Tuder et al., 2003b) for a review). However, the addition of activated caspase 3 (Aoshiba et al., 2003b), or the more upstream regulator ceramide (Petrache et al., 2005) will also induce apoptosis. In addition, ceramide inhibition will prevent VEGFR2 receptor blockade induced apoptosis and oxidative stress. (Petrache et al., 2005), both of which appear to be integral to the genesis of emphysema in this model (Tuder et al., 2003c).

It must now be asked whether the above animal data have any translation into human disease, particularly emphysema induced by cigarette smoke. Increased apoptosis has been found in the alveoli of patients with emphysema (Yokohori et al., 2004; Calabrese et al., 2005; Imai et al., 2005), although it is largely restricted to lungs with severe emphysema. There are increased levels of ceremide in smokers, both with and without emphysema (Petrache et al., 2005). In one study (Marwick et al., 2006), VEGFR2 protein expression levels were similar in the lungs of smokers and nonsmokers, although there was decreased expression of VEGFR2 in the smokers with COPD, while another (Lee et al., 2001) demonstrated decreased VEGF and VEGFR2 protein, but only in severe emphysema.

While these studies certainly suggest a link between human cigarette smoke induced emphysema and abnormalities of the lung maintenance and repair system, it remains troubling that other than ceremide, the abnormalities are only present in severe emphysema. Further, the presence of apoptotic alveolar epithelial cells does not in itself imply that apoptosis drives emphysema, since apoptosis of epithelial cells could equally well be secondary to matrix destruction by proteases. In this context it should be noted that alveolar cell apoptosis is also increased in fibrotic lung diseases such as usual interstitial pneumonia, and adult respiratory distress syndrome.

Emphysema as an Immunlogical Process

There is increasing speculation about the role of autoimmunity in the genesis of emphysema (Agusti et al., 2003; Cosio, 2004; Grumelli et al., 2004), and Figure 3 provides a simplistic outline of the possibilities. Firstly, many studies have found increased numbers of T lymphocytes, not only in the airways (Bosken et al., 1992; O’Shaughnessy et al., 1997; Saetta et al., 1998; Sun et al., 1998; Lams et al., 1999, 2000; Turato et al., 2002; Fabbri et al., 2003; Hogg et al., 2004), but also in the emphysematous lung parenchyma (Finkelstein et al., 1995; Saetta et al., 1999; Majo et al., 2001; Cosio et al., 2002; Grumelli et al., 2004).

Both CD4 and CD8 T cells are increased, and more recently, the T cells have been shown to have a Th1 phenotype (Grumelli et al., 2004) with increased expression of interferon-γ and activation of the STAT-4 system (Di Stefano et al., 2004). Furthermore, one study demonstrated oligoclonal CD4 T cell populations in severe emphysema (Sullivan et al., 2005). It is intriguing that increases in T cell numbers appear to be associated with increases in MMP-12 release (Grumelli et al., 2004). Finally, a recent study has reemphasized the role of the B lymphocyte, with oligoclonal proliferation found in both parenchyma and airways (van der Strate et al., 2006).

There are several findingsthat imply that there is an adaptive immune response involved in the genesis or perpetuation of the emphysematous process (Agusti et al., 2003; Barnes et al., 2004; Cosio, 2004). Firstly the magnitude of the inflammatory infiltrate appears to roughly correlate with emphysema (Finkelstein et al., 1995), and also parallels the level of apoptosis (Majo et al., 2001). Secondly, the inflammatory infiltrate persists, often after many years, after smoking cessation (Retamales et al., 2001); and third, although many types of inflammatory cells appear to be participants, the expansion of dendritic cells, B cells and T cells suggests activation of the innate and adaptive immune response. Finally, emphysema can be induced by the transfer of CD4 lymphocytes from an animal immunized against endothelial cell VEGFR2 (Taraseviciene-Stewart et al., 2005).

If activation of the immune response is important in emphysematous lung destruction, one must also ask what mechanism would instigate or drive such a response. Tobacco smoke itself is a complex xenobiotic amalgam of proteins, polysaccharides, and free radicals which could act as antigens to induce or perpetuate an immune response. Smoke could also damage the lung and alter or expose lung proteins that, in turn, could act as targets, resulting in loss of tolerance to self-epitopes. Animal indicate that antibodies against VEGFR2 on the endothelial cells will induce apoptosis and emphysema (Taraseviciene-Stewart et al., 2005), but thus far there is no evidence that such antibodies exist in humans.

Finally, when a cell undergoes apoptosis, it ordinarily exposes phosphatidylserine on the plasma membrane, resulting in its uptake and removal (Fadok et al., 2001; Hoffmann et al., 2001), but oxidation of membrane phospholipids appears to result in the formation of a recognition ligand that actually promotes an inflammatory response (discussed in references Fadok et al., 2001; Bratton et al., 2005). Thus, oxidants within cigarette smoke could act by not only inducing apoptosis, but by altering the exposed plasma membrane lipids. Finally, it is quite possible that colonization of the airways by bacteria, or persistence of viral antigens in the lung parenchyma could represent a continued source of new antigens to perpetuate the inflammatory reaction (Agusti et al., 2003).

Summary

The mechanisms involved in the genesis of emphysema are not fully understood. While this review has divided possible mechanisms into distinct areas, it is quite likely that emphysema may be a result of the interplay of the inflammatory and immunological systems, resulting in abnormal maintenance and repair of the lung.