Stress Responses Affecting Homeostasis of the Alveolar Capillary Unit

Abstract

The maintenance of the alveolar structure is required throughout life. To accomplish this goal, alveolar cells, including endothelial, epithelial, and fibroblastic cells, provide key molecules with broad survival and antiapoptotic effects. These complex interactions are disrupted by cigarette smoke, leading to emphysema. Smoke imposes an environmental stress to the lung with the activation of “sensor-like” molecular signaling. Activation of RTP801, leading to mTOR inhibition, is paradigmatic of these responses. The accumulation of cellular damage, with the generation of endogenous mediators of inflammation, may proceed toward an aging phenotype. These alterations may impose significant challenges to cell-based regenerative or pharmacological therapies.

The destruction of the alveolar structure is a key component of a large number of pulmonary diseases. A distinct pathological phenotype is the disappearance of alveolar septa, which is characteristic of emphysema, particularly in the setting of chronic obstructive pulmonary disease (COPD) caused by cigarette smoke. Adult-onset emphysema shares the morphological features present in the chronic stage of bronchopulmonary dysplasia (1), which results largely from a disruption of lung development and maturation and growth. These shared features suggest that similar molecular mechanisms may account for the emphysematous alveolar phenotype in both disease settings. Lung development and the expansion of alveolar surface areas require a complex set of coordinated molecular signals. Although alveolar formation ceases in the first decade of life, continuous maintenance of alveolar structure is required for proper preservation of gas exchange structures, including alveolar capillaries, later in life. We have proposed that disruption of the molecular signals involved in alveolar structural maintenance may play a central role in emphysema development due to cigarette smoke (2). Many of the molecular signals involved in lung development and growth are required to maintain the lung throughout adulthood, which therefore can explain the similar pathological manifestation of emphysema in COPD and bronchopulmonary dysplasia, with reduction of alveolar structures and decrease in overall gas exchange surface.

COPD has been considered a disease resulting from chronic inflammation, leading to extracellular matrix remodeling (3). We have recently summarized that the interplay between injurious agents may synergize with endogenously released danger-associated molecular patterns (DAMPS) in promoting pulmonary injury in several lung diseases (4) (Figure 1), many of which are also characterized by inflammation, such as COPD. In fact, several DAMPS represent cellular components, such as high-mobility group box 1 proteins (HMGB-1) or mitochondria components, components of extracellular matrix molecules (e.g., collagen or elastin fragments), or signaling molecules involved in pathogen recognition (4). The conceptual advances derived from the identification of DAMPS in chronic inflammatory states have led to pathogenetic breakthroughs in the understanding of degenerative diseases, with the design of novel therapies based on these mechanisms (5). Moreover, aging may accelerate the development of self-sustained inflammatory responses. It is therefore conceivable to conclude that the generation of DAMPS may result from a global failure of cellular maintenance.

Figure 1.

Outline of the spectrum of airway-mediated injurious agents and activation of danger-associated molecular patterns (DAMPS) (see Reference 4) operative in multiple lung diseases, including chronic obstructive pulmonary disease. Many of injurious agents engage toll-like receptors and may also act as DAMPS. Some of these signaling events engage signaling systems involved in protection of the host from environmental stresses, including hypoxia and deprivation to nutrients. Progression and persistence of alveolar injury would involve the release of several endogenous mediators, which converge in promoting organ dysfunction and persistent inflammation (alveolar outline based on Reference 74). EC = extracellular; NLR = nucleotide oligomerization domain-like receptor; PGP = proline-glycine-proline; TLR = toll-like receptors.

We present the concept that lung tissue injury by cigarette smoke, particularly in COPD, involves the following three stages: stage 1, initiation; stage 2, progression; and stage 3, propagation of cell damage (6). There is growing evidence that a fourth stage of “consolidation” of the pronounced cellular damage exists, leading to an aging phenotype, the ultimate manifestation of a global failure of lung maintenance (Figure 2) (7). Teleologically, the molecular players involved in the initial reaction of host stress responses caused by tobacco smoke, the direct organ damage causing progression of injury, and propagation of tissue injury are intimately involved in the process of aging. Previous reviews have emphasized the extent to which the pathology and molecular events underlying COPD emulate similar events underlying organ aging (8, 9). In accord with this concept, stresses, stress sensing, and activation of these endogenous molecular pathogenetic processes are intimately associated with the current paradigms of aging and senescence (10). These processes lead to inflammation, which becomes a consequence (rather than the cause) of all these destructive processes. This complex pathobiologic scenario may impart a formidable task to develop cell-based or pharmacological therapies once these processes are operational in the COPD lung.

Figure 2.

Endogenous signals that can propagate lung inflammation in emphysema.

This review highlights three specific aspects related to diseases that alter the steady-state homeostasis of the alveolar structure, with a particular focus on the impact of cigarette smoke on lung injury. We discuss (1) how the lung handles the stress imposed by cigarette smoke (i.e., initiation stage), leading to progression of injury; (2) the propagation of damage responses, with the activation of endogenous mediators of tissue injury, including DAMPS; and (3) the aging-related phenotype of COPD lungs (i.e., consolidation stage).

Initiation of cigarette smoke–induced lung injury: RTP801/Redd1 and lung injury due to cigarette smoke

Over the past 12 years, novel insights have provided support for the hypothesis that the disruption of alveolar structural and molecular maintenance programs by cigarette smoke contributes to the pathogenesis of emphysema (6). Based on observations that alveolar cell apoptosis interacts with oxidative stress in the causation of emphysema induced by vascular endothelial growth factor receptor (VEGFR) blockade (11) (and subsequently confirmed by genetic deletion of lung VEGF [12]), we proposed that a feed-forward interaction among apoptosis, protease/antiprotease imbalance, and oxidative stress mediates alveolar destruction in emphysema (11).

The eventual disruption of organ maintenance by these processes is mediated by a prototypic organ response to the inhalation of cigarette smoke. Environmental stresses, such as those caused by cigarette smoke and pollutants, activate trigger molecules that govern harmful alveolar septal cell responses characterized by oxidative stress, apoptosis, and alveolar inflammation. In line with this overall paradigm, we undertook mechanistic studies targeting the cellular stress response gene RTP801 (also known as Redd1 for “regulated in development and DNA damage responses”), which was initially discovered based on a screening for genes up-regulated by hypoxia-inducible factor 1α (HIF-1α) and shown to induce oxidative stress and apoptosis (13, 14) (Figure 3).

Figure 3.

RTP801 may act as a stress sensor, amplifying alveolar injury triggered by cigarette smoke with cellular processes of apoptosis, inflammation, and oxidative stress. Because these properties are dependent on RTP801-mediated inhibition of mTOR, the role of RTP801 can be reproduced by rapamycin, which phenocopies the role of RTP801 overexpression in wild-type mice when administered to RTP801 knockout mice.

In subsequent studies, however, the mutual interaction between RTP801 expression and oxidative stress was underscored by the documentation that arsenate induces RTP801 expression via activation of C/CAAT regions in the RTP801 promoter (15). This interaction was also supported by the report of induction of RTP801 expression in hypoxic neurons in vivo (13) and the protection of RTP801 KO mice from apoptosis in the neural retina and central vascular obliteration secondary to endothelial cell apoptosis in a model of retinopathy of prematurity (16). Conversely, forced overexpression of RTP801 in cultured cells and in the lung has been linked to apoptosis (13) and enhanced production of reactive oxygen species (14). We therefore hypothesized that RTP801 would function as a “switch,” integrating the lung's reaction to cigarette smoke acting as an environmental stressor (Figure 3). Cell signaling involving RTP801 would provide a supplementary level of control of key cellular events, particularly inflammation, in addition to traditional inducers of inflammation and cell injury (e.g., toll-like receptors [TLRs] and the ensuing cytokines and chemokines).

We observed that smokers with mild COPD had increased lung levels of RTP801 mRNA when compared with normal individuals and patients with severe COPD (17). However, RTP801 protein levels were increased in this latter group of patients. RTP801 mRNA and protein were also augmented in lungs of mice exposed acutely to cigarette smoke, preferentially in type II cells (Figure 2). Although the initial rise in lung transcript levels was followed by a decrease to baseline over 48 hours postexposure to cigarette smoke, RTP801 proteins levels remained high. These findings agree with the observed RTP801 protein stabilization in cells exposed to arsenite or oxidants (18). We did not detect RTP801 in inflammatory cells retrieved by bronchopulmonary lavage. The up-regulation of RTP801 by cigarette smoke was dependent of oxidative stress in vivo, based on its blockade with N-acetyl cysteine supplementation, and in vitro, which required the C/CAAT response element (17).

RTP801 expression was necessary to activate inflammation due to cigarette smoke in vivo and in vitro. RTP801 knockout mice were remarkably protected against inflammation, apoptosis, and oxidative stress caused by short-term exposure to cigarette smoke; consistent with these observations, these mice did not develop emphysema after 6 months of exposure to cigarette smoke when compared with wild-type mice (17).

RTP801 acts as an inhibitory switch of the serine–threonine kinase mTOR signaling in mammalian cells (19) and in drosophila (20). Nutrients, particularly amino acids (e.g., glutamine or leucine), lipids, and glucose, activate mTOR, which drives hypertrophic and hyperplastic cell growth responses. Conversely, mTOR is suppressed by the complex TSC–1/TSC–2 under adverse conditions (including low energetic states), by hypoxia, and by cigarette smoke (21). mTOR is a central component of two well defined, evolutionary conserved complexes named mammalian TORC1 and TORC2. mTORC1 involves the association of mTOR with Raptor (for “regulatory associated protein of TOR”), mSLT8/GβL, Deptor, and PRAS40 and is primarily involved in cell growth signals. mTORC2, a multiprotein complex containing mTOR and Rictor (for “rapamycin insensitive companion of TOR”), activates AKT and mediates the reorganization of the cytoskeleton.

In line with these observations, we noted that RTP801 knockout mice have increased baseline lung mTOR signaling when compared with wild-type littermates. The increased mTOR in the knockout mice mediated alveolar protection, as its blockade with rapamycin abrogated the protection against cigarette smoke and triggered alveolar inflammation and apoptosis in wild-type mice.

RTP801 overexpression in ambient air–exposed mice triggered alveolar inflammation, oxidative stress, and alveolar cell apoptosis (Figure 4). These effects were related to the mTOR inhibitory activity of RTP801 (and are in accord with the effects of rapamycin in the lung) because a RTP801 mutant protein (variant RPAA), unable to bind to TSC-2 and displace the 14–3–3 adaptor (22), did not cause the lung pathologies seen with the wild-type construct (17). These findings confirm the report that blockade of mTOR, such as that achieved by rapamycin or RTP801 overexpression, can be proinflammatory (23–25). In line with these findings, HIF-1α, an inducer of RTP801 expression, activates NF-κB under conditions of hypoxia (26). These observations agree with the reports that rapamycin causes lymphocytic pneumonitis in transplant recipients (27). However, the impact of mTOR activation on cigarette smoke–induced lung injury can be cell-, disease-, and organ-context specific. In contrast with the findings in the RTP801 knockout mice, we observed that rapamycin protected against acute cigarette smoke–induced inflammation in RTP801 wild-type mice, which might have been due to mTORC1-driven IKKβ activation, particularly in inflammatory cells (28). In accord with these observations, rapamycin-induced protection was observed in experimental asthma (29).

Figure 4.

RTP801 expression in mouse lungs exposed to cigarette smoke for 3 days (red), costained with (A) prosurfactant protein C (green, type II cells), (B) T1α (green, type I cell), and (C) endothelial cells (green, thrombomodulin). (D–F) Expression of RTP801 (red) in type II cells (green) in vector control (D), mutant RTP801 unable to inhibit mTOR (RPAA) (E), and human RTP801 (F) in transduced mice lung (3 days postinoculation). (G–I) Lung macrophages (immunohistochemistry, brown) in vector control (G), RPAA- (H), and RTP801 transduced mouse lungs exposed to ambient air (I) (based on Ref. 17).

The finding that reduced mTORC1 signaling accounts for most of the pathology of RTP801 deregulation (17) may have broader pathogenetic and potential therapeutic implications for COPD. Muscle wasting due to hypoxia (30), a major contributing factor on morbidity and mortality in advanced COPD, has also been linked to up-regulated RTP801 expression (30), cell apoptosis, and oxidative stress (31). Furthermore, mTOR is a key blocker of autophagy, a process of cell defense due to accumulation of cellular compartments within lysosomal-like structures, through sensing the abundance of nutrients in the cellular microenvironment. In this setting, mTORC1 binds and inhibits the initial molecular activators of autophagy, involving ULK-1/2, Atg13, and FIP200. Rapamycin inhibits mTORC1 and therefore simulates a state similar to that caused by nutrient deprivation or stress responses (including hypoxia and oxidative stress), causing autophagy. The observation of excessive autophagy in the lungs of patients with emphysema (32) appears to be detrimental because blockade of autophagy in mice lacking Atg8 (microtubule-associated light chain 3) confers protection against cigarette smoke–induced emphysema (33). Taken together, these data may link baseline mTOR signaling in the protection against cigarette smoke–induced lung injury and its broad antiautophagic role.

Progression of alveolar injury into emphysema

Continuous lung injury leads to the destructive disappearance of alveolar structures, characteristic of emphysema. Although this destruction results from a chronic exposure to cigarette smoke, cigarette smoke also triggers acute inflammatory responses and episodes of oxidative stress. It is conceivable that emphysema would develop acutely if the influx of inflammatory cells and the ensuing protease/antiprotease imbalance were the sole trigger of alveolar destruction. Experimental data suggest that these events, though critical in the pathogenesis of the disease, may require chronic disruption of alveolar cell and molecular maintenance, overcoming protective mechanisms that allow preservation of alveolar architecture. This concept has been supported by the observations that (1) epithelial cell–specific deletion of VEGF coreceptor neuropilin-1 predisposes to accelerated cigarette smoke–induced alveolar cell death and emphysema (34), (2) mice lacking the master transcription factor nuclear erythroid-related factor 2 develop more severe and early-onset emphysema due to cigarette smoke exposure (35), and (3) mice prone to early aging have accelerated emphysema associated with alveolar cell death when compared with wild-type littermates (36). Chronic cigarette smoke may act on the lung to progressively impair its cellular and maintenance program (6), including down-regulation of VEGF and VEGF coreceptors (37). The down-regulation of VEGF receptor signaling or decreases in lung VEGF may then lead to apoptosis, with the potential activation of proinflammatory destructive mechanisms (12, 38, 39).

Based on data that showed the feedback interplay between oxidative stress and apoptosis in the VEGF receptor rat model of emphysema, we postulated that the mutual interaction among these destructive processes (11), in addition to the protease/antiprotease imbalance, might set in motion and amplify the alveolar tissue destruction (40). These interactions were also verified when endothelial cell apoptosis was directly targeted by a lung homing peptide linked to a mitochondria-pore forming peptide, in which apoptosis was accompanied by increased macrophage influx and oxidative stress (41). These experiments indicated that the disruption of alveolar integrity based on capillary endothelial cell death reproduced several of the key findings seen with cigarette smoke.

The vascular compartment within alveolar septa plays a central role in maintaining lung structure during the attack promoted by chronic cigarette smoke exposure. In opposition to the documented regenerative ability of type II cells, capillary endothelial cells may not be able to reconstitute the native alveolar capillary network once cell death ensues. Whether either one of the cells coordinates the final fate of alveolar septa in emphysema is an important question that will require clear and stringent pursuit. Although the “vascular hypothesis” of emphysema was introduced more than 50 years ago (42), it has been recently reinforced by the finding of increased levels of apoptotic endothelial cell microparticles in the circulation of patients with COPD with no or minimal emphysema but with decreased Dl_CO (43). Notwithstanding the evidence that type II cell apoptosis might also occur in emphysema (44), endothelial cells might represent the key cell driving the pathogenesis of destructive airspace enlargement. This interpretation might be supported by the evidence that type II cell apoptosis seen in other lung disease processes, such as acute lung injury (34), are not associated with emphysema development.

Propagation of alveolar injury: role of endogenous amplification of alveolar tissue destruction

A vexing observation pertaining to the so-called inflammation-centric hypothesis of COPD is the lack of clear therapeutic responses with immunosuppressive drugs and persistent inflammation despite smoking cessation (45). An alternative hypothesis is that an autoimmune mechanism is engaged (46), notably in advanced stages of the disease (47). The authors propose that infectious, environmental, and endogenous mediators (released due to cell injury caused by cigarette smoking and known as DAMPS) may amplify lung inflammation by binding TLR or alternative signaling receptors, such as the receptor for advanced glycation end products (RAGE) (4) (Figure 1). Like the skin and the gastrointestinal tract, the lung serves as a conduit for several infectious and environmental agents that may trigger chronic illnesses. Infectious stimuli can activate shared inflammatory pathways via pathogen-associated molecular patterns (PAMPS). This infection-triggered paradigm is supported by several examples of chronic pulmonary diseases involving bacteria, viruses, and parasitic agents. Infections by bacteria (e.g., Hemophilus influenza) and viruses (e.g., influenza and parainfluenza) lead to significant disease worsening and increased mortality in COPD and in disease progression due to asthma (48); γ-herpes viruses have been linked to the pathogenesis of fibrotic lung diseases (49) and pulmonary hypertension (50), and the infection by the parasite Schistosoma mansoni has been associated with pulmonary hypertension (51).

The lung is uniquely affected by environmental agents, such as cigarette smoke and pollution, which can activate a series of molecular pathways involved in COPD pathogenesis (52). Shared inflammatory pathways have been proposed among infectious and environmental agents, particularly those initiated by the activation by PAMPS (53). Furthermore, infection and environmental triggers interact to promote tissue damage via the activation of several endogenous molecules and pathways. These processes enhance proinflammatory signals and increase inflammatory cell influx and expression of cytokines and chemokines (Figures 1 and 2). Products derived from dying cells acting as DAMPS elicit potent inflammatory stimuli, which synergize with infectious agents in the promotion of tissue injury. Indeed, many DAMPS can be triggered by infectious agents and environmental exposures, underscoring the potential for DAMPS or their downstream signaling events as attractive target(s) for disease modification and treatment.

A growing number of DAMPS have been linked to COPD (Figures 1 and 2). Elastin fragments attract monocytes (54); the tripeptide proline-glycine-proline, originated from degradation of collagen, activates the CXCR2 receptor (55). HMGB1 can activate TLR4 and RAGE (56). Many of these products are generated from dying cells, including apoptotic cells inefficiently cleared in the COPD lung (57). These processes may be amplified by ceramide, a lipid second messenger implicated in cigarette smoke–induced emphysema. We have highlighted the potential amplifying role of ceramide, a second messenger lipid signaling molecule, in emphysema because ceramide is generated during stresses (including cigarette smoke) (58). Recently, Petrache and collaborators showed that ceramide not only increases the rate of cell death but also disrupts apoptotic cell clearance (59). These data and the concept of accelerated aging underscore the alternative view that lung inflammation in COPD is driven from the “within” rather than from the stimuli originated from cigarette smoke (Figure 3).

Consolidation of lung injury: Shared mechanisms between aging and COPD

Studies performed in rats with pancreatic elastase-induced emphysema demonstrated that all-trans retinoic acid (ATRA) rescued the alveolar destruction (60). These promising results were followed by clinical trials aimed at addressing whether ATRA would also provide a therapy for cigarette smoke–induced emphysema (61). Because these expectations were not realized (62), it is important to determine why ATRA did not work and, by extension, whether cell regenerative therapies might provide relief from the progressive nature of emphysema. Several studies have recently revealed that the emphysematous lung contains several features seen in aged organs and senescent cells (7), including shortened telomeres, activation of senescence-associated cyclin kinase inhibitors (such as p16^Ink4a, p19^Arf, and p21^{CIP1/WAF1/SDI1}), and macromolecular modifications indicative of oxidative stress (63, 64) (Figure 3). These molecular signatures of aging and senescence complement earlier observations of marked extracellular matrix modification, including exaggerated collagen deposition in advanced emphysema (65). We have recently reported increased DNA strand breaks in the promoter regions of the VEGF gene that might explain the decreased VEGF levels seen in COPD lungs (66). In light of these alterations, chronic cigarette smoking substantially modifies the lung structure, disrupts its maintenance program, and curtails its repair ability, suggesting that regenerative therapies might not be feasible or efficacious.

There are additional important implications as the lung becomes prematurely aged. Senescent cells assume a senescence-associated secretory phenotype with the production of cytokines, notably of IL-6 and IL-8, which are increased in patients with COPD. These cytokines may further drive the senescence phenotype in the emphysematous lungs (67). A recent observation has linked senescence, dysregulated cytokine production, Klotho (an anti-aging molecule that protects against development of emphysema [68]), and activation of viral PAMP-associated signaling. Klotho, which is decreased in aged cells, blocks the activation of retinoic-inducible helicase (RIG-I). RIG-I recognizes viral double-stranded RNA and activates a signaling process resulting in NF-κB activation and NF-κB–dependent cytokine IL-6 (69). Inappropriate RIG-I activation by poly I:C and influenza virus mediate accelerated emphysema in mice subjected for a short exposure of cigarette smoke (70), potentially providing additional insights into the link between cigarette smoke, viral PAMPs, and COPD-associated lung aging.

Finally, the recent demonstration that telomerase-deficient mice are more susceptible to cigarette smoke–induced lung injury provides compelling evidence that key molecular mechanisms involved in aging are also central in the emphysema pathogenesis (71). These data provide elegant mechanistic insights of erosion of telomere lengths in alveolar cells in COPD lungs (72) and peripheral blood mononuclear cells (73).

Conclusions

Cigarette smoke may promote injury not only because of the generation of thousands of oxidants per puff but also because it activates powerful endogenous destructive mediators. These regulators of innate immunity, with the aid of acquired immunity targeted at lung antigens or to infectious agents, promote the relentless destruction of the lung. The overall paradigm imposes significant challenges for therapy development. Late-stage disease may have a multitude of pathogenetic processes that cannot be counteracted by single-target approaches. An alternative resides in fortifying antipathogenetic processes that limit these amplifying processes. The better understanding of which processes are activated and how they are activated may indicate how best to intervene to modify the course of the disease.