Fundamental research discoveries inevitably drive conceptual shifts that bring us closer to scientific truths. In medicine, this effort translates into better health and living standards. However, as in most if not all branches of human knowledge, these discoveries carry a price. Traditional conceptual frameworks or paradigms cannot always explain newly generated knowledge obtained by experimentation and careful interpretation of scientific data. What once was new and revolutionary now becomes outdated and needs to be readdressed within a new conceptual framework. In fact, the past and the inevitable future clash and, from this dialectic (i.e., conflict of opposites), a new understanding emerges.

The article by Yao and coworkers in this issue of the AJRCMB (pp. 7–18), while exploring well-established paradigms of cigarette smoke–induced lung injury, provides data that support a potential conceptual shift regarding how the lung is damaged by persistent exposure to cigarette smoke (1). This paradigm has evolved based on the resemblance between emphysema and aging lung processes (4), the impact of lung injuries promoted by lifelong exposure to cigarette smoke, and the potential for an autoimmune component of inflammation in advanced stages of the disease (5). The inflammation and protease–antiprotease hypothesis of chronic obstructive lung disease represents the traditional paradigm or the status quo that aligns linearly cigarette smoke and inflammation (6). This hypothesis can be dated to the identification of α₁-antitrypsin deficiency (7), which accordingly would allow for the unopposed destructive actions of neutrophil elastase toward the extracellular matrix. The inflammation/protease–antiprotease hypothesis was further supported by the identification of the role of metalloproteases, particularly of metalloprotease-12 (8), and the requirement for neutrophils (9), macrophages (10), and more recently, cytotoxic lymphocytes in the alveolar destruction promoted by cigarette smoke (11). It is always tempting to relate these cellular events to the paradigm of protease–antiprotease imbalance, that is, the pathway of “less resistance” and easier acceptance. However, may a rigid interpretation of the inflammation/protease–antiprotease hypothesis blind one to a more accurate interpretation of the true pathobiological events in lung injury caused by cigarette smoke? Concepts such as the influx and pathogenetic role of inflammatory cells can be explained by an autoimmune attack on lung components damaged by cigarette smoke (5, 12) or superimposed viral infections (13), which cast doubt on the linearity of cigarette smoke as a mere “proinflammatory” agent. Indeed, anti-inflammatory strategies remain to be demonstrated as effective treatments for chronic obstructive pulmonary disease (COPD). Even the traditional concept that elastases are the main and sole target of α₁-antitrypsin is being expanded by the evidence that this serpin blocks apoptosis by binding and inhibiting active caspase 3 (14, 15), has direct anti-inflammatory effects (16), and neutralizes the activation of protease activated receptor 1 by thrombin (17).

The work by Yao and colleagues reveals that the cyclin-dependent kinase inhibitor p21^{CIP1/WAF1/SDI1} (designated hereafter as p21), which has been linked to cell cycle growth arrest due to stress or “danger” cell responses, may modulate alveolar inflammation and alveolar destruction, and thus enlightens our present understanding of how the lung “senses” injury due to cigarette smoke and integrates these responses with those that activate inflammatory pathways potentially harmful to the lung (Figure 1). Furthermore, the interplay of p21 and cellular processes involving cell senescence and the imbalance of cell proliferation/apoptosis may provide us with a more logical explanation of how p21, acting as a sensor of cellular stress, might have such potent and wide roles in lung responses triggered by cigarette smoke. Molecular switches, ontologically designed for the protection of the host, are now “hijacked” by injurious stresses (such as cigarette smoke), leading to organ damage. This concept, which is shared with the underlying processes that lead to aging, may also apply to several diseases, including COPD (18). It is apparent that a novel paradigm is emerging. This paradigm requires that we consider how cigarette smoke alters the molecular controls of lung cell homeostasis, leading to disruption of alveolar maintenance (19), with the destructive consequences of failure to repair and terminal senescence. These pathobiological processes may provide an alternative or be complementary to the models of unopposed apoptosis and extracellular matrix destruction.

Cell cycle progression to the S phase is regulated by complexes of cyclins and cyclin-dependent kinases (Cdks) (20). p21 acting as a central downstream target of p53 activation inhibits the cyclin E-CDK2 and cyclin D1-CDK4 complexes, and therefore can induce G1 arrest and block entry into the S phase of the cell cycle (21). Other family members in the Cip/Kip group of Cdk inhibitors include p27^Kip1 and p57^Kip2, which share significant sequence homology but have different cyclin-CDK targets (22). In the cytoplasm, p21 remains in quaternary complexes with cyclin and Cdks (23). p21 also interacts with proliferating cell nuclear antigen (PCNA), an accessory protein of DNA polymerase, and thereby inhibits DNA synthesis (24).

The p21 gene is located at 6p21.2. Gene expression of p21 is activated by p53, growth factors including transforming growth factor β (TGF-β), cytokines, retinoids, along with many other factors, including the transcription factors Sp1/Sp3, Smads, Ap2, STAT, BRCA1, E2F-1/E2F-3, and CAAT/enhancer binding proteins a and h (25). c-Myc and c-jun both inhibit p21 expression (25). The biological actions of p21 are determined by its expression levels, affected by transcriptional factor regulation of the p21 promoter, transcript stability, and post-translational controls (25).

The central tenets of the concept of aging and activation of molecular switches involved in “danger” signals have been recently documented in cigarette smoke–induced emphysema. Emphysematous lungs exposed to cigarette smoke express the hypoxia-inducible, oxidative stress modulatory molecule RTP-801 (18). Human emphysematous lungs also have decreased telomere length, enhanced expression of p21, p16^Ink4a, p19^Arf, and the cytochemical marker of cellular senescence reliant on low pH-activated β-galactosidase activity (33). Furthermore, these results are complemented by evidence that these alterations are systemic since peripheral blood mononuclear cells share the telomere shortening seen in the lung (34) and alveolar macrophages from smokers have increased expression of p21 (35). The links among cell senescence, aging, and predisposition to cigarette smoke–induced emphysema were strengthened by the documentation that prematurely aged mice lacking the anti-aging protein senescence marker protein (SMP)-30 have increased alveolar injury and emphysema caused by cigarette smoke (36). This paradigm may not be restricted to cigarette smoke as recent reports have linked mutations in telomerase and the RNA component of telomerase (TERC) to interstitial lung disease, also shown to be associated with cigarette smoke (37). These data indicate that biological processes involved in aging, particularly oxidative stress, may hold the key to the unrelenting progression of COPD despite smoking cessation. Ceramide, which is increased in lungs of smokers and in mice exposed to cigarette smoke and mediates experimental emphysema (38), has been also shown to cause senescence (39) and promote a feedforward amplification of its pro-oxidative stress and apoptotic actions in vitro (40).

It is tempting to interpret that the pathogenic role of p21 in mice exposed to cigarette smoke may be related to its pro-senescence properties, as suggested by Yao and coworkers (1). The paradigm centered on the role of p21 in senescence reveals facets neither predicted nor investigated by the proponents of the inflammation/protease–antiprotease hypothesis. Senescent cells are more prone to produce cytokines and therefore promote inflammation in vitro (41) and in vivo in late generation mice lacking TERC (26).

Yao and coworkers expand the scope of their investigation by using inhaled LPS or fMLP. Although LPS is often used to induce acute lung injury, long-term LPS treatment can mimic the COPD phenotype (45), while short-term LPS exposure has shown to induce bronchial epithelial cell apoptosis (46). LPS-induced inflammation is different from cigarette smoke–induced inflammation in that it does not involve metalloprotease-12 activity (47). fMLP, a bacterial peptide that functions as a chemoattractant for phagocytes, induces a full-scale neutrophil response including production of reactive oxygen species, granule enzyme release, and activation of various downstream signaling pathways, including those involving phophoplipase A and D, phosphotidylinositol 3-(PI3) kinase (PI3Kinase), protein kinase C (PKC), and mitogen-activated protein kinase (MAPK). Consistent with its relevance in cigarette smoke–induced lung disease, the fMLP receptor is increased in neutrophils of patients with emphysema and in smokers (48). Both LPS and fMLP might cause alveolar destruction by enhancing neutrophil recruitment to the lung via the action of the collagen-derived peptide N-acetyl proline-glycin-proline (PGP). Degradation of the extracellular matrix generates PGP, which could therefore activate the CXCR2 receptor. This receptor is activated by ELR motif-CXC chemokines (IL-8, GRO-α, -β, and -γ, KC, MIP-1α) (49), which have been linked to recruitment of inflammatory cells to lungs exposed to cigarette smoke (50).

How does p21 regulate an otherwise classical inflammatory pathway? Among all agents of lung stress and injury triggered by cigarette smoke, oxidative stress is a leading and unifying agent of emphysema (51). Not only is cigarette smoke a major source of oxidants, but infiltrating inflammatory cells and alveolar cell apoptosis themselves generate potent free radicals (52). Furthermore, oxidative stress leads to enhanced inflammation (51), apoptosis (52), and potentially senescence via the p53/p21 pathway (41). Cigarette smoke damages DNA (53), which could activate a p53/p21 stress response as suggested by the co-expression of p53 and p21 in alveolar macrophages of smokers (35). The data of Yao and colleagues suggest that p21 enhances oxidative damage, as they demonstrate that p21 knockout mice have decreased levels of superoxide and hydroperoxides in cells retrieved by bronchoalveolar lavage or in lung tissue lysates. These findings are in agreement with data that forced in vitro expression of p21, but not the cell cycle inhibitor p16, triggers enhanced oxidative stress, which conversely up-regulates p21 expression (54).

Although it is unclear how p21 modulates oxidative stress, the authors suggest that p21-dependent kinase (PAK) links p21 with the activation of cytokines and NF-κB–dependent signaling. PAK(s) represent a new mechanism for p21 to influence a diverse array of intracellular processes (55). PAK(s) are serine/threonine kinases that control cytoskeleton dynamics and actin depolymerization. In addition to p21, PAK(s) are activated by GTPases and other signaling molecules.

As in most centrally placed signaling molecules, there is a “biological price” to pay if these molecules are blocked or overexpressed. p21 plays a central role in stem cell maintenance, particularly during situations of stress. Too much or too little p21 leads to a decreased stem cell pool or stem cell exhaustion, respectively (26). The role of p21 remains ambiguous as some reports have described either worse or improved phenotypes in p21 knockout mice. In line with this potential discrepancy, we have observed that p21 knockout newborn mice showed significant lung damage to hyperoxia, including alveolar enlargement and evidence of alveolar cell apoptosis, when compared with wild-type pups (61).