|Home | About | Journals | Submit | Contact Us | Français|
The neutrophil is a powerful cellular defender of the vulnerable interface between the environment and pulmonary tissues. This cell’s potent weapons are carefully calibrated in the healthy state to maximize effectiveness in fighting pathogens while minimizing tissue damage and allowing for repair of what damage does occur. The three related chronic airway disorders of cystic fibrosis, non–cystic fibrosis bronchiectasis, and alpha-1 antitrypsin deficiency all demonstrate significant derangements of this homeostatic system that result in their respective pathologies. An important shared feature among them is the inefficient resolution of chronic inflammation that serves as a central means for neutrophil-driven lung damage resulting in disease progression. Examining the commonalities and divergences between these diseases in the light of their immunopathology is informative and may help guide us toward future therapeutics designed to modulate the neutrophil’s interplay with the pulmonary environment.
The human pulmonary microenvironment serves as an important location for the interactions of external antigens and elements constituting the host response, and is a challenging environment in which to coordinate host defense. A large volume (approximately 15,000 L) of air replete with noxious particles and organisms is inhaled daily and exposed to the enormous alveolar surface area of the lung (approximately 160 m2), which is juxtaposed to a rich capillary network of equal size (1). As such, a multitude of innate immune mechanisms work via distinct pathways to reinforce lung homeostasis and protect against injury and infection. These mechanisms include, but are not limited to, pattern recognition receptors (Toll-like receptors, dectin-1, and CD14), surfactant proteins A and D, transforming growth factor-β activation, chemokine production, and extracellular matrix destruction (2). These modalities collectively serve to induce local inflammation when injury or invasion occurs so that defensive and reparative processes can be initiated.
During both acute and nonresolving chronic inflammatory responses, polymorphonuclear cells (PMNs), or neutrophils, are recruited via specific signals to sites of injury. PMNs are the most numerous leukocytes in adult peripheral blood, and represent more than 95% of the granulocytes, which also include eosinophils and basophils. They are produced from stem cells in the bone marrow and mature over a period of about 2 weeks (3).
PMNs exhibit a variety of biologic functions critical for maintenance of airway homeostasis. Two of the most classic features of PMNs are their ability to move by chemotaxis to sites of inflammation and their ability to kill invading microorganisms. Circulating PMNs can be directed to sites of inflammation by migrating over chemotactic gradients. Over the past 50 years, a host of ligands and PMN receptors have been found to be capable of inducing PMN chemotaxis. The interaction of these ligands and receptors induces intracellular actin polymerization/depolymerization, causing the neutrophil to form a “leading edge” toward the chemotactic molecule (4), which culminates in locomotion toward the stimulus. The circulating PMN must also traverse the vascular endothelium to enter the extravascular site of inflammation; this is in part regulated by the interplay of L-selectin on PMNs and P- or E-selectins on inflamed endothelia (5). Once in the interstitium, the neutrophil extrudes proteases to break apart extracellular matrix in fleeting bursts termed “quantum proteolysis” facilitating the cell’s advancement toward the site of interest (6).
When a pathogen has been reached, the ability to effectively kill the invading microorganism is a cardinal feature of the PMN. The most direct mechanism is through phagocytosis of the invading organism, which can be triggered by Fc receptors, complement, or pattern recognition receptors. This leads to the development of the phagolysosome and the deployment of toxic products from intracellular granules designed to effectively kill hostile microorganisms within the cell (6). Another mechanism by which PMNs destroy invading organisms is through the release into the extracellular environment of PMN granules (degranulation) such as defensins, lactoferrin, myeloperoxidase, and proteases (2, 7).
Although much is already known about these mechanisms of neutrophil function, our understanding of neutrophil biology continues to advance. In 2004, an entirely new mechanism by which neutrophils kill was described by Mayer-Scholl and colleagues and Brinkmann and colleagues (8, 9). It was shown that PMNs can actively extrude DNA, leading to the development of neutrophil extracellular traps (NETs) in a process that leads to programmed cell death but is independent of apoptosis. This network of extracellular DNA is coupled to a variety of neutrophil-derived antimicrobial proteins (such as proteases and histones) at high concentrations, which actively kill microorganisms independent of degranulation or phagocytosis (10).
These functions and perhaps other as yet unappreciated mechanisms have evolved over time to allow PMNs to carefully control the extent of the local immune response, with both proinflammatory and antiinflammatory mechanisms actively at work in the neutrophil-inflamed lung. When the complex neutrophil regulatory systems fail, these powerful cells can mediate an overexuberant inflammatory response, leading to a chronic immune response resulting in damage to both cellular and extracellular components of the lung. When extracellular matrix components and the cells that maintain them are degraded too rapidly (or if damaging processes are sustained for too long), the natural reparative mechanisms that maintain normal pulmonary structures are overwhelmed and pulmonary structures become pathologically altered. This is the essence of the current understanding of pathological remodeling such as occurs in emphysema (characterized by damage to alveolar structures) and bronchiectasis (characterized by damage to airways). Although other cells play important roles in this cycle of damage, the dysregulation of neutrophilic inflammation is central to the pathology of cystic fibrosis (CF), non-CF bronchiectasis (NCFBE), and alpha-1 antitrypsin deficiency (A1ATD). Moreover, these conditions share a paradoxical impairment of neutrophil effectiveness, feeding into a vicious cycle of infection and/or inflammation that leads to ongoing lung damage (see Figure 1).
CF is the most common inherited genetic disorder in the white population worldwide. It is due to loss-of-function mutations in the cystic fibrosis transmembrane conductance regulator gene (CFTR), disrupting normal epithelial cell function in multiple organs. The most prominent manifestations are observed in the lungs, where the loss of mucociliary clearance in the airway leads to mucostasis, infection, and ongoing PMN-predominant inflammation (2). Although there is relatively little evidence that PMN recruitment into the lung is impaired, PMNs in the CF lung display a myriad of functional alterations that likely contribute to the progression of lung disease. Importantly, there is emerging evidence that PMNs recruited to the lung are actively “tuned” by the lung microenvironment.
A critical feature of the CF airway is the persistence of bacteria colonizing the airways (both planktonic and within biofilms) as a downstream feature of loss of cystic fibrosis transmembrane conductance regulator protein (CFTR). Airway PMNs in CF have been shown to produce reduced levels of respiratory burst (10). Indeed, one mechanism by which PMN bacterial killing is attenuated is through cleavage of CXCR1 by degranulated neutrophil proteases including neutrophil elastase (NE), leading to reduced PMN oxidative burst (11). Similarly, PMNs can undergo proteolytic loss of CD14 and CD16, receptors important for bacteria phagocytosis (12, 13), and these changes are observed in the CF airway because of local neutrophil protease activity (14). Among these proteases are matrix metalloproteinase (MMP)-8 and MMP-9, which have been shown to be elevated in the CF airway, especially during exacerbations (15). These enzymes, in conjunction with the serine protease prolyl endopeptidase, are capable of inducing further local neutrophil recruitment via the production from collagen of small peptide breakdown products (termed “matrikines”), such as the bioactive tripeptide proline-glycine-proline, which has chemotactic and inflammatory activity toward neutrophils (16). As a result of these emerging data a clinical trial has been performed, examining doxycycline as an antiprotease with potential to attenuate this neutrophilic positive feedback loop in acute exacerbations of cystic fibrosis (17). Overall, these features lead to a neutrophil that is actively recruited to the CF airway from the vascular space but then undergoes local changes that lead to reduced capacity for phagocytosis (14) and bacterial killing, contributing to ongoing bacterial colonization and thus persistent inflammation.
PMNs that transmigrate to the CF airway from the circulation undergo local priming from inflammatory mediators that stimulate degranulation. The CF airway PMN, in particular, displays signs of increased granule mobilization in the CF airway, with increased surface markers for CD66b and CD11b, indicating increased release of secondary and tertiary granules. There is also increased CD63 cellular expression, highlighting the release of primary granules (14). The primary granule contains both NE and myeloperoxidase, and its poorly regulated release is particularly capable of “bystander” damage to the CF airway. As PMNs are longer lived in the CF airspace (18), it is entirely possible that these PMNs endure “waves” of degranulation before undergoing apoptosis. Evidence suggests that the intrinsic loss of CFTR function in the PMN may contribute to these changes in degranulation, as the use of the CFTR corrector ivacaftor has shown improvement in these effects (19, 20).
NETs are a relatively understudied area of research in CF lung disease, although they are considered to be a critical fate of the PMN in CF airways. NETs have been found in CF airway secretions (21) and play an important role in ongoing airway inflammation (22). NETs can be stimulated for release by Pseudomonas aeruginosa (22) and, more specifically, by pyocyanin derived from the bacteria (23); in both cases, NET release is NADPH oxidase dependent. In addition to these bacterial products, host mediators also display the ability to induce NET formation (24, 25). Although NET release in the CF airway may be considered beneficial, this mechanism does not seem to effectively eradicate certain bacteria, including Pseudomonas aeruginosa (26). Although the reason for this is unknown, one theory is that the antibacterial effects of NETs may not work equally well on all bacterial species. This may lead to the development of microcolonization, which could encourage the development of a dominant species in the airway (27).
Bronchiectasis is defined by dilation of the cartilage-containing airways exceeding the diameter of the corresponding pulmonary arterial diameter on high-resolution computed tomography (28). In the modern era, there has been increasing recognition of bronchiectasis caused by a host of often-overlapping etiologies (29). Mendelian and non-Mendelian conditions including diseases that impair host defense or mucociliary clearance, or that result in chronic inflammation, all result in bronchiectasis in susceptible subjects by setting in motion a chronic inflammatory or bacterial state that hones immunity to the lung (28, 30, 31). Because of the diversity of causes of bronchiectasis and for the purpose of clinical investigation, causes of bronchiectasis not due to CF are typically classified as non-CF bronchiectasis (NCFBE). Similar to CF, the neutrophil plays an important role in the pathology of NCFBE.
Despite a spectrum of etiologies in NCFBE, key pathological insults appear to be responsible for the development of this syndrome (31). Similar to the acute response to infections, cellular-mediated immune responses, primarily mediated by PMN influx, are stimulated on an ongoing basis in bronchiectasis (32–34). PMNs are the predominant cellular component of sputum and bronchoalveolar lavage in patients with NCFBE (35, 36). The noncellular sputum component of airway secretions in patients with NCFBE is enriched in potent neutrophilic chemoattractant cytokines, including IL-1β, tumor necrosis factor-α, IL-8, and leukotriene B4 produced by the stimulatory effects of hyperactive alveolar macrophages (37–39). This persistent cytokine milieu results in chronic neutrophilic recruitment and migration to the alveoli and distal airways in NCFBE, as in CF. This feature also partially accounts for the common clinical findings of daily, mucopurulent sputum in patients with NCFBE (40).
Chronic neutrophil migration causes a recurrent cycle of innate (mostly neutrophilic) and adaptive immune responses that result in progression of disease. Notably, chronic airway inflammation and/or infection appear to cause a persistent imbalance of repair and damage leading to destruction that manifests as dilated airways, further predisposing the system to ongoing infection and inflammation. This “vicious cycle” thus creates the opportunity for imbalance of inflammation due to chronic airway inflammation, a concept described by Cole in bronchiectasis (33). Clinically, key mediators of airway damage are found at elevated levels in NCFBE, including neutrophil elastase and myeloperoxidase, compared with control subjects (41). This response was further heightened in subjects with chronic Pseudomonas infection (42). Thus, there is abundant evidence that the neutrophil is the key cell mediator in the “vicious cycle” that is thought to drive the development and progression of bronchiectasis.
Additional insights into the pathways of PMN recruitment and killing provide evidence for a defective yet hyperactive response to stimuli in NCFBE. The exuberant and continued recruitment of PMNs to the airways is driven by persistently high concentrations of neutrophil chemoattractants in the airway secretions and in the bloodstream. Elevated levels of key endothelial adhesion molecules have been detected in NCFBE as well as CF (42). The character of this response does appear to be distinct from that of CF. Neutrophil surface expression of CD11b/CD18 is normal in NCFBE in contrast to CF, in which it is up-regulated (30). Also, L-selectin shedding is decreased in stimulated CF neutrophils whereas it is normal in NCFBE (42). Finally, elevations of key proteases including MMP-8 and MMP-9 have been found in exhaled breath condensates of children with NCFBE (43), which may contribute to perpetuation of local neutrophil recruitment via matrikine release, as occurs in CF (15). However, little is known about the relative activity of this key pathway in NCFBE. Together, this evidence suggests that, in general, PMN migration and recruitment are normal in NCFBE but hyperstimulated due to persistently elevated levels of chemoattractants for migration of these cells.
Despite normal recruitment, there is evidence for abnormal neutrophil activation and function in NCFBE, in parallel to CF. However, whether neutrophil fates are deranged in NCFBE remains under investigation. Bacterial killing via phagocytosis is indeed impaired in NCFBE. Much of the current literature focuses on the role of NE, due to unadulterated degranulation of granules containing proteolytic enzymes from airway PMNs (44). The effect of elevated NE in the airway is impaired phagocytosis via cleavage of key complement receptors of opsonized bacteria (45, 46) and by causing opsonin/receptor imparity (47). Innate reduction of phagocytic receptors on the neutrophil also likely impairs this process in NCFBE (46). In addition, unchecked neutrophilic degranulation ,among other mechanisms, may directly impair the neutrophil oxidative burst in some patients, leading to defective killing after phagocytosis (48). Despite its deleterious effects, NE is crucial for initiating host defense to key bacterial infections including Pseudomonas aeruginosa (49). Thus an imbalance of NE levels and/or function, among other key neutrophilic by-products, including human neutrophil peptides (46), likely contributes to defective killing in the setting of normal and/or enhanced neutrophil recruitment to the site of infection.
In addition to the primary role of neutrophil phagocytosis, the formation of coordinated neutrophil networks or NETs has been increasingly recognized as a key function of recruited neutrophils (22). NET formation and function are clearly impaired in CF bronchiectasis (26). In addition, aged patients acquire defects in NET formation, but implications of deranged NET formation and function have not been investigated in NCFBE (50).
A1ATD is a rare lung disease that often presents with early-onset emphysema, although presentation can be quite variable (51). Loss of the antiprotease alpha-1 antitrypsin (alpha-1 AT) leads to increased serine protease activity, causing progressive lung parenchymal damage. Although it is tempting to consider A1ATD as a pure protease/antiprotease imbalance leading to local damage and pulmonary pathology, it is now recognized that the loss of alpha-1 AT impacts immune function both directly and through complex indirect mechanisms in the damaged lung, and that this is a critical component of the disease process (52). With this more nuanced understanding of A1ATD it becomes clear that this disease is, at heart, a disorder of innate immunity centered on the neutrophil much like CF and NCFBE. It is worth noting that a subset of A1ATD exhibits bronchiectasis and therefore could be considered a subcategory of NCFBE (which as mentioned earlier is a heterogeneous clinical phenotype caused by varied pathophysiological mechanisms). This incomplete overlap in presentation between A1ATD and the bronchiectatic disorders is especially interesting in light of the considerable homology between neutrophilic derangements seen in these conditions, supporting the importance of this cell type in the pathology seen in these diseases. Conversely, the fact that the majority of patients with A1ATD do not exhibit prominent bronchiectasis (52) belies the point that the interplay between clinical phenotype and a given patient’s pathophysiology is complex and incompletely understood.
Numerous mechanisms increase the burden of neutrophils in the lungs of individuals with A1ATD, both in the interstitial space and in the airways (53–56). Increased pulmonary NE activity in this disease acts through numerous pathways to encourage neutrophils to traffic into the lungs (55, 57, 58). NE is capable of acting directly on airway epithelial cells to increase IL-8 production (59) and especially on pulmonary macrophages to release leukotriene B4 (53), two powerful neutrophil chemokines. The sensitivity of these two pulmonary cell types to the action of NE perhaps explains in part the particular vulnerability of the lungs to this systemic disorder. As NE is itself produced by neutrophils, this establishes a local feed-forward process in the A1ATD airway. Furthermore, the presence of NE has been shown to amplify the response of airway epithelial cells to the oxidants present in cigarette smoke extract (60). The result is increased release of IL-8 induced by cigarette smoke. This further sensitizes the A1ATD smoker to the neutrophilic inflammation caused by tobacco smoke and perhaps other noxious stimuli. A1ATD also increases the release of IL-8 by the neutrophil itself (61); this IL-8 release creates another positive feedback loop in the deficient state that favors ongoing recruitment of neutrophils to sites of established neutrophilic inflammation. The increased presence of neutrophils can accelerate itself even more as the latter destroy proteins of the interstitial matrix, potentially leading to the release of neutrophil-stimulating matrikines (62, 63). Yet another potential mechanism contributing to the pulmonary neutrophilic burden lies in alpha-1 AT polymers, which tend to aggregate in the pulmonary interstitium (64). These polymers are now known to potently attract PMNs (65) and seem likely to augment pulmonary neutrophilia, although the biologic relevance of this observation is not yet certain.
The behavior of neutrophils in A1ATD also appears to be abnormal, especially chemotaxis. Chemotaxis toward endotoxin is diminished in infants and children with the genetic disorder, through an as yet unknown mechanism (66). However, as mentioned previously, pulmonary chemotactic stimuli are increased in A1ATD, and the neutrophils of deficient individuals are actually hyperresponsive to at least some chemotactic stimuli. This relates to a decrease in the suppressive role played by the alpha-1 AT protein on the response of neutrophils to IL-8 and soluble immune complexes (67), as well as perhaps to cathepsin G (68), another neutrophil-derived serine protease that is inhibited by alpha-1 AT and encourages neutrophil chemotaxis. In addition, A1ATD encourages neutrophil endothelial adhesion through an unknown protease-independent process (69).
Furthermore, there is emerging evidence that alpha-1 AT itself may have a multitude of immunomodulatory properties (summarized in Reference 70), and its loss may impact other aspects of PMN biology through tumor necrosis factor-α–dependent mechanisms including changes in PMN apoptosis (71) and degranulation (72). Interestingly, the serine protease site of the enzyme does not mediate these effects (69, 71), which underscores the pleiotropic nature of this protein’s neutrophil-modulating capacity. The net effect of these changes in A1ATD appears to be a derangement in neutrophil behavior that leads to an increase in tissue neutrophilia most pronounced in areas of the body, such as the lung, where environmental stimuli (and perhaps pathological stimuli such as alveolar alpha-1 AT polymer deposits) trigger sustained neutrophil recruitment.
However, despite these increases in both neutrophil density in pulmonary tissue and neutrophil responsiveness to certain chemotactic stimuli, in close analogy to the aforementioned bronchiectatic disorders it appears that the neutrophils in A1ATD may be less effective in their primary duty, killing pathogens. In a manner reminiscent of CF, the proteolytic derangements induced by A1ATD (especially increased airway NE levels) function to impair the ability of the neutrophil to efficiently kill bacteria. One major mechanism for this phenomenon is thought to be NE-mediated inactivation of CXCR1, as previously mentioned for CF. Indeed, decreased neutrophil surface expression of CXCR1 impairs the respiratory burst and has been shown to correlate with reduced PMN bactericidal capacity in patients with CF, chronic obstructive pulmonary disease, and bronchiectasis (10). This effect correlates with airway levels of NE. The cleavage products of CXCR1 can also induce IL-8 release through Toll-like receptor 2 signaling (73), acting as yet another mechanism to increase neutrophil density in the tissues, leading downstream to the release of still more NE. Proteolytic loss of CD14 and CD16 may also contribute to impaired bactericidal capacity of airway immune cells in A1ATD. Excessive airspace NE also encourages bacterial virulence through local tissue destruction and resultant disruption of airway epithelial integrity, degradation of endogenous antimicrobial peptides, and possibly enhanced bacterial invasion of epithelial cells (74). Supporting the relevance of these findings in vivo, pulmonary expression of transgenic human alpha-1 AT in a mouse model of Pseudomonas aeruginosa pneumonia dramatically reduced mortality (75). Effects of the alpha-1 AT protein beyond NE inhibition also appear to play a role. Alpha-1 AT appears to act as an inhibitor of matriptase, an activator of epithelial sodium channels, which function to resorb sodium from mucus (76). This would be expected to cause decreased airway surface liquid depth and impaired mucociliary clearance in A1ATD, setting the stage for chronic infection and bronchiectasis, although this mechanism has not yet been studied in vivo. In the future, assays of neutrophil bactericidal activity and microbiological studies may confirm these in vitro observations that suggest a vulnerability to infection among individuals with A1ATD. Importantly, such impaired bacterial killing might contribute to lung damage by encouraging exacerbations and by allowing bacterial presence to serve as a persistent stimulus for ongoing local inflammation, as it does in the other disorders discussed in this article.
Viewed collectively, the evidence suggests that neutrophils in patients with chronic lung disease are predisposed to traffic to the interface with pathogens or irritants (i.e., the airway), but paradoxically function to impair one another’s ability to terminate the inflammatory stimulus. This sets the stage for unchecked release of less effective but locally destructive mediators of the neutrophilic inflammatory response, leading to progressive lung damage. In some circumstances the PMN may display intrinsic defects, but for the most part the neutrophil acquires an altered phenotype on recruitment to the airway from the circulation. Understanding the role of the PMN in response to the signals generated in the lung will provide a more dynamic appreciation for these PMN fates during the course of disease. Harnessing the dynamics of this complex interaction can potentially allow for more effective therapies for these chronic lung disorders.
Supported by the NIH (HL102371 and HL126596 to A.G.), the Veterans Administration (1 I01 BX001756 to A.G.), and the Cystic Fibrosis Foundation (CLANCY09Y0 and SORSCH15RO to G.M.S.).
Author Contributions: D.W.R., A.G., and G.M.S. all contributed to manuscript preparation and final editing. All authors approved the final manuscript draft before submission.