Hyperinflammation in airways of cystic fibrosis patients: what’s new?

The genetic disorder cystic fibrosis (CF) is the most common lethal monogenic disease in Caucasian populations, estimated to affect one out of 2500–4000 newborns. While CF affects all exocrine organs throughout the body resulting from mutations in the CF transmembrane conductance regulator (CFTR), severe and persistent lung complications (e.g., excessive inflammation and bacterial infection) still remain a major cause of morbidity and mortality. Approximately 70% of patients with CF are affected by the F508del mutation, and the resulting improperly folded CFTR protein cannot reach the plasma membrane of epithelial cells. In patients with CF, lack of CFTR Cl^-channel function leads to progressive pulmonary damage frequently associated with a severe and persistent neutrophil-dominated endobronchial inflammation and bacterial infection [1].

How the mutation or absence of CFTR, which is primarily expressed in cilated [2] and submucosal gland [3] epithelial cells of the respiratory tract, promotes pulmonary inflammation is still incompletely understood. The molecular mechanisms connecting abnormal CFTR function in airway epithelial cells to excessive lung neutrophilic inflammation have not been fully elucidated.

What is the early pathological event in CF airways: neutrophilic inflammation or bacterial infection?

CF disease results from a failure of pulmonary innate immune functions such as loss of airway antibacterial activity, reduced mucociliary clearance and failure of immune mechanisms to successfully opsonize and kill bacteria [4]. Neutrophil-predominant airway inflammation in CF begins early during the neonatal period, increases throughout childhood and adolescence and is associated with persistent airway infection. However, whether it is a cause or a consequence of infection remains conjectural. Abnormal CFTR functional activity results in airway surface dehydration and a modification of the properties of mucus clearance that participate in the susceptibility to chronic infection with pathogens such as Staphylococcus aureus and Pseudomonas aeruginosa[4–6]. The resistance to chronic P. aeruginosa lung infection requires CFTR-modulated IL-1β release and signaling through the IL-1 receptor, a process that is deficient in cells lacking functional CFTR [7].

There are now a number of endogenous factors that have been recently identified as important modifiers in the susceptibility of CF airway cells to P. aeruginosa. Elevated levels of furin in CF airway cells cause hypersusceptibility to P. aeruginosa exotoxin A-induced cytotoxicity [8]. Ceramide accumulation in CF airway cells mediates bacterial infection, inflammation and cell death in the respiratory tract of cftr-deficient mice [9].

In the lungs of CF patients, the inflammatory response to a defined bacterial load appears to be greater and more excessive than in normal lung [4,10]. This lung inflammation is characterized by a sustained accumulation of neutrophils, high proteolytic activity and elevated levels of chemokines, such as IL-8 [11,12]. However, despite intensive investigation, the relative importance of these multiple factors to the clinical manifestations of CF remains uncertain.

The sequence of events at the onset of airway inflammation has been the subject of debate following the finding of neutrophil-dominated inflammation in the absence of bacterial or viral pathogens in bronchial lavages obtained from CF infants [13,14]. The proof that CF lung inflammation can occur prior to any infection came from naive human CF lung grafts developed in severe combined immunodeficiency mice [15]. In this ex vivo model, we had demonstrated that sterile human CF lungs secreted high levels of IL-8 prior to any infection, and host (mouse) neutrophils progressively accumulated in CF lungs leading to tissue destruction. Consequently, these observations have led to the proposal that increased neutrophil-dominated inflammation, or alternatively the failure to downregulate inflammation, is intrinsic to CF and operates independently of an infectious stimulus.

Why is it important to identify molecular factor(s) associated with the absence of CFTR activity?

Key studies have shown that neutrophilic inflammation occurs early in the course of CF lung disease and that neutrophil-derived proteases, most probably elastase, play a crucial pathological role. The exact mechanism of neutrophilic inflammation in airways of CF patients remains unclear. The airway epithelium is an important source of chemokine IL-8 and is the principal neutrophil chemoattractant in the CF lung [16]. During the past decade, cell culture systems using primary human bronchial epithelial cells and respiratory cell lines have revealed that CF cells exhibit enhanced proinflammatory signaling compared with non-CF cells [17–23]. This intrinsic inflammation was mainly characterized by an increased level of inflammatory mediators (e.g., IL-8 and IL-6) compared with non-CF bronchial epithelial cells, either in the absence [19,21,24] or presence of bacterial stimulation [25–28]. This enhanced production of proinflammatory cytokines released at the airway epithelium surface might also facilitate bacterial proliferation [29] and could increase their adhesion in the presence of IL-6 and IL-8, which contribute to an enhanced level of sialyl-Lewis x and 6-sulfo-sialyl-Lewis x epitopes in human airway mucins of CF patients [30].

In airways of CF patients, the origin(s) of the proinflammatory state support(s) the concept of enhanced NF-κB signaling in epithelial cells that leads to enhanced IL-8 production [25,31–36]. Interestingly, some studies found that CF airways are relatively deficient in the counter-regulatory molecules IL-10 and NO, both of which preserve the function of IκBa, the major inhibitor of NF-κB. In situations in which a lack of IL-10 or NO occurred, an imbalance between levels of IκBα and NF-κB activity in airway epithelia would result in the prolonged and excessive production of the mediators responsible for the damaging inflammation [37,38]. Interestingly, we had shown that the de novo expression of IκBα protein could be obtained in CF human bronchial epithelial cells following treatment with the flavonoid genistein [25]. Perez and colleagues have also observed that cultured primary normal human bronchial epithelial cells have significantly increased basal secretion of proinflammatory cytokines, nuclear NF-κB translocation and stimulated secretion when these cells are first treated with a new class of specific inhibitor of CFTR chloride conductance, CFTR_inh-172[39]. These data support the hypothesis that lack of CFTR activity in airway epithelial cells is responsible for the onset of the inflammatory cascade in the CF lung.

Despite the fact that the link between NF-κB signaling and inflammatory gene activation is now (at least partly) established, the proteins bound to F508del CFTR in CF airway epithelial cells leading to an elevated susceptibility to NF-κB activation remain to be elucidated.

Other pathways could be involved in promoting the inflammatory response in the CF airway. The deficiency in the activity of PPAR-γ has been observed, therefore contributing to the imbalance between IκBα and NF-κB in CF airway cells [40]. Signaling depending on MAPK/ERK and AP-1 has also been described to be hyperactivated [41], particularly when CF airway epithelial cells are exposed to oxidative stress [42–44].

Neutrophils in CF airways: are they nonapoptotic, apoptotic and/or necrotic cells?

Neutrophils killed by P. aeruginosa present in airways release high amounts of proteases that disable any still viable neighboring neutrophils [45]. Recently, it has been shown that in the airways of CF patients (and possibly patients with other chronic lung diseases), IL-8 continuously recruits neutrophils to the lung, where the CXCR-1 is immediately cleaved from their cell surface by free elastase, thereby disarming neutrophils of their antibacterial capacity. This cleavage step releases soluble CXCR1, which stimulates IL-8 production by airway epithelial cells, thereby initiating a feedback circuit that amplifies neutrophil recruitment within the CF airways [46]. Extracellular neutrophil elastase can also [47]:

• • Induce mucin overproduction by airway glands: potentiating obstruction;

• • Cleave phagocytic receptors on macrophages: potentiating infection;

• • Cleave scavenger receptors on macrophages and epithelial cells: inhibiting clearance of apoptotic neutrophils;

• • Induce IL-8 secretion by epithelial cells and neutrophils: potentiating inflammation.

In CF airways, the vicious cycle of inflammation and infection makes it very difficult to assess the dynamic sequence of events that cause cell death and clearance of neutrophils, leaving a number of questions unanswered.

Which cells cause more damage to the airway epithelium: activated viable or necrotic neutrophils?

Neutrophils are attracted to CF lungs very early on and in high numbers due to pre-existing unknown stress conditions (e.g., upregulated cytokine secretion, pH, osmolarity and so on). Once in the lungs, the majority of neutrophils are believed to undergo rapid apoptosis and necrosis; and large quantities of DNA, actin, granule-derived myeloperoxidase and proteases spill out into the lung lumen, which increases redox stress, mucus viscosity and induce airway damage [48].

The circulating neutrophils from CF patients display a profound modification of gene expression compared with healthy donors, but there was a very limited difference between neutrophils isolated from CF blood and airways [49]. In the blood of CF patients, the level of apoptosis of neutrophils appears similar to the level of neutrophil apoptosis of healthy subjects, but the interaction of CF airway neutrophils with CF airway epithelial cells reduces their level of apoptosis compared with non-CF neutrophil–airway epithelial cell interaction [50]. Interestingly, Tirouvanziam and colleagues recently established that CF airways contain a significant fraction of nonapoptotic viable neutrophils [51]. They demonstrated that profound functional and signaling changes readily occur within viable neutrophils when recruited to CF airways. Compared with their blood counterparts, airway neutrophils have undergone conventional activation, as shown by decreased intracellular glutathione, increased lipid raft assembly, surface mobilization of CD11b⁺ and CD66b⁺ granules, and increased levels of the cytoskeleton-associated phospho-Syk kinase. Unexpectedly, they also mobilize CD63⁺ elastase-rich granules to the surface (usually confined intracellularly) and lose surface expression of CD16 and CD14 (both key receptors in phagocytosis). Furthermore, they express CD80, MHC-II and the prostaglandin D2 receptor CD294, which are all normally associated with other lineages, reflecting functional reprogramming. This notion is reinforced by their decreased total phosphotyrosine levels, mirroring a postactivated stage and their increased levels of the phospho-S6 ribosomal protein. Thus, these authors have identified a subset of neutrophils within CF airways with a viable, but dysfunctional, phenotype. This subset provides a possible therapeutic target and indicates a need to revisit current paradigms of CF airway disease.

Conclusion & future outlook

In the past decade, important advances have been made in the understanding of the mechanisms underlying lung disease in CF. It is currently believed that inflammatory signaling in airway epithelium plays a critical role in orchestrating the response of the lungs to a broad variety of insults in the context of CF. In addition, the NF-κB pathway in respiratory epithelial cells appears to be a focal point for the control of lung inflammation through regulated production of mediators that participate in recruitment and activation of neutrophils, modulation of apoptosis and control of epithelial barrier integrity. Consequently, dysregulated NF-κB activation in the epithelium may provide a common pathway for driving the excessive inflammatory response in CF. New emerging challenges for consideration include whether the airway epithelium may be an important and feasible target to limit lung inflammation in CF and how reducing NF-κB activation may provide a therapeutic solution.

The challenge for researchers over the next few years will be to determine CF lung biomarker(s) of inflammation that are critical in the drug development process for tracking rapid disease progression and capturing response to treatment.

Yet, many questions remain unanswered. How do live airway neutrophils release toxic effectors and interact with other components of the lung (i.e., epithelial cells, macrophages, bacteria and other neutrophils) and how can we regulate pathological alterations to these processes? These are some of the most pressing issues that need to be clarified in the near future in the context of CF lung pathology. Considering live CF lung neutrophils not only as a transient fraction designed for rapid death but also as potential contributors to the disease, with a lifespan of several days, should considerably change the focus of research on neutrophils homing into the airways of CF patients.

Therefore, we consider that CF is not only an epithelial disease, but also a neutrophil disease. A comprehensive approach that will include a better understanding of the function of CFTR protein as a regulator of the airway inflammation and the knowledge of the fraction of viable nonapoptotic neutrophils homing into the airways of patients with CF disease will provide useful information for a better treatment of lung inflammation of patients with CF.

Financial & competing interests disclosure

We acknowledge research support from the French Cystic Fibrosis Association (Vaincre la Mucoviscidose, Paris, France), the Inserm and Université Pierre et Marie Curie Paris 6.

The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript.

No writing assistance was utilized in the production of this manuscript.