Lymphocytes in cystic fibrosis lung disease: a tale of two immunities

After over half a century of research a clear, comprehensive and correct understanding of the pathogenesis of the chronic bronchocentric pulmonary infection and inflammation that characterizes cystic fibrosis (CF) lung disease continues to elude us. The earliest hypothesis, postulating the critical role of impaired mucociliary clearance by dessicated, hyperviscous airway fluids, is alive and well, having been updated by intensive study of the ion transport regulatory properties of the cystic fibrosis transmembrane conductance regulator (CFTR). Mutations in the CFTR gene on both alleles result in defective chloride secretion and sodium hyperabsorption in the conducting airways with resulting volume depletion of the airway surface liquid [1,2]. However, in the case of CF the application of Occam's razor, seeking a parsimonious cause-and-effect relationship between CFTR's well-described chloride channel function and all the downstream pathogenetic effects, has been undermined completely by the discovery of the astounding multi-functionality of this protein, the pleotropic effects of its dysfunction on a large number of other genes and cell functions, and the contributory effects of modifying genes in this supposedly monogenic disease [3–5]. To cite just one example among several that have potentially important clinical implications for immune responses, CFTR directly influences expression of the regulated upon activation, normal T cell expressed, and presumably secreted (RANTES) chemokine [6]. Thus over the last decade the net has again widened to examine how CFTR mutations may result in a variety of host defence defects leading to the disease we observe as clinicians. The task is complicated by the onset of lung disease in early infancy, making primary CFTR-related defects difficult to distinguish from secondary effects [7]. Furthermore, in many cases (for example, in vitro studies of cytokine and chemokine secretion by CF respiratory epithelial cells at baseline and upon activation) understanding is also obscured by the substantial impact of methodological variables such as cell source, culture conditions, stimulus and kinetics [8,9]. All this counsels a healthy dose of scepticism and great caution in interpreting any particular set of results without much confirmatory work.

For several decades the problem of chronic airways infection with a restricted spectrum of bacterial pathogens, especially Pseudomonas aeruginosa, led to a large effort to find underlying defects in the adaptive immune response in CF. The lack of systemic infection in CF suggests strongly that any such defects are local, and thus likely as secondary effects of established infection and/or inflammation. Indeed, a large literature exists that demonstrates myriad immunomodulatory effects arising from established CF lung disease, such as proteolytic degradation of opsonins, effects of chronic antigenic stimulation upon antibody responses and mutual bacterial–host interactive effects [10–12]. Direct effects of P. aeruginosa exoproducts upon lymphocyte function probably account for early reports of suppressed cell-mediated immunity in CF patients [13,14], but overall responses probably represent a complex net effect of both suppressive and demonstrably stimulatory effects of P. aeruginosa or other microbes, as well as the accompanying proinflammatory milieu [15].

In the last several years much attention has turned to the innate immune system of the respiratory tract, as this can link primary changes in CF epithelia to first-line host defences and ultimately to changes in adaptive immunity seen in CF patients [1,16]. While salt-sensitive defensin deficiency is less likely to play the key role suggested several years ago [2,3,17], multiple other changes in innate defences have now been described − including defects in interleukin-10 production and nitric oxide synthase 2 expression and increased hypophyosphorylated nuclear factor-κB expression − that may explain the susceptibility to chronic inflammation, vulnerability to serious early viral respiratory infection and altered adaptive immune responses [18–23]. Of particular interest in this regard, these altered pathways may represent downstream effects of a CF-associated down-regulation of signal transducer and activator of transcription-1 (STAT-1) activity [24]. Working out the relative roles of altered innate immune pathways and mechanisms in CF respiratory epithelium should keep investigators busy for some time to come.

In this issue of Clinical and Experimental Immunology, Hubeau et al. describe altered cytokine production by CD3⁺ T cells in whole blood obtained from young, clinically stable CF patients [25]. Altered cytokine production has been described previously in CF-derived ciliated respiratory epithelial cells and submucosal glands, peripheral blood mononuclear cells and monocyte-derived macrophages, neutrophils and lymphocytes [8,9,18,26–31]. Hubeau et al.'s data are of interest in that eight of 10 patients were free of P. aeruginosa infection, seven of 10 had normal pulmonary function (forced expiratory flow in 1 s >80% predicted), none were on antibiotic or anti-inflammatory therapy, and all were in stable condition both clinically and in terms of blood inflammatory surrogates such as white blood count and C-reactive protein levels. The results showing higher IL-2 and lower IL-8 production in the CF T cells compared to controls differs from previous studies finding lower IL-10 and IFN-γ levels [27,29,30] but, as pointed out by the authors, both the clinical patient sample and the experimental conditions varied considerably from the earlier studies, with Hubeau et al.'s small sample representing younger, healthier, less P. aeruginosa-infected patients. What is perhaps more salient is the additional evidence of altered cytokine regulation in the circulating cells of the adaptive immune system in a variety of clinical settings and experimental methodologies. It remains unclear whether the altered cytokine profile seen in CF lymphocytes represents a direct effect of CFTR (which is expressed at levels in lymphocytes that are low but comparable to those in ciliated respiratory epithelial cells) [27,32] or another secondary disease effect. Results in epithelial cells suggests that the former is a realistic possibility. Preliminary evidence from pharmacological blockade of T cell chloride flux suggests that CFTR-dependent channel function may be involved [33].

Are the altered cytokine patterns seen in CF lymphocytes a clinically relevant problem or merely an interesting epiphenomenom? We do not know, but a good starting-place to look for the answer is in the pathology of CF lung disease. While most people interested in CF think of it quite reasonably as neutrophil-mediated disease, based upon the tremendous influx of neutrophils into the airway lumen and the deleterious effects of numerous neutrophil exoproducts and cellular constituents on airway integrity, what is usually overlooked is the fact that at the tissue level, in the airway wall, CF is lymphocytic disease [34–37]. In a model of human fetal airway xenografts placed into severe combined immunodeficient mice, Tirouvanzian et al. have shown that CF but not normal grafts undergo consistent accumulation of host-derived Ly5⁺ leucocytes in the subepithelial mucosa after apical bacterial stimulation in the proximal airways but spontaneously in the distal airways [38,39]. A previous study by Hubeau et al. in human fetal tissue showed late gestational accumulation of macrophages and mast cells in CF but not control airways mucosa [40]. These findings suggest that the CF genotype produces a primary alteration of leucocyte trafficking to the airways that may be in part or wholly responsible for the earliest stages of inflammatory lung disease. Functional studies of submucosal non-resident inflammatory cells in both central and distal airways mucosa of young CF patients would be necessary to elucidate this possibility further, but obtaining such tissue is extremely difficult. Perhaps CF murine models can provide further useful information about this, as a variety of immunoregulatory abnormalities extending beyond the epithelial cell have been documented in such animals [41,42].

The signals responsible for this abnormal accumulation of non-resident cells that mediate adaptive immunity and inflammation are not yet identified, but recent work provides evidence suggesting that CF epithelial cells secrete markedly elevated levels of chemokines that could provide the missing link between the disordered innate and adaptive immune systems in CF [43]. Thus two dysregulated immunities in CF may yet be woven into one troublesome tale.