Polymorphonuclear leukocyte-dominated airway inflammation is a major component of cystic fibrosis (CF) lung disease and may be associated with CF transmembrane conductance regulator (CFTR) dysfunction as well as infection. Mutant ΔF508 CFTR is mistrafficked, accumulates in the endoplasmic reticulum (ER), and may cause “cell stress” and activation of nuclear factor (NF)-κB. G551D mutants also lack Cl− channel function, but CFTR is trafficked normally. We compared the effects of CFTR mutations on the endogenous activation of an NF-κB reporter construct. In transfected Chinese hamster ovary cells, the mistrafficked ΔF508 allele caused a sevenfold activation of NF-κB compared with wild-type CFTR or the G551D mutant (P < 0.001). NF-κB was also activated in 9/HTEo−/pCep-R cells and in 16HBE/pcftrantisense cell lines, which lack CFTR Cl− channel function but do not accumulate mutant protein in the ER. This endogenous activation of NF-κB was associated with elevated interleukin-8 expression. Impaired CFTR Cl− channel activity as well as cell stress due to accumulation of mistrafficked CFTR in the ER contributes to the endogenous activation of NF-κB in cells with the CFTR mutation.
- nuclear factor-κB
- cystic fibrosis transmembrane conductance regulator
- chloride channel
- inflammatory response
- intracellular calcium
- mitogen-activated protein kinase
cystic fibrosis (CF) transmembrane conductance regulator (CFTR) mutations have many effects on the physiology of respiratory epithelial cells in addition to the expected effects on Cl−transport. One of the major clinical manifestations of CFTR mutations is excessive airway inflammation (14-17), implying that CFTR affects the immune function of airway epithelial cells, particularly the expression of nuclear factor (NF)-κB-dependent proinflammatory chemokines and cytokines. Several lines of experimental evidence suggest that CFTR dysfunction influences the exogenous activation of epithelial immune function (8, 30). Bacteria persist in the respiratory tract of CF patients due to decreased antimicrobial activity in CF airway surface fluid and impair bacterial clearance due to mucus plugging (25). Cells with CFTR mutations have increased numbers of asialoglycolipid receptors for common bacterial pathogens (24). Ligation of these receptors activates NF-κB-dependent epithelial interleukin (IL)-8 production to recruit polymorphonuclear leukocytes (PMNs) from the circulation into the airways (8). Epithelial expression of IL-8 after bacterial stimulation is increased in cells with CFTR dysfunction, a finding demonstrated in both matched CF and corrected cell lines (5, 9, 27, 30) and in cftr-deficient [cftr(−/−)] mice compared with normal control mice (29). How CFTR dysfunction might affect the endogenous expression of NF-κB-dependent genes is less well documented.
NF-κB is a transcription factor required for the expression of many genes important in inflammation. It is complexed in the cell cytoplasm with IκB-α and -β, which are selectively phosphorylated, ubiquitinated, and degraded in the proteasome in response to stimuli (2). There is both signal-dependent and signal-independent IκB-α phosphorylation (20), a distinction that may be relevant to the endogenous activity of NF-κB as opposed to the activation in response to a superficial stimulus such as bacterial attachment to surface receptors. Ligation of superficial asialo-GM1 receptors on epithelial surfaces by bacterial adhesins stimulates Ca2+-dependent kinase activity through mitogen-activated protein kinase phosphorylation and NF-κB activation (23a). NF-κB translocation may be regulated by other intracellular factors that lead to a different pattern of IκB-α phosphorylation than that elicited by extracellular stimuli (4). It has been proposed that the abnormal trafficking of ΔF508 CFTR and its accumulation in the endoplasmic reticulum (ER) provides sufficient intracellular stress to cause NF-κB activation (2).
The most common CFTR mutation responsible for CF in ∼70% of patients is the deletion of a phenylalanine residue (ΔF508) in the nucleotide binding domain. The ΔF508 mutant CFTR not only fails to transport Cl− in response to cAMP but accumulates in the ER due to abnormal maturation and folding (1, 7, 32). From experimental data generated with adenovirus mutants targeted for retention in the ER, Pahl and colleagues (18, 19) proposed that homozygous ΔF508 mutations in CFTR would similarly cause Ca2+ release, activation of NF-κB, and resultant transcription of proinflammatory cytokines and chemokines. Comparison of the endogenous activation of NF-κB in IB3 cells (W1282X/ΔF508), a trafficking mutant similar to the homozygous ΔF508 mutation, with that in corrected C-38 cells, which express a functional but truncated form of CFTR, was consistent with this hypothesis. The transcriptionally active p65 component of NF-κB was found in the nuclei of unstimulated CF cells but not of the corrected cells by immunofluorescence labeling and was confirmed by gel shift assays (8). Manipulations to increase CFTR trafficking to the apical surface of the cell, such as growth at low temperature or the addition of glycerol, decreased the amount of p65 found in the IB3 cell nuclei. Although these observations were consistent with the hypothesis that abnormal CFTR proteins cause endogenous activation of NF-κB, they did not differentiate between two possible mechanisms: that “cell stress” associated with CFTR mistrafficking leads to changes in intracellular Ca2+ concentration ([Ca2+]i) and stimulation of NF-κB translocation or that CFTR Cl− channel dysfunction by itself is associated with endogenous activation of NF-κB.
Other types of CFTR mutations cause “CF-like” physiology, i.e., lack of Cl− secretion in response to cAMP, without causing trafficking abnormalities or resulting in the accumulation of mutant proteins in the ER (7). Analysis of the activation of NF-κB in such cell lines would provide the opportunity to determine whether Cl− channel dysfunction alone is associated with the regulation of NF-κB. The G551D CFTR mutant, in which a conserved glycine in the ATP binding cassette is mutated, lacks Cl−channel function, but the protein is properly folded and trafficked (12). Patients with the G551D mutation have a clinical disease that is indistinguishable from that caused by the more common ΔF508 mutation (10). The 9/HTEo− human tracheal epithelial cell line transfected with a plasmid directing constitutive expression of the CFTR regulatory (R) domain also fails to transport Cl− in response to cAMP while expressing low levels of wild-type CFTR message detected by RT-PCR (21). These 9/HTEo−/pCep-R cells are readily activated byPseudomonas aeruginosa and express significantly more IL-8 than control cells containing the empty vector (5). The impact of impaired CFTR Cl− channel activity on the normal interaction of IκBs and NF-κB is not known. Recently constructed 16HBE cell lines expressing plasmid-encoded cftr in the sense or antisense orientation also provide the opportunity to test the impact of CFTR Cl− channel function on the activation of NF-κB in airway cells without the confounding issues of mistrafficked CFTR protein (23).
In the experiments detailed in this report, several matched cell lines with CFTR dysfunction due to lack of Cl− channel function, CFTR misfolding, mistrafficking, or both were tested for endogenous activation of NF-κB and basal IL-8 expression. Although CFTR mistrafficking typical of the common ΔF508 mutation and associated cell stress may be responsible for the increased expression of proinflammatory cytokines and chemokines associated with CF cells, we sought to establish whether CFTR Cl− channel function as well as the effects of CFTR mistrafficking contributes to endogenous activation of NF-κB and the excessive inflammation that characterizes CF airway pathology.
MATERIALS AND METHODS
Epithelial Cell Lines
9/HTEo− cells, a human tracheal epithelial cell line (11), were grown in DMEM supplemented with 10% FCS and were stably transfected with either pCep, a vector control, or pCep-R, providing constitutive expression of the R-domain of CFTR resulting in typical CF-like physiology, i.e., failure of Cl− secretion in response to cAMP (21) and altered superficial sialylation (5). IB3 cells (ΔF508/W1282X), a human bronchial epithelial cell line, and C-38 cells, “corrected” cells with normal physiology that express an episomal truncated form of CFTR (33), were obtained from P. Zeitlin (Johns Hopkins University, Baltimore, MD) and grown in LHC-8 medium (Biofluids, Rockville, MD) plus 10% FCS. The IB3 cells used in these studies contain the empty vector. 16HBE cells transfected with plasmids expressing the first 131 nucleotides of cftr in either the sense or antisense orientation were grown in MEM plus 10% FCS plus l-glutamine and have been previously described (23). Cells expressing the antisense plasmid failed to respond to cAMP agonists with Cl− secretion. Stably transfected Chinese hamster ovary (CHO)-K1 cells that express ΔF508 CFTR or G551D CFTR were obtained from J. Riordan (Mayo Clinic Scottsdale, Scottsdale, AZ) and grown in αMEM plus 200 μM methotrexate plus 8% FCS. CHO-K1 cells were obtained from the American Type Culture Collection (ATCC) and lipofected with pCep plasmid constructs expressing either wild-type CFTR, G551D CFTR, ΔF508 CFTR, or the vector alone, provided by M. Drumm (Case Western Reserve University, Cleveland OH). CHO cells were grown in MEM plus 10% FCS. CFTR expression in the cells was verified by immunodetection with COOH terminus-specific mouse monoclonal anti-CFTR antibody (Genzyme, Minneapolis, MN).
Total RNA was isolated from confluent monolayers of IB3 and C-38 cells grown in 10-cm plates with RNeasy Mini Kit, and RT was performed with Omniscript RT (both from QIAGEN, Valencia, CA). PCR primers for human IL-8 were 5′-TACTCCAAACCTTTCCAACCC-3′ and 5′-AACTTCTCCACAACCCTCTG-3′. The cells were stimulated by the addition of 0.5 ml of P. aeruginosa PAO1 (5 × 109 colony-stimulating units/ml) scraped from an agar plate and resuspended in cell culture medium.
IL-8 in epithelial cell culture supernatants was measured with ELISA (R&D Systems, Minneapolis, MN). The cells were weaned from serum for 18 h, and the supernatants were harvested 18 h later for IL-8 ELISA. Duplicate wells were treated with trypan blue to assess epithelial viability during the assay, which was >75%. Each IL-8 data point was determined in quintuplicate, and means ± SD were calculated. Significance was evaluated with a one-way ANOVA with Dunnett's posttest (GraphPad Instat version 3.00, GraphPad Software, San Diego, CA) to test the null hypothesis that there was no difference in the amount of IL-8 produced by the epithelial cells with different CFTR function.
Activation of NF-κB Detected by Luciferase Reporter Constructs
Cells were grown in six-well plates to 80–90% confluence, washed once with PBS, and replated with serum- and antibiotic-free medium. The cells were transiently transfected with LipofectAMINE 2000 reagent (Life Technologies, Gaithersburg, MD), 1 μg of each plasmid DNA, pNF-κB-Luc (Stratagene, La Jolla, CA) and a constitutively active pRL-TK (Promega, Madison, WI), to control for transfection efficiency and incubated at 37°C in 5% CO2 for 18 h. Lipofection of the CHO cells was done with an additional 3 μg of pCep-CFTR vector DNA. The cells were washed once, lysed, and harvested with passive lysis buffer (Promega). Luciferase assays were performed with the reagents and protocol for the dual-luciferase reporter assay system (Promega) and analyzed with a luminometer. After standardization for transfection efficiency, data were plotted as the means of quadruplicate samples and are representative of at least three independent experiments. Statistical analysis was performed with an unpaired t-test for the experiments involving epithelial cells and a one-way ANOVA to analyze the results obtained with the CHO cells to accommodate a larger sample size.
[Ca2+]i in IB3 and C-38 Cells
Relative levels of [Ca2+]i were estimated in fluo 3-loaded IB3 and C-38 cells by fluorescence imaging. Monolayers of C-38 or IB3 cells were grown for 2 days on Lab-TekII chambered slides (Nalge Nunc International, Naperville, IL) in LHC-8 medium (Biofluids) supplemented with 10% FCS. The monolayers were washed and dye loaded with fluo 3-AM (Bio-Rad, Hercules, CA) for 0.5 h at room temperature in the dark in modified Ringer solution (145 mM NaCl, 5 mM KCl, 1 mM Na2HPO4, 2 mM CaCl2, 10 mM glucose, 20 mM HEPES, and 1 mM MgSO4, pH 7.4). The cells were rinsed in Ringer solution and incubated for 30 min to allow continued cleavage of the acetoxymethyl ester. The monolayers were imaged on a Zeiss Axiovert microscope at ×40 magnification with Inovision software. Images were taken at 10-s intervals before and after the addition of stimuli, and each image is the average of eight scans. Final measurements were taken after the addition of the ionophore 4-bromo-A-23187 (Bio-Rad) to allow for correction of differences in dye loading. Images were analyzed with Scion image software (Scion, Frederick, MD). Significance was determined with an unpaired t-test to compare the mean relative fluorescence of the two populations of epithelial cells.
Airway epithelial cells were grown to near confluence on Lab-TekII eight-chambered glass slides (Nalge Nunc International). They were washed with PBS, fixed with 4% paraformaldehyde in PBS for 20 min at room temperature, washed three times with PBS, and blocked in 10% goat serum in Ca2+/Mg2+ PBS for 1 h at room temperature. After blocking, the cells were washed three times in PBS and incubated with a 1:100 dilution of rabbit polyclonal antibody p65 (Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h at room temperature, washed, and then incubated with goat anti-rabbit IgG conjugated to tetramethylrhodamine B isothiocyanate (TRITC; 1:50; Zymed, South San Francisco, CA) in PBS plus 0.1% BSA plus an equal volume of bis-benzimide (50 μg/ml) for 1 h at room temperature. The cells were imaged on a Zeiss Axiovert S100 microscope with a Hamamatsu Orca II digital camera and Inovision software. 560D15 and 400DF15 band-pass filters were used for TRITC and bis-benzimide imaging, respectively. Emission data were collected with a triple-pass filter at wavelengths of 460 nm for bis-benzimide and 602 nm for TRITC.
Activation of NF-κB in Epithelial Cells With CFTR Dysfunction
To determine whether lack of CFTR-associated Cl−channel function affects the basal activation of NF-κB in airway epithelial cells, two cell lines with defective CFTR function but without mutations associated with mistrafficking were studied. 9/HTEo−/pCep cells and 9/HTEo−/pCep-R cells, which have constitutive overexpression of the CFTR R-domain and fail to transport Cl− in response to cAMP, were transfected with an NF-κB-luciferase reporter construct. The 9/HTEo−/pCep-R cells had 5.5-fold greater activation of NF-κB than 9/HTEo−/pCep cells expressing the vector control (Fig. 1). Similarly, 16HBE cells expressing cftr antisense, previously shown to lack CFTR-dependent Cl− channel activity, had fivefold greater activation of the NF-κB reporter than the cells expressingcftr in the sense orientation (P < 0.001 for both).
Nuclear p65 (Rel A) in Cells With CFTR Dysfunction Detected by Immunofluorescence
The presence of the p65 (Rel A) component of NF-κB in the cell nucleus is generally indicative of NF-κB activation and gene transcription. To verify that activation of NF-κB as detected by the reporter constructs was indicative of nuclear localization of the p65 component, the cell lines were permeabilized, treated with antibody to p65 that was detected with a goat anti-human IgG conjugated to TRITC with bis-benzimide to stain the nuclei. The control cells with normal CFTR function, either 9/HTEo−/pCep or 16HBE/cftr sense, had predominantly cytosolic p65 fluorescence (Fig. 2). In contrast, within the fields of 9/HTEo−/Cep-R and 16HBE/cftr antisense cells, there were significantly more cells with nuclear p65 TRITC fluorescence colocalizing with the blue bis-benzimide nuclear stain.
Activation of NF-κB in CHO Cells Expressing Wild-Type and Mutant CFTR
CHO cells that were transiently transfected with the pCep empty vector or pCep vector expressing wild-type CFTR had low levels of basal NF-κB activation as did the cells transfected with the plasmid expressing the mutant G551D CFTR (Fig.3 A). This was in contrast to the threefold increase in luciferase activity associated with CHO cells expressing mutant ΔF508 CFTR (P < 0.001 for ΔF508 compared with the pCep vector control). Experiments done with stably transfected CHO cells yielded similar results (Fig. 3 C); expression of ΔF508 CFTR was associated with a sevenfold increase in reporter activation compared with that in the CHO cells expressing the G551D mutation (P < 0.001) or CHO cells transfected with the reporter construct alone (P < 0.001). The relative expression of CFTR in the CHO cells was roughly equivalent (Fig. 3, B and D). In the CHO cells, expression of the ΔF508 CFTR mutation was sufficient to activate NF-κB, whereas the presence of comparable amounts of the mutant G551D CFTR was not.
IL-8 Expression Is Increased in Cell Lines With CFTR Dysfunction
The nuclear localization of NF-κB is associated with the transcription of several genes such as IL-8, a major PMN chemokine, that are important in inflammation. Increased expression of IL-8 has been frequently reported to be characteristic of CF cells after bacterial stimulation as well as under basal conditions (8). To determine whether the activation of NF-κB seen in the epithelial cell lines with dysfunctional CFTR is associated with an IL-8 response, IL-8 expression was assessed with RT-PCR (Fig.4). (These unstimulated cells did not produce sufficient amounts of IL-8 in serum-free medium to detect with a standard ELISA.) There was increased IL-8 mRNA expression detected in the IB3 cells compared with the endogenous levels in the C-38 cells, and both cell lines increased IL-8 mRNA expression in response to bacterial stimulation, included as a positive control.
IL-8 production could be reliably measured by ELISA in 9/HTEo− and 16HBE cell culture supernatants accumulated over 18 h after being weaned from serum. The 9/HTEo−/pCep-R cells produced more IL-8 than those expressing the vector control (P < 0.001), and lack of CFTR expression in the 16HBE cells expressing cftr antisense was also associated with increased IL-8 expression (P< 0.001; Fig. 5). Thus cell lines with defective CFTR Cl− channel activity, regardless of the nature of the CFTR defect, had increased IL-8 production compared with the corresponding cell line with normal CFTR activity.
[Ca2+]i in Cells With ER Overload
Epithelial cells with mistrafficked CFTR and “ER overload” have been postulated to have increased release of Ca2+ from ER stores as a consequence, a response that could activate Ca2+-dependent kinase activity. Relative [Ca2+]i was estimated in subconfluent IB3 and C-38 cells loaded with the fluorescent indicator fluo 3-AM (Fig.6). When the populations of the different cell types were compared, there was a consistent trend of increased [Ca2+]i associated with the IB3 cells, with relative basal fluorescence ranging from 0.6 to 0.85 compared with that in the population of C-38 cells, which had basal [Ca2+]i in the range of 0.5 to 0.75. Comparing the relative fluorescence of 10 individual cells at a single time point (30 s), the IB3 cells had a relative mean fluorescence of 0.758 ± 0.012 (SD) compared with 0.628 ± 0.08 for the C-38 cells (P < 0.001). Both cell types responded to the addition of P. aeruginosa with increased [Ca2+]i.
Airway inflammation is a major factor in the pathogenesis of CF pulmonary disease. Clinical studies have clearly established a direct correlation between infection, IL-8 secretion, and the accumulation of PMNs and their toxic products in the CF lung (14-17). Bacterial stimulation of cells with CFTR mutations activates significantly greater cytokine expression than matched cells with normal CFTR function (5, 8, 29, 30). Several reports (8, 28) suggest that cells with CFTR dysfunction have endogenously upregulated expression of proinflammatory cytokines and chemokines. Young infants with CF, even without detectable evidence of prior infection, have PMNs and increased amounts of IL-8 in their airways (14), and proinflammatory cytokine expression is increased in cells cultured directly from uninfected CF tissues (28). These findings are consistent with the in vitro studies demonstrating endogenous activation of NF-κB in CF but not in corrected cell lines. As indicated by the data in this report, lack of CFTR Cl− channel function and the physiological consequences of mistrafficked mutant CFTR both appear to contribute to the endogenous activation of NF-κB in CF airway cells, which is manifested by the increased expression of IL-8 in these cells, even in the absence of bacterial stimulation.
There is evidence of endogenous activation of NF-κB in cells with CFTR dysfunction due to several different mechanisms. In previous studies, IB3 cells, which express the W1282X/ΔF508 mutation associated with mistrafficked CFTR that accumulates in the ER, were found to have significant amounts of the p65 component of NF-κB in nuclei under basal conditions in which there was no nuclear NF-κB in the corrected C-38 cells (8). Because the translocation of NF-κB can be associated with many types of stimuli, additional consequences of accumulated mutant CFTR in the ER were also demonstrated in the mutant cell line. The IB3 cells were found to have higher basal [Ca2+]i, a predicted consequence of “ER stress” (2, 18). The presence of nuclear NF-κB in the IB3 cells could be correlated with increased IL-8 expression as estimated by RT-PCR compared with that in the control C-38 cell line. The concentration of IL-8 in the tissue culture supernatant was minimal in these cells after an 18-h incubation in serum-free medium; however, the relative amount of IL-8 message was consistent with a previous report by DiMango et al. (8). Not all investigators have found increased endogenous levels of IL-8 associated with CFTR mutations (30). Cell culture conditions, effects of FCS, and handling of the monolayers all can affect the translocation of NF-κB and endogenous expression of IL-8. Thus it seems relevant to demonstrate that the presence of nuclear p65 appears to be associated with the expected biological response, namely IL-8 expression.
The activation of other NF-κB-dependent pathways is not necessarily increased in IB3 compared with C-38 cells. Stimulation of NF-κB translocation is associated with an antiapoptotic effect in many cell lines including respiratory epithelial cells. Rates of apoptosis can be increased by proteasome inhibitors that inhibit IκB degradation and block NF-κB translocation (31). In a study published recently (23), the rates of IB3 and C-38 cell apoptosis were equivalent both under control conditions and after bacterial stimulation despite the observed differences in endogenous nuclear NF-κB. Thus the association between CFTR dysfunction, activation of NF-κB, and transcription of NF-κB-dependent genes is somewhat selective and is not a global response.
The consequences of CFTR mistrafficking and the activation of NF-κB were further demonstrated in CHO cells that do not normally express CFTR. The presence of the mistrafficked mutant ΔF508 CFTR was a significant stimulus for NF-κB activation in CHO cells. The G551D CFTR, which does not function appropriately as a Cl−channel but is trafficked normally to the apical surface of the respiratory epithelial cell (10), did not stimulate NF-κB. Expression of normal CFTR from the same plasmid construct did not cause NF-κB activation in CHO cells nor did transfection with the empty vector or the luciferase reporter construct. Although both the G551D and ΔF508 CFTR mutations are associated with clinical disease and lack of Cl− secretion in response to cAMP (10), the major difference between these mutants is the mistrafficking and accumulation of ΔF508 CFTR within the ER. Thus it appears that the consequences of mistrafficking of ΔF508 are sufficient to activate NF-κB in CHO cells.
In epithelial cells, unlike CHO cells, there appears to be a requirement for normal CFTR function in regulating NF-κB-dependent gene transcription. The homozygous ΔF508 or the compound ΔF508/W1282X mutation could activate NF-κB by two independent mechanisms: cell stress associated with mistrafficking or effects directly due to lack of CFTR Cl− channel activity. Cell lines that lack CFTR Cl− function due to either the overproduction of the CFTR R-domain or the expression of CFTR antisense had activated NF-κB and significant amounts of endogenous IL-8 production compared with the corresponding control cell lines. The amount of IL-8 endogenously expressed by these cell lines was greater than that of the IB3 or C38 cells and could be quantified by a standard ELISA assay. The presence of the p65 component of NF-κB in nuclei of undisturbed cells grown on coverslips as well as the relative activation of an NF-κB reporter construct compared with appropriate controls indicates that their activated state is unlikely to be an artifact of either cell transformation or manipulations occurring during cell culture. Because neither the 9/HTEo−/pCep-R or the 16HBE/cftr antisense cells have excessive CFTR accumulation in the ER, the lack of CFTR Cl− channel function in these epithelial cells appears to be responsible for the observed increase in NF-κB activation and IL-8 expression.
Exactly how CFTR dysfunction contributes to the activation and nuclear localization of NF-κB is unclear. There are multiple mechanisms for NF-κB activation that can be divided into signal-dependent and signal-independent pathways (20). In airway epithelial cells, bacterial attachment to asialo-GM1 activates a Ca2+-dependent signaling pathway that results in NF-κB activation and IL-8 transcription through a cascade that includes phosphorylation of both the p38 and extracellular signal-regulated kinase families of mitogen-activated protein kinases (23a). A signal-independent pathway, perhaps involving IκB or NF-κB regulation directly, may be affected by mistrafficked CFTR. IκB-β has been shown to be increased in IB3 cells under basal conditions and is hypophosphorylated, enabling nuclear NF-κB with bound IκB-β complexes to direct transcription (30). Hypophosphorylated IκB-β may be involved in either signal-dependent or the persistent activation of NF-κB (26). Diminished levels of cytosolic IκB-α was associated with CF (ΔF508/ΔF508) bronchial gland cells in primary culture and correlated with the presence of nuclear NF-κB Rel A (27). It is possible that CFTR function, perhaps by affecting pH in specific intracellular compartments (3), affects relevant kinase or phosphatase activity. Abnormalities in signal transducer and activator of transduction-1 phosphorylation in the same 9/HTEo−/pCep-R cells have been described recently (13), indicating that CFTR may have a role in multiple signaling pathways. These observations reflect the complexities of CFTR involvement in the normal physiology of the epithelial cell. Strategies such as gene therapy to correct CFTR Cl− channel function alone, which do not ameliorate the effects of mutant CFTR mistrafficking in the majority of airway epithelial cells, may not be sufficient to control the inflammatory component of CF lung disease.
These consequences of CFTR dysfunction in activating translocation of NF-κB and stimulating the expression of proinflammatory cytokines are important in not only understanding the many roles of CFTR in normal cell physiology but are also clinically relevant. The majority of CF patients with the most common ΔF508 CFTR alleles have multiple reasons for increased expression of inflammatory mediators in their airways. The effects of CFTR mistrafficking and ER stress, lack of CFTR Cl− channel function, and exogenous stimulation due to airway infection all act to increase epithelial NF-κB translocation and the expression of proinflammatory genes. This epithelial immune activation does provide multiple targets for therapy. It may be possible to devise a therapy to modulate the endogenous upregulation of NF-κB-dependent gene transcription without compromising the epithelial barrier function or the host inflammatory response to exogenous airway pathogens.
Expert technical assistance was provided by Robert Adamo.
This work was supported by National Heart, Lung, and Blood Institute Grants HL-56194 (to A. Prince) and P50-HL-60293 (to P. Davis, Case Western Reserve University, Cleveland, OH). Imaging studies were performed in the Optical Microscopy Facility of the Herbert Irving Cancer Center at Columbia University (New York, NY), supported by National Center for Research Resources Grant 1-S10-RR-10506.
Original submission in response to a special call for papers on “CFTR Trafficking and Signaling in Respiratory Epithelium.”
Address for reprint requests and other correspondence: A. S. Prince, Black Bldg. 416, Columbia Univ., 650 West 168th St., New York, NY 10032 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society