help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on December 11, 2003, doi:10.1164/rccm.200207-765OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow Online Supplement
Right arrow All Versions of this Article:
200207-765OCv1
169/5/645    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Becker, M. N.
Right arrow Articles by Randell, S. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Becker, M. N.
Right arrow Articles by Randell, S. H.
American Journal of Respiratory and Critical Care Medicine Vol 169. pp. 645-653, (2004)
© 2004 American Thoracic Society

Cytokine Secretion by Cystic Fibrosis Airway Epithelial Cells

Marie N. Becker, Mariam S. Sauer, Marianne S. Muhlebach, Andrew J. Hirsh, Qi Wu, Margrith W. Verghese and Scott H. Randell

Cystic Fibrosis/Pulmonary Research and Treatment Center, Department of Medicine; Department of Cellular and Molecular Physiology; and Department of Pediatrics, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

Correspondence and requests for reprints should be addressed to Scott H. Randell, Ph.D., UNC CF Center, CB 7248, 4011 Thurston-Bowles, Chapel Hill, NC 27599. E-mail: randell{at}med.unc.edu


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
It is controversial whether mutations in cystic fibrosis transmembrane conductance regulator intrinsically dysregulate inflammation. We characterized passage 2 human tracheobronchial epithelial cell cultures morphologically and physiologically and determined whether cytokine production or nuclear factor-{kappa}B activation was systematically altered in cystic fibrosis (CF) cells. Non-CF and CF cells originating from a total of 33 and 25 lungs, respectively, were available for culture on plastic or at an air–liquid interface until well differentiated. Forskolin-stimulated short-circuit currents were present in representative polarized non-CF cultures and were absent in CF cultures, whereas uridine 5'-triphosphate–stimulated currents were present in both. Constitutive or interleukin (IL)-1ß–induced IL-8 or IL-6 secretion or nuclear factor-{kappa}B activity was not significantly different between non-CF and CF cells. The cytokines regulated upon activation, normal T cell expressed and secreted (RANTES) and IL-10 were not detectable. Stimulation with tumor necrosis factor-{alpha} or a synthetic toll-like receptor 2 agonist or variable doses and times of Staphylococcus aureus culture filtrate revealed a single dose- and time-dependent difference in IL-8 production by CF cells. Interestingly, although IL-8 secretion after stimulation with Pseudomonas aeruginosa filtrates was not greater in CF cells in the absence of human serum, it was variably greater in its presence. Thus, although exaggerated responses may develop under certain conditions, our results do not support an overall intrinsically hyperinflammatory phenotype in CF cells.

Morbidity and mortality in cystic fibrosis (CF) are largely due to chronic endobronchial bacterial infection with severe neutrophilic inflammation. Ongoing infections result from impaired mucociliary clearance subsequent to defective ion transport, which depletes the periciliary liquid layer and raises mucus viscosity (1). Defects in antimicrobial activity (2), diminished binding and uptake of Pseudomonas aeruginosa by epithelial cells (3), or defective neutrophil phagocytosis (4) may also contribute. The efficacy of antiinflammatory therapies (5, 6) underscores the importance of inflammation to cause loss of lung function. Increased interleukin (IL)-8 and neutrophils found in CF bronchial lavage fluid without apparent infection (7, 8) or when normalized for bacterial burden (9, 10) suggest that defective CF transmembrane conductance regulator (CFTR) might directly dysregulate inflammation. It is reported that fetal human CF airways and lung tissues, even when sterile, accumulate more leukocytes compared with non-CF airways when implanted in immune-deficient mice (11, 12). However, the notion of inflammation without infection as a primary defect in CF infants has been challenged (13).

Several cell culture studies support the concept of dysregulated inflammatory responses in CF epithelial cells. IL-8 and IL-6 secretion in response to P. aeruginosa was elevated and more sustained in CFTR mutant cell lines (14, 15), and higher baseline nuclear factor-{kappa}B (NF-{kappa}B) activation was reversed by the introduction of normal CFTR or by culture at permissive temperature (16). Excessive cytokine production may reflect a cell stress response caused by mutant, misfolded CFTR accumulation in the endoplasmic reticulum, but cells with the G551D mutation, which express and traffic inactive CFTR to the cell surface, also had enhanced Ca2+ signaling and greater NF-{kappa}B activity (17). Similarly, two groups (18, 19) found that CF tracheal gland cells exhibit elevated basal and stimulated levels of IL-8 and IL-6. The CF gland cells had reduced levels of the inhibitor of nuclear factor {kappa}B{alpha} (I{kappa}B{alpha}), which was reversed by genistein, presumably because of enhanced delivery of mutant CFTR to the cell surface (20). Hypertonic stress also had a greater IL-8–inducing effect on CF gland cells (21), perhaps indicating a lower "stress threshold." Furthermore, a recent report showed equal baseline but higher tumor necrosis factor-{alpha} (TNF-{alpha})–stimulated NF-{kappa}B activity and IL-8 production in a CF cell line that was ascribed to altered I{kappa}Bß, rather than I{kappa}B{alpha}, activity (22).

Other studies suggest no or only small differences related to inflammation in CF cells. In primary human epithelial cells or cell lines on plastic, release of regulated upon activation, normal T cell expressed and secreted (RANTES) was dependent on CFTR expression, but basal or stimulated secretion of IL-8 or monocyte chemoattractant protein-1 was independent of CFTR status (23). No differences between CF and non-CF derived cell lines were observed for secretion of IL-8 or IL-6 induced by a variety of stimuli, including P. aeruginosa products (24). Similarly, no differences between CF and non-CF cell lines were found for IL-8 production in experiments in which correction of the CFTR defect normalized Cl- secretion and P. aeruginosa adherence (25). Still another study failed to detect differences in TNF-{alpha}–stimulated IL-8 secretion between paired cell lines where CFTR deficiency was corrected by an adenoviral vector (26). Reduced, as opposed to increased, IL-8 production by CF cell lines was reported that was reversed by the introduction of wild-type CFTR (27). Cells freshly harvested from inflamed CF lungs showed evidence for enhanced cytokine production (28), but no difference in baseline or stimulated IL-8 secretion due to CFTR status was detected after time in primary culture (26). No significant differences in basal IL-8 production or NF-{kappa}B activation were observed in primary nasal epithelial cells from CF versus non-CF donors, but P. aeruginosa adhered more readily and thus elicited a greater IL-8 response in CF cells (29). Thus, it remains controversial whether intrinsically exaggerated inflammatory responses result from mutations in CFTR.

Passaged and cryopreserved primary human tracheobronchial epithelial (hTBE) cells can be grown at an air–liquid interface (ALI), where they become well differentiated, resembling the in vivo morphology. We sought to establish whether passaged CF and non-CF hTBE cells grown at an ALI until well differentiated were similar morphologically and displayed physiologic properties consistent with their genotype. We then examined whether cytokine production or NF-{kappa}B activation was systematically altered in CF cells at baseline or in response to several relevant challenges, including IL-1ß, sterile culture filtrates of Staphylococcus aureus and P. aeruginosa, a synthetic toll-like receptor (TLR)-2 agonist or TNF-{alpha}.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Culture
Human lung tissue was procured under an institutional review board–approved protocol, and epithelial cell harvest and culture were performed using established procedures (30, 31). Samples originated from a total of 33 non-CF and 25 CF lungs (Table 1) . Cryopreserved passage 1 cells were cultured in bronchial epithelial growth medium on Vitrogen-coated plastic dishes. At 75–90% confluence, passage 2 cells were transferred to type IV collagen-coated Snapwells (Corning Costar, Cambridge, MA) for use in Ussing chambers or 0.4-µm T-Clear (Corning Costar) or Millicell CM membranes (Millipore, Bedford, MA) in low-endotoxin ALI medium (32). Beginning at Days 7–10, cells were maintained at an ALI. Histology was performed on formalin-fixed, paraffin-embedded cells at Day 21. Passage 2 cells were also cultured on 96-well plastic plates (35,000 cells/well) in bronchial epithelial growth medium. For all experiments, the minimum sample size was cultures from four individuals, but the sample size ranged up to 11. Samples were obtained from triplicate culture wells.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Demographics of airway tissue donors and guide to experimental use of cells

 
Bioelectric Properties
Six and 21 days after formation of an ALI, Snapwell inserts were mounted in Ussing chambers (Physiologic Instruments Inc., San Diego, CA). The epithelium was voltage clamped, and short-circuit current (Isc) and transepithelial resistance measured (Physiologic Instruments). Data were analyzed using Acquire and Analysis (version 1.2) software (Physiologic Instruments). Solutions and compound additions are given in the online supplement.

Bacterial Filtrates
S. aureus strain ATCC 29213 and P. aeruginosa strain ATCC 27853 were grown in trypticase soy broth for 72 hours at 37° with shaking at 250 rpm. Cultures were centrifuged at 5,500 x g (4°C) for 30 minutes, and supernatants were 0.45 µm filtered, aliquoted, and stored at -20°C.

Cell Stimulation and Biochemical Measurements
Well differentiated cells on membranes and poorly differentiated cells grown on plastic were challenged with IL-1ß, S. aureus, or P. aeruginosa filtrates, the TLR-2 agonist S-[2,3-bis(palmitoyloxy)-(2-RS)-propyl]-N-palmitoyl-(R)-Cys-(S)-Ser-Lys4-OH, trihydrochloride (Pam3Cys), or TNF-{alpha} as indicated. Cytokines, lactate dehydrogenase (LDH), and DNA were assayed as described in the online supplement.

Electrophoretic Mobility Shift Assays
hTBE cells on 24-mm inserts were treated with buffer or IL-1ß (5 ng/ml) for 1 hour, and nuclear extracts were prepared from three replicate wells and processed as described previously (32). Loading of gels was normalized for the protein content of the nuclear extracts.

Statistical Analysis
Data are presented as the means ± SEM. Raw cytokine concentrations, fold changes versus control subjects, or values normalized for DNA are given as indicated. Log transformation was used to achieve a normal distribution, and analysis of variance with standard weighted means was performed using VassarStats software (http://faculty.vassar.edu./lowry/VassarStats.html). Where appropriate, significance between groups was determined using Tukey's test and was accepted if p was less than 0.05.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
It is controversial whether the CF defect intrinsically alters the production of inflammatory mediators by airway epithelial cells. Although many prior investigations used cell lines, our approach was to study well differentiated, passaged, primary non-CF and CF hTBE cells.

Morphologic and Electrophysiologic Properties of Well Differentiated hTBE Cells
In an initial experiment, we simultaneously thawed cryopreserved primary epithelial cells derived from six non-CF and six CF airway specimens. All 12 cultures were initiated simultaneously, thus minimizing one source of experimental variability, and all cultures grew well as passage 1 cells on plastic and were transferred to porous supports as passage 2 cells. All passage 2 cells became confluent within 7–10 days, at which point an ALI was established. By 21 days after ALI, both CF and non-CF hTBE cells differentiated into a pseudostratified mucociliary epithelium resembling the in vivo phenotype. Although there was some variability in the thickness of the cell layer, the percentage of mucous goblet cells, and the extent of ciliation, there were no systematic morphologic differences related to derivation from non-CF versus CF airways. Representative photomicrographs are shown in Figure 1 .



View larger version (81K):
[in this window]
[in a new window]
  		Figure 1. Representative histology of non-cystic fibrosis (CF) (A) and CF (B) well differentiated Day 21 air–liquid interface (ALI) human tracheobronchial epithelial (hTBE) cell cultures. Both cell types reproduced a pseudostratified mucociliary epithelium that resembled the in vivo phenotype. The T-Clear porous support is visible below the cells. Original magnification x200, hematoxylin and eosin stain.

 
Using replicate cultures from the same 12 individuals, we determined whether polarized hTBE cells were physiologically consistent with their CF versus non-CF genotype. We performed Ussing chamber studies of all specimens at 6 days after ALI and in five of six CF and four of six non-CF specimens at 21 days after ALI. Representative tracings and the mean data are given in Figure 2 and Table 2 . At Day 6, forskolin did not stimulate Isc in any of the CF cultures but did in all six of the non-CF cultures, whereas both cell types responded to uridine 5'-triphosphate. At 21 days after ALI, CF and non-CF cells demonstrated similar basal Isc, and the Isc from the CF cells had a significantly larger amiloride-sensitive component compared with normal subjects. Similar to the earlier time point, forskolin did not stimulate Isc in CF cells but did in non-CF cells, whereas both cell types responded to uridine 5'-triphosphate. Thus, the electrophysiologic properties of the polarized cultures reproduce important features of the CF airway epithelial phenotype.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Representative tracings and summary data from Ussing chamber studies of Day 6 (A and C) and Day 21 (B and D) hTBE cell cultures. At Day 6, a forskolin (Forsk)-stimulated current was present in all six of the non-CF cultures but was absent from all CF cultures, whereas both cell types responded to uridine 5'-triphosphate (UTP). At Day 21, a greater proportion of the baseline Isc was amiloride (Amil) sensitive in CF cells, and a forskolin response was present only in non-CF cells, whereas both cell types responded to UTP. In C, n = 6, except for the UTP bars, where n = 2 and 1 for CF and normal subjects, respectively. In D, n = 5 and 4 for CF and normal subjects, respectively, except for the UTP bars, where n = 3 and 4 for CF and normal subjects, respectively. Isc = short-circuit current.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Basal electrophysiologic characteristics of noncystic fibrosis and cystic fibrosis cells measured in ussing chambers as described

 
Cytokine Secretion at Baseline and in Response to IL-1ß
Excessive basal or stimulated cytokine secretion has been implicated in the pathogenesis of CF airway inflammation. Therefore, we quantitated IL-8 or IL-6 secretion by replicate wells of the same set of cultures that we had characterized morphologically and electrophysiologically (Figure 3) . We used 5 ng/ml of IL-1ß, a commonly used dose shown to be just beyond the beginning of the plateau response level in preliminary dose–response experiments (data not shown). Unstimulated non-CF and CF cultures produced similar amounts of IL-8 (2.3 ± 0.5 and 1.6 ± 0.4 ng/ml, respectively). Sham treatment (phosphate-buffered saline [PBS] instead of PBS plus IL-1ß) did not increase IL-8 secretion in either type of culture, whereas stimulation with IL-1ß increased the levels of IL-8 nearly 10-fold. However, there were no statistically significant differences in IL-8 secretion between non-CF and CF cells (13.9 ± 1.8 and 16.1 ± 1.6 ng/ml, respectively), even when expressed as fold changes (8.26 ± 1.65 and 12.26 ± 3.81, respectively). In this experiment, baseline secretion of IL-6 was undetectable in most cultures, and stimulation with IL-1ß increased IL-6 somewhat more in CF than in non-CF cultures (417 ± 80 and 183 ± 63 pg/ml, respectively), but the change was not significant by analysis of variance of log-transformed values (p = 0.16). Fold change comparisons for IL-6 were not possible because so many of the starting values were 0. Neither IL-10 nor RANTES was detectable in these cultures.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 3. Box plot representation of interleukin (IL)-8 (A) and IL-6 (B) production by Day 21 hTBE cells as measured by ELISA. Basal medium was sampled before and after a 17-hour IL-1ß exposure. Sham treatment was performed with phosphate-buffered saline (PBS) instead of PBS plus IL-1ß. The top, middle, and bottom lines of the boxes correspond to the 75th, 50th, and 25th percentiles, and the whiskers are the 90th and 10th percentiles. The filled rectangle is the arithmetic mean. Triplicate wells were used, and each sample was assayed in duplicate (n = cells from six different individuals). After log transformation to achieve normality, analysis of variance indicated that the differences between non-CF and CF cells were not significant.

 
Activation of NF-{kappa}B at Baseline and in Response to IL-1ß
Because the promoter region of several important inflammatory mediators, including IL-8 and IL-6, contain NF-{kappa}B consensus sites and because of reports of constitutive activation of NF-{kappa}B in CF cells, we performed NF-{kappa}B electrophoretic mobility shift assay in the same set of CF and non-CF hTBE cultures in which morphology, physiology, and cytokine release were studied (Figure 4) . Both CF and non-CF nuclei contained low levels of NF-{kappa}B that were increased substantially after IL-1ß stimulation. There was no evidence that CF hTBE cells expressed higher levels of the active p65/p50 heterodimer (upper band) than non-CF hTBE cells, either before or after stimulation. The unlabeled NF-{kappa}B consensus but not the mutant oligonucleotide competed for binding with the labeled oligonucleotide, demonstrating the specificity for NF-{kappa}B, and antibodies against p50 and p65 produced the expected supershifts.



View larger version (111K):
[in this window]
[in a new window]
  		Figure 4. Nuclear factor-{kappa}B (NF-{kappa}B) electrophoretic mobility shift assay. Nuclear extracts were prepared from well differentiated non-CF and CF cells with or without IL-1ß stimulation. Each lane originated from an equal amount of nuclear protein extract. The bands were specific for NF-{kappa}B, as demonstrated by competition with nonradioactive consensus NF-{kappa}B oligomer and the lack of competition by a mutant consensus sequence. Antibodies specific to p50 and p65 NF-{kappa}B subunits resulted in a supershift. Both CF and non-CF cells exhibited low basal levels of nuclear NF-{kappa}B. After stimulation with IL-1ß, both CF and non-CF cells shifted NF-{kappa}B to the nucleus, and the predominant band was composed of p50/p65 heterodimers.

 
Cytokine Production in Response to Exogenous Stimuli
Because airway infection is a hallmark of CF, we performed experiments in which hTBE cells were exposed to soluble products from relevant bacterial pathogens. To model in vivo exposure, we added sterile filtrates of late stationary phase S. aureus or P. aeruginosa cultures, or similarly treated bacterial growth medium (trypticase soy broth) as the control, to the apical surface of polarized cells. Doses were chosen based on preliminary studies showing lack of overt toxicity of S. aureus and P. aeruginosa filtrates up to 10% and 20%, respectively, as indicated by maintenance of a patent ALI. S. aureus filtrates appeared more damaging to the cultures as manifest by fluid leak upon 48-hour exposure to the 20% dose. The S. aureus filtrates caused modest dose-dependent cellular cytotoxicity, as measured by LDH release, and non-CF cells seemed more susceptible (see Figure E1 in the online supplement). Forty-eight–hour treatment with P. aeruginosa filtrate, at any tested dose up to 20%, did not cause LDH release over control values. Thus, 10% S. aureus or 20% P. aeruginosa doses were chosen for most subsequent experiments. During a 24-hour exposure, 10% S. aureus or 20% P. aeruginosa filtrate both induced an approximately fourfold increase in IL-8 secretion (Table 3) . There were no statistically significant differences between non-CF and CF cells (Table 3, experiment 3.1). Human serum, as a source of LPS binding protein and soluble CD14, greatly enhances the response of human monocytes to the TLR-4 agonist LPS (33). Culture filtrates from the Gram-negative organism P. aeruginosa likely contain LPS among other active components. Therefore, we stimulated well differentiated hTBE cell cultures from seven non-CF or seven CF individuals with P. aeruginosa filtrates or the bacterial medium control (trypticase soy broth) in the presence of 10% human serum (Table 3, experiment 3.2). Human serum alone increased IL-8 levels (compare Table 3, experiments 3.1 and 3.2), but there was still an approximately fourfold increase after P. aeruginosa exposure. In this experiment, there was great variability in the range of both baseline values and fold increases, and there were no significant differences between non-CF and CF cells. The bacterial supernatants contain a mixture of compounds that can activate cells through TLRs and other mechanisms. Because our previous mRNA analysis indicated that hTBE cells express TLR-2 (32), we also exposed the cells to the synthetic TLR-2 agonist Pam3Cys. To accommodate a more extensive comparison between CF and non-CF cultures (n = 11 each), we chose the inactive OH3Cys analogue as the control rather than media alone. IL-8 levels in the OH3Cys-treated cells were higher than in previous control cultures. Nevertheless, hTBE cells responded specifically to Pam3Cys with robust IL-8 production, and the difference between non-CF and CF cells was insignificant (Table 3, experiment 3.3).


View this table:
[in this window]
[in a new window]
  		TABLE 3. INTERLEUKIN-8 secretion (ng/ml or average fold increase as indicated) in response to bacterial products or a synthetic toll-like receptor 2 agonist

 
Secretion of IL-6 in response to bacterial supernatants was examined in the absence or presence of human serum (Table 4 , experiments 4.1 or 4.2, respectively). In accordance with the IL-8 data, human serum alone increased IL-6 secretion, and P. aeruginosa or S. aureus filtrates further increased IL-6 secretion at least threefold under all conditions. Interestingly, secretion of IL-6 in the trypticase soy broth control was substantially lower in the CF groups, resulting in a greater fold increase in IL-6 in CF cells. However, because of variability between donors, the difference between CF and non-CF cultures was not significant.


View this table:
[in this window]
[in a new window]
  		TABLE 4. INTERLEUKIN-6 secretion (pg/ml or average fold increase as indicated) in response to bacterial products

 
Cytokine Production in Poorly Differentiated hTBE Cells in Response to Endogenous and Exogenous Stimuli
Because many previously reported studies involving non-CF versus CF cytokine production were performed using poorly differentiated cells or cell lines, we determined whether differences were detectable in primary cells before they assumed a mucociliary phenotype. For this purpose, hTBE cells were cultured on plastic until confluence and then exposed to media alone, the endogenous cytokines IL-1ß, TNF-{alpha}, or the synthetic TLR-2 agonist Pam3Cys and its control, OH3-Cys. As indicated in Figure 5 , there were no statistically significant differences in IL-8 production between poorly differentiated non-CF and CF cells.



View larger version (52K):
[in this window]
[in a new window]
  		Figure 5. IL-8 production by poorly differentiated hTBE cultured on plastic. hTBE cells were treated with media alone, stimulated as described with OH3Cys (inactive analogue of Pam3Cys, 25 µg/ml), Pam3Cys (25 µg/ml), tumor necrosis factor-{alpha} (TNF-{alpha}) (25 ng/ml), or IL-1ß (10 ng/ml) for 24 hours when supernatants were collected for IL-8 quantitation by ELISA. The data are means ± SEM from triplicate wells of cells derived from 11 different non-CF and CF lungs. Statistically significant differences between non-CF and CF groups were not present as indicated by analysis of variance.

 
Dose- and Time-dependent IL-8 Production in Well Differentiated CF and non-CF hTBE Cultures
A study comparing CFTR-deficient with CFTR-sufficient cell lines indicated that hyperinflammatory responses in CF cells became more apparent on prolonged stimulation (15). Because our previous studies might have missed dose- and/or time-dependent differences between non-CF and CF cells, we stimulated well differentiated hTBE cultures with varying concentrations of bacterial filtrates, sampling the basolateral medium at 8, 24, and 48 hours. IL-8 secretion is presented as fold increase over the trypticase soy broth control at 8 hours (Figure 6) . S. aureus treatment was performed in the absence of human serum, and P. aeruginosa challenges were performed in both the absence and presence of 10% human serum. Control IL-8 secretion was not different between non-CF and CF cultures, and similar increases were observed over time (note the different scales at 8, 24, and 48 hours in Figure 6). Except for the 20% group at 8 hours, S. aureus filtrates elicited similar IL-8 secretion in non-CF and CF cells at all time points; 20% P. aeruginosa filtrate in the absence of serum stimulated 10- to 20-fold increases in IL-8 at 24 and 48 hours, respectively, and there were no statistically significant differences between non-CF and CF cells. In contrast, when P. aeruginosa was added in the presence of human serum, in this experiment, IL-8 secretion by CF cells was significantly greater at all time points. The differences became most apparent at the 48-hour time point where only modest increases in IL-8 over control values were seen in non-CF cells, compared with a doubling at the higher P. aeruginosa doses in CF cells. Because cell proliferation during 48 hours of culture may differ between non-CF and CF cells, we measured the DNA at the end of the 48-hour exposure and normalized IL-8 levels for DNA content (Figure 7) . Again, CF versus non-CF differences were not significantly different, except for P. aeruginosa stimulation in the presence of human serum. Thus, normalization for DNA content did not alter the results.



View larger version (48K):
[in this window]
[in a new window]
  		Figure 6. IL-8 production by well differentiated non-CF and CF hTBE cells. Time course and dose response to S. aureus (S. a.) or P. aeruginosa (Ps. a.) in the absence of human serum (top) or P. aeruginosa in the presence of 10% human serum (bottom). ALI hTBE cell cultures were apically challenged with the indicated concentrations of bacterial filtrates. IL-8 in the basolateral media was measured by ELISA at 8, 24, or 48 hours. All data are presented as the means ± SEM of fold increases over the corresponding control value (0% bacterial filtrate, 20% trypticase soy broth) at 8 hours. Absolute levels are indicated on the 8-hour panels. n = 8 for the non-CF group and 9 for the CF group except for the 2.5% and 20% doses of S. aureus, the 20% P. aeruginosa dose without serum, and the 2.5% and 10% dose of P. aeruginosa with serum where n = 4 and 5 for non-CF and CF, respectively. Differences between non-CF and CF groups were determined by analysis of variance and Tukey'spost hoc test and are denoted by *p < 0.05 or **p < 0.01, respectively.

 


View larger version (38K):
[in this window]
[in a new window]
  		Figure 7. IL-8 production in well differentiated non-CF and CF hTBE cells after normalization for DNA content per culture well. ALI hTBE cell cultures were apically challenged with 10% S. aureus or P. aeruginosa in the absence of human serum or 20% P. aeruginosa in the presence of 10% human serum as indicated. IL-8 in the basolateral medium and DNA in the cell layer was measured by ELISA and a fluorescence-based kit, respectively. n = 8 for the non-CF group and 9 for the CF group except for P. aeruginosa without serum where n = 4 and 5, respectively. Statistically significant differences between non-CF and CF groups were determined by analysis of variance and Tukey's post hoc test and are denoted **p < 0.01.

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Evidence for excessive neutrophil dominated inflammation in bronchial lavage fluid from CF patients when compared with non-CF individuals has been documented (6–8), but the cause is not clear. In vitro studies performed with nasal or bronchial epithelial cells from CF or non-CF individuals or with immortalized cell lines in which CFTR status has been manipulated give conflicting results, ranging from increased NF-{kappa}B activation and production of IL-8 to no changes or even decreased levels of inflammatory cytokines (14–17, 22–27). Thus, it is controversial whether excessive pulmonary inflammation is an intrinsic property of the CFTR defect or whether it is secondary to the unique environment of the CF lung.

Our studies focused on primary airway epithelial cells. The culture methods employed included an expansion to a second passage. Samples from a total of 33 non-CF and 25 CF lungs were available for comparison. When cultured at an ALI until well differentiated, the cells mimicked many features (polarity, Isc, cAMP-sensitive Cl- conductance, fluid absorption, mucus transport) expected of normal and CF airway epithelium (Figure 1) (1). Unlike transfected cell lines, these passage 2 hTBE cells are likely to express physiologically relevant numbers of mutant and normal CFTR channels, as manifested by their electrophysiologic properties. Under these conditions, there was no evidence that CF-derived cells contained more activated NF-{kappa}B or secreted more IL-8 than non-CF cells, either in their basal state or when stimulated with IL-1ß. IL-10 or RANTES was not detectable in these cultures. IL-6 production in CF cells stimulated with IL-1ß was somewhat greater than in non-CF cells, but the difference was not statistically significant.

Airway epithelial cells express mRNA for many of the TLRs, including TLR-1, TLR-2, TLR-3, TLR-4, TLR-5, and TLR-6 (32). The bacterial products used in our studies likely stimulate hTBE cells through these receptors, activating characteristic downstream signal transduction pathways. In the absence of human serum, neither S. aureus or P. aeruginosa filtrates (with exception of the 20% S. aureus dose at 8 hours) nor a synthetic TLR-2 agonist revealed any differences in IL-8 or IL-6 secretion in well-differentiated hTBE derived from non-CF or CF lungs. The lack of differential sensitivity to relevant proinflammatory agents was not a result of prolonged culture to achieve a mucociliary phenotype, as poorly differentiated cells on plastic did not show differences in cytokine- or TLR-2 agonist–stimulated IL-8 secretion due to CFTR status.

Interestingly, when hTBE cultures were exposed to P. aeruginosa filtrate in the presence of human serum, significant differences sometimes became apparent between non-CF and CF cells. IL-8 secretion by both cell types continued to increase over time. CF cells demonstrated continuous dose responsiveness to P. aeruginosa filtrate, whereas the stimulatory effects of P. aeruginosa on non-CF cells seemed to wane, especially at the 48-hour time point. An initial experiment failed to reveal significant differences between CF and non-CF hTBE cells to P. aeruginosa filtrates at 24 hours even when human serum was present (Table 4). It is possible that variability between donors, including coincidentally clustered hyperresponders in the non-CF group, masked any differences in this particular experiment. Non-CF versus CF differences may have become apparent if this initial experiment was extended to 48 hours. However, there was a clear increase in P. aeruginosa–stimulated IL-8 production by CF cells in the presence of human serum in a second experiment. Overall, our data indicate that inflammatory responses of non-CF and CF cells are not globally different but that exaggerated responses may develop in CF cells under specific conditions. The significant differences in our experiments were never greater than 2.5-fold. CF cells may be more sensitive to synergism between serum and P. aeruginosa–derived factors, and it is possible that when multiple proinflammatory pathways are activated, non-CF cells are more capable of limiting their responsiveness than CF cells. As the studies presented here were under review, data were published comparing intercellular adhesion molecule-1 upregulation or IL-8 secretion in response to TNF-{alpha}, IL-1ß, killed whole Haemophilus influenzae or P. aeruginosa in primary hTBE cells from eight non-CF and eight CF donors (34). These studies support our findings that differences between non-CF and CF cells are apparent only under specific conditions and that donor-to-donor variability complicates their detection.

It is important to note that all in vitro cell culture systems are only an approximation of the actual in vivo physiologic state. The electrophysiologic properties of the CF epithelium in vivo include a higher baseline potential difference thought to represent hyperactivity of epithelial sodium channels and higher Isc responses to calcium mobilizing agonists such as uridine 5'-triphosphate (35, 36). Although the non-CF and CF passage 2 cultures we studied were absolutely faithful to their genotype regarding the absence or presence of cAMP stimulated currents, the CF cells did not exhibit higher baseline potential differences or exaggerated uridine 5'-triphosphate responses. One could argue that the intrinsic hyperinflammatory defect is linked to the sodium channel and calcium-activated chloride channel abnormalities. In vitro cultures displaying hyperactivity of sodium and calcium activated chloride channels will be necessary to resolve this question.

Studies using cell lines offer the attractive feature of manipulating CFTR status on an isogenic background. The exact cause for paradoxical results between the many reports using different paired CF and non-CF cell lines is unknown. In the recently published study of Aldallal and colleagues (34), significant differences in IL-8 secretion ascribed to CFTR correction could be demonstrated in one but not another set of paired cell lines. The differences may relate to altered gene expression patterns or changes in signal transduction pathways induced during the generation of the specific cell lines. Many cell lines are unstable and aneuploid because of the action of the transforming oncogenes used in their creation. In many cases, cell lines may be separated by many passages and may have accumulated differences other than CFTR status.

In summary, IL-8 secretion and NF-{kappa}B activation, at baseline or in response to a diverse set of relevant stimuli, were generally similar in a representative sample of passaged primary CF and non-CF hTBE cells that morphologically and physiologically reproduce many features of the in vivo bronchial epithelium. Only under certain conditions, such as P. aeruginosa in the presence but not in the absence of serum, was there an exaggerated and sustained but somewhat variable IL-8 response in CF cells. Our results suggest that intrinsic baseline differences in inflammation due to mutant CFTR per se are not a primary cause for the hyperinflammatory status of the CF lung. More likely, severe inflammation occurs in response to bacteria, their products, and host factors that accumulate in the unique environment created by defective mucociliary clearance in the CF lung. Under these circumstances, it is possible that CF cells are less capable of downregulating proinflammatory responses.


   Acknowledgments
 
The authors thank the Tissue Procurement/Cell Culture and Histology Cores of the University of North Carolina CF/Pulmonary Research and Treatment Center for excellent service. They also thank Dr. Harry Hurd of the University of North Carolina Department of Statistics for expert consultation.


   FOOTNOTES
 
Supported by the Cystic Fibrosis Foundation and National Institutes of Health grants HL58345, HL 60280, and HL 51818.

This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org

Conflict of Interest Statement: M.N.B. has no declared conflict of interest; M.S.S. has no declared conflict of interest; M.S.M. has no declared conflict of interest; A.J.H. has no declared conflict of interest; Q.W. has no declared conflict of interest; M.W.V. has no declared conflict of interest; S.H.R. serves as a consultant for Vertex Pharmaceuticals and has received a speaking fee for work not related to the subject of this manuscript.

Received in original form July 29, 2002; accepted in final form December 9, 2003


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Matsui H, Grubb BR, Tarran R, Randell SH, Gatzy JT, Davis CW, Boucher RC. Evidence for periciliary liquid layer depletion, not abnormal ion composition, in the pathogenesis of cystic fibrosis airways disease. Cell 1998;95:1005–1015.[CrossRef][Medline]
  2. Ganz T. Antimicrobial polypeptides in host defense of the respiratory tract. J Clin Invest 2002;109:693–697.[CrossRef][Medline]
  3. Pier GB. Role of the cystic fibrosis transmembrane conductance regulator in innate immunity to Pseudomonas aeruginosa infections. Proc Natl Acad Sci USA 2000;97:8822–8828.[Abstract/Free Full Text]
  4. Berger M, Sorensen RU, Tosi MF, Dearborn DG, Doering G. Complement receptor expression on neutrophils at an inflammatory site, the Pseudomonas-infected lung in cystic fibrosis. J Clin Invest 1989;84:1302–1313.
  5. Eigen H, Rosenstein BJ, FitzSimmons S, Schidlow DV. A multicenter study of alternate-day prednisone therapy in patients with cystic fibrosis: Cystic Fibrosis Foundation Prednisone Trial Group. J Pediatr 1995;126:515–523.[CrossRef][Medline]
  6. Konstan MW, Byard PJ, Hoppel CL, Davis PB. Effect of high-dose ibuprofen in patients with cystic fibrosis. N Engl J Med 1995;332:848–854.[Abstract/Free Full Text]
  7. Khan TZ, Wagener JS, Bost T, Martinez J, Accurso FJ, Riches DWH. Early pulmonary inflammation in infants with cystic fibrosis. Am J Respir Crit Care Med 1995;151:1075–1082.[Abstract]
  8. Balough K, McCubbin M, Weinberger M, Smits W, Ahrens R, Fick R. The relationship between infection and inflammation in the early stages of lung disease from cystic fibrosis. Pediatr Pulmonol 1995;20:63–70.[Medline]
  9. Muhlebach MS, Stewart PW, Leigh MW, Noah TL. Quantitation of inflammatory responses to bacteria in young cystic fibrosis and control patients. Am J Respir Crit Care Med 1999;160:186–191.[Abstract/Free Full Text]
  10. Muhlebach MS, Noah TL. Endotoxin activity and inflammatory markers in the airways of young patients with cystic fibrosis. Am J Respir Crit Care Med 2002;165:911–915.[Abstract/Free Full Text]
  11. Tirouvanziam R, de Bentzmann S, Hubeau C, Hinnrasky J, Jacquot J, Peault B, Puchelle E. Inflammation and infection in naive human cystic fibrosis airway grafts. Am J Respir Cell Mol Biol 2000;23:121–127.[Abstract/Free Full Text]
  12. Tirouvanziam R, Khazaal I, Peault B. Primary inflammation in human cystic fibrosis small airways. Am J Physiol 2002;283:L445–L451.
  13. Armstrong DS, Grimwood K, Carzino R, Carlin JB, Olinsky A, Phelan PD. Lower respiratory infection and inflammation in infants with newly diagnosed cystic fibrosis. BMJ 1995;310:1571–1572.[Free Full Text]
  14. DiMango E, Zar HJ, Bryan R, Prince A. Diverse Pseudomonas aeruginosa gene products stimulate respiratory epithelial cells to produce interleukin-8. J Clin Invest 1995;96:2204–2210.
  15. Kube D, Sontich U, Fletcher D, Davis PB. Proinflammatory cytokine responses to P. aeruginosa infection in human airway epithelial cell lines. Am J Physiol 2001;280:L493–L502.
  16. Stecenko AA, King G, Torii K, Breyer RM, Dworski R, Blackwell TS, Christman JW, Brigham KL. Dysregulated cytokine production in human cystic fibrosis bronchial epithelial cells. Inflammation 2001;25:145–155.[CrossRef][Medline]
  17. Weber AJ, Soong G, Bryan R, Saba S, Prince A. Activation of NF-kappaB in airway epithelial cells is dependent on CFTR trafficking and Cl- channel function. Am J Physiol 2001;281:L71–L78.
  18. Kammouni W, Figarella C, Marchand S, Merten M. Altered cytokine production by cystic fibrosis tracheal gland serous cells. Infect Immun 1997;65:5176–5183.[Abstract]
  19. Tabary O, Zahm JM, Hinnrasky J, Couetil JP, Cornillet P, Guenounou M, Gaillard D, Puchelle E, Jacquot J. Selective up-regulation of chemokine IL-8 expression in cystic fibrosis bronchial gland cells in vivo and in vitro. Am J Pathol 1998;153:921–930.[Abstract/Free Full Text]
  20. Tabary O, Escotte S, Couetil JP, Hubert D, Dusser D, Puchelle E, Jacquot J. Genistein inhibits constitutive and inducible NFkappaB activation and decreases IL-8 production by human cystic fibrosis bronchial gland cells. Am J Pathol 1999;155:473–481.[Abstract/Free Full Text]
  21. Tabary O, Escotte S, Couetil JP, Hubert D, Dusser D, Puchelle E, Jacquot J. High susceptibility for cystic fibrosis human airway gland cells to produce IL-8 through the I kappa B kinase alpha pathway in response to extracellular NaCl content. J Immunol 2000;164:3377–3384.[Abstract/Free Full Text]
  22. Venkatakrishnan A, Stecenko AA, King G, Blackwell TR, Brigham KL, Christman JW, Blackwell TS. Exaggerated activation of nuclear factor-kappaB and altered IkappaB-beta processing in cystic fibrosis bronchial epithelial cells. Am J Respir Cell Mol Biol 2000;23:396–403.[Abstract/Free Full Text]
  23. Schwiebert LM, Estell K, Propst SM. Chemokine expression in CF epithelia: implications for the role of CFTR in RANTES expression. Am J Physiol 1999;276:C700–C710.
  24. Bedard M, McClure CD, Schiller NL, Francoeur C, Cantin A, Denis M. Release of interleukin-8, interleukin-6, and colony-stimulating factors by upper airway epithelial cells: implications for cystic fibrosis. Am J Respir Cell Mol Biol 1993;9:455–462.
  25. Pizurki L, Morris MA, Chanson M, Solomon M, Pavirani A, Bouchardy I, Suter S. Cystic fibrosis transmembrane conductance regulator does not affect neutrophil migration across cystic fibrosis airway epithelial monolayers. Am J Pathol 2000;156:1407–1416.[Abstract/Free Full Text]
  26. Black HR, Yankaskas JR, Johnson LG, Noah TL. Interleukin-8 production by cystic fibrosis nasal epithelial cells after tumor necrosis factor-alpha and respiratory syncytial virus stimulation. Am J Respir Cell Mol Biol 1998;19:210–215.[Abstract/Free Full Text]
  27. Massengale AR, Quinn F Jr, Yankaskas J, Weissman D, McClellan WT, Cuff C, Aronoff SC. Reduced interleukin-8 production by cystic fibrosis airway epithelial cells. Am J Respir Cell Mol Biol 1999;20:1073–1080.[Abstract/Free Full Text]
  28. Bonfield TL, Konstan MW, Berger M. Altered respiratory epithelial cell cytokine production in cystic fibrosis. J Allergy Clin Immunol 1999;104:72–78.[CrossRef][Medline]
  29. Scheid P, Kempster L, Griesenbach U, Davies JC, Dewar A, Weber PP, Colledge WH, Evans MJ, Geddes DM, Alton EW. Inflammation in cystic fibrosis airways: relationship to increased bacterial adherence. Eur Respir J 2002;17:27–35.
  30. Bernacki SH, Nelson AL, Abdullah L, Sheehan JK, Harris A, Davis CW, Randell SH. Mucin gene expression during differentiation of human airway epithelia in vitro: muc4 and muc5b are strongly induced. Am J Respir Cell Mol Biol 1999;20:595–604.[Abstract/Free Full Text]
  31. Randell SH, Walstad L, Schwab UE, Grubb BR, Yankaskas JR. Isolation and culture of airway epithelial cells from chronically infected human lungs. In Vitro Cell Dev Biol Anim 2001;37:480–489.[CrossRef][Medline]
  32. Becker MN, Diamond G, Verghese MW, Randell SH. CD14-dependent lipopolysaccharide-induced beta-defensin-2 expression in human tracheobronchial epithelium. J Biol Chem 2000;275:29731–29736.[Abstract/Free Full Text]
  33. Beutler B. Tlr4: central component of the sole mammalian LPS sensor. Curr Opin Immunol 2000;12:20–26.[CrossRef][Medline]
  34. Aldallal N, McNaughton EE, Manzel LJ, Richards AM, Zabner J, Ferkol TW, Look DC. Inflammatory response in airway epithelial cells isolated from patients with cystic fibrosis. Am J Respir Crit Care Med 2002;166:1248–1256.[Abstract/Free Full Text]
  35. Knowles M, Gatzy J, Boucher R. Increased bioelectric potential difference across respiratory epithelia in cystic fibrosis. N Engl J Med 1981;305:1489–1495.[Abstract]
  36. Knowles MR, Clarke LL, Boucher RC. Activation by extracellular nucleotides of chloride secretion in the airway epithelia of patients with cystic fibrosis. N Engl J Med 1991;325:533–538.[Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow Online Supplement
Right arrow All Versions of this Article:
200207-765OCv1
169/5/645    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Becker, M. N.
Right arrow Articles by Randell, S. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Becker, M. N.
Right arrow Articles by Randell, S. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Cell Mol. Biol.
Copyright © 2004 American Thoracic Society