NF-κB Mediates Up-regulation of CFTR Gene Expression in Calu-3 Cells by Interleukin-1β*

Inflammation of the airways is a major feature of the inherited disease cystic fibrosis. Previous studies have shown that the pro-inflammatory cytokines tumor necrosis factor α and interferon γ reduce the expression of the cystic fibrosis transmembrane conductance regulator (CFTR) gene (CFTR) in HT-29 and T84 cells by acting post-transcriptionally. We have investigated the effect of the pro-inflammatory peptide interleukin 1β (IL-1β) on the expression of the CFTR in Calu-3 cells. IL-1β increased the production of CFTR mRNA in a dose- and time-dependent manner. Its action was inhibited by inhibitors of the NF-κB pathway, includingN-acetyl-l-cysteine, pyrrolidine dithiocarbamate, and a synthetic cell-permeable peptide containing the NF-κB nuclear localization signal sequence. Gel shift analysis showed that IL-1β activated NF-κB in Calu-3 cells, and transfection experiments using p50 and RelA expressing vectors showed that exogenous transfected NF-κB subunits increased the concentration of CFTR mRNA. Gel shift analysis with antibody supershifting also showed that IL-1β caused the binding of NF-κB to a κB-like response element at position −1103 to −1093 in the CFTR5′-flanking region. Transfection experiments using −2150 to +52CFTR reporter gene constructs showed that the activity of the CFTR promoter is enhanced by exogenous transfected NF-κB and IL-1β and that this enhancement is due, at least in part, to the −1103 to −1093 κB site. We conclude that the intracellular signaling that leads to increased CFTR mRNA in response to IL-1β in Calu-3 cells includes the binding of NF-κB to the −1103 κB element and a subsequent increase in CFTR promoter activity.


Inflammation of the airways is a major feature of the inherited disease cystic fibrosis. Previous studies have shown that the pro-inflammatory cytokines tumor necrosis factor ␣ and interferon ␥ reduce the expression of the cystic fibrosis transmembrane conductance regulator (CFTR) gene (CFTR) in HT-29 and T84 cells by acting post-transcriptionally. We have investigated the effect of the pro-inflammatory peptide interleukin 1␤ (IL-1␤) on the expression of the CFTR in Calu-3 cells. IL-1␤ increased the production of CFTR mRNA in a dose-and time-dependent manner. Its action was inhibited by inhibitors of the NF-B pathway, including N-acetyl-L-cysteine, pyrrolidine dithiocarbamate, and a synthetic cellpermeable peptide containing the NF-B nuclear localization signal sequence. Gel shift analysis showed that IL-1␤ activated NF-B in Calu-3 cells, and transfection experiments using p50 and RelA expressing vectors
showed that exogenous transfected NF-B subunits increased the concentration of CFTR mRNA. Gel shift analysis with antibody supershifting also showed that IL-1␤ caused the binding of NF-B to a B-like response element at position ؊1103 to ؊1093 in the CFTR 5-flanking region. Transfection experiments using ؊2150 to ؉52 CFTR reporter gene constructs showed that the activity of the CFTR promoter is enhanced by exogenous transfected NF-B and IL-1␤ and that this enhancement is due, at least in part, to the ؊1103 to ؊1093 B site. We conclude that the intracellular signaling that leads to increased CFTR mRNA in response to IL-1␤ in Calu-3 cells includes the binding of NF-B to the ؊1103 B element and a subsequent increase in CFTR promoter activity.
Cystic fibrosis (CF) 1 is an inherited multisystem disease caused by a number of mutations that affect the gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR) (1)(2)(3), an integral membrane protein that functions as a cAMP-regulated chloride channel (4). Despite our increased understanding of the structure and function of CFTR, the mechanisms by which the absence or dysfunction of CFTR causes numerous disorders is less well documented. One of these puzzling disorders is the destruction of the lung by infective-inflammatory injury, which is the most common cause of morbidity in cystic fibrosis. Several clinical studies of the airways of young patients with cystic fibrosis have found excessive amounts of pro-inflammatory cytokines, even in the absence of any clinical lung disease or detectable infection (5). At least two of these pro-inflammatory peptides, TNF␣ and INF␥, modulate the expression of the CFTR by acting post-transcriptionally (6,7). However, the effects of other pro-inflammatory cytokines, such as interleukin-1␤ (IL-1␤), on CFTR expression are poorly documented (8). TNF␣ and IL-1␤, which have overlapping and synergistic effects on cell function, should be considered to be proximal or primary cytokines, in many respects, because they are produced early in the response to infection and determine the pattern of later cytokine production and secretion in the inflammatory response. The signaling pathways mediated by the receptors for both IL-1␤ and TNF␣ are complex and involve multiple coordinated kinases, including JNK, SAP/p38 and ERK2 MAP kinases. These activate transcription factors such as AP-1 and NF-IL6 (C/EBP␤). The receptor-signaling pathways ultimately converge upon the NF-B-inducing kinase, which activates the IB kinase complex. Phosphorylation of the NF-B cytoplasmic inhibitory binding protein, IB, by IB kinase triggers its ubiquitination/degradation and allows the release of the active form of the NF-B factor. Thus, we need to know whether IL-1␤, like TNF␣, modulates CFTR expression, and whether this effect depends on transcription factors such as NF-B.
NF-B is clearly involved in the inducible regulation of genes in the immune system and the inflammatory response. NF-B consists of dimeric complexes of Rel/NF-B proteins sequestered in the cytosol by the inhibitory proteins IB. The phosphorylation and degradation of IB leads to the translocation of NF-B to the nucleus, where it binds to specific cis-regulatory elements located in transcriptional regulatory regions of its target genes.
Previous studies on CFTR transcription have shown that the cell specificity and the low concentration of CFTR are at least partly dictated by the genomic sequences 5Ј upstream of the transcription initiation region (9,10). Analysis of the CFTR 5Ј-region revealed that the CFTR promoter is a TATA-less promoter, which probably explains why transcription is initiated at multiple start points (9 -11) and has a high G/C content. Studies on the CFTR promoter have provided differing results on the location of the major transcription initiation site and the * This work was supported in part by the INSERM, CNRS, Association Française de Lutte contre la Mucoviscidose. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ Supported by a fellowship from the Association Française de Lutte contre la Mucoviscidose and Fondation pour la Recherche Médicale.
We have therefore investigated the effect of IL-1␤ on the expression of the CFTR in Calu-3 cells. Northern blot analysis showed that IL-1␤ stimulated the production of CFTR mRNA via an NF-B-dependent mechanism. Electrophoretic mobility shift assays and reporter transfection assays indicated that a B element at position Ϫ1103 of the CFTR 5Ј-flanking region was involved in this regulation.
Northern Blot Analysis-Aliquots (15 g) of total RNA isolated from Calu-3 or T84 cells using the TRIzol reagent (Life Technologies) were size-fractionated by agarose gel electrophoresis. The RNA was then transferred to a nylon membrane (Promega, Charbonnières, France) by capillary blotting, fixed by heating, and hybridized under standard conditions with the Quick Hyb protocol provided by Stratagene (Ozyme, Les Ulis, France). The 32 P-labeled CFTR probe consisted of the 1.5-kb EcoRI-EcoRI fragment of human CFTR cDNA labeled by random priming. The membrane was also hybridized with a human ␤-actin cDNA probe from Oncogene Science (France Biochem, Meudon, France). The mRNAs were quantified by densitometric scanning of the autoradiograms using an ImageMaster VSD (Amersham Pharmacia Biotech, Orsay, France) Nuclear Extracts-Cells were rinsed twice with cold phosphate-buffered saline (PBS), pH 7.4, scraped off into cold PBS, and centrifuged at 600 ϫ g for 3 min at 4°C. The resulting cell pellet was suspended in lysis buffer A (10 mM HEPES-NaOH (pH 7.9), 1.5 mM MgCl 2 , 10 mM KCl, 0.5 mM dithiothreitol (DTT), 0.5 mM EDTA, and supplemented with an antiprotease mixture (Roche Molecular Biochemicals). IGEPAL CA 630 (Sigma) detergent was then added (0.05%, v/v), and the cells were left on ice for 10 min. The crude nuclei released by lysis were pelleted by centrifugation at 1100 ϫ g for 10 min at 4°C, washed in lysis buffer A, and suspended in buffer C (20 mM HEPES (pH 7.9), 25% (v/v) glycerol, 420 mM NaCl, 1.5 mM MgCl 2 , 0.5 mM DTT, 0.5 mM EDTA and antiprotease mixture (Roche Molecular Biochemicals) by vigorous pipetting. The lysis of nuclei was checked under a phase-contrast microscope. The nuclei were left on ice for 15 min, vortexed, and clarified by centrifugation at 15,000 ϫ g for 5 min at 4°C. The protein concentration (ϳ5 mg/ml) was determined by the Lowry method, and the nuclear extracts were rapidly frozen and stored at Ϫ80°C.
Electrophoretic Mobility Shift Assay-Calu-3 cells were placed in medium without serum for 16 -18 h, and IL-1␤ (Calbiochem, France) (2 ng/ml) was then added to the medium. Oligonucleotides, synthesized by Eurobio (France), were annealed, end-labeled with [␥-32 P]dATP (50 Ci at 3000 Ci/mmol, Amersham Pharmacia Biotech) using the T4 polynucleotide kinase (Life Technologies) and purified on micro-spin purification columns (Qiagen, France). Binding reactions were done by mixing ϳ40 fmol of double-stranded, end-labeled oligonucleotides with 10 g of nuclear extract proteins for 20 min at room temperature in a final volume of 20 l. The mix contained EMSA binding buffer (10 mM HEPES, pH 7.9, 10 mM KCl, 0.5 mM DTT, 0.5 mM EDTA, 5% glycerol, and 0.2-1 g of poly(dI-dC)). Unlabeled double-stranded competitors were added to the binding reaction mixture 10 min prior to adding labeled probe. Supershift analyses were done by incubating nuclear extracts with 1.5 ng of the appropriate antibody (anti p50, RelA, or cRel from Santa Cruz Biotechnology, Inc., Tebu, France) for 30 min on ice prior to adding 32 P-labeled probe. Samples were loaded onto a 5% nondenaturing polyacrylamide gel in 0.5ϫ Tris borate EDTA, without any loading dye, and electrophoresed at 125 V for 2.5 h. The separated DNA⅐protein complexes were visualized on gels by autoradiography at Ϫ80°C for 4 -16 h. All experiments were done at least three times.
Plasmid Constructions-The (Ϫ2150/ϩ52)pGL3 construct was obtained by subcloning the SacI/HindIII 2.2-kb fragment of the (Ϫ2150/ ϩ52)-luc construct, generously provided by Dr. G. S. McKnight (University of Washington, Seattle, WA), into the pGL3 basic vector (Promega, Charbonnières, France). In vitro site-directed mutagenesis was performed on this plasmid using the Stratagene mutagenesis system following the manufacturer's instructions. The sequences of the (Ϫ2150/ϩ52)pGL3 constructs were confirmed by DNA sequencing (Genome Express, France). Two independent clones were tested in transfection experiments and gave similar luciferase activities. CMV-driven expression plasmids for p50 and RelA were kindly provided by Dr. Bauerle (Tularik, San Francisco, CA). A ␤-galactosidase-encoding vector, CMV-␤-galactosidase, was purchased from CLONTECH.
Transfections and Assays of Reporter Gene Constructs-Plasmids for transfection were purified with the Qiagen endo-free plasmid Mega kit. To avoid any possible influence of the quality of plasmid preparation on transfection efficiency, each series of transfection experiments was performed using products from at least two different amplifications. Calu-3 and T84 cells were transiently transfected in six-well dishes with LipofectAMINE Plus reagent (Life Technologies) and 2 g/well DNA in opti-MEM medium (Life Technologies) for 20 h, according to the manufacturer's directions. The transfected cells were washed three times with cold PBS and scraped off into 240 l of lysis buffer (Promega). Each lysate was mixed vigorously and clarified by centrifugation at 12,000 ϫ g for 3 min at 4°C. Supernatants were used for reporter assays. The luciferase activity in 20-or 30-l extract was evaluated with the Luciferase Reporter assay system (Promega) and a Berthold Biolumat LB9500 luminometer.

RESULTS
Effect of IL-1␤ on CFTR mRNA in Calu-3 Cells-Two proinflammatory cytokines, TNF␣ and INF␥, have been reported to down-regulate CFTR expression in colonic cell lines (6,7). We determined whether this response to pro-inflammatory mediators was a general phenomenon by testing the ability of TNF␣ and IL-1␤ to modulate CFTR expression in colonic (HT-29) and pulmonary (Calu-3) cell lines. Cultures of HT-29 and Calu-3 cells were stimulated for 24 h with TNF␣ or IL-1␤, and CFTR mRNA was measured by Northern hybridization analysis. TNF␣ reduced the amount of CFTR mRNA in HT-29, as it has already been described (6, 7), but increased it slightly in Calu-3 cells (Fig. 1A). These results contrast with those obtained with IL-1␤, which significantly increased the CFTR mRNA in both cell lines. The increase in CFTR mRNA in Calu-3 cells induced by IL-1␤ treatment was always greater than that induced by TNF␣, under our experimental conditions. Treatment of Calu-3 cells with various doses of IL-1␤ (0.5, 1.0, and 2.5 ng/ml) increased the amount of CFTR mRNA in a dose-dependent manner (Fig. 1B, upper bands), whereas the mRNA of the constitutively expressed control, ␤-actin, was stable under all conditions (Fig. 1B, lower bands). We investigated the time course of this effect by incubating Calu-3 cells with IL-1␤ (2 ng/ml) for 0, 3, 6, 9, 15, and 24 h and analyzing the CFTR mRNA by Northern blotting (Fig. 1C). Fig. 1C (right  panel) shows the kinetic of the CFTR mRNA increase in response to IL-1␤. IL-1␤ caused a rapid increase in steady-state levels of CFTR mRNA and continued to exert its effect for up to 24 h thereafter. Thus, these data indicated that the effect of IL-1␤ on CFTR mRNA production is dose-and time-dependent.
Influence of IL-1␤ on CFTR Gene Expression: Role of an NF-B-dependent Pathway-Because IL-1␤ is known to activate NF-B in several cell types, we determined whether the IL-1␤-induced expression of CFTR mRNA was mediated by an NF-B-dependent mechanism. In the first series of experiments, the cells were incubated with N-acetyl-L-cysteine (NAC) and pyrrolidine dithiocarbamate (PDTC), two radical scavengers that inhibit IB phosphorylation/degradation and, consequently, NF-B activation. In the second series of experiments, the cells were pretreated with a synthetic peptide containing the NF-B nuclear localization signal (NLS) fused to a membrane-permeable hydrophobic region, which specifically prevents NF-B nuclear translocation, or with its inactive analogue (15,16). The cells were then stimulated with IL-1␤. NAC and PDTC did not modulate the amount of constitutive CFTR mRNA but markedly reduced the IL-1␤-mediated increase in CFTR mRNA (Fig. 1B). Similar results were obtained with cells pretreated with the NF-B NLS but not those pretreated with its inactive form. These results suggest that IL-1␤ acts on CFTR expression, at least in part, via an NF-B-dependent pathway. We investigated the link between IL-1␤, NF-B, and the expression of the CFTR using two experimental strategies. We first measured the NF-B activity by EMSA using nuclear extracts from Calu-3 cells treated with IL-1␤ and a labeled consensus B probe (mouse immunoglobulin kappa light chain enhancer, B element). IL-1␤ activated NF-B, as indicated by the appearance of DNA⅐NF-B complexes (Fig. 2). We also transiently transfected Calu-3 and T84 cells (a human intestinal cell line that also expresses the CFTR constitutively) with CMV-driven RelA (p65) and p50 expressing vectors, or the empty vector (mock control), and then subjected them to Northern blot analysis. We checked that this transfection procedure mimicked the endogenous activation of the NF-B factor by looking for specific factors binding the B element in the nu-cleus of the p50/p65-transfected cells (Fig. 2). The relative amounts of CFTR mRNA in NF-B-expressing cells were greater than those of the control cells (Fig. 3)

. Northern blot analysis of CFTR mRNA in Calu-3 and HT-29 cells following stimulation with IL-1␤ and/or TNF␣. A, HT-29 and Calu-3 cells
were incubated with or without 15 ng/ml TNF␣ or 2.5 ng/ml IL-1␤ for 24 h. Total RNA was then prepared, fractionated on agarose gel, transferred onto nylon membranes, and hybridized with the appropriate cDNA probe (CFTR, ␤-actin). A Northern blot autoradiograph representative of three independent experiments is shown.  5) and by the incapacity of the same molar excess of an unrelated oligomer (AP-2 response element) to produce the same effect (lanes 3 and 6). The specificity of the labeled complexes is also shown by supershift analysis using an anti-IB antibody as control (lane 7) and an anti-RelA antibody (lane 8), which prevents its formation. Finally, the nuclear extracts from p50/RelA-transfected cells (lane 10) produced the same shifted complex, further indicating that this complex corresponds to NF-B. The arrow denotes the NF-B-specific complex. that the CFTR 5Ј-flanking region is sufficient to trigger the tissue-specific, low activity of the CFTR (9, 10). If the NF-B factor acts on CFTR transcription, there must be an NF-B response element in a region regulating the transcription of the CFTR gene, and in particular, in the 5Ј-flanking region. Analysis of the sequence of this region revealed a putative NF-B response element between Ϫ1103 and Ϫ1093 (5Ј-GGGAAT-GCCC-3Ј) that differed by 1 bp from the consensus NF-B response element (GGGNNTYYCC) (17). We next investigated whether the action of NF-B on CFTR mRNA production involved the binding of NF-B to this putative cis-regulatory sequence. We used an electrophoretic mobility shift assay (EMSA) to determine the capacity of the Ϫ1103 to Ϫ1093 sequence to bind NF-B proteins using the Ϫ1111 to Ϫ1090 region of CFTR as probe and nuclear extracts from Calu-3 cells treated with IL-1␤ (2 ng/ml) (Fig. 4). Cells incubated with IL-1␤ formed a single shifted complex. This complex was displaced by excess unlabeled consensus B element from the mouse immunoglobulin kappa light chain enhancer (B site), whereas the same excess of unlabeled consensus AP-2 element did not change the binding of the labeled probe (Fig. 4, compare lanes  4 and 5). Thus, a specific IL-1␤-induced factor that has a high affinity for the consensus B element binds to the Ϫ1111 to Ϫ1090 probe. We attempted to identify the proteins that formed the shifted complex in a supershift analysis using antibodies to three members of the NF-B/Rel family (p50, RelA (p65), and cRel). The formation of the shifted complex was prevented by anti-p50 or anti-RelA but not by anti-cRel antibodies (Fig. 4, lanes 7-9).
Enhanced CFTR Promoter Activity by NF-B-NF-B (probably the p50/RelA heterodimer) bound the Ϫ1103 B element in vitro. Thus, this sequence could constitute a B regulatory element in the CFTR 5Ј-flanking region. We investigated this and determined whether NF-B acted on the transcription of the CFTR by constructing a human CFTR promoter-luc plasmid, (Ϫ2150/ϩ52)pGL3, and transfecting Calu-3 cells with it (Fig. 5A). Cotransfection experiments using this construct plus the CMV-p50-and RelA-expressing vectors, or the mock plasmids as control, showed a 2.0-fold increase in the promoter activity when the cells were cotransfected with the CMV-p50 and RelA plasmids. Thus the CFTR promoter activity was stimulated by NF-B subunits. We then generated a site-specific mutation of the Ϫ1103 B element in the (Ϫ2150/ ϩ52)pGL3 construct by changing two nucleotides within the Ϫ1103 to Ϫ1093 sequence (from 5Ј-GGGAATGCCC-3Ј to 5Ј-GGGCATTTCT-3Ј). This base transition was sufficient to de-stroy the binding capacity of the NF-B proteins (data not shown). The activity of mutated CFTR promoter was less strongly stimulated (1.4-fold induction) than was the wild type one when this reporter construct was cotransfected with p50and RelA-expressing vectors (Fig. 5A). Hence, the Ϫ1103 to Ϫ1093 B element of the CFTR 5Ј-flanking region may take part in the up-regulation of the CFTR by exogenous transfected NF-B factor. Finally, we performed transfection experiments with the (Ϫ2150/ϩ52)pGL3 reporter gene constructs and stimulated cells with IL-1␤ to determine whether NF-B activation mediated by IL-1␤ involved this B element. IL-1␤ produced more luciferase activity in cells transfected with the wild type constructs than in cells transfected with the mutated one (Fig.  5B). These results indicated that IL-1␤ stimulation reproduced the results obtained with exogenous NF-B factor The reproducible 1.4-to 2-fold induction of CFTR promoter activity after cotransfection with p50/p65 expressing vectors or after IL-1␤ treatment is within the range of induction reported for the same promoter after stimulation with forskolin (12,13). The modest increase caused by Il-1␤ was abolished when the Ϫ1103 B element was mutated. Thus, the transfection data support the Northern blot analyses presented in Figs. 2 and 3. Therefore, we believe that this inducibility of the CFTR promoter could be biologically important and conclude that the Ϫ1103 B element of the CFTR takes part in the induction of CFTR mRNA production in response to IL-1␤ and in the subsequent NF-B activation in Calu-3 cells. DISCUSSION The development of an inflammatory response is a complex biological process that involves many changes in gene expression in populations of interacting cells, all with different time courses. In the microenvironment of the inflammation site, this complex cascade of events produces extracellular signals, such as IL-1␤ and TNF␣, that tightly control the changes in gene expression. This response helps to restore the equilibrium disturbed by the initial injury (bacterial, viral, or parasitic infection), but excessive production of inflammatory mediators may have negative effects and can lead to the destruction of the damaged tissue. Such a failure to modulate the inflammatory response occurs in cystic fibrosis. The abundance of pro-inflammatory cytokines in the airways of CF patients reflects the dramatic lung inflammatory injury that is often lethal in cystic fibrosis. Several studies have focused on the effect of cytokines in the inflamed CF airways on the expression of the CFTR (6,7). These studies have shown that CFTR expression is modulated by inflammatory signals, suggesting that CFTR contributes to the change in cell functions caused by the inflammatory stress. We have now obtained evidence that IL-1␤ stimulates the production of CFTR mRNA in a dose-, time-, and NF-Bdependent manner in a pulmonary cell line derived from the serous cells of submucosal glands. The EMSA and promoterreporter gene transfection assays indicate that IL-1␤ causes NF-B to bind to the CFTR 5Ј-flanking region, leading to increased CFTR transcription. This is of particular interest, because molecules, such as forskolin, that increase intracellular cAMP were the only extracellular stimuli that increased CFTR transcription until now. The cells were incubated with IL-1␤ for close to 20 h in most experiments, so that the observed effect on CFTR expression reflects primary and possibly secondary Il-1␤-induced transduction pathways.
IL-1␤ and TNF␣ have broadly overlapping, often synergistic, effects on cell functions. Not surprisingly, these cytokines also share signaling pathways, such as those that activate the NF-B transcription factor and the p38, JNK, and ERK MAP kinases. Our data for Calu-3 cells show that TNF␣ and IL-1␤ stimulate the expression of CFTR mRNA but to different extents. Surprisingly, the two cytokines have opposing effects in HT-29 cells, TNF␣ decreases CFTR mRNA expression, whereas IL-1␤ increases it. This finding suggests that IL-1␤ involves signaling events that are not shared with TNF␣. One possible explanation is that TNF␣ and IL-1␤ signaling pathways stimulate different signals and/or activate the same signal but confer on it different properties under some conditions.
Studies on the IL-1␤ signaling pathway have often focused on the complex cascade of intracellular molecular events that converge ultimately on the activation of other transcription factors, including AP-1 (18) and C/EBP (19). However, the IL-1 receptor-mediated signaling pathway also activates the NF-B transcription factor (20,21), which is of central importance to immune and inflammatory responses. Our findings provide evidence that the IL-1␤-induced expression of the CFTR is mediated by NF-B activation and binding to the CFTR 5Јflanking region. Nonetheless, activation of NF-B does not appear to be sufficient, by itself, to activate CFTR transcription, as indicated by the effects of TNF␣ in HT-29 and Calu-3 cells. Also, TNF␣ is a strong activator of NF-B in both cell lines (data not shown), it inhibits CFTR mRNA expression in HT-29 cells and only slightly stimulates CFTR expression in Calu-3 cells. These findings suggest that regulation of CFTR transcription by IL-1␤, which we unmasked in Calu-3 cells, involves different IL-1␤-induced transcription factors in addition to NF-B. This is supported by the presence of several putative binding sequences for IL-1-induced transcription factors in the CFTR 5Ј-flanking region, including two AP-1 response elements in the vicinity of the Ϫ1103 B element (13). There is strong evidence for interactions between NF-B and bZIP transcription factors, such AP-1 and C/EBP, influencing the ability of NF-B to regulate gene expression in a selective manner (22)(23)(24)(25)(26). Thus, interactions between these factors could lead to a synergistic activation of the CFTR transcription.
The subtle difference between the Ϫ1103 B element and the consensus B motif probably explains the slight difference in NF-B binding affinity, between the Ϫ1103 B element and the canonical B element from the immunoglobulin kappa light chain enhancer (in vitro assay, data not shown). Nevertheless such variant NF-B binding sites have been found to be important in gene transcription (26), because protein⅐protein interactions with other transcription factors are believed to stabilize their binding to DNA. Thus, interactions between NF-B and other IL-1␤-induced transcription factors could stabilize NF-B binding to the CFTR 5Ј-flanking region.
The increased CFTR expression in response to NF-B activation could be of pathophysiological significance in cystic fibrosis, because it was recently demonstrated that NF-B is endogenously activated in CF cells. The endogenous activation of transcription factor NF-B has been correlated with the constitutive inflammatory response in CF cells (27). It is suggested that the presence of the F508del-mutated CFTR proteins in the cells produces an endogenous stress that results in increased NF-B activation and thus an increased production of endogenous pro-inflammatory cytokines. The present study shows that the activation of NF-B can also lead to increased CFTR activity in airway epithelial cells. Thus, the F508del protein could exert a positive influence on its own synthesis in FIG. 5. Activities of control and mutated CFTR promoter in Calu-3 cells. A, Calu-3 cells were transiently transfected using 1.4 g of (Ϫ2150/ϩ52)pGL3 or mutated (Ϫ2150/ϩ52)pGL3, or the pGL3 basic vector (control) in combination with 0.6 g/well CMV-driven expression vectors for p50 (0.2 g/well) and for RelA (0.4 g/well), or with 0.6 g/well of the empty CMV vector alone to keep the concentration of transfected DNA constant. The results are expressed as the ratio of the luciferase activities in the (Ϫ2150/ϩ52)pGL3 constructs to the activities in the control constructs. Promoter activity is given as the mean of triplicate assays for one of two experiments. The parentheses show the -fold increase in luciferase activity in cells transfected with p50/RelA NF-B plasmids compared with the luciferase activity in the cells transfected with the reporter constructs without p50-and RelA-expressing vectors. B, Calu-3 cells were transiently transfected as previously with the same pGL3 vectors in combination with a CMV-␤ gal vector to normalize for transfection efficiency. 36-h post-transfection cells were stimulated with IL-1␤ for 16 h in a medium deprived of serum. Promoter activity is expressed as the ratio of the luciferase/␤galactosidase activities in the (Ϫ2150/ϩ52)pGL3 constructs to the activities in the control constructs. Data are expressed as the means Ϯ S.E. of three independent experiments, each in triplicate. **, p Ͻ 0.005, using Student's t test for unpaired values. some conditions by activating the NF-B system.