INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is characterized by chronic airway inflammation, with infiltration of eosinophils, lymphocytes, and monocytes/macrophages, and is associated with the increased expression of several inflammatory proteins, including cytokines, enzymes, and adhesion molecules in the airways (1, 2). The molecular regulatory pathways involved in the induction of chronic cytokine expression and the recruitment and activation of inflammatory cells in asthma are not well understood, but there is an increasing recognition that these processes involve increased transcription of inflammatory genes, and that this is regulated by transcription factors (3). One such transcription factor, nuclear factor-kappa B (NF-B), regulates the expression of a wide range of genes involved in immune and inflammatory response (4). The prototypic NF-B heterodimer consists of the p50 and p65 (Rel A) subunits, but other complexes have been described with different members of the Rel family modulating DNA binding and transactivation. In its inactive state, the NF-B dimer is present in the cytosol, where it is bound to an inhibitory protein, I-B (7). Activation of NF-B by several stimuli, including cytokines, reactive oxygen species, and microorganisms, induces the release and degradation of the inhibitory protein I-B from the dimeric complex (4, 7), followed by translocation of NF-B to the nucleus (4, 8, 9). The p50/p65 heterodimer is largely responsible for the inducible transcription-activating potential of NF-B (10). In synergy with other transcription factors, nuclear-translocated NF-B rapidly initiates the transcription of genes that have high-affinity binding motifs for NF-B in their enhancer or promoter regions (11).

Whether NF-B plays a role in asthma is not known, but several lines of evidence suggest that it may play a pivotal role. A distinct profile of inflammatory proteins, particularly cytokines and enzymes, show increased expression in asthmatic airways, including the cytokines tumor necrosis factor- (TNF-), interleukin-1 (IL-1), IL-4, and IL-5; the chemotactic cytokines (chemokines) regulated on activation, normal T-cell expressed and secreted (RANTES), macrophage chemotactic protein-1 (MCP-1), and macrophage inflammatory protein-1 (MIP-1); granulocyte-macrophage colony stimulating factor (GM-CSF); the inflammatory enzymes inducible nitric oxide synthase (iNOS) and inducible cyclooxygenase (COX-2); and the adhesion molecules intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) (12). The genes for these proteins are regulated by NF-B. Activation of NF-B by proinflammatory cytokines, such as TNF- and IL-1, has been demonstrated in various cells (4, 8), including human lung and blood monocytes, with electrophoretic mobility shift assays (19, 20). The observation that some cytokines regulated by NF-B, such as TNF- and IL-1, can also activate NF-B expression, indicates that this may represent a feed-forward amplifying loop that could form the basis for the persistence of the chronic inflammatory process in asthma. Although NF-B has been hitherto considered as mediating acute inflammatory responses, it has been shown to be chronically activated in chronic inflammatory conditions such as inflammatory arthritis and atherosclerosis (21, 22). Further support for the idea that NF-B may be important in asthma is that glucocorticosteroids, the most effective treatment for asthma, are potent inhibitors of this transcription factor (20, 23, 24).

In the present study, we hypothesized that an increase in NF-B expression and activation in asthma could be the basis for the chronic inflammation observed in this disease. We determined the binding of NF-B from nuclear extracts of cells obtained in induced sputum and in bronchial biopsy specimens from patients with mild asthma with electrophoretic-mobility-shift assays. In addition, we determined the immunohistochemical localization and protein expression of NF-B in bronchial biopsy specimens with an antibody that recognizes the activated p65 subunit of NF-B. Our study demonstrates the presence of activated NF-B in airway epithelial cells and macrophages of these asthmatic patients.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Three separate groups of asthmatic and normal subjects were recruited, with one group undergoing sputum induction and the other two undergoing fiberoptic bronchoscopy in order to obtain biopsy specimens for immunohistochemistry or for isolation of nuclear proteins with which to perform electrophoretic-mobility-shift assays (EMSAs). The characteristics of the three groups are summarized in Table 1. The study was approved by the Royal Brompton Hospital Ethics Committee, and all subjects gave informed consent. All asthmatic patients had stable asthma and had not been receiving inhaled or oral corticosteroid therapy for at least 1 yr, and were using only inhaled -adrenergic medication intermittently for relief of breakthrough symptoms. The provocative concentration of methacholine needed to cause a 20% decline in baseline FEV₁ (PC₂₀) was > 32 mg/ml in normal subjects and < 8 mg/ml in asthmatic patients. In order to determine the specificity of any changes in NF-B for asthma, we also studied induced sputum samples collected from five smoking patients with chronic obstructive pulmonary disease (COPD) who had no history of asthma and no airway response to a course of prednisolone at 40 mg/day for 1 wk (Table 1).

Sputum Induction

Sputum was induced with a 3.5% solution of hypertonic saline generated from an ultrasonic nebulizer (DeVilbiss UltraNeb 99; DeVilbiss Co., Heston, UK) and inhaled by tidal breathing, as previously described (25). Subjects were encouraged to cough deeply after 5 min and at 3-min intervals thereafter. The sputum obtained (3.5 ml ± 0.41 ml) was processed within 1 h of expectoration. The sputum was homogenized by the addition of 10% dithiothreitol (DTT; 10%; 200 µl) and Hanks' balanced salt solution (HBSS) (1.8 ml) without calcium or magnesium. The solution was vortexed to obtain maximal cell dispersion, and the cells were then washed in HBSS. Cell counts were performed and cytospin preparations were obtained and stained with May-Gunwald-Giemsa in order to obtain differential cell counts. The remaining cells were centrifuged (250 × g for 10 min at 4° C), and the cell pellet was used for extraction of nuclear proteins.

Fiberoptic Bronchoscopy

After an overnight fast, subjects were pretreated with nebulized albuterol (2.5 mg) and midazolam (5 to 10 mg intravenously). Oxygen (3 L/min) was administered via nasal prongs. Following the induction of local anesthesia of the upper airways and larynx with lidocaine (4%), a fiberoptic bronchoscope (BF10; Olympus Corp., Tokyo, Japan) was passed through the nasal passages into the trachea. Further lidocaine (2%) was sprayed into the lower airways, and three or four bronchial mucosal biopsy specimens were taken from segmental and subsegmental airways of the right lower and upper lobes, using a size 19 cupped forceps. Bronchial biopsy specimens were mounted in ornithyl carbamyl transferase (OCT) medium, snap-frozen in isopentane, and kept in liquid nitrogen for immunohistochemistry.

Electrophoretic-mobility-shift Assay

Nuclear proteins were extracted from cells obtained from sputum or from bronchial biopsy specimens through detergent lysis according to the method of Gough (26). Cell lysis was induced with Nonidet P-40 (NP-40; Sigma Co., Poole, Dorset, UK) at 4° C for 3 min, and soluble nuclear extracts were obtained after osmotic lysis of the nuclear envelope. Nuclear proteins (5 µg) were incubated for 1 h with ³²P-labeled double-stranded oligonucleotides (0.0175 pM) containing a tandem repeat of the consensus sequence for the NF-B-DNA binding site (5'-AGTTGAGGGGACTTTCCCAGGC-3'). We also used binding of the noninducible transcription factor Oct-1, with the consensus sequence (5'-TGTCGAATGCAAATCACTAGAA-3'), as an internal standard, expressing the amount of NF-B binding as a ratio of Oct-1 binding. DNA-protein complexes were resolved on a 6% nondenaturing polyacrylamide gel (37:1 acrylamide:bis-acrylamide) in 0.25× Tris-borate- ethylenediamine tetraacetic acid (EDTA) buffer. Gels were dried and autoradiographed at 70° C using Kodak X-OMATS film. The retarded bands were quantified with laser densitometry (Protein Databases Inc., New York, NY), and band-density measurements were then expressed as a ratio of NF-B to Oct-1 retarded bands. Specificity was determined by addition of a 100-fold excess of unlabeled double-stranded NF-B or Oct-1 oligonucleotide. To confirm the identity of the components of the retarded complexes, supershift experiments were conducted, using anti-p50 and anti-p65 human antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Normal rabbit IgG was used as a negative control. These antibodies were added to the reaction mixture 2 h before the addition of labeled oligonucleotide.

Immunostaining for NF-B

Frozen sections (6 µm) of the bronchial biopsy specimens were cut on a cryostat and placed on poly-L-lysine-coated microscope slides. Sections were fixed in cold 4% paraformaldehyde solution, followed by washing in Tris-buffered saline (TBS). The sections were treated with 0.1% saponin in TBS. Endogenous peroxidase activity was blocked by incubating slides in 1% hydrogen peroxide and 0.02% sodium azide in TBS for 1 h, followed by washing in TBS. After blocking with 10% normal swine serum, sections were incubated overnight at 4° C with a rabbit polyclonal anti-p65 antibody (Santa Cruz Biotechnology) at dilutions of 1:200 of a 100 µg/ml solution. This antibody recognizes an epitope corresponding to amino acids mapping within the N-terminal domain of p65. For the negative control sections, normal rabbit immunoglobulin (Dako, High Wycombe, UK) was used at the same protein concentration as the antibody. Following application of the primary antibody, we applied a biotinylated swine antirabbit antibody, (1:100 dilution; Dako) followed by the addition of peroxidase-conjugated avidin (1:100; Dako) for 45 minutes at room temperature. Chromogen fast diaminobenzidine (DAB; Sigma Co.) was used for 5 min, and the slides were counterstained in hematoxylin and mounted on mounting medium (DPX). For immunostaining of cytospins preparations obtained from sputum cells, we visualized the activated form of NF-B, using the glucose oxidase-DAB-nickel method (27).

As a positive control, we used the pattern of staining of A549 cells, a human lung epithelial-cell line (American Type Culture Collection, Rockville, MD) cultured to confluence in multichamber glass slides in Dulbecco's modified essential medium (DMEM) containing 10% fetal calf serum (FCS), L-glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 µg/ml), and amphotericin (95.6 mg/L). When subconfluent, cells were incubated for 24 h in DMEM without FCS prior to addition of IL-1 (1 ng/ml). After washing, the slides were air dried, fixed in 4% paraformaldehyde, and stained with the anti-NF-B antibody.

Immunostaining for Inflammatory Cells

In order to stain for the presence of inflammatory cells, the following antibodies were used on adjacent sections: for eosinophils, a mouse monoclonal antihuman antibody to major basic protein (MBP) (28); for T-cells, mouse monoclonal antihuman CD4 and CD8 antibodies (Dako); and for macrophages, a mouse monoclonal antibody to CD68 (Dako). Following application of the primary antibody, we applied a biotinylated rabbit antimouse immunoglobulin (1:100) and then peroxidase-conjugated avidin (1:200). Chromogen-fast DAB was used for 5 min, after which the slides were counterstained in hematoxylin and mounted.

Cell Counts

Counts of positive cells were made on all biopsy sections, and were divided according to whether the positive cells were in the airway epithelium or beneath the epithelium to a depth of 175 µm. Counts were made only in areas of intact epithelium. For NF-B-positive cells, the number of positive cells was expressed as a percentage of nucleated cells in the epithelium and in the subepithelium in at least 4 fields at ×400 magnification. For the inflammatory cells, the number of positive cells was expressed as the number per field. For the epithelium, one field was defined as a length of 175 µm, and for the subepithelium, one field was defined as an area of 175 µm². At least four fields were counted for each subject for the epithelium and subepithelium. All counts were made by an experienced observer unaware of the origin of the sections.

Data Analysis

Data are presented as mean ± SEM. Differences between normal and asthmatic subjects were assessed with the Mann-Whitney U test, and a value of p < 0.05 was taken as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NF-B-DNA Binding and Expression in Cells in Induced Sputum

Differential cell counts made on induced sputum showed a significant increase in the percentage of eosinophils recovered in sputum of asthmatic subjects as compared with normals (2.5 ± 0.7% versus 0.28 ± 0.12%, respectively; p < 0.01), whereas the percentages of macrophages and neutrophils were similar in both groups (76.6 ± 3.7% versus 76.0 ± 5.4% and 21.2 ± 4.0 versus 23.5 ± 5.5%, respectively). A significant increase in NF-B-DNA binding, expressed as a ratio with Oct-1-DNA binding, was observed in the nuclear protein extracts from sputum cells of asthmatic as compared with normal subjects, with a mean 2.5-fold increase (p < 0.003; Figure 1A). Excess cold double-stranded NF-B oligonucleotide, but not Oct-1 oligonucleotide, outcompeted NF-B for DNA-binding in these samples, indicating the specific nature of NF-B binding (Figure 1A). Supershift assay confirmed the presence of p50 and p65 subunits of NF-B (Figure 1B). There was a significant correlation between the percent eosinophils in induced sputum and the NF-B-DNA binding (r = 0.72; p < 0.02).

View larger version (57K): [in this window] [in a new window]   View larger version (55K): [in this window] [in a new window]   View larger version (43K): [in this window] [in a new window]   Figure 1.   (A) Representative electrophoretic-mobility-shift assays of nuclear proteins from induced sputum obtained from one normal and two asthmatic subjects, and two patients with COPD. The left panel shows increased NF-B-DNA binding, as demonstrated by the intensity of the retarded band (arrow). Specificity of the binding was confirmed by the addition of an 100-fold excess of cold double-stranded NF-B oligonucleotide (XS NF-B), which inhibited binding as compared with a sample having excess cold double-stranded Oct-1 oligonucleotide (Lane: XS Oct-1). The right panel shows similar DNA-binding on the noninducible transcription factor Oct-1 in these three subject groups. There was loss of the retarded Oct-1 band in the presence of excessive nonradiolabeled Oct-1 (XS Oct-1). (B) Supershift assay of nuclear proteins from induced sputum of normal (lanes 7 to 9; N) and asthmatic subjects (lanes 1 to 6; Asthmatic). Baseline NF-B binding is shown in lanes 1 and 7, and supershifted bands due to association with p50 are shown in lanes 2 and 8. A supershifted band with anti-p65 antibody is observed in lane 3. Lane 4 is a negative control incubated with preimmune serum with no supershifted band. Lanes 5 and 6 show the effect of excess unlabeled NF-B and Oct-1 oligonucleotides, respectively, and confirm the specificity of the retarded bands seen in the other lanes. (C ) Macrophages from induced sputum of a normal (N) and an asthmatic (A) subject, immunostained with an anti-p65 antibody and showing the activated form of p65 as black staining with the glucose oxidase-DAB-nickel method. Positive nuclear staining was stronger in the macrophages from the asthmatic subject. Original magnification: ×1,000.

Differential cell counts in induced sputum from COPD patients showed 76.2 ± 2.3% neutrophils, 2.2 ± 0.9% eosinophils, and 13.5 ± 3.6% macrophages. There was no increased NF-B- DNA binding in cells from induced sputum of patients with chronic obstructive airways disease as compared with controls (Figure 1A). Noninducible Oct-1-DNA binding activity was not different in controls, asthma patients, and COPD patients.

Immunostaining of cytospin preparations of cells from induced sputum revealed more intense nuclear staining in nuclei of macrophages from asthma patients than from normal subjects; there was also positive staining in the cytoplasm of macrophages from asthma patients (Figure 1C). No staining was observed in eosinophils. There was little staining of macrophages from patients with COPD (not shown).

NF-B-DNA Binding and Expression in Bronchial Biopsies

Because of low protein yields, pooled nuclear proteins for normal and asthmatic subjects (eight in each group; 2 µg per biopsy) were compared. There was greater NF-B-DNA binding, of 44.4%, in the pooled nuclear proteins from asthmatic than from normal subjects. We determined the components of the NF-B binding-protein complex by performing a supershift assay with antibodies to the p50 and p65 subunits of human NF-B. This assay confirmed the presence of p50 and p65 subunits (Figure 2). There was no supershift observed with antibodies to other members of the rel family, namely Rel b and c Rel (data not shown).

View larger version (72K): [in this window] [in a new window]   Figure 2.   Electrophoretic-mobility-shift-assay of the pooled extracts of biopsy specimens from eight asthmatic (2 µg protein from each; Asthmatic; lanes 2 to 7) and eight normal subjects (2 µg protein from each; N; lane 1), with a 44% greater expression in asthmatic (lane 2) than in normal subjects (lane 1). Lanes 3 and 4 show supershifted complexes with anti-p50 and anti-p65 antibodies, respectively, and lane 5 is a negative control incubated with normal rabbit IgG. Specificity of the binding is shown in lanes 6 and 7, reflecting incubation with excess cold NF-B and Oct-1 oligonucleotides, respectively. There is a clear reduction in the band in lane 6 with excess cold NF-B oligonucleotide.

Unstimulated A549 cells showed a small degree of staining in the cytoplasm, without any staining in the nuclei. However, after stimulation with IL-1, there was intense nuclear staining, with increased staining in the cytoplasm (Figure 3A). In the bronchial biopsies, there was no background staining in the absence of the antibody to the p65 component of NF-B. Staining was observed in airway epithelium, and also occasionally in submucosal cells in biopsies from normal and asthmatic subjects (Figure 3B). In normal subjects, 92.6 ± 2.1% of epithelial cells had cytoplasmic staining, as compared with 96.6 ± 1.3% in asthmatic subjects (p = NS). However, 45.1 ± 6.8% of epithelial cells expressed nuclear staining in asthmatic subjects, as compared with 20.7 ± 3.7% in normal subjects (p < 0.01). In the submucosa, the percentages of cells showing nuclear staining with the anti-p65 antibody were 2.7 ± 1.1% and 5.6 ± 2.0% (p = NS) in normal and asthmatic subjects respectively. However, there was an increase in cytoplasmic staining in submucosal cells (asthmatic subjects: 17.1 ± 3.2%, versus normal subjects: 6.3 ± 1.8%; p < 0.02). There was a significant increase in CD68⁺ macrophages in both the epithelium and subepithelium, and of MBP⁺ eosinophils, which were found only in the submucosa, in asthmatic subjects as compared with normal subjects (Table 2).

View larger version (109K): [in this window] [in a new window]   View larger version (136K): [in this window] [in a new window]   Figure 3.   (Upper panels) Localization of NF-B by immunohistochemistry. A549 cells at confluence were stained with anti-p65 antibody followed by peroxidase-conjugated avidin. There is very little staining at baseline (a), but after IL-1 stimulation (b) there is increased expression of p65 in cytoplasm, and particularly in the nuclei. Original magnification: ×400. (Lower panels) High-power photomicrographs (original magnification: ×1,000) of bronchial biopsies from a normal (N) and an asthmatic (A) subject, showing dark-brown positive nuclear staining with anti-p65 antibody in epithelial cells. Nuclei are counterstained blue with hematoxylin. The asthmatic epithelial cells show a greater percentage of positive nuclear staining. No nuclear staining but some cytoplasmic staining is observed in the submucosal cells.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our study demonstrates for the first time the presence of activated NF-B in asthmatic airways and inflammatory cells. We found that in cells obtained from induced sputum and in bronchial biopsy specimens from patients with mild asthma, there was an increased expression of NF-B as measured with the NF-B-DNA binding assay and immunohistochemical staining. However, cells from induced sputum from patients with COPD did not show evidence of increased activation. These studies indicate that NF-B activation is increased particularly in airway epithelial cells and submucosal cells, probably macrophages, in asthmatic airways. Because increased NF-B binding is associated with increased expression of the genes for many cytokines or enzymes of relevance in asthma, our findings suggest that NF-B may have an important role in maintaining the chronic inflammatory response in asthma.

We assessed the activation of NF-B by immunostaining, with an antibody to the p65 subunit. NF-B was expressed in human airways, particularly in airway epithelial cells and to a lesser degree in airway submucosal cells, with both cytoplasmic and nuclear localization. Asthmatic subjects had a significantly greater expression of p65 in the nuclei of airway epithelial cells than did normal subjects, indicating an overexpression of the translocated activated p65 from NF-B. In addition, there was a threefold increase in the number of submucosal cells expressing p65 NF-B in the cytoplasm in patients with asthma, indicating an overexpression of inactive NF-B. The increased number of epithelial cells showing nuclear staining was associated with an increased number of eosinophils and macrophages in the submucosa, as is typical of asthma (29). The anti-p65 antibody stained epithelial A549 cells in culture to only a minimal extent, but after stimulation with IL-1, there was increased staining in both the cytoplasm and nuclei of these cells, indicating that the antibody recognized the activated form of NF-B in the nucleus. Another cell showing increased NF-B binding in asthma is the macrophage. Cells recovered from induced sputum consist mainly of macrophages, and these showed increased DNA binding, as confirmed by the localization of nuclear p65 through immunostaining.

A redistribution of NF-B subunits from cytosolic to nuclear extracts is indicative of activation of NF-B (30). An increase in NF-B nuclear expression can be induced in cells in which the p65 subunit has been overexpressed (31). Thus, the constitutive cytoplasmic and nuclear staining observed particularly in the airway epithelium of normal subjects indicates some activation of NF-B, which may reflect the state of activity of differentiation of the epithelial cells, and also stimulation by external factors, such as ozone and other inhaled oxidants. Constitutive NF-B activity has been described in lymphoid cells such as mature B-cells and macrophages (32), and in certain subsets of neurons (33).

Supershift analysis of the retarded bands from asthmatic subjects showed the presence of both p50 and p65 subunits of NF-B in both retarded bands. The presence of the inducible and transcriptionally active forms of NF-B in increased amounts within samples from asthmatic subjects indicates that this may underlie the enhanced expression of certain cytokines and enzymes seen in asthmatic subjects. Although the increase in NF-B-DNA binding was small, we have previously shown that a twofold change in DNA-binding corresponds to an equivalent change in the number of DNA-binding species. Thus, a 44% increase in NF-B binding in bronchial biopsy specimens, and a 2.5-fold increase in macrophages from induced sputum, as we found, may be enough to cause an increase in gene induction (24).

The increase in the number of cells with nuclear localization of the p65 subunit of NF-B in patients with mild asthma indicates that there is activation of epithelial cells in these patients' airways. Because there was no significant increase in nuclear localization of p65 in submucosal cells, it appears that the state of enhanced activation was confined to the epithelium. The epithelium in asthma has been the site of enhanced expression of several proteins, including cytokines such as GM-CSF, RANTES, and MCP-1; enzymes such as iNOS and COX-2; and adhesion molecules such as ICAM-1 (12, 14, 16, 34). The genes encoding such proteins have NF-B-response elements in their enhancer or promoter regions, and an increase in activation of NF-B in the epithelium could form the basis for the overexpression of such diverse proteins in asthma, particularly in the epithelium. A previous study has reported increased expression of c-fos, a component of the transcription factor activated protein-1 (AP-1), in the epithelium of biopsy samples from asthmatic patients (35). It is possible that upregulation of a discrete number of transcription factors in asthma accounts for the profile of increased cytokine and enzyme expression in this disease.

The augmented activity of NF-B in asthma raises several issues about the pathogenesis of the chronic inflammation in asthma. First, the cause of the sustained increase in NF-B activity is unknown. NF-B binding is increased by several cytokines, such as TNF-, IL-1, and GM-CSF in a time- and dose-dependent manner in peripheral blood mononuclear cells and human lung tissue (19, 20). External stimuli such as viruses or oxidizing pollutants may also induce an increased expression of NF-B (7, 36, 37). The increased NF-B activity in the airway epithelium of asthmatic subjects could reflect an increase in sensitivity to a given stimulus in these subject's epithelial cells. The interaction of regulatory transcription factors and cytokines may represent an autoregulatory loop responsible for perpetuation of the chronic airway inflammatory process in asthma. Second, if NF-B activation leads to chronicity of airway inflammation, how may this occur? NF-B is usually not sufficient on its own to activate gene transcription, and interacts with other transcription factors that are activated within the cell at the same time, such as AP-1. Thus, NF-B, by interacting with a specific set of transcription factors, may lead to a pattern of expression of inflammatory genes characteristic of asthma.

We have shown that the chronic airway inflammatory process of asthma is associated with an increased expression of activated NF-B, which may play an important role in inflammatory and immune responses. Although we have shown that activated NF-B is present in asthma but not in COPD, it has also been found in other inflammatory conditions, such as inflammatory arthritis and atherosclerosis (21, 22). NF-B is therefore not specific for asthma, but may act in synergy with other transcription factors to induce maximal gene expression. The specificity of the inflammatory and cellular responses in asthma may depend on the interaction of NF-B with other transcription factors. Further studies are needed to characterize the roles of the NF-B family of transcription factors in this interaction in asthma, and to determine whether the levels of these transcription factors determine the severity and course of disease.