Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Asthma is a prevalent chronic pulmonary disease that is characterized by airway inflammation, bronchoconstriction, and airway hyperreactivity. Prominent features of asthmatic airway inflammation include mucus cell hyperplasia; epithelial desquamation; and the influx of leukocytes into the airway, including eosinophils and mast cells. Allergic asthmatics constitute a large proportion of patients with this disease, and hallmarks of allergic asthma are increased amounts of serum immunoglobulin (Ig) E and skin-test reactivity to aeroallergens (1). The immunologic basis of asthma has come under intense investigation (2). T lymphocytes, especially CD4+ T helper (Th) cells of the Th2 subset, are now known to be major participants in the development and maintenance of this disease (2). Th2 cells release a characteristic cytokine profile upon activation by relevant allergens and costimulatory molecules, including interleukin (IL)-4, -5, and -13 (3). The expression of Th2 cytokines is increased in asthma, and these cytokines contribute to many of the pathophysiologic features of allergic airway inflammation (2, 4). The factors responsible for a Th2 bias in asthma are unknown.

We have noted strain-related differences in the susceptibility of mice to aeroallergen-induced airway inflammation and hyperresponsiveness (5, 6). For example, we have extensively characterized airway responses in A/J and C3H/HeJ (C3H) mice, and found that allergen susceptibility in this model correlates with strain-dependent expression of the proallergic Th2 cytokine IL-4 (6). In these experiments, we found that A/J mice develop airway hyperreactivity, pulmonary eosinophilia, mucus cell hyperplasia, elevated serum IgE levels, and increased IL-4 and IL-5 production in response to allergen challenge (6, 8). Depletion of CD4+ T cells with monoclonal antibody prevented hyperreactivity and eosinophilia in this model, indicating that Th cells were critical for development of the allergic response (7). In a subsequent study, blockade of the IL-4 receptor completely prevented airway responses in antigen-challenged A/J mice, confirming the importance of Th2 cytokines in this model (8). In contrast to asthma-susceptible A/J mice, C3H mice are resistant to antigen- induced airway inflammation and hyperresponsiveness (6). Interestingly, administration of anti-IL-12 antibodies converted resistant C3H mice into asthma-susceptible mice, and led to the production of Th2 cytokines (6). This suggests that airway inflammation is suppressed in asthma- resistant C3H mice in part due to IL-12-dependent inhibition of Th2 cytokine gene expression.

Dysregulated IL-4 production can explain many of the pathophysiologic features of airway inflammation in human subjects with allergic asthma and in antigen-challenged A/J mice. First, IL-4 is one of the major factors driving the differentiation of Th2 cells from naive precursors (3). Th2 cells are a major source of cytokines, such as IL-13, that contribute to allergic airway inflammation and hyperreactivity (9). Second, IL-4 induces IgE isotype switching in B cells, and IgE levels are known to correlate with the incidence of asthma in epidemiologic surveys (1). Third, IL-4 can activate airway epithelial cells (10). Fourth, the recruitment of lymphocytes to the airway is dependent upon T-cell expression of IL-4 (11). In addition, by virtue of its ability to induce chemokine and adhesion molecule expression (12, 13), IL-4 can also enhance the recruitment of other inflammatory leukocytes, such as eosinophils. Finally, polymorphisms in the IL-4 promoter are associated with asthma severity in some human subjects (14). Together, these results suggest that the susceptibility to allergen-induced airway inflammation and hyperresponsiveness involves dysregulated expression of the IL-4 gene. We hypothesized that by studying the factors responsible for enhanced IL-4 gene expression in A/J mice, we would gain insights into IL-4 dysregulation in allergic asthma.

IL-4 gene expression is now known to be regulated at the level of transcription (15). Experiments using transgenic mice expressing reporter genes driven by different IL-4 promoter constructs have shown that the proximal IL-4 promoter is expressed in a tissue- and lineage-specific manner in lymphocytes and Th2 cells (16). Additional regulatory elements may be needed for maximal expression of the IL-4 gene, including still poorly characterized elements in the IL-4/IL-13 intergenic region (17). Many transcription factors are known to bind to the promoter of the IL-4 gene and regulate its expression (15, 18). Within the IL-4 promoter are at least six binding sites for the nuclear factor of activated T cells (NFAT) (19, 20). These purine-rich sites, known as P elements, bind NFAT as well as other associated transcription factors such as c-Maf (21). The two most proximal P elements, P0 and P1, appear to be particularly critical for the induction and Th2 restriction of IL-4 gene expression (16).

Of the four known NFAT proteins, NFATp and NFATc predominate in mature T cells (see Reference 22 for review). NFAT proteins are structurally homologous (especially in the DNA-binding domain), and dephosphorylation by the calcium-activated phosphatase calcineurin is sufficient to induce their rapid nuclear translocation. Experiments using knockout mice revealed strikingly different roles for individual NFAT proteins in regulating immunity. Although NFATc knockout animals died in utero secondary to cardiac valve agenesis, chimeric Rag1-deficient mice in which T cells were reconstituted from NFATc-deficient precursors were unable to produce IL-4 or other Th2 cytokines (23). Disruption of the NFATp gene, conversely, resulted in lymphoproliferation and increased production of Th2 cytokines, including IL-4, -5, and -13 (24). Thus, NFATp and NFATc appear to mediate qualitatively distinct immune responses.

In the present report we show that splenocytes stimulated ex vivo with a nonspecific mitogen, concanavalin A (Con A), recapitulated the airway cytokine phenotype of allergen-challenged A/J and C3H mice. We hypothesized that strain-dependent differences in IL-4 production between A/J and C3H mice involved differential regulation of IL-4 gene transcription. This was supported by the observation that Con A induced de novo IL-4 gene transcription in stimulated splenocytes, as determined by nuclear run-on analysis. This could be due to polymorphisms in IL-4 promoter regulatory sequences, or to the activation of transcription factors necessary for IL-4 promoter transactivation in a strain-dependent manner. To test these hypotheses, we first sequenced the IL-4 promoter using genomic DNA isolated from both mouse strains. In addition, we extracted DNA-binding proteins from splenocytes isolated from A/J and C3H mice and analyzed their ability to interact with relevant IL-4 promoter sequences using electrophoretic mobility shift assays (EMSAs). Here we report that NFATc is preferentially activated in splenocytes isolated from A/J mice.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Male A/J and C3H mice, 6 wk old, were obtained from the Jackson Laboratory (Bar Harbor, ME). Animals were housed under laminar flow hoods in an environmentally controlled, pathogen-free animal facility for the duration of experiments. They were given free access to tap water and ovalbumin-free rodent chow. The studies reported here conformed to the principles for laboratory animal research outlined by the Animal Welfare Act and the Department of Health, Education and Welfare (National Institutes of Health) guidelines for the experimental use of animals.

Splenocyte Isolation

Spleens were isolated aseptically in a laminar flow hood. Cells were teased apart between two sterile slides and resuspended in 5 ml RPMI plus L-glutamine (BioFluids, Camarillo, CA). Cells were washed twice with RPMI medium and pelleted at 1,800 rpm, 4°C, for 8 min to remove cellular debris. Cells were plated in RPMI supplemented with 2 mM L-glutamine (Sigma, St. Louis, MO), 10% heat inactivated fetal bovine serum (BioFluids), 50 mM 2- mercaptoethanol (Sigma), 50 mg/ml gentamycin sulfate (BioFluids), and 0.1 M N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid (Hepes) (BioFluids) at 1.25 × 10⁶ cells/ml, and treated with either 4 µg/ml Con A (Sigma) or phosphate-buffered saline (PBS) (BioFluids) control for varying time periods at 37°C, 5% CO₂.

Quantitation of IL-4 Protein Levels in Splenocyte Cultures

Splenocytes were isolated from naive A/J and C3H mice (n = 5 mice/group) as described earlier and were cultured in the presence of Con A (4 µg/ml). Supernatants were harvested 24 or 48 h after stimulation. IL-4 protein levels were measured in duplicate samples (100 µl) from each animal by sandwich enzyme-linked immunosorbent assay (ELISA) as previously described (7). ELISAs were conducted using matched antibody pairs obtained from PharMingen (San Diego, CA) according to the manufacturer's instructions. Optical density readings of samples were converted to picograms per milliliter using values obtained from standard curves generated with varying concentrations of mouse recombinant IL-4 (5 to 2,000 pg/ml). The limit of detection was 15 pg/ml.

Nuclear Run-On Analysis

Splenocytes were pooled from three A/J and three C3H mice to obtain 50 × 10⁶ cells. Cells were then stimulated with either Con A (4 µg/ml) or PBS control for 48 h. Nuclei were then isolated according to methods previously described (25). In vitro transcription in the presence of [³²P]uridine triphosphate (200 Ci/mmol; Amersham, Piscataway, NJ), was performed for 30 min at 30°C. Nuclear RNA was isolated using the RNeasy kit (Qiagen, Valencia, CA), resuspended in 100 µl dH₂O, and denatured for 5 min at 80°C. RNA was then hybridized for 4 h at 68°C to a previously prehybridized nitrocellulose membrane (Bio-Rad, Hercules, CA), on which complementary DNA corresponding to the coding regions of murine IL-4 and IL-13 (15 µg each) and -actin genes (2 µg) were immobilized using a slot-blot apparatus (BioRad, Hercules, CA). Membranes were first washed twice for 15 min each with 2× saline sodium citrate (SSC) plus 0.1% sodium dodecyl sulfate (SDS) at room temperature, then twice with 0.2× SSC plus 0.1% SDS at 55°C. After incubation, membranes were exposed to a phosphoimager plate (Packard Instruments, Meriden, CT) for 3 d. Hybridization was determined using QuantCount software.

Genomic DNA Isolation and DNA Sequencing

Genomic DNA was isolated from whole blood obtained from A/J and C3H mice using the Qiagen Midi Kit (Qiagen) according to manufacturer's instructions. IL-4 promoter sequences spanning nucleotides 717 to +62 relative to the transcription start site were amplified from 100 ng DNA using the polymerase chain reaction (PCR) and the following primers: 5'TGCACATAGATACACACATGATCACAT3' (717 to 690) and 5' TATATTCTAGAAATAGCTCTGTGCCGTCAGT3' (+62 to +32).

PCR products were then ligated into the SrfI site of PCR-Script (Stratagene, La Jolla, CA) using standard techniques, followed by DNA sequence analysis of both coding and noncoding strands using the dideoxy method. Genomic DNA isolated from at least three animals per strain were analyzed and compared with published sequences for both BALB/c and C57Bl6 mice.

Nuclear Protein Extraction and EMSA

After treatment, 1.0 × 10⁷ cells were washed twice in 10 ml cold Tris-buffered saline (TBS) (BioFluids), and pelleted at 1,600 rpm for 5 min at room temperature. Cells were transferred to 1.5-ml microcentrifuge tubes and washed again in 1.0 ml TBS. Pellets were then resuspended in 400 µl cold buffer A (10 mM Hepes, 1 mM dithiothreitol [DTT], 0.1 mM ethylenediaminetetraacetic acid [EDTA], 0.1 mM ethyleneglycol-bis-(-aminoethyl ether)-N,N'-tetraacetic acid [EGTA], 10 mM KCl, 0.1 mM phenylmethylsulfonyl fluoride [PMSF], 1 mg/ml leupeptin, and 1 mg/ml aprotinin) and incubated on ice for 15 min. Cells were lysed using 25 µl of 10% Nonidet P-40 (Sigma) and vortexed, and nuclear pellets were recovered by centrifugation at 14,000 rpm for 30 s. The nuclear pellet was resuspended in 50 µl cold buffer C (20 mM HEPES, 1 mM DTT, 1 mM EDTA, 1 mM EGTA, 420 mM KCl, 0.1 mM PMSF, 1 mg/ml leupeptin, and 1 mg/ml aprotinin) and incubated on ice for 15 min. Debris was removed by centrifugation at 14,000 rpm at 4°C for 5 min, and aliquots were stored at 70°C until use. Nuclear protein extracts (4 µg/lane) were incubated with 40% glycerol and either poly dIdC or dGdC (0.2 µg) (Pharmacia, San Diego, CA), as indicated, for 20 min on ice. [³²P]-labeled probes (3 × 10⁴ counts per min/lane) corresponding to the IL-4 promoter elements were then added to cell extracts and incubated for 20 min at room temperature. The sequences of the oligonucleotides used were: 5'-ATTGCTGAAACCGAGGGAAAATGAGTTTACATTG-3' (P0, 69 to 36) and 5'-TGAGTTTACATTGGAAATTTTCGTTACACCAGATTG-3' (P1, 92 to 60).

In competition experiments, extracts were preincubated for 20 min at room temperature with 50-fold molar excess of the P1 probe, as well as the following commercially available consensus oligonucleotides: AP1, 5'-CGCTTGATGAGTCAGCCGGAA-3'; and AP2, 5'-GATCGAACTGACCGCCCGCGGCCCGT-3'.

In supershift assays, cell extracts were incubated with anti-NFATp (Santa Cruz Biotechnology, Santa Cruz, CA), anti-NFATc (Affinity BioReagents, Golden, CO), or isotype control (mouse IgG; Sigma) antibodies for 10 min on ice before addition of probe. Extracts were analyzed by 5% polyacrylamide gel electrophoresis run at 150 V in 0.5× TBE, followed by autoradiography.

Western Blotting

Nuclear and cytoplasmic proteins were separated on a 7.5% polyacrylamide gel at 100 volts for 90 min. Proteins were then transferred to a polyvinylidene difluoride membrane (Bio-Rad) for 1 h at 90 V. Membranes were blocked overnight at 4°C, then probed with anti-NFATc antibodies (7A6, 1:2,000; Affinity Bioreagents) for 1 h at room temperature and washed twice in TBS plus 0.1% Tween 20 for 5 min each wash. Membranes were then incubated with antimouse IgG horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. Blots were developed using the Amersham ECL Plus chemiluminescent detection kit.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Increased IL-4 Production Is an Inherent Property of A/J Lymphocytes

We have previously reported that allergen challenge induces a significant increase in bronchoalveolar lavage (BAL) IL-4 levels in A/J but not C3H mice (6). This correlated with the development of airway inflammation and hyperresponsiveness (6). To determine whether increased IL-4 production was an inherent property of A/J lymphocytes, we isolated splenocytes from each strain and studied their ability to secrete IL-4 protein when stimulated ex vivo with the nonspecific agonist Con A. Figure 1 shows that splenocytes isolated from A/J mice produced significantly more IL-4 after stimulation with Con A than did C3H splenocytes, particularly at later time points. Thus, purified splenocytes stimulated with a nonspecific agonist exhibited the same phenotype as that observed in allergen-challenged animals. To exclude the possibility that Con A caused a selective expansion of A/J CD4+ T cells, we analyzed resting and Con A-stimulated splenocytes from both strains using immunofluorescence and flow cytometry. These experiments revealed that the fraction of CD4+ T cells was the same in A/J and C3H splenocytes both before and after Con A treatment (data not shown). These data are consistent with our previous observation that Con A-induced splenocyte proliferation did not differ between A/J and C3H mice (26). Thus, a selective increase in CD4+ T cells in A/J mice did not explain increased IL-4 production.

View larger version (20K): [in this window] [in a new window]   Figure 1.   Strain-related differences in splenocyte production of IL-4 protein. Con A-stimulated levels of IL-4 were measured in the supernatants of A/J and C3H splenocytes by ELISA as described (see MATERIALS AND METHODS). Results shown are the means ± standard error of the mean of IL-4 levels from each of 5 mice per group. *Values significantly different from those of A/J mice (P < 0.0004).

IL-4 Gene Expression in Con A-Stimulated Splenocytes Is Controlled at the Level of Transcription

To determine whether IL-4 expression in Con A-stimulated lymphocytes required de novo gene transcription, we isolated nuclei from A/J and C3H splenocytes and studied their ability to initiate transcription of the IL-4 gene using nuclear run-on assays. In these experiments we compared the expression of IL-4 and its close congener IL-13 to -actin as a control in resting and activated cell nuclei. Figure 2 shows that upon activation with Con A, transcription of both the IL-4 and IL-13 genes was induced. A weak signal was also detected using C3H splenocytes (Figure 2, right). Therefore, in this model, IL-4 gene expression was regulated at the transcriptional level.

View larger version (30K): [in this window] [in a new window]   Figure 2.   Nuclei were obtained from A/J splenocytes (left panel) or C3H splenocytes (right panel) either unstimulated or stimulated for 48 h with Con A. In vitro transcription was performed in the presence of [32P]deoxyuridine triphosphate for 30 min. RNA was then isolated and equal amounts were hybridized to a nitrocellulose membrane containing immobilized cDNA specific for the coding regions of the IL-4, IL-13, or -actin genes. Upon stimulation with Con A, de novo IL-4 and IL-13 gene transcription was detected in both A/J and C3H splenocytes. Transcription of -actin control was detected in nuclei isolated from both unstimulated and stimulated splenocytes. Results are representative of one additional experiment.

IL-4 Promoter Sequences Are Identical Between Strains

One potential mechanism for differences in IL-4 transcription between A/J and C3H splenocytes could be polymorphisms in regulatory DNA sequences. Differences at this level could lead to enhanced IL-4 gene expression by altering the affinity of transcription-factor binding sites. To exclude this possibility, we sequenced more than 700 base pairs (bp) spanning the proximal IL-4 promoter region and transcription start site, using genomic DNA isolated from both strains (see MATERIALS AND METHODS). No differences between these two strains were detected in these experiments (Table 1).

A Novel Complex Binds the P0 Element in Both A/J and C3H Mice

We next speculated that enhanced IL-4 production in A/J mice may be due to increased expression and/or activation of transcription factors necessary for IL-4 promoter transactivation. The proximal IL-4 promoter contains six NFAT binding sites (termed the P elements), and NFAT proteins are known to be critical regulators of IL-4 gene expression. We therefore extracted nuclear proteins from resting and activated splenocytes isolated from A/J and C3H mice, and used EMSAs to analyze their ability to bind different IL-4 promoter P elements. Because the P0 and P1 sites are particularly important in regulating IL-4 expression in T lymphocytes and Th cell subsets, we initially focused our attention on these two elements. Figure 3 shows the results of an EMSA using nuclear extracts isolated from A/J and C3H splenocytes stimulated with Con A for 48 h and a radiolabeled oligonucleotide encompassing the P0 element. This oligonucleotide supported the binding of both a constitutive nuclear protein (Figure 3, II) and an activation-induced complex (Figure 3, lanes 2 and 4, asterisk). Neither complex was affected by antisera directed against factors known to bind in this region, including NFATp, NFATc, C/EBP, Maf, or YY1 (data not shown). Note, however, that binding of these factors was not different between the two strains.

View larger version (60K): [in this window] [in a new window]   Figure 3.   Nuclear extracts isolated from A/J and C3H splenocytes either unstimulated or after stimulation for 48 h with Con A were incubated with a radiolabeled oligonucleotide probe from the P0 IL-4 NFAT element and analyzed by EMSA. The P0 element supported binding of both a constitutive factor (complex II) and an activation-induced complex (*) that formed equally well using extracts isolated from both strains. Neither complex was affected by antisera directed against factors known to bind in this region, including NFATp, NFATc, Maf, C/EBP, and YY1. No strain-specific or activation-induced complexes were detected using extracts from cells stimulated at earlier time points (data not shown).

Strain-Dependent Binding of a Transcription Factor Complex to the P1 Element Occurs after 48 h

We next analyzed nuclear extracts isolated from resting and activated splenocytes using a radiolabeled oligonucleotide containing the P1 element in EMSA. Two constitutively binding factors (complex I and II) were detected in these experiments (Figure 4A). When analyzing extracts obtained from splenocytes stimulated for 1 h with Con A, an additional band (activation-induced complex [AIC] 1) was also detected, but AIC1 was expressed similarly in splenocytes from the two strains (Figure 4A, lanes 2 and 4). In addition, the formation of complex I was also slightly enhanced upon cell stimulation (Figure 4A, lane 2).

View larger version (61K): [in this window] [in a new window]   View larger version (60K): [in this window] [in a new window]   View larger version (38K): [in this window] [in a new window]   View larger version (56K): [in this window] [in a new window]   View larger version (58K): [in this window] [in a new window]   Figure 4.   Nuclear extracts were isolated from A/J and C3H splenocytes either unstimulated or after stimulation with Con A (see MATERIALS AND METHODS), and then analyzed by EMSA using a radiolabeled oligonucleotide probe from the P1 IL-4 NFAT element. (A) Two constitutively forming complexes are indicated (I and II). An additional low-mobility complex also formed using extracts isolated from cells stimulated for 1 h with Con A (AIC1), although it was equally expressed in both strains. (B) Two constitutively bound complexes (II and III) were again detected in splenocyte extracts from both strains. In contrast, a low-mobility complex preferentially formed upon activation of A/J splenocytes with Con A for 48 h (AIC48, lane 2). This experiment was performed using poly dIdC as competitor DNA (see MATERIALS AND METHODS). (C) Splenocyte nuclear proteins were extracted as in B and analyzed by EMSA in the presence of poly dIdC or poly dGdC as indicated. Note that AIC48 again preferentially formed using A/J nuclear extracts (lanes 1 and 3). The formation of complex III in mice was inhibited in the presence of poly dIdC (lanes 3 and 4). This experiment was performed using different litters of each strain than in B. (D) AIC48 is specific for the IL-4 P1 element because its binding was completely inhibited in the presence of excess self (lane 2) but not unrelated oligonucleotides (lanes 3 and 4). (E) AIC48 contains predominantly NFATc. Nuclear extracts were isolated from Con A-stimulated splenocytes and analyzed as in A, in the presence of antibodies directed against NFATc, NFATp, or species-matched control antisera. AIC48 is almost completely supershifted by the anti-NFATc antibody (lane 2, s.s.), and is therefore labeled NFATc. Neither this complex nor complex II was affected by control antibody anti-NFATp. The NFATc antibody alone did not react with the oligonucleotide (not shown). (s.s., supershift; mIgG, mouse IgG control; gIgG, goat IgG control).

Kinetics of IL-4 protein production indicated that A/J splenocytes preferentially express more IL-4 protein at later time points than do C3H splenocytes (Figure 1). We therefore obtained nuclear extracts from splenocytes stimulated for 48 h with Con A, and analyzed them by EMSA with the P1 oligonucleotide probe. A slowly migrating, activation-induced complex (AIC48) was detected in these experiments (Figure 4B). Interestingly, AIC48 was preferentially detected using nuclear extracts isolated from A/J mice, and was barely detectable (Figure 4B) or undetectable (Figure 4C) in extracts obtained from C3H mice studied under identical conditions. This complex was not significantly affected by the composition of competitor DNA in the EMSA reaction (Figure 4C). Complex III, however, was inhibited in the presence of poly dIdC (Figure 4C), although it formed equally well using splenocyte extracts from both strains. Preferential expression of AIC48 in stimulated A/J splenocytes was observed using multiple different litters of both strains (Figures 4B and 4C, and data not shown). The complex was specific for the IL-4 P1 element because it was completely inhibited in the presence of excess unlabeled P1 oligonucleotide (Figure 4D, lane 2) but not unrelated competitors (Figure 4D, lanes 3 and 4).

AIC48 Contains NFATc

To determine the identity of AIC48, stimulated and unstimulated splenocyte nuclear extracts were incubated with anti-NFATp, anti-NFATc, or isotype-matched control antibodies, and analyzed by EMSA. As shown in Figure 4E, anti-NFATc antibodies completely supershifted this complex (lane 2), whereas neither anti-NFATp nor isotype control affected its formation (lanes 3-5). These results indicate that the nuclear expression and DNA-binding ability of NFATc is preferentially induced in A/J splenocytes upon cell activation.

Nuclear Translocation of NFATc Is Increased in A/J Splenocytes

The increase in NFATc binding to the P1 element could result from increased cellular concentration of NFATc in A/J mice or from increased translocation of NFATc into the nucleus in activated cells. To distinguish these two possibilities, we examined NFATc protein expression in A/J and C3H cytoplasmic and nuclear fractions by Western blotting. As shown in Figure 5, NFATc was detected in both nuclear and cytoplasmic extracts after 48 h stimulation with Con A. The overall concentration of NFATc was similar between the two strains. However, nuclear translocation of NFATc was increased in A/J (Figure 5, lane 2) when compared with C3H (Figure 5, lane 4) mice. This provides a mechanism for increased NFATc binding to the IL-4 promoter (Figure 3).

View larger version (29K): [in this window] [in a new window]   Figure 5.   Equal amounts of nuclear proteins (lanes 1-4) or cytoplasmic proteins (lanes 5-8) isolated from A/J and C3H splenocytes either unstimulated or stimulated for 48 h with Con A were analyzed by Western blotting using a monoclonal anti-NFATc antibody. NFATc protein was detected after stimulation with Con A in both strains (even-numbered lanes), and the overall cellular concentration was similar between strains. However, nuclear translocation of NFATc was enhanced in A/J splenocytes (compare lanes 2 and 6) as opposed to C3H splenocytes (compare lanes 4 and 8). Two NFATc immunoreactive complexes of approximately 80 and 100 kD were detected (even-numbered lanes). Results are representative of three additional experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The molecular basis for enhanced Th2 cytokine gene expression in allergic diseases is not well understood. Growing evidence suggests that dysregulation of IL-4 gene transcription is a fundamental component of this process. We analyzed in detail the molecular pathways involved in the regulation of IL-4 transcription in a well-characterized mouse model of allergic asthma. In this model, asthma susceptibility in A/J mice is known to correlate with the production of Th2 cytokines, whereas C3H mice fail to induce Th2 cytokines and are resistant to allergen-induced responses. Here we report that the transcription factor NFATc is preferentially activated in splenocytes from asthma-susceptible A/J mice. Activated NFATc was identified using EMSAs and an oligonucleotide probe containing the IL-4 promoter P1 element. Importantly, this regulatory element is known to be critical for the induction and Th2- restriction of IL-4 gene expression. Activation of NFATc in A/J mice appears to involve enhanced nuclear translocation. Together with the observation that Th2 cytokine production was hindered in mice expressing NFATc-deficient T cells (23), our results suggest that NFATc is an important regulator of Th2-driven allergic immune responses.

The genetic basis for strain-dependent differences in Th phenotype is complex and involves differential expression of multiple gene products. We found that A/J and C3H splenocytes stimulated ex vivo with the nonspecific mitogen Con A expressed a similar cytokine profile to that observed in BAL fluids of antigen-challenged animals (Figure 1 and Reference 6). This suggested that enhanced Th2 cytokine production was an intrinsic property of A/J lymphocytes and was not due, for example, to differences in antigen presentation or cellular trafficking between the two strains. Using nuclear run-on analysis, we found that IL-4 is transcriptionally regulated in our model (Figure 2). Interestingly, we observed coordinate induction of IL-4 and IL-13 transcription after Con A stimulation using these assays (Figure 2). Growing evidence suggests that the expression of these cytokines is regulated by a common molecular mechanism (17). Studies are underway to investigate the transcriptional regulation of IL-13 in our model. We also detected ongoing cytokine gene transcription in stimulated C3H splenocytes (Figure 2). This was not surprising, given (1) the ability of C3H splenocytes to produce small quantities of IL-4 protein (Figure 1), and (2) the detection of some NFATc in C3H nuclear extracts by both EMSA (Figure 4) and Western analysis (Figure 5). Thus, although we cannot completely exclude a role for post-transcriptional regulation of IL-4, our data are consistent with a large body of literature highlighting the critical importance of the initiation of transcription in regulating IL-4 gene expression (15, 16, 18, 23, 24, 27).

We next investigated the molecular mechanisms controlling differential gene transcription upon cell activation. Differential expression of the IL-4 gene could involve preferential activation of IL-4-specific transcription factors or genetic alterations in regulatory DNA sequences that control IL-4 expression. In support of the latter possibility, the murine IL-4 promoter, like its human counterpart (14), has been reported to be polymorphic (Reference 28; see Table 1). However, after sequencing over 700 bp spanning the IL-4 transcription start site using genomic DNA isolated from A/J and C3H mice, we did not detect any differences in nucleotide sequence between the two strains. A lack of polymorphisms in the IL-4 promoter is consistent with our finding that antigen-induced airway hyperresponsiveness in progeny of A/J and C3H mice is not linked with markers in the region of chromosome 11 containing the IL-4 gene (36).

We used EMSAs to screen for transcription factors that were preferentially induced in nuclear extracts of A/J but not C3H splenocytes. Using the P1 element as an oligonucleotide probe, we detected activated NFATc after stimulation with Con A for 48 h when the preferential expression of IL-4 protein was most apparent (Figures 1 and 4B-4E). Interestingly, the transcriptional activity mediated by the P1 element was found to be highly induced in effector Th2 cells (16). When we studied the P0 element in EMSA, we observed the formation of an apparently novel activation-induced complex that was not related to factors previously shown to interact in this region (Figure 3). The identity of this factor, which was expressed equally in splenocytes from both strains, is currently unknown. Both the P0 and P1 oligonucleotides supported the binding of constitutive nuclear factors. Complex III (Figure 4C) appears to be related to HMGI/Y proteins which are known to interact with the P1 element, inhibit the binding of NFAT, and bind in a poly dIdC-sensitive manner (27). However, this and the other constitutive factors observed were expressed equally in splenocytes from both strains.

Induction of Th2 cytokine gene expression is a highly regulated process, involving the activation of ubiquitous and Th2-restricted transcription factors such as Stat6, GATA3, and c-Maf (21, 29). These factors appear to mediate distinct phases of this process. For example, Stat6 is required for modulation of chromatin structure near the Th2 cytokine gene locus (30), but not for direct transactivation of the full-length IL-4 promoter (31). Additionally, GATA3 was found to activate transcription driven by the IL-5 promoter, and not the IL-4 promoter, in co-transfection assays (32). This is supported by the recent observation that GATA3 expression was increased in the airway of atopic asthmatics, and correlated with IL-5 messenger RNA expression (33). The P0 oligonucleotide used in our experiments is analogous to the c-Maf response element (MARE) described by Ho and colleagues (21), although we did not detect binding of Maf proteins by EMSA in our experiments. Because Ho and associates detected the MARE using deoxyribonuclease footprinting and not EMSA, we cannot exclude a role for c-Maf in our model (21). Growing evidence points to the importance of NFAT proteins in regulating IL-4 gene expression in Th cells (22). On the basis of the data described in this report, we conclude that NFATc contributes to preferential IL-4 promoter activation in a naturally occurring, asthma-susceptible mouse strain.

NFAT comprises a family of transcription factors whose expression and function are regulated at multiple levels (22). The first level of regulation involves the control of NFAT gene expression. This is best studied in the case of NFATc, which is expressed constitutively at low levels in T cells but induced upon cell stimulation with calcium ionophores or by crosslinking CD28 (22). Second, NFAT proteins are regulated at the level of nuclear translocation by the calcium-sensitive phosphatase calcineurin. This process appears to be highly conserved between different NFAT species, involves unmasking of a nuclear localization sequence, and is inhibited by cyclosporin A (22). Third, nuclear export of NFAT also appears to be a highly regulated process (34). Finally, NFAT can be regulated by additional post-transcriptional modification. For example, there are multiple isoforms of NFATc which arise due to alternative splicing of a common gene product (35). Emerging data suggest that these isoforms posses distinct transactivation abilities (25).

In this report we used EMSA and Western blot analysis to detect differential expression of activated NFATc in splenocyte nuclear extracts obtained from A/J mice. Combining these two techniques allowed for the detection of a functional NFAT complex, and was motivated by the observation that nuclear translocation of NFAT does not always correlate with its DNA binding ability. However, in our experiments we found a close correlation between the nuclear translocation of NFATc as detected by Western blot and its ability to bind the IL-4 P1 element in EMSA. This argues that additional post-translational modification of NFATc is not required after nuclear translocation to achieve full DNA binding ability. On the IL-2 promoter, NFATc is thought to associate with nuclear cofactors to form a high-affinity DNA binding complex (22). However, the observations that (1) the IL-4 P1 element in particular is a high-affinity binding site capable of binding NFAT alone, and (2) AIC48 was completely supershifted by antibodies directed against NFATc (Figure 4E), argue against this possibility in the case of the IL-4 promoter.

The precise biochemical basis for the enhanced activation of NFATc in A/J mice is currently unknown. Future studies will be needed to determine whether this is due to increased activation of calcium-sensitive signaling pathways, reduced nuclear export, or altered structure of NFATc itself. In addition, studying the role of signals provided by other cell types in mediating this response represents a promising area for future research. Elucidating the mechanism of enhanced activation of NFATc in this model should further our understanding of Th2 responses and allergic diseases.