IL-12 suppresses proallergic Th2-type cytokine production and induces Th1-type cytokine production by peripheral blood T cells from subjects with allergic disease. The objective of the present study was to examine the relevance of these findings to target organ T cell responses in human asthma. Bronchoalveolar lavage (BAL) and PBMC were collected from atopic asthmatics 24 h after fiberoptic allergen challenge of a segmental bronchus. BAL T cells and PBMC were cultured with allergen in the presence of recombinant IL-12 or IFN-γ, and cytokines were measured in culture supernatants after 6 days. IL-5 production by BAL T cells and PBMC was inhibited by IL-12 and, to a lesser extent, by IFN-γ. IL-12 also induced IFN-γ production by BAL T cells and PBMC. The effects of IL-12 nor IFN-γ on IL-5 production could not be reversed by neutralizing anti-IFN-γ or anti-IL-12 mAbs, respectively. Thus, the effect of neither IL-12 nor IFN-γ appeared to be mediated through induction of the other cytokine. In situ hybridization revealed that approximately one-third of BAL T cells expressed mRNA transcripts encoding the IL-12R β2 subunit following allergen challenge. Thus, human T cells obtained from BAL during asthmatic late responses, like T cells in the peripheral circulation, remain susceptible to immunomodulation by IL-12. These findings raise the possibility that IL-12 may hold therapeutic potential in allergic diseases such as asthma.
Allergen-specific T cells from the peripheral blood of patients with allergic asthma or rhinitis are characterized by elevated expression of the Th2-type cytokines, IL-5, IL-4, and IL-13 (1, 2, 3, 4, 5), and mRNA encoding these cytokines is up-regulated within the target organ on allergen exposure (6, 7, 8). Of the Th2-type cytokines, IL-5 is considered to be of particular relevance to asthma pathogenesis by virtue of its multiple proeosinophilic properties. For example, IL-5 promotes eosinophil maturation, activation, survival (9, 10, 11), release from bone marrow (12), and enhanced responsiveness to chemokines such as RANTES (13) and eotaxin (14). In murine models, the allergen-induced asthmatic late response is dependent on IL-5 and eosinophils (15).
In vitro studies suggest that the ability of allergen-specific T cells to produce Th2-type or Th1-type cytokines can be regulated by exposure to cytokines such as IL-12, IFN-γ, and IL-4. Specifically, addition of IL-12 (and to a lesser extent IFN-γ) to freshly isolated peripheral blood T cells obtained from atopic subjects shifts the expression of cytokines away from a Th2-type profile and toward a Th1-type profile (16, 17). Conversely, addition of IL-4 results in a shift from Th1- to Th2-type cytokine production in vitro (18). The effects of IL-12 on Th2 cytokine-producing T cells that have undergone recent stimulation by Ag are more controversial: it has been reported that certain Th2-type T cells have the capacity, when activated and expanded in vitro (i.e., as T cell lines or clones), to become resistant to the cytokine-modulating effects of IL-12 when restimulated (17, 19, 20). This has been linked to loss of function in the IL-12 signal transduction pathway, as evidenced by a failure to phosphorylate STAT4 on tyrosine residues (19, 20), and more recently, through failure to express the β2 subunit of the IL-12R (IL-12Rβ2) (21, 22).
These observations raise the question of whether T cells present in human tissues exposed to allergen become refractory to IL-12. There is indirect evidence to suggest that IL-12 and IFN-γ may be able to suppress allergic inflammation locally. First, numbers of IL-12 mRNA-expressing cells are elevated in bronchial biopsies from nonasthmatics as compared with asthmatic subjects, and treatment of the latter with corticosteroids is associated with a rise in the numbers of IL-12 mRNA-expressing cells (23). Second, following allergen immunotherapy, a reduction in the cutaneous late phase response to allergen is associated with elevated local expression of IL-12 and IFN-γ mRNA (24, 25). However, recent data suggest that fewer lung T cells from patients with asthma are recognized by an anti-IL-12Rβ2 mAb than in patients with sarcoidosis (a putative Th1 disease) (26), consistent with down-regulation in asthma, up-regulation in sarcoidosis, or both. We therefore examined the effects of IL-12 (and IFN-γ) on cytokine production by bronchoalveolar lavage (BAL)3 T cells isolated from the bronchial lumen during an allergen-induced late response, and compared these with the responses of PBMC obtained from the same patients. We hypothesized that IL-5 production by both BAL T cells and T cells in PBMC would be inhibited by IL-12. Furthermore, we have used in situ hybridization to examine whether mRNA transcripts encoding the IL-12Rβ2 subunit can be detected in enriched BAL T cells collected from asthmatic patients during the allergen-induced late response.
Materials and Methods
All subjects were required to have a forced expiratory volume in 1 s (FEV1) >80% predicted and a methacholine PC20 (that concentration of inhaled methacholine resulting in a 20% reduction in baseline FEV1) of >1 mg/ml, but <32 mg/ml. All of these subjects had ≥5-mm diameter wheal at 15 min following skin-prick testing with Phleum pratense (timothy grass pollen) or Dermatophagoides pteronyssinus (house dust mite) extract in the presence of negative diluent and positive histamine controls. These subjects were also required to have elevated concentrations of serum IgE Abs specific for P. pratense or D. pteronyssinus (radioallergosorbent test, RAST >0.70 IU/ml; CAP system; Pharmacia Diagnostics, Uppsala, Sweden). All subjects participating in this study were nonsmokers and, where appropriate, inhaled corticosteroid therapy was withheld 2 wk before bronchoscopy. The study was approved by the Ethics Committee of the Royal Brompton Hospital (London, U.K.) and all subjects gave written informed consent.
Fiberoptic bronchoscopy and allergen challenge
All subjects were premedicated with 2.5 mg nebulized albuterol, and 0.6 mg atropine and 5–10 mg midazolam administered i.v. Local anesthesia of the vocal cords and trachea was induced with 2–4% lidocaine. After inspection of the bronchial tree, the tip of the bronchoscope (an Olympus BFP20; Olympus, London, U.K.) was wedged at random in a segmental bronchus of the lingula or middle lobe, and BAL was performed by sequentially instilling two 60-ml aliquots of sterile warmed saline, followed by gentle aspiration into a sterile glass bottle (baseline BAL). Allergen challenge was then performed in a segmental bronchus of the lingula or middle lobe (contralateral to that which had been lavaged) by instilling 100 biological units (BU) of P. pratense or D. pteronyssinus (Aquagen extract, kindly provided by ALK Abelló, Horshølm, Denmark) made up in 5 ml of sterile saline. The challenge site was observed for an additional 5 min, and in the absence of excessive local bronchoconstriction, a further 400 BU of allergen was introduced in 5 ml of sterile saline. All subjects were subsequently detained in hospital overnight for observation. During this period, nebulized bronchodilator (5 mg albuterol) was administered as necessary to the asthmatics to maintain FEV1 >80% of the predicted value. A second bronchoscopy was repeated after 24 h. Just before premedication for the second bronchoscopy, a sample of peripheral venous blood was collected in a sterile heparinized syringe. BAL (two 60-ml aliquots of saline) was then performed in the allergen-challenged segment, as described above.
PBMC were isolated from heparinized blood samples by density-gradient centrifugation over Histopaque (Sigma, Poole, U.K.), washed twice in HEPES-buffered RPMI (Sigma), and resuspended in RPMI supplemented with 5% human AB serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM l-glutamine (all Sigma). This supplemented medium (complete medium) was used for all cell culture experiments. BAL cells were isolated by passing BAL fluid through two layers of sterile gauze to remove mucus, washed twice in HEPES-buffered RPMI, and resuspended in complete medium. A differential cell count was performed on a cytospin of BAL cells using May-Grünwald-Giemsa stain. Mononuclear cells were enriched from BAL cells by density-gradient centrifugation over Histopaque, washed twice, and resuspended in HEPES-buffered RPMI. T cells were then enriched from BAL mononuclear cells by passage of cells through a T cell enrichment column containing anti-human Ig-coated glass beads (R&D Systems, Abingdon, U.K.). Lymphocytes constituted >90% of such preparations, as judged by morphology, and showed good viability (>90%), as determined by trypan blue exclusion. Purity of BAL T cells enriched over these columns is 85–90%, as judged by flow cytometry with anti-CD3 mAb (27). The BAL T cell enrichment protocol typically yielded about 0.15–0.5 × 106 cells.
BAL T cell cultures
BAL T cell concentrations in culture were restricted by the relatively low numbers of enriched BAL T cells that could be obtained from human volunteers. Therefore, BAL T cells were resuspended at 8 × 104 cells/ml in 96-well round-bottom plates (Nunc, Roskilde, Denmark) with irradiated (3000 rad) autologous PBMC at 4 × 105 cells/ml as APCs, in the presence and absence of 10 μg/ml allergen extract (Aquagen extract; ALK Abelló). All cultures were performed in 250 μl vol. In all cases, control cultures were performed with APC with allergen (i.e., without BAL T cells) to confirm the absence of a background response in the irradiated PBMC population. We previously showed that under these conditions, allergen-induced IL-5 production is detectable in cultures of enriched BAL T cells from atopic asthmatics, but not peripheral blood T cells from the same patients, or BAL T cells from nonatopic normal subjects. Moreover, we reported that allergen-induced IL-5 production by BAL T cells correlated with allergen-induced eosinophilia in the bronchi (28).
Allergen-stimulated BAL T cell cultures were supplemented with 0.5, 5, and 50 U/ml human rIL-12 (R&D Systems), and 5, 50, and 500 U/ml human rIFN-γ (R&D Systems) or vehicle control. All cultures were performed in a minimum of triplicate. In some experiments, cultures were supplemented with 10 μg/ml of neutralizing anti-IFN-γ mAb (R&D Systems) to examine whether the effects of IL-12 were IFN-γ dependent. Culture supernatants were removed from each microculture well on day 6 for ELISA measurements.
Detection of IL-5 production by peripheral blood T cells necessitated different culture conditions from those that could be used to detect IL-5 production by enriched BAL T cells. To elicit allergen-induced IL-5 production by peripheral blood T cells, PBMC were resuspended at 5 × 106 cell/ml in triplicate 250 μl vol in 96-well flat-bottom cell culture plates (Nunc) in the presence or absence of 10 μg/ml of allergen extract. We previously showed that allergen-induced IL-5 production is optimal at this cell density, is dependent on CD4+ T cells, and is specifically elevated in patients with atopic asthma or allergic rhinitis, but not asymptomatic atopic or nonatopic controls (2). PBMC cultures were supplemented with 0.5, 5, and 50 U/ml human rIL-12 (R&D Systems), and 5, 50, and 500 U/ml human rIFN-γ (R&D Systems) or vehicle control. In some experiments, cultures were supplemented with 10 μg/ml of neutralizing anti-IFN-γ mAb (R&D Systems) or 50 μg/ml of neutralizing anti-IL-12 mAb (R&D Systems) to examine whether the effects of IL-12 were IFN-γ dependent and whether the effects of IFN-γ were IL-12 dependent. Optimal concentrations of anti-IFN-γ and anti-IL-12 mAbs were determined by titration of the concentration of Ab required to reverse the inhibition of allergen-induced IL-5 production by PBMC by IFN-γ and IL-12, respectively. Culture supernatants were removed from each microculture well on day 6 for ELISA measurements.
T cell clone culture
The AC1 T cell clone specific for house dust mite (Der p 2) was isolated as previously described (29). To elicit IL-5 production, T cell clones (105 cells/ml) were stimulated with and without a synthetic peptide (homologous for a Der p 2 epitope) in the presence of irradiated autologous EBV-transformed B cells and human rIL-12 (R&D Systems). Culture supernatants were collected after 48 h for the determination of IL-5 concentrations by ELISA.
IL-5 concentrations in BAL T cell culture supernatants were measured in duplicate using a commercial assay sensitive above 1 pg/ml (R&D Systems). IL-5 in PBMC culture supernatants were determined in duplicate using commercially available Ab pairs (PharMingen, Cowley, U.K.), sensitive to 10 pg/ml. The same human rIL-5 standard (R&D Systems) was used in both assays. IFN-γ concentrations were measured by ELISA, also in duplicate, using commercially available Ab pairs (PharMingen), sensitive to 10 pg/ml.
In situ hybridization
Riboprobes, antisense and sense, were prepared from cDNA encoding IL-12Rβ2 (generous gift from Dr. Gubler, Hoffmann-LaRoche, Nutley, NJ). IL-12Rβ2 cDNA was subcloned into the bluescript SKII− vector. cDNA was linearized with appropriate enzymes before transcription. Transcription was performed in the presence of [35S]UTP) and the appropriate T7 or T3 RNA polymerases. In situ hybridization was performed on paraformaldehyde-fixed cytospin preparations of enriched BAL T cells, isolated as above. Briefly, cytospins were permeabilized with Triton X-100 in PBS, followed by proteinase K digestion. To inhibit nonspecific binding of 35S, sections were treated with iodoacetamide and N-ethylmaleimide and then in acetic anhydride/triethanolamine before hybridization with 35S-labeled riboprobes. As a negative control, sections were hybridized with the sense probe or treated with RNase A solution before the prehybridization step with antisense probes. Specific hybridization was recognized as clear dense deposits of silver grains in the photographic emulsion overlying cytospins.
Results are expressed as mean ± SEM. Statistical analysis of data was performed using a one-way ANOVA or paired Student’s t test, as described in Fig. 1. All analyses were performed using a commercially available statistical package (Minitab, State College, PA), and p < 0.05 was considered as significant.
Effects of IL-12/IFN-γ on IL-5 production by peripheral blood T cells
Initial experiments were performed to determine the optimal concentration ranges of exogenous IL-12 and IFN-γ required to down-regulate IL-5 production by PBMC stimulated with allergen. PBMC were isolated from four patients with allergic asthma and/or rhinitis (not the same patients who later underwent segmental allergen challenge) and stimulated under conditions that were previously shown to result in optimal CD4+ T cell-dependent IL-5 production (2). Addition of human rIL-12 significantly affected allergen-induced IL-5 production (Fig. 1,a; p = 0.0005 by ANOVA). Comparison of individual IL-12 concentrations with baseline showed that IL-12 significantly inhibited allergen-induced IL-5 production at 0.5, 5, and 50 U/ml. Based on these results, the concentrations of IL-12 chosen for use in subsequent cultures with BAL T cells were 0.5, 5, and 50 U/ml. rIFN-γ also had a significant effect on allergen-induced IL-5 production by PBMC (Fig. 1 b; p = 0.001 by ANOVA), and this was statistically significant compared with baseline at 5, 50, and 500 U/ml IFN-γ. The concentrations of IFN-γ chosen for use in subsequent cultures with BAL T cells were 5, 50, and 500 U/ml.
Effects of IL-12 on IL-5 production by an established allergen-specific T cell clone
To confirm previous reports that production of Th2 cytokines by allergen-specific T cell lines and clones can be refractory to inhibition by IL-12 (17, 20), the effect of IL-12 on IL-5 production by a T cell clone specific for Der p 2 (derived from house dust mite) was also examined. T cell clones were stimulated with Der p 2-derived peptides and APCs (EBV-transformed B cells) in the presence of 0.005, 0.05, 0.5, 5, and 50 U/ml IL-12. In contrast to PBMC, IL-5 production by the T cell clone appeared to be resistant to the effects of IL-12 at these concentrations (Fig. 2).
Effects of IL-12 and IFN-γ on cytokine production by BAL T cells: comparison with effects on peripheral blood T cells
BAL T cells were enriched and cultured from four atopic asthmatics 24 h after segmental challenge with grass pollen allergen. The characteristics of the subjects and the cell populations recovered in BAL fluid are described in Table I and Table II, respectively (subjects 1 to 4). Two of the four subjects had current bronchial hyperreactivity (histamine PC20 < 8 mg/ml), whereas two patients with pollen-induced asthma were asymptomatic out of season, with histamine PC20 values >8 mg/ml. All four subjects were highly sensitized to P. pratense grass pollen, as evidenced by markedly raised allergen-specific IgE levels. Both the total numbers of BAL cells and the percentages of eosinophils were increased following segmental allergen challenge in all subjects (Table II).
|Patient .||Age (yr) .||Sex .||Methacholine PC20 (mg/ml) .||RAST (IU/ml) .||PEFR % Falla .|
|1||43||F||15||>100 (GPb )||33|
|Patient .||Age (yr) .||Sex .||Methacholine PC20 (mg/ml) .||RAST (IU/ml) .||PEFR % Falla .|
|1||43||F||15||>100 (GPb )||33|
Maximal fall in peak expiratory flow rate within 12 h of allergen provocation.
GP, grass pollen; HDM, house dust mite.
|Patient .||Sample .||Total cells (× 106) .||Lymphocytes (%) .||Eosinophils (%) .||Neutrophils (%) .||Macrophages (%) .|
|Patient .||Sample .||Total cells (× 106) .||Lymphocytes (%) .||Eosinophils (%) .||Neutrophils (%) .||Macrophages (%) .|
In each of the four patients tested, culture of enriched BAL T cells with allergen and autologous irradiated PBMC (as APCs) was accompanied by detectable IL-5 production. In three of the four patients tested, IL-5 production appeared to be clearly induced by the allergen, whereas in the other patient, allergen had only a minor effect in up-regulating IL-5 secretion (Fig. 3). In all four patients, addition of exogenous recombinant IL-12 or IFN-γ resulted in inhibition of IL-5 production (Fig. 3). In contrast to IL-5, production of IFN-γ by enriched BAL T cells was not clearly induced by allergen in any of the four patients examined, but increased in the presence of IL-12 in all cases (Fig. 4). IL-5 and IFN-γ could not be detected in culture supernatants of irradiated PBMC alone (either with or without allergen).
To facilitate direct comparison of the responsiveness of T cells in BAL and PBMC to IL-12 and IFN-γ, data have also been expressed as a percentage of the maximal response (Fig. 5). IL-12 had significant effects on both IL-5 and IFN-γ production by enriched BAL T cells (p = 0.002 and p = 0.0005 by ANOVA, respectively) and PBMC (p = 0.001 and p = 0.004 by ANOVA, respectively). Similarly, IFN-γ also significantly affected IL-5 production by enriched BAL T cells and PBMC (both p = 0.02, by ANOVA). The degree of inhibition of the IL-5 response was comparable in enriched BAL T cells and PBMC. At 50 U/ml IL-12, IL-5 production by allergen-stimulated BAL T cells and PBMC was 27.6 ± 6.6% and 30.1 ± 9.7% of the maximal observed response, respectively (p = 0.88 for BAL vs PBMC, by paired t test). At 500 U/ml IFN-γ, IL-5 production by allergen-stimulated BAL T cells and PBMC was 48.1 ± 4.1% and 37.7 ± 12.5% of the maximal observed response, respectively (p = 0.32 for BAL vs PBMC, by paired t test). Similarly, for the purposes of comparing responses of BAL T cell and PBMC cultures with IL-12, IFN-γ concentrations in cultures were also expressed as a percentage of the maximal response (Fig. 5). For enriched BAL T cells and PBMC, maximal IFN-γ production was observed at 50 U/ml IL-12, and the percentage of this maximal response observed without IL-12 was 11 ± 6.6% and 11 ± 8.9%, respectively (p = 1 for BAL vs PBMC, by paired t test). Thus, both IL-12 and IFN-γ inhibited IL-5 production by enriched BAL T cells and PBMC to a similar degree. Similarly, IL-12-induced IFN-γ production, though not clearly up-regulated by the presence of allergen alone, was up-regulated in BAL T cells and PBMC by IL-12 to similar degrees.
Inhibition of IL-5 responses by IL-12 is not affected by neutralizing endogenous IFN-γ production
To address the possibility that IL-12 indirectly suppresses allergen-induced IL-5 production through its ability to up-regulate IFN-γ expression, cultures from three asthmatics (PBMC in two cases; enriched BAL T cells in one case) were supplemented with neutralizing anti-IFN-γ mAb. The Ab reversed the inhibition of IL-5 production caused by addition of exogenous rIFN-γ (Fig. 6), confirming the validity of the Ab. In contrast, the neutralizing anti-IFN-γ mAb failed to reverse the inhibitory effects of IL-12 in any of the three experiments. Furthermore, to address the possibility that the effects of IFN-γ on allergen-induced IL-5 production are mediated through up-regulation of IL-12 production, PBMC cultures from two patients were supplemented with neutralizing anti-IL-12 mAb. Although anti-IL-12 mAb reversed the effects of IL-12 on allergen-induced IL-5 production, neutralization of IL-12 did not reduce the inhibitory effect of IFN-γ on IL-5 production (Fig. 6).
Expression of IL-12Rβ2 subunit mRNA by BAL T cells
BAL T cells were isolated from five asthmatics (patients 3 to 7; Tables I and II) before and after segmental allergen challenge using T cell enrichment columns. Expression of IL-12Rβ2 subunit mRNA transcripts was examined in cytospin preparations of these cells by in situ hybridization (Fig. 7). The percentages of cells expressing IL-12Rβ2 mRNA were 68 (±12)% and 33 (±15)%, respectively. Comparison of samples obtained before and after segmental challenge revealed that there was a trend for the percentage of enriched BAL T cells expressing mRNA for the IL-12Rβ2 subunit to decrease following allergen provocation (p = 0.047 for baseline vs postallergen challenge, by paired t test).
Allergen-specific T cells freshly isolated from the peripheral blood of subjects with active atopic disease are characterized by production of Th2-type cytokines, indicating some degree of in vivo differentiation. In these T cells, production of IL-5, IL-13 (4), IL-4, and IFN-γ (16, 17) is consistently sensitive to the effects of IL-12. According to several reports, allergen-specific T cell clones and lines, activated and expanded in vitro from peripheral blood T cells, can undergo commitment to the Th2-type lineage through loss of sensitivity to the regulatory cytokine IL-12 (17, 19, 20). Inthe present study, we tested the hypothesis that IL-5 production by T cells present in the lungs of asthmatics during the allergen-induced late response, thought to be at least partly driven by T cell-derived IL-5 (15, 30), is sensitive to inhibition by IL-12. There is good reason that the functional status of such cells might be ambiguous in relation to freshly isolated peripheral blood T cells and peripheral blood T cells that have been activated and expanded with Ag in vitro. First, the allergen-induced late asthmatic response is characterized by an increase in expression of IL-5, IL-4, and IL-13 by bronchial mucosal or luminal T cells (6, 7, 31) and an increased frequency of T cells expressing surface activation markers, such as CD25 (7). This suggests some degree of allergen-driven T cell activation and differentiation. Second, bronchial T cells are phenotypically and functionally distinct from freshly isolated peripheral blood T cells: for example, BAL T cells are almost exclusively CD45RO+ memory T cells (32), show an increased tendency to undergo apoptosis (33), and, when collected 24 h aftersegmental allergen challenge, BAL T cells produce significantly greater amounts of IL-5 than equivalent numbers of freshly isolated peripheral blood T cells when stimulated with allergen in vitro (28).
In the present study, we chose to study T cells in BAL fluid rather than biopsy tissue specimens because it was our aim to investigate the responses of T cells that had been freshly isolated in an ex vivo culture model: the number of enriched T cells in free suspension that can be obtained from bronchial biopsy tissue is much lower than with BAL, necessitating prolonged expansion in vitro (34). We previously reported that IL-5 production by BAL, but not peripheral blood T cells stimulated with allergen and irradiated PBMC (as APC) correlates with the degree of airway eosinophilia and is higher in asthmatics than control subjects (28). Although the APC types in PBMC are unlikely to reflect the exact composition of APC in the lung, a previous report did not find differences in IL-12 responsiveness by allergen-specific peripheralblood T cells stimulated by different APC types (17). We conclude, therefore, that the conditions adopted for this study are of physiological relevance and are preferable to alternative means, such as use of nonspecific mitogens or expanded T cell clones or lines. The relationship between the properties of BAL T cells and bronchial mucosal tissue T cells does, however, remain to be established conclusively. Nevertheless, there is strong evidence from human studies to suggest that the properties of BAL T cells are highly relevant to asthma: first, CD3+ BAL T cells, like CD3+ T cells in the bronchial mucosa, are characterized by elevated expression of IL-5 and IL-4 mRNA (35). Second, the proportion of BAL T cells that are activated and express Th2-type cytokine mRNA has been correlated with asthma symptoms (36). Third, following inhalational allergen challenge, the numbers of BAL T cells expressing Th2-type cytokine mRNA are increased, correlating with the numbers of recruited eosinophils (7). Finally, improvement in asthma symptoms and decreased BAL eosinophilia following steroid therapy is accompanied by decreased expression of Th2-type cytokine mRNA by BAL T cells (37).
To our knowledge, this is the first study to directly examine the effects of IL-12 and IFN-γ on cytokine responses by T cells isolated from human asthmatic airways. BAL T cells collected 24 h after a segmental allergen challenge produced IL-5 in response to allergen stimulation and, as with IL-5 production by T cells in PBMC, this was inhibited by IL-12 in a concentration-dependent manner. Furthermore, the degree of inhibition at particular IL-12 concentrations, expressed in percentage terms, was similar for T cells from both sources (Fig. 5). Addition of IFN-γ to cultures also resulted in similar patterns of inhibition of IL-5 production by enriched BAL T cells and PBMC (Fig. 5). Furthermore, in addition to inhibiting IL-5 production, IL-12 also up-regulated IFN-γ production in BAL and PBMC cultures.
Collectively, these data suggest that in terms of their ability to respond to IL-12 and IFN-γ, cultured T cells derived from asthmatic late responses closely resemble T cells in the peripheral circulation. Thus, T cells present in the airways of asthmatics following allergen exposure, although predisposed to produce cytokines such as IL-5, appear not to have undergone commitment to loss of IL-12 sensitivity. It was previously reported that human naive cord blood T cells or CD45RA+ peripheral blood T cells show induction of IL-12Rβ2 mRNA and surface protein on stimulation with anti-CD3 mAb, and that this is potentiated by IL-12 (22, 26). It is therefore possible, in theory, that the responsiveness of BAL T cells to IL-12 in our study was due to up-regulation of the IL-12Rβ2 subunit on in vitro restimulation with allergen and IL-12, and that the same cells would have been unresponsive to IL-12 in vivo. There are a number of observations, however, that suggest that this is unlikely to be the case. First, Marshall et al. (17) restimulated allergen-specific T cells from atopic patients with specific allergen through a number of expansion cycles in the presence or absence of IL-12 before analyzing cytokine production. The effects of adding IL-12 at the time of restimulation with allergen through the TCR progressively diminished through each cycle until there was no effect on IL-4 and IFN-γ production. Second, Hilkens et al. (20) stimulated allergen-specific Th2 clones with anti-CD3 and anti-CD28 mAb in the presence of IL-12 and also reported no effect on production of cytokines. These findings are consistent with our own data obtained with an allergen-specific T cell clone (Fig. 2). Thus, IL-12 responsiveness does not appear to be a global feature of all human memory allergen-specific T cells restimulated through the TCR in vitro.
Current evidence strongly suggests that failure to express IL-12Rβ2 mRNA correlates with IL-12 insensitivity in Th2 cells (21, 22), although recent studies suggest, at least in mice, that restoration of IL-12Rβ2/STAT4 signaling in committed Th2 cells may not reverse IL-12 insensitivity (38, 39). In the light of this correlation, our aim was to seek support for our hypothesis that airway T cells in asthmatics are IL-12 responsive by determining whether IL-12Rβ2 mRNA transcripts are detectable in enriched BAL CD3+ T cells after segmental allergen provocation. In the event, significant numbers of BAL T cells could be demonstrated to contain IL-12Rβ2 mRNA transcripts (mean = 33.1%) following segmental allergen challenge (Fig. 7). Although we are unable to determine whether or not these cells were allergen specific, these data are consistent with the observed IL-12 responsiveness of these T cells when stimulated with allergen in vitro. Although the proportion of BAL T cells expressing IL-12Rβ2 mRNA was higher before provocation with allergen, we cannot be certain whether this reflects in situ down-regulation of IL-12Rβ2 mRNA expression in Th2-type cells, or a dilution of the IL-12Rβ2-expressing population by T cells recruited during the asthmatic late response. As with the in vitro experiments, we elected to perform these measurements on the entire CD3+ population since studies of human cord blood CD8+ and CD4+ T cell lines induced to produce type 1 and type 2 cytokines suggest that the proposed restricted patterns of IL-12Rβ2 are applicable to both CD8+ and CD4+ T cells (26). Furthermore, studies on BAL and bronchial biopsies from asthmatics have indicated that CD8+ cells producing type 2 cytokines (Tc2 cells), as well as CD4+ T cells, are present in the airways of these subjects (40, 41). Thus, studying the entire CD3+ population may be more relevant than studying CD4+ cells in isolation.
Although IL-5 production by BAL T cells was up-regulated by culture with allergen, this was not the case with IFN-γ. It was not apparent whether the increased IFN-γ production in response to IL-12 was spontaneous or allergen induced in specific T cells because the small numbers of cells that could be isolated precluded setting up cultures in the absence of allergen. Nevertheless, it could be argued that for IFN-γ such a distinction would be unimportant since in the context of the lung in vivo, it is the net effect of IL-12 on IFN-γ production by all T cells that would be biologically important. A further issue relates to whether the effects of IL-12 on IL-5 production are independent of endogenously produced IFN-γ, and similarly, whether the effects of IFN-γ are IL-12 independent. In experiments performed with allergen-stimulated PBMC and BAL T cells from atopic patients, the effects of IL-12 on IL-5 production could not be reversed by addition of a neutralizing anti-IFN-γ Ab (Fig. 6). Similarly, the effects of IFN-γ on IL-5 production by allergen-stimulated PBMC were not reversed when endogenous IL-12 production was neutralized with a specific Ab (Fig. 6). Thus, under the conditions tested, the effects of neither IL-12 nor IFN-γ appear to be mediated through induction of the other cytokine.
One question raised by our findings concerns the role of extinction of IL-12 sensitivity in Th2 immune responses if this is not a property of effector T cells at the site of inflammation. Although entirely speculative, one possibility is that extinction plays a role in long-term memory of Th2-type immune responses and occurs only in a subpopulation of long-lived T cells that are selected by prolonged in vitro culture during the establishment of T cell clones and lines.
This study, which suggests that the presence of IL-12 and IFN-γ in the mucosa could decrease local production of IL-5 by T cells even during an ongoing asthmatic late response, is consistent with previous reports suggesting that the bronchial mucosa (23) and peripheral blood (42) of asthmatics are characterized by abnormally low IL-12 expression. Furthermore, clinical improvement following specific allergen immunotherapy (24) or corticosteroid therapy (23) may be associated with increased numbers of IL-12 mRNA-expressing cells in target organ tissue. As corticosteroids appear to inhibit IL-12 mRNA expression at the single cell level (43), it seems likely that corticosteroids bring about an increase in numbers of IL-12 mRNA-expressing cells in asthmatics through an effect on local cytokine networks or cell recruitment.
In animal models of asthma, systemic IL-12 decreases allergen-induced airway eosinophilia, hyperresponsiveness, and/or expression of IL-4 and IL-5 (44, 45, 46, 47). Of particular interest are animal studies showing that nasal administration of IL-12 (48) or transient IL-12 gene transfer to the mucosa (49) also effectively inhibits airway eosinophilia and hyperresponsiveness. Similar findings have also been reported with nebulized IFN-γ (50). The present study provides the first evidence, however, that production of proallergic cytokines by T cells from the site of the disease in human asthma is susceptible to down-regulation by IL-12 or IFN-γ. We believe that these findings provide a valuable insight into the properties of IL-5-producing T cells in the target organ in human allergic disease.
We thank Fiona O’Brien and Wendy Noble for their expert technical assistance.
This study was supported by National Asthma Campaign and Medical Research Council (U.K.).
Abbreviations used in this paper: BAL, bronchoalveolar lavage; FEV1, forced expiratory volume in 1 s; PC20, that concentration of inhaled methacholine resulting in a 20% reduction in baseline FEV1.