INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Recent studies have shown that Th2 lymphocytes play an important role in the generation of airway eosinophilia in asthma through their ability to produce interleukin (IL)-5 (1, 2). Increased capacity of such cells to produce IL-5 has been demonstrated in the airways and peripheral blood of both atopic and nonatopic people with asthma. Furthermore, expression and production of IL-5, but not IL-4 or other Th2 cytokines, in bronchoalveolar lavage (BAL) and bronchial biopsies from asthmatics are consistently related both to the infiltration of the airways with eosinophils and to clinical parameters of disease severity (3).

There is a substantial difference in IL-5 mRNA expression in bronchial biopsies and IL-5 production by BAL cells between atopic people with asthma and atopic people without asthma. The difference is apparent both at baseline and after allergen inhalation challenge (7). These findings are compatible with reports of higher numbers and activation of eosinophils in the airways of atopic people with asthma compared with atopic people without asthma (10, 11). In contrast to the airway observations, the difference in peripheral blood T-cell responses to in vitro allergen stimulation is less marked (6, 12, 13). These findings suggest that T-cell IL-5 response to allergen may be differentially regulated by local mechanisms, so that not all atopic people develop asthma.

Work over the last several years has demonstrated that different antigen-presenting cell (APC)-related factors in the microenvironment of a T cell may determine its profile and quantity of cytokine expression and production (14). This raises the possibility that induction of protective or pathogenic immunity in the airways of atopic individuals may be mediated by such resident APC-derived factors. The effects of alveolar macrophage (AM) products on allergen-specific T-cell cytokine responses, particularly on IL-5 production, in the airways of atopic people with and without asthma are of special interest in this respect. It has been shown that the AM plays an important role in maintaining local immunological homeostasis in the airways of normal subjects through suppressing T-cell responses to local soluble antigens (21), but AM may play a contrary role in the development of allergic inflammation in the airways of atopic people with asthma (24).

In this study, we hypothesized that the effect of AM on IL-5 production by CD4+ T cells in atopic people with asthma differs from that in atopic people without asthma. To test this, we assessed IL-5 production in cocultures of CD4+ T cells with AM or peripheral blood monocytes (MN) taken from atopic people with asthma, as well as atopic and nonatopic people without asthma. We then investigated the effects of the AM-derived factors, IL-1, IL-1, IL-6, and tumor necrosis factor- (TNF-), on IL-5 production by allergen-stimulated CD4+ cells in atopic people with asthma through a series of cytokine-blocking experiments.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Ten atopic people with asthma, nine atopic people without asthma, and eight nonatopic normal subjects were enrolled in the study. Their clinical and demographic characteristics are shown in Table 1. Subjects with atopic asthma met the American Thoracic Society criteria for the diagnosis of asthma (27). They had documented reversible airflow obstruction (15% improvement in FEV₁ either spontaneously or in response to inhaled ₂-agonist). All the atopic asthmatics had mild asthma and were using only ₂-agonist as relief medication. None had received inhaled or oral corticosteroid therapy in the 3 mo before the study. All the atopics had two or more positive skin responses to a panel of common environmental allergens, including either rye grass pollen or house dust mite (HDM). Nonasthmatic individuals had no past or present asthmatic or other allergic disease symptoms, and they had normal pulmonary function. None of the subjects were currently smoking, had smoked for more than 10 pack-years, or had smoked within the previous 12 mo. Any subject with a history compatible with respiratory infection in the 6 wk preceding the study was excluded.

This study was approved by The Alfred Hospital Ethics Committee, and informed consent was given by all patients and control subjects.

Bronchoscopy Procedure

Fiberoptic bronchoscopy was undertaken according to the guidelines of the American Thoracic Society (28). Subjects were premedicated with nebulized albuterol (2.5 mg), intravenous atropine (0.6 mg), and intravenous midazolam (5 to 10 mg). They received supplemental oxygen at 4 L/min throughout and were monitored by continuous pulse oximetry. Subjects were given 2% and 4% lignocain for local anesthesia. Bronchoalveolar lavage (BAL) was performed using three aliquots, 60 ml each, of phosphate-buffered saline (PBS) at 37° C introduced by gentle hand pressure into the middle lobe and recovered by suction at less than 80 mm Hg. The BAL fluid was collected into polypropylene tubes and immediately placed on ice. On the same occasion, 40 ml of venous blood was drawn into a heparinized container for later separation of monocytes and CD4+ T cells.

Isolation of Peripheral Blood CD4+ T Cells, Monocytes, and Alveolar Macrophages

Peripheral blood CD4+ T cells. Peripheral blood mononuclear cells (PBMC) were isolated from freshly drawn peripheral blood by density-gradient centrifugation on Ficoll-Paque equipment (Pharmacia, Uppsala, Sweden). To obtain CD4+ T cells, magnetic beads coated with monoclonal antibody against CD4 (Dynal A. S., Oslo, Norway) were added to the PBMC suspension at a ratio of 4:1 (beads to CD4+ T cells) and incubated at 4° C for 1 h with gentle tilting and rotation. The rosetted CD4+ T cells were isolated by placing the test tube on a magnet for 2 to 3 min, and then washing 3 times with PBS containing 2% fetal calf serum (FCS). Beads were removed from CD4+ T cells by using DETACHaBEAD (Dynal A. S.). This procedure resulted in a CD4+ T-cell fraction of > 99% purity by flow cytometry and > 99% viability by trypan blue-dye exclusion. Less than 1% of cells in the CD4+ purified-cell fraction stained with the monocyte marker CD14.

Monocytes and alveolar macrophages. Peripheral blood monocytes were prepared by adherence of the CD4+ T-cell-free fraction of PBMC on plastic dishes for 1 h at 37° C in a 5% CO₂ humidified atmosphere. The nonadherent cells were discarded, and the adherent cells were removed by gently scraping with a plastic scraper, then resuspended in RPMI 1640 medium supplemented with 5% heat-inactivated FCS, 2 mM L-glutamine, 100 units/ml penicillin, 100 µg/ml streptomycin, and 125 µg/ml gentamicin. A population comprising > 95% MN with a viability of  95%, as judged by Quick Dip and trypan blue staining, was normally yielded by two rounds of adherent incubation. Alveolar macrophages were isolated directly from BAL cell suspensions by adherence, with the resulting AM population containing > 95% AM with a viability of > 90%.

Cell Cultures

Purified peripheral blood CD4+ T cells were cultured (1 × 10⁶/ml/ well) alone in RPMI 1640 medium or cocultured with autologous MN or AM at a 2:1 ratio of T cells to MN or AM. (In preliminary experiments, CD4+ T cells from three atopic people with asthma were cocultured at ratios of 2:1, 1:1, 1:2, and 1:4 with autologous AM, and no significant differences were found in mitogen-stimulated IL-5 production among these cocultures.) The cultures were performed with and without mitogen (24 h) or allergen stimulation (5 d). Rye grass pollen (20 µg/ml) or HDM (10 µg/ml) was used for atopic subjects according to their skin-test sensitivity and for normal controls at random. The same batches of rye grass pollen and HDM were used for all experiments (Greer Laboratories, Lenoir, NC). Phorbol 12-myristate 13-acetate (PMA) and ionomycin were used at 10 ng/ml and 1 µmol, respectively (Sigma, Sydney, Australia), for mitogen stimulation. Brefeldin A (10 µg/ml) (Sigma) was added to all the cultures 10 h before cell harvesting to block intracellular transport mechanisms leading to an accumulation of cytokines in the Golgi complex.

Intracellular Interleukin-5 Assay by Flow Cytometry

Cells (5 × 10⁵/50 µl) were incubated at 4° C with fluorescein isothiocyanate (FITC)-conjugated CD4 monoclonal antibody (Becton Dickinson, Mountain View, CA) for 30 min. After washing in PBS, cells were fixed with 4% paraformaldehyde at 4° C for 20 min. The fixed cells were thoroughly resuspended in 50 µl of permeabilization buffer containing 1% saponin (Sigma) and incubated with phycoerythrin (PE)-conjugated IL-5 monoclonal antibody (PharMingen, San Diego, CA) at a concentration of 20 µg/ml at 4° C for 30 min. After a final wash in saponin buffer, cells were resuspended in PBS and their fluorescence was analyzed by flow cytometry (FACScan; Becton Dickinson). The number of immunofluorescence-positive cells was determined per 10,000 analyzed cells. To ensure that only intracellular proteins were quantified, control cells were fixed but not permeabilized, giving < 0.5% PE-positive cells. Irrelevant isotype-matched control antibodies produced < 0.3% fluorescent cells. In four experiments, preincubating the anti-IL-5 antibody for 1 h with 20 µg of recombinant human IL-5 (PharMingen) resulted in > 90% inhibition of PE-positive cells, confirming IL-5 specificity.

Interleukin-5 ELISA

Quantitative determination of IL-5 levels in supernatants was performed by ELISA using paired antibodies (PharMingen) according to the procedure recommended by the manufacturer. Recombinant human IL-5 was used as a standard. The detection limit was 31 pg/ml.

Blocking Studies

Blocking experiments were performed in allergen-stimulated AM- CD4+ T-cell cocultures for seven atopic people with asthma. The AM were preincubated with 8 µg of anti-IL-1, anti-IL-1, anti-IL-6, or anti-TNF- (R&D Systems, Minneapolis, MN) monoclonal antibodies at 37° C for 1 h to block the biological activities of these cytokines. The percentages of IL-5-positive CD4+ T cells and levels of IL-5 production in the cocultures, with or without different blocking regimens, were compared. To examine the effects of these AM-derived cytokines on CD4+ T-cell activation, the percentages of CD4+ CD25+ T cells were detected as before by double-labeling with FITC-conjugated CD4 monoclonal antibody and PE-conjugated CD25 monoclonal antibody (Becton Dickinson).

Statistical Analysis

Significant differences between groups and within a group were assessed by using the Mann-Whitney U test and the Wilcoxon signed rank test, respectively. The relationship between two parameters was analyzed by Spearman's rank correlation. For statistical analysis of the percentage of cells positive for IL-5 by flow cytometry, a value of 0.05% was assigned when values were less than 0.5%; 0.1 pg/ml was designated if IL-5 was undetectable in the supernatant of cell cultures.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Interleukin-5 production in cultures of CD4+ T cells alone, with and without allergen stimulation. Interleukin-5 production by resting CD4+ cells alone differed significantly among the three groups studied (Figure 1). The percentage of IL-5- producing cells in atopic people with asthma (median, 1.1%; range, 0.05-2.1%) and in atopic people without asthma (median, 0.6%; range, 0.05-1.0%) was higher than that in nonatopic normal subjects (median, 0.05%; range, 0.05-0.7%; p < 0.05). Consistent with this finding, the level of IL-5 in the supernatant of resting CD4+ cell cultures was higher in those with asthma (median, 470.8 pg/ml; range, 122.1-731.8 pg/ml), than in atopic people without asthma (median, 197.2 pg/ml; range, 0.1-260 pg/ml; p < 0.01) and in nonatopic normal subjects (median, 33.1 pg/ml; range, 0.1-254 pg/ml; p < 0.001).

View larger version (34K): [in this window] [in a new window]   Figure 1.   Percentages of IL-5-producing cells (top) and concentrations of supernatant IL-5 (bottom) in CD4+ cell cultures, with or without allergen stimulation (rye grass pollen, 20 µg/ml or HDM, 10 µg/ml), in the presence or absence of either MN or AM in ten atopic people with asthma (AA), nine atopic people without asthma (AN), and eight nonatopic normal subjects (N). Data are shown as medians with SEM. Probability values: #p < 0.05, compared with culture with unstimulated CD4+ cells alone; *p < 0.05, **p < 0.01, compared with culture with stimulated CD4+ cells alone; &p < 0.05, compared with CD4+ cell plus MN coculture with allergen stimulation.

Upon allergen stimulation, no significant increases were observed in IL-5 production in cultures of CD4+ cells alone in the two nonasthmatic groups, but there was an increase in the percentage of IL-5-producing cells in atopic asthmatics (to a median 1.5%; range, 0.5-6.2%; p < 0.05).

Effect of monocytes on interleukin-5 production by allergen-stimulated CD4+ T cells. The effect of peripheral blood MN on IL-5 production by CD4+ cells in response to allergen stimulation is also shown in Figure 1. The median percentages of IL-5-producing CD4+ cells were increased to 3.0% (range, 1.0-6.9%) in atopic people with asthma (p < 0.05) and to 2.0% (range, 0.6-3.9%) in atopic people without asthma (p < 0.05) by the addition of MN, but there was no change in nonatopic normal subjects. Likewise, for IL-5 levels in culture supernatants, there was an increase in atopic people with asthma to 1,048 pg/ml (range, 590-1,642 pg/ml; p < 0.01) and in atopic people without asthma to 560 pg/ml (range, 349-1,432 pg/ml; p < 0.01). A small increase in the median level of supernatant IL-5 to 167 pg/ml (range, 0.1-422.2 pg/ml) was also observed in nonatopic normal subjects, but this was not statistically significant.

Effect of alveolar macrophages on interleukin-5 production by allergen-stimulated CD4+ T cells. The effects of AM on CD4+ cell IL-5 responses were markedly different in the three subject groups (Figure 1). In atopic asthmatics, the addition of AM to allergen-stimulated CD4+ cell cultures greatly enhanced the median percentage of IL-5-producing cells to 4.6% (range, 2.7-11.8%; p < 0.01) and the median level of supernatant IL-5 to 1,290 pg/ml (range, 646-1,832 pg/ml; p < 0.01). However, in the two nonasthmatic groups, IL-5 responses with the addition of AM to CD4+ cell cultures were not significantly different from control values (p = 0.2 and 0.1 for nonatopic and atopic control subjects, respectively).

In the two atopic groups, the effects of AM on CD4+ cell IL-5 production differed from those due to MN. Further increases in the percentage of IL-5-producing CD4+ cells (p < 0.05) and the level of supernatant IL-5 (p = 0.05) were observed in the cocultures with AM for atopic people with asthma, compared with the cocultures with MN. In contrast, for atopic people without asthma, AM tended to inhibit the production of IL-5 when compared to MN cocultures, although this effect did not quite reach conventional significance (p = 0.07 and 0.1, for the percent of IL-5-producing cells and supernatant IL-5 level, respectively).

Effect of monocytes or alveolar macrophages on interleukin-5 production by mitogen-stimulated CD4+ T cells. As shown in Figure 2, the effects of MN and AM on mitogen- induced IL-5 production by CD4+ cells, as assessed by flow cytometry, were qualitatively similar to those observed in the allergen-stimulated cultures for the three groups. The presence of MN in the cultures enhanced the median percentage of IL-5-producing CD4+ cells in all three groups, but this was significant only in atopic people without asthma (p < 0.05). In contrast, the presence of AM significantly enhanced the median percentage of IL-5-producing CD4+ cells from 1.8% (range, 0.05-6.8%) to 4.0% (range, 0.8-10.7%) in atopic asthmatics (p < 0.05), but no change was induced by AM in the two nonasthmatic groups.

View larger version (26K): [in this window] [in a new window]   Figure 2.   Percentages of IL-5-producing cells in CD4+ cell cultures stimulated by PMA (10 ng/ml) and ionomycin (1 µM), in the presence or absence of either MN or AM in ten atopic people with asthma (AA), nine atopic people without asthma (AN), and eight nonatopic normal subjects (N). Data are shown as medians with SEM. Probability values: *p < 0.05, compared with culture with stimulated CD4+ cells alone; &p < 0.05, compared with CD4+ cell plus MN coculture with mitogen stimulation.

Compared with the effects of MN, AM further increased the median percentage of IL-5-producing CD4+ cells in response to mitogen in atopic people with asthma (p < 0.05), but in the two nonasthmatic groups, AM again inhibited IL-5 responses (p = 0.05 for nonatopic control subjects; p < 0.05 for atopic control subjects).

Effects of alveolar macrophage-derived cytokines on interleukin-5 production by allergen-stimulated CD4+ T cells. To investigate the effects of AM-derived cytokines on IL-5 production, a blocking study was performed. As shown in Figure 3a, the median percentage of IL-5 producing cells in cocultures of CD4+ cells with AM was reduced from 3.5% (range, 1.1-11.0%) to 2.4% (range, 0.9-5.2%), 2.2% (range, 0.6-3.2%), 2.1% (range, 1.0-3.7%), and 2.9% (range, 1.3- 5.2%) by anti-IL-1, anti-IL-1, anti-IL-6 and anti-TNF-, respectively, but only the decrease for anti-IL-1 was statistically significant (p < 0.05). A significant decrease in the median levels of IL-5 in these culture supernatants was induced by anti-IL-1, from 1,560 pg/ml (range, 1,040-2,126 pg/ml) to 633.9 pg/ml (range, 300-780 pg/ml; p < 0.05), and by anti-IL-6, to 580.0 pg/ml (range, 200-1,060 pg/ml; p < 0.05). No significant change in supernatant IL-5 level occurred with anti-IL-1 or anti-TNF- (Figure 3b). For cell cultures from three atopic people with asthma, the combined effect of the four monoclonal antibodies was tested, but no additive effect was observed. The addition of the cytokine-blocking antibodies to CD4+ cell plus AM cocultures caused a decrease in the percentages of CD4+ CD25+ cells (Figure 3c), but this was significant only for the anti-IL-1 and anti-IL-6 antibodies (p < 0.05).

View larger version (30K): [in this window] [in a new window]   Figure 3.   Effect of addition of IL-1, IL-1, IL-6, and TNF- antibodies to CD4+ cell and AM cocultures stimulated with allergen in seven atopic asthmatics. Data are shown as medians with SEM. Percentages of IL-5-producing cells (a), levels of IL-5 in culture supernatants (b), and percentages of "activated" CD4+ CD25+ cells (c) are shown. Probability values: *p < 0.05, compared with CD4+ cell plus AM coculture without blocking antibodies.

There was a significant relationship between the percentages of IL-5-producing CD4+ cells and the levels of IL-5 in the supernatants (r = 0.566, p < 0.001) when the data from all culture studies were combined.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our studies demonstrate for the first time that IL-5 production by peripheral blood CD4+ T cells, either in response to allergen or to mitogen, is differentially modified by peripheral blood MN and AM in atopic people with asthma, atopic people without asthma, and nonatopic normal subjects. In response to allergen stimulation, IL-5 production was significantly enhanced by the addition of MN to CD4+ cell cultures in atopic people with or without asthma, but not in nonatopic normal subjects. In mitogen-stimulated MN cocultures, there was a small increase in IL-5 production in all three groups. In contrast, the addition of AM to parallel cultures significantly increased IL-5 production in atopic people with asthma, but not in atopic people without asthma or nonatopic normal subjects. Furthermore, in atopic people with asthma, IL-5 production in cocultures of CD4+ cells plus AM, stimulated by allergen or mitogen, was significantly higher than that in the cocultures with MN. However, in the two nonasthmatic groups, IL-5 production from AM cocultures was lower than that from MN cocultures. These data demonstrate that the capacity of allergen- or mitogen-stimulated CD4+ cells to produce IL-5 in vitro was markedly, but differentially, influenced by the interaction with MN or AM.

As in our previously published studies (6, 7, 9), we have found a significant difference in spontaneous IL-5 production by resting CD4+ cells among the three groups studied. Atopic people with asthma had the highest IL-5 production in unstimulated cultures of CD4+ cells alone, whereas nonatopic normal subjects showed the lowest, with atopic people without asthma at an intermediate level. These observations may well reflect different in vivo activation of these cells in the three groups. It has been clearly shown that in vivo production of IL-5 by T cells is closely associated with clinical characteristics of asthma (1). However, it is not clear whether T cells from people with asthma, compared with those from people without, have an intrinsically higher responsiveness that causes IL-5 overproduction. In the present study, when CD4+ cells were cultured alone, allergen stimulation did not result in a significantly greater elevation of IL-5 production in atopic people with asthma than in atopic people without asthma, although there was a small increase in the percentage of IL-5- producing cells in the former. This was also the case for mitogen stimulation when comparing atopic people who have asthma with atopic people who don't or with nonatopic normal subjects. These limited differences under single-cell culture conditions might suggest that CD4+ T cell overproduction of IL-5 in vivo in atopic asthmatics is unlikely to be due only to an intrinsic increase in T cell reactivity, but rather is associated with additional local regulatory factors.

This concept was further supported by the observations from cocultures of CD4+ cells plus MN. In the presence of MN, the actual increases in IL-5 production by allergen-stimulated CD4+ cells from atopic people with or without asthma were almost identical. Thus, the intrinsic potential of CD4+ cells to produce IL-5 in response to allergen appeared again to be similar between atopic people with asthma and atopic people without asthma when uniformly conditioned by their respective MN.

Elevation of IL-5 production by the addition of MN to allergen-stimulated CD4+ cell cultures was probably achieved either by priming naive CD4+ cells or by activating memory CD4+ cells in an allergen-specific manner restricted to the class II major histocompatibility complex (MHC; 15, 16). It is most likely that MN functioned as APC during this process i.e., they provided antigen-specific signals mediated through T-cell receptor activation by antigen/class II MHC and nonantigen-specific costimulatory signals. These signals are essential for generating fully activated effector CD4+ cells (14, 18, 29, 30). Supporting this idea, allergen stimulation led to a much smaller increase, or indeed, no change in production of IL-5 by CD4+ cells in both atopic groups when cultured alone. Furthermore, the observation that the addition of MN to mitogen-stimulated cultures heightened the percentages of IL-5-producing CD4+ cells in all three groups may reflect the activity of nonantigen-specific costimulatory signals expressed by MN.

Interleukin-5 production by CD4+ T cells, whether stimulated by allergen or mitogen, was not augmented by autologous AM in either nonasthmatic group. Furthermore, IL-5 production in these AM cocultures for these two groups was lower than that in the parallel cocultures with MN. The fact that AM from these subjects functioned poorly as APC when compared with MN may represent a local inhibitory protective mechanism in the airways. Previous studies have suggested that AM downregulate cellular immune responses in the lungs of normal individuals. Alveolar macrophages suppress T-cell blastogenic responses to phytohemagglutinin (21, 22) and a variety of antigens (31) in physiological situations. It has been noted that these immunosuppressive properties are particular to AM, especially as compared with autologous blood MN (21, 22, 26). In the present study, we did not find a direct inhibition of IL-5 production when AM were added to CD4+ cell cultures from the two nonasthmatic groups. This finding appears to be compatible with the studies suggesting that AM inhibit T-cell activation at an early stage before cell division (21, 23), selectively suppressing their proliferation and expansion, but without affecting T-cell cytokine production (23, 32).

In striking contrast to the effects of AM from the two nonasthmatic groups, AM from atopic people with asthma showed more stimulation than MN on CD4+ cell IL-5 production. These data suggest a key role for AM in immunopathogenic responses, characterized by increased IL-5 expression and production (2, 7), in the airways of atopic people with asthma. Recent studies have demonstrated that different APC-related factors in the microenvironment of a T cell may determine its profile and quantity of cytokine expression and production (16, 17, 19, 29, 30, 32). We could speculate that AM from atopic people with asthma upregulated the IL-5 response of CD4+ cells by expressing and producing the molecules that favor a Th2 response or selectively favor a specific IL-5 response. It has been shown more recently that IL-4 and IL-5 are not always coordinately regulated and produced (33, 34).

This study was unable to fully define the principal AM- derived mechanisms that amplify IL-5 production in atopic people with asthma. Our blocking studies demonstrated an important role for IL-1 and IL-6 in promoting IL-5 secretion by allergen-stimulated CD4+ T cells. A similar, but less marked, effect of these AM-derived cytokines on the number of IL-5-producing CD4+ T cells was seen. Although IL-1 and IL-6 have been previously shown to be critical mediators in T-cell activation and proliferation, and contribute to the onset of a Th2 response in maturing human T cells (15, 20, 35), it is unclear whether or not other AM-derived factors participate or have a more important role in regulating the differentiation of allergen-stimulated CD4+ T cells. Recent studies have demonstrated that IL-12 (an accessory cell cytokine) plays a major role in the development of the Th1 phenotype, with a complementary suppression of Th2 responses (17, 19, 20). A similar role has also been described for the costimulatory CD28 ligand B7-1 (16). The relative contributions of these factors, compared with IL-1 and IL-6, in local regulation of T-cell responses in the airways of atopic people with and without asthma are under current investigation.

Finally, the fact that there was a significant difference in IL-5 production by CD4+ cells in response to allergen stimulation between atopic people without asthma and nonatopic normal subjects, most marked in the MN cocultures, demonstrated a higher responsiveness of CD4+ cells to allergen in atopic people without asthma than in nonatopic normal subjects. Our present study suggests that the immunosuppressive role of AM in atopic people without asthma may be crucial in downregulating IL-5 production in the airways to a relatively low level, insufficient to induce the pathogenic degree of local eosinophilia necessary to elicit clinical asthma.