INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Atopy is a spreading condition defined by a susceptibility to becoming sensitized to allergens in an IgE-dependent way. Many studies in neonates (1, 2) and children (3) have suggested that an immune deviation to a T helper cell, type 2 (Th2) pattern of cytokines produced by T lymphocytes was the primary abnormality of atopy. A defective production of interferon- (IFN-) and/or an increased production of interleukin-4 (IL-4) by T lymphocytes characterize this biased immune response. The increase of IL-4/IFN- ratio would lead to aberrant IgE production in response to aeroallergens and sensitizations.

Atopy predisposes to atopic dermatitis, allergic rhinitis, and asthma. In asthma, an increase of cells containing messenger RNA (mRNA) coding for the Th2 cytokines and of corresponding proteins is found in the bronchi, in bronchoalveolar lavage fluid (BALF), and also in sera and blood cell cultures. This Th2 activation is mainly ascribed to CD4⁺ T cells (7). In this view, inflammation in asthma appears as a result of the Th2 commitment characterizing atopy, amplified by exposure to aeroallergens.

Except for cytokine assays in sera (6), most assays in blood were done in artificial culture media after purification of peripheral blood mononuclear cells (PBMC). This is likely to modify the cell activation and give an incomplete view of immune reactions. Flow cytometry provides a sensitive method for investigating cytokine production at the single cell level without prior purification (8). Therefore, flow cytometry permits the investigation of cytokine production by cells being cultured in blood. This prompted us to reassess in whole blood cultures the Th1/Th2 paradigm in adult atopy and asthma.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Sixty-nine subjects (23 males and 46 females, mean age 39 ± 13) were included in this study. Thirty-seven were consecutive patients referred to the allergy clinic for a history compatible with asthma. Thirty-two were healthy control subjects. A blood sample was collected by peripheral venous puncture in each patient for cell culture. The ethical committee of Marseille-I approved the protocol, and each subject gave an informed consent. Patients were not included if they were smokers, had displayed an infectious episode, or taken systemic corticosteroids or antihistamines during the last month.

Diagnosis of Atopy

The diagnosis of atopy was allowed if at least one skin prick test to a common environmental aeroallergen from a standard battery of 35 extracts (Laboratoire des Stallergenes, Paris, France) was positive. The skin test battery included the following allergens: house dust (100 IR, i.e., index of reactivity that is a concentration which induced a 7-mm wheal in a group of sensitized individuals), Dermatophagoides pteronyssinus and farinae (100 IR), feather (1/10 wt/vol), cat and dog extracts (100 IR), German cockroach (1/10 wt/vol), Alternaria, Aspergillus, Cladosporium, and Penicillium (1/10, wt/vol), mixed grass pollen (poa, fescue, timothy, rye, orchard) (100 IR), Bermuda grass (1/20 wt/vol), Parietaria (1/20 wt/vol), alders (100 IR), birch tree (100 IR), hornbeam, hazel, olive tree, ash, privet, oak, mimosa, poplar, false acacia, lime tree, mulberry, nettle tree (1/20 wt/vol), mixed weeds (100 IR) including ragweed, mugwort, amaranth, goosefoot, sorrel, and plantain. Prick tests were considered positive if the wheal diameter 20 min after allergen injection reached at least half the wheal diameter induced by codeine phosphate and more than the wheal induced by saline 9%. In any case, a positive (codeine phosphate 9%) and a negative test were performed.

Diagnosis of Asthma

The diagnosis of asthma was confirmed on the basis of a history of dyspnea and wheezes, either with a reversible airflow obstruction or a positive methacholine challenge test. Reversible airways obstruction was characterized by a 20% increase in forced expiratory volume in one second (FEV₁) after the inhalation of 200 µg of albuterol. Methacholine challenge tests were performed when spirometric data were normal. Cumulative doses of methacholine were administered through a dosimeter (ME-FAR dosimeter; Elletromedically, Brescia, Italy); specific airway resistance (SRaw) and FEV₁ measurements were made after each dose in an 830-L constant body plethysmograph (model Master Lab Jaeger, Wurzburg, Germany). Bronchial hyperreactivity was defined by a 100% increase of SRaw at 200 µg of methacholine or less. The dose of methacholine inducing a decrease of 20% of the FEV₁ (PD₂₀) was determined for each positive test.

Severity of asthma was graded from I (intermittent asthma) to IV (severe persistent asthma) according to the classification of the Global Initiative for Asthma (GINA) (9).

Control Subjects

The control groups consisted of healthy volunteers without previous history of asthma or allergic disease. The absence of asthma was confirmed in these subjects by the stability of FEV₁ and SRaw after the inhalation of a total dose of 500 µg of methacholine. The negativity of all 35 skin prick tests confirmed the absence of atopy.

Total Serum IgE and Blood Eosinophil Measurements

Serum IgE were assayed on the sera of the patients using the paper radioimmunosorbent technique (PRIST) (Pharmacia, Uppsala, Sweden). Blood eosinophils were counted after coloration with eosin 2%.

Cell Culture

Each sample collected was divided in two parts, one for soluble cytokine assays in culture supernatants using a flow cytometer microsphere-based technique and the other for intracytoplasmic detection of cytokines. Cell culture was performed at 37° C in an atmosphere containing 5% CO₂. For soluble cytokine assays, cells were cultured in 6-well plates in a volume of 1 ml of whole blood, in the presence or not of 100 ng/ml of phorbol 12-myristate acetate (PMA) and 2 µg/ml of ionomycin. After 6 h of culture, cells were harvested, centrifuged, and the supernatants were frozen and kept at 80° C until assay. For intracytoplasmic stainings, cells were cultured in 96-well plates in a volume of 50 µl of whole blood in the presence or not of PMA and ionomycin at the same concentrations as above, and in the presence of brefeldin (20 µg/ml) and monensin (2 µmol/L). After 6 h of culture, cells were harvested, centrifuged, washed, and resuspended in phosphate-buffered saline (PBS).

Soluble Cytokine Assay

The flow cytometer microsphere-based assay (FMBA) allows simultaneous quantitative determination of multiple analyses in a sample. It makes use of sets of immunoassays performed on fluorescent and antibody-coated microspheres, developed by Immunotech (Marseilles, France) (10). Each microsphere set is designed for an individual assay and is coded by the intensity of the green fluorescent emitter. Each cytokine assay involves anticytokine antibody-coated microspheres and a complementary biotinylated antibody to capture the cytokine in the sample. Detection is achieved with a red streptavidin conjugate emitter as reporter molecule. After the assay, the microspheres are individually analyzed in a flow cytometer. Measurement of green (FL1: 525 nm) and red (FL4: 675 nm) fluorescence yields both the specificity of the microsphere sets and the amounts of immune complex formed respectively on their surface. Simultaneous assays of IL-4, IFN-, and IL-12 (gp70) (Immunotech) were performed. Fifty microliters of sample were incubated in a membrane filter bottomed microplate (Nunc, Roskilde, Denmark) for 2 h with 10 µl of the mixture of coated microspheres (100 µg/ml) and 50 µl of the mixture of biotinylated antibodies (1 µg/ml) at room temperature with shaking. After two washes by filtration, 100 µl (0.5 µg/ml) of streptavidin-PE-Cy5 (PC5) (Immunotech) conjugate was incubated with microspheres for 30 min at room temperature with shaking, and washed twice by filtration. The microspheres were then transferred in phosphate buffer to tubes for analysis on a Coulter EPICS XL/MCL flow cytometer (Beckman Coulter, Miami, FL).

The photomultiplier tube (PMT) instrument was carefully set to provide optimal discrimination for FL1-coded microspheres, separated into discrete population, and the optimal range for FL4-binding. Acquisition was performed at high speed and forward scatter/side scatter (FSC/SCC) gating was used to ensure that only single beads were analyzed. Data for 500 events per cytokines microspheres set (1,500 gated events for our assay) were collected for each sample, which is enough to yield a good precision of the fluorescence signal. Indeed, each event represents an individual assay of the cytokine considered. The resulting FSC files were analyzed by in-house software which calculates mean fluorescence for each individual assay. Cytokine concentrations were then subsequently calculated using spline-fitted standard curves for each individual assay.

Staining

Cells were incubated with 10 µl of anti-CD3 monoclonal antibody (mAb) (IgG₁, clone UCHT1; Immunotech) or anti-CD8 (IgG₁, clone 13B8.2; Immunotech) for 15 min in dark. Cells were then fixed and permeabilized by using the IntraPrep reagents (Immunotech) as indicated by the manufacturer. Cells were then incubated for 15 min with 40 ng of fluorescein isothiocyanate (FITC)-coupled anti-IFN- mAb (clone 4S B3, IgG₁; Pharmingen, CA) and with 20 µl of PE-coupled anti-IL-4 mAb (clone 4D9, IgG₁; Immunotech). FITC- and PE-conjugated control isotypes (IgG₁, clone 679.1 Mc7; Immunotech) were used in control experiments. After a wash in PBS, cells were centrifuged and resuspended in 250 µl of PBS 0.5% formaldehyde. Blocking the reaction with increasing concentrations of recombinant cytokines (Pharmingen) assessed specificity of the stainings. Optimal concentrations of antibodies were determined in preliminary experiments.

Intracellular Cytokine Detection by Flow Cytometry

Flow cytometric data were immediately acquired on a Coulter EPICS XL/MCL flow cytometer (Beckman Coulter, Miami, FL), using XL II TM software. Dead cells, monocytes, and polymorphonuclear cells were excluded by forward and side scatter gating. Acquisition was gated on the CD3⁺ cells, or the CD8⁺_bright cells, and a minimum of 20,000 T cells was acquired. The gating on the CD8⁺_bright cells was set to exclude from the analysis natural killer (NK) and T cells, which express the CD8 subunit and are CD8⁺_low. Statistical markers were set using the negative control as reference. Results are expressed as the percentage of IFN-- or IL-4-expressing CD3⁺ or CD8⁺ cells.

Statistical Analysis

Results are expressed as mean ± SEM. Average percentages of positive cells and cytokine concentrations were compared between groups (controls, atopic asthmatics, nonatopic asthmatics, and atopic nonasthmatics) using the analysis of variance (ANOVA) if the normality test of Kolmogorov-Smirnov allowed the use of parametric tests. In other cases, data were log-transformed in order to obtain a Gaussian-shaped distribution and the ANOVA performed on log-transformed data.

When the ANOVA showed a statistical difference between groups, a multiple linear regression analysis was done to identify if atopy, asthma, or both could explain the variable studied. To further analyze the data, multiple comparisons versus controls using the Bonferroni t test were performed. Correlations between variables were done using Pearson's correlation test for data distributed normally and Spearman's nonparametric test in other cases. Analyses were performed with SigmaStat software for Windows (SPSS Inc., Chicago, IL).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of Patients, IgE Levels and Eosinophil Counts

Characteristics of the patients are indicated in Table 1. Among the 69 included patients, 31 were atopic asthmatics, six were nonatopic asthmatics, 13 were atopic nonasthmatics and 19 were nonatopic control subjects. None of the subjects was a smoker; none of the asthmatics was treated with oral or parenteral steroids or antihistaminics. IgE levels were significantly increased in atopic and asthmatic subjects compared with control subjects. Eosinophil counts were significantly higher in asthmatics than in control subjects, but not in atopic nonasthmatics. Eleven asthmatics were intermittent (Grade I), four were mild persistent (Grade II), 18 were moderate persistent (Grade III), and four were severe persistent (Grade IV).

IL-4 Production in Atopy and Asthma

IL-4 assay in whole blood culture supernatants. After stimulation, IL-4 concentrations differed between groups (p < 0.001). The multiple linear regression analysis showed that IL-4 production was significantly linked to atopy (p = 0.006), but not to asthma (p = 0.136) (Table 2).

Multiple comparisons versus controls showed that IL-4 concentrations were significantly increased in atopic asthma compared with control subjects (602.70 ± 52.05 pg/ml versus 304.7 ± 53.85 pg/ml, p < 0.001). In atopic nonasthmatics, IL-4 concentrations tended to be higher than in control subjects (503.8 ± 51 pg/ml) but the difference was not significant (p = 0.11). In nonatopic asthma, IL-4 concentrations were not different from control subjects (409.87 ± 147.19 pg/ml) (Figure 1A).

View larger version (14K): [in this window] [in a new window]   Figure 1.   IL-4 concentrations in stimulated culture supernatants. Cells were cultured in whole blood for 6 h in the presence of PMA and ionomycine, the supernatant harvested, and IL-4 assayed by the microsphere-based method. IL-4 concentrations are expressed in pg/ml. IL-4 concentrations were compared according to diagnosis (A) and were correlated to total IgE levels (B).

IL-4 concentrations were correlated with total IgE levels (r = 0.36, p = 0.0037) (Figure 1B).

Intracytoplasmic IL-4 detection. In unstimulated T cells, no IL-4 was detected by intracellular stainings (number of positive cells < 0.05%). After stimulation, IL-4 was detected in all samples. The number of IL-4-producing T cells varied from 0.05 to 1% of T cells, mean 0.32 ± 0.03. According to diagnosis, the number of IL-4-producing T cells was not different between groups. This number was 0.35 ± 0.06% in control subjects, 0.30 ± 0.07% in atopic nonasthmatics, 0.27 ± 0.08% in nonatopic asthmatics, and 0.32 ± 0.04% in atopic asthmatics. CD8⁺ T cells did not produce IL-4 either spontaneously or under stimulation.

IFN- Production in Atopy and Asthma

IFN- assay in whole blood culture supernatants. After stimulation, IFN- levels differed significantly (p = 0.002) between groups. The multiple linear regression analysis showed that IFN- concentrations were highly linked to asthma (p = 0.003) but not to atopy (p = 0.178) (Table 2).

Multiple comparisons versus control subjects showed that IFN- concentrations were higher in atopic and nonatopic asthmatics compared with controls (respectively, 29,157.7 ± 2,643.7 pg/ml, p = 0.05, and 43,508.7 ± 5,824.8 pg/ml, p < 0.001, versus 18,352.3 ± 3,103 pg/ml) (Figure 2A). The IFN- concentrations in nonasthmatic atopics were not different from control subjects (16,807.4 ± 3,780 pg/ml).

View larger version (16K): [in this window] [in a new window]   Figure 2.   IFN- production in stimulated whole blood culture. (A) IFN- assay in supernatants. Cells were cultured in whole blood for 6 h in the presence of PMA and ionomycine, the supernatant harvested, and IFN- assayed by the microsphere-based method. Results are expressed in pg/ml. (B) IFN--producing T cells. Cells were cultured in whole blood for 6 h in the presence of PMA and ionomycine and of brefeldin and monensin, harvested and stained with anti-CD3 and anti-IFN- antibodies. The proportion of IFN--producing T cells was determined among CD3+ cells by flow cytometry. (C ) IFN--producing CD8+ T cells. Cells were cultured as described in (B). Cells were harvested and stained with anti-CD8 and anti-IFN- antibodies. The proportion of IFN--producing CD8+ T cells was determined among CD8+bright cells by flow cytometry.

Intracytoplasmic IFN- detection: total T-cell population. In unstimulated T cells, no IFN- was detected by intracellular stainings (number of positive cells < 0.5%). After stimulation, IFN- was detected in all samples. The number of IFN--producing T cells varied from 4.2% to 41.9% of T cells (mean 15.8 ± 7.3%). T helper cell type 0 (Th0) cells, producing simultaneously IFN- and IL-4, were not detected.

According to diagnosis, the number of IFN--producing T cells was significantly different between groups (p = 0.014). The multiple linear regression showed that the number of IFN--producing T cells was positively associated with asthma (p = 0.004) and negatively with atopy (p = 0.005) (Table 2).

Multiple comparisons versus controls showed that the number of IFN--secreting T cells was not significantly different in patient groups compared with control subjects. This number tended to be higher in nonatopic asthma (25.7 ± 11.7%) and lower in nonasthmatic atopics (11.7 ± 1.4%) compared with controls (16.7 ± 8.9%, p = 0.221 and p = 0.146, respectively). In atopic asthmatics, it was similar to control subjects (15.6 ± 0.9) (Figure 2B).

In atopic nonasthmatics, a negative correlation was found between the number of IFN--producing T cells and the number of positive skin tests (r = 0.63, p = 0.03).

Intracytoplasmic IFN- detection: CD8⁺ T cells. After stimulation, the number of IFN--producing CD8⁺ T cells varied from 1.5% to 52% of T cells (mean 11.5 ± 1.4%). This number significantly varied according to diagnosis (p = 0.025). The multiple linear regression analysis showed that the number of IFN--producing CD8⁺ T cells was highly linked to asthma (p = 0.005) but not to atopy (p = 0.672) (Table 2).

Multiple comparisons versus control subjects showed that the number of IFN--producing CD8⁺ T cells was significantly higher in atopic asthmatics (15.07 ± 2.35% versus 6.74 ± 0.87%, p = 0.05). In nonatopic asthma, this number was higher than in control subjects (11.47 ± 2.06%), but the difference did not reach statistical significance. In atopic nonasthmatics, the number of IFN--producing CD8⁺ T cells was similar to control subjects (6.02 ± 0.97%) (Figure 2C).

The percentage of IFN--producing CD8⁺ T cells was related to various features of asthma: it was higher in moderate and severe than in intermittent and mild asthma (18.46 ± 2.7% versus 6.5 ± 6.2%, p = 0.02). It was positively correlated to eosinophil counts (r = 0.35, p = 0.007) (Figure 3A). It was inversely correlated to PD₂₀ (r = 0.59, p = 0.04) (Figure 3B).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Correlations between percentages of IFN--producing CD8+ T cells and blood eosinophilia, bronchial hyperresponsiveness, and IL-12 production. For determination of number of IFN--producing CD8+ cells, cells were cultured and stained as described in Figure 2C. (A) Correlation between IFN--producing CD8+ T cells and blood eosinophilia in asthma. Blood eosinophilia was determined by counting after coloration with eosin 2%. (B) Correlation between IFN--producing CD8+ T cells and bronchial hyperresponsiveness in asthma. Bronchial hyperresponsiveness was measured during a methacholine challenge test to determine PD20. (C ) Correlation between IFN--producing CD8+ T cells and IL-12 (p70) in blood culture supernatants. IL-12 was assayed in culture supernatants by the microsphere-based technique. The number of IFN--producing CD8+ T cells, blood eosinophilia, and PD20 are log-transformed to obtain a Gaussian-shaped distribution of data. IL-12 results are expressed in pg/ml.

IL-12 Production in Atopy and Asthma

The concentrations of IL-12 in whole blood culture supernatants were not different between groups. This number was 10.39 ± 1.34 pg/ml in control subjects, 29.55 ± 15.38 pg/ml in atopic nonasthmatics, 10.08 ± 3.14 pg/ml in nonatopic asthmatics, and 13.41 ± 1.75 pg/ml in atopic asthmatics.

There was a high correlation between IL-12 concentrations and percentages of IFN--positive T cells (r = 0.48, p = 0.0002) (Figure 3C).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this work, we have shown in whole blood culture a different pattern of T-cell activation between atopy and asthma. Atopy is linked to an increase of IL-4 in blood culture supernatants, as shown by the multiple linear regression analysis. This increase is correlated to total IgE levels, a feature linked to atopy. The source of IL-4 in atopy is unclear. We did not find any difference between groups with regard to T-cell IL-4 production by flow cytometry. However, a very low number of positive cells were detected, so that the power of the method was probably insufficient to detect differences. Therefore, we cannot exclude that T cells are involved in the overproduction of IL-4 in atopy. Because cells were cultured in whole blood, basophils or eosinophils could also produce IL-4.

A decreased capacity of T cells to produce IFN- in response to PMA and ionomycin was also associated with atopy, as shown by multiple linear regression analysis. Because this impairment was not found in CD8⁺ T cells, we can conclude that it is limited to CD4⁺ T cells. The direct study of this subset was not possible, because of the internalization of CD4 upon stimulation by PMA and ionomycin. The number of IFN--producing T cells was inversely correlated to the number of positive skin tests in nonasthmatic atopics. This suggests that this impairment of CD4⁺ T cells to produce IFN- is directly related to atopy. Furthermore, this correlation suggests that atopy is not an all-or-nothing condition, but that various degrees exist from nonatopic with negative skin tests and normal T-cell IFN- production, to highly atopic with many positive prick tests and a high impairment of T-cell IFN- production. Our results in atopy are consistent with previous studies in children, which suggest that atopy is the result of an impaired IFN- production leading to IL-4 synthesis and IgE- dependent sensitization to allergens (3).

In asthma, an increased production of IL-4 is found if atopy is present. As previously reported, it is not found in nonatopic asthma (11). Asthma is characterized by an overproduction of IFN-, due at least in part to CD8⁺ T cells. This results in an IFN- production by the whole T-cell population in atopic asthmatics that is similar to control subjects. The increased capacity of CD8⁺ T cells to produce IFN- is a characteristic of asthma. Indeed, it is related to severity, bronchial hyperresponsiveness, and blood eosinophilia.

These results are in discrepancy with previous data obtained in children in which a decrease of IFN- in peripheral blood from asthmatic atopics was found (3). This can be the result of a higher expression of atopy in children. Indeed, Krug and coworkers have found in a similar study a higher proportion of IL-4-producing T cells in atopic asthmatic children but not in adults (6). However, the cell culture conditions can also explain such discrepancies, because we cultured cells in whole blood albeit all discordant studies used separated PBMC. Jung and coworkers studied 23 patients (20 children and three adults in their third decade) displaying asthma or atopic dermatitis (12). As we did, they stimulated T cells with PMA and ionomycin and studied the IFN- production at the single cell level in CD3⁺ and CD8⁺ cells. They found a decrease in the number of T cells producing IFN-, and no elevation of CD8⁺ IFN--producing cells. The major difference in their protocol was that they have stimulated PBMC and not whole blood. We have chosen to culture cells in whole blood to keep T lymphocytes in their original environment, i.e., in the presence of cytokines and cells already present in the patient. T cells were therefore cultured in the presence of polymorphonuclear cells, notably eosinophils. These cells, which are the main effector cells in asthma, are able to produce IL-12 and stimulate CD8⁺ T cells to produce IFN- (13). This IL-12 could have stimulated the IFN- production from CD8⁺ T cells. Supporting this hypothesis, we found that the IFN--producing CD8⁺ population was related to blood eosinophilia and to IL-12 concentrations.

However, a series of previous studies are concordant with our results. Krug and coworkers performed a study in whole blood culture at the single cell level and in adult atopic asthma and did not find atopic impairment of IFN- production by T cells (14). These investigators did not study the CD8⁺ population. An increased concentration of IFN- in asthmatic sera (15), notably in severe asthma (16), was found previously. In addition, IFN- was found elevated in culture supernatants of asthmatic bronchoalveolar lavage (BAL) cells (17). However, in situ hybridization experiments failed to demonstrate an increase in IFN- mRNA-expressing cells (18). Krug and coworkers, in the study cited previously (14), showed at the single cell level an increase of the proportion of IFN--producing BAL T cells after stimulation by PMA and ionomycin. They did not study the T-cell subpopulations, so that it is not known whether the IFN--producing T cells in situ belong to the CD8⁺ subset. In atopic dermatitis (19), and in nasal polyposis (20), diseases having a pathogenesis very close to asthma, an IFN- production in situ was described.

The role of IFN--producing CD8⁺ T cells in asthma is unknown. Depending on the conditions of stimulation, CD8⁺ T cells are able to produce Th1-like or Th2-like cytokines, so that a distinction in Tc1 (c for cytotoxic) and Tc2 cells was proposed for this subset (21). In asthma and atopy, several studies have shown Tc2-cell activation, recently reviewed by Kalish (22). However, many experimental works have shown that CD8⁺ T cells can inhibit allergic reactions. Transfer of antigen-primed CD4⁺ but not CD8⁺ T cells induces allergic airway response (23). Moreover, in mice sensitized to ovalbumin, transfer of sensitized CD8⁺ T cells inhibits the IgE production and normalizes the airway responsiveness (24). In a model of in vivo induction of tolerance, McMenamin and Holt have shown that this tolerance is transferable to naive animals via CD8⁺ T cells (25). In these two latter works, CD8⁺ T cells were producers of IFN-. Several works have shown in mice that IFN- was a negative regulator of asthmalike inflammation. Administered by aerosol, IFN- inhibited the antigen- induced eosinophil recruitment by suppressing the infiltration of CD4⁺ cells (26). In another study, nebulized IFN- inhibited the antigen-induced airway hyperresponsiveness (AHR) (27). Lastly, Li and coworkers showed elegantly that mucosal IFN- gene transfer inhibits pulmonary allergic responses in mice (28). In humans, atopy patch tests revealed a strong Th1 response appearing 48 to 72 h after challenge, this reaction being preceded by a Th2 response in the first 24 h (29). This observation suggests that the Th1-like reaction is induced in response to the initial Th2 activation. Some studies suggest that IFN- could represent a therapeutic approach in atopic diseases. In severe atopic dermatitis, IFN- was administered successfully in several works (30). In mild asthma, no convincing results were shown, but studies are required in severe cases (31). Taken together, these data suggest that IFN--producing CD8⁺ T cells have a regulatory role in atopic diseases, suppressing the eosinophilic and IgE-dependent inflammation. In nonatopic asthma, these cells could have a similar role. Indeed, it was clearly shown that interleukin-5 (IL-5), a Th2, pro-eosinophilic cytokine inhibited by IFN-, was involved in nonatopic asthma as well as in atopic asthma (32).

The strong correlation between IL-12 concentrations in whole blood culture supernatants and the number of IFN-producing CD8⁺ T cells suggests that the activation of IFN--producing CD8⁺ T cells is dependent on IL-12 secretion. This result further supports the hypothesis of a regulatory role for IFN--producing CD8⁺ T cells in asthma. Indeed, IL-12 is a major stimulant of IFN- production, and was shown to inhibit antigen-induced AHR, inflammation, and Th2 cytokine expression in a mouse model of asthma (33).

The downregulation of allergic inflammation by IFN- was obtained after the administration of exogenous IFN-, CD8⁺ T cells, or IL-12, or after a gene transfer leading to a high local production of IFN-. However, the spontaneous production of IFN- is not necessarily beneficial in asthma. Indeed, several works suggest that endogenous IFN- could induce rather than suppress airway responsiveness. Hessel and coworkers (34) have elegantly shown that the eosinophilic infiltration of ovalbumin-sensitized mice can be dissociated from AHR by a prior treatment of animals with anti-IL-5 or anti-IFN- antibodies. In anti-IFN--treated mice, the response to methacholine was abolished, but not the eosinophil infiltration, albeit in anti-IL-5 treated animals, the eosinophil infiltration was inhibited, but not the AHR. The source of the IFN--inducing AHR could be CD8⁺ T cells, because depletion of CD8⁺ T cells in sensitized mice led to the incapacity to develop AHR (35). In this view, IFN- could be a link between allergic inflammation and bronchial hyperresponsiveness. Such a role for IFN- was previously proposed in skin between allergic inflammation and chronic eczema (19).

We did not culture cells in artificial media after prior purification. This allowed the study of mononuclear cells in presence of other circulating cells, i.e., in conditions supposed to better reflect the reality. It is of note, however, that a strong artificial cell activation was required to detect cytokine production. Further studies are therefore necessary, notably using allergen-specific stimulation.

In conclusion, the assessment of the Th1/Th2 paradigm in whole blood cell culture reveals that atopy is characterized by a Th2 immune deviation, with high IL-4 production and an impairment of CD4⁺ T cells to produce IFN-. This impairment is related to the number of positive skin tests. In atopic asthma, the Th2 deviation is associated with an overproduction of IFN-, notably by CD8⁺ T cells. This IFN--producing CD8⁺ population is related to the severity of asthma and probably controlled by IL-12. Further studies are required to elucidate to what extent IFN--producing CD8⁺ T cells in asthma downregulate the primary Th2 inflammation and what is their role in the inception of bronchial hyperreactivity.