Although our knowledge about the initiation of the allergic response has rapidly expanded in recent years [9], nothing is known about why the inflammatory response cannot be turned off in atopic persons. Because immune reactions in general are potentially dangerous in a physiologic setting [10], they must be carefully controlled (if the antigenic stimulus becomes too great) or extinguished (if the damaging agent is successfully eliminated). One important mechanism of lymphocyte control is programmed cell death (apoptosis), which occurs in every immune response [11, 12], is energy dependent, and is associated with endonuclease activation. The biochemical hallmark of apoptosis is the fragmentation of chromatin into oligonucleosomes with subsequent changes in nuclear structure (picnosis) and alterations in the surface membrane of the apoptotic cell; this allows the cell to be eliminated from the site of inflammation by phagocytosis without damage to the surrounding host tissue [12].

T cells have at least two apoptotic pathways: active death, which is antigen driven, and passive death, which occurs at the end of an immune response and is due to lymphokine withdrawal or other mechanisms [13]. These two forms of death are molecularly distinct because the cell surface molecule Fas/APO-1 (CD95), a 45-kD protein belonging to the tumor necrosis factor (TNF) receptor family [14], and TNF itself are major participants in active, but not passive, death. The cross-linking of Fas receptor with its recently identified ligand (Fas ligand) leads to the induction of apoptosis in lymphocytes [15] and provides a mechanism for the removal of antigen-activated T cells [16, 17]. This leads to resolution of the inflammatory response [18], protecting the host against the detrimental effects that would ensue if cell disintegration or necrosis were to occur. Spontaneous remission of respiratory allergic disease invariably following withdrawal of allergen [19, 20] underlies the mechanism of passive death.

Because Fas and TNF are major participants in active antigen-driven death [13], we investigated whether T cells from the lungs of untreated persons with asthma express Fas receptor. We show that the persistence of allergen-specific T cells at the mucosal surfaces of atopic persons results from local defective surface expression of Fas receptor and subsequent impairment in active, allergen-driven cell death.

Methods

Top Methods Results Discussion Author & Article Info References

Study Participants

Twelve patients (6 male and 6 female; age range, 8 to 54 years) with newly diagnosed, mildly symptomatic chronic asthma who underwent bronchoalveolar lavage were included in the study. At the time of recruitment, all patients had stable pulmonary function and an FEV₁ at least 70% of that predicted for their age and height. None was receiving inhaled or oral corticosteroids, sodium chromoglycate, theophylline, or ß₂ agonists. All had increased airway responsiveness to methacholine (concentration producing a decrease of 20% from baseline in FEV₁, <8 mg/mL); were atopic, as defined by positive skin prick tests done with purified Dermatophagoides pteronyssinus allergen extract (Neo Abello, Madrid, Spain); and had positive results on enzyme-linked immunosorbent assay (DPC Corp., Los Angeles, California) for circulating allergen-specific IgE. No patients smoked or had had an upper respiratory tract infection in the 8 weeks before the study.

Ten age- and sex-matched normal nonsmoking persons served as controls. No control had a history of asthma, systemic illness, or recent respiratory illness or any evidence of airway hyperresponsiveness on methacholine challenge. All had normal results on pulmonary function tests; none had positive results on skin prick tests to a panel of allergens that included house dust mites and pollens.

The study protocol was approved by the institutional review board for human studies of our university, and informed consent was given by each patient or by the patient's relatives if the patient was a child. The study was conducted according to local ethical committee guidelines and the principles of the Declaration of Helsinki.

Collection of Bronchoalveolar Lavage Fluid

Fiberoptic bronchoscopy was performed as reported elsewhere [21]. Lavage of the right middle lobe was done with three consecutive 20-mL aliquots of prewarmed physiologic saline. Recovered bronchoalveolar lavage cells were spun immediately at 200 g for 10 minutes at 4°C. After resuspension, cytospins were prepared for differential cell counts.

Reagents and Flow Cytometry

Immunophenotyping was done simultaneously on pulmonary and peripheral blood T cells from patients and controls by using the following monoclonal antibodies in various combinations: phycoerythrin-conjugated anti-CD3 (OKT3, Ortho, Raritan, New Jersey), which recognizes up to 90% of T cells bearing the ß or T-cell receptor heterodimer; anti-CD4 and anti-CD8 (OKT4 and OKT8, Ortho), which stain helper/inducer and cytotoxic T-cell subsets; anti-CD25 (anti-interleukin-2 receptor, Becton-Dickinson, Mountain View, California) and anti-HLA-DR, which recognize, respectively, the interleukin-2 receptor chain and the major histocompatibility complex class II molecule (both referring to activated T cells); anti-CD45R0 (Immunotech, Marseille, France), a common leukocyte antigen present on activated/memory T cells; fluorescein-conjugated anti-Fas receptor (Immunotech); and phycoerythrin-conjugated anti-TCR [T-cell receptor] delta1 (T-Cell Sciences, Cambridge, Massachusetts), a pan-reactive T-cell reagent. For staining, 5000 to 10 000 cells were resuspended in 50 µL of saline, incubated at 4°C for 30 minutes, washed, and analyzed by flow cytometry (FACScan, Becton-Dickinson). For analysis of two-color cytofluorometric data, an electronic gate was set on the lymphocyte population based on the forward-angle versus the right-angle light scatter histogram. Quadrant markers in fluorescence histograms were set by using matched isotype controls. The Lysis II program (Becton-Dickinson) was used to optimize gating of lymphocytes and to provide an objective way to exclude both debris (noncellular events due to particulate matter) and other cells from the lymphocyte gate.

Separation of Enriched T-Cell Populations

Bronchoalveolar lavage mononuclear cells from patients and controls were separated into sheep erythrocyte rosette-enriched (containing T cells) and rosette-depleted subsets [22]. On the basis of their reactivity with the anti-CD3 monoclonal antibody OKT3 (>98% by immunofluorescence), the rosetted mononuclear cells were considered to be highly purified T-cell subsets and were used for Fas messenger RNA (mRNA) polymerase chain reaction (PCR) analysis and apoptotic cell death experiments.

Modulation of Surface Fas Receptor on Pulmonary T Cells

T-cell-enriched populations from patients and controls were cultured in vitro on 24-well plastic plates (Kostar, Denmark) coated with anti-CD3 monoclonal antibody for 24 hours in the presence of recombinant human interleukin-2 (50 U/mL; Serotec, Oxford, United Kingdom) or interleukin-4 (400 IU/mL; Serotec). At the end of the culture time, cells were analyzed for surface expression of Fas molecule by double-color immunofluorescence.

Evaluation of Fas-Induced Programmed Cell Death

Enriched T lymphocytes derived from bronchoalveolar lavage were washed and resuspended in RPMI-1640 (Gibco, Grand Island, New York) supplemented with 10% fetal calf serum, 2 mmol of L-glutamine per L, 10 mmol of HEPES per L, 50 U of penicillin per L, and 50 µg of streptomycin at 10⁶ cells/mL. All samples were incubated for 18 hours with medium alone or with anti-Fas IgM (a specific antibody to the Fas receptor that can induce apoptosis of Fas⁺ target cells), washed, centrifuged at 200 g for 10 minutes, and dissolved in hypotonic lysing buffer (100 mmol of NaCl per L, 10 mmol of Tris per L, 1 mmol of EDTA per L, 1% sodium dodecyl sulfate, 200 µg of proteinase K per mL, pH of 7.5). A standard DNA electrophoresis assay was used for qualitative evaluation of apoptotic cells, as described elsewhere [23]. Quantitative analysis of spontaneous and anti-Fas-induced programmed cell death was done by cytofluorometry with a fluorochrome solution containing propidium iodide [23].

Fas Messenger RNA Expression

Expression of mRNA was detected by reverse transcription PCR. With the guanidinium isothiocyanate acid phenol method, RNA was isolated from bronchoalveolar lavage-enriched T cells from patients, controls, and the U937 cell line (as a positive control) [24]. First-strand complementary DNA synthesis was done by using total RNA, random hexadeoxynucleotide primers, and Moloney murine leukemia virus reverse transcriptase (Gibco BRL, Life Technologies, Milano, Italy), as described elsewhere [25]. Primers for Fas and ß-actin PCR amplification were designed according to previously published sequences [25]. After amplification, one tenth of the PCR product was run on 2% agarose gel and stained with ethidium bromide.

Statistical Analysis

Patients were stratified according to age. Pairwise comparisons for measuring the significance of the differences among mean values (±SD) calculated in the various groups (patients with asthma compared with controls) were done by using the Mann-Whitney U test. P values less than 0.05 were chosen for rejection of the null hypothesis. Statistical computations were done by using SPS, version 4.0 (SPS, Chicago, Illinois).

Role of Funding Sources

The funding sources, including the Italian Ministry of Scientific Research and Technology, had no role in the collection, analysis, or interpretation of the data or in the decision to submit the final manuscript for publication.

Results

Top Methods Results Discussion Author & Article Info References

Flow cytometric analysis of bronchoalveolar lavage-enriched T cells from the untreated patients with asthma were stained with hypotonic propidium iodide solution and showed a negligible proportion of hypodiploid DNA peaks (ranging from 2% to 5%; data not shown) in all samples that were tested basally in medium alone. Therefore, we excluded the presence of significant spontaneous apoptosis due to bronchoalveolar lavage sampling, a phenomenon that cannot be shown by the trypan blue dye exclusion test because apoptotic cells, although they have damaged DNA, can exhibit intact cytoplasmic membrane. Although more than 50% of normal T lymphocytes died after exposure to anti-Fas monoclonal antibodies, the same antibodies did not induce substantial cell death in purified pulmonary T cells from patients with asthma (Figure 1).

View larger version (32K): [in this window] [in a new window]   Figure 1. In vitro induction of programmed cell death by IgM anti-Fas monoclonal antibody. Top. Quantitative assay with propidium iodide staining and flow cytometry. Enriched bronchoalveolar lavage T cells from patients with asthma and from controls were incubated for 24 hours with IgM anti-Fas monoclonal antibody, 0.5 µg/mL (Immunotech, Marseille, France), and then stained with hypotonic propidium iodide solution. Culturing in medium alone for 24 hours did not induce apoptosis in significant percentages of cells, but in vitro stimulation with anti-Fas monoclonal antibody determined the appearance of significant percentages of cells with hypodiploid DNA in samples from controls. Black bars represent patients with asthma; white bars represent controls. *P < 0.001 for comparison with patients with asthma. Bottom. A representative qualitative assay of DNA fragmentation of T-cell samples from patients with asthma and controls loaded into wells of 1% agarose gel. Ethidium bromide staining was used to visualize DNA. Lane 1: control, medium alone; lane 2: patient with asthma, anti-Fas monoclonal antibody; lane 3: control, anti-Fas monoclonal antibody; lane 4: patient with asthma, anti-Fas monoclonal antibody; lane 5: control, anti-Fas monoclonal antibody. A clear DNA ladder, characteristic of apoptosis, is seen for anti-Fas-treated T cells from controls but not patients with asthma. Numbers in first column are the molecular weight.

The continued viability of the patients' T cells was not due to lack of activation: Surface expression of activation or memory markers, such as CD25 (22.5% ± 6.3%), HLA-DR (7.4% ± 3.5%), and CD45R0 (36.5 ± 5.8), was greater in patients than in controls (CD25, 5.2% ± 1.4% [P < 0.001]; HLA-DR, 2.5% ± 1.6% [P < 0.05]; CD45R0, 18.7% ± 4.2% [P < 0.05]). This suggests that Fas receptor expression is defective on pulmonary T cells from patients with asthma.

Cell-surface cytofluorometry showed that Fas receptor was not expressed by pulmonary T cells from persons with asthma, but it was found on CD3⁺ T cells from controls that were double-stained with anti-CD3 and anti-Fas monoclonal antibodies (Table 1). Like all other CD3⁺ T cells, the bronchoalveolar lavage T lymphocytes from our patients with asthma did not express the membrane Fas receptor. The membrane defect seemed to be confined to pulmonary T cells because the percentage of Fas expression shown by circulating CD3⁺ cells from patients was similar to that in controls (Table 1).

In experiments to test the expression of Fas mRNA, pulmonary T lymphocytes were examined by reverse transcription PCR. Because Fas mRNA was detected in all control samples but in no samples from patients with asthma (Figure 2), our results strongly support the data from the immunofluorescence studies.

View larger version (57K): [in this window] [in a new window]   Figure 2. Reverse transcription polymerase chain reaction. Top. Fas messenger RNA (mRNA) isolated from enriched bronchoalveolar lavage T cells. Lane 1: control; lanes 2, 3, and 4: patients with asthma; lane 5: control; lane 6: U937 cell line as a positive control. Samples from patients with asthma, unlike those from controls, lack Fas mRNA. Bottom. ß-actin mRNA from the same samples. Numbers in first column are the molecular weight.

In testing the effect of Th2-type cytokines, such as interleukin-4, on downregulation of surface Fas expression on in vitro activated enriched pulmonary T cells, we found that anti-CD3-triggered pulmonary T cells alone from both patients and controls can upregulate Fas receptor. However, although interleukin-2 does not affect expression of the Fas receptor, interleukin-4 incubation caused a downmodulation of the receptor in allergic patients (Figure 3).

View larger version (22K): [in this window] [in a new window]   Figure 3. Modulation of surface Fas expression by cytokines. Representative FACscan profiles (Becton-Dickinson, Mountain View, California) obtained in a patient with asthma (top) and a control (bottom). Pulmonary enriched T cells were cultured for 24 hours with plastic-coated anti-CD3 monoclonal antibody plus recombinant human interleukin-2, 50 U/mL (top left and bottom left), or recombinant human interleukin-4, 400 U/mL (top right and bottom right). Subsequent surface Fas expression was measured on cytofluorometry by gating positively stained cells (M1) in the FL2 channel of the FACScan. Relative percentages and fluorescence intensity are shown. In vitro anti-CD3 stimulation was able to upregulate Fas molecule on enriched T cells. However, although interleukin-2 maintained upregulation of the molecule, interleukin-4 determined its partial downregulation. This was seen in patients with asthma but was almost absent from controls. One experiment representative of all those performed on both patients and controls is shown.

Discussion

Top Methods Results Discussion Author & Article Info References

We show that Fas mRNA and its surface receptor (CD95/APO-1) are defectively expressed on the pulmonary T lymphocytes of patients with allergic asthma, a condition that impairs in vivo activation-induced programmed cell death. This defect is not inherited: We show the presence of Fas receptor on the surface of peripheral blood T lymphocytes from patients with asthma and the upregulation of Fas receptor expression on pulmonary T cells after in vitro stimulation with anti-CD3 and interleukin-2. On the contrary, incubation with interleukin-4 downmodulated surface Fas expression, indicating that the local presence of large amounts of this cytokine may explain the observed in vivo findings.

The pathogenesis of bronchial asthma involves several genetic and environmental factors. Important roles are played by inherited features (such as bronchial hyperreactivity, elevated specific IgE levels, and interleukin-4 production [4], many of which map on chromosomes 5q31-q33, 11q, and 14q [3]) and the cytologic characteristics of allergen-driven inflammatory bronchial infiltrate [26]. These characteristics may vary according to the clinical severity of the disease or allergen exposure, but they invariably correlate with the presence of eosinophils [27] and allergen-specific Th2-type T lymphocytes [6]. These often bear the T-cell receptor [28].

Pulmonary allergen-specific T lymphocytes have the characteristics of chronically activated T cells, but, lacking the Fas receptor on their surface, they are resistant to in vitro and perhaps in vivo delivered apoptotic signals. However, lung transplantation studies [29] have shown that allergic bronchial asthma, even if genetically inherited [3-5], is an organ-specific disease. In fact, pulmonary T lymphocytes from affected patients are phenotypically [26] and functionally [30] different from those circulating in the bloodstream. It is therefore possible that an environmental factor acting at the level of the mucosal surface, such as inhaled allergens, air pollutants [31], common respiratory viruses [2], and a specific cytokine [1], could negatively influence the ability of mucosal T lymphocytes to express Fas mRNA and Fas receptor on their surfaces after in vivo allergen activation. Our results support this hypothesis: The presence of interleukin-4 significantly reduces surface expression of Fas on in vitro anti-CD3-triggered T cells from patients with asthma, whereas circulating T cells from the same patients express the receptor in a way similar to that of controls.

The mechanism by which Th2 cytokines, particularly interleukin-4 and interleukin-5, may interfere with Fas mRNA synthesis or with the transcription of the surface protein is still not understood. Signaling through the interleukin-4 receptor leads to various biochemical reactions able to interfere with cell growth and gene expression. Interleukin-4 by itself does not induce proliferation of T cells but does regulate gene expression, including that of its own receptor [32]. Whether one of these genes controls expression of the Fas receptor deserves further study. The anti-apoptotic properties of interleukin-4 have been described in various experimental models [33, 34]; therefore, downmodulation of Fas receptor would not be surprising.

Increasing evidence indicates that the effector mechanisms responsible for protective immunity can also injure host tissues. In some situations, specific immune responses have little or no protective value, and harmful consequences predominate [10, 13]. Examples of this include autoimmune diseases caused by a pathologic immune response against self-antigens and, as reported here, allergic diseases in which the harmful T cell-driven IgE-mediated immune response cannot be turned off because of the lack of Fas receptors on mucosal T lymphocytes.

Apoptosis is the mechanism that controls the size of a given cell population in order to allow its proper function. Cell-mediated and humoral response effector cells die through apoptosis [35], and expression of Fas receptor is important in this process [36, 37]. The physiologic role of the receptor in antigen-induced and harmful bystander T-cell activation in vivo is supported by the demonstration of the constitutive expression of the Fas receptor in normal human gut lamina propria T lymphocytes. The presence of these lymphocytes may provide a possible explanation of the mechanism for intestinal tolerance to ingested food antigens [38]. In addition, it was recently shown [39] that apoptosis contributes to the elimination of pulmonary lymphocytes during physiologic immune responses in mice and that this process is mediated by CD95 (Fas) expression on pulmonary CD3⁺ T cells. Our controls also showed "normal" Fas expression on the surface of pulmonary T cells; this provides a possible explanation for their "tolerance" of airborne allergens.

Hypoexpression of Fas mRNA and Fas receptor by pulmonary T cells, with subsequent impairment of active death in patients with allergic asthma, may be the molecular basis for the development and persistence of inflammatory infiltrate in the mucosa of the respiratory tract. However, from a clinical point of view, remission of disease depends on allergen avoidance [19, 20] or the administration of drugs capable of inducing the death of lymphocytes and eosinophils. We previously showed that the positive therapeutic effect of systemic corticosteroids in this disease is mediated by the induction of apoptotic cell death in allergen-specific CD4 (^-) CD8^- and CD4⁺ T lymphocytes [23, 26]. In addition, the induction of programmed cell death has been shown in eosinophils incubated in vitro with corticosteroids [40]. This mechanism of action may explain the clinical efficacy of corticosteroids in controlling the symptoms of asthma and preventing its exacerbation. Future therapeutic strategies should consider the molecular characteristics of pulmonary T lymphocytes and their role in the pathogenesis of bronchial asthma.

Drs. Forenza and Bertotto: Istituto di Clinica Pediatrica, Universita di Perugia, Policlinico Monteluce, I-06122 Perugia, Italy.

Dr. Bassotti: Dipartimento di Medicina Clinica e Sperimentale, Sezione di Gastroenterologia, Universita di Perugia, Policlinico Monteluce, I-06122 Perugia, Italy.