- Email: [email protected]
T cells and eosinophils cooperate in the induction of bronchial epithelial cell apoptosis in asthma
T cells and eosinophils cooperate in the induction of bronchial epithelial cell apoptosis in asthma Axel Trautmann, MD,a Peter Schmid-Grendelmeier, MD,a Katja Krüger,a Reto Crameri, PhD,a Mübeccel Akdis, MD,a Ahmet Akkaya, MD,b Eva-B. Bröcker, MD,c Kurt Blaser, PhD,a and Cezmi A. Akdis, MDa Davos, Switzerland, Isparta, Turkey, and Würzburg, Germany
From athe Swiss Institute of Allergy and Asthma Research (SIAF), bthe Department of Pneumology, Süleyman Demirel University, and cthe Department of Dermatology, University of Würzburg. Supported by the Swiss National Foundation (32-65661-01 and 31-63381-00) and the Saurer Foundation. A.T. is a recipient of a grant from the Deutsche Forschungsgemeinschaft (TR460/1-1), P.S.-G. is a recipient of a grant for the promotion of academic young people of the University of Zurich. Received for publication August 20, 2001; revised October 25, 2001; accepted October 31, 2001. Reprint requests: A. Trautmann, MD, Swiss Institute of Allergy and Asthma Research (SIAF), Obere Strasse 22, CH-7270 Davos, Switzerland. Copyright © 2002 by Mosby, Inc. 0091-6749/2002 $35.00 + 0 1/83/121460 doi:10.1067/mai.2002.121460
Key words: Apoptosis, asthma, bronchial epithelial cell, eosinophil, Fas, interferon-gamma, T cell, tumor necrosis factor
Asthma is characterized by chronic inflammation of the bronchial mucosa where T cells and eosinophils are prominent. Shedding of bronchial epithelial cells (ECs) is an important histologic feature observed in bronchial biopsy specimens from asthmatic patients.1,2 Shedding of ECs is characterized by loss of the normal bronchial pseudostratified epithelium and the maintenance of a few basal cells on a thickened basement membrane.3 Activated T cells and eosinophils can be detected in the bronchial mucosa, and their numbers correlate with disease severity.4 Extensive evidence has been provided that T cells play a role in chronic asthma.5,6 A major advance in the understanding of the functional capabilities of eosinophils has been the demonstration that they are a source of pro-inflammatory cytokines including GMCSF, IL-4, IL-5, IL-6, IL-8, and TNF-α.7 The aim of this study was to investigate whether, and by which mechanism, T cells and eosinophils can cause damage to airway ECs. Recently we demonstrated that activated T cells infiltrating the skin in atopic dermatitis and allergic contact dermatitis induce apoptosis of keratinocytes and the characteristic spongiosis pattern in the epidermis.8,9 Less is understood regarding the specific roles of immune effector cells in the damage of bronchial ECs in asthma. The present study demonstrates that both T cells and eosinophils contribute to the induction of EC apoptosis by secretion of IFN-γ and TNF-α and triggering of death receptors on bronchial ECs. Moreover, we show that TNF receptor (TNFR) but not Fas is predominantly involved in apoptosis of ECs. Our data suggest that apoptosis represents an essential mechanism for the development of EC injury and shedding in asthma.
METHODS Subjects Approval for the use of human tissues was granted by the ethics committee of Davos. The fiberoptic bronchial biopsy specimens of 7 asthma patients (mild persistent asthma according to the Global Initiative for Asthma guidelines) were fixed in 4% paraformaldehyde and embedded in paraffin. Control bronchial mucosa was obtained from 3 patients with malignant lung tumors. T cells and eosinophils were purified from 11 atopic and 6 healthy donors. 329
Basic and clinical immunology
Background: Asthma is an inflammatory airway disease associated with an infiltration of T cells and eosinophils, increased levels of pro-inflammatory cytokines, and shedding of bronchial epithelial cells (ECs). Objective: Shedding of bronchial ECs is characterized by loss of the normal bronchial pseudostratified epithelium and the maintenance of a few basal cells on a thickened basement membrane. The aim of this study was to investigate whether, and by which mechanism, T cells and eosinophils can cause damage to airway ECs. Methods: Bronchial ECs, cultured and exposed to cytokines, eosinophil cationic protein, activated T cells, and eosinophils were studied for the expression of apoptosis receptors (flow cytometry, immunoblotting, and RNA expression) and for the susceptibility for undergoing apoptosis. In addition, bronchial biopsy specimens from patients with asthma were evaluated for EC apoptosis. Results: We demonstrate herein that the respiratory epithelium is an essential target of the inflammatory attack by T cells and eosinophils. Bronchial ECs underwent cytokine-induced cell death with DNA fragmentation and morphologic characteristics of apoptosis mediated by activated T cells and eosinophils. T cell- and eosinophil-induced EC apoptosis was blocked by inhibition of IFN-γ and TNF-α; the Fas ligand-Fas pathway appears to be less important. Recombinant eosinophil cationic protein induced mainly necrosis of ECs. Furthermore, we demonstrated in situ apoptotic features of ECs in bronchial biopsy specimens of asthmatic patients. Conclusion: T cell- and eosinophil-induced apoptosis represents a key pathogenic event leading to EC shedding in asthma. (J Allergy Clin Immunol 2002;109:329-37.)
330 Trautmann et al
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
Abbreviations used BEBM: Bronchial epithelial cell basal medium EC: Epithelial cell ECP: Eosinophil cationic protein HE: Hematoxylin-eosin TNFR: TNF receptor TUNEL: Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling
Reagents
Basic and clinical immunology
Human recombinant IFN-γ was a gift from Novartis Pharma (Basel, Switzerland); IL-5 and GM-CSF were from R&D Systems Europe Ltd (Abingdon, UK). Ethidium bromide was purchased from Sigma Chemical Co (St Louis, Mo). The following antibodies were used: anti-Fas-PE, anti-Fas ligand-FITC, (both from Alexis Corp., San Diego, Calif); neutralizing anti-TNF-α (R&D Systems Europe Ltd); anti-TNFR-I, anti-TNFR-II, and anti-TRAMP/deathR (DR) 3 (Oncogene, Cambridge, Mass); anti-TRAILR-I/DR4 and anti-TRAILR-II/DR5 (Calbiochem, San Diego, Calif); anti-βcatenin (Santa Cruz Biotechnology Inc, Santa Cruz, Calif); apoptosis-inducing anti-Fas mAb (CH-11, Immunotech, Marseilles, France); and neutralizing anti-IFN-γ (45-15, gift from Novartis Pharma). Human recombinant soluble Fas ligand, TRAIL, TNF-α, and TWEAK were from Alexis Corp. The caspase inhibitor AcAsp-Glu-Val-Asp-aldehyde (Ac-DEVD-CHO) was from Bachem AG (Bubendorf, Switzerland).
Culture of human airway ECs Primary normal human bronchial ECs were purchased from Clonetics Corp. (San Diego, Calif). Cells were grown in bronchial EC basal medium (BEBM) supplemented as prescribed by the manufacturer. ECs were subcultured and used between passages 3 and 7.
Flow cytometry ECs were incubated with appropriate FITC- or PE-labeled mAb for 30 min at 4°C and then fixed in 4% paraformaldehyde (pH 7.4). Negative control cells were prepared in a similar fashion with isotype control IgG1-FITC/PE mAb. Fluorescence intensity was measured by flow cytometry (EPICS XL-MCL flow cytometer, Beckmann Coulter Int. S.A., Nyon, Switzerland).
Isolation of
CD45RO+
effector T cells
Mononuclear cells were isolated by Ficoll (Biochrom KG, Berlin, Germany) density gradient centrifugation of venous blood. CD45RO+ T cells were isolated with use of the MACS system according to the instructions of the manufacturer (Miltenyi Biotec, Bergisch Gladbach, Germany) as described.8,9 T cells, 1 × 106 CD45RO+/1 mL BEBM, were stimulated with a combination of soluble anti-CD2 (0.5 (µg/mL), anti-CD3 (1.0 (µg/mL), and anti-CD28 (0.5 (µg/mL) mAb, and supernatants were harvested after 3 days.
Isolation of eosinophils Granulocyte-enriched cell pellets were isolated by Ficoll (Biochrom KG) density gradient centrifugation of venous blood and depleted of erythrocytes by use of hypotonic saline lysis. Eosinophils were negatively selected by using the MACS system (Miltenyi Biotec) as described.10 The viability of freshly isolated eosinophils was >95% (by trypan blue exclusion) and the purity was >99% (Diff-Quick staining; Dade AG, Düdingen, Switzerland). Eosinophils, 1 × 106/1 mL BEBM, were stimulated with IFN-γ (100
ng/mL), IL-5, or GM-CSF (25 ng/mL) or both, and supernatants were harvested after 1 day. To obtain eosinophil lysate, cells were washed twice and exposed to twice-repeated freeze-thaw cycles. Ultrasound destruction was applied for 1 min at 4°C, and lysates were centrifuged (14.400 U/min, 4 min) at 4°C.
Cloning, expression, and purification of human eosinophil cationic protein (ECP) Messenger ribonucleic acid from 107 purified eosinophils was isolated by using the Oligotex Direct mRNA Midi Kit (Qiagen AG, Basel, Switzerland) followed by oligo-dT primed reverse transcription. This cDNA was used to amplify a fragment containing the open reading frame encoding mature ECP by PCR using the following primers: 5´-primer 5´-CG GGA TCC AGA CCC CCA CAG TTT ACG AGG G and 3´-primer 5´ GGG AAG CTT TTA GAT GGT GGT ATC CAG GTG AAC TGG. The PCR product was gel purified and restricted with BamHI and HindIII incorporated into the amplification primers (underlined sequences). Restricted fragments were gel purified and ligated into BamHI/HindIII-restricted pQE30 (Qiagen AG) expression vector to produce hexahistidinetagged recombinant protein. The ligation mixture was transformed into Escherichia coli strain M15 by electroporation,11 and the nucleotide sequence was confirmed to code the reported amino acid sequence of human ECP.12 Correct constructs were inoculated into 2 × YT broth and induced with 2 mmol/L isopropyl-β-D-thiogalactoside at an OD595 of 0.7 to produce recombinant ECP. The protein was found in insoluble inclusion bodies, which were solubilized in guanidinium hydrochloride and one-step purified over Ni2+-NTA metal affinity chromatography resin (Qiagen AG) under denaturing conditions. The eluted protein was dialyzed against water containing 0.05% glacial acetic acid and 5 mmol/L EDTA. The protein was devoid of relevant contaminants as determined by SDS-PAGE.
Coculture experiments ECs were first seeded into 6-well or 48-well tissue culture plates in BEBM, and were incubated for 1 day to allow attachment and formation of a monolayer. Fifty percent of the medium was then removed and replaced with supernatants from CD45RO+ T cells or eosinophil lysates. CD45RO+ T cells or eosinophils were stimulated and washed twice before direct coculture with ECs.
Demonstration of EC viability and apoptosis in cultures and in situ. For the analysis of apoptosis in EC cultures, cells were examined under phase-contrast microscopy to evaluate the morphologic characteristics. EC viability was evaluated by means of ethidium bromide (1 µmol/L) uptake and flow cytometry. Annexin V is a phosphatidylserine binding protein that was also used to detect apoptotic cells.8,13 Hoechst staining was done as described.8,9 For the demonstration of apoptosis by DNA fragmentation, we isolated total genomic DNA of EC with the DNeasy kit according to the protocol of the manufacturer (Qiagen AG). The ApoAlert LM-PCR ladder assay kit (Clontech Laboratories AG, Basel, Switzerland) was used for the detection of nucleosomal ladders in apoptotic EC.14,15 After deparaffination and rehydration of the slides, apoptotic cells were identified in situ by staining double-stranded DNA breaks. Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) was employed as described.8,9,16
Immunoprecipitation and immunoblotting EC lysates were prepared, and immunoprecipitation was performed as described.9,17 Lysates were quantified with the Bio-Rad Protein assay (Bio-Rad Laboratories, Glattbrugg, Switzerland) to ensure equivalent gel loading. Immunoblots were probed with
A
B
C
D
FIG 1. A, Death receptor expression on ECs. ECs were cultured in the absence (Control) or presence of IFNγ, TNF-α, or both cytokines. The percentage of stained ECs after 1 day is indicated. B, Immunoprecipitation and immunoblotting with EC lysates were performed with the same mAb. β-Catenin is a control for equivalent gel loading. C, Gene expression array. cDNA probes of EC stimulated with 10 ng/mL IFN-γ and 10 ng/mL TNF-α were hybridized to gene-specific cDNA fragments spotted on GEArray membranes in duplicates. D, Sensitivity of ECs to apoptosis-inducing stimuli. ECs were treated for 3 days with apoptosis-inducing anti-Fas mAb, TNF-α, TRAIL, and TWEAK either in the absence (Alone) or presence of IFN-γ. EC viability (%) was determined by ethidium bromide exclusion and flow cytometry. Results shown in A-D represent 3 experiments. *P < .05.
appropriate antibodies followed by peroxidase-conjugated antimouse, anti-rabbit, or anti-goat IgG (Dako A/S, Glostrup, Denmark). Detection was performed by use of the ECL system (Amersham Pharmacia Biotech Ltd., Buckinghamshire, UK).
Gene expression array Biotinylated cDNA probes were hybridized to gene-specific cDNA fragments spotted on GEArray membranes as described by the manufacturer (GEA human TNF&Fas network, SuperArray Inc, Bethesda, Md). The relative expression level of genes was detected by chemiluminescence.
Statistical analysis Results are shown as mean ± SD. The paired Student t test was used for comparison of paired conditions.
RESULTS Apoptosis receptor expression and induction of apoptosis in bronchial ECs To assess EC surface molecules, we examined ECs by using flow cytometry (Fig 1, A). Both Fas and Fas ligand
are expressed on ECs and were upregulated mainly by TNF-α. TNFR-I and TNFR-II expression was higher and substantially upregulated by IFN-γ and TNF-α. By using immunoprecipitation and immunoblotting of EC lysates, we detected specific bands corresponding to Fas, TNFRI, and TNFR-II proteins (Fig 1, B). In contrast, apoptosis receptors DR3 (TRAMP), DR4 (TRAILR-I), and DR5 (TRAILR-II) were absent. IFN-γ– and TNF-α–stimulated ECs showed mRNA transcripts of Fas, TNFR-I, and TNFR-II. It is noteworthy that ECs express significant amounts of TNF-α mRNA (Fig 1, C). The apoptotic response of ECs was analyzed by using anti-Fas mAb, soluble Fas ligand, TNF-α, TRAIL, and TWEAK as apoptosis-inducing ligands. Anti-Fas mAb or soluble Fas ligand, either alone or after pretreatment with IFN-γ, did not trigger >5% of the ECs to undergo apoptosis (Fig 1, D). ECs showed significant death only in response to TNF-α after pretreatment with IFN-γ. ECs treated only with 10 ng/mL TNF-α showed a reduced viability compared to untreated cells.
Basic and clinical immunology
Trautmann et al 331
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
332 Trautmann et al
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
Basic and clinical immunology
Supernatants from stimulated T cells (1 × 106/mL) contain IFN-γ (156.3 ± 12.3 IU/mL) and TNF-α (140.3 ± 17.0 pg/mL), as determined by ELISA. The levels of IFN-γ released from activated CD45RO+ T cells obtained from atopic donors were slightly lower than those of nonatopic donors but still above 100 IU/mL. Control ECs had a polygonal shape and maintained confluence during 3 days. Supernatants from unstimulated T cells did not alter this morphology (Fig 2, A). After 3 days of treatment with supernatants from stimulated T cells, dead ECs appearing rounded and granulated were found to be detached from the bottom of the well. To demonstrate the type of cell death, cellular DNA was visualized with Hoechst staining. The nuclei of ECs were condensed and showed intense fluorescence when undergoing apoptosis. Quantitative assessment of viability confirmed the observed morphologic changes. As shown in Fig 2, A, EC apoptosis could be totally inhibited by pretreatment with neutralizing anti-IFN-γ mAb. The inhibitory effect of anti-TNF-α mAb was smaller but still significant. The caspase inhibitor Ac-DEVD-CHO prevented also EC apoptosis caused by stimulated T cells. EC apoptosis induced by supernatants from stimulated T cells was not prevented by blocking Fas or Fas ligand (data not shown). As a next step, we monitored the viability of ECs in direct contact with unstimulated and stimulated CD45RO+ T cells in cocultures. As displayed in Fig 2, B, destruction of the EC monolayer was observed after 3 days incubation with stimulated T cells. Anti-IFN-γ and antiTNF-α mAbs significantly rescued the EC from apoptosis.
Eosinophils induce both apoptosis and necrosis of ECs
FIG 2. Anti-IFN-γ and anti-TNF-α mAb inhibit EC apoptosis induced by stimulated T cells. A, Phase contrast morphology (×200) of ECs 3 days after exposure to supernatants from unstimulated and stimulated CD45RO+ T cells. In parallel, ECs were stained with Hoechst (×400) to visualize all nuclei. B, Phase contrast morphology (×400) of ECs after 3 days direct coculture with unstimulated and stimulated CD45RO+ T cells. A, B, EC viability (%) was determined by ethidium bromide exclusion and flow cytometry. Mean ± SD of four separate experiments are shown in parentheses. *P < .05.
T cell–mediated EC apoptosis To investigate the role of T cells in EC apoptosis, we exposed ECs grown as a monolayer to supernatants of unstimulated or stimulated CD45RO+ T cells for 3 days.
Eosinophils comprise major cells in the inflammatory infiltrate of asthma. Therefore, the role of eosinophils in EC apoptosis was further investigated. Eosinophils could only induce apoptosis in IFN-γ-pretreated ECs. After 3 days treatment with lysates from IFN-γ stimulated eosinophils, significant numbers of ECs demonstrated typical features of apoptosis (Fig 3, A). In contrast, IL-5 and GM-CSF did not activate eosinophils to induce EC apoptosis. Quantitative assessment of viability confirmed the observed morphologic changes. Lysates from IFN-γstimulated eosinophils (1 × 106/mL) contain 64.2 ± 1.3 pg/mL TNF-α, as determined by ELISA. The EC nuclei undergoing apoptosis were condensed and showed intense fluorescence with use of Hoechst staining. When pretreated with anti-TNF-α mAb, significantly less ECs died. In repeated experiments the apoptosis-inducing capacity of T cells and eosinophils did not differ between atopic and nonatopic donors. In addition to cytokines, eosinophils contain and release potent cytotoxic substances into the inflammatory microenvironment.7 Significant amounts of ECP were detected in lysates of eosinophils. Representative values for eosinophils, 1 × 106/mL, after 1 day culture were 0.83 µg/mL in unstimulated, 1.84 µg/mL in IL-5 stimulated, 0.77 µg/mL in INF-γ stimulated, and 1.86 µg/mL in INFγ– and IL-5–stimulated eosinophils. By using human recombinant ECP, we noted that the viability of ECs
Trautmann et al 333
Basic and clinical immunology
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
FIG 3. A, Anti-TNF-α mAb inhibits EC apoptosis induced by stimulated eosinophils. Phase contrast morphology (×200) of ECs pretreated with 10 ng/mL IFN-γ 3 days after exposure to lysates from unstimulated and stimulated eosinophils. In parallel, ECs were stained with Hoechst (×400) to visualize all nuclei. B, ECPmediated EC necrosis. Phase contrast morphology (×400) of ECs 6 hours after exposure to the indicated concentrations of ECP. C, Annexin V/propidiumiodide staining of ECs 6 hours after exposure to 10 µg/mL ECP indicate necrotic cells in quadrant 4. A, B, EC viability (%) was determined by ethidium bromide exclusion and flow cytometry. Mean ± SD of 3 separate experiments are shown in parentheses. *P < .05.
decreased significantly at 10 µg/mL doses (Fig 3, B). The type of cell death was mainly necrosis, as determined with annexin V/propidiumiodide staining. As early as 6 hours after ECP treatment, ECs were stained with propidiumiodide but did not show any membrane phosphatidylserine inversion (Fig 3, C).
Assessment of EC apoptosis by use of DNA fragmentation and annexin V staining A hallmark of apoptosis is the characteristic DNA fragmentation (200 to 500 bp) in gel electrophoresis. Viable control ECs did not exhibit any fragmentation (Fig 4, A, control). Deoxyribonucleic acid fragmentation as a characteristic feature of apoptotic cells was observed 3 days after addition of stimulated T cell supernatants. Similarly, DNA fragmentation was induced by IFN-γ and TNF-α treatments. It is remarkable that DNA fragmentation was absent in ECs in which stimulated T cell-induced apopto-
sis had been blocked by anti-IFN-γ. ECs pretreated with IFN-γ and exposed to lysates from IFN-γ-stimulated eosinophils also showed DNA fragmentation (Fig 4, A). Eosinophil-induced DNA fragmentation in ECs was inhibited by neutralizing TNF-α. Because annexin V is a marker of apoptosis, we also stained ECs which were exposed to T cells and eosinophils with annexin V-FITC. In correlation with the DNA fragmentation data, there was substantial staining with annexin V in ECs, which were exposed to stimulated T cells and eosinophils (Fig 4, B).
Demonstration of EC apoptosis in patients with asthma These results demonstrated that both T cells and eosinophils contribute to EC apoptosis. The in vivo relevance of these findings was investigated by using bronchial biopsy specimens. Airway mucosa from the
334 Trautmann et al
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
FIG 4. Demonstration of apoptosic features of ECs. A, Agarose gel electrophoresis of EC DNA. ECs were cultured for 3 days with supernatants from stimulated CD45RO+ T cells or with lysates from unstimulated and stimulated eosinophils. Control, ECs alone. B, ECs were cultured with supernatants from T cells and lysates from eosinophils. The percentage of annexin V-stained ECs after 3 days is indicated. Results shown in A and B represent 3 experiments. *P < .05.
Basic and clinical immunology FIG 5. Demonstration of apoptotic ECs in asthma. A, In addition to routine HE staining, sections were subjected to TUNEL with hematoxylin counterstaining and Hoechst staining. A, Normal bronchial mucosa as a negative control. B and C, Red nuclei indicate positive TUNEL staining; condensed and bright nuclei indicate positive Hoechst staining of apoptotic ECs in asthma. B through D, Note the detachment of apoptotic ECs with almost complete exfoliation and only a few viable residual basal cells. Results are representative of 7 asthma specimens.
proximal conducting airways was examined for histologic changes and EC apoptosis. Examination of control mucosa revealed a pseudostratified epithelium consisting of basal cells, ciliated cells, and secretory goblet cells (Fig 5, A). In contrast, the mucosa of asthma patients showed marked damage to the epithelial layer, with exfoliation of ECs (Fig 5, B-D). Homogeneous thickening of the basement membrane was another key finding in asthmatic bronchi. As shown, damaged ECs of asthma patients were strongly TUNEL positive (Fig 5, B and C). Hoechst staining was used as a second method, complementing TUNEL staining. Condensed and fragmented nuclei of apoptotic ECs were demonstrated during progressive EC shedding (Fig 5, B and C). Hematoxylineosin stained overall histology of the sections showed that the cells stained by TUNEL and Hoechst appeared to be damaged ECs. During the process of EC shedding, finally only a few viable basal cells are retained on the basement membrane (Fig 5, D).
DISCUSSION The present study demonstrates that ECs are targets of activated T cells and eosinophils that invade the bronchial wall in asthma. Among the growing complexity in the pathogenic mechanisms of asthma, the potential role of T cells and eosinophils is suggested. It is proposed that ECs can be the target of the immunologic injury in certain inflammatory disorders, such as eczematous dermatitis and asthma.8,9,18 A morphologic study of the bronchial mucosa from patients with asthma demonstrated 54.5% lymphocytes, 22.1% eosinophils, 4.9% neutrophils, and 4.6% mast cells in the inflammatory infiltrate.19 It was previously suggested that T cells may contribute to airway inflammation, mainly by activating and prolonging the life span of eosinophils via IL-5, IL-3, and GM-CSF.5,6,20 The results of our study point to a direct pathogenic role for T cells. Activated effector T cells secrete IFN-γ and sensitize ECs for TNF-α-mediated apoptosis. IFN-γinduced increase in TNFR expression may be one key event in the control of TNF-α-induced EC apoptosis. A role for IFN-γ and TNF-α in damaging ECs was shown in other experimental systems,21,22 but these findings were not linked to asthma until the present study. It is interesting that we observed that activated T cells and eosinophils from healthy and atopic individuals do not differ in their capacity for inducing EC apoptosis. Apparently, bronchial ECs and skin keratinocytes utilize different predominant apoptosis pathways in response to allergic inflammation. Epidermal keratinocytes do not express Fas ligand in the course of eczema, and upregulation of Fas mainly by IFN-γ renders the cells susceptible to apoptosis, induced by Fas ligand provided by activated skin-infiltrating T cells.8 In contrast, the TNFR-related apoptotic pathway is predominant in bronchial EC. Approximately 70% of ECs expressed TNFR-I and TNFR-II by IFN-γ and TNF-α stimulation, whereas Fas expression was observed in only 20% of the cells. It is possible that ECs from atopic
Trautmann et al 335
versus healthy donors may show differences in sensitivity to different apoptosis-inducing ligands. The mucosa in asthma is distinguished from other inflammatory lung diseases such as cystic fibrosis by the presence of large numbers of activated eosinophils. The present study demonstrates that EC apoptosis induced by exposure to lysates from IFN-γ-stimulated eosinophils is mediated mainly by TNF-α. In addition to activated eosinophils, other sources of TNF-α in asthmatic airway inflammation might exist, such as macrophages, mast cells, and stimulated ECs.23 Tumor necrosis factor-α induces opposing cellular responses that result from activation of two different pathways: cytokine release via NF-κB activation or apoptosis via FADD-caspase 8 activation.24 However, the mechanisms by which these opposing programs are selected remain unclear. Substantial evidence supports the view that engagement of TNFR-I can trigger apoptosis in many cell types.24 Engagement of TNFR-II has been found to stimulate TNFR-I-mediated apoptosis and to be able to autonomously induce cell death.25,26 Other potential pathways of tissue injury by eosinophils exist, such as release of cytotoxic proteins including ECP, eosinophil-derived neurotoxin, eosinophil peroxidase, and major basic protein.27 Recombinant human ECP induced mainly necrosis of ECs in the pres-ent study. Eosinophil cationic protein might damage cells by a colloid-osmotic process, as it was shown that ECP can introduce non-ion selective pores in both cellular and synthetic membranes.28 To induce toxic EC death, doses far higher than those measured in the lysates of eosinophils were required. Eosinophil cationic protein induced EC necrosis at as early as 6 hours of incubation. However, activated T cells and eosinophils induce EC apoptosis after 3 days. Therefore, the present data do not fully support the possibility of a major role for ECP in EC apoptosis. The blocking effect of anti-TNF-α mAb suggests that major basic protein, eosinophil-derived neurotoxin, and eosinophil peroxidase may also not contribute predominantly to EC apoptosis. Apoptotic ECs may be cleared from the epithelial layer in vivo by macrophages and shedding. Apoptotic cells that have not been phagocytized and removed progress to secondary necrosis with loss of cell membrane integrity and the release of cytoplasmic contents. Increased DNA fragmentation shown by TUNEL and Hoechst staining in biopsy samples significantly correlated with the morphologic findings of EC damage. The highest number of apoptotic ECs was encountered in areas of initial EC detachment. In the cell culture experiments, loosening of cell-cell contacts is obvious when ECs become apoptotic. Druilhe et al29 discuss that their failure to detect differences in the number of TUNELpositive ECs between healthy and asthmatic subjects may be the result of the detachment of ECs into the bronchial lumen. In addition, the problem of real and artifactual EC shedding in evaluations of bronchoscopic airway biopsy specimens has to be considered.
Basic and clinical immunology
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
336 Trautmann et al
Basic and clinical immunology
Recent studies challenge the concept that the eosinophil is the central effector cell in asthma.6,30,31 Furthermore, the issue of whether the asthmatic inflammation is fully mediated by type 2 T cells should be revisited.6,32 In humans there is not a straight bias for type 1 and type 2 differentiated T-cell subsets in disease. Even allergen-specific T-cell clones isolated from the airways of asthmatic subjects often release significant levels of IFN-γ along with IL-4.33,34 The amount of IFN-γ and TNF-α is elevated in supernatants from cultures of unstimulated and stimulated bronchoalveolar-lavagefluid cells from asthma patients.35 During exacerbations of asthma, viral infections may also contribute to the IFN-γ level in the airway mucosa.6,23 In the present study ECs exhibited a relatively low threshold in response to IFN-γ, in the range of 1 to 10 ng/mL. Several approaches to the inhibition of TNF-α effects are now under investigation in asthma, including mAb to TNF-α and soluble TNF-α R.36,37 Bronchial ECs exposed to IFN-γ and TNF-α secrete increased amounts of bFGF and TGF-β1 as determined by ELISA (authors’ unpublished observation). It is likely, that the stressed epithelium may contribute to the structural remodeling of the airways including thickening of the basement membrane and induction of myofibroblasts.23,38 The demonstration of bronchial EC apoptosis in cocultures with T cells and eosinophils as well as bronchial biopsies establishes the essential role of T cells and eosinophils in bronchial EC injury and destruction in asthma. This pathologic process, which represents a main direct tissue damage in asthma, is mediated by IFNγ and TNF-α and characterized by apoptosis and shedding of bronchial ECs. The knowledge of this molecular basis is pivotal in understanding the development of pathology in asthma, and opens a future for more focused therapeutic applications. REFERENCES 1. Laitinen LA, Heino M, Laitinen A, Kava T, Haahtela T. Damage of the airway epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 1985;131:599-606. 2. Jeffery PK, Wardlaw AJ, Nelson FC, Collins JV, Kay AB. Bronchial biopsies in asthma: an ultrastructural, quantitative study and correlation with hyperreactivity. Am Rev Respir Dis 1989;140:1745-53. 3. Montefort S, Roche WR, Roberts JA, Holgate ST. Bronchial epithelial shedding in asthmatics and non-asthmatics. Respir Med 1989;87:9-11. 4. Bentley AM, Menz G, Storz C, Robinson DS, Bradley B, Jeffery PK, Durham SR, et al. Identification of T lymphocytes, macrophages, and activated eosinophils in the bronchial mucosa in intrinsic asthma. Am Rev Respir Dis 1992;146:500-6. 5. Howarth PH, Bradding P, Montefort S, Peroni D, Djukanovic R, Carroll MP, et al. Mucosal inflammation and asthma. Am J Respir Crit Care Med 1994;150:S18-22. 6. Busse WW, Lemanske RF. Asthma. New Engl J Med 2001;344:350-62. 7. Weller PF. Human eosinophils. J Allergy Clin Immunol 1997;100:283-7. 8. Trautmann A, Akdis M, Kleemann D, Altznauer F, Simon HU, Graeve T, et al. T cell-mediated Fas-induced keratinocyte apoptosis plays a key pathogenetic role in eczematous dermatitis. J Clin Invest 2000;106:25-35. 9. Trautmann A, Altznauer F, Akdis M, Simon HU, Disch R, Bröcker EB, et al. The differential fate of cadherins during T cell-induced keratinocyte apoptosis leads to spongiosis in eczematous dermatitis. J Invest Dermatol 2001;117:927-34. 10. Hansel TT, de Vries IJM, Carballido JM, Braun RK, Carballido-Perrig N,
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
22.
23.
24. 25.
26.
27. 28.
29.
30.
31. 32.
Rihs S, et al. Induction and function of eosinophil intercellular adhesion molecule-1 and HLA-DR. J Immunol 1992;149:2130-6. Crameri R, Jaussi R, Menz G, Blaser K. Display of expression products of cDNA libraries on phage surfaces. A versatile screening system for selective isolation of genes by specific gene-product/ligand interaction. Eur J Biochem 1994;226:53-8. Rosenberg HF, Ackerman SJ, Tenen DG. Human eosinophil cationic protein. Molecular cloning of a cytotoxin and helminthotoxin with ribonuclease activity. J Exp Med 1989;170:163-76. Vermes I, Haanen C, Steffens-Nakken H, Reutlingsberger CDM. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labeled Annexin V. J Immunol Method 1995;184:39-51. Trautmann A, Akdis M, Schmid-Grendelmeier P, Disch R, Bröcker EB, Blaser K, et al. Targeting keratinocyte apoptosis in the treatment of atopic dermatitis and allergic contact dermatitis. J Allergy Clin Immunol 2001;108:839-46. Staley K, Blaschke AJ, Chun J. Apoptotic DNA fragmentation is detected by a semiquantitative ligation-mediated PCR of blunt DNA ends. Cell Death Diff 1997;4:66-75. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992;119:493-501. Akdis CA, Joss A, Akdis M, Faith A, Blaser K. A molecular basis for T cell suppression by IL-10: CD28-associated IL-10 receptor inhibits CD28 tyrosine phosphorylation and phosphatidylinositol 3-kinase binding. FASEB J 2000;14:1666-8. Montefort S, Herbert CA, Robinson C, Holgate ST. The bronchial epithelium as a target for the inflammatory attack in asthma. Clin Exp Allergy 1992;22:511-20. Ohashi Y, Motojima S, Fukuda T, Makino S. Airway hyperresponsiveness, increased intracellular spaces of bronchial epithelium, and increased infiltration of eosinophils and lymphocytes in bronchial mucosa in asthma. Am Rev Respir Dis 1992;145:1469-76. Simon HU, Blaser K. Inhibition of programmed eosinophil death—a key pathogenic event for eosinophilia? Immunol Today 1995;16:53-5. Kampf C, Relova AJ, Sandler S, Roomans GM. Effects of TNF-α, IFNγ and IL-1-β on normal human bronchial epithelial cells. Eur Respir J 1999;14:84-91. Gitter A, Bendfeldt K, Schulzke JD, Fromm M. Leaks in the epithelial barrier caused by spontaneous and TNF-α-induced single-cell apoptosis. FASEB J 2000;14:1749-53. Bousquet J, Jeffery PK, Busse WW, Johnson M, Vignola AM. Asthma. From bronchoconstriction to airways inflammation and remodeling. Am J Respir Crit Care Med 2000;161:1720-45. Pimentel-Muinos FX, Seed B. Regulated commitment of TNF receptor signaling: a molecular switch for death or activation. Immunity 1999;11:783-93. Weiss T, Grell M, Hessabi B, Bourteele S, Müller G, Scheurich P, et al. Enhancement of TNF receptor p60 mediated cytotoxicity by TNF receptor p80: requirement of the TNF receptor-associated factor-2 binding site. J Immunol 1997;158:2398-404. Haridas V, Darnay BG, Natarajan K, Heller R, Aggarwal BB. Overexpression of the p80 TNF receptor leads to TNF-dependent apoptosis, nuclear factor-κB activation, and c-Jun kinase activation. J Immunol 1998;160:3152-62. Gleich GJ. Mechanisms of eosinophil-associated inflammation. J Allergy Clin Immunol 2000;105:651-63. Young JDE, Peterson CGB, Venge P, Cohn ZA. Mechanism of membrane damage mediated by human eosinophil cationic protein. Nature 1986;321:613-14. Druilhe A, Wallaert B, Tsicopoulos A, Lapa e Silva JR, Tillie-Leblond I, Tonnel AB, et al. Apoptosis, proliferation, and expression of Bcl-2, Fas, and Fas ligand in bronchial biopsies from asthmatics. Am J Respir Cell Mol Biol 1998;19:747-57. Lee NA, Gelfand EW, Lee JJ. Pulmonary T cells and eosinophils: coconspirators or independent triggers of allergic respiratory pathology. J Allergy Clin Immunol 2001;107:945-57. Boushey HA, Fahy JV. Targeting cytokines in asthma therapy: round one. Lancet 2000;356:2114-16. Holtzman MJ, Sampath D, Castro M, Look DC, Jayaraman S. The onetwo of T helper cells: does interferon-γ knock out the Th2 hypothesis for asthma? Am J Respir Cell Mol Biol 1996;14:316-18.
33. Krug N, Madden J, Redington AE, Lackie P, Djukanovic R, Schauer U, et al. T-cell cytokine profile evaluated at the single cell level in BAL and blood in allergic asthma. Am J Respir Cell Mol Biol 1996;14:319-26. 34. Parronchi P, Macchia D, Piccini MP, Biswas P, Simonelli C, Maggi E, et al. Allergen- and bacterial antigen-specific T-cell clones established from atopic donors show a different profile of cytokine production. Proc Natl Acad Sci USA 1991;88:4538-42. 35. Cembrzynska-Nowak M, Szklarz E, Inglot AD, Teodorczyk-Injeyan JA. Elevated release of tumor necrosis factor-α and interferon-γ by bron-
Trautmann et al 337
choalveolar leukocytes from patients with asthma. Am Rev Respir Dis 1993;147:291-5. 36. Barnes PJ. New directions in allergic diseases: mechanism-based antiinflammatory therapies. J Allergy Clin Immunol 2000;106:5-16. 37. Stirling RG, Chung KF. New immunological approaches and cytokine targets in asthma and allergy. Eur Respir J 2000;16:1158-74. 38. Morishima Y, Nomura A, Uchida Y, Noguchi Y, Sakamoto T, Ishii Y, et al. Triggering the induction of myofibroblasts and fibrogenesis by airway epithelial shedding. Am J Respir Cell Mol Biol 2001;24:1-11.
Basic and clinical immunology
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
From athe Swiss Institute of Allergy and Asthma Research (SIAF), bthe Department of Pneumology, Süleyman Demirel University, and cthe Department of Dermatology, University of Würzburg. Supported by the Swiss National Foundation (32-65661-01 and 31-63381-00) and the Saurer Foundation. A.T. is a recipient of a grant from the Deutsche Forschungsgemeinschaft (TR460/1-1), P.S.-G. is a recipient of a grant for the promotion of academic young people of the University of Zurich. Received for publication August 20, 2001; revised October 25, 2001; accepted October 31, 2001. Reprint requests: A. Trautmann, MD, Swiss Institute of Allergy and Asthma Research (SIAF), Obere Strasse 22, CH-7270 Davos, Switzerland. Copyright © 2002 by Mosby, Inc. 0091-6749/2002 $35.00 + 0 1/83/121460 doi:10.1067/mai.2002.121460
Key words: Apoptosis, asthma, bronchial epithelial cell, eosinophil, Fas, interferon-gamma, T cell, tumor necrosis factor
Asthma is characterized by chronic inflammation of the bronchial mucosa where T cells and eosinophils are prominent. Shedding of bronchial epithelial cells (ECs) is an important histologic feature observed in bronchial biopsy specimens from asthmatic patients.1,2 Shedding of ECs is characterized by loss of the normal bronchial pseudostratified epithelium and the maintenance of a few basal cells on a thickened basement membrane.3 Activated T cells and eosinophils can be detected in the bronchial mucosa, and their numbers correlate with disease severity.4 Extensive evidence has been provided that T cells play a role in chronic asthma.5,6 A major advance in the understanding of the functional capabilities of eosinophils has been the demonstration that they are a source of pro-inflammatory cytokines including GMCSF, IL-4, IL-5, IL-6, IL-8, and TNF-α.7 The aim of this study was to investigate whether, and by which mechanism, T cells and eosinophils can cause damage to airway ECs. Recently we demonstrated that activated T cells infiltrating the skin in atopic dermatitis and allergic contact dermatitis induce apoptosis of keratinocytes and the characteristic spongiosis pattern in the epidermis.8,9 Less is understood regarding the specific roles of immune effector cells in the damage of bronchial ECs in asthma. The present study demonstrates that both T cells and eosinophils contribute to the induction of EC apoptosis by secretion of IFN-γ and TNF-α and triggering of death receptors on bronchial ECs. Moreover, we show that TNF receptor (TNFR) but not Fas is predominantly involved in apoptosis of ECs. Our data suggest that apoptosis represents an essential mechanism for the development of EC injury and shedding in asthma.
METHODS Subjects Approval for the use of human tissues was granted by the ethics committee of Davos. The fiberoptic bronchial biopsy specimens of 7 asthma patients (mild persistent asthma according to the Global Initiative for Asthma guidelines) were fixed in 4% paraformaldehyde and embedded in paraffin. Control bronchial mucosa was obtained from 3 patients with malignant lung tumors. T cells and eosinophils were purified from 11 atopic and 6 healthy donors. 329
Basic and clinical immunology
Background: Asthma is an inflammatory airway disease associated with an infiltration of T cells and eosinophils, increased levels of pro-inflammatory cytokines, and shedding of bronchial epithelial cells (ECs). Objective: Shedding of bronchial ECs is characterized by loss of the normal bronchial pseudostratified epithelium and the maintenance of a few basal cells on a thickened basement membrane. The aim of this study was to investigate whether, and by which mechanism, T cells and eosinophils can cause damage to airway ECs. Methods: Bronchial ECs, cultured and exposed to cytokines, eosinophil cationic protein, activated T cells, and eosinophils were studied for the expression of apoptosis receptors (flow cytometry, immunoblotting, and RNA expression) and for the susceptibility for undergoing apoptosis. In addition, bronchial biopsy specimens from patients with asthma were evaluated for EC apoptosis. Results: We demonstrate herein that the respiratory epithelium is an essential target of the inflammatory attack by T cells and eosinophils. Bronchial ECs underwent cytokine-induced cell death with DNA fragmentation and morphologic characteristics of apoptosis mediated by activated T cells and eosinophils. T cell- and eosinophil-induced EC apoptosis was blocked by inhibition of IFN-γ and TNF-α; the Fas ligand-Fas pathway appears to be less important. Recombinant eosinophil cationic protein induced mainly necrosis of ECs. Furthermore, we demonstrated in situ apoptotic features of ECs in bronchial biopsy specimens of asthmatic patients. Conclusion: T cell- and eosinophil-induced apoptosis represents a key pathogenic event leading to EC shedding in asthma. (J Allergy Clin Immunol 2002;109:329-37.)
330 Trautmann et al
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
Abbreviations used BEBM: Bronchial epithelial cell basal medium EC: Epithelial cell ECP: Eosinophil cationic protein HE: Hematoxylin-eosin TNFR: TNF receptor TUNEL: Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling
Reagents
Basic and clinical immunology
Human recombinant IFN-γ was a gift from Novartis Pharma (Basel, Switzerland); IL-5 and GM-CSF were from R&D Systems Europe Ltd (Abingdon, UK). Ethidium bromide was purchased from Sigma Chemical Co (St Louis, Mo). The following antibodies were used: anti-Fas-PE, anti-Fas ligand-FITC, (both from Alexis Corp., San Diego, Calif); neutralizing anti-TNF-α (R&D Systems Europe Ltd); anti-TNFR-I, anti-TNFR-II, and anti-TRAMP/deathR (DR) 3 (Oncogene, Cambridge, Mass); anti-TRAILR-I/DR4 and anti-TRAILR-II/DR5 (Calbiochem, San Diego, Calif); anti-βcatenin (Santa Cruz Biotechnology Inc, Santa Cruz, Calif); apoptosis-inducing anti-Fas mAb (CH-11, Immunotech, Marseilles, France); and neutralizing anti-IFN-γ (45-15, gift from Novartis Pharma). Human recombinant soluble Fas ligand, TRAIL, TNF-α, and TWEAK were from Alexis Corp. The caspase inhibitor AcAsp-Glu-Val-Asp-aldehyde (Ac-DEVD-CHO) was from Bachem AG (Bubendorf, Switzerland).
Culture of human airway ECs Primary normal human bronchial ECs were purchased from Clonetics Corp. (San Diego, Calif). Cells were grown in bronchial EC basal medium (BEBM) supplemented as prescribed by the manufacturer. ECs were subcultured and used between passages 3 and 7.
Flow cytometry ECs were incubated with appropriate FITC- or PE-labeled mAb for 30 min at 4°C and then fixed in 4% paraformaldehyde (pH 7.4). Negative control cells were prepared in a similar fashion with isotype control IgG1-FITC/PE mAb. Fluorescence intensity was measured by flow cytometry (EPICS XL-MCL flow cytometer, Beckmann Coulter Int. S.A., Nyon, Switzerland).
Isolation of
CD45RO+
effector T cells
Mononuclear cells were isolated by Ficoll (Biochrom KG, Berlin, Germany) density gradient centrifugation of venous blood. CD45RO+ T cells were isolated with use of the MACS system according to the instructions of the manufacturer (Miltenyi Biotec, Bergisch Gladbach, Germany) as described.8,9 T cells, 1 × 106 CD45RO+/1 mL BEBM, were stimulated with a combination of soluble anti-CD2 (0.5 (µg/mL), anti-CD3 (1.0 (µg/mL), and anti-CD28 (0.5 (µg/mL) mAb, and supernatants were harvested after 3 days.
Isolation of eosinophils Granulocyte-enriched cell pellets were isolated by Ficoll (Biochrom KG) density gradient centrifugation of venous blood and depleted of erythrocytes by use of hypotonic saline lysis. Eosinophils were negatively selected by using the MACS system (Miltenyi Biotec) as described.10 The viability of freshly isolated eosinophils was >95% (by trypan blue exclusion) and the purity was >99% (Diff-Quick staining; Dade AG, Düdingen, Switzerland). Eosinophils, 1 × 106/1 mL BEBM, were stimulated with IFN-γ (100
ng/mL), IL-5, or GM-CSF (25 ng/mL) or both, and supernatants were harvested after 1 day. To obtain eosinophil lysate, cells were washed twice and exposed to twice-repeated freeze-thaw cycles. Ultrasound destruction was applied for 1 min at 4°C, and lysates were centrifuged (14.400 U/min, 4 min) at 4°C.
Cloning, expression, and purification of human eosinophil cationic protein (ECP) Messenger ribonucleic acid from 107 purified eosinophils was isolated by using the Oligotex Direct mRNA Midi Kit (Qiagen AG, Basel, Switzerland) followed by oligo-dT primed reverse transcription. This cDNA was used to amplify a fragment containing the open reading frame encoding mature ECP by PCR using the following primers: 5´-primer 5´-CG GGA TCC AGA CCC CCA CAG TTT ACG AGG G and 3´-primer 5´ GGG AAG CTT TTA GAT GGT GGT ATC CAG GTG AAC TGG. The PCR product was gel purified and restricted with BamHI and HindIII incorporated into the amplification primers (underlined sequences). Restricted fragments were gel purified and ligated into BamHI/HindIII-restricted pQE30 (Qiagen AG) expression vector to produce hexahistidinetagged recombinant protein. The ligation mixture was transformed into Escherichia coli strain M15 by electroporation,11 and the nucleotide sequence was confirmed to code the reported amino acid sequence of human ECP.12 Correct constructs were inoculated into 2 × YT broth and induced with 2 mmol/L isopropyl-β-D-thiogalactoside at an OD595 of 0.7 to produce recombinant ECP. The protein was found in insoluble inclusion bodies, which were solubilized in guanidinium hydrochloride and one-step purified over Ni2+-NTA metal affinity chromatography resin (Qiagen AG) under denaturing conditions. The eluted protein was dialyzed against water containing 0.05% glacial acetic acid and 5 mmol/L EDTA. The protein was devoid of relevant contaminants as determined by SDS-PAGE.
Coculture experiments ECs were first seeded into 6-well or 48-well tissue culture plates in BEBM, and were incubated for 1 day to allow attachment and formation of a monolayer. Fifty percent of the medium was then removed and replaced with supernatants from CD45RO+ T cells or eosinophil lysates. CD45RO+ T cells or eosinophils were stimulated and washed twice before direct coculture with ECs.
Demonstration of EC viability and apoptosis in cultures and in situ. For the analysis of apoptosis in EC cultures, cells were examined under phase-contrast microscopy to evaluate the morphologic characteristics. EC viability was evaluated by means of ethidium bromide (1 µmol/L) uptake and flow cytometry. Annexin V is a phosphatidylserine binding protein that was also used to detect apoptotic cells.8,13 Hoechst staining was done as described.8,9 For the demonstration of apoptosis by DNA fragmentation, we isolated total genomic DNA of EC with the DNeasy kit according to the protocol of the manufacturer (Qiagen AG). The ApoAlert LM-PCR ladder assay kit (Clontech Laboratories AG, Basel, Switzerland) was used for the detection of nucleosomal ladders in apoptotic EC.14,15 After deparaffination and rehydration of the slides, apoptotic cells were identified in situ by staining double-stranded DNA breaks. Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) was employed as described.8,9,16
Immunoprecipitation and immunoblotting EC lysates were prepared, and immunoprecipitation was performed as described.9,17 Lysates were quantified with the Bio-Rad Protein assay (Bio-Rad Laboratories, Glattbrugg, Switzerland) to ensure equivalent gel loading. Immunoblots were probed with
A
B
C
D
FIG 1. A, Death receptor expression on ECs. ECs were cultured in the absence (Control) or presence of IFNγ, TNF-α, or both cytokines. The percentage of stained ECs after 1 day is indicated. B, Immunoprecipitation and immunoblotting with EC lysates were performed with the same mAb. β-Catenin is a control for equivalent gel loading. C, Gene expression array. cDNA probes of EC stimulated with 10 ng/mL IFN-γ and 10 ng/mL TNF-α were hybridized to gene-specific cDNA fragments spotted on GEArray membranes in duplicates. D, Sensitivity of ECs to apoptosis-inducing stimuli. ECs were treated for 3 days with apoptosis-inducing anti-Fas mAb, TNF-α, TRAIL, and TWEAK either in the absence (Alone) or presence of IFN-γ. EC viability (%) was determined by ethidium bromide exclusion and flow cytometry. Results shown in A-D represent 3 experiments. *P < .05.
appropriate antibodies followed by peroxidase-conjugated antimouse, anti-rabbit, or anti-goat IgG (Dako A/S, Glostrup, Denmark). Detection was performed by use of the ECL system (Amersham Pharmacia Biotech Ltd., Buckinghamshire, UK).
Gene expression array Biotinylated cDNA probes were hybridized to gene-specific cDNA fragments spotted on GEArray membranes as described by the manufacturer (GEA human TNF&Fas network, SuperArray Inc, Bethesda, Md). The relative expression level of genes was detected by chemiluminescence.
Statistical analysis Results are shown as mean ± SD. The paired Student t test was used for comparison of paired conditions.
RESULTS Apoptosis receptor expression and induction of apoptosis in bronchial ECs To assess EC surface molecules, we examined ECs by using flow cytometry (Fig 1, A). Both Fas and Fas ligand
are expressed on ECs and were upregulated mainly by TNF-α. TNFR-I and TNFR-II expression was higher and substantially upregulated by IFN-γ and TNF-α. By using immunoprecipitation and immunoblotting of EC lysates, we detected specific bands corresponding to Fas, TNFRI, and TNFR-II proteins (Fig 1, B). In contrast, apoptosis receptors DR3 (TRAMP), DR4 (TRAILR-I), and DR5 (TRAILR-II) were absent. IFN-γ– and TNF-α–stimulated ECs showed mRNA transcripts of Fas, TNFR-I, and TNFR-II. It is noteworthy that ECs express significant amounts of TNF-α mRNA (Fig 1, C). The apoptotic response of ECs was analyzed by using anti-Fas mAb, soluble Fas ligand, TNF-α, TRAIL, and TWEAK as apoptosis-inducing ligands. Anti-Fas mAb or soluble Fas ligand, either alone or after pretreatment with IFN-γ, did not trigger >5% of the ECs to undergo apoptosis (Fig 1, D). ECs showed significant death only in response to TNF-α after pretreatment with IFN-γ. ECs treated only with 10 ng/mL TNF-α showed a reduced viability compared to untreated cells.
Basic and clinical immunology
Trautmann et al 331
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
332 Trautmann et al
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
Basic and clinical immunology
Supernatants from stimulated T cells (1 × 106/mL) contain IFN-γ (156.3 ± 12.3 IU/mL) and TNF-α (140.3 ± 17.0 pg/mL), as determined by ELISA. The levels of IFN-γ released from activated CD45RO+ T cells obtained from atopic donors were slightly lower than those of nonatopic donors but still above 100 IU/mL. Control ECs had a polygonal shape and maintained confluence during 3 days. Supernatants from unstimulated T cells did not alter this morphology (Fig 2, A). After 3 days of treatment with supernatants from stimulated T cells, dead ECs appearing rounded and granulated were found to be detached from the bottom of the well. To demonstrate the type of cell death, cellular DNA was visualized with Hoechst staining. The nuclei of ECs were condensed and showed intense fluorescence when undergoing apoptosis. Quantitative assessment of viability confirmed the observed morphologic changes. As shown in Fig 2, A, EC apoptosis could be totally inhibited by pretreatment with neutralizing anti-IFN-γ mAb. The inhibitory effect of anti-TNF-α mAb was smaller but still significant. The caspase inhibitor Ac-DEVD-CHO prevented also EC apoptosis caused by stimulated T cells. EC apoptosis induced by supernatants from stimulated T cells was not prevented by blocking Fas or Fas ligand (data not shown). As a next step, we monitored the viability of ECs in direct contact with unstimulated and stimulated CD45RO+ T cells in cocultures. As displayed in Fig 2, B, destruction of the EC monolayer was observed after 3 days incubation with stimulated T cells. Anti-IFN-γ and antiTNF-α mAbs significantly rescued the EC from apoptosis.
Eosinophils induce both apoptosis and necrosis of ECs
FIG 2. Anti-IFN-γ and anti-TNF-α mAb inhibit EC apoptosis induced by stimulated T cells. A, Phase contrast morphology (×200) of ECs 3 days after exposure to supernatants from unstimulated and stimulated CD45RO+ T cells. In parallel, ECs were stained with Hoechst (×400) to visualize all nuclei. B, Phase contrast morphology (×400) of ECs after 3 days direct coculture with unstimulated and stimulated CD45RO+ T cells. A, B, EC viability (%) was determined by ethidium bromide exclusion and flow cytometry. Mean ± SD of four separate experiments are shown in parentheses. *P < .05.
T cell–mediated EC apoptosis To investigate the role of T cells in EC apoptosis, we exposed ECs grown as a monolayer to supernatants of unstimulated or stimulated CD45RO+ T cells for 3 days.
Eosinophils comprise major cells in the inflammatory infiltrate of asthma. Therefore, the role of eosinophils in EC apoptosis was further investigated. Eosinophils could only induce apoptosis in IFN-γ-pretreated ECs. After 3 days treatment with lysates from IFN-γ stimulated eosinophils, significant numbers of ECs demonstrated typical features of apoptosis (Fig 3, A). In contrast, IL-5 and GM-CSF did not activate eosinophils to induce EC apoptosis. Quantitative assessment of viability confirmed the observed morphologic changes. Lysates from IFN-γstimulated eosinophils (1 × 106/mL) contain 64.2 ± 1.3 pg/mL TNF-α, as determined by ELISA. The EC nuclei undergoing apoptosis were condensed and showed intense fluorescence with use of Hoechst staining. When pretreated with anti-TNF-α mAb, significantly less ECs died. In repeated experiments the apoptosis-inducing capacity of T cells and eosinophils did not differ between atopic and nonatopic donors. In addition to cytokines, eosinophils contain and release potent cytotoxic substances into the inflammatory microenvironment.7 Significant amounts of ECP were detected in lysates of eosinophils. Representative values for eosinophils, 1 × 106/mL, after 1 day culture were 0.83 µg/mL in unstimulated, 1.84 µg/mL in IL-5 stimulated, 0.77 µg/mL in INF-γ stimulated, and 1.86 µg/mL in INFγ– and IL-5–stimulated eosinophils. By using human recombinant ECP, we noted that the viability of ECs
Trautmann et al 333
Basic and clinical immunology
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
FIG 3. A, Anti-TNF-α mAb inhibits EC apoptosis induced by stimulated eosinophils. Phase contrast morphology (×200) of ECs pretreated with 10 ng/mL IFN-γ 3 days after exposure to lysates from unstimulated and stimulated eosinophils. In parallel, ECs were stained with Hoechst (×400) to visualize all nuclei. B, ECPmediated EC necrosis. Phase contrast morphology (×400) of ECs 6 hours after exposure to the indicated concentrations of ECP. C, Annexin V/propidiumiodide staining of ECs 6 hours after exposure to 10 µg/mL ECP indicate necrotic cells in quadrant 4. A, B, EC viability (%) was determined by ethidium bromide exclusion and flow cytometry. Mean ± SD of 3 separate experiments are shown in parentheses. *P < .05.
decreased significantly at 10 µg/mL doses (Fig 3, B). The type of cell death was mainly necrosis, as determined with annexin V/propidiumiodide staining. As early as 6 hours after ECP treatment, ECs were stained with propidiumiodide but did not show any membrane phosphatidylserine inversion (Fig 3, C).
Assessment of EC apoptosis by use of DNA fragmentation and annexin V staining A hallmark of apoptosis is the characteristic DNA fragmentation (200 to 500 bp) in gel electrophoresis. Viable control ECs did not exhibit any fragmentation (Fig 4, A, control). Deoxyribonucleic acid fragmentation as a characteristic feature of apoptotic cells was observed 3 days after addition of stimulated T cell supernatants. Similarly, DNA fragmentation was induced by IFN-γ and TNF-α treatments. It is remarkable that DNA fragmentation was absent in ECs in which stimulated T cell-induced apopto-
sis had been blocked by anti-IFN-γ. ECs pretreated with IFN-γ and exposed to lysates from IFN-γ-stimulated eosinophils also showed DNA fragmentation (Fig 4, A). Eosinophil-induced DNA fragmentation in ECs was inhibited by neutralizing TNF-α. Because annexin V is a marker of apoptosis, we also stained ECs which were exposed to T cells and eosinophils with annexin V-FITC. In correlation with the DNA fragmentation data, there was substantial staining with annexin V in ECs, which were exposed to stimulated T cells and eosinophils (Fig 4, B).
Demonstration of EC apoptosis in patients with asthma These results demonstrated that both T cells and eosinophils contribute to EC apoptosis. The in vivo relevance of these findings was investigated by using bronchial biopsy specimens. Airway mucosa from the
334 Trautmann et al
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
FIG 4. Demonstration of apoptosic features of ECs. A, Agarose gel electrophoresis of EC DNA. ECs were cultured for 3 days with supernatants from stimulated CD45RO+ T cells or with lysates from unstimulated and stimulated eosinophils. Control, ECs alone. B, ECs were cultured with supernatants from T cells and lysates from eosinophils. The percentage of annexin V-stained ECs after 3 days is indicated. Results shown in A and B represent 3 experiments. *P < .05.
Basic and clinical immunology FIG 5. Demonstration of apoptotic ECs in asthma. A, In addition to routine HE staining, sections were subjected to TUNEL with hematoxylin counterstaining and Hoechst staining. A, Normal bronchial mucosa as a negative control. B and C, Red nuclei indicate positive TUNEL staining; condensed and bright nuclei indicate positive Hoechst staining of apoptotic ECs in asthma. B through D, Note the detachment of apoptotic ECs with almost complete exfoliation and only a few viable residual basal cells. Results are representative of 7 asthma specimens.
proximal conducting airways was examined for histologic changes and EC apoptosis. Examination of control mucosa revealed a pseudostratified epithelium consisting of basal cells, ciliated cells, and secretory goblet cells (Fig 5, A). In contrast, the mucosa of asthma patients showed marked damage to the epithelial layer, with exfoliation of ECs (Fig 5, B-D). Homogeneous thickening of the basement membrane was another key finding in asthmatic bronchi. As shown, damaged ECs of asthma patients were strongly TUNEL positive (Fig 5, B and C). Hoechst staining was used as a second method, complementing TUNEL staining. Condensed and fragmented nuclei of apoptotic ECs were demonstrated during progressive EC shedding (Fig 5, B and C). Hematoxylineosin stained overall histology of the sections showed that the cells stained by TUNEL and Hoechst appeared to be damaged ECs. During the process of EC shedding, finally only a few viable basal cells are retained on the basement membrane (Fig 5, D).
DISCUSSION The present study demonstrates that ECs are targets of activated T cells and eosinophils that invade the bronchial wall in asthma. Among the growing complexity in the pathogenic mechanisms of asthma, the potential role of T cells and eosinophils is suggested. It is proposed that ECs can be the target of the immunologic injury in certain inflammatory disorders, such as eczematous dermatitis and asthma.8,9,18 A morphologic study of the bronchial mucosa from patients with asthma demonstrated 54.5% lymphocytes, 22.1% eosinophils, 4.9% neutrophils, and 4.6% mast cells in the inflammatory infiltrate.19 It was previously suggested that T cells may contribute to airway inflammation, mainly by activating and prolonging the life span of eosinophils via IL-5, IL-3, and GM-CSF.5,6,20 The results of our study point to a direct pathogenic role for T cells. Activated effector T cells secrete IFN-γ and sensitize ECs for TNF-α-mediated apoptosis. IFN-γinduced increase in TNFR expression may be one key event in the control of TNF-α-induced EC apoptosis. A role for IFN-γ and TNF-α in damaging ECs was shown in other experimental systems,21,22 but these findings were not linked to asthma until the present study. It is interesting that we observed that activated T cells and eosinophils from healthy and atopic individuals do not differ in their capacity for inducing EC apoptosis. Apparently, bronchial ECs and skin keratinocytes utilize different predominant apoptosis pathways in response to allergic inflammation. Epidermal keratinocytes do not express Fas ligand in the course of eczema, and upregulation of Fas mainly by IFN-γ renders the cells susceptible to apoptosis, induced by Fas ligand provided by activated skin-infiltrating T cells.8 In contrast, the TNFR-related apoptotic pathway is predominant in bronchial EC. Approximately 70% of ECs expressed TNFR-I and TNFR-II by IFN-γ and TNF-α stimulation, whereas Fas expression was observed in only 20% of the cells. It is possible that ECs from atopic
Trautmann et al 335
versus healthy donors may show differences in sensitivity to different apoptosis-inducing ligands. The mucosa in asthma is distinguished from other inflammatory lung diseases such as cystic fibrosis by the presence of large numbers of activated eosinophils. The present study demonstrates that EC apoptosis induced by exposure to lysates from IFN-γ-stimulated eosinophils is mediated mainly by TNF-α. In addition to activated eosinophils, other sources of TNF-α in asthmatic airway inflammation might exist, such as macrophages, mast cells, and stimulated ECs.23 Tumor necrosis factor-α induces opposing cellular responses that result from activation of two different pathways: cytokine release via NF-κB activation or apoptosis via FADD-caspase 8 activation.24 However, the mechanisms by which these opposing programs are selected remain unclear. Substantial evidence supports the view that engagement of TNFR-I can trigger apoptosis in many cell types.24 Engagement of TNFR-II has been found to stimulate TNFR-I-mediated apoptosis and to be able to autonomously induce cell death.25,26 Other potential pathways of tissue injury by eosinophils exist, such as release of cytotoxic proteins including ECP, eosinophil-derived neurotoxin, eosinophil peroxidase, and major basic protein.27 Recombinant human ECP induced mainly necrosis of ECs in the pres-ent study. Eosinophil cationic protein might damage cells by a colloid-osmotic process, as it was shown that ECP can introduce non-ion selective pores in both cellular and synthetic membranes.28 To induce toxic EC death, doses far higher than those measured in the lysates of eosinophils were required. Eosinophil cationic protein induced EC necrosis at as early as 6 hours of incubation. However, activated T cells and eosinophils induce EC apoptosis after 3 days. Therefore, the present data do not fully support the possibility of a major role for ECP in EC apoptosis. The blocking effect of anti-TNF-α mAb suggests that major basic protein, eosinophil-derived neurotoxin, and eosinophil peroxidase may also not contribute predominantly to EC apoptosis. Apoptotic ECs may be cleared from the epithelial layer in vivo by macrophages and shedding. Apoptotic cells that have not been phagocytized and removed progress to secondary necrosis with loss of cell membrane integrity and the release of cytoplasmic contents. Increased DNA fragmentation shown by TUNEL and Hoechst staining in biopsy samples significantly correlated with the morphologic findings of EC damage. The highest number of apoptotic ECs was encountered in areas of initial EC detachment. In the cell culture experiments, loosening of cell-cell contacts is obvious when ECs become apoptotic. Druilhe et al29 discuss that their failure to detect differences in the number of TUNELpositive ECs between healthy and asthmatic subjects may be the result of the detachment of ECs into the bronchial lumen. In addition, the problem of real and artifactual EC shedding in evaluations of bronchoscopic airway biopsy specimens has to be considered.
Basic and clinical immunology
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2
336 Trautmann et al
Basic and clinical immunology
Recent studies challenge the concept that the eosinophil is the central effector cell in asthma.6,30,31 Furthermore, the issue of whether the asthmatic inflammation is fully mediated by type 2 T cells should be revisited.6,32 In humans there is not a straight bias for type 1 and type 2 differentiated T-cell subsets in disease. Even allergen-specific T-cell clones isolated from the airways of asthmatic subjects often release significant levels of IFN-γ along with IL-4.33,34 The amount of IFN-γ and TNF-α is elevated in supernatants from cultures of unstimulated and stimulated bronchoalveolar-lavagefluid cells from asthma patients.35 During exacerbations of asthma, viral infections may also contribute to the IFN-γ level in the airway mucosa.6,23 In the present study ECs exhibited a relatively low threshold in response to IFN-γ, in the range of 1 to 10 ng/mL. Several approaches to the inhibition of TNF-α effects are now under investigation in asthma, including mAb to TNF-α and soluble TNF-α R.36,37 Bronchial ECs exposed to IFN-γ and TNF-α secrete increased amounts of bFGF and TGF-β1 as determined by ELISA (authors’ unpublished observation). It is likely, that the stressed epithelium may contribute to the structural remodeling of the airways including thickening of the basement membrane and induction of myofibroblasts.23,38 The demonstration of bronchial EC apoptosis in cocultures with T cells and eosinophils as well as bronchial biopsies establishes the essential role of T cells and eosinophils in bronchial EC injury and destruction in asthma. This pathologic process, which represents a main direct tissue damage in asthma, is mediated by IFNγ and TNF-α and characterized by apoptosis and shedding of bronchial ECs. The knowledge of this molecular basis is pivotal in understanding the development of pathology in asthma, and opens a future for more focused therapeutic applications. REFERENCES 1. Laitinen LA, Heino M, Laitinen A, Kava T, Haahtela T. Damage of the airway epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 1985;131:599-606. 2. Jeffery PK, Wardlaw AJ, Nelson FC, Collins JV, Kay AB. Bronchial biopsies in asthma: an ultrastructural, quantitative study and correlation with hyperreactivity. Am Rev Respir Dis 1989;140:1745-53. 3. Montefort S, Roche WR, Roberts JA, Holgate ST. Bronchial epithelial shedding in asthmatics and non-asthmatics. Respir Med 1989;87:9-11. 4. Bentley AM, Menz G, Storz C, Robinson DS, Bradley B, Jeffery PK, Durham SR, et al. Identification of T lymphocytes, macrophages, and activated eosinophils in the bronchial mucosa in intrinsic asthma. Am Rev Respir Dis 1992;146:500-6. 5. Howarth PH, Bradding P, Montefort S, Peroni D, Djukanovic R, Carroll MP, et al. Mucosal inflammation and asthma. Am J Respir Crit Care Med 1994;150:S18-22. 6. Busse WW, Lemanske RF. Asthma. New Engl J Med 2001;344:350-62. 7. Weller PF. Human eosinophils. J Allergy Clin Immunol 1997;100:283-7. 8. Trautmann A, Akdis M, Kleemann D, Altznauer F, Simon HU, Graeve T, et al. T cell-mediated Fas-induced keratinocyte apoptosis plays a key pathogenetic role in eczematous dermatitis. J Clin Invest 2000;106:25-35. 9. Trautmann A, Altznauer F, Akdis M, Simon HU, Disch R, Bröcker EB, et al. The differential fate of cadherins during T cell-induced keratinocyte apoptosis leads to spongiosis in eczematous dermatitis. J Invest Dermatol 2001;117:927-34. 10. Hansel TT, de Vries IJM, Carballido JM, Braun RK, Carballido-Perrig N,
J ALLERGY CLIN IMMUNOL FEBRUARY 2002
11.
12.
13.
14.
15.
16.
17.
18.
19.
20. 21.
22.
23.
24. 25.
26.
27. 28.
29.
30.
31. 32.
Rihs S, et al. Induction and function of eosinophil intercellular adhesion molecule-1 and HLA-DR. J Immunol 1992;149:2130-6. Crameri R, Jaussi R, Menz G, Blaser K. Display of expression products of cDNA libraries on phage surfaces. A versatile screening system for selective isolation of genes by specific gene-product/ligand interaction. Eur J Biochem 1994;226:53-8. Rosenberg HF, Ackerman SJ, Tenen DG. Human eosinophil cationic protein. Molecular cloning of a cytotoxin and helminthotoxin with ribonuclease activity. J Exp Med 1989;170:163-76. Vermes I, Haanen C, Steffens-Nakken H, Reutlingsberger CDM. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labeled Annexin V. J Immunol Method 1995;184:39-51. Trautmann A, Akdis M, Schmid-Grendelmeier P, Disch R, Bröcker EB, Blaser K, et al. Targeting keratinocyte apoptosis in the treatment of atopic dermatitis and allergic contact dermatitis. J Allergy Clin Immunol 2001;108:839-46. Staley K, Blaschke AJ, Chun J. Apoptotic DNA fragmentation is detected by a semiquantitative ligation-mediated PCR of blunt DNA ends. Cell Death Diff 1997;4:66-75. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 1992;119:493-501. Akdis CA, Joss A, Akdis M, Faith A, Blaser K. A molecular basis for T cell suppression by IL-10: CD28-associated IL-10 receptor inhibits CD28 tyrosine phosphorylation and phosphatidylinositol 3-kinase binding. FASEB J 2000;14:1666-8. Montefort S, Herbert CA, Robinson C, Holgate ST. The bronchial epithelium as a target for the inflammatory attack in asthma. Clin Exp Allergy 1992;22:511-20. Ohashi Y, Motojima S, Fukuda T, Makino S. Airway hyperresponsiveness, increased intracellular spaces of bronchial epithelium, and increased infiltration of eosinophils and lymphocytes in bronchial mucosa in asthma. Am Rev Respir Dis 1992;145:1469-76. Simon HU, Blaser K. Inhibition of programmed eosinophil death—a key pathogenic event for eosinophilia? Immunol Today 1995;16:53-5. Kampf C, Relova AJ, Sandler S, Roomans GM. Effects of TNF-α, IFNγ and IL-1-β on normal human bronchial epithelial cells. Eur Respir J 1999;14:84-91. Gitter A, Bendfeldt K, Schulzke JD, Fromm M. Leaks in the epithelial barrier caused by spontaneous and TNF-α-induced single-cell apoptosis. FASEB J 2000;14:1749-53. Bousquet J, Jeffery PK, Busse WW, Johnson M, Vignola AM. Asthma. From bronchoconstriction to airways inflammation and remodeling. Am J Respir Crit Care Med 2000;161:1720-45. Pimentel-Muinos FX, Seed B. Regulated commitment of TNF receptor signaling: a molecular switch for death or activation. Immunity 1999;11:783-93. Weiss T, Grell M, Hessabi B, Bourteele S, Müller G, Scheurich P, et al. Enhancement of TNF receptor p60 mediated cytotoxicity by TNF receptor p80: requirement of the TNF receptor-associated factor-2 binding site. J Immunol 1997;158:2398-404. Haridas V, Darnay BG, Natarajan K, Heller R, Aggarwal BB. Overexpression of the p80 TNF receptor leads to TNF-dependent apoptosis, nuclear factor-κB activation, and c-Jun kinase activation. J Immunol 1998;160:3152-62. Gleich GJ. Mechanisms of eosinophil-associated inflammation. J Allergy Clin Immunol 2000;105:651-63. Young JDE, Peterson CGB, Venge P, Cohn ZA. Mechanism of membrane damage mediated by human eosinophil cationic protein. Nature 1986;321:613-14. Druilhe A, Wallaert B, Tsicopoulos A, Lapa e Silva JR, Tillie-Leblond I, Tonnel AB, et al. Apoptosis, proliferation, and expression of Bcl-2, Fas, and Fas ligand in bronchial biopsies from asthmatics. Am J Respir Cell Mol Biol 1998;19:747-57. Lee NA, Gelfand EW, Lee JJ. Pulmonary T cells and eosinophils: coconspirators or independent triggers of allergic respiratory pathology. J Allergy Clin Immunol 2001;107:945-57. Boushey HA, Fahy JV. Targeting cytokines in asthma therapy: round one. Lancet 2000;356:2114-16. Holtzman MJ, Sampath D, Castro M, Look DC, Jayaraman S. The onetwo of T helper cells: does interferon-γ knock out the Th2 hypothesis for asthma? Am J Respir Cell Mol Biol 1996;14:316-18.
33. Krug N, Madden J, Redington AE, Lackie P, Djukanovic R, Schauer U, et al. T-cell cytokine profile evaluated at the single cell level in BAL and blood in allergic asthma. Am J Respir Cell Mol Biol 1996;14:319-26. 34. Parronchi P, Macchia D, Piccini MP, Biswas P, Simonelli C, Maggi E, et al. Allergen- and bacterial antigen-specific T-cell clones established from atopic donors show a different profile of cytokine production. Proc Natl Acad Sci USA 1991;88:4538-42. 35. Cembrzynska-Nowak M, Szklarz E, Inglot AD, Teodorczyk-Injeyan JA. Elevated release of tumor necrosis factor-α and interferon-γ by bron-
Trautmann et al 337
choalveolar leukocytes from patients with asthma. Am Rev Respir Dis 1993;147:291-5. 36. Barnes PJ. New directions in allergic diseases: mechanism-based antiinflammatory therapies. J Allergy Clin Immunol 2000;106:5-16. 37. Stirling RG, Chung KF. New immunological approaches and cytokine targets in asthma and allergy. Eur Respir J 2000;16:1158-74. 38. Morishima Y, Nomura A, Uchida Y, Noguchi Y, Sakamoto T, Ishii Y, et al. Triggering the induction of myofibroblasts and fibrogenesis by airway epithelial shedding. Am J Respir Cell Mol Biol 2001;24:1-11.
Basic and clinical immunology
J ALLERGY CLIN IMMUNOL VOLUME 109, NUMBER 2