Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Therapeutic irradiation of thoracic malignancies often induces radiation pneumonitis and interstitial lung fibrosis. This is a limiting factor for radiotherapy (1). Pathological characteristics of pneumonitis are interstitial edema, interstitial and alveolar cellular inflammation, and an increase in number of type II pneumocytes (2, 3). The fibrotic process is characterized by increased interstitial collagen deposition, thickening of vascular walls, vascular occlusion (4), and a concomitant augmentation of interstitial macrophages, as well as a pan-leukocytic infiltration of the interstitium (5). The time course of interstitial inflammation following irradiation has recently been studied in detail (6, 7). Among the various inflammatory cell types, macrophages are most prominent. They are also found in the lumen of airways and in the alveolar space (8). Macrophages increase in number shortly after irradiation. They have been shown to produce a wide spectrum of cytokines which regulate fibroblast proliferation and the production of extracellular matrix (9). These include platelet-derived growth factor (PDGF) (10), interleukin-1 (IL-1) (11), tumor necrosis factor  (TNF) (12), basic fibroblast growth factor (bFGF) (13), and transforming growth factor  (TGF) (14). These cytokines not only regulate fibroblast function and collagen synthesis, but also recruit other inflammatory cells and stimulate their cytokine production. The role of individual cytokines released early in the inflammatory phase of pulmonary fibrosis remains to be elucidated. In addition to TGF and TNF, IL-4 is shown to induce collagen synthesis by cultured fibroblasts (15, 16). Therefore, we were interested to document the presence of IL-4 during the development of irradiation-induced pulmonary fibrosis. Here we show that IL-4 is produced in lung tissue during the early inflammatory response to radiation. IL-4 production continues for a long period of time concomitantly to the generation of interstitial fibrosis. Furthermore, we demonstrate that under these conditions pulmonary macrophages are a source of IL-4.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals

CDF^R (F-344)/CrlBR male Fischer rats were purchased from Charles River (Sulzfeld, Germany). The right lung was irradiated by a single dose of ultrahard X-rays (20 Gray, linear accelerator, 9-MeV Photons). The contralateral lung was protected against direct radiation by absorbers.

Cells and Cell Culture

Rat basophil leukemia cells of the 2H3 subline (RBL-2H3) were a gift from Dr. M. Kent (Arthritis and Rheumatism Branch, National Institute of Arthritis and Musculoskeletal Skin Diseases, National Institute of Health, Bethesda, MD). 7 × 10⁵/ml RBL-2H3 cells were cultured in Quadriperm dishes (Heraeus, Hanau, Germany) and maintained in DMEM containing 2 mM glutamin (Biochrom, Berlin, Germany), 10% FCS (Biochrom), 100 U/ml penicillin, and 100 µg/ml streptomycin.

Bronchoalveolar Lavage Cells

Bronchoalveolar lavage (BAL) cells were isolated by repeated lavage of the airways with a total of 50 ml Hank's balanced salt solution (HBSS; Gibco, Eggenstein, Germany). Cytospins were prepared using a Shandon cytocentrifuge (Shandon GmbH, Frankfurt/Main, Germany).

Dot Blotting Studies

The reactivity of rabbit anti-human IL-4 IgG (BL4P; Genzyme, Rüsselsheim, Germany) was tested by dot blot analysis on recombinant human IL-4 (1 × 10⁸ U/mg, 10 µg/ml, Genzyme; and 5 × 10⁶ U/mg, 100 µg/ml, Pharmingen, San Diego, CA), on recombinant murine IL-4 (5,000 U/ml, 5 µg; Boehringer Mannheim, Mannheim, Germany), on recombinant rat IL-4 obtained from unpurified conditioned medium of a Chinese hamster ovary cell line (CHO cells) transfected with cDNA for rat IL-4 (Genzyme) as well as on cell lysate extracts from 1 × 10⁷ rat mast cells (RBL-2H3) as controls. RBL-2H3 cells were grown to confluence in 75 cm² tissue culture flasks (Greiner, Frickenhausen, Germany). Culture medium was removed by washing twice with PBS and the cells were lysed in situ with 1 ml of 50 mM sodium borate, pH 8.0, containing 1% NP-40, 150 mM NaCl, 0.5% sodium deoxycholate, and protease inhibitors (0.1 mg/ml PMSF, 1 µg/ml aprotinin, 1 µg/ml leupeptin), and sonicated (3 pulses, 30 s each) on ice. After 30 min, the lysates were spun by centrifugation at 12,000 × g for 10 min at 4°C. Supernatants were spotted onto nylon membranes at different concentrations containing 10 ng/µl, 1 ng/µl, 100 pg/µl, and 10 pg/µl of protein. Recombinant rat IL-4 and cell lysates were applied undiluted and at a 1:10 dilution. Membranes were washed in 0.1 M maleinic acid (0.15 M NaCl), pH 7.5, followed by incubation in 1% blocking reagent (Boehringer Mannheim) for 60 min at 21°C and covered with 40 µg/ml BL4P in 1% blocking reagent overnight at 4°C. Unbound Ab was removed by washing twice for 15 min with 0.1 M maleinic acid (0.15 M NaCl), pH 7.5. Filters were covered with alkaline phosphatase (AP) conjugated goat anti-rabbit IgG (1:5,000; Dianova, Hamburg, Germany) for 30 min at 21°C. The unbound Ab conjugate was removed by washing twice for 15 min with 0.1 M maleinic acid (0.15 M NaCl), pH 7.5, at 21°C. The membrane was incubated in 0.1 M Tris/HCl, pH 9.5, 0.1 M NaCl, 0.05 M MgCl₂ (pH 9.5), and the signals were detected by incubation in 10 ml staining solution consisting of 45 µl NBT (75 mg/ml nitroblue tetrazolium salt in dimethylformamide, 70%; Boehringer Mannheim), 35 µl X-phosphate solution (50 mg/ml 5-bromo-4-chloro-3-indolyl phosphate, toluidinium salt in dimethylformamide, 70%; Boehringer Mannheim), and 200 µl levamisole (0.05 M; Serva, Heidelberg, Germany) in 0.1 M Tris/HCl, pH 9.5. The spots appeared after 5 min, and the development was stopped by washing the filters for 5 min with 50 ml 0.01 M Tris/HCl (0.001 M EDTA), pH 8.0.

Binding Inhibition

Binding inhibition studies were performed both by dot blot studies and by immuno-histochemistry. The BL4P Ab (2.5 µg/ml) was preincubated with recombinant human IL-4 (0.3 µg/ml; Genzyme) for 60 min at 37°C prior to addition to RBL-2H3 rat mast cell lysate (undiluted, 1:10), respectively, recombinant human IL-4 (10 ng/µl up to 10 pg/µl; Genzyme) in the dot blot.

The BL4P Ab (20 µg/ml) was preincubated with purified recombinant human IL-4 (3 µg/ml; Genzyme) or with undiluted culture supernatant containing rat IL-4 for 120 min at 37°C prior to addition to cryostat sections of rat lung tissue prepared at various time points after irradiation.

Ribonuclease Protection Assay

RNA isolation. Total RNA was extracted from rat lungs using the Total RNA Separator Kit from Clontech Laboratories, Palo Alto, CA.

Probe synthesis. cDNA of rat IL-4 was kindly provided by Dr. A. Neil Barclay (MRC Cellular Immunology Unit, Sir William Dunn School of Pathology, University of Oxford, Oxford, UK). 520 bp antisense riboprobes were generated after linearization of rat IL-4 cDNA cloned in pT7T319U with EcoR1 (Gibco) for antisense orientation. GAPDH riboprobes were generated from a commercially available template (Ambion, Austin, TX). Linearized cDNAs were labeled with 50 µCi [-³²P] CTP, 800 Ci/mmol, 10 mCi/ml (DuPont, Bad Homburg, Germany), and 3 µl of 0.05 mM CTP by in vitro transcription with T7 polymerase (for IL-4 cRNA) and T3 polymerase (for GAPDH cRNA) to a specific activity of 3-5 × 10⁸ cpm/µg by using the MAXIscript^TM In Vitro Transcription Kit from Ambion. The efficiency of the transcription reaction was determined by measuring the radioactivity incorporated into an aliquot after trichloroacetic acid precipitation. The probes were purified on a 0.75 mm thick 5% acrylamide gel/8 M urea to get full-length probes. The expected sizes of the transcripts were about 520 bases (IL-4) and 355 bases (GAPDH).

Hybridization. The hybridization was performed utilizing a modification of Ambion's HybSpeed^TM RPA. Briefly, 1 × 10⁵ cpm of each probe were mixed with 5 µg of RNA supplemented to a total of 50 µg with yeast RNA (ITC, Heidelberg, Germany), and ethanol precipitated for 15 min at 20°C. Ten µl of pre-heated (95°C) HybSpeed Hybridization Buffer (Ambion) was added to pelleted probes and sample RNAs. After 15 min of incubation at 68°C, the reaction was incubated with 50 µl of RNaseA/T1 mix (1:100; ITC) for 30 min at 37°C. The reaction was inactivated and samples were precipitated with 150 µl stop solution (Guardian^TM RNase Protection Assay; ITC) for 20 min at 20°C. Following centrifugation at 18,000 rpm for 15 min at 4°C the pellets were dissolved in 10 µl Gel Loading Buffer (Ambion), vortexed, and loaded on a 5% acrylamide/8M urea denaturing gel. Separation was performed in TBE buffer (0.09 M Tris, 0.09 M boric acid, 2 mM EDTA) at 450 volts for about 90 min at 20°C. The gel was transferred to filter paper, covered with plastic wrap, dried at 75°C, and exposed 3 h to a Phosphor Screen (PhosphorImager SI^TM; Molecular Dynamics GmbH, Krefeld, Germany).

Western Blotting

Irradiated right lungs, non-irradiated contralateral left lungs, and lungs obtained from untreated animals at the same time points as for ribonuclease protection assay were homogenized using an ultraturrax in ice-cold extraction buffer containing 50 mM Tris, pH 7.5, 150 mM NaCl, 1% Triton, 10 µM leupeptin, 1 mM PMSF, 1 mM EDTA, 0.1 U aprotinin, and 0.2 mM DTT. The homogenates were kept under gentle stirring for 60 min on ice. Insoluble material was removed by centrifugation for 60 min at 50,000 × g at 4°C. The protein concentration was determined according to Bradford (17).

Proteins of the extracts normalized for protein content were denatured and reduced by treatment with 50 mM Tris/HCl, pH 7.4, 10 mM SDS, 50% glycerol, 5% bromophenolblue, and 200 mM 2-mercaptoethanol at 95°C for 5 min. The proteins were separated by electrophoresis on a 4-20% SDS polyacrylamide gel and blotted onto nitrocellulose membranes. Because purified rat IL-4 was not available, 200 ng of crossreacting purified recombinant human IL-4 were taken as standard. IL-4 was detected by a monoclonal mouse anti-human IL-4 antibody cross-reacting with rat IL-4 (Genzyme) using an amplifying Biotin-AP System (Boehringer Mannheim). Unspecific binding was blocked by incubation with 10% nonfat milk in 50 mM Tris/HCl, pH 7.5, 150 mM NaCl containing 0.25% Tween-20 overnight at 4°C. The blots were evaluated by densitometry on a Gel Doc System 1000 UV (BioRad, München, Germany).

In Situ Hybridization

Rat IL-4 cDNA was kindly provided by Dr. A. Neil Barclay (MRC Cellular Immunology Unit Sir William Dunn School of Pathology, University of Oxford, Oxford, UK). 520 bp antisense and sense riboprobes were generated after linearization of rat IL-4 cDNA cloned in pT7T319U with EcoR1 (Gibco) for antisense orientation and with BamH1 (Gibco) for sense orientation. Linearized cDNAs were labeled with digoxigenin by in vitro transcription with T7 and T3 polymerase by using the DIG RNA Labeling Kit from Boehringer. Transcription products were checked in a formaldehyde gel containing 1% agarose. The efficiency of labeling reaction was routinely tested by dot blotting.

Sections of frozen lung tissue (12 µm) were prepared, mounted on silane-coated slides pretreated with 2% solution of 3-amino-propyltrithoxysilane (18). Cytospin preparations for in situ hybridization were obtained by centrifugation of 1 × 10⁴ RBL-2H3 or BAL cells onto silane-coated slides using a cytocentrifuge (Shandon). The sections and cytospins were heated for 2 min at 50°C to fix the RNA, dried for 30 min, and then fixed in 4% phosphate-buffered paraformaldehyde for 7 min at room temperature.

Tissue sections and cytospins were washed once in PBS, pH 7.4, for 3 min, twice in 2× SSC (0.3 M NaCl, 0.03 M sodium citrate, pH 7.0) for 5 min, and prehybridized in 60% formamide/2× SSC; 0.1 M DDT; 500 µg/ml salmon sperm DNA; 250 µg/ml yeast tRNA; 5% dextran sulfate; 10× Denhardt's solution for 2 h at 42°C and hybridized overnight at 42°C using 3 µg/ml cRNA diluted in 100 µl prehybridization solution in a sealed humidified chamber. To remove the unbound cRNA probe under high stringency conditions the slides were washed once in 2× SSC for 5 min, three times with 60% formamide in 0.2× SSC for 5 min, and twice in 2× SSC for 5 min under gentle agitation. In order to demonstrate the specificity of the in situ hybridization, control slides were pretreated prior to hybridization with RNase A1 (20 µg/ml in RNase buffer containing 0.5 M NaCl, 10 mM Tris, pH 7.5, 1 mM EDTA; 30 min at 37°C). In addition, slides were treated with RNase A1 (20 µg/ml in RNase buffer, 30 min at 37°C) after hybridization and washing. The specimens were rinsed for 5 min at room temperature with 100 mM Tris/HCl, pH 7.5, and 150 mM NaCl, and incubated with 1% blocking reagent (pH 7.5; Boehringer Mannheim) for 1 h followed by reaction with AP-conjugated anti-digoxigenin polyclonal Ig (1: 100 in 0.1% blocking reagent in 0.1 M maleinic acid) for 1 h at 21°C. The bound Ab was detected by adding NBT/BCIP as substrate for an incubation period of 17 h. Sections were counterstained with nuclear Fast red and mounted. The following preparations were used as controls: tissue sections pretreated with RNase A1 (20 µg/ml in RNase buffer, 30 min at 37°C), sections hybridized with IL-4 sense probes (3 µg/ml cRNA; 100 µl per section), and sections treated with the prehybridization solution without consecutive hybridization. The rat mast cells RBL-2H3 were hybridized in the same manner and used as a positive control.

Immunofluorescence Staining

Since mouse monoclonal antibodies were used to identify rat macrophages in two-color fluorescence staining, the polyclonal rabbit anti-human IL-4 Ab BL4P was chosen for the detection of IL-4. Tissue sections on silane-coated slides were incubated with 10% normal goat serum (The Binding Site, Heidelberg, Germany) in PBS/0.5% BSA for 30 min at 21°C to block Fc receptor binding sites, followed by incubation with 5 µg/ml polyclonal rabbit anti-human IL-4-Ab (BL4P) diluted in PBS containing 0.5% BSA for 1 h at 21°C or at 4°C overnight. After washing in PBS/BSA the sections were covered with DTAF (dichlorotriazinyl-aminofluorescein)-conjugated F(ab')2 goat anti-rabbit Ig (Dianova) at a 1:40 dilution in PBS/0.5% BSA for 20 min at 4°C and embedded in anti-fading solution consisting of 9 parts glycerol, 1 part 1 M PBS, pH 7.5, 2% 1.4-diazo-bicyclo-[2,2,2]-octane. In control experiments, unspecific Ig derived from the same species as the primary Ab was used at corresponding concentrations.

Double Immunofluorescence Staining

In order to assess IL-4 localization in macrophages, tissue sections were incubated simultaneously with anti-IL-4 Ig (BL4P) and with rat macrophage specific mouse monoclonal Abs ED1, ED2, and RM1. After washing, bound antibodies were detected by DTAF conjugated F(ab')2 goat anti-rabbit Ig (Dianova) and indocarbocyanin (Cy3)-labeled F(ab')2 goat anti-mouse Ig (Dianova) as second antibodies.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

IL-4 Gene Expression Following Irradiation

In a first approach, we determined the IL-4 gene activation as well as the production of IL-4 in lung tissue. For this purpose right (irradiated) and left (non-irradiated) lungs were prepared at day 1, 7, 14, 21, 28, 56, and 84 following irradiation and total RNA was extracted. Lung tissue from untreated animals served as an additional control. IL-4 mRNA was evaluated by the ribonuclease protection assay. The expected sizes for IL-4 and GAPDH hybrids were 520 bases and 355 bases, respectively. Labeled RNA molecular mass markers (RNA Century Marker Template Set; Ambion) were utilized to assess the size of protected fragments. As shown in Figure 1, IL-4 mRNA was upregulated in the irradiated lung tissue whereas the contralateral non-irradiated lungs and lungs from untreated animals did not show changes in the IL-4 mRNA concentration.

View larger version (54K): [in this window] [in a new window]   Figure 1.   Evaluation of IL-4 mRNA by ribonuclease protection assay. Rat IL-4 mRNA was detected in 10 µg of total RNA prepared from right (irradiated) and left (non-irradiated) rat lungs at various intervals after irradiation. Total RNA from rat lungs of untreated animals served as an additional control. Antisense RNA probes of high specific activity (3-5 × 108 cpm/µg) to IL-4 and GAPDH were synthesized and 1 × 105 cpm of each probe was processed in a multiple probe ribonuclease protection assay using the HybSpeedTM RPA Kit (Ambion). Panel A shows a comparison of the relative expression levels of each probe within RNA extracted from rat lung tissue. The expected sizes of IL-4 and GAPDH hybrids were 520 bases and 355 bases, respectively. Probe concentrations utilized were found to be in excess of target for both IL-4 and GAPDH. Control lanes on the left side show yeast RNA treated with RNase and IL-4 and GAPDH RNA without RNase (total probe). Only 40% of the total probes were loaded onto the gel to avoid overexposure. (Panel B) The autoradiographs were evaluated by densitometry on ImageQuaNTTM analysis (Molecular Dynamics GmbH, Krefeld, Germany). The amount of RNase-treated hybridization products of IL-4 (shown in autoradiographs in A) was calculated in percent of the obtained activity with the undigested IL-4 probe. The variation in concentration of total RNA was assessed by the mean value of the GAPDH-standard after hybridization.

The production of IL-4 protein of one animal per time point was determined by Western blotting and is documented in Figure 2. When 200 µg of total lung protein were separated on a SDS-polyacrylamide gel, IL-4 protein could only be detected in irradiated lungs (Figure 2). The amount of IL-4 found in single animals at various time points after irradiation increased. The highest concentration was found in a lung prepared at day 84 after irradiation. The same amount of protein obtained from non-irradiated lungs and from untreated lungs of control animals did not contain a detectable amount of IL-4 at any time point tested.

View larger version (57K): [in this window] [in a new window]   Figure 2.   Western blot analysis of IL-4 protein production in irradiated rat lung tissue at various time points after irradiation. 200 µg of total lung protein was separated on a SDS polyacrylamide gel and blotted. IL-4 was detected by an AP-conjugated mouse monoclonal anti-human IL-4 antibody crossreactive with rat IL-4. The density of single bands was determined and calculated as percent of the density obtained with 200 ng of purified recombinant human IL-4 separated and blotted under identical conditions.

Cellular Distribution of IL-4 Gene Expression

Having demonstrated IL-4 synthesis as a consequence of lung irradiation it was of interest to identify the cellular origin of IL-4. Non-isotopic in situ hybridization with IL-4 specific cRNA (antisense and sense) was performed on frozen lung tissue sections prepared from irradiated, non-irradiated, and untreated animals as well as on cytospin preparations from BAL cells. In order to validate the in situ hybridization technique for the detection of IL-4 mRNA, the following control experiments were performed. Cytospin preparations of the rat mast cell line RBL-2H3 known to secrete IL-4 were hybridized with the antisense probe. As shown in Figure 3A, a strong signal was obtained with these cells, whereas hybridization with the sense probe (Figure 3B) remained negative. In addition, when tissue sections of irradiated lungs were treated with RNase prior to hybridization, no binding of the antisense probe was detected (Figure 3C). Unspecific binding of the anti-digoxigenin antibody conjugated to alkaline phosphatase could be excluded by staining of tissue where hybridization with the antisense probe has been omitted (Figure 3D). The IL-4 specific hybridization pattern of irradiated lung sections as detected by immunophosphatase staining showed a strong reactivity in peribronchiolar cells, interstitial cells and epithelial cells of alveolarseptal regions, and in free cells located in the alveolar lumen which, by morphological criteria, were identified as macrophages (Figure 4). Lungs developing injury showed a more intense peribronchial and alveolarseptal staining than non-irradiated lungs and lungs from untreated animals. IL-4 mRNA expression analyzed one day after irradiation was less pronounced in alveolarseptal areas than in peribronchial tissue (Figure 4C). This reactivity pattern became more obvious with progression of fibrosis. The peribronchiolar and parenchymal signals were particularly strong in fibrotic areas 2 and 3 months following irradiation (Figure 4D and E). No upregulation of IL-4 gene was observed in the contralateral, non-irradiated lungs nor in lungs from untreated animals (Figure 4A and B). Since no difference of staining pattern in the non-irradiated lungs at various time points following irradiation was found, only the results of one time point are shown.

View larger version (96K): [in this window] [in a new window]   Figure 3.   Positive and negative controls of in situ hybridization experiments. (Panel A) RBL-2H3 mast cells. In situ hybridization with antisense IL-4 probe. Original magnification: ×400. (Panel B) RBL-2H3 mast cells. In situ hybridization with sense IL-4 probe. Original magnification: ×400. (Panel C) Section of irradiated lung (1 d after irradiation) following RNase treatment prior to hybridization. Original magnification: ×250. (Panel D) Irradiated lung tissue (day 1) with omitted antisense IL-4 probe. Original magnification: ×400. Counterstaining of cells in panels A-D with Fast Red.

View larger version (106K): [in this window] [in a new window]   Figure 4.   Localization of IL-4 mRNA as determined by immunoalkaline phophatase reaction using NBT/BCIP as substrate in formalin-fixed rat lung tissue sections following in situ hybridization with RNase treatment prior to hybridization. Note the increased purple-blue staining of parenchymal lung cells in panels D and E in comparison with that of lungs of untreated animals (panels A and B). (A) Lung tissue of untreated animals with faint staining in parenchymal cells (arrows). (B) Left, non-irradiated lung sections at day 1 following irradiation with slight staining (arrows). (C) At day 1 after irradiation, the purple-blue stain increased; however, the IL-4 mRNA expression was less pronounced in alveolarseptal areas (arrow) than in peribronchial tissue (arrowheads). (D) Strong infiltration of the parenchymal tissue by intensely stained cells and thickening of vascular walls (arrows) observed 2 months after irradiation. (E) Tissue alterations are characterized by vascular occlusion with particularly strong signals in fibrotic areas (arrows) 3 months after irradiation. Counterstaining with Fast Red. Original magnification of panels A-E: ×400.

In order to further analyze the cellular distribution of IL-4 mRNA, BAL cells were collected from lungs of irradiated rats at various intervals after irradiation. As shown in Figure 5A, IL-4 mRNA was mainly found in cells resembling alveolar macrophages and in lymphocytes. The proportion of BAL cells expressing positive IL-4 mRNA signals 21 d after irradiation was about 30% with a relative proportion of monocytes/macrophages and of lymphocytes of 2.5 to 1. In BAL cells prepared from untreated control rats, no IL-4 message was detected (Figure 5B).

View larger version (82K): [in this window] [in a new window]   Figure 5.   In situ hybridization of BAL with an antisense IL-4 probe. (A) BAL cells prepared at day 21 following irradiation. IL-4 mRNA positive alveolar macrophages (arrows) and lymphocytes (arrowheads) are indicated. (B) BAL cells prepared from untreated control rats. Counterstained with Fast Red. Original magnification: ×400.

Irradiation Induces IL-4 Production in Lung Macrophages

In order to investigate in more detail the question as to whether lung macrophages are involved to a substantial degree in IL-4 production after irradiation, a double color fluorescence analysis was performed on lung tissue sections and on BAL cells. Macrophages were identified by the monoclonal antibodies ED1, ED2, and RM-1 which are known to define antigens differentially expressed on alveolar and interstitial macrophages in rats. The cells were simultaneously stained with the polyclonal antibody BL4P raised against human recombinant IL-4 which, in accordance with our experience, cross-reacts with rat IL-4. The specific recognition of rat IL-4 by this antibody was confirmed by binding and binding inhibition studies in the dot blot technique using IL-4 from various species. As can be seen from Figure 6A, the BL4P antibody reacted with two different specimens of human IL-4 and with recombinant rat IL-4. However, it did not recognize murine IL-4. In addition, it stained the cell lysate of the rat mast cell line RBL-2H3 known to produce IL-4. This reaction was completely blocked by preincubation of the BL4P antibody with human IL-4. By preincubation of the BL4P antibody with unpurified recombinant rat IL-4, a significant reduction of binding was observed; however, a complete inhibition was not obtained probably due to the insufficient concentration of the rat IL-4 in the supernatant of the transfected CHO-cells (Figure 6B). When the BL4P anti IL-4 antibody was used for staining of lung tissue in indirect immunofluorescence only few labeled cells were seen in normal lungs from untreated animals, whereas in irradiated lung tissue IL-4 could be detected in a great number of small sized cells already one day after irradiation (Figure 7A and C). The staining pattern of sections of the contralateral non-irradiated lung from the same animal did not differ from that of untreated controls (Figure 7B). The increased number of IL-4 positive cells in irradiated lung tissue was still seen at day 84 after irradiation. At this time, the IL-4 positive cells were concentrated in form of large clusters within the fibrotic tissue (Figure 7D). In contrast to the homogeneous staining of small sized cells early after irradiation, the labeled cells at the latest stage showed an extended cytoplasm and a speckled fluorescence. The specific detection of rat IL-4 by the BL4P antibody in tissue sections was again confirmed by preincubation of the antibody with either human recombinant IL-4 or rat recombinant IL-4. Human recombinant IL-4 at a final concentration of 3 µg/ml completely blocked the binding activity of the antibody. Staining of cells was only partially blocked when the BL4P Ab was preincubated with a supernatant containing rat IL-4 (the only form of recombinant rat IL-4 which was commercially available) (Figure 7E and F).

View larger version (47K): [in this window] [in a new window]   Figure 6.   Recognition of rat IL-4 by the polyclonal antibody BL4P raised against human IL-4. Binding and binding inhibition studies were performed in the dot blot technique to confirm the specific recognition of rat IL-4 by the polyclonal BL4P. (Panel A) Dot blot analysis of purified recombinant human IL-4 and recombinant murine IL-4, unpurified recombinant rat IL-4 and mast cell lysates with polyclonal rabbit anti-human IL-4. 10 ng, 1 ng, 100 pg, and 10 pg of recombinant human and murine IL-4 were spotted onto the nylon membrane. The protein concentrations of rat IL-4 and mast cell lysates could not be determined. The polyclonal anti-IL-4 is reactive with all samples (lanes 1, 2, 4, and 5) except for recombinant murine IL-4 (lane 3). (Panel B) Pretreatment of the antibody BL4P at 2.5 µg/ml with 0.3 µg/ml recombinant human IL-4 was sufficient to completely block the reaction of BL4P with recombinant human IL-4 and rat IL-4 from RBL-2H3 mast cell lysate (lanes 2 and 5). Lanes 3 and 6 show the incompletely blocked reaction of BL4P pretreated with recombinant rat IL-4. Lanes 1 and 4 show the reaction of the Ab with recombinant human IL-4 as well as rat IL-4 without pretreatment of Ab.

View larger version (32K): [in this window] [in a new window]   Figure 7.   Indirect immunofluorescence staining with the polyclonal rabbit anti-human IL-4 Ab on rat lung tissue. Staining for IL-4 is indicated by green fluorescence. Immunofluorescence staining of lung tissue sections of untreated animals (A) and of contralateral non- irradiated lung tissue sections (B). Immunofluorescence staining of lung tissue sections 1 d after irradiation (C). Immunofluorescence on lung tissue sections at day 84 after irradiation (D). The staining pattern following reaction with BL4P Ab at day 1 after irradiation pretreated with recombinant human IL-4 is shown in panel E. Panel F indicates the staining pattern of BL4P Ab one day after irradiation pretreated with recombinant rat IL-4. Original magnification: ×250.

When the IL-4 staining with the BL4P antibody was combined with the labeling of cell surface structures selectively expressed on macrophages (mAb ED1, ED2, RM1) it became evident that a considerable number of IL-4 positive cells in irradiated lungs were macrophages. Figure 8A shows cells (21 d after irradiation) stained by the anti IL-4 antibody and the RM1 antibody which recognizes an antigen on rat macrophages and dendritic cells. Since the green fluorescence of the IL-4 staining is often hardly seen when the same cell is stained by Cy3-labeled antibody (Figure 8A) confocal laser microscopy was applied in order to distinguish the intracytoplasmic IL-4 staining and the labeling of a membrane structure. This is illustrated in Figure 8B, where the intracytoplasmic IL-4 staining was combined with the membraneous labeling by the Cy3 coupled secondary antibody to the ED2 monoclonal antibody (56 d after irradiation).

View larger version (47K): [in this window] [in a new window]   Figure 8.   Double color fluorescence analysis of lung macrophages (lung tissue and BAL) involved in IL-4 production after irradiation. Macrophages were identified by the monoclonal Ab ED1, ED2, and RM1 (red fluorescence) and were simultaneously stained with the BL4P anti-IL-4 Ab (green fluorescence). (Panel A) Arrow shows an IL-4/RM1 double positive macrophage in irradiated rat lung tissue (21 d after irradiation). (Panel B) Confocal laser microscopy was applied in order to distiguish the intracytoplasmic IL-4 staining (green) and the labelling of a membrane structure (red) by the ED2 monoclonal Ab (arrow) following double immunofluorescence (56 d after irradiation). Original magnification: ×400. (Panel C) IL-4 producing macrophages were identified in BAL cell preparation of irradiated animals (56 d after irradiation). The macrophages on the cytospin were stained with the BL4P Ab and with ED1. Arrows indicate IL-4 positive cells. Original magnification: ×400.

Also in BAL cell preparations obtained after irradiation, IL-4 producing macrophages were identified by two-color immunofluorescence staining as documented in Figure 8C (56 d after irradiation). It shows a cytospin stained with the anti IL-4 antibody (green fluorescence) and with mAb ED1 (red fluorescence of the Cy3-labeled secondary antibody) that has been reported to react with alveolar macrophages.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The aim of this study was to document the presence of IL-4 in the pathogenesis of radiation pneumonitis and radiation lung fibrosis. We chose the rat model since this animal proved to be appropriate for the selective irradiation of one lung with the linear accelerator leaving the contralateral lung protected. By this procedure, it was possible to determine the local induction of cytokines and to correlate it with the irradiation by comparison with the contralateral shielded lung. After a single dose of 20 Gy applied to one lung, first histological signs of developing interstitial fibrosis can be observed after about 4 wk. As shown in this report, IL-4 gene activation and IL-4 production were determined at various time intervals after irradiation both on total lung tissue as well as on a single cell level in the irradiated lung and, for comparison, in the contralateral protected lung and in lungs of untreated rats. For detection of IL-4 mRNA in total lung tissue we used the ribonuclease protection assay because it proved to be more sensitive than Northern hybridization (19). When animals were killed at different time points after irradiation and compared with each other, an increase of the relative amount of IL-4 mRNA was observed up to day 28 where a plateau level was reached. This level did not markedly change until 3 months after irradiation where the analysis was stopped. The amount of IL-4 protein in the irradiated lungs increased continuously, attaining the highest concentration at the end of the experimental period of 3 months after irradiation. In the contralateral lungs, no increase of IL-4 gene activation or of IL-4 production was observed. This indicates a correlation between local IL-4 protein expression and the development of radiation-induced pulmonary fibrosis. The IL-4 mRNA levels and the IL-4 protein levels do not closely correlate in the late stages of the development of pulmonary fibrosis. This might point to an intracellular storage of IL-4 protein similar to the reported documentation of TNF in mast cells (20, 21).

In order to localize the source of IL-4 in the irradiated lung tissue, both in situ hybridization and immunofluorescence analysis were performed. We found IL-4 gene transcription and IL-4 production concentrated in peribronchial and perivascular areas as well as in cells located in the alveolar lumen and in the alveolar septa. This pattern corresponds to the regions of cellular infiltration following radiation injury. IL-4 is the master cytokine of the Th2 type immune response (22). It has been found that it is the IL-4 itself which programs CD4⁺ T-lymphocytes for IL-4 production (23, 24). There is, however, no clearcut evidence for the origin of the initial IL-4 nor is it known what signals trigger the secretion of this primary inducing IL-4. Besides CD4⁺ T-lymphocytes, various cell types have been shown to produce IL-4: mast cells, basophilic granulocytes and, particularly in mice, NK 1.1⁺ T cells (25). Mast cells have been described to increase in number during the development of radiation fibrosis of the lung (26). In our analysis, a great deal of IL-4 positive cells morphologically appeared to be macrophages. In order to confirm this finding, two-color immunofluorescence staining of tissue sections and of cytospin preparations of bronchoalveolar lavage cells was performed using the monoclonal antibodies ED1, ED2, and RM1 which specifically recognize rat macrophage and dendritic cell populations (27, 28). The staining was performed simultaneously with the polyclonal anti-human IL-4 antibody, BL4P. The crossreaction with rat IL-4 was confirmed by binding and binding inhibition studies in the dot blot analysis. Early after irradiation IL-4-producing RM1 positive cells were found in sections of irradiated lungs. These cells disappeared at later stages of developing fibrosis. Cells stained by the mAb ED1 which was reported to label dendritic cells, monocytes, and newly recruited as well as resident macrophages mainly within the alveolar space (29) appeared as large clusters and showed an extended cytoplasm with a speckled fluorescent staining pattern. Increasing numbers of ED1/IL-4 double positive cells were found in the course of developing fibrosis. In addition, IL-4-producing cells stained by the mAb ED2 which defines highly differentiated macrophages appeared only during advanced stages of fibrosis. By confocal laser microscopy, we were able to unequivocally show that ED2 surface positive cells expressed IL-4 intracellularly. Taken together, the data obtained by two-color fluorescence labeling not only show for the first time that macrophages participate in the production of IL-4 during the generation of pneumonitis and fibrosis following irradiation but also suggest that different subpopulations of macrophages are involved in the pathogenesis of radiation fibrosis.

It will be of interest to see how the IL-4 gene is induced following lung irradiation and to look at INF as frequent counterpart of IL-4 in order to get more information about the cytokine network controlling inflammation and fibrosis in the lung.