Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Lung transplantation is a therapeutic modality for the treatment of many end stage pulmonary diseases. However, rejection occurs more often in the lung than in other solid organ allografts, and is the leading cause of death in lung allograft recipients (1, 2). Rejection is initiated by the recognition of allogeneic (donor) major histocompatibility complex (MHC) molecules by host T lymphocytes, leading to upregulated cellular and humoral immunity (3). Immunosuppressive agents often fail to prevent continued rejection episodes, and therefore, the ultimate goal of inducing indefinite acceptance of the allograft, known as immunologic tolerance, remains elusive. Because allogeneic MHC molecules are the stimulus and target of the immune response during rejection, MHC-derived peptides or synthetic peptides that may be homologous to MHC antigens have been the focus of investigations attempting to induce immunologic tolerance to allografts (4, 5).

Because recognition of polymorphic regions of donor MHC molecules is usually the stimulus for allo-immune responses, immunologic tolerance induced by peptides derived from the donor MHC is often specific to the allele of the donor MHC molecules. Therefore, identification of proteins/peptides that are highly conserved among individuals and induce immunologic tolerance across multiple MHC alleles may be of great benefit for the allograft recipient. However, the use of such proteins/peptides for induction of immunologic tolerance to lung allografts has not been evaluated.

In our murine model that reproduces the histology and immunology of acute lung allograft rejection, the perivascular and peribronchiolar connective tissues are the sites of antibody deposition (6, 7). These same tissues are the sites of rejection activity in human lung allograft recipients (1). Collagen type V (col[V]) is present in the lung (8) and is located in the peribronchiolar connective tissues (9), alveolar interstitium (10), and capillary basement membranes (9). The 1 chain of type V collagen (1[V]) is nearly 76% homologous to the 2 chain of type XI collagen (2[XI]) (11). The gene for 2(XI) is located in the MHC class II loci of mice and humans (12), and shares amino acid sequences with MHC class II (13). Because col(V) may have MHC-like sequences and MHC-derived peptides have been used to induce tolerance in allografts other than the lung, the current study determined the ability of col(V) to modulate the immune response to lung alloantigens in vivo.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mice

Six- to 8-wk-old female C57BL/6 (I-a^b, H-2^b) and BALB/c (I-a^d, H-2^d) mice were obtained from Harlan Sprague Dawley Inc. (Indianapolis, IN) and housed in micro isolator cages in the Laboratory Animal Resource Center at the Indiana University School of Medicine in accordance with institutional guidelines as previously reported (6, 7, 14).

Collection and Phenotype of Donor Mice Bronchoalveolar Lavage Cells

Donor bronchoalveolar lavage (BAL) cells were obtained by BAL (6, 7, 14). In brief, anesthesia was induced in C57BL/6 and BALB/c mice by an intramuscular injection of a mixture of ketamine (80 to 100 mg/kg), acepromazine (8 to 10 mg/kg), and atropine (0.5 mg/kg). After isolation of the trachea by dissection and opening the thoracic cavity by midline incision, an 18-gauge Teflon catheter was inserted into the trachea and secured by suture. The lungs were lavaged with a total of 20 ml of phosphate-buffered saline (PBS) at 37°C, and cells were isolated from lavaged specimens by centrifugation. BAL cells were resuspended in PBS at a concentration of 1 × 10⁶/ml. Immunocytochemical examination of cytospin preparations showed that macrophages and dendritic cells make up 96 and 2% of BAL cells, respectively (6, 7, 14).

Distribution of Instilled BAL Cells in Lungs of Recipient Mice

Similar to prior reports, two different methods were used to determine if the cells that were instilled by nasal insufflation reached the alveolar space (6, 7, 14). In brief, colloidal carbon (100 µl of a 5% saline solution; Pelikan, Hanover, Germany) was instilled by nasal insufflation into the lower respiratory tract of C57BL/6 mice. After a 2-h incubation, the recipient mice underwent BAL and the number of carbon-loaded BAL cells was detected by examination of cytospin preparations using light microscopy. These carbon-loaded BAL cells (1.5 × 10⁵/mouse) were then instilled into the airway of anesthetized BALB/c mice by nasal insufflation. After 2 h, the recipient mice were killed, and the thoracic organs were harvested en bloc, fixed by inflation and immersion in 6% glutaraldehyde, sectioned, stained with eosin, and examined using light microscopy for the presence of carbon-loaded BAL cells in the alveolar spaces.

Alternatively, donor DC (C57BL/6, I-a^b) were identified in the alveoli of BALB/c mice by immunohistochemistry. In brief, 4 h after nasal insufflation of the lung cells of C57BL/6 mice, the recipient lungs (BALB/c) were prepared for immunohistochemistry by intratracheal instillation of optimal cutting temperature (OCT) compound (Tissue-Tek, Elkhart, IN) diluted 1:1 with PBS. After freezing in liquid nitrogen, the lungs were stored at 80°C. Localization of donor cells (I-a^b+) was performed on cryostat lung sections using biotinylated mouse antimouse I-a^b antibodies (PharMingen). Antibody detection was performed using streptavidin alkaline phosphatase and ABTS substrate following the manufacturer's directions (Kirkegaard and Perry). These studies also confirmed that donor cells (C57BL/6) reached the alveoli of recipient mice (BALB/c).

Preparation of Collagens

Collagen type II (col[II]) was isolated from canine cartilage as previously reported (15) or was purchased from Collaborative Biomedical Products (Bedford, MA). Both preparations were solubilized in 0.5 M acetic acid, then dialyzed to yield a final concentration of 0.5 mg/ml in PBS.

Bovine collagen type XI (col[XI]) from fetal calf cartilage (16) was the generous gift of Nicolas P. Morris, Ph.D. (Shriners Hospital for Crippled Children, Portland, OR), or was purchased from Biogenesis (Sandown, NH). Both col(XI) preparations were solubilized in 50 µM Tris, 0.2 M NaCl, pH 7.5.

Human col(V), extracted from human placenta and purified by differential NaCl precipitation (17), was a generous gift from Jerome Seyer, Ph.D. (Veterans Administration Hospital, Hampton, VA). In brief, placental tissues were minced, washed, and suspended in 0.5 M acetic acid containing 0.2 M NaCl, and digested by pepsin at 4°C. Supernatants were aspirated from centrifuged specimens, the pellet collected, and the extraction procedure repeated. The supernatants were combined from the two digests, and col(V) was purified from the supernatants by differential NaCl precipitation from 0.5 M acetic acid (15, 17). Col(V) was soluble in 0.7 M NaCl and precipitated in 1.2 M NaCl. In separate experiments requiring purified (V) chains, the cycle of solubilization in acetic acid and NaCl precipitation was repeated until a type V preparation with an -chain ratio, 1(V)/2(V), of approximately two was obtained as determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (15). Separation of 1(V) from 2(V) was achieved by chromatography on diethylaminoethyl cellulose (17). The 1(V) and 2(V) chains were eluted from the column, and purity was confirmed by SDS-PAGE as previously reported (15). Intact col(V) or 1(V) and 2(V) chains were diluted in PBS (0.5 mg/ml) until use. Col(V) was also purchased from Collaborative Biomedical Products.

The quantity of collagen types II, XI, V, 1(V), and 2(V) were assessed by determination of the hydroxyproline content in the samples as previously reported (18, 19). Experimental outcomes were not affected by the source of col(II), col(XI), or col(V).

Murine Treatment Groups

Anesthetized BALB/c mice received either 1.5 × 10⁵ C57BL/6 (allogeneic) or BALB/c (autologous) BAL cells in 100 µl of PBS by nasal insufflation weekly for 4 wk (6, 7, 14). Our previous study (6) and experiments performed in the current study demonstrated that nasal insufflation of 100 µl of PBS or autologous (1.5 × 10⁵) BAL cells weekly for 4 wk had no effect on histology, BAL differential cell counts, or cytokine levels in the lungs of recipient mice.

In other experiments, BALB/c mice received col(II), col(XI), col(V), or purified 1(V) or 2(V) chains by intratracheal instillation weekly for 4 wk followed by four weekly instillations of allogeneic (C57BL/6) BAL cells by nasal insufflation weekly for 4 wk. In brief, BALB/c mice were anesthetized, the ventral surface of the neck shaved, and the trachea isolated by blunt dissection and cannulated with a 22-gauge Teflon catheter. Each mouse received 50 µg of col(II), col(V), or purified 1(V) or 2(V) chains in 100 µl of PBS; or 50 µg of col(XI) in 100 µl of diluent (0.01 M acetic acid, 0.2 M NaCl, pH 7.5) intratracheally, weekly for 4 wk. Intratracheal instillations were performed to ensure introduction of the collagens into the lower respiratory tract. Preliminary studies confirmed that repeated instillations of collagen (50 µg) did not induce pathologic lesions or alterations in differential cell counts in recipient lungs. At the completion of the four weekly instillations of collagen, BALB/c mice received 1.5 × 10⁵ C57BL/6 BAL cells in 100 µl of PBS by nasal insufflation weekly for 4 wk.

In separate experiments, BALB/c mice received four weekly instillations of allogeneic (C57BL/6) BAL cells followed by a 4-wk recovery period. At the end of this 8-wk period, BALB/c mice received 1.5 × 10⁵ autologous BAL cells (BALB/c) that had been pulsed with col(II), col(V), or col(XI) into the lung by nasal insufflation weekly for 4 wk. In brief, BALB/c BAL cells (1.5 × 10⁵/ml) were incubated with collagen types II, V, or XI (100 µg/ml) in complete media (RPMI 1640 supplemented with 10% heat- inactivated fetal calf serum in RPMI 1640 with 25 mM N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid, 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml streptomycin [all from GIBCO BRL, Grand Island, NY]) for 24 h at 37°C, 5%. After washing, cells were resuspended in PBS, and 1.5 × 10⁵ cells were instilled into the lung by nasal insufflation. Viability exceeded 90% for all donor cells.

Collection of Serum and BAL at Completion of Experimental Period

At the end of the study period for each group, mice were anesthetized with ketamine, acepromazine, and atropine. The thoracic and abdominal cavities were opened and mice exsanguinated by cardiac and inferior vena cava puncture. Serum was collected from centrifuged specimens. After the trachea was dissected and transected, an 18-gauge catheter was inserted into the trachea and the lungs were lavaged with a total of 2.5 ml of sterile PBS. In brief, a 0.5- to 1.0-ml aliquot of PBS was instilled into the trachea and aspirated five times before placement into a specimen container. Cell-free BAL supernatants were obtained by centrifugation of BAL fluids. All serum and BAL supernatants were stored at 80°C until use. Differential cell counts were performed on cytospin preparations of BAL cells.

Lung Histology

The thoracic organs of recipient mice were removed en bloc after BAL and were fixed by an intratracheal instillation of 6% glutaraldehyde. Preliminary studies demonstrated that the greatest deposition of donor cells occurred in the midlung zones (perihilar distribution). Therefore, four to six cryostat sections were obtained from the perihilar regions of recipient lungs, stained with hematoxylin and eosin, examined under light microscopy, and graded according to the histologic criteria established by the Lung Rejection Study Group (20) and as previously reported (6, 7, 14). In addition, all grading of histologic sections was done in blinded fashion by one of the authors (O.W.C.) without prior knowledge of the treatment group (6, 7, 14).

Detection of Apoptosis

Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL) assay kits (In Situ Cell Death Detection Kit; Boehringer Mannheim, Indianapolis, IN) were used to detect apoptosis in lung tissue sections. In brief, xylene was used to remove wax from paraffin-embedded tissues, and serial incubations with decreasing ethanol solutions were used to rehydrate tissue sections. After placement in Sequenza chambers (Shandon Lipshaw, Pittsburgh, PA) and immersion in Tris-buffered saline (0.05 M Tris/HCl, pH 7.5-8.0 in 0.15 M saline), slides were treated with proteinase K solution (20 µg/ml in 10 mM Tris/HCl, pH 7.4-8.0) for 30 min at 37°C. Tissue sections treated with deoxyribonuclease I (1 mg/ml; Boehringer Mannheim) for 30 min at room temperature were used for positive controls for the TUNEL assay. The labeling and alkaline phosphatase conversion steps were performed per protocol supplied by the manufacturer. Substrate solution (BrdU Labeling and Detection Kit; Boehringer Mannheim) was used to develop reaction products per manufacturer's directions. Slides were cover-slipped and examined by light microscopy.

Isolation of Lung Lymphocytes

Because too few BAL T lymphocytes were available for these studies, lymphocytes were isolated from the lung parenchyma of normal or recipient BALB/c mice (14). In brief, mice were injected with 150 µl of heparin (1,000 U/ml; Upjohn, Kalamazoo, MI) and anesthetized using intramuscular injection of a mixture of ketamine (80 to 100 mg/kg), acepromazine (8 to 10 mg/kg), and atropine (0.5 mg/kg). After opening the chest and abdominal cavity by midline incision and exsanguination by transection of the inferior vena cava, the trachea was isolated and BAL was performed using five 1-ml aliquots of sterile PBS (37°C). Once the pulmonary vasculature was perfused with 10 ml of PBS (37°C) to remove peripheral blood, lungs were removed avoiding all lymph node tissue, placed in a petri dish, and minced into 1-mm cubes. The lung tissue was further digested by stirring in a flask containing collagenase/DNAase solution for 90 min at 37°C. (Collagenase/DNAase solution, 52.5 mg collagenase [Boehringer Mannheim], was added to medium, 10 ml RPMI [GIBCO BRL] and 10% heat-inactivated fetal calf serum [Hyclone, Logan, UT]. A total of 1 ml of DNAase [Boehringer Mannheim] stock solution [3 mg/ml] was added to each 5 ml of collagenase solution.) To remove particulate matter from the collagenase preparation and to obtain individual lung T lymphocytes, the cells were centrifuged over a Percoll gradient (Pharmacia, Piscataway, NJ), incubated on a nylon wool column for 1 h at 37°C, and eluted from the column using complete medium. Any remaining red blood cells were removed by lysis using ammonium chloride. Immunocytochemical analysis of cytospin preparations confirmed that the cells obtained were T lymphocytes (> 90% Thy-1+).

Mixed Leukocyte Reaction

The ability of mitomycin C-treated C57BL/6 splenocytes to induce proliferation in T lymphocytes isolated from the lungs of BALB/c mice that received weekly instillations of col(V) followed by instillations of C57BL/6 BAL cells or T lymphocytes from normal BALB/c mice was determined by a mixed leukocyte reaction (14). In brief, C57BL/6 splenocytes, which were used as a source of antigen-presenting cells (APCs), were treated with mitomycin C (Sigma, St. Louis, MO) and cocultured in varying ratios with lung T lymphocytes (3 × 10⁵/well) in 200 µl of medium (RPMI, 2 mM L-glutamine, 5 × 10⁵ M 2-mercaptoethanol, 100 U/ml penicillin, 100 µl/ml streptomycin, 10% heat-inactivated fetal calf serum) in flat-bottom, 96-well microtiter plates (Costar, Cambridge, MA). Eighteen hours before the completion of a 3-d incubation at 37°C (5% CO₂), 0.5 µCi/ml of ³H (Amersham Corp., Arlington Heights, IL) was added to each well. Cultures were harvested with an automated cell harvester (Brandel, Gaithersburg, MD) and analyzed in a liquid scintillation counter (Beckman, Arlington Heights, IL). Cellular proliferation was determined as the mean of counts per minute of [³H]thymidine incorporation in triplicate cultures and reported as a stimulation index as described in RESULTS.

Cytokine Enzyme-Linked Immunosorbent Assay

Tumor necrosis factor (TNF)- was measured in unconcentrated BAL fluid from recipient mice using commercial enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN) per manufacturer's protocol. The sensitivity of this assay was 23.4 pg/ml. Interleukin (IL)-4 and IL-10 were measured by ELISA in unconcentrated BAL fluid of recipient mice as previously reported (14).

Statistics

The data were initially assessed to confirm normality using a Shapiro-Wilk statistic. Comparisons between groups were analyzed for each of the dependent variables using a one-way analysis of variance with four interventions. For those comparisons demonstrating significance, a post hoc Student-Newman-Keuls was performed to determine differences between interventions. Where multiple comparisons were performed, a Bonferonni correction was applied and the significance level was 0.02. For other comparisons, the level of significance was 0.05.

Differences in TNF- levels between groups were determined using Mann-Whitney U test for unpaired data. Values of P < 0.05 were considered to be significant.

Data regarding the presence or absence of pathologic lesions in the lungs of mice that received allogeneic cells alone or collagen-pulsed autologous BAL cells were determined by using a log-likelihood G statistic. Values of P < 0.05 were considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Initial experiments determined if intrapulmonary instillation of col(V) before instillation of allogeneic cells would prevent the development of histology analogous to the rejection response in recipient lungs. BALB/c mice received 100 µl of PBS, 50 µg of purified 1(V) chains or 2(V) chains, or intact col(II), col(V), or col(XI) in 100 µl of diluent (see MATERIALS AND METHODS) by intratracheal instillation weekly for 4 wk. This was followed by four weekly instillations of 1.5 × 10⁵ C57BL/6 BAL cells in 100 µl of PBS into the lungs by nasal insufflation. Col(II), a major component of articular cartilage (15), is not present in the lung and, therefore, served as a control for these studies. Col(XI), a minor component within cartilage (16), has not been demonstrated within lung parenchyma. At the completion of the experimental period, BAL was performed, and the lungs of recipient mice were harvested, fixed, and examined for rejection pathology. Figure 1 shows mononuclear cell infiltrates in the peribronchiolar and perivascular tissues analogous to grades 1 and 2 acute rejection in the lungs of mice that received instillations of allogeneic cells alone (Figure 1A); or col(II) or col(XI) before instillations of allogeneic cells (Figures 1B and 1C, respectively). In contrast, Figure 2 shows that intrapulmonary instillations of purified 1(V) or 2(V) chains before allogeneic BAL cells only induced perivascular edema without perivascular or peribronchiolar mononuclear cell infiltrates (Figures 2A and 2B, respectively). Similar data was observed in experiments using intact col(V) (data not shown). Preliminary experiments showed that four weekly instillations of collagens, diluent for collagens, or PBS alone did not induce any alterations in the histology or differential cell counts in recipient lungs. Consistent with limited mononuclear cell infiltrates in lungs of mice that received col(V) before allogeneic cells, Figure 3 shows significantly fewer macrophages and lymphocytes in BAL fluid of these same mice compared with mice that received allogeneic cells alone, or col(II) or col(XI) before allogeneic cells (P < 0.002 and P < 0.0005, for macrophages and lymphocytes, respectively). Instillation of purified 1(V) or 2(V) chains before allogeneic cells resulted in differential cell counts similar to that observed using intact col(V). BAL differential cell counts for macrophages and lymphocytes were comparable in mice that received allogeneic cells alone, or col(II) or col(XI) before allogeneic cells (Figure 3).

View larger version (150K): [in this window] [in a new window]   View larger version (157K): [in this window] [in a new window]   View larger version (145K): [in this window] [in a new window]   Figure 1.   Lung histology in BALB/c mice after four weekly instillations of 1.5 × 105 allogeneic (C57BL/6) BAL cells alone, col(II), or col(XI) (50 µg each) followed by four weekly instillations of C57BL/6 BAL cells. (A) Peribronchiolar and perivascular mononuclear cell infiltrates are shown in the lungs of BALB/c mice that received instillations of BAL cells from C57BL/6 mice. Similar pathologic lesions were observed in the lungs of BALB/c mice that received weekly instillations of col(II) (B) or col(XI) (C) before instillations of C57BL/6 cells. Histology is representative of five to eight mice in each group (original magnification: ×200).

View larger version (156K): [in this window] [in a new window]   View larger version (157K): [in this window] [in a new window]   Figure 2.   Lung histology in BALB/c mice after four weekly instillations of purified 1(V) or 2(V) chains (50 µg each) followed by four weekly instillations of 1.5 × 105 allogeneic (C57BL/6) BAL cells alone. (A) The presence of perivascular edema, but the absence of perivascular or peribronchiolar mononuclear cell infiltrates, is shown in mice that received instillations of 1(V) before C57BL/6 BAL cells. (B) Similar histology in the lungs of BALB/c mice that received instillations of 2(V) before instillations of C57BL/6 BAL cells is shown. Histology is representative of five to eight mice in each group (original magnification: ×200).

View larger version (21K): [in this window] [in a new window]   Figure 3.   Differential cell counts in BAL fluid of recipient mice. BALB/c mice received 1.5 × 105 allogeneic (C57BL/6) BAL cells alone, or col(II), col(XI), or col(V) (50 µg each) weekly for 4 wk followed by four weekly instillations of C57BL/6 BAL cells. At the completion of the experimental period, recipient mice under BAL and differential cell counts were performed by counting 300 cells per high powered field on cytospin preparations. Data represent the mean ± standard deviation of four to six recipient mice in each group (*P < 0.002 for macrophages and P < 0.0005 for lymphocytes).

Previous reports have shown that prevention of allograft rejection by pretransplant immunization with peptides that may be similar to donor MHC molecules is in part due to impaired proliferative responses of host T lymphocytes to donor alloantigens (4, 5). Therefore, we next determined if instillation of col(V) into the lung before instillation of allogeneic BAL cells prevented T-lymphocyte proliferation in response to donor alloantigen. Because instillation of 1(V), 2(V), or intact col(V) before instillations of allogeneic cells all induced comparable pathology and BAL differential cell counts in the lung, all subsequent studies involving col(V) used intact col(V) proteins and not individual  chains. BALB/c mice received intrapulmonary instillations of col(V) (50 µg) for 4 wk followed by four weekly instillations of C57BL/6 BAL cells. At the completion of the study period, the ability of C57BL/6 splenocytes to induce proliferation in lung T lymphocytes isolated from recipient or normal BALB/c mice was determined by [³H]thymidine incorporation. Figure 4 shows that donor (C57BL/6) splenocytes induced dose-dependent proliferation in T lymphocytes isolated from the lungs of normal BALB/c mice. In contrast, T lymphocytes isolated from the lungs of BALB/c mice pretreated with col(V) before instillation of allogeneic BAL cells did not proliferate in response to donor alloantigen, and at a stimulator/responder ratio of 1:1, proliferation was inhibited (Figure 4).

View larger version (16K): [in this window] [in a new window]   Figure 4.   Mixed leukocyte reaction. Variable quantities of mitomycin C-treated C57BL/6 splenocytes (stimulators) were incubated with 3 × 105 T lymphocytes (responders) isolated from the lungs of normal BALB/c mice (Normal) or BALB/c mice that received four weekly instillations of col(V) (50 µg) followed by four weekly instillations of C57BL/6 BAL cells. Eighteen hours before the completion of a 72-h incubation, the cells were pulsed with 3H and proliferation determined by counts/minute (cpm) of thymidine incorporation. Stimulation index equals the multiples of proliferation in lung lymphocytes induced by varying quantity of stimulator cells relative to proliferation of lung lymphocytes alone. Data are representative of three experiments.

Upregulated TNF- production has been shown to play a key role in the pathogenesis of lung allograft rejection in vivo and alloimmune responses in vitro (21, 22). Therefore, we next determined if instillations of col(V) into the lung before instillations of allogeneic BAL cells downregulate local production of TNF-. Table 1 shows that instillation of allogeneic BAL cells induces the vigorous production of TNF- locally in recipient lungs. In contrast to instillations of allogeneic BAL cells alone, instillations of col(V) before allogeneic BAL cells induced the local production of TNF- in only one of five recipient mice (P < 0.016). TNF- was not detected in the BAL fluid of normal mice or mice that received instillates of autologous cells.

Using TUNEL assays that detect DNA strand scission, a commonly used method to identify apoptotic cells, we next determined if apoptosis contributed to the airway pathology and vasculopathy induced by the instillation of allogeneic BAL cells into recipient lungs. Figure 5 shows that instillation of C57BL/6 BAL cells into the lungs of BALB/c mice weekly for 4 wk induced apoptosis in airway epithelium and vascular endothelium in recipient BALB/c mouse lungs. There were no apoptotic cells present in the lungs of normal mice (data not shown). To determine if col(V) downregulated rejection pathology by preventing apoptosis in response to allogeneic BAL cells, BALB/c mice received four weekly instillations of col(V) (50 µg) into the lung followed by four weekly instillations of C57BL/6 BAL cells. At the completion of the experimental period, apoptosis was detected in tissue sections of recipient lungs by TUNEL assay. In contrast to mice that received allogeneic BAL cells, Figure 6 shows that apoptotic cells were not detected in the lungs of mice that received instillations of col(V) before instillations of allogeneic cells.

View larger version (159K): [in this window] [in a new window]   Figure 5.   Instillation of allogeneic BAL cells induces apoptosis in recipient lungs. BALB/c mice received instillations of 1.5 × 105 allogeneic (C57BL/6) BAL cells into the lung weekly for 4 wk. At the completion of the experimental period, TUNEL assays were used (see MATERIALS AND METHODS) to detect apoptotic cells in lung tissue sections. Apoptotic cells indicated by dark blue nuclei (arrows) are present in airway epithelium and vascular endothelium. Tissue section is representative of 10 recipient mice in this group (original magnification: ×200).

View larger version (165K): [in this window] [in a new window]   Figure 6.   Detection of apoptotic cells in the lungs of mice that received col(V) before allogeneic BAL cells. BALB/c mice received 50 µg of col(V) into the lung weekly for 4 wk followed by four weekly instillations of 1.5 × 105 allogeneic (C57BL/6) BAL cells. At the completion of the experimental period, TUNEL assays were used (see MATERIALS AND METHODS) to detect apoptotic cells in lung tissue sections. In contrast to Figure 5, in Figure 6 no apoptotic cells were detectable in the lungs of mice that received instillations of col(V) before instillations of allogeneic BAL cells. Tissue sections are representative of eight recipient mice (original magnification: ×200).

We next determined if repeated instillations of col(II), col(XI), or col(V) in the lung would perpetuate the perivascular and peribronchiolar mononuclear cell infiltrates induced initially by instillations of allogeneic BAL cells. Preliminary studies showed that the vasculopathy and bronchitis induced by allogeneic BAL cells resolves completely within 5 wk after the last instillation of cells. Therefore, BALB/c mice received instillations of 1.5 × 10⁵ C57BL/6 BAL cells into the lung weekly for 4 wk followed by a 4-wk recovery period. At the completion of this 8-wk period, recipient mice did not receive any further interventions or received four weekly intrapulmonary instillations of autologous BAL cells (BALB/c, 1.5 × 10⁵) that had been pulsed with 100 µg of either col(II), col(XI), or col(V) as described in MATERIALS AND METHODS. At the completion of this 12-wk experimental period, the lungs were harvested and examined for the development of rejection pathology. Figure 7 shows normal histology in the lungs of mice that received allogeneic cells only in the initial 4-wk period (Figure 7A) and in the lungs of mice that received allogeneic cells followed by autologous BAL cells pulsed with col(II) (Figure 7B) or col(XI) (Figure 7C). In contrast, Figure 7D shows that instillation of allogeneic BAL cells followed by autologous BAL cells pulsed with col(V) induced perivascular and peribronchiolar infiltrates analogous to grades 1 and 2 acute rejection in recipient lungs. Table 2 shows that three of five mice that received col(V)-pulsed autologous BAL cells developed pathologic lesions in the lung compared with the absence of lesions in the lungs of mice that received col(II)- or col(XI)-pulsed autologous cells (P < 0.048).

View larger version (155K): [in this window] [in a new window]   View larger version (155K): [in this window] [in a new window]   View larger version (153K): [in this window] [in a new window]   View larger version (157K): [in this window] [in a new window]   Figure 7.   Effect of instillation of collagen-pulsed autologous BAL cells into the lungs of mice primed with allogeneic BAL cells. BALB/c mice received four weekly instillations of 1.5 × 105 allogeneic (C57BL/6) BAL cells into the lung followed by a 4-wk recovery period. At the completion of this 8-wk period, mice received no further instillations or instillations of 1.5 × 105 autologous (BALB/c) BAL cells that had been pulsed with either col(II), col(XI), or col(V) into the lung weekly for 4 wk. Histology is shown for the different treatment groups at the completion of the 12-wk experimental period. (A) Normal histology in the lungs of mice that received allogeneic cells alone for the first 4 wk is shown. (B) Normal histology in lungs of mice that received allogeneic BAL cells, followed by instillations of col(II)-pulsed autologous BAL cells and (C) similar findings in the lungs of mice that received allogeneic BAL cells followed by instillations of col(XI)-pulsed autologous BAL cells are shown. (D) Instillations of allogeneic BAL cells followed by instillations of col(V)- pulsed autologous BAL cells induced perivascular and peribronchiolar mononuclear cell infiltrates in recipient lungs. Tissue sections are representative of five mice in each group (original magnification: ×200).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Using our murine model that reproduces the histology and immunology of acute lung allograft rejection (6, 7, 14), data in the current study show that prior instillation of col(V), but not col(II) or col(XI), prevents development of rejection pathology in response to allogeneic BAL cells, downregulates proliferative responses of lung T lymphocytes to donor alloantigens, inhibits development of apoptosis in the recipient lung, and abrogates TNF- production locally. Finally, instillation of autologous BAL cells pulsed with col(V), but not col(II) or col(XI), perpetuates the rejection pathology induced by prior instillations of allogeneic BAL cells.

Col(V) is a minor collagen in the lung and is located in the perivascular and peribronchiolar connective tissues (8), which are the same sites of rejection activity. Our prior studies showed the selective deposition of immunoglobulin G2a antibodies in these same tissues of mice that have received weekly instillations of allogeneic BAL cells (6, 7). Humoral responses during allograft rejection are directed against donor MHC antigens. However, because col(V) may be MHC-like, data in the current study suggest that molecular mimicry (4) between MHC molecules and non-MHC proteins, such as col(V), may be active in the pathogenesis of lung allograft rejection.

The ability to induce tolerance to donor alloantigens is crucial to the prevention of allograft rejection (4, 5, 23). Tolerance may result from induction of anergy in T lymphocytes to alloantigens or clonal deletion of alloantigen-specific T lymphocytes (4, 5). Because col(V) may be MHC-like, then data in the current study showing that instillations of col(V) prevented proliferative responses to alloantigens and prevented development of pathology analogous to acute rejection in recipient lungs suggest that col(V) may induce anergy to donor alloantigens. Alternatively, the lack of proliferative responses to donor antigens may be due to clonal deletion of alloantigen-specific lung lymphocytes or suppressor cell activity. However, a role for col(V)-induced clonal deletion of alloantigen-specific lymphocytes or stimulation of suppressor cell activity as mechanisms of tolerance induction in this model is unknown and the subject of ongoing investigations. Significantly, these data are similar to reports from investigators studying different animal models who showed that tolerance to alloantigens and prevention of allograft rejection induced by injections of MHC- derived peptides was in part due to impaired proliferative responses of host lymphocytes to donor antigens (4, 5).

Previous reports have shown that the induction of anergy or clonal deletion may vary relative to the use of peptides derived from MHC class I or class II molecules (4, 5). For example, proposed mechanisms for tolerance induced by MHC class I-derived peptides include binding to heat shock proteins (24), modulation of heme oxygenase activity (25), or possible inhibition of natural killer cell activity (4). MHC class II-derived peptides have been reported to induce tolerance by competition for target antigens during indirect allorecognition (4), blockade of cell-cycle progression (4), and inhibition of CD4 receptor function (4). More recently, Murphy and coworkers (26) reported that peptides from nonpolymorphic regions of HLA-DQA1 inhibited alloimmune responses by possible blockade of MHC class II molecules, which induced apoptosis in responding lymphocytes. In the current study, the possible similarity of peptides derived from col(V) to MHC class II could suggest blockade of MHC class II and T-cell receptor interactions. However, data showing col(V) perpetuates the rejection response in lungs primed by alloantigen makes this mechanism less likely. Although col(V) may have similarities to MHC class II, previous reports have shown that mismatches at either MHC class I or II loci between donor and recipient may be sufficient to induce the rejection response (3). Because donor (C57BL/6, H-2^b, I-a^b) and recipient (BALB/c, H-2^d, I-a^d) mice used in the current study are fully mismatched at MHC class I and II loci, and either donor MHC class I or II are able to induce the rejection response in recipient lungs (unpublished observations, manuscript in preparation), then it is unlikely that blockade of MHC class II by col(V), alone, is the mechanism by which col(V) downregulates alloimmune responses in vivo.

Several cytokines have been implicated in the pathogenesis of lung allograft rejection (1). Of these, only blockade of TNF- has been shown to downregulate lung allograft rejection (22). The mechanism of TNF--induced allograft destruction likely includes enhanced alloimmune responses (1, 22) and induction of apoptosis in the allograft (26). In the current study, instillation of allogeneic BAL cells induced the vigorous production of TNF- and apoptosis in recipient lungs. Therefore, data in the current study showing that instillation of col(V) prevented pathologic lesions and apoptosis in recipient lungs may have been due to col(V)-induced downregulation of local TNF- production. However, the specific mechanism of TNF- in stimulating lung allograft rejection remains to be determined.

In contrast to the local production of TNF-, the production of IL-4 and/or IL-10 has been associated with induction of tolerance to antigens (23). However, neither IL-4 nor IL-10 was detected by ELISA in BAL fluid from the lungs of mice that received instillations of col(V) before instillations of allogeneic BAL cells (data not shown). Transforming growth factor (TGF)- production has also been associated with tolerance induction in several experimental systems (23). Although TGF- production was not examined in the current study, ongoing investigations are attempting to determine if either local or systemic production of TGF- has a role in tolerance induction during lung allograft rejection.

Data in the current study showing that rejection pathology is perpetuated by instilling col(V)-pulsed autologous APCs into the lungs of mice primed previously with instillations of allogeneic BAL cells suggest that indirect allorecognition has a role in the pathogenesis of lung allograft rejection. These data are similar to a very recent report by SivaSai and colleagues (27) who showed that peripheral blood mononuclear cells from lung allograft recipients undergoing chronic rejection proliferated in response to donor-derived MHC class I peptides presented by host antigen-presenting cells. In addition, these data and those in the current study indicate that indirect allorecognition may have a key role in the prevention and induction of lung allograft rejection.

Different techniques have been used to induce tolerance to solid organ allografts, such as donor-specific blood transfusion, thymic injection with donor-derived APCs, or systemic immunization with peptides derived from donor MHC molecules before transplantation (4). For each of these techniques to be effective, the specific donor MHC molecules must be known several weeks before transplantation to allow sufficient time, i.e., weeks to months, for tolerance induction to occur. However, only a few hours exist between identification of a potential donor and the transplantation surgery, and therefore, time is not available for tolerance induction by these techniques. A very recent report demonstrated that a nonpolymorphic synthetic peptide derived from MHC molecules induced tolerance to multiple allogeneic MHC molecules in vitro (26). The current study showed that instillation of col(V) before allogeneic BAL cells downregulates T-lymphocyte responses to alloantigen in the lung in vivo. These data raise the possibility that prior treatment of prospective transplant recipients with MHC-like peptides that are not derived specifically from the MHC molecule of the donor may be a potential therapy to prevent rejection in lung allograft recipients despite incompatibilities between MHCs of the donor and recipient.