Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Lung transplantation is a modality commonly used for the treatment of end-stage pulmonary diseases. However, allograft rejection remains a major cause of morbidity and mortality in lung transplant patients (1). Although the introduction of potent immunosuppressive agents has improved the early graft survival, repeated acute rejection results in chronic rejection, which is the leading cause of death in lung allograft recipients (1). During graft rejection, allogeneic (donor) major histocompatibility complex (MHC) molecules are usually the stimulus and the target of the immune response (2). However, in allografts other than the lung, non-MHC antigens may also be immune targets during rejection episodes (3).

Immunologic tolerance is defined as immune unresponsiveness to an antigen implicated in causing the disease (4). Although different routes can be used for administering the antigens for the induction of immunologic tolerance, oral tolerance refers to oral administration of antigens, which has resulted in suppression of disease activity in various models of autoimmune diseases, including experimental allergic encephalomyelitis, uveitis, diabetes, myasthenia gravis, and arthritis (reviewed in Ref. 4).

In rodent models of cardiac and corneal transplantation, immunization with MHC-derived peptides or other proteins/peptides that are targets of the rejection response have been used to induce immunologic tolerance and prevent allograft rejection (5). In particular, oral administration of alloantigens has been shown to downregulate the systemic cell-mediated immune response against histocompatibility antigens and to induce donor-specific immunosuppression in transplantation of organs other than the lung (8, 9). In addition to the prevention of rejection activities, immune suppression induced by oral tolerance has also been quantitated by downregulation of delayed-type hypersensitivity (DTH) responses to target antigens (4, 10).

We have recently reported that type V collagen [col(V)] is a target of the alloimmune response in a murine model that reproduces the histology and immunology of lung allograft rejection (11). In the current study, we examine the utility of col(V) to induce oral tolerance and prevent lung allograft rejection. The data show that oral administration of col(V) to lung allograft recipients before transplantation downregulates acute lung allograft rejection. Diminished DTH responses to donor alloantigens and intact responses to nominal antigens in col(V)-fed recipients show that feeding col(V) prevented allograft rejection by inducing immunologic tolerance to donor antigens. Feeding col(V) induced stimulation of systemic production of transforming growth factor (TGF)-, but not interleukin (IL)-4 or IL-10. Neutralizing TGF- recovered the DTH response to donor antigen in col(V)-fed allograft recipients.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Pathogen-free, MHC (RT1)-incompatible male rats were used for the study: Wistar Kyoto (WKY, RT1^l), Fischer 344 (F344, RT1^lv1), and Brown Norway (BN, RT1ⁿ) rats (250 to 300 g at the time of transplantation). All rats were purchased from Harlan Sprague Dawley (Indianapolis, IN) and housed in the Laboratory Animal Resource Center at Indiana University School of Medicine (Indianapolis, IN) in accordance with institutional guidelines.

Preparation of Collagens

Collagen type II [col(II)] was isolated from canine cartilage as previously reported (11, 12), or purchased from Collaborative Biomedical Products (Bedford, MA). Both preparations were solubilized in 0.005 M acetic acid and dialyzed to yield a final concentration of 0.5 mg/ml.

Bovine collagen type XI [col(XI)] from fetal calf cartilage (13) was purchased from Biogenesis (Sandown, NH), diluted in 0.005 M acetic acid (0.5 mg/ml), and stored at 4°C until use.

Human col(V), extracted from human placenta and purified by differential NaCl precipitation (11), was a gift from Dr. Jerome Seyer (VA Hospital, Hampton, VA). In brief, placental tissues were minced, washed, suspended in 0.5 M acetic acid containing 0.5 M NaCl, and digested by pepsin at 4°C. Supernatants were aspirated from centrifuged specimens, the pellet was collected and the extraction procedure was 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 (11, 12). The intact col(V) was diluted in 0.005 M acetic acid (0.5 mg/ml) until use.

The quantities of collagens were assessed by determination of the hydroxyproline content in the samples as previously reported (11).

Oral Administration of Collagen

WKY rats were fed with 10 µg of col(II), col(V), or col(XI) dissolved in 0.5 ml of saline by a gastric gavage used a 16-gauge, ball-point, stainless-steel animal feeding needle (Braintree Scientific, Braintree, MA). Control animals were fed with diluent alone. Animals were fed every other day for eight feedings. This dose of collagen was chosen because of its effectiveness in oral tolerance induction in nontransplantation studies in rats (14). At 7 d after the last feeding, these rats were used as recipients of lung allografts (described later).

Transplantation Model

The orthotopic transplantation of left-lung isografts (WKY WKY) or allografts (F344 WKY) was performed as previously reported (15), using a procedure initially described by Marck and colleagues (16), and Prop and associates (17). In brief, after the donor rats (F344 or WKY) were anesthetized with an intramuscular injection of ketamine (40 mg/kg) and xylazine (5 mg/kg), the chest was shaved, a sternotomy incision was made, and the heart and lungs were removed en bloc. The left lung was than resected, and heparinized Lactated Ringer\Qs solution was infused into the pulmonary artery. The donor lung was wrapped in sterile gauze saturated with saline and placed on ice (4°C) in a sterile beaker until transplantation.

The recipient rats were anesthetized with subcutaneous (s.c.) injection of atropine (0.05 mg/kg), followed by an inhalation of 2% halothane. The airway was cannulated with a 14-gauge Teflon catheter and the rat was mechanically ventilated with a rodent ventilator (Analytical Specialties Co., St. Louis, MO) using 100% oxygen and the inhalation of 1.5 to 2% isoflurane for maintenance anesthesia. Once a thoracotomy incision was made in the left fourth intercostal space and hemostats were placed on the left pulmonary vessels and bronchus, the left lung was resected. The pulmonary vessels of the donor lung were anastomosed to the recipient by a plastic cuff and 7-0 silk sutures (Kono, Chiba, Japan). The donor and recipient bronchi were sutured together with 8-0 Prolene sutures (Ethicon, Sommerville, NJ). Immediately after completion of the anastomosis of the bronchus, the hemostat was removed and ventilation was restored. After the left thoracotomy incision was closed over a 16-gauge chest tube with 3-0 silk suture (Ethicon), maintenance anesthesia was discontinued and the animal was allowed to recover. Once spontaneous respiration resumed, the cannula was removed from the airway and the chest tube was removed. The ischemic time of the donor lung was approximately 1 h and the total operating time for harvesting and transplanting the donor lung was approximately 2 h. All transplantation procedures were performed by one author (K.Y.) under a surgical microscope (Micro Tech, Colorado Springs, CO) under sterile conditions. The F344 WKY transplant model is associated with the development of mild acute rejection by the end of the first week and moderate to severe acute rejection by the end of the second week (18). Similar to our prior report (15), survival exceeded 90% in all transplantation groups. No immunosuppressive therapy was given at any time during the experimental period.

Transplanted lungs were monitored by serial chest radiographs on Days 1, 6, and 13 after transplantation. The radiographic changes were graded as follows: grade 1, normal; grade 2, mild infiltrates; grade 3, moderate infiltrates; and grade 4, severe infiltrates or complete opacification.

Five transplantation groups were studied: lungs from WKY rats transplanted into WKY recipients (control isografts); F344 lungs transplanted into diluent-fed WKY recipients (control allografts); F344 lungs transplanted into col(V)-fed WKY recipients [col(V)-fed allografts]; F344 lungs transplanted into col(II)- fed WKY recipients [col(II)-fed allografts]; and F344 lungs transplanted into col(XI)-fed WKY recipients [col(XI)-fed allografts].

Collection of Bronchoalveolar Lavage Fluid

Collection of bronchoalveolar lavage (BAL) fluid (BALF) was performed in ketamine-anesthetized lung transplant recipients 2 wk after transplantation as previously reported (15). In brief, BAL of native and transplanted lungs was performed by selective cannulation of right and left mainstem bronchi with a 16-gauge catheter secured by suture. While clamping the contralateral bronchus, 3-ml aliquots of sterile phosphate-buffered saline (PBS) (37°C) were instilled into each main stem bronchus and aspirated. Cell-free BALF supernatants obtained from centrifuged specimens were stored at 70°C until use. BALF differential cell counts were performed using light microscopy to count 300 cells per high-power field on cytospin preparations to determine the quantity of macrophages, lymphocytes, and polymorphonuclear (PMN) cells.

DTH Response

DTH responses were determined by a modification of a procedure described by Yamagami and coworkers (19). In brief, 2 wk after lung transplantation, control or col(V)-fed WKY rats received 10⁷ irradiated (3,000 rad) donor-derived F344 or third-party (BN) splenocytes in 30 µl of PBS into the right pinnae by s.c. injection using a 26-gauge needle. The left pinnae received an equal volume of diluent and served as the control site. Naive WKY rats were negative controls. A separate group of naive or allograft recipient WKY rats were tested with 15 µg of col(II), col(V), or col(XI) in 30 µl volume injected into the right pinnae and diluent into the left. The ear thickness was measured with a micrometer caliper (Mitutoyo; Field Tool Supply, Chicago, IL) in a blinded fashion immediately before and 24 h after injection. Antigen-specific DTH response was calculated according to the following formula: Specific ear swelling = (right ear thickness at 24 h  right ear thickness at 0 h)  (left ear thickness at 24 h  left ear thickness at 0 h) × 10³ mm (19). All data are reported as the mean of triplicate measurements.

In separate experiments, naive and col(V)-fed WKY rats were primed with 100 µg of low endotoxin bovine serum albumin (BSA) (Sigma, St. Louis, MO) dissolved in 100 µl of an emulsion of adjuvant (Titermax; CytRx Corp., Norcross, GA). Each rat was primed subcutaneously with the emulsion at the base of the tail. At 7 d later rats were challenged with 2% heat-aggregated BSA solution into the right pinnae and diluent into the left (20). Unprimed rats were controls for these studies. The ear thickness was measured immediately before and 24 h after injection and the specific ear swelling was calculated as described earlier.

Neutralization of TGF-, IL-4, and IL-10 in DTH Assays

Neutralization of TGF- at the DTH site was performed by a modification of a procedure described by Bickerstaff and colleagues (21). In brief, 2 wk after lung transplantation, col(V)-fed WKY rats received 10⁷ irradiated (3,000 rad) donor-derived F344 splenocytes mixed with 5 µg of polyclonal chicken antirat TGF- antibody (Ab), or 5 µg of polyclonal goat antirat IL-4 or IL-10 Ab (all from R&D Systems, Minneapolis, MN) in 30 µl of PBS into the right pinnae by s.c. injection using a 26-gauge needle. The left pinnae received an equal volume of diluent and served as the control site. For negative controls, a separate group of col(V)-fed allografts received 10⁷ irradiated (3,000 rad) donor-derived F344 splenocytes mixed with 5 µg of control chicken immunoglobulins (Igs) or control goat Igs (R&D Systems) into the right pinnae and diluent into the left. The specific ear swelling was determined as described earlier. Control Igs had no effect on the DTH response.

Rat Model of Acute Lung Injury

Alveolar or intravenous instillation of lipopolysaccharide (LPS) (Sigma) was performed by a modification of a procedure described by O'Leary and colleagues (22). Briefly, normal WKY rats or col(V)-fed WKY rats, 1 wk after last feeding, were anesthetized with a s.c. injection of atropine (0.05 mg/kg) followed by an inhalation of 2% halothane. The airway was cannulated with a 14-gauge Teflon catheter and the rat was mechanically ventilated with a rodent ventilator (Analytical Specialties) using 100% oxygen, and the inhalation of 1.5 to 2% isoflurane for maintenance anesthesia. After disappearance of spontaneous respiration, LPS (1 mg/kg at 1 mg/ml) was instilled into the airway and mechanically ventilated for 10 min. Maintenance anesthesia was discontinued and the animal was allowed to regain consciousness. Once spontaneous respiration resumed, the cannula was removed from the airway. In separate experiments, rats were injected intravenously into the tail veins with LPS (4 mg/kg at 1 mg/ml). At 24 h after challenge, BAL was performed and the lungs were harvested for assessment of pathology.

Quantitation of Cytokines

TGF- levels in serum of the experimental groups were quantitated by enzyme-linked immunosorbent assay (ELISA) using the TGF-₁ immunoassay system (Promega, Madison, WI) per manufacturer's protocol. IL-4 and IL-10 levels in serum were quantitated by ELISA using Cytoscreen immunoassay kits (BioSource International, Camarillo, CA) per manufacturer's protocol. The sensitivities of the TGF-, IL-4, and IL-10 assays were 32, 2, and 5 pg/ml, respectively.

Pathologic Grading

Native and transplanted lungs from each group were harvested, fixed, sectioned, stained, and graded for rejection pathology using standard criteria (23) by a pathologist (one author [O.W.C.]) in a blinded fashion without prior knowledge of the transplantation group, as previously reported (15).

Statistics

Analyses of PMN cell and lymphocyte counts in BALF were performed initially by analysis of variance (ANOVA) to determine whether differences were present among groups. If differences were found, then a post hoc analysis using a Student-Newman- Keuls test was performed to determine which group was different. Because data for DTH in control allograft and naive WKY rats challenged with different antigens were found to be non-normally distributed, a rank-sum two-way ANOVA with interaction was used to determine differences among groups. Differences in DTH responses to donor alloantigens between control allografts and col(V)-fed allografts were determined using a Mann-Whitney U test. In all of these analyses, P values < 0.05 were determined to be significant. Differences between airway and vascular pathologic scores were determined initially utilizing the Kruskal- Wallis test followed by a post hoc analysis using the Mann-Whitney U test. P values < 0.03 were determined to be significant. Student's t test for multiple comparison was used for analysis of cytokines. P values < 0.05 were determined to be significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

DTH responses to donor antigens, an in vivo test of cellular immunity, have been reported to correlate with the extent of rejection in various rodent models of organ transplantation other than the lung (24, 25). We have previously reported that col(V) is a target of the local immune response to lung alloantigens in mice (11). Therefore, we determined the systemic DTH response to alloantigen in naive rats and lung allograft recipients to determine whether col(V) is recognized as an antigen during lung allograft rejection. We also determined whether lung allograft recipients develop a DTH response to col(V). DTH responses to F344 (donor) splenocytes and col(V) were examined in WKY rats 2 wk after receiving F344 lung allografts, the time at which severe acute rejection begins to develop (18), and in naive, nontransplanted WKY rats. To determine the specificity of the DTH response to alloantigens and col(V), we also determined DTH responses to col(II), col(XI), and third-party antigens, BN splenocytes. Col(II), a major component of the articular cartilage, is not present in the lung and is not homologous to col(V) (12). In contrast, col(XI) has homology to col(V) (13) but, similar to col(II), it is found in articular cartilage and is not present in the lung. For these reasons, col(II) and col(XI) served as controls for col(V).

Using specific ear swelling as a measurement of DTH responses, Figure 1 shows that control allograft recipients developed significant DTH responses to F344 splenocytes and col(V) 2 wk after transplantation [P < 0.0001 compared with naive WKY rats challenged with F344 splenocytes or col(V), and P < 0.0001 compared with naive WKY rats challenged with col(V) or F344 splenocytes] (Figure 1). In contrast, control allografts did not have DTH responses to third-party (BN) antigens, col(II), or col(XI) (Figure 1). Naive WKY rats did not have DTH responses to col(II), or col(XI) (data not shown). These data confirm other studies showing that DTH responses are indicative of immune activation during allograft rejection, which is specific to donor but not third-party alloantigens (24). In addition, it confirms our prior report in mice (11) that col(V), but not col(II) or col(XI), is a target of the immune response to lung alloantigens.

View larger version (13K): [in this window] [in a new window]   Figure 1.   DTH responses to donor alloantigens, col(II), col(V), col(XI), and third-party alloantigens in control allograft recipients 2 wk after transplantation. Animals received 107 irradiated (3,000 rad) donor-derived F344 splenocytes, third-party (BN) splenocytes, or 15 µg of col(II), col(V), or col(XI) into the right pinnae and diluent into the left pinnae. The ear thickness was measured with a micrometer caliper in a blinded fashion immediately before and 24 h after injection and the specific ear swelling was calculated as described in MATERIALS AND METHODS. Spl, splenocytes. Data represent means ± standard error of the mean (SEM) of specific ear swelling in mm × 103 of four rats in each group. *P < 0.0001 compared with naive WKY rats challenged with F344 splenocytes or col(V);  P < 0.0001 compared with naive WKY rats challenged with col(V) or F344 splenocytes.

Prior reports have shown that oral administration of antigens that are targets of the immune response during rejection of allografts, other than the lung, induces tolerance to the donor organ (8, 9). To determine whether oral administration of collagens to lung allograft recipients before transplantation induces immunologic tolerance to the donor lung, WKY recipients were fed col(II), col(V), or col(XI) before transplantation as described in MATERIALS AND METHODS, followed by an assessment of serial chest X-rays, allograft BALF differential cell counts, pathologic grading, and DTH responses to donor antigens.

Figure 2 shows the differential cell counts in BALF from the experimental groups 2 wk after transplantation, which is the time of onset of severe acute rejection (18), and in normal WKY rats. There were no differences in BALF differential cell counts in normal compared with isograft lungs. Similar to prior reports (17, 26), PMN cells and lymphocytes were significantly increased in control allograft BALF compared with normal or isograft lungs (P < 0.00001 for lymphocytes and P < 0.038 for PMN cells compared with normal or isograft lungs) (Figure 2). In contrast, feeding col(V) before transplantation resulted in a significant reduction in BALF PMN cells and lymphocytes compared with control allografts (P < 0.0001 for lymphocytes and P < 0.023 for PMN cells compared with control allografts) (Figure 2).

View larger version (21K): [in this window] [in a new window]   Figure 2.   BALF differential cell counts in normal WKY lungs, control isograft lungs, control allograft lungs, and col(V)-fed allograft lungs. At 2 wk after transplantation, transplanted lungs underwent BAL as described in MATERIALS AND METHODS. Differential cell counts were determined by counting 300 cells/field on cytospin preparations using light microscopy. Mac, macrophages; Lym, lymphocytes. Data represent means ± SEM in percent of total cell counts of four normal WKY lungs, four control isografts, five control allografts, and five col(V)-fed allografts. *P < 0.00001 for lymphocytes and  P < 0.038 for PMN cells compared with normal or isograft lungs;  P < 0.0001 for lymphocytes and § P < 0.023 for PMN cells compared with control allografts.

Acute allograft rejection is usually associated with an increase of total cell counts in allograft BALF (18). However, at 2 wk after transplant, the control WKY allograft lungs were usually undergoing severe rejection and, due to destruction of the allograft, sufficient BAL could not be performed reliably to determine BALF total cell counts. In contrast, col(V)-fed allograft recipients showed less severe rejection, which allowed easier BAL, resulting in higher cell counts. For these reasons, comparison of total cell counts between the groups could not be done. Collectively, these data demonstrate that oral immunization with col(V) is associated with fewer PMN cells and lymphocytes in allograft BALF during acute rejection.

Diminished PMN-cell and lymphocyte counts in allograft BALF is usually associated with less severe radiographic and histologic lesions during acute lung allograft rejection. To determine the rate of progression of lung infiltrates, transplant recipients were monitored by serial chest radiographs on Days 1, 6, and 13 after transplantation and graded as described in MATERIALS AND METHODS. As shown in Figure 3, control isografts did not have any pulmonary infiltrates at all monitored time points (grade 1) (Figure 3A). In the control allografts, serial X-rays revealed gradual development of mild infiltrates (grade 2) in the left lung at 6 d after transplant (data not shown), which resulted in severe infiltrates and complete opacification of the allograft (grade 4) by the end of the second week (Figure 3B). However, in col(V)-fed allografts, the development of infiltrates was much slower compared with controls. The X-rays were normal (grade 1) at Day 6 and only mild infiltrates (grade 2) were present at 2 wk after transplantation (Figure 3C).

View larger version (81K): [in this window] [in a new window]   Figure 3.   Serial chest X-rays of transplant recipients 2 wk after transplantation. The arrows in the left lung field point to the cuffs used for vascular anastomosis. Control isograft recipients show normal chest X-rays (grade 1) (A). X-rays of control allograft recipients reveal severe infiltrates and complete opacification of the allograft, indicative of severe rejection (grade 4) (B). However, col(V)-fed allograft recipients show only mild infiltrates at 2 wk after transplantation (grade 2) (C). Chest X-rays are representative of five rats in each group.

The upper panels of Figure 4 show the gross anatomy of the native and isograft WKY lungs, and the native and allograft lungs from control allograft and col(V)-fed allograft rats harvested at 2 wk after transplantation. The isograft (left-L) and the native lung (right-R) were normal in appearance (Figure 4A). The transplanted lung in control allograft recipients was dark brown in color and shrunken compared with the native lung (Figure 4B). In contrast, the transplanted left lung in col(V)-fed allograft recipients (Figure 4C) had the appearance of the native (normal) or isograft lung (Figure 4A).

View larger version (94K): [in this window] [in a new window]   Figure 4.   Upper panel: Gross anatomy of control isograft lungs (A), control allograft lungs (B), and col(V)-fed allograft lungs (C) 2 wk after transplantation (posterior view). The left (L) lung is the transplanted lung and the right (R) is the native lung in each panel. The control allograft lung (L in B) is dark brown in color and shrunken compared with the native lung. However, the col(V)-fed allograft lung (L in C) has the appearance of the isograft lung (L in A). Control isograft lungs (A) show no pathologic lesions and are identical to normal WKY lungs (data not shown). Photographs are representative of five rats in each group. Lower panel: Histology of control isografts (D), control allografts (E), and col(V)-fed allografts (F) 2 wk after transplantation. Control isografts show normal airway and vascular structures (D). Control allografts show extensive perivascular, peribronchial, and alveolar mononuclear cell infiltrates, consistent with severe rejection (E). In contrast, col(V)-fed allografts show only mild to moderate perivascular and peribronchial mononuclear cell infiltrates (F). Photomicrographs are representative of five rats in each group (original magnification: ×100).

Fewer PMN cells and lymphocytes in allograft BALF, less severe infiltrates in the allografts on chest X-ray, and preserved gross anatomy suggest that feeding col(V) before transplantation downregulated development of rejection pathology. The lower panels of Figure 4 show the representative histology of control isografts, control allografts, and col(V)-fed allografts 2 wk after transplantation. Similar to prior reports (26), all control isograft lungs had normal histology without signs of rejection (Figure 4D). Control allografts revealed extensive perivascular, peribronchial, and alveolar mononuclear cell infiltrates, consistent with severe acute rejection (Figure 4E). In contrast, only mild to moderate perivascular and peribronchial infiltration were detected in the col(V)-fed allograft lungs (Figure 4F).

Table 1 shows the grading of rejection pathology at 2 wk after transplantation. Acute rejection was graded A0 to A4 according to the presence and extent of perivascular and interstitial mononuclear cell infiltrates, and B0 to B4 according to the extent and intensity of the airway inflammation (23). All control isograft lungs revealed normal histology of the lung (A0 ± 0, B0 ± 0). The control allografts had severe vascular and airway rejection (A3.8 ± 0.2, B4 ± 0, respectively). In contrast, col(V)-fed allografts showed mild to moderate vascular and airway rejection (A2.8 ± 0.2, B2.6 ± 0.2, respectively) (P < 0.028 for A scores and P < 0.009 for B scores compared with control allografts) (Table 1). Experiments showed that feeding col(II) or col(XI) had no effect on development of allograft pathology compared with control allografts (Table 1). These data show that feeding col(V) downregulated acute rejection pathology.

Data showing that feeding col(V) downregulates lung allograft rejection suggest that orally tolerized lung allograft recipients should have diminished DTH responses to donor alloantigens. To determine whether col(V) feeding diminished immune responses to alloantigens, control allograft recipients and col(V)-fed allograft recipients were challenged in the right pinnae with whole allogeneic F344 splenocytes and PBS in the left pinnae and DTH responses were determined. As shown in Figure 5 and previously shown in Figure 1, untreated control allograft recipients had a strong DTH response after challenge with donor antigen. In contrast, compared with control allograft recipients, the DTH response to donor antigen was reduced significantly in col(V)-fed allograft recipients (P < 0.02) (Figure 5).

View larger version (16K): [in this window] [in a new window]   Figure 5.   Reduction of DTH responses to donor alloantigens by oral administration of col(V). Control allograft recipients and col(V)-fed allograft recipients were challenged in the right pinnae with 107 irradiated (3,000 rad) donor-derived F344 splenocytes, and diluent in the left pinnae 2 wk after transplantation. The ear thickness was measured with a micrometer caliper in a blinded fashion immediately before and 24 h after injection and the specific ear swelling was calculated as described in MATERIALS AND METHODS. Spl, splenocytes. Data represent means ± SEM of specific ear swelling in mm × 103 of four rats in each group (*P < 0.02 compared with control allografts).

The impaired immune response to alloantigen induced by col(V) could have been due to global immune hyporesponsiveness (4) and not immune tolerance. Therefore, to determine whether col(V)-fed WKY rats could respond to other antigens, these rats received LPS either intratracheally (1 mg/kg) or intravenously (4 mg/kg) at doses known to induce severe inflammatory reactions in the lung and systemically 24 h after challenge (22). The disease induced is analogous to pneumonia and sepsis caused by gram-negative bacteria (22). Similar to normal WKY rats, instillation of LPS into lungs or injected intravenously into col(V)-fed WKY rats induced severe illness (ruffled fur and prostration) and massive influx of PMN cells and lymphocytes into the lung as observed in BALF differential cell counts and pathology (data not shown).

To investigate further whether the impaired immune response induced by col(V) feeding was antigen-specific, we determined whether feeding col(V) affected DTH responses to an unrelated nominal antigen, BSA, a T lymphocyte-dependent antigen in rats (20). Naive and col(V)- fed WKY rats were primed by s.c. injection of 100 µg of BSA in adjuvant, and 7 d later were challenged with 2% heat-aggregated BSA solution in the right pinnae and diluent in the left. DTH responses were determined 24 h after injection. Unprimed WKY rats served as controls for these studies. As shown in Figure 6, injection of BSA into the pinnae of unprimed rats did not induce significant ear swelling. In contrast, injecting BSA into unfed primed WKY rats induced significant ear swelling (P < 0.018 compared with unprimed naive WKY rats) (Figure 6). However, col(V) feeding did not affect the DTH responses to BSA (P > 0.05 compared with primed WKY rats) (Figure 6). Collectively, these data show that col(V)-induced suppression of lung allograft rejection is mediated by immune tolerance and not by global immune hyporesponsiveness.

View larger version (17K): [in this window] [in a new window]   Figure 6.   DTH responses to BSA in naive and col(V)- fed WKY rats. Naive and col(V)-fed WKY rats were primed by s.c. injection of 100 µg of BSA in adjuvant and 7 d later were challenged with 2% heat-aggregated BSA solution in the right pinnae and diluent in the left. The ear thickness was measured with a micrometer caliper in a blinded fashion immediately before and 24 h after injection and the specific ear swelling was calculated as described in MATERIALS AND METHODS. Unprimed WKY rats served as controls. Data represent means ± SEM of specific ear swelling in mm × 103 of four rats in each group (*P < 0.018 compared with unprimed naive WKY rats;  P > 0.05 compared with primed WKY rats).

Systemic production of TGF-, IL-4, and IL-10 are cited frequently as cytokines responsible for suppressing immune responses in oral tolerance (4). Therefore, we next determined whether oral tolerance induced by col(V) is associated with upregulated production of TGF-, IL-4, and IL-10 during lung allograft rejection. Using commercial ELISAs, TGF-, IL-4, and IL-10 were quantitated in serum of the experimental groups. Figure 7 shows the serum TGF- levels in normal WKY rats, control allografts, and col(V)-fed allografts 2 wk after transplantation. As expected, low levels of TGF- were present in the serum of normal WKY rats (27). There was a slight increase of TGF- in control allografts. In contrast, TGF- levels were upregulated markedly in serum of col(V)-fed allografts (P < 0.05 compared with control allograft recipients) (Figure 7). Neither IL-4 nor IL-10 was detectable in serum of the same rats (data not shown).

View larger version (9K): [in this window] [in a new window]   Figure 7.   TGF- levels in serum of normal WKY rats, control allograft recipients, and col(V)-fed allograft recipients at 2 wk after transplantation. Levels of TGF- in serum were quantitated by ELISA per manufacturer's protocol. Data represent means ± SEM of TGF- in pg/ml of four rats in each group (*P < 0.05 compared with control allograft recipients).

Although IL-4 and IL-10 were not detected in the serum, this did not preclude their activity systemically in downregulating cellular immune responses to donor alloantigens. To determine whether TGF-, IL-4, or IL-10 had a role in suppression of immune responses to alloantigens, we used neutralizing Abs to these cytokines in the DTH assay to donor antigens. Using a modification of a procedure reported by Bickerstaff and associates (21), at 2 wk after lung transplantation col(V)-fed WKY rats received 10⁷ irradiated (3,000 rad) donor-derived F344 splenocytes mixed with 5 µg of polyclonal anti-TGF- Ab or 5 µg of polyclonal anti-IL-4 or -IL-10 Ab in PBS into the right pinnae. The left pinnae received an equal volume of diluent plus splenocytes and served as the control site. For negative controls, a separate group of col(V)-fed allografts received control Igs with splenocytes into the right pinnae and an equal volume of diluent plus splenocytes into the left pinnae. As shown in Figure 8 and previously shown in Figures 1 and 5, after challenge with donor antigen untreated control allograft recipients had a strong DTH response that was significantly reduced in col(V)-fed allograft recipients. However, col(V)-fed allografts significantly recovered DTH responses when anti-TGF- Abs were mixed with donor splenocytes and injected into the pinnae of the ears [P < 0.03 compared with col(V)-fed allografts challenged with antigens mixed with control Ig] (Figure 8). In contrast, mixing donor splenocytes with neutralizing Abs to IL-4 or IL-10 was less effective in restoring DTH responses to donor antigens [P > 0.05 compared with col(V)-fed allografts challenged with antigens mixed with control Ig] (Figure 8). The restoration of the DTH responses in col(V)-fed allografts with anti-TGF-, anti-IL-4, and anti-IL-10 Abs relative to control allografts was 75.7, 24.3, and 39.9%, respectively (Figure 8).

View larger version (16K): [in this window] [in a new window]   Figure 8.   Neutralization of TGF- restored DTH responses to donor alloantigens in col(V)-fed allograft recipients. Col(V)-fed WKY rats were challenged in the right pinnae with 107 irradiated (3,000 rad) donor-derived F344 splenocytes mixed with either 5 µg of polyclonal anti-TGF- Ab or anti-IL-4 or -IL-10 Ab in PBS 2 wk after transplantation. The left pinnae received an equal volume of diluent plus splenocytes, and served as the control site. For negative controls, a separate group of col(V)-fed allografts received control Igs with splenocytes into the right pinnae and an equal volume of diluent plus splenocytes into the left pinnae. The ear thickness was measured with a micrometer caliper in a blinded fashion immediately before and 24 h after injection and the specific ear swelling was calculated as described in MATERIALS AND METHODS. Spl, splenocytes. Data represent means ± SEM of specific ear swelling in mm × 103 of four rats in each group. *P < 0.03 and , P > 0.05 compared with col(V)-fed allografts challenged with antigens mixed with control Ig. The restoration of the DTH responses in col(V)-fed allografts with anti-TGF-, anti-IL-4, and anti-IL-10 antibodies relative to control allografts was 75.7, 24.3, and 39.9%, respectively.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Oral tolerance has been shown to be of benefit in downregulating alloreactivity in organ transplants other than the lung (8, 9). However, oral tolerization in lung transplantation has not previously been reported. Using a rat model of lung transplantation, data in the current study show that oral administration of col(V) to lung allograft recipients before lung transplantation downregulates rejection responses. Immunologic, radiologic, and histologic analysis of col(V)-fed compared with control allograft recipients showed that feeding col(V) was associated with diminished PMN-cell and lymphocyte counts in allograft BALF, less severe infiltrates in the allografts on chest X-ray, preservation of gross anatomy of the allograft, and reduction of rejection pathology. Feeding col(V) induced upregulated production of TGF- in the serum. Finally, orally tolerized allograft recipients showed diminished DTH responses to donor alloantigens, which were recovered by neutralizing TGF-.

In our prior study we showed that col(V) is a target of the local immune response to alloantigens in mice (11). Col(V) is a minor collagen in the lung (28) which is located in the peribronchiolar connective tissues (29), alveolar interstitium (30), and capillary basement membrane (29). These tissues have been shown to be sites of pathologic lesions in response to alloantigens in our murine model (31) and are sites of rejection activity in human lung allograft recipients (1). Because prior reports have shown that MHC-derived peptides or other proteins/peptides that are targets of the immune response can be used in the induction of immunologic tolerance (5), col(V) was used in this study for tolerance induction. Our current data show that col(V) may be a strong candidate for the induction of immunologic tolerance and prevention of allograft rejection.

The oral administration of antigens is an effective method of inducing peripheral T-lymphocyte tolerance. This phenomenon, often referred to as oral tolerance, has been well studied in various models of autoimmune diseases in animals, including experimental allergic encephalomyelitis, uveitis, diabetes, myasthenia gravis, and arthritis (reviewed in Ref. 4). Early results from clinical trials in humans suggest that oral tolerance is effective in autoimmune uveitis, diabetes, nickel allergy, and possibly multiple sclerosis (4). However, there are few studies reporting oral tolerance induction in organ transplantation (8, 9). In each report, tolerance was induced by feeding donor MHC-derived peptides or feeding allogeneic cells before transplantation, which were effective in preventing rejection of cardiac and corneal allografts (8, 9; reviewed in Ref. 4).

There are three mechanisms for the induction of oral tolerance: active suppression of antigen-specific cells, clonal anergy of antigen-specific cells, and clonal deletion of antigen-specific cells (4). Although all three mechanisms can be operative simultaneously in response to oral tolerance, active suppression and clonal anergy are the key mechanisms of immune suppression induced by oral tolerance (4). In addition, different mechanisms of tolerance are induced when antigens are given orally at different doses (4). Generally, higher doses of antigen are reported to induce anergy or clonal deletion, whereas low doses induce cytokine regulation and active suppression (4).

Active suppression is described as the regulation of one lymphocyte subset by another in an antigen-specific manner (4). Depending on the antigen and disease state, the suppressor cells may be CD4+ and/or CD8+ T lymphocytes that migrate from peripheral lymphoid tissues, such as spleen and peripheral lymph nodes, to the sites of disease activity (4). Adoptive transfer of these cells to naive recipients has confirmed the role of these cells in active suppression in rodent models of ovalbumin-induced hypersensitivity and multiple sclerosis (4). In our model of oral tolerance, preliminary data show that adoptive transfer of splenic T lymphocytes from col(V)-fed allograft recipients prevents lung allograft rejection in naive allograft recipients. The specific cell types inducing immune suppression will be examined in future studies to confirm the development of regulatory cells.

Clonal anergy refers to unresponsiveness of antigen-specific T lymphocytes, which is characterized by diminished proliferation after exposure to an antigen and is involved in oral tolerance in several animal models (4). Anergy could be the result of production of soluble suppressive factors by CD4+ or CD8+ T lymphocytes themselves, by other T lymphocytes or cells in the local enviroment, or as a result of decreased expression of appropriate costimulatory molecules (4). To determine whether T lymphocytes from col(V)-fed tolerant lung allograft recipients are anergic, proliferative responses of T lymphocytes from control allograft recipients and col(V)-fed allograft recipients will be compared in a mixed leukocyte reaction in future studies. Experiments have begun to isolate alloreactive T lymphocytes for use in studies to examine the role of clonal deletion (32) as a mechanism of col(V)-induced tolerance.

The soluble mediators that suppress the immune response during oral tolerance are derived mainly from regulatory or suppressor T lymphocytes (4). There are four types of T lymphocytes, described by the cytokines they produce: T helper (Th) 1-type that produce IL-2 and interferon (IFN)- (4, 10); Th2-type that produce IL-4 and IL-10 (4, 10); Th3-type that produce high levels of TGF-, alone or in conjunction with very low levels of IL-4, IL-10, or IFN- (4, 10); and Tr1 cells that produce high levels of IL-10 in conjunction with low levels of TGF- (4, 10). Because Th2, Th3, and Tr1 T lymphocytes have been shown to be the major mediators of active suppression induced by oral tolerance, TGF-, IL-4, and IL-10 are believed to be key cytokines in this process (4, 10). Data in the current study showing that antibodies to TGF-, but not IL-4 or IL-10, recovered the DTH responses to donor alloantigens in orally tolerized animals suggest that Th3 cell-induced active suppression may be responsible for oral tolerance in response to col(V). However, other mediators may be involved and will be examined in future studies.

In experimental autoimmune models of oral tolerance, regulatory cells after oral tolerization are triggered in an antigen-specific fashion but suppress in an antigen nonspecific fashion (4). In transplantation research, different techniques have been used to induce transplantation tolerance. Donor-specific blood transfusion (33), bone-marrow transplantation (34), thymic injection of allogeneic cells (35), or systemic immunization with donor MHC-derived peptides (5) have been shown to induce transplantation tolerance in various animal models. However, these techniques would have limited utility in the potential lung allograft recipient due to the fact that the donor cells used for tolerance induction would not be available in sufficient time to induce tolerance before transplantation. Our model of oral tolerance in lung transplantation showed that orally administered col(V), which is not donor-specific, is capable of suppressing alloreactivity and inducing transplantation tolerance. Therefore, it may not be necessary to identify the target autoantigen itself, but necessary only to administer orally a protein capable of inducing regulatory cells that secretes suppressive cytokines. Accordingly, data presented in the current study suggest that col(V)-induced oral tolerance may be a potential therapy for preventing lung allograft rejection.