|
|
||||||||
J Thorac Cardiovasc Surg 2000;120:1070-1077
© 2000 The American Association for Thoracic Surgery
General Thoracic Surgery |
From the Department of Surgery,a Division of Cardiothoracic Surgery, and the Department of Pathology and Laboratory Medicine, b Hospital of the University of Pennsylvania, Philadelphia, Pa.
Supported in part by the American Surgical Association, the Charles E. Culpeper Foundation, and The American Association for Thoracic Surgery, Dwight Harken Research Scholarship.
Address for reprints: Bruce R. Rosengard, MD, Division of Cardiothoracic Surgery, Department of Surgery, Hospital of the University of Pennsylvania, 3400 Spruce St, Silverstein 6th Floor, Philadelphia, PA 19104(E-mail: brosenga{at}mail.med.upenn.edu).
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The pathogenesis of OB remains poorly understood, but the histologic characteristics and the absence of the process in experimental animals rendered tolerant have led many to speculate that OB is the primary manifestation of chronic rejection in lung transplantation. Clinical observations support the notion that OB results from alloimmune-mediated mechanisms. Acute rejection has been demonstrated to be a major risk factor for the subsequent development of OB.
5-7 Up-regulation of major histocompatibility complex (MHC) class I and class II antigens on bronchial epithelium and infiltration of donor-specific lymphocyte subsets have been found in grafts with OB.
8-10 The presence of anti-HLA class I antibodies in lung transplant recipients is associated with OB, and recently anti-HLA class I antibodies have been shown to lead to protein phosphorylation and proliferation of cultured lung epithelium.
11 Furthermore, increased immunosuppression has been shown to stabilize lung function in some patients.
12,13
Bone marrowderived antigen-presenting cells (APCs) play a central role in triggering recipient T cells.
14 However, little is known about how allorecognition pathways affect the development of OB. Responses to allogeneic tissue are unique in that both donor and recipient APCs are capable of triggering rejection, termed direct and indirect allorecognition, respectively. Both recipient CD4+ and CD8+ T cells can be triggered directly by donor APCs, whereas recipient APCs stimulate only CD4+ T cells by presenting allopeptides in the context of self MHC class II molecules. Understanding the role of allorecognition pathways in OB may lead to the development of rational strategies to prevent this malady.
The goal of our study was to investigate whether direct allorecognition triggered by donor APCs is important in the development of OB, and, if so, which pathways are involved. We used a heterotopic mouse tracheal transplant model to study the development of obliterative airway disease, which resembles OB histologically.
15 To eliminate the direct allorecognition pathway, we created chimeric B6 tracheas bearing recipient C3H APCs, denoted hereafter as B6(C3H), using allogeneic bone marrow transplantation (BMT),
16,17 and transplanted them into C3H recipients. Furthermore, tracheas from B6 MHC class I knockout (B6I) mice and B6 MHC class II knockout (B6II) mice were transplanted to examine the contributions of CD4+ and CD8+ direct allorecognition pathways. Naive or autologously reconstituted B6 and C3H tracheas were transplanted into C3H recipients as our positive and negative controls, respectively. The autologously reconstituted B6 tracheas, denoted hereafter as B6(B6), behaved similarly to the naive B6 and served as controls for the BMT tracheas. We report here that donor bone marrowderived APCs are important in the development of obliterative airway disease and that donor APC expression of either MHC class I or class II antigen is sufficient to trigger the development of obliterative airway disease.
| Materials and methods |
|---|
|
|
|---|
BMT
Chimeric tracheas were created by BMT as previously described.
16,17 In brief, BMT recipient mice were lethally irradiated (10 Gy) and then reconstituted with 1 x 107 unfractionated bone marrow cells from a donor, injected in 1 mL of medium via the tail vein. The BMT recipients were treated with 2 weeks of oral antibiotics and allowed to engraft for 90 days before their use as trachea donors.
Heterotopic tracheal transplantation
By means of a technique first described by Hertz and associates,
15 1 to 4 tracheal segments were transplanted subcutaneously into the dorsa of C3H recipients in experimental groups as shown in Table I. Recipients with multiple tracheal segments contained grafts from the same genotypes. No immunosuppression was administered to any group. Donor mice were anesthetized with enflurane, and their skin was prepared with ethyl alcohol. In each case, the thoracic cavity was entered through a midline incision, and the animal was exsanguinated. The thymus and the heart were removed, so that the trachea and the main bronchi were exposed. With sharp dissection, the trachea from the larynx to the carina was removed and placed in cold Dulbecco's phosphate-buffered saline solution (Gibco, Grand Island, NY).
|
Confirmation of chimerism with flow cytometry
To confirm the status of bone marrow chimeras, we analyzed flow cytometry as previously described by our laboratory.
16,17 We have previously shown that the degree of APC replacement in the spleen consequent to BMT mirrors the degree of APC replacement in other organs. This method has been further validated by reverse transcriptasepolymerase chain reaction and the mixed lymphocyte reaction. Spleens of donor chimeric animals were harvested at the time of transplantation. The spleens were mechanically disrupted and filtered over a 70-µm sterile nylon mesh. The spleen cells were treated with ACK buffer (BioWhittaker, Walkersville, Md), washed with cold phosphate-buffered saline solution, and counted. All antibodies were obtained from Pharmingen (San Diego, Calif). Fluorescein isothiocyanateconjugated mouse anti-mouse I-Ab (clone 25-9-17, immunoglobulin G2a) and fluorescein isothiocyanateconjugated mouse anti-mouse I-Ak (11-5.2, immunoglobulin G2b) were used to label spleen cells from chimeras for flow cytometry analysis. Flow cytometry analysis was performed on a FACScan flow cytometer equipped with CELLQuest software (Becton Dickinson, San Jose, Calif).
Graft removal and data analysis
Tracheal grafts were removed 14 and 28 days after transplantation. In recipients with multiple grafts, the tracheal segments were removed at the same time points. Specimens were fixed in 10% formalin and submitted for staining with hematoxylin and eosin. All tracheal specimens were evaluated for their degree of luminal occlusion, the morphology of the respiratory epithelium, and the presence of intraepithelial and peritracheal cellular infiltrates by a pathologist blinded to the protocol. For recipients with multiple tracheal segments, the mean luminal occlusion of the grafts was reported. The Student t test was used to compare the mean luminal occlusion in all groups versus that of the naive and autologously reconstituted B6 allograft group.
| Results |
|---|
|
|
|---|
Luminal occlusion
Luminal occlusion was determined by microscopy and expressed as a percentage of the cross-sectional area of the tracheal lumen. The mean luminal occlusion of the groups was compared as shown inFig 1. No statistically significant difference existed between the mean occlusion of the naive B6 and the autologously reconstituted B6(B6) grafts at day 28 (P = .43). Hence, all B6 and B6(B6) grafts were placed into the allograft group. At days 14 and 28, none of the isografts showed evidence of luminal occlusion. In the allograft group at day 14, the luminal occlusion was 5.7% ± 5.4% (mean ± standard error of the mean). Specimens from 14 of 15 recipients demonstrated minimal occlusion (ranging from 0% to 5%), and 1 of 15 demonstrated marked occlusion (80%). At day 28, the luminal occlusion was 69.5% ± 11.6%. Specimens from 10 of 15 recipients demonstrated marked occlusion (ranging from 90% to 100%), and 5 of 15 demonstrated minimal to moderate occlusion (ranging from 0% to 50%). In the B6I and B6II groups, tracheas showed minimal occlusion at day 14 (ranging from 0% to 5%). At day 28 in the B6I group, the luminal occlusion was 53.0% ± 16.3%. Specimens from 4 of 8 recipients demonstrated marked occlusion (ranging from 70% to 100%), and 4 of 8 had moderate occlusion (ranging from 0% to 55%). Similarly, the luminal occlusion in the B6II group at day 28 was 52.2% ± 15.9%. Specimens from 4 of 9 recipients had marked occlusion (100%), and 5 of 9 demonstrated moderate occlusion (ranging from 0% to 50%). Tracheas from the B6I and B6II groups did not demonstrate significant decreases in luminal occlusion at day 28 when compared with the allograft group (P = .20, .19, respectively). In the B6(C3H) group at day 14, the luminal occlusion was 0.7% ± 0.7%. Specimens from all 7 recipients demonstrated minimal occlusion (ranging from 0% to 5%). At day 28, the luminal occlusion was 33.6% ± 16.2%. Specimens from 6 of 9 recipients demonstrated minimal occlusion (ranging from 0% to 5%), and 3 of 9 demonstrated marked occlusion (ranging from 90% to 100%). B6(C3H) tracheas demonstrated a significant decrease in luminal occlusion when compared with the allograft group at day 28 (P = .039).
|
|
| Discussion |
|---|
|
|
|---|
Evidence from human studies has suggested that OB is an alloimmune-mediated process against the bronchial epithelium. Increased immunosuppression has been shown to stop the progression of OB in some patients.
12,13 Presentation of MHC class I and class II antigens on the bronchial epithelium has been demonstrated to be a possible mechanism in the development of OB.
8-11,19-21 Professional APCs such as dendritic cells are associated with bronchial tissues of patients with OB and have been postulated to be involved in the development of OB.
22
Animal models have also been described to study the relationship between the immune responses and the development of OB. Orthotopic allogeneic lung transplants in rats and pigs develop the lesions of OB.
23,24 However, these models require substantial surgical expertise, and OB does not consistently develop in the small animal models. The mouse heterotopic tracheal allograft model of obliterative airway disease was first described by Hertz and associates
15 in 1993. This model reproduces some of the histopathologic characteristics of OB over a period of 3 to 4 weeks.
Data from animal studies support the notion that OB is a result of histoincompatibility. Expression of MHC class II has been shown to be increased on the bronchial epithelium in rat lung allografts with OB.
23,24 Furthermore, Pham and colleagues
25 demonstrated that induction of tolerance by means of mixed hematopoietic chimerism prevents OB in rat orthotopic lung allografts. Similarly, Kelly, Hertz, and Mueller
26 demonstrated that alloimmune responses directed against class I antigens on bronchial epithelium may be responsible for the development of obliterative airway disease. Moreover, several groups have shown that the use of immunosuppression can prevent the development of obliterative airway disease.
27,28
The present study provides further evidence that OB is an immune-mediated process. Consistent with the findings in human and animal studies, our data suggest that professional APCs may play a central role in the development of obliterative airway disease.
22-24 By replacing donor-type APCs with recipient-type APCs, we eliminated direct allorecognition of allopeptides presented by donor-type APCs and prevented the development of obliterative airway disease. Similar to the B6 and B6(B6) allografts, the B6(C3H) specimens exhibited evidence of an early inflammatory event at day 14 that subsided by day 28. However, in contrast to the B6 and B6(B6) allografts, luminal occlusion of the B6(C3H) specimens remained minimal at day 28, and although flattened and nonciliated, the respiratory epithelium remained intact. Interestingly, immunosuppressive regimens including cyclosporine (INN: ciclosporin) and sirolimus have been shown to inhibit the development of obliterative airway disease and have resulted in similar epithelial changes with minimal luminal occlusion.
27,28 The pathogenesis of OB is thought to involve an early epithelial injury followed by aberrant epithelial regeneration related to the alloimmune response. Our findings are consistent with this paradigm since B6(C3H) tracheas demonstrate early epithelial injury but undergo a near-normal process of repair, presumably as a consequence of reduced immunogenicity of the grafts. Tracheas from autologously reconstituted B6 mice confirmed that the lack of obliterative airway disease was not due to the effects of irradiation or BMT.
Our findings involving the B6I and the B6II tracheas have further demonstrated that "professional" APCs expressing either MHC class I or class II antigens are sufficient to trigger the development of obliterative airway disease. Previous human and animal studies have suggested that MHC class I and class II antigens on bronchial epithelium (a "nonprofessional" APC) are sufficient to drive the development of OB.
8-11,19-21,23,26 In contrast, our data suggest that direct presentation by the bronchial epithelium or indirect presentation of donor MHC class I and class II antigens by recipient APCs is not sufficient to drive the development of obliterative airway disease.
Increasing evidence suggests that chronic rejection is mediated by persistent indirect presentation of allopeptides. Vella and coworkers
29 demonstrated that lymphocytes from renal transplant recipients with chronic rejection had a significant increase in proliferation to HLA peptides when compared with controls. Ciubotariu and associates
30 showed that the incidence of cardiac allograft vasculopathy was significantly higher in patients whose lymphocytes displayed persistent allopeptide reactivity. Last, our laboratory has previously shown that chimeric rat cardiac allografts bearing recipient-type APCs in a low responder strain combination survive long term but have severe allograft vasculopathy.
17
In our experiments, obliterative airway disease does not develop in chimeric tracheas in the absence of donor-type APCs, despite the presence of bronchial epithelial MHC class I and class II antigens. Therefore, we believe that obliterative airway disease is a model of acute rejection, rather than one of chronic rejection. It has been postulated that the vigor of acute allograft rejection is due to the "unnatural" stimulation of recipient T cells by donor-type APCs (direct allorecognition). The ability of graft-resident bone marrowderived APCs to induce allosensitization was first demonstrated by Steinmuller.
31 Subsequently, Lechler and Batchelor
14 showed that depletion of donor APCs led to prolonged allograft survival and that perioperative adoptive transfer of donor-type APCs restored acute rejection. Recently, our laboratory has reported that donor APC replacement in the same mouse strain combination alters the tempo and pathogenesis of cardiac allograft rejection.
17 One caveat is that obliterative airway disease might eventually develop in chimeric tracheas if subsequent time points were assessed. However, the inability in our laboratory to harvest intact chimeric specimens (previously irradiated) with consistently reliable architecture beyond 28 days makes it difficult to address this possibility experimentally.
Even though the fibroproliferative process in obliterative airway disease resembles that found in OB histologically, the murine heterotopic tracheal transplant model has several limitations. The lesions of OB are primarily found in the terminal airways in human beings, whereas the obliterative airway disease lesion occurs more proximally in the airway. In addition, the heterotopic tracheal transplant model depends on neovascularization, in contrast to immediately revascularized orthotopic lung transplant models. Although this system has been used as a model of OB, the failure of therapies that effectively prevent obliterative airway disease to make an impact on clinical OB further suggests that this model should be used in studies of acute rejection.
In summary, our study demonstrates that donor-type APCs are important in the development of obliterative airway disease in a heterotopic tracheal transplant model. Furthermore, donor APC expression of either MHC class I or class II antigens is sufficient to trigger the development of obliterative airway disease. In terms of clinical applicability, the concept of APC depletion to reduce immunogenicity of allografts has already led to clinical trials with anti-CD45 used in renal transplantation.
32,33 On the basis of these studies, we would speculate that APC depletion might reduce the incidence and severity of acute lung rejection.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Boehler and M. Estenne Post-transplant bronchiolitis obliterans Eur. Respir. J., December 1, 2003; 22(6): 1007 - 1018. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Chalermskulrat, I. P. Neuringer, W. J. Brickey, N. J. Felix, S. H. Randell, J. P. Ting, and R. M. Aris Hierarchical Contributions of Allorecognition Pathways in Chronic Lung Rejection Am. J. Respir. Crit. Care Med., April 1, 2003; 167(7): 999 - 1007. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |