INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Chronic rejection, characterized by migration and proliferation of mesenchymal cells in the vascular and airway lumen, is the main cause of morbidity and mortality among long-term survivors of cardiac and lung allograft recipients (1). Despite advances in understanding the pathobiology of chronic rejection, the inciting and effector mechanisms underlying this fibroproliferative disorder are incompletely known. Observations in experimental cardiac and tracheal allografts support the concept that the immunological pathway in the development of chronic rejection is mediated by a delayed-type hypersensitivity-like reaction, in which activated macrophages and CD4⁺ T cells have a major initiating role and the final effector molecules, cytokines and growth factors, are similar to those operating in other fibroproliferative disorders (2).

Previous investigations of the pathogenesis of chronic rejection have mainly focused on the role of acquired immunity, i.e., cell-mediated and humoral immunity. To date, there are no studies examining how natural immunity, i.e., complement (C) activation and phagocytosis, may regulate cellular signaling in chronic rejection. Complement is one of the major mediators of acute inflammatory responses such as hyperacute allograft and xenograft rejection causing thrombosis, edema, hemorrhages, and rapid graft loss (6). There are two pathways of complement activation, the classic and the alternative pathway (10). The classic pathway is activated by antigen-antibody complexes containing IgM or IgG. C1 binds to the Fc portion of antigen-bound antibodies, and becomes an enzyme that can split C4 and C2 to form C4b2a, the C3 convertase of the classic pathway (10). The alternative pathway is activated by covalent attachment to a surface via an intramolecular thioester bond of C3, and the attached C3 reacts with factors D and B to form C3bBb, the C3 convertase of the alternative pathway (10). The next step in both pathways is the binding of C3b to the C3 convertase, forming C5 convertase. Once C5 is cleaved, both pathways share the same terminal steps, i.e., the sequential binding of C6, C7, C8, and C9 to form C5b-9, a lipid-soluble pore structure called the membrane attact complex (MAC), resulting in osmotic lysis of cells (10).

Of endogenous regulatory proteins of complement, complement receptor type 1 (CR1) has the best inhibitory potential to block C3 and C5 convertases of both classic and alternative pathways, but its restriction to a few cell lines limits its function in vivo. This limitation was overcome by recombinant soluble complement receptor type 1 (sCR1), which had a complement inhibitory and antiinflammatory activity in a rat model of postischemic myocardial necrosis (11). sCR1 was demonstrated to prolong graft survival in models of hyperacute allograft (8) and xenograft (7, 9) rejection. In an unsensitized model of rat renal allograft rejection, sCR1 treatment partially inhibited vascular injury and leukocyte infiltration (12).

This study was undertaken to investigate whether complement is induced during the development of experimental obliterative bronchiolitis (OB), a manifestation of chronic rejection in rat tracheal allografts, and to investigate the biological role of complement activation in this fibroproliferative disorder by inhibition of complement with human recombinant sCR1 in this model.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Heterotopic Tracheal Transplantation Model and Study Design

Specific pathogen-free, 2- to 3-mo-old inbred male DA (AG-B4, RT1^a) and WF (AG-B2, RT1^u) rats (Laboratory Animal Center, University of Helsinki, Helsinki, Finland) weighing 200-300 g were used. The animals received humane care in compliance with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals prepared and formulated by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1985). Heterotopic tracheal transplantations were performed as described (3). In short, a 2- to 3-cm-long segment of donor trachea was removed, perfused with phosphate-buffered saline containing penicillin (10,000 IU/ml) and streptomycin (1,000 µg/ml), and stored in the same solution at 4° C until used as a transplant. After an upper abdominal midline incision was performed, the trachea was placed into the greater omentum under chloral hydrate anesthesia (240 mg/kg, intraperitoneal). Buprenorphine (Temgesic, 0.25 kg/kg, subcutaneous; Reckitt & Colman, Hull, UK) was used for postoperative pain relief. Allogeneic transplantation was performed from DA donor to WF recipient rats and syngeneic controls from DA to DA rats. Five groups were formed: (1) nontransplanted normal DA tracheas, (2) nontreated syngeneic controls, (3) nontreated allogeneic controls, (4) allografts treated with suboptimal dose of cyclosporine A (CsA) and vehicle, and (5) allografts treated with suboptimal dose of CsA and sCR1. Complement activation in the development of OB was characterized in all study groups. To investigate the pathophysiology of complement activation in OB, allograft recipients were given either sCR1 or vehicle under suboptimal CsA immunosuppression, used to prevent fulminant cellular immune response in the graft (3). The grafts were removed 3, 10, and 30 d after transplantation to determine histological changes and in vivo cell proliferation. A detailed immunohistochemical analysis was performed to examine how sCR1 modulates inflammatory cell infiltration and production of mediators of inflammation in tracheal allografts.

Drugs Used

Cyclosporine A (CsA; Sandimmune; Novartis, Basel, Switzerland) was dissolved in Intralipid (200 mg/kg; KabiVitrum, Stockholm, Sweden) to a final concentration of 1 mg/ml and was given subcutaneously at dosage of 1 mg/kg/d. sCR1 (Avant Immunotherapeutics, Needham, MA) was diluted to a concentration of 5.0 mg/ml in buffer containing mannitol (20 mg/ml) and was administered intraperitoneally at a dosage of 20 mg/kg/d, starting preoperatively and continuing for 7 d. The vehicle contained mannitol (20 mg/ml), monobasic potassium phosphate (1.09 mg/ml), dibasic potassium phosphate (1.44 mg/ml), and sodium chloride (1.46 mg/ml) and was administered similarly as sCR1.

Histological Evaluation

The grafted trachea was excised, embedded in Tissue-Tek (Miles, Elkhart, IN), snap-frozen in liquid nitrogen, and stored at 70° C until use. For histological evaluation, frozen sections were stained with Mayer's hematoxylin-eosin (H&E). The histological changes in respiratory epithelium were evaluated as a percentage of normal respiratory, abnormal cuboidal, and squamous epithelium; necrotic epithelium; or loss of epithelium. The lumenal occlusion was evaluated as reduction of lumenal area using an ocular grid. The intensity of neutrophil infiltration was scored from 0 to 3 from H&E-stained sections. All analyses were done in a blinded review by two observers, and the scores of these two observers were highly correlated (r² = 0.98). The degree of airway wall inflammation, the cell types in the myofibroproliferative lesion, and the number of proliferating inflammatory and myofibroproliferative cells were analyzed by immunohistochemistry.

Immunostaining

Serial frozen sections (4-6 µm) were air dried on Silane-coated slides, fixed in acetone at 20° C for 20 min, and stored at 20° C until use. Before immunostaining, the slides were refixed with chloroform and then air dried. After incubation with appropriate 1.5% nonimmune serum (Vector Laboratories, Burlingame, CA) for monoclonal antibodies or polyclonal antibodies, frozen sections were incubated with a mouse monoclonal antibody at room temperature for 30 to 60 min, or with a rabbit or goat polyclonal antibody at 4° C for 12 h. The primary antibodies were diluted in phosphate-buffered saline (PBS) with 1% bovine serum albumin (BSA) and appropriate 3% nonimmune serum. With intervening washes in TRIS-buffered saline, the slides were immersed in biotinylated horse anti-mouse, goat anti-rabbit, or rabbit anti-goat rat-absorbed antibodies at room temperature for 30 min followed by avidin-biotinylated horseradish complex (Vectastain Elite ABC kit; Vector Laboratories) in PBS at room temperature for 30 min. The reaction was revealed by chromogen 3-amino-9-ethylcarbazole (AEC; Sigma, St. Louis, MO) containing 0.1% hydrogen peroxidase, yielding a brown-red reaction product. The specimens were counterstained with hematoxylin and coverslips were aquamounted (Aquamount; BDH, Poole, UK).

Antibodies Used

Complement activation was determined with mouse monoclonal antibodies to rat C3 (ED11, diluted 1:200; Serotec, Oxford, UK), human and rat C5b-9 (aE11, diluted 1:50; Dako A/S, Glostrup, Denmark), rat IgM (MARM-4; diluted 1:1,000; Serotec), and rat IgG₁ heavy chain (MARG1-2, diluted 1:100; Serotec). Other mouse monoclonal antibodies were directed to rat macrophages (ED1; Serotec), activated macrophages (ED3; Serotec), CD4⁺ (W3/25; Sera-Lab, Sussex, UK) and CD8⁺ (OX8; Sera-Lab) T cell subsets, CD45⁺ B cells (OX-33; Serotec), MHC class II common determinant (OX6; Sera-Lab), IL-2R (PharMingen, San Diego, CA), IFN- (CY-047; Innogenetics, Zwijndrecht, Belgium), TNF- (CY-051; Innogenetics), IL-4 (MRC OX-81; Serotec), human IL-8, cross-reactive with rat IL-8-like molecule (13) (1:20; Genzyme, Cambridge, MA), and intercellular adhesion molecule 1 (ICAM-1) (CD54) (1A29, diluted 1:100; Seikagaku, Tokyo, Japan). Rabbit polyclonal antibodies were directed to mouse/rat/human P-selectin (09361, diluted 1:100; PharMingen), IL-1 (LP-712; Genzyme), rat monocyte chemoatractant protein 1 (MCP-1) (AAR12Z, diluted 1:50; Serotec), rat macrophage inflammatory protein 2 (MIP-2) (AAR11Z, diluted 1:20; Serotec), human and rat PDGF-AA (ZP-214; Genzyme), human and rat PDGF-BB (ZP-215; Genzyme), and human pan-TGF- (AB-100-NA; R&D Systems, Minneapolis, MN). Goat polyclonal antibodies were directed to IL-2 (SC-1786; Santa Cruz Biotechnology, Santa Cruz, CA) and IL-10 (AB-417-NA; R&D Systems).

Specificity Controls of Immunostainings

Specificity controls were performed by using the same immunoglobulin concentration of species and isotype-matched antibodies; mouse monoclonal IgG₁ (X931; Dako A/S) and rabbit polyclonal immunoglobulin fraction (X936; Dako A/S) for monoclonal and polyclonal antibodies, respectively. Additional specificity controls involved the use of a working dilution of the polyclonal antibody after overnight incubation with a 10 to 20 M excess of recombinant cytokine or peptide. None of these control stainings showed any immunoreactivity.

Quantification of Immunohis

The immunohistochemical analysis was done in a blind review by two observers and the score assigned was determined by consensus. The intensity of the staining was scored from 0 to 3 as follows: 0, no visible staining; 1, few cells with faint staining; 2, moderate intensity with multifocal staining; and 3, intense diffuse staining of the cells analyzed. Positive staining for IL-2, IL-4, and IL-10 was scored as the number of positive cells in cross-section.

In Vivo Labeling for Cell Proliferation

All recipients were injected intravenously with 400 µl of a concentrated solution of 5-bromo-2'-deoxyuridine (BrdU, 3 mg/ml) and 5-fluoro-2'-deoxyuridine (0.3 mg/ml) (Zymed Laboratories, San Francisco, CA) 3 h before sacrifice. Cell proliferation in frozen sections was revealed by an IgG₁ mouse monoclonal antibody to BrdU (M744, diluted 1:20; Dako A/S) and the Vectastain Elite ABC kit method as described above. Before staining, the frozen sections were fixed with buffered formalin for 15 min. After a 10-min wash in PBS, the sections were microwave treated (500 W) in 0.1 M citrate buffer, pH 6, for 5 min to break the double-stranded DNA, followed by another 10-min wash in PBS. Cell proliferation was measured by counting the number of labeled nuclei separately in the airway wall, i.e., adventitia and submucosa and in the myofibroproliferative lesion. The scores are given as BrdU⁺ cells per cross-section.

Statistical Analyses

All data are expressed as means ± SEM. The nonparametric Mann- Whitney U test (Statview 512+ program: Brain Power, Calabasas, CA) was used to evaluate the significances. p Values of < 0.05 were regarded as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histology

New nonimmunosuppressed syngeneic and allogeneic grafts were done to validate the model to those previously described in detail (3, 4), and thus the histological results of these controls are given here only briefly. Three days after transplantation the respiratory epithelium of syngeneic grafts was slightly damaged because of ischemic injury before revascularization of the graft. Thirty days after transplantation the respiratory epithelium of syngeneic grafts was fully recovered and the grafts were filled with mucus, a sign of normal function of the epithelium. In syngeneic grafts, no neutrophilic cell infiltration was observed, no myofibroproliferation occurred, and the overall histology resembled that of nontransplanted tracheas. In nontreated allografts, there was a progressive loss of respiratory epithelium and infiltration of neutrophils into the epithelium and submucosa. Airway wall inflammation peaked 10 d after tranplantation, subsiding thereafter. In nontreated allografts, the lumenal surface was already reduced by proliferation of -smooth muscle actin immunoreactive cells at 10 d, and at 30 d intense myofibroproliferation nearly totally occluded the lumen. Suboptimal CsA treatment did not prevent progressive loss of normal respiratory epithelium and infiltration of neutrophils into the subepithelial space, or myofibroproliferation and airway occlusion. Treatment with sCR1 significantly reduced epithelial necrosis (Figures 1A and 2) and neutrophil infiltration (Table 1 and Figure 2) at 3 d, and the development of airway occlusion at 3, 10, and 30 d (Figures 1B and 3).

View larger version (26K): [in this window] [in a new window]   Figure 1.   Histological features of CsA + vehicle-treated and CsA + sCR1-treated allografts. (A) The percentage of epithelial necrosis, (B) degree of airway occlusion, (C ) airway wall inflammatory cell proliferation, and (D) myofibroproliferation. Airway wall inflammatory cell proliferation and myofibroproliferation were determined by BrdU immunoreactivity and the number of BrdU+ cells per cross-section was counted. Data are given as means ± SEM. *p < 0.05 and p < 0.01 by Mann-Whitney U test, when compared with CsA + vehicle-treated allografts.

View larger version (109K): [in this window] [in a new window]   Figure 2.   Photomicrographs of graft epithelium, airway wall cell proliferation, and myofibroproliferation 3 d after transplantation. (A) Epithelial necrosis and infiltration of neutrophils into subepithelial space and submucosa of allografts treated with CsA + vehicle. (C ) Well-preserved respiratory epithelium and no neutrophil infiltrates in CsA + sCR1-treated allografts (H&E staining; original magnification, ×200). (B) BrdU incorporation into proliferating myofibroblast-like cells of allografts treated with CsA + vehicle has already occurred by 3 d, whereas in (D) CsA + sCR1-treated allografts, myofibroproliferation was mild and proliferation occurred mainly in the well-preserved respiratory epithelium. Immunostainings with IgG1 mouse monoclonal antibody to BrdU; original magnification, ×200.

Effect of sCR1 on Graft Complement and Immunoglobulin Expression

As shown in Table 2 and Figure 4, in nontransplanted DA tracheas and syngeneic grafts, the expression of C3, C5b-9, IgM, and IgG was nonexistent to mild. In nontreated allografts and allografts treated with a suboptimal dose of CsA, C3 expression was mainly located in the subepithelial areas of the airway wall, C5b-9 expression was detected mainly around mononuclear inflammatory cell infiltrates and in vascular structures, whereas IgM and IgG expression was diffuse in the airway wall. Allograft C3, C5b-9, and IgG, but not IgM, expression was increased compared with syngeneic grafts. Treatment with sCR1 significantly inhibited intragraft C5b-9 and IgG depositions 3 d after transplantation (Table 2 and Figure 4).

View larger version (148K): [in this window] [in a new window]   Figure 4.   Photomicrographs of C3, C5b-9, and IgG expression in a syngeneic graft, and nontreated, CsA + vehicle-treated, and CsA + sCR1-treated allografts 3 d after transplantation (original magnification, ×100; insets, ×1,000).

Effect of sCR1 on In Vivo Cell Proliferation

In allografts treated with a suboptimal dose of CsA, a strong and early airway wall inflammatory cell proliferation occurred at 3 and 10 d, followed by a myofibroproliferative response and development of tracheal occlusion at 30 d. Treatment with sCR1 significantly inhibited airway wall inflammatory cell proliferation at 10 d and a myofibroproliferation at 30 d (Figures 1C, 1D, and 2).

Effect of sCR1 on Airway Wall Inflammation and Immune Activation

In allografts treated with a suboptimal dose of CsA, a prominent airway wall infiltration of ED1⁺ and ED3⁺ macrophages and CD4⁺ T cells, and to lesser extent of CD8⁺ T cells and CD45⁺ B cells, was observed 3 and 10 d after transplantation, subsiding thereafter (Table 1). Many of the inflammatory cells expressed IL-2R and MHC class II. Also, moderate epithelial MHC class II expression was observed. Treatment with sCR1 did not downregulate the number of graft-infiltrating inflammatory cell subsets or alloimmune activation determined as IL-2R or MHC class II expression (Table 1).

Effect of sCR1 on Chemokine and Adhesion Molecule Expression

In allografts treated with a suboptimal dose of CsA, mononuclear inflammatory cells expressed MIP-2 and MCP-1 with moderate intensity in the airway wall and early in the myofibroproliferative lesion. Moderate to strong IL-8-like molecule expression was detected in subepithelial mononuclear inflammatory cells and neutrophils (Table 1). Mild ICAM-1 expression was located in mononuclear inflammatory cells and capillaries, and P-selectin was located only in capillaries of the airway wall and myofibroproliferative lesion. Treatment with sCR1 significantly downregulated IL-8-like molecule expression at 10 d, and ICAM-1 expression at 3 d, but did not affect MIP-2, MCP-1, or P-selectin expression (Table 1).

Effect of sCR1 on Cytokine and Growth Factor Expression

In allografts treated with a suboptimal dose of CsA, mild to moderate IL-1, IFN-, TNF-, and IL-2 expression was observed in mononuclear inflammatory cells, whereas mild PDGF-AA, PDGF-BB, and TGF- expression was found also in epithelial cells and smooth muscle cells. sCR1 treatment significantly reduced mononuclear inflammatory cell TNF- and IL-2 expression at 3 d, and upregulated Th2 cytokine IL-10 expression 3 and 10 d after transplantation (Table 1 and Figure 5). Treatment with sCR1 also significantly reduced the PDGF-AA expression of myofibroblast-like cells in the myofibroproliferative lesion at 10 d, but did not alter the PDGF-BB and TGF- expression profiles (Table 1).

View larger version (95K): [in this window] [in a new window]   Figure 5.   Photomicrographs of IL-8-like molecule, ICAM-1, TNF-, IL-2, and IL-10 stainings of CsA + vehicle-treated and CsA + sCR1-treated allografts 3 d after transplantation. Treatment with sCR1 inhibited IL-8-like molecule, a neutrophil chemokine, and ICAM-1 expression. In addition, sCR1 treatment inhibited TNF- and IL-2 expression, whereas IL-10 expression was upregulated, indicating a switch from Th1-type response toward Th2-type immune response (original magnification, ×100; insets, ×1,000).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The immunobiology of OB as a manifestation of chronic rejection remains ill defined in humans. Observations in rodents demonstrate that heterotopic tracheal allografts develop epithelial damage, low-level airway wall inflammation, subsequent proliferation of -smooth muscle actin-immunoreactive myofibroblast-like cells, and gradual occlusion of the airway lumen, similar to those seen in small and large airways of human lung allografts with OB (2, 4, 14). At the early phase of OB, the respiratory epithelium of tracheal allografts strongly expresses MHC class II, enabling a direct presentation of foreign antigens to alloreactive T cells (3). In addition, graft-infiltrating host professional antigen-presenting cells may process donor antigens for indirect T and B cell activation (3, 15). The recruitement of CD4⁺ and CD8⁺ T cells and macrophages into the airway wall induces prominent local Th1 cell and macrophage-derived cytokine and growth factor production leading to myofibroproliferation and OB. On the other hand, in syngeneic grafts and allografts sufficiently immunosuppressed with CsA, obliterative changes do not develop, indicating the proximal role of alloimmune response in OB development (3).

We believe that epithelial injury initiates the T and B cell-driven alloimmune response leading to uncontrolled myofibroproliferation. We hypothesized that complement activation may contribute to this injury. Complement activation is an important mediator in many acute inflammatory responses, including those that are clearly T cell driven (6). In the context of transplantation, previous studies of complement have focused on antibody-associated rejection in presensitized patients and xenotransplantation models. The characterization of complement activation has not previously been described in the pathogenesis of chronic rejection. The biological role of complement activation in experimental OB as a manifestation of chronic rejection was tested by inhibition of the complement activation cascade by sCR1, known to effectively protect against complement-mediated tissue damage in ischemia/reperfusion (11, 16), immune complex (17), thermally and cobra venom factor (17) induced injury models.

Complement activation may occur early in the development of experimental OB via the classic pathway by alloantibodies, and via the alternative and classic pathways by ischemia-reperfusion injury (6). In this article we demonstrate that complement activation has a destructive role in OB development. In syngeneic grafts, complement and immunoglobulin deposits were nonexistent to mild and the expression resembled that of normal nontransplanted tracheas, indicating that ischemia was not responsible for complement activation in this model. In nontreated allografts, an increase in C3 and C5b-9 expression was demonstrated during progressive loss of respiratory epithelium and airway occlusion. Suboptimal CsA treatment did not significantly downregulate this expression. In contrast, only a 7-d treatment with sCR1 inhibited the early C5b-9 and IgG deposits. Treatment with sCR1 resulted in downregulation of epithelial injury, intragraft neutrophil infiltration, inflammatory cell and myofibroproblast proliferation, and the development of tracheal allograft occlusion.

There are several mechanisms by which complement components may facilitate rejection. Complement may assist in antigen presentation to T cells (18), enhance alloantigen and IL-2-induced T cell proliferation (19, 20), and induce B cell proliferation and antibody production (21). Complement is chemotactic to neutrophils (24) and upregulates ICAM-1 and E- and P-selectin expression (25, 26) as well as endothelial IL-8 production (27), thereby enhancing leukocyte chemotaxis and adhesion. Complement also induces expression of cytokines IL-1 and TNF- (24, 28, 29) and chemokines MCP, MIP-1, and IL-8 by monocytes and macrophages (30, 31). These proinflammatory factors lead to autocrine stimulation of macrophages, further upregulate adhesion molecule expression, and enhance leukocyte chemotaxis and tissue damage, which induce local reparative processes and mesenchymal cell proliferation.

In this study, sCR1 treatment did not affect the magnitude of intragraft infiltration of macrophages, T cells, or B cells, or the level of immune activation of these cells. Treatment with sCR1 modulated the immune response toward Th2 cytokine IL-10 expression, which has been linked with prolonged graft survival and tolerance induction in rat renal allografts (32) and mice cardiac allografts (33), as well as inhibition of experimental OB (4). sCR1 treatment decreased intragraft neutrophil infiltration, mediated possibly by downregulation of IL-8-like molecule production and ICAM-1 expression (34, 35). Bronchoalveolar lavage neutrophilia and IL-8 production have been associated with OB in humans (36).

In conclusion, these data demonstrate that inhibition of complement activation with sCR1 attenuates the development of epithelial injury and airway occlusion, the hallmarks of OB, suggesting a role for complement in the pathogenesis of OB and chronic rejection. Inhibition of experimental OB with sCR1 was associated with downregulation in IL-8-like molecule and ICAM-1 expression and neutrophil infiltration, indicating a role for neutrophils in the early phase of the development of experimental OB. In addition, blockade of complement activation modulates the alloimmune response toward Th2 cytokines, with increased IL-10 expression, which may have an antiproliferative role in fibroproliferative disorders.