Published ahead of print on August 17, 2006, doi:10.1164/rccm.200601-044OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200601-044OC
Role of Platelet-derived Growth Factor and Vascular Endothelial Growth Factor in Obliterative Airway DiseaseCardiopulmonary Research Group, Transplantation Laboratory, University of Helsinki/Helsinki University Central Hospital; Division of Nephrology, Department of Medicine, and Department of Cardiothoracic Surgery, Helsinki University Central Hospital, Helsinki, Finland; and Novartis, Basel, Switzerland Correspondence and requests for reprints should be addressed to Jussi Tikkanen, M.D., Ph.D., Cardiopulmonary Research Group, Transplantation Laboratory, University of Helsinki and Helsinki University Central Hospital, P.O. Box 21 (Haartmaninkatu 3), FIN-00014 Helsinki, Finland. E-mail: jussi.tikkanen{at}helsinki.fi
Rationale: Platelet-derived growth factor (PDGF) is an important smooth muscle cell mitogen, and vascular endothelial growth factor (VEGF) is a known angiogenic and proinflammatory growth factor. We hypothesized that specific therapy aimed at these growth factors might inhibit the development of experimental obliterative airway disease (OAD). Methods: In fully mismatched rat tracheal allografts, we used imatinib and PTK/ZK, either alone or in combination, to block PDGF and VEGF receptor protein tyrosine kinase (RTK) action, respectively. Prophylaxis was initiated at the time of transplantation. Early treatment was commenced on Day 7 during the inflammatory phase and late treatment on Day 14 during the fibroproliferative phase of OAD. No immunosuppression was administered. Measurements and Main Results: Prophylaxis with either PTK/ZK or imatinib alone significantly reduced OAD, and combined prophylaxis completely prevented its development. Early treatment with PTK/ZK and imatinib also effectively reduced the development of OAD. Late treatment failed to show significant efficacy. Blocking VEGF RTK action with PTK/ZK reduced the activation of allograft blood vessels and the number of lymph vessels in the allograft airway wall, and significantly diminished allograft inflammation, whereas PDGF blockade with imatinib inhibited the growth of smooth muscle cells in the proliferating lesion. Conclusions: Combined prophylactic PDGF and VEGF RTK blockade completely prevents the development of OAD. Also, when early treatment with PTK/ZK and imatinib is commenced during the inflammatory phase of OAD development, it significantly attenuates the development of tracheal occlusion, suggesting that these drugs could potentially be used to treat bronchiolitis obliterans syndrome in its early phase.
Key Words: lung transplantation • obliterative bronchiolitis • chronic rejection • angiogenic growth factors
The leading cause of lung allograft loss is obliterative bronchiolitis (OB). After lung transplantation, ischemia-reperfusion injury and acute alloimmune response cause damage to the lung allograft, leading to a response-to-injury reaction (18, 24). Accumulating evidence underlines the importance of growth factors in the development of the fibroproliferative lesion that gradually occludes the airways (2, 10, 11). Currently, there is no treatment available for OB, and immunosuppressive agents have little, if any, effect on the progress of this disorder. Recent studies focusing on the pathobiology of OB have suggested that its development can be divided into different phases: an early inflammatory phase, where both nonalloimmune and alloimmune activation result in strong immune activation that damages the allograft, and that induces the reparatory fibroproliferative phase, leading to scarring of the airways (18). It seems that, during the fibroproliferative phase of OB development, the growth of the fibroproliferative lesion is no longer responsive to augmented immunosuppression. Therefore, new therapeutic approaches for the prevention and treatment of OB are warranted. Advances both in cancer research and rational drug design have led to the emergence of a novel drug family consisting of receptor protein tyrosine kinase (RTK) inhibitors of a variety of hormone and growth factor families. Prevention of RTK signaling may be attained by blocking ligand binding to RTK with specific binding proteins or antibodies, inhibition of RTK expression using antisense oligonucleotides, or inhibiting RTK activity using small molecule inhibitors. Clinical trials have focused on the use of small molecule inhibitors, with promising results in advanced cancer (12, 21). However, studies with RTK inhibitors have also revealed that there is significant intracellular cross-talk between different RTKs, suggesting that inhibition of a single RTK may also alter the activity of other RTKs (16). On the other hand, blocking one single RTK pathway may not suffice to completely block the biological activity of a receptor, because other RTKs may serve as secondary signal mediators and may offer a route to circumvent the blockade of a single RTK.
This study used two different RTK inhibitors, PTK/ZK and imatinib, either alone or in combination to prevent or treat rat tracheal allograft obliterative airway disease (OAD). PTK/ZK inhibits vascular endothelial growth factor (VEGF) RTK activity but may also inhibit other class III kinases, such as platelet-derived growth factor receptor
Tracheal Transplantations Specific pathogen–free inbred male Dark Agouti (DA) (AG-B4, RT1a) and Wistar Furth (WF) (AG-B2, RT1u) rats weighing 200 to 300 g and of 2 to 3 mo of age (Harlan, Horst, The Netherlands) were used. Syngeneic tracheal grafts were transplanted heterotopically from DA donors to DA recipients and allografts from fully mismatched DA donors to WF recipients into the recipient's greater omentum as described (22). The rats were killed 10 and 30 d after transplantation for immunohistochemical and histologic analyses. Permission for animal experimentation was obtained from the State Provincial Office of Southern Finland. The rats received care in compliance with the "Guide for the Care and Use of Laboratory Animals" (18).
Drug Regimens For the prophylaxis experiment, tracheal allograft recipients received PTK/ZK, imatinib, or a combination of these (n = 6–10/group) starting on the day of transplantation, whereas control animals were treated with polyethylene glycol. In the treatment experiments, the drug treatment was initiated at 7 and 14 d after transplantation for the early (inflammatory phase) and late (fibroproliferative phase) treatment groups, respectively.
Histology
Immunohistochemical and Immunofluorescence Staining
Statistical Analyses
Tracheal Allograft Expression of VEGF and PDGF Ligands and Receptors There was abundant expression of VEGF in allografts 10 d after transplantation. Allografts expressed VEGF strongly in the inflammatory cells of the airway wall and blood vessels but less so in the epithelium, which was already undergoing necrosis at 10 d. Moderate VEGFR-1 expression was observed mainly in the medial SMC-like cells of allograft arterioles and, to lesser extent, in endothelial and mononuclear inflammatory cells. VEGFR-2 expression was localized to endothelial cells of small blood vessels (Figures 1A–1C).
PDGF-A expression was localized to the fibroproliferative lesion, SMCs, capillary endothelium, and inflammatory cells (Figure 1D). PDGFR-  was observed in medial cells of arterioles as well as in SMC-like cells of the myofibroproliferative lesion (Figure 1E). PDGF-B ligand immunoreactivity was seen mainly in airway wall SMCs and inflammatory cells (Figure 1F). PDGFR- immunoreactivity was relatively weak and concentrated in allograft blood vessels (Figure 1G). Neither PTK/ZK nor imatinib prophylaxis had a significant effect on allograft protein expression of VEGF, PDGF-A, or PDGF-B at 10 or 30 d according to immunohistochemical analyses (data not shown).
Effect of VEGF and PDGF Receptor Tyrosine Kinase Inhibition on Allograft Histology
Early treatment. At 7 d, untreated allografts showed marked epithelial necrosis (29 ± 6%). The remaining epithelium had lost its normal ciliated structure and was morphologically cuboidal or squamous. However, only low-level luminal occlusion (9 ± 5%) was observed. When early treatment was initiated on Day 7, both PTK/ZK and imatinib resulted in a 30 to 40% reduction in luminal occlusion at 30 d but only imatinib treatment achieved statistical significance. The combination of these drugs resulted in a more prominent inhibition of luminal occlusion than either drug alone. None of the regimens prevented total epithelial loss from developing (Figure 2).
Late treatment.
Effect of Receptor Tyrosine Kinase Inhibition on Allograft Inflammation
Effect of PTK/ZK and Imatinib on Cell Proliferation We measured cell proliferation in the allograft using the cell proliferation marker Ki-67. Both PTK/ZK and imatinib significantly reduced cell proliferation in the proliferative lesion and subepithelial space of the allografts at 10 d (Figure 3D). Using immunofluorescence double-staining, we observed that Ki-67–positive cells expressed either -SMC actin or CD45, indicating that the proliferating cells were either SMCs or inflammatory cells.
Effect of PTK/ZK and Imatinib on Allograft Angiogenesis and Lymphangiogenesis
Lymphangiogenesis was investigated by immunostaining of the lymph vessel marker LYVE-1. In normal trachea, lymph vessels were restricted mainly to the subepithelial space, whereas after syngeneic transplantation, a mild increase in LYVE-1–positive lymph vessels was observed at 10 d. At 30 d, LYVE-1 expression was very similar in the syngeneic grafts compared with that of normal trachea. Allografts showed a distinctively different profile, with a dramatic increase in the number of lymph vessels in the airway wall at 10 d. At 30 d, the number of LYVE-1–positive vessels was still elevated and LYVE-1–positive lymph vessels were evident in the fibroproliferative lesion occluding the tracheal lumen. PTK, but not imatinib prophylaxis, halved the number of LYVE-1–positive lymph vessels in tracheal allografts 10 d after transplantation compared with vehicle-treated controls (Figure 4D).
In the tracheal transplantation model, the epithelium of syngeneic grafts suffers minor damage but recovers after establishing an adequate blood supply from the microcirculation of the surrounding tissue (11). The tracheal lumen remains completely open and the trachea is lined with normal, mucus-secreting epithelium at 30 d after transplantation. In untreated allografts, the epithelium sustains progressive damage leading to nearly total epithelial necrosis 10 d after transplantation (7, 14, 22). Coinciding with the strong Th1-dominated alloimmune response and increasing epithelial loss, prominent expression of growth factors, such as PDGF, VEGF, endothelin-1, transforming growth factor- , and fibroblast growth factor, is observed together with intense SMC proliferation (2, 11, 14, 23) and formation of new blood and lymph vessels mainly into the surrounding airway wall. The process culminates in the development of a fibroproliferative lesion consisting mainly of -SMC actin–positive cells and extracellular matrix. If the epithelium has undergone necrosis, initiation of conventional immunosuppressive therapy does not influence the development of OAD (1). Furthermore, if untreated allografts are removed on Day 14 and retransplanted into syngeneic recipients, OAD develops in the absence of alloimmune responses. However, if the same procedure is performed on Day 7 before loss of epithelium, OAD does not develop (13). This study sought to prevent obliterative changes without using any background immunosuppressive medication by combining two distinctly acting RTK inhibitors. Both PTK/ZK and imatinib attenuated the development of tracheal occlusion, but dual inhibition of VEGF and PDGF receptor protein tyrosine kinase activity totally prevented the development of the fibroproliferative lesion, and thereby tracheal allograft occlusion, despite total necrosis of the epithelium, underlining the potency of this combination in the present model. The result suggests distinct mechanisms of action for these two drugs. VEGF ligand and receptor expression is prominent during the development of OAD (15). PTK/ZK-mediated VEGF RTK inhibition did not inhibit tracheal allograft vascularization (RECA-1–positive vessels) but inhibited the activation of allograft blood vessels depicted as reduced HMW-MAA and VCAM-1 expression. Therefore, VEGF regulates vascular remodeling, and possibly permeability, but is not rate-limiting for angiogenesis per se. Supporting this, Belperio and coworkers showed recently that anti-VEGF antibodies do not inhibit angiogenesis in murine trachea, whereas inhibition of the chemokine CXCR2/CXCR2 ligand interaction reduces angiogenesis (4). Thus, it is unlikely that the drugs used in this study would induce major anatomic alterations in the tracheal allografts but that their effect would be mediated through interruption of the alloimmune activation and SMC proliferation. PTK/ZK prophylaxis also dramatically reduced the number of LYVE-1–positive lymph vessels, thus possibly interfering with the formation of the lymphatic network required for adequate alloantigen presentation. PTK/ZK inhibits VEGFR-3 RTK activity to some extent and may directly reduce lymphangiogenesis. Also, the antiinflammatory effects of PTK/ZK may have also led to reduced lymph vessel formation (25). Although VEGF mediates its mitogenic, angiogenic, and permeability-enhancing effects through VEGFR-2 (9, 20), PTK/ZK treatment may also directly reduce the number of allograft-infiltrating cells via inhibition of VEGFR-1–mediated macrophage chemotaxis (3). Therefore, the beneficial effects of PTK/ZK prophylaxis and early treatment initiated on Day 7 during intense alloimmune activation are probably related to the antiinflammatory effects of VEGF RTK inhibition. These seem to include inhibition of vascular activation but not angiogenesis itself, reduced lymphangiogenesis, and possibly direct inhibition of macrophage/monocyte chemotaxis.
We have previously shown that PDGF-A and PDGF-R In this study, no base immunosuppression was used. Despite this, both PTK/ZK and imatinib prophylaxis initiated on the day of transplantation were effective in inhibiting the development of OAD, and their combination totally prevented the fibroproliferative lesion from developing. This finding underlines the potency of the two drugs in this disease process. When early treatment of PTK/ZK and imatinib was initiated on Day 7, during intense alloimmune activation but before total loss of epithelium, a clear beneficial effect could still be observed. However, late treatment commencing on Day 14 failed to prevent OAD. At this time point, one sees near-total epithelial necrosis and severe allograft injury and it seems that the fibroproliferation cannot be reversed even by intense treatment as in our study or by removal of the alloimmune environment (13). In the attempt to extrapolate the findings of this study to the clinical situation, one has to take into account the limitations of the tracheal allograft model. First of all, the anatomy of the trachea is considerably different from that of bronchioli. The trachea is surrounded by cartilage not seen in bronchioli. Furthermore, the obliterative changes affecting bronchioles do not extend to large airways in humans. In addition, the tracheal allograft is not vascularized, which makes the interpretation of findings related to ischemia difficult. The tracheal allograft has no airflow and is not in contact with foreign pathogens. The trachea contains less lymphoid tissue than lung allografts and may reduce the impact of direct allorecognition in this model. Finally, in our model, the obliterative lesion develops in 1 mo compared with years in lung transplant patients. However, the obliterative lesion seen in tracheal allografts is similar to that seen in bronchioles in humans, and our tracheal allograft model thus forms a reproducible and simple model for investigation of the pathogenesis of OAD. This study shows that both PTK/ZK and imatinib effectively inhibit the development of experimental OAD through distinct growth factor pathways. Importantly, the combination of these drugs nearly completely prevents tracheal occlusion even when drug treatment is initiated during the inflammatory phase of OAD development. The results suggest a potential therapeutic role for RTK inhibitors in the prevention and treatment of bronchiolitis obliterans syndrome in the clinic.
The authors thank Mrs. Eeva Rouvinen, R.N., for her excellent technical assistance and Mrs. Leena Saraste for her help in preparing the manuscript. They also thank Dr. Elisabeth Buchdunger (Novartis) for her valuable expertise on the subject.
Supported by grants from the Helsinki University Central Hospital Research Funds, the Sigrid Juselius Foundation, the Finnish Life and Pension Insurance companies, the Academy of Finland, Finska Läkaresällskapet, the Jalmari and Rauha Ahokas Foundation, the Finnish Pulmonary Association, the Emil Aaltonen Foundation, the Farmos Research Foundation, and the University of Helsinki. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200601-044OC on August 17, 2006 Conflict of Interest Statement: J.M.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. M.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. A.I.N. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.W. is a full-time employee of Novartis and supplied the compounds used in the study and gave professional input into the study and manuscript. P.K.K does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. K.B.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form January 11, 2006; accepted in final form August 24, 2006
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
p < 0.01 compared with vehicle-treated controls. Statistical analyses were performed using analysis of variance (ANOVA) and Bonferroni correction.
