Journal Logo

Original Basic Science—General

The Lymphatic Phenotype of Lung Allografts in Patients With Bronchiolitis Obliterans Syndrome and Restrictive Allograft Syndrome

Traxler, Denise; Schweiger, Thomas MD; Schwarz, Stefan MD; Schuster, Magdalena Maria; Jaksch, Peter MD; Lang, Gyoergy MD; Birner, Peter MD, MSc; Klepetko, Walter MD; Ankersmit, Hendrik Jan MD; Hoetzenecker, Konrad MD, PhD

Author Information
doi: 10.1097/TP.0000000000001263

Lung transplantation currently constitutes the only therapeutic option for end-stage pulmonary diseases. Even though outcome after lung transplantation has significantly improved over the past decades, long-term survival is still worse than in other solid organ transplants.1,2 The major obstacle of long-term survival is chronic lung allograft dysfunction (CLAD). Bronchiolitis obliterans syndrome (BOS) and restrictive allograft syndrome (RAS) represent the most important forms of CLAD.3

Bronchiolitis obliterans syndrome is histologically characterised by obliterative bronchiolitis (OB). Obliterative bronchiolitis is a fibrotic/inflammatory process affecting small noncartilaginous airways leading to either total or subtotal occlusion of bronchioles.2 Additionally to fibrosis of the subepithelial region, atrophy of smooth muscle and destruction of the elastic parts of the airway wall is described.4 Nevertheless, transbronchial biopsy is not a sufficient diagnostic tool as OB is distributed irregularly and false negative results may occur.5 A common clinical characteristic of BOS is the deterioration of pulmonary function, especially a decrease in the forced expiratory volume in 1 second unexplained by other confounding factors such as acute rejection or infection. A classification system reflecting the extent of decrease of lung function has been proposed by the International Society of Heart and Lung Transplantation.6

Recently, it has been proposed that RAS should be considered another subtype of CLAD that contributes to about 25% to 35% of CLAD. Restrictive allograft syndrome represents restrictive functional changes and is defined as a persistent decline in total lung capacity additionally to a decline in forced expiratory volume in 1 second. It is histologically characterised by diffuse alveolar damage followed by fibroelastosis in the alveolar interstitium, interlobular septa, and visceral pleura and can also contain scattered OB lesions.3,7,8

The pathogenesis of both BOS and RAS is still obscure. In BOS, a strong association with recurrent infections on the one hand and episodes of acute rejections on the other hand indicate an important role of alloimmune T cell reactivity in the disease process.5 The development of BOS is characterized by 2 stages: first persistent alloimmune injury occurs, which is then followed by chronic airway inflammation.9 Several risk factors contributing to RAS have been described, such as airway neutrophilia, viral infection, and previous episodes of acute rejection.3,10 Currently, it is not clear if underlying mechanisms of BOS may also influence development of RAS. Specific pathophysiological mechanisms, which determine the focus of inflammation and fibrosis in BOS and RAS have not been identified.3

Lymphatic vessels (LVs) are a crucial factor of the immune defence. They serve as collecting vessels for extravasated leukocytes, activated antigen-presenting cells and soluble antigens. Another important function of LVs is the regulation of fluid homeostasis.11 The growth of LVs is promoted by vascular endothelial growth factor C (VEGF-C) and VEGF-D via VEGF receptor-3 (VEGFR-3). VEGFR-3 is located on lymphatic endothelial cells, also under inflammatory conditions.12 Increased lymphangiogenesis may act as a compensatory mechanism in inflammation to clear inflammatory infiltrates and reduce tissue oedema.11 However, in contrast to acute inflammation, antigens are not removed after transplantation, which leads to persistent antigen presentation.

In previous research a role for the LV system in rejection of solid organ transplants has been proposed. In human kidney transplantation Kerjaschki et al13 demonstrated the presence of LVs in kidney grafts and an increase in regions with nodular infiltrates. Additionally, in almost all patients with chronic graft rejection, LV density (LVD) was increased. In a rat renal transplant model of chronic rejection high LVD significantly correlated with poor graft outcome.14 In rat cardiac allografts, chronic rejection led to a doubling of the myocardial LVD.15 Additionally, an increase in LVD has been proposed in acute rejection in kidney, liver, and heart transplantation.16–18 Dashkevich et al19 could previously demonstrate an increase of LVs in acute rejection of lung grafts.

Although recent animal studies indicate a role for lymphangiogenesis in BOS, to the best of our knowledge, this issue has not been addressed in humans to date.20,21 We postulated increased lymphangiogenesis in the peribronchiolar region in CLAD patients. To test this hypothesis, we performed immunohistochemical analysis of podoplanin on lung tissue specimens of CLAD and non-CLAD patients and correlated our findings with clinical parameters.


Study Population

The study was approved by the ethics committee of the Medical University of Vienna (EK 1269/14) and was conducted according to the declaration of Helsinki. Twenty-two patients who underwent retransplantation at the Division of Thoracic Surgery, Medical University of Vienna for CLAD (14 BOS, 8 RAS) between 2003 and 2013 were included in this retrospective study. Time to BOS III and RAS diagnosis was defined as the time from primary transplantation to the date of BOS III and RAS diagnosis (verified by lung function testing according to the revised classification system for BOS [2001]6 and the recently stated classification for RAS8). Established clinicopathological parameters were gathered during lung transplantation follow-up visits at our department. Thirteen patients suffering from peripherally located lung cancer who underwent lobectomy or pneumectomy were assigned to the control group likewise to a previous study by Mori et al.22 The tumor was located 5 cm or greater from the lung area used for the immunohistochemical analysis. All patients used as controls had no evidence of parenchymal diseases (eg, chronic obstructive pulmonary disease, fibrosis, sarcoidosis), which could interfere with the number of LVs.22-24

Immunohistochemistry and Immunofluorescence

Immunohistochemical and immunofluorescence staining were performed on formalin-fixed, paraffin-embedded tissue specimens according to the manufacturer's protocol as described previously.25 Antipodoplanin (1:50, clone D2-40; Cell Marque Corporation, Rocklin, CA), anti-VEGFR-3 (1:50; ReliaTech, San Pablo, CA), and anti-LV endothelial hyaluronan receptor (LYVE)-1 (1:50, clone C-20; Santa Cruz Biotechnology, Santa Cruz, CA) were used as primary antibodies. As a negative control primary antibodies were omitted.

Determination of LVD

The number of LVs around small noncartilaginous bronchioles was counted and referred to the number of bronchioles and mm cumulative bronchial epithelium (measured lumen-sided). The cumulative length of lymphatic endothelium was referred to mm bronchial epithelium as previously described.22 Samples were evaluated for cellular immunological infiltrates. Sections were scanned with an automated scanning microscope (TissueFAXs; TissueGnostics, Vienna, Austria) and regions of interest were selected using the TissueFAXs Viewer 4.2 software (TissueGnostics). Mean values were calculated from 2 independently counted densities. In case of strong interobserver discrepancy, the slide was reevaluated. On an average, 6 bronchi per sample were evaluated. All measurements were performed using ImageJ version 1.46r software. Specificity of podoplanin staining was verified by double staining with VEGFR-3 and LYVE-1 (Fig. S1, SDC, within 5 patients.

Statistical Analysis

Data obtained were statistically evaluated using IBM SPSS Statistics version 21 (SPSS Inc., Chicago, IL) and GraphPad Prism 6 software (GraphPad Software Inc., LA Jolla, CA). Mann-Whitney U test and Kruskal-Wallis test were used to compare nonparametric variables and expressed as median, first and third quartile. Student t test and 1-way analysis of variance was used to compare parametric variables and expressed as mean ± SD. Kaplan-Meier curves and log-rank test were used to compare time to BOS III or RAS diagnosis between a low and high LVD (median value served as cutoff). Spearman rank correlation was used to assess a relationship between 2 nonparametric ranked variables. Fisher exact test was used to compare nominal variables. Dunn correction was used for post hoc tests in multiple group comparison. All tests were performed in a 2-sided manner. P values equal or below 0.05 were considered as statistically significant.


Study Cohort

Tissue specimens of 22 CLAD patients phenotyped as either BOS (14 patients, 64%) or RAS (8 patients, 36%) and 13 control subjects were available. Mean age of BOS patients was 36.8 ± 14.9 years, or RAS patients 30.1 ± 11.5 and of control subjects 68.6 ± 12.2 years, respectively (P < 0.001). Furthermore, no statistical significant difference between the groups was present. A summary of the clinicopathological parameters of included patients from the first transplant is depicted in Table 1.

Characteristics of the investigated groups from the first transplant

LVD Is Not Altered in BOS and RAS

To assess LVD, lung tissue specimens of BOS patients, RAS patients, and control subjects were stained for podoplanin (Figures 1A-C). Specific staining of LVs by podoplanin was verified by immunofluorescence double staining of LYVE-1 (Fig. S1A-C, SDC, and VEGFR-3 (Fig. S1D-E, SDC, No difference in LVD in the peribronchiolar region expressed as LVs per bronchiole (median, 4.75 (all CLAD), 4.25 (control) P = 0.94; 4.75 (BOS), 6.47 RAS), P = 0.97, Figure 2A) could be demonstrated. Additionally, we observed that LVs per mm bronchial epithelium (median = 2.20 (all CLAD), 1.98 (control), P = 0.72; 2.61 (BOS), 1.64 (RAS), P = 0.64, Figure 2B) as well as μm lymphatic endothelium per mm bronchial epithelium (median, 434.85 [all CLAD], 435.17 [control], P = 0.95; 496.27 [BOS], 318.87 [RAS], P = 0.72, Figure 2C) did not differ between groups.

Representative images showing peribronchiolar regions with LVs (arrows) of a BOS patient (A), RAS patient (B) and control subject (C).
LVD is not altered in patients suffering from BOS and RAS and does not serve as an indicator for time to BOS III and RAS diagnosis. LVD assessed as LVs per bronchiole (A), number of LVs per mm bronchiole (B) and μm lymphatic endothelium per mm bronchial epithelium (C) was equal in all groups. LVD expressed as LVs per bronchiole (D), number of LVs per mm bronchiole (E) and μm lymphatic endothelium per mm bronchial epithelium (F) was not associated with the time to BOS III or RAS diagnosis.

Kerjaschki et al13 had previously demonstrated an increase of LVs in kidney allografts with nodular inflammatory infiltrates. We therefore investigated whether or not LVD might be altered in bronchioles with inflammatory infiltrates compared to bronchioles without inflammatory infiltrates. Each section was evaluated for cellular immunological infiltrates. In the BOS group, samples with infiltrates were found in 9 patients and in the RAS group infiltrates were observed in 5 patients. In the control group, bronchioles with infiltrates were present in 3 patients. No difference in LVD in the peribronchiolar region with or without inflammatory infiltrates could be found in our study groups (LVs per bronchiole: with infiltrates: median, 5.00 [all CLAD], 4.00 [control], P = 0.69; 5.00 [BOS], 9.00 [RAS], P = 0.62, without infiltrates: median = 4.08 [all CLAD], 4.56 [control], P = 0.70; 4.50 [BOS], 0.00 [RAS], P = 0.74, Fig. S2A, D, SDC,; LVs per mm bronchial epithelium: with infiltrates: median = 2.30 [all CLAD], 2.07 [control], P = 0.63; 2.30 [BOS], 1.85 [RAS], P = 0.74, without infiltrates: median, 2.52 [all CLAD], 1.97 [control], P = 0.83; 3.64 [BOS], 0.00 [RAS], P = 0.74, Fig. S2B, E, SDC,; μm lymphatic endothelium per mm bronchial epithelium: with infiltrates: median = 384.72 [all CLAD], 435.17 [control], P = 0.91; 384.72 [BOS], 379.97 [RAS], P = 0.98, without infiltrates: median, 554.00 [all CLAD], 462.16 [control], P = 0.90; 600.44 [BOS], 0.00 [RAS], P = 0.56, Fig. S2C, F, SDC,

LVD Is Not Associated With the Time to Development of BOS and RAS

We assessed a possible association of LVD with the time to BOS III and RAS diagnosis. The density of LVs was independent from the time to development of BOS III after lung transplantation (LVs per bronchiole, low vs high: median, 38.0 vs. 86.0 [range, 7-89 vs 8-139] months, P = 0.15, Figure 2D; LVs per mm bronchial epithelium, low vs high: median, 25.5 vs 65.5 [range, 7-73 vs 8-139) months, P = 0.09, Figure 2E; μm lymphatic endothelium per mm bronchial epithelium, low vs high: median, 80.0 vs 39.5 [range, 11-139 vs 7-89] months, P = 0.10, Figure 2F). Furthermore, in the subgroup of RAS patients, LVD was not a relevant factor for the time to RAS development (LVs per bronchiole, low vs high: median, 60.5 vs 69.5 [range, 21-161 vs 39-134) months, P = 0.80, Figure 2D; LVs per mm bronchial epithelium, low vs high: median, 62.0 vs 77.0 [range, 21-161 vs 39-134] months, P = 0.98, Figure 2E; μm lymphatic endothelium per mm bronchial epithelium, low vs high: median, 62.0 vs 77.0 [range, 21-161 vs 39-134] months, P = 0.98, Figure 2F).


To the best of our knowledge, this is the first study in humans investigating angiogenesis of LVs in CLAD. Recent findings indicate a role of lymphangiogenesis in acute and chronic rejection of solid organ grafts. In rat models of kidney, heart, and lung transplant, an increase of LVD was demonstrated during chronic rejection.14,15,20,21 In human kidney transplantation, chronic graft loss is associated with a high LVD.13 An increase of LVD has been demonstrated in acute rejection of liver, heart, and kidney transplants.16-18 However, it is unclear if it may act in a beneficial or harmful way. Increased lymphangiogenesis could influence graft survival in a positive way by clearing inflammatory infiltrates or in contrast lead to worse graft survival through continuous alloantigen presentation in draining lymph nodes. This could promote a persisting alloimmune effector T cell response leading to chronic allograft rejection.

We demonstrated that lymphangiogenesis expressed as LVs per bronchiole, LVs per mm bronchial epithelium and μm lymphatic endothelium per mm bronchial epithelium is equal in CLAD and non-CLAD patients. Because Stuht et al26 have found an increased number of LVs in renal allografts with cellular infiltrates, we evaluated our specimens for the presence of cellular immunological infiltrates in the peribronchiolar region and divided them into samples with and without cellular infiltrates. Distribution of samples with and without cellular immunological infiltrates did not differ significantly between CLAD patients and control subjects. In addition, no differences in LVD could be seen. Although we included 14 BOS and 8 RAS patients, which is a comparably high number of patients in regard of the strict conditions for retransplantation, the sample size may still be too small to reach statistical significance.

In addition, we evaluated if the number of LVs may serve as an indicator for the time to BOS III or RAS diagnosis. However, we could not demonstrate a significant difference in time to BOS III or RAS diagnosis between low or high LVD patients.

The number of LVs in a newly transplanted lung has been thoroughly examined before.19 Big draining LVs are naturally cut at the level of the main bronchi during transplantation. It takes 7 to 14 days to reestablish new lymphatic connections after the disruption. These newly built LVs are sufficient to fulfil their tasks such as tissue fluid homeostasis, resolving tissue oedema, and immune cell trafficking. No further increase of LVs can be seen 3 months after the transplantation, neither in acutely rejected nor nonrejected lungs.19 In contrast, recent data by Cui et al27 report a reduction of LVs during acute lung transplant rejection. This decline could not be reversed after an alleviation of the rejection. As the lung is a highly immunologically active organ and exposed to environmental factors, in contrast to the kidney and heart, a further increase of the natural number LVs during rejection might not be necessary. Episodes of acute rejection predispose for later development of both BOS and RAS.5,10 We therefore evaluated the impact of acute rejection on the number of LVs in CLAD patients, but LVD did not differ in patients exhibiting episodes of acute rejection (Fig. S3A-C, SDC,

A possible confounder in our study might be the antilymphangiogenic effect of administered immunosuppressive drugs. Interestingly, a recent study demonstrated an increase of LVs in rat model of obliterative airway disease, which was reversed by administration of cyclosporine A (Cy A) in a dose-dependent way.21 Cyclosporine A is a known inhibitor of nuclear factor of activated T cells 1. Nuclear factor of activated T cells 1 is also found on lymphatic endothelial cells and involved in the expression of podoplanin. In addition, VEGF-A–induced pulmonary lymphangiogenesis is dependent on calcineurin activation. Embryonic mice exposed to Cy A show poorly organized sprouting of LVs from jugular lymph sacs.28 On the other hand, Cy A does not impact local proinflammatory cytokine release, which is an important pathophysiologic factor in the development of BOS.29 Eight of 22 CLAD patients in our study received a Cy A–based immunosuppressive therapy regimen. However, no difference in lymphangiogenesis could be demonstrated between patients with or without Cy A (Fig. S4A-C, SDC,

One strength of this study is that tissue specimens were obtained from explanted BOS and RAS organs during retransplantation. Tissue specimens from transbronchial biopsy are small and allow only an inadequate assessment of pulmonary lymphangiogenesis.19 In our study, the size of obtained tissue samples was approximately 20 × 25 mm. Images of full face sections could be recorded, which allowed us to draw a representative image of peribronchiolar lymphangiogenesis in BOS and RAS patients (Figure 1). Thus, the confounding factor of specimen size could be reduced to a minimum.

The use of lung tissue from patients, who underwent major lung resection (lobectomy, pneumectomy) for peripheral lung cancer, as a control is an important limitation of this study. Healthy lung tissue distant to the tumor from patients with lung cancer is a widely accepted control, mainly due to a lack of better alternatives. It is well known that cancer develops in an inflammatory environment and even though our samples were obtained from tissue greater than 5 cm distant from the tumor, an impact of the tumor on peribronchiolar LVs cannot be excluded. Another limitation of this study is the difference in age between the CLAD and non-CLAD groups.22,30 However, we could not find an association of age and lymphangiogenesis (Fig. S5A-C, SDC,

In conclusion, this is the first study in humans investigating lymphangiogenesis in patients with BOS and RAS. We found no difference in LVD in CLAD patients, and the number of peribronchiolar LVs was not associated with graft survival. Further studies are needed to confirm our findings in a larger patients cohort, including different stages of BOS.


The authors thank the Core Facility Imaging, Medical University of Vienna, Austria (Marion Gröger, PhD) for providing the TissueFAXs scanning microscopy and TissueFAXs Viewer software.


1. Yusen RD, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: Thirty-first Adult Lung and Heart-Lung Transplant Report-2014; Focus Theme: Retransplantation. J Heart Lung Transplant. 2014;33:1009–1024.
2. Weigt SS, DerHovanessian A, Wallace WD, et al. Bronchiolitis obliterans syndrome: the Achilles' heel of lung transplantation. Semin Respir Crit Care Med. 2013;34:336–351.
3. Sato M. Chronic lung allograft dysfunction. Gen Thorac Cardiovasc Surg. 2014;3:71–78.
4. Stewart S, Fishbein MC, Snell GI, et al. Revision of the 1996 working formulation for the standardization of nomenclature in the diagnosis of lung rejection. J Heart Lung Transplant. 2007;26:1229–1242.
5. Todd JL, Palmer SM. Bronchiolitis obliterans syndrome: the final frontier for lung transplantation. Chest. 2011;140:502–508.
6. Estenne M, Maurer JR, Boehler A, et al. Bronchiolitis obliterans syndrome 2001: an update of the diagnostic criteria. J Heart Lung Transplant. 2002;21:297–310.
7. Verleden GM, Raghu G, Meyer KC, et al. A new classification system for chronic lung allograft dysfunction. J Heart Lung Transplant. 2014;33:127–133.
8. Sato M, Waddell TK, Wagnetz U, et al. Restrictive allograft syndrome (RAS): a novel form of chronic lung allograft dysfunction. J Heart Lung Transplant. 2011;30:735–742.
9. Trulock EP. Lung transplantation. Am J Respir Crit Care Med. 1997;155:789–818.
10. Verleden SE, Ruttens D, Vandermeulen E, et al. Restrictive chronic lung allograft dysfunction: where are we now? J Heart Lung Transplant. 2015;34:625–630.
11. Tammela T, Alitalo K. Lymphangiogenesis: molecular mechanisms and future promise. Cell. 2010;140:460–476.
12. Baluk P, Tammela T, Ator E, et al. Pathogenesis of persistent lymphatic vessel hyperplasia in chronic airway inflammation. J Clin Invest. 2005;115:247–257.
13. Kerjaschki D, Regele HM, Moosberger I, et al. Lymphatic neoangiogenesis in human kidney transplants is associated with immunologically active lymphocytic infiltrates. J Am Soc Nephrol. 2004;15:603–612.
14. Palin NK, Savikko J, Koskinen PK. Sirolimus inhibits lymphangiogenesis in rat renal allografts, a novel mechanism to prevent chronic kidney allograft injury. Transpl Int. 2013;26:195–205.
15. Nykänen AI, Sandelin H, Krebs R, et al. Targeting lymphatic vessel activation and CCL21 production by vascular endothelial growth factor receptor-3 inhibition has novel immunomodulatory and antiarteriosclerotic effects in cardiac allografts. Circulation. 2010;121:1413–1422.
16. Yamamoto I, Yamaguchi Y, Yamamoto H, et al. A pathological analysis of lymphatic vessels in early renal allograft. Transplant Proc. 2006;38:3300–3303.
17. Geissler HJ, Dashkevich A, Fischer UM, et al. First year changes of myocardial lymphatic endothelial markers in heart transplant recipients. Eur J Cardiothorac Surg. 2006;29:767–771.
18. Ishii E, Shimizu A, Kuwahara N, et al. Lymphangiogenesis associated with acute cellular rejection in rat liver transplantation. Transplant Proc. 2010;42:4282–4285.
19. Dashkevich A, Heilmann C, Kayser G, et al. Lymph angiogenesis after lung transplantation and relation to acute organ rejection in humans. Ann Thorac Surg. 2010;90:406–411.
20. Krebs R, Tikkanen JM, Nykanen AI, et al. Dual role of vascular endothelial growth factor in experimental obliterative bronchiolitis. Am J Respir Crit Care Med. 2005;171:1421–1429.
21. Krebs R, Tikkanen JM, Ropponen JO, et al. Critical role of VEGF-C/VEGFR-3 signaling in innate and adaptive immune responses in experimental obliterative bronchiolitis. Am J Pathol. 2012;181:1607–1620.
22. Mori M, Andersson CK, Graham GJ, et al. Increased number and altered phenotype of lymphatic vessels in peripheral lung compartments of patients with COPD. Respir Res. 2013;14:65.
23. Yamashita M, Mouri T, Niisato M, et al. Heterogeneous characteristics of lymphatic microvasculatures associated with pulmonary sarcoid granulomas. Ann Am Thorac Soc. 2013;10:90–97.
24. El-Chemaly S, Malide D, Zudaire E, et al. Abnormal lymphangiogenesis in idiopathic pulmonary fibrosis with insights into cellular and molecular mechanisms. Proc Natl Acad Sci U S A. 2009;106:3958–3963.
25. Schweiger T, Kollmann D, Nikolowsky C, et al. Carbonic anhydrase IX is associated with early pulmonary spreading of primary colorectal carcinoma and tobacco smoking. Eur J Cardiothorac Surg. 2014;46:92–99.
26. Stuht S, Gwinner W, Franz I, et al. Lymphatic neoangiogenesis in human renal allografts: results from sequential protocol biopsies. Am J Transplant. 2007;7:377–384.
27. Cui Y, Liu K, Monzon-Medina ME, et al. Therapeutic lymphangiogenesis ameliorates established acute lung allograft rejection. J Clin Invest. 2015;125:4255–4268.
28. Kulkarni RM, Greenberg JM, Akeson AL. NFATc1 regulates lymphatic endothelial development. Mech Dev. 2009;126:350–365.
29. Floreth T, Stern E, Tu Y, et al. Differentiated transplant derived airway epithelial cell cytokine secretion is not regulated by cyclosporine. Respir Res. 2011;12:44.
30. Coussens LM, Werb Z. Inflammation and cancer. Nature. 2002;420:860–867.
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.