ORIGINAL RESEARCH published: 09 October 2019 doi: 10.3389/fimmu.2019.02339 Acute Graft-vs.-Host Disease-Associated Endothelial Activation in vitro Is Prevented by Defibrotide Julia Martinez-Sanchez 1,2,3 , Hannah Hamelmann 4,5,6 , Marta Palomo 1,2,3 , Enrique Mir 1,2,3 , Ana Belen Moreno-Castaño 2 , Sergi Torramade 2 , Montserrat Rovira 7 , Ginés Escolar 2 , Steffen Cordes 4,5,6 , Martina Kalupa 4,5,6 , Sarah Mertlitz 4,5,6 , Katarina Riesner 4,5,6 , Enric Carreras 1,3 , Olaf Penack 4,5,6 and Maribel Diaz-Ricart 2,3* 1 Josep Carreras Leukaemia Research Institute, Hospital Clinic, University of Barcelona, Barcelona, Spain, 2 Department of Hematopathology, Biomedical Diagnosis Center (CDB), Institute of Biomedical Research August Pi i Sunyer (IDIBAPS), Edited by: Hospital Clinic of Barcelona, University of Barcelona, Barcelona, Spain, 3 Barcelona Endothelium Team, Josep Carreras Hildegard Theresia Greinix, Leukaemia Research Institute, Barcelona, Spain, 4 Hematology, Oncology and Tumor Immunology Department, Charité Medical University of Graz, Austria Universitätsmedizin Berlin, Berlin, Germany, 5 Berlin-Brandenburg Center for Regenerative Therapies, Charité Medical University of Berlin, Berlin, Germany, 6 Department of Hematology and Oncology, Berlin Institute of Health, Berlin, Germany, Reviewed by: 7 Stem Cell Transplantation Unit, IDIBAPS, Hospital Clinic, University of Barcelona, Barcelona, Spain John Magenau, University of Michigan, United States Nuala Mooney, Angiogenesis and endothelial activation and dysfunction have been associated with Centre National de la Recherche acute graft-vs.-host disease (aGVHD), pointing to the endothelium as a potential target for Scientifique (CNRS), France pharmacological intervention. Defibrotide (DF) is a drug with an endothelium-protective *Correspondence: Maribel Diaz-Ricart effect that has been approved for the treatment of veno-occlusive disease/sinusoidal mdiaz@clinic.cat obstruction syndrome after allogeneic hematopoietic cell transplantation. Clinical data suggest that DF also reduces the incidence of aGVHD; however, the mechanisms Specialty section: This article was submitted to of DF-mediated aGVHD regulation have not been examined. To investigate possible Alloimmunity and Transplantation, DF-mediated prophylactic and therapeutic mechanisms in aGVHD, we performed in a section of the journal Frontiers in Immunology vitro studies using endothelial cell (EC) lines. We found that DF significantly and Received: 25 April 2019 dose-dependently suppressed EC proliferation and notably reduced their ability to form Accepted: 17 September 2019 vascular tubes in Matrigel. To explore whether DF administered prophylactically or Published: 09 October 2019 therapeutically has a significant effect on aGVHD endothelial dysfunction, ECs were Citation: exposed to media containing sera from patients with aGVHD (n = 22) in the absence or Martinez-Sanchez J, Hamelmann H, Palomo M, Mir E, presence of DF and from patients that did not develop aGVHD (n = 13). ECs upregulated Moreno-Castaño AB, Torramade S, adhesion molecules (vascular cell adhesion molecule 1, intercellular adhesion molecule Rovira M, Escolar G, Cordes S, Kalupa M, Mertlitz S, Riesner K, 1), the adherence junction protein VE-cadherin, von Willebrand factor (VWF), and Carreras E, Penack O and Akt phosphorylation in response to aGVHD sera. These responses were suppressed Diaz-Ricart M (2019) Acute upon treatment with DF. In summary, DF inhibits vascular angiogenesis and endothelial Graft-vs.-Host Disease-Associated Endothelial Activation in vitro Is activation induced by sera from aGVHD patients. Our results support the view that DF Prevented by Defibrotide. has notable positive effects on endothelial biology during aGVHD. Front. Immunol. 10:2339. doi: 10.3389/fimmu.2019.02339 Keywords: endothelial dysfunction, angiogenesis, acute GVHD, hematopoietic cell transplantation, defibrotide Frontiers in Immunology | www.frontiersin.org 1 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction INTRODUCTION prevention of aGVHD, is currently being evaluated for GVHD prevention in a clinical trial in adults (20). Hematopoietic cell transplantation (HCT) has demonstrated to The effect of DF has never been evaluated in an in vitro be an effective treatment for various hematological and non- aGVHD-related endothelial damage scenario. The present study hematological diseases. Allogeneic HCT (allo-HCT) is associated aimed to explore the effect of DF on endothelial cell proliferation with various early complications, among which acute graft-vs.- and tube formation and its potential as a prophylactic or/and host disease (aGVHD) is the major cause of morbidity and therapeutic agent for aGVHD-associated endothelial damage. mortality after this treatment (1–5). aGVHD is characterized by tissue and organ injury mainly affecting the liver, gut, and skin (1). First-line treatment for patients undergoing aGVHD is MATERIALS AND METHODS limited to corticosteroid therapy and immunosuppressive drugs, Endothelial Cell Culture such as calcineurin or mTOR inhibitors (1, 6, 7). Although Primary human umbilical vein endothelial cells (HUVEC) were aGVHD progression can be partly or completely reverted used to assess DF effect in proliferation and tube formation through the effects of these therapies on immune cells, they have assays. Human microvascular endothelial cells (HMEC-1) a detrimental impact on the endothelium (8). (ATCC, Manassas, VA, USA) were used to investigate DF The endothelial cells (ECs) that cover the vascular tree effect on a GVHD-endothelial dysfunction in vitro model. are directly exposed to all the damaging factors occurring HUVEC were cultured in endothelial basal medium 2 (Lonza, during the onset and development of aGVHD. Our group Walkersville, MD, USA) supplemented with endothelial cell has previously demonstrated that allo-HCT is associated with growth medium SingleQuots TM (EGM-2) (Lonza), human endothelial dysfunction (9, 10) and that endothelial damage epidermal growth factor, hydrocortisone, GA-1000 (gentamicin, is aggravated with the development of aGVHD. Humoral amphotericin-B), 2% fetal bovine serum, vascular endothelial factors in the sera of allo-HCT recipients developing aGVHD growth factor, human basic fibroblast growth factor, R3 insulin induced a marked proinflammatory and prothrombotic state growth factor-1, ascorbic acid, and heparin. HMEC-1 were in cultured ECs (11). Plasma levels of von Willebrand grown in MCDB131 (Gibco-BRL, Madrid, Spain) medium factor (VWF) and tumor necrosis factor receptor 1 were supplemented with 15% fetal bovine serum, 4% L-glutamine, 1% higher in patients developing aGVHD than in allo-HCT penicillin-streptomycin (Gibco-BRL), 1 µg/ml hydrocortisone recipients not developing this complication. Other endothelial (Sigma-Aldrich, Madrid, Spain), and 10 ng/ml epidermal growth damage markers, including angiopoietin-2, were increased factor (BD Biosciences, Erembodegem, Belgium). Both HUVEC even in pretransplant patients who later developed aGVHD, and HMEC-1 were cultured under humidified conditions at 37◦ C corroborating that the endothelium plays a role in the and in the presence of 5% CO2 . pathogenesis of aGVHD (12). In addition, it has been considered that angiogenesis, the formation of new blood vessels, might play MTT Proliferation Assay a major role in aGVHD development. It has been demonstrated The proliferation rate of cells was assessed by the 3-[4,5- that, in the early inflammatory phase of GVHD, there is a dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT) noticeable neovascularization that facilitates the migration of assay using the MTT Cell Proliferation Assay Kit (ATCC). inflammatory cells to target organs (13). Together, these findings HUVEC were seeded in three 96-well plates (Sarstedt, highlight the endothelium as a potential target for new therapies Nümbrecht, Germany) at 1–2 × 104 cells/well in 100 µl of to prevent and treat aGVHD (11, 14, 15). EGM-2. DF was added at 10, 100, or 500 µg/ml. The first two Defibrotide (DF), an endothelial-protective drug (16), has concentrations (10 and 100 µg/ml) were selected due to their demonstrated to be effective in the prevention and treatment clinical relevance (21, 22) and the last one (500 µg/ml) to analyze of veno-occlusive disease/sinusoidal obstruction syndrome if maximum concentrations, not expected in the clinical routine, (VOD/SOS), an early complication of HCT proven to be of have an additional effect. Then, the plates were incubated at 37◦ C endothelial origin (17). Our previous studies demonstrated the in for 24 h (D + 1), 48 h (D + 2), or 72 h (D + 3), and 10 µl of MTT vitro preventive effect of DF on the development of endothelial reagent was added to each well. After 3 h, 100 µl of Detergent damage in association with HCT (18) and the mechanisms of Reagent was added to each well, and the plates were incubated action involved (19). DF, designated as an orphan drug for the in the dark (3–4 h or overnight) until the precipitated formazan crystals were dissolved. The absorbance at 570 nm was measured using a SpectraMax Plus 384 mL plate reader (Molecular Devices, San Francisco, CA, USA). The mean value from quadruplicate Abbreviations: aGVHD, acute graft-vs.-host disease; allo-HCT, allogeneic hematopoietic cell transplantation; DF, defibrotide; DF-PT, defibrotide— readings was determined. prophylaxis plus treatment; DF-T, defibrotide—therapeutic; EC, endothelial cell; ECM, extracellular matrix; HCT, hematopoietic cell transplantation; Tube Formation Assay HMEC-1, human microvascular endothelial cells; HUVEC, human umbilical The angiogenic activity of cells was determined by a tube vein endothelial cells; ICAM-1, intercellular adhesion molecule 1; MFI, mean formation assay. Forty microliters of defrosted Matrigel R fluorescence intensity; SEM, standard error of the mean; TNFR1, tumor necrosis factor receptor 1; VCAM-1, vascular cell adhesion molecule 1; (Corning, Corning, NY, USA) was pipetted into the wells of a VOD/SOS, veno-occlusive disease/sinusoidal obstruction syndrome; VWF, 96-well plate (Sarstedt), using a cut 10- or 100-µl pipette tip. von Willebrand factor. The plate was left to stand for 30–45 min until the Matrigel Frontiers in Immunology | www.frontiersin.org 2 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction was solidified. HUVEC (1 × 104 cells) in 100 µL of medium TABLE 1 | Characteristics of patients with or without acute GVHD. containing DF at different concentrations were added to Groups GVHD NoGVHD each well. Cells incubated in EGM-2 and endothelial basal medium 2 medium were used as a positive and negative control, Patients respectively. Tube formation was evaluated every hour, and after Number (% of the whole group) 22 (63) 13 (27) 6 h, micrographs were taken of three to five random fields per Gender well. The micrographs were analyzed using the Angiogenesis Male/female 11/11 10/3 Analyzer software developed by Gilles Carpentier (http:// Diagnostic image.bio.methods.free.fr/ImageJ/?Angiogenesis-Analyzer-for- ALL 3 3 ImageJ&lang=en) in ImageJ Fiji (Bethesda, Rockville, MD, AML 8 5 USA) (23). The following features were assessed: number of NHL/HL 6 3 junctions, number of master junctions, number of segments, PCD 2 1 number of master segments, number of meshes, mean mesh CLL/PLL 3 1 size, and total meshes area. The mean value from quadruplicate Donor type readings from two independent experiments was determined Sibling (% by group) 8 (36) 11 (85) for each parameter. Student’s unpaired t-test was applied for Unrelated (% by group) 14 (64) 2 (15) the statistical analysis of all data using the statistics program Conditioning GraphPad Prism, version 4.02 (GraphPad Software Inc., La Jolla, MAC (% by group) 12 (54) 9 (69) CA, USA). The values were adjusted to the positive control RIC (% by group) 10 (46) 4 (31) and shown as percentage; the value of the positive control was GVHD Prophylaxis adjusted to 100%. Values were shown as mean ± standard error CyA 2 – of the mean (SEM). CyA/MTX 10 4 CyA/MMF 9 8* Exposure of Endothelial Cells in Culture to Tacro/Rapa 1 1 Acute GVHD II–IV aGVHD Media Onset day, median (range) 16 (7–28) NA The effect of DF on aGVHD was assessed in vitro. HMEC- 1 (24) were incubated in growth medium supplemented with Chronic GVHD 20% of serum pools, each pool containing combinations of >at day+180 (% by group) 10 (78) 6 (46) sera from five different patients. Serum samples were collected *One case with CD34+ positive selection. from patients with aGVHD on the day of diagnosis. For AML, acute myeloid leukemia; ALL, acute lymphoblastic leukemia; NHL, non-Hodgkin’s experimental treatments, DF (100 µg/ml) was added in two lymphoma; HL, Hodgkin’s lymphoma; PCD, plasma cell disease; CLL, chronic lymphocytic leukemia; PLL, prolymphocytic leukemia; MAC, myeloablative conditioning; modalities: (i) prophylaxis plus treatment (DF-PT), in which RIC, reduced intensity conditioning; CyA, cyclosporine; MTX, methotrexate; MMF, ECs were incubated with DF 24 h before exposure to aGVHD mycophenolate mofetil; Rapa, rapamicine; Tacro, tacrolimus. medium, and then fresh DF was added with medium replacement every 24 h, and (ii) therapeutic (DF-T), in which ECs were exposed to aGVHD medium, and DF was added 24 h later and Immunofluorescence Analysis at growth medium replacement every 24 h. The control group The expression of the adhesion molecules intercellular adhesion consisted of serum samples from allo-HCT recipients that did not molecule 1 (ICAM-1) and vascular cell adhesion molecule 1 develop aGVHD (NoGVHD). (VCAM-1) on the cell surface, VE-cadherin at cell junctions, and VWF in the extracellular matrix (ECM) was assessed by immunofluorescence. HMEC-1 monolayers were incubated Patients and Specimen Collection in medium containing 20% pooled serum from the aGVHD Patients undergoing allo-HCT at the Hospital Clinic of Barcelona group, in the absence or presence of DF (DF-PT or DF- were included consecutively between 2008 and 2010. The T), and from NoGVHD group in eight-well microplates experimental group consisted of patients who developed aGVHD (Nunklabtech, Madrid, Spain) for 48 h. ECs were fixed with grade II–IV within the first 3 months post allo-HCT (aGVHD 4% paraformaldehyde and incubated with specific monoclonal group, n = 22). The control group consisted of 13 patients antibodies to VCAM-1 (MA5-11447, 1/10; Invitrogen, Carlsbad, without aGVHD (NoGVHD group). Blood samples were CA, USA) and ICAM-1 (sc-107, 1/50; SantaCruz Biotech, Dallas, collected on the day of aGVHD diagnosis (before any change in TX, USA) or permeabilized (0.05% Triton X-100, 3 min at room treatment) for the aGVHD group and on day 21 post allo-HCT temperature) and then incubated with a specific monoclonal for the NoGVHD group. Patients with other early complications antibody to VE-cadherin (ab33168, 1/1,000; Abcam, Cambridge, associated with allo-HCT such as VOD/SOS, thrombotic UK) or a specific polyclonal antibody to VWF (A0082, 1/2,000; microangiopathy, capillary leak syndrome, or engraftment Dako/Agilent, Santa Clara, CA, USA), at room temperature for syndrome were excluded for this study. Serum was obtained by 1 h. Then, the cells were incubated with secondary antibodies centrifugation at 2,500 rpm for 15 min and was stored at −80◦ C. conjugated with Alexa Fluor 488 or 594 (Molecular Probes, Patient characteristics are provided in Table 1. Eugene, OR, USA) at room temperature for 1 h. Cell monolayers Frontiers in Immunology | www.frontiersin.org 3 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction were evaluated using a DM4000 B epifluorescence equipped Activation of proteins induced by NoGVHD sera was used microscope and images captured through a video camera (Leica, as control. Barcelona, Spain). Analysis of fluorescence was carried out using ImageJ Fiji software in 20 fields per coverslip, by independent Statistical Analysis technicians blinded to the treatments. Data are reported as the mean ± SEM. Data from aGVHD and NoGVHD exposure experiments following a normal distribution Reactivity of the ECM Generated by were analyzed by paired t-tests. Data from proliferation and tube formation experiments were analyzed by unpaired t-tests. HMEC-1 Exposed to Study Conditions All statistical analyses were conducted using GraphPad Prism, ECs were cultured for 7 days on 18 × 18 mm2 coverslips in version 4.02. P < 0.05 was considered significant. six-well plates in medium supplemented with 20% serum pools. The monolayers were exposed to 9% ethyleneglycoltetraacetic Ethical Aspects acid at 37◦ C for 1 h to remove ECs. To evaluate the reactivity of Informed consent for blood utilization was obtained from all the ECM toward circulating platelets, the ECM-coated coverslips patients. The study was approved by the Ethics committee of were perfused with citrated whole blood from healthy donors the Hospital Clinic of Barcelona (CEIC 2008/3353) and was in a parallel-plate perfusion chamber (shear rate of 800/s, carried out in accordance with the principles of the Declaration 5 min). After perfusion, the coverslips were fixed with 0.5% of Helsinki. glutaraldehyde at 4◦ C and stained with 0.02% toluidine blue. Images were captured using the DM4000 B epifluorescence microscope. The surface covered by platelets (%) was quantified RESULTS en face in 20 fields per well using ImageJ Fiji by independent DF Suppresses EC Proliferation and Tube technicians blinded to the treatments. Formation As aGVHD has been associated with angiogenesis, we Western Blotting Analysis of investigated the effect of DF on the angiogenic potential of Phosphorylated Intracellular Proteins ECs. First, we assessed the impact of DF on EC proliferation Confluent EC monolayers, after starving with medium by MTT assay. DF reduced EC proliferation in a time- and containing 2% serum for 24 h, were exposed to the dose-dependent manner (Figure 1). Of note, even at the serum pools under study with or without DF for 15 min. lowest concentration of 10 µg/ml, DF significantly inhibited The ECs were lysed with Laemmli’s buffer (68.5 mM EC proliferation. The reduction in the proliferation was more tris[hydroxymethyl]aminomethane-HCl, 2% sodium dodecyl notable at 100 µg/ml (Figures 1B,C corresponding to MTT assay sulfate, 10% glycerol, 5 mM β-mercaptoethanol, and 0.003% day + 2 and day + 3), and no differences were observed between bromophenol blue) containing 2 mM sodium orthovanadate, 100 and 500 µg/ml concentrations. and 0.625 mg/ml N-ethylmaleimide. Proteins were resolved by Next, we assessed the impact of DF on the angiogenic activity 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis of ECs by a tube formation assay. Cells were incubated in the and transferred to nitrocellulose membranes (BioRad, Hercules, presence of 10, 100, or 500 µg/ml DF in a 96-well plate coated Canada), which were probed with specific antibodies to with Matrigel and analyzed after 6 h (micrographs in Figure 2). phosphorylated and unphosphorylated Akt, p38 MAPK, Erk1/2, DF significantly suppressed tube formation in a dose-dependent and SAPK/JNK (Cell Signaling Technology, Danvers, MA, USA). manner for all features assessed: junctions (Figure 2A), master FIGURE 1 | Defibrotide (DF) suppresses endothelial cell (EC) proliferation. An 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay was used to quantify the effect of DF (10, 100, or 500 µg/ml) for (A) 24 h (D + 1), (B) 48 h (D + 2), or (C) 72 h (D + 3) on EC proliferation. Bar diagrams represent the optical density (OD) normalized to control data (%) expressed as the mean ± SEM of three experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Frontiers in Immunology | www.frontiersin.org 4 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction FIGURE 2 | Defibrotide (DF) suppresses endothelial cell (EC) tube formation. Representative images of angiogenic activity of ECs by a tube formation assay incubated with DF at different concentrations. Bar diagrams represent different characteristics measured during tube formation: numbers of (A) junctions and (B) master junctions, numbers of (C) segments and (D) master segments, and (E) number, (F) size, and (G) area of meshes, as indicated. Data are expressed as the mean ± SEM of three experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Frontiers in Immunology | www.frontiersin.org 5 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction junctions (Figure 2B), segments (Figure 2C), master segments and then fresh DF was added with medium replacement every (Figure 2D), meshes (Figure 2E), mean mesh size (Figure 2F), 24 h; and (ii) therapeutic (DF-T), in which ECs were exposed and total meshes area (Figure 2G). Thus, DF significantly to aGVHD medium, and DF was added 24 h later and at suppressed EC proliferation and tube formation in vitro. growth medium replacement every 24 h. Figure 3 shows that aGVHD serum-induced expression of VCAM-1 and ICAM-1 was significantly suppressed in the DF-PT group (to 37.1 ± 2.1 DF Prevents aGVHD Serum-Induced and 24.8 ± 3, respectively, P < 0.05) and the DF-T group (to 31.7 Expression of Adhesion Molecules and ± 4.1 and 26.6 ± 2.4, respectively, P < 0.01). VE-Cadherin on ECs Next, we analyzed the expression of VE-cadherin, an To analyze the effect of soluble factors present in aGVHD patient adherence junction protein involved in angiogenesis and vascular serum on ECs, HMEC-1 were cultured with pooled aGVHD or integrity, on ECs exposed to aGVHD or NoGVHD serum. NoGVHD serum samples for 48 h, and the expression of surface aGVHD serum enhanced VE-cadherin expression compared to proteins was analyzed by immunofluorescence. VCAM-1 and NoGVHD serum (MFI of eight different experiments, 15.14 ± ICAM-1 levels were significantly increased in the presence of 1.34 vs. 8.3 ± 2, P < 0.05). DF suppressed the VE-cadherin aGVHD serum compared to NoGVHD serum (Figure 3). The upregulation in the prophylactic as well as the therapeutic setting mean fluorescence intensity (MFI of eight different experiments) (DF-PT, 12.8 ± 2.1; DF-T, 11.5 ± 1.1, P < 0.05) (Figure 4). Thus, for VCAM-1 and ICAM-1 was 41.9 ± 3.1 and 34.3 ± 3.3, DF reduced aGVHD serum-induced upregulation of adhesion respectively, in the presence of aGVHD serum vs. 27 ± 3.2 molecule and VE-cadherin expression on ECs in vitro. and 22.3 ± 3.2, respectively, in NoGVHD (P < 0.05 and P < 0.01, respectively). To analyze whether DF can suppress the DF Inhibits the Release of VWF From ECs aGVHD-induced upregulation of adhesion molecule expression, to the ECM we incubated ECs with aGVHD serum and DF in two regimens: To investigate the potential antithrombotic effect of DF, (i) prophylaxis plus treatment (DF-PT), in which ECs were either DF-PT or DF-T, in aGVHD endothelial damage, VWF incubated with DF 24 h before exposure to aGVHD medium, expression in ECM generated by ECs grown in the presence FIGURE 3 | Defibrotide (DF) has an anti-inflammatory effect on endothelial cells (ECs) activated and injured after exposure to acute graft-vs.-host disease (aGVHD) serum. Representative images of (A) vascular cell adhesion molecule 1 (VCAM-1) and (B) intercellular adhesion molecule 1 (ICAM-1) expression on EC surfaces in response to aGVHD serum collected on the day of onset in the absence (GVHD-NoDF) or presence of DF (prophylaxis and treatment, GVHD + DF-PT; treatment, GVHD + DF-T). An image representative of results with NoGVHD medium is also included. Bar diagrams show the quantitative analysis of VCAM-1 and ICAM-1 for GVHD-NoDF (black bars), GVHD + DF-PT (gray bars), GVHD + DF-T (dashed bars), and NoGVHD (white bars). Values represent mean ± SEM of MFI from eight different experiments. *P < 0.05, **P < 0.01. Original magnification, 400× for all micrographs. Frontiers in Immunology | www.frontiersin.org 6 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction FIGURE 4 | Defibrotide (DF) attenuates the increase in VE-cadherin expression on the surface of endothelial cells (ECs) exposed to acute graft-vs.-host disease serum. VE-cadherin staining in ECs exposed to aGVHD serum in the absence (GVHD-NoDF) or presence of DF. An image representative of results with NoGVHD medium is also included. Bar diagrams show the quantitative analysis of VE-cadherin expression for GVHD-NoDF (black bar), GVHD + DF-PT (gray bar), GVHD + DF-T (dashed bar), and NoGVHD (white bar). Values indicate the mean ± SEM of MFI from eight different experiments. *P < 0.05. Original magnification, 400× for all micrographs. of aGVHD or NoGVHD with or without DF was evaluated by activation of p38, SAPK/JNK, and Erk1/2 to 1.3 ± 0.1, 1.13 ± immunofluorescence. aGVHD serum enhanced VWF expression 0.1, and 1.05 ± 0.1, respectively, and DF-T to 1.14 ± 0.16, 1.1 ± compared to NoGVHD serum. The MFI of VWF, based on eight 0.06, and 1.37 ± 0.05, respectively (Figures 6B–D). Differences different experiments, was 47.1 ± 5.6 for aGVHD vs. 31.9 ± 2.1 between groups were only statistically significant for Akt. for NoGVHD (P < 0.01). Both DF-PT and DF-T significantly suppressed VWF production, as indicated by MFIs of 37.4 ± 5.2 for DF-PT (P < 0.05) and 40.1 ± 5.9 for DF-T (Figure 5A). DISCUSSION DF Reduces the Increased Reactivity of the We investigated the effect of DF on angiogenesis and its potential preventive and restorative effects on endothelial injury caused by ECM Toward Platelets Caused by Exposure aGVHD in an in vitro cell culture model. Experiments performed to aGVHD Serum to explore the effect of DF on angiogenesis showed that this The effects of aGVHD serum and DF on ECM thrombogenicity compound suppressed both proliferation and tube formation were studied under flow conditions. ECMs generated by ECs capability in ECs. In addition, DF prevented and partially grown in the presence of aGVHD or NoGVHD serum for 7 days reversed the endothelial inflammatory and prothrombotic were exposed to circulating citrated blood from healthy donors, phenotypes induced by aGVHD serum. Interestingly, our in and the surface covered by platelets was determined. The covered vitro experiments demonstrated that DF is capable of restoring surface (expressed in percent with respect to the total area of the endothelial damage when administered therapeutically. Our field analyzed, mean ± SEM, n = 8) was higher in the aGVHD results suggest a potential role of this drug as a prophylactic than in the NoGVHD group (38.4 ± 1.3 vs. 27 ± 1%, P < and therapeutic agent for endothelial dysfunction associated 0.01). DF-PT and DF-T significantly reduced surface coverage with aGVHD. to 30.4 ± 0.8% (P < 0.001) and to 32.1 ± 2.6% (P < 0.05), Owing to advances in the knowledge of aGVHD respectively (Figure 5B). pathophysiology in the last decade, the vascular endothelium was identified as a new target for the prevention and treatment DF Suppresses the Activation of Akt in ECs of this complication (6, 15, 25–28). In the initial phases of Exposed to Serum From aGVHD Patients aGVHD development, angiogenesis precedes key processes, such The effects of aGVHD serum and DF on phosphorylation- as inflammatory cell infiltration in typical aGVHD target organs mediated activation of Akt, p38 MAPK, SAPK/JNK, and Erk1/2 (29). However, blood vessels suffer fibrosis and rarefaction during were investigated by Western blotting. Akt phosphorylation was later stages of aGVHD due to the attack of alloreactive donor significantly higher in cells exposed to aGVHD serum (fold T cells on the vasculature (14, 30). Inhibition of angiogenesis increase of 2.1 ± 0.3 in aGVHD vs. NoGVHD, P < 0.01). DF- during the early phases of aGVHD may suppress the recruitment PT and DF-T significantly suppressed Akt phosphorylation in of infiltrating leukocytes and, consequently, inflammation, response to aGVHD serum to 1.4 ± 0.2 and 1.6 ± 0.4, respectively and thus may improve aGVHD (13, 29). Moreover, there is (Figure 6A). Similar trends were observed for p38, SAPK/JNK, evidence for the association between angiogenesis and the and Erk1/2, with fold increases of 1.46 ± 0.2, 1.36 ± 0.3, and 1.4 expression of VE-cadherin, the main endothelial cell-to-cell ± 0.1 in aGVHD vs. NoGVHD, respectively. DF-PT reduced the junction transmembrane protein that functions in vascular Frontiers in Immunology | www.frontiersin.org 7 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction FIGURE 5 | Defibrotide (DF) reduces von Willebrand factor (VWF) expression and platelet adhesion on extracellular matrix (ECM) produced by endothelial cells (ECs) exposed to aGVHD serum. (A) Micrographs showing expression of VWF on ECM produced by ECs exposed to aGVHD serum in the absence (GVHD-NoDF) or presence of DF (DF-PT and DF-T). An image representative of results with NoGVHD medium is also included. Bar diagrams show the quantitative analysis of VWF expression for GVHD-NoDF (black bar), GVHD + DF-PT (gray bar), GVHD + DF-T (dashed bar), and NoGVHD (white bar). Data are the mean ± SEM of MFI from eight different experiments. *P < 0.05. Original magnification, 400× for all micrographs. (B) Representative images showing platelet adhesion after exposing ECM produced under the indicated conditions to flowing blood (800/s, 5 min). Bar diagrams show the surface covered by platelets (%) for GVHD-NoDF (black bar) GVHD + DF-PT (gray bar), GVHD + DF-T (dashed bar), and NoGVHD (white bar). Data are the mean ± SEM of MFI from eight different experiments. *P < 0.05, **P < 0.001. Original magnification, 200× for all micrographs. morphogenesis and endothelial survival, on ECs (31, 32). In ICAM-1. DF had an anti-inflammatory effect as it suppressed the this regard, Medinger et al. (33) reported higher microvessel expression of these receptors upon therapeutic and prophylactic density and VE-cadherin expression in bone marrow sections plus therapeutic administration. This strategy has already been from patients with aGVHD than from patients without aGVHD. explored in the context of aGVHD; studies in mouse models have In line herewith, we also found that VE-cadherin expression shown that suppression of effector T cell recruitment by blocking was elevated in ECs exposed to aGVHD serum. In addition, chemokine receptors and their ligands might be an interesting VE-cadherin was clearly downregulated by DF, mostly when ECs approach to reducing GVHD (37–39). were incubated with DF after 24 h of exposure to aGVHD serum. ECs grown in the presence of sera from HCT recipients The effect of DF on angiogenesis has been controversial. On generate an ECM enriched with VWF and tissue factor, which are the one hand, DF has been shown to bind to and protect basic responsible for increased reactivity toward circulating platelets fibroblast growth factor, a key angio-protective protein, thus (9). This prothrombotic phenotype was even more notable promoting EC mitogenesis in vitro (34), and to stimulate hepatic in ECs exposed to sera from aGVHD patients than in ECs sinusoidal EC angiogenesis in a dose-dependent manner (35). exposed to NoGVHD serum. DF was able to suppress this On the other hand, it has been demonstrated that DF directly overproduction and/or release of VWF to the ECM. These interferes with EC migration and tube formation in vitro and in results are in line with the antithrombotic–thrombolytic behavior animal models (36). Our results regarding new vessel formation of DF demonstrated in several experimental models (16) and VE-cadherin expression support the antiangiogenic role of and with results from clinical trials that demonstrated that this drug. DF restores the thrombolytic–fibrinolytic balance in patients The current study also provided new evidence indicating that suffering from VOD/SOS (40). DF has been also proposed for the treatment of ECs with DF is also effective in restoring the the treatment of thrombosis associated with various pathologies, inflammatory phenotype induced by aGVHD. The exposure of including cerebral malaria, in which DF treatment leads to the ECs to pooled sera from aGVHD recipients induced increases in maintenance of thrombomodulin and the suppression of platelet cell-surface expression of the adhesion receptors VCAM-1 and aggregation (41). Frontiers in Immunology | www.frontiersin.org 8 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction FIGURE 6 | Defibrotide (DF) suppresses the activation of Akt in endothelial cells (ECs) exposed to serum from aGVHD patients. ECs were exposed for 15 min to aGVHD serum in the absence (GVHD-NoDF) or presence of DF (DF-PT and DF-T) to evaluate the effect of DF on the phosphorylation of (A) Akt (P-Akt), (B) p38 MAPK (P-p38), (C) SAPK/JNK (P-SAPK/JNK), and (D) Erk1/2 (P-Erk1/2) by Western blotting. Images are representative of four experiments. Bar diagrams represent the increases in phosphorylation with respect to the phosphorylation levels in cells exposed to NoGVHD serum. Data are the mean ± SEM from four different experiments **P < 0.01. Results corresponding to NoGVHD are represented by a dotted line. Sera from aGVHD patients disrupted endothelial homeostasis to NoGVHD serum. DF prevented and reduced the activation of and induced strong activation of the PI3K-Akt-mTOR stress- all these proteins. signaling pathway. DF suppressed phosphorylation-dependent The present study demonstrated that DF inhibits EC Akt activation nearly to control levels. Akt is related to processes proliferation and tube formation. Considering that angiogenesis devoted to EC stabilization during a dysfunctional state (42, 43), precedes the onset of aGVHD, our results suggest that DF and its dysregulation has been extensively studied in aGVHD, might be useful for the prevention and treatment of aGVHD. although not in endothelium. PI3K/Akt pathway inhibitors can In addition, DF ameliorates the dysfunctional endothelial interfere with T cell activation, and their therapeutic utility in phenotype caused by aGVHD serum by preventing inflammatory aGVHD has been explored (44, 45). The endothelial damage and prothrombotic cell reactions and suppresses the activation caused by leukocyte transmigration implies the activation of of the stress-related protein Akt. Interestingly, DF is able to the inflammatory protein p38, the apoptosis-related protein restore established endothelial damage. The present study was SAPK/JNK, and the cell proliferation, differentiation, and performed in vitro, and more research is needed to corroborate survival protein Erk1/2 (46). In the current study, all these our findings. However, our results warrant further study of the proteins were consistently, to different extents, more strongly usefulness of DF for the prevention and treatment of aGVHD. activated in cells exposed to aGVHD serum than in cells exposed Clinical trials on aGVHD prophylaxis by DF are currently going. Frontiers in Immunology | www.frontiersin.org 9 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction DATA AVAILABILITY STATEMENT patient data. MR, SC, MK, SM, KR, AM-C, ST, OP, and GE reviewed the data and the manuscript. EC, OP, and MD-R The raw data supporting the conclusions of this manuscript will designed and supervised the project and reviewed the data be made available by the authors, without undue reservation, to and the manuscript. All authors read, edited, and approved any qualified researcher. the manuscript. ETHICS STATEMENT FUNDING This study was carried out in accordance with the This study was supported in part by Jazz Pharmaceuticals Plc recommendations of Ethics committee of the Hospital Clinic (IST-16-10355), German Jose Carreras Leukaemia Foundation of Barcelona with written informed consent from all subjects. (11R/2016 and 03R/2019), Integrated Project in Health Institutes All subjects gave written informed consent in accordance with (PIE15/00027), Technology Development Projects in Health the Declaration of Helsinki. The protocol was approved by the 2016 (DTS16/00133 and FIS PI19/00888) from the Instituto Hospital Clinic of Barcelona (CEIC 2008/3353). de Salud Carlos III, Spanish Government; 2017-SGR671 from Generalitat de Catalunya, and CERCA Programme of the Generalitat de Catalunya. AUTHOR CONTRIBUTIONS JM-S, EM, and HH designed and performed the experiments, ACKNOWLEDGMENTS analyzed the results, and prepared the images. JM-S and EM wrote the manuscript. MP, ST, and AM-C contributed to We would like to thank the Primary Hemostasis Laboratory experimental design, image preparation, and reviewed the data Group at Hospital Clinic de Barcelona for the contribution to the and the manuscript. MR provided the samples and clinical performing of some experiments and technical support. REFERENCES 11. Mir E, Palomo M, Rovira M, Pereira A, Escolar G, Penack O, et al. Endothelial damage is aggravated in acute GvHD and could predict its development. Bone 1. Holtan SG, Pasquini M, Weisdorf DJ. Acute graft-versus-host Marrow Transplant. (2017) 52:1317–25. doi: 10.1038/bmt.2017.121 disease: a bench-to-bedside update. Blood. (2014) 124:363–73. 12. Dietrich S, Falk CS, Benner A, Karamustafa S, Hahn E, Andrulis M, doi: 10.1182/blood-2014-01-514786 et al. Endothelial vulnerability and endothelial damage are associated 2. Carreras E, Diaz-Ricart M. The role of the endothelium in the short- with risk of graft-versus-host disease and response to steroid treatment. term complications of hematopoietic SCT. Bone Marrow Transplant. (2011) Biol Blood Marrow Transplant. (2013) 19:22–7. doi: 10.1016/j.bbmt.2012. 46:1495–502. doi: 10.1038/bmt.2011.65 09.018 3. Jodele S, Laskin BL, Dandoy CE, Myers KC, El-Bietar J, Davies SM, et al. A 13. Penack O, Henke E, Suh D, King CG, Smith OM, Na I-K, et al. Inhibition new paradigm: diagnosis and management of HSCT-associated thrombotic of neovascularization to simultaneously ameliorate graft-vs-host disease microangiopathy as multi-system endothelial injury. Blood Rev. (2015) and decrease tumor growth. J Natl Cancer Inst. (2010) 102:894–908. 29:191–204. doi: 10.1016/j.blre.2014.11.001 doi: 10.1093/jnci/djq172 4. Lee Y-H, Rah W-J. Pre-engraftment syndrome: clinical significance and 14. Biedermann BC, Sahner S, Gregor M, Tsakiris DA, Jeanneret C, Pober JS, pathophysiology. Blood Res. (2016) 51:152–4. doi: 10.5045/br.2016.51.3.152 et al. Endothelial injury mediated by cytotoxic T lymphocytes and loss of 5. Lucchini G, Willasch AM, Daniel J, Soerensen J, Jarisch A, Bakhtiar S. microvessels in chronic graft versus host disease. Lancet. (2002) 359:2078–83. Epidemiology, risk factors, and prognosis of capillary leak syndrome in doi: 10.1016/S0140-6736(02)08907-9 pediatric recipients of stem cell transplants: a retrospective single-center 15. Tichelli A, Gratwohl A. Vascular endothelium as ‘novel’ target of graft- cohort study. Pediatr Transplant. (2016) 20:1132–6. doi: 10.1111/petr.12831 versus-host disease. Best Pract Res Clin Haematol. (2008) 21:139–48. 6. Markey KA, MacDonald KPA, Hill GR. The biology of graft-versus-host doi: 10.1016/j.beha.2008.02.002 disease: experimental systems instructing clinical practice. Blood. (2014) 16. Pescador R, Capuzzi L, Mantovani M, Fulgenzi A, Ferrero ME. Defibrotide: 124:354–62. doi: 10.1182/blood-2014-02-514745 properties and clinical use of an old/new drug. Vascul Pharmacol. (2013) 7. Garnett C, Apperley JF, Pavlu J. Treatment and management of graft-versus- 59:1–10. doi: 10.1016/j.vph.2013.05.001 host disease: improving response and survival. Ther Adv Hematol. (2013) 17. Corbacioglu S, Cesaro S, Faraci M, Valteau-Couanet D, Gruhn B, 4:366. doi: 10.1177/2040620713489842 Rovelli A, et al. Defibrotide for prophylaxis of hepatic veno-occlusive 8. Carmona A, Díaz-Ricart M, Palomo M, Molina P, Pino M, Rovira disease in paediatric haemopoietic stem-cell transplantation: an open- M, et al. Distinct deleterious effects of cyclosporine and tacrolimus label, phase 3, randomised controlled trial. Lancet. (2012) 379:1301–9. and combined tacrolimus–sirolimus on endothelial cells: protective doi: 10.1016/S0140-6736(11)61938-7 effect of defibrotide. Biol Blood Marrow Transplant. (2013) 19:1439–45. 18. Palomo M, Diaz-Ricart M, Rovira M, Escolar G, Carreras E. Defibrotide doi: 10.1016/j.bbmt.2013.07.001 prevents the activation of macrovascular and microvascular endothelia 9. Palomo M, Diaz-Ricart M, Carbo C, Rovira M, Fernandez-Aviles F, Escolar caused by soluble factors released to blood by autologous hematopoietic G, et al. The release of soluble factors contributing to endothelial activation stem cell transplantation. Biol Blood Marrow Transplant. (2011) 17:497–506. and damage after hematopoietic stem cell transplantation is not limited to doi: 10.1016/j.bbmt.2010.11.019 the allogeneic setting and involves several pathogenic mechanisms. Biol Blood 19. Palomo M, Mir E, Rovira M, Escolar G, Carreras E, Diaz-Ricart M. Marrow Transplant. (2009) 15:537–46. doi: 10.1016/j.bbmt.2009.01.013 What is going on between defibrotide and endothelial cells? Snapshots 10. Palomo M, Diaz-Ricart M, Carbo C, Rovira M, Fernandez-Aviles F, Martine C, reveal the hot spots of their romance. Blood. (2016) 127:1719–27. et al. Endothelial dysfunction after hematopoietic stem cell transplantation: doi: 10.1182/blood-2015-10-676114 role of the conditioning regimen and the type of transplantation. Biol Blood 20. Tekgündüz E, Kaya AH, Bozdag SC, Koçubaba S, Kayikçi Ö, Namdaroglu Marrow Transplant. (2010) 16:985–93. doi: 10.1016/j.bbmt.2010.02.008 S, et al. Does defibrotide prophylaxis decrease the risk of acute graft versus Frontiers in Immunology | www.frontiersin.org 10 October 2019 | Volume 10 | Article 2339 Martinez-Sanchez et al. Defibrotide Prevents aGVHD Endothelial Dysfunction host disease following allogeneic hematopoietic cell transplantation? Transfus 37. He S, Cao Q, Qiu Y, Mi J, Zhang JZ, Jin M, et al. A new approach to Apher Sci. (2016) 54:30–4. doi: 10.1016/j.transci.2016.01.009 the blocking of alloreactive T cell-mediated graft-versus-host disease by in 21. Noseda G, Fragiacomo C, Ferrari D. Pharmacokinetics of defibrotide vivo administration of anti-CXCR3 neutralizing antibody. J Immunol. (2008) in healthy volunteers. Haemostasis. (1986) 16(Suppl. 1):26–30. 181:7581–92. doi: 10.4049/jimmunol.181.11.7581 doi: 10.1159/000215336 38. Wysocki CA, Jiang Q, Panoskaltsis-Mortari A, Taylor PA, McKinnon KP, Su L, 22. Tocchetti P, Tudone E, Marier JF, Marbury TC, Zomorodi K, Eller M. et al. Critical role for CCR5 in the function of donor CD4+ CD25+ regulatory Pharmacokinetic profile of defibrotide in patients with renal impairment. T cells during acute graft-versus-host disease. Blood. (2005) 106:3300–7. Drug Des Devel Ther. (2016) 10:2631–41. doi: 10.2147/DDDT.S112181 doi: 10.1182/blood-2005-04-1632 23. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch 39. Ueha S, Murai M, Yoneyama H, Kitabatake M, Imai T, Shimaoka T, T, et al. Fiji: an open-source platform for biological-image analysis. Nature et al. Intervention of MAdCAM-1 or fractalkine alleviates graft-versus- Methods. (2012) 9:676–82. doi: 10.1038/nmeth.2019 host reaction associated intestinal injury while preserving graft-versus-tumor 24. Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse effects. J Leukoc Biol. (2007) 81:176–85. doi: 10.1189/jlb.0306231 DC, et al. HMEC-1: establishment of an immortalized human 40. Richardson PG, Soiffer RJ, Antin JH, Uno H, Jin Z, Kurtzberg J, et al. microvascular endothelial cell line. J Invest Dermatol. (1992) 99:683–90. Defibrotide for the treatment of severe hepatic veno-occlusive disease and doi: 10.1111/1523-1747.ep12613748 multiorgan failure after stem cell transplantation: a multicenter, randomized, 25. Petersdorf EW. The major histocompatibility complex: a model for dose-finding trial. Biol Blood Marrow Transplant. (2010) 16:1005–17. understanding graft-versus-host disease. Blood. (2013) 122:1863–72. doi: 10.1016/j.bbmt.2010.02.009 doi: 10.1182/blood-2013-05-355982 41. Moore JM, Avery JW. Defibrotide: a Swiss Army knife intervention in the 26. Ferrara JLM, Levine JE, Reddy P, Holler E. Graft-versus-host disease. Lancet. battle against cerebral malaria. Arterioscler Thromb Vasc Biol. (2012) 32:541–4. (2009) 373:1550–61. doi: 10.1016/S0140-6736(09)60237-3 doi: 10.1161/ATVBAHA.111.242776 27. Blazar BR, Murphy WJ, Abedi M. Advances in graft-versus-host 42. Burgering BMT, Coffer PJ. Protein kinase B (c-Akt) in phosphatidylinositol- disease biology and therapy. Nat Rev Immunol. (2012) 12:443–58. 3-OH kinase signal transduction. Nature. (1995) 376:599–602. doi: 10.1038/nri3212 doi: 10.1038/376599a0 28. Penack O, Holler E, van den Brink MRM. Graft-versus-host disease: 43. Manning BD, Toker A. AKT/PKB signaling: navigating the network. Cell. regulation by microbe-associated molecules and innate immune receptors. (2017) 169:381–405. doi: 10.1016/j.cell.2017.04.001 Blood. (2010) 115:1865–72. doi: 10.1182/blood-2009-09-242784 44. Herrero-Sánchez MC, Rodríguez-Serrano C, Almeida J, San Segundo L, 29. Riesner K, Shi Y, Jacobi A, Kräter M, Kalupa M, McGearey A, et al. Initiation Inogés S, Santos-Briz A, et al. Targeting of PI3K/AKT/MTOR pathway to of acute graft-versus-host disease by angiogenesis. Blood. (2017) 129:2021–32. inhibit T cell activation and prevent graft-versus-host disease development. doi: 10.1182/blood-2016-08-736314 J Hematol Oncol. (2016) 9:113. doi: 10.1186/s13045-016-0343-5 30. Penack O, Socié G, van den Brink MRM. The importance 45. Long J, Chang L, Shen Y, Gao W-H, Wu Y-N, Dou H-B. valproic of neovascularization and its inhibition for allogeneic acid ameliorates graft-versus-host disease by downregulating Th1 and hematopoietic stem cell transplantation. Blood. (2011) 117:4181–9. Th17 cells. J Immunol. (2015) 195:1849–57. doi: 10.4049/jimmunol.15 doi: 10.1182/blood-2010-10-312934 00578 31. Giannotta M, Trani M, Dejana E. VE-cadherin and endothelial adherens 46. Zarbock A, Ley K. Mechanisms and consequences of neutrophil interaction junctions: active guardians of vascular integrity. Dev Cell. (2013) 26:441–54. with the endothelium. Am J Pathol. (2008) 172:1–7. doi: 10.2353/ajpath.2008.0 doi: 10.1016/j.devcel.2013.08.020 70502 32. Wallez Y, Vilgrain I, Huber P. Angiogenesis: the VE-cadherin switch. Trends in Cardiovasc Med. (2006) 16:55–9. doi: 10.1016/j.tcm.2005.11.008 Conflict of Interest: This research has been partially supported by Jazz 33. Medinger M, Tichelli A, Bucher C, Halter J, Dirnhofer S, Rovo A, Pharmaceuticals plc/Gentium Inc. MP declares conflict of interest with Jazz et al. GVHD after allogeneic haematopoietic SCT for AML: angiogenesis, Pharmaceuticals plc/Gentium Inc. in the form of speaker’s fee for symposiums. vascular endothelial growth factor and VEGF receptor expression in EC, OP, and MD-R received research funding and speaker’s fees from Jazz the BM. Bone Marrow Transplant. (2013) 48:715–21. doi: 10.1038/bmt. Pharmaceuticals. 2012.200 34. Benimetskaya L, Wu S, Voskresenskiy AM, Echart C, Zhou J-F, Shin J, et al. The remaining authors declare that the research was conducted in the absence of Angiogenesis alteration by defibrotide: implications for its mechanism of any commercial or financial relationships that could be construed as a potential action in severe hepatic veno-occlusive disease. Blood. (2008) 112:4343–52. conflict of interest. doi: 10.1182/blood-2008-04-149682 35. Wang X, Pan B, Hashimoto Y, Ohkawara H, Xu K, Zeng L, et al. Copyright © 2019 Martinez-Sanchez, Hamelmann, Palomo, Mir, Moreno-Castaño, Defibrotide stimulates angiogenesis and protects endothelial cells from Torramade, Rovira, Escolar, Cordes, Kalupa, Mertlitz, Riesner, Carreras, Penack and calcineurin inhibitor-induced apoptosis via upregulation of AKT/Bcl-XL. Diaz-Ricart. This is an open-access article distributed under the terms of the Creative Thromb Haemost. (2018) 118:161–73. doi: 10.1160/TH17-04-0275 Commons Attribution License (CC BY). The use, distribution or reproduction in 36. Koehl GE, Geissler EK, Iacobelli M, Frei C, Burger V, Haffner S, et al. other forums is permitted, provided the original author(s) and the copyright owner(s) Defibrotide: an endothelium protecting and stabilizing drug, has an anti- are credited and that the original publication in this journal is cited, in accordance angiogenic potential in vitro and in vivo. Cancer Biol Ther. (2007) 6:686–90. with accepted academic practice. No use, distribution or reproduction is permitted doi: 10.4161/cbt.6.5.3959 which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 11 October 2019 | Volume 10 | Article 2339