Transplant Toxicities

Thrombomodulin blocks calcineurin inhibitor-induced vascular permeability via inhibition of Src/VE-cadherin axis

Article metrics

Abstract

Recombinant human soluble thrombomodulin (rTM) counteracted capillary leakage and alleviated edema in individuals with sinusoidal obstruction syndrome and engraftment syndrome after hematopoietic stem cell transplantation. We previously showed that rTM increased levels of antiapoptotic protein Mcl-1 and protected endothelial cells from calcineurin inhibitor cyclosporine A (CsA)-induced apoptosis. However, the molecular mechanisms by which rTM enhances barrier function in vascular endothelial cells remain unknown. Here we show that exposure of vascular endothelial EA.hy926 cells to CsA induced phosphorylation of Src/vascular endothelial cadherin (VE-cadherin) and translocation of VE-cadherin from cell surface to cytoplasm, resulting in an increase in vascular permeability. In addition, CsA increased production of inflammatory cytokines, including interleukin (IL)-1β and IL-6, associated with an increase in nuclear levels of nuclear factor-κB (NF-κB) which also enhanced vascular permeability. Importantly, the fourth and fifth regions of epidermal growth factor-like domain of TM (TME45) attenuated CsA-induced p-Src/VE-cadherin and vascular permeability in parallel with a decrease in nuclear levels of NF-κB and cytokine production in EA.hy926 cells. In conclusion, TM, especially TME45, maintains vascular integrity, at least in part, via Src signaling.

Introduction

Adherens junctions regulate vascular permeability in which cell surface expressed vascular endothelial cadherin (VE-cadherin) plays a critical role by interacting with β-catenin through its cytoplasmic tail.1 An increase in phosphorylation of VE-cadherin (p-VE-cadherin) results in disruption of VE-cadherin/β-catenin binding and causes disassembly of endothelial adherens junctions.1 Src is one of the regulators of p-VE-cadherin; bradykinin and histamine causes phosphorylation of Src and VE-cadherin, resulting in internalization of VE-cadherin and its degradation by ubiqutination. Blockade of Src attenuates phosphorylation of VE-cadherin and inhibits vascular permeability.2

Nuclear factor κB (NF-κB) is one of transcription factors that play a role in inflammation as well as cell proliferation by regulation of the expression of numerous genes.3, 4 The inactive form of NF-κB binds to its scaffold protein IκBα and stays in the cytoplasm. Once IκBα is phosphorylated by IκBα kinase, it is degraded and releases NF-κB. The liberated NF-κB translocates into the nucleus and activates its target genes.

Cyclosporine A (CsA), a calcineurin inhibitor, is widely used for prophylaxis or treatment of GVHD or graft rejection. However, its clinical use is often limited by nephrotoxicity, systemic arterial hypertension and posterior reversible encephalopathy syndrome that is associated with capillary leakage.5, 6 Furthermore, CsA and tacrolimus (FK506), another calcineurin inhibitor, are supposed to cause transplantation-associated thrombotic microangiopathy (TA-TMA) as a result of endothelial injury in hematopoietic stem cell transplantation recipients.7 Withdrawal of the calcineurin inhibitors is effective treatment for TMA in some cases.8

Thrombomodulin (TM), a receptor for thrombin, is mainly expressed on the cell surface of vascular endothelial cells and plays an important role in the coagulation system by generating activated protein C.9, 10, 11, 12 Curiously, the lectin-like domain of TM possesses anti-inflammatory functions; this region inactivates NF-κB and mitogen-activated protein kinase pathways and inhibits adhesion of neutrophils to endothelial cells.13 In addition, this domain degrades the inflammatory mediator high mobility group box 1 that is released from necrotic cells or activated inflammatory cells.14

Recombinant human soluble thrombomodulin (rTM), comprising the active, extracellular domain of TM, was approved for the treatment of disseminated intravascular coagulation in Japan in 2008.15 We have since successfully treated individuals with disseminated intravascular coagulation complicated by lethal hematopoietic stem cell transplantation-associated complications, including TA-TMA and sinusoidal obstructive syndrome, as well as engraftment syndrome.16, 17, 18 The pathogenesis of these transplantation-associated complications is based on the endothelial cell insults caused by calcineurin inhibitors, exaggerated production of cytokines, anti-leukemia agents and so on. We also showed that the use of rTM for treatment of coagulopathy developed after hematopoietic stem cell transplantation improved overall survival of patients by improving underlying diseases, including sinusoidal obstruction syndrome and TA-TMA, compared with those who did not receive rTM.19 The in vitro study found that rTM protected human umbilical vessel cells from CsA-induced apoptosis associated with an increase in levels of antiapoptotic Mcl-1 proteins.20 Interestingly, this cytoprotective effect was mediated by the fourth and fifth regions of TM (TME45) in an activated protein C-independent manner.20

This study explored the molecular mechanisms by which CsA caused insult to endothelial cells and increased vascular permeability. Also possible roles of rTM in vascular barrier function were investigated by utilizing human umbilical vein EA.hy926 cells.

Materials and methods

Cells

EA.hy926 cells were purchased from Lonza Walkersville Inc. (Walkersville, MD, USA) and cultured with DMEM culture medium supplemented with 10% fetal bovine serum.

Reagents

CsA, FK506 and IL-1β were purchased from Sigma Chemical (St Louis, MO, USA). IL-6 was purchased from PeproTech (Rocky Hill, NJ, USA). Dehydroxymethylepoxyquinomicin (DHMEQ), an inhibitor of NF-κB, was synthesized as previously described and dissolved in DMSO.21 Tocilizumab, a humanized anti-human IL-6R monoclonal antibody of the IgG1κ subtype, was obtained from Chugai Pharmaceutical (Tokyo, Japan). rTM was provided by Asahi Kasei Pharma (Tokyo, Japan).

RNA isolation and real-time RT-PCR

RNA isolation and cDNA preparation followed our previous study.22 Primers for PCR are shown in Table 1. RT-PCR was repeated at least three times

Table 1 RT-PCR primers

ELISA

Culture media of human umbilical vessel cells were collected and applied to ELISA kit to measure the levels of IL-1β and IL-6 according to the manufacturer’s protocol (R&D system, Minneapolis, MN, USA). This assay was repeated at least three times in a duplicate plate.

Western blot analysis

Western blot analysis was performed as described previously.23, 24 Anti-IκBα (Cell Signaling Technology, #2859, Beverly, MA, USA), the anti-p65 subunit of NF-κB (Santa Cruz Biotechnology, sc-373, Santa Cruz, CA, USA), anti-histone H1 (Santa Cruz Biotechnology, sc-8030), anti-p-VE-cadherin (Tyr 731) (Bioworld Technology, Inc., BS4242, St Louis Park, MN, USA), anti-VE-cadherin (Cell Signaling Technology, #2500) and anti-GAPDH (Abcam, ab181602, Tokyo, Japan) antibodies were used. Western blot analysis was repeated at least three times.

Immunoprecipitation

Whole-cell lysates were extracted from cells and immunoprecipitated with an anti-Src antibody and subjected to western blot analyses. The membrane was probed with an anti-p-Src (Y418) (Abcam) antibody.

Vascular permeability assay in vitro

Effects of test agents on vascular permeability were measured by a vascular permeability assay kit (Millipore, Billerica, MA, USA) as described previously.20, 25 Briefly, EA.hy926 cells were plated onto collagen-coated inserts and cultured for 72 h to allow them to form a tight monolayer. After 24 h culture without fetal calf serum, cells were exposed to test agents for 12 h and then FITC-Dextran was added to the culture. The extent of vascular permeability was quantified by measuring the fluorescence of bottom plate solution affected by permeated FITC-Dextran.

Vascular permeability assay in vivo

All murine experiments were approved by the Institutional Animal Care and Use Committee of Kochi University. All animal studies were performed in accordance with the ARRIVE guidelines for reporting experiments involving animals.26 Specific pathogen-free 6-week-old female ICR mice were purchased from Japan SLC, Inc. (Tokyo, Japan) and were maintained in a temperature-controlled environment with a 12 h light/dark cycle. Mice (n=3) were IV injected with 10 mL/kg of 1% Evans blue dye solution in normal saline, followed by SC injection of 0.1% BSA (10 μL) or CsA (10−4 m in 10 μL 0.1% BSA) either with or without rTM (10 μg in 10 μL 0.1% BSA). After 30 min, vascular leakage was assessed. The skin area with blue color was quantified by imageJ software (Wayne Rasband, NIH).

Fluorescence imaging and analysis

EA.hy926 cells were stained with the anti-VE-cadherin and anti-mouse PE-Cy5-linked secondary antibody. Images were recorded with a Zeiss LSM 510 confocal laser scanning microscope.

TME45 protein

The fourth and fifth regions of epidermal growth factor-like domain of TM was designed and cloned into the FLAG-TEV–tagged pcDNA3.1/V5-His-TOPO expression vector (Invitrogen, Tokyo, Japan), as previously described.20 TME45 proteins were produced by COS-1 cells and purified, as previously described.20

VE-cadherin translocation assay

After treatment of the cells with test agents, membrane and cytosolic fractions were prepared as previously described.27 Proteins extracted from cytosolic and membrane fractions were subjected to western blot analysis with anti-VE-cadherin antibody.

Statistical analysis

Statistical analysis of the differences between two groups under multiple conditions was performed using one-way ANOVA followed by Bonferroni multiple comparison tests using PRISM statistical analysis software (GraphPad Software ver. 6, San Diego, CA, USA).

Results

CsA increased vascular permeability in association with VE-cadherin translocation in EA.hy926 cells

The exposure of EA.hy926 cells to CsA increased their vascular permeability in a dose-dependent manner (Figure 1a). We next examined whether CsA-mediated vascular permeability was associated with internalization of VE-cadherin in endothelial cells. Exposure of EA.hy926 cells to CsA (2.5–10 μm, 48 h) increased the levels of p-VE-cadherin in a dose-dependent manner without affecting the total amount of VE-cadherin (Figure 1b). Furthermore, as expected, CsA stimulated translocation of VE-cadherin from cell surface to the cytosolic fraction (Figure 1c), suggesting that CsA-induced vascular permeability might be linked with phosphorylation and translocation of VE-cadherin in EA.hy926 cells.

Figure 1
figure1

CsA increases vascular permeability in association with VE-cadherin translocation in EA.hy926 cells. Vascular permeability assay (a). EA.hy926 monolayer was exposed to CsA (2.5–10 μm) for 12 h and then FITC-Dextran was added to the culture media. The extent of vascular permeability was quantified by measuring the fluorescence of bottom plate solution affected by permeated FITC-Dextran. Results represent fold of control well without agents. CsA causes phosphorylation and internalization of VE-cadherin in EA.hy926 cells. Western blot analysis (b, c). EA.hy926 cells were plated in six-well plates and cultured with CsA (2.5–10 μm). After 48 h, whole-cell lysates (b), membrane and cytoplasmic proteins (c) were extracted and subjected to western blot analyses. The membrane was sequentially probed with the indicated antibodies. Thirty micrograms of proteins were loaded to each lane. Levels of GAPDH were measured as a loading control. CsA increases production of inflammatory cytokines. Real-time RT-PCR (d). EA.hy926 cells were plated in 12-well plates and cultured in the presence of CsA (2.5–10 μm) (d, upper panel). After 48 h, RNA was extracted and subjected to real-time RT-PCR to measure the levels of indicated cytokines. Transcripts of cytokines were normalized according to the amount of 18S transcript. (d, lower panel) ELISA. EA.hy926 cells were plated in 12-well plates and cultured with CsA (2.5–10 μm). After 48 h, the levels of IL-1β and IL-6 in the culture media of EA.hy926 cells were measured by ELISA. CsA decreases cytoplasmic levels of IκBα and increases nuclear levels of NF-κB in EA.hy926 cells. Western blot analysis (e). EA.hy926 cells were exposed to various concentrations of CsA (2.5–10 μm). After 48 h, cytoplasmic and nuclear proteins were extracted and subjected to western blot analysis. Levels of GAPDH or Histone H1 were measured as a loading control. The pictures of western blots represent one of the three experiments performed independently with similar results. Data in the bar graph represent means±s.d. of at least four experiments. *P<0.05, **P<0.01.

CsA increased production of IL-1β and IL-6 in EA.hy926 cells

We examined whether CsA increased inflammatory cytokines such as IL-1β and IL-6, which were shown to induce vascular permeability in association with disruption of VE-cadherin cell-surface localization in human dermal microvascular endothelial cells (HMVEC-d).28 As shown in Figure 1d (upper panel), exposure of EA.hy926 cells to CsA (2.5–10 μm, 48 h) significantly increased levels of IL-1β and IL-6 in a dose-dependent manner as measured by real-time RT-PCR. CsA (2.5–5 μm, 48 h) also increased levels of IL-1α, IFN-α and TNF-α in a dose-dependent manner; however, the highest dose of CsA (10 μm) was not able to increase the levels of these cytokines (figure not shown). Likewise, exposure of EA.hy926 cells to FK506 (1.25–5 μm, 48 h) also increased levels of IL-1β and IL-6 in a dose-dependent manner (figure not shown). In parallel with an increase in mRNA levels of IL-1β and IL-6 in EA.hy926 cells, production of IL-1β and IL-6 proteins in culture media was significantly increased when EA.hy926 cells were cultured in the presence of CsA (Figure 1d, lower panel). All of these observations suggested that CsA-mediated production of inflammatory cytokines such as IL-1β and IL-6 might be involved in increasing vascular permeability in EA.hy926 cells.

CsA increased nuclear levels of NF-κB in EA.hy926 cells

Expression of cytokine genes is intimately regulated by NF-κB. To elucidate the effect of CsA on activity of NF-κB, we measured the levels of NF-κB proteins in the nuclei and the levels of IκBα proteins in the cytoplasm in EA.hy926 cells after exposure to CsA. Exposure of EA.hy926 cells to CsA increased nuclear levels of NF-κB in a dose-dependent manner in association with a decrease in levels of IκBα in cytoplasm (Figure 1e), suggesting that CsA stimulated nuclear translocation of NF-κB.

Blockade of NF-κB inhibited CsA-induced production of cytokines and attenuated vascular permeability

We next explored whether blockade of nuclear translocation of NF-κB by DHMEQ blocked CsA-mediated production of cytokines and loss of cell-surface VE-cadherin localization in EA.hy926 cells. As shown in Figure 2a, DHMEQ inhibited nuclear translocation of NF-κB stimulated by CsA. Furthermore, DHMEQ potently blocked CsA-induced production of IL-1β and IL-6 in EA.hy926 cells (Figure 2b). Importantly, CsA-induced vascular permeability and loss of cell-surface VE-cadherin were attenuated in the presence of DHMEQ (Figures 2c and d). To elucidate the effect of cytokines on EA.hy926 cells, we performed vascular permeability assays. IL-1β and IL-6 increased vascular permeability in a dose-dependent manner (figure not shown). To further explore whether blockade of cytokine signaling pathways, such as IL-6 signaling, hampered CsA induced vascular permeability, we inhibited the signaling mediated by IL-6 by a humanized anti-human IL-6R monoclonal antibody tocilizumab (10 μg/mL, 4 h), and found that CsA-induced vascular permeability and loss of VE-cadherin on cell surface of EA.hy926 cells were attenuated in the presence of tocilizumab (Figures 2e and f), suggesting that CsA increased endothelial cell permeability at least in part via NF-κB-mediated cytokine production.

Figure 2
figure2

Blockade of NF-κB inhibits CsA-stimulated cytokine production and vascular permeability in EA.hy926 cells. Western blot analysis (a). EA.hy926 cells were plated in six-well plates and cultured with either CsA (10 μm) and/or DMHEQ (1 μg/mL). After 48 h, cytoplasmic and nuclear proteins were extracted and subjected to western blot analysis. The membrane was sequentially probed with the indicated antibodies. Levels of GAPDH or Histone H1 were measured as a loading control. Real-time RT-PCR (b, upper panel). EA.hy926 cells were plated in 12-well plates and cultured in the presence of CsA and/or DHMEQ. After 48 h, RNA was extracted and subjected to real-time RT-PCR to measure the levels of indicated cytokines. Transcripts of cytokines were normalized according to the amount of 18S transcript. ELISA (b, lower panel). The levels of IL-1β and IL-6 in the culture media of EA.hy926 cells were measured by ELISA. Vascular permeability assay (c). EA.hy926 monolayer was cultured with either CsA (10 μm) and/or DMHEQ (1 μg/mL) for 12 h and then FITC-Dextran was added to the culture media. The extent of vascular permeability was quantified by measuring the fluorescence of bottom plate solution affected by permeated FITC-Dextran. Results represent fold of control well without agents. Western blot analysis (d). EA.hy926 cells were plated in six-well plates and cultured with CsA and/or DHMEQ. After 48 h, both membrane and cytoplasmic proteins were extracted and subjected to western blot analysis. The membrane was sequentially probed with the indicated antibodies. Thirty micrograms of proteins were loaded to each lane. CsA, cyclosporine A. Vascular permeability assay (e). EA.hy926 cell monolayer was exposed to tocilizumab (10 μg/mL). After 4 h, cells were washed with PBS twice and exposed to CsA (10 μm). After 12 h, FITC-Dextran was added to the culture media. The extent of vascular permeability was quantified by measuring the fluorescence of bottom plate solution affected by permeated FITC-Dextran. **P<0.01. (f) Western blot analysis. EA.hy926 cells were exposed to tocilizumab (10 μg/mL). After 4 h, cells were washed with PBS for twice and exposed to CsA (10 μm). After 48 h, proteins were extracted and subjected to western blot analysis. The membrane was sequentially probed with the indicated antibodies. Thirty microgramsg of proteins were loaded to each lane. The pictures of western blots represent one of the three experiments performed independently with similar results. Data in the bar graph represent means±s.d. of at least four experiments. **P<0.01.

rTM attenuated CsA-induced loss of cell-surface localization of VE-cadherin

We assessed whether rTM affected cell-surface localization of VE-cadherin in endothelial cells. As expected, rTM (30 ng/mL, 48 h) counteracted CsA-induced internalization of VE-cadherin in EA.hy926 cells (Figures 3a and b). Importantly, compared with SC injection of BSA, SC injection of CsA induced capillary leakage in ICR mice (Figure 3c, middle arrow). Of note, injection of rTM in combination with CsA apparently alleviated CsA-induced capillary leakage (Figure 3c, right arrow). We next explored the effect of rTM on the levels of cytokines produced by EA.hy926 cells after exposure to CsA. Exposure of EA.hy926 cells to rTM (30 ng/mL, 48 h) potently inhibited CsA-stimulated production of both IL-1β and IL-6 in these cells (Figure 3d). In parallel, a decrease in cytoplasmic levels of IκBα and an increase in nuclear levels of NF-κB in EA.hy926 cells after exposure to CsA were attenuated when these cells were exposed to CsA and rTM in combination (Figure 3e).

Figure 3
figure3

rTM inhibits CsA-induced loss of cell-surface VE-cadherin in EA.hy926 cells. Immunocytochemistry (a). EA.hy926 cells were stained with the anti-VE-cadherin and anti-mouse-PE-Cy5-linked secondary antibodies, counterstained with DAPI and examined under a microscope. Western blot analysis (b). EA.hy926 cells were cultured with either CsA (10 μm) and/or rTM (30 ng/mL). After 48 h, both membrane and cytoplasmic proteins were extracted and subjected to western blot analysis. The membrane was sequentially probed with the indicated antibodies. Thirty micrograms of proteins were loaded to each lane. rTM alleviates capillary leakage in vivo (c). ICR mice were IV injected with 10 mL/kg of 1% Evans blue dye solution, followed by SC injection of 0.1% BSA (10 μL) or CsA (10−4 m in 10 μL 0.1% BSA) either with or without rTM (10 μg in 10 μL 0.1% BSA). After 30 min, vascular leakage was assessed. The picture represents one of the three experiments performed independently with similar results. CsA, cyclosporine. rTM inhibits NF-κB and cytokine production. ELISA (d). EA.hy926 cells were cultured with either CsA (10 μm) and/or rTM (30 ng/mL). After 48 h, the levels of IL-1β and IL-6 in the culture media of EA.hy926 cells were measured by ELISA. **P<0.01. Western blot analysis (e). EA.hy926 cells were exposed to either CsA (10 μm) and/or rTM (30 ng/mL). Cells were harvested at the indicated time point and both cytoplasmic and nuclear proteins were extracted and subjected to western blot analysis. The membrane was probed sequentially with the indicated antibodies. Levels of GAPDH or Histone H1 were measured as a loading control. The pictures of western blots represent one of the three experiments performed independently with similar results. Data in the bar graph represent means±s.d. of at least four experiments. **P<0.01. CsA=cyclosporine A; rTM=recombinant human soluble thrombomodulin. A full color version of this figure is available online at the Bone Marrow Transplantation website.

Fourth and fifth regions of EGF-like domain of TM inhibited NF-κB via inhibition of Src

We explored whether TME45 affected the ability of CsA to increase vascular permeability. As expected, TME45 counteracted CsA-induced vascular permeability in EA.hy926 cells (figure not shown). Previous study showed that Src is also one of the regulators of p-VE-cadherin; blockade of Src attenuates phosphorylation of VE-cadherin and inhibits vascular permeability.2 We examined whether TME45 prevented CsA-induced vascular permeability via affecting p-Src/VE-cadherin in EA.hy926 cells. As shown in Figure 4a, the levels of p-Src and p-VE-cadherin were potently increased after exposure of EA.hy926 cells to CsA. Interestingly, the phosphorylation of Src and VE-cadherin in EA.hy926 cells caused by CsA was attenuated when these cells were exposed to CsA in combination with TME45 (30 ng/mL). Intriguingly, CsA-mediated upregulation of NF-κB in the nucleus and downregulation of IκBα in the cytoplasm of EA.hy926 cells were hampered in the presence of TME45 (30 ng/mL) (Figure 4). Moreover, CsA-stimulated production of cytokines in EA.hy926 cells was also blocked in the presence of TME45 (Figure 4). We expected that TME45 inhibited NF-κB and cytokine production via inhibition of Src in endothelial cells. To test this hypothesis, we exposed EA.hy926 cells to an Src inhibitor dasatinib (10 nm), which potently inhibited phosphorylation of Src and VE-cadherin caused by CsA. CsA-stimulated expression of IL-1β and IL-6 was significantly attenuated in association with a decrease in nuclear levels of NF-κB in EA.hy926 cells (Supplementary Figure 1).

Figure 4
figure4

TME45 inhibits phosphorylation of Src/VE-cadherin, NF-κB and cytokine production in EA.hy926 cells. EA.hy926 cells were exposed to either CsA (10 μm) and/or TME45 (30 ng/mL). After 48 h, cells were harvested and whole-cell lysates as well as each fraction of proteins were extracted and subjected to immunoprecipitation and western blot analysis, respectively. The membrane was sequentially probed with the indicated antibodies. Levels of α-tubulin, GAPDH or Histone H1 were measured as a loading control. At the same time, RNA was extracted from cells and subjected to real-time RT-PCR to measure the levels of the indicated cytokines. Transcripts of cytokines were normalized according to the amount of 18S transcript. Data represent means±s.d. of the four independent experiments performed in duplicate (n=4). **P<0.01. TME45=the fourth and fifth regions of EGF-like domain of thrombomodulin.

Discussion

This study found that CsA caused nuclear translocation of NF-κB, which stimulated production of IL-1β and IL-6 in EA.hy926 cells (Figure 1). These cytokines probably induced phosphorylation of VE-cadherin and its internalization, resulting in an increase in vascular permeability in EA.hy926 cells. In fact, exposure of EA.hy926 cells to both IL-1β and IL-6 increased vascular permeability in a dose-dependent manner and downregulation of these cytokines by an inhibitor of NF-κB blocked CsA-induced vascular permeability in association with inhibition of VE-cadherin internalization in EA.hy926 cells (Figures 2b–d). In addition, blockade of IL-6 signaling by tocilizumab counteracted CsA-induced loss of cell-surface localization of VE-cadherin and significantly attenuated vascular permeability (Figures 2e and f). These observations suggested the important roles of inflammatory cytokines in CsA-induced vascular permeability. Similarly, previous studies showed that IL-1β and IL-6 secreted by A2058 melanoma cells disrupted the VE-cadherin junction in human umbilical vessel cell.29

It is noticeable that CsA-induced nephrotoxicity was attenuated by provinol, a red wine polyphenol, via inhibition of reactive oxygen species, inducible nitric oxide synthase and NF-κB expression.30 Furthermore, CsA augmented IL-6 production in human umbilical vessel cells which in turn inhibited their proliferation.31 These observations suggest that NF-κB and cytokines may be promising therapeutic targets to maintain vascular integrity.

The present study found that rTM decreased CsA-induced nuclear levels of NF-κB and cytokine production in EA.hy926 cells (Figure 3). We expected that inhibition of NF-κB by rTM was mediated by an anti-inflammatory function of the lectin-like domain of TM, as it inhibited adhesion of neutrophils to vascular cells via inhibition of NF-κB.13 Contrary to our expectation, surprisingly, TME45, a mutant form of TM lacking the lectin-like domain of TM, blocked activation of this nuclear transcription factor (Figure 4). Activated protein C, produced by TM after binding to thrombin, possesses anti-inflammatory function; however, TME45 was not able to produce activated protein C.20 The inhibitor of Src, dasatinib, blocked CsA-induced phosphorylation of Src/VE-cadherin in parallel with inhibition of NF-κB and cytokine production (Supplementary Figure 1). TME45 also blocked CsA-induced p-Src/VE-cadherin in EA.hy926 cells (Figure 4). These observations suggested that TME45 might inactivate NF-κB via inhibition of Src (Figure 5).

Figure 5
figure5

The schema depicts the possible mechanisms by which CsA causes vascular permeability. CsA=cyclosporine A; p-VE-cadherin=phosphorylated forms of vascular endothelial cadherin; p-Src=phosphorylated forms of Src; TME45=the fourth and fifth regions of EGF-like domain of thrombomodulin. A full color version of this figure is available online at the Bone Marrow Transplantation website.

Intriguingly, forced expression of TM in melanoma A2058 cells that lack endogenous TM induced their clustered colony formation. On the other hand, when A2058 cells were transfected with the lectin-like domain-deleted TM, these cells grew singly without cell-to-cell contact. Further study showed that anti-lectin-like domain antibody disrupted the close clustering of the endogenous TM-expressing keratinocytes. These observations suggest that the lectin-like domain of TM plays a crucial role in maintaining the integrity of cell-to-cell contact, although molecular mechanisms whereby the lectin-like domain of TM affected adhesion molecules on cancer cells remain unknown.32

Taken together, CsA increases vascular permeability in association with phosphorylation of Src/VE-cadherin, activation of NF-κB and an increase in expression of cytokines in EA.hy926 cells. rTM, especially TME45, counteracted CsA-induced vascular permeability by inhibition of phosphorylation of Src/VE-cadherin, NF-κB and cytokine production in vascular endothelial cells.

References

  1. 1

    Dejana E, Orsenigo F, Lampugnani MG . The role of adherens junctions and VE-cadherin in the control of vascular permeability. J Cell Sci 2008; 121: 2115–2122.

  2. 2

    Orsenigo F, Giampietro C, Ferrari A, Corada M, Galaup A, Sigismund S et al. Phosphorylation of VE-cadherin is modulated by haemodynamic forces and contributes to the regulation of vascular permeability in vivo. Nat Commun 2012; 3: 1208.

  3. 3

    Hayden MS, Ghosh S . Shared principles in NF-kappaB signaling. Cell 2008; 132: 344–362.

  4. 4

    Vallabhapurapu S, Karin M . Regulation and function of NFkappaB transcription factors in the immune system. Annu Rev Immunol 2009; 27: 693–733.

  5. 5

    Takahata M, Hashino S, Izumiyama K, Chiba K, Suzuki S, Asaka M . Cyclosporin A-induced encephalopathy after allogeneic bone marrow transplantation with prevention of graft-versus-host disease by tacrolimus. Bone Marrow Transplant 2001; 28: 713–715.

  6. 6

    Sarkodee-Adoo C, Sotirescu D, Sensenbrenner L, Rapoport AP, Cottler-Fox M, Tricot G et al. Thrombotic microangiopathy in blood and marrow transplant patients receiving tacrolimus or cyclosporine A. Transfusion 2003; 43: 78–84.

  7. 7

    Minn AY, Fisher PG, Barnes PD, Dahl GV . A syndrome of irreversible leukoencephalopathy following pediatric allogeneic bone marrow transplantation. Pediatr Blood Cancer 2007; 48: 213–217.

  8. 8

    Furlong T, Storb R, Anasetti C, Appelbaum FR, Deeg HJ, Doney K et al. Clinical outcome after conversion to FK 506 (tacrolimus) therapy for acute graft-versus-host disease resistant to cyclosporine or for cyclosporine-associated toxicities. Bone Marrow Transplant 2000; 26: 985–991.

  9. 9

    Ikezoe T . Thrombomodulin/activated protein C system in septic disseminated intravascular coagulation. J Intensive Care 2015; 3: 1.

  10. 10

    Suzuki K, Kusumoto H, Deyashiki Y, Nishioka J, Maruyama I, Zushi M et al. Structure and expression of human thrombomodulin, a thrombin receptor on endothelium acting as a cofactor for protein C activation. EMBO J 1987; 6: 1891–1897.

  11. 11

    Dittman WA, Majerus PW . Structure and function of thrombomodulin: a natural anticoagulant. Blood 1990; 75: 329–336.

  12. 12

    Dahlback B, Villoutreix BO . The anticoagulation protein C pathway. FEBS Lett 2005; 579: 3310–3316.

  13. 13

    Conway EM, Van de Wouwer M, Pollefeyt S, Jurk K, Van Aken H, De Vriese A et al. The lectin-like domain of thrombomodulin confers protection from neutrophil-mediated tissue damage by suppressing adhesion molecule expression via nuclear factor kappaB and mitogen-activated protein kinase pathways. J Exp Med 2002; 196: 565–577.

  14. 14

    Abeyama K, Stern DM, Ito Y, Kawahara K, Yoshimoto Y, Tanaka M et al. The N-terminal domain of thrombomodulin sequesters high-mobility group-B1 protein, a novel antiinflammatory mechanism. J Clin Invest 2005; 115: 1267–1274.

  15. 15

    Ikezoe T . Pathogenesis of disseminated intravascular coagulation in patients with acute promyelocytic leukemia, and its treatment using recombinant human soluble thrombomodulin. Int J Hematol 2014; 100: 27–37.

  16. 16

    Sakai M, Ikezoe T, Bandobashi K, Togitani K, Yokoyama A . Successful treatment of transplantation-associated thrombotic microangiopathy with recombinant human soluble thrombomodulin. Bone Marrow Transplant 2010; 45: 803–805.

  17. 17

    Ikezoe T, Togitani K, Komatsu N, Isaka M, Yokoyama A . Successful treatment of sinusoidal obstructive syndrome after hematopoietic stem cell transplantation with recombinant human soluble thrombomodulin. Bone Marrow Transplant 2010; 45: 783–785.

  18. 18

    Ikezoe T, Takeuchi A, Taniguchi A, Togitani K, Yokoyama A . Recombinant human soluble thrombomodulin counteracts capillary leakage associated with engraftment syndrome. Bone Marrow Transplant 2011; 46: 616–618.

  19. 19

    Ikezoe T, Takeuchi A, Chi S, Takaoka M, Anabuki K, Kim T et al. Effect of recombinant human soluble thrombomodulin on clinical outcomes of patients with coagulopathy after hematopoietic stem cell transplantation. Eur J Haematol 2013; 91: 442–447.

  20. 20

    Ikezoe T, Yang J, Nishioka C, Honda G, Furihata M, Yokoyama A . Thrombomodulin protects endothelial cells from a calcineurin inhibitor-induced cytotoxicity by upregulation of extracellular signal-regulated kinase/myeloid leukemia cell-1 signaling. Arterioscler Thromb Vasc Biol 2012; 32: 2259–2270.

  21. 21

    Suzuki Y, Sugiyama C, Ohno O, Umezawa K . Preparation and biological activities of optically active dehydroxymethylepoxyquinomicin, a novel NF-κB inhibitor. Tetrahedron 2004; 60: 7061–7066.

  22. 22

    Ikezoe T, Tanosaki S, Krug U, Liu B, Cohen P, Taguchi H et al. Insulin-like growth factor binding protein-3 antagonizes the effects of retinoids in myeloid leukemia cells. Blood 2004; 104: 237–242.

  23. 23

    Nishioka C, Ikezoe T, Jing Y, Umezawa K, Yokoyama A . DHMEQ, a novel nuclear factor-kappaB inhibitor, induces selective depletion of alloreactive or phytohaemagglutinin-stimulated peripheral blood mononuclear cells, decreases production of T helper type 1 cytokines, and blocks maturation of dendritic cells. Immunology 2008; 124: 198–205.

  24. 24

    Ikezoe T, Yang Y, Heber D, Taguchi H, Koeffler HP . PC-SPES: potent inhibitor of nuclear factor-kappa B rescues mice from lipopolysaccharide-induced septic shock. Mol Pharmacol 2003; 64: 1521–1529.

  25. 25

    Zhang Y, Furumura M, Morita E . Distinct signaling pathways confer different vascular responses to VEGF 121 and VEGF 165. Growth Factors 2008; 26: 125–131.

  26. 26

    Kilkenny C, Browne WJ, Cuthill IC, Emerson M, Altman DG . Animal research: reporting in vivo experiments: The ARRIVE guidelines. PLoS Biol 2010; 8: e1000412.

  27. 27

    Côté M, Payet MD, Dufour MN, Guillon G, Gallo-Payet N . Association of the G protein alpha(q)/alpha11-subunit with cytoskeleton in adrenal glomerulosa cells: role in receptor-effector coupling. Endocrinology 1997; 138: 3299–3307.

  28. 28

    Zhu W, London NR, Gibson CC, Davis CT, Tong Z, Sorensen LK et al. Interleukin receptor activates a MYD88-ARNO-ARF6 cascade to disrupt vascular stability. Nature 2012; 492: 252–255.

  29. 29

    Khanna P, Yunkunis T, Muddana HS, Peng HH, August A, Dong C . p38 MAP kinase is necessary for melanoma-mediated regulation of VE-cadherin disassembly. Am J Physiol Cell Physiol 2010; 298: C1140–C1150.

  30. 30

    Buffoli B, Pechánová O, Kojsová S, Andriantsitohaina R, Giugno L, Bianchi R et al. Provinol prevents CsA-induced nephrotoxicity by reducing reactive oxygen species, iNOS, and NF-kB expression. J Histochem Cytochem 2005; 53: 1459–1468.

  31. 31

    Storogenko M, Pech-Amsellem MA, Kerdine S, Rousselet F, Pallardy M . Cyclosporin-A inhibits human endothelial cells proliferation through interleukin-6-dependent mechanisms. Life Sci 1997; 60: 1487–1496.

  32. 32

    Huang HC, Shi GY, Jiang SJ, Shi CS, Wu CM, Yang HY et al. Thrombomodulin-mediated cell adhesion: involvement of its lectin-like domain. J Biol Chem 2003; 278: 46750–46759.

Download references

Acknowledgements

This study was supported by grants from Uehara Memorial Foundation, SENSHIN Medical Research Foundation and KAKENHI (23591421 and 26461406).

Author contributions

TI contributed to the concept and design, interpreted and analyzed the data, and wrote the article; JY performed the experiments and wrote the article; CN performed the experiments; KU synthesized DHMEQ and critical revision; and AY provided important intellectual content and gave final approval.

Author information

Correspondence to T Ikezoe.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on Bone Marrow Transplantation website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ikezoe, T., Yang, J., Nishioka, C. et al. Thrombomodulin blocks calcineurin inhibitor-induced vascular permeability via inhibition of Src/VE-cadherin axis. Bone Marrow Transplant 52, 245–251 (2017) doi:10.1038/bmt.2016.241

Download citation

Further reading