Signals of the -1−MET Axis and Cues of Mechanical Force Exertion Converge to Elicit Inflammatory Activation in Coherent Endothelial Cells This information is current as of September 27, 2021. Maryam Rezaei, Ana C. Martins Cavaco, Jochen Seebach, Stephan Niland, Jana Zimmermann, Eva-Maria Hanschmann, Rupert Hallmann, Hermann Schillers and Johannes A. Eble

J Immunol published online 28 January 2019 Downloaded from http://www.jimmunol.org/content/early/2019/01/27/jimmun ol.1801346 http://www.jimmunol.org/ Supplementary http://www.jimmunol.org/content/suppl/2019/01/27/jimmunol.180134 Material 6.DCSupplemental

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2019 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published January 28, 2019, doi:10.4049/jimmunol.1801346 The Journal of Immunology

Signals of the Neuropilin-1–MET Axis and Cues of Mechanical Force Exertion Converge to Elicit Inflammatory Activation in Coherent Endothelial Cells

Maryam Rezaei,* Ana C. Martins Cavaco,* Jochen Seebach,† Stephan Niland,* Jana Zimmermann,* Eva-Maria Hanschmann,‡ Rupert Hallmann,* Hermann Schillers,x and Johannes A. Eble*

The neuropilin-1 (NRP1)-MET signaling axis regulates the motility of individual endothelial cells (ECs). It is unknown how this signaling pathway affects the endothelial barrier in coherent ECs forming a tight monolayer. We hypothesized that it is involved both in modulation of the endothelial barrier and in EC activation. To investigate the role of NRP1–MET signaling in inflammatory processes (e.g., systemic inflammatory response syndrome [SIRS] or snakebite-induced SIRS-like conditions), we employed the Downloaded from C-type lectin-related rhodocetin-ab (RCab) as a specific trigger of this signal axis in ECs in vitro. In coherent HUVECs, RCab reinforced the actin cytoskeleton and increased cell stiffness, thus favoring vascular endothelial cadherin–mediated trans- mission of intercellular forces. Increased cell stiffness was associated with enhanced activation of RhoA and nuclear translocation of NF-kB. Simultaneously, RCab-triggered signaling via the NRP1–MET axis increased EC monolayer permeability, induced transcription of proinflammatory such as ICAM-1 and, consequently, leukocyte tethering. The RCab-induced transcriptome differed from that induced by , although in both cases the same tyrosine kinase, MET, was involved. This http://www.jimmunol.org/ was due to RCab-mediated recruitment of the MET coreceptor NRP1 and additional Rho-mediated activation of the actomyosin system. RCab induced similar transcriptional and cellular changes if external shear forces were applied. These data highlight the modulatory role of NRP1 as MET coreceptor, and they explain how some snake venoms induce SIRS-like conditions. Addi- tionally, this study demonstrates that inflammatory activation of coherent ECs is triggered by converging signals that are induced by NRP1–MET signaling and influenced by intercellular forces. The Journal of Immunology, 2019, 202: 000–000.

he inner surface of blood vessels is lined by a monolayer The interaction of leukocytes with ECs during TEM is a mul-

of endothelial cells (ECs). Tightly connected to each tistep procedure, involving exposure of leukocyte-adhesive pro- by guest on September 27, 2021 T other, they form a barrier between the blood compartment teins, such as ICAM-1, on the luminal surface of ECs (5, 6). Being and surrounding vascular and stromal cells (1). Moreover, this EC a marker of this EC activation, ICAM-1 expression is regulated by monolayer controls the transport of biochemical compounds and various inflammatory cytokines (5, 7). Characteristically activated of cells between bloodstream and vessel-surrounding stroma. during inflammation (8), NF-kB is a transcription factor that Leukocytes cross the monolayer of tightly interconnected ECs regulates transcription of the ICAM-1 (5, 7). To fulfill their during immune surveillance and inflammatory defense (2). Dys- functions, ICAM-1 molecules, which are stored in intracellular function of this barrier increases vessel permeability and leakage vesicles, have to be exposed on the cell surface. Moreover, for and disturbs transendothelial migration (TEM) of leukocytes. This leukocyte attachment to ICAM-1–exposing ECs, the ICAM-1 becomes manifest in edema formation and malfunctioning of the molecules have to be arrayed in a certain pattern on the cell immune defense, such as in chronic inflammatory diseases. A surface, a process that depends on the stiffness of the underlying local inflammatory response may become life-threatening if it matrix (9). reaches out to the entire vasculature, resulting in systemic inflam- ECs need to withstand mechanical shear forces induced by the matory response symptoms/sepsis (3, 4). bloodstream. They are able to do this because of their actomyosin

*Institute of Physiological Chemistry and Pathobiochemistry, Westfalian Wilhelm Uni- The microarray data presented in this article have been submitted to the Gene Ex- versity Mu¨nster, 48149 Mu¨nster, Germany; †Institute of Anatomy and Vascular Biology, pression Omnibus database (http://www.ncbi.nlm.nih.gov/geo/query/) under acces- Westfalian Wilhelm University Mu¨nster, 48149 Mu¨nster, Germany; ‡Department of sion number GSE121297. Neurology, Medical Faculty, Heinrich-Heine University Du¨sseldorf, 40225 Du¨sseldorf, x Address correspondence and reprint requests to Dr. Johannes A. Eble, Institute of Germany; and Institute of Physiology II, Westfalian Wilhelm University Mu¨nster, 48149 Physiological Chemistry and Pathobiochemistry, University of Mu¨nster, Waldeyerstraße15, Mu¨nster, Germany 48149 Mu¨nster, Germany. E-mail address: [email protected] ORCIDs: 0000-0002-2055-8656 (S.N.); 0000-0003-3288-1706 (R.H.); 0000-0001- The online version of this article contains supplemental material. 6090-4771 (H.S.); 0000-0001-9156-2137 (J.A.E.). Abbreviations used in this article: AFM, atomic force microscopy; CBT, Cell Border Received for publication October 9, 2018. Accepted for publication December 27, Tracker; EC, endothelial cell; ECM, extracellular matrix; HGF, hepatocyte growth 2018. factor; NRP1, neuropilin-1; pfp, percentage of false predictions; PKB/Akt, protein This work was supported by Deutsche Forschungsgemeinschaft (DFG) Grant kinase B; RCab, rhodocetin-ab; ROCK, Rho-associated kinase; siRNA, small inter- SFB1009 A09 (to J.A.E.) and Wilhelm Sander-Stiftung Grant 2016.113.1 (to J.A.E.). fering RNA; SIRS, systemic inflammatory response syndrome; TEM, transendothe- As a member of the DFG-funded Cells-in-Motion Cluster of Excellence, J.A.E. also lial migration; TER, transendothelial electrical resistance; VE-cadherin, vascular received financial support from there (EXC1003-CiM). J.S. received funding from endothelial cadherin; VEGF, vascular endothelial growth factor. the Cells-in-Motion Cluster of Excellence (FF-2016-15). Copyright Ó 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1801346 2 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE system and their anchoring , which connect ECs to the thereby promoting (29). In addition, in an in vivo underlying extracellular matrix (ECM) and neighboring cells tumor model, RCab selectively disrupted blood vessels of the (10, 11). The actomyosin machinery is a complex network of tumor but not of healthy tissues (31). F-actin bundles, actin-organizing proteins, such as members of the To better understand these surprising findings, the effect of Rho family, and motor proteins (12, 13). Mechanical forces be- RCab on monolayers of confluent ECs was investigated in an tween ECs and their surrounding are transmitted via transmem- in vitro system at transcriptional, protein-chemical, and cellular brane proteins such as and cadherins. These proteins level. anchor an EC to the ECM and neighboring cells, respectively. Integrins are heterodimeric proteins that bind to ECM proteins and are connected to the actin cytoskeleton via adapter proteins (14). Materials and Methods They are part of adhesomes, supramolecular complexes in which Cell culture and silencing of NRP1 and MET several adapter and signaling proteins help integrins to transmit HUVECs (Promocell, Heidelberg, Germany) were cultivated, unless mechanical forces between a cell and its ECM (15). Two main otherwise stated, up to passage 5 on rat tail tendon collagen-I (Corning, types of adhesomes are distinguished: focal complexes and focal Tewksbury, MA) in complete EC growth medium (Promocell) supple- mented with 2% heat-inactivated FBS and 1% penicillin/streptomycin adhesions. The latter can bear higher mechanical forces and (Invitrogen Life Technologies). contain force-sensing vinculin (16). Intercellular contacts between U937 cells were maintained in RPMI 1640 containing 10% FBS. For neighboring ECs are mainly mediated by vascular endothelial rollingassays,U937cellswereresuspendedinbEndmedium(DMEM, cadherin (VE-cadherin) (17). Not only intercellular force trans- 4.5 g/ml glucose, stabilized glutamine; PAN-Biotech, Aidenbach, Germany) supplemented with 10% FBS (Biochrom, Berlin, Germany), 0.5 mM pyruvate Downloaded from mission and endothelial integrity but also endothelial permeability (PAN-Biotech), 5 ml of nonessential amino acids (1003 mix; PAN-Biotech), depend significantly on VE-cadherin. Although VE-cadherins in- and 5 mM 2-ME (Life Technologies/Thermo Fisher Scientific, Schwerte, teract in a homophilic and Ca2+-dependent manner and thus connect Germany). neighboring cells, their cytoplasmic domain is linked via a-and RCab was isolated from Calloselasma rhodostoma crude venom as b-catenin, as well as vinculin, to the actin cytoskeleton (12), thus described previously (32). Absence of endotoxins was ensured using a HEK-Blue LPS Detection Kit 2 (InvivoGen, San Diego, CA). RCab allowing perception and transmission of intercellular forces (18). contained 0.02 endotoxin units per mg protein (0.002–0.004 ng/ml), which

In analogy to vinculin-containing focal adhesions, the force- was unaffected by 100 mg/ml endotoxin-binding antibiotic (polymyxin b). http://www.jimmunol.org/ transmitting array of vinculin-recruiting VE-cadherin–catenin MET inhibitor SU11274 was from Selleckchem/BIOZOL (Eching, 5 complexes are also referred to as focal adherens junctions (18). Germany). To silence NRP1 and MET, 1–2 3 10 HUVECsin2mlof medium were transfected (Magnet-Assisted Transfection; Promokine, Hei- Their interrupted pattern, orthogonally oriented to the cell mem- delberg, Germany) with small interfering RNAs (siRNAs) and matched brane, depends on intercellular forces and is a dynamic structure controls (SMARTpools; Dharmacon, Lafayette, CO) according to the man- over time (19). ufacturer’s protocol. Cytoskeletal contractility is further modulated by various in- Transendothelial electrical resistance flammatory mediators, such as TNF-a, thrombin, and vascular endothelial growth factor (VEGF), and depends on activation of The transendothelial electrical resistance (TER) of interconnected ECs RhoA (20, 21). Various actin-binding proteins and their regulatory forming a tight monolayer on cross-linked collagen was monitored by by guest on September 27, 2021 impedance spectroscopy (MOS-Technologies, Telgte, Germany) for enzymes are targets of Rho GTPases (20, 22). Moreover, RhoA 30 min at 37˚C (33). Subsequently, 400 nM RCab was added, and activates NF-kB signaling (23), which is crucial in innate and the time course of the impedance change was recorded. adaptive immunity (24) and, furthermore, contributes to leukocyte extravasation and (10, 18). NF-kB is also a RhoA- Collagen gel contraction dependent mechanosensor that is activated by increased mechan- HUVECs (3 3 105) were embedded in 0.6 ml of collagen-I gels (1 mg/ml ical tension within the monolayer and then translocated into the rat tail tendon collagen-I in EC basal medium [PAA Laboratories, nucleus (25). Pasching, Austria] supplemented with 10% FBS) containing 400 nM RCab, 10 ng/ml TNF-a, or PBS as control. After 30 min at 21˚C, gels Cytoskeletal contractility stiffens ECs, which may result in were detached from the well walls with a scalpel, and 0.6 ml EC basal chronic hypertension (26). It also induces EC activation and medium containing 10% FBS and 1% penicillin/streptomycin was added. promotes inflammatory processes, such as tethering and ex- The time course of gel contraction was recorded photographically. travasation of leukocytes, increased vessel permeability, and an Atomic force microscopy enhanced tendency for thrombosis (7). In diverse vasculo- pathies, such as atherosclerosis, acute coronary syndrome, and Atomic force microscopy (AFM) was performed in HEPES-Ringer buffer at systemic inflammatory response syndrome (SIRS), prolonged 37˚C in PeakForce QNM mode (34) using a BioScope Catalyst (Bruker Nano Surfaces, Santa Barbara, CA) and a PeakForce QNM–Live Cell EC activation causes endothelial inflammation and dysfunction probe (tip radius 65 nm; Bruker AFM Probes, Camarillo, CA). The can- (3, 7, 20, 27). Similar to SIRS, snake venoms often increase tilever’s spring constant (0.0611 N/m) was determined interferometrically leukocyte extravasation and endothelial permeability systemically (OFV-551; Polytec, Waldbronn, Germany). Deflection sensitivity was and not just locally at the site of the snakebite (3, 4). However, the adjusted as published (35). PeakForce QNM data sets, obtained at 100 3 100 mm with 512 3 512 pixels and 0.2-Hz scan frequency, 0.5-kHz underlying mechanisms are still largely unknown (28). PeakForce oscillation frequency, an amplitude of 300 nm, and a PeakForce In a previous work, we identified rhodocetin-ab (RCab) as the setpoint of 750 pN, were analyzed using Nanoscope Analysis (v1.60; Bruker first nonenzymatic component of a snake venom that recognizes Nano Surfaces). neuropilin-1 (NRP1) on ECs (29). NRP1 is a transmembrane Force mapping was performed in HEPES-Ringer buffer at 37˚C using a expressed inter alia in ECs, , and immune JPK NanoWizard 3 AFM (JPK Instruments, Berlin, Germany) in closed- loop mode with 2-mm ramp size, 500-pN maximum force, and 1-mm/s tip cells (30). There, NRP1 functions as coreceptor of VEGF re- velocity. Each force map contained 16 3 16 force-distance cycles over an ceptor family members and of the hepatocyte growth factor area of 100 3 100 mm. Measurements were performed using a colloidal (HGF) MET. RCab binds to NRP1 and induces the probe cantilever with a 5-mm polystyrene bead (Novascan, Boone, IA). formation of a ternary complex with MET (29). In vitro, this The cantilever’s spring constant (0.0491 N/m) was determined inter- ferometrically and the deflection sensitivity adjusted accordingly (35). leads, in ECs without adjacent cells, to a reorganization of Force-indentation curves were analyzed with Protein Unfolding and the actin cytoskeleton and to restructuring of the cell matrix– Nano-Indentation Analysis Software (http://punias.free.fr/) using the anchoring complexes from focal adhesions into focal contacts, linearized Hertz model (36). The Journal of Immunology 3

Pull-down of active RhoA additionally compare the genes whose transcription is affected by HGF or RCab, HUVECs were stimulated for 8 h under static conditions with RhoA activation was analyzed using a RhoA pull-down activation assay 200 nM RCab or 200 ng/ml HGF and subsequently analyzed for gene according to the manufacturer’s instructions (Cytoskeleton, Denver, expression (MFT-Services). Per array, 100 ng of total RNA was ampli- 3 6 CO). Thereto, 5 10 HUVECs were cultivated for 24 h on 10-cm fied and labeled with the Affymetrix 39 IVT Express Kit, according to ab dishes and then treated with 400 nM RC or PBS. Activated RhoA was the manufacturer’s instructions. Samples were hybridized, stained, and pulled down from cell lysates with rhotekin domain-functionalized scanned with an Affymetrix U219 Perfect Match array beads. The bead eluate was analyzed by 12% SDS-PAGE, followed strip kit. Differential expression was analyzed with the microarray by immunoblotting (ECL Western Blotting Substrate [Pierce/Thermo analysis software Mayday. The data of transcriptome analysis were de- Fisher Scientific]; ImageQuant LAS 4000 [GE Healthcare, Freiburg, posited at the Omnibus database under accession Germany]). The Rho–Rho-associated kinase (ROCK)–actomyosin path- number GSE121297 (accessible at: https://www.ncbi.nlm.nih.gov/geo/ m way was blocked in HUVECs at the level of Rho or ROCK with 1 g/ml query/). A total of 23,520 genes were examined for differential expres- C3 transferase (Cytoskeleton) for 4 h and 10 mM Y-27632 for 10 min, sion. After log2-transformation, values were normalized to the averaged respectively. cumulative intensity value and filtered with a threshold of an absolute log2-transformed fold change of 1 (fold expression change .2or,0.5). Transcriptome analysis of RCab-treated HUVECs Additionally, a product test was carried out. To correct for effects of p For comparative transcriptome analysis, HUVECs, which had been freshly multiple testing, values were controlled using the false discovery rate isolated according to (13), were used in accordance to principles outlined or the percentage of false predictions (pfp). The significance level for the p # in the Declaration of Helsinki and approved by the ethics board of the pfp value was set to 0.05. Both criteria, logFC and pfp, were si- multaneously applied to filter differentially expressed genes. Institute of Anatomy and Vascular Biology, University of Mu¨nster (2009- 537-f-S). Only cells of passage 1 were used for these experiments. Fluid RNA isolation and RT-PCR shear stress experiments were performed in a cone plate-based system (BioTechFlow; MOS-Technologies) using collagen-I–coated plates as de- Total RNA from 2 3 105 cells was isolated using the RNeasy Mini Kit Downloaded from scribed (37, 38). After treatment with or without 200 nM RCab, statically (QIAGEN, Hilden, Germany) according to the manufacturer’s protocol. incubated and shear-subjected cells were harvested, and their transcriptomes Total RNA was reverse-transcribed (Reverse Transcriptase Core Kit; were analyzed (MFT-Services, University of Tu¨bingen, Germany). To Eurogentec, Seraing, Belgium) and quantified in duplicate by real-time http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 1. RCab alters the morphology of confluent ECs, concomitant with a redistribution of VE-cadherin into focal adherens junctions and re- cruitment of vinculin. (A) Immunofluorescent staining of VE-cadherin (green) revealed the cell borders of coherent HUVECs that had been treated with RCab (400 nM) for 6 h. Nuclei (blue) were stained with DAPI. Scale bars, 20 mm. (B) Cell roundness and circularity were quantified from images such as those shown in (A). For the evaluation of cell roundness, the numbers of cells analyzed were as follows: PBS, n = 285; RCab, n = 201; VEGF-A165 (VEGF-a), n = 194; and TNF-a, n = 58. For cell circularity, numbers of cells analyzed were as follows: PBS, n = 285; RCab, n = 189; VEGF-a, n = 206; and TNF-a, n =71.Means6 SD are shown. **p # 0.01, ****p # 0.0001. (C) Double immunofluorescent staining for VE-cadherin (green) and vinculin (red) of coherent HUVECs after 6 h treatment with RCab (400 nM) or PBS as control (Ctrl) revealed an overall reduction of VE-cadherin and its concentration into punctual intercellular contacts, which were positive for vinculin. Scale bars, 20 mm. (D) Intensities of VE-cadherin and vinculin at cell junctions were quantified with ImageJ and CBT in three independent experiments, shown as means 6 SD. For the calculation of VE-cadherin junctional intensity, we evaluated n = 64 Ctrl cells and n =42RCab cells at 3 h and n = 58 Ctrl cells and n =77RCab cells at 6 h. Evaluation of vinculin junctional intensity has been done for n = 195 Ctrl cells and n = 185 RCab cells at 3 h and n = 148 Ctrl cells and n =191RCab cells at 6 h. *p # 0.05, **p # 0.01. (E) Fluorescence line scan from images of (C) quantifying the presence of VE-cadherin and vinculin along a cell border [indicated by white lines in (C)] without and with RCab.(F)RCab increased permeability of a HUVEC monolayer and decreased TER. TER was measured by impedance spectroscopy with HUVECs from different donors treated with PBS as control or RCab (400 nM). TER values normalized to the initial ones were plotted against time. Means 6 SD are shown. *p # 0.05, **p # 0.01. 4 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE Downloaded from http://www.jimmunol.org/

FIGURE 2. RCab increases stress fiber formation and rigidity of ECs, stimulates EC-mediated gel contraction, and enhances Rho-kinase activity. (A) Immunofluorescent staining for VE-cadherin (green) and phalloidin (red) of confluent HUVECs treated with PBS (control [Ctrl]) or RCab (400 nM). Scale by guest on September 27, 2021 bars, 20 mm. Phalloidin intensity, which correlates with the number of stress fibers, was quantified after 3 and 6 h with ImageJ (right). (B and C) AFM deflection images and time-lapse elastic modulus maps of live ECs before (Ctrl) and after adding 400 nM RCab. Scale bars, 20 mm. The change in elastic modulus over time is shown in (C). (D and E) Contraction of a type I collagen gel by embedded HUVECs in the presence of PBS (Ctrl), RCab, or TNF-a and its quantification after 24 h as area of contracted gel normalized to the original area. Means 6 SD are shown (n = 3). *p # 0.05, **p # 0.01. (F) Rho activity was determined by pull-down assay. The GTP-bound active form (top) and the total amount of RhoA (bottom) were detected in immunoblots of electrophoretically separated lysates of HUVECs treated with PBS (Ctrl) or RCab.

PCR (Rotor-Gene Q and QuantiFast SYBR Green PCR Kit; QIAGEN) with 400 nM RCab, 200 ng/ml HGF, 20 ng/ml VEGF-A165, or PBS for using the following primer pairs: ICAM-1, 59-AGGGTAAGGTTCTT- 6 h. Then cells were fixed (2% paraformaldehyde/PBS, pH 7.5), per- GCCCAC-39,59-TGATGGGCAGTCAACAGCTA-39; E-, 59-GA- meabilized (0.1% Triton X-100 [Thermo Fisher Scientific], 10 min at 4˚C), GTGCACATCTCAGGGACA-39,59-ACTGCCAGGCTTGAACATTT-39; and stained overnight at 4˚C with primary Abs against VE-cadherin CCL2, 59-AGGTGACTGGGGCATTGAT-39,59-GCCTCCAGCATGAA- (1:100; Santa Cruz Biotechnology), ICAM-1, selectin/CD62E (1:50; AGTCTC-39; CXCL5, 59-AAACTTTTCCATGCGTGCTC-39,59-TTGT- R&D Systems), NF-kB p65(F-6) (1:50; Santa Cruz Biotechnology), and CTTGATCCAGAAGCCC-39;MET,59-GCTGCAAAGCTGTGGTAAACT- vinculin (1:100; Sigma-Aldrich, Deisenhofen, Germany). Subsequently, 39,59-CTCCAGCATTTTTACGGACC-39; NRP1, 59-TTGCAGTCTCTGT- cells were stained with Alexa Fluor 488– or Alexa Fluor 568–labeled Abs. CCTCCAA-39,59-GAAAAATGCGAATGGCTGAT-39,andTOP1,59-CC- Actin was labeled with phalloidin–Alexa Fluor 568 (Invitrogen, Karlsruhe, AGACGGAAGCTCGGAAAC-39,59-GTCCAGGAGGCTCTATCTTGAA- Germany). Afterward, specimens were counterstained with DAPI, 39. Cycle threshold (Ct) values were normalized by the DDCt method (39) mounted (Dako fluorescent mounting medium; Agilent, Du¨sseldorf, and TOP1 as a housekeeping gene. Germany), and confocal images were acquired (LSM 700; Zeiss). Morphology and junctional protein localization in VE-cadherin–stained cells Immunoblotting were quantified with Cell Border Tracker (CBT) (41). Cell roundness and circularity were determined with ImageJ. Both of these values Cell lysates were separated on 10% SDS gels and transferred by wet blotting compare the shape of the cells with a circle. They are based on different onto nitrocellulose (Whatman, Dassel, Germany). Membranes were incu- mathematical expressions. Cell roundness is the ratio of actual cell area bated with anti-human VE-cadherin (1:500; Santa Cruz Biotechnology, to a theoretical circular area calculated from a diameter corresponding Dallas, TX), anti-RhoA (1:500; Cytoskeleton), and anti-human ICAM-1/ to the long axis of the cell, whereas cell circularity is the ratio of actual CD54 (1:250; R&D Systems, Minneapolis, MN) Abs and afterward with cell area to a theoretical circular area with the same perimeter as HRP-conjugated secondary IgG (1:1000; Dako, Hamburg, Germany). the cell. Bands were detected with ECL Western Blotting Substrate (Pierce/Thermo To quantify the amount of junctional proteins, CBT-generated outlines Fisher Scientific) and ImageQuant LAS 4000 (GE Healthcare). Relative were broadened to 1 mm, and the fluorescence intensity of the respective band intensities were determined using ImageJ (40). protein in such a junctional region of interest was quantified. The relative Immunofluorescence staining and morphological analysis VE-cadherin concentration is defined as the ratio of VE-cadherin intensity along the cell border (determined by CBT) and the cell border length. The Cells (2 3 105) were seeded on rat tail tendon collagen-I–coated intensity of NF-kB in DAPI-stained nuclear regions of interest was mea- (100 mg/ml; Life Technologies) permanox chamber slides (Nunc/Thermo sured with Fiji and normalized to the total image intensity. At least three Fisher Scientific, Braunschweig, Germany). After 24 h, cells were treated independent experiments were quantified. The Journal of Immunology 5 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 3. RCab induces the translocation of NF-kB into the nucleus, thereby activating NF-kB–regulated genes. (A–C) Time course of nuclear translocation of NF-kB in confluent HUVECs after addition of RCab. After treatment for 3 h (PBS: n = 132; RCab: n = 282; TNF-a: n = 134 cells) (A) and for 6 h (PBS: n = 99; RCab: n = 222; TNF-a: n = 211 cells) (B), images from three independent experiments were analyzed for NF-kB–p65/relA+ nuclei. Scale bars, 20 mm. (C) Biometric evaluation. Means 6 SD are shown. *p # 0.05, **p # 0.01, ****p # 0.0001. (D) Expression of E-selectin (SELE) and ICAM-1 in coherent HUVECs after treatment for 6 h with different concentrations of RCab. The real-time PCR data were normalized to the level of TOP1 mRNA as an internal control. Results from three independent experiments are given as means 6 SD. *p # 0.05. (E) Flow cytometric analysis showed that RCab increased surface presentation of ICAM-1 but only slightly that of E-selectin, whereas TNF-a–treated cells exposed high amounts of both ICAM-1 and E-selectin. (F) Immunofluorescent staining of phalloidin (red), NF-kB–p65 (green), and VE-cadherin (blue) in HUVECs treated for 4 h with membrane- permeable C3 transferase to inhibit Rho and for 10 min with Y-27632 to inhibit ROCK. (G) Quantification of nuclear translocation of NF-kB. RCab: n =112 cells; ROCK inhibitor: n = 71 cells; and C3 transferase: n = 158 cells. Results from three independent experiments are given as means 6 SD. *p # 0.05.

Flow cytometry Pineda, Berlin, Germany) and anti-rat Alexa Fluor 568 Ab (1:200; BD Transduction Laboratories, Heidelberg, Germany). After 24 h of cultivation on collagen-I–coated (100 mg/ml) plates, 2 3 105 HUVECs were treated for 6 h with 400 nM RCab, 200 ng/ml HGF, Leukocyte adhesion and rolling 20 ng/ml VEGF-A165,or10ng/mlTNF-a. Subsequently, they were harvested with accutase (Chemicon/Millipore, Darmstadt, Germany) according to U937 cells were labeled for 5 min at 37˚C with 10 mg/ml TAMRA the manufacturer’s instructions, centrifuged at 250 3 g for 4 min, (succinimidyl-ester of TAMRA, mixed isomers; Invitrogen). HUVECs resuspendedin0.1%BSA/PBS,andstainedwithFITC-labeledCD54Ab were cultivated for 24 h to form confluent monolayers in ibidi mVI0.4 (1:20; Immunotools, Friesoythe, Germany), FITC-labeled monoclonal slides. They were treated with RCab (200 or 400 nM), HGF (200 ng/ml), MET Ab (1:20, eBioclone 7; Thermo Fisher Scientific), and IgG2a-FITC VEGF-A165 (20 ng/ml), or TNF-a (10 ng/ml) for 6 h before tethering of control Ab. NRP1 was detected with rat polyclonal NRP1 Ab (1:20; leukocytes was measured. In the incubation chamber of an AxioObserver. 6 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE Downloaded from

FIGURE 4. RCab treatment of coherent ECs promotes attachment and rolling of monocytes. (A) A suspension of fluorescently labeled U937 monocytic cells (red) was pumped at a defined flow rate over a confluent monolayer of HUVECs pretreated for 6 h with different concentrations of RCab. Phase contrast images (top) and fluorescence images (middle and bottom) at the start of experiment (top and middle row) and after 30 s at a shear rate of 2 dyn/cm2 http://www.jimmunol.org/ (bottom row). (B and C) Rolling and adhesion of U937 cells on HUVECs were analyzed using a self-written ImageJ macro. The numbers of rolling and adhering cells, respectively, were normalized to the initial cell count at the beginning of a video (at 0 dyn/cm2, time 0). Each data point represents one of three independent experiments. Data are means 6 SD. *p # 0.05, **p # 0.01.

Z1 (Zeiss) inverted live-cell imaging microscope, which was thermostated Fig. 1C, 6 h). Although the total amount of VE-cadherin decreased 6 to 37˚C, the slide was connected to a pump. After 5 min, 1 3 10 immune along the cell–cell borders (Fig. 1C, 1D), colocalization of vinculin with cells per ml (in bEnd medium/10 mM HEPES, pH 7.4) were perfused at 2 dyn/cm2 over the monolayer for 30 s and recorded at five frames per s VE-cadherin appeared to increase in such junctional adhesions (Fig. 1E, (Plan-Neofluar 103/0.30 Ph103 objective and Hamamatsu Orca-ER fluorescence intensity scans along the white lines marked in Fig. 1C). camera). At least three experiments were performed, each with more This suggested RCab-induced relocalization of VE-cadherin from an by guest on September 27, 2021 than 1000 U937 cells per data set. The images were processed with Fiji initially homogeneous distribution to concentrated vinculin-containing (42) using a self-written macro. Briefly, the initial leukocyte number was spots along the cell–cell borders. This vinculin recruitment suggested an taken from the first frame of the video. After 15 s, stable flow conditions had been reached, and then cells were tracked with the plugin MTrack2. intercellular point-to-point force transmission that was accompanied The acquired data were evaluated with Microsoft Excel. Cells slower with ECs elongation (Fig. 1A, 1B). Hence, we referred to these force- than 0.5 mm/s and visible for more than 14 s were considered adherent, and transmitting intercellular junctions as focal adherens junctions according their number was normalized to the initial cell number. Cells moving in the to a previous work (18). Moreover, we observed a sustained perme- direction of flow with 0.5–20 mm/s and trackable for at least 5 s (25 frames) were considered rolling and normalized to the initial cell number. ability increase of the EC cell monolayer, despite a temporary increase in TER within the first 150 min of RCab treatment(Fig.1F).Thisalso Statistics indicated that RCab weakened the EC barrier function functionally, For all experiments, data were compared with unpaired t tests by GraphPad along with the redistribution of the VE-cadherin at the cell–cell borders. software (GraphPad Prism 7.0). Data were considered statistically signif- icant if p values were #0.05. Error bars denote SD. In coherent ECs, RCab treatment enhanced stress fibers and increased cell stiffness Results As RCab caused a redistribution of the force-transmitting ab In EC monolayer, RC induced cell elongation and caused cytoskeletal adapter vinculin into punctiform focal adherens density reduction and redistribution of vinculin at intercellular junctions, we analyzed its effect on the actin cytoskeleton. cell borders Time-dependently, RCab markedly induced the formation of Cohesion of ECs within an EC layer is mediated by intercellular phalloidin-binding F-actin stress fibers (Fig. 2A, red signal), contacts that significantly determine vascular function. Coherent which are able to bear tensile forces. In fact, after RCab treat- ECs were treated under static conditions with RCab, VEGF-A165, ment, these stress fibers were under higher tension, resulting in TNF-a, or PBS as control for 6 h. Thereafter, the morphological increased cell stiffness, which was detectable by increased AFM- parameters cell roundness and circularity of VE-cadherin–stained determined cell elastic modulus (Fig. 2B, 2C). If embedded in cells were quantified with CBT (41) from confocal images collagen-I gels, RCab-treated ECs contracted the gel even more a (Fig. 1A). RCab, like VEGF-A165 and TNF-a, significantly re- strongly than TNF- –treated ECs through the force-exerting ac- duced both roundness and circularity (Fig. 1B). At the same time, tomyosin system (Fig. 2D, 2E). As RhoA controls the force ex- RCab decreased the amount of VE-cadherin along the cell–cell ertion of the actomyosin system, the effect of RCab on RhoA borders within 3 h (Supplemental Fig. 1) and more prominently activation was investigated by immunoblotting. Consistent with after 6 h (Fig. 1C). During the same period, vinculin, a marker of the increase in stress fibers and contractility, 30 min after addition force-transmitting cell–matrix and cell–cell contacts, shifted from fo- of RCab, the amount of GTP-bound RhoA was significantly in- cal adhesions to intercellular contact sites (Supplemental Fig. 1, 3 h; creased despite an unchanged total amount of RhoA (Fig. 2F). The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 5. EC activation occurs by signaling via the NRP1–MET axis. (A–C) The reduction of surface expression of MET (A) and of NRP1 (B) after transfection with the corresponding siRNAs was verified by flow cytometry. Quantification of the flow cytometric data is illustrated in (C). Means 6 SD are shown. *p # 0.05, ***p # 0.001. (D–F) The RCab-induced increase of ICAM-1 expression in HUVECs was distinctly reduced by treatment with NRP1 siRNA and, less effectively, with MET siRNA. NRP1 siRNA–reduced ICAM-1 expression was measured at the mRNA level by quantitative PCR (D), at the protein level by immunoblot of cell lysates (E), and at the level of cell surface exposure by flow cytometry (F). (G) MET inhibitor SU11274 (inh) reduced the RCab- or TNF-a–stimulated increase in ICAM-1 mRNA as detected by quantitative PCR. (H) Immunofluorescent staining of VE-cadherin (green) in HUVECs was used to quantify their morphological changes after treatment with RCab or TNF-a in the presence or absence of inh. (I) Cell roundness was determined using CBT on representative images (H) from three independent experiments. Ctrl: n = 57 cells; Ctrl + inhibitor: n = 62 cells; RCab: n =65 cells; RCab + inhibitor: n = 74 cells; TNF-a: n = 54 cells; and TNF-a + inhibitor: n = 82 cells. Means 6 SD show that the MET inhibitor significantly affected shape changes of HUVECs treated with RCab and TNF-a, respectively. ****p # 0.0001.

RCab-induced force exertion between interconnected ECs TNF-a. Dose-dependently, RCab-induced nuclear translocation caused translocation of NF-kB into the nucleus of the transcription factor NF-kB enhanced transcription of the Concomitant with RhoA activation and increased cell stiffness, the proinflammatory marker genes ICAM-1 and E-selectin (Fig. 3D). ab transcription factor NF-kB translocated into the nuclei of RCab- By flow cytometry of ECs derived from RC -treated mono- treated coherent ECs, similarly to treatment with TNF-a (Fig. 3A, layers, a significant increase in ICAM-1, but only a slight increase 3B, green signal). Quantification of NF-kB subunit RelA in the of E-selectin, was measured on the cell surface at the protein level nuclei of RCab-treated ECs showed a significant increase over (Fig. 3E). Exposure of ICAM-1 on the EC surface is a sign of EC PBS-treated cells after 3 and 6 h (Fig. 3C), a reaction similar to the activation. The RCab-induced translocation of NF-kB to the nu- positive control, TNF-a. However, ECs responded to RCab with a cleus required the Rho–ROCK–actomyosin pathway, as nuclear greater variance. Some single cells showed a higher nuclear translocation of RelA was diminished by inhibition of either Rho intensity of RelA in response to RCab compared with TNF-a, with C3 transferase or ROCK with Y-27632 when applied prior to whereas other cells responded much more weakly to RCab than to RCab treatment. Both inhibitors reduced the number of stress 8 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 6. RCab, unlike HGF, activates NF-kB signaling. (A and B) VE-cadherin in coherent ECs was immunohistochemically labeled (green) to observe the influence of HGF and RCab, respectively, on cell morphology, which was biometrically quantified by cell roundness. Ctrl: n = 285; HGF: n = 267; and RCab: n = 201 cells. Means 6 SD. ****p # 0.0001. (C) Nuclear translocation of NF-kB–p65/relA was induced within 6 h by RCab (400 nM) and TNF-a (10 ng/ml), respectively, but not by HGF (200 ng/ml). Cell borders were marked by VE-cadherin (green). (D) Quantification of NF-kB–p65/ relA+ nuclei. Ctrl: n = 73 cells; RCab: n = 125 cells; HGF: n = 144 cells; and TNF-a: n = 115 cells. Means 6 SD. (Figure legend continues) The Journal of Immunology 9

fibers and attenuated translocation of NF-kB into the nucleus with respect to the phosphorylation of serine residue S473 of (Fig. 3F, 3G). protein kinase B (PKB/Akt), a downstream target of MET (Fig. 6H). The observation that only RCab treatment was able ab RC -treated ECs promote leukocyte adhesion and rolling to significantly activate PKB/Akt by S473 phosphorylation Because of increased ICAM-1 expression on RCab-treated ECs, showed that the signaling pathways of RCab and HGF are tetheringofleukocytestosuchanECmonolayershouldbe different, irrespective of their common starting point, the strengthened. In fact, in an in vitro cell interaction test, tethering of tyrosine kinase MET. fluorescently labeled U973 monocytes to ECs was enhanced after the RCab and HGF induce different transcriptional profiles of EC monolayer had been treated with RCab (Fig. 4A). U973 cells confluent ECs rolled in a dose-dependent manner and adhered to RCab-treated EC monolayers (Fig. 4B, 4C). Similarly, TNF-a, as a positive control, To identify additional molecules that are regulated by RCab- and stimulated rolling and adhesion of U973 cells (Supplemental Fig. 2). HGF-triggered signal transduction, coherent ECs treated with RCab, HGF, or PBS as control were subjected to transcriptional ab In interconnected ECs, RC also triggers signaling along the profiling (Fig. 7). This also provided information on the different NRP1–MET axis signaling pathways of the NRP1–MET complex versus MET alone. To find out if RCab in interconnected ECs elicits the same sig- Transcriptome analysis showed that RCab and HGF induced 234 naling as previously disclosed for scattered ECs, NRP1 and MET and 293 genes, respectively (Fig. 7A). Both MET ligands, HGF and were individually knocked down in ECs (Fig. 5A–C). Subse- RCab, induced 107 genes (25.5%) similarly (Fig. 7A). In contrast, quently, these cells, after having reached confluency, were treated transcription of 186 genes (44.3%) was upregulated specif- Downloaded from with RCab and then analyzed for ICAM-1 at the transcriptional ically by HGF, whereas transcription of 127 genes (30.2%) was and translational levels. ICAM-1 mRNA synthesis was reduced to upregulated exclusively by RCab (Fig. 7A). Remarkably, un- background levels in NRP1-deficient ECs, whereas it was just like HGF, RCab stimulated the transcription of cytokines and slightly reduced in MET-deficient ECs (Fig. 5D), although the chemokines relevant to innate immunity, such as CCL20, siRNAs successfully reduced surface expression of the corre- CXCL6, CXCL1, CXCL2, and CCL2 (Supplemental Fig. 3A, ab sponding target proteins, NRP1 and MET, on the ECs (Fig. 5A–C). Table I). In addition, RC ,butnotHGF,alsopromotedthe http://www.jimmunol.org/ Also, at the protein level, ICAM-1 expression was only abrogated by transcription of adhesion molecules such as ICAM-1 and knockdown of NRP1 but not of MET (Fig. 5E). Nevertheless, RCab- VCAM-1 (Supplemental Fig. 3A). Thus, RCab mainly pro- induced surface expression of ICAM-1 was reduced in coherent moted proinflammatory immune and defense responses, as ECs after individual knockdowns of either NRP1 or MET well as responses to injuries (Fig. 7B). In contrast, HGF- (Fig.5F).Asanalternativeto siRNA, we employed the MET regulated genes are involved in the negative regulation of inhibitor SU11274 to characterize RCab-induced signaling in various processes and also in particular in the response to me- coherent ECs. This MET inhibitor effectively reduced tran- chanical stimuli (Fig. 7C). scription of ICAM-1, as well as in control experiments with Under shear forces, RCab also increases elongation of TNF-a (Fig. 5G). To investigate possible morphological differ- by guest on September 27, 2021 coherent ECs, causes redistribution of VE-cadherin, ences between RCab-andTNF-a–treated ECs in the presence or and promotes stress fiber formation absence of this MET inhibitor, VE-cadherin at cell borders was immunofluorescently labeled, and cell roundness was deter- On the one hand, RCab in interconnected ECs stimulated the mined as biometric parameter. Both RCab and TNF-a caused force-generating actomyosin system and increased cell rigidity. cell elongation, which was counteracted by the MET inhibitor On the other hand, ECs are constantly exposed to shear forces (Fig. 5H, 5I). These experiments show that RCab also signals of the bloodstream. Therefore, the influence of RCab on shear along the NRP1–MET axis in coherent ECs. forces–exposed ECs in a monolayer were investigated in an in vitro flow system. Under shear stress, RCab affected cell ab RC caused EC reactions other than the canonical MET morphology (Fig. 8A), similarly to static conditions (Fig. 1A). ligand, HGF Also, under shear forces, RCab induced ECs to assume an Although binding to the same , MET, elongated shape as quantified by decreased cell roundness and RCab and HGF appeared to signal differently and elicited circularity (Fig. 8A). Likely because of cell elongation, RCab different reactions. In contrast to RCab,HGFfailedtocause decreased the relative density of VE-cadherin along the cell morphological changes of coherent ECs (Fig. 6A, 6B); neither perimeter (Fig. 8B) and more frequently appeared in a spotlike, did HGF recruit activated NF-kB into the nuclei of confluent noncontiguous manner (cropped region in Fig. 8B). Also, the HUVECs, in contrast to RCab and TNF-a (Fig. 6C, 6D). In enhanced formation of actin stress fiber (Fig. 8C) after RCab addition, HGF treatment did not increase the surface ex- treatment in shear force–exposed ECs is similar to the reaction pression of the inflammation marker ICAM-1, unlike RCab of ECs in a static condition. (Fig. 6E, 6F). Therefore, U937 cells neither rolled on nor RCab-induced transcription profiles of shear-exposed ECs adhered to an HGF-treated EC monolayer, as opposed to andofECsunderstaticconditions, both within a coherent RCab-treated ECs (Fig. 6G). At the molecular level, ECs dif- monolayer, were compared (Fig. 9). Shear forces significantly ferently responded to treatment with HGF, RCab,orTNF-a changed the expression of 135 RCab-inducible genes (Fig. 9A).

**p # 0.01, ****p # 0.0001. (E)RCab, but not HGF, induced ICAM-1 on the surface of confluent ECs, as determined by flow cytometry. (F) ICAM-1+ cells were quantified by flow cytometry in three independent experiments after treatment with RCab, HGF, and TNF-a, respectively. Means 6 SD are shown. ****p # 0.0001. (G) For comparison, HUVECs were pretreated with 400 nM RCab and 200 ng/ml HGF before tethering and rolling of U937 cells was measured as shown in Fig. 4. Means 6 SD are shown. *p # 0.05, **p # 0.01. (H)RCab, but not HGF, activated PKB/Akt by phosphorylation at S473. HUVECs were stimulated for 6 h with RCab (400 nM), HGF (200 ng/ml), or TNF-a (10 ng/ml). Cell lysates were assessed by ELISA for Akt (p-S473). Data were measured as duplicates and shown as means 6 SD. *p # 0.05. 10 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE Downloaded from

FIGURE 7. The transcriptome of RCab-treated ECs differs from that of HGF-treated ECs. (A) Venn diagram illustrating the total number of genes that were regulated by either HGF or RCab or by both. (B and C) of biological processes (GOBP) representing biological reactions and regulatory connections that are affected by RCab (B) or HGF (C).

Among them, 53 genes (39.3%) were specifically induced and Discussion http://www.jimmunol.org/ 1 gene (0.7%) was repressed in the presence of shear force It has been hypothesized that the NRP1–MET signaling axis is in- (Fig. 9B), whereas 34 genes (25.2%) were induced only in its volved in inflammatory processes such as SIRS or snakebite-induced absence (Fig. 9A). Nevertheless, the transcriptomes of RCab- SIRS-like conditions. To unravel the molecular mechanism of this treated interconnected ECs were qualitatively similar, irrespective signaling pathway, the C-type lectin-related protein RCab,asthe of additional shear forces (Supplemental Fig. 3B). In particular, the specific trigger of this signaling axis in ECs, was studied in vitro. proinflammatory effect of RCab was pronounced under both con- RCab increased 1) mechanical tension between and within ECs, ditions, albeit with some quantitative differences. With shear ap- resulting in increased permeability of the vasculature, and 2) TEM of plied, RCab-induced mRNA levels of ICAM-1, E-selectin, CCL2, leukocytes at vascular sites of inflammation (9, 20, 43). Both and CXCL5 were significantly increased in comparison with processes are highly relevant in inflammatory diseases such by guest on September 27, 2021 static conditions (Supplemental Fig. 3C, Table I). Correspond- as atherosclerosis and the life-threating SIRS (3, 4). The ingly, RCab-treated and shear-exposed confluent ECs showed present study shows that the venom component RCab like- significantly increased ICAM-1 protein levels (Supplemental wise induces stiffening of ECs that are tightly interlinked Fig. 3D). It is noteworthy that the protein levels of E-selectin with adjacent cells in a monolayer. This stiffening explains in RCab-treated ECs were also significantly increased under the SIRS-like symptoms caused by some snake venoms (44) shear and showed a diffuse distribution on the cell surface at the inter- and intracellular level, and it provides insight (Supplemental Fig. 3D), which is consistent with low amounts into the underlying signaling through the NRP1–MET axis of E-selectin on the surface of ECs treated with RCab in the and the resulting increased mechanotransduction within a absence of shear force (Fig. 3E). confluent monolayer of tightly interlinked ECs.

Table I. RCab-induced genes related to innate immunity

Gene Symbol GenBank Accession No. Full Gene Name Fold Change Static Fold Change Shear Cytokines and chemokines CCL2 NM_002982 Chemokine (C-C motif) ligand 2 2.88 5.30 CCL20 NM_004591 Chemokine (C-C motif) ligand 20 1.90 3.97 CCL8 NM_005623 Chemokine (C-C motif) ligand 8 2.96 3.15 CX3CL1 NM_002996 Chemokine (C-X3-C motif) ligand 1 1.94 1.68 CXCL1 NM_001511 Chemokine (C-X-C motif) ligand 1 2.66 2.97 CXCL10 NM_001565 Chemokine (C-X-C motif) ligand 10 1.60 3.53 CXCL11 NM_005409 Chemokine (C-X-C motif) ligand 11 1.65 1.78 CXCL2 NM_002089 Cchemokine (C-X-C motif) ligand 2 3.81 3.78 CXCL3 NM_002090 Chemokine (C-X-C motif) ligand 3 3.70 3.60 CXCL5 NM_002994 Chemokine (C-X-C motif) ligand 5 4.58 4.03 CXCL6 NM_002993 Chemokine (C-X-C motif) ligand 6 4.36 3.23 CXCL8 NM_000584 Chemokine (C-X-C motif) ligand 8 1.24 1.62 Adhesion molecules SELE NM_000450 Selectin E 6.24 6.92 VCAM1 NM_001078 VCAM 1 4.60 6.13 ICAM1 NM_000201 ICAM 1 2.89 2.94 VCAN NM_004385 Versican NS 1.56 JAM2 NM_021219 Junctional adhesion molecule 2 NS 1.15 The Journal of Immunology 11 Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 8. RCab also affects coherent ECs under mechanical shear forces. (A) HUVECs preconditioned to shear stress (15 dyn/cm2) were treated with RCab (400 nM) or PBS as control (Ctrl) at constant shear force (15 dyn/cm2). Under shear, RCab-induced cell elongation increased, as quantified by the biometric parameters cell roundness and circularity using VE-cadherin labeling of cell–cell borders (green). To determine cell roundness and circularity, the following numbers of cells were evaluated. For cell roundness: Shear-Ctrl, n = 132 cells, and Shear-RCab, n = 132 cells; for cell circularity: Shear-Ctrl, n = 132 cells, and Shear-RCab, n = 92 cells. Data from three independent experiments are shown as means 6 SD. ****p # 0.0001. (B) Also, with applied shear forces, RCab reduced VE-cadherin density along the cell perimeters, measured as relative fluorescence intensity along the cell junctions. VE-cadherin junctional intensity was evaluated with n = 109 and 96 cells for the conditions Shear-Ctrl and Shear-RCab (Figure legend continues) 12 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 9. The presence or absence of shear force affects the transcriptome of RCab-treated ECs. (A) Venn diagram of all RCab-regulated genes whose expression depends on shear forces. (B) Venn diagrams of RCab-regulated genes whose expression is stimulated (left) or reduced (right) in response to shear forces. (C and D) Gene ontology of biological processes (GOBP) representing biological reactions and regulatory connections that are susceptible to RCab and, in addition, depend on shear force. (C) Static: 81 genes; (D) shear: 101 genes.

Within an endothelial monolayer, the molecular mode of action actomyosin system, consistent with activation of RhoA as a of RCab is based on its binding to NRP1 and, thus, the formation marker for increased force exertion and EC stiffness (20). Re- of a ternary complex with MET as signaling receptor tyrosine markably, focal adhesions as sites of force transmission between kinase (29). Because of increased phosphorylation of paxillin, this ECs and their ECM disappeared in RCab-treated ECs within causes disassembly of focal adhesions, which are the major force- a contiguous monolayer, like in isolated ECs. This left VE- transmitting adhesion contacts of ECs with their ECM (29). This cadherin–containing intercellular contacts between coherent mechanism has already been observed in single ECs seeded on ECs the major sites of force transmission. Interestingly, treat- ECM proteins, such as collagen. In the present work, the effect of ment of coherent ECs with RCab resulted in a redistribution of RCab was investigated on interconnected ECs, which formed a VE-cadherin in dotlike structures at intercellular contact faces, to monolayer and, thus, more closely resembled the endothelial which also vinculin was recruited. Such structures within ECs were lining of blood vessels. It was demonstrated by siRNA-based described as force-dependent focal adherens junctions by Huve- knockdown and by using a MET inhibitor that coherent ECs neers and coworkers (18). Vinculin is considered a marker protein forming a contiguous monolayer respond to RCab also via the of such sites of intercellular force transmission (18) and serves as NRP1–MET axis and consequently disassemble their focal adhe- an anchoring site for cell-spanning F-actin stress fibers. The RCab- sions. In contrast to isolated ECs, confluent ECs formed signifi- induced relocalization and accumulation of VE-cadherin into cantly more F-actin stress fibers and exhibited considerably higher focal adherens junctions was accompanied by a decrease in cell stiffness. Both indicate increased force generation by the the overall homogeneous distribution of VE-cadherin along

respectively. Cropped areas are indicated by white rectangles and highlight the redistribution of VE-cadherin to noncontiguous cell–cell contact sites, which appeared more frequently after RCab-treatment. Data from three independent experiments are shown as means 6 SD. *p # 0.05. (C)A synergistic effect of externally applied shear forces on RCab-induced stress fiber formation and increased anchorage to focal adherens junctions (yellow colocalization signal) was revealed by (immuno)fluorescence staining of VE-cadherin (green) and phalloidin (red) in confluent EC monolayers. Scale bars, 20 mm. The Journal of Immunology 13 intercellular contact faces, which presumably correlates with tissues because of the abluminal expression of its ligand, NRP1, the increased perimeters of elongated cells (13) Consequently, on ECs (31). In the event of a snakebite envenomation, metallo- permeability increased, and transendothelial resistance of EC and serine proteinases in the venom destroy the tight endothelial monolayers decreased (17). These RCab-caused effects reflect vessel lining, and the vasculature becomes leaky (53). Under such in vitro the SIRS-associated increased vessel permeability as a conditions, RCab potentiates the vessel wall barrier-breaching vascular response to inflammation. activity of other venom components and, thus, contributes sig- Focal adhesion junctions transmit forces between ECs and nificantly to the SIRS-like symptoms of snakebite envenomation contribute to cell stiffening, which is notably pronounced in RCab- (3, 4). RCab upregulates ICAM-1 on the luminal face of ECs, treated ECs. Moreover, cell stiffening and RhoA activation are key thereby promoting leukocyte TEM (54, 55). In contrast to HGF, signals by which cells sense mechanical load and promote inter which weakens leukocyte adhesion to untreated ECs comparable alia nuclear translocation of the NF-kB transcription factor to VEGF-A165–induced ECs (37), RCab-triggered cell stiffening (20, 45, 46). NF-kB activated transcription of the EC activation enhances TEM, as leukocytes are durotactically stimulated to marker ICAM-1 and several other genes, most of which are related extravasate (20). Further attachment of leukocytes to ECs in turn to the inflammatory response of ECs (7, 27). From previous increases their cell tension (5). Thus, RCab promotes TEM in a studies, it is known that ICAM-1 expression of ECs is finely positive feedback loop. In summary, our data depict in vitro a regulated by the rigidity of the ECM in a ROCK-independent SIRS-like scenario. manner, with both very soft and very stiff substrates increasing ab ICAM-1 expression (9). In contrast, in our study, RC -dependent Acknowledgments disassembly of focal adhesions abrogated any force transmission Downloaded from We thank B. Schedding for technical assistance. We thank Hans Schnittler between ECs and their ECM. Consequently, mechanical forces for providing equipment and expertise on the fluid shear experiments with between neighboring cells were mainly transmitted via intercellular the cone plate-based system. The Schnittler laboratory also provided the VE-cadherin–mediated contacts, with the Rho–NF-kBsignaling ECs for these experiments. We thank A. Ahle from the Institute of Anatomy pathway actively involved in this redirection of mechanical forces. and Vascular Biology, as well as D. Vestweber and M. Meyer zu Brickwedde Moreover, in RCab-treated coherent ECs, Rho–NF-kBsignaling from the Max Planck Institute of Molecular Biomedicine, Mu¨nster,

and signaling via the ternary RCab–NRP1–MET complex seem Germany, for providing HUVECs. We acknowledge Jakob Matthes for help http://www.jimmunol.org/ to converge. and expertise in transcriptome analysis at MFT-Services, University Notably, MET, when stimulated by its canonical ligand, HGF, of Tu¨bingen. did not elicit an inflammatory response in ECs. HGF treatment of ECs has even been reported to reduce ICAM-1 expression (47). Disclosures Although HGF and RCab use the same receptor tyrosine kinase, The authors have no financial conflicts of interest. MET, to trigger an intracellular signaling cascade in ECs, the respective cellular responses were very different. This is likely References because NRP1 can modulate the way by which MET signals. 1. Radeva, M. Y., and J. Waschke. 2018. Mind the gap: mechanisms regulating the ab by guest on September 27, 2021 Moreover, MET signaling in RC -treated EC monolayers likely endothelial barrier. Acta Physiol. (Oxf.) 222: e12860. is modulated by the cytoskeletal rearrangement of F-actin stress 2. Vestweber, D. 2015. How leukocytes cross the vascular . Nat. Rev. fibers and by the increased intercellular force transmission via Immunol. 15: 692–704. 3. Ait-Oufella, H., E. Maury, S. Lehoux, B. Guidet, and G. Offenstadt. 2010. The VE-cadherin– and vinculin-containing cell–cell contacts. Cell endothelium: physiological functions and role in microcirculatory failure during stiffening, Rho–NF-kB signaling, and NF-kB activation contribute severe sepsis. Intensive Care Med. 36: 1286–1298. ab 4. Opal, S. M., and T. van der Poll. 2015. Endothelial barrier dysfunction in septic to the differential gene expression in response to RC as com- shock. J. Intern. Med. 277: 277–293. pared with HGF. Furthermore, NF-kB has been reported to be 5. Lyck, R., and G. Enzmann. 2015. The physiological roles of ICAM-1 and activated by MET signaling via phosphoinositide-3-kinase, PKB/ ICAM-2 in neutrophil migration into tissues. Curr. Opin. Hematol. 22: 53–59. k 6. Schnoor, M., P. Alcaide, M. B. Voisin, and J. D. van Buul. 2015. Crossing the Akt, and I B phosphorylation (48–50). Consistent with this, we vascular wall: common and unique mechanisms exploited by different leukocyte demonstrated activation of PKB/Akt by phosphorylation of the subsets during extravasation. Mediators Inflamm. 2015: 946509. essential residue S473 (51). In addition, our data showed that 7. Xiao, L., Y. Liu, and N. Wang. 2014. New paradigms in inflammatory sig- naling in vascular endothelial cells. Am. J. Physiol. Heart Circ. Physiol. 306: stimulation of MET by HGF alone is insufficient to induce NF-kB H317–H325. translocation into the nucleus. In fact, only a minority of 25.5% of 8. Siebenlist, U., G. Franzoso, and K. Brown. 1994. Structure, regulation and function of NF-kappa B. Annu. Rev. Cell Biol. 10: 405–455. the differentially regulated genes are similarly activated by either 9. Scott, H. A., B. Quach, X. Yang, S. Ardekani, A. P. Cabrera, R. Wilson, HGF or RCab. Therefore, Rho activation and cell stiffening ap- I. Messaoudi-Powers, and K. Ghosh. 2016. Matrix stiffness exerts biphasic pear to be essential and additive factors for NF-kB translocation control over monocyte-endothelial adhesion via Rho-mediated ICAM-1 clus- tering. Integr. Biol. 8: 869–878. into the nucleus, and, together with the NRP1 coreceptor, account 10. Kruse, K., and Y. A. Komarova. 2015. Tension across adherens junctions: when for the different signaling outcomes of the NRP1–MET complex less is more. Oncotarget 6: 30433–30434. and MET alone. The presence or absence of shear forces had few 11. Efimova, N., and T. M. Svitkina. 2018. Branched actin networks push against each other at adherens junctions to maintain cell-. J. Cell Biol. 217: qualitative effects on RCab-triggered NRP1–MET signaling and 1827–1845. gene expression in interconnected ECs. RCab treatment reinforced 12. Tojkander, S., G. Gateva, and P. Lappalainen. 2012. Actin stress fibers--assem- bly, dynamics and biological roles. J. Cell Sci. 125: 1855–1864. shear forces–induced intracellular force transmission and, thus, 13. Cao, J., M. Ehling, S. Ma¨rz, J. Seebach, K. Tarbashevich, T. Sixta, boosted gene expression and cellular responses induced by shear M. E. Pitulescu, A. C. Werner, B. Flach, E. Montanez, et al. 2017. Polarized forces alone (52). This emphasizes the importance of intercellular actin and VE-cadherin dynamics regulate junctional remodelling and cell mi- gration during sprouting angiogenesis. Nat. Commun. 8: 2210–2230. force-transmitting contacts between ECs in signaling processes. 14. Ru¨egg, C., and G. C. Alghisi. 2010. Vascular integrins: therapeutic and imaging Notably, several inflammatory genes were also activated by RCab, targets of tumor angiogenesis. Recent Results Res. 180: 83–101. consistent with the involvement of NF-kB in the perception of 15. Geiger, B., and K. M. Yamada. 2011. Molecular architecture and function of matrix adhesions. Cold Spring Harb. Perspect. Biol. 3: a005033. external forces in an activated endothelium (6, 20, 23). 16. Bays, J. L., and K. A. DeMali. 2017. Vinculin in cell-cell and cell-matrix ad- RCab is the first component isolated from a snake venom that hesions. Cell. Mol. Life Sci. 74: 2999–3009. 17. Dejana, E., F. Orsenigo, and M. G. Lampugnani. 2008. The role of adherens activates ECs via the NRP1–MET axis. Remarkably, isolated junctions and VE-cadherin in the control of vascular permeability. J. Cell Sci. RCab does not show any effect on intact blood vessels of normal 121: 2115–2122. 14 NRP1–MET SIGNALS AND CUES FROM INTERCELLULAR FORCES CONVERGE

18. Huveneers, S., J. Oldenburg, E. Spanjaard, G. van der Krogt, I. Grigoriev, 36. Carl, P., and H. Schillers. 2008. Elasticity measurement of living cells with an A. Akhmanova, H. Rehmann, and J. de Rooij. 2012. Vinculin associates with atomic force microscope: data acquisition and processing. Pflugers Arch. 457: endothelial VE-cadherin junctions to control force-dependent remodeling. 551–559. J. Cell Biol. 196: 641–652. 37. Buschmann, M. H., P. Dieterich, N. A. Adams, and H. J. Schnittler. 2005. 19. Lampugnani, M. G., M. Corada, L. Caveda, F. Breviario, O. Ayalon, B. Geiger, Analysis of flow in a cone-and-plate apparatus with respect to spatial and tem- and E. Dejana. 1995. The molecular organization of endothelial cell to cell poral effects on endothelial cells. Biotechnol. Bioeng. 89: 493–502. junctions: differential association of plakoglobin, beta-catenin, and alpha-catenin 38. Schnittler, H. J., R. P. Franke, U. Akbay, C. Mrowietz, and D. Drenckhahn. 1993. with vascular endothelial cadherin (VE-cadherin). J. Cell Biol. 129: 203–217. Improved in vitro rheological system for studying the effect of fluid shear stress 20. Huveneers, S., M. J. Daemen, and P. L. Hordijk. 2015. Between Rho(k) and a on cultured cells. Am. J. Physiol. 265: C289–C298. hard place: the relation between vessel wall stiffness, endothelial contractility, 39. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression and cardiovascular disease. Circ. Res. 116: 895–908. data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. 21. Mammoto, A., T. Mammoto, and D. E. Ingber. 2008. Rho signaling and me- Methods 25: 402–408. chanical control of vascular development. Curr. Opin. Hematol. 15: 228–234. 40. Rueden, C. T., J. Schindelin, M. C. Hiner, B. E. DeZonia, A. E. Walter, 22. Lum, H., and K. A. Roebuck. 2001. Oxidant stress and endothelial cell dys- E. T. Arena, and K. W. Eliceiri. 2017. ImageJ2: imageJ for the next generation of function. Am. J. Physiol. Cell Physiol. 280: C719–C741. scientific image data. BMC Bioinformatics 18: 529. 23. Matoba, K., D. Kawanami, M. Tsukamoto, J. Kinoshita, T. Ito, S. Ishizawa, 41. Seebach, J., A. A. Taha, J. Lenk, N. Lindemann, X. Jiang, K. Brinkmann, Y. Kanazawa, T. Yokota, N. Murai, S. Matsufuji, and K. Utsunomiya. 2014. Rho- S. Bogdan, and H. J. Schnittler. 2015. The CellBorderTracker, a novel tool to kinase regulation of TNF-a-induced nuclear translocation of NF-kB RelA/p65 quantitatively analyze spatiotemporal endothelial junction dynamics at the and M-CSF expression via p38 MAPK in mesangial cells. Am. J. Physiol. Renal subcellular level. Histochem. Cell Biol. 144: 517–532. Physiol. 307: F571–F580. 42. Schindelin, J., I. Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, 24. Sun, S. C., J. H. Chang, and J. Jin. 2013. Regulation of nuclear factor-kBin S. Preibisch, C. Rueden, S. Saalfeld, B. Schmid, et al. 2012. Fiji: an open-source autoimmunity. Trends Immunol. 34: 282–289. platform for biological-image analysis. Nat. Methods 9: 676–682. 25. Tong, L., and V. Tergaonkar. 2014. Rho protein GTPases and their interactions 43. Urbano, R. L., C. Furia, S. Basehore, and A. M. Clyne. 2017. Stiff substrates with NFkB: crossroads of inflammation and matrix biology. Biosci. Rep. 34: increase inflammation-induced endothelial monolayer tension and permeability. 283–295. Biophys. J. 113: 645–655. 26. Prystopiuk, V., B. Fels, C. S. Simon, I. Liashkovich, D. Pasrednik, C. Kronlage, 44. Sanhajariya, S., S. B. Duffull, and G. K. Isbister. 2018. Pharmacokinetics of Downloaded from R. Wedlich-So¨ldner, H. Oberleithner, and J. Fels. 2018. A two-phase response of snake venom. Toxins (Basel) 10: 73. endothelial cells to hydrostatic pressure. J. Cell Sci. 131: jcs206920. 45. Perona, R., S. Montaner, L. Saniger, I. Sa´nchez-Pe´rez, R. Bravo, and J. C. Lacal. 27. Pober, J. S., and W. C. Sessa. 2007. Evolving functions of endothelial cells in 1997. Activation of the nuclear factor-kappaB by Rho, CDC42, and Rac-1 inflammation. Nat. Rev. Immunol. 7: 803–815. proteins. Genes Dev. 11: 463–475. 28. Leo´n, G., L. Sa´nchez, A. Herna´ndez, M. Villalta, M. Herrera, A. Segura, 46. Yu, O. M., and J. H. Brown. 2015. G protein-coupled receptor and RhoA- R. Estrada, and J. M. Gutie´rrez. 2011. Immune response towards snake venoms. stimulated transcriptional responses: links to inflammation, differentiation, and Inflamm. Allergy Drug Targets 10: 381–398. cell proliferation. Mol. Pharmacol. 88: 171–180.

29. Niland, S., B. Ditkowski, D. Parrandier, L. Roth, H. Augustin, and J. A. Eble. 47. Min, J. K., Y. M. Lee, J. H. Kim, Y. M. Kim, S. W. Kim, S. Y. Lee, Y. S. Gho, http://www.jimmunol.org/ 2013. Rhodocetin-ab-induced neuropilin-1-cMet association triggers restruc- G. T. Oh, and Y. G. Kwon. 2005. Hepatocyte growth factor suppresses vascular turing of matrix contacts in endothelial cells. [Published erratum appears in 2013 endothelial growth factor-induced expression of endothelial ICAM-1 and Arterioscler. Thromb. Vasc. Biol. 33: e113.] Arterioscler. Thromb. Vasc. Biol. 33: VCAM-1 by inhibiting the nuclear factor-kappaB pathway. Circ. Res. 96: 300– 544–554. 307. 30. Sulpice, E., J. Ploue¨t, M. Berge´, D. Allanic, G. Tobelem, and T. Merkulova- 48. Trusolino, L., A. Bertotti, and P. M. Comoglio. 2010. MET signalling: principles Rainon. 2008. Neuropilin-1 and neuropilin-2 act as coreceptors, potentiating and functions in development, organ regeneration and cancer. Nat. Rev. Mol. Cell proangiogenic activity. Blood 111: 2036–2045. Biol. 11: 834–848. 31. Niland, S., D. Komljenovic, J. Macas, T. Bracht, T. Ba¨uerle, S. Liebner, and 49. Fan, S., M. Gao, Q. Meng, J. J. Laterra, M. H. Symons, S. Coniglio, J. A. Eble. 2018. Rhodocetin-ab selectively breaks the endothelial barrier of the R. G. Pestell, I. D. Goldberg, and E. M. Rosen. 2005. Role of NF-kappaB sig- tumor vasculature in HT1080 fibrosarcoma and A431 epidermoid carcinoma naling in hepatocyte growth factor/scatter factor-mediated cell protection. On- tumor models. Oncotarget 9: 22406–22422. cogene 24: 1749–1766.

32. Eble, J. A., M. McDougall, G. L. Orriss, S. Niland, B. Johanningmeier, 50. Mu¨ller, M., A. Morotti, and C. Ponzetto. 2002. Activation of NF-kappaB is by guest on September 27, 2021 G. Pohlentz, M. Meier, S. Karrasch, M. I. Esteva˜o-Costa, A. Martins Lima, and essential for hepatocyte growth factor-mediated proliferation and tubulogenesis. J. Stetefeld. 2017. Dramatic and concerted conformational changes enable Mol. Cell. Biol. 22: 1060–1072. rhodocetin to block a2b1 selectively. [Published erratum appears in 51. Risso, G., M. Blaustein, B. Pozzi, P. Mammi, and A. Srebrow. 2015. Akt/PKB: 2017 PLoS Biol. 15: e1002613.] PLoS Biol. 15: e2001492. one kinase, many modifications. Biochem. J. 468: 203–214. 33. Seebach, J., G. Donnert, R. Kronstein, S. Werth, B. Wojciak-Stothard, 52. Kronstein, R., J. Seebach, S. Grossklaus, C. Minten, B. Engelhardt, M. Drab, D. Falzarano, C. Mrowietz, S. W. Hell, and H. J. Schnittler. 2007. Regulation of S. Liebner, Y. Arsenijevic, A. A. Taha, T. Afanasieva, and H. J. Schnittler. 2012. endothelial barrier function during flow-induced conversion to an arterial phe- Caveolin-1 opens endothelial cell junctions by targeting catenins. Cardiovasc. notype. Cardiovasc. Res. 75: 596–607. Res. 93: 130–140. 34.Schillers,H.,I.Medalsy,S.Hu,A.L.Slade,andJ.E.Shaw.2016.PeakForce 53. Gutie´rrez, J. M., J. J. Calvete, A. G. Habib, R. A. Harrison, D. J. Williams, and tapping resolves individual microvilli on living cells. J. Mol. Recognit. 29: D. A. Warrell. 2017. Snakebite envenoming. Nat. Rev. Dis. Primers 3: 17063. 95–101. 54. Vestweber, D. 2012. Relevance of endothelial junctions in leukocyte extrava- 35. Schillers, H., C. Rianna, J. Scha¨pe, T. Luque, H. Doschke, M. Wa¨lte, sation and vascular permeability. Ann. N. Y. Acad. Sci. 1257: 184–192. J. J. Uriarte, N. Campillo, G. P. A. Michanetzis, J. Bobrowska, et al. 2017. 55. Nourshargh, S., P. L. Hordijk, and M. Sixt. 2010. Breaching multiple barriers: Standardized nanomechanical atomic force microscopy procedure (SNAP) for leukocyte motility through venular walls and the interstitium. Nat. Rev. Mol. Cell measuring soft and biological samples. Sci. Rep. 7: 5117. Biol. 11: 366–378. A VE-cadherin Vinculin Merge B Ctrl VE-cadherin Vinculin 150 50 µm

l 100 r t C

- 50 h 3 0 0 10 20 30 20 µm

RCαβ 150 VE-cadherin Vinculin β

α 100 C R - 50 h 3 0 0 10 20 30 20 µm Distance[μm]

Supplementary Fig. 1 Supplementary figure S1. RCαβ induces recruitment of VE-cadherin and vinculin into focal adherens junctions, already within 3 h. (A) RCαβ reduced VE-cadherin at intercellular contact faces of elongated HUVECs within 3 h. Simultaneously, VE-cadherin was concentrated at punctual intercellular focal adherens junctions, which also contained vinculin. Vinculin was concentrated at cell borders after RCαβ treatment. Double immunofluorescence staining of VE-cadherin (green) and vinculin (red) in confluent HUVECs after 3h treatment with PBS (control) and RCαβ (400 nM). Nuclei (blue) were stained with 4’, 6’-diamidino-2-phenylindole (DAPI). Scale bar: 20 μm. (B) Fluorescence intensity scans along the intercellular contact faces, indicated by white lines in (A). A B

Supplementary Fig. 2 Supplementary figure S2. TNF-α-treatment of HUVECs dose-dependently increases tethering and rolling of monocytic U937 cells. Percentage of (A) rolling and (B) adhering U937 cells was determined as described in Materials and Methods. Bar charts from a representative experiment show means ± SD from 4 videos. A 2 β

α 1 C R 0

F -1 G H -2

B l r

t 6 C c

. i s t v

a 4 t β α S C R

l 2 r t C

. w s 0 v o

l β F α

C -2 R

C D o

t

o t e

ICAM-1 SELE r e a 80 r 200 a p p m l m

150 l

o 60 r o r e t c t c

c c c

- ICAM-1 n - n 40 n

c 100 c o o 2.0 e i i i i t t y s Shear-Ctrl Shear-RCαβ c s t a a s i s t t s 20 e 1.5

s 50 s e s r e r r n p o d e p x 1.0 t

0 a u e x

l l 0 l n

l c fl e β r r β β r i d β

r -

t t

l t t α α α α l 0.5 E d

C C o

C C a l - C - C - C - C t V c r F r o R R c i R R - 0.0 - a - i o t t - a - l F e r r 1 a c c e T r i a a i - β t t h t t h a t S a e e α S a M C t h S t S - h 20 µm r C S S S S A R a -

C e

r I h a o

o Shear-Ctrl t Shear-RCαβ S e t

h e e CCL2 CXCL5 e S r r c a a 40 n

p 300 p 2.5 SELE e d y c m m a l t 2.0 l i o s r c o r

30 t c - s e t

c c

r 1.5

n E

- 200 c n - o c n e o V i i t

c 1.0 -

20 u t o i s i n t a s fl E i t

s

e 0.5 a l L s 100 t s r

10 a E

s 20 µm e p t 0.0 r x S o e p l T

β r x 0 α d 0 t

l l l e β β l β l β C r r r r C α t α o α - t t α t R

d r

C C F - l C C C C - - C C a r R R - - o r R R c - - c - r - e a i a r i a h F t c t c r e a i e a a i e a S h t t h t t a e a h e S S t S h S t S h S S S S Supplementary Fig. 3 Supplementary figure S3. Microarray-based transcriptome analysis and validation of statistically significant DEGs whose expression is regulated in EC monolayers. (A) by RCαβ and HGF or (B) by RCαβ under static and flow conditions. Their log2 fold changes (FDR < 0.05) are illustrated in heatmaps. Green and red color represent a decrease and increase, respectively, of fold change according to the scale on the right. (C) Quantification of mRNAs for ICAM-1, E-selectin (SELE), CCL2, and CXCL5 in RCαβ-treated and untreated confluent HUVECs under static and flow conditions. Means ± SD are shown in the bar charts (*, p ≤ 0.05; **, p ≤ 0.01). (D) HUVECs, treated in the presence of shear force with RCαβ and PBS as control (Ctrl), respectively, were immunofluorescentlystained for VE-cadherin (green), ICAM-1 (red), and SELE (red). Nuclei were counterstained with DAPI. Shear force-exposed HUVECs responded to RCαβ with increased surface expression of ICAM-1 and E-selectin. Scale bar: 20 µm. Shear force- induced changes in total fluorescent intensity of ICAM- 1 and E-Selectin were quantified using such immunofluorescence micrographs. Means ± SD are shown in the bar charts (*, p ≤ 0.05; **, p ≤ 0.01).