doi:10.1182/blood-2009-11-254557Prepublished online July 14, 2010;2010 116: 3356-3366

Peter CarmelietNy, Michael Simons, Mieke Dewerchin, Stefan Schulte-Merker, Elisabetta Dejana, Kari Alitalo and

AnneliiFrederik De Smet, Evisa Gjini, Kristof Anthonis, Bin Ren, Dontcho Kerjaschki, Monica Autiero, Karlien Hermans, Filip Claes, Wouter Vandevelde, Wei Zheng, Ilse Geudens, Fabrizio Orsenigo, Role of synectin in lymphatic development in zebrafish and frogs

http://bloodjournal.hematologylibrary.org/content/116/17/3356.full.htmlUpdated information and services can be found at:

(333 articles)Vascular Biology (291 articles)Phagocytes, Granulocytes, and Myelopoiesis

Articles on similar topics can be found in the following Blood collections

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Washington DC 20036.by the American Society of Hematology, 2021 L St, NW, Suite 900, Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly

For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

VASCULAR BIOLOGY

Role of synectin in lymphatic development in zebrafish and frogs*Karlien Hermans,1,2 *Filip Claes,1,2 Wouter Vandevelde,1,2 Wei Zheng,3 Ilse Geudens,1,2 Fabrizio Orsenigo,4

Frederik De Smet,1,2 Evisa Gjini,5 Kristof Anthonis,1,2 Bin Ren,6 Dontcho Kerjaschki,7 Monica Autiero,1,2 Annelii Ny,1,2

Michael Simons,6 Mieke Dewerchin,1,2 Stefan Schulte-Merker,5 Elisabetta Dejana,4,8 Kari Alitalo,3 and Peter Carmeliet1,2

1Vesalius Research Center, Vlaams Iustituut voor Biotechnologie (VIB), Leuven, Belgium; 2Vesalius Research Center, Katholieke Universiteit Leuven,Leuven, Belgium; 3Molecular Cancer Biology Program, Biomedicum Helsinki, Haartman Institute, University of Helsinki, Helsinki, Finland; 4IFOM, FIRCInstitute of Molecular Oncology, Milan, Italy; 5Hubrecht Institute-KNAW and UMC, Utrecht, The Netherlands; 6Section of Cardiovascular Medicine, YaleUniversity School of Medicine, New Haven, CT; 7Institute of Clinical Pathology, Medical University of Vienna, Vienna, Austria; and 8Department ofBiomolecular Sciences and Biotechnologies, University of Milan, Milan, Italy

The molecular basis of lymphangiogen-esis remains incompletely character-ized. Here, we document a novel role forthe PDZ domain-containing scaffold pro-tein synectin in lymphangiogenesis us-ing genetic studies in zebrafish andtadpoles. In zebrafish, the thoracic ductarises from parachordal lymphangio-blast cells, which in turn derive fromsecondary lymphangiogenic sprouts

from the posterior cardinal vein. Morpho-lino knockdown of synectin in zebrafishimpaired formation of the thoracic duct,due to selective defects in lymphangio-genic but not angiogenic sprouting. Syn-ectin genetically interacted with Vegfr3and neuropilin-2a in regulating lym-phangiogenesis. Silencing of synectinin tadpoles caused lymphatic defectsdue to an underdevelopment and im-

paired migration of Prox-1! lymphaticendothelial cells. Molecular analysis fur-ther revealed that synectin regulatedSox18-induced expression of Prox-1 andvascular endothelial growth factor C–induced migration of lymphatic endothe-lial cells in vitro. These findings reveal anovel role for synectin in lymphaticdevelopment. (Blood. 2010;116(17):3356-3366)

Introduction

The lymphatic vasculature regulates interstitial fluid homeosta-sis, fat resorption, immune defense, inflammation and cancermetastasis. The molecular basis of the lymphatic developmentremains incompletely understood.1 Lymphatic endothelialcells (LECs) derive from venous blood endothelial cells(BECs).2-5 Sox186 and Prox-1,2-5 as well as miRNAs,7 play akey role in this process. Intriguingly, despite their venousorigin, formation of lymphatics relies in part on signals thatparticipate in arterial development. For instance, the forkheadtranscription factors Foxc1/2 are required for arterial specifica-tion and sprouting of LECs.8,9 In addition, the PDZ interactionsite in EphrinB2, itself a marker of arterial endothelial cells(ECs), is essential for lymphatic development.10 Similarly,Dll4/Notch signaling regulates arterial and lymphaticdevelopment.11

Recently, we identified the PDZ domain–containing scaffoldprotein synectin (GIPC1) as a regulator of arterial but not venousgrowth,12 at least in part through control of Vegfa signaling.13

Prompted by the finding that similar factors control arterial andlymphatic development, we studied whether synectin also regu-lated lymphatic development. Although PDZ domain–dependentsignaling is crucial for lymphatic development,10 a role for synectinin lymphatic development has not been documented yet. Wetherefore explored a role of synectin in this process using zebrafishand Xenopus tadpole models.

Methods

Zebrafish analysis

Fli1:eGFPy1,14 Gata-1:DsRed15 zebrafish and PLC!y10 16 zebrafish weremaintained under standard conditions. All morpholinos were previouslyreported and purchased from Gene Tools (supplemental Table 1;available on the Blood Web site; see the Supplemental Materials link atthe top of the online article). Different doses of morpholinos wereinjected into single- to 4-cell stage embryos, as previously described.12

Phenotyping is described in supplemental material. All animal experi-mentation was approved by the Katholieke Universiteit Leuven institu-tional ethical committee.

Xenopus analysis

The generation and characterization of transgenic Flk1:eGFP Xenopuslaevis frogs will be reported elsewhere (manuscript in preparation). Eggswere obtained by natural mating of hormonally induced Flk1:eGFPfemales and wild-type males and injected with different doses of synectin-specific or control morpholinos (Gene Tools, supplemental Table 1) into the2-cell stage.17 Nonoverlapping translational blocking synectin morpholinoswere designed based on published GenBank Xenopus laevis sequences(NM_001088594) and morpholino efficiency was tested using an in vitroluciferase reporter assay17 and immunoblotting (supplemental Methods).Tadpoles were kept in tadpole growth medium (0.1 ! MMR) at 18°C untilgastrulation was completed and from then on at 22°C.17 Phenotyping isdescribed in supplemental Methods.

Submitted November 16, 2009; accepted June 30, 2010. Prepublishedonline as Blood First Edition paper, July 14, 2010; DOI 10.1182/blood-2009-11-254557.

*K.H. and F.C. contributed equally to this work.

The online version of this article contains a data supplement.

The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 USC section 1734.

© 2010 by The American Society of Hematology

3356 BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17

For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

Immunohistochemistry and in situ hybridization

Fli1:eGFPy1 zebrafish embryos were fixed overnight in 4% paraformalde-hyde/PBS at 4°C and processed for paraffin- or cryosectioning. Sections(7 "m) were stained using a polyclonal chicken anti–human synectinantibody (Abcam) and a rabbit anti-GFP antibody (Torrey-Pines Biolabs),which were then detected using biotinylated anti-chicken (Abcam) andAlexa Fluor 488–conjugated anti-rabbit secondary antibodies, respectively.Sections were mounted with Vectashield DAPI (4#,6-diamidino-2-phenylindole; Vector Laboratories).

For Xenopus in situ hybridizations, embryos were fixed in Memfafixative and whole-mount in situ hybridization using antisense probes forProx-117 and synectin (primer sequence, supplemental Table 1) wasperformed as described.17 Analysis of Prox-1–stained areas in the tadpoletail was performed as reported (supplemental Methods).17 For zebrafishwhole-mount in situ hybridization, dechorionated embryos were fixedovernight in 4% paraformaldehyde at 4°C. In situ hybridization wasperformed as described,12 using antisense probes for EphrinB2a,12 Gridlock(Grl),18 EphB4,19 Flt4,12 and Tie-2.20 Both sections and whole-mounts werevisualized on a Zeiss Axioplan 2 imaging microscope.

Cell culture experiments

Human microvascular lung endothelial cells (HMVEC-Lly; Lonza,Invitrogen) and adult dermal LECs (HMVEC-dLyAd; Lonza, Invitrogen),telomerase transfected dermal LECs (hTERT-HDLECs21), immortalized (i)LECs, and human umbilical vein ECs (Lonza, Invitrogen) were grown inendothelial growth medium (EGM)-2–MV medium (Lonza, Invitrogen) at37°C. Human dermal LEC (HDLEC) cells were obtained from Promocelland cultured in endothelial medium provided by the supplier. Synectinexpression was evaluated by quantitative reverse-transcription polymerasechain reaction (qRT-PCR; supplemental Table 1 primer sequences). TheLEC spheroid and Sox18-mediated lymphatic reprogramming assays aredescribed in supplemental Methods.

Statistical analysis

Absolute values were used to calculate mean $ SEM. Significance levelswere calculated by unpaired Student t test, univariate, or multivariateanalysis with the treatment groups as fixed factor and experiment ascovariate. To determine the penetrance of the zebrafish and tadpolephenotypes, we counted the number of embryos, exhibiting the (differentseverities of) morphant phenotype. We used %2 analysis to determinewhether the severity distribution differed between treatment groups.

Results

Synectin is expressed in LECs in zebrafish

Because the expression of synectin in lymphatics has not beendocumented yet, we analyzed its expression pattern during lym-phangiogenesis in zebrafish embryos. For reasons of clarity, wewill briefly explain first how lymphatics develop in zebrafish(supplemental Figure 1).11,20,22-25 The thoracic duct (TD) is the firstperfused lymphatic, located between the dorsal aorta (DA) andposterior cardinal vein (PCV). TD development initiates between30 and 50 hours postfertilization (hpf) when secondary sproutsbranch off dorsally from the PCV. Half of these sprouts areangiogenic, as they connect to primary (arterial) intersomiticvessels (aISVs), which thereby become venous ISVs (vISVs). Theother half of these secondary sprouts are lymphangiogenic, as theymigrate dorsally to the horizontal myoseptum, where they form (by36-60 hpf) a transient, nonlumenized string of parachordal lym-phangioblast (PL) cells, termed so because they are precursors ofLECs forming the TD. Beyond 60 hpf, PL cells navigate ventrallyand dorsally alongside aISVs, with ventral sprouts migrating to

their location between the DA and PCV to fuse and establish theTD (3–6 days postfertilization [dpf]).

We and others previously reported that synectin is expressedin the PCV when secondary lymphangiogenic sprouts form,12,26

but did not characterize synectin expression in lymphaticvessels. When using Fli1:eGFPy1 embryos, which expressenhanced green fluorescent protein (eGFP) in blood and lymphvessels,14,25 we found that synectin was expressed in the DA,PCV, PL cells, and TD (Figure 1A-F) by double immunostainingfor GFP and synectin in 3- and 7-dpf embryos. Furthermore, inagreement with previous findings,12,26 synectin was detectable inthe head (not shown) and in the neural tube, pronephric duct,somites, and gut (Figure 1A-F). Widespread expression ofsynectin has been observed in human, mouse, and Xenopustissues.27-29 Of note, synectin was also detectable in primary andimmortalized human LECs (supplemental Figure 2), consistentwith previous observations.30

Incomplete silencing of synectin in zebrafish does notaffect angiogenesis

To study the role of synectin in lymphangiogenesis, we silencedits expression (synectinKD) in Fli1:eGFPy1 zebrafish embryosusing previously characterized SynATG1 and SynATG2 morpholi-nos.12 As synectin regulates angiogenesis, we first determined arange of submaximal morpholino doses, which only minimallyaffected blood vessel development, to avoid that angiogenicdefects secondarily caused lymphatic defects. At a dose of 9 ngSynATG1 and 6 ng SynATG2 per embryo, only 7% of synectinKD

embryos displayed subtle vascular defects, including a slightlythinner DA and a few malformed ISVs (Figure 2A-B andsupplemental Table 2). The remainder of the blood vasculaturein the head, trunk, tail and subintestinal vessels developednormally, had a normal size and shape, and exhibited compa-rable branching and density (supplemental Figure 3). Blood flowin large axial vessels and ISVs was normal in 2-dpf synectinKD

Gata-1:DsRed embryos, harboring DsRed& erythrocytes15 (Fig-ure 2C-D). Furthermore, arterio-venous differentiation of largeaxial vessels and primary ISVs was normal in synectinKD

embryos, as evidenced by in situ hybridization for arterial andvenous markers at 28 hpf (Figure 2E-L). Only morphantembryos with normal trunk circulation and body size andwithout developmental delay, tissue malformations, edema, ortoxic signs were analyzed. (We refer to synectinKD embryoswithout mentioning that silencing was incomplete.)

Silencing of synectin in zebrafish impairs TD formation

We studied TD formation by measuring its length at 7 dpf, whenthis vessel was completely formed in control embryos, usingprevious methods.11 Briefly, the TD length was measured in10 somites and expressed as percentage of this trunk fragment(supplemental Figure 4).11 Because the penetrance of the lymphaticphenotype was variable (supplemental Note 1), we counted thefraction of embryos with severe, intermediate, or subtle lymphaticdefects. For SynATG1 and SynATG2, the severity and penetrance ofthe knockdown phenotypes were dose-dependent (supplementalTables 3-4); for reasons of brevity, only the most penetrantphenotype is shown (9 ng of SynATG1). In addition, lymphaticdevelopment in zebrafish occurs in a metameric pattern, sincelymphangiogenic sprouts, and hence the PL and TD, form,statistically, at every second unilateral somite segment, giving rise

SYNECTIN IN LYMPHATIC DEVELOPMENT 3357BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

to discontinuous TD fragments that fuse with each other toestablish a continuous perfused TD (supplemental Figure 1).11

In control embryos, injected with a 5-base-pair mismatchmorpholino (SynCtr), the TD developed as a continuous lymphaticby 7 dpf (Figure 3A,C,E and supplemental Table 3). In contrast,32% of synectinKD embryos completely lacked a TD by 7 dpf(“severe” defects; Figure 3B,D-E and supplemental Table 3).Follow-up studies further revealed that TD development was notrescued at 10 dpf (supplemental Figure 3E-F), indicating thatlymphatic formation was aborted and not merely delayed in thesesynectinKD embryos. In another 14% of morphants, TD rudimentsdeveloped over only 1 to 3 somites into an incomplete string ofdisconnected segments (“intermediate” defects; Figure 3E andsupplemental Table 3). Incomplete TD formation over 30%-90% ofits normal length was further observed in 36% of synectinKD

embryos (“subtle” defects; Figure 3E and supplemental Table 3),while TD development was normal in the remainder of themorphant embryos. Because the TD develops in a metamericpattern (see above), the finding that the TD formed over forinstance only 20% of its length implies that TD development wasaborted in the 8 other of the 10 somites analyzed. Thus, silencing ofsynectin aborted lymphatic development completely in all somitesin a third of all morphants, and partially in a fraction of the somitesin another 50% of the morphants.

Knockdown of synectin impairs early lymphaticdevelopment in zebrafish

We then explored whether the TD defect in synectinKD morphantswas related to a defect in PL formation by quantifying its length in10 somites in Fli1:eGFPy1 embryos at 60 hpf (using a similarmethod as used for the TD). In control embryos, PL cells weredetected in nearly every somite segment (Figure 4A,C and supple-mental Table 4). By contrast, the PL was absent in 28% ofsynectinKD embryos (Figure 4B-C and supplemental Table 4) orformed over only 10%-30% of its normal length in another 27% ofmorphants (Figure 4C and supplemental Table 4). An additional40% of synectinKD embryos displayed PL cells in only 3 to9 somites, while in the remaining 5%, the PL string developednormally. The close correlation between TD and PL defectssuggests that silencing of synectin impaired early lymphaticformation.

Knockdown of synectin impairs secondary sprout formation

We then evaluated whether absence of PL cells resulted fromlymphangiogenic sprout defects. In case lymphangiogenic butnot angiogenic sprout formation would be defective, we ex-pected that 50% of all secondary sprouts would be affected/absent. Several assays were used to address this issue. First, we

Figure 1. Synectin is expressed in lymphatic vessels in zebrafish embryos. Scale bars represent 10 "m in panels A-F. In all panels, the dorsal side of the embryo isat the top of the figure. G indicates gut; NT, neural tube; P, pronephric duct; and S, somite. (A-C) Transverse sections through the trunk of a 3-dpf wild-type Fli1:eGFPy1

zebrafish embryo, stained by immunohistochemistry using a polyclonal anti–human synectin antibody, and by DAPI nuclear stain, revealing prominent synectinexpression in the NT, G, S, dorsal aorta (DA, white arrow), posterior cardinal vein (PCV, green arrow) and parachordal lymphangioblast (PL) cells (white arrowhead).Insets show magnification of the boxed areas (PL cells) in each panel. (D-F) Using the same staining procedure on transverse sections through the rostral trunk of a 7dpf wild-type Fli1:eGFPy1 zebrafish embryo, similar widespread synectin expression in the NT, G, S, P, DA (white arrow), PCV (green arrow), and thoracic duct (whitearrowhead) was observed.

3358 HERMANS et al BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

used whole-mount in situ staining for Tie-2 to identify bothangiogenic and lymphangiogenic secondary sprouts at 48 hpf. In81% of control embryos (n ' 66), 8 to 10 secondary sproutsdeveloped in the 10 unilateral somites analyzed, (ie, in nearlyevery somite [Figure 4D-D#]); in another 17%, 6 to 8 secondarysprouts developed, while very rarely (in 2%), only half thenumber of secondary (4 to 5) sprouts formed. By contrast, in22% of synectinKD embryos, only half of the normal number ofsecondary sprouts formed (n ' 74; P ( .001; Figure 4E-E#),while in another 41%, only 6 to 8 Tie-2& sprouts developed(N ' 74; P ( .001 vs control), while the remainder fractiondisplayed no secondary sprouting defects. Notably, comparablefractions of morphant embryos suffered severe secondary sproutdefects, and lacked a PL and TD, suggesting that the latter andformer processes were linked.

Second, we used the genetic PLC!1y10 model to visualizesecondary sprouts in isolation of primary ISVs. Indeed, in thesemutant embryos, aISVs fail to form, while secondary sproutsemanate normally.16 Synectin silencing in PLC!1y10xFli1:eGFPy1 embryos showed that only half the normal number ofsecondary sprouts developed by 48 hpf (sprouts in 10 somites:7.3 $ 0.9 in control vs 4.0 $ 0.8 in synectinKD; N ' 53-57embryos, P ( .01; Figure 4F-H). Thus, also with this method,approximately 50% fewer secondary sprouts branched off insynectinKD embryos.

Silencing of synectin does not alter secondaryangiogenic sprouting

We then assessed whether synectin silencing impaired lymphangio-genic sprouting using an indirect method (ie, by counting thenumber of ISVs that were connected to the PCV [vISVs]), as thelatter are established when angiogenic secondary sprouts connectto primary ISVs. Because a comparable fraction of lymphangio-genic and angiogenic sprouts normally emanates from the PCV, anormal number of vIVSs, in combination with a reduced totalnumber of secondary sprouts, would be evidence for defectiveformation of lymphangiogenic sprouts. In control 4 dpf Fli1:eGFPy1 embryos (N ' 29), 55% of the ISVs were connected to thePCV and thus venous, while the remaining fraction was connectedto the DA and hence arterial (Figure 4I). Notably, in the mostseverely affected synectinKD embryos, lacking the PL completelyor forming only a partial PL in 30% of the somites (N ' 18), half ofthe ISVs was still connected to the PCV (Figure 4I), showing thatangiogenic sprouts normally emanated from the PCV and implyingthat lymphangiogenic sprouting was impaired in synectinKD em-bryos. As a lymphangiogenic sprouting defect might reflect a defectin migration and/or emergence of the lymphatic lineage, but thedynamics of Prox-1 expression during lymphatic differentiationcan be better visualized in frog than zebrafish embryos, we resortedto tadpoles, a previously validated model to study lymphangiogen-esis,17 to dissect the underlying mechanism in more detail.

Figure 2. Normal blood vessel development in synectinKD embryos. In all panels, head faces left and dorsal is up. Scale bars represent 50 "m in panels A-Band E-L, and 150 "m in panels C-D. DA indicates dorsal aorta; DLAV, dorsal longitudinal anastomosing vessel; and PCV, posterior cardinal vein.(A-B) Confocal side views of the trunk of 48-hpf control (A) and synectinKD (B) Fli1:eGFPy1 zebrafish embryos, showing normal vascular morphogenesis in the morphantembryo. Large arrow indicates the DLAV while the small arrow indicates an intersegmental vessel. (C-D) Macroscopic side views of 48-hpf control (C) and synectinKD

(D) Gata-1:DsRed embryos expressing DsRed specifically in erythrocytes. SynectinKD embryos show a normal distribution of erythrocytes throughout the body,indicating a fully functional circulation. (E-L) Bright-field side views of the trunk of 28-hpf control (E-F,I-J) and synectinKD (G-H,K-L) embryos stained by whole-mount insitu hybridization for a panel of arterial (Grl, E,G; EphrinB2a, F,H) and venous markers (Flt4, I,K; EphB4, J,L). No difference in expression of any of these markers couldbe observed in the synectinKD embryos, indicating that arterio-venous differentiation had occurred normally in these embryos. Red and blue lines denote DA and PCV,respectively.

SYNECTIN IN LYMPHATIC DEVELOPMENT 3359BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

Knockdown of synectin impairs lymphangiogenesis in tadpoles

We first analyzed synectin expression in tadpoles. In agreementwith previous findings,28 whole-mount in situ hybridization at stage35-40 revealed expression of synectin in the neural tube, eye, brain,otic vesicle, somites, and branchial arches, as well as at locations ofthe DA, PCV, and dorsal longitudinal anastomosing vessel (DLAV),where the ventral caudal lymph vessel (VCLV) and dorsal caudallymph vessel (DCLV) develop in the frog embryo (not shown andsupplemental Figure 5A-F). We also used a novel ‘LEC labeling’technique in transgenic Flk1:eGFP frogs, that express eGFP inblood and lymph vessels (generation of this line will be reportedelsewhere). Upon intracardial injection and extravasation fromblood vessels, tetramethyl rhodamine isothiocyanate (TRITC)-dextran is selectively taken up by LECs, allowing isolation ofdoubly labeled yellow LECs and singly labeled green BECs byflow cytometry. RT-PCR showed that sorted BECs and LECsexpressed synectin (synectin/105 )-actin mRNA copies: 874 $ 625for BECs vs 1415 $ 822 for LECs; N ' 4).

We then evaluated whether silencing of synectin impairedlymphatic development. In tadpoles, BECs from the PCV, express-

ing Prox-1, migrate ventrally to form the VCLV, or dorsally to thelevel of the DLAV to establish the DCLV.17 To knockdownsynectin, Flk1:eGFP tadpoles were injected with the nonoverlap-ping morpholinos xSynATG1 or xSynATG2, targeting the ATG of theXenopus laevis synectin mRNA. As both morpholinos blockedtranslation (supplemental Figure 6) and caused indistinguishablephenotypes (supplemental Table 5), only results for xSynATG1 areshown. We first determined a range of submaximal morpholinodoses, that minimally affected vascular development. Screening ofGFP& tadpoles at stage 45 (5 dpf) showed that, at 30 ng morpholinoper embryo, 17% of GFP& tadpoles had only subtle vasculardefects (reduced complexity of capillary network interconnectingISVs; Figure 5A-B and supplemental Table 5). In line herewith,blood flow in axial vessels was normal in most tadpoles (supplemen-tal Table 5).

To characterize lymphatic development, we focused on theformation of the VCLV and DCLV. LECs in Flk1:eGFP tadpoleswere labeled with TRITC-dextran to distinguish them from GFP&

BECs. Imaging of control tadpoles at stage 47 revealed that theDCLV and VCLV formed a regular continuous vessel over its entirelength (Figure 5A,C,E). By contrast, in 57% of synectinKD tad-poles, the DCLV was incompletely formed and dysmorphogenic,consisting of isolated LEC clusters, distributed discontinuouslyover the trunk (Figure 5B,D and supplemental Table 5). The VCLVappeared largely normal in most synectinKD tadpoles, at leastmorphologically, while other morphants showed an irregularstructure (Figure 5B,F). Lymphangiography at stage 45 revealedthat the DCLV and VCLV were functional in all control tadpoles(not shown and Figure 5G). When analyzing the DCLV in15 synectinKD tadpoles, the dye was not drained at all or only over ashort distance, indicating that this lymphatic was dysfunctional(not shown). Similarly, despite its apparent normal morphologicappearance, the VCLV failed drain the dye (Figure 5H). Wepreviously noticed that the DCLV is more severely affected thanthe VCLV upon silencing of Vegfr3,17,31 Vegfc (unpublished) orLiprin)-1,32 presumably because LECs, arising from the PCV, haveto migrate further to form the distant DCLV than the nearby VCLV.In line with these findings, 32% of synectinKD tadpoles sufferedlymphedema around the heart and gut (Figure 5I-J and supplemen-tal Table 5). Thus, synectin also regulates lymphatic developmentin the tadpole model.

Synectin regulates lymphatic differentiation and migrationin tadpoles

The tadpole model allows study of the effects on LEC differentia-tion and migration.17 To explore whether lymphatic differentiationor migration were impaired in synectinKD tadpoles, we stainedstage 35/36 embryos by whole-mount in situ hybridization forProx-1 to measure the accumulation of Prox-1& cells in the ventraltrunk as an index of lymphatic lineage emergence, and in the moredorsal trunk area as a measure of LEC migration using previouslyestablished methods.17 Morphometric quantification revealed adose-dependent decrease of both Prox-1& areas in synectinKD

tadpoles (by 40% at the highest morpholino dose; Figure 5K-N).Thus, synectin regulates both lymphatic lineage formation andLEC migration in tadpoles.

Synectin genetically interacts with Vegfc/Vegfr3

We then explored how synectin regulated LEC migration, using theFli1:eGFPy1 zebrafish model. Among the possible interactingpartners of synectin, we focused on neuropilin-2 (Nrp2), as it

Figure 3. Knockdown of synectin disrupts thoracic duct formation in zebrafish.In all panels, the head of the embryo faces left and dorsal is up. Scale bars represent100 "m in panels A-B and 50 "m in panels C-D. DA indicates dorsal aorta; and PCV,posterior cardinal vein. (A-D) Confocal images of GFP& vessels in the trunk of 7-dpfFli1:eGFPy1 zebrafish embryos, showing the formation of a normal lymphatic thoracic duct(TD) in the control embryo (A,C) but not in the synectinKD embryo (B,D). Panels C andD represent close-ups of the boxed areas in panels A and B, between the dorsal aorta(DA) and posterior cardinal vein (PCV). In these latter panels, arrows highlight the TD, whileasterisks denote absence of TD. (E) Quantification of the TD formation defects afterinjection of SynATG1 at 7 dpf. Percentages of embryos displaying complete lack of TD,TD formation over 10%-30% or 30%-90% of its normal length, and a normal TD arerepresented for each treatment group (see also supplemental Table 3). We quantitativelyanalyzed TD formation by scoring its presence in 10 consecutive somite segments (fromsomite 5 to somite 15).

3360 HERMANS et al BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

contains a PDZ-domain33 and regulates, as a coreceptor ofVegfr3,34,35 lymphatic sprouting in mice.36,37 Vegfc stimulates LECmigration through Vegfr3 signaling in mice,1,38,39 zebrafish20,40 andtadpoles.17,31 Because Vegfr3 is the signaling receptor of theVegfr3/Nrp2 complex, and silencing of synectin phenocopies thelymphatic defects induced by silencing of Vegfc or Vegfr3 inzebrafish20,40 and tadpoles,17,31 we first assessed whether synectinand Vegfr3 genetically interacted, by testing whether a combinationof low morpholino doses of synectin (2.5 ng/embryo) and Vegfr3(1.25 ng/embryo), which by themselves induced only a minimaleffect, caused a more significant lymphatic defect (the lowmorpholino doses were denoted by L-KD, as in synectinL-KD, todistinguish them from the dose used in the incomplete silencingstudies). To silence Vegfr3, we used the previously characterizedVegfr3ATG1 morpholino.40 This analysis revealed that, comparedwith each single knockdown, the combined knockdown caused moresevere defects in TD and PL formation in synectinL-KDxVegfr3L-KD

morphants (P ( .05 vs single knockdown; Figure 6A-B).

We also assessed whether synectin genetically interacted withVegfr3 in regulating lymphangiogenic sprouting in the PLC!y10

zebrafish model. A low SynATG1 dose did not reduce the number ofsecondary sprouts (7.4 $ 0.3 in control vs 7.7 $ 0.3 in SynL-KD;N ' 49-62, P ' .40; Figure 6C), while a low dose of Vegfr3ATG1

slightly inhibited secondary sprouting (5.9 $ 0.3 in Vegfr3L-KD;N ' 49-62, P ( .001; Figure 6C). However, the combination ofboth morpholinos impaired secondary sprouting significantly more(4.9 $ 0.2 in synectinL-KDxVegfr3L-KD; N ' 70, P ( .001 vs synect-inL-KD; P ( .05 vs Vegfr3L-KD; Figure 6C). We therefore assessedwhether angiogenic or lymphangiogenic secondary sprouts wereaffected in synectinL-KDxVegfr3L-KD embryos, by counting thenumber of vISVs in 4 dpf Fli1:eGFPy1 embryos (for rationale, seeabove).

In line with reports that Vegfr3 regulates angiogenic secondarysprouting,40 the fraction of vISVs was reduced by a low dose ofVegfr3ATG1 but not of SynATG1 (% vISVs: 26 $ 12% in Vegfr3L-KD

vs 52 $ 7% in synectinL-KD or 54 $ 9% in control; N ' 17-29;

Figure 4. Knockdown of synectin impairs early lymphatic development in zebrafish. In all panels, the head of the embryo faces left and dorsal is up. Scale bars represent50 "m. DA indicates dorsal aorta; ISV, intersomitic vessel; and PCV, posterior cardinal vein. (A-B) Confocal images of 60-hpf control (A) and synectinKD (B) Fli1:eGFPy1

embryos revealing a reduced formation of the parachordal lymphangioblast (PL) string (arrows in A) upon synectin knockdown. Asterisks in panel B denote absence of PL cells.(C) Quantification of the PL cells in control and synectinKD Fli1:eGFPy1 zebrafish embryos at 60 hpf. The percentages of embryos lacking all PL cells and displaying PL stringformation over 10%-30%, 30%-90%, and 100% of its normal length are indicated per treatment group (see also supplemental Table 4). Formation of the PL was scored persomite in 10 consecutive somites between somites 5 and 15. (D-E) Whole-mount in situ hybridization of 48-hpf control (D) and synectinKD (E) embryos for Tie-2, labeling allsecondary sprouts (arrows) emerging from the PCV. In synectinKD embryos the number of secondary sprouts was markedly reduced, by approximately 50%. Panels D# andE# are magnifications of panel D and E, respectively. In synectinKD embryos somites lacking a secondary sprout are indicated with an asterisk. (F-G) Confocal images of 48-hpfFli1:eGFPy1xPLC!1y10 control (F) and synectinKD(G) embryos showing a reduction of secondary sprouts upon synectin knockdown. Small arrows indicate unilateral secondarysprouts; long arrow denotes PL cells. (H) Quantification of the number of unilateral secondary sprouts in a 10-somite region of 48-hpf Fli1:eGFPy1xPLC!1y10 embryosconfirmed a significant reduction of nearly 50% in secondary sprouts budding from the PCV upon synectin knockdown. (I) Quantification of the fraction of venous ISVs,identified by their connection to the PCV upon confocal screening, in a 10-somite region in 4-dpf Fli1:eGFPy1 embryos revealed a normal ratio of venous ISVs upon synectinknockdown. Error bars in panels H-I represent SEM; *P ( .05 by univariate analysis.

SYNECTIN IN LYMPHATIC DEVELOPMENT 3361BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

P ( .001). Notably, coknockdown of Vegfr3 and synectin did notfurther reduce the fraction of vISVs (% vISVs: 25 $ 12% insynectinL-KDxVegfr3L-KD; N ' 27; P ' .968 vs Vegfr3L-KD), indicat-ing that angiogenic sprouting was not under the control of asynectin/Vegfr3 interaction. Thus, the normal number of angio-genic sprouts, in combination with the reduced total number ofsecondary sprouts and impaired formation of PL and TD, insynectinL-KDxVegfr3L-KD embryos indicates that the genetic synectin/Vegfr3 interaction did not control angiogenic but lymphangiogenicsprouting. Because other experiments revealed that Vegf/Vegfr2signaling did not affect lymphatic development in zebrafish(supplemental Note 2), we did not explore an interaction withsynectin.

Synectin genetically interacts with Nrp2a in vivo

We then analyzed whether synectin genetically interacted withNrp2. We first examined whether the zebrafish Nrp2a and Nrp2bcoorthologues regulated lymphatic development, using previouslyvalidated morpholinos for Nrp2a (Nrp2aATG1) and Nrp2b(Nrp2bATG1).41 Consistent with previous findings,42 a maximal doseof 10 ng Nrp2aATG1 caused hind brain edema and arteriovenous(AV) shunts in a small fraction of embryos, while a similar dose ofNrp2bATG1 induced cardiac edema formation and mild DA dysmor-

phogenesis in some morphants (not shown). As reported,42 theseblood vessel phenotypes were confined to the extreme caudalregion of the trunk (not shown); morphant embryos with edema ormorphologic anomalies were excluded. The residual Nrp2aKD andNrp2bKD embryos developed normally by 6 dpf with no or onlyminimal blood vessel or circulatory defects in the mid-trunk (wherewe scored lymphatic development; ie, more rostral than the caudalregion in which AV shunts/DA defects were present; Figure 6E,G).While lymphatic development was normal in Nrp2bKD embryos(supplemental Figure 7A), Nrp2aKD dose-dependently caused PLand TD defects (Figure 6D-K).

We then analyzed whether synectin genetically interacted withNrp2a. A low dose of Nrp2aATG1 (Nrp2aL-KD) or SynATG1 (SynL-KD)minimally affected TD formation (Figure 6L). In contrast, knock-down of both genes by combined injection of a low dose of bothmorpholinos severely impaired lymphatic development, as the TDwas absent in 30% of synectinL-KDxNrp2aL-KD morphants (P ( .001vs single knockdown; Figure 6L). Similar findings were obtainedwhen analyzing the PL (not shown).

Notably, however, Nrp2a knockdown did not affect secondarysprout formation in PLC!1y10xFli1:eGFPy1 embryos (supplemen-tal Figure 7B). Coknockdown analysis did also not reveal a geneticinteraction between synectin and Nrp2a in secondary sprout

Figure 5. Synectin knockdown impairs lymphatics in tadpoles. In all panels, the head of the tadpoles faces left and dorsal is up. Scale bars represent 100 "m in panelsA-B and K-L; 50 "m in panels C-F; 150 "m in panels G-H; and 400 "m in panels I-J. DCLV indicates dorsal caudal lymph vessel; DLAV, dorsal anastomosing longitudinalvessel; PCV, posterior cardinal vein; and VCLV, ventral caudal lymph vessel. (A-F) Fluorescent analysis of blood and lymph vessels in stage 47 transgenic Flk1:eGFP tadpoles.Lymph vessels are labeled with TRITC-dextran that was intracardially injected at stage 45 and taken up by LECs after extravasation from the blood vessels. Hence, GFP&

BECs can be easily distinguished from doubly labeled (yellow) LECs. Areas in the trunk corresponding to the boxed areas in panels A or B are shown in panels C-F. In controltadpoles, both the DCLV (C) and the VCLV (E) developed normally. Injection of 30 ng of synectin morpholino resulted in a hypoplastic, disorganized, and discontinuous DCLV(D), while the VCLV (F) had a grossly normal appearance. (G-H) Lymphangiography of stage 45 Flk1:eGFP tadpoles, revealing dysfunction of the VCLV after synectinknockdown (H) in contrast to normal controls (G). Whereas in control injected embryos, the dye is drained normally in a rostral direction (white arrows in G), no dye uptake wasobserved in the VCLV in synectinKD tadpoles. White asterisks indicate injection site of the fluorescent dye. (I-J) Bright-field pictures of live embryos at stage 45 (5 dpf), showinglymphedema (arrows) in a synectinKD tadpole (J) compared with a control tadpole (I). (K-L) Whole mount Prox-1 in situ hybridization of stage35/36 tadpoles, revealing the presence of fewer Prox-1& LECs in the area ventrally to the dorsal margin of the endoderm (reflecting lymphatic lineage emergence) and in thearea dorsally to this margin (reflecting LEC budding/migration) in the posterior trunk of synectinKD-morphant tadpoles (L) compared control (K) tadpoles. Dotted line: dorsalmargin of the endoderm. (M-N) Morphometric measurement revealed a dose-dependent decrease of the Prox-1& area both ventrally (M) and dorsally (N) of the dorsal marginof the endoderm in the trunk of Prox-1–stained synectinKD compared with control tadpoles at stage 35/36, indicating that lymphatic lineage development and migration isaffected upon synectin knockdown. Values expressed relative to control. *P ( .001 by multivariate analysis.

3362 HERMANS et al BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

formation (supplemental Figure 7E; see below for discussion).Thus, our findings are consistent with a model, whereby synectingenetically interacts with Vegfr3 and Nrp2a during lymphaticdevelopment, but only beyond the initial stage of lymphangiogenicsprouting.

Synectin mediates VEGFC/VEGFR3-driven LECsprouting in vitro

Because synectin regulates LEC migration in tadpoles and geneti-cally interacts with Vegfr3, known to stimulate LEC migra-tion,1,31,39 we examined whether synectin knockdown inhibitedsprouting of cultured HDLECs in response to vascular endothelialgrowth factor C (VEGFC). We silenced synectin expression inLECs using simultaneously 2 nonoverlapping synectin–specificsiRNAs. qRT-PCR analysis showed that synectin mRNA levels inHDLECs were reduced by 80% (not shown). HDLECs treated with

control or synectin-specific siRNA were aggregated as spheroidsand lymphatic sprouts were visualized by CD31 immunostaining.VEGFC stimulated lymphatic sprouting in control spheroids(N ' 17; Figure 7A,C,E), but synectinKD HDLECs formed substan-tially fewer sprouts in response to VEGFC (N ' 26, P ( .001;Figure 7B,D-E).

Initial evidence for a role of synectin lymphatic differentiationin vitro

The reduced number of Prox-1& cells in the ventral region insynectinKD tadpoles suggested a role for synectin in lymphaticdifferentiation. In mice, Sox18 acts upstream of Prox-1 during thisprocess and lymphatic reprogramming is mimicked in vitro byoverexpression of Sox18 in venous ECs.3,6 We therefore overex-pressed Sox18 in murine H5V BECs, transfected them with controlor synectin-specific siRNA at day 0, and analyzed lymphatic

Figure 6. Synectin modulates Vegfr3 signaling in lymphatic development in vivo. In all panels, the head of the embryo faces left and dorsal is up. Scale bars represent50 "m in panels D-E and 100 "m in panels F-I. DA indicates dorsal aorta; ISV, intersomitic vessel; and PCV, posterior cardinal vein. (A) Quantification of parachordallymphangioblast (PL) cell defects after injection of SynATG1 (2.5 ng; synL-KD; N ' 79), Vegfr3ATG1 (1.25 ng; Vegfr3L-KD; N ' 62) or both (N ' 50) at 60 hpf. Percentages ofembryos displaying complete lack of PL cells, PL string over 10%-30% or 30%-90% of its normal length and a normal PL string are shown for each group. Compared with singleknockdown groups, the PL formation was more severely impaired in double morphants (P ( .001). (B) Quantification of TD defects after injection of SynATG1 (2.5 ng; synL-KD;N ' 101), Vegfr3ATG1 (1.25 ng; Vegfr3L-KD; N ' 107) or both (N ' 86) at 7 dpf. Percentages of embryos displaying complete lack of TD, TD formation over 10%-30% or30%-90% of its normal length and a normal TD are represented for each treatment group. Compared with single knockdown groups, the TD was more severely impaired indouble morphants (P ( .001 vs synectinL-KD; P ( .05 vs Vegfr3L-KD). (C) Quantification of the number of unilateral secondary sprouts in a 10-somite region of 48-hpfFli1:eGFPy1xPLC!1y10 embryos revealed that coknockdown of synectin and Vegfr3 significantly aggravated the secondary sprouting defects compared with single knockdownof either gene when using suboptimal doses of SynATG1 (2.5 ng; synL-KD) and Vegfr3ATG1 (2.5 ng; Vegfr3L-KD); (N ' 45, 37, 48, and 63 for control, Vegfr3L-KD, synectinL-KD, andcoknockdown, respectively; *P ( .05; **P ( .01; ***P ( .001). (D-E) Confocal images of 60-hpf control (D) and Nrp2aKD (E) Fli1:eGFPy1 embryos revealing impaired formationof the PL string (arrows) upon Nrp2a knockdown. Asterisks denote absence of PL cells. (F-I) Confocal images of GFP& vessels in the trunk of 7-dpf Fli1:eGFPy1 zebrafishembryos, showing formation of a normal TD in a control embryo (F,H) but not in a Nrp2aKD embryo (G,I). Panels H and I represent close-up magnifications of the boxed areas inpanels F and G; arrows denote TD, asterisks denote absence of TD. (J) Quantification of PL cells in control and Nrp2aKD Fli1:eGFPy1 zebrafish embryos at 60 hpf. Thepercentages of embryos lacking PL cells and displaying PL string over 10%-30%, 30%-90%, and 100% of its normal length are indicated per treatment group. Formation of thePL string was scored per somite in 10 consecutive somites between somite 5 and 15 (N ' 106, 152, and 60 for 0, 5, and 10 ng of Nrp2aATG1, respectively). (K) Quantification ofTD in control and Nrp2aKD Fli1:eGFPy1 zebrafish embryos at 7 dpf. The percentages of embryos lacking TD and displaying TD formation over 10%-30%, 30%-90%, and 100%of its normal length are indicated per treatment group. Formation of the TD was scored per somite in 10 consecutive somites between somite 5 and 15 (N ' 99, 164, and 56 for0, 5, and 10 ng of Nrp2aATG1, respectively). (L) Quantification of TD formation after injection of Nrp2aATG1 (5 ng; Nrp2aL-KD; N ' 74), SynATG1 (2.5 ng; synL-KD; N ' 98) or both(N ' 41) revealed that coknockdown impaired lymphatic development more severely than single synectinL-KD (P ( .001) or Nrp2aL-KD (P ( .001).

SYNECTIN IN LYMPHATIC DEVELOPMENT 3363BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

marker expression on day 7, as reported.6 The siRNA transfectionprocedure was repeated on day 3 to maintain prolonged synectinsilencing until 7 days, when synectin mRNA levels were reducedby 60% (not shown).

As reported,6 Prox-1 transcript levels were higher in BECstransfected with Sox18-lentivirus (BECssox18) than control GFP-virus (BECsGFP), when treated with control siRNA; expression ofthe blood vascular marker Nrp1 was not altered (Figure 7F). Uponsynectin knockdown, induction of Prox-1, but not of Nrp1,expression in BECsSox18 was inhibited (Figure 7F). Knockdown ofsynectin did not alter gene expression in BECsGFP (relative mRNAlevels vs control: 1.07 $ 0.03 for Sox18; 0.84 $ 0.12 for Prox1;and 0.97 $ 0.09 for Nrp1; N ' 3; P * .05). Thus, silencing ofsynectin impaired Sox18-driven lymphatic reprogramming in vitro.

Discussion

The present study provides genetic evidence for a role of synectinin early lymphatic development. In zebrafish and frog embryos,knockdown of synectin dose-dependently impaired and, in themost severely affected embryos, even aborted lymphatic formation.Several lines of evidence suggest that TD defects in synectinKD

zebrafish were due to defects in lymphangiogenic sprout formationfrom the venous system. First, lymphangiogenic sprouts that giverise to PL cells (ie, precursors of LECs forming the TD) wereunderdeveloped or, in the most severely affected morphants, didnot form at all in synectinKD embryos. In the ccbe1, Vegfc, or Vegfr3mutants, lack of formation of these lymphangiogenic sprouts also

aborts further lymphatic development.20,40 Second, the fraction ofsynectinKD embryos without TD closely correlated with the fractionof morphants without lymphangiogenic sprouts and PL cells. Third,the missing secondary sprouts in synectinKD embryos were lym-phangiogenic and not angiogenic, since these embryos developedat the expected ratio vISVs, the formation of which requires normalangiogenic secondary sprouting.

So far, silencing of only a few other genes (Ccbe1, Vegfc, orVegfr3) is known to abort lymphangiogenic sprouting, the forma-tion of the PL and TD.20,40 However, in these morphant embryos,branching of angiogenic sprouts from the PCV was also impaired.20

By contrast, in Dll4/Notch hypomorphants, the fraction of angio-genic sprouts is increased at the expense of lymphangiogenicsprouts.11 It is thus noteworthy that synectin silencing selectivelyimpaired lymphangiogenic without affecting angiogenic secondarysprouting. The identification of synectin as the first selectiveregulator of lymphangiogenic sprouting suggests that formation oflymphangiogenic versus angiogenic sprouts is under distinct ge-netic control.

Synectin silencing inhibited migration of cultured LECs inresponse to VEGFC and reduced the dorsal Prox-1& area, aparameter of LEC migration, in tadpoles. Similar LEC migrationdefects contribute to lymphatic impairment in tadpoles or micelacking VEGFC or VEGFR3.1,17,31,38,39 Moreover, silencing ofsynectin in zebrafish aggravated the lymphatic defects, induced bysilencing of Vegfr3 as well as of Nrp2a, the coreceptor ofVegfr3,34,35 that contains a PDZ-binding domain,33 through whichsynectin interacts with its partners.27 These data thus illustrate agenetic interaction of synectin with Vegfr3 as well as with Nrp2a.

Figure 7. Synectin regulates Vegfc-driven lymphatic sprouting and Sox18-mediated lymphatic reprogramming in vitro. (A-D) CD31 immunostaining of untreated(A-B) or VEGFC-stimulated (C-D; 200 ng/mL) LEC spheroids in fibrin gels, revealing reduced VEGFC sprouting response by LECs transfected with synectin-specific(B,D; synectinKD) compared with control siRNA (A,C). Scale bar represents 100 "m. (E) Quantification of the sprout number per LEC spheroid, revealing that synectinknockdown impaired lymphatic sprouting in response to VEGFC (N ' 17 for control; N ' 26 for synectinKD), while baseline sprouting of control (N ' 17) and synectinKD LECs(N ' 35) was comparable. *P ( .001 by univariate analysis. (F) RT-PCR of control and synectinKD H5V blood vascular endothelial cells (BECs), transfected with Sox18-virus(BECsSox18) or GFP-virus (BECsGFP), revealing that knockdown of synectin impaired up-regulation of Prox-1 in BECsSox18 without affecting Nrp1 expression. Results arerepresented as fold change vs control BECsGFP. Dashed line indicates baseline expression levels in BECsGFP. N ' 3; *P ( .01 by Student t test. Error bars represent SEM.

3364 HERMANS et al BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

While Vegfr3/synectin participated already in lymphangiogenicsprouting, Nrp2a/synectin only regulated subsequent formation ofthe PL and TD. In mice, Nrp2 is also dispensable for initiallymphangiogenic budding from veins, but required for subsequentlymphatic development.36 Together, these genetic studies suggest amodel whereby synectin modulates Vegfr3 signaling in lymphaticdevelopment in a Nrp2a-independent manner (during lymphangio-genic sprouting) or Nrp2a-dependent manner (PL and TDformation).

How synectin regulates Vegfr3 signaling is an intriguingquestion. During arterial development, synectin promotes intracel-lular Vegfr2 trafficking to early endosomes, required for down-stream signaling, via Nrp1-dependent and -independent mecha-nisms.13,43 By analogy with Vegfr2/Nrp1, it is tempting to speculatethat synectin might perhaps also participate in Vegfr3 endocytosis,at least partially, via an interaction with Nrp2. Supporting thismodel, Vegfr3 not only colocalizes with Nrp2 during endocytosisupon stimulation by Vegfc and Vegfd,35 but internalization is alsorequired for proper Vegfr3 signaling.44 Because synectin modulatesVegfr3-driven lymphangiogenic but not angiogenic sprouting, it istempting to speculate that synectin-mediated endocytosis regulatesdistinct Vegfr3 signaling pathways. Elucidation of the underlyingmolecular details will require future study.

Lymphatic defects were unlikely to be secondary to angiogenicalterations, as the PCV formed normally and expressed propervenous markers in synectinKD embryos, consistent with findingsthat synectin is dispensable for venous morphogenesis.12 In addi-tion, the DA and primary aISVs formed normally and expressedarterial markers upon incomplete silencing of synectin. In fact,alterations in arterial identity or morphogenesis are unlikely toperturb lymphangiogenic sprouting or cause the type of lymphaticdefects, observed upon synectin knockdown, since secondarysprouts and PL cells formed normally in PLC!1y10 mutants, whicheven lack primary ISVs and exhibit an underdeveloped andimproperly differentiated DA.16,25 In addition, we and others22

observed that lymphatic development is nearly normal uponsilencing of Vegfaa- or Kdr-like, 2 well-established regulators ofarterial morphogenesis and specification.45,46

A limitation of our study is that Prox-1 levels in the PCV at thetime of lymphatic differentiation in zebrafish are below thethreshold of detection (unpublished findings, P.C. and S.S.M.;personal communication, H. Gerhardt [London Research Institute,Cancer Research UK, London, United Kingdom] and F. Cotelli[University of Milan, Milan, Italy]), thereby precluding us fromanalyzing whether synectin silencing impairs LEC differentiationin zebrafish. We therefore used alternative models in vivo and invitro. Indeed, our genetic analysis in tadpoles shows that silencingof synectin caused an underdevelopment of Prox-1& cells at sites

previously implicated in differentiation of LECs from venousECs.17 In addition, silencing of synectin reduced lymphatic repro-gramming of venous ECs in vitro upon overexpression of Sox18,an established regulator of lymphatic differentiation in mice.6 Therole of Sox18 in lymphatic development in tadpoles and zebrafishhas not yet been documented but, interestingly, silencing of Sox18(in combination with Sox7) in zebrafish impaired venous ECs fromwhich LECs differentiate.47-49 Notably, because PDZ-dependentbinding has been reported for some of the SoxA family members,50

a possible interaction of synectin with Sox18 is an excitingpossibility that remains to be further investigated.

In conclusion, synectin regulates lymphangiogenesis in zebra-fish and tadpoles through genetic interactions with Vegfr3 andNrp2a.

Acknowledgments

The authors thank K. Brepoels, A. Carton, M. De Mol, E. Janssens,S. Louwette, A. Manderveld, M. Peeters, J. Souffreau, B. Tembuyser,A. Van den Eynde, A. Van Nuffelen, B. Vanwetswinkel,S. Verstraeten, S. Vinckier, and S. Wyns for technical assistance.

K.H., W.V., K.A., I.G., and F.D.S. are sponsored by a PhD grantof the Institute for the promotion of Innovation through Scienceand Technology (IWT) in Flanders (IWT-Vlaanderen), Belgium.E.G. and S.S.-M. are supported by the Royal Netherlands Academyof Arts and Sciences (KNAW). This work is supported bylong-term structural Methusalem Funding by the Flemish Govern-ment to P.C., Research Foundation Flanders (FWO) ResearchProject Funding by the Flemish Government to P.C., Interuniver-sity Attraction Poles–Belgian Science Policy (IUAP P6/20) toM.D., and European Commission Lymphangiogenomics Consor-tium Funding (LSHG-CT-2004-503573) to P.C.

Authorship

Contribution: K.H., F.C., A.N., M.A., M.S., M.D., S.S.-M., E.D.,K.A., and P.C. designed research; K.H., F.C., W.V., W.Z., I.G., F.O.,A.N., F.D.S., E.G., K.A., and B.R. performed the experiments andanalyzed the data; D.K. provided vital reagents; and K.H., F.C.,M.D., and P.C. wrote the paper.

Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.

Correspondence: Peter Carmeliet, Vesalius Research Center,Vlaams Iustituut voor Biotechnologie, Katholieke UniversiteitLeuven, Campus Gasthuisberg, Herestraat 49, B-3000, Belgium;e-mail: Peter.Carmeliet@vib-kuleuven.be.

References1. Tammela T, Alitalo K. Lymphangiogenesis: Mo-

lecular mechanisms and future promise. Cell.2010;140(4):460-476.

2. Wigle JT, Oliver G. Prox1 function is required forthe development of the murine lymphatic system.Cell. 1999;98(6):769-778.

3. Wigle JT, Harvey N, Detmar M, et al. An essentialrole for Prox1 in the induction of the lymphaticendothelial cell phenotype. EMBO J. 2002;21(7):1505-1513.

4. Hong YK, Harvey N, Noh YH, et al. Prox1 is amaster control gene in the program specifyinglymphatic endothelial cell fate. Dev Dyn. 2002;225(3):351-357.

5. Petrova TV, Makinen T, Makela TP, et al. Lym-phatic endothelial reprogramming of vascular en-dothelial cells by the Prox-1 homeobox transcrip-tion factor. EMBO J. 2002;21(17):4593-4599.

6. Francois M, Caprini A, Hosking B, et al. Sox18induces development of the lymphatic vascula-ture in mice. Nature. 2008;456(7222):643-647.

7. Leslie Pedrioli DM, Karpanen T, Dabouras V, etal. miR-31 functions as a negative regulator oflymphatic vascular lineage-specific differentiationin vitro and vascular development in vivo. MolCell Biol. 2010;30(14):3620-3634.

8. Petrova TV, Karpanen T, Norrmen C, et al. Defec-tive valves and abnormal mural cell recruitment

underlie lymphatic vascular failure in lymphed-ema distichiasis. Nat Med. 2004;10(9):974-981.

9. Seo S, Fujita H, Nakano A, Kang M, Duarte A,Kume T. The forkhead transcription factors,Foxc1 and Foxc2, are required for arterial specifi-cation and lymphatic sprouting during vasculardevelopment. Dev Biol. 2006;294(2):458-470.

10. Makinen T, Adams RH, Bailey J, et al. PDZ inter-action site in ephrinB2 is required for the remod-eling of lymphatic vasculature. Genes Dev. 2005;19(3):397-410.

11. Geudens I, Herpers R, Hermans K, et al. Role ofDll4/Notch in the Formation and Wiring of theLymphatic Network in Zebra Fish. ArteriosclerThromb Vasc Biol. 2010;30(9):1695-1702.

SYNECTIN IN LYMPHATIC DEVELOPMENT 3365BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom

12. Chittenden TW, Claes F, Lanahan AA, et al. Se-lective regulation of arterial branching morpho-genesis by synectin. Dev Cell. 2006;10(6):783-795.

13. Lanahan AA, Hermans K, Claes F, et al. VEGFreceptor 2 endocytic trafficking regulates arterialmorphogenesis. Dev Cell. 2010;18(5):713-724.

14. Lawson ND, Weinstein BM. In vivo imaging ofembryonic vascular development using trans-genic zebrafish. Dev Biol. 2002;248(2):307-318.

15. Traver D, Paw BH, Poss KD, Penberthy WT, LinS, Zon LI. Transplantation and in vivo imaging ofmultilineage engraftment in zebrafish bloodlessmutants. Nat Immunol. 2003;4(12):1238-1246.

16. Lawson ND, Mugford JW, Diamond BA,Weinstein BM. phospholipase C gamma-1 is re-quired downstream of vascular endothelialgrowth factor during arterial development. GenesDev. 2003;17(11):1346-1351.

17. Ny A, Koch M, Schneider M, et al. A geneticXenopus laevis tadpole model to study lym-phangiogenesis. Nat Med. 2005;11(9):998-1004.

18. Zhong TP, Rosenberg M, Mohideen MA,Weinstein B, Fishman MC. gridlock, an HLH generequired for assembly of the aorta in zebrafish.Science. 2000;287(5459):1820-1824.

19. Cooke J, Moens C, Roth L, et al. Eph signallingfunctions downstream of Val to regulate cell sort-ing and boundary formation in the caudal hind-brain. Development. 2001;128(4):571-580.

20. Hogan BM, Bos FL, Bussmann J, et al. ccbe1 isrequired for embryonic lymphangiogenesis andvenous sprouting. Nat Genet. 2009;41(4):396-398.

21. Nisato RE, Harrison JA, Buser R, et al. Genera-tion and characterization of telomerase-trans-fected human lymphatic endothelial cells with anextended life span. Am J Pathol. 2004;165(1):11-24.

22. Bussmann J, Bos FL, Urasaki A, Kawakami K,Duckers HJ, Schulte-Merker S. Arteries provideessential guidance cues for lymphatic endothelialcells in the zebrafish trunk. Development. 2010;Inpress.

23. Isogai S, Lawson ND, Torrealday S, Horiguchi M,Weinstein BM. Angiogenic network formation inthe developing vertebrate trunk. Development.2003;130(21):5281-5290.

24. Kuchler AM, Gjini E, Peterson-Maduro J,Cancilla B, Wolburg H, Schulte-Merker S. Devel-opment of the zebrafish lymphatic system re-quires VEGFC signaling. Curr Biol. 2006;16(12):1244-1248.

25. Yaniv K, Isogai S, Castranova D, Dye L, Hitomi J,Weinstein BM. Live imaging of lymphatic devel-opment in the zebrafish. Nat Med. 2006;12(6):711-716.

26. Wang L, Mukhopadhyay D, Xu X. C terminus ofRGS-GAIP-interacting protein conveys neuro-pilin-1-mediated signaling during angiogenesis.FASEB J. 2006;20(9):1513-1515.

27. Katoh M. GIPC gene family (Review). Int J MolMed. 2002;9(6):585-589.

28. Tan C, Deardorff MA, Saint-Jeannet JP, Yang J,Arzoumanian A, Klein PS. Kermit, a frizzled inter-acting protein, regulates frizzled 3 signaling inneural crest development. Development. 2001;128(19):3665-3674.

29. Zhang Y, Chittenden T, Simons M. Characteriza-tion of synectin expression and promoter activity.Gene. 2004;342(1):29-34.

30. Podgrabinska S, Braun P, Velasco P, Kloos B,Pepper MS, Skobe M. Molecular characterizationof lymphatic endothelial cells. Proc Natl Acad SciU S A. 2002;99(25):16069-16074.

31. Ny A, Koch M, Vandevelde W, et al. Role ofVEGF-D and VEGFR-3 in developmental lym-phangiogenesis, a chemicogenetic study inXenopus tadpoles. Blood. 2008;112(5):1740-1749.

32. Norrmen C, Vandevelde W, Ny A, et al. Liprin(beta)1 is highly expressed in lymphatic vascula-ture and is important for lymphatic vessel integ-rity. Blood. 2010;115(4):906-909.

33. Nakamura F, Goshima Y. Structural and func-tional relation of neuropilins. Adv Exp Med Biol.2002;515:55-69.

34. Favier B, Alam A, Barron P, et al. Neuropilin-2 in-teracts with VEGFR-2 and VEGFR-3 and pro-motes human endothelial cell survival and migra-tion. Blood. 2006;108(4):1243-1250.

35. Karpanen T, Heckman CA, Keskitalo S, et al.Functional interaction of VEGF-C and VEGF-Dwith neuropilin receptors. FASEB J. 2006;20(9):1462-1472.

36. Yuan L, Moyon D, Pardanaud L, et al. Abnormallymphatic vessel development in neuropilin 2 mu-tant mice. Development. 2002;129(20):4797-4806.

37. Xu Y, Yuan L, Mak J, et al. Neuropilin-2 mediatesVEGF-C-induced lymphatic sprouting togetherwith VEGFR3. J Cell Biol. 2010;188(1):115-130.

38. Karkkainen MJ, Haiko P, Sainio K, et al. Vascularendothelial growth factor C is required for sprout-

ing of the first lymphatic vessels from embryonicveins. Nat Immunol. 2004;5(1):74-80.

39. Makinen T, Jussila L, Veikkola T, et al. Inhibitionof lymphangiogenesis with resulting lymphedemain transgenic mice expressing soluble VEGF re-ceptor-3. Nat Med. 2001;7(2):199-205.

40. Hogan BM, Herpers R, Witte M, et al. Vegfc/Flt4signalling is suppressed by Dll4 in developingzebrafish intersegmental arteries. Development.2009;136(23):4001-4009.

41. Yu HH, Moens CB. Semaphorin signaling guidescranial neural crest cell migration in zebrafish.Dev Biol. 2005;280(2):373-385.

42. Martyn U, Schulte-Merker S. Zebrafish neuro-pilins are differentially expressed and interactwith vascular endothelial growth factor during em-bryonic vascular development. Dev Dyn. 2004;231(1):33-42.

43. Salikhova A, Wang L, Lanahan AA, et al. Vascularendothelial growth factor and semaphorin induceneuropilin-1 endocytosis via separate pathways.Circ Res. 2008;103(6):71-79.

44. Wang Y, Nakayama M, Pitulescu ME, et al.Ephrin-B2 controls VEGF-induced angiogenesisand lymphangiogenesis. Nature.2010;465(7297):483-486.

45. Covassin LD, Villefranc JA, Kacergis MC,Weinstein BM, Lawson ND. Distinct genetic inter-actions between multiple Vegf receptors are re-quired for development of different blood vesseltypes in zebrafish. Proc Natl Acad Sci U S A.2006;103(17):6554-6559.

46. Lawson ND, Vogel AM, Weinstein BM. sonichedgehog and vascular endothelial growth factoract upstream of the Notch pathway during arterialendothelial differentiation. Dev Cell. 2002;3(1):127-136.

47. Cermenati S, Moleri S, Cimbro S, et al. Sox18and Sox7 play redundant roles in vascular devel-opment. Blood. 2008;111(5):2657-2666.

48. Herpers R, van de Kamp E, Duckers HJ, Schulte-Merker S. Redundant roles for sox7 and sox18 inarteriovenous specification in zebrafish. Circ Res.2008;102(1):12-15.

49. Pendeville H, Winandy M, Manfroid I, et al.Zebrafish Sox7 and Sox18 function together tocontrol arterial-venous identity. Dev Biol. 2008;317(2):405-416.

50. Bernard P, Harley VR. Acquisition of SOX tran-scription factor specificity through protein-proteininteraction, modulation of Wnt signalling andpost-translational modification. Int J Biochem CellBiol. 2010;42(3):400-410.

3366 HERMANS et al BLOOD, 28 OCTOBER 2010 ! VOLUME 116, NUMBER 17 For personal use only.2012. at TERKKO NATIONAL LIBRARY OF HEALTH SCIENCES on April 4,bloodjournal.hematologylibrary.orgFrom