Lymphatic vascular specification and its modulation during embryonic development

YMVRE-03458; No. of pages: 7; 4C: Microvascular Research xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Microvascular Research journal homepage: www.elsevier.com/locate/ymvre

Lymphatic vascular specification and its modulation during embryonic development Cathy Pichol-Thievend, Benjamin M. Hogan ⁎,1, Mathias Francois ⁎,1 Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland 4072, Australia

a r t i c l e

i n f o

Article history: Accepted 29 July 2014 Available online xxxx Keywords: Lymphatic endothelial cell Vascular Specification Transcription factors Signalling pathways Zebrafish Mouse Lymphoedema

a b s t r a c t Despite its essential roles in development and disease, the lymphatic vascular system has been less studied than the blood vascular network. In recent years, significant advances have been made in understanding the mechanisms that regulate lymphatic vessel formation, both during development and in pathological conditions. Remarkably, lymphatic endothelial cells are specified as a subpopulation of pre-existing venous endothelial cells. Here, we summarize the current knowledge of the transcription factor pathways responsible for lymphatic specification and we also focus on the factors that promote or restrict this event. © 2014 Published by Elsevier Inc.

Introduction The specification and sprouting of the earliest lymphatic vessels The lymphatic vasculature and its origins The lymphatic vasculature is crucial for interstitial fluid drainage, immune surveillance and lipid absorption. The lymphatic vessel network is composed of a hierarchy of blind-ended lymphatic capillaries, precollecting lymphatic vessels and collecting lymphatic vessels. Capillaries absorb fluid, proteins and extravasated macromolecules. The resulting intravascular lymph is then delivered, through the collecting vessels, to the lymph nodes (where 40% of fluid absorption occurs) or further via the subclavian veins and returned to the blood circulation (Földi, 2004). The disruption of the establishment or the maintenance of this homeostatic process leads to a variety of pathologies that prominently include lymphoedema and inflammatory disorders (Alitalo, 2011). Although the lymphatic vasculature was identified centuries ago, its formation during embryonic development has only been characterized in depth from the beginning of the 20th century. Two main hypotheses have been proposed to explain the origin of the lymphatic vasculature in the embryo. Using dye-injection experiments in pig embryos, Florence Sabin proposed that endothelial cells bud from veins to form lymph ⁎ Corresponding authors at: Division of Molecular Genetics and Development, Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland 4072, Australia. E-mail addresses: [email protected] (B.M. Hogan), [email protected] (M. Francois). 1 Equal contributors.

sacs (LS), which in turn sprout in a centrifugal pattern to form the entire lymphatic vascular network (Sabin, 1902). In contrast, Huntington proposed that the lymphatic vessels arise from mesenchymal cells, independently of the vein (Huntington, 1908). Since these initial observations, experiments performed largely using the mouse model system but also more recently in zebrafish embryos, have demonstrated that most lymphatic endothelial cells (LECs) differentiate from the embryonic veins, and most prominently from the cardinal vein (CV) (Wigle and Oliver, 1999; Srinivasan et al., 2007; Yaniv et al., 2006). Cellular lymphangiogenesis in the embryo For some time, it has been known that the initiation of expression of the transcription factor Prospero-related homeobox domain 1 (PROX1) marks the initial steps of LEC emergence from approximately 9.5 days postcoitum (dpc) of mouse development. PROX1 is first expressed in a subpopulation of venous endothelial cells in a polarized manner in the dorsal wall of the CV (Wigle and Oliver, 1999). By 10.5 dpc, these cells can be observed delaminating from the wall of the CV. These initial LECs will form the primary lymphatic structures that include the LS. Recent work using high-resolution imaging has described in more detail the cellular events that drive lymphatic vessel formation during mouse development (François et al., 2012; Hagerling et al., 2013; Yang et al., 2012). Using confocal and electron microscopy, Yang et al. has shown that LEC progenitors bud from the CV as an interconnected group of cells and that they can also arise from the intersomitic veins. Importantly, by examining adherens junctions during this process, it was found that the integrity of the cardinal vein is maintained by the presence of junctions between venous endothelial cells and LEC

http://dx.doi.org/10.1016/j.mvr.2014.07.011 0026-2862/© 2014 Published by Elsevier Inc.

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011

2

C. Pichol-Thievend et al. / Microvascular Research xxx (2014) xxx–xxx

progenitors (Yang et al., 2012). A similar observation was made by Francois et al., who also observed LEC progenitors that delaminate as streams of cells from the CV. This second study additionally suggested that LEC progenitors are localized in discrete clusters arranged along the anteroposterior axis of the CV. These pre-lymphatic clusters (PLCs) are proposed to undergo progressive ballooning to generate primitive LS that initially contain blood cells (François et al., 2012). Finally a third study used ultramicroscopy to visualize early lymphangiogenesis in the embryo and suggested a number of significant changes to existing models. Hagerling et al. proposed that two populations of LECs arising from the CV contribute to the establishment of the first lymphatic structures. LECs emerging from the CV are seen as loosely attached networks consistent with the studies of Yang et al. and Francois et al., but they are observed to accumulate and condense close to the first lateral branch of the intersegmental vessels (ISV) to form a structure termed the peripheral longitudinal lymphatic vessel (PLLV), which subsequently gives rise to superficial lymphatic vessels that will probably contribute dermal lymphatics. At the same time, LECs located in closer proximity to the CV aggregate to form the primordial thoracic duct (pTD), which was suggested to be the structure previously defined as the LS (Hagerling et al., 2013). In addition to these observations, perhaps the most unexpected finding was that there is an additional source of LECs from a structure that is separate from the CV. This structure, dubbed the superficial venous plexus (sVP), appears to be a source of PROX1-positive LECs that is endothelial and likely venous in nature (Hagerling et al., 2013). This may suggest that multiple venous endothelial beds can give rise to LECs, a significant change to our way of thinking about the lymphatic vasculature as exclusively of CV origin. It is well worthy of note that there remain differences between all of the models recently described. This is telling and suggests that our ability to visualize early lymphangiogenesis is limited in the mouse model. A model synthesizing the current view of how lymphangiogenesis occurs in the embryo is presented in Fig. 1. In the zebrafish embryo, lymphangiogenesis can be directly visualized by live-imaging (Yaniv et al., 2006). This advantage has led to the appreciation that lymphangiogenesis occurs concurrently with the angiogenic sprouting of the venous ISVs (vISVs) from the posterior cardinal vein (PCV). Half of the venous sprouts connect to arterial intersegmental vessels (aISV) and the other half migrate more dorsally to the horizontal myoseptum, producing a pool of cells dubbed as parachordal lymphangioblasts (PLs) based on their anatomical location and developmental potential (Hogan et al., 2009a; Yaniv et al., 2006). Later in

development, after an initial period within the myoseptum, PLs migrate dorsally or ventrally alongside the arterial ISVs (Bussmann et al., 2010; Cha et al., 2012), to form the intersegmental lymphatic vessels (ISLVs), dorsal longitudinal lymphatic vessel (DLLV) and the thoracic duct (TD) (reviewed in (Koltowska et al., 2013; van Impel and Schulte-Merker, 2014; van Impel et al., 2014). Recent work has characterized three other lymphatic networks in zebrafish, the facial lymphatics (FL), the lateral lymphatics and the intestinal lymphatics (Okuda et al., 2012). Interestingly, LEC-progenitors that contribute to the FL originate from a number of different blood vessel origins, together with the mouse studies of the sVP suggesting that there may be multiple ways to form LECs in the embryo. While the optical advantages of the zebrafish has led to many of the observations above, it is worth noting that the weakness here is the absence of well-defined markers of cell state in the zebrafish, it remains difficult to define specified and differentiated cell populations in this model (van Impel et al., 2014), in contrast to the mouse. Transcriptional control of lymphatic endothelial cell specification Prospero-related homeobox domain 1 (Prox1) Prox1 is considered the master regulator of both specification and maintenance of the lymphatic endothelial cell phenotype. The first indication that lymphangiogenesis has begun is the specific expression of PROX1 in a restricted subpopulation of endothelial cells in the CV (Wigle et al., 2002). Prox1−/− embryos are completely devoid of lymphatic vessels and die between 14 and 15 dpc (Johnson et al., 2008; Wigle and Oliver, 1999). PROX1 activity is essential for the emergence of the LEC progenitors at the level of the CV (Yang et al., 2012). In addition, PROX1 is sufficient when overexpressed in cultured blood vascular endothelial cells, to induce the expression of lymphatic specific markers (podoplanin and VEGFR3) and is capable of suppressing 40% of blood endothelial cell-specific genes (Hong et al., 2002; Petrova et al., 2002). Furthermore, ectopic expression of Prox1 under the control of Tie1 promoter in vivo induces the expression of LEC markers in some blood vessels (Kim et al., 2010). During zebrafish development, Prox1 also labels the lymphatic vasculature (Dunworth et al., 2014; van Impel et al., 2014; Yaniv et al., 2006). However, it has been recently shown that, in contrast to mammals, Prox1 is not necessary for the development of all lymphatic vessels but just a percentage of lymphatic vessels. Prox1a/b mutants show a reduction in lymphatic vessel development but still form up to 70% (in length) of the thoracic duct (TD). This study may suggest divergence

Fig. 1. LEC specification in embryonic veins during development. (left panel) During mouse embryonic development, LEC specification is marked at around 9.5 dpc with the induction of PROX1 expression in a subset of endothelial cells (red) in the dorsal wall of the cardinal vein (CV). (middle panel) At 10.5 dpc, LEC progenitors migrate out of the wall of the CV and acquire more lymphatic specific markers (NRP2, podoplanin) (green). LEC progenitors are localized in defined clusters arranged according to an anteroposterior axis along the CV (PLC). LEC specification is not restricted to the CV and also occurs at the level of intersegmental veins (vISV). (right panel) At around 11 dpc, migrating LECs (green) invade the surrounding environment to form the primary lymphatic structures. LEC progenitors (red) are detected in the superficial venous plexus (sVP) and probably contribute to the formation of the superficial lymphatic network.

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011

C. Pichol-Thievend et al. / Microvascular Research xxx (2014) xxx–xxx

or redundancy with other Prox1-related transcription factors (van Impel et al., 2014). SRY-related HMG domain 18 (Sox18) SOX18 belongs to the highly conserved family of SRY-related HMG domain transcription factors and together with SOX7 and SOX17, constitutes the SOX-F group. SOX18 is expressed in the vascular endothelium and hair follicles during mouse embryo development (Pennisi et al., 2000). Point mutations in Sox18, resulting in a dominant negative form of this protein, cause the cardiovascular and hair follicle phenotype responsible of the ragged opossum mouse phenotype (Pennisi et al., 2000). Furthermore, the equivalent mutations in the human SOX18 gene, which also generate a dominant negative SOX18 variant, are associated with the syndrome hypotrichosis–lymphedema–telangiectasia (HLT) (Irrthum et al., 2003; Moalem et al., 2014). HLT is characterized by alopecia and lymphedema. During development, the expression of SOX18 in the venous endothelium will commit this cell lineage towards a lymphatic fate. Genetic inactivation of Sox18 in a pure C57BL6 background results in a total absence of PROX1-positive LEC progenitors. In vivo and in vitro, SOX18 was shown to directly activate the transcription of Prox1 by binding to its promoter (François et al., 2008). Interestingly, SOX-Fs have redundant roles during development and can compensate for each other's loss in certain conditions. SOX7 and SOX17 act as modifiers of SOX18 function in some genetic backgrounds in mice (Hosking et al., 2009). Likewise, in zebrafish, redundant roles of Sox7 and Sox18 establish arteriovenous identity (Cermenati et al., 2008; Herpers et al., 2008; Pendeville et al., 2008). Chicken ovalbumin upstream promoter-transcription factor II (COUP-TF II) COUP-TFII is an orphan receptor transcription factor highly expressed in mesenchymal and venous endothelial cells. Coup-TFII−/− mutant embryos die at 10.0 dpc due to abnormal venous development (Pereira et al., 1999). Conditional endothelial knockout (e.g. using Tie2Cre) results in a loss of venous identity, which is associated with a dramatic decrease of LEC progenitors (Srinivasan et al., 2007; You et al., 2005). COUP-TFII regulates venous identity by inhibiting Notch activity in endothelial cells (You et al., 2005). Deletion of both Coup-TFII and the transcriptional effector of Notch signalling pathway Rbpj partially rescues venous identity (Srinivasan et al., 2010). In this context, Prox1 expression remained reduced, suggesting that COUP-TFII is required for LEC specification during development. Furthermore, COUP-TFII is able to bind to the Prox1 promoter to activate its transcription during LEC

3

specification (Srinivasan et al., 2010). Interestingly, COUP-TFII also controls the early maintenance of PROX1 expression between 10.5 and 12.5 dpc, serving as more than just an early inducer of LEC fate (Lin et al., 2010). This contrasts the role of SOX18, which is thought to act only during the initial induction stage. Recent mechanistic insights on COUP-TFII function elegantly showed that COUP-TFII homodimerisation induces venous endothelial cell (VEC) gene expression, whereas its heterodimerisation with PROX1 results in LEC gene specification (Aranguren et al., 2013). These three factors above are the most well-defined transcriptional regulators of LEC fate. They each have their own distinct roles, but essentially they function as a team. Future work is needed to understand their combinatorial actions and determine if they act exclusively as a trio or are part of a larger complex in the earliest LEC nucleus.

Factors increasing specified LECs and promoting emergence from embryonic veins While much is known about the signalling pathways that control lymphangiogenesis and has been reviewed in detail elsewhere (Coso et al., 2014), some pathways have emerged that 'tune' the number of LECs that leave the veins and can influence the transcriptional control of LEC fate in a direct manner. In the second part of this review, we will focus on factors involved in the promotion and/or restriction of PROX1-positive LEC progenitors during embryonic development and will highlight how they may directly influence LEC fate (Fig. 2).

VEGFC-D/CCBE1/VEGFR3 signalling Vascular growth factor C (VEGFC), together with its receptor VEGFR3, are essential for lymphangiogenesis during both embryonic and adult vessel formation and this pathway represents the most centrally and selectively lymphangiogenic pathway known. In pathological contexts, mutations in VEGFC and VEGFR3 have been shown to result in lymphoedema (Gordon et al., 2013; Irrthum et al., 2000; Karkkainen et al., 2000), whereas reactivation of this signalling pathway in tumours stimulates de novo lymphangiogenesis leading to metastasis (He et al., 2005; Karnezis et al., 2012). The role of VEGFC in promoting lymphatic vessel formation has been studied in diverse models in vivo and in vitro uncovering the function of VEGFC in promoting LEC proliferation, migration and survival through activation of VEGFR3, which in turn stimulates downstream AKT/ERK signalling pathways (reviewed in (Zheng et al., 2014).

Fig. 2. Molecular mechanisms of LEC emergence during embryonic development. Schematic overview of the molecular pathways that control LEC specification and modulate the process at the level of the cardinal vein. Transcriptional control of LEC specification is controlled by SOX18 and COUPTF-II, which are expressed at the level of the CV and induce PROX1 expression. PROX1 expression defines the LEC as specified. Migrating lymphatics acquire specific markers such as podoplanin (PDP) and NRP2. PROX1 and COUPTF-II expression are also required for LEC fate maintenance. VEGFC, CCBE1 and ERK have been identified as direct activators of LEC emergence by increasing the number of PROX1-positive cells at the level of the CV. CCBE1 is required for VEGFC processing and secretion, whereas ERK activation induces SOX18 expression in vitro, suggesting a crosstalk between growth factor signalling and transcriptional control. On the opposite, Notch, TGFb signalling and CYP26B1 restricts the number of LEC progenitors (restrictors) during embryonic mouse development. However, the mechanisms by which these factors restrict LEC specifications need to be determined.

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011

4

C. Pichol-Thievend et al. / Microvascular Research xxx (2014) xxx–xxx

Studies on Vegfc−/− mutants revealed that they are completely devoid of lymphatic vessels, which is due to the incapacity of LEC progenitors to bud from the CV to form the primary lymphatic structures. Interestingly, PROX1-positive LEC progenitors are still detected in the CV, suggesting that LEC transcriptional specification is not sufficient to induce endothelial cell migration out of the CV. VEGFR3 is initially expressed in endothelial cells during the early stages of development and becomes highly enriched in the lymphatic vasculature at around 11.5 dpc. VEGFR3 is essential for the formation and remodelling of the blood vascular network and Vegfr3−/− embryos die during embryogenesis. This early lethality is not reproduced in VEGFC/VEGFD double knockout animals (Haiko et al., 2008), indicative of the broader not-exclusively lymphatic functions of VEGFR3. Intriguingly, Vegfr3 mutant mice lacking the ligand-binding domain show a distinct phenotype from Vegfr3 mutant animals kinase domain activity (Zhang et al., 2010). Sprouting lymphangiogenesis is abolished in both mutants, whereas lymph sacs still form in the ligand-binding mutant, suggesting that other VEGFs or ligand-independent signalling might play roles in some contexts. VEGF signalling in lymphangiogenesis is also propagated through interaction of VEGFR3 with its co-receptor neuropilin-2 (NRP2), this interaction is essential to induce lymphatic vessel sprouting (Yuan et al., 2002). In mouse, loss of NRP2 function alters the number of small lymphatic vessels. VEGFC/VEGFR3 signalling function is evolutionary conserved as it is also required for LEC emergence during zebrafish development (Hogan et al., 2009b; Küchler et al., 2006; Yaniv et al., 2006). Furthermore, during zebrafish and Xenopus development, the interaction between Nrp2 and Synectin promotes the sprouting of LEC precursors from the veins (Hermans et al., 2010). CCBE1 is an extracellular matrix protein that was originally identified by genetic screening in the zebrafish (Hogan et al., 2009a) and is mutated in patients presenting with Hennekam syndrome, a rare disease characterized by lymphedema and lymphangectasia (Alders et al., 2009; Connell et al., 2010). During development, CCBE1 is indispensable for lymphangiogenesis in both zebrafish and mice (Bos et al., 2011; Hogan et al., 2009a). Ccbe1 acts non-cell autonomously to regulate the initial budding of lymphangioblasts from the CV (Bos et al., 2011; Hogan et al., 2009a) and it does this by enhancing the processing and secretion of VEGFC (Jeltsch et al., 2014; Le Guen et al., 2014). Ccbe1 is necessary for Vegfr3-dependent Erk signalling in veins (Le Guen et al., 2014) and achieves this, as shown in vitro, by enhancing the activity of the A disintegrin and metalloprotease with thrombospondin motifs-3 (ADAMTS-3), which functions to cleave VEGFC (Jeltsch et al., 2014). VEGFD is an additional member of the Vegf family and has the potential to bind to VEGFR3, but this interaction does not promote lymphangiogenesis during embryonic development (Baldwin et al., 2005). However, VEGFD upregulation promotes pathological neolymphangiogenesis (Koch et al., 2009; Stacker et al., 2001) and affects blood vascular angiogenesis (Song et al., 2007). Particularly interesting in relation to the process of LEC specification, recent work in both mice and zebrafish has shown that VEGFD is necessary for normal blood vascular development in the absence of SOX18 (Duong et al., 2014). In this study, double VEGFD/SOX18 mutant mice and morphant zebrafish uncovered a synergistic genetic interaction in the formation of the early blood vasculature. Further mechanistic analysis revealed that the transcriptional activity of SOX18 is specifically enhanced at target promoters in vitro and in vivo, and that SOX18 protein distribution in the nucleus is enhanced by VEGF signalling through ERK (Duong et al., 2014). Combined with another zebrafish knockdown study that earlier showed a synergistic phenotypic interaction between Sox18 and Vegfc (Cermenati et al., 2013), this work indicates a level of interaction between growth factors and the lymphangiogenic transcription factor SOX18. Importantly, these pathways had been previously considered mutually exclusive and so these studies suggest a mechanism that may act to integrate the inductive roles of transcription factors in LEC fate with the inductive roles of growth factors in promoting LEC sprouting from veins.

RAF1/MEK/ERK intracellular signalling The Ras/Raf/MEK/ERK cascade couples signals from cell surface receptors to transcription factors, which regulate gene expression. The signalling pathway is activated by the interaction between the Ras family of GTPases (Hras, Nras and Kras) Ras and RAF1, which in turn activates MEK kinases and then ERK1/2. Disruption of Ras genes in endothelial cells led to lymphatic vessel hypoplasia in vivo and downregulated VEGFR3 expression in vitro suggesting that the RAF1/MEK/ERK pathway regulates lymphatic vessel growth by modulating Vegfr3 signalling in LECs (Ichise et al., 2010). Furthermore, during blood vessel formation in zebrafish embryos, MAPK/ERK signalling and the PI3/AKT pathway exert opposite effects to control arteriovenous differentiation (Hong et al., 2006; Ren et al., 2010). Whereas ERK signalling promotes arterial specification, PI3Kinase has an opposing effect by blocking ERK activation via AKT1-dependent phosphorylation of RAF1 on Ser259 (Ren et al., 2010). It was recently demonstrated quite directly, that the RAF1/MEK/ERK pathway is involved in LEC specification. An endothelial-specific, inducible, transgenic mouse was generated carrying a specific mutation in RAF1 (S259A), which inhibits the crosstalk between RAF1 and AKT, thereby leading to ERK activation (Deng et al., 2013). Transgenic embryos are characterized by subcutaneous edema due to a massive enlargement and malformation of the jugular lymph sac and subcutaneous vessels. More importantly, this mutation results in an increase of PROX1-positive cells at the level of the CV, suggesting a role for ERK signalling in LEC specification. At 10.5 dpc, whereas few PROX1-positive cells have migrated out of the CV in normal conditions, in transgenic embryos with constitutively active RAF many PROX1-positive cells have migrated out of the CV and are detectable both at the level of the CV and the DA. Furthermore, this increase in PROX1-positive cells correlated with stronger Sox18 expression at the level of the blood vasculature (arteries and veins) and in lymphatic vessels at 14.5 dpc. In vitro, RAF1S259A is capable of inducing Sox18 expression, thus demonstrating that RAF1/MEK/ERK signalling regulates LEC commitment and subsequent specification (Deng et al., 2013). This mechanistic study demonstrates a level of direct crosstalk between growth factor-induced signalling and transcriptional regulation of LEC fate. The above studies combined have warranted a revisiting of the concept of separate VEGF-mediated induction of sprouting and distinct transcriptional control of LEC fate, highlighting far more integrated molecular processes that concurrently drive fate induction and emergence from veins. Factors decreasing specified LECs during emergence from embryonic veins Notch signalling The Notch pathway is an evolutionarily conserved signalling system that plays a major role during blood vessel development. Notch genes encode membrane receptors (NOTCH1 to NOTCH4) that bind to their ligands of the Jagged (JAG1 and JAG2) or Delta-like (DLL1, DLL3 and DLL4) families. The interaction of Notch receptors and ligands leads to the intracellular cleavage of Notch by a γ secretase, releasing the intracellular domain, which then translocates to the nucleus to subsequently activate target genes via the activity of the RBPJ transcription factor. During blood vessel formation, Notch signalling regulates arteriovenous differentiation by maintaining arterial identity and inhibiting venous differentiation (Gridley, 2010). During sprouting angiogenesis, lateral inhibition through DLL4 and the Notch pathway leads to the restriction of the number of tip cells, maintaining the identity of the stalk cells and restricting angiogenesis (Benedito et al., 2009; Hellstrom et al., 2007; Jakobsson et al., 2010; Lobov et al., 2007). Cells that are Notch signalling-positive are less responsive to VEGF signalling, hence Notch dampens the response of endothelial cells to VEGFs. Despite the many known functions in blood vascular angiogenesis and development, a role for Notch signalling remained elusive in lymphangiogenesis for some time. In vitro studies using LEC-coated beads in a 3D fibrin gel demonstrated that Notch signalling inhibition

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011

C. Pichol-Thievend et al. / Microvascular Research xxx (2014) xxx–xxx

induces lymphangiogenic tip cell formation via VEGFR2 and VEGFR3 signalling, strongly suggesting an in vivo role (Zheng et al., 2011). Moreover, Notch signalling downregulates the expression of PROX1 and COUPTF-II in primary LECs (Kang et al., 2010). Studies in vivo both in zebrafish and mouse have recently shown that Notch signalling is required for establishing normal embryonic lymphangiogenesis. Work from zebrafish has shown that during lymphatic development, inhibition of Dll4 and Notch (1b and 6) leads to a decrease in the number of lymphatic precursors at the level of the myoseptum, suggesting that Notch regulates the initial step of lymphangiogenesis (Geudens et al., 2010). Similarly, inhibiting DLL4/Notch signalling pathway during mouse postnatal lymphangiogenesis of the tail, ear and intestine using blocking antibodies results in decreased overall lymphatic density (Niessen et al., 2011). These studies contrast work in blood vascular Notch signalling and in vitro, where reduced Notch leads to increased vessel sprouting. However, both studies were complicated by phenotypes in adjacent vessel beds, which made the interpretation of observations challenging. Recent studies in mouse demonstrated the role of Notch signalling in restricting LEC specification during embryonic development (Fatima et al., 2014; Murtomaki et al., 2013). Deleting Notch1 specifically in PROX1-positive cells in mouse embryos or inactivating the Notchdependent transcriptional complex both result in an increase of LEC progenitors migrating out of the CV. Furthermore, the lymphatic markers LYVE-1 and podoplanin, which are normally restricted to the lymphatic endothelium, become misexpressed in this setting (Murtomaki et al., 2013). In addition, lack of Notch1 resulted in enhanced lymphatic sprouting along with increased LEC proliferation and survival in vivo (Fatima et al., 2014). Thus, while there remain some contradictions in the recent literature relating to the role of Notch in LECs, these particular in vivo studies indicated that Notch signalling restricts LEC fate within the wall of the CV and restricts sprouting of LECs in a manner that is similar to blood vessel angiogenesis. Prox1 expression is likely repressed in high Notch scenarios, but it remains unclear how this controls the specification of LECs at the mechanistic, transcriptional level. Certainly, it seems that there is activity in the right cells, at the right developmental stages for there to be a relatively direct mechanism at play. Cyp26b1 Retinoic acid (RA) is the active derivative of vitamin A. RA has pleiotropic functions during development, it regulates embryonic anterior-posterior patterning and a balance between degradation and synthesis of RA is important for organogenesis in a number of organs. It has been shown that during cardiovascular system development, the appropriate level of RA is essential for the establishment of the heart (reviewed in (Niederreither and Dollé, 2008). Throughout the body, the amount of RA is tightly regulated by RA-synthesising enzymes (retinaldehyde dehydrogenases; RALDH1, -2 and -3) and RA-catabolising enzymes (cytochrome P450 enzymes, Cyp26A1, -26B1 and -26C1). The involvement of RA during initial lymphatic sprouting from veins has been recently demonstrated in several settings. Firstly, in vitro, RA and its derivatives have been shown to activate proliferation, migration and tube formation of human lymphatic endothelial cells (hLECs) (Choi et al., 2012). Secondly, incubation of mouse embryoid bodies with RA result in an increase of CD31 +/LYVE1+ positive structures mediated via RA receptor (RAR-a) signalling (Marino et al., 2011). Thirdly, in an in vivo model where mouse and Xenopus embryos were exposed to excessive concentrations of RA (10 μM), lymphatic markers (LYVE 1 and VEGFR3) were upregulated throughout the developing vasculature (Marino et al., 2011). Most recently, an analysis by Bowles et al. further revealed the role of RA during mouse lymphatic vascular development. Using a retinoic acid transgenic reporter mouse (RARE-LacZ), this approach revealed that RA-induced gene activity is restricted to the ventral

5

part of the CV – opposite to the location of PROX1-positive cells. This observation suggested that LEC progenitors are normally subject to low RA concentrations in vivo (Bowles et al., 2014). Cyp26b1 knockout embryos are deficient for a key RA degrading enzyme and express excessive levels of RA. These animals presented with an enlarged region of PROX1-positive/LYVE1-positive progenitors that expand ectopically to the ventral and medial part of the CV during development. Conversely, transgenic embryos overexpressing Cyp26b1, which have systemically reduced RA levels, display smaller jugular lymph sacs (Bowles et al., 2014). Altogether, these data suggest that a window of RA concentration in vivo is critical to properly pattern the lymphatic vessels during the initial steps of lymphangiogenesis. Importantly, this is occurring in the wall of the CV concomitant with the specification of PROX1-positive precursors. The question of whether or not RA signalling transduces to the nucleus, to modulate transcription factors activity, or acts through other modulatory pathways remains to be addressed. TGFb/BMP signalling The TGFb signalling pathway is activated in a large number of cell types to regulate growth, differentiation, migration, adhesion and apoptosis. TGF/BMPs ligands bind to TGFb type II receptors (BMPR2, ACVR2A and ACVR2B) inducing the phosphorylation of type I receptors (ALK1-7), which in turn activate the SMAD transcription factors (SMAD 1, 5 and 8). SMAD transcription factors interact with SMAD4, translocate to the nucleus and activate the transcription of specific target genes (reviewed in (von Bubnoff and Cho, 2001). The TGFb signalling pathway is involved in developmental sprouting angiogenesis and in the maintenance of vascular integrity in the adult. In humans, various genetic alterations within this protein family cause vascular disorders, with phenotypes including the disintegration of vascular integrity in which hereditary haemorrhagic telangiectasia leads to abnormal blood vessel formation in the lungs, brain, liver and gastrointestinal tract (Lebrin et al., 2005; Park et al., 2008). BMP9 and 10 have been shown to regulate postnatal angiogenesis of the mouse retina (Ricard et al., 2012), and in vitro BMP9 induces proliferation and migration of endothelial cells (Suzuki et al., 2010). BMP9 and BMP2 have also been recently identified as negative regulators of developmental lymphangiogenesis (Dunworth et al., 2014; Yoshimatsu et al., 2013). Knockout of Bmp9 or Alk1 in mouse leads to lymphatic vessel enlargement in multiple organs, suggesting enhanced proliferation of LECs. In vitro, BMP9 is able to directly downregulate PROX1 expression via ALK1 (Yoshimatsu et al., 2013). During zebrafish development, overexpression of bmp2b (an ortholog for mammalian BMP2) results in the almost complete absence of the TD at 4 dpf. In addition, the absence of LECs at 60 hpf, prior to TD formation suggests that Bmp2 negatively regulates LEC emergence from the CV in zebrafish. Mechanistically, Bmp2 signalling strongly inhibits Prox1 expression, in this case by induction of the expression of micro-RNAs (miR-31 and miR181a) in an SMAD-dependent manner (Dunworth et al., 2014). In addition, recent work has shown that conditional deletion of TGFb receptors Tgfbr1 and Tgfbr2 in LECs during development reduces lymphatic network coverage in the skin and impaired tip cell formation, in parallel with slightly increased LEC numbers suggesting that TGFb signalling is involved in the patterning of the lymphatic network (James et al., 2013). Altogether, these studies showed quite clearly that TGFb/ BMP signalling plays important roles in lymphatic vessel formation during embryonic development. However, the precise role of TGFb receptors and their contribution in early LEC specification and emergence from the CV has not been studied in detail. Future directions Altogether, the studies highlighted above demonstrate a recent bounty of significant advances in understanding the mechanisms controlling lymphatic endothelial cell specification. The identification of

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011

6

C. Pichol-Thievend et al. / Microvascular Research xxx (2014) xxx–xxx

novel factors involved in the promotion or restriction of LEC emergence and the first insights into how these pathways interact with transcriptional regulators of specification has provided a more comprehensive and mature view of this process. Until recently, the literature has suggested that LEC transcriptional specification and LEC behavior during sprouting angiogenesis were regulated independently; however, it now seems clear that the processes are more integrated and may in fact regulate each other. Recent studies have provided new insights suggesting the existence of genetic interactions and the mechanistic interplay between transcription factor and vascular growth factor-induced signalling (e.g. Sox18 and ERK/VEGFC-D). A series of new signalling events have been implicated to act at the same time and in the same cells as LEC specification is occurring (e.g. Notch, TGFb/BMP and RA). Furthermore, a series of detailed cellular studies, which have ridden on the back of emerging and improving imaging technologies, have shown the concomitant nature of cellular sprouting and morphogenesis events with the acquisition of cell fate markers. It is now clear that further studies are required to better understand the level of integration of the molecular and cellular mechanisms at play in lymphangiogenesis and to fully uncover their influence over cell fate induction. We have for some time now understood the medical importance and potential therapeutic utility of the lymphatic vascular system in a wide range of pathologies. Beyond uncovering how this fascinating biological process occurs, a deeper understanding of the mechanistic control of LEC fate induction holds promise to uncover medical treatment strategies in contexts from lymphedema to inflammation and cancer metastasis. Acknowledgments MF was supported by a National Health and Medical Research Council (Australia) Career Development Award (APP1011242) and BMH was supported by an Australian Research Council Future Fellowship (FT100100165). References Alders, M., et al., 2009. Mutations in CCBE1 cause generalized lymph vessel dysplasia in humans. 41, 1272–1274. Alitalo, K., 2011. The lymphatic vasculature in disease. Nat. Med. 17, 1371–1380. Aranguren, X.L., et al., 2013. COUP-TFII orchestrates venous and lymphatic endothelial identity by homo- or hetero-dimerisation with PROX1. 126, 1164–1175. Baldwin, M.E., et al., 2005. Vascular endothelial growth factor D is dispensable for development of the lymphatic System. 25, 2441–2449. Benedito, R., et al., 2009. The notch ligands Dll4 and Jagged1 have opposing effects on angiogenesis. Cell 137, 1124–1135. Bos, F.L., et al., 2011. CCBE1 is essential for mammalian lymphatic vascular development and enhances the lymphangiogenic effect of vascular endothelial growth factor-C In Vivo. 109, 486–491. Bowles, J., et al., 2014. Control of retinoid levels by CYP26B1 is important for lymphatic vascular development in the mouse embryo. 386, 25–33. Bussmann, J., et al., 2010. Arteries provide essential guidance cues for lymphatic endothelial cells in the zebrafish trunk. Development 137, 2653–2657. Cermenati, S., et al., 2008. Sox18 and Sox7 play redundant roles in vascular, development. 111, 2657–2666. Cermenati, S., et al., 2013. Sox18 genetically interacts with VegfC to regulate lymphangiogenesis in zebrafish. 33, 1238–1247. Cha, Y.R., et al., 2012. Chemokine signaling directs trunk lymphatic network formation along the preexisting blood vasculature. Dev. Cell 22, 824–836. Choi, I., et al., 2012. 9-Cis retinoic acid promotes lymphangiogenesis and enhances lymphatic vessel regeneration: therapeutic implications of 9-Cis retinoic acid for secondary lymphedema. 125, 872–882. Connell, F., et al., 2010. Linkage and sequence analysis indicate that CCBE1 is mutated in recessively inherited generalised lymphatic dysplasia. Hum. Genet. 127, 231–241. Coso, S., et al., 2014. Pressing the right buttons: signaling in lymphangiogenesis. Deng, Y., et al., 2013. Endothelial ERK signaling controls lymphatic fate specification. 123, 1202–1215. Dunworth, W.P., et al., 2014. Bone morphogenetic protein 2 signaling negatively modulates lymphatic development in vertebrate embryos. 114, 56–66. Duong, T., et al., 2014. VEGFD regulates blood vascular development by modulating SOX18 activity. 123, 1102–1112. Fatima, A., et al., 2014. Murine Notch1is required for lymphatic vascular morphogenesis during development. Role of Notch1 in lymphatic vessel development (n/a-n/a).

Földi, M.,Strossenreuther, R., 2004. Foundations of Manual Lymph Drainage, Third edition. Elsevier, New York. François, M., et al., 2008. Sox18 induces development of the lymphatic vasculature in mice. 456, 643–647. François, M., et al., 2012. Segmental territories along the cardinal veins generate lymph sacs via a ballooning mechanism during embryonic lymphangiogenesis in mice. 364, 89–98. Geudens, I., et al., 2010. Role of Delta-like-4/Notch in the formation and wiring of the lymphatic network in zebrafish. 30, 1695–1702. Gordon, K., et al., 2013. Mutation in vascular endothelial growth factor-C, a ligand for vascular endothelial growth factor receptor-3, is associated with autosomal dominant Milroy-like primary lymphedema. 112, 956–960. Gridley, T., 2010. Current Topics in Developmental Biology. Elsevier. Hagerling, R., et al., 2013. A novel multistep mechanism for initial lymphangiogenesis in mouse embryos based on ultramicroscopy. EMBO J. 32, 629–644. Haiko, P., et al., 2008. Deletion of vascular endothelial growth factor C (VEGF-C) and VEGF-D is not equivalent to VEGF receptor 3 deletion in mouse embryos. Mol. Cell. Biol. 28, 4843–4850. He, Y., et al., 2005. Vascular endothelial cell growth factor receptor 3-mediated activation of lymphatic endothelium is crucial for tumor cell entry and spread via lymphatic vessels. Cancer Res. 65, 4739–4746. Hellstrom, M., et al., 2007. Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis. Nature 445, 776–780. Hermans, K., et al., 2010. Role of synectin in lymphatic development in zebrafish and frogs. 116, 3356–3366. Herpers, R., et al., 2008. Redundant roles for Sox7 and Sox18 in arteriovenous specification in zebrafish. 102, 12–15. Hogan, B.M., et al., 2009a. ccbe1 is required for embryonic lymphangiogenesis and venous sprouting. 41, 396–398. Hogan, B.M., et al., 2009b. Vegfc/Flt4 signalling is suppressed by Dll4 in developing zebrafish intersegmental arteries. Development 136, 4001–4009. Hong, Y.-K., et al., 2002. Prox1 is a master control gene in the program specifying lymphatic endothelial cell fate. 225, 351–357. Hong, C.C., et al., 2006. Artery/vein specification is governed by opposing phosphatidylinositol-3 kinase and MAP kinase/ERK signaling. 16, 1366–1372. Hosking, B., et al., 2009. Sox7 and Sox17 are strain-specific modifiers of the lymphangiogenic defects caused by Sox18 dysfunction in mice. Development 136, 2385–2391. Huntington, G.S., 1908. The genetic interpretation of the development of the lymphatic system. Anat. Rec. 2, 19–46. Ichise, T., et al., 2010. H-, N- and Kras cooperatively regulate lymphatic vessel growth by modulating VEGFR3 expression in lymphatic endothelial cells in mice. 137, 1003–1013. Irrthum, A., et al., 2000. Congenital hereditary lymphedema caused by a mutation that inactivates VEGFR3 tyrosine kinase. 67, 295–301. Irrthum, A., et al., 2003. Mutations in the transcription factor gene SOX18 underlie recessive and dominant forms of hypotrichosis-lymphedema-telangiectasia. 72, 1470–1478. Jakobsson, L., et al., 2010. Endothelial cells dynamically compete for the tip cell position during angiogenic sprouting. 12, 943–953. James, J.M.,Nalbandian, A.,Mukouyama, Y.S., 2013. TGF signaling is required for sprouting lymphangiogenesis during lymphatic network development in the skin. Development 140 (18), 3903–3914. http://dx.doi.org/10.1242/dev.095026. Jeltsch, M., et al., 2014. CCBE1 enhances lymphangiogenesis via ADAMTS3-mediated VEGF-C activation. Johnson, N.C., et al., 2008. Lymphatic endothelial cell identity is reversible and its maintenance requires Prox1 activity. Genes Dev. 22, 3282–3291. Kang, J., et al., 2010. An exquisite cross-control mechanism among endothelial cell fate regulators directs the plasticity and heterogeneity of lymphatic endothelial cells. Blood 116, 140–150. Karkkainen, M.J., et al., 2000. Missense mutations interfere with VEGFR-3 signalling in primary lymphoedema. 25, 153–159. Karnezis, T., et al., 2012. VEGF-D promotes tumor metastasis by regulating prostaglandins produced by the collecting lymphatic endothelium. Cancer Cell 21, 181–195. Kim, H., et al., 2010. Embryonic vascular endothelial cells are malleable to reprogramming via Prox1 to a lymphatic gene signature. 10, 72. Koch, M., et al., 2009. VEGF-D deficiency in mice does not affect embryonic or postnatal lymphangiogenesis but reduces lymphatic metastasis. 219, 356–364. Koltowska, K., et al., 2013. Getting out and about: the emergence and morphogenesis of the vertebrate lymphatic vasculature. 140, 1857–1870. Küchler, A.M., et al., 2006. Development of the zebrafish lymphatic system requires Vegfc signaling. 16, 1244–1248. Le Guen, L., et al., 2014. Ccbe1 regulates Vegfc-mediated induction of Vegfr3 signaling during embryonic lymphangiogenesis. 141, 1239–1249. Lebrin, F., et al., 2005. TGF-beta receptor function in the endothelium. Cardiovasc. Res. 65, 599–608. Lin, F.J., et al., 2010. Direct transcriptional regulation of neuropilin-2 by COUP-TFII modulates multiple steps in murine lymphatic vessel development. J. Clin. Invest. 120, 1694–1707. Lobov, I.B., et al., 2007. Delta-like ligand 4 (Dll4) is induced by VEGF as a negative regulator of angiogenic sprouting. Proc. Natl. Acad. Sci. U. S. A. 104, 3219–3224. Marino, D., et al., 2011. A role for all-trans-retinoic acid in the early steps of lymphatic vasculature development. 48, 236–251. Moalem, S., Brouillard, P.,Kuypers, D., Legius, E., Harvey, E.,Taylor, G., Francois, M., Vikkula, M., Chitayat, D., 2014. Hypotrichosis-lymphedema-telangiectasia-renal defect associated with a truncating mutation in the SOX18 gene. Clin Genet. http://dx.doi.org/10. 1111/cge.12388 (Epub ahead of print).

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011

C. Pichol-Thievend et al. / Microvascular Research xxx (2014) xxx–xxx Murtomaki, A., et al., 2013. Notch1 functions as a negative regulator of lymphatic endothelial cell differentiation in the venous endothelium. 140, 2365–2376. Niederreither, K., Dollé, P., 2008. Retinoic acid in development: towards an integrated view. 9, 541–553. Niessen, K., et al., 2011. The Notch1-Dll4 signaling pathway regulates mouse postnatal lymphatic development. 118, 1989–1997. Okuda, K.S., et al., 2012. lyve1 expression reveals novel lymphatic vessels and new mechanisms for lymphatic vessel development in zebrafish. 139, 2381–2391. Park, S.O., et al., 2008. ALK5- and TGFBR2-independent role of ALK1 in the pathogenesis of hereditary hemorrhagic telangiectasia type 2. Blood 111, 633–642. Pendeville, H., et al., 2008. Zebrafish Sox7 and Sox18 function together to control arterial– venous identity. 317, 405–416. Pennisi, D., et al., 2000. Mutations in Sox18 underlie cardiovascular and hair follicle defects in ragged mice. 24, 434–437. Pereira, F.A., et al., 1999. The orphan nuclear receptor COUP-TFII is required for angiogenesis and heart development. 13, 1037–1049. Petrova, T.V., et al., 2002. Lymphatic endothelial reprogramming of vascular endothelial cells by the Prox-1 homeobox transcription factor. 21, 4593–4599. Ren, B., et al., 2010. ERK1/2-Akt1 crosstalk regulates arteriogenesis in mice and zebrafish. 120, 1217–1228. Ricard, N., et al., 2012. BMP9 and BMP10 are critical for postnatal retinal vascular remodeling. 119, 6162–6171. Sabin, F.R., 1902. On the origin of the lymphatic system from the veins and the development of the lymph hearts and thoracic duct in the pig. Am. J. Anat. 1, 367–389. Song, M., et al., 2007. A critical role of vascular endothelial growth factor D in zebrafish embryonic vasculogenesis and angiogenesis. 357, 924–930. Srinivasan, R.S., et al., 2007. Lineage tracing demonstrates the venous origin of the mammalian lymphatic vasculature. 21, 2422–2432. Srinivasan, R.S., et al., 2010. The nuclear hormone receptor Coup-TFII is required for the initiation and early maintenance of Prox1 expression in lymphatic endothelial cells. 24, 696–707. Stacker, S.A., et al., 2001. VEGF-D promotes the metastatic spread of tumor cells via the lymphatics. 7, 186–191.

7

Suzuki, Y., et al., 2010. BMP-9 induces proliferation of multiple types of endothelial cells in vitro and in vivo. J. Cell Sci. 123, 1684–1692. van Impel, A., Schulte-Merker, S., 2014. A fisheye view on lymphangiogenesis. Adv. Anat. Embryol. Cell Biol. 214, 153–165. van Impel, A., et al., 2014. Divergence of zebrafish and mouse lymphatic cell fate specification pathways. 141, 1228–1238. von Bubnoff, A., Cho, K.W., 2001. Intracellular BMP signaling regulation in vertebrates: pathway or network? Dev. Biol. 239, 1–14. Wigle, J.T., Oliver, G., 1999. Prox1 function is required for the development of the murine lymphatic system. 98, pp. 769–778. Wigle, J.T., et al., 2002. An essential role for Prox1 in the induction of the lymphatic endothelial cell phenotype. 21, 1505–1513. Yang, Y., et al., 2012. Lymphatic endothelial progenitors bud from the cardinal vein and intersomitic vessels in mammalian embryos. 120, 2340–2348. Yaniv, K., et al., 2006. Live imaging of lymphatic development in the zebrafish. 12, 711–716. Yoshimatsu, Y., et al., 2013. Bone morphogenetic protein-9 inhibits lymphatic vessel formation via activin receptor-like kinase 1 during development and cancer progression. 110, 18940–18945. You, L.R., et al., 2005. Suppression of Notch signalling by the COUP-TFII transcription factor regulates vein identity. Nature 435, 98–104. Yuan, L., et al., 2002. Abnormal lymphatic vessel development in neuropilin 2 mutant mice. 129, 4797–4806. Zhang, L., et al., 2010. VEGFR-3 ligand-binding and kinase activity are required for lymphangiogenesis but not for angiogenesis. Cell Res. 20, 1319–1331. Zheng, W., et al., 2011. Notch restricts lymphatic vessel sprouting induced by vascular endothelial growth factor. 118, 1154–1162. Zheng, W., et al., 2014. Lymphangiogenic factors, mechanisms, and applications. 124, 878–887.

Please cite this article as: Pichol-Thievend, C., et al., Lymphatic vascular specification and its modulation during embryonic development, Microvasc. Res. (2014), http://dx.doi.org/10.1016/j.mvr.2014.07.011