Cell Tissue Res (2014) 355:515–522 DOI 10.1007/s00441-014-1843-7
AT-A-GLANCE
VE-cadherin at a glance Luca Bravi & Elisabetta Dejana & Maria Grazia Lampugnani
Received: 10 December 2013 / Accepted: 4 February 2014 / Published online: 20 March 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Although being a monolayer the vascular endothelium controls fundamental vessel functions such as permeability, leukocyte extravasation and angiogenesis. The endothelial selective transmembrane constituent of adherens junctions, Vascular Endothelial- (VE-) cadherin plays a crucial role in the regulation of such activities. The signaling pathways controlled by VE-cadherin as well as the ones that regulate VEcadherin activity start to be elucidated. This delineates a complex network of molecular and functional interactions that can be altered in pathologies. Keywords VE-cadherin . Cell-to-cell junctions . Endothelium . Permeability . Angiogenesis
by the endothelium and which was localized at cell-to-cell junctions. We then identified this protein through affinity purification and N-terminal sequencing as cadherin 5 (Suzuki et al. 1991) or CD144. Because of this and due to its cell-type specificity, we proposed the name of vascular endothelial cadherin (VEcadherin; Lampugnani et al. 1992), which has been commonly used since then in more than 1,600 papers (PubMed data, November 2013). VE-cadherin is the major transmembrane component of endothelial adherens junctions and it is expressed in vivo in any type of vessels, arterial, venous and lymphatic (Fig. 1). Its expression represents an early step in the differentiation of the endothelial phenotype (Breier et al. 1996). Although VE-cadherin is not necessary for the acquisition of endothelial traits, its absence deeply impairs the physiological progression of the vascular program (Carmeliet et al. 1999).
A little bit of ‘history’ The ultra-structural architecture of endothelial cell-to-cell junctions was first described in the 1970s by Simionescu et al. (1976). They identified specialized regions of the endothelial plasma membrane greatly resembling the ones present in the epithelia: the tight, adherens and gap junctions. However, the definition of the molecular composition of such membrane domains in the endothelium had to wait until the 1990s. In particular, our group, in a screening of monoclonal antibodies directed to endothelial cells, identified an antigen selectively and ubiquitously expressed L. Bravi : E. Dejana (*) : M. G. Lampugnani (*) FIRC Institute of Molecular Oncology (IFOM) Fondazione, Via Adamello 16, 20139 Milan, Italy e-mail: [email protected] e-mail: [email protected] E. Dejana Department of Biosciences, School of Sciences, Milan University, Milan, Italy M. G. Lampugnani Mario Negri Institute for Pharmacological Research, Milan, Italy
VE-cadherin and its interactors: a never-ending story As a classical cadherin, VE-cadherin is tightly bound to cytoplasmic catenins: beta-catenin, plakoglobin and p120 (Nelson 2008), which have a strong impact on the activity of VEcadherin. This is the consequence of the role of catenins in both tethering VE-cadherin to the cortical actin cytoskeleton (as well as controlling cytoskeleton functions) and in connecting VE-cadherin with several different molecules for the constitution of signaling clusters (Fig. 2). In addition, some catenins are important regulators of transcription, although the relationship between junctional and nuclear catenins is still not defined in mammals. A truncated form of VE-cadherin [truncation of the carboxy-terminal amino acids (aa) 703-784] is unable to bind beta-catenin and plakoglobin (but still binding p120) and loses most of its regulatory properties in vitro, although it still localizes at junctions and retains adhesiveness comparable to wt (Navarro et al. 1995). In addition, it produces a lethal vascular phenotype in the homozygous mutant mouse embryo
516
Fig. 1 VE-cadherin is ubiquitously expressed in the vascular endothelium and localizes to cell-to-cell junctions. a An artery (A), a vein (V) and smaller vessels branching from them in the retina of a 7-dpn mouse pup stained with an antibody to VE-cadherin. Bar 100 um. b, c Magnifications of the boxed areas in (a). Arrowheads point to VE-cadherin positive endothelial junctions in artery (A) and vein (V), respectively. Bar 30 um. Immunofluorescence confocal microscopy was used to produce this figure. A similar result is observed staining any vascularized organs starting from the earliest stages of endothelial differentiation in the embryo (Breier et al 1996). Lymphatic vessels are also positive to VEcadherin (Baluk et al 2007). Junctional distribution and expression of VEcadherin can be modulated in response to physiological and pathological stimuli (Dejana et al 2009). See text for details on this issue
comparable to the null mutation (death at 9.0 dpc; Carmeliet et al. 1999). A mutant VE-cadherin unable to bind p120 after deletion of the p120 binding region (juxta-membrane domain aa 621702) or mutation to alanine of each of three core aa of the p120 binding domain (Thoreson et al. 2000) (aa-aa EMD, 652-654 of human and mouse VE-cadherin) is also regularly distributed to cell-to-cell junctions (Lampugnani et al. 2002; Corada et al., unpublished results). In addition, the EMD mutant shows a steady-state surface expression level comparable to the wt (Nanes et al. 2012). However, the p120-binding domaindeleted VE-cadherin did not recruit the Rac activator Tiam (Rac guanine nucleotide exchange factor, GEF), neither activated Rac, nor induced organization of actin stress fibers (Lampugnani et al. 2002). This behavior mimicked the effect of VE-cadherin null mutation (and the truncation of the beta-catenin-binding domain), suggesting abnormal signaling from junctional VEcadherin in the absence of associated p120. In agreement, homozygous mice with endothelial-specific null mutation of p120 die around 11.5 dpc with disorganized and hemorrhagic vessels (Oas et al. 2010). All in all, these data indicate that the correct association between VE-cadherin and catenins is crucial for the effective functioning of endothelial junctions. The first molecule reported to bind cadherins using betacatenin is alfa-catenin (Nelson 2008). Such association is functionally crucial for VE-cadherin. Indeed, a-catenin can tether VE-cadherin to the cortical acto-myosin cytoskeleton (see Pokutta et al. 2008 for the updated model of alfa-catenin and actin interaction and functional regulation) that can contract in
Cell Tissue Res (2014) 355:515–522 Fig. 2 Some of the molecules that have been demonstrated to form ab complex with VE-cadherin are reported. The regions of interaction with VE-cadherin, identified by amino acid (aa) position, are indicated, when known, along the linear structure of VE-cadherin (mouse sequence, upper part of the figure). Several molecules require beta-catenin to associate to VE-cadherin. It is reasonable to envisage that distinct complexes are formed, possibly defining specific membrane subdomains at the adherens junctions. Whether heterogeneity in the molecules associated to VEcadherin can define functional subdomains in different vessels and how physiological or pathological stimuli can module such complexes is still an unexplored, although extremely interesting, topic. Also, some tyrosine (two out of the nine present in the mouse sequence) and serine residues of VE-cadherin, observed phosphorylated in the whole organism, are highlighted (red P). Phosho-Ser665 (phosphorylated by PAK, for example in response to VEGF) is the binding site of beta2-arrestin that promotes clathrin-dependent internalization of VE-cadherin (Gavard and Gutkind 2006). Tyrosine residues 658 and 685 are phosphorylated in vivo in stable venous vessels (low-shear stress activates Src-mediated phosphorylation) (Orsenigo et al. 2012). Although not impairing the barrier properties of unstimulated vessels, the phosphorylation of these two residues facilitates VE-cadherin internalization in response to permeability increasing agents. If and how the association of other molecules can be modulated by the phosphorylation of tyrosine residues of VEcadherin is still debated (Potter et al 2005; Adam et al 2010). Among the extracellular interactors also VE-cadherin is indicated. Cadherins can form cis-homodimers (through ECI-ECIV domains) that interact in trans with similar dimers present on adjoining cells to support cell-to-cell recognition and adhesion. VE-PTP requires ECV domain of VE-cadherin for binding. Binding to VE-cadherin via: *CCM1 associated to betacatenin; ^beta-catenin and MAGI1; °704-784 aa region of VE-cadherin; #phosphodiesterase 4D cAMP specific (PDE4D); &co-immunoprecipitation with VE-cadherin, binding domain not determined
response to stimulation of Rho small GTPase that activates Rock to phosphorylate MLC2. This process results in pulling apart the cell-to-cell junctions of adjoining cells, with formation of gaps and increased permeability (Lampugnani 2012). In contrast, in most cases, activation of Rac GTPase strengthens endothelial junctions inhibiting Rho and stabilizing the actin cytoskeleton (Spindler et al. 2010) (for exceptions, see van Wetering et al. 2002 who reported increased permeability due to production of oxygen species by NADPH in response to Rac activation and Gavard and Gutkind 2006, see below). VEcadherin can contribute to Rac activation recruiting at junctions Tiam, a Rac GEF (Lampugnani et al. 2002). This aspect is discussed in further detail in the following paragraph. Besides these traditional companions, a full cohort of molecules has been shown to be associated to VE-cadherin. They can be both cytoplasmic and trasmembrane molecules and bind VEcadherin in the vast majority trough beta-catenin, although some examples of direct binding have been reported (see Fig. 2 for a list, although not comprehensive, of interactors and its legend for further details). In face of the huge number of interactors, often bound either to common regions of VE-cadherin or via identical molecular bridging, it is reasonable to hypothesize that distinct complexes exist, possibly determining subdomains at adherens junctions supporting specific functions. In addition, it is reasonable to imagine that specific complexes could be selectively present in distinct sites of the vascular network
Cell Tissue Res (2014) 355:515–522
contributing to the distinct function of the endothelium in different tissues and organs. These issues, although crucial for a comprehensive description of the molecular organization of endothelial adherens junctions and for envisaging specific therapeutic interventions, remain still almost unexplored at the present time. We have observed heterogeneity of VE-cadherin phosphorylation at the venous versus arterious site in vivo, with
517
VE-cadherin phosphorylated on tyrosine 658 and 685 only in veins and small veins (Orsenigo et al. 2012). This issue is discussed in detail in the following paragraph. We have reported up to now examples of association to the cytoplasmic domain of VE-cadherin, often mediated by betacatenin, (see Fig. 2), for which a vast literature exists. The endothelial-specific phosphatase VE-PTP, on the contrary,
518
directly associates to the extracellular domain of VE-cadherin through its extracellular domain. When associated to VEcadherin at junctions, it cooperates for the maintenance of stable, non-permeable junctions (Nawroth et al. 2002). Besides VE-cadherin, VE-PTP can dephosphorylate (Fukuhara et al. 2008; Saharinen et al. 2008) Tie2, plakoglobin and junction associated-VEGFR2 (Hayashi et al. 2013).
VE-cadherin as a functional hub: adhesion and signaling Homotypic cell recognition and adhesion is a primary function of VE-cadherin. VE-cadherin can be recruited in early junctions by nectins previously engaged in adhesive interactions at junctions (Miyoshi and Takai 2007). Control of permeability is the significant consequence of the engagement of VE-cadherin molecules in cis- and trans-homophilic binding (Dejana et al. 2009). This function can be both the direct result of the adhesive recognition between VE-cadherin molecules and also its indirect outcome. Indeed, VE-cadherin clustered to cell-to-cell junctions activates a signaling pathway that upregulates the transcription and the deposition at junctions of claudin-5, the major transmembrane component of the tight junctions (Taddei et al. 2008), which seals the intercellular space to the passage of very small solutes (less than 800 D; Nitta et al. 2003). The mechanism of transcriptional regulation of claudin-5 represents an example of the signaling activity of VE-cadherin that has been studied in detail in our laboratory. When clustered in stable junctions, VE-cadherin recruits PI3K, through beta-catenin. This kinase locally activates Akt that phosphorylates FoxO1, a major repressor of claudin-5 transcription in association to beta-catenin. The phosphorylated FoxO1 is excluded from the nucleus (phosphorylation on Thr24 promotes its exclusion from the nucleus and phosphorylation on Ser256 inhibits its binding to DNA), thus relieving the inhibition of claudin-5 transcription (Taddei et al. 2008). VE-cadherin, when associated in stable junctions, besides contributing to the barrier function of the endothelium, enhances protection from apoptosis and moderates proliferation, two primary features of endothelial cells in adult stable vessels. (Dejana et al. 2009). To this aim, the association between VE-cadherin and two central managers of endothelial behavior, such as VEGFR2 and TGFbetaR2 (as well as Alk1 and Alk5), is crucial. Also, in this case, beta-catenin is required for the formation of the complex between VE-cadherin and these receptors (Lampugnani et al. 2006; Rudini et al. 2008). Further molecular details are still missing. Recently, FGFR1 has also been found associated to VE-cadherin, which reduces its phosphorylation associating the phosphate Dep1 in a multiprotein complex (Giampietro et al. 2012). In contrast, in cells with VE-cadherin engaged in either unstable or earlyestablished junctions, as during angiogenesis, the mitogenic
Cell Tissue Res (2014) 355:515–522
and migratory responses to VEGFR2 and TGFRbeta2 are enhanced, while their anti-apoptotic signaling is switched off. Both activities are crucial for the growth and remodeling of a new vascular network. The examples briefly discussed above suggest that the signaling activity of VE-cadherin is modulated by the quality of its clustering at cell-to-cell junctions, which in molecular terms regulates and allows specific interactions among molecules associated to VE-cadherin. The association of adherens junctions to the cortical actin cytoskeleton is critical to determine the stability of junctions (Nelson 2008), as introduced in the previous paragraph. The small GTPase Rap1, associated to VE-cadherin through CCM1 bound to beta-catenin (Glading et al. 2007), is critical to maintain VE-cadherin in stable junctions with the cooperation of the other members of the CCM complex, CCM2 and CCM3 (Stahl et al. 2008). A possible mechanism for the stabilization of VE-cadherin at cell-to-cell junctions would involve inhibition of acto-myosin contraction by Rac, activated by Rap1. On its side, Rap1 would be activated at adherens junctions by the association to VE-cadherin of PDZ-GEF1 and Epac, both Rap1 guanine-nucleotide-exchange factors (GEFs) (Pannekoek et al. 2011). VE-cadherin, in association to CCMs molecules, controls another central function of the endothelium that is the apical– basal polarity. The distribution to junctions of two components of the polarity complex, Par3 and aPKz, depends on VEcadherin (Lampugnani et al. 2010). Par3 directly associates to VE-cadherin (classII PDZ binding motif in the five terminal aa, Vestweber) and CCM1 is required for the activation of aPKz (Lampugnani et al. 2010). Polarization of the endothelium is fundamental in adhesion and trafficking of the immune cells as well as in the transport of molecules to and from the tissues. Another important consequence of the engagement of VEcadherin in stable adherens junctions is the regulation of the level of nuclear beta-catenin. Although the mechanism of this process is not yet settled, it is consistent with the observation that, when VE-cadherin participates in stable junctions, betacatenin is excluded from the nucleus, while when junctions are weak, beta-catenin concentrates in the nucleus where it is transcriptionally active in association to Tcf transcription factors (Clevers and Nusse 2012). It is not yet clear whether the beta-catenin that accumulates in the nucleus derives from junctions after detachment from VE-cadherin. A report in epithelial cells using photo-activable GFP-tagged betacatenin would indicate that beta-catenin could dissociate from E-cadherin and re-localize to the nucleus (Kam and Quaranta 2009). Beta-catenin can also be found at junctions not associated to E-cadherin (Maher et al. 2009). It is not known whether this beta-catenin was previously associated to Ecadherin, or whether it represents a cadherin-independent pool, as observed in Drosophila (Sanson et al. 1996) and
Cell Tissue Res (2014) 355:515–522
confirmed in C. elegans (Korswagen et al. 2000). In addition, it has been observed in epithelial cells that components of the machinery that drives degradation of beta-catenin (phosphodestruction complex: namely axin, APC2 and GSK3beta) are associated to cell-to-cell contacts and cadherins can promote N-terminal phosphorylation of beta-catenin, targeting it to proteosomal degradation and limiting the cytoplasmic pool of beta-catenin and as a consequence its localization to the nucleus. Therefore, the stability of junctions would regulate the activity of such machinery and the degradation of betacatenin (Maher et al. 2009). The consequence would be that the transcriptional activity of beta-catenin could be regulated by the state of junctions and be appropriate to it. It remains to be tested whether a similar mechanism can also operate in endothelial cells. Distribution of VE-cadherin to junctions can be regulated both in acute and chronic ways. Acute and generally reversible, disorganization of junctional VE-cadherin occurs in response to permeability increasing agents and short-term stimulation with growth factors. The mechanism involves stimulation of Rho GTPase, activation of the downstream effector ROCK that results in contraction of peripheral acto-myosin bundles and opening of intercellular gaps, as already introduced above. Chronic and non-reversible dissociation of VE-cadherin from junctions can be observed in certain pathologies and can be the consequence of the constitutive presence of angiogenic and inflammatory stimuli, of mutation in regulators of the stability of junctional VE-cadherin and of downregulation of the transcription and expression level of VE-cadherin. The permanence of clustered VE-cadherin at cell-to-cell junctions can also be modulated by post-transcriptional modification of the cytoplasmic domain of VE-cadherin. In particular, phosphorylation on tyrosine 658 and 685 (Y658 and Y685) can modulate the rate of internalization of VE-cadherin through a clathrin-dependent mechanism and can contribute to modify the stability, composition and signaling activity of adherens junctions. These post-transcriptional modifications correlate with a more prompt internalization and loss of barrier function in response to permeability increasing agents such as bradykinin and histamine. The composition of the molecular complexes formed by VE-cadherin phosphorylated on those tyrosine residues remains to be analyzed. This takes us to the ‘vexata quaestio’ of how post-transcriptional modification of VEcadherin, in particular tyrosine phosphorylation, can modulate the activity of VE-cadherin. Eight tyrosine residues (in the human form and nine in the murine one) could be potential targets of tyrosine kinases (Potter et al. 2005: fig. 2). Although, in the current model, tyrosine phosphorylation of VE-cadherin is considered to characterize weak and permeable junctions (Turowski et al. 2008; Potter et al. 2005), it
519
may, however, not be sufficient alone and other signals could cooperate to the weakening of the barrier (Adam et al. 2010; Orsenigo et al. 2012; see previous comment). As far as the function of other residues phosphorylated in vivo in the organism are concerned, the residue S665 of VE-cadherin is the docking site for beta-2 arrestin that promotes clathrin-dependent internalization of VE-cadherin (Gavard and Gutkind 2006). S665 has been reported to be phosphorylated in VE-cadherin of mouse skin vessels (Gavard et al. 2008), although the regulation of such phosphorylation has been prevalently studied in in vitro-cultured cells. Ser665 is phosphorylated by PAK, activated in response to VEGF, which sequentially activates Src, Vav2, a Rac GEF and Rac itself to trigger PAK . The weakening of endothelial cell-to-cell junctions ensuing such signaling sequence can be blocked by Angiopoietin-1 that sequesters Src through mDia. While Src has also been demonstrated to play an important, although not direct, role in the molecular mechanism of VEcadherin phosphorylation on tyrosine residues (see also Orsenigo et al. 2012), the phosphatases that regulate the phosphorylation of VE-cadherin are still poorly defined. Among the phosphatases associated to VE-cadherin, VEPTP can use VE-cadherin as a substrate (Saharinen et al. 2008; Fukuhara et al. 2008) while SHP2 does not (Ukropec et al. 2000). However, it is not known which phospho-tyrosine residues of VE-cadherin may be direct substrates of VE-PTP. To conclude, an increased level of junctional VEcadherin means more stable and less permeable endothelium. This has been observed in transgenic mice with haplo-insufficiency of the oxygen sensor PDH2 (Mazzone et al. 2009). In this condition, VE-cadherin transcript and protein increase and reshape and strengthen the endothelial lining to increase the control of permeability and cell extravasation. Tumor cell invasion, intravasation and metastasis were inhibited in these mutant mice (Mazzone et al. 2009).
VE-cadherin and pathologies While mutations in the VE-cadherin gene have not yet been reported in human patients, modifications in the VE-cadherin ‘situation’ can be observed in human pathologies. What has been reported up to now is that the distribution of VE-cadherin to adherens junctions, the transcription of the VE-cadherin gene and the stability of the protein can be modulated. While molecular mechanisms have been described to regulate each of these steps in experimental systems, as described in the previous paragraph, it still remains poorly defined how VEcadherin is de-regulated in pathologies. A typical example of disorganized and weak endothelial junctions is observed in tumor vessels, in which the local microenviroment is rich in angiogenic factors (typically VEGF produced in response to the hypoxic tumor
520
environment; Carmeliet and Jain 2011) and permeability increasing agents (such as inflammatory cytokines released either by tumor cells or infiltrating cells of the immune system), which determine dismantling of adherens junctions and permeable endothelial barrier (Baluk et al. 2005). Interestingly, in tumors vessels, VE-cadherin can expose a distinct epitope that can be specifically recognized by a monoclonal antibody. Such an antibody developed at ImClone, NY, US, besides specifically targeting tumor vessels could also induce inhibition of tumor growth and regression in mouse models of syngenic tumors or xenographs (Controllare) (May et al. 2005). Another human disease characterized by disorganized endothelial junctions is the cerebral cavernous malformation (CCM), a vascular pathology that presents vascular malformations typically localized at the venous side of the central nervous system vessels (Boulday et al. 2011; Maddaluno et al. 2013). The products of the three genes mutated in CCM pathology, CCM1, 2 and 3, form a complex (Stahl et al. 2008) that can associate to VE-cadherin through the direct interaction of β-catenin and CCM1 (Glading et al. 2007). CCM1 also binds Rap1 that, as mentioned before, exerts a fundamental action in stabilizing adherens junctions in endothelial cells (Kooistra et al. 2005). As a consequence of the ablation of any of the CCM gene, adherens junctions are disorganized and the transcription of the VE-cadherin gene can also be reduced. Such transcriptional regulation sets in the process of endothelial–mesenchymal transition that follows the loss of CCM proteins (Maddaluno et al. 2013) and involves a typical cadherin switch with increased expression of N-cadherin (Taddei et al. 2008; Giampietro et al. 2012) and mesenchymal markers. This process of de-differentiation represents a crucial step in the establishment of the vascular lesion and is determined by the de-regulated signaling of TGF-beta and beta-catenin pathways (Maddaluno et al. 2013; Bravi, unpublished).
Perspectives: from research to therapeutic application Several issues in the basic regulation of VE-cadherin organization and function remain to be defined, as has been pointed out in the text. Regarding the translational aspect, a series of crucial questions should be answered: Are there perspectives of using VEcadherin for a targeted vascular therapy and how to open and close adherens junctions and possibly reversibly, targeting VE-cadherin? Some drugs have been shown to stabilize adherens junctions and reduce vascular permeability both in wild-type and pathological (CCM1 and 2 heterozygous) animal models, blocking Rock fasudil that competes for ATP binding (Yamashita et al. 2007; Stockton et al. 2010) and membrane-
Cell Tissue Res (2014) 355:515–522
association of Rho (statins that inhibit isoprenylation of Rho GTPase; Collisson et al. 2002; Park et al. 2002; Whitehead et al. 2009). However, drugs able to weaken junctions targeting VE-cadherin have not yet been reported to our knowledge. Opening endothelial junctions could be particularly useful in ischemic pathologies of many organs or in pathologies of the central nervous system where the tissue is protected by the blood–brain barrier. A possible direct approach could be to down-regulate VE-cadherin transcription using either siRNA or shRNA. This method, which is quite effective in in vitro systems of cells in culture, would require precise targeting, as a generalized down-regulation of VEcadherin expression could have dangerous consequences, as described above. Local delivery would require the identification of specific epitopes expressed by endothelial cells in the particular environment that needs to be targeted. This implies definition of the local phenotype of endothelial cells and the development of tools for specific delivery (for example, nanoparticles) (Davis et al. 2010; Thanou and Gedroyc 2013). Acknowledgements This work is supported by grants from the Fondation Leducq Transatlantic Network of Excellence, Associazione Italiana per la Ricerca sul Cancro (AIRC) and ‘Special ProgramMolecular Clinical Oncology 5X1,000 to AGIMM (AIRC-Gruppo Italiano Malattie Mieloproliferative), the European Community (ENDOSTEM-HEALTH2009-241440; JUSTBRAIN-HEALTH-2009-241861; ITN Vessels), the European Research Council (ERC) and CARIPLO Foundation.
References Adam AP, Sharenko AL, Pumiglia K, Vincent PA (2010) Src-induced tyrosine phosphorylation of VE-cadherin is not sufficient to decrease barrier function of endothelial monolayers. J Biol Chem 285:7045–7055 Baluk P et al (2007) Functionally specialized junctions between endothelial cells of lymphatic vessels. J Exp Med 204:2349–2362 Baluk P, Hashizume H, McDonald DM (2005) Cellular abnormalities of blood vessels as targets in cancer. Curr Opin Genet Dev 15:102–111 Baumeister U et al (2005) Association of Csk to VE-cadherin and inhibition of cell proliferation. Embo J 24:1686–1695 Boulday G et al (2011) Developmental timing of CCM2 loss influences cerebral cavernous malformations in mice. J Exp Med 208:1835–1847 Breier G et al (1996) Molecular cloning and expression of murine vascular endothelial-cadherin in early stage development of cardiovascular system. Blood 87:630–641 Carmeliet P, Jain RK (2011) Principles and mechanisms of vessel normalization for cancer and other angiogenic diseases. Nat Rev Drug Discov 10:417–427 Carmeliet P et al (1999) Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell 98:147–157 Clevers H, Nusse R (2012) Wnt/beta-catenin signaling and disease. Cell 149:1192–1205 Collisson EA, Carranza DC, Chen IY, Kolodney MS (2002) Isoprenylation is necessary for the full invasive potential of RhoA overexpression in human melanoma cells. J Invest Dermatol 119: 1172–1176
Cell Tissue Res (2014) 355:515–522 Davis ME et al (2010) Evidence of RNAi in humans from systemically administered siRNAvia targeted nanoparticles. Nature 464:1067–1070 Dejana E, Tournier-Lasserve E, Weinstein BM (2009) The control of vascular integrity by endothelial cell junctions: molecular basis and pathological implications. Dev Cell 16:209–221 Fukuhara S et al (2008) Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1. Nat Cell Biol 10:513–526 Gavard J, Gutkind JS (2006) VEGF controls endothelial-cell permeability by promoting the beta-arrestin-dependent endocytosis of VEcadherin. Nat Cell Biol 8:1223–1234 Gavard J, Patel V, GutkindJS (2008) Angiopoietin-1 prevents VEGFinduced endothelial permeability by sequestering Src through mDia. Dev Cell 14:25–36 Giampietro C et al (2012) Overlapping and divergent signaling pathways of N-cadherin and VE-cadherin in endothelial cells. Blood 119: 2159–2170 Glading A, Han J, Stockton RA, Ginsberg MH (2007) KRIT-1/CCM1 is a Rap1 effector that regulates endothelial cell cell junctions. J Cell Biol 179:247–254 Hayashi M et al (2013) VE-PTP regulates VEGFR2 activity in stalk cells to establish endothelial cell polarity and lumen formation. Nat Commun 4:1672 Iden S et al (2006) A distinct PAR complex associates physically with VEcadherin in vertebrate endothelial cells. EMBO Rep 7:1239–1246 Kam Y, Quaranta V (2009) Cadherin-bound beta-catenin feeds into the Wnt pathway upon adherens junctions dissociation: evidence for an intersection between beta-catenin pools. PLoS ONE 4:e4580 Kooistra MR, Corada M, Dejana E, Bos JL (2005) Epac1 regulates integrity of endothelial cell junctions through VE-cadherin. FEBS Lett 579:4966–4972 Korswagen HC, Herman MA, Clevers HC (2000) Distinct beta-catenins mediate adhesion and signalling functions in C. elegans. Nature 406: 527–532 Kronstein R et al (2012) Caveolin-1 opens endothelial cell junctions by targeting catenins. Cardiovasc Res 93:130–140 Lampugnani MG (2012) Endothelial cell-to-cell junctions: adhesion and signaling in physiology and pathology. Cold Spring Harb Perspect Med 2 Lampugnani MG et al (1997) Cell confluence regulates tyrosine phosphorylation of adherens junction components in endothelial cells. J Cell Sci 110(Pt 17):2065–2077 Lampugnani MG et al (1995) The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, betacatenin, and alpha-catenin with vascular endothelial cadherin (VEcadherin). J Cell Biol 129:203–217 Lampugnani MG, Orsenigo F, Gagliani MC, Tacchetti C, Dejana E (2006) Vascular endothelial cadherin controls VEGFR-2 internalization and signaling from intracellular compartments. J Cell Biol 174:593–604 Lampugnani MG et al (2010) CCM1 regulates vascular-lumen organization by inducing endothelial polarity. J Cell Sci 123:1073–1080 Lampugnani MG et al (1992) A novel endothelial-specific membrane protein is a marker of cell-cell contacts. J Cell Biol 118:1511–1522 Lampugnani MG et al (2002) VE-cadherin regulates endothelial actin activating Rac and increasing membrane association of Tiam. Mol Biol Cell 13:1175–1189 Maddaluno L et al (2013) EndMT contributes to the onset and progression of cerebral cavernous malformations. Nature 498:492–496 Maher MT, Flozak AS, Stocker AM, Chenn A, Gottardi CJ (2009) Activity of the beta-catenin phosphodestruction complex at cellcell contacts is enhanced by cadherin-based adhesion. J Cell Biol 186:219–228 May C et al (2005) Identification of a transiently exposed VE-cadherin epitope that allows for specific targeting of an antibody to the tumor neovasculature. Blood 105:4337–4344
521 Mazzone M et al (2009) Heterozygous deficiency of PHD2 restores tumor oxygenation and inhibits metastasis via endothelial normalization. Cell 136:839–851 Miyoshi J, Takai Y (2007) Nectin and nectin-like molecules: biology and pathology. Am J Nephrol 27:590–604 Nanes BA et al (2012) p120-catenin binding masks an endocytic signal conserved in classical cadherins. J Cell Biol 199:365–380 Navarro P et al (1995) Catenin-dependent and -independent functions of vascular endothelial cadherin. J Biol Chem 270:30965–30972 Nawroth R et al (2002) VE-PTP and VE-cadherin ectodomains interact to facilitate regulation of phosphorylation and cell contacts. Embo J 21: 4885–4895 Nelson WJ (2008) Regulation of cell-cell adhesion by the cadherincatenin complex. Biochem Soc Trans 36:149–155 Nitta T et al (2003) Size-selective loosening of the blood-brain barrier in claudin-5-deficient mice. J Cell Biol 161:653–660 Oas RG et al (2010) p120-Catenin is required for mouse vascular development. Circ Res 106:941–951 Orsenigo F et al (2012) Phosphorylation of VE-cadherin is modulated by haemodynamic forces and contributes to the regulation of vascular permeability in vivo. Nat Commun 3:1208 Pannekoek WJ et al (2011) Epac1 and PDZ-GEF cooperate in Rap1 mediated endothelial junction control. Cell Signal 23: 2056–2064 Park HJ et al (2002) 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors interfere with angiogenesis by inhibiting the geranylgeranylation of RhoA. Circ Res 91:143–150 Pokutta S, Drees F, Yamada S, Nelson WJ, Weis WI (2008) Biochemical and structural analysis of alpha-catenin in cell-cell contacts. Biochem Soc Trans 36:141–147 Potter MD, Barbero S, Cheresh DA (2005) Tyrosine phosphorylation of VE-cadherin prevents binding of p120- and beta-catenin and maintains the cellular mesenchymal state. J Biol Chem 280:31906– 31912 Rampersad SN et al (2010) Cyclic AMP phosphodiesterase 4D (PDE4D) Tethers EPAC1 in a vascular endothelial cadherin (VE-Cad)-based signaling complex and controls cAMP-mediated vascular permeability. J Biol Chem 285:33614–33622 Rudini N et al (2008) VE-cadherin is a critical endothelial regulator of TGF-beta signalling. EMBO J 27:993–1004 Saharinen P et al (2008) Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell-cell and cell-matrix contacts. Nat Cell Biol 10:527–537 Sakurai A et al (2006) MAGI-1 is required for Rap1 activation upon cellcell contact and for enhancement of vascular endothelial cadherinmediated cell adhesion. Mol Biol Cell 17:966–976 Sanson B, White P, Vincent JP (1996) Uncoupling cadherin-based adhesion from wingless signalling in Drosophila. Nature 383:627–630 Simionescu M, Simionescu N, Palade GE (1976) Characteristic endothelial junctions in different segments of the vascular system. Thromb Res 8:247–256 Spindler V, Schlegel N, Waschke J (2010) Role of GTPases in control of microvascular permeability. Cardiovasc Res 87:243–253 Stahl S et al (2008) Novel CCM1, CCM2, and CCM3 mutations in patients with cerebral cavernous malformations: in-frame deletion in CCM2 prevents formation of a CCM1/CCM2/CCM3 protein complex. Hum Mutat 29:709–717 Stockton RA, Shenkar R, Awad IA, Ginsberg MH (2010) Cerebral cavernous malformations proteins inhibit Rho kinase to stabilize vascular integrity. J Exp Med 207:881–896 Suzuki S, Sano K, Tanihara H (1991) Diversity of the cadherin family: evidence for eight new cadherins in nervous tissue. Cell Regul 2: 261–270 Taddei A et al (2008) Endothelial adherens junctions control tight junctions by VE-cadherin-mediated upregulation of claudin-5. Nat Cell Biol 10:923–934
522 Thanou M, Gedroyc W (2013) MRI-Guided Focused Ultrasound as a New Method of Drug Delivery. J Drug Deliv 2013:616197 Thoreson MA et al (2000) Selective uncoupling of p120(ctn) from Ecadherin disrupts strong adhesion. J Cell Biol 148:189–202 Turowski P et al (2008) Phosphorylation of vascular endothelial cadherin controls lymphocyte emigration. J Cell Sci 121:29–37 Tyler RC, Peterson FC, Volkman BF (2010) Distal interactions within the par3-VE-cadherin complex. Biochemistry 49:951–957 Ukropec JA, Hollinger MK, Salva SM, Woolkalis MJ (2000) SHP2 association with VE-cadherin complexes in human endothelial cells is regulated by thrombin. J Biol Chem 275:5983–5986 van Wetering S et al (2002) Reactive oxygen species mediate Racinduced loss of cell-cell adhesion in primary human endothelial cells. J Cell Sci 115:1837–1846
Cell Tissue Res (2014) 355:515–522 Whitehead KJ et al (2009) The cerebral cavernous malformation signaling pathway promotes vascular integrity via Rho GTPases. Nat Med 15:177–184 Yamaoka-Tojo M et al (2006) IQGAP1 mediates VE-cadherin-based cell-cell contacts and VEGF signaling at adherence junctions linked to angiogenesis. Arterioscler Thromb Vasc Biol 26: 1991–1997 Yamashita K et al (2007) Fasudil, a Rho kinase (ROCK) inhibitor, protects against ischemic neuronal damage in vitro and in vivo by acting directly on neurons. Brain Res 1154:215–224 Zanetti A et al (2002) Vascular endothelial growth factor induces SHC association with vascular endothelial cadherin: a potential feedback mechanism to control vascular endothelial growth factor receptor-2 signaling. Arterioscler Thromb Vasc Biol 22:617–622
AT-A-GLANCE
VE-cadherin at a glance Luca Bravi & Elisabetta Dejana & Maria Grazia Lampugnani
Received: 10 December 2013 / Accepted: 4 February 2014 / Published online: 20 March 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Although being a monolayer the vascular endothelium controls fundamental vessel functions such as permeability, leukocyte extravasation and angiogenesis. The endothelial selective transmembrane constituent of adherens junctions, Vascular Endothelial- (VE-) cadherin plays a crucial role in the regulation of such activities. The signaling pathways controlled by VE-cadherin as well as the ones that regulate VEcadherin activity start to be elucidated. This delineates a complex network of molecular and functional interactions that can be altered in pathologies. Keywords VE-cadherin . Cell-to-cell junctions . Endothelium . Permeability . Angiogenesis
by the endothelium and which was localized at cell-to-cell junctions. We then identified this protein through affinity purification and N-terminal sequencing as cadherin 5 (Suzuki et al. 1991) or CD144. Because of this and due to its cell-type specificity, we proposed the name of vascular endothelial cadherin (VEcadherin; Lampugnani et al. 1992), which has been commonly used since then in more than 1,600 papers (PubMed data, November 2013). VE-cadherin is the major transmembrane component of endothelial adherens junctions and it is expressed in vivo in any type of vessels, arterial, venous and lymphatic (Fig. 1). Its expression represents an early step in the differentiation of the endothelial phenotype (Breier et al. 1996). Although VE-cadherin is not necessary for the acquisition of endothelial traits, its absence deeply impairs the physiological progression of the vascular program (Carmeliet et al. 1999).
A little bit of ‘history’ The ultra-structural architecture of endothelial cell-to-cell junctions was first described in the 1970s by Simionescu et al. (1976). They identified specialized regions of the endothelial plasma membrane greatly resembling the ones present in the epithelia: the tight, adherens and gap junctions. However, the definition of the molecular composition of such membrane domains in the endothelium had to wait until the 1990s. In particular, our group, in a screening of monoclonal antibodies directed to endothelial cells, identified an antigen selectively and ubiquitously expressed L. Bravi : E. Dejana (*) : M. G. Lampugnani (*) FIRC Institute of Molecular Oncology (IFOM) Fondazione, Via Adamello 16, 20139 Milan, Italy e-mail: [email protected] e-mail: [email protected] E. Dejana Department of Biosciences, School of Sciences, Milan University, Milan, Italy M. G. Lampugnani Mario Negri Institute for Pharmacological Research, Milan, Italy
VE-cadherin and its interactors: a never-ending story As a classical cadherin, VE-cadherin is tightly bound to cytoplasmic catenins: beta-catenin, plakoglobin and p120 (Nelson 2008), which have a strong impact on the activity of VEcadherin. This is the consequence of the role of catenins in both tethering VE-cadherin to the cortical actin cytoskeleton (as well as controlling cytoskeleton functions) and in connecting VE-cadherin with several different molecules for the constitution of signaling clusters (Fig. 2). In addition, some catenins are important regulators of transcription, although the relationship between junctional and nuclear catenins is still not defined in mammals. A truncated form of VE-cadherin [truncation of the carboxy-terminal amino acids (aa) 703-784] is unable to bind beta-catenin and plakoglobin (but still binding p120) and loses most of its regulatory properties in vitro, although it still localizes at junctions and retains adhesiveness comparable to wt (Navarro et al. 1995). In addition, it produces a lethal vascular phenotype in the homozygous mutant mouse embryo
516
Fig. 1 VE-cadherin is ubiquitously expressed in the vascular endothelium and localizes to cell-to-cell junctions. a An artery (A), a vein (V) and smaller vessels branching from them in the retina of a 7-dpn mouse pup stained with an antibody to VE-cadherin. Bar 100 um. b, c Magnifications of the boxed areas in (a). Arrowheads point to VE-cadherin positive endothelial junctions in artery (A) and vein (V), respectively. Bar 30 um. Immunofluorescence confocal microscopy was used to produce this figure. A similar result is observed staining any vascularized organs starting from the earliest stages of endothelial differentiation in the embryo (Breier et al 1996). Lymphatic vessels are also positive to VEcadherin (Baluk et al 2007). Junctional distribution and expression of VEcadherin can be modulated in response to physiological and pathological stimuli (Dejana et al 2009). See text for details on this issue
comparable to the null mutation (death at 9.0 dpc; Carmeliet et al. 1999). A mutant VE-cadherin unable to bind p120 after deletion of the p120 binding region (juxta-membrane domain aa 621702) or mutation to alanine of each of three core aa of the p120 binding domain (Thoreson et al. 2000) (aa-aa EMD, 652-654 of human and mouse VE-cadherin) is also regularly distributed to cell-to-cell junctions (Lampugnani et al. 2002; Corada et al., unpublished results). In addition, the EMD mutant shows a steady-state surface expression level comparable to the wt (Nanes et al. 2012). However, the p120-binding domaindeleted VE-cadherin did not recruit the Rac activator Tiam (Rac guanine nucleotide exchange factor, GEF), neither activated Rac, nor induced organization of actin stress fibers (Lampugnani et al. 2002). This behavior mimicked the effect of VE-cadherin null mutation (and the truncation of the beta-catenin-binding domain), suggesting abnormal signaling from junctional VEcadherin in the absence of associated p120. In agreement, homozygous mice with endothelial-specific null mutation of p120 die around 11.5 dpc with disorganized and hemorrhagic vessels (Oas et al. 2010). All in all, these data indicate that the correct association between VE-cadherin and catenins is crucial for the effective functioning of endothelial junctions. The first molecule reported to bind cadherins using betacatenin is alfa-catenin (Nelson 2008). Such association is functionally crucial for VE-cadherin. Indeed, a-catenin can tether VE-cadherin to the cortical acto-myosin cytoskeleton (see Pokutta et al. 2008 for the updated model of alfa-catenin and actin interaction and functional regulation) that can contract in
Cell Tissue Res (2014) 355:515–522 Fig. 2 Some of the molecules that have been demonstrated to form ab complex with VE-cadherin are reported. The regions of interaction with VE-cadherin, identified by amino acid (aa) position, are indicated, when known, along the linear structure of VE-cadherin (mouse sequence, upper part of the figure). Several molecules require beta-catenin to associate to VE-cadherin. It is reasonable to envisage that distinct complexes are formed, possibly defining specific membrane subdomains at the adherens junctions. Whether heterogeneity in the molecules associated to VEcadherin can define functional subdomains in different vessels and how physiological or pathological stimuli can module such complexes is still an unexplored, although extremely interesting, topic. Also, some tyrosine (two out of the nine present in the mouse sequence) and serine residues of VE-cadherin, observed phosphorylated in the whole organism, are highlighted (red P). Phosho-Ser665 (phosphorylated by PAK, for example in response to VEGF) is the binding site of beta2-arrestin that promotes clathrin-dependent internalization of VE-cadherin (Gavard and Gutkind 2006). Tyrosine residues 658 and 685 are phosphorylated in vivo in stable venous vessels (low-shear stress activates Src-mediated phosphorylation) (Orsenigo et al. 2012). Although not impairing the barrier properties of unstimulated vessels, the phosphorylation of these two residues facilitates VE-cadherin internalization in response to permeability increasing agents. If and how the association of other molecules can be modulated by the phosphorylation of tyrosine residues of VEcadherin is still debated (Potter et al 2005; Adam et al 2010). Among the extracellular interactors also VE-cadherin is indicated. Cadherins can form cis-homodimers (through ECI-ECIV domains) that interact in trans with similar dimers present on adjoining cells to support cell-to-cell recognition and adhesion. VE-PTP requires ECV domain of VE-cadherin for binding. Binding to VE-cadherin via: *CCM1 associated to betacatenin; ^beta-catenin and MAGI1; °704-784 aa region of VE-cadherin; #phosphodiesterase 4D cAMP specific (PDE4D); &co-immunoprecipitation with VE-cadherin, binding domain not determined
response to stimulation of Rho small GTPase that activates Rock to phosphorylate MLC2. This process results in pulling apart the cell-to-cell junctions of adjoining cells, with formation of gaps and increased permeability (Lampugnani 2012). In contrast, in most cases, activation of Rac GTPase strengthens endothelial junctions inhibiting Rho and stabilizing the actin cytoskeleton (Spindler et al. 2010) (for exceptions, see van Wetering et al. 2002 who reported increased permeability due to production of oxygen species by NADPH in response to Rac activation and Gavard and Gutkind 2006, see below). VEcadherin can contribute to Rac activation recruiting at junctions Tiam, a Rac GEF (Lampugnani et al. 2002). This aspect is discussed in further detail in the following paragraph. Besides these traditional companions, a full cohort of molecules has been shown to be associated to VE-cadherin. They can be both cytoplasmic and trasmembrane molecules and bind VEcadherin in the vast majority trough beta-catenin, although some examples of direct binding have been reported (see Fig. 2 for a list, although not comprehensive, of interactors and its legend for further details). In face of the huge number of interactors, often bound either to common regions of VE-cadherin or via identical molecular bridging, it is reasonable to hypothesize that distinct complexes exist, possibly determining subdomains at adherens junctions supporting specific functions. In addition, it is reasonable to imagine that specific complexes could be selectively present in distinct sites of the vascular network
Cell Tissue Res (2014) 355:515–522
contributing to the distinct function of the endothelium in different tissues and organs. These issues, although crucial for a comprehensive description of the molecular organization of endothelial adherens junctions and for envisaging specific therapeutic interventions, remain still almost unexplored at the present time. We have observed heterogeneity of VE-cadherin phosphorylation at the venous versus arterious site in vivo, with
517
VE-cadherin phosphorylated on tyrosine 658 and 685 only in veins and small veins (Orsenigo et al. 2012). This issue is discussed in detail in the following paragraph. We have reported up to now examples of association to the cytoplasmic domain of VE-cadherin, often mediated by betacatenin, (see Fig. 2), for which a vast literature exists. The endothelial-specific phosphatase VE-PTP, on the contrary,
518
directly associates to the extracellular domain of VE-cadherin through its extracellular domain. When associated to VEcadherin at junctions, it cooperates for the maintenance of stable, non-permeable junctions (Nawroth et al. 2002). Besides VE-cadherin, VE-PTP can dephosphorylate (Fukuhara et al. 2008; Saharinen et al. 2008) Tie2, plakoglobin and junction associated-VEGFR2 (Hayashi et al. 2013).
VE-cadherin as a functional hub: adhesion and signaling Homotypic cell recognition and adhesion is a primary function of VE-cadherin. VE-cadherin can be recruited in early junctions by nectins previously engaged in adhesive interactions at junctions (Miyoshi and Takai 2007). Control of permeability is the significant consequence of the engagement of VE-cadherin molecules in cis- and trans-homophilic binding (Dejana et al. 2009). This function can be both the direct result of the adhesive recognition between VE-cadherin molecules and also its indirect outcome. Indeed, VE-cadherin clustered to cell-to-cell junctions activates a signaling pathway that upregulates the transcription and the deposition at junctions of claudin-5, the major transmembrane component of the tight junctions (Taddei et al. 2008), which seals the intercellular space to the passage of very small solutes (less than 800 D; Nitta et al. 2003). The mechanism of transcriptional regulation of claudin-5 represents an example of the signaling activity of VE-cadherin that has been studied in detail in our laboratory. When clustered in stable junctions, VE-cadherin recruits PI3K, through beta-catenin. This kinase locally activates Akt that phosphorylates FoxO1, a major repressor of claudin-5 transcription in association to beta-catenin. The phosphorylated FoxO1 is excluded from the nucleus (phosphorylation on Thr24 promotes its exclusion from the nucleus and phosphorylation on Ser256 inhibits its binding to DNA), thus relieving the inhibition of claudin-5 transcription (Taddei et al. 2008). VE-cadherin, when associated in stable junctions, besides contributing to the barrier function of the endothelium, enhances protection from apoptosis and moderates proliferation, two primary features of endothelial cells in adult stable vessels. (Dejana et al. 2009). To this aim, the association between VE-cadherin and two central managers of endothelial behavior, such as VEGFR2 and TGFbetaR2 (as well as Alk1 and Alk5), is crucial. Also, in this case, beta-catenin is required for the formation of the complex between VE-cadherin and these receptors (Lampugnani et al. 2006; Rudini et al. 2008). Further molecular details are still missing. Recently, FGFR1 has also been found associated to VE-cadherin, which reduces its phosphorylation associating the phosphate Dep1 in a multiprotein complex (Giampietro et al. 2012). In contrast, in cells with VE-cadherin engaged in either unstable or earlyestablished junctions, as during angiogenesis, the mitogenic
Cell Tissue Res (2014) 355:515–522
and migratory responses to VEGFR2 and TGFRbeta2 are enhanced, while their anti-apoptotic signaling is switched off. Both activities are crucial for the growth and remodeling of a new vascular network. The examples briefly discussed above suggest that the signaling activity of VE-cadherin is modulated by the quality of its clustering at cell-to-cell junctions, which in molecular terms regulates and allows specific interactions among molecules associated to VE-cadherin. The association of adherens junctions to the cortical actin cytoskeleton is critical to determine the stability of junctions (Nelson 2008), as introduced in the previous paragraph. The small GTPase Rap1, associated to VE-cadherin through CCM1 bound to beta-catenin (Glading et al. 2007), is critical to maintain VE-cadherin in stable junctions with the cooperation of the other members of the CCM complex, CCM2 and CCM3 (Stahl et al. 2008). A possible mechanism for the stabilization of VE-cadherin at cell-to-cell junctions would involve inhibition of acto-myosin contraction by Rac, activated by Rap1. On its side, Rap1 would be activated at adherens junctions by the association to VE-cadherin of PDZ-GEF1 and Epac, both Rap1 guanine-nucleotide-exchange factors (GEFs) (Pannekoek et al. 2011). VE-cadherin, in association to CCMs molecules, controls another central function of the endothelium that is the apical– basal polarity. The distribution to junctions of two components of the polarity complex, Par3 and aPKz, depends on VEcadherin (Lampugnani et al. 2010). Par3 directly associates to VE-cadherin (classII PDZ binding motif in the five terminal aa, Vestweber) and CCM1 is required for the activation of aPKz (Lampugnani et al. 2010). Polarization of the endothelium is fundamental in adhesion and trafficking of the immune cells as well as in the transport of molecules to and from the tissues. Another important consequence of the engagement of VEcadherin in stable adherens junctions is the regulation of the level of nuclear beta-catenin. Although the mechanism of this process is not yet settled, it is consistent with the observation that, when VE-cadherin participates in stable junctions, betacatenin is excluded from the nucleus, while when junctions are weak, beta-catenin concentrates in the nucleus where it is transcriptionally active in association to Tcf transcription factors (Clevers and Nusse 2012). It is not yet clear whether the beta-catenin that accumulates in the nucleus derives from junctions after detachment from VE-cadherin. A report in epithelial cells using photo-activable GFP-tagged betacatenin would indicate that beta-catenin could dissociate from E-cadherin and re-localize to the nucleus (Kam and Quaranta 2009). Beta-catenin can also be found at junctions not associated to E-cadherin (Maher et al. 2009). It is not known whether this beta-catenin was previously associated to Ecadherin, or whether it represents a cadherin-independent pool, as observed in Drosophila (Sanson et al. 1996) and
Cell Tissue Res (2014) 355:515–522
confirmed in C. elegans (Korswagen et al. 2000). In addition, it has been observed in epithelial cells that components of the machinery that drives degradation of beta-catenin (phosphodestruction complex: namely axin, APC2 and GSK3beta) are associated to cell-to-cell contacts and cadherins can promote N-terminal phosphorylation of beta-catenin, targeting it to proteosomal degradation and limiting the cytoplasmic pool of beta-catenin and as a consequence its localization to the nucleus. Therefore, the stability of junctions would regulate the activity of such machinery and the degradation of betacatenin (Maher et al. 2009). The consequence would be that the transcriptional activity of beta-catenin could be regulated by the state of junctions and be appropriate to it. It remains to be tested whether a similar mechanism can also operate in endothelial cells. Distribution of VE-cadherin to junctions can be regulated both in acute and chronic ways. Acute and generally reversible, disorganization of junctional VE-cadherin occurs in response to permeability increasing agents and short-term stimulation with growth factors. The mechanism involves stimulation of Rho GTPase, activation of the downstream effector ROCK that results in contraction of peripheral acto-myosin bundles and opening of intercellular gaps, as already introduced above. Chronic and non-reversible dissociation of VE-cadherin from junctions can be observed in certain pathologies and can be the consequence of the constitutive presence of angiogenic and inflammatory stimuli, of mutation in regulators of the stability of junctional VE-cadherin and of downregulation of the transcription and expression level of VE-cadherin. The permanence of clustered VE-cadherin at cell-to-cell junctions can also be modulated by post-transcriptional modification of the cytoplasmic domain of VE-cadherin. In particular, phosphorylation on tyrosine 658 and 685 (Y658 and Y685) can modulate the rate of internalization of VE-cadherin through a clathrin-dependent mechanism and can contribute to modify the stability, composition and signaling activity of adherens junctions. These post-transcriptional modifications correlate with a more prompt internalization and loss of barrier function in response to permeability increasing agents such as bradykinin and histamine. The composition of the molecular complexes formed by VE-cadherin phosphorylated on those tyrosine residues remains to be analyzed. This takes us to the ‘vexata quaestio’ of how post-transcriptional modification of VEcadherin, in particular tyrosine phosphorylation, can modulate the activity of VE-cadherin. Eight tyrosine residues (in the human form and nine in the murine one) could be potential targets of tyrosine kinases (Potter et al. 2005: fig. 2). Although, in the current model, tyrosine phosphorylation of VE-cadherin is considered to characterize weak and permeable junctions (Turowski et al. 2008; Potter et al. 2005), it
519
may, however, not be sufficient alone and other signals could cooperate to the weakening of the barrier (Adam et al. 2010; Orsenigo et al. 2012; see previous comment). As far as the function of other residues phosphorylated in vivo in the organism are concerned, the residue S665 of VE-cadherin is the docking site for beta-2 arrestin that promotes clathrin-dependent internalization of VE-cadherin (Gavard and Gutkind 2006). S665 has been reported to be phosphorylated in VE-cadherin of mouse skin vessels (Gavard et al. 2008), although the regulation of such phosphorylation has been prevalently studied in in vitro-cultured cells. Ser665 is phosphorylated by PAK, activated in response to VEGF, which sequentially activates Src, Vav2, a Rac GEF and Rac itself to trigger PAK . The weakening of endothelial cell-to-cell junctions ensuing such signaling sequence can be blocked by Angiopoietin-1 that sequesters Src through mDia. While Src has also been demonstrated to play an important, although not direct, role in the molecular mechanism of VEcadherin phosphorylation on tyrosine residues (see also Orsenigo et al. 2012), the phosphatases that regulate the phosphorylation of VE-cadherin are still poorly defined. Among the phosphatases associated to VE-cadherin, VEPTP can use VE-cadherin as a substrate (Saharinen et al. 2008; Fukuhara et al. 2008) while SHP2 does not (Ukropec et al. 2000). However, it is not known which phospho-tyrosine residues of VE-cadherin may be direct substrates of VE-PTP. To conclude, an increased level of junctional VEcadherin means more stable and less permeable endothelium. This has been observed in transgenic mice with haplo-insufficiency of the oxygen sensor PDH2 (Mazzone et al. 2009). In this condition, VE-cadherin transcript and protein increase and reshape and strengthen the endothelial lining to increase the control of permeability and cell extravasation. Tumor cell invasion, intravasation and metastasis were inhibited in these mutant mice (Mazzone et al. 2009).
VE-cadherin and pathologies While mutations in the VE-cadherin gene have not yet been reported in human patients, modifications in the VE-cadherin ‘situation’ can be observed in human pathologies. What has been reported up to now is that the distribution of VE-cadherin to adherens junctions, the transcription of the VE-cadherin gene and the stability of the protein can be modulated. While molecular mechanisms have been described to regulate each of these steps in experimental systems, as described in the previous paragraph, it still remains poorly defined how VEcadherin is de-regulated in pathologies. A typical example of disorganized and weak endothelial junctions is observed in tumor vessels, in which the local microenviroment is rich in angiogenic factors (typically VEGF produced in response to the hypoxic tumor
520
environment; Carmeliet and Jain 2011) and permeability increasing agents (such as inflammatory cytokines released either by tumor cells or infiltrating cells of the immune system), which determine dismantling of adherens junctions and permeable endothelial barrier (Baluk et al. 2005). Interestingly, in tumors vessels, VE-cadherin can expose a distinct epitope that can be specifically recognized by a monoclonal antibody. Such an antibody developed at ImClone, NY, US, besides specifically targeting tumor vessels could also induce inhibition of tumor growth and regression in mouse models of syngenic tumors or xenographs (Controllare) (May et al. 2005). Another human disease characterized by disorganized endothelial junctions is the cerebral cavernous malformation (CCM), a vascular pathology that presents vascular malformations typically localized at the venous side of the central nervous system vessels (Boulday et al. 2011; Maddaluno et al. 2013). The products of the three genes mutated in CCM pathology, CCM1, 2 and 3, form a complex (Stahl et al. 2008) that can associate to VE-cadherin through the direct interaction of β-catenin and CCM1 (Glading et al. 2007). CCM1 also binds Rap1 that, as mentioned before, exerts a fundamental action in stabilizing adherens junctions in endothelial cells (Kooistra et al. 2005). As a consequence of the ablation of any of the CCM gene, adherens junctions are disorganized and the transcription of the VE-cadherin gene can also be reduced. Such transcriptional regulation sets in the process of endothelial–mesenchymal transition that follows the loss of CCM proteins (Maddaluno et al. 2013) and involves a typical cadherin switch with increased expression of N-cadherin (Taddei et al. 2008; Giampietro et al. 2012) and mesenchymal markers. This process of de-differentiation represents a crucial step in the establishment of the vascular lesion and is determined by the de-regulated signaling of TGF-beta and beta-catenin pathways (Maddaluno et al. 2013; Bravi, unpublished).
Perspectives: from research to therapeutic application Several issues in the basic regulation of VE-cadherin organization and function remain to be defined, as has been pointed out in the text. Regarding the translational aspect, a series of crucial questions should be answered: Are there perspectives of using VEcadherin for a targeted vascular therapy and how to open and close adherens junctions and possibly reversibly, targeting VE-cadherin? Some drugs have been shown to stabilize adherens junctions and reduce vascular permeability both in wild-type and pathological (CCM1 and 2 heterozygous) animal models, blocking Rock fasudil that competes for ATP binding (Yamashita et al. 2007; Stockton et al. 2010) and membrane-
Cell Tissue Res (2014) 355:515–522
association of Rho (statins that inhibit isoprenylation of Rho GTPase; Collisson et al. 2002; Park et al. 2002; Whitehead et al. 2009). However, drugs able to weaken junctions targeting VE-cadherin have not yet been reported to our knowledge. Opening endothelial junctions could be particularly useful in ischemic pathologies of many organs or in pathologies of the central nervous system where the tissue is protected by the blood–brain barrier. A possible direct approach could be to down-regulate VE-cadherin transcription using either siRNA or shRNA. This method, which is quite effective in in vitro systems of cells in culture, would require precise targeting, as a generalized down-regulation of VEcadherin expression could have dangerous consequences, as described above. Local delivery would require the identification of specific epitopes expressed by endothelial cells in the particular environment that needs to be targeted. This implies definition of the local phenotype of endothelial cells and the development of tools for specific delivery (for example, nanoparticles) (Davis et al. 2010; Thanou and Gedroyc 2013). Acknowledgements This work is supported by grants from the Fondation Leducq Transatlantic Network of Excellence, Associazione Italiana per la Ricerca sul Cancro (AIRC) and ‘Special ProgramMolecular Clinical Oncology 5X1,000 to AGIMM (AIRC-Gruppo Italiano Malattie Mieloproliferative), the European Community (ENDOSTEM-HEALTH2009-241440; JUSTBRAIN-HEALTH-2009-241861; ITN Vessels), the European Research Council (ERC) and CARIPLO Foundation.
References Adam AP, Sharenko AL, Pumiglia K, Vincent PA (2010) Src-induced tyrosine phosphorylation of VE-cadherin is not sufficient to decrease barrier function of endothelial monolayers. J Biol Chem 285:7045–7055 Baluk P et al (2007) Functionally specialized junctions between endothelial cells of lymphatic vessels. J Exp Med 204:2349–2362 Baluk P, Hashizume H, McDonald DM (2005) Cellular abnormalities of blood vessels as targets in cancer. Curr Opin Genet Dev 15:102–111 Baumeister U et al (2005) Association of Csk to VE-cadherin and inhibition of cell proliferation. Embo J 24:1686–1695 Boulday G et al (2011) Developmental timing of CCM2 loss influences cerebral cavernous malformations in mice. J Exp Med 208:1835–1847 Breier G et al (1996) Molecular cloning and expression of murine vascular endothelial-cadherin in early stage development of cardiovascular system. Blood 87:630–641 Carmeliet P, Jain RK (2011) Principles and mechanisms of vessel normalization for cancer and other angiogenic diseases. Nat Rev Drug Discov 10:417–427 Carmeliet P et al (1999) Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGF-mediated endothelial survival and angiogenesis. Cell 98:147–157 Clevers H, Nusse R (2012) Wnt/beta-catenin signaling and disease. Cell 149:1192–1205 Collisson EA, Carranza DC, Chen IY, Kolodney MS (2002) Isoprenylation is necessary for the full invasive potential of RhoA overexpression in human melanoma cells. J Invest Dermatol 119: 1172–1176
Cell Tissue Res (2014) 355:515–522 Davis ME et al (2010) Evidence of RNAi in humans from systemically administered siRNAvia targeted nanoparticles. Nature 464:1067–1070 Dejana E, Tournier-Lasserve E, Weinstein BM (2009) The control of vascular integrity by endothelial cell junctions: molecular basis and pathological implications. Dev Cell 16:209–221 Fukuhara S et al (2008) Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1. Nat Cell Biol 10:513–526 Gavard J, Gutkind JS (2006) VEGF controls endothelial-cell permeability by promoting the beta-arrestin-dependent endocytosis of VEcadherin. Nat Cell Biol 8:1223–1234 Gavard J, Patel V, GutkindJS (2008) Angiopoietin-1 prevents VEGFinduced endothelial permeability by sequestering Src through mDia. Dev Cell 14:25–36 Giampietro C et al (2012) Overlapping and divergent signaling pathways of N-cadherin and VE-cadherin in endothelial cells. Blood 119: 2159–2170 Glading A, Han J, Stockton RA, Ginsberg MH (2007) KRIT-1/CCM1 is a Rap1 effector that regulates endothelial cell cell junctions. J Cell Biol 179:247–254 Hayashi M et al (2013) VE-PTP regulates VEGFR2 activity in stalk cells to establish endothelial cell polarity and lumen formation. Nat Commun 4:1672 Iden S et al (2006) A distinct PAR complex associates physically with VEcadherin in vertebrate endothelial cells. EMBO Rep 7:1239–1246 Kam Y, Quaranta V (2009) Cadherin-bound beta-catenin feeds into the Wnt pathway upon adherens junctions dissociation: evidence for an intersection between beta-catenin pools. PLoS ONE 4:e4580 Kooistra MR, Corada M, Dejana E, Bos JL (2005) Epac1 regulates integrity of endothelial cell junctions through VE-cadherin. FEBS Lett 579:4966–4972 Korswagen HC, Herman MA, Clevers HC (2000) Distinct beta-catenins mediate adhesion and signalling functions in C. elegans. Nature 406: 527–532 Kronstein R et al (2012) Caveolin-1 opens endothelial cell junctions by targeting catenins. Cardiovasc Res 93:130–140 Lampugnani MG (2012) Endothelial cell-to-cell junctions: adhesion and signaling in physiology and pathology. Cold Spring Harb Perspect Med 2 Lampugnani MG et al (1997) Cell confluence regulates tyrosine phosphorylation of adherens junction components in endothelial cells. J Cell Sci 110(Pt 17):2065–2077 Lampugnani MG et al (1995) The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, betacatenin, and alpha-catenin with vascular endothelial cadherin (VEcadherin). J Cell Biol 129:203–217 Lampugnani MG, Orsenigo F, Gagliani MC, Tacchetti C, Dejana E (2006) Vascular endothelial cadherin controls VEGFR-2 internalization and signaling from intracellular compartments. J Cell Biol 174:593–604 Lampugnani MG et al (2010) CCM1 regulates vascular-lumen organization by inducing endothelial polarity. J Cell Sci 123:1073–1080 Lampugnani MG et al (1992) A novel endothelial-specific membrane protein is a marker of cell-cell contacts. J Cell Biol 118:1511–1522 Lampugnani MG et al (2002) VE-cadherin regulates endothelial actin activating Rac and increasing membrane association of Tiam. Mol Biol Cell 13:1175–1189 Maddaluno L et al (2013) EndMT contributes to the onset and progression of cerebral cavernous malformations. Nature 498:492–496 Maher MT, Flozak AS, Stocker AM, Chenn A, Gottardi CJ (2009) Activity of the beta-catenin phosphodestruction complex at cellcell contacts is enhanced by cadherin-based adhesion. J Cell Biol 186:219–228 May C et al (2005) Identification of a transiently exposed VE-cadherin epitope that allows for specific targeting of an antibody to the tumor neovasculature. Blood 105:4337–4344
521 Mazzone M et al (2009) Heterozygous deficiency of PHD2 restores tumor oxygenation and inhibits metastasis via endothelial normalization. Cell 136:839–851 Miyoshi J, Takai Y (2007) Nectin and nectin-like molecules: biology and pathology. Am J Nephrol 27:590–604 Nanes BA et al (2012) p120-catenin binding masks an endocytic signal conserved in classical cadherins. J Cell Biol 199:365–380 Navarro P et al (1995) Catenin-dependent and -independent functions of vascular endothelial cadherin. J Biol Chem 270:30965–30972 Nawroth R et al (2002) VE-PTP and VE-cadherin ectodomains interact to facilitate regulation of phosphorylation and cell contacts. Embo J 21: 4885–4895 Nelson WJ (2008) Regulation of cell-cell adhesion by the cadherincatenin complex. Biochem Soc Trans 36:149–155 Nitta T et al (2003) Size-selective loosening of the blood-brain barrier in claudin-5-deficient mice. J Cell Biol 161:653–660 Oas RG et al (2010) p120-Catenin is required for mouse vascular development. Circ Res 106:941–951 Orsenigo F et al (2012) Phosphorylation of VE-cadherin is modulated by haemodynamic forces and contributes to the regulation of vascular permeability in vivo. Nat Commun 3:1208 Pannekoek WJ et al (2011) Epac1 and PDZ-GEF cooperate in Rap1 mediated endothelial junction control. Cell Signal 23: 2056–2064 Park HJ et al (2002) 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors interfere with angiogenesis by inhibiting the geranylgeranylation of RhoA. Circ Res 91:143–150 Pokutta S, Drees F, Yamada S, Nelson WJ, Weis WI (2008) Biochemical and structural analysis of alpha-catenin in cell-cell contacts. Biochem Soc Trans 36:141–147 Potter MD, Barbero S, Cheresh DA (2005) Tyrosine phosphorylation of VE-cadherin prevents binding of p120- and beta-catenin and maintains the cellular mesenchymal state. J Biol Chem 280:31906– 31912 Rampersad SN et al (2010) Cyclic AMP phosphodiesterase 4D (PDE4D) Tethers EPAC1 in a vascular endothelial cadherin (VE-Cad)-based signaling complex and controls cAMP-mediated vascular permeability. J Biol Chem 285:33614–33622 Rudini N et al (2008) VE-cadherin is a critical endothelial regulator of TGF-beta signalling. EMBO J 27:993–1004 Saharinen P et al (2008) Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell-cell and cell-matrix contacts. Nat Cell Biol 10:527–537 Sakurai A et al (2006) MAGI-1 is required for Rap1 activation upon cellcell contact and for enhancement of vascular endothelial cadherinmediated cell adhesion. Mol Biol Cell 17:966–976 Sanson B, White P, Vincent JP (1996) Uncoupling cadherin-based adhesion from wingless signalling in Drosophila. Nature 383:627–630 Simionescu M, Simionescu N, Palade GE (1976) Characteristic endothelial junctions in different segments of the vascular system. Thromb Res 8:247–256 Spindler V, Schlegel N, Waschke J (2010) Role of GTPases in control of microvascular permeability. Cardiovasc Res 87:243–253 Stahl S et al (2008) Novel CCM1, CCM2, and CCM3 mutations in patients with cerebral cavernous malformations: in-frame deletion in CCM2 prevents formation of a CCM1/CCM2/CCM3 protein complex. Hum Mutat 29:709–717 Stockton RA, Shenkar R, Awad IA, Ginsberg MH (2010) Cerebral cavernous malformations proteins inhibit Rho kinase to stabilize vascular integrity. J Exp Med 207:881–896 Suzuki S, Sano K, Tanihara H (1991) Diversity of the cadherin family: evidence for eight new cadherins in nervous tissue. Cell Regul 2: 261–270 Taddei A et al (2008) Endothelial adherens junctions control tight junctions by VE-cadherin-mediated upregulation of claudin-5. Nat Cell Biol 10:923–934
522 Thanou M, Gedroyc W (2013) MRI-Guided Focused Ultrasound as a New Method of Drug Delivery. J Drug Deliv 2013:616197 Thoreson MA et al (2000) Selective uncoupling of p120(ctn) from Ecadherin disrupts strong adhesion. J Cell Biol 148:189–202 Turowski P et al (2008) Phosphorylation of vascular endothelial cadherin controls lymphocyte emigration. J Cell Sci 121:29–37 Tyler RC, Peterson FC, Volkman BF (2010) Distal interactions within the par3-VE-cadherin complex. Biochemistry 49:951–957 Ukropec JA, Hollinger MK, Salva SM, Woolkalis MJ (2000) SHP2 association with VE-cadherin complexes in human endothelial cells is regulated by thrombin. J Biol Chem 275:5983–5986 van Wetering S et al (2002) Reactive oxygen species mediate Racinduced loss of cell-cell adhesion in primary human endothelial cells. J Cell Sci 115:1837–1846
Cell Tissue Res (2014) 355:515–522 Whitehead KJ et al (2009) The cerebral cavernous malformation signaling pathway promotes vascular integrity via Rho GTPases. Nat Med 15:177–184 Yamaoka-Tojo M et al (2006) IQGAP1 mediates VE-cadherin-based cell-cell contacts and VEGF signaling at adherence junctions linked to angiogenesis. Arterioscler Thromb Vasc Biol 26: 1991–1997 Yamashita K et al (2007) Fasudil, a Rho kinase (ROCK) inhibitor, protects against ischemic neuronal damage in vitro and in vivo by acting directly on neurons. Brain Res 1154:215–224 Zanetti A et al (2002) Vascular endothelial growth factor induces SHC association with vascular endothelial cadherin: a potential feedback mechanism to control vascular endothelial growth factor receptor-2 signaling. Arterioscler Thromb Vasc Biol 22:617–622
