This article is a US Government work and, as such, is in the public domain in the United States of America.
Patterns & Phenotypes
Tissue-specific venous expression of the EPH family receptor EphB1 in the skin vasculature
Version of Record online: 18 JUN 2013
Copyright © 2013 Wiley Periodicals, Inc.
Volume 242, Issue 8, pages 976–988, August 2013
How to Cite
Li, W. and Mukouyama, Y.-s. (2013), Tissue-specific venous expression of the EPH family receptor EphB1 in the skin vasculature. Dev. Dyn., 242: 976–988. doi: 10.1002/dvdy.23985
- Issue online: 17 JUL 2013
- Version of Record online: 18 JUN 2013
- Accepted manuscript online: 7 MAY 2013 12:00AM EST
- Manuscript Accepted: 24 APR 2013
- Manuscript Revised: 23 APR 2013
- Manuscript Received: 20 NOV 2012
- Intramural Research Program of the National Heart, Lung, and Blood Institute, National Institutes of Health. Grant Number: HL005702-07
- venous endothelial cell marker;
- venous development;
Background: The major arteries and veins are formed early during development. The molecular tools to identify arterial and venous endothelial cells improve our understanding of arterial–venous differentiation and branching morphogenesis. Compared with arterial differentiation, relatively little is known about what controls venous development, due to lack of definitive molecular markers for venous endothelial cells. Results: Here we report that the antibody against EphB1, an EphB class receptor, makes it possible to establish a reliable whole-mount immunohistochemical analysis of venous identity with greater resolution than previously possible in embryonic and adult skin vasculature models. EphB1 expression is restricted to the entire venous vasculature throughout embryonic development to adulthood, whereas the previously established venous marker EphB4 is also detectable in lymphatic vasculature. This venous-restricted expression of EphB1 is established after the vascular remodeling of the primary capillary plexus has occurred. Compared with its venous-specific expression in the skin, however, EphB1 is not restricted to the venous vasculature in yolk sac, trunk and lung. Conclusions: These studies introduce EphB1 as a new venous-restricted marker in a tissue-specific and time-dependent manner. Developmental Dynamics 242:976–988, 2013. © 2013 Wiley Periodicals, Inc.
Blood vessels form a highly branched hierarchical vascular tree composed of arteries and veins. The primitive vascular network (capillary plexus) is remodeled to form a well-defined, hierarchical branching network in peripheral tissues (Risau, 1997). In this process, vascular endothelial growth factor (VEGF) and Notch signaling directs arterial differentiation (Mukouyama et al., 2002; Visconti et al., 2002; Duarte et al., 2004; Gale et al., 2004), whereas chicken ovalbumin upstream-transcription factor II (COUP-TFII) maintains venous identity by inhibiting Notch activation (You et al., 2005). Our understanding of a genetic program specifying arterial identity has been advanced by the identification and functional analysis of signaling molecules including Neuropilin-1 (Nrp1), Dll4, Notch1, Notch4, Hey1, Hey2, ephrinB2, Cx37, and Cx40 (Rocha and Adams, 2009; Chong et al., 2011). However, relatively little is known about what controls venous development, due to lack of definitive molecular markers for venous endothelial cells.
EphB4 is the first gene to be described that is expressed in a venous-specific manner from early in embryogenesis into adulthood (Wang et al., 1998), and which is functionally essential for angiogenesis as well (Gerety et al., 1999). EphB4taulacZ knock-in mice were used to show EphB4 expression in veins (Gerety et al., 1999). Although EphB4 is widely accepted as a venous marker, weak expression is also detectable in some arterial and lymphatic endothelial cells during development (Shin et al., 2001; Shin and Anderson, 2005; Srinivasan et al., 2007). COUP-TFII (You et al., 2005), Neuropilin-2 (Nrp2), (Yuan et al., 2002), and VEGFR3 (Kaipainen et al., 1995) are more abundantly expressed in veins than arteries, but these genes are not restricted to veins throughout development. Thus, the study of venous branching morphogenesis has been hindered by the confounding specificity of these markers' expression.
Here we show that EphB1, an EphB class receptor, is specifically expressed by venous endothelial cells, but not arterial and lymphatic endothelial cells, in embryonic and adult skin vasculature. This expression provides a directly observable vascular network with an anatomically recognizable pattern. The antibody against EphB1 makes it possible to establish a reliable whole-mount immunohistochemical analysis of venous identity with greater resolution than previously possible. Interestingly, our extensive comparison of EphB1 and EphB4 expression in embryonic skins reveals that both EphB1 and EphB4 are uniformly expressed in the primitive capillary network at embryonic day (E) 13.5, a stage when no or fewer arterial markers such as ephrinB2 and Nrp1 are expressed. By embryonic day (E) 14.5, vascular remodeling occurs and both markers are expressed in remodeled veins. By E15.5, EphB1 expression is restricted to the entire venous vasculature, whereas EphB4 is also expressed in lymphatic vasculature. The venous-specific expression of EphB1 persists into adulthood in the skin vasculature. Compared with the venous-specific expression in the skin, however, EphB1 is not restricted to venous vasculature in yolk sac, trunk, and lung. These data indeed suggest EphB1 is a new venous-specific marker that functions in a tissue-specific and time-dependent manner.
EphB1 Is Expressed in Veins in the Developing Skin Vasculature
Previous studies demonstrated that ephrinB2, a transmembrane ligand, is expressed by arteries but not veins (Wang et al., 1998), whereas one of its receptors, the tyrosine kinase EphB4, is more abundantly expressed by veins than by arteries (Wang et al., 1998; Gerety et al., 1999). Whole-mount staining of EphB4taulacZ/+ heterozygous limb skin with antibodies to β-galactosidase (β-gal) revealed that at E15.5, EphB4taulacZ expression was mainly detected in veins (Fig. 1A–A″,C–C″,D–D″, open arrows) but weak EphB4taulacZ expression was also detectable in lymphatic-like vessels which do not form a smooth vessel structure and express relatively weak levels of the pan-endothelial cell marker PECAM-1 (Fig. 1A–A″,C–C″,D–D″, open arrowheads; see Fig. 2A,D,G). This observation prompted us to explore genes specifically expressed in venous endothelial cells in the developing skin vasculature. We initially examined the expression patterns of those EphB receptors (EphB1, EphB2, EphB3) reported to bind the arterial marker ephrinB2 (Flanagan and Vanderhaeghen, 1998), using whole-mount immunohistochemical analysis of limb skins with antibodies to those EphB receptors. Among them, only EphB1 antibody yielded venous-specific staining for the limb skin vasculature (Fig. 1A–A″,B–B″,C–C″, arrowheads). EphB1 expression appeared to overlap with EphB4taulacZ expression in large-diameter veins, whereas EphB1 expression was much more clearly detectable in branched middle-diameter veins and venous capillaries than EphB4taulacZ expression (Fig. 1A–A″,C–C″). Importantly, no weak EphB1 expression was detectable in EphB4taulacZ-expressing lymphatic-like vessels (Fig. 1A–A″,C–C″).
EphB1 Is Expressed by Veins but Not Lymphatics and Arteries in the Developing Skin Vasculature
These observations led us to precisely characterize the expression of EphB1 in combination with the lymphatic endothelial cell markers LYVE-1 and Neuropilin-2 (Nrp2). As mentioned above, EphB4taulacZ expression was also detectable in LYVE-1+ lymphatic vasculature in EphB4taulacZ/+ heterozygous limb skin (Fig. 2A,D,G, open arrowheads). However, no significant expression of EphB1 was detectable in the LYVE-1+and Nrp2+ lymphatic vasculature in the limb skin at E15.5 (Fig. 2B,C,E,F,H,I, arrowheads vs. open arrowheads).
We next sought to examine the expression of EphB1 in arterial vasculature, by performing whole-mount limb skin staining with the arterial endothelial cell markers ephrinB2 and Neuropilin-1 (Nrp1). From late embryonic day (E) 13.5 onward, in conjunction with vascular remodeling of the primary capillary plexus, ephrinB2taulacZ expression was detectable in branched vessels in ephrinB2taulacZ/+ heterozygous limb skin (Fig. 3A,B,I,J, open arrowheads). At E14.0, some ephrinB2taulacZ-expressing honeycomb-like vessels appear to be in the process of vascular remodeling and express EphB1 (Fig. 3B,F,J, open arrows). By E15.5, after vascular remodeling was complete, no significant expression of EphB1 was observed in the ephrinB2taulacZ-expressing arterial vasculature (Fig. 3C,G,K, arrowheads vs. open arrowheads). We confirmed that no significant EphB1 expression was detectable in the Nrp1+ arterial vasculature (Fig. 3D,H,L, arrowheads vs. open arrowheads). Taken together, these data clearly demonstrate that EphB1 expression is restricted to the venous vasculature but not the lymphatic and arterial vasculature. Given that none of the commercial anti-EphB4 antibodies was adequately specific for whole-mount staining of limb skin vasculature and that EphB4taulacZ expression is not absolutely restricted to venous endothelial cells, EphB1 represents a venous-restricted marker that allows us to analyze venous identity with a significant resolution.
Venous Differentiation in the Developing Skin Vasculature
We next examined venous fate establishment in the developing skin vasculature. By E13.5, a primary capillary plexus is established in the skin. Subsequently, the primitive vascular system undergoes intensive vascular remodeling triggered by blood flow and develops into a hierarchical vascular branching network including arteries and veins. Our previous whole-mount immunohistochemical studies demonstrated that no vascular expression of arterial markers such as ephrinB2 and Nrp1 is detectable in the primitive vasculature at E13.5 (Mukouyama et al., 2002). Surprisingly, expression of EphB1 and EphB4taulacZ was detectable throughout the primary capillary plexus at E13.5 (Fig. 4A,D,G,J). We further confirmed that EphB1 expression was detectable at earlier stage (Fig. 4D inset, E13.0). Given the prominent expression of EphB1 and EphB4taulacZ in the primary capillary plexus, we cannot presently discern whether this expression may be considered an indicator of immature vessels, or whether this expression pattern correlates with prespecification of venous identity.
By E14.5, vascular remodeling had occurred and arterial markers were detectable in some vessels (Mukouyama et al., 2002). Expression of EphB1 and EphB4taulacZ progressively decreased in the remodeled arterial vessels (Fig. 4B,E,K, PECAM-1 single positive vessels) and became detectable in the developing venous vasculature (Fig. 4B,E, H,K, arrowheads and open arrows). Significant EphB1 expression but weak or faint EphB4taulacZ expression was detectable in some smaller vessels, suggesting that these EphB1+ vessels are in the process of vascular remodeling to form larger-diameter veins (Fig. 4B,E,H). By E15.5, the venous expression of EphB1 and EphB4taulacZ largely persisted (Fig. 4C,F,I,L, arrowheads and open arrows), although EphB4taulacZ expression was also detectable in lymphatic vessels (Fig. 4C,I,L, open arrowheads). Our detailed time-course analysis of vascular EphB1 expression suggests that EphB1 expression is detectable throughout the primary capillary plexus, and after vascular remodeling. EphB1 expression is restricted to the venous vasculature in a time-dependent manner. Furthermore, the venous expression of EphB1 allows detection of a previously unrecognized intricate venous network in developing skin vasculature.
Given that arterial branches are densely covered with αSMA+ vascular smooth muscle cells (VSMCs), whereas less or no VSMC coverage occurs in venous branches (Mukouyama et al., 2002), we further confirmed whether EphB1+ vessels exhibit venous structure. Indeed, EphB1 expression was detectable in a typical venous vasculature with less or no VSMC coverage (Fig. 5A,D,G, arrowheads vs. open arrowheads). In contrast, no significant EphB1 expression was detectable in a typical arterial vasculature with dense VSMC coverage (Fig. 5A,D, open arrowheads). Moreover, venous expression of EphB1 was observed in back and head skin vasculatures (Fig. 5B,E,H, back skin; C,F,I, head skin, arrowheads vs. open arrowheads). Of note, venous remodeling from the primary capillary plexus appears to be in process at E15.5, albeit arterial remodeling has been completed to form a hierarchical arterial branching. These data led to a more general understanding of the venous expression of EphB1 in embryonic skin vasculature.
Venous Expression of EphB1 in the Adult Skin Vasculature
We next examined the expression of EphB1 in adult skin. Our whole-mount immunohistochemical analysis of adult ear skin revealed that EphB1 expression persisted in the adult skin vasculature (Fig. 6A,C,E, arrowheads). The cellular resolution of EphB1 expression afforded by the immunohistochemical staining with the antibody to EphB1 allowed us to visualize the highly branched vascular network. Although EphB1 had previously been reported to be largely expressed in the nervous system (Flanagan and Vanderhaeghen, 1998), no significant expression of EphB1 was detectable in peripherin+ nerves in the adult ear skin (Fig. 6B,B′,D,D′,F,F′, arrows vs. arrowheads).
Some EphB1+ vessels express the venous marker EphB4taulacZ (Fig. 7A,D,G, arrowheads and open arrows). EphB1 expression was detected in the entire venous network (Fig. 7A,D,G, arrowheads), whereas EphB4taulacZ expression was partially detected in venous and lymphatic-like vessels (Fig. 7A,D,G, open arrows and open arrowheads). No significant expression of the arterial marker ephrinB2taulacZ and the lymphatic marker LYVE-1 was detected in EphB1+ vessels (Fig. 7B, E, H, ephrinB2taulacZ, arrowheads vs. open arrows; Fig. 7C,F,I, LYVE-1, arrowheads vs. open arrowheads). Taken together, these data demonstrate that the preferential expression of EphB1 in the skin venous vasculature persists into adulthood.
EphB1 Expression Is Not Restricted to Veins in Yolk Sac, Trunk, and Lung
The preceding observations raised the question of whether EphB1 expression is restricted to veins in other tissues. To address this question, we examined the expression of EphB1 in the dorsal aorta and cardinal vein in E9.5 and E13.5 mouse embryos (Fig. 8). Double immunohistochemical analysis of transverse sections of E9.5 and E13.5 trunks with antibodies to EphB1 and PECAM-1 revealed that EphB1 expression was detectable in PECAM-1+ endothelial cells in E9.5 trunk vasculature including both the dorsal aorta and cardinal vein (Fig. 8A,D, arrowheads). The pan-endothelial expression of EphB1 persisted in E13.5 trunk vasculature (Fig. 8B,E, arrowheads), but the expression was not detectable in LYVE-1+ lymphatic vessels (Fig. 8C,F, arrowheads vs. open arrowheads).
We further examined the expression of EphB1 in yolk sac and lung vasculature (Fig. 9). EphB1 expression was detectable in yolk sac capillary plexus at E8.0, as seen in E13.5 skin capillary plexus (Fig. 9A,C). At E9.5, after the vascular remodeling had occurred, EphB1 expression persisted in all yolk sac vasculature including Nrp1+ remodeled arteries (Fig. 9B,D, open arrowheads), Nrp1− remodeled veins and capillaries (Fig. 9B,D, arrowheads). In embryonic lung vasculature, EphB1 expression was detectable in EphB4taulacZ-expressing vessels, which appear to form the majority of lung microvasculature (Fig. 9E,G, arrowheads), whereas EphB1 expression was also detectable in ephrinB2taulacZ-expressing large-diameter arteries (Fig. 9F,H, open arrowheads).
These data suggest that EphB1 may be considered as a tissue-specific venous endothelial cell marker in skin vasculature, rather than as a general venous-restricted marker.
In this study, we have described in detail the venous expression of EphB1 in mouse embryonic and adult skin vasculature. Whole-mount staining of skin with the antibody to EphB1 allows us to study venous vessel identity and the intricate branching network with greater resolution. Our findings provide important implications for the use of EphB1 as a venous-restricted marker in the analysis of venous differentiation and branching patterns in the embryonic skin and the analysis of neo-vascularization and tumor angiogenesis in the adult skin.
EphB receptors are expressed in the developing vasculature and play critical roles in embryonic angiogenesis. Previous studies demonstrated that the expression of EphB2, EphB3, and EphB4 mRNA is detectable in embryonic blood vessels (Adams et al., 1999) and the prominent venous expression of EphB4 was detected by anti-β-galactosidase antibody staining in EphB4taulacZ/+ heterozygous mice (Gerety et al., 1999). Interestingly EphB4taulacZ/taulacZ homozygous mutants (Gerety et al., 1999) and EphB2−/−;EphB3−/− double homozygous mutants (Adams et al., 1999; Williams et al., 2003) exhibit angiogenic remodeling defects, resulting in the failure of organized vessel architecture during early embryogenesis (∼E9.5). In contrast, EphB1−/− homozygous mutants do not show any detectable phenotypic defects in vascular development (Williams et al., 2003). Because no vascular defect is detectable in single EphB2−/− or EphB3−/− homozygous mutants as well, these EphB receptors may be functionally redundant such that EphB2, EphB3, or EphB4 may compensate for the lack of EphB1 in the homozygous mutants.
Recent studies have revealed that ephrinB2 regulates the endocytosis of VEGFR2 and VEGFR3 in sprouting angiogenesis and lymphangiogenesis (Sawamiphak et al., 2010; Wang et al., 2010). The internalization of VEGFR2 and VEGFR3 is necessary for activation and downstream signaling of the receptor and is required for VEGF-mediated endothelial sprouting. Given that EphB1 and EphB4 are differentially expressed by blood and lymphatic endothelial cells, one intriguing scenario is that endothelial ephrinB2 may be activated by either EphB1 or EphB4 during sprouting angiogenesis, whereas ephrinB2 activation in lymphatic endothelial cells may be mainly conducted by EphB4 during sprouting lymphangiogenesis.
Our skin vasculature model system provides a directly observable vascular network with an anatomically recognizable pattern (Li and Mukouyama, 2011). The time course analysis allows us to study the process of development from primary vascular network to a well-defined, hierarchical branching network. Previously, we discovered that in embryonic limb skin, arterial differentiation and the branching pattern of arterial vessels are controlled by that of peripheral sensory nerves (Mukouyama et al., 2002). We further discovered that nerve-derived VEGF-A is necessary for proper arterial differentiation (Mukouyama et al., 2005), and nerve-derived Cxcl12 controls vessel branching and alignment with nerves (Li et al., 2013). Given that few molecular tools are available to identify venous endothelial cells, however, what controls venous differentiation and branching patterns still remains elusive. The use of EphB1 makes it possible to image venous endothelial cells in the skin with greater resolution and higher reliability and to delineate defective venous development in mouse mutants.
Our detailed analysis of embryonic EphB1 expression provides the first evidence that in the skin EphB1 is expressed throughout the primitive capillary plexus, but after vascular remodeling has occurred the expression is restricted to veins. EphB1 expression is robustly reduced in the arterial marker ephrinB2+ vessels, underlying the mechanism that negatively regulates EphB1 expression during arterial differentiation. These data raised the question of whether EphB1 expression at the stage of the primitive capillary plexus is correlated with prespecification of venous identity. One scenario is that venous specification might simply require the maintenance of venous-restricted genes, whereas arterial specification might require the induction of arterial specific genes and the repression of venous-restricted genes. Resolution of this issue will require an extensive comparison of gene expression patterns between EphB1+ endothelial cells from the primary capillary plexus and EphB1+ venous endothelial cells. Overall, the use of EphB1 as a venous-restricted marker in future studies should not only provide insights into our understanding of venous development but also refine our understanding of arterial-venous branching network formation.
Whole-Mount Immunohistochemistry of Limb Skin
Staining was performed essentially as described previously (Mukouyama et al., 2002; Li and Mukouyama, 2011). Briefly, forelimb skin tissue was dissected from embryos (E13.5–E15.5), fixed in 4% paraformaldehyde/phosphate buffered saline (PBS) at 4°C overnight, and dehydrated in 100% methanol at −20°C. Staining was performed using anti-PECAM-1 antibody (rat monoclonal antibody, clone MEC13.3, BD Pharmingen, 1:300 overnight at 4°C) to detect endothelial cells; anti-Nrp1 antibody (rabbit polyclonal antibody, A.L. Kolodkin, 1:3,000, overnight at 4°C) was used as an arterial endothelial cell marker; β-gal antibody (rabbit polyclonal antibody, Cappel, 1:500, overnight at 4°C) as arterial (ephrinB2taulacZ) or venous (EphB4taulacZ) endothelial cell markers; anti-LYVE1 antibody (rat monoclonal antibody, MBL, 1:300, overnight at 4°C) and anti-Nrp2 (rabbit polyclonal antibody, Cell Signaling 1:100, overnight at 4°C) to detect lymphatic endothelial cells; and anti-EphB1 antibody (goat polyclonal antibody, M-19, Santa Cruz, 1:100) to detect venous endothelial cells. For immunofluorescence detection, either Alexa-488-, Alexa-568-, Cy3- or Dylight 649-conjugated secondary antibodies (Invitrogen 1:250 or Jackson, 1:300, 1 hr at room temperature) were used. All confocal microscopy was carried out on a Leica TCS SP5 confocal (Leica).
Whole-Mount Immunohistochemistry of Adult Ear Skin
Murine adult ear skin tissue was dissected and fixed in 4% paraformaldehyde/PBS room temperature for 1 hour, and washed with PBS. Staining procedures were performed essentially as described previously (Feldheim et al., 1998; Flanagan and Vanderhaeghen, 1998; Nagao et al., 2009) and staining with antibodies and use of reagents were performed as described above. All confocal microscopy was carried out on a Leica TCS SP5 confocal (Leica).
Section staining was performed as described previously (Mukouyama et al., 2005). In brief, embryos (E9.5 and E13.5) were fixed with 4% paraformaldehyde/PBS at 4°C overnight, immersed in 30% sucrose/PBS at 4°C overnight, and then embedded in OCT compound. Embryos were cryosectioned at 10–12 μm thickness and collected on precleaned slides (Matsunami, Japan). Staining was performed using anti-PECAM-1 antibody and anti-EphB1 to detect endothelial cells. For immunofluorescence detection, either Alexa-568- or Dylight 649-conjugated secondary antibodies (Invitrogen 1:250 or Jackson, 1:300, 1 hr at room temperature) were used. All confocal microscopy was carried out on a Leica TCS SP5 confocal (Leica).
We thank S. Motegi and K. Zukosky for technical assistance with whole-mount ear skin staining. We also thank M. Udey, R. S. Adelstein, A. Michelson, and R. S. Balaban for invaluable help and discussion, and other members of the Laboratory of Stem Cell and Neuro-Vascular Biology for technical help and thoughtful discussions. Thanks to J. Hawkins and the staff of NIH Bldg50 animal facility for assistance with mouse breeding and care, K. Gill for laboratory management and technical support, M. Conti for editorial advice on the manuscript, and Y. Carter, L. Oundo, and R. Reed for administrative assistance. None of the authors have any financial or other conflicts of interest.
- 1999. Roles of ephrinB ligands and EphB receptors in cardiovascular development: demarcation of arterial/venous domains, vascular morphogenesis, and sprouting angiogenesis. Genes Dev 13:295–306. , , , , , , , .
- 2011. Stepwise arteriovenous fate acquisition during mammalian vasculogenesis. Dev Dyn 240:2153–2165. , , , , .
- 2004. Dosage-sensitive requirement for mouse Dll4 in artery development. Genes Dev 18:2474–2478. , , , , , , , , .
- 1998. Topographic guidance labels in a sensory projection to the forebrain. Neuron 21:1303–1313. , , , , , , .
- 1998. The ephrins and Eph receptors in neural development. Ann Rev Neurosci 21:309–345. , .
- 2004. Haploinsufficiency of delta-like 4 ligand results in embryonic lethality due to major defects in arterial and vascular development. Proc Natl Acad Sci U S A 101:15949–15954. , , , , , , , , , , , .
- 1999. Symmetrical mutant phenotypes of the receptor EphB4 and its specific transmembrane ligand ephrin-B2 in cardiovascular development. Mol Cell 4:403–414. , , , .
- 1995. Expression of the fms-like tyrosine kinase 4 gene becomes restricted to lymphatic endothelium during development. Proc Natl Acad Sci U S A 92:3566–3570. , , , , , , , .
- 2011. Whole-mount immunohistochemical analysis for embryonic limb skin vasculature: a model system to study vascular branching morphogenesis in embryo. J Vis Exp pii 2620. , .
- 2013. Peripheral nerve-derived CXCL12 and VEGF-A regulate the patterning of arterial vessel branching in developing limb skin. Dev Cell 24:359–371. , , , , , , , , .
- 2002. Sensory nerves determine the pattern of arterial differentiation and blood vessel branching in the skin. Cell 109:693–705. , , , , .
- 2005. Peripheral nerve-derived VEGF promotes arterial differentiation via neuropilin 1-mediated positive feedback. Development 132:941–952. , , , , .
- 2009. Murine epidermal Langerhans cells and langerin-expressing dermal dendritic cells are unrelated and exhibit distinct functions. Proc Natl Acad Sci U S A 106:3312–3317. , , , , , , , .
- 1997. Mechanisms of angiogenesis. Nature 386:671–674. .
- 2009. Molecular differentiation and specialization of vascular beds. Angiogenesis 12:139–147. , .
- 2010. Ephrin-B2 regulates VEGFR2 function in developmental and tumour angiogenesis. Nature 465:487–491. , , , , , , .
- 2005. Isolation of arterial-specific genes by subtractive hybridization reveals molecular heterogeneity among arterial endothelial cells. Dev Dyn 233:1589–1604. , .
- 2001. Expression of ephrinB2 identifies a stable genetic difference between arterial and venous vascular smooth muscle as well as endothelial cells, and marks subsets of microvessels at sites of adult neovascularization. Dev Biol 230:139–150. , , , , , , , , , .
- 2007. Lineage tracing demonstrates the venous origin of the mammalian lymphatic vasculature. Genes Dev 21:2422–2432. , , , , , , , .
- 2002. Orchestration of angiogenesis and arteriovenous contribution by angiopoietins and vascular endothelial growth factor (VEGF). Proc Natl Acad Sci U S A 99:8219–8224. , , .
- 1998. Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by ephrin-B2 and its receptor Eph-B4. Cell 93:741–753. , , .
- 2010. Ephrin-B2 controls VEGF-induced angiogenesis and lymphangiogenesis. Nature 465:483–486. , , , , , , , , , , , , , , .
- 2003. Ephrin-B2 and EphB1 mediate retinal axon divergence at the optic chiasm. Neuron 39:919–935. , , , , , , , , , .
- 2005. Suppression of Notch signalling by the COUP-TFII transcription factor regulates vein identity. Nature 435:98–104. , , , , , .
- 2002. Abnormal lymphatic vessel development in neuropilin 2 mutant mice. Development 129:4797–4806. , , , , , , .