Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. 2007 Feb 27; 104(9): 3225–3230.
Published online 2007 Feb 12. doi:  10.1073/pnas.0611177104
PMCID: PMC1805603
Developmental Biology

The Notch ligand Delta-like 4 negatively regulates endothelial tip cell formation and vessel branching


Delta-like 4 (Dll4) is a transmembrane ligand for Notch receptors that is expressed in arterial blood vessels and sprouting endothelial cells. Here we show that Dll4 regulates vessel branching during development by inhibiting endothelial tip cell formation. Heterozygous deletion of dll4 or pharmacological inhibition of Notch signaling using γ-secretase inhibitor revealed a striking vascular phenotype, with greatly increased numbers of filopodia-extending endothelial tip cells and increased expression of tip cell marker genes compared with controls. Filopodia extension in dll4+/− retinal vessels required the vascular growth factor VEGF and was inhibited when VEGF signaling was blocked. Although VEGF expression was not significantly altered in dll4+/− retinas, dll4+/− vessels showed increased expression of VEGF receptor 2 and decreased expression of VEGF receptor 1 compared with wild-type, suggesting they could be more responsive to VEGF stimulation. In addition, expression of dll4 in wild-type tip cells was itself decreased when VEGF signaling was blocked, indicating that dll4 may act downstream of VEGF as a “brake” on VEGF-mediated angiogenic sprouting. Taken together, these data reveal Dll4 as a negative regulator of vascular sprouting and vessel branching that is required for normal vascular network formation during development.

Keywords: angiogenesis, vascular development, VEGF, sprouting, guidance

Notch signaling controls cell fate specification in a variety of cell contexts during embryonic and postnatal development (1). Genetic deletion of multiple components of the Notch pathway has revealed a critical role for Notch in vascular development (2). In mice, the absence of Notch signaling results in defective yolk sac vascular remodeling and aberrant formation of arterial-venous circuits in the embryo, often leading to embryonic death (2). Studies in zebrafish and in vitro cultured cells have implicated Notch signaling in arterial-venous specification through regulation of arterial genes such as EphrinB2 (efnb2) (2, 3) and have identified Notch as a downstream target of the vascular growth factor VEGF (4). Delta-like 4 (Dll4) is a transmembrane ligand for Notch receptors that shows restricted expression to endothelial cells (ECs), in particular to arteries and capillaries (58). Deletion of a single dll4 allele in mice results in early embryonic death [from embryonic day (E)9.5] associated with major defects in vascular remodeling in the yolk sac and embryo (911). Haploinsufficiency within the vascular system has previously been observed only for VEGF (12, 13), suggesting that an appropriate dosage of both of these genes is critical for correct vascular development. The degree of lethality associated with heterozygous dll4 deletion depends on the mouse strain, and a small percentage of outbred CD1 dll4+/− embryos can survive, allowing for more detailed analysis of roles for dll4 at later stages of vascular development. Here we describe a previously uncharacterized function for dll4 during sprouting angiogenesis, with dll4 suppressing endothelial tip cell formation and vessel branching by inhibiting the response of sprouting ECs to VEGF.


dll4+/− Embryos Show Vessel Branching Defects.

We compared outbred E10.5 heterozygous dll4 embryos carrying a β-gal reporter gene insertion (11) to either heterozygous unc5b β-gal-reporter embryos, which show similar vascular expression to dll4+/− but with normal vascular development (14), or to platelet–EC adhesion molecule 1 (PECAM-1)-stained wild-type littermates (Fig. 1 A–D). In dll4+/− embryos, the major arteries such as the internal carotid artery appeared normal but displayed increased numbers of smaller arterial branches compared with unc5b+/− or PECAM-1-stained wild-type embryos (Fig. 1 A–D). All dll4+/− embryos examined showed branching defects but with varying severity [compare Fig. 1B and supporting information (SI) Fig. 5A], indicating that unknown genetic modifiers also influence the dll4+/− phenotype. To exclude the possibility that the defects observed in vessel branching were secondary to other defects in the embryo, e.g., disrupted blood flow, we cultured aortic explants from dll4+/− and wild-type mice in collagen gels. Sprouting from dll4+/− aortas occurred earlier and more profusely than from wild-type aortas (Fig. 1 E–G), indicating that the dll4+/− vessel defect is intrinsic to ECs and independent of alterations to flow or cardiac output.

Fig. 1.
Vessel branching defect in dll4+/− embryos. (A–D) Staining for β-gal activity in E10.5 unc5b+/− and dll4+/− embryos and for PECAM-1 in E10.5 wild-type and dll4+/− embryos. dll4+/− embryos show increased ...

To establish the underlying cause of the dll4+/− vessel-branching defects, we performed high-resolution confocal microscopy on isolectin B4-stained dll4+/− and wild-type E11.5 hindbrains. Staining for β-gal in dll4+/− hindbrains showed expression in virtually all ECs (data not shown). Vessel branching was significantly increased in dll4+/− compared with wild-type hindbrains (Fig. 1 H–J). Strikingly, vessels from dll4+/− hindbrains extended numerous filopodia from the whole length of the vessel surface, whereas vessels from wild-type hindbrains extended only a few filopodia at scattered points (Fig. 1 K–N). Filopodia extension is a morphological characteristic of specialized ECs called “tip cells,” which are lumenless ECs present at the leading edge of vascular sprouts that integrate directional cues from their environment and so define the direction in which the new sprout grows (15). Tip cells also find and create connections with adjacent sprouts and so generate functional vascular networks (16). The profusion of filopodia in dll4+/− hindbrains suggested that most, if not all, ECs in the hindbrain vessels were acting like tip cells, leading to increased connections between adjacent vessels (i.e., branching).

dll4+/− Retinal Vessels Are Hyperbranched with Increased Tip Cell Numbers.

To further evaluate whether loss of dll4 leads to increased endothelial tip cell formation, we analyzed postnatal retinal vascular development of surviving dll4+/− and wild-type pups (approximately one-third of dll4+/− embryos survive to birth). The retina vasculature at postnatal day (P)4–6 allows for simultaneous visualization of angiogenic sprouting at the vascular front (where most endothelial tip cells are located) and remodeling of the nascent vasculature within the vascular plexus (17). In wild-type retinas, dll4 expression was detected predominantly in arteries and in tip cells at the vascular front, with lower levels observed in vessels within the vascular plexus (SI Fig. 5). In dll4+/− vessels, lower levels of dll4 were detectable by in situ hybridization (ISH; SI Fig. 5) and quantitative PCR (qPCR; 51% of wild-type levels), although the expression pattern was comparable to wild-type (SI Fig. 5). As with hindbrains, dll4+/− retinal vessels showed severe patterning defects, forming a hyperbranched, hyperfused plexus behind the vascular front (Fig. 2 A and B). Numerous filopodia extended from dll4+/− vessels at the vascular front and also within the vascular plexus in both arterial and venous zones (Fig. 2D; SI Fig. 6). In contrast, wild-type vessels extended few filopodia in regions away from the vascular front (Fig. 2C; SI Fig. 6). Quantification (see SI Fig. 6) showed 75% more filopodia extensions at the vascular front, 125% more branchpoints within the vascular plexus, and an 80% increase in the area covered by vessels in dll4+/− compared with wild-type (Fig. 2 E–G). This increase in the vessel coverage suggested that dll4+/− vessels may also have proliferation defects. Quantification of BrdU-labeled (S-phase cells) or phospho-histone-H3-stained ECs (M-phase cells) indicated a modest (1.16-fold, nonsignificant) increase in proliferating ECs in dll4+/− vessels compared with wild-type (Fig. 2H; SI Fig. 6). Although this small increase in EC proliferation could contribute to the overall dll4+/− phenotype, it is unlikely to be the causative defect. Smooth muscle and pericyte coverage, as well as distribution of retinal astrocytes ahead of the vascular plexus, appeared normal between wild-type and dll4+/− vessels (SI Fig. 6). Retinal astrocytes define the pattern of vascular development in the retina through the production of VEGF gradients in response to hypoxia (15, 18, 19). Assessment of hypoxia and VEGF gradients showed no significant differences between wild-type and dll4+/− retinas (SI Fig. 7). dll4+/− vessels were perfused normally (SI Fig. 7), and there was no evidence of plasma leakage. Taken together, these observations suggest that dll4+/− retinal vessels are functional, and that the major defect in vessel patterning is due to inappropriate and excessive sprouting.

Fig. 2.
Increased tip cell formation in dll4+/− retinal vessels. (A and B) Isolectin B4-stained P5 dll4+/− retinal vessels show a hyperfused plexus compared with wild-type (a, artery; v, vein). (C and D) dll4+/− vessels extend many more ...

In addition to dll4, other genes are also expressed at high levels in endothelial tip cells in the retina, including pdgfb (15) and unc5b (14). Compared with wild-type, dll4+/− retinal vessels expressed pdgfb (Fig. 2 I, J, M, and N) and unc5b (Fig. 2 K, L, O, and P), over an expanded area, especially in the hyperfused plexus. Thus, vessels from dll4+/− retinas display genetic as well as morphological (filopodia) and behavioral (hyperfused vessels) indicators of an expansion in the number of ECs that have a “tip cell” phenotype, suggesting that Dll4 normally functions to suppress tip cell formation in growing vessels.

dll4 Negatively Regulates VEGF-Induced Sprouting Angiogenesis.

Normal angiogenic sprouting in the retina requires local gradients of VEGF (15, 18, 19). If VEGF stimulation is blocked, tip cell filopodia retract, and progression of the vascular sprout halts. In dll4+/− retinas, increased tip cell formation occurs in the absence of an altered VEGF gradient, suggesting that Dll4 could function downstream of VEGF. To test whether dll4 expression depended on VEGF signaling, we injected the eyes of P5 wild-type pups with soluble VEGF receptor 1 (VEGFR1; sFlt1, VEGF sink). In addition to the expected loss of tip cell filopodia (15), dll4 mRNA was down-regulated in a significant proportion of tip cells (Fig. 3 A–D, dll4 expression in 165/183 tip cells in uninjected retinas, and 30/172 tip cells in sFlt1-injected retinas; n = 7 per group), indicating that dll4 expression in tip cells is downstream of VEGF in vivo, as previously shown in vitro (20). To confirm that filopodia extension in dll4+/− vessels still required VEGF signal, we injected eyes with receptor-blocking antibodies against VEGFR2 (Flk-1), the major mediator of VEGF function, and VEGFR1, a decoy receptor that negatively modulates VEGF signaling (21). VEGFR2-blocking antibody induced filopodia retraction in wild-type and dll4+/− retinas (Fig. 3 E, F, and I), whereas VEGFR1-blocking antibody had no significant effect (Fig. 3 G–I). Soluble Flt1 also blocked filopodial extension in dll4+/− vessels (Fig. 3I). Thus, absence of a dll4 allele does not alter the requirement for VEGF and VEGFR2 activation for tip cell filopodia extension. Because the VEGF gradient is intact in dll4+/− retinas, excessive sprouting in dll4+/− vessels could therefore be due to an increased response to normal levels of VEGF signal, perhaps through altered VEGFR expression. In support of this hypothesis, activated Notch signaling in ECs has been shown to down-regulate the expression of VEGFR2 in vitro (2224). Thus, we asked whether loss of a dll4 allele would lead to increased VEGFR2 expression. Because VEGFR2 is expressed strongly in the neuronal retinal layer (25, 26), we performed qPCR analysis on isolated retinal ECs and found increased vegfr2 expression in dll4+/− compared with wild-type (Fig. 3J). ISH also showed an expanded domain of vegfr2 expression in dll4+/− retinal vessels compared with wild-type, especially in the region of the hyperfused plexus (Fig. 3 K, L, O, and P). In addition, we observed decreased vegfr1 expression in dll4+/− retinal vessels by both qPCR and ISH (Fig. 3 J, M, and N, Q and R). Thus, dll4+/− retinal vessels express higher levels of VEGFR2 and lower levels of VEGFR1, potentially increasing their responsiveness to VEGF. Such alterations in VEGFR levels could therefore provide a rational explanation for the increased sprouting observed in dll4+/− retinas.

Fig. 3.
Regulation of dll4+/− tip cells by VEGF. (A–D) Loss of dll4 mRNA expression after intraocular injection of soluble VEGFR1 (sFlt1) in wild-type pups. (A and B) Isolectin B4 staining shows tip cell filopodia (A, white arrowheads), and whole-mount ...

Disruption of Notch Signaling Recapitulates the dll4+/− Retinal Vessel Defect.

To confirm that the dll4+/− vascular defects were consistent with a disruption of Notch signaling, we measured expression levels of the Notch target genes hey1 and hey2 by qPCR. Both were down-regulated in dll4+/− retinas compared with wild-type (by 15% and 21%, respectively). Pharmacological disruption of Notch signaling in wild-type retinas using the γ-secretase inhibitor N-[N-(3,5-difluorophenacetyl-l-alanyl)]-S-phenylglycine t-butyl ester (DAPT) produced vessel abnormalities similar to the dll4+/− phenotype (Fig. 4 A–D), including a hyperfused plexus and numerous ectopic filopodia extending from vessels (40% increase in filopodia-extending cells, 150% increase in branchpoints, and 100% increase in vascular area). The expression of tip cell marker genes, including pdgfb and dll4 itself, was also up-regulated in DAPT-treated retinal vessels (Fig. 4 E–L), indicating that sprouting cannot be suppressed when Notch signaling is blocked, even in the presence of increased dll4. The vascular defects observed in dll4+/− mice are thus consistent with a disruption in Notch signaling. Likely candidate Notch receptors include Notch1 and Notch4, both of which are expressed in retinal vessels (7).

Fig. 4.
Inhibition of Notch signaling in retinas. (A and B) Isolectin B4 staining of wild-type P5 retinal vessels after 2 days of treatment with γ-secretase inhibitor DAPT or DMSO control. (C and D) Quantification of vessel branchpoints and percentage ...


Previous studies showed that genetic deletion of dll4 or other Notch signaling components in mice produces profound vascular defects (9%11, 27%%%31), indicating the importance of these molecules in vascular development but providing little insight into their function. Studies in zebrafish and in vitro cultured cells indicate that some of the vascular abnormalities observed in dll4-deficient mice may be attributable to defects in arterial-venous specification (3, 4, 20). However, arteries from dll4+/− embryos and retinas appear morphologically normal, and dll4+/− retinas express arterial markers such as unc5b (Fig. 2 L and P) and efnb2 (data not shown). Thus, we find the vessel branching defects observed in dll4+/− retinas inconsistent with an arterial specification problem. Instead, our data suggest a model in which Dll4, expressed in endothelial tip cells, inhibits the angiogenic response of adjacent ECs to VEGF stimulation, most likely through Notch signaling. This mechanism would permit an asymmetric cellular response to VEGF stimulation during vascular sprouting by allowing some ECs to respond to a local VEGF gradient by forming a sprout, while, through up-regulation of dll4 expression, inhibiting adjacent cells from also forming sprouts. When even a single dll4 allele is absent, or when Notch signaling is blocked, this suppression is lost, resulting in increased sprout formation and tip cell filopodia. This mechanism provides an elegant negative feedback system intrinsic to ECs to control their response to VEGF and suggests that vascular network formation is coordinated by VEGF and Dll4/Notch signaling.

Materials and Methods

Generation of Mice and Collection of Tissues.

Outbred CD1 dll4+/− male mice (11) were crossed with CD1 wild-type females (Charles River Breeding Laboratories, Wilmington, MA) to generate dll4+/− offspring. Approximately 25% of pups born carried the β-gal transgene (expected frequency, 50%). CD1 unc5b+/− embryos were generated as described (14). Embryos were briefly fixed in 4% paraformaldehyde (Sigma, St. Louis, MO) for X-Gal (Invitrogen, Carlsbad, CA) staining or fixed in DMSO/methanol for whole-mount PECAM-1 (Mec13.3; BD PharMingen, San Diego, CA) staining. Hindbrains and retinas were dissected as described (14, 15).

Immunostaining and ISH.

Immunostaining was performed as described (14, 15), using biotinylated isolectin B4 (1:50; Vector Laboratories, Burlingame, CA), anti-phospho-histone-H3 (1:100; Abcam, Cambridge, MA), anti-NG2 (1:40; Chemicon, Hampshire, U.K.), anti-α-smooth muscle actin-Alexa488 (1:100; Sigma), anti-GFAP (1:75; Dako, Carpinteria, CA), anti-rabbit Alexa488 (1:100; Invitrogen), and streptavidin Cy-3 or Cy-2 (1:100; Amersham Biosciences, Piscataway, NJ). BrdU (1 mg/kg; Roche, Indianapolis, IN) was injected i.p. 2 h before death and retinas stained with anti-BrdU-Alexa546 (1:50; Invitrogen). Hypoxia was assessed by injection of 60 mg/kg pimonidazole (Chemicon) 90 min before death, and retinas stained with Hypoxyprobe Mab1-FITC (1:100; Chemicon). For perfusion assays, FITC-dextran (20 mg/ml, Mr 200,000; Sigma) was injected into the tail vein of anesthetized P5 pups. Confocal images were acquired by using a Leica (Deerfield, IL) TCS SP2 microscope. Whole-mount ISH of retinas was performed as described (32), except for treatment with proteinase K at 8 mg/ml for 10 min.

Vessel branchpoints and filopodia bursts were counted manually (by two independent observers); vascular area was calculated by using Adobe Photoshop 7.0 (Adobe Systems, San Jose, CA; SI Fig. 6). Quantifications were performed on two images per hindbrain (n = four wild-type; five dll4+/−) and three to four images per retina (n = six wild-type; seven dll4+/−). Proliferating ECs were counted manually on four images per retina (n = five wild-type; five dll4+/−). Proliferating non-ECs were not counted.

Aortic Ring Assay.

Aortas from P5 pups were dissected, cut into short segments, and incubated overnight in EC growth medium-2 (Promocell, Heidelberg, Germany.) containing 10% FBS before being placed into collagen gels (Roche). Aortic explants were imaged at 48 h after embedding by using an inverted microscope (Leica) and sprout area quantified by using Metaview imaging software (Princeton, MA).

DAPT Treatment and Eye Injections.

The γ-secretase inhibitor DAPT (100 mg/kg per mouse; ref. 33) (Calbiochem, La Jolla, CA; dissolved in DMSO) was injected i.p. into wild-type pups once per day at P3 and P4 and retinas dissected at P5. Vessel quantifications were performed on three to four images per retina (n = four per group). Intraocular injections were performed as described (14, 15). Mice were killed 4–5 h after injection. sFlt-1 (R&D Systems, Minneapolis, MN) was injected at 1 μg/μl, VEGFR2-blocking antibody (DC101; Imclone, New York, NY) at 5.7 μg/μl, and VEGFR1-blocking antibody (MF1; Imclone) at 5.2 μg/μl. The number of injected eyes was sFlt-1, 12 wild type, five dll4+/−; VEGFR2-blocking antibody, seven wild type, five dll4+/−; and VEGFR1-blocking antibody, four wild type, three dll4+/−. Number and length of filopodia (in pixels) were quantified from three to four fields per retina by using MetaView software imaging.


Retinal ECs were isolated by digestion of dissected retinas for 45 min with 1 mg/ml collagenase (Gibco, Carlsbad, CA), incubation for 15 min at 4°C with biotinylated anti-PECAM-1 antibody (1:500 dilution), and separation by using streptavidin-coated magnetic beads (Miltenyi Biotec, Auburn, CA). Total RNA was extracted by using the RNeasy mini kit (Qiagen, Valencia, CA) and reverse-transcribed to cDNA by using the SuperScript III First Strand Synthesis kit (Invitrogen). qPCR was performed in triplicate by using an ABI PRISM 7300 (Applied Biosystems, Foster City, CA) and SYBR green chemistry. Gene expression was normalized to β-actin and relative differences determined based on six separate reactions.

Supplementary Material

Supporting Figures:


We thank G. Breier (Technical University, Dresden, Germany) for the vegf probe and R. del Toro and B. deLafarge for help with quantifications. This work was supported by grants from Institut National de la Santé et de la Recherche Médicale (Avenir), Ministère de la Recherche (Agence Nationale de la Recherche (ANR) Programme Cardiovasculaire, Obésité, Diabète, No. 05-022-03), the Association pour la Recherche sur le Cancer (No. 3124), the Institut de France (Cino Del Duca), and the European Community (Grant LSHG-CT-2004-503573). S.S. is supported by INSERM (poste vert). C.F. is supported by Fundação para a Ciência e a Tecnologia (Fellowship No. 20225/2004).


delta-like 4
endothelial cell
in situ hybridization
embryonic day n
platelet-EC adhesion molecule 1
postnatal day n
quantitative PCR
N-[N-(3,5-difluorophenacetyl-l-alanyl)]-S-phenylglycine t-butyl ester
VEGF receptor 1/2.


The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0611177104/DC1.


1. Artavanis-Tsakonas S, Rand MD, Lake RJ. Science. 1999;284:770–776. [PubMed]
2. Shawber CJ, Kitajewski J. BioEssays. 2004;26:225–234. [PubMed]
3. Lawson ND, Scheer N, Pham VN, Kim CH, Chitnis AB, Campos-Ortega JA, Weinstein BM. Development. Vol. 128. Cambridge, UK: 2001. pp. 3675–3683. [PubMed]
4. Lawson ND, Vogel AM, Weinstein BM. Dev Cell. 2002;3:127–136. [PubMed]
5. Villa N, Walker L, Lindsell CE, Gasson J, Iruela-Arispe ML, Weinmaster G. Mech Dev. 2001;108:161–164. [PubMed]
6. Shutter JR, Scully S, Fan W, Richards WG, Kitajewski J, Deblandre GA, Kintner CR, Stark KL. Genes Dev. 2000;14:1313–1318. [PMC free article] [PubMed]
7. Claxton S, Fruttiger M. Gene Expr Patterns. 2004;5:123–127. [PubMed]
8. Benedito R, Duarte A. Gene Expr Patterns. 2005;5:750–755. [PubMed]
9. Gale NW, Dominguez MG, Noguera I, Pan L, Hughes V, Valenzuela DM, Murphy AJ, Adams NC, Lin HC, Holash J, et al. Proc Natl Acad Sci USA. 2004;101:15949–15954. [PMC free article] [PubMed]
10. Krebs LT, Shutter JR, Tanigaki K, Honjo T, Stark KL, Gridley T. Genes Dev. 2004;18:2469–2473. [PMC free article] [PubMed]
11. Duarte A, Hirashima M, Benedito R, Trindade A, Diniz P, Bekman E, Costa L, Henrique D, Rossant J. Genes Dev. 2004;18:2474–2478. [PMC free article] [PubMed]
12. Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, O'Shea KS, Powell-Braxton L, Hillan KJ, Moore MW. Nature. 1996;380:439–442. [PubMed]
13. Carmeliet P, Ferreira V, Breier G, Pollefeyt S, Kieckens L, Gertsenstein M, Fahrig M, Vandenhoeck A, Harpal K, Eberhardt C, et al. Nature. 1996;380:435–439. [PubMed]
14. Lu X, Le Noble F, Yuan L, Jiang Q, De Lafarge B, Sugiyama D, Breant C, Claes F, De Smet F, Thomas JL, et al. Nature. 2004;432:179–186. [PubMed]
15. Gerhardt H, Golding M, Fruttiger M, Ruhrberg C, Lundkvist A, Abramsson A, Jeltsch M, Mitchell C, Alitalo K, Shima D, et al. J Cell Biol. 2003;161:1163–1177. [PMC free article] [PubMed]
16. Isogai S, Lawson ND, Torrealday S, Horiguchi M, Weinstein BM. Development. Vol. 130. Cambridge, UK: 2003. pp. 5281–5290. [PubMed]
17. Gariano RF, Gardner TW. Nature. 2005;438:960–966. [PubMed]
18. West H, Richardson WD, Fruttiger M. Development. Vol. 132. Cambridge, UK: 2005. pp. 1855–1862. [PubMed]
19. Stone J, Itin A, Alon T, Pe'er J, Gnessin H, Chan-Ling T, Keshet E. J Neurosci. 1995;15:4738–4747. [PubMed]
20. Liu ZJ, Shirakawa T, Li Y, Soma A, Oka M, Dotto GP, Fairman RM, Velazquez OC, Herlyn M. Mol Cell Biol. 2003;23:14–25. [PMC free article] [PubMed]
21. Olsson AK, Dimberg A, Kreuger J, Claesson-Welsh L. Nat Rev Mol Cell Biol. 2006;7:359–371. [PubMed]
22. Williams CK, Li JL, Murga M, Harris AL, Tosato G. Blood. 2005 [PMC free article] [PubMed]
23. MacKenzie F, Duriez P, Larrivee B, Chang L, Pollet I, Wong F, Yip C, Karsan A. Blood. 2004;104:1760–1768. [PubMed]
24. Taylor KL, Henderson AM, Hughes CC. Microvasc Res. 2002;64:372–383. [PubMed]
25. Yang K, Cepko CL. J Neurosci. 1996;16:6089–6099. [PubMed]
26. Gariano RF, Hu D, Helms J. Gene Expr Patterns. 2006;6:187–192. [PubMed]
27. Fischer A, Schumacher N, Maier M, Sendtner M, Gessler M. Genes Dev. 2004;18:901–911. [PMC free article] [PubMed]
28. Krebs LT, Xue Y, Norton CR, Shutter JR, Maguire M, Sundberg JP, Gallahan D, Closson V, Kitajewski J, Callahan R, et al. Genes Dev. 2000;14:1343–1352. [PMC free article] [PubMed]
29. Limbourg FP, Takeshita K, Radtke F, Bronson RT, Chin MT, Liao JK. Circulation. 2005;111:1826–1832. [PMC free article] [PubMed]
30. Xue Y, Gao X, Lindsell CE, Norton CR, Chang B, Hicks C, Gendron-Maguire M, Rand EB, Weinmaster G, Gridley T. Hum Mol Genet. 1999;8:723–730. [PubMed]
31. Nakajima M, Yuasa S, Ueno M, Takakura N, Koseki H, Shirasawa T. Mech Dev. 2003;120:657–667. [PubMed]
32. Moyon D, Pardanaud L, Yuan L, Breant C, Eichmann A. Development. Vol. 128. Cambridge, UK: 2001. pp. 3359–3370. [PubMed]
33. Dovey HF, John V, Anderson JP, Chen LZ, de Saint Andrieu P, Fang LY, Freedman SB, Folmer B, Goldbach E, Holsztynska EJ, et al. J Neurochem. 2001;76:173–181. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...