QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gerety, S. S. Articles by Anderson, D. J. Search for Related Content PubMed PubMed Citation Articles by Gerety, S. S. Articles by Anderson, D. J.

Cardiovascular ephrinB2 function is essential for embryonic angiogenesis

Sebastian S. Gerety^1, and David J. Anderson^1,2,*

Present address: Department of Developmental Neurobiology, NIMR, The RIdgeway, Mill Hill, London NW7 1AA, UK
¹ Division of Biology 216-76, California Institute of Technology, Pasadena, CA 91125, USA
² Howard Hughes Medical Institute, California Institute of Technology, Pasadena, CA 91125, USA

View larger version (40K): [in a new window]   Fig. 1. Generation of the Floxed ephrinB2 locus. (A) Restriction maps of the wild-type ephrinB2 locus, the targeting vector, the initial targeted locus, floxed locus after neo deletion and targeted locus after complete deletion. The targeting vector contains loxP sites (arrowheads) flanking the first coding exon (gray bar). It also contains a floxed PGK-neomycin (‘PGKneo’) selection cassette that was subsequently removed by transient Cre expression to avoid disturbing normal ephrinB2 transcription (see C). (B) Confirmation of homologous recombination of the targeting vector by Southern blotting. The ES cell genomic DNA has been digested with HindIII, and hybridized with Southern probe A (see A). Wild-type (6 kb) and targeted (4 kb) loci differ by a HindIII site flanking the 5' loxP site (see A). (C) Identification of ES cells that have undergone PGKneo cassette deletion (see A) after transient Cre recombinase expression. Genomic DNA was digested with HindIII, and hybridized with southern probe B (see A). ‘Neo deleted’ indicates loss of PGKneo cassette with retention of the first exon. Deletion of the entire region or deletion of the floxed exon are not distinguishable from wild type in this Southern blot. (D) Confirmation of ephrinB2 exon 1 deletion in mice. Progeny of an ephrinB2loxP/+ X CMV-Cre cross show intact (long) or deleted (short) PCR products with primers specific for the loxP allele.

View larger version (99K): [in a new window]   Fig. 2. Tie2-Cre activity in early embryogenesis is restricted to the vasculature. (A-E) Widespread vascular activity of Tie2-Cre in the progeny of a Tie2-Cre X R26R lacZ reporter cross at E8.25, revealed by X-gal staining (blue color). (A) The highly vascularized yolk sac (YS) shows intense Tie2-Cre activity. The primitive vasculature throughout the embryo proper (arrowheads) also shows Tie2-Cre activity. (B) Close-up photograph of intersomitic sprouts (arrows) from the dorsal aorta (DA) of a similar embryo to (A) shows Tie2-Cre activity in vessels undergoing angiogenic sprouting. (C-E) Sections of littermates of A, showing Tie2-Cre activity in vessels (arrowheads) of the head (C), hindbrain (D) and trunk (E). The endocardial lining of the heart is also positive (E, arrows), as expected (Kisanuki et al., 2000). Sections were counterstained with Hematoxylin.

View larger version (119K): [in a new window]   Fig. 3. Gross morphology of conditional ephrinB2 knockout embryos and littermates. Comparison of embryos from an ephrinB2lacZ/+;Tie2-Cre+ X ephrinB2loxP/+ cross (A,C-F), and from an ephrinB2lacZ/+ intercross (B) by anti-PECAM1 staining at E9.5. The conditional ephrinB2 knockout embryo (C) is growth retarded compared with its wild-type littermates (A) and littermates that lack either the Tie2-Cre allele or one of the targeted ephrinB2 alleles (D-F). An ephrinB2-lacZ homozygous embryo (B) shows developmental arrest similar to the conditional knockout embryo (compare with C).

View larger version (114K): [in a new window]   Fig. 4. Yolk sac angiogenesis is defective in conditional ephrinB2 knockout embryos. (A-F) Restriction of ephrinB2 mRNA expression to endothelial cells of the E9 yolk sac is revealed by in situ hybridization with ephrinB2 (A-C) and Flk1 (an endothelial-specific marker, D-F) RNA probes on sections of ephrinB2lacZ/+ control (A,D) ephrinB2lacZ/lacZ (B,E) and conditional ephrinB2 knockout (C,F) yolk sacs at E9. EphrinB2 mRNA in controls is expressed in Flk1-positive endothelial cells (A versus D), and is lost in both conventional (B) and conditional (C) ephrinB2 mutants. (G-I) Identical defects in yolk sac vessel remodeling are seen in ephrinB2lacZ/lacZ (H) and conditional ephrinB2 (I) knockout embryos compared with ephrinB2lacZ/+ controls (G) are revealed by whole-mount anti-PECAM1 staining of E9.5 yolk sacs. The large (arrow) and small (arrowheads) branches in control yolk sacs (G) are instead a plexus of equally sized capillaries (arrowheads) in both mutants (H,I).

View larger version (91K): [in a new window]   Fig. 5. Defective remodeling of the anterior cardinal vein (ACV) of conditionally deleted ephrinB2 embryos. (A-C) EphrinB2 expression is widespread in the hindbrain. EphrinB2lacZ/+ mice at E9.5 were sectioned and stained for PECAM-1 (A and C, green channel) and ß-gal (A and B, red channel). A merged image in A shows ephrinB2 expression in both arterial endothelium (yellow, arrows) and non-endothelial mesenchyme (red channel, black arrowheads) and neuroepithelium (red channel, white arrowheads). (D-F) Tie2-Cre activity is restricted to endothelial cells of the hindbrain. Embryos from a Tie2-Cre X Rosa-lacZ reporter cross at E9.5, sectioned and stained for PECAM-1 (D and F, green channel) and ß-gal (D and E, red channel) show complete overlap (arrows). A merged image is shown in (D). (G-L) EphrinB2 mRNA expression is lost specifically in the hindbrain vasculature of conditional knockout embryos. Sections through ephrinB2lacZ/+ (G,J), ephrinB2lacZ/lacZ (H,K) and ephrinB2lacZ/loxP;Tie2-Cre+ (conditional knockout, I,L) hindbrain regions were hybridized with RNA in situ probes to ephrinB2 (G-I) and Flk1 (J-L), and show loss of ephrinB2 in vessels of conditional knockout embryos (compare G and J with I and L, arrows). EphrinB2 mRNA is completely absent in the ephrinB2lacZ/lacZ conventional knockout (H) but remains in the mesenchyme (white arrowheads) and neuroepithelium (black arrowheads) of conditional knockout embryos (compare G versus I). (M-O) Whole-mount PECAM-1 staining shows a failure of assembly of the ACV in conditional knockout embryos compared with ephrinB2lacZ/+ controls (compare O with M, arrows). EphrinB2lacZ/lacZ embryos show an identical vascular phenotype (compare O with N). Image in O is a close-up of embryo in Fig. 3C. (P-R) Vascular sprouts connect the ACV to the dorsal aorta at E8.5. Whole-mount staining for PECAM1 (P,R, green channel) and ß-gal (P,Q, red channel) of ephrinB2lacZ/+ mice show multiple vascular branches (P and R, white arrowheads) interconnecting the ACV primordium (P-R, outlined by broken white lines) and the ephrinB2 expressing dorsal aorta (arrows). A merged image in (P) shows mesenchymal ephrinB2 expression (red, black arrowheads) surrounding the immature ACV plexus (green channel, outlined by broken white lines).

View larger version (91K): [in a new window]   Fig. 6. Defective angiogenesis in the heads of conditionally deleted ephrinB2 embryos. (A-C) EphrinB2 expression is widespread in the head. EphrinB2lacZ/+ mice at E9.5 were sectioned and stained for PECAM1 (A,C, green channel) and ß-gal (A,B, red channel). A merged image in A shows ephrinB2 expression in both endothelial (yellow) and non-endothelial (red) mesenchymal (black arrowheads) and neuroepithelial (white arrowheads) tissues; ICA, internal carotid artery. (D-F) Tie2-Cre activity in the head is restricted to endothelial cells. Embryos from a Tie2-Cre X Rosa-lacZ reporter cross at E9.5, sectioned and stained for PECAM-1 (D,F, green channel) and ß-gal (D,E, red channel), show complete overlap. A merged image is shown in D. (G-L) EphrinB2 mRNA expression is lost specifically in the vessels of conditional knockout embryos. Sections through ephrinB2lacZ/+ (G,J), ephrinB2lacZ/lacZ (H,K) and ephrinB2lacZ/loxP;Tie2-Cre+ (conditional knockout, I,L) were hybridized with RNA in situ probes to ephrinB2 (G-I) and Flk1 (J-L), and show loss of ephrinB2 in vessels of conditional knockout embryos (compare G and J with I and L, arrows). ephrinB2 mRNA is completely lost in the ephrinB2lacZ/lacZ conventional knockout (H), but remains in the mesenchyme (white arrowheads) and neuroepithelium (black arrowheads) of conditional knockout embryos (compare G with I). (M-O) Whole-mount PECAM1 staining shows arrested vascular remodeling in the heads of conditional ephrinB2 knockouts compared with littermate controls (compare O with M). ephrinB2lacZ/lacZ heads show an identical vascular phenotype (compare O with N). Images in M,O are close-ups of embryos in Fig. 3F,C, respectively.

View larger version (111K): [in a new window]   Fig. 7. Angiogenic arrest of the intersomitic vasculature in conditional ephrinB2 knockout embryos. (A-C) EphrinB2 expression is present in vascular and non-vascular trunk tissues. EphrinB2lacZ/+ mice at E8.5 were stained in whole-mount for PECAM1 (A,C, green channel) and ß-gal (A,B, red channel). The dorsal aorta and its intersomitic sprouts (A-C, arrows) express ephrinB2-lacZ (red channel), as does the caudal half of each somite (A,B arrowheads). (D-F) Tie2-Cre activity is restricted to endothelial cells in the trunk. Whole-mount staining for PECAM1 (D,F, green channel) and ß-gal (D,E, red channel) of E8.5 progeny of a Tie2-Cre X R26R lacZ reporter cross shows that ß-gal expression, reflecting Tie2-Cre activity (E), is restricted to endothelial cells of the dorsal aorta and intersomitic vessels (D-F, arrows), as seen in merged image (D, yellow). (G-L) EphrinB2 mRNA is still expressed in the somites of the conditional ephrinB2 knockout, as revealed by in situ hybridization with ephrinB2 (G-I) and Flk1 (J-L) RNA probes in E9 ephrinB2lacZ/+ control (G,J), ephrinB2lacZ/lacZ mutant (H,K) and conditional ephrinB2 knockout (I,L) embryos. Somite expression of ephrinB2 in caudal half of somites (G-I, arrowheads) is completely lost in conventional knockout (H) but is still present in the vessel-specific knockout (I), compared with control embryos (G). Insets in G-L confirm ephrinB2 expression in ISVs of ephrinB2lacZ/+ control embryos (G versus J, arrows), and its absence from these vessels in conventional (H versus K, arrows) and conditional mutants (I versus L, arrows); S, somite; V, vessel. (M-O) Absence of intersomitic vessel remodeling at E9.5 is revealed by whole-mount staining for PECAM1 in control ephrinB2lacZ/+ (M), ephrinB2lacZ/lacZ (N) and conditional knockout (O) embryos. The ISV network is fused dorsally in both conventional (N) and conditional (O) ephrinB2 mutants, when compared with the elaborated network in control embryos (compare N and O with M). ISV guidance appears normal in both conventional and conditional mutant embryos (M-O, arrowheads). Images in M,N are close-ups of embryos in Fig. 3F,B, respectively.

View larger version (115K): [in a new window]   Fig. 8. Defective heart morphogenesis in both ephrinB2 conventional and conditional mutant embryos. (A-C) EphrinB2 is primarily expressed in the endocardial lining (arrows) of the heart, as revealed by sections of E9.5 ephrinB2lacZ/+ embryos stained for PECAM1 (A,C, green channel) and ß-gal (A,B, red channel). Some non-endocardial staining is seen (A-C, red channel not overlapping with green, arrowheads). (D-F) In the heart, restriction of Tie2-Cre activity to the endocardial lining is revealed by E9.5 sections of Tie2-Cre X R26R lacZ reporter cross embryos double-labeled for PECAM1 (D,F, green channel) and ß-gal (D,E, red channel). A merged image (D) shows Tie2-Cre activity is found only in the PECAM1-positive endocardial lining (arrows). (G-I) EphrinB2 conventional and conditional embryos show swelling and defective looping of the heart compared with control ephrinB2lacZ/+ embryos (compare H and I with G), revealed by whole-mount anti-PECAM1 staining at E9.5. (J-L) PECAM1 stained sections of hearts reveal little or no myocardial trabeculation (arrowheads) in both the conditional and conventional mutants (compare L with K) compared with control embryos (compare K and L with J, arrowheads) at E9.5.