First published online 20 November 2007
doi: 10.1242/jcs.013169
Journal of Cell Science 120, 4269-4277 (2007)
Published by The Company of Biologists 2007
Compensatory signalling induced in the yolk sac vasculature by deletion of TGFβ receptors in mice
Rita L. C. Carvalho1,*,
Fumiko Itoh2,*,
Marie-Jose Goumans3,
,
Franck Lebrin1,
Mitsuyasu Kato2,
Satoru Takahashi4,
Masatsugu Ema4,
Susumu Itoh2,
Marga van Rooijen1,
Philippe Bertolino3,
Peter ten Dijke3,5 and
Christine L. Mummery1,
1 Hubrecht Laboratory, Netherlands Institute for Developmental Biology, Uppsalalaan 8, 3584CT Utrecht, The Netherlands
2 Department of Experimental Pathology, Graduate School of Comprehensive Human Sciences, University of Tsukuba, 1-1-1, Tennodai, Tsukuba, Ibaraki 305-8575, Japan
3 Division of Cellular Biochemistry, The Netherlands Cancer Institute, Plesmanlaan 121, 1066CX Amsterdam, The Netherlands
4 Department of Anatomy and Embryology, Graduate School of Comprehensive Human Sciences, University of Tsukuba, 1-1-1, Tennodai, Tsukuba, Ibaraki 305-8575, Japan
5 Leiden University Medical Center, Department of Molecular Cell Biology, Leiden, The Netherlands

View larger version (108K):
[in this window]
[in a new window]
|
Fig. 1. Morphology analysis of embryos obtained from Tie1-Cre-TβRIIfl/fl and Tie1-Cre-ALK5fl/fl mice. (A-C) Tie1-Cre-ALK5fl/fl littermate embryos collected at E9.5. (A) Wild type (WT); (B) Tie1-Cre-ALK5fl/fl embryo showing defects in the heart; (C) Tie1-Cre-ALK5fl/fl embryo that is approximately 1-day retarded. (D-F) Tie1-Cre-TβRIIfl/fl littermate embryos collected at E9.5. (D) Wild type; (E) Tie1-Cre-TβRIIfl/fl embryo showing defects in the heart and branchial arch; (F) Tie1-Cre-TβRIIfl/fl embryo that is approximately 1-day retarded. a, allantois; ba, branchial arch; ht, heart; ov, optic vesicle; s, somites. Scale bars, 0.5 mm.
|
|

View larger version (93K):
[in this window]
[in a new window]
|
Fig. 2. Morphology of embryos obtained from SM22-Cre-TβRIIfl/fl mice. (A-D) SM22-Cre-TβRIIfl/fl littermate embryos collected at E9.5. (A) Wild type (WT); (B) wild type stained for β-galactosidase (blue); (C) SM22-Cre-TβRIIfl/fl embryo; (D) SM22-Cre-TβRIIfl/fl embryo stained for β-galactosidase (blue). (A,C) Arrowheads indicate the well-developed vasculature in the yolk sac. (E-H) SM22-Cre-TβRIIfl/fl littermate embryos collected at E12.5. (E) Wild type; (F) SM22-Cre-TβRIIfl/fl embryo and yolk sac showing vasculature defects when compared with wild type (arrowhead in E); (G) SM22-Cre-TβRIIfl/fl embryo exhibiting defects in the heart; (H) SM22-Cre-TβRIIfl/fl embryo approximately 1-day retarded. (H) Arrowhead indicates underdeveloped heart. e, eye; ht, heart; s, somites. Scale bars, 0.5 mm.
|
|

View larger version (107K):
[in this window]
[in a new window]
|
Fig. 3. ALK5 knock-in mice show an embryonic lethality similar to ALK5–/– but express PSMAD1 protein. (A,B) Embryos from conventional ALK5–/– mice (KO) die in mid-embryogenesis. At E9.5, the yolk sac from ALK5–/– mice shows impaired blood vessel development when compared with yolk sac from wild-type (WT) mice (arrow). The embryos exhibit cardiac effusion (arrowheads). (C,D) Similarly, ALK5(D266A) knock-in (KI) embryos show defects in the yolk sac and in heart development. (E-J) PSMAD1 (PS1) and PSMAD2 (PS2) were detected by immunohistochemistry in both the endothelial (EC, black arrowhead) and mesothelial (mes, red arrowhead) cell layer of wild-type (E,F), ALK5–/– (G,H) and ALK5(D266A) knock-in (I,J) yolk sacs. In the yolk sac from ALK5–/– mice, PSMAD1 was not detected in the endothelial layer (G), but PSMAD2 was observed (H). Scale bars, 0.5 mm.
|
|

View larger version (41K):
[in this window]
[in a new window]
|
Fig. 4. SMAD2 phosphorylation and SMA in yolk sacs from wild-type and ALK5–/– embryos. (A) PSMAD2 (PS2) was detected by immunohistochemistry in both the endothelial (EC) and mesothelial (mes) cell layer of wild-type (WT) yolk sacs. (B,C) In the Tie1-Cre-ALK5fl/fl yolk sac, PSMAD2 was not detected in either of the cell layers (B), but phosphorylation of SMAD2 was observed in yolk sacs from ALK5–/– embryos (C). (D) Smooth muscle actin expression in wild-type yolk sacs. (E,F) In the Tie1-Cre-ALK5–/– yolk sac, SMA was not detected (E), but it was observed in ALK5–/– yolk sacs (F). (G,H) Real-time reverse-transcriptase (RT)-PCR analysis of ALK4 RNA expression in wild-type, ALK5–/– (G) and Tie1-Cre-ALK5–/– (H) yolk sacs. Expression of ALK4 RNA was compared to β-actin (as loading control) in wild-type and mutant yolk sacs. The number of yolk sacs analyzed is shown (n). Contamination with genomic DNA was not evident in samples without RT (not shown). The Wilcoxon test was used for statistical analysis of significance; medians are shown as a black horizontal line (*P<0.05, **P<0.2). EC, endothelial cell layer; end, endoderm; mes, mesothelial cell layer. Scale bars, 0.5 mm.
|
|

View larger version (41K):
[in this window]
[in a new window]
|
Fig. 5. ALK4 co-localizes with VSMCs in yolk sacs from ALK5–/– embryos. (A-F) Cryosections of yolk sacs from wild-type embryos at E9.5 stained with antibodies against ALK4 (B,E; red) in combination either with PECAM as a marker of ECs (A; green) or SMA , as a marker of VSMCs (D; green). The overlays (C,F) show that ALK4 is expressed in VSMCs (F) and, to a lesser extent, in ECs. (G) The number of SMA -positive cells compared to the number of endoderm cells, as an internal control, in individual yolk sacs (*P<0.05). EC, endothelial cell layer; end, endoderm; mes, mesothelial cell layer.
|
|

View larger version (56K):
[in this window]
[in a new window]
|
Fig. 8. Schematic representation of TGFβ signalling in ECs and VSMCs in TGFβ-receptor mutant embryos. See key (bottom) for details. (A) In wild-type embryos, SMAD2 is phosphorylated by ALK5 in both ECs and VSMCs, whereas SMAD1 is phosphorylated by ALK1 in ECs. (B) When ALK5 is deleted specifically in ECs, PSMAD2 is no longer detected in either cell type. Additionally, PSMAD1 is also lost, because ALK1 activity requires ALK5. (C) In ALK5–/– yolk sacs, ALK4 is upregulated in both ECs and VSMCs; this rescues SMAD2 phosphorylation, but ALK1 is still not able to phosphorylate SMAD1. (D) However, in the presence of the ALK5 mutant (D266A), ALK1 is able to phosphorylate SMAD1 in ECs. In VSMCs, SMAD2 is again phosphorylated via the upregulation of ALK4.
|
|
© The Company of Biologists Ltd 2007