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

First published online 16 August 2006
doi: 10.1242/dev.02519

This Article Summary Full Text Full Text (PDF) Supplementary Material 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 Rivera-Feliciano, J. Articles by Pu, W. T. Search for Related Content PubMed PubMed Citation Articles by Rivera-Feliciano, J. Articles by Pu, W. T.

Development of heart valves requires Gata4 expression in endothelial-derived cells

José Rivera-Feliciano¹, Kyu-Ho Lee^2,*, Sek Won Kong², Satish Rajagopal², Qing Ma², Zhangli Springer², Seigo Izumo³, Clifford J. Tabin¹ and William T. Pu^2,

¹ Department of Genetics, Harvard Medical School, Boston, MA 02115, USA.
² Department of Cardiology, Children's Hospital and Department of Pediatrics, Harvard Medical School, Boston, MA 02115, USA.
³ Cardiovascular Division, Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA 02215, USA.

View larger version (102K): [in a new window]   Fig. 1. Endocardial-restricted inactivation of Gata4. (A) Structure of wild-type (wt), hypomorphic (H), floxed, and deleted (ex2) Gata4 alleles. (B) Fate mapping of T2Cre-recombined endothelial cells. (B1,B2) In T2Cre+; R26RstoplacZ embryos at E9.5, ß-galactosidase reporter expression (blue) was seen in the endocardium (arrow) and EC mesenchyme (arrowhead) of the AVC, and in the OT endothelium. (B3) By E11.5, the ß-galactosidase-positive cells populated the AV cushions (asterisks). Endothelial-derived OT cushion mesenchyme (yellow arrow) was largely confined to the most proximal portion of the OT cushion. Most of the outflow cushion was not recombined by T2Cre (green arrow). (C) Endocardial-restricted inactivation of Gata4 by T2Cre at E9.5. Control (C1,C2) and Gata4T2del (C3,C4) embryos, hybridized to a Gata4 exon 2-specific probe (red pseudocolor). Blue, DAPI counterstain. The Gata4 exon 2 in situ hybridization signal was present in the endocardium of control embryos but not of mutant embryos (white arrowheads). Gata4 expression in the myocardium (yellow arrowheads) and proepicardium (star) was unaffected. (D) Inactivation of Gata4 in endothelium and endothelium-derived cells at E11.5. In control embryos (D1), Gata4 exon 2 in situ hybridization signal was present in the endocardium (white arrowhead), myocardium (yellow arrowhead), and AV and OT cushion mesenchyme. In mutant embryos (D2), Gata4 was not detected in endothelium (white arrowhead) or AVC mesenchyme. Strong signal was still present in mid and distal OT cushion mesenchyme (green arrow) and myocardium (yellow arrowhead). a, atria; v, ventricle; ot, OT. Scale bars in C and D: 100 µm.

View larger version (105K): [in a new window]   Fig. 2. Phenotype of embryos with endothelial-restricted inactivation of Gata4 (Gata4T2del). (A) Gata4T2del embryos die by E12.5-E13.5. The incidence of Gata4T2del embryos is plotted against gestational age; expected incidence was 25%. Numbers indicate total number of embryos genotyped. (B) Gross appearance of E12.5 control and Gata4T2del littermates. Mutants showed liver hypoplasia (L) and variable growth retardation, peripheral hemorrhage and pericardial effusion (arrowhead). (C) E12.5 Gata4T2del embryo with severe pericardial effusion (arrowhead). Scale bars: 50 µm. (D-G) Formation of AV cushion mesenchyme requires Gata4 activity in endothelial-derived cells. By E10.5, the AV cushions of control embryos were populated by numerous mesenchymal cells (asterisk, D), whereas in Gata4T2del embryos the AV cushions were severely hypocellular (asterisk, F). The endocardium overlying the AV cushions became several cell layers thick in mutant embryos (arrow, G), whereas it was one cell layer thick in control embryos (E).

View larger version (134K): [in a new window]   Fig. 3. Fate map of T2Cre-recombined cells in Gata4T2del hearts. (A-D) T2Cre-mediated activation of the R26RstoplacZ reporter (blue) was mapped in E11.5 control (A,C) or Gata4T2del (B,D) embryo hearts. Whole-mount embryos were X-gal stained, then sectioned in the sagittal plane. Sections through the AV (A,B) and OT (C,D) cushions are shown. In control hearts, T2Cre-recombined cells of endothelial origin gave rise to the AV cushion mesenchyme (asterisk), as well as the proximal tip of the OT cushion (yellow arrow). The mid- and distal OT cushions were not recombined by T2Cre (red arrows). In Gata4T2del embryos, the corresponding regions were markedly hypocellular. Residual cells in the AV cushions were descended from T2Cre-recombined endothelium (arrowhead). Scale bar: 100 µm.

View larger version (48K): [in a new window]   Fig. 4. Endocardial Gata4 activity is required for mesenchyme formation in AV cushion explants. (A,B) Gata4T2del AVC explants produced markedly fewer mesenchymal cells than control explants. (C) Quantitation of the number of cells with mesenchymal morphology from control and Gata4T2del explants (n=3). (D,E) Failure of endocardial activation in Gata4T2del AV explant cultures. In control embryos, Pecam-expressing endothelial cells lost cell-cell contacts and adopted a rounded shape (arrow). Some rounded cells expressed SMA (white arrowhead) and acquired a mesenchymal morphology (yellow arrowhead). In mutant explants, endothelial cells remained in a sheet (star) and did not adopt a rounded shape. No mesenchymal cells or cells positive for both Pecam and SMA were detected (n=4). Myocardium (M) of control and mutant explants stained positive for SMA.

View larger version (63K): [in a new window]   Fig. 5. Expression of myocardial and endocardial markers in Gata4T2del mutant hearts. (A,B) In situ hybridization for Nppa (A) and Tbx2 (B) in E9.5 embryos, showing intact expression in Gata4T2del hearts. Asterisk indicates the EC. (C) In situ hybridization for Notch1 in E10.5 embryos, showing unperturbed expression in AV endocardium (arrowheads). (D) Alcian Blue staining of acidic glycosaminoglycans, including hyaluronic acid, in E10.5 control and mutant cushions. (E,F) Cdh5 immunostaining of E9.5 embryo sections. Note the bunched up phenotype of the mutant AV endocardium compared with control AV endocardium (arrowheads). Panels in F are higher magnifications of the sections shown in E.

View larger version (41K): [in a new window]   Fig. 6. Gata4 regulates Erbb3 expression. (A) qRT-PCR of control and mutant AVC samples. Gene expression was normalized to Gapdh. *P<0.05, n=3 per group. (B) In situ hybridization demonstrating downregulation of Thbs1, Plxnc1 and Erbb3 in Gata4T2del EC. Arrowheads indicate AV cushions. (C) Downregulation of Erbb3 protein in AVC whole cell lysates by western blotting. Each sample contained AVCs pooled from 10 hearts. Expression was quantitated by densitometry and normalized to Gapdh. (D) Adenoviral expression of Gata4 upregulated Erbb3 in HUVEC cells, compared with GFP expression from a control virus. Erbb3 expression was measured by qRT-PCR and normalized to Gapdh. The graph shows the average of three independent experiments. (E) BT20 cells were co-transfected with an EGFP expression vector and either Gata4DBD-engrailed or an empty expression vector. After sorting for transfected (GFP-expressing) cells, Erbb3 expression was measured by qRT-PCR and normalized to Gapdh. Results are representative of two independent experiments. (F) Constructs containing Erbb3 promoter (-1356 to +214) and intron 1 enhancer (+104 to +3076) sequences driving luciferase were cotransfected with Gata4 or Gata4ex2 expression constructs into BT20 cells. Luciferase activity was normalized to pRL-null. Results are the average of five (enhancer) or eight (promoter) independent experiments (*P<0.05).

View larger version (82K): [in a new window]   Fig. 7. Gata4 regulates an Erbb3-Erk pathway required for EMT. (A) E9.5 control or Gata4T2del embryos were cultured in M199 containing 100 ng/ml heregulin. A phosphospecific Erk antibody (pErk; green staining) showed defective Erk ativation in the endocardium of mutant embryos (arrowheads, bottom panel) compared with control embryos (arrowheads, middle panel). Myocardial pErk levels were unchanged between genotypes. To demonstrate antibody specificity, a control embryo was incubated with U0126, a selective inhibitor of Erk activation (top panel). (B) U0126 (10 µM) blocked the formation of mesenchymal cells in wild-type AV explants. (C) Synthetic phenotype resulting from the combination of reduced Gata4 expression and partial Erk inhibition. Explants from embryos homozygous for a hypomorphic Gata4 allele (H/H) underwent EMT at the same rate as controls in the presence of vehicle (NS, no significant difference). However, in the presence of U0126 at a 50% inhibitory dose, EMT was strongly inhibited in hypomorphic explants (*P<0.05 compared with control-treated H/H or U0126-treated wt/wt; n=4). The extent of EMT was measured by the number of cells with mesenchymal morphology that invaded the collagen gel.

View larger version (124K): [in a new window]   Fig. 8. AV valve remodeling requires Gata4 expression in endocardium and its derivatives. Gata4T2del/Ki (Gata4Ki/flox; Tie2Cre+) embryos are deficient for Gata4-Fog interaction in the endocardium and its derivatives. (A-D) AV cushions were normally populated by mesenchymal cells at E12.5 (asterisks, C,D), indicating that the Gata4-Fog interaction is not essential for EMT. The liver (L) was normal in size and morphology. Boxed regions of A,B are enlarged in C,D. (E,F) A common atrioventricular canal (asterisk) was present in all Gata4T2del/Ki embryos at E14.5, whereas in control embryos the AVC was septated by the fusion of the superior and inferior cushions. (G) Decreased proliferation of cushion mesenchyme in Gata4T2del/Ki embryos. Proliferation was measured as the percentage of nuclei that stained positive for BrdU (red). Nuclei were stained with DAPI. Myocyctes were labeled with desmin (green). Three sections from three different E12.5 embryos were examined per group by a blinded observer (*P<0.05).