Transcriptional activation of BMP-4 and regulation of mammalian organogenesis by GATA-4 and -6

Available online at www.sciencedirect.com R

Developmental Biology 254 (2003) 131–148

www.elsevier.com/locate/ydbio

Transcriptional activation of BMP-4 and regulation of mammalian organogenesis by GATA-4 and -6 Georges Nemer and Mona Nemer* Laboratoire de de´veloppement et diffe´renciation cardiaques, Institut de recherches cliniques de Montre´al (IRCM), and De´partement de Pharmacologie, Universite´ de Montre´al, 110 des Pins Ouest, Montre´al Quebec, Canada H2W 1R7 Received for publication 15 May 2002, revised 21 October 2002, accepted 23 October 2002

Abstract Transcription factors GATA-4, -5, and -6 constitute an evolutionary conserved subfamily of vertebrate zinc finger regulators highly expressed in the developing heart and gut. Genetic evidence suggests that each protein is essential for embryonic development, but their exact functions are not fully elucidated. Moreover, because all three proteins share similar transcriptional properties in vitro, and because transcripts for two or more GATA genes are present in similar tissues, the molecular basis underlying in vivo specificity of GATA factors remains undefined. Knowledge of the exact cell types expressing each protein and identification of downstream targets would greatly help define their function. We have used high-resolution immunohistochemistry to precisely determine the cellular distribution of the GATA-4, -5, and -6 proteins in murine embryogenesis. The results reveal novel sites of expression in mesodermal and ectodermal cells. In particular, GATA-4 and -6 expression was closely associated with yolk sac vasculogenesis and early endoderm–mesoderm signaling. Additionally, GATA-6 was strongly expressed in the embryonic ectoderm, neural tube, and neural crest-derived cells. This pattern of expression closely paralled that of BMP-4, and the BMP-4 gene was identified as a direct downstream target for GATA-4 and -6. These findings offer new insight into the function of GATA-4 and -6 during early stages of embryogenesis and reveal the existence of a positive cross-regulatory loop between BMP-4 and GATA-4. They also raise the possibility that part of the early defects in GATA-4 and/or GATA-6 null embryos may be due to impaired BMP-4 signaling. © 2003 Elsevier Science (USA). All rights reserved. Keywords: GATA factors; Embryonic development; Heart; Vasculogenesis; BMP-4

Introduction GATA proteins are zinc fingers, DNA-binding transcription factors that play critical roles in cell differentiation and embryogenesis. In vertebrates, six members of the GATA family have been identified and subdivided into two subgroups, based on sequence homology and expression pattern (Lowry and Atchley, 2000). The first subgroup includes GATA-1, -2, and -3, which are present mainly in hematopoietic cells where they play essential nonredundant functions (Weiss and Orkin, 1995). Transcripts for GATA-4, -5, and -6 are present predominantly in cells of the heart and the digestive system as well as in the extraembryonic endoderm

* Corresponding author. Fax: ⫹514-987-5575. E-mail address: [email protected] (M. Nemer).

(Arceci et al., 1993; Gre´pin et al., 1994; Heikinheimo et al., 1994; Jiang and Evans, 1996a; Laverriere et al., 1994; Morrisey et al., 1996; White and Clark, 1993). Within these tissues, expression of two or more GATA genes is often detected, raising the possibility that they may have redundant functions. This was further supported by the finding that all three proteins bound to similar DNA elements on numerous cardiac (Charron et al., 1999; Molkentin et al., 1994; Morrisey et al., 1996; Nemer et al., 1999) and some gastric (Gao et al., 1998; Nishi et al., 1997; Tamura et al., 1993) promoters and potently enhanced their activity. Moreover, ectopic expression of each gene in Xenopus oocytes resulted in upregulation of cardiac markers (Jiang and Evans, 1996b), while downregulation of either GATA-4 or GATA-6 proteins in cardiomyocytes similarly inhibited several cardiac genes (Charron et al., 1999). Despite these

0012-1606/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0012-1606(02)00026-0

Fig. 1. Specificity of the anti-GATA-4, -5, and -6 antibodies. (A) Aminoacid homology between the GATA-4, -5, and -6 proteins. Note how the C-terminal domain is the least conserved region across the different family members. Antibodies were directed against this region. (B) Transient expression in CV-I cells of GATA-4, -5, and -6 was detected by immunohistochemistry using the respective antibodies (␣G4, ␣G5, and ␣G6). All pictures were taken at a 20⫻ magnification.

Fig. 2. Expression of the GATA-4 and -6 proteins during early murine development. Serial consecutive transverse sections of mice embryos at e7.5 showing differential cellular expression of GATA-4 and GATA-6 in early embryonic and extraembryonic tissues. (A) Both proteins are expressed in precardiac mesoderm (open arrow). Little GATA-4 is detected in the embryonic endoderm (solid arrow) and it is restricted to cells underlying the precardiac mesoderm, whereas GATA-6 labels many more endodermal and mesodermal cells. In addition, GATA-6 is expressed in the embryonic ectoderm (*). The strongest expression of both proteins is in the extraembryonic endoderm. Note how GATA-6 is expressed in both visceral (VE) and parietal endoderm (PE), whereas GATA-4 is more prominent in VE. The inset is a higher magnification (40⫻) showing labeling in blood islands (BI) and mesothelial cells (MC). (B) Other serial sections of e7.5 embryo showing the allantois where GATA-4- and -6-positive cells are detected. Note the difference in the cells staining for GATA-4 and -6, both in the allantois and in the VE. In the low magnification (10⫻), cells stained with the anti-GATA-6 antibody are clearly visible in the embryonic ectoderm and mesoderm as well as in the amnion, whereas GATA-4-positive cells are restricted to the lateral plate mesoderm.

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

133

134

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

predictions, inactivation of the GATA-4, -5, and -6 genes in mice resulted in unique phenotypes arguing for specialized nonredundant functions for each GATA factor. In particular, inactivation of the GATA-4 gene caused defects in ventral morphogenesis and heart tube formation, leading to developmental arrest between embryonic days 7 and 9.5 in the majority of homozygous embryos, while 30 – 40% failed to gastrulate (Kuo et al., 1997; Molkentin et al., 1997). GATA-4 null embryonic stem cells were defective in visceral endoderm differentiation, suggesting an essential role for GATA-4 in extraembryonic tissues (Narita et al., 1997; Soudais et al., 1995). An intrinsic role for GATA-4 in heart morphogenesis was also demonstrated in mice homozygous for a mutation in GATA-4 that disrupts interaction with its cofactor, FOG-2 (Crispino et al., 2001). While many cardiac genes were shown to be activated in vitro by GATA-4, the downstream transcription targets of GATA-4 during early stages of embryogenesis remain largely unknown. In addition to GATA-4, GATA-6 also appears to be essential for visceral endoderm differentiation, and inactivation of the GATA-6 gene resulted in early embryonic lethality (e5.5– e7.5) and differentiation defects in the ectoderm and part of the visceral endoderm, where HNF4 was identified as a GATA-6 target gene. The reason for the observed apoptosis in the ectoderm remains elusive, given that GATA-6 expression was not reported therein (Koutsourakis et al., 1999; Morrisey et al., 1998). In contrast, targeted mutation of the GATA-5 gene in mice did not interfere with embryonic or heart development, but resulted in mild defects in female genitourinary tract (Molkentin et al., 2000), making GATA-5 the only mammalian GATA gene that is dispensable for embryonic development (Pandolfi et al., 1995; Pevny et al., 1991; Tsai et al., 1994). This phenotype was unexpected, given that, in zebrafish, GATA-5 is required for survival and differentiation of myocardial and endocardial precursors, and mutations in the GATA-5 gene lead to the faust mutant characterized by abnormal heart development (Reiter et al., 1999). While genetic approaches in different species clearly established the essential role of GATA factors for embryonic development, the molecular basis underlying their functions and their specificity have remained unclear. In particular, the inability of different GATA factors to compensate for each other in vivo remains unexplained. It may reflect their cross-regulation (Kuo et al., 1997; Molkentin et al., 1997; Morrisey et al., 1998), the presence of two or more GATA factors in common regulatory pathways (Charron et al., 1999), or differential interaction with cofactors (Durocher et al., 1997; Morin et al., 2000). Alternatively, the proteins may not be expressed as presumed from the in situ hybridization studies. Posttranscriptional and posttranslational regulation of transcription factors is widely documented as best exemplified by the presence of caudal mRNA throughout the embryo, whereas the protein is only detected at the posterior end (Mlodzik and Gehring, 1987). Other examples of posttranscriptional regulation include

alternative or defective splicing, resulting in no protein or in isoforms with different activities or cellular localization (Belaguli et al., 1999; Condie et al., 1990; Martin et al., 1994; Mumberg et al., 1991). It is also possible that the different GATA factors may be expressed in a spatially distinct manner within the same tissue. The existence of heterogeneity in GATA-4- and GATA-6-expressing cells is suggested from the heterogeneous distribution of GATA-6positive cells in the 3.5-dpc mouse inner cell mass (Koutsourakis et al., 1999) and the absence of GATA-6 in P19derived cardiomyocyte precursors (Gre´ pin et al., 1997). Thus, precise information on cellular distribution of GATA-4, -5, and -6 proteins and identification of downstream targets/effectors are prerequisite to elucidating their mechanisms of action and their in vivo functions. In this paper, we have analyzed the expression of the GATA-4, -5, and -6 proteins at the single cell level by using high-resolution immunohistochemistry with specific antibodies. The results suggest that GATA-4 and GATA-6 likely play critical roles in early endoderm–mesoderm signaling as revealed from their complementary expression pattern in several extraembryonic structures closely associated with yolk sac vasculogenesis, such as the visceral endoderm, the allantois, blood islands, and mesothelial cells. Additionally, GATA-6 was present from the earliest developmental stages in the ectoderm and in neuroectodermal derivatives, raising the possibility that it may have a previously unexplored function in neurogenesis. Within the embryo, the three proteins differentially labeled several cells of the cardiovascular and digestive systems, implicating them in the control of cardiovascular and metabolic homeostasis. BMP-4, an activator of GATA-4 whose expression closely parallels that of GATA-4 in mesoderm cells, was identified as a GATA-4 transcription target, suggesting that, at the least, some of the early developmental defects in the GATA-4 null embryos may be due to impaired BMP-4 signaling.

Materials and methods Antibodies used Anti-GATA-5 and anti-GATA-6 antibodies were prepared as follow. cDNAs corresponding to amino acids 300 – 402 of the mouse GATA-5 protein and amino acids 380 – 441 of the mouse GATA-6 protein were fused to the GST sequences in the pGEX-4T3 vector (Pharmacia 27-458301). The recombinant, Coomassie blue-stained GATA-5 and GATA-6 fusion protein bands were eluted electrophorically from an SDS–PAGE, emulsified with complete Freund’s adjuvant, and injected into New Zealand white rabbits. Another injection with incomplete Freund’s adjuvant was given 21 and 42 days after the first injection. Two weeks after the last injection, heart puncture was performed on the animals and serum was collected. Immunoglobin

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

135

purification was done by using Protein A-Sepharose CL-4B (Pharmacia 17-0780-01). The anti-GATA-3, GATA-4, FOG2, SRF, Nkx2.5, dHand, HNF1, and PCNA antibodies were purchased from Santa Cruz Biotechnology (catalog numbers, resepectively: sc-268x, sc1237x, sc-10755x, sc335x, sc-8697x, sc-9411x, sc8986x, and sc-56). The biotinylated anti-mouse, rabbit, and goat IgGs antibodies were purchased from Vector (catalog numbers: BA2000, BA1000, and BA5000). The Rhodamine Avidin D and Fluoroscein Avidin D antibodies were from Vector (catalog numbers: A3562, A2002, and A2001). The mouse alkaline phosphatase and Glucagon antibodies were obtained from Sigma (A3562 and G2654), and the smooth muscle monoclonal antibody was from Dako (M0851). The neurofilament antibody (2H3) was a kind gift from Dr. T.M. Jessell.

generated as described previously (Gre´ pin et al., 1997). Rat neonatal cardiomyocytes were infected with an adenovirus overexpressing GATA-4 as described in Charron et al. (2001).

Nuclear extracts of neonatal cardiomyocytes and 293 cells transiently overexpressing GATA-4 or GATA-6 were obtained as described previously (Charron et al., 1999). EMSA was carried out as previously described (Gre´ pin et al., 1994) by using 3–5 ␮g of nuclear extracts. The probe used for GATA binding corresponded to the ⫺232 GATA site of the mouse BMP-4 promoter (5⬘-GTTTCATTATCTCTAATT-3⬘).

DNA transfections

RNA extraction and PCR analysis

The GATA-4, -5, and -6 entire coding sequences were subcloned in CMV-driven expression vector (pCDNA3) as previously described (Charron et al., 1999; Nemer et al., 1999). CV1 cells were maintained in culture and transfected as described previously (Gre´ pin et al., 1997). For immunohistochemistry, 30,000 cells per well were plated in 12-well dishes and transfected with 10 ␮g of expression vector. Forty-eight hours after transfection, cells were washed twice with PBS and then fixed at room temperature for 15 min with a 4% paraformaldehyde solution. Cells were quenched for endogenous peroxidase activity by treating them for 5 min with 3% H2O2 and were then incubated with primary antibodies overnight at 4°C. Cells were washed 3 times with PBS containing 0.2% Tween, and then incubated with secondary antibodies (anti-goat peroxidase conjugate A9452 and anti-rabbit peroxidase conjugate A6154 from Sigma) at a dilution of 1/250 for 1 h at room temperature. Revelation was carried out by using the 3,3⬘-Diamino-Benzidine (DAB) chromagen (Sigma D-5637) in 0.05 M Tris, pH 7.4, for 5 min at room temperature. Cotransfections of GATA-expressing vectors with the BMP-4/luc constructs were carried out as previously described (Nemer et al., 1999). The ⫺1.3-kbp mouse BMP-4 promoter fused to luciferase was a kind gift from Dr. H. Stephen (Feng et al., 1995). Mutations of the GATA elements in the context of the ⫺1.3-kpb promoter were generated by using PCR-mediated site-directed mutagenesis. All constructs were confirmed by DNA sequencing. The ⫺95-bp promoter was generated from the longer construct by using SmaI. NIH3T3 cells were transfected by using the calcium phosphate precipitation method. Briefly, 30,000 cells per well were plated on a 12-well plate. A total of 1 ␮g of reporter gene was used per well, and total DNA was kept constant at 2 ␮g. Luciferase activity was measured 36 h after transfection by an LKB illuminometer. The results are the mean of three independent experiments, each done in duplicate. GATA-4 stably transfected P19 and C2C12 cells were

Total cellular RNA was extracted according to the thiocyanate–phenol– chloroforme method. cDNAs were generated from 5 ␮g of total RNA by using an oligonucleotide dT12–18 in the presence of AMV-RT (Promega). Semiquantitative PCR was conducted by using specific oligonucleotides for each gene, and a dose-response assay was carried out to determine the optimal amount of cDNA to be used for PCR amplification by using the following: 3 min at 94°C, 30 s at 94°C, 30 s annealing temperature for each oligonucleotide pair, and 1 min/kb at 72°C repeated 29 cycles. Amplification of tubulin was used as an internal control. PCR products were resolved on 1.2% agarose gels. The analysis was carried out in duplicate with RNA isolated from at least two different experiments. The sequences of the oligonucleotides for GATA-4 and tubulin were previously described (Nemer and Nemer, 2002). For BMP- 4, the following oligonucleotides were used: forward primer, 5⬘ACCTCAACTCAACCAACC-3⬘, and reverse primer, 5⬘GTGTGGTATGTGTAGGTGG-3⬘.

Gel shift assays

Immunohistochemistry on tissue sections Embryos were dissected from CD-1 mice (Charles Rivers, Canada) at different stages of gestation and fixed overnight with 4% paraformaldehyde solution and then embedded in paraffin and sectioned 5 ␮m thick. Rehydration was carried out as described previously (Viger et al., 1998), and incubation with primary antibodies was done as mentioned above. Coupling to a biotinylated secondary antibody (biotinylated anti-rabbit BA-1000 and biotinylated anti-goat BA-5000 from Vector Laboratories) was done for 1 h at room temperature at a dilution of 1/250 followed by three washes in PBS, 0.2% Tween. Sections were then incubated for 45 min at room temperature with a Streptavidin–HRP conjugate (NEL750 from NEN Life Science) at a dilution of 1/500. Revelation was done as described above. Some sections were counterstained with Methyl-green as previously described (Paradis et al., 2000).

Fig. 3. GATA-4, -5, and -6 in the developing heart. Serial sagittal sections of dissected embryonic heart (A) and frontal sections (B) of embryos at e11.5. (A) GATA-4 and GATA-5 are highly expressed in the endocardial cushions (EC). Note the high expression of GATA-5 in the endocardium and some underlying myocardial cells. GATA-4 expression is detected both in the endocardium and myocardium. Upper panels photos are taken at 10⫻ magnification of the original size, while the lower panels are at 40⫻. (B) Predominant expression of GATA-6 in the left ventricle (LV) at e11.5. Note the difference in the number of GATA-6-stained nuclei (brown color) between left and right ventricles. Staining for dHand was used as control; dHand expression is more intense in the right ventricle, where it is highly enriched in the endocardium. Photos are taken at 20⫻ magnification of their original size.

Fig. 4. Expression of GATA-4 and GATA-6 during heart morphogenesis. (A) GATA-4 and GATA-6 are expressed in the bulbis cordis (BC), common atrium (A), and sinus venosis (SV) as early as e10.5. The expression of the GATA proteins in the third branchial arch artery (3rd BA) along that of dHand and FOG2 is noteworthy. In addition, please note the expression of GATA-4 and FOG2 in the mesenchyme surrounding the lung bud (lu) and the differential distribution of GATA-4 and GATA-6 in the liver (li). GATA-6 protein was also detected in the dorsal aorta (Ao). Magnification is 5⫻ the original size. (B) Expression of GATA-4, -6, and FOG2 during valvuloseptal formation. Note the strong expression of all three proteins in the outflow tract of the left ventricle leading to the pulmonary artery (PA). GATA-6 expression follows a gradient-like pattern in the ventricles with less expression in the trabeculae compared with the ventricular wall. Note the lower staining in the atrioventricular canal regions. GATA-4 and GATA-6 are also present in the diaphragm (d) and the liver (li), while GATA-6 and FOG2 are coexpressed in the aorta (Ao). Magnification is 10⫻ the original size. (C) Coexpression of GATA-4 and -6 in the e10.5 bulbis cordis showing the nuclear staining for both. In blue, control staining for the nuclei with Hoechst, in green GATA-4, and in red GATA-6. Note how all red nuclei (GATA-6⫹) are also positive for GATA-4 (green). Magnification is 20⫻ the original size.

138

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

To analyze expression of the GATA-4, -5, and -6 proteins at the cell level, we used immunohistochemistry with specific antibodies directed at the C-terminal domain of each protein. This region is the least homologous across the different family members (Fig. 1A). Additionally, it is present in the different GATA-5 and -6 isoforms that arise from alternative splicing or alternative promoters and translation initiation sites (Brewer et al., 1999; MacNeill et al., 1997). The specificity of the antibodies was ascertained by immunohistochemistry and Western and gel shift analyses; all were found to be highly specific and to not cross-react with other members of the cardiac or hematopoietic GATA factors (Fig. 1B, and data not shown).

stage were extraembryonic mesoderm and endoderm. Close examination of serial sagital sections revealed differential distribution with GATA-4 being highly expressed in the visceral vs the parietal endoderm, while GATA-6 was present in both layers (Fig. 2A). Moreover, the two proteins were not coexpressed in all cells, suggesting cellular heterogeneity (Fig. 2A and B), a finding consistent with the reported heterogeneity for GATA-6 expression (using a lacZ knock-in in GATA-6 heterozygote embryos) in cells of the inner cell mass (Koutsourakis et al., 1999). Other extraembryonic structures also expressed GATA-4 and -6, including the allantois, where several cells showed positive GATA-4 or -6 staining (Fig. 2B); these mesodermal cells are the hemangioblasts capable of differentiating into endothelial and hemopoietic cells (Caprioli et al., 2001). In addition, there was consistent labeling of cells within the blood islands (Fig. 2A, inset). Labeling for GATA-4 and -6 was also observed in mesothelial cells surrounding the blood islands or lining the visceral endoderm between blood islands (Fig. 2A); these cells are thought to give rise to vascular smooth muscle cells of the large vitelline vessels. A few GATA-4- and GATA-6-positive cells were also consistently detected in the amnion (Fig. 2B). Together, these findings reveal a close association of GATA-4 and -6 with early endoderm–mesoderm signaling and yolk sac vasculogenesis and hemopoiesis. They also raise the intriguing possibility that early defects in vasculogenesis and/or hemopoiesis may underlie the developmental arrest of GATA-4 and/or GATA-6 null embryos.

Expression of GATA proteins in multiple extraembryonic structures

Distinct expression of GATA-4, -5, and -6 in the developing heart

The presence of immunoreactive GATA-4, -5, and -6 proteins was first analyzed at e7.5. At this stage, presumptive cardiac mesodermal cells are arranged in two bilateral primordial structures of the lateral splanchnic mesoderm. Staining for GATA-4, -5, and -6 was reproducibly detected bilaterally in the lateral plate mesoderm (Fig. 2). GATA-5 and -6 immunoreactivity extended to cells in the intermediate mesoderm, which gives rise to urogenital organs. Weak staining for GATA-4 was observed in a few cells of the intermediate mesoderm. Expression of GATA-4 in the embryonic endoderm was restricted at this stage to very few cells underlying the presumptive cardiogenic mesoderm, and GATA-5 was not reproducibly detected in the embryonic endoderm. In contrast, the GATA-6 protein was broadly distributed at this stage in the three embryonic layers: ectoderm, endoderm, and mesoderm, with notable expression in the neural tube. The presence of previously unreported GATA-6 in ectodermal cells may explain why apoptotic cells were detected in the ectoderm of the Gata6⫺/⫺ embryos, and why the ectoderm in these mice showed abnormal development (Koutsourakis et al., 1999). While GATA-4 and -6 were clearly detected in embryonic tissues, by far, the major sites of expression at this

Expression of the three GATA proteins was further examined in the developing heart starting from e8.5 just after formation of the heart tube, which, at this stage, is composed of a common atrial chamber attached to the sinus venosus and a common ventricular chamber which is continuous to the conus at the most anterior part. At e8.5, sagital sections of the heart showed positive staining for GATA-4 and -6 in atrial and ventricular myocardium with the sharpest signal at the posterior part, mainly at the junction of the atrial chamber to the sinus venosus and the septum transversum (data not shown). Following looping, and just prior to valvular septation of the heart, a few myocytes— underlying the endocardium—showed positive staining for GATA-5; however, the cells of the endocardium and endocardial cushion were the major site of GATA-5 expression (Fig. 3, right); these cells also coexpressed GATA-4 (Fig. 3, left). GATA-4 and -5 were also clearly detected in pericardial cells, which are thought to contribute to interstitial cardiac fibroblasts, coronary smooth muscle, and endothelial cells (Mikawa and Gourdie, 1996). From e9.5 to e12.5, GATA-4 and, to a lesser extent, GATA-6 were present in the myocardium, the endocardium, and the endocardial cushions (Fig. 3A and B). The two proteins

For GATA-4/GATA-6 coimmunostaining, incubation with the anti-GATA-4 antibody was carried out overnight and the signal was amplified by Fluorescein Avidin D (1/ 500). The anti-GATA-6 antibody was added the following day for an incubation time of 2 h at room temperature, and the signal was amplified by Rhodamine Avidin D (1/500). For the GATA-4/Glucagon coimmunostaining, an antimouse alkaline phosphatase antibody (1/250) was used to detect the signal for Glucagon. Incubation with 4-Nitro Blue Tetrazolium Chloride (NBT) and 5-Bromo-Indolyl-Phosphate (BCIP) (Boehringer) was carried out in the dark for 15 min to detect the blue staining.

Results

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

were also widely distributed in cells that will contribute to the different heart structures, including the third branchial arch artery, the common atria, the bulbus cordis, the sinus venosis, and the venous valves (Fig. 4A). GATA-4 protein was also abundantly expressed in the septum transversum, as well as the lung mesenchyme. Expression of the GATAinteracting protein FOG-2 (Svensson et al., 1999) was found to colocalize with GATA-4 and -6 in the myocardium and in the outflow tract and with GATA-6 in the aorta (Fig. 4B). Within the myocardium, almost all GATA-6-positive myocytes contained GATA-4, but a significant percent of GATA-4-positive cells did not colabel for GATA-6 (Fig. 4C), suggesting the existence of distinct myocyte subpopulations. Expression of other cardiac transcription factors, like MEF2C, Nkx2.5, and SRF, overlapped with that of GATA-4 and -6 only in the myocardium throughout embryonic heart development (data not shown). Interestingly, the basic helix–loop– helix transcription factor dHand strongly labeled endocardial cells which contained GATA-4 (Fig. 3B). These findings are consistent with a role for GATA factors in early septal and valvular development. Extracardiac expression of GATA-4, -5, and -6 The main extracardiac sites of expression of the GATA-4, -5, and -6 transcripts are the gonads and the endodermal cells of the gut. We examined the cellular distribution of all three GATA proteins in the developing digestive and urogenital systems. As described below, GATA-5 showed the most restricted expression in extracardiac tissues, being detectable only in the intestine, the bladder and, at low levels, in the lung. In contrast, GATA-6 was widely distributed in organs of the digestive, urogenital, and respiratory systems as well as in the brain. Nevertheless, GATA-6 marked specific cells within these organs that, at times, overlapped with GATA-4- or GATA-5-expressing cells. With respect to the digestive system, the three GATA proteins were very weakly expressed at the anterior intestinal port at e8 (data not shown). However, by e9.5– e10.5, robust expression of GATA-4 and -6, but not GATA-5, was observed in the septum transversum and the liver bud as well as in the epithelial cells of the future hindgut (Figs. 3A and 5A). Later, between e13.5 and e14.5, expression of GATA-4 and -6 marked distinct cells of the developing liver. GATA-4 was restricted to the epithelial cells lining the liver, where it colocalized with dHand. Additionally, scattered GATA-4-positive cells were detected within the liver; many were lining hepatic veins and were costained with smooth muscle actin (Fig. 5A). There was little, if any, coexpression of GATA-4 with the hepatocyte-specific transcription factor HNF1 (Nagy et al., 1994). In contrast, GATA-6 was restricted to the hepatocytes as evidenced from colocalization with HNF1 (Fig. 5B). GATA-5 was never detected in the embryonic liver. GATA-4 and -6 proteins were also expressed in the developing stomach,

139

where they marked the gastric epithelium (Fig. 5C, and data not shown). GATA-4 was also strongly expressed in an anteroposterior gradient in mesenchymal cells surrounding the junction of the stomach with the diaphragm. Later (e17.5), GATA-4 and -6 were present in the glandular but not in the mucosal gastric cells (Fig. 5C). GATA-6 protein was present, though to a lower extent, in both gastric epithelial and smooth muscle cells and in the esophagus (data not shown). GATA-4 and -6 proteins were also found in the exocrine cells of the pancreas (Fig. 5C). Finally, all three GATA proteins were present early (beginning e11.5) in the endodermal cells of the intestine, where their expression pattern was partially overlapping in the epithelial cells of the villi. GATA-6 also marked the smooth muscle cells of the intestine (Fig. 5D). Given that the endodermal epithelium contains many types of endocrine cells (Skipper and Lewis, 2000), it will be interesting in the future to determine whether the different GATA factors label specific hormonesecreting cells of the gut. In the urogenital system, the only site of GATA-5 expression was in the smooth muscle cells of the bladder (Fig. 6A) and neither the kidney, gonad, nor adrenal showed detectable GATA-5 protein at all stages examined (e14 – e17). GATA-6, but not GATA-4, was also present in the bladder, where it labeled both the epithelial and smooth muscle cells (Fig. 6A). As expected (Viger et al., 1998), GATA-4 and -6 proteins were present in the gonadal component of the genital ridge; additionally, GATA-6 was strongly expressed in the mesonephric component, where it labeled the tubules (Fig. 5B). Finally, GATA-4 and -6 were present early in adrenal cells. Cells in the adrenal cortex are derived from the splanchnic mesoderm, whereas those of the medulla have a neural crest origin. GATA-4 labeled presumed adrenal cortex cells at e13.5 that were negative for such neural crest markers as dHand and neurofilament (Fig. 6B, left). Later (e17.5), GATA-4 became restricted to the adrenal capsule (Fig. 6B, top right); this suggests a role for GATA-4 in adrenal cortical cell proliferation and is consistent with the finding of GATA-4 in murine adrenal cortical tumors and in human adrenal cortical carcinomas, but not in normal adult adrenal cortex (Kiiveri et al., 1999). In contrast, GATA-6 strongly labeled cells of the medulla, which are neural crest-derived and some cells in the cortex (Fig. 6B). These were not the only GATA-6-expressing neuronal cells; as mentioned earlier, GATA-6 immunoreactivity was detected in the neural tube as early as e7.5 (Fig. 2), and GATA-6 positive cells persisted throughout brain development, where GATA-6 colocalized with FOG2 in numerous area, including the cortex and the forebrain (Fig. 6C). Finally, the GATA-6 gene was shown to be expressed in the lung, where it plays an essential role in branching morphogenesis (Keijzer et al., 2001). We found that GATA-6 was expressed in numerous pulmonary structures, including smooth muscle cells, bronchi, mesenchyme, and arteries (Fig. 7). These findings further extend the reported

140

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

Fig. 5. Expression of the GATA-4, -5, and -6 proteins in the gut endoderm. (A) GATA-4 is expressed in the liver as early as e9.5 in scattered cells in the septum transversum, the liver bud, and the epithelium of the gut (left panel, magnification 20⫻). From e10.5 to e14.5, GATA-4 expression in the liver is restricted to the surface epithelial cells as well as to those lining the vessels. Of note, the coexpression with dHand (magnification 20⫻). GATA-4-expressing cells do not colocalize with those positive for the hepatocyte marker HNF1. Some cells expressing GATA-4 colocalize with smooth muscle actin (sAc) around the hepatic vessels. (B) The pattern of expression of GATA-6 in the liver (li) overlaps with HNF1 at e14.5. Lower magnification 10⫻, higher magnification 40⫻. Note expression in the gonadal ridge and the kidney. No correlation could be established between GATA-6 and PCNA (a marker of proliferation). (C)

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

141

In the stomach, GATA-4 is expressed in the epithelial cells at e14.5, and at later stages (e17.5), becomes confined to the glandular cells (G). No expression of GATA-4 or GATA-6 is detected in the mucosal cells (M) of the stomach at e17.5. Note the expression of both GATA-4 and GATA-6 in the embryonic pancreas (Pan); in the adult pancreas, GATA-4 and -6 are expressed in the exocrine cells, while cells staining for Glucagon (Glu) are negative for GATA-4 (20⫻). (D) Expression in the epithelial cells of the small intestine at e13.5 (top panel; 5⫻) and e17.5 (middle panel, 20⫻), and lower panel, 40⫻. Note how GATA-4- and -6-positive cells are almost complementary. GATA-4 and -5 have an only partially overlapping pattern.

142

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

Fig. 6. GATA proteins in the urogenital system and in neural ectoderm. (A) GATA-4 is absent from the bladder (e17.5), whereas GATA-5 and GATA-6 are expressed in the smooth muscle layer (black arrow). In addition, GATA-6 is expressed in the epithelial cells (open arrow) (magnification 10⫻). (B) GATA-4 is expressed in a subset of cells of the adrenal gland at el4.5 that are distinct from those expressing dHand and neurofilament (blue color) (magnification 20⫻). At later stages (e17.5), GATA-4 is expressed only in the epithelium lining the adrenal, whereas GATA-6 is expressed in cells of the cortex (C) and medulla (M) (magnification 40⫻). (C) Expression of GATA-6 in the neural tube (NT) at e11.5 through e14.5 (magnification 20⫻). Note the colocalization with FOG2 in the brain cortex (C) (magnification 40⫻).

distribution of GATA-6 transcripts in the lung (Keijzer et al., 2001) and reveal that GATA-6 is expressed in endodermally as well as mesodermally derived pulmonary cells. GATA-4 and -5 were also clearly detected in the lung but were restricted to mesodermally derived cells. For example, GATA-4 was expressed in the mesenchymal cells surrounding the lung bud of e11.5 embryos (Fig. 4); later (e14.5), its expression was restricted to the pulmonary artery and veins, where it partially overlapped with GATA-6 and SRF (Fig. 7). On the other hand, GATA-5 was restricted to the mesenchyme (data not shown). The absence of GATA-4 and -5 from the pulmonary endoderm may explain the dependence on GATA-6 for lung epithelial cell differentiation (Keijzer et al., 2001). The presence of three GATA factors in different cells of the lung points to the importance of this class

of regulators in pulmonary development and, possibly, diseases. The BMP-4 gene is a downstream GATA target As stated earlier, 30 – 40% of GATA-4 null embryos fail to gastrulate (Molkentin et al., 1997), and the remaining embryos arrest around e9.5 with defects in yolk sac development and ventral folding. Altered differentiation and/or proliferation of mesodermal precursors or defective mesoderm– endoderm signaling have been suggested as possible mechanisms underlying the phenotype of GATA-4 null embryos. Among the signaling molecules required for mesoderm differentiation and gastrulation are bone morphogenetic proteins 2 and 4 (BMP-2 and BMP-4), which are

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

143

Fig. 7. Expression of GATA factors in multiple pulmonary cell types. Sagital consecutive sections of a mouse lung at e17.5 show that GATA-4 is expressed in the smooth muscle cells of the pulmonary artery (PA), whereas GATA-6 and SRF are expressed in the PA, bronchi (B), and mesenchymal cells (M). All sections were counterstained with methyl green. Magnification 10⫻ for the left panels and 20⫻ for the right panels.

members of the TGF␤ superfamily of secreted peptides (Winnier et al., 1995; Zhang and Bradley, 1996). Interestingly, expression of BMP-2 and BMP-4 during early murine development closely parallels that of GATA-4 in the extramesodermal cells of the allantois and amnion, and in the precardiac mesoderm. We tested the possibility that BMP-4

might be a downstream target for GATA-4. The BMP-4 promoter contains two highty conserved GATA elements (Fig. 8A) and is activated in a dose-dependent manner by cotransfection with GATA-4 or GATA-6 expression vectors (Fig. 8B). Both GATA elements are able to interact specifically with recombinant GATA-4 and GATA-6 as well as

144

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

with endogenous GATA-4 present in cardiac extracts (Fig. 8C). In addition, binding and competition experiments indicate that the distal ⫺526-bp GATA element has a higher affinity for GATA-4 and -6 than the ⫺392-bp element (Fig. 8C, and data not shown). Mutation of the proximal element in the context of the full-length BMP-4 promoter significantly reduces GATA-dependent transactivation, whereas mutation of either the distal element or of both GATA sites abrogates the response to GATA-4 or GATA-6 (Fig. 8D). Ectopic expression of GATA-4 in P19 embryonic stem cells, in C2C12 myoblasts, or in cardiac myocytes, results in marked induction of endogenous BMP-4 transcripts (Fig. 8E). Together, these results indicate that BMP-4 is a direct transcription target for GATA-4 and raise the possibility that impaired BMP-4 production and mesoderm differentiation may underlie some of the early defects seen in the GATA-4 (and possibly GATA-6) null embryos.

Discussion In the present study, we have determined the spatial and temporal distribution of the GATA-4, -5, and -6 proteins at the single cell level. We also present evidence suggesting that BMP-4 is a direct GATA-4 target. The results offer new insight into the in vivo function of an important subfamily of transcriptional regulators and provide an essential framework for further analysis of the role of GATA-4, -5, and -6 in embryogenesis. Amplification of the BMP-4/GATA-4 pathway through a positive feedback loop Previous studies have shown that BMP-4 is an activator of GATA-4 and suggested that GATA-4 may mediate the cardiogenic effects of BMP-4 as well as its role in hepatic development (Schultheiss et al., 1997; Rossi et al., 2001). The data presented here reveal that GATA-4 is a transcriptional activator of BMP-4, suggesting the existence of a positive BMP-4/GATA-4 feedback loop. This loop may serve to amplify the BMP/GATA pathway in several embryonic cell types where the two proteins and the BMP receptors are coexpressed, notably in the extraembryonic

145

and embryonic mesoderm. BMP-4 and its receptors also play key inductive roles in lung morphogenesis (reviewed in Warburton et al., 2000); given the results of this study and the reported requirement for GATA-6 for lung morphogenesis, it is tempting to speculate that altered BMP-4 signaling may underlie the requirement of GATA-6 in lung development. Little is known about the transcription factors involved in early induction of BMP-4. They include the AP-1 proteins cjun and cfos (Knochel et al., 2000) and the homeobox protein Vent2 (Onichtchouk et al., 1996). The present work identifies GATA-4/-6 as potent regulators of the BMP-4 gene. GATA-4 activation of BMP-4 is consistent with the reported autoactivatory effect of BMP-4 on its promoter (Metz et al., 1998); moreover, GATA-4, whose expression is upregulated by retinoids (Kostetskii et al., 1999; Wilson et al., 1993), may mediate the indirect effect of retinoic acid on BMP-4 transcription (Helvering et al., 2000). It will be interesting in the future to test whether these or other regulators of BMP-4 converge on GATA-4 and whether GATA proteins interact with the other transcription activators of BMP-4. GATA factors in extraembryonic tissues: a role in vasculogenesis? The finding that BMP-4 is a transcription target for GATA-4 (and possibly GATA-6) implicates GATA-4 and -6 in early mesoderm and endoderm signaling and yolk sac vasculogenesis. The presence of these GATA factors in mesoderm and endoderm suggests that they may influence mesoderm– endoderm signaling and early vasculogesis at several levels. In the visceral endoderm, GATA-4 and -6 may control expression of other growth factors that are essential for signaling to the underlying extraembryonic mesoderm, such as Indian Hedgehog (Ihh) and vascular endothelial growth factor (VEGF), which are essential paracrine signals for differentiation of the underlying extraembryonic mesoderm into hemangioblasts and endothelial cells at the onset of gastrulation (Belaoussoff et al., 1998; Byrd et al., 2002; Carmeliet et al., 1996). In this respect, it is noteworthy that a consensus high affinity GATA binding site is present in

Fig. 8. Activation of BMP-4 transcription by GATA-4 and GATA-6. (A) Schematic representation of the BMP-4 promoter. Two consensus GATA sites (in reverse orientation) centred at positions ⫺526 bp and ⫺392 bp are conserved across species. (B) GATA-4 and GATA-6 activate the BMP4 promoter in a dose response manner (25, 50, and 250 ng) when cotransfected in NIH3T3 cells. The minimal BMP-4 promoter lacking the two GATA binding sites is unresponsive to GATA-4 or GATA-6. (C) Gel shift analysis using the two GATA elements of the BMP-4 promoter shows specific binding to endogenous GATA-4 in cardiomyocytes nuclear extracts (blocked by the addition of the GATA-4 antibody) and to recombinant GATA-4 or GATA-6 ectopically expressed in 293 cells. Competitor unlabeled oligonucleotides were added at 100-fold excess. S, self competitor; M, oligonucleotide containing the mutated GATA site; C, consensus GATA binding site from the ANF promoter. Note how the ⫺526-bp probe has a higher affinity for GATA-4 than the ⫺392 element. (D) Both GATA sites mediate GATA-dependent activation of the BMP-4 promoter. Mutation of the ⫺392-bp (M1) or of the ⫺526-bp (M2) GATA site in the context of the ⫺1300-bp promoter decreases or abolishes GATA-dependent activation. DM, mutation of both sites. Transfection conditions and DNA concentrations are the same as in (B). The GATA-4 expression vector was used at 50 ng/well. (E) BMP-4 is a direct in vivo target for GATA-4. RT-PCR analysis of BMP-4 transcripts in cells overexpressing GATA-4. P19 and C2C12 cells stably expressing GATA-4 show marked upregulation of BMP-4 transcripts. mRNA of rat neonatal cardiomyocytes (cardios) infected with a modified adenovirus expressing GATA-4 also show induction of BMP-4 transcripts.

146

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

both mouse (Accession No. U41B83) and human (Accession No. AF095785) VEGF promoters, suggesting that VEGF may indeed be a direct GATA transcription target. Thus, the defect in visceral endoderm differentiation in GATA-4 and -6 null embryos may translate into impaired yolk sac vasculogenesis, leading to early embryonic lethality. Early defects in hematopoiesis or vasculogenesis may also account for the developmental arrest in double GATA4/GATA-6 heterozygote mice, which die prior to e12.5 (Molkentin, 2000). Support for this hypothesis comes from the finding that Gata4⫺/⫺ embryoid bodies are defective in primitive hematopoiesis and vasculogenesis (Bielinska et al., 1996). In fact, the pattern of expression of GATA-4 and -6 in early embryogenesis raises the possibility for multiple functions of these proteins in early vasculogeonesis and an essential role in extraembryonic endoderm and mesoderm interactions. For example, the two proteins are present in an almost complementary manner in the allantois, which will give rise to the umbilical blood vessels. GATA-6, and to a lesser extent GATA-4, is also present in the blood islands, the site of early hemangioblasts, which will differentiate into hematopoietic and endothelial cells (Dieterlen-Lievre, 1998). Of note, GATA-6 transcripts were also detected in the blood islands of the developing Xenopus (Gove et al., 1997), indicating an evolutionary conserved function in these cells. Finally, both GATA-4 and -6 were detected in mesothelial cells lining the visceral endoderm between blood islands, which give rise to the smooth muscle cells that are recruited to the forming vessels. Thus, it would be worthwhile to examine vascular and hematopoietic development in mice lacking a Gata4 and Gata6 allele as well as the endothelial and hematopoietic differentiation potential of ES cells homozygous for the Gata6 gene deletion. Finally, and in support for a role of GATA-4 and -6 in the vasculature, it is noteworthy to point out the presence of GATA-4 in endothelial cells of the adult heart (our unpublished data) and in vascular smooth muscle cells in the developing lung and liver (Figs. 6 and 8), as well as its reported expression in endothelial cell progenitors derived from mesodermal cells of e7.5 mouse embryos (Hatzopoulos et al., 1998). Similarly, GATA-6 transcripts were also reported in mouse endothelial progenitors (Hatzopoulos et al., 1998) and in a human umbilical cord-derived endothelial cell line in which transcriptional regulation of the endothelial vascular cell adhesion molecule-1 (VCAM-1) was dependent on GATA-6 protein (Umetani et al., 2001). Taken together, these observations suggest a previously unexplored role of GATA-4 and -6 in vasculogenesis and possibly angiogenesis. GATA factors as regulators of ectodermal cells The high resolution of immunohistochemistry allowed clear detection of GATA-6 in previously unreported sites. These include the embryonic ectoderm, where dependence on GATA-6 for proper development could help explain the

reported widespread apoptosis observed in the embryonic ectoderm of GATA-6 null mice (Morrisey et al., 1998) as well as abnormal ectoderm development in them (Koutsourakis et al., 1999). Later, GATA-6 was detected in numerous ectodermal derivatives, including neural tube and neural crest-derived cells such as adrenal medulla and motor neurons. Thus, GATA-6 may play a previously unsuspected role in survival and/or differentiation of the ectodermal lineage. Although the functions of GATA factors have been most intensely studied in cells of the endodermal and mesodermal lineages, accumulating evidence points to an important role of this class of transcription factors in differentiation of ectodermally derived cells, including neurons. The presence of GATA-2 and -3 expression has been recently documented in the central nervous system of many species (Debacker et al., 1999; Sheng and Stern, 1999). Moreover, targeted mutagenesis of the Gata2 and Gata3 genes in mice revealed an essential role for GATA-2 in the generation of V2 interneurons and for GATA-3 in the differentiation of the sympathetic nervous system and the cephalic neural crest cells (Zhou et al., 2000; Lim et al., 2000). Thus, GATA-6 is the third member of the GATA family expressed in neuronal cells. Its highly localized distribution within the brain and in neural crest-derived cells warrants further investigation of its regulatory role in neuronal development. Acknowledgments We are grateful to Annie Valle´ e for the preparation of histologic sections, to Lise Laroche for secretarial assistance, to Chantal Lefebvre for technical help, and to members of the Nemer laboratory for helpful discussions. We thank Dr. Stephen Harris for generously providing the BMP-4 promoter. This work was supported by a grant from the Canadian Institutes of Health Research (MOP-36382). G.N. was the recipient of GIBCO-IRCM and Boehringer Ingelheim Canada studentships; M.N. holds a Canada Research Chair in Molecular Biology. References Arceci, R.J., King, A.A., Simon, M.C., Orkin, S.H., Wilson, D.B., 1993. Mouse GATA-4: a retinoic acid-inducible GATA-binding transcription factor expressed in endodermally derived tissues and heart. Mol. Cell. Biol. 13, 2235–2246. Belaguli, N.S., Zhou, W., Trinh, T.H., Majesky, M.W., Schwartz, R.J., 1999. Dominant negative murine serum response factor: alternative splicing within the activation domain inhibits transactivation of serum response factor binding targets. Mol. Cell. Biol. 19, 4582– 4591. Belaoussoff, M., Farrington, S.M., Baron, M.H., 1998. Hematopoietic induction and respecification of A–P identity by visceral endoderm signaling in the mouse embryo. Development 125, 5009 –5018. Bielinska, M., Narita, N., Heikinheimo, M., Porter, S.B., Wilson, D.B., 1996. Erythropoiesis and vasculogenesis in embryoid bodies lacking visceral yolk sac endoderm. Blood 88, 3720 –3730.

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148 Brewer, A., Gove, C., Davies, A., McNulty, C., Barrow, D., Koutsourakis, M., Farzaneh, F., Pizzey, J., Bomford, A., Patient, R., 1999. The human and mouse GATA-6 genes utilize two promoters and two initiation codons. J. Biol. Chem. 274, 38004 –38016. Byrd, N., Becker, S., Maye, P., Narasimhaiah, R., St Jacques, B., Zhang, X., McMahon, J., McMahon, A., Grabel, L., 2002. Hedgehog is required for murine yolk sac angiogenesis. Development 129, 361–372. Caprioli, A., Minko, K., Drevon, C., Eichmann, A., Dieterlen-Lievre, F., Jaffredo, T., 2001. Hemangioblast commitment in the avian allantois: cellular and molecular aspects. Dev. Biol. 238, 64 –78. Carmeliet, P., Ferreira, V., Breier, G., Pollefeyt, S., Kieckens, L., Gertsenstein, M., Fahrig, M., Vandenhoeck, A., Harpal, K., Eberhardt, C., Declercq, C., Pawling, J., Moons, L., Collen, D., Risau, W., Nagy, A., 1996. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. Nature 380, 435– 439. Charron, F., Paradis, P., Bronchain, O., Nemer, G., Nemer, M., 1999. Cooperative interaction between GATA-4 and GATA-6 regulates myocardial gene expression. Mol. Cell. Biol. 19, 4355– 4365. Charron, F., Tsimiklis, G., Arcand, M., Robitaille, L., Liang, Q., Molkentin, J.D., Meloche, S., Nemer, M., 2001. Tissue-specific GATA factors are transcriptional effectors of the small GTPase RhoA. Genes Dev. 15, 2702–2719. Condie, B.G., Brivanlou, A.H., Harland, R.M., 1990. Most of the homeobox-containing Xhox 36 transcripts in early Xenopus embryos cannot encode a homeodomain protein. Mol. Cell. Biol. 10, 3376 – 3385. Crispino, J.D., Lodish, M.B., Thurberg, B.L., Litovsky, S.H., Collins, T., Molkentin, J.D., Orkin, S.H., 2001. Proper coronary vascular development and heart morphogenesis depend on interaction of GATA-4 with FOG cofactors. Genes Dev. 15, 839 – 844. Debacker, C., Catala, M., Labastie, M.C., 1999. Embryonic expression of the human GATA-3 gene. Mech. Dev. 85, 183–187. Dieterlen-Lievre, F., 1998. Hematopoiesis: progenitors and their genetic program. Curr. Biol. 8, R727–R730. Durocher, D., Charron, F., Warren, R., Schwartz, R.J., Nemer, M., 1997. The cardiac transcription factors Nkx2–5 and GATA-4 are mutual cofactors. EMBO J. 16, 5687–5696. Feng, J.Q., Chen, D., Cooney, A.J., Tsai, M.J., Harris, M.A., Tsai, S.Y., Feng, M., Mundy, G.R., Harris, S.E., 1995. The mouse bone morphogenetic protein-4 gene. Analysis of promoter utilization in fetal rat calvarial osteoblasts and regulation by COUP-TFI orphan receptor. J. Biol. Chem. 270, 28364 –28373. Gao, X., Sedgwick, T., Shi, Y.B., Evans, T., 1998. Distinct functions are implicated for the GATA-4, -5, and -6 transcription factors in the regulation of intestine epithelial cell differentiation. Mol. Biol. Cell. 18, 2901–2911. Gove, C., Walmsley, M., Nijjar, S., Bertwistle, D., Guille, M., Partington, G., Bomford, A., Patient, R., 1997. Over-expression of GATA-6 in Xenopus embryos blocks differentiation of heart precursors. EMBO J. 16, 355–368. Gre´ pin, C., Dagnino, L., Robitaille, L., Haberstroh, L., Antakly, T., Nemer, M., 1994. A hormone-encoding gene identifies a pathway for cardiac but not skeletal muscle gene transcription. Mol. Cell. Biol. 14, 3115– 3129. Gre´ pin, C., Nemer, G., Nemer, M., 1997. Enhanced cardiogenesis in embryonic stem cells overexpressing the GATA-4 transcription factor. Development 124, 2387–2395. Harzopoulos, A.K., Folkman, J., Vasile, E., Eiselen, G.K., Rosenberg, R.D., 1998. Isolation and characterization of endothelial progenitor cells from mouse embryos. Development 125, 1457–1468. Heikinheimo, M., Scandrett, J.M., Wilson, D.B., 1994. Localization of transcription factor GATA-4 to regions of the mouse embryo involved in cardiac development. Dev. Biol. 164, 361–373. Helvering, I.M., Sharp, R.I., Ou, X., Geiser, A.G., 2000. Regulation of the promoters for the human bone morphogenetic protein 2 and 4 genes. Gene 256, 123–138.

147

Jiang, Y., Evans, T., 1996a. The Xenopus GATA-4/5/6 genes are associated with cardiac specification and can regulate cardiac-specific transcription during embryogenesis. Dev. Biol. 174, 258 –270. Jiang, Y.M., Evans, Y., 1996b. The Xenopus GATA-4/5/6 genes are associated with cardiac specification and can regulate cardiac-specific transcription during embryogenesis. Dev. Biol. 174, 258 –270. Koijzer, R., van Tuyl, M., Meijers, C., Post, M., Tibboel, D., Grosveld, F., Koutsourakis, M., 2001. The transcription factor GATA6 is essential for branching morphogenesis and epithelial cell differentiation during fetal pulmonary development. Development 128, 503–511. Kiiveri, S., Siltanen, S., Rahman, N., Bielinska, M., Lehto, V.P., Huhtaniemi, I.T., Muglia, L.J., Wilson, D.B., Heikinheimo, M., 1999. Reciprocal changes in the expression of transcription factors GATA-4 and GATA-6 accompany adrenocortical tumorigenesis in mice and humans. Mol. Med. 5, 490 –501. Knochel, S., Schuler-Metz, A., Knochel, W., 2000. c-Jun (AP-1) activates BMP-4 transcription in Xenopus embryos. Mech. Dev. 98, 29 –36. Kostetskii, I., Jiang, Y., Kostetskaia, E., Yuan, S., Evans, T., Zile, M., 1999. Retinoid signaling required for normal heart development regulates GATA-4 in a pathway distinct from cardiomyocyte differentiation. Dev. Biol. 206, 206 –218. Koutsourakis, M., Langeveld, A., Patient, R., Beddington, R., Grosveld, F., 1999. The transcription factor GATA6 is essential for early extraembryonic development. Development 126, 723–732. Kuo, C.T., Morrisey, E.E., Anandappa, R., Sigrist, K., Lu, M.M., Parmacek, M.S., Soudais, C., Leiden, J.M., 1997. GATA4 transcription factor is required for ventral morphogenesis and heart tube formation. Genes Dev. 11, 1048 –1060. Laverriere, A.C., MacNeill, C., Mueller, C., Poelmann, R.E., Burch, J.B., Evans, T., 1994. GATA-4/5/6, a subfamily of three transcription factors transcribed in developing heart and gut. J. Biol. Chem. 269, 23177– 23184. Lim, K.C., Lakshmanan, G., Crawford, S.E., Gu, Y., Grosveld, F., Engel, J.D., 2000. Gata3 loss leads to embryonic lethality due to noradrenaline deficiency of the sympathetic nervous system. Nat. Genet. 25, 209 – 212. Lowry, J.A., Atchley, W.R., 2000. Molecular evolution of the GATA family of transcription factors: conservation within the DNA-binding domain. J. Mol. Evol. 50, 103–115. MacNeill, C., Ayres, B., Laverriere, A.C., Burch, J.B., 1997. Transcripts for functionally distinct isoforms of chicken GATA-5 are differentially expressed from alternative first exons. J. Biol. Chem. 272, 8396 – 8401. Martin, J.F., Miano, J.M., Hustad, C.M., Copeland, N.G., Jenkins, N.A., Olson, E.N., 1994. A MEF2 gene that generates a muscle-specific isoform via alternative mRNA splicing. Mol. Cell. Biol. 14, 1647– 1656. Metz, A., Kno¨ chel, S., Bu¨ chler, P., Ko¨ ster, M., Knutzon, D.S., 1998. Structural and functional analysis of the BMP-4 promoter in early embryos of Xenopus laevis. Mech. Dev. 74, 29 –39. Mikawa, T., Gourdie, R.G., 1996. Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ. Dev. Biol. 174, 221–232. Mlodzik, M., Gehring, W.J., 1987. Expression of the caudal gene in the germ line of Drosophila: formation of an RNA and protein gradient during early embryogenesis. Cell 48, 465– 478. Molkentin, J.D., 2000. The zinc finger-containing transcription factors GATA-4, -5, and -6 —Ubiquitously expressed regulators of tissuespecific gene expression. J. Biol. Chem. 275, 38949 –38952. Molkentin, J.D., Kalvakolanu, D.V., Markham, B.E., 1994. Transcription factor GATA-4 regulates cardiac muscle-specific expression of the a-myosin heavy-chain gene. Mol. Cell. Biol. 14, 4947– 4957. Molkentin, J.D., Lin, Q., Duncan, S.A., Olson, E.N., 1997. Requirement of the transcription factor GATA4 for heart tube formation and ventral morphogenesis. Genes Dev. 11, 1061–1072. Molkentin, J.D., Tymitz, K.M., Richardson, J.A., Olson, E.N., 2000. Abnormalities of the genitourinary tract in female mice lacking GATA5. Mol. Cell Biol. 20, 5256 –5260.

148

G. Nemer, M. Nemer / Developmental Biology 254 (2003) 131–148

Morin, S., Charron, F., Robitaille, L., Nemer, M., 2000. GATA-dependent recruitment of MEF2 proteins to target promoters. EMBO J. 19, 2046 – 2055. Morrisey, E.E., Ip, H.S., Lu, M.M., Parmacek, M.S., 1996. GATA-6 —a zinc finger transcription factor that is expressed in multiple cell lineages derived from lateral mesoderm. Dev. Biol. 177, 309 –322. Morrisey, E.E., Tang, Z., Sigrist, K., Lu, M.M., Jiang, F., Ip, H.S., Parmacek, M.S., 1998. GATA6 regulates HNF4 and is required for differentiation of visceral endoderm in the mouse embryo. Genes Dev. 12, 3579 –3590. Mumberg, D., Lucibello, F.C., Schuermann, M., Muller, R., 1991. Alternative splicing of fosB transcripts results in differentially expressed mRNAs encoding functionally antagonistic proteins. Genes Dev. 5, 1212–1223. Nagy, P., Bisgaard, H.C., Thorgeirsson, S.S., 1998. Expression of hepatic transcription factors during liver development and oval cell differentiation. J. Cell Biol. 126, 223–233. Narita, N., Bielinska, M., Wilson, D.B., 1997. Wild-type endoderm abrogates the ventral developmental defects associated with GATA-4 deficiency in the mouse. Dev. Biol. 189, 270 –274. Nemer, G., Nemer, M., 2002. Cooperative interaction between GATA-5 and NF-ATc regulates endothelial– endocardial differentiation of cardiogenic cells. Development 129, 4045– 4055. Nemer, G., Qureshi, S.A., Malo, D., Nemer, M., 1999. Functional analysis and chromosomal mapping of GATA5, a gene encoding a zinc finger DNA-binding protein. Mamm. Genome 10, 993–999. Nishi, T., Kubo, K., Hasebe, M., Maeda, M., Futai, M., 1997. Transcriptional activation of H⫹/K⫹⫺ATPase genes by gastric GATA binding proteins. J. Biochem. 121, 922–929. Onichtchouk, D., Gawantka, V., Dosch, R., Delius, H., Hirschfeld, K., Blumenstock, C., Niehrs, C., 1996. The Xvent-2 homeobox gene is part of the BMP-4 signalling pathway controlling [correction of controling] dorsoventral patterning of Xenopus mesoderm [published erratum appears in Development 1997 Jan;124(1):preceding table of contents]. Development 122, 3045–3053. Pandolfi, P.P., Roth, M.E., Karis, A., Leonard, M.W., Dzierzak, E., Grosveld, F.G., Engel, J.D., Lindenbaum, M.H., 1995. Targeted disruption of the GATA3 gene causes severe abnormalities in the nervous system and in fetal liver haematopoiesis. Nat. Genet. 11, 40 – 44. Paradis, P., Dali-Youcef, N., Paradis, F.W., Thibault, G., Nemer, M., 2000. Overexpression of angiotensin II type 1 receptor in cardiomyocytes induces cardiac hypertrophy and remodeling. Proc. Natl. Acad. Sci. USA 97, 931–936. Pevny, L., Simon, M.C., Robertson, E., Klein, W.H., Tsai, S.F., D’Agati, V., Orkin, S.H., Costantini, F., 1991. Erythroid differentiation in chimaeric mice blocked by a targeted mutation in the gene for transcription factor GATA-1. Nature 349, 257–260. Reiter, J.F., Alexander, J., Rodaway, A., Yelon, D., Patient, R., Holder, N., Stainier, D.Y., 1999. Gata5 is required for the development of the heart and endoderm in zebrafish. Genes Dev. 13, 2983–2995. Rossi, J.M., Dunn, N.R., Hogan, B.L., Zaret, K.S., 2001. Distinct mesodermal signals, including BMPs from the septum transversum mesen-

chyme, are required in combination for hepatogenesis from the endoderm. Genes Dev. 15, 1998 –2009. Schultheiss, T.M., Burch, J.B., Lassar, A.B., 1997. A role for bone morphogenetic proteins in the induction of cardiac myogenesis. Genes Dev. 11, 451– 62 26. Sheng, G., Stern, C.D., 1999. Gata2 and Gata3: novel markers for early embryonic polarity and for non-neural ectoderm in the chick embryo. Mech. Dev. 87, 213–216. Skipper, M., Lewis, J., 2000. Getting to the guts of enteroendocrine differentiation. Nat. Genet. 24, 3– 4. Soudais, C., Bielinska, M., Heikinheimo, M., MacArthur, C.A., Narita, N., Saffitz, J.E., Simon, M.C., Leiden, J.M., Wilson, D.B., 1995. Targeted mutagenesis of the transcription factor GATA-4 gene in mouse embryonic stem cells disrupts visceral endoderm differentiation in vitro. Development 121, 3877–3888. Svensson, E.C., Tufts, R.L., Polk, C.E., Leiden, J.M., 1999. Molecular cloning of FOG-2: a modulator of transcription factor GATA-4 in cardiomyocytes. Proc. Natl. Acad. Sci. USA 96, 956 –961. Tamura, S., Wang, X.H., Maeda, M., Futai, M., 1993. Gastric DNA-binding proteins recognize upstream sequence motifs of parietal cell-specific genes [published erratum appears in Proc Natl Acad Sci USA 1994 May 10; 91(10):4609]. Proc. Natl. Acad. Sci. USA 90, 10876–10880. Tsai, F.Y., Keller, G., Kuo, F.C., Weiss, M., Chen, J., Rosenblatt, M., Alt, F.W., Orkin, S.H., 1994. An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371, 221–226. Umetani, M., Mataki, C., Minegishi, N., Yamamoto, M., Hamakubo, T., Kodama, T., 2001. Function of GATA transcription factors in induction of endothelial vascular cell adhesion molecule-1 by tumor necrosis factor-alpha. Arterioscler. Thromb. Vasc. Biol. 21, 917–922. Viger, R.S., Mertineit, C., Trasler, J.M., Nemer, M., 1998. Transcription factor GATA-4 is expressed in a sexually dimorphic pattern during mouse gonadal development and is a potent activator of the Mu¨ llerian inhibiting substance promoter. Development 125, 2665–2675. Warburton, D., Schwarz, M., Tefft, D., Flores-Delgado, G., Anderson, K.D., Cardoso, W.V., 2000. The molecular basis of lung morphogenesis. Mech. Dev. 92, 55– 81. Weiss, M.J., Orkin, S.H., 1995. GATA transcription factors: key regulators of hematopoiesis. Exp. Hematol. 23, 99 –107. White, A., Clark, A.J., 1993. The cellular and molecular basis of the ectopic ACTH syndrome. Clin. Endocrinol. 39, 131–141. Wilson, T.E., Mouw, A.R., Weaver, C.A., Milbrandt, J., Parker, K.L., 1993. The orphan nuclear receptor NGFI-B regulates expression of the gene encoding steroid 21-hydroxylase. Mol. Cell. Biol. 13, 861– 868. Winnier, G., Blessing, M., Labosky, P.A., Hogan, B.L., 1995. Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 9, 2105–2116. Zhang, H., Bradley, A., 1996. Mice deficient for BMP2 are nonviable and have defects in amnion/chorion and cardiac development. Development 122, 2977–2986. Zhou, Y., Yamamoto, M., Engel, J.D., 2000. GATA2 is required for the generation of V2 interneurons. Development 127, 3829 –3838.