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

First published online 29 June 2005
doi: 10.1242/dev.01913

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 Rojas, A. Articles by Black, B. L. Search for Related Content PubMed PubMed Citation Articles by Rojas, A. Articles by Black, B. L.

Gata4 expression in lateral mesoderm is downstream of BMP4 and is activated directly by Forkhead and GATA transcription factors through a distal enhancer element

¹ Cardiovascular Research Institute, University of California, San Francisco, CA 94143-0130, USA
² Department of Pediatrics, University of California, San Francisco, CA 94143-0130, USA
³ Genome Sciences Department, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA
⁴ Department of Biochemistry and Biophysics, University of California, San Francisco, CA 94143, USA

View larger version (10K): [in a new window]   Fig. 1. Schematic representation of the Gata4 locus and the Gata4 G2-lacZ transgene. The top line represents a 103 kb region of the mouse Gata4 locus, including its seven exons (black vertical lines). The arrow represents the transcriptional start; exon 2 is the first coding exon of the Gata4 gene. The red boxes (G1-G5) represent five regions of strong conservation between human and mouse Gata4 within noncoding sequences. The lower line depicts the transgene construct G2-lacZ, which contains the 4368 bp G2 fragment of Gata4 subcloned into the transgenic reporter plasmid HSP68-lacZ.

View larger version (104K): [in a new window]   Fig. 2. The Gata4 G2-lacZ transgene is expressed in the lateral mesoderm and septum transversum during mouse embryonic development. Whole-mount (A,B,F), transverse (E,M,O) and sagittal (J) sections of X-gal-stained G2-lacZ transgenic embryos are shown. For comparison, whole-mount (C,D,G,H), sagittal (I,K,L) and transverse (N,P) section in situ hybridization with different mesodermal markers is shown. (D,H,L) Foxf1; (G,K) Bmp4; (C,I,N) Gata4. (A,B) At 7.75 dpc and 8.25 dpc, lacZ expression directed by the Gata4 G2 enhancer is present in the lateral mesoderm (LM) and allantois (al). (E) A transverse section at midgut level at 8.0 dpc, ß-galactosidase activity is evident in both the somatic mesoderm (SM) and visceral mesoderm (VM) of the transgenic embryo. (F,J,M,O) By 9.5 dpc, lacZ expression is very robust in the mesodermal component of the septum transversum (ST) and in the visceral mesoderm surrounding the gut. Note that transgene expression is completely absent in the hepatic endoderm (HE) within the ST (F,J), and in the visceral endoderm (VE) of the gut (O). The expression directed by the Gata4 enhancer overlaps the expression of the mesodermal forkhead gene Foxf1, Bmp4 and endogenous Gata4 at all time points examined. NT, neural tube. Asterisks denote expression of the G2-lacZ transgene and endogenous Gata4 in the allantois at 8.25 dpc. Scale bars: 100 µm.

View larger version (72K): [in a new window]   Fig. 3. Expression directed by the Gata4 lateral mesoderm enhancer becomes restricted to the mesenchyme surrounding the liver. Representative X-gal-stained transgenic embryos (A-C) and dissected livers from transgenic animals (E-H) are shown. (A-C) Expression directed by the Gata4 enhancer is present exclusively in the liver (L) of transgenic embryos at 11.5 dpc (A), and this expression is restricted to the mesenchyme surrounding the liver, as observed in sagittal (B) and transverse (C) sections. The liver mesenchyme expression directed by the enhancer is identical to the expression of endogenous Gata4 in the liver, which was evident in transverse sections at 11.5 dpc (D). (E-G) Expression directed by the Gata4 enhancer in the liver mesenchyme was strong at 11.5 dpc (E) but began to diminish by 13.5 dpc (F) and 16.5 dpc (G). No transgene expression was observed in the adult liver (H). NT, neural tube. The arrows in C and D denote expression of G2-lacZ and endogenous Gata4 in the mesenchyme surrounding the liver at 11.5 dpc. Scale bars: in B-D, 100 µm; in H, 1 cm.

View larger version (34K): [in a new window]   Fig. 4. Deletional analysis of the Gata4 lateral mesoderm enhancer identifies a highly conserved element that is necessary and sufficient for enhancer function in vivo. (A) Schematic diagram of the deletion constructs of the Gata4 lateral mesoderm enhancer. The genomic organization of the G2 region of Gata4 is depicted at the top. Red boxes represent three regions of high sequence homology between the opossum and mouse Gata4 genes, denoted as CR1, CR2 and CR3. The nucleotide positions of each deletion construct, relative to G2, are denoted on the left, and mesoderm expression directed by each construct is summarized by a plus (mesodermal expression) or a minus (no detectable mesodermal expression) to the right of the line representing each construct. The column on the far right indicates the number of independent transgenic lines or F0 embryos that expressed lacZ in the septum transversum and lateral mesoderm as a fraction of the total number of transgene-positive F0 embryos or lines examined. (B-H) Representative transgenic embryos for each of the deletion constructs depicted in A were collected at 9.5 dpc and X-gal stained. G2, G22 and G24 directed strong expression in the septum transversum (ST) and visceral mesoderm (B,D,F). G24 encompasses only the region surrounding the highly conserved CR2. G25 contains only 308 bp of CR2 and was sufficient to direct only very weak expression in septum transversum and gut mesoderm (G). Deletion of CR2 (2495-3738) from G2 to generate G26 completely ablated transgene activity (H). Arrowheads indicate visceral mesoderm.

View larger version (51K): [in a new window]   Fig. 5. The Gata4 lateral mesoderm enhancer contains three conserved, candidate FOX-binding sites, two conserved, candidate SMAD-binding sites, and three perfectly conserved, candidate GATA factor binding sites. ClustalW analysis comparing the sequence of the conserved enhancer region from mouse, human and opossum, identified: three conserved candidate binding sites for FOX transcription factors (blue boxes), denoted as Fox I, Fox II and Fox III; two conserved, candidate binding sites for SMAD transcription factors (yellow boxes), denoted as Smad I and Smad II; and three perfectly conserved, candidate binding sites for GATA factors (red boxes), denoted as gata I, gata II and gata III. Asterisks denote nucleotides that have been perfectly conserved among all three species.

View larger version (61K): [in a new window]   Fig. 6. The Gata4 lateral mesoderm enhancer contains a high-affinity Forkhead-binding site. (A) Recombinant FOXF1 protein was transcribed and translated in vitro and used in EMSA with a radiolabeled double-stranded oligonucleotide encompassing the Gata4 Fox I site (lanes 2-5). Lane 1 contains reticulocyte lysate without recombinant FOXF1 protein (represented by a minus sign). FOXF1 efficiently bound to the Gata4 Fox I site (lane 2) and this binding was efficiently competed by a range of excess unlabeled Gata4 Fox I oligonucleotides (lane 3-5). A nonspecific lysate-derived band is denoted. (B) Recombinant FOXF1 protein was transcribed and translated in vitro, and used in EMSA with a radiolabeled double-stranded oligonucleotide representing a canonical FOXF1 site (Control FoxFI site). Lane 1 contains reticulocyte lysate without recombinant FOXF1 protein (represented by a minus sign). FOXF1 efficiently bound to the control Fox I site (lane 2). This binding was efficiently competed by an excess of unlabeled control probe (lane 3) and by an excess of Gata4 Fox I probe (lanes 4-6), even when only a 10-fold excess of the competitor was present (lane 6). A mutant version of the Gata4 Fox I site failed to compete for FOXF1 binding even at a 100-fold excess (lane 7). (C) Recombinant FOXA2 protein was transcribed and translated in vitro, and used in EMSA with a radiolabeled double-stranded oligonucleotide representing a canonical control FOXA2 site (lanes 2-5) or the Gata4 Fox I site (lanes 7-10). Lanes 1 and 6 contain reticulocyte lysate without recombinant FOXA2 protein (represented by a minus sign). FOXA2 efficiently bound to the control FOXA2 site (lane 2) and to the Gata4 Fox I site (lane 7), and this binding was efficiently competed by an excess of unlabeled control FOXA2 probe (lanes 3 and 8) and by an excess of Gata4 Fox I probe (lanes 4 and 9), but not by an excess of unlabeled mutant Gata4 Fox I probe (lanes 5 and 10).

View larger version (40K): [in a new window]   Fig. 7. The Gata4 lateral mesoderm enhancer contains two high-affinity GATA-binding sites. (A,B) Recombinant GATA4, GATA5 and GATA6 proteins were transcribed and translated in vitro, and used in EMSA with radiolabeled double-stranded oligonucleotides encompassing the Gata4 gata I site (A, lanes 2-7) or the gata II site (B, lanes 2-7). In A and B, lane 1 contains reticulocyte lysate without recombinant protein (represented by a minus sign). GATA4 efficiently bound to both the gata I and gata II sites (A and B, lane 2) and this binding was specifically competed by excess unlabeled gata I (A, lane 3) and gata II probes (B, lane 3). GATA5 and GATA6 proteins also bound, although more weakly than GATA4 protein, to the gata I (A, lanes 4,6) and gata II sites (B, lanes 4,6), and binding by GATA5 and GATA6 was specifically competed by excess unlabeled gata I (A, lanes 5,7) and gata II probes (B, lanes 5,7). Approximately equivalent amounts of GATA4, GATA5 and GATA6 proteins were used in each sample. (C) Recombinant GATA4 protein was transcribed and translated in vitro and used in EMSA with a radiolabeled double-stranded oligonucleotide encompassing the Gata4 gata I site (lanes 1-5) or the gata II site (lanes 6-10). Lanes 1 and 6 contain reticulocyte lysate without recombinant GATA4 protein (represented by a minus sign). GATA4 efficiently bound to both the gata I and gata II sites (lanes 2,7). Binding of GATA4 to the Gata4 gata I site was competed by an excess of unlabeled gata I site (I, lane 4) and by an excess of an unlabeled control GATA site from the Nkx2.5 gene (C, lane 3), but not by an excess of a mutant version of the Gata4 gata I site (mI, lane 5). Likewise, the binding of GATA4 to the Gata4 gata II site was specifically competed by excess unlabeled gata II probe (II, lane 9) and by excess unlabeled control probe (C, lane 8), but not by an excess of a mutant version of the gata II probe (mII, lane 10).

View larger version (94K): [in a new window]   Fig. 8. The Gata4 lateral mesoderm enhancer is dependent on conserved Forkhead and GATA sites for its function in vivo. The wild-type Gata4 enhancer transgene construct G22 (wt; A,D,G) and transgenes containing mutations either in the Fox I site (mFox; B,E,H), or in both the gata I and gata II sites (mGATA; C,F,I), in the context of G22, were used to generate transgenic embryos. Representative transgenic embryos are shown at 7.75 dpc (A-C), 9.5 dpc (D-F) and 11.5 dpc (G-I). The wild-type construct directed strong expression in the lateral mesoderm (arrowhead) and allantois (al) at 7.75 dpc (A), in the septum transversum (ST) and visceral mesoderm (arrowhead) at 9.5 dpc (D), and in the mesenchyme surrounding the liver (L) at 11.5 dpc (G). Mutations in the Fox I site completely eliminated transgene expression in all of the fifteen independent transgenic embryos analyzed (B,E,H). Similarly, mutation of both GATA sites completely eliminated lacZ expression in the fourteen independent transgenic embryos analyzed (C,F,I). Arrowheads in A-C indicate lateral mesoderm; arrowheads in D-F indicate visceral mesoderm.

View larger version (85K): [in a new window]   Fig. 9. Activity of the Gata4 lateral mesoderm enhancer requires BMP4. (A,B) The G2-lacZ transgene was crossed into either a Bmp4+/– or a Bmp4–/– background, and embryos were collected at 7.5 dpc and stained with X-gal. Heterozygous Bmp4+/–; G2-lacZ Tg/0 embryos displayed strong expression of lacZ in the lateral mesoderm (LM) and allantois (al; A). By contrast, expression directed by the Gata4 lateral mesoderm enhancer in Bmp4–/–; G2-lacZ Tg/0 embryos was dramatically attenuated in all embryos examined (B), indicating that transgene activity is dependent on BMP4. (C-F) Explanted tissue containing the heart (hrt) and septum transversum (ST) of embryos from a single G2-lacZ stable transgenic line (C,D) or a single SMaa-lacZ stable transgenic line (E,F) was collected at 9.5 dpc and cultured for 48 hours in the presence of BSA (C,E) or recombinant Noggin (D,F). Following incubation, explants were X-gal stained. Gata4 lateral mesoderm enhancer activity was significantly reduced in all 15 viable embryo explants treated with Noggin when compared with the 15 viable embryo explants treated with BSA. By contrast, no differences were observed in X-gal staining in the heart or somites in SMaa-lacZ explants treated with BSA or Noggin (compare E and F), indicating that embryo explants were not in general crisis in the presence of Noggin. Viability of embryo explants was assessed by the observation of a continually beating heart.

View larger version (18K): [in a new window]   Fig. 10. A reinforcing transcriptional network for gene activation in the lateral mesoderm dependent on GATA4. In this model, GATA4 expression is activated indirectly by BMP4 (indicated by a dashed arrow) and directly by Forkhead factors, such as FOXF1. BMP4 and FOXF1 reciprocally activate the expression of one another, which further reinforces GATA4 expression. GATA4 then further reinforces the program by activating its own transcription and the transcription of downstream mesodermal genes. Red arrows represent evidence provided in the current study. Black arrows represent evidence provided by previously published studies. Dashed arrows indicate either a direct or indirect activation; solid arrows represent direct activation through direct enhancer binding.