A role for GATA5 in Xenopus endoderm specification.

The endoderm gives rise to the gut and tissues that develop as outgrowths of the gut tube, including the lungs, liver and pancreas. Here we show that GATA5, a zinc-finger transcription factor, is expressed in the yolk-rich vegetal cells of Xenopus embryos from the early gastrula stage onwards, when these cells become committed to form endoderm. At mid-gastrula stages, GATA5 is restricted to the sub-blastoporal endoderm and is the first molecular marker for this subset of endodermal cells so far identified. We show that GATA4 and GATA5 are potent inducers of endodermal marker genes in animal cap assays, while other GATA factors induce these genes only weakly, if at all. When injected into the dorsal marginal zone, GATA5 respecifies prospective mesoderm towards an endodermal fate, thereby disrupting the convergence and extension movements normally undergone by the dorsal mesoderm. The resulting phenotype is very similar to those seen after injection of dominant negative versions of the FGF-receptor or the T-box transcription factor, Xbra and can be rescued by eFGF. The ability of GATA5 to respecify ectodermal and mesodermal cells towards endoderm suggests an important role for GATA5 in the formation of this germlayer. In animal cap assays, GATA5 is induced by concentrations of activin above those known to induce dorsal mesoderm and heart, in an FGF-independent manner. These data indicate that the emerging view for endodermal induction in general, namely that it is specified by high levels of TGF-beta in the absence of FGF signalling, is specifically true for sub-blastoporal endoderm.

Insights into early vasculogenesis revealed by expression of the ETS-domain transcription factor Fli-1 in wild-type and mutant zebrafish embryos.

Fli-1 is an ETS-domain transcription factor whose locus is disrupted in Ewing's Sarcoma and F-MuLV induced erythroleukaemia. To gain a better understanding of its normal function, we have isolated the zebrafish homologue. Similarities with other vertebrates, in the amino acid sequence and DNA binding properties of Fli-1 from zebrafish, suggest that its function has been conserved during vertebrate evolution. The initial expression of zebrafish fli-1 in the posterior lateral mesoderm overlaps with that of gata2 in a potential haemangioblast population which likely contains precursors of blood and endothelium. Subsequently, fli-1 and gata2 expression patterns diverge, with separate fli-1 and gata2 expression domains arising in the developing vasculature and in sites of blood formation respectively. Elsewhere in the embryo, fli-1 is expressed in sites of vasculogenesis. The expression of fli-1 was investigated in a number of zebrafish mutants, which affect the circulatory system. In cloche, endothelium is absent and blood is drastically reduced. In contrast to the blood and endothelial markers that have been studied previously, fli-1 expression was initiated normally in cloche embryos, indicating that induction of fli-1 is one of the earliest indicators of haemangioblast formation. Furthermore, although fli-1 expression in the trunk was not maintained, the normal expression pattern in the anterior half of the embryo was retained. These anterior cells did not, however, condense to form blood vessels. These data indicate that cloche has previously unsuspected roles at multiple stages in the formation of the vasculature. Analysis of fli-1 expression in midline patterning mutants floating head and squint, confirms a requirement for the notochord in the formation of the dorsal-aorta. The formation of endothelium in one-eyed pinhead, cyclops and squint embryos indicates a novel role for the endoderm in the formation of the axial vein. The phenotype of sonic-you mutants implies a likely role for Sonic Hedgehog in mediating these processes.

Phosphorylation of GATA-1 increases its DNA-binding affinity and is correlated with induction of human K562 erythroleukaemia cells.

We have investigated by electrophoretic mobility shift assay (EMSA) the level of GATA-1 DNA-binding activity in nuclear extracts prepared from the human erythroleukaemic cell line, K562, after erythroid induction by hemin, sodium butyrate (NaB) or Trichostatin A or treatment with N -acetylcysteine (NAC). Relative to extract from untreated cells, GATA-1 binding activity increased markedly in all cases. However, immunoblot analysis revealed unchanged levels of GATA-1 protein after induction. Incubation of induced but not uninduced K562 extracts with phosphatase prior to EMSA weakened the binding activity, suggesting that the increase in GATA-1 binding following induction of K562 cells was a consequence of phosphorylation. When the mouse erythroleukaemic cell line MEL was induced with dimethylsulphoxide (DMSO), NaB or NAC, GATA-1 binding activity fell with DMSO, rose significantly with NaB and remained at about the same level in NAC-induced cells. In this case immunoblotting revealed that GATA-1 protein levels were in accord with the EMSA data. The DNA-binding activities of induced and uninduced MEL cell nuclear extracts were decreased by incubation with phosphatase, showing that phosphoryl-ation and DNA binding of GATA-1 are already optimalin these cells. The DNA-binding activity of affinity-purified GATA-1 from MEL cells was also reduced by phosphatase treatment, showing that phosphorylation/dephosphorylation is directly affecting the factor. Furthermore, when a comparison was made by EMSA of nuclear extracts prepared from K562 and MEL cells untreated or incubated with okadaic acid, a phosphatase inhibitor, GATA-1 binding was seen to increase with K562 cells, whereas with MEL cells there was no change in GATA-1 binding. Overall the results suggest that the level of GATA-1 phosphorylation increases after the induction of K562, but not MEL cells, where GATA-1 is already highly phosphorylated. Furthermore, phosphorylation increases the binding affinity of GATA-1 for a canonical binding site.

Suppression of GATA factor activity causes axis duplication in Xenopus.

In Xenopus, the dorsoventral axis is patterned by the interplay between active signalling in ventral territories, and secreted antagonists from Spemann's organiser. Two signals are important in ventral cells, bone morphogenetic protein-4 (BMP-4) and Wnt-8. BMP-4 plays a conserved role in patterning the vertebrate dorsoventral axis, whilst the precise role of Wnt-8 and its relationship with BMP-4, are still unclear. Here we have investigated the role played by the GATA family of transcription factors, which are expressed in ventral mesendoderm during gastrulation and are required for the differentiation of blood and endodermal tissues. Injection ventrally of a dominant-interfering GATA factor (called G2en) induced the formation of secondary axes that phenocopy those induced by the dominant-negative BMP receptor. However, unlike inhibiting BMP signalling, inhibiting GATA activity in the ectoderm does not lead to neuralisation. In addition, analysis of gene expression in G2en injected embryos reveals that at least one known target gene for BMP-4, the homeobox gene Vent-2, is unaffected. In contrast, the expression of Wnt-8 and the homeobox gene Vent-1 is suppressed by G2en, whilst the organiser-secreted BMP antagonist chordin becomes ectopically expressed. These data therefore suggest that GATA activity is essential for ventral cell fate and that subsets of ventralising and dorsalising genes require GATA activity for their expression and suppression, respectively. Finally, using G2en, we show that suppression of Wnt-8 expression, in conjunction with blocked BMP signalling, does not lead to head formation, suggesting that the head-suppressing Wnt signal may not be Wnt-8.

Evidence for non-axial A/P patterning in the nonneural ectoderm of Xenopus and zebrafish pregastrula embryos.

Recent studies in early Xenopus and zebrafish embryos have demonstrated that posteriorizing, non-axial signals arising from outside the organizer (or shield) contribute to A/P patterning of the neural axis, in contradiction to the classical Spemann model in which such signals were proposed to be solely organizer derived. Our studies on the early expression of the transcription factors GATA-2 and 3 in both Xenopus and zebrafish nonneural ectoderm lend support to the existence of such non-axial signaling in the A/P axis. Thus we find that the earliest expression of GATA-2 and 3 is located in nonneural ectoderm and is strongly patterned in a graded manner along the A/P axis, being high anteriorly and absent from the most posterior regions. This results by early neurula stages in three broad zones: an anterior region which is positive for both GATA-2 and 3, a middle region which is positive for GATA-2 alone and a posterior region in which neither gene is expressed. These regions correspond to head, trunk and tail ectoderm and may represent the beginnings of functional segmentation of nonneural ectoderm, as suggested in the concept of the 'ectomere'. We find that A/P patterning of GATA expression in nonneural ectoderm may occur as early as late blastula/early gastrula stages. We investigate which posteriorizing signals might contribute to such distinct non axial ectodermal patterning in the A/P axis and provide evidence that both FGF and a Wnt family member contribute towards the final A/P pattern of GATA expression in nonneural ectoderm.

Factor binding to the human gamma-globin gene distal CCAAT site: candidates for repression of the normal gene or activation of HPFH mutants.

We have examined factor binding to the distal human gamma-globin CCAAT site and three naturally occurring hereditary persistence of fetal haemoglobin (HPFH) mutations of this site. Factor binding was examined using nuclear extracts from the erythroleukaemic cell lines K562 and MEL, and from A4 cells, a non-transformed mouse bone marrow stem cell line, using the electrophoretic mobility shift assay. Under standard binding conditions, in addition to the previously reported binding by a CCAAT factor (CP1) and GATA-1, the wild-type (wt) sequence bound high mobility factors which appeared to be GATA-2 isoforms. However, when the non-specific competitor conditions were varied, the binding profile with K562, but not MEL nuclear extract, was substantially altered. CP1 and GATA-1 were absent, and two new factors were detected, one of which bound preferentially to the Greek and Japanese non-deletion HPFH mutants. However, binding by the GATA-2 isoforms to the wt sequence was maintained with both cell types, as it was using the A4 cell line. With modified binding conditions, in A4 cells the two non-deletion and the Black deletion HPFH mutants each had a different protein binding profile which was lost on erythroid induction of the cells. We discuss the possibility that the GATA-2 isoforms bound to the wt sequence may function to suppress wt gamma gene expression in the bone marrow. Additionally, those factors which bind preferentially either to the deletion or non-deletion HPFH mutants may play positive roles in establishing an active chromatin structure.

The maternal CCAAT box transcription factor which controls GATA-2 expression is novel and developmentally regulated and contains a double-stranded-RNA-binding subunit.

The transcription factor GATA-2 is expressed at high levels in the nonneural ectoderm of the Xenopus embryo at neurula stages, with lower amounts of RNA present in the ventral mesoderm and endoderm. The promoter of the GATA-2 gene contains an inverted CCAAT box conserved among Xenopus laevis, humans, chickens, and mice. We have shown that this sequence is essential for GATA-2 transcription during early development and that the factor binding it is maternal. The DNA-binding activity of this factor is detectable in nuclei and chromatin bound only when zygotic GATA-2 transcription starts. Here we report the characterization of this factor, which we call CBTF (CCAAT box transcription factor). CBTF activity mainly appears late in oogenesis, when it is nuclear, and the complex has multiple subunits. We have identified one subunit of the factor as p122, a Xenopus double-stranded-RNA-binding protein. The p122 protein is perinuclear during early embryonic development but moves from the cytoplasm into the nuclei of embryonic cells at stage 9, prior to the detection of CBTF activity in the nucleus. Thus, the accumulation of CBTF activity in the nucleus is a multistep process. We show that the p122 protein is expressed mainly in the ectoderm. Expression of p122 mRNA is more restricted, mainly to the anterior ectoderm and mesoderm and to the neural tube. Two properties of CBTF, its dual role and its cytoplasm-to-nucleus translocation, are shared with other vertebrate maternal transcription factors and may be general properties of these proteins.

The SCL gene specifies haemangioblast development from early mesoderm.

The SCL gene encodes a basic helix-loop-helix (bHLH) transcription factor that is essential for the development of all haematopoietic lineages. SCL is also expressed in endothelial cells, but its function is not essential for specification of endothelial progenitors and the role of SCL in endothelial development is obscure. We isolated the zebrafish SCL homologue and show that it was co-expressed in early mesoderm with markers of haematopoietic, endothelial and pronephric progenitors. Ectopic expression of SCL mRNA in zebrafish embryos resulted in overproduction of common haematopoietic and endothelial precursors, perturbation of vasculogenesis and concomitant loss of pronephric duct and somitic tissue. Notochord and neural tube formation were unaffected. These results provide the first evidence that SCL specifies formation of haemangioblasts, the proposed common precursor of blood and endothelial lineages. Our data also underline the striking similarities between the role of SCL in haematopoiesis/vasculogenesis and the function of other bHLH proteins in muscle and neural development.

GATA-1 DNA binding activity is down-regulated in late S phase in erythroid cells.

We have set out to test a model for tissue-specific gene expression that relies on the early replication of expressed genes to sequester limiting activating transcription factors. Using an erythroid cell line, we have tested the changes in the DNA binding activity of the lineage-restricted transcription factor GATA-1 through the cell cycle. We find that GATA-1 activity is low in G1, peaks in mid-S phase, and then decreases in G2/M. In contrast, the binding activities of two ubiquitous transcription factors, Oct1 and Sp1, remain high in G2/M. GATA-1 protein and mRNA vary in a similar manner through the cell cycle, suggesting that the expression of the gene or the stability of its message is regulated. Although a number of transcription factors involved in the control of the cell cycle or DNA replication have been shown to peak in S phase, this is the first example of a lineage-restricted transcription factor displaying S phase-specific DNA binding activity. One interpretation of these data leads to a model in which the peak in GATA-1 DNA binding amplifies the effect of early replication on the activation of erythroid-specific genes at the same time as preventing activation of non-erythroid genes containing GATA-responsive elements. These results may also relate to recent data implicating GATA-1 function in apoptosis and cell cycle progression.

GATA-2 is a maternal transcription factor present in Xenopus oocytes as a nuclear complex which is maintained throughout early development.

We show that Xenopus oocytes and embryos contain GATA-2, stored in nuclei as a non-chromatin-bound complex. Its binding site specificity is different from that of GATA-1, having a much higher affinity for the motif with a core GATC sequence. This binding site preference was markedly reduced by either release of the factor with deoxycholate or purification on a DNA affinity column, suggesting a role for a cofactor(s). The identity of the maternal GATA factor was established as GATA-2 in two ways: (1) binding to an oligonucleotide probe was abolished by inclusion of either of two GATA-2 monoclonal antibodies, and (2) a protein of correct molecular weight for GATA-2 was detected by immunoblotting with a polyclonal antibody raised against a Xenopus GATA-2-specific peptide. Although predominantly complexed, some of the oocyte GATA-2 is functional as a transcription factor because the transcriptional activity of the chicken betaH-globin promoter injected into oocytes was reduced by mutation of either of two GATA binding sites. This effect was more pronounced when the stronger of the two sites was mutated. Butyrate treatment of oocytes stimulated cap-site initiation by up to 17-fold with both normal promoter and GATA site mutant constructs, showing that the mechanism of butyrate stimulation is not via GATA-2. The possible significance of regulating the availability of maternal GATA-2 during early development is discussed.

GATA factors and the origins of adult and embryonic blood in Xenopus: responses to retinoic acid.

The transcription factors, GATA-1, -2 and -3 play essential roles in the differentiation of haematopoietic cells. To study the process of blood formation during vertebrate development we have used the expression of these GATA factors to locate haematopoietic cells in Xenopus embryos and to act as sensors for the effects of all-trans retinoic acid (RA), a signalling molecule which influences both anteroposterior patterning and haematopoietic differentiation. GATA factor expression was detected in the leading edge of the gastrulating mesoderm, in the ventral blood island (VBI) and dorsolateral plate (DLP) mesoderms and in a population of cells between the VBI and DLP. The VBI contributes to both embryonic and adult blood, whereas the DLP contains precursors of adult blood only, which have not been identified previously with molecular markers. The possibility that the GATA-2-expressing cells between the VBI and DLP were haematopoietic progenitors migrating from the VBI to the DLP was ruled out by transplantation analysis. Differential effects of RA on the expression of GATA-1 and GATA-2 suggest that RA has a direct action on haematopoietic differentiation, rather than on the formation of haematopoietic mesoderm.

Nuclear translocation of a maternal CCAAT factor at the start of gastrulation activates Xenopus GATA-2 transcription.

The transcription factor GATA-2 is present in blood cell precursors and plays a pivotal role in the control of erythroid differentiation. In Xenopus embryos, low levels of GATA-2 mRNA are maternally derived, while the onset of zygotic GATA-2 expression coincides with commitment to haematopoietic lineages. However, its initial transcriptional activation is not restricted to the presumptive blood islands, but occurs throughout ventral and lateral regions, in all three germ layers. In order to determine how this expression pattern is controlled, we have isolated and characterized the Xenopus GATA-2 gene. We show that 1.65 kb of 5' flanking sequences are sufficient to direct both correct transcriptional initiation in oocytes and appropriate temporal and spatial gene expression in early embryos. The transgene is activated during gastrulation and by neurula stages in predominantly expressed in the ventral hemisphere. We demonstrate that a CCAAT element is necessary for gene activity in both systems and that extracts prepared from oocytes and embryos contain a factor which specifically recognizes this element. We also show that cytoplasmic localization inhibits the function of this CCAAT factor until the beginning of gastrulation, when the zygotic GATA-2 gene is activated. These observations extend our understanding of the mechanisms by which maternal factors control the temporal activation of transcription in early vertebrate embryos.

Negative control of Xenopus GATA-2 by activin and noggin with eventual expression in precursors of the ventral blood islands.

To increase our understanding of haematopoiesis during early vertebrate development, we have studied the expression pattern of the transcription factor GATA-2 in Xenopus embryos, and asked how this is regulated. We show that the blood island precursors of the ventral mesoderm express GATA-2 RNA at neural tube stages, some 5 hours before globin RNA is detected in their derivatives. Prior to this however, GATA-2 is expressed much more widely within the embryo. Maternal transcripts are uniformly distributed, and zygotic transcription is activated during gastrulation throughout ventral and lateral regions of the embryo, with expression highest in the sensorial ectoderm and only weak in the ventral mesoderm. The domain of GATA-2 expression in neurulae outlines the region of the neural plate and suggests a possible wider role in dorsoventral patterning. To identify the signals involved in regulating this pattern of expression, we performed experiments with embryo explants. GATA-2 is activated autonomously in isolated animal caps and this activation is suppressed by the mesoderm-inducing factor activin, but not by FGF. Thus, the down-regulation of GATA-2 observed in the region of the Spemann organiser may be a response to an activin-like signal emanating from the dorsal-vegetal region or Nieuwkoop centre. GATA-2 activation in animal caps and ventral marginal zones was suppressed by co-culturing with dorsal marginal zones, suggesting that a signal from the Spemann organiser is involved in suppression of GATA-2 in the dorsal region of the embryo. Expression of a candidate for this signal, noggin, had the same effect. Taken together, the observations presented here suggest that GATA-2 activation occurs by default in the absence of signals, that the restriction of its expression within the early embryo is controlled by negative signals emanating from the Nieuwkoop centre and the organiser, and that noggin and activin-like molecules play a role in these signalling pathways.

Quantitative and qualitative analysis of exogenous gene expression by the S1 nuclease protection assay.

A method for determining the beginning and end of gene transcripts along with the position of intron/exon boundaries by S1 nuclease protection is described. Probe strategies and hybridization conditions are considered in detail. Under conditions of probe excess the method is quantitative.

DNA replication facilitates the action of transcriptional enhancers in transient expression assays.

We demonstrate a general role for DNA replication in the activation of gene transcription in transient transfection assays. The effect is observed for a wide range of genes and cell types, transfected by a number of protocols and is independent of increased template copy number. Replication does not stimulate transcription driven by proximal promoter elements alone but requires a functional enhancer element. This synergy between an active replication origin and an enhancer is not confined to elements from viruses such as SV40, which undergo an early to late switch in gene expression that is tightly coupled to replication, since the enhancer-containing long terminal repeats from retroviruses are strongly stimulated by replication. Furthermore, synthetic enhancers consisting of multimerised binding sites for one or two factors are also subject to replication-activation. The diversity of synthetic and natural enhancers used in this study suggests that replication and transcription do not share a common protein factor. We propose that replication leads to chromatin modifications that facilitate enhancer action.

Evidence for torsional stress in transcriptionally activated chromatin.

The existence of torsional stress in eukaryotic chromatin has been controversial. To determine whether it could be detected, we probed the structure of an alternating AT tract. These sequences adopt cruciform geometry when the DNA helix is torsionally strained by negative supercoiling. The single-strand-specific nuclease P1 was used to determine the structure of an alternating AT sequence upstream of the Xenopus beta-globin gene when assembled into chromatin in microinjected Xenopus oocytes. The pattern of cleavage by P1 nuclease strongly suggests that the DNA in this chromatin template is under torsional stress. The cruciform was detected specifically in the most fully reconstituted templates at later stages of chromatin assembly, suggesting that negative supercoiling is associated with chromatin maturation. Furthermore, the number of torsionally strained templates increased dramatically at the time when transcription of assembled chromatin templates began. Transcription itself has been shown to induce supercoiling, but the requisite negative supercoiling for cruciform extrusion by (AT)n in oocytes was not generated in this way since the characteristic P1 cutting pattern was retained even when RNA polymerase elongation was blocked with alpha-amanitin. Thus, torsional stress is associated with transcriptional activation of chromatin templates in the absence of ongoing transcription.

A chicken red cell inhibitor of transcription associated with the terminally differentiated state.

When a red cell nuclear extract (RCE) from adult chickens was injected into Xenopus oocytes along with the chicken beta globin gene, transcript levels were dramatically reduced compared to injection of DNA alone. The inhibitory action of the RCE was not specific to the beta globin gene since the Herpes thymidine kinase and Xenopus 5S RNA gene transcript levels were similarly reduced. Transcriptional repression was observed even after passage of the RCE through oocyte cytoplasm to the nucleus. The inhibitory activity binds to DNA cellulose, which suggests that the inhibitor either binds to DNA or associates with DNA-binding proteins. Nuclease digestion of the chromatin assembled on injected beta globin DNA revealed that inhibition was not associated with local changes in chromatin structure. Extracts from 9-d chicken embryonic erythroid cells, in which the endogenous beta globin gene is actively expressed, did not inhibit transcription. The inhibitory activity is, therefore, restricted to transcriptionally quiescent, adult erythrocytes. Since the inhibitory effects were seen with both polymerase II and III directed genes, we speculate that the activity may be part of the extreme transcriptional repression which occurs in the terminally differentiated erythrocyte.

Quantitative and Qualitative Analysis of Exogenous Gene Expression by the S1 Nuclease ProtectionAssay.

The Sinuclease protection procedure allows the precise definition of the beginning and end of gene transcripts as well as the position of intron/ exon boundaries in a gene. Alternatively, when these parameters are already known, the technique can be used to quantify transcript levels in a variety of expression systems.

Primer extension analysis of mRNA isolated from transfected cell lines.

Primer extension is a relatively quick and convenient means by which transcription from a gene transfected into tissue-culture cells can be monitored. The technique can be used to accurately determine the site of transcription initiation or to quantify the amount of cap-site-specific message produced.