Search

Atypical E2F transcription factors (E2F7 and E2F8) function as key regulators of cell cycle progression and their inactivation leads to spontaneous cancer formation in mice. However, the mechanism of the tumor suppressor functions of E2F7/8 remain obscure. In this study we discovered that atypical E2Fs control tumor angiogenesis, one of the hallmarks of cancer. We genetically inactivated atypical E2Fs in epithelial and mesenchymal neoplasm and analyzed blood vessel formation in three different animal models of cancer. Tumor formation was either induced by application of 7,12-Dimethylbenz(a)anthracene/12-O-Tetradecanoylphorbol-13-acetate or by Myc/Ras overexpression. To our surprise, atypical E2Fs suppressed tumor angiogenesis in all three cancer models, which is in a sharp contrast to previous findings showing that atypical E2Fs promote angiogenesis during fetal development in mice and zebrafish. Real-time imaging in zebrafish displayed that fluorescent-labeled blood vessels showed enhanced intratumoral branching in xenografted E2f7/8-deficient neoplasms compared with E2f7/8-proficient neoplasms. DLL4 expression, a key negative inhibitor of vascular branching, was decreased in E2f7/8-deficient neoplastic cells, indicating that E2F7/8 might inhibit intratumoral vessel branching via induction of DLL4.

Background: We have used the advantages of the zebrafish model system to demonstrate which of the vertebrate Myosin Light Chain Kinase (MLCK) genes is expressed in thrombocytes and important for thrombus formation. Methods and Results: Here we report that Mlck1a is an essential component of thrombus formation. Phylogenetic data revealed four zebrafish orthologous for three human MLCKgenes. To investigate expression of the zebrafish mlckgenes in thrombocytes we compared GFP‐tagged platelets with other cells by microarray analysis, and showed that mlck1aexpression was 4.5‐fold enriched in platelets. Furthermore, mlck1amRNA and mRNA for the platelet‐specific cd41co‐localized in thrombi. Expression of other mlcksubtypes was lower in GFP‐tagged platelets (mlck1b; 0.77‐fold enriched) and absent in thrombi (mlck1b, ‐2, ‐3). To investigate the role of Mlck1a in thrombus formation, we knocked down mlck1ausing two morpholinos. This resulted in impaired morphology changes of platelets adhering on fibrinogen. In a thrombosis model, in which thrombocytes adhere to the vessel wall damaged by laser irradiation, thrombus formation was slowed down in mlck1a‐deficient embryos. Conclusion: We conclude that Mlck1a is the subtype of MLCK that contributes to platelet shape change and thrombus formation.

Classical embryological studies have implied the existence of an apical ectodermal maintenance factor (AEMF) that sustains signaling from the apical ectodermal ridge (AER) during vertebrate limb development. Recent evidence suggests that AEMF activity is composed of different signals involving both a sonic hedgehog (Shh) signal and a fibroblast growth factor 10 (Fgf10) signal from the mesenchyme. In this study we show that the product of the dackel (dak) gene is one of the components that acts in the epidermis of the zebrafish pectoral fin bud to maintain signaling from the apical fold, which is homologous to the AER of tetrapods. dak acts synergistically with Shh to induce fgf4 and fgf8 expression but independently of Shh in promoting apical fold morphogenesis. The failure of dak mutant fin buds to progress from the initial fin induction phase to the autonomous outgrowth phase causes loss of both AER and Shh activity, and subsequently results in a proximodistal truncation of the fin, similar to the result obtained by ridge ablation experiments in the chicken. Further analysis of the dak mutant phenotype indicates that the activity of the transcription factor engrailed 1 (En1) in the ventral non-ridge ectoderm also depends on a maintenance signal probably provided by the ridge. This result uncovers a new interaction between the AER and the dorsoventral organizer in the zebrafish pectoral fin bud.

Bone morphogenetic proteins (Bmps) are key regulators of dorsoventral (DV) patterning. Within the ectoderm, Bmp activity has been shown to inhibit neural development, promote epidermal differentiation and influence the specification of dorsal neurons and neural crest. In this study, we examine the patterning of neural tissue in mutant zebrafish embryos with compromised Bmp signalling activity. We find that although Bmp activity does not influence anteroposterior (AP) patterning, it does affect DV patterning at all AP levels of the neural plate. Thus, we show that Bmp activity is required for specification of cell fates around the margin of the entire neural plate, including forebrain regions that do not form neural crest. Surprisingly, we find that Bmp activity is also required for patterning neurons at all DV levels of the CNS. In swirl/bmp2b(-) (swr(-)) embryos, laterally positioned sensory neurons are absent whereas more medial interneuron populations are hugely expanded. However, in somitabun(-) (sbn(-)) embryos, which probably retain higher residual Bmp activity, it is the sensory neurons and not the interneurons that are expanded. Conversely, in severely Bmp depleted embryos, both interneurons and sensory neurons are absent and it is the most medial neurons that are expanded. These results are consistent with there being a gradient of Bmp-dependent positional information extending throughout the entire neural and non-neural ectoderm.

Sonic hedgehog (Shh) is expressed in the posterior vertebrate limb bud mesenchyme and directs anteroposterior patterning and growth during limb development. Here we report an analysis of the pectoral fin phenotype of zebrafish sonic you mutants, which disrupt the shh gene. We show that Shh is required for the establishment of some aspects of anteroposterior polarity, while other aspects of anteroposterior polarity are established independently of Shh, and only later come to depend on Shh for their maintenance. We also demonstrate that Shh is required for the activation of posterior HoxD genes by retinoic acid. Finally, we show that Shh is required for normal development of the apical ectodermal fold, for growth of the fin bud, and for formation of the fin endoskeleton.

Activin and Vg1, two members of the TGF‐beta family, are believed to play roles in mesoderm induction and axis formation in the amphibian embryo. Both molecules are provided maternally, either as protein (activin) or as RNA and protein (Vg1), and experiments with a truncated form of a type IIB activin receptor have led to the conclusion that activin is required for induction of mesoderm in vivo. In this paper we first show that truncated versions of two different Xenopus activin receptors also have severe effects on the activity of the mature region of Vg1, suggesting that such receptors may block the function of several members of the TGF‐beta family. We go on to demonstrate that follistatin, a secreted protein which binds activin and blocks its activity, does not interfere with Vg1 signalling. Furthermore, overexpression of follistatin mRNA in Xenopus embryos does not perturb mesoderm formation. Taken together, our data show that the effects of truncated activin receptors on Xenopus development can be explained by the inhibition of Vg1 activity, while the lack of effect of follistatin argues against a function for activin in mesoderm induction.

Activin and Vg1, two members of the TGF‐beta family, are believed to play roles in mesoderm induction and axis formation in the amphibian embryo. Both molecules are provided maternally, either as protein (activin) or as RNA and protein (Vg1), and experiments with a truncated form of a type IIB activin receptor have led to the conclusion that activin is required for induction of mesoderm in vivo. In this paper we first show that truncated versions of two different Xenopus activin receptors also have severe effects on the activity of the mature region of Vg1, suggesting that such receptors may block the function of several members of the TGF‐beta family. We go on to demonstrate that follistatin, a secreted protein which binds activin and blocks its activity, does not interfere with Vg1 signalling. Furthermore, overexpression of follistatin mRNA in Xenopus embryos does not perturb mesoderm formation. Taken together, our data show that the effects of truncated activin receptors on Xenopus development can be explained by the inhibition of Vg1 activity, while the lack of effect of follistatin argues against a function for activin in mesoderm induction.

Embryos mutant for the T gene, in mice, make insufficient mesoderm and fail to develop a notochord. We report the cloning and sequencing of the T gene in the zebrafish (Brachydanio rerio) and show the nuclear localization of the protein product. Both RNA and protein are found in cells of the germ ring, including enveloping layer cells, prior to and during gastrulation of zebrafish embryos. Nuclei of the yolk syncytial layer do not express Zf-T. High levels of expression are maintained throughout early development in the notochord, while in paraxial mesoderm cells the gene is turned off during gastrulation. Exposure of animal cap cells to activinA induces Zf-T expression, as does transplantation into the germ ring.

Early dorsoventral pattern formation in vertebrate embryos is regulated by opposing activities of ventralizing bone morphogenetic proteins (BMPs) and dorsal-specific BMP antagonists such as Chordin, Noggin and Follistatin. Specific defects in early dorsoventral patterning have been recently found in a number of zebrafish mutants, which exhibit either a ventralized or dorsalized phenotype. One of these, the ventralized mutant chordino (originally called dino) is caused by a mutation in the zebrafish chordin homologue and interacts genetically with the dorsalized mutant swirl. In swirl mutant embryos, dorsal structures such as notochord and somites are expanded while ventral structures such as blood and nephros are missing. Here we demonstrate that the swirl phenotype is caused by mutations in the zebrafish bmp2 gene (zbmp2). While injection of mRNAs encoded by the mutant alleles has no ventralizing effect, injection of wild-type zbmp2 mRNA leads to a complete rescue of the swirl mutant phenotype. Fertile adult mutant fish were obtained, showing that development after gastrulation is not dependent on zbmp2 function. In addition zBMP2 has no maternal role in mesoderm induction. Our analysis shows that swirl/BMP2, unlike mouse BMP2 but like mouse BMP4, is required for early dorsoventral patterning of the zebrafish embryo.

Embryos mutant for the T gene, in mice, make insufficient mesoderm and fail to develop a notochord. We report the cloning and sequencing of the T gene in the zebrafish (Brachydanio rerio) and show the nuclear localization of the protein product. Both RNA and protein are found in cells of the germ ring, including enveloping layer cells, prior to and during gastrulation of zebrafish embryos. Nuclei of the yolk syncytial layer do not express Zf-T. High levels of expression are maintained throughout early development in the notochord, while in paraxial mesoderm cells the gene is turned off during gastrulation. Exposure of animal cap cells to activinA induces Zf-T expression, as does transplantation into the germ ring.

goosecoid is an immediate early gene expressed at the dorsal blastoporal lip of the Xenopus gastrula. Microinjection experiments have suggested a direct role for goosecoid in organizing the dorsoventral axis of the frog embryo. Here we characterize the zebrafish homologue of goosecoid (gsc) and compare its expression to that of Brachyury or no tail (ntl), another immediate early gene required in developing mesoderm. We show that gsc exhibits two independent phases of expression: an early one in cells anterior to the presumptive notochord, but not in cells of the notochord itself, and a later one in neural crest derivatives in the larval head. Zygotic gsc transcripts are detected soon after the midblastula transition, and at the blastula stage form a gradient with a maximum at the dorsal side. Use of gsc as a dorsal marker allowed us to demonstrate that ntl expression is initially activated at the dorsal side of the blastula. At this early stage, gsc and ntl show overlapping domains of expression and are co-expressed in cells at the dorsal midline of the early gastrula. However, gsc- and ntl-expressing cells become separated in the course of gastrulation, with gsc being expressed in the axial hypoblast (prechordal plate) anterior to the ntl-expressing presumptive notochord cells. Studies with mutant embryos suggest that gsc is independent of ntl function in vivo.

The mouse T (Brachyury) gene is required for normal mesoderm development and the extension of the body axis. Recently, two mutant alleles of a zebrafish gene, no tail (ntl), have been isolated (Halpern, M. E., Ho., R. K., Walker, C. and Kimmel, C. B. (1993) Cell 75, 99-111). ntl mutant embryos resemble mouse T/T mutant embryos in that they lack a differentiated notochord and the caudal region of their bodies. We report here that this phenotype is caused by mutation of the zebrafish homologue of the T gene. While ntl embryos express mutant mRNA, they show no nuclear protein product. Later, expression of mRNA in mutants, but not in wild types, is greatly reduced along the dorsal midline where the notochord normally forms. This suggests that the protein is required for maintaining transcription of its own gene.

We have identified and characterized zebrafish eve1, a novel member of the Drosophila even-skipped (eve) gene family. eve1 RNAs are expressed initially in late blastulae with a peak during the gastrula stage, at which time expression is confined to ventral and lateral cells of the marginal zone of the zebrafish embryo. Later, eve1 transcripts are located in the most posterior part of the extending tail tip. We show that LiCl, known to dorsalize Xenopus embryos, has the same effect in zebrafish, resulting in embryos with exaggerated dorsoanterior structures. In LiCl-treated embryos, eve1 transcripts are completely absent. eve1 is therefore a marker of ventral and posterior cells. In the light of its ventroposterior expression domain, the localization of eve1 transcripts was analysed in spadetail (spt) and no tail (ntl), two mutants with abnormal caudal development. In sptb140 homozygous mutants, there is an accumulation of cells in the tail region, resulting from inadequate migratory behaviour of precursors to the trunk somites. These cells, in their abnormal environment, express eve1, emphasizing the correlation between ventroposterior position and eve1 expression. In homozygous mutant embryos for the gene ntl (the homologue of mouse Brachyury, originally called Zf-T), posterior structures are missing (M. E. Halpern, C. B. Kimmel, R. K. Ho and C. Walker, 1993; Cell In press). While mutant and wild-type embryos do not differ in their eve1 transcript distribution during gastrulation, eve1 expression is absent in the caudal region of mutant ntl embryos during early somitogenesis, indicating a requirement for ntl in the maintenance of eve1 expression during tail extension. Our findings suggest that eve1 expression is correlated with a ventral and posterior cell fate, and provide first insights into its regulation.

The mouse T (Brachyury) gene is required for normal mesoderm development and the extension of the body axis. Recently, two mutant alleles of a zebrafish gene, no tail (ntl), have been isolated (Halpern, M. E., Ho., R. K., Walker, C. and Kimmel, C. B. (1993) Cell 75, 99-111). ntl mutant embryos resemble mouse T/T mutant embryos in that they lack a differentiated notochord and the caudal region of their bodies. We report here that this phenotype is caused by mutation of the zebrafish homologue of the T gene. While ntl embryos express mutant mRNA, they show no nuclear protein product. Later, expression of mRNA in mutants, but not in wild types, is greatly reduced along the dorsal midline where the notochord normally forms. This suggests that the protein is required for maintaining transcription of its own gene.

goosecoid is an immediate early gene expressed at the dorsal blastoporal lip of the Xenopus gastrula. Microinjection experiments have suggested a direct role for goosecoid in organizing the dorsoventral axis of the frog embryo. Here we characterize the zebrafish homologue of goosecoid (gsc) and compare its expression to that of Brachyury or no tail (ntl), another immediate early gene required in developing mesoderm. We show that gsc exhibits two independent phases of expression: an early one in cells anterior to the presumptive notochord, but not in cells of the notochord itself, and a later one in neural crest derivatives in the larval head. Zygotic gsc transcripts are detected soon after the midblastula transition, and at the blastula stage form a gradient with a maximum at the dorsal side. Use of gsc as a dorsal marker allowed us to demonstrate that ntl expression is initially activated at the dorsal side of the blastula. At this early stage, gsc and ntl show overlapping domains of expression and are co-expressed in cells at the dorsal midline of the early gastrula. However, gsc- and ntl-expressing cells become separated in the course of gastrulation, with gsc being expressed in the axial hypoblast (prechordal plate) anterior to the ntl-expressing presumptive notochord cells. Studies with mutant embryos suggest that gsc is independent of ntl function in vivo.