XB-ART-9107Int J Dev Biol 2001 Apr 01;452:387-96.
Show Gene links Show Anatomy links
Two-step induction of primitive erythrocytes in Xenopus laevis embryos: signals from the vegetal endoderm and the overlying ectoderm.
Primitive blood cells differentiate from the ventral mesoderm blood islands in Xenopus embryos. In order to determine the tissue interactions that propagate blood formation in early embryogenesis, we used embryos that had the ventral cytoplasm removed. These embryos gastrulated normally, formed a mesodermal layer and lacked axial structures, but displayed a marked enhancement of alpha-globin expression. Early ventral markers, such as msx-1, vent-1 and vent-2 were highly expressed at the gastrula stage, while a dorsal marker, goosecoid, was diminished. Several lines of experimental evidence demonstrate the critical role of animal pole-derived ectoderm in blood cell formation: 1) Mesoderm derived from dorsal blastomeres injected with beta-galactosidase mRNA (as a lineage tracer) expressed alpha-globin when interfaced with an animal pole-derived ectodermal layer; 2) Embryos in which the animal pole tissue had been removed by dissection at the blastula stage failed to express alpha-globin; 3) Exogastrulated embryos that lacked an interaction between the mesodermal and ectodermal layers failed to form blood cells, while muscle cells were observed in these embryos. Using dominant-negative forms of the BMP-4 and ALK-4 receptors, we showed that activin and BMP-4 signaling is necessary for blood cell differentiation in ventral marginal zone explants, while FGF signaling is not essential. In ventralized embryos, inactivation of the BMP-4 signal within a localized area of the ectoderm led to suppression of globin expression in the adjacent mesoderm layer, but inactivation of the activin signal did not have this effect. These observations suggest that mesodermal cells, derived from a default pathway that is induced by the activin signal, need an additional BMP-4-dependent factor from the overlying ectoderm for further differentiation into a blood cell lineage.
PubMed ID: 11330858
Article link: Int J Dev Biol
Species referenced: Xenopus laevis
Genes referenced: acvr1b alk bmp4 gsc hba1 hba3 msx1 tbxt ventx1.2 ventx2.2
Article Images: [+] show captions
|Fig. 1. Removal of the vegetal pole cytoplasm caused an enhanced expression of a-globin in the embryo. (A) Schematic image of the depletion experiment. The vegetal pole cytoplasm was removed by pricking the embryo at about 0.3 NT. Such embryos showed a radial symmetrical shape (DAI=0), (Kao and Elinson, 1988) and expressed a large amount of a-globin. (B) Northern blot analysis showing that expression of a-globin in the cytoplasm-removed embryos (lane 4) was drastically increased in comparison with that in untreated normal embryos (lane 1). The same effect was also observed in BMP- 4 (lane 2)- or msx-1 (lane 3) -injected embryos (5 ng RNA/embryos). The expression of EF-1a served as a loading control.|
|Fig. 2. Whole-mount in situ hybridization in control and cytoplasm-removed embryos showing the expression of early dorsal, ventral and panmesodermal markers. Expression of goosecoid (A,F), Xmsx-1 (B,G), vent-1 (C,H), vent-2 (D,I) and Xbra (E,J) was examined in control (A-E) and cytoplasm-depleted embryos at the early gastrula stage (stage 10+). In treated embryos, the expression level of a dorsal marker (goosecoid) was decreased and expression levels of the ventral markers, Xmsx-1, vent-1, vent-2, were increased. Expression of the pan mesodermal marker (Xbra) was not affected by this treatment. All pictures are vegetal views with the dorsal region to the top. Scale bar in J indicates 0.5 mm.|
|Fig. 3. Sagittal sections of pricked embryos and a control embryo, showing the origin of globinpositive mesodermal cells. (A) Schematic image of the tracing experiment. Two prospective dorsal blastomeres of either pricked or intact embryos at the 16-cell stage were injected with b-galactosidase RNA (0.5 ng/embryo) and cultured for 2 days (stage 35/36). (B) Expression of a-globin in a pricked embryo. An embryo cultured for 2 days (stage 35/36) was processed for whole-mount in situ hybridization. The globin-positive cells exist in the involuted mesodermal cells, which face the animal pole-derived ectoderm layer. (C and D) Cell lineage was traced by injection of b-galactosidase RNA, followed by staining with X-gal. Note that labeled cells in the pricked embryos (C, arrowheads) have migrated to the same area as that of globin-expressing cells in B, while those in the intact embryos (D) have become the muscle and surrounding mesenchyme. Scale bar in D indicates 0.5 mm.|
|Fig. 4. Effect of animal pole tissue on a-globin expression in control and cytoplasm-depleted embryos. (A) Schematic image of the experimental procedure. Pricked and untreated embryos were allowed to develop until stage 9, and a small (20% of the animal hemisphere) (C,G), middle (50% of the animal hemisphere) (D,H), or large (100% of the animal hemisphere) (E,I) part of the animal pole tissue was removed as described in Materials and Methods. Such embryos were further cultured for 2 days and were used for the whole-mount in situ hybridization analysis to detect the expression of a-globin. (B-I) Typical examples from each experimental group are shown. Note that removal of the animal pole tissue caused a drastic reduction in a-globin expression in both control (B,C,D,E) and cytoplasm-depleted (F,G,H,I) embryos. Scale bar in I indicates 2 mm.|
|Fig. 5. Effect of animal pole tissue on a-globin expression in control and exogastrula embryos. Treated embryos (B) and untreated embryos (A) were allowed to develop until stage 7 and then cultured in 1.4å~MMR as described in Materials and Methods. Such embryos were further cultured for 2 days (stage 35/36) and were used for whole-mount in situ hybridization analysis (A,B) and Northern blotting analysis (C) to detect the expression of a- globin. In exogastrula embryos (B), no expression of a-globin was observed. Scale bar in B indicates 2 mm. In addition, the expression of a-actin was not affected in the exogastrula embryos. The expression of EF-1a served as a loading control.|
|Fig. 6. FGF signaling is not involved in primitive blood cell differentiation in VMZ explants. (A) Experimental protocol. Two ventral blastomeres of 4-cell-stage embryos were injected with RNA, and DMZ was excised at stage 10+, and cultured for 2 days. Explants were used either for Western blot analysis (B) or for histological study (C). (B) Western blot analysis to detect a-globin production in VMZ explants. The induction of a-globin was completely suppressed in the explants injected with tBR or tALK4 RNA, but retained in the explants injected with XFD RNA. (C) Histological observation of RNA-injected explants. A tBR-injected VMZ explant contained welldifferentiated axial structures such as neural tissue (N) and muscle (M). A tALK4-injected explant contained homogenous undifferentiated cells. Blood-like cells (B) are shown in the explants injected with XFD. A thick epidermis layer with many melanin granules is also characteristic of XFD-injected explants.|
|Fig. 7. BMP-4 signaling is involved in the stimulation process from the epidermis to the blood precursor cells in the adjacent mesoderm. At the 16-cell stage, truncated BMP-4 receptor (tBR) RNA (A) or truncated activin receptor (tALK4) RNA (B) was injected into a single animal cell of embryos in which the vegetal pole cytoplasm had been removed (pricked embryos). Each cell lineage was simultaneously traced by co-injection of b-galactosidase RNA. On the second day of culture after injection, embryos were fixed and processed for X-gal staining (blue signals) followed by wholemount in situ hybridization. The globin-expressing cells were visualized with fast red (red signals). b- galactosidase-positive cells are largely overlying the globin-expressing cells in the control and tALK4-injected embryos, whereas b-galactosidase-positive cells are apart from the globin-expressing cells in tBR –injected embryos. Two arrowheads beside the top embryos of each panel indicate the overlapping regions. Embryo shown in the top of panel A was judged as an overlapped embryo in Table 1. Scale bar = 1 mm.|
|Fig. 8. A model of the blood differentiation program in the ventral mesoderm. As is generally accepted, endoderm-derived factors, such as activin (or an activin-like factor), induce the formation of a default mesoderm in the marginal zone of the embryo. This default mesoderm needs interaction with the overlaying ectoderm to differentiate into blood cells. This factor(s) is thought to be BMP-4 itself or a secretory factor(s) activated by the BMP-4 signaling (BMP-4-dependent factor).|