February 1, 2003;
The prevailing model of dorsal ventral
patterning of the amphibian embryo
predicts that the prospective mesoderm
is regionalized at gastrulation in response to a gradient of signals. This gradient is established by diffusible BMP and Wnt inhibitors secreted dorsally in the Spemann organizer
. An interesting question is whether ventrolateral tissue
passively reads graded levels of ventralizing signals, or whether local self-organizing regulatory circuits may exist on the ventral
side to control cell behavior and differentiation at a distance from the Organizer
. We provide evidence that sizzled
, a secreted Frizzled-related protein expressed ventrally during and after gastrulation, functions in a negative feedback loop that limits allocation of mesodermal cells to the extreme ventral
fate, with direct consequences for morphogenesis and formation of the blood
islands. Morpholino-mediated knockdown of Sizzled
protein results in expansion of ventral posterior mesoderm
and the ventral blood
islands, indicating that this negative regulation is required for proper patterning of the ventral mesoderm
. The biochemical activity of sizzled
is apparently very different from that of other secreted Frizzled-related proteins, and does not involve inhibition of Wnt8
. Our data are consistent with the existence of some limited self-organizing properties of the extreme ventral mesoderm
[+] show captions
Fig. 3. Szl negatively regulates development of the ventral blood islands (VBI). (A) Benzidine staining of hemoglobin-containing cells. Embryos were injected with the morpholino oligo MOSZL in both blastomeres at the two-cell stage, or with Szl RNA in the ventral marginal zone at the four-cell stage. (B) In situ hybridization analysis of hematopoietic markers in embryos injected with 20 ng of the morpholino oligos MOSZL or MOC2, or with 40 pg of wSzl RNA in the ventral marginal zone. (a) Expression of xSCL at stage 24. Embryos are shown from the ventral side with anterior upwards. (b) Expression of xGata1 at stage 30. Embryos are shown from the ventral side with anterior downwards. Note that detection of xGata1 in MOC2 injected embryos required a longer reaction time, resulting in darker background staining of head structures. (c) Expression of xSCL at stage 28. (d) Expression of larval globin at stage 33/34.
Fig. 5. Patterns of gene expression in szl morphants and Szl-overexpressing embryos. (A) In situ hybridization analysis of marker genes in embryos injected with 20 ng of MOSZL. (a,b) Expression of Xbra at stage 10.5; vegetal view. (c,d) Expression of Chordin (blue), and Sox17β (magenta) at stage 11. Vegetal view, dorsal is upwards. (e,f) Expression of Vent2/Xom at stage 12. Vegetal/posterior view, dorsal is upwards. (g,h) Expression of Xenopus posterior (Xpo) at stage 10.5. Vegetal view, dorsal is upwards. (i,j) Double in situ hybridization for xMyoD (blue) and szl (magenta) at stage 11. Vegetal view, dorsal is upwards. (k,l) Expression of xMyoD (blue) and szl (magenta) at stage 18. In each panel, the embryo on the left is seen from the top, with anterior towards the left; the embryo on the right is seen form the posterior side, with dorsal upwards. (m,n) Szl expression at stage 24. Embryos are seen from the ventral side, with anterior upwards. (B) RT-PCR analysis of szl expression in embryos injected with 20 ng of the morpholino oligo MOSZL, or increasing amounts of wSzl RNA. Endogenous szl was selectively amplified with primers that do not recognize the injected wSzl RNA. EF1α served as an internal control for RNA levels. (C) In situ hybridization of lateral mesoderm markers in embryos injected with 20 ng of MOSZL. (a) Expression of Xlim-1 at stage 20 in lateral (top) and dorsal view (bottom). Arrowheads indicate the point where expression is shifted dorsally in szl morphants. (b-d) Expression of Flk-1 at stage 26 in lateral view and corresponding transverse section. Arrowheads indicate the presumptive vitelline (yellow) and cardinal (black) veins.
Fig. 7. Effects of Wnt8 and Wnt11 on development of the VBI. Embryos were injected in the marginal zone at the four-cell stage with the indicated amounts of expression vectors encoding xWnt8 or xWnt11 (pCS2-xWnt8 and pCS2-xWnt11), or the indicated amounts of RNA encoding dominant-negative xWnt8 or xWnt11. Embryonic αT3 globin was detected by in situ hybridization at stage 30/35.
Fig. 8. Comparative analysis of szl and other secreted Wnt inhibitors. (A-C) Increasing amounts of RNA encoding Myc-tagged versions of Xenopus Szl, Frzb-1 (Frzb), Crescent (Cres) and Dikkopf1 (Dkk1) were injected in the marginal zone of four-cell stage embryos. (A) Expression of the erythroid markers xGata1 and αT3 globin was analyzed by RT-PCR at stage 33/34. EF1α was used as an internal control. (B) Expression of the injected Myc-tagged proteins was analyzed by immunoblotting at stage 11 on embryos from the same round of injections. The same blot was probed for actin as a loading control. (C) At the same time, morphology of the embryos and shape of the VBI were visualized by whole-mount in situ hybridization for embryonic αT3 globin. (D) Activity of the injected proteins as inhibitors of the canonical Wnt pathway was assessed in a Luciferase reporter assay. Embryos were injected in the animal pole with 200 pg of pTOP-FLASH, 4 pg of xWnt8 RNA, and increasing amounts of RNA encoding Myc-tagged Wnt inhibitors. Relative LUC activity is normalized to activation of the reporter by xWnt8 alone, and amounts of injected sFRPs are plotted on a logarithmic scale. For each RNA dilution, injected proteins were expressed at comparable levels; i.e. Frzb-MT RNA was injected at a higher concentration to compensate for poor translation.