July 14, 1995;
A nodal-related gene defines a physical and functional domain within the Spemann organizer.
A functional screen for gene products that rescue dorsal development in ventralized Xenopus embryos has yielded Xenopus nodal
-related 3 (Xnr3
), a diverged member of the TGF beta superfamily. Xnr3
is specifically expressed in the Spemann organizer
and is only expressed in the epithelial layer of the organizer
immediately preceding and extending through gastrulation. Like noggin
can induce muscle
in ventral mesoderm
explants, consistent with a role in patterning the gastrula
. In other ways, the activity of Xnr3
is different from noggin
. Embryos receiving injections of Xnr3
, particularly in the animal pole, send out tube-like extensions of tissue
from the site of injection. These protrusions usually contain no axial mesoderm
and only occasionally are positive for neural markers. It has previously been proposed that the epithelial layer of the organizer
initiates and coordinates the morphogenetic movements at gastrulation. The protrusions observed may reflect an activity of Xnr3
in promoting morphogenesis.
[+] show captions
Figure 1. Alignment of Xnr3 with Nodal and Xenopus BMP7 Mature
Sequences are shown starting at the first conserved cysteine. Identities
between the sequences are highlighted. The seven conserved
cysteines found in TGFB family members are marked with asterisks.
Figure 2. Developmental Expression of Xnr3 in Staged Normal, Dorsalized,
and Ventralired Embryos
The RNA equivalent of 2.5 embryos was loaded per lane. The filter
was previously hybridized with an EFla probe to control for RNA loading
(Smith and Harland, 1991).
Figure 3. In Situ Hybridization to Normal, Dorsalized, and Ventralized
Late blastula- and gastrula-stage embryos were hybridized with digoxigenin-
labeled RNA probes. (A)-(G) show Xnr3 expression.
(A) Vegetal view of late blastula-stage embryo (stage 9). Newly transcribed
XnA message is restricted to nuclei on dorsal side of embryo.
(6) Vegetal view of early gastrula-stage embryo (stage 10). Arrow
marks the dorsal lip of the blastopore. Unlike other organizer-specific
genes, hybridizing cells can be found on both sides of the blastopore
(C) Vegetal view of midgastrula embryo (stages 10.5-l 1). Hybridization
signal is significantly reduced from early gastrula stage and is
restricted to cells around the blastopore.
(D) Animal pole view of dorsalized (Lit-treated) embryo at the late
blastula stage (stage 9).
(E) Side view of embryo in (D).
(F) Vegetal pole view of ventralized (UV-treated) embryo at the late
blastula stage. No specific XnA hybridization could be detected.
(G) Sagittal section of stage 9 embryo through the dorsal-ventral axis.
XnA hybridization is found only in the epithelial layer of the organizer.
(H) Section as in (G), but with noggin hybridization probe. Hybridization
is detected in both the epithelial and deep layers of the organizer.
(I) Section as in (G), but with goosecoid hybridization probe. Hybridization
is excluded from the epithelial layer of the organizer. The outer
edge of the epithelial layer is marked by the broken line.
(J) Section as in(G), but with Erachyury hybridization probe. Hybridization
is similar to that seen with noggin, with the exception that it is
found on both dorsal and ventral sides of the embryo.
Figure 4. Regulation of Xnr3 Expression by Xwnt-8, Activin, and
We injected l-cell embryos at the animal pole with 100 pg of either
f3-galactosidase RNA (A), Xwnt-8 RNA (B), noggin RNA (C), or activin
RNA(D). The expression of Xnr3 was assayed by in situ hybridization
at stage 9. Only embryos injected with Xwnt-8 RNA (6) show expanded
expression of Xnr3 (all these embryos are stained throughout the animal
hemisphere; two embryos are lying on their sides). All embryos
showed extreme phenotypic effects of RNA injection. noggin-injected
embryos developed only heads; activin-injected embryos underwent
apparent formation of bottle cells over the entire animal pole (one
stage-l 1 embryo is shown for this panel, in which the residual pigment
has concentrated to the animal pole). For this experiment, pigmented
embryos were injected and bleached after staining.
Figure 5. Effects of Xnr3 RNA Injection into Embryos
(A-D) UV-ventratlzed embryos were injected at the 32-M stage with 100 pg each of Xnr3 and 6galactosidase RNAs. Single cells were injected
at either the animal tier (A), the two marginal tiers (S), or the vegetal tier (C). In (D), the embryo was injected in the animal tier with B-galactosidase
RNA only. At about stage 25, the embryos were ftxed and stained with X-Gal and 12/101.
(E-L) Normal embryos (not UV-treated) were injected with 166 pg of Xnr3 RNA at either the animal tier (E and I), the two marginal tiers (F and
J), or the vegetal tier (0 and K). (H) and (L) show embryos injected in the marginal zone with 196 pg of noggin RNA. Embryos were fixed at about
stage 25 and were either photographed without further processing (E-H) or stained with an anti-muscle antibody (I-L). Induced ectopic patches
of muscle are marked with arrows.
(M-P) In situ hybridizations. Embryos were injected at the l-cell stage with 166 pg of Xnr3 RNA at the animal pole. Induced protrusions are marked
with open arrows. (M) Embryos stained at approximately stage 25 for collagen type II RNA expression. (N) Embryos stained at approximately
stage 35 for N-CAM and nrpl RNA. The upper embryo represents the class that stains weakly for neural markers in the protrusion; the lower
embryo shows extremely weak staining, despite the overstaining for the primary neural axis. (0) Embryos stained at approximately stage 25 for
HNF3alXFKH2 RNA. When examined in all orientations, all protrusions are negative, although edge effects and staining in the endoderm sometimes
give an impression of staining (embryo on right) (P) noggin RNA-injected embryo (160 pg; about stage 25) stained for N-CAM and nrpl RNA.
Figure 6. Xnr3 Lacks Direct Mesoderm-Inducing Activity
(Top) We injected l-cell embryos with Xnf3 or activin RNAs. Animal
caps were dissected at stage 6 (midblastula) and grown to stage 20
(early tailbud). RNA was assayed for muscle actin expression by
RNase protection assay. Each sample had ten animal caps. The lane
marked /acZ RNA is 6galactosidase RNA.
(Bottom) Embryos were injected at l-cell stage with XnA RNA, Xnr3
expression plasmid (Xnr3-pXEX), or activin RNA. Animal caps, prepared
as in (A), were analyzed at stage 11 for expression of Brachyury
by RNase protection assay.
Figure 7. Induction of Muscle Actin Transcript in Ventral Marginal
Zones by Xnr3 and noggin Expression Plasmids
Embryos were injected at the 4-M stage in the ventral marginal zone
with the indicated expression plasmids. Ventral marginal zone explants
removed at early gastrula were grown until the early tailbud
stage. RNA isolated from the explants was assayed for the presence
of muscle actin transcript by RNase protection. The muscle actin probe
also protects a smaller fragment of a cytoskeletal actin transcript,
which serves as a control for total RNA input. /ecZ is P-galactosidase.