December 1, 1996;
Expression cloning of a Xenopus T-related gene (Xombi) involved in mesodermal patterning and blastopore lip formation.
We have used a functional assay to identify a putative T-box transcription factor (Xombi
) that has the ability to induce sites of invagination in the ectoderm
of Xenopus embryos that resemble the blastopore
lip. Maternal Xombi
transcript is first localized to the oocyte''s vegetal cortex and cytoplasm
, early sources of mesoderm
-inducing signals. Soon after zygotic transcription begins, there is a wave of Xombi
expression, beginning in dorsal mesoderm
and then extending to lateral
and ventral mesoderm
, that precedes and parallels blastopore
lip formation at the border between the mesoderm
. Transcripts encoding brachyury
colocalize with Xombi
transcripts within the marginal zone; ectopic expression of Xombi
induces expression of all three mesodermal genes. In ectodermal explants, Xombi
expression is induced by the secreted mesoderm
inducers activinA, activinB, Xnrl and eFGF
, and by brachyury
, another Xenopus T-box containing gene. The time course and location of Xombi
expression, its biological activities and the partial dependence of Xombi
expression and blastopore
lip formation on fibroblast
growth factor (FGF) signaling suggest that Xombi
contributes to a traveling wave of morphogenesis and differentiation during Xenopus gastrulation.
[+] show captions
Fig. 1. Ectopic lip formation by Xombi in normal and ventralized
embryos. Xombi RNA (100 pg or 300 pg) was injected at the fourcell
stage of development into the animal pole region of a ventral
blastomere. (A) Lateral view of a normal early gastrula-stage embryo
with a slightly curved ectopic lip. (B) Lateral view of a normal late
gastrula-stage embryo with an almost circular ectopic lip. At this
stage, the endogenous blastopore is almost closed. (C) Sagittal
section through a ventralized embryo showing both endogenous and
ectopic lips. (D) A higher magnification view of the endogenous lip
shown in C. (E) A higher magnification view of the ectopic lip
shown in C. In A, B and C, the endogenous blastopore is indicated
by a white arrowhead and the ectopic lip is indicated by a black
Fig. 2. Xombi encodes a T-box protein related
to Xenopus brachyury, mouse Tbx2 and
Drosophila optomotor-blind. (A) DNA and
putative amino acid sequence of the Xombi
cDNA. The putative DNA-binding domain (the
T-box) is underlined. (B) Alignment of the
putative DNA-binding domains of several Trelated
proteins. Amino acid residues identical
to those in the putative Xombi T-box are boxed
Fig. 3. Xombi is expressed maternally and zygotically. Embryos
were harvested at the indicated stage and assayed by RT-PCR for
expression of Xombi. Elongation factor 1-a (EF-1a) expression was
also assayed as a loading control (after blastula stages) and to
illustrate the timing of the onset of general zygotic transcription.
Each PCR reaction (23 cycles) contained the cDNA equivalent of
one-tenth of an embryo.
Fig. 4. Spatial expression of Xombi
during normal development. Oocytes or
embryos were hybridized with a
digoxigenin-labeled Xombi antisense
RNA probe. (A) Animal-vegetal section
of a stage III oocyte. (B) Higher
magnification view of a different section
from the same oocyte. (C) Vegetal view
of a stage V oocyte. (D) Animal-vegetal
(side) view of a stage VI oocyte.
(E) Vegetal view of a stage 9.6 embryo.
(F) Vegetal view of a stage 10+ embryo.
The arrow denotes the blastopore groove.
(G) Vegetal view of a stage 10.5 embryo.
(H) Vegetal view of a stage 11 to 11.5
embryo. (I) Parasagittal section through a
mid-gastrula embryo. (J) Parasagittal
view of a mid-gastrula embryo that has
been split open. In I and J, the black
arrows denote the forming archenteron
and the white arrows denote the
boundary of Xombi expression.
(K) Dorsal view of a mid-gastrula stage
stained for Xombi (light blue stain) and brachyury (magenta stain). Where both co-localize in the marginal zone, a dark blue stain is detected.
In dorsal notochord-forming cells, where only brachyury is expressed, only a magenta stain is detected (arrow at anterior end). (L) Dorsal view
of a neurula stage embryo. Note that Xombi is still specifically excluded from the notochord. (M) Lateral view of an early tailbud stage embryo.
(N) Differential interference contrast (DIC) image of a posterior transverse section of an early tailbud stage embryo showing Xombi staining
only within cells in the neural tube. In A, B and D, the animal pole is at the top. In E, F, G, H, I and J, dorsal is to the right. In K, L and M,
anterior is to the right. In M and N, dorsal is at the top. All oocytes and embryos except for those in G, H, J and K were cleared in a 2:1 mixture
of benzyl benzoate and benzyl alcohol.
Fig. 5. Xombi expression in ventralized and dorsalized embryos.
Embryos were hybridized with a digoxigenin-labeled Xombi
antisense RNA probe. (A-C) Uninjected embryos. (D-F) Dorsalized
embryos. (G-I) Ventralized embryos. (A,D,G) Vegetal view of stage
9.6 embryos (about 90 minutes prior to the onset of gastrulation).
(B,E,H) Vegetal view of stage 10.5 embryos. (C, F,I) Posterior view
of stage 13 embryos. Except for A, D and G, which show cleared
sibling albino embryos, sibling pigmented embryos were used. The
arrows in B and E indicate the location of the blastopore pigment
Fig. 6. Xombi is a general mesoderm inducer. (A) Presumptive
ectoderm was explanted from embryos injected with the indicated
dose of Xombi RNA and assayed by RT-PCR for the expression of
brachyury (Xbra), Xwnt8 or goosecoid (Gsc). EF-1a expression was
assayed as a loading control. (B) Lateral view of a Xombi RNAinjected
gastrula embryo that has been stained by in situ
hybridization for Xwnt8. Xwnt8 expression (light blue stain) is
detected in the endogenous marginal zone above the forming
blastopore lip as well as in a region of cells above the forming
ectopic lip. Dorsal is to the left.
Fig. 7. Induction of Xombi expression by mesoderm-inducing
proteins. Ectodermal explants were harvested from uninjected
blastula-stage embryos and then treated with the indicated
concentration of activin A protein. In parallel, presumptive ectoderm
was also explanted from embryos that had been injected in the
animal pole at the two-cell stage with the indicated amount of
activinbB, Xnr1, eFGF or brachyury (Xbra) RNA. Explants were
harvested when sibling embryos reached the mid-gastrula stage, and
then assayed by RT-PCR for Xombi expression. EF-1a expression
was assayed as a loading control.
Fig. 8 Xombi expression and blastopore lip formation are FGFindependent
early in gastrulation but FGF-dependent late in
gastrulation. Embryos were co-stained by double in situ hybridization
for Xombi (magenta stain) and brachyury (light blue stain). A dark
blue stain is detected wherever the genes are co-expressed.
(A-C) Uninjected embryos. (D-F) Embryos injected dorsally with
XFD RNA (1 or 2 ng). (G-I) Embryos injected ventrally with XFD
RNA (1 or 2 ng). (A,D,G) Stage 10 embryos in which apically
pigmented superficial cells are just becoming apparent. (B,E,H) Stage
10.25 embryos in which the dorsal lip has formed but not extended
very far laterally. (C,F,I) Stage 11.5 embryos. All embryos are shown
in vegetal view with dorsal at the top.
Fig. 9. Kinetic model of Xombi expression and blastopore lip
formation. A vegetal view of a schematized late blastula to early
gastrula Xenopus embryo is shown. Xombi expression (blue) begins
dorsally during late blastula stages and has moved circumferentially
towards the ventral side of the embryo by early gastrula stages.
Blastopore lip formation (red) follows this traveling wave although
the site of lip formation is always several cell diameters from the
border of Xombi expression in the mesoderm. As mesoderm
involution proceeds through the nascent lip (green), Xombi is
expressed in a standing wave that forms orthogonal to the traveling
wave of lip formation.