June 1, 1997;
Bmp-4 acts as a morphogen in dorsoventral mesoderm patterning in Xenopus.
The marginal zone
is a ring of tissue
that gives rise to a characteristic dorsoventral pattern of mesoderm
in amphibian embryos. Bmp-4
is thought to play an important role in specifying ventral
mesodermal fate. Here we show (1) that different doses of Bmp-4
are sufficient to pattern four distinct mesodermal cell
types and to pattern gene expression in the early gastrula marginal zone
into three domains, (2) that there is a graded requirement for a Bmp signal in mesodermal patterning, and (3) that Bmp-4
has long-range activity which can become graded in the marginal zone
by the antagonizing action of noggin
. The results argue that Bmp-4
acts as a morphogen in dorsoventral patterning of mesoderm
[+] show captions
Fig. 1. Different doses of Bmp-4 are necessary and sufficient for
patterning terminal differentiation of mesoderm. 4-cell-stage
embryos were microinjected into all blastomeres with the indicated
amount of mRNA. Dorsal marginal zones (DMZ) were excised
(encompassing the 60° dorsal lip only) from embryos injected with
Bmp-4 mRNA and ventral marginal zones (VMZ) were excised from
embryos injected with DmTFR11. Note the successive peaks of
differentiation. For each injected dose at least 12 explants were
analyzed for the presence of the indicated marker tissues, the
frequency of which is given in percent. Identification of welldifferentiated
tissues was by morphological criteria by two
individuals scoring independently. 100% is the total number of
explants scored and each individual explant may contain multiple
tissues. For the highest dose of injected DmTFR11 mRNA, 4-cell
embryos were only injected into two ventral blastomeres in such a
way that the injections left little scars. This permitted safe
identification of the ventral side even in gastrula embryos that
contained secondary dorsal lips.
Fig. 2. XMyf-5 is a molecular marker for the dorsolateral marginal zone. Whole-mount in
situ hybridisations of XMyf-5, XMyoD, Xvent-1, Xvent-2 and noggin. Xmyf-5 and XMyoD
expression at the indicated stages are shown in the top two panels. Stage 10+ embryos are
in vegetal view with the dorsal side facing up, stage 12 embryos in dorsal view with
anterior facing left. The XMyf-5 expression domain always remains dorsolateral until late
gastrulation, while XmyoD is expressed additionally in the lateroventral domain of the
gastrula and the circumblastoporal collar in stage 12 embryos. Lower panel shows stage
10+ gastrulae. Note that the expression of Xvent-1 is nested within the domain of Xvent-2
expression. These expression patterns molecularly define a dorsal, dorsolateral and
lateroventral domain, as shown schematically in the lower right. The domains correspond
to expression of noggin (blue), XMyf-5 (white) and Xvent-1 (red). All embryos are shown
in vegetal view with the dorsal side up.
Fig. 3. Different doses of Bmp-4 are necessary and sufficient for
patterning of the marginal zone. 4-cell-stage embryos were
microinjected into all blastomeres with either 50 (low) or 150 pg
(high) Bmp-4 mRNA, or either 100 (low) or 1000 pg (high) dominant
negative Bmp receptor (DmTFR11) mRNA per blastomere as
indicated. Embryos were processed for whole-mount in situ
hybridisation at stage 10+. All embryos are shown in vegetal view
with the dorsal side up. Typically, two sets of titration experiments by
microinjections were carried out with two-fold dose increments and at
least five embryos per experiment. Of a population of injected
embryos, a representative specimen is shown for the respective dose.
For simplicity, only key doses are displayed (wt, uninjected controls).
Fig. 4. Dose-dependent modulation of Bmp-4 activity by noggin.
(A) 4-cell-stage embryos were microinjected into all blastomeres
with noggin or Bmp-4 mRNA or both with the indicated dose in ng
per blastomere. Dorsal (DMZ, top) or ventral (VMZ, bottom)
marginal zones were excised at the gastrula stage and cultured until
sibling embryos reached late gastrula. Explants were analyzed for
expression of EF1-a, gsc and Xvent-1, as indicated. WE, whole
embryo; -RT, -reverse transcription. (B) 4-cell-stage embryos were
microinjected into all blastomeres with 30 (low) or 500 pg (high)
noggin mRNA per blastomere. Embryos were processed for wholemount
in situ hybridisation at stage 10+ with the probes indicated.
All embryos are shown in vegetal view with the dorsal side up.
Fig. 5. Direct and long-range action of Bmp-4 and noggin. (A) (Top) 32-cell-stage embryos
were microinjected with 0.2 ng b-galactosidase lineage tracer mRNA containing a nuclear
localization signal and either without (lacZ) or with 30 pg Bmp-4 or 500 pg Xsmad1 (XMad-1)
mRNA as indicated into two marginal zone blastomeres, followed by treatment with LiCl.
Embryos were processed for whole-mount in situ hybridisation at stage 10+ with Xvent-1. Note
that Xvent-1 expression extends beyond the injected cells (light blue nuclei) in the Bmp-4-
injected embryo, while it is restricted to the b-galactosidase-positive cells in Xsmad1-injected
embryos. Embryos are shown in lateral view with the animal side up. The insets show a
magnification of the highlighted area. (Bottom) 4-cell-stage embryos were uninjected (control)
or microinjected into all blastomeres with Bmp-4 or Xsmad1 mRNA. Note that the phenotype of
both Bmp-4- as well as Xsmad1-microinjected embryos is complete Bauchstück-type
ventralization. (B) 32-cell-stage embryos were microinjected with 0.2 ng b-galactosidase
lineage tracer mRNA containing a nuclear localization signal and either without (lacZ) or with
25 pg (low) or 100 pg (high) noggin mRNA, as indicated, into two marginal zone blastomeres.
Embryos were processed for whole-mount in situ hybridisation at stage 10+ with Xvent-1 or
Xvent-2 as indicated. Note that the Xvent-1 and Xvent-2 repression halos extend beyond the
injected cells (light blue nuclei) in the noggin-injected embryos and that the size of the halo is
dose dependent. Embryos are shown in lateral view with the animal side up.