November 1, 2001;
Proteolytic cleavage of Chordin as a switch for the dual activities of Twisted gastrulation in BMP signaling.
Dorsoventral patterning is regulated by a system of interacting secreted proteins involving BMP, Chordin
and Twisted gastrulation
). We have analyzed the molecular mechanism by which Tsg
regulates BMP signaling. Overexpression of Tsg
mRNA in Xenopus embryos has ventralizing effects similar to Xolloid
, a metalloprotease that cleaves Chordin
. In embryos dorsalized by LiCl treatment, microinjection of Xolloid
mRNA restores the formation of trunk
structures, indicating an increase in BMP signaling. Microinjection of Tsg
mRNA leads to the degradation of endogenous Chordin
fragments generated by Xolloid
. The ventralizing activities of Tsg
require an endogenous Xolloid-like
activity, as they can be blocked by a dominant-negative Xolloid
mutant. A BMP-receptor binding assay revealed that Tsg
has two distinct and sequential activities on BMP signaling. First, Tsg
a better BMP antagonist by forming a ternary complex that prevents binding of BMP to its cognate receptor. Second, after cleavage
competes the residual anti-BMP activity of Chordin
fragments and facilitates their degradation. This molecular pathway, in which Xolloid
switches the activity of Tsg
from a BMP antagonist to a pro-BMP signal once all endogenous full-length Chordin
is degraded, may help explain how sharp borders between embryonic territories are generated.
[+] show captions
Fig. 1. Tsg and Xolloid ventralize the Xenopus embryo. (A) Uninjected stage 18 control embryos stained for krox20 and otx2. (B) Embryos microinjected at the four-cell stage four times at the animal pole with 100 pg Xolloid, (C) 250 pg mouse Tsg or (D) both mRNAs. Same results were obtained using Xenopus Tsg mRNA (data not shown). For each mRNA combination at least 25 embryos were analyzed. (E-H) LiCl-treated embryos. (E) Radially dorsalized LiCl-treated embryo (n=40; dorsoanterior index, DAI=9.5); (F) embryo microinjected into a single blastomere of the marginal zone at the 16-cell stage with 200 pg Xolloid (26% with trunk/tail structures, n=23, DAI=8); (G) 500 pg Xenopus Tsg (32%, n=33, DAI=8.1); or (H) both mRNAs (51%, n=27, DAI=7). Lineage tracing with lacZ and Red-Gal shows that the cells injected with Xenopus Tsg or Xolloid mRNA contributed mostly to ventroposterior mesoderm in the rescued tail region.
Fig. 3. The ventralizing activity of Tsg is dependent on endogenous Xolloid. (A) Albino embryos were microinjected once into an animal cell with 500 pg Xenopus Tsg and lacZ mRNA at the 32-cell stage and krox20 in situ hybridization performed. Note that krox20 expression is reduced on the injected side. (B) krox20 in situ hybridization of uninjected embryos at neural plate stage. (C) Embryos injected with 250 pg Xenopus Tsg mRNA, (D) 250 pg dominant negative (dn) Xld mRNA or, (E) co-injected with Xenopus Tsg and dnXld mRNAs (n=25 or more for each mRNA combination). All embryos were injected 4 times in the animal pole at the four-cell stage. Similar results were obtained using mouse Tsg mRNA (data not shown). (F) Ventral injection of 5 pg mouse Chd mRNA induces secondary axes; (G) injection of 5 pg mouse Chd and 500 pg mouse Tsg mRNA; (H) injection of 5 pg mouse Chd and 500 pg dnXld mRNA, and (I) 5 pg mouse Chd, 500 pg mouse Tsg and 500 pg dnXld mRNAs. (J,K) Injection of 500 pg of dnXld mRNA(J) and uninjected controls (K). Similar results were obtained using Xenopus Tsg and Xenopus chd mRNA (data not shown). Injection of 5 pg of chd mRNA induced strong secondary axes in 47% of the cases (F); these axes were not seen after chd and Tsg co-injection (G), but in 14-50% of the embryos injected with chd, Tsg and dnXld mRNA, strong secondary axes were rescued (I). Note in J that dnXld was unable to induce secondary axes on its own. (L) Ventral injection of 5 pg Xenopus Chd mRNA induces secondary axes (44%). (M) Co-injection of 500 pg dnTsg mRNA reduced the axis-forming activity of Xenopus Chd (14%). (N) 20 pg Xenopus CR1 mRNA induced weak secondary axes. (O) Co-injection of dnTsg mRNA enhanced the secondary axis phenotype caused by Xenopus CR1. For all injections, at least 35 embryos were analyzed. (P) RT-PCR analysis of animal cap explants injected with the indicated combinations of mRNAs and analyzed at stage 25; total amounts of mRNA injected per embryo were 800 pg Xenopus Tsg, 40 pg Xenopus chd and 1 ng dnXld. NCAM is a pan-neural marker, α-Glo (α-globin), a ventral mesoderm marker, and EF1α was used as a loading control.