180
C. Niehrs et al.
Xvent-} and Xvent-2 share many functional features but what distinguishes them is their effect following wild-type overexpression, where
Xvent-2 is able to ventralize embryos in a dose-dependent manner, while
Xvent-} causes milder ventralization and lethality at a higher dose
(Gawantka et al. 1995; Onichtchouk et al. 1996). In addition, unlike
VPXvent-2, VPXvent-} induces secondary embryonic axes that frequently contain notochords (Onichtchouk et al. 1998). Yet, Xvents have
very similar DNA binding domains and in both gain- and loss-of-function experiments affect expression of a similar set of marker genes
suggesting that they share the same targets. They possibly recruit distinct cofactors via their unique N-terminal domains.
The functional similarities taken together with the distinct d/v expression domains of Xvent-} and -2 support a scenario where the sum of
Xvent activity determines the specification of mesodermal cells
(Fig. 4G). The dorsolateral boundaries of Xvent gene expression are
determined by the local activity of BMP-4 (Dosch et al. 1997). In
ventrolateral mesoderm both Xvents are expressed and the sum of Xvent
activity is high, suppressing muscle and notochord and allowing mesenchyme and blood differentiation. In dorsolateral mesoderm, only
Xvent-2 is expressed which suppresses notochord but allows muscle
differentiation. Firstly, this model is supported by the finding, that
Xvents act in an additive fashion both in gain- and loss-of-function
experiments. Secondly, a high dose of Bmp-4 mRNA injection induces
both Xvent genes and represses muscle and notochord differentiation,
while a low dose of Bmp-4 induces Xvent-2 and muscle and represses
notochord differentiation (Dosch et al. 1997).
Figure 5 shows a diagram of the BMP signal transduction pathway. It
is interesting that a variety of components of the pathway appear to be
regulated by BMP signalling itself, since they show the characteristic
complex expression pattern of Bmp-4 at tadpole stage (Gawantka et al.
1998; Hata et al. 1998; Nakayama et al. 1998a; Takase et al. 1998;
Bhushan et al. 1998; Casellas and Hemmati-Brivanlou 1998; Nakayama
et al. 1998b). Indeed, there are positive feedback loops between Bmp-4
and Xvent-2 in Xenopus (Onichtchouk et al. 1996; Schmidt et al. 1996),
as well as bmp-4 and bmp-2 in zebrafish (Kishimoto et al. 1997; Nguyen
et al. 1998). The diagnostic expression pattern of Bmp-4 may therefore
be useful to screen for novel components of the pathway as was the case
for Xvent-2 (Onichtchouk et al. 1996). We have also observed the
C. Niehrs et al.
Xvent-} and Xvent-2 share many functional features but what distinguishes them is their effect following wild-type overexpression, where
Xvent-2 is able to ventralize embryos in a dose-dependent manner, while
Xvent-} causes milder ventralization and lethality at a higher dose
(Gawantka et al. 1995; Onichtchouk et al. 1996). In addition, unlike
VPXvent-2, VPXvent-} induces secondary embryonic axes that frequently contain notochords (Onichtchouk et al. 1998). Yet, Xvents have
very similar DNA binding domains and in both gain- and loss-of-function experiments affect expression of a similar set of marker genes
suggesting that they share the same targets. They possibly recruit distinct cofactors via their unique N-terminal domains.
The functional similarities taken together with the distinct d/v expression domains of Xvent-} and -2 support a scenario where the sum of
Xvent activity determines the specification of mesodermal cells
(Fig. 4G). The dorsolateral boundaries of Xvent gene expression are
determined by the local activity of BMP-4 (Dosch et al. 1997). In
ventrolateral mesoderm both Xvents are expressed and the sum of Xvent
activity is high, suppressing muscle and notochord and allowing mesenchyme and blood differentiation. In dorsolateral mesoderm, only
Xvent-2 is expressed which suppresses notochord but allows muscle
differentiation. Firstly, this model is supported by the finding, that
Xvents act in an additive fashion both in gain- and loss-of-function
experiments. Secondly, a high dose of Bmp-4 mRNA injection induces
both Xvent genes and represses muscle and notochord differentiation,
while a low dose of Bmp-4 induces Xvent-2 and muscle and represses
notochord differentiation (Dosch et al. 1997).
Figure 5 shows a diagram of the BMP signal transduction pathway. It
is interesting that a variety of components of the pathway appear to be
regulated by BMP signalling itself, since they show the characteristic
complex expression pattern of Bmp-4 at tadpole stage (Gawantka et al.
1998; Hata et al. 1998; Nakayama et al. 1998a; Takase et al. 1998;
Bhushan et al. 1998; Casellas and Hemmati-Brivanlou 1998; Nakayama
et al. 1998b). Indeed, there are positive feedback loops between Bmp-4
and Xvent-2 in Xenopus (Onichtchouk et al. 1996; Schmidt et al. 1996),
as well as bmp-4 and bmp-2 in zebrafish (Kishimoto et al. 1997; Nguyen
et al. 1998). The diagnostic expression pattern of Bmp-4 may therefore
be useful to screen for novel components of the pathway as was the case
for Xvent-2 (Onichtchouk et al. 1996). We have also observed the