Main Page Dynamic Development
The Foundations of Developmental
From Sperm and Egg to Embryo
Genetic Regulation of Development
Organizing the Multicellular
Generating Cell Diversity
Dynamic Development at a
The Developmental Biology Journal
Developmental Biology Tutorial
Overview of the role of ß-catenin in specification of the dorsal-ventral
axis of Xenopus
Randall T. Moon
Howard Hughes Medical Institute and Department of Pharmacology
University of Washington School of Medicine
Seattle, WA 98195-7370
The issue of how a fertilized egg develops dorsal-ventral polarity during
the cleavage stages has attracted many investigators, as solution of this
question would lead to understanding the hierarchy of signals leading to
formation of the gastrula organizer. It is clear that in Xenopus
embryos the multi-functional protein ß-catenin is required for formation
of the endogenous gastrula organizer, based on elegant antisense oligo-knockout
experiments of Janet Heasman and Chris Wylie. Four important issues regarding
the role of ß-catenin in axis specification are considered below:
Is endogenous ß-catenin expressed in a manner consistent with a
role in axis specification? This is an important question, as it is
irrelevant that ectopic ß-catenin induces an axis if endogenous ß-catenin
does not show greater levels, or activity, on the future dorsal side of
the embryo. Similarly, the loss-of-function experiments are hard to interpret
when you don't know where the protein even functions.
This question has now been resolved by confocal microscopy of the early
embryo. Larabell et al. in J.
of Cell Biology show that ß-catenin accumulates opposite the sperm
entry point by the end of the first cell cycle. It continues to accumulate
in dorsal (i.e., opposite the sperm entry point) but not ventral cytoplasm
through the early cleavage stages. By the 16- to 32- cell stages it accumulates
in dorsal but not ventral nuclei. Remarkably, the pattern of dorsal accumulation
of ß-catenin closely parallels the ability of transplanted dorsal
cells to induce an axis when implanted into host embryos (reviewed by Larabell
et al., 1997). Thus, ß-catenin is the first signaling molecule
to show a dorso-ventral polarity in the early embryo. Combined with the
loss-of-function data from Heasman et al., it is now clear that when fertilization
elicits a cortical rotation, and displacement of material and organelles
to the future dorsal side (Rowning
et al., 1997, Proc. Natl. Acad. Sci. USA 94: 1224-1229), this
leads to a dorso-ventral asymmetry in ß-catenin, which is required
for axis formation.
The next question is "what establishes this asymmetry in ß-catenin?"
This is partially answered by Yost, Torres et al. in Genes
and Development. The results show that a serine/threonine kinase, glycogen
synthase kinase-3 (Xgsk-3), directly phosphorylates ß-catenin in
vitro at an amino terminal site. Deletion or mutation of this site greatly
reduces phosphorylation of ß-catenin in embryos, and inhibition of
endogenous Xgsk-3 with a dominant negative protein also reduces phosphorylation
of ß-catenin. The experiments show further that phosphorylation of
ß-catenin by Xgsk-3 targets it for rapid degradation in the embryo.
When this phosphorylation by Xgsk-3 is inhibited in the embryo, ß-catenin
accumulates in both the cytoplasm and in the nucleus. The nuclear ß-catenin
likely functions in conjunction with HMG-box architectural transcription
factors to modulate expression of specific genes (Molenaar
et al., Cell, August, 1996).
Thus, Yost, Torres et al. propose that after fertilization of Xenopus
eggs, dorsal-ventral differences in Xgsk-3 activity may arise, due to presently
unknown mechanisms, though Rowning et al. (1997) suggest how this
could be linked to cortical rotation. These differences in kinase activity
lead to dorsal-ventral differences in ß-catenin phosphorylation and
thus steady-state levels. Greater dorsal ß-catenin then participates
in establishing dorsal mesoderm, which leads to formation of the gastrula
The third question is "what lies downstream of ß-catenin?"
Restated, we know that injection of Wnts (McMahon and Moon, 1989, Cell 58:
1075-1084), or ß-catenin (work of Gumbiner and colleagues), on the
ventral side of an embryo, induces a large number of genes and an ectopic
axis. Therefore, any or all of these genes might normally be regulated by
ß-catenin in the unperturbed embryo. However, are these genes a direct
target of ß-catenin, or is this highly indirect, perhaps working without
ß-catenin ever entering a nucleus?
The issue of whether any dorsally expressed genes are the direct target
of ß-catenin has just been answered. Brannon et al. in Genes
and Development (in press) show that the HMG Box factor XTcf-3 directly
binds the siamois promoter. In the absence of ß-catenin, XTcf-3
inhibits gene expression. However, on the dorsal side of the embryo, ß-catenin
binds the XTcf-3, and now activates the gene. This is notable because siamois
is a homeobox gene likely playing a major role in specifying formation of
Spemann's Organizer. So, a dorso-ventral difference in ß-catenin forms
within an hour or two of fertilization, directly regulating a key homeobox
gene in the blastula, thus contributing to forming Spemann's Organizer on
the dorsal side of the gastrula.
The next issue is "where in the cell does ß-catenin function?"
From the discussion above, you would probably say that since ß-catenin
accumulates in dorsal nuclei, and directly interacts with a transcription
factor, ß-catenin probably works in the nucleus. However, keep in
mind that ß-catenin is found at the plasma membrane in association
with cadherins, at the plasma membrane in association with the tumor suppressor
promoter APC and microtubles, in the cytoplasm, and in the nucleus. So,
it cannot be presumed that it must work only in the nucleus to "signal"
a Wnt response.
Recently, Klymkowsky and colleagues (Merriam et al.. 1997. Develop.
Biol.185: 67-81) found that ectopic membrane-tethered plakoglobin (related
to ß-catenin) was active in axis induction. They speculated that plakoglobin,
and ß-catenin, work by keeping a repressor of dorsal cell fate
out of the nucleus, and speculate this repressor is XTcf-3 (Molenaar et
al., 1996). Miller
and Moon (J. Cell Biol., 1997, in press) performed similar experiments
with ectopic membrane tethered ß-catenin, and likewise found it to
be active in axis induction. However, they came to a very different conclusion
and interpretation. They find that ectopic membrane-tethered ß-catenin,
and even ectopic wild-type plakoglobin, bind to endogenous APC. Since APC
normally plays a role in degrading ß-catenin, this results in the
artefactual stabilization and accumulation of ß-catenin. As ß-catenin
levels rise, it accumulates in nuclei. So, if this nuclear ß-catenin
were responsible for the axis inducing activity of membrane-tethered ß-catenin,
you might expect that co-expression of cadherin (which would titrate the
free ß-catenin to the membrane) would block the axis inducing activity
of membrane tethered ß-catenin. On the other hand, if ß-catenin
functions to keep a repressor out of the nucleus, then cadherin probably
should not block its axis inducing activity. The results show that cadherin
does block the axis inducing activity of membrane-tethered ß-catenin.
A final answer of where ß-catenin functions in the cell during Wnt
signaling is under investigation in several labs, in a genetic background
where ß-catenin is not present.
Return to Initiating the Embryonic
Body Plan: Dorsalization of the Xenopus Embryo.
A current model of Wnt signaling has recently been posted on the Moon laboratory Web site.