Download Dishevelled: at the crossroads of divergent

Mechanisms of Development 83 (1999) 27±37
Dishevelled: at the crossroads of divergent intracellular
signaling pathways
Michael Boutros, Marek Mlodzik*
Developmental Biology Programme, European Molecular Biology Laboratory, Meyerhofstrasse 169117, Heidelberg, Germany
Received 19 February 1999; received in revised form 26 February 1999; accepted 26 February 1999
Abstract
During the development of multicellular organisms the formation of complex patterns relies on speci®c cell-cell signaling events. For
tissues to become spatially organized and cells to become committed to specialized fates it is absolutely crucial for proper development that
the underlying signaling systems receive and route information correctly. Recently, a wealth of genetic and biochemical experimental data
has been collected about prevalent evolutionary conserved signaling families, such as the Wnts, Dpp/BMPs, and Hedgehogs, in ¯ies, worms,
and vertebrates. Paradoxically, members of a particular signaling family often have receptors with similar biochemical binding properties,
though they activate different intracellular pathways in vivo and can be phenotypically distinguished. How are their speci®c biological
responses then generated? With respect to signaling speci®city in Wnt pathways, Dishevelled is an intriguing protein; in Drosophila
melanogaster it is required in two distinct signaling pathways, that share Frizzled receptors of similar structure, but have distinct intracellular
signaling routes. Recent results suggest that Dishevelled is a multifunctional protein at the crossroads of divergent Wnt/Fz pathways.
Dishevelled appears to be a key factor in Wnt signaling to `read' signals coming from the plasma membrane and route them into the correct
intracellular pathways. q 1999 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Dishevelled; Frizzled; Signaling speci®city; Wnt signaling; Planar polarity
1. Introduction
Understanding the molecular basis for signal transduction
in pattern formation and cell fate speci®cation is of key
importance in modern biology. During ontogenesis of
animals, cell proliferation and differentiation are coordinated to generate ordered patterns and cell fates are induced
in uncommitted precursor cells in a spatially and temporally
correct fashion, often through a complex system of intercellular communication routes. The resulting activation of
intracellular signaling pathways leads to highly speci®c
responses, such as cell proliferation, cell migration, or terminal differentiation. The response systems are therefore
among the most important elements that govern many
aspects in the development of higher eucaryotes. Signaling
cascades and often even the context in which they are used
are well conserved between vertebrates and invertebrates.
The dissection of signaling mechanisms in genetically
amenable model organisms like Drosophila melanogaster
or Caenorhabditis elegans has identi®ed new components;
epistasis analysis has been used to establish the ¯ow of
* Corresponding author. Tel.: 1 49-6221-387 303; fax: 1 49-6221-387.
E-mail address: [email protected] (M. Mlodzik)
information in vivo. Although many of the signaling molecules have been identi®ed, the biological networks and the
connection and insulation of signaling pathways are still
poorly understood.
Signaling is often context speci®c. In response to different upstream signals, cells often use the same signaling
molecule for different purposes, as in the example of
Dishevelled that we discuss in this review. Also, the `activation' of the same protein can route information into
distinct downstream signaling cascades. A well described
example is the function of members of the p21 small
GTPase Ras-superfamily in cellular processes. The small
GTPase RhoA can be activated by different extracellular
stimuli and then relay the signal to different effector pathways leading to either activation of transcriptional
responses or cytoskeletal rearrangements (reviewed in
Van Aelst and D'Souza-Schorey, 1997). Interestingly,
effector proteins bind to overlapping regions of the small
GTPase and can be singled out by speci®c point mutations
(Sahai et al., 1998), raising the question how downstream
proteins might know, when they should bind.
The secreted signaling molecules of the Hedgehog, TGFb or Wnt families can induce very speci®c cellular
responses from proliferation and morphogenesis to cell
0925-4773/99/$ - see front matterq 1999 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0925-477 3(99)00046-5
28
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
fate induction and terminal differentiation. Although the
signaling factors cannot always be distinguished by their
properties in vitro assays, the outcome of the signaling
events in vivo is often very speci®c. Signaling molecules
of the Wnt family are representative examples. Seven-pass
transmembrane receptors of the Frizzled (Fz) family have
recently been identi®ed as receptors for the Wnt family of
secreted growth factors (reviewed in Cadigan and Nusse,
1997). Biochemical binding assays have suggested that
many members of the Wnt family can bind most (if not
all) of the Fz receptors. However, when co-injected into
Xenopus embryos only some Wnt-Fz combinations give a
biological response (Moon et al., 1997). In another example,
patterning of the apical ectodermal ridge during vertebrate
limb development, members of the Wnt family can induce
distinct cellular responses by using different intracellular
signaling pathways (Kengaku et al., 1998; and see below).
What can be the mechanisms by which speci®city is generated?
Dishevelled (Dsh) is a signaling molecule that functions
in at least two distinct pathways in Drosophila, where it was
originally identi®ed. It is required for the transmission of
wingless (Wg) signals and for the generation of epithelial
planar polarity (tissue polarity). Genetic and biochemical
data indicate that Dsh acts downstream of Fz receptors in
both pathways. The planar polarity and the Wg pathways
might require different members of the Fz-receptor family.
Loss-of-function mutations in the Frizzled gene cause
planar polarity phenotypes, while Dfz2 was shown by overexpression studies to function as a Wg receptor during
imaginal disc development (Bhanot et al., 1996; Cadigan
et al., 1998; Zhang and Carthew, 1998). Recent reports
show that to abolish Wg signaling in embryonic development, one needs to remove both Fz and Fz2 activity (Bhat,
1998; Kennerdell and Carthew, 1998; Mueller et al., 1999).
Although previous reports have suggested a common Wnt/
Fz signaling mechanism in wg and planar polarity signaling
(Tomlinson et al., 1997), recent experiments implicate two
distinct pathways downstream of Frizzled receptors for
planar polarity and Wg signaling (Axelrod et al., 1998;
Boutros et al., 1998). This raises the question of how Dsh
as a common protein downstream of related receptors can
activate distinct effector pathways.
In this review, we will address the role of Dsh in Wnt
signaling. We will ®rst give an overview of the known
biological functions of Dsh during the development of
different organisms. Subsequently, we will discuss the role
of Dsh in signaling pathways mediated by the Fz-receptor
family, and the speci®c functions Dsh has in activating its
downstream effector cascades. We will examine the
proposed molecular models for Dsh in participating and
distinguishing between the planar polarity and Wg signaling
pathways. There is increasing evidence that these different
modes of action of Dsh are conserved throughout the animal
kingdom. A further understanding of its functions could
elucidate some principle mechanisms at a molecular level
for how biological networks achieve speci®city.
2. Dishevelled in wingless and epithelial planar polarity
signaling
2.1. Dishevelled is required for wg signaling
The ®rst dishevelled (dsh) mutant reported in Drosophila
was a viable allele, dsh 1 (Fahmy and Fahmy, 1959), that
causes defects in the arrangement of bristles on the wing
and thorax, hence its name, that were later shown to result
from planar polarity defects. Further genetic screens have
produced null alleles that are embryonic lethal when lacking
both maternal and zygotic function (Perrimon and Mahowald, 1987), and have phenotypes that are reminiscent of
wingless and armadillo mutant embryos (NuÈsslein-Volhard
and Wieschaus, 1980). These similarities suggested that dsh
has a function in a signaling pathway with wg and armadillo
(arm) (Fig. 1A). Thus, the dsh gene appears to function in
two different cellular responses during Drosophila development.
The wingless (wg) gene has been shown to be required
during early and late stages of Drosophila development. In
the embryo, wg is involved in segmentation and patterning
of the embryonic epidermis, in the development of head
structures, the nervous system, the midgut and the malphigian tubules (reviewed in Klingensmith and Nusse, 1994).
In the adult ¯y, wg is required during imaginal disc development, for example, for dorso-ventral patterning in leg and
wing development, in bristle speci®cation and in eye development (Klingensmith and Nusse, 1994; Ma and Moses,
1995; Treisman and Rubin, 1995; Heberlein et al., 1998;
Wehrli et al., 1998).
Genetic studies of dsh have revealed similar phenotypes
to those of wg alleles and demonstrated its requirement for
wg signaling throughout development (Klingensmith et al.,
1994; Theisen et al., 1994). Besides dsh, zeste-white-3 (zw3)
and arm are required for wg signaling. arm, zw3 and dsh
have strictly cell autonomous phenotypes indicating a function in the interpretation and intracellular relay of wg signaling. In order to establish the genetic epistasis, Noordermeer
et al., (1994) analyzed double mutants generated by expressing wg in combination with the absence of maternal and
zygotic dsh and arm. Ectopic expression of Wg leads to an
expanded engrailed (en) expression domain, which, the
authors reasoned, would be modi®able by mutations in
genes which are genetically downstream, and therefore are
required for the reception and intracellular transmission
rather than for the establishment of the signal (Fig. 1A).
The experiments showed indeed that the ectopic en domain
was modi®ed in dsh and arm germline clones; dsh and arm
mutants displayed the same phenotype with or without ectopic Wg expression. These experiments led to the conclusion
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
that dsh and arm are downstream of wg and are required for
the intracellular transduction of the signal.
2.2. Dsh is required in the epithelial planar polarity
signaling pathway
The dsh 1 allele is fully viable with no apparent phenotypes besides the planar polarity defects. Subsequent clonal
analysis of dsh null alleles in imaginal discs has revealed
similar severe defects in planar polarity. This showed that
the dsh gene is required for polarity generation and dsh 1 is a
strong planar polarity speci®c allele (Theisen et al., 1994).
Epithelial planar polarity phenotypes in Drosophila are
characterized by the misorientation of cells within epithelia,
in the wings, thorax and eyes (e.g. Fig. 1B) (Adler, 1992;
29
Gubb, 1993; Theisen et al., 1994; Zheng et al., 1995). In the
wing, each cell orients itself along the proximal-to-distal
axis and generates a hair pointing distally. Mutations in
the tissue polarity genes result in abnormal hair orientation
(Adler, 1992). In the eye, polarity is re¯ected in the mirrorsymmetric arrangement of ommatidial units relative to the
dorso-ventral midline (the equator). This pattern is generated early in larval development when ommatidial preclusters rotate 908 towards the equator, adopting opposite
chirality depending on their dorsal or ventral positions.
Planar polarity defects result in the loss of mirror-image
symmetry, with the ommatidia misrotating and adopting
random chirality or remaining symmetrical (Fig. 1B)
(Adler, 1992; Zheng et al., 1995; Strutt et al., 1997; Boutros
et al., 1998).
Fig. 1. The roles of dishevelled in Drosophila and Xenopus development. (A) dsh mutant phenotype during embryogenesis reveals defects in the establishment
of segment borders (segment polarity class) and is very similar to wg mutants. A wild-type embryo is shown for comparison. The left panel shows the genetic
pathway for wg signaling in this process. (B) The planar polarity phenotypes of the dsh 1 allele. Left panels show the ommatidial arrangement in a dorsal area of
the eye (wild type on top and dsh 1 mutant below), with schematic presentation on the side (black arrow: correct alignment, red arrow: misrotated ommatidium,
green arrow: wrong or no chiral form; arrows point towards the dorso-ventral midline of symmetry, the equator). Middle panels show the planar polarity wing
phenotype; and right panel shows the genetic pathway required for the interpretation of the planar polarity signal. (C) The role of Dsh in Xenopus axis
induction. Overexpression assays of Xenopus components of the Wnt signaling pathway causes axis bifurcation; compare WT with the injected Xdsh (OEXdsh) embryos. The Wnt pathway implicated in axis generation in Xenopus is similar to Wg signaling in Drosophila.
30
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
The phenotypes of different planar polarity mutants also
underline differences in the mechanistic aspects of polarity
generation in the wing and the eye. In the wing every cell
adjusts its polarity independently and reorganizes its cytoskeleton so that an actin spike that will give rise to the ®nal
hair in the adult is generated at the distal vertex of each cell
(Fig. 1B). Planar polarity defects are visible by the mislocalization of the actin spike and in the worst case by the
generation of multiple spikes and ultimately hairs. Thus,
several reports have suggested that planar polarity signaling
directly affects cytoskeletal rearrangement (Eaton et al.,
1996), for example, by asymmetric subcellular localization
of a `nucleation' component. In the eye, however, a group of
cells makes a coordinated movement. Each developing
ommatidial precluster consists of 5-8 cells at the time of
the interpretation of a `polarity signal'. In the eye, planar
polarity signaling affects the organization of groups of cells
and generates a difference among the R3/R4 photoreceptor
cells of the cluster, suggesting that the primary response
includes cell fate speci®cation and cell-cell communication
(Fig. 1B) (Zheng et al., 1995). Supporting this, it has been
shown that the Delta gene, a ligand for the Notch receptor, is
a transcriptional target of Fz/Dsh signaling in R3/R4 cell
speci®cation (Fanto and Mlodzik, 1999). The phenotypic
differences of planar polarity in different tissues might
re¯ect different cellular programs downstream of a general
`polarity signal'.
Based on mutant phenotypes, prickle-spiny legs (pk-sple),
rhoA and strabismus (stbm) are required for planar polarity
generation in all tissues (Adler et al., 1990; Gubb, 1993;
Strutt et al., 1997; Wolff and Rubin, 1998). Unlike dsh,
these genes do not display wg phenotypes. In addition,
several genes have been identi®ed that are required for
planar polarity generation in speci®c tissues. For example,
mutations in the genes fuzzy, inturned and multiple wing
hairs (mwh) only affect polarity in the wing (Adler et al.,
1994), whereas mutations in nemo and roulette appear only
to cause defects in the eye (Choi and Benzer, 1994). These
observations indicate that whereas a set of genes, including
fz and dsh, are required for all aspects of planar polarity
generation and is probably involved in the general mechanism of `reading' a polarity signal, other genes like fuzzy or
nemo are acting as effectors of the polarity signaling pathway in speci®c tissues.
Genetic epistasis analysis has placed dsh downstream of
fz in polarity signaling (Adler et al., 1994; Krasnow et al.,
1995). However, in contrast to dsh, other components of Wg
signaling pathway, such as Zw-3, Arm and Pan, do not
display planar polarity defects (Axelrod et al., 1998;
Boutros et al., 1998). A combination of genetic and
biochemical studies has demonstrated that the planar polarity signaling pathway downstream of Fz consists of Dsh and
the small GTPases of the Rho subfamily (RhoA and Rac),
and leads to the activation of a JNK-type MAPK module
(Strutt et al., 1997; Boutros et al., 1998). Thus, a signaling
cascade that is distinct from the `classical' Wg/Wnt path-
way has emerged for Fz signaling in planar polarity (Fig.
1B). However, it remains unknown how these components
are linked, e.g. what the intermediates are between Fz and
Dsh, and between Dsh and the small GTPases (see below).
3. Functional studies of Dishevelled
3.1. Molecular features of Dishevelled
Despite the genetic requirement for Dsh in wg and polarity signaling, its molecular functions remain unclear. In situ
analysis reveals that dsh mRNA is ubiquitously expressed in
the embryo and imaginal discs at all stages of development.
The dsh gene encodes a 623 amino acid protein with no
obvious similarities to proteins with known functions (Klingensmith et al., 1994; Theisen et al., 1994; Yanagawa et al.,
1995). Several parts of Dsh have homologous sequences in
other proteins, which might give some ideas about its molecular functions. Moreover, homologs of Dsh have been
identi®ed in other organisms, in nematodes as well as in
vertebrates. In vertebrates several Dsh homologs have
been described so far, one in Xenopus, and three in
mammals (see sequence alignment and phylogenetic tree,
Figs. 2 and 3).
All Dsh genes have three highly conserved domains. An
N-terminal DIX (Dishevelled-Axin) domain extends for 80
amino acids and was also found in the Axin protein, which
seems to be a scaffolding factor for Wnt signaling (Zeng et
al., 1997; Behrens et al., 1998). Sequence analysis reveals
that Dsh has a PDZ domain. Many other signaling proteins
contain PDZ domains which can bind to a conserved stretch
of amino acids at the C-termini of receptors or form homotypic complexes with other PDZ domains. Multi-domain
PDZ domain proteins have been proposed to integrate
signaling molecules into larger complexes. For instance,
InaD and GRIP can cluster receptors and signaling proteins
(reviewed in Ponting et al., 1997). A third domain at the Cterminal part of Dsh, called the DEP domain (DishevelledEGL-10-Pleckstrin), can be found in signaling factors such
as the RGS protein family, that are involved in G protein
signaling (Ponting and Bork, 1996). It is not yet known
whether DEP domains play an active role in regulating G
proteins. Although Dsh and its domains are highly
conserved from Drosophila and C. elegans to mouse and
human, Dsh does not have an apparent catalytic biochemical
activity, nor have any clear interacting partners for Dsh been
identi®ed.
3.2. Activities of Dishevelled in wg and planar polarity
signaling
Genetic and biochemical analysis of wg signaling in
Drosophila suggests that wg activates dsh, which inactivates
the zw3/sgg kinase. In the absence of signaling, Arm is
targeted for proteolytic degradation through phosphorylation by Zw3. Upon Wg signaling, Arm accumulates in the
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
31
Fig. 2. Alignment of Dishevelled protein sequences. Three Dishevelled proteins have been identi®ed in mice and humans. The CeDsh sequence was shortened
by 100 non-homologous amino acids at the N-terminus. Sequences were aligned with ClustalX.
32
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
Fig. 3. (A) Phylogenetic tree of Dishevelled in different species (calculated by ClustalX). Schematic representation of protein domains in Dishevelled and the
behavior of deletion constructs in assays for wg, planar polarity signaling in Drosophila, secondary axis induction and membrane localization. Required
sequences are annotated in blue; grey boxes indicate sequences that are partially required or which functions are unclear.
cytoplasm and nucleus, where it activates Wg target genes
by complexing with Pan/TCF transcription factors. What is
the role of Dsh in this cascade? Yanagawa et al. (1995) have
shown, that overexpression of Dsh in tissue culture cells is
suf®cient to stabilize Arm independent of a Wg signal. This
suggests, that Dsh can autonomously activate downstream
signaling molecules once cells have suf®cient Dsh protein
levels at the presumably correct subcellular localization.
Further deletion analysis experiments showed that the Dsh
DEP domain is not required to induce Arm accumulation.
Furthermore, Yanagawa et al. (1995) observed that Wg
signaling leads to phosphorylation of Dsh on Ser and Thr
residues and enrichment of the phosphorylated species in
the membrane fraction. Membrane relocalization of Dsh has
also been described for its mammalian counterparts in Wnt
responsive cell lines (Steitz et al., 1996). These experiments
suggest that upon Wnt signaling part of the cytoplasmic
pool of Dsh is phosphorylated and recruited to the
membrane.
The observed phosphorylation of Dsh concurrent with Wg
pathway activation suggests that Dsh becomes `activated' by
an unknown Ser/Thr kinase. By biochemical puri®cation,
Willert et al. (1997) identi®ed casein kinase 2 as a kinase
that binds and phosphorylates Dsh in vitro and in embryo
extracts. However, the functional relevance of Dsh phosphorylation is unclear, as no mutants have been tested in
vivo. The causal relationship between Dsh `activation' and
its phosphorylation is also obscured by the fact that by overexpressing Dfz2 in cell lines, Dsh phosphorylation and stabilization of Arm are uncoupled (Willert et al., 1997). Thus,
although Dsh phosphorylation is concurrent with Wg pathway activation, its biological role remains unclear.
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
Interestingly, the intracellular signaling pathways downstream of Dsh are different in planar polarity and wg signaling. The similarity of Frizzled and Dfz2 and their ability to
bind other Wnt molecules suggests that the polarity signal
might be a Wnt and that wg and planar polarity signaling
might use the same intracellular signaling pathway.
However, analysis of wg signaling-de®cient arm, pan/TCF
and zw3/sgg clonal tissue in the wing (Axelrod et al., 1998)
and eye (Boutros et al., 1998) reveals no planar polarity
related phenotypes, indicating that the wg and polarity pathways are distinct downstream of Dsh. Dominant genetic
interactions and rescue experiments in the eye imaginal
disc suggest that signaling is rather routed through a Rho/
Rac and JNK/SAPK-like kinase cascade (Strutt et al., 1997,
Boutros et al., 1998). Furthermore, overexpression of Dsh
induces JNK-cascade activation (Boutros et al., 1998; Li et
al., 1999). These experiments indicate that Dsh is involved
in two distinct Wnt signaling pathways, and raises the question of how Dsh routes the information to the right effector
cascade in each case.
3.3. Modular nature of Dsh
One possibility is that the mechanism whereby Dsh
routes information into the different intracellular pathways
lies in its domain composition (Fig. 3). By an in vivo
structure-function approach, using transgenic Drosophila
speci®cally expressing domain deletion mutants of Dsh
that rescue either the embryonic wg-pathway related dsh
phenotype or the planar polarity defects, it has been shown
that the DEP domain is crucial for polarity signaling,
whereas the DIX domain is dispensible (Axelrod et al.,
1998; Boutros et al., 1998). This conclusion has been
further supported by the molecular nature of the dsh 1 allele,
which has a single amino acid substitution in a conserved
residue of the DEP domain (Axelrod et al., 1998; Boutros
et al., 1998). The DEP domain requirement has also been
demonstrated in tissue culture experiments, in which Dsh
mutants without a DEP domain do not activate the JNK
pathway (Boutros et al., 1998; Li et al., 1999). The DIX
domain is essential for wg signaling both in vivo and in
vitro (Yanagawa et al., 1995;Axelrod et al., 1998; Boutros
et al., 1998;). In contrast, whereas the DEP domain (or
parts thereof) appears required for in vivo rescue of the
wg-pathway related dsh phenotype, it is dispensible for
Arm/b -cat stabilization in cell culture overexpression
assays (Yanagawa et al., 1995). Further deletion analysis
revealed that most of the DEP domain is indeed dispensible
for wg signaling (Axelrod et al., 1998, Rothbacher et al.,
1999). In summary, these experiments show that Dsh uses
different domains in a modular way to activate distinct
downstream effector cascades: the DIX domain is essential
for Arm stabilization and dispensible for planar polarity
signaling, whereas the DEP domain is essential for planar
polarity signaling.
The requirements of the distinct domains for membrane
33
association have also been analyzed. Injection assays in
Xenopus animal caps with either Drosophila or Xenopus
Dsh have shown that the DEP domain is essential and suf®cient for its membrane localization, whereas neither the DIX
nor the PDZ domains appear to have a role in this process
(Axelrod et al., 1998; Rothbacher et al., 1999). This
suggests that membrane localization is important for polarity signaling. Although the DEP domain is crucial for
membrane localization, it is puzzling that the amino acid
mutation in the DEP domain of the dsh 1 allele has little
effect on membrane localization (Axelrod et al., 1998).
4. Dishevelled in vertebrates
4.1. Can Dsh play a similar dual role in Wnt signaling in
vertebrates?
Evidence for the involvement of Dsh in Wnt signaling has
also been demonstrated in Xenopus. Wnt signaling is
instructive for dorsal-axis formation in Xenopus development. Injection of Wnt signaling factors into the developing
ventral mesoderm causes the formation of a complete
secondary axis (although the endogenous Wnt protein is
still unknown; for review see Moon et al., 1997). It has
been shown (Sokol et al., 1995) that injecting Dsh into the
ventral mesoderm is suf®cient to induce a secondary axis
and mimic the effect of Wnt proteins (Fig. 1C), which
suggests that Dsh function is conserved between Drosophila
and Xenopus. The biochemical activity of Drosophila Dsh is
conserved, since injection of the Drosophila Dsh has the
same effect as Xenopus Dsh (Rothbacher et al., 1995).
Furthermore, different levels of injected Dsh can elicit
different cell fate speci®cation responses (Itoh and Sokol,
1997).
If Dsh has a dual function in Drosophila, how might this
be re¯ected in vertebrates? Vertebrate phenotypes that
correspond to planar polarity are still to be identi®ed,
and there is no obvious assay in which Dsh can be tested.
Rothbacher et al. 1999 have recently proposed a dual role
for Dsh in Xenopus development. In Xenopus, Dsh is
predominantly phosphorylated at the dorsal side of the
embryo. By injecting Dsh domain deletion constructs,
they have shown that its phosphorylation and membrane
localization proporties can be separated from its function in
Wnt-dependent axis induction (see Fig. 3). The `modularity' in Dsh is similar to that in Drosophila: whereas the
DIX domain is crucial for transducing a signal through
Arm/b -catenin, the DEP domain seems to be required for
membrane localization (Axelrod et al., 1998; Rothbacher et
al., 1999). Although these molecular requirements match
the `modularity' of Dsh in planar polarity signaling, further
studies are needed to understand what the DEP domaindependent membrane localization and phosphorylation
mean for its signaling functions.
34
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
4.2. Biological roles of Dishevelled in mammals
Although only one Dsh gene has been reported in Xenopus, three highly related Dishevelled (Dvl) proteins have
been isolated in mice and humans (Lijam and Sussman,
1995; Semenov and Snyder, 1997). The expression patterns
of mDvl1, mDvl2 and mDvl3 during mouse development
overlap signi®cantly, suggesting that they have redundant
functions. The mouse Dvl1 gene is highly expressed in adult
cerebellar tissues (Sussman et al., 1994) and has been
disrupted by targeted mutagenesis. Surprisingly, homozygous Dvl1 2/2 animals are fully viable and fertile (Lijam
et al., 1997). No developmental defects have been observed
in these mice, although they exhibit abnormal social behavior, such as de®cits in whisker trimming, nest-building and
huddling contacts. Mutations in the other mouse Dvl genes
have not yet been reported, and thus it remains elusive
whether they have redundant functions and what the phenotypes of the other single mutant and double mutant combinations might be. By analogy to the role of Dsh in
Drosophila, and lethal Wnt phenotypes in mice, a lethal
phenotype might be expected for mice lacking all Dvl
genes. Thus, at this time, the role of Dishevelled in mammalian development remains unclear.
the EMS cell (Rocheleau et al., 1997; Thorpe et al., 1997).
However, only mom-2 and mom-5 are involved in the orientation of the mitotic spindle of the ABar cell (Rocheleau et
al., 1997); there is no requirement for the `classical' Wg
downstream components.
The generation of cell polarity by a Fz/Dsh-dependent
mechanism seems to be conserved between nematodes
and ¯ies. During bristle development in Drosophila, sensory
organ precursor cells (SOP) undergo two rounds of asymmetric cell divisions; a similar process as described for
asymmetric cell division of the ABar cell in C. elegans,
by a Wnt/Fz dependent mechanism. Recently, Fz and Dsh
signaling was found to be required for orientation of the
spindle apparatus and asymmetric distribution of the intracellular components in SOP (Knoblich, 1997; Gho and
Schweisguth, 1998). Thus, a conserved pathway for inducing asymmetries in cells may exist in nematodes and vertebrates. These ®ndings could be an exciting avenue, linking
epithelial planar polarity signals to mechanisms whereby
cells establish (and inherit) asymmetries. Whether planar
polarity in the wing and ommatidial rotation in the eye
also rely on spindle orientation is currently unknown.
5.1. Are there other signaling pathways involved?
5. Multiple intracellular communication routes for Wnt
signals
Besides the Drosophila wg and planar polarity signaling
experiments discussed above, evidence for `alternative'
Wnt signaling pathways is also accumulating in other organisms. From early on, vertebrate Wnt genes have been categorized in different groups according to their ability to
transform mammary epithelial cells and whether they
induce secondary axes during frog development (Moon et
al., 1993; Wong et al., 1994). Recent experiments have
suggested that XWnt5a, which does not cause axis bifurcation by itself, can signal through heterotrimeric G-proteins
(Moon et al., 1997; Slusarski et al., 1997).
Several Wnt genes are expressed during chick limb development. Wnt3a, expressed at an early stage, is able, when
misexpressed, to induce expression of genes required for
apical ectodermal ridge (AER) development. Kengaku et
al. (1998) found that Wnt7a, when misexpressed, does not
affect AER formation or induce expression of AER speci®c
genes, but instead induces a different set of marker genes.
Misexpression of activated b -catenin or dominant negative
Lef1/TCF was found to mimic only Wnt3a, but not Wnt7a
effects, suggesting that the responses are mediated by different intracellular signaling pathways (Kengaku et al., 1998).
Evidence for Wnt pathways independent of Arm/b -catenin comes also from genetic observations in C. elegans.
Herman et al. (1995) have shown that mom-2 (a Wnt
gene), mom-5 (a Fz homologue), WRM-1 (Arm/b -catenin)
and pop-1 (Pan/TCF) are all required for the polarization of
Besides its role in the Wg and planar polarity pathways (Fig. 4), Dsh has been linked to Notch signaling
(Axelrod et al., 1998). Genetic analysis has shown that
dsh mutants interact antagonistically with Notch and
Delta (one of the known Notch ligands). Notch encodes
an evolutionarily conserved receptor that controls a broad
spectrum of cell fate decisions, from permissive roles in
many cell-cell interactions to instructive signaling during
lateral inhibition processes (reviewed in Artavanis-Tsakonas et al., 1995).
The relevance of the genetic interactions between Dsh
and Notch has been supported by evidence for a direct
physical interaction between these two proteins in the
yeast two hybrid assay: the N-terminal part of Dsh (including the DIX and PDZ domains) binds the most C-terminal
cytoplasmic region of Notch, which contains a PDZ domain
binding consensus sequence (Axelrod et al., 1998). These
results are particularly intriguing since several observations
suggest a link between the Notch and Wg pathways. Mutations in Notch and Wg often affect the same tissues and
cells, they have related phenotypes in loss-of-function
mutants, and screens for genetic modi®ers of Notch have
led to the isolation of wg alleles and vice versa (e.g.
reviewed in Martinez Arias, 1998). The direct interaction
of Dsh with the Notch C-terminus and the antagonistic
genetic interactions suggest that Dsh may block Notch
signaling directly, possibly interfering with binding of
another cytoplasmic protein interacting with Notch.
However, this physical interaction has yet to be con®rmed
in vivo, and other reports have suggested that Notch itself is
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
35
Fig. 4. Model of how Dishevelled discriminates between Wg and planar polarity signaling. See text for details.
involved in the reception of the Wg signal (as part of a
receptor complex) (Couso and Martinez Arias, 1994).
Thus several different interpretations of the presumed
Notch-Dsh interaction are possible.
6. Perspectives
The recent progress on the modular nature of Dsh in Wnt/
Fz signaling pathways suggest that the intracellular routes
Wnt signals take are more complex than the initial models
would suggest. Dsh appears to be a key factor that `reads'
signals coming from the membrane and route them into the
correct intracellular pathways. The discovery of the distinct
requirements of the different conserved domains of Dsh is
the ®rst step towards the understanding of the biochemical
mechanisms of Dsh function. However, the lack of knowledge of its physical interaction partners are still hampering
our understanding of how signals are routed.
Attempts to show that Frizzled family receptors bind to
Dsh by their putative C-terminal PDZ-binding consensus
have been unsuccessful, and Dsh has also not been shown
to bind to other characterized components of the wg or
planar polarity pathway. Thus interaction partners for Dsh
remain to be identi®ed. The role of Dsh phosphorylation for
either membrane localization or b -catenin/TCF and JNK
signaling also remains unclear. Does phosphorylation of
Dsh affect its subcellular localization or its capacity to
physically interact with other proteins? Genetic screens
have led to the discovery of many Wg pathway components,
similarly genetics have been successful in the search for
novel planar polarity signaling components. However,
genetic and biochemical experiments will be needed to
conclusively address these questions.
The use of Dsh in a modular way to convey different
signals cannot explain a priori how these differences originate. Whereas Drosophila Fz and Fz2 can both bind Wg in
vitro and transduce it during embryogenesis, they do not
activate both pathways in imaginal discs tissues (Cadigan
et al., 1998; Zhang and Carthew, 1998; our unpublished
results), suggesting that they can assemble qualitatively
different signaling complexes at the plasma membrane.
The recent studies implicating both Fz and Fz2 in the transduction of wg signals during embryonic development is
puzzling, since the overexpression of Fz2 in imaginal disc
does not phenocopy Fz gain-of-function defects, and vice
versa. However, the `speci®city' is lost when truncated
receptors are used; FzN and Fz2N can block both pathways.
The issue of which Fz is the receptor of speci®c Wnt ligands
might get more complicated considering the presence of two
more Fz genes in Drosophila (Fz3 and Fz4; our unpublished
results). One might speculate that Fz receptors are only part
of a receptor-transducer complex, which assembles in a
tissue-speci®c context. One could speculate about a speci®city co-receptor, as shown for RAMPs as speci®city switching factors of the CGRP receptor in mammals (McLatchie et
al., 1998). Signaling factors such as Dishevelled might
assemble `submembranous transduciosomes' whereby
signals are routed into different intracellular signaling pathways.
36
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
At this point, there are many gaps in the understanding
of Wnt signaling pathways still to be ®lled; identifying
the missing pathway components and dissecting the way
they are interconnected will be of key importance to understand how these signaling networks are regulated in vivo.
Acknowledgements
We are most grateful to Ute Rothbaecher and Scott
Fraser for the permission to cite unpublished results, to
Sergei Sokol for helpful suggestions, and Jeff Axelrod
and Sergei Sokol for pictures of ¯y and frog embryos
shown in Fig. 1. We would like to thank Jennifer Curtiss
for critical and helpful comments to improve this manuscript and Toby Gibson for help with sequence alignments.
Micheal Boutros was supported by a fellowship from the
Boehringer Ingelheim Fonds and the Studienstiftung des
deutschen Volkes.
References
Adler, P.N., 1992. The genetic control of tissue polarity in Drosophila.
BioEssays 14, 735±741.
Adler, P.N., Charlton, J., Park, W.J., 1994. The Drosophila tissue polarity
gene inturned functions prior to wing hair morphogenesis in the regulation of hair polarity and number. Genetics 137, 1±8.
Adler, P.N., Vinson, C., Park, W.J., Conover, S., Klein, L., 1990. Molecular
structure of frizzled, a Drosophila tissue polarity gene. Genetics 126,
401±416.
Artavanis-Tsakonas, S., Matsuno, K., Fortini, M., 1995. Notch signalling.
Science 268, 225±232.
Axelrod, J., Matsuno, K., Artavanis-Tsakonas, S., Perrimon, N., 1998.
Interaction between Wingless and Notch signaling pathways mediated
by Dishevelled. Science 271, 1826±1832.
Behrens, J., Jerchow, B.A., Wurtele, M., Grimm, J., Asbrand, C., Wirtz, R.,
Kuhl, M., Wedlich, D., Birchmeier, W., 1998. Functional interaction of
an axin homolog, conductin, with beta- catenin, APC, and GSK3beta.
Science 280, 596±599.
Bhanot, P., Brink, M., Samos, C.H., Hsieh, J.C., Wang, Y., Macke, J.P.,
Andrew, D., Nathans, J., Nusse, R., 1996. A new member of the frizzled
family from Drosophila functions as a Wingless receptor. Nature 382,
225±230.
Bhat, K.M., 1998. Frizzled and frizzled 2 play a partially redundant role in
wingless signaling and have similar requirements to wingless in neurogenesis. Cell 95, 1027±1036.
Boutros, M., Paricio, N., Strutt, D.I., Mlodzik, M., 1998. Dishevelled activates JNK and discriminates between JNK pathways in planar polarity
and wingless signaling. Cell 94, 109±118.
Cadigan, K.M., Fish, M.P., Rulifson, E.J., Nusse, R., 1998. Wingless
repression of Drosophila frizzled 2 expression shapes the Wingless
morphogen gradient in the wing. Cell 93, 767±777.
Cadigan, K.M., Nusse, R., 1997. Wnt signaling: a common theme in animal
development. Genes Dev. 11, 3286±3305.
Choi, K.W., Benzer, S., 1994. Rotation of photoreceptor clusters in the
developing Drosophila eye requires the nemo gene. Cell 78, 125±136.
Couso, J.P., Martinez Arias, A., 1994. Notch is required for wingless signaling in Drosophila. Cell 79, 259±272.
Eaton, S., Wepf, R., Simons, K., 1996. Roles for Rac1 and Cdc42 in planar
polarization and hair outgrowth in the wing of Drosophila. J. Cell Biol.
135, 1277±1289.
Fahmy, O.G., Fahmy, M., 1959. New mutants report. Drosophila Information Service 33, 83±94.
Fanto, M., Mlodzik, M., 1999. Asymmetric Notch activation speci®es
photoreceptors R3 and R4 and planar polarity in the Drosophila eye.
Nature, in press.
Gho, M., Schweisguth, F., 1998. Frizzled signalling controls orientation of
asymmetric sense organ precursor cell divisions in Drosophila. Nature
393, 178±181.
Gubb, D., 1993. Genes controlling cellular polarity in Drosophila. Development Suppl. 1993, 269±277.
Heberlein, U., Borod, E.R., Chanut, F., 1998. Dorsoventral patterning in the
Drosophila retina by wingless. Development 125, 567±577.
Herman, M.A., Vassilieva, L.L., Horvitz, H.R., Shaw, J.E., Herman, R.K.,
The, C., 1995. elegans gene lin-44, which controls the polarity of
certain asymmetric cell divisions, encodes a Wnt protein and acts cell
nonautonomously. Cell 83, 101±110.
Itoh, K., Sokol, S.Y., 1997. Graded amounts of Xenopus dishevelled
specify discrete anteroposterior cell fates in prospective ectoderm.
Mech. Dev. 61, 113±125.
Kengaku, M., Capdevila, J., Rodriguez-Esteban, C., De La Pena, J., Johnson, R.L., Belmonte, J.C.I., Tabin, C.J., 1998. Distinct WNT pathways
regulating AER formation and dorsoventral polarity in the chick limb
bud. Science 280, 1274±1277.
Kennerdell, J.R., Carthew, R.W., 1998. Use of dsRNA-mediated genetic
interference to demonstrate that frizzled and frizzled 2 act in the wingless pathway. Cell 95, 1017±1026.
Klingensmith, J., Nusse, R., 1994. Signaling by wingless in Drosophila.
Dev. Biol. 166, 396±414.
Klingensmith, J., Nusse, R., Perrimon, N., 1994. The Drosophila segment
polarity gene dishevelled encodes a novel protein required for response
to the wingless signal. Genes Dev. 8, 118±130.
Knoblich, J.A., 1997. Mechanisms of asymmetric cell division during
animal development. Curr. Opin. Cell Biol. 9, 833±841.
Krasnow, R.E., Wong, L.L., Adler, P.N., 1995. Dishevelled is a component
of the frizzled signaling pathway in Drosophila. Development 121,
4095±4102.
Li, L., Yuan, H., Xie, W., Mao, J., Caruso, A.M., McMahon, A., Sussman,
D.J., Wu, D., 1999. Dishevelled proteins lead to two signalling pathways. Regulations of LEF-1 and c-Jun N-terminal kinase in mammalian
cells. J. Biol. Chem. 274 (1), 129±134.
Lijam, N., Paylor, R., McDonald, M.P., Crawley, J.N., Deng, C.X., Herrup,
K., Stevens, K.E., Maccaferri, G., McBain, C.J., Sussman, D.J.,
Wynshaw-Boris, A., 1997. Social interaction and sensorimotor gating
abnormalities in mice lacking Dvl1. Cell 90, 895±905.
Lijam, N., Sussman, D.J., 1995. Organization and promoter analysis of the
mouse dishevelled-1 gene. Genome Res. 5, 116±124.
Ma, C., Moses, K., 1995. Wingless and patched are negative regulators of
the morphogenetic furrow and can affect tissue polarity in the developing Drosophila compound eye. Development 121, 2279±2289.
Martinez Arias, A., 1998. Interactions between Wingless and Notch during
the assignation of cell fates in Drosophila. Int. J. Dev. Biol. 42, 325±
333.
McLatchie, L.M., Fraser, N.J., Main, M.J., Wise, A., Brown, J., Thompson,
N., Solari, R., Lee, M.G., Foord, S.M., 1998. RAMPs regulate the
transport and ligand speci®city of the calcitonin- receptor-like receptor.
Nature 393, 333±339.
Moon, R.T., Brown, J.D., Torres, M., 1997. WNTs modulate cell fate
and behavior during vertebrate development. Trends Genet. 13, 157±
162.
Moon, R.T., Campbell, R.M., Christian, J.L., McGrew, L.L., Shih, J.,
Fraser, S., 1993. Xwnt-5A: a maternal Wnt that affects morphogenetic
movements after overexpression in embryos of Xenopus laevis. Development 119, 97±111.
Mueller, H., Samanta, R., Wieschaus, E., 1999. Wingless signaling in the
Drosophila embryo: zygotic requirements and the role of the frizzled
genes. Development 126, 577±586.
Noordermeer, J., Klingensmith, J., Perrimon, N., Nusse, R., 1994. Dishev-
M. Boutros, M. Mlodzik / Mechanisms of Development 83 (1999) 27±37
elled and armadillo act in the wingless signalling pathway in Drosophila. Nature 367, 80±83.
NuÈsslein-Volhard, C., Wieschaus, E., 1980. Mutations affecting segment
number and polarity in Drosophila. Nature 287, 795±801.
Perrimon, N., Mahowald, A.P., 1987. Multiple functions of segment polarity genes in Drosophila. Dev. Biol. 119, 587±600.
Ponting, C.P., Bork, P., 1996. Pleckstrin's repeat performance: a novel
domain in G-protein signaling?. Trends Biochem Sci 21, 245±246.
Ponting, C.P., Phillips, C., Davies, K.E., Blake, D.J., 1997. PDZ domains:
targeting signalling molecules to sub-membranous sites. Bioessays 19,
469±479.
Rocheleau, C.E., Downs, W.D., Lin, R., Wittmann, C., Bei, Y., Cha, Y.H.,
Ali, M., Priess, J.R., Mello, C.C., 1997. Wnt signaling and an APCrelated gene specify endoderm in early C. elegans embryos. Cell 90,
707±716.
Rothbacher, U., Laurent, M.N., Blitz, I.L., Watabe, T., Marsh, J.L., Cho,
K.W., 1995. Functional conservation of the Wnt signaling pathway
revealed by ectopic expression of Drosophila Dishevelled in Xenopus.
Dev. Biol. 170, 717±721.
Rothbacher, U., Laurent, M.N., Deardorff, M.A., Klein, P.., Cho, K.W.,
Fraser, S.E., 1999 A dual role for Dishevelled protein and its involvement in early dorsal-ventral patterning of Xenopus, submitted.
Sahai, E., Alberts, A.S., Treisman, R., 1998. RhoA effector mutants reveal
distinct effector pathways for cytoskeletal reorganization. SRF activation and transformation. EMBO J. 17, 1350±1361.
Semenov, M.V., Snyder, M., 1997. Human dishevelled genes constitute a
DHR-containing multigene family. Genomics 42, 302±310.
Slusarski, D.C., Corces, V.G., Moon, R.T., 1997. Interaction of Wnt and a
Frizzled homologue triggers G-protein-linked phosphatidylinositol
signalling. Nature 390, 410±413.
Sokol, S.Y., Klingensmith, J., Perrimon, N., Itoh, K., 1995. Dorsalizing and
neuralizing properties of Xdsh, a maternally expressed Xenopus homolog of dishevelled. Development 121, 3487.
Steitz, S.A., Tsang, M., Sussman, D.J., 1996. Wnt-mediated relocalization of dishevelled proteins. In Vitro Cell Dev. Biol. Anim. 32, 441±
445.
Strutt, D.I., Weber, U., Mlodzik, M., 1997. The role of RhoA in tissue
polarity and Frizzled signalling. Nature 387, 292±295.
Sussman, D.J., Klingensmith, J., Salinas, P., Adams, P.S., Nusse, R., Perrimon, N., 1994. Isolation and characterization of a mouse homolog of
37
the Drosophila segment polarity gene dishevelled. Dev. Biol. 166, 73±
86.
Theisen, H., Purcell, J., Bennett, M., Kansagara, D., Syed, A., Marsh, J.L.,
1994. dishevelled is required during wingless signalling to establish
both cell polarity and cell identity. Development 120, 347±360.
Thorpe, C.J., Schlesinger, A., Carter, J.C., Bowerman, B., 1997. Wnt
signaling polarizes an early C. elegans blastomere to distinguish endoderm from mesoderm. Cell 90, 695±705.
Tomlinson, A., Strapps, W.R., Heemskerk, J., 1997. Linking Frizzled and
Wnt signaling in Drosophila development. Development 124, 4515±
4521.
Treisman, J.E., Rubin, G.M., 1995. Wingless inhibits phogenetic furrow
movement in the Drosophila eye disc. Development 121, 3519±3527.
Van Aelst, L., D'Souza-Schorey, C., 1997. Rho GTPases and signaling
networks. Genes Dev. 11, 2295±2322.
Wehrli, M., Tomlinson, A., 1998. Independent regulation of anterior/
posterior and equatorial/polar polarity in the Drosophila eye; evidence
for the involvement of Wnt signaling in the equatorial/polar axis.
Development, 125, 1421±1432.
Willert, K., Brink, M., Wodarz, A., Varmus, H., Nusse, R., 1997. Casein
kinase 2 associates with and phosphorylates dishevelled. EMBO J. 16,
3089±3096.
Wolff, T., Rubin, G.M., 1998. Strabismus, a novel gene that regulates tissue
polarity and cell fate decisions in Drosophila. Development 125, 1149±
1159.
Wong, G.T., Gavin, B.J., McMahon, A.P., 1994. Differential transformation of mammary epithelial cells by Wnt genes. Mol. Cell. Biol. 14,
6278±6286.
Yanagawa, S., van Leeuwen, F., Wodarz, A., Klingensmith, J., Nusse, R.,
1995. The dishevelled protein is modi®ed by wingless signaling in
Drosophila. Genes Dev. 9, 1087±1097.
Zeng, L., Fagotto, F., Zhang, T., Hsu, W., Vasicek, T.J., Perry, W.L.R., Lee,
J.J., Tilghman, S.M., Gumbiner, B.M., Costantini, F., 1997. The mouse
Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway
that regulates embryonic axis formation. Cell 90, 181±192.
Zhang, J., Carthew, R.W., 1998. Interactions between Wingless and DFz2
during Drosophila wing development. Development 125, 3075±3085.
Zheng, L., Zhang, J., Carthew, R.W., 1995. frizzled regulates mirrorsymmetric pattern formation in the Drosophila eye. Development
121, 3045±3055.