Sfrp1, Sfrp2, and Sfrp5 Regulate the Wnt/b-Catenin
and the Planar Cell Polarity Pathways During Early Trunk
Formation in Mouse
Wataru Satoh,1,2Makoto Matsuyama,1Hiromasa Takemura,1,2Shinichi Aizawa,1
and Akihiko Shimono1*
1Vertebrate Body Plan, Center for Developmental Biology, RIKEN Kobe, Minatojima-Minami, Chuou-ku, Kobe, Japan
2The Graduate School of Biological Sciences, Nara Institute of Science and Technology, Ikoma, Nara, Japan
Received 23 July 2007; Revised 30 November 2007; Accepted 2 December 2007
Summary: Sfrp is a secreted Wnt antagonist that directly
interacts with Wnt ligand. We show here that inactiva-
tion of Sfrp1, Sfrp2, and Sfrp5 leads to fused somites
formation in early-somite mouse embryos, simultane-
ously resulting in defective convergent extension (CE),
which causes severe shortening of the anteroposterior
axis. These observations indicate the redundant roles of
Sfrp1, Sfrp2, and Sfrp5 in early trunk formation. The
roles of the Sfrps were genetically distinguished in
terms of the regulation of Wnt pathways. Genetic analy-
revealed the involvement of Sfrps in CE through the reg-
ulation of the planar cell polarity pathway. Furthermore,
Dkk1-deficient embryos carrying Sfrp1 homozygous and
Sfrp2 heterozygous mutations display irregular somites
and indistinct intersomitic boundaries, which indicates
that Sfrps-mediated inhibition of the Wnt/b-catenin
pathway is necessary for somitogenesis. Our results
suggest that Sfrps regulation of the canonical and non-
canonical pathways is essential for proper trunk forma-
tion. genesis 46:92–103, 2008.
C 2008 Wiley-Liss, Inc.
Key words: Sfrp;
polarity pathway; convergent extension; somitogenesis
Wnt is a family of secreted glycoproteins (Prud’homme
et al., 2002; http://www-leland.stanford.edu/ ~ rnusse/
wntwindow.html). The Wnt family activates different
intercellular signaling cascades through its receptor,
Frizzled. These cascades are classified as either the ca-
nonical Wnt/b-catenin pathway or the noncanonical
Wnt pathways, which include the Wnt/Ca21and the pla-
nar cell polarity (PCP) pathways (Widelitz, 2005).
Secreted Frizzled-related protein (Sfrp) possesses a
structure similar to the cysteine-rich domain (CRD) of
Frizzled and directly interacts with Wnt ligands (Kawano
and Kypta, 2003). This ligand interaction theoretically
suggests that Sfrp can regulate the canonical and nonca-
nonical pathways (Kawano and Kypta, 2003). The Sfrp
gene family consists of five members in the mouse and
human genomes (Jones and Jomary, 2002). Crescent and
Sizzled are unique members in Xenopus, zebrafish, and
chicken (Jones and Jomary, 2002). In particular, Sfrp1,
Sfrp2, and Sfrp5 comprise a subfamily (referred to here-
after as Sfrps), which is suggestive of the similar molecu-
lar function of Sfrps. Indeed, Sfrps inhibit the Wnt/b-cat-
enin pathway in vitro; moreover, Sfrps demonstrate dis-
tinct characteristics in comparison to Sfrp3 and Sfrp4
probably due to ligand specificity (Galli et al., 2006;
Suzuki et al., 2004). Furthermore, a double homozygous
mutation of Sfrp1 and Sfrp2 (Sfrp12/2 Sfrp22/2)
results in cell migration abnormalities in the presomitic
mesoderm (PSM) and incomplete somite segmentation
albeit only in the thoracic region (Satoh et al., 2006).
This observation suggested functional redundancy of
Sfrp1, Sfrp2, and Sfrp5; therefore, the role of Sfrps dur-
ing embryonic development remained unclear. More-
over, how Sfrps control mesoderm cell movement and
somitogenesis with respect to regulation of Wnt path-
ways was unknown.
The canonical Wnt pathway plays critical roles with
respect to the somite segmentation process. Wnt3a is
thought to determine the segmentation boundary
through its own protein gradient and/or via direct activa-
tion of Fgf8 expression (Aulehla and Herrmann, 2004).
Current address for Hiromasa Takemura: Laboratory of Animal Experi-
ments for Regeneration, Institute for Frontier Medical Sciences, Kyoto Uni-
versity, 53 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan.
Published online 6 February 2008 in
Wiley InterScience (www.interscience.wiley.com).
This article contains supplementary material available via the Internet at
*Correspondence to: Akihiko Shimono, Vertebrate Body Plan, Center for
Developmental Biology, RIKEN Kobe, Minatojima Minami, Chuou-ku, Kobe
' 2008 Wiley-Liss, Inc.genesis 46:92–103 (2008)
Furthermore, somite formation is coupled to oscillatory
Notch-related gene expression in the PSM. In this con-
text, Wnt/b-catenin signaling activates Dll1 expression
and modulates Notch signaling (Galceran et al., 2004;
Hofmann et al., 2004). The canonical Wnt pathway is
essential for primitive streak and mesoderm formation at
earlier stages (Liu et al., 1999). Moreover, mesoderm
migration is dependent on the function of T, a down-
stream target of the canonical pathway (Wilson et al.,
1993; Yamaguchi et al., 1999a). In contrast, studies in
Xenopus and zebrafish demonstrated that the PCP path-
way is required for convergent extension (CE), which is
the process consisting of simultaneous and interdepend-
ent mediolateral narrowing and anteroposterior elonga-
tion of embryonic tissue (Keller, 2002; Torban et al.,
2004). Recent studies involving mice carrying mutations
of PCP pathway components, including double homozy-
gous mutant mice of Dvl1 and Dvl2 as well as Loop-tail
(Lp) mice (characterized by the Stbm/Vangl2 mutation),
afforded evidence pertaining to regulation of CE by the
PCP pathway in higher vertebrates (Wang et al., 2006;
Ybot-Gonzalez et al., 2007).
In this study, Sfrp1, Sfrp2, and Sfrp5 triple homozy-
gous mutant embryos were generated to reveal the
redundant roles in embryonic development. Moreover,
crossing compound mutant mice with Dkk1-deficient
mice, a mouse line displaying a defective canonical Wnt
pathway, and with Lp mice, a strain exhibiting a defec-
tive noncanonical pathway, permitted genetic manipula-
tion of the balance of Sfrps regulation in the canonical
and noncanonical Wnt pathways. We provide here
genetic evidence regarding Sfrps regulation of the Wnt/
b-catenin and PCP pathways in association with somito-
genesis and CE during early trunk formation.
Generation of Sfrp1, Sfrp2, and Sfrp5 Compound
To elucidate the role of Sfrps in embryogenesis, single
Sfrp1, Sfrp2, and Sfrp5 knock-out (KO) mouse lines
were generated initially; these murine lines were viable
and fertile (Satoh et al., 2006) (Fig. S1a,b; see the Supple-
mentary Data). The expression pattern during early
embryogenesis is suggestive of the redundant role of
Sfrps. In streak stages embryos, overlapping expression
of Sfrp1 and Sfrp5 is observed in the anterior visceral
endoderm (AVE) (Finley et al., 2003; Hoang et al., 1998).
At later stages, e.g., embryonic day (E) 8.5, expression of
both Sfrp1 and Sfrp2 occurs in the neuroectoderm along
the anteroposterior (a-p) axis (Leimeister et al., 1998),
whereas simultaneous Sfrp1, Sfrp2, and Sfrp5 expres-
sion is apparent in the neural tube caudal to the hind-
brain (Finley et al., 2003) (Fig. 1a).
To identify the redundant roles, the following five ge-
notypes were produced by crosses between mutant
mice and, subsequently, compound mutant mice: (a)
Sfrp12/2 Sfrp52/2, (b) Sfrp22/2 Sfrp52/2, (c)
Sfrp11/2 Sfrp21/2 Sfrp52/2, (d) Sfrp11/2 Sfrp22/
2 Sfrp52/2, and (e) Sfrp12/2 Sfrp21/2 Sfrp52/2.
These compound mutant mouse lines (a–e) were viable
and fertile. No obvious abnormalities in gross morphol-
ogy, with the exception of hindlimb syndactyly and tail
kinks during embryogenesis in a minor population of
those compound mutant mice carrying both mutated
Sfrp2 alleles as previously reported (Cox et al., 2006;
Satoh et al., 2006), were observed. In contrast, highly
specific spatial and temporal expression patterns of
Sfrps are observed in many developing organs (Lei-
meister et al., 1998). Hence, the long-range effect of Sfrp
as a soluble factor could compensate for the functions of
other Sfrps in those mutants.
Subsequently, to revel redundant roles of Sfrp1,
Sfrp2, and Sfrp5 in embryonic development, embryos
carrying the triple homozygous mutation of Sfrp1,
Sfrp2, and Sfrp5 (Sfrp12/2 Sfrp22/2 Sfrp52/2)
were obtained by intercrosses of Sfrp12/2 Sfrp21/2
embryos were lethal at around E12.5 and characterized
by a severe a-p axis elongation defect in the trunk (data
not shown, Fig. 1b,c). In Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos at E9.5, an open neural tube was observed
consistently (Fig. 1b). The neural tissue possessed dor-
sal–ventral patterning as indicated by the expressions of
Msx1 (Robert et al., 1989) in the roof and Shh (Echelard
et al., 1993) in the floor plate of the neural tube (Fig.
1b). At these stages, embryonic growth was delayed,
probably due to abnormalities in embryonic turning and
umbilical cord formation. These observations suggest
that Sfrp1, Sfrp2, and Sfrp5 are functionally redundant
during early trunk formation.
Somitogenesis and a-p Axis Elongation Defects in
Sfrp1, Sfrp2, and Sfrp5 Triple Knock-Out Embryos
Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos at E8.5
exhibited compression of the somites along the a-p axis
and lateral expansion (Fig. 1c, between arrowheads), a
phenotype more severe than that observed in Sfrp12/2
Sfrp22/2 embryos (Fig. 1c). The Sfrp12/2 Sfrp22/2
Sfrp52/2 embryos display the fused stripes of the
Uncx4.1 expression domain, a marker occurring in the
posterior half of the somite (Mansouri et al., 1997;
Fig. 1c). In addition, histological analysis revealed fused
small somites in E8.5 Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos (Fig. 1d, arrows).
To identify a defect in the expression of cyclic gene
during somitogenesis, Lfng expression in Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos was examined at E8.25,
prior to the point where delayed embryonic growth
became apparent. Lfng, which is a direct target of Notch
signaling, encodes a glycosyltransferase that acts as a
Notch signaling inhibitor; furthermore, Lfng expression
oscillates in the PSM (Cole et al., 2002; Dale et al., 2003;
Morales et al., 2002). The expression patterns at E8.25
were typically classified into three phases, which
THE ROLES OF SFRPS IN EARLY TRUNK FORMATION
appeared with similar frequencies in control embryos
(phase I, 28%; phase II, 38%; phase III, 34%; Fig. 1e). In
the PSM of Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos at
E8.25, which corresponds to the 8–11 somites stage, the
asymmetric stripe of Lfng expression was frequently
observed in the anterior-most region of the PSM in
Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos (Fig. 1e, aster-
isks). In addition, the appearance frequency of phase I
defect. Arrow, midbrain-hindbrain boundary. Sections of the embryos (inset) generated at the broken line. Arrowhead, floor plate; arrow,
notochord; Fl, forelimb bud. (c) Compression of somites along the a-p axis and lateral spreading (between arrowheads) in Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos at E8.5. (Inset) Higher magnification of Uncx4.1 expression domains in Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos. (d) Horizontal section of control and Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos at E8.5. Arrows, fused somites. A, anterior; P, poste-
rior; Sc, spinal cord; PSM, presomitic mesoderm. (e) The oscillation cycles of Lfng are altered in the PSM of Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos at the stage corresponding to 8–11 somites (lower panels) in comparison with control embryos (upper panels). A fraction indicates
the appearance frequency of each phase.
(a) Expression of Sfrp1, Sfrp2, and Sfrp5 at E8.5. (b) Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos at E9.5 show open neural tube
SATOH ET AL.
in Lfng oscillatory cycles was significantly increased
(58%; 15 of 26 embryos), which suggests altered oscilla-
tory expression of Lfng in a manner consistent with
observations in double homozygous Sfrp12/2 Sfrp22/2
embryos (Satoh et al., 2006; Fig. 1e).
The defect in a-p axis elongation as well as the open
neural tube in Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos
appeared to be related to CE. Regionalization along the
a-p axis at E8.5–9, which was monitored by regional
markers in the brain and spinal cord, occurred in the
neural tissue (Fig. S2a–f), although the rhombomere
appeared to be affected and compressed along the a-p
axis (Fig. S2d). Furthermore, the primitive streak was
maintained in E8.5 Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos, which was indicated by markers expression
(Fig. S2g–i). Interestingly, the Sfrp12/2 Sfrp22/2
Sfrp52/2 embryos displayed ectopic colonization of
axial mesoderm characterized by Shh expression (Fig.
S2j, arrowheads), which is suggestive of abnormality in
distribution of the axial mesoderm.
The defect of axial mesoderm distribution and noto-
chord lengthening is the evidence of disruption of CE
(Wang et al., 2006). Therefore, we observed the expres-
sion of Shh in the Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos at earlier stages to determine when axial meso-
derm distribution became defective. At the late head-fold
stages, when Shh expression was distributed as a stripe
along the a-p axis in control embryos, the expression
was laterally expanded in the midline of the embryo
(Fig. 2c,d; arrow). At the late bud stage, Shh expression
was widely distributed in the axial region of Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos in comparison with that
in control embryos, even before morphological abnor-
malities in the embryos became apparent (Fig. 2a,b;
Next, the axial elongation abnormalities were directly
monitored with respect to cells derived from the ante-
rior end of the primitive streak or the node at the late
bud/early head-fold stage. To conduct this analysis, cells
were labeled with 1, 10-dioctadecyl-3, 3, 30, 30-tetrame-
thylindocarbocyanine perchlorate (DiI) (Fig. 2e,g; t 5 0);
and then the embryos were cultured for 16 h (Fig. 2f,h;
t 5 16). The labeled axial tissue including the axial mes-
oderm expanded along the a-p axis in control embryos
after the culture (Fig. 2e,f, arrowheads; n 5 3). In con-
trast, expansion was greatly reduced in Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos; additionally the labeled
cells were laterally distributed (Fig. 2g,h, arrowheads;
n 5 3). These observations suggest that Sfrp1, Sfrp2,
and Sfrp5 are functionally redundant during CE as well
as during somitogenesis.
Genetic Analysis of Sfrps Function in Canonical
and Noncanonical Wnt Pathways
The fused somites may be caused by an abnormality in
the canonical Wnt pathway during somitogenesis in
Sfrp12/2 Sfrp22/2 (Satoh et al., 2006) and Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos. However, Sfrps theoreti-
cally regulate the canonical and noncanonical pathways;
consequently, the requirement of Sfrps with respect to
Wnt pathway regulation in terms of CE and somitogene-
sis is difficult to distinguish. To differentiate Sfrp regula-
tion of the Wnt pathways, the balance of the signaling
regulation was genetically manipulated using compound
mutant mice crossed with Lp mice, a strain displaying a
defective noncanonical pathway, and with Dkk1-defi-
cient mice, a mutant mouse line exhibiting a defective
canonical Wnt pathway.
Sfrps Regulate the PCP Pathway
Lp is a mouse line carrying a mutation in Stbm/Vangl2
(also known as Ltap), which encodes a membrane pro-
tein (Torban et al., 2004). Lp/1 Sfrp12/2 Sfrp21/2
Sfrp52/2 embryos at E14.5 were obtained from crosses
between Sfrp12/2 Sfrp21/2 Sfrp52/2 female and
Lp/1 Sfrp12/2 Sfrp52/2 male mice. The Lp heterozy-
gous mutation led to a local neural tube closure defect,
spina bifida, in approximately 13% of the embryos char-
acterized by the Sfrp12/2 Sfrp52/2 background (total
n 5 16) as well as in the wild type background as
reported previously (Lu et al., 2004). The frequency of
spina bifida increased to 62% in Lp/1 Sfrp12/2
Sfrp21/2 Sfrp52/2 embryos (total n 5 13) (Fig.
S3a,b). Thus, Sfrps genetically interact with Stbm/
In addition, Lp/1 Sfrp12/2 Sfrp22/2 Sfrp51/2
embryos (at E9.5) were obtained by crossing Sfrp12/2
Sfrp21/2 females with Lp/1 Sfrp12/2 Sfrp21/2
Sfrp52/2 males. The embryos consistently displayed
of CE. (a–d) Shh expression in control (a,c) and Sfrp12/2 Sfrp22/2
Sfrp52/2 (b,d) embryos at the late bud (a,b) and late head fold
stage (c,d). (e–h), DiI-labeling assay of the node in control (e,f) and
Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos (g,h) at the early bud stage.
A, anterior; P, posterior; Hf, head fold.
Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos exhibit disruption
THE ROLES OF SFRPS IN EARLY TRUNK FORMATION
craniorachischisis (n 5 17), which was not observed in
the Sfrp12/2 Sfrp22/2 Sfrp51/2 (n 5 17) or in the
Lp/1 Sfrp12/2 Sfrp21/2 Sfrp51/2 embryos (n 5
41) (Fig. 3a). An abnormality in somitogenesis was appa-
rent in Lp/1 Sfrp12/2 Sfrp22/2 Sfrp51/2 embryos
in a region behind the forelimb, similarly found in
Sfrp12/2 Sfrp22/2 Sfrp51/2 embryos (Fig. 3d,
bracket). The Sfrp12/2 Sfrp22/2 Sfrp51/2 embryos
were morphologically indistinguishable from Sfrp12/2
Sfrp22/2 embryos at E9.5 (see Fig. 3). On the basis of
Stbm/Vangl2. Lp/1 Sfrp12/2 Sfrp22/2
Sfrp51/2 embryos at E9.5 display cra-
niorachischisis (a). T (b), Dll1 (c), and
Uncx4.1 (d) expression suggests that
defective a-p axis elongation in Lp/1
Sfrp12/2 Sfrp22/2 Sfrp51/2 embryos
was enhanced in the trunk in compari-
son tothat observed in
dorsal view). Arrow, anterior boundary of
the PSM; arrowhead, the tail bud end. Fl,
revealed wide distribution of axial meso-
derm in the midline of Lp/1 Sfrp12/2
Sfrp22/2 Sfrp51/2 embryos at the late
head fold stage (upper, frontal view;
lower, lateral view).
Sfrps genetically interact with
SATOH ET AL.
the expression of Dll1 in the PSM (Bettenhausen et al.,
1995) and the distance between the stripes of Uncx4.1
expression, a defect in a-p axis elongation in Lp/1
Sfrp12/2 Sfrp22/2 Sfrp51/2 embryos (n 5 3 for
Dll1, n 5 3 for Uncx4.1) was enhanced in the trunk rela-
tive to that observed in Sfrp12/2 Sfrp22/2 Sfrp51/2
embryos (n 5 3 for Dll1, n 5 4 for Uncx4.1) (Fig. 3c,d).
Furthermore, Lp/1 Sfrp12/2 Sfrp22/2 Sfrp51/2
embryos at the late head-fold stage (n 5 2) displayed the
abnormalities in CE, which was evidenced by the wider
distribution of the axial mesoderm expressing Shh (Fig.
3e, between arrowheads). Given that Lp/1 Sfrp12/2
Sfrp22/2 Sfrp51/2 embryos exhibited an open neural
tube and that Stbm/Vangl2 is necessary for both meso-
dermal and neural tissue morphogenesis, it is highly
possible that Sfrps are involved in CE of neural ectoderm
in later embryonic stages as well as in CE of mesoderm
during gastrulation (Ybot-Gonzalez et al., 2007). Hence,
the genetic analysis demonstrates that Sfrps regulate the
PCP pathway in association with CE and a-p axis
Inhibition of Wnt/b-Catenin Signaling by Sfrps Is
Required for Somitogenesis
Dkk1 is a secreted antagonist, which typically inhib-
its the Wnt/b-catenin pathway due to an interaction
with Wnt co-receptor Lrp5 and 6 (Niehrs, 2006). Sev-
eral lines of evidence indicate that Dkk1 inhibits the
Wnt/b-catenin pathway via interaction with the core-
ceptor Lrp6 during somitogenesis (Carter et al., 2007;
MacDonald et al., 2004; Niehrs, 2006). Dkk1 is
expressed in the posterior primitive streak at the late
head-fold stage; moreover, Dkkl oscillates in the PSM at
E8.25 to E9.25 (Dequeant et al., 2006) (Fig. S4a,b). At
later stages (e.g., E9.75), posterior Dkk1 expression is
restricted to the ventral tail bud (MacDonald et al.,
2004). The oscillatory pattern of Dkk1 was not congru-
ent with that of Axin2, the Wnt-related cyclic gene
(Aulehla et al., 2003; Dequeant et al., 2006) (Fig. S4c–
c@). In addition, the oscillation was out of phase with
Lfng in the tail region of E8.75 embryos. On the basis
of the stripes (s0 and s-1) of Lfng expression, the ante-
rior boundary of Dkk1 expression is located in the PSM
about two somites posterior to the newly generated
somite (Fig. S4d–d0,e).
Previously, we documented higher levels of activated
b-catenin in the Sfrp12/2 Sfrp22/2 tail bud (Satoh
et al., 2006). However, whether inhibition of the Wnt/b-
catenin pathway by Sfrps is actually required for somito-
genesis remained unclear. To distinguish pathway regula-
tion, compound mutant mice of Sfrps and Dkk1 were
generated. Relative to Sfrp12/2 Sfrp22/2 embryos
at E9.5, Dkk12/2 Sfrp12/2 Sfrp22/2 embryos dis-
played more severe phenotypes such as incomplete
embryonic turning and reduction of the PSM, as
evidenced by expression of Tbx6 (Chapman et al.,
1996) and Dll1 (Fig. S5e; data not shown). However,
T expression suggested the presence of the tail bud
and notochord (Wilkinson et al., 1990) (data not
shown). In contrast, the PSM in the Dkk12/2 Sfrp12/2
Sfrp22/2 embryos was maintained around E9.0 (Fig.
S5j). Sfrp12/2 Sfrp22/2 mutants demonstrate incom-
plete somite segmentation (Satoh et al., 2006; Fig. 4b,g);
as a result, a defect in somitogenesis is expected in the
embryo carrying the Dkk12/2 homozygous mutation
in a genetic background of Sfrp12/2 Sfrp22/2.
Indeed, Dkk12/2 Sfrp12/2 Sfrp22/2 embryos at
E9.5 exhibited indistinct intersomitic boundaries poste-
rior to the forelimb bud, as suggested by Pax3 expres-
sion (Fig. 4e,m). The Uncx4.1 expression domain
occurred as fused stripes in the region displaying the
indistinct intersomitic boundaries (Fig. 4j; brackets). The
expression pattern of Dkk1 differs from that of Sfrp1
and Sfrp2; however, we could not identify a clear differ-
ence between the somite defects found in Sfrp12/2
Sfrp22/2 Sfrp52/2 andDkk12/2 Sfrp12/2 Sfrp22/2
embryos. It is notable that expression of Pax1 and Myo-
genin was diminished in the posterior region of
Dkk12/2 Sfrp12/2 Sfrp22/2 embryos in contrast to
that in Sfrp12/2 Sfrp22/2 embryos (data not shown;
Satoh et al., 2006), which is suggestive of the require-
ment of canonical Wnt signaling inhibition during
somite differentiation as previously reported (Capdevila
et al., 1998).
Importantly, irregular somitic segmentation and indis-
tinct intersomitic boundaries were observed in half
(52%; total n 5 25) of Sfrp2 heterozygous mutant
embryos characterized by a Dkk12/2 Sfrp12/2 back-
ground at E9.5 (Fig. 4d,i; bracket). Incomplete pene-
trance could be attributable to influence of the genetic
background (Leaf et al., 2006). Sections of the Dkk12/2
Sfrp12/2 Sfrp21/2 embryo exhibiting Pax3 expres-
sion (Goulding et al., 1991) revealed the absence of
intersomitic boundaries in the posterior portion (Fig. 4l,
asterisk). In addition, Dkk12/2 Sfrp12/2 Sfrp21/2
embryos displayed irregular stripes of the Uncx4.1 do-
main (Fig. 4i; arrows); subsequently, the pattern of
stripes became obscure in the posterior embryonic
region (Fig. 4i, inset; arrowheads). Dkk12/2, Dkk12/2
Sfrp12/2 and Dkk11/2 Sfrp12/2 Sfrp21/2 em-
bryos did not demonstrate somite segmentation abnor-
malities at E9.5; consequently, the phenotype found in
Dkk12/2 Sfrp12/2 Sfrp21/2 is indicative of genetic
interaction of Dkk1 and Sfrp2 during somitogenesis.
To elucidate the molecular events associated with
Dkk12/2, Dkk12/2 Sfrp12/2, Dkk12/2 Sfrp12/2
embryos at E8.75, which corresponds to the 13–19
somites stage. Examination at this stage can exclude the
possibility that reduction of the PSM may result in abnor-
mal somitogenesis in Dkk12/2 Sfrp12/2 Sfrp22/2
embryos. Additionally, the expression of Wnt3a and
Fgf8 was normally observed in the tail bud of the
Dkk12/2 Sfrp12/2 Sfrp22/2 embryos (data not
shown). In Dkk12/2 embryos at E8.75, Lfng expres-
sion and the oscillatory cycles appeared to be normal
THE ROLES OF SFRPS IN EARLY TRUNK FORMATION
(Fig. 5a,b). At E8.75, the majority of Dkk12/2 Sfrp12/2
Sfrp22/2 embryos displayed an Lfng expression pat-
tern clarified as phase III (87%; 12 of 14 embryos), sug-
gesting a defect of the oscillatory expression (Fig. 5c). In
contrast, the overall expression levels of Lfng appeared
to change periodically in the PMS of Dkk12/2 Sfrp12/2
Sfrp21/2 embryos; however, the patterns were not
congruent with those expected in terms of control
embryos in half of the embryos (55%; total n 5 18; Fig.
5d). Moreover, the stripes of Lfng expression exhibited
asymmetric distribution in the Dkk12/2 Sfrp12/2
Sfrp21/2 embryos (Fig. 5d). The abnormal pattern was
detected in Dkk12/2 Sfrp12/2 embryos at a lower fre-
quency (31%; total n 5 26; data not shown), which sug-
gests the presence of milder abnormalities in the process
of somite segmentation.
The Lfng expression pattern in Dkk12/2 Sfrp12/2
Sfrp22/2 embryos was distinct from that in Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos. The appearance fre-
quency of phase I in Lfng oscillatory cycles was
increased in the PSM of Sfrp12/2 Sfrp22/2 Sfrp52/2
embryos; in contrast, phase III was found in the majority
of Dkk12/2 Sfrp12/2 Sfrp22/2 embryos (Figs. 1e
and 5c). Oscillatory expression of Dkk1 was evident in
Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos corresponding
to the 8–11 somites stage (n 5 24; data not shown).
Therefore, the difference is probably attributable to a
discrepancy in expression patterns of Sfrps, which are
strong in the middle of the trunk and weaker in the pos-
terior region including the PSM (Fig. 1a), and Dkk1,
which is found in the posterior PSM (Fig. S4). The obser-
vations regarding Lfng expression suggest that inhibition
of the canonical Wnt pathway is necessary for the main-
tenance of oscillatory changes of Notch signaling in the
PSM, because Lfng acts as a Notch signaling inhibitor
(Dale et al., 2003). Thus, genetic analysis combining
Sfrps and Dkk1 mutations indicates that Sfrps-mediated
inhibition of the Wnt/b-catenin pathway profoundly
impacts somite segmentation.
Sfrps Regulate the PCP Pathway
The inactivation of Sfrp1, Sfrp2, and Sfrp5 results in
defective CE from the late bud stage. With correlation to
the timing of the defect in Sfrp12/2 Sfrp22/2
Sfrp52/2 embryos, Sfrp1 is expressed in the anterior
mesendoderm (Satoh et al., 2006), whereas Sfrp2
expression is observed in the embryonic ectoderm at
was examined in control (a,f; Dkk11/? Sfrp12/2 Sfrp21/?), Dkk11/? Sfrp12/2 Sfrp22/2 (b,g), Dkk12/2 Sfrp12/2 (c,h), Dkk12/2
Sfrp12/2 Sfrp21/2 (d,i), and Dkk12/2 Sfrp12/2 Sfrp22/2 (e,j) embryos at E9.5. The ‘‘?’’ indicates 1 or 2. Insets show higher magnifica-
tion of dorsal view. (k–m) Sections of Dkk12/2 Sfrp12/2 Sfrp21/2 (l) and Dkk12/2 Sfrp12/2 Sfrp22/2 (m) embryos display indistinct
intersomitic boundaries (*). Fl, forelimb; A, anterior; P, posterior; Sc, spinal cord. The bracket indicates a unit of the somite.
Wnt/b-catenin signaling inhibition is required for proper somitogenesis. (a–j) Expression of Pax3 (a–e) and Uncx4.1 (f–j) in somites
SATOH ET AL.
the late bud to the early head fold stages (Mukhopadhyay
et al., 2003); furthermore, Sfrp5 expression is detected
in the foregut endoderm underlying the cardiac meso-
derm at around early head-fold stage (Finley et al.,
In addition, our data indicate the genetic interaction
between Sfrps and Stbm/Vangl2. Stbm/Vangl2 is a
mouse orthologue of the core PCP component Strabis-
mus/Van Gogh in Drosophila (Torban et al., 2004). In
Drosophila, PCP signaling leads to asymmetric distribu-
tion of its own molecular component, i.e., Frizzled (Fz)
and Strabismus (Stbm); the subcellular distribution is
required for the PCP pathway (Seifert and Mlodzik,
2007). Fz interacts with a cytoplasmic protein, Dishev-
elled (Dvl); moreover, Dvl membrane localization
reflects activation of the pathway (Axelrod, 2001). Stbm
physically interacts with Prickled (Pk), a cytoplasmic
protein (Jenny et al., 2003). The Stbm-Pk complex
reduces membrane localization of Dvl, which is indica-
tive of the inhibition of Fz-Dvl signaling (Tree et al.,
2002). Therefore, evidence regarding genetic interaction
of Sfrps and Stbm/Vanfl2 may imply that Sfrps inhibit
Wnt5a and Wnt11 modulate the PCP pathway in ver-
tebrates (Carreira-Barbosa et al., 2003; Heisenberg et al.,
2000; Qian et al., 2007). Recent studies revealed that the
Wnt5a signaling pathway is distinct from Wnt11, and
that Wnt5a activates JNK through Ror2 receptor (Oishi
et al., 2003; Schambony and Wedlich, 2007). Wnt5a is
expressed in the primitive streak during mouse gastrula-
tion (Yamaguchi et al., 1999b). Moreover, Wnt5a homo-
zygous mutant mouse embryos as well as Ror2 homozy-
gous mutants display a-p axis elongation defects (Oishi
et al., 2003; Yamaguchi et al., 1999b). The phenotype in
terms of the a-p body axis is similar to that observed in
Sfrp12/2 Sfrp22/2 embryos (Satoh et al., 2006).
Therefore, it is likely that Sfrps inhibit Wnt5a signaling;
furthermore, this inhibition may play a role in regulation
of the PCP pathway.
Abnormality in the regulation of the PCP pathway
might affect somite segmentation as Stbm expression
and the Wnt5a/Ror2 pathway repress the canonical
Wnt/b-catenin pathway in cell culture (Mikels and
Nusse, 2006; Park and Moon, 2002). Distortion of CE
was enhanced at the earlier stages in Lp/1 Sfrp12/2
Sfrp22/2 Sfrp51/2 embryos; however, the region in
which the defect was observed during somitogenesis
was not altered in comparison to Sfrp12/2 Sfrp22/2
Sfrp51/2 embryos. The mutation in the PCP pathway
component does not directly affect somite segmentation
(Henry et al., 2000). Furthermore, no apparent abnor-
mality in somite segmentation has been reported in
Wnt5a mutants of mouse and Zebrafish (Carreira-Bar-
bosa et al., 2003; Yamaguchi et al., 1999b).
Sfrps Inhibit Wnt/b-Catenin Signaling
Somites are periodically generated from the PSM. Peri-
odic somite formation is controlled by the molecular os-
cillator, i.e., the segmentation clock, which controls the
expression of cyclic genes in the PSM. The cyclic genes
have been identified as Notch-related genes, such as
Lfng, c-hairy1, Hes1, and Hes7, as well as Wnt-related
genes, such as Axin2, Nkx2, and Dkk1 (Aulehla and
Herrmann, 2004). A negative feedback loop in Notch sig-
nalling established by Lfng, which underlies the segmen-
tation clock in the posterior PSM, has been proposed
(Dale et al., 2003).
In Sfrp12/2 Sfrp22/2,Sfrp12/2 Sfrp22/2 Sfrp52/2,
Sfrp12/2 Sfrp22/2 embryos, fused somites were
observed. In addition, the cyclic expression of Lfng was
altered in Sfrp12/2 Sfrp22/2 and Sfrp12/2 Sfrp22/2
Sfrp52/2 embryos. Furthermore, the cyclic expression
was defective in Dkk12/2 Sfrp12/2 Sfrp22/2
embryos. Lfng acts as a Notch signaling inhibitor (Dale
embryos at the 13–19 somites stage. (b) Lfng expression oscillates
in the PSM of Dkk12/2 embryos corresponding to the 13–19
somites stage. (c) The oscillation cycles are defective in the PSM of
Dkk12/2 Sfrp12/2 Sfrp22/2 embryos at stages similar to those
shown in (a) and (b). Embryonic stages were identified on the basis
of somite numbers in the majority of the other littermates. (d) Lfng
expression patterns are aberrant in Dkk12/2 Sfrp12/2 Sfrp21/2
embryos corresponding to the 13–19 somites stage. While arrows
denote unusual Lfng expression pattern in the PSM. A fraction indi-
cates the appearance frequency of each phase.
(a) Lfng expression oscillates in the PSM of control
THE ROLES OF SFRPS IN EARLY TRUNK FORMATION
et al., 2003); thus, these observations suggest that Wnt/
b-catenin signaling inhibition in the PSM is required for
Notch signaling oscillation. A defect of Notch signaling
oscillation in the PSM could result in fused somites for-
mation. Additionally, segmentation of the somite at the
interface of the positive and negative domains of Notch
signalling in the anterior PSM, which is established
by Lfng following induction by Mesp2 transcription
factor, has been demonstrated (Morimoto et al., 2005).
Dkk12/2 Sfrp12/2 Sfrp21/2 embryos display an
abnormal Lfng expression pattern in anterior PSM. This
finding may be suggestive of the involvement of Wnt/b-
catenin signaling inhibition by Sfrps and Dkk1 in seg-
mentation of the somite in a manner corresponding to
the determination of a-p polarity within somites.
Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos at E8.5 ex-
hibit strong defects in CE. The possibility exist that ele-
vated Wnt/b-catenin signaling interferes with CE; how-
ever, genetic analysis of Lp mice highlights the role of
Sfrps in the regulation of the PCP pathway. As previously
reported, activation of Wnt/b-catenin induced by the
inactivation of Tcf-3 negative transcriptional repressor
leads to an expansion of the primitive streak and partial
duplication of the axial structure in early mouse
embryos (Merrill et al., 2004); additionally, similar find-
ings are observed in mutant embryos with respect to the
negative regulator Axin (Zeng et al., 1997). The pheno-
type found in these mouse mutant embryos indicates
that excess Wnt/b-catenin activity affects primitive streak
Sfrp12/2 Sfrp22/2 embryos at E9.5 displayed severe
reduction in PSM cell numbers. Although we did not
address this issue in greater detail, the phenotype in the
PSM of Dkk12/2 Sfrp12/2 Sfrp22/2 embryos at E9.5
was distinct from that associated with shortening of the
PSM region in Sfrp12/2 Sfrp22/2 and Sfrp12/2
Sfrp22/2 Sfrp52/2 embryos as well as in Lp/1
Sfrp12/2 Sfrp22/2 Sfrp51/2 embryos. Dkk1 typi-
cally inhibits the Wnt/b-catenin pathway by inhibiting
Wnt coreceptor Lrp5 and 6 (Niehrs, 2006). Thus, inhibi-
tion of Wnt/b-catenin activity in the posterior PSM might
be necessary for maintenance of PSM cells.
The Role of Sfrps in Early Trunk Formation
During mouse gastrulation, the precursors of the para-
xial mesoderm start to derive from the primitive streak
at around the mid-streak stage (Kinder et al., 1999);
moreover, the timing is nearly identical to that of axial
mesoderm generation and ingression (Kinder et al.,
2001). Subsequently, the paraxial mesoderm is allocated
in the proper position anterior to the node from the late
streak stage to the head-fold stage (Yamamoto et al.,
2000). Following the commencement of somitogenesis,
the paraxial mesoderm is continuously supplied from
the primitive streak and integrated into the PSM. Soon af-
ter the generation of the axial and paraxial mesoderm,
mouse PCP mutant embryos display disruption of CE
(Wang et al., 2006). This observation suggests that CE
follows the generation of axial and paraxial mesoderm.
This study demonstrated Sfrps inhibition of the Wnt/
b-catenin signaling pathway during somitogenesis (see
Fig. 6). A descending gradient of Sfrps from anterior to
posterior in E7.5–E9 embryos, in conjunction with
Dkk1, could control the optimal activity of canonical
Wnt signaling in the PSM (Finley et al., 2003; Hoang
et al., 1998), which may be required for maintenance of
oscillatory changes of Notch signaling in the PSM. In
addition Sfrps regulate the PCP pathway (see Fig. 6).
This regulation governs CE, which leads to the elonga-
tion of the a-p axis. In Sfrp12/2 Sfrp22/2 embryos
(Satoh et al., 2006) as well as in Sfrp12/2 Sfrp22/2
Sfrp52/2 embryos, the defect in a-p axis elongation is
strongly associated with somite segmentation abnormal-
ities. Our observations suggest that the timed regulation
of the Wnt/b-catenin and PCP pathways is mediated by
Sfrp1, Sfrp2, and Sfrp5 during early trunk formation.
Interestingly, the point at which the defects become
apparent in Sfrp12/2 Sfrp22/2 Sfrp52/2 embryos is
much earlier than that in Sfrp12/2 Sfrp22/2 embryos,
which suggests that Sfrps cooperatively regulate the
canonical and noncanonical Wnt pathways utilizing
those expression patterns.
129 strain BAC clones of Sfrp5 gene were obtained
from BACPAC resources. To generate Sfrp5 KO vector, a
13-kbp Asp718 fragment containing the Sfrp5 gene was
subcloned into a plasmid. A 505.2-kbp HindIII-SmaI frag-
ment and a 303.9-kbp SpeI-EcoRI fragment were ligated
pathways conducted by Sfrp1, Sfrp2, and Sfrp5 in early trunk for-
mation. PS, primitive streak. See the text for details.
The regulation of the canonical and noncanonical Wnt
SATOH ET AL.
with a cassette containing PGKneobpAloxA (Satoh
et al., 2006), resulting in deletion of the 50coding region
in the first exon encoding the CRD. The cassette also
contained IRES (internal ribosomal entry sequence) and
a mutated yellow fluorescent protein sequence (YFP*). A
frame shift mutation was introduced by ligation with
IRES, which led to the generation of multiple stop co-
dons. An SV40 polyA sequence (SV) served to terminate
transcription (Fig. S1a). A MC1DtpA-negative selection
marker cassette was added to the 50homologous arm to
enrich for homologous recombinants (Yagi et al., 1993).
Electrophoration of the vectors into embryonic stem
(ES) cells and identification of homologous recombina-
tion events were performed with the similar strategy of
Shimono and Behringer (2003). Sfrps compound mutant
mice were maintained in a 129 and C57BL/6 mixed
Sfrp22/2 Sfrp52/2 embryos were determined based
on head morphology. Lp mice of the LPT/LeJ stock were
acquired from the Jackson Laboratory and genotypes
were as described (Lu et al., 2004). Dkk1 KO mice (Fur-
ushima et al., 2007) (Acc. No. CDB0030K) were utilized
for the production of Dkk1 and Sfrps compound mutant
mice. Embryonic stages of Dkk1 and Sfrps compound
mutant were determined from somite numbers. Embry-
onic staging from E7.0 to E8.0 was done according to
Downs and Davis (1993).
In Situ hybridization
Whole mount in situ hybridization was performed
according to the description of Wilkinson (1992). An
Sfrp1 cDNA fragment was obtained from cDNA subtrac-
tion screening (Shimono and Behringer, 1999). Dkk1
cDNA, which isolated from an E8.5 cDNA library (gift of
Dr. H. Hamada, Osaka University), was used in order to
detect Dkk1 expression. Sfrp2 and Sfrp5 cDNA clones
were obtained as I.M.A.G.E. clones (Invitrogen); the
Axin2 probe was generated from a FANTOM cDNA
clone (FANTOM Consortium, 2005).
DiI Injection and Whole Embryo Culture
Whole-embryo culture and DiI labeling were per-
formed according to the method of Shimono and Beh-
ringer (2003). DiI was injected at the primitive node
region at non-allantoic/early bud stages.
We express our appreciation to Dr. A. Bradley for provid-
ing the AB1 ES and NSL cell lines and to Dr. R. Behringer
for the PGKneobpAloxA cassettes. We wish to thank
Drs. R. Johnson, P. Gruss, C. Wright, E. Olson, A. McMa-
hon, A. Gossler, B. Herrmann, H. Kokubo and Y. Saga for
probes. Dr. K. Nakao, Laboratory for Animal Resources
and Genetic Engineering at CDB, kindly assisted us in
chimera production. We also thank Drs. H. Sasaki and H.
Kokubo for helpful comments regarding the manuscript.
Aulehla A, Herrmann BG. 2004. Segmentation in vertebrates: Clock and
gradient finally joined. Genes Dev 18:2060–2067.
Aulehla A, Wehrle C, Brand-Saberi B, Kemler R, Gossler A, Kanzler B,
Herrmann BG. 2003. Wnt3a plays a major role in the segmentation
clock controlling somitogenesis. Dev Cell 4:395–406.
Axelrod JD. 2001. Unipolar membrane association of Dishevelled medi-
ates Frizzled planar cell polarity signaling. Genes Dev 15:1182–
Bettenhausen B, Hrabe de Angelis M, Simon D, Guenet JL, Gossler A.
1995. Transient and restricted expression during mouse embryo-
genesis of Dll1, a murine gene closely related to Drosophila Delta.
Capdevila J, Tabin C, Johnson RL. 1998. Control of dorsoventral somite
patterning by Wnt-1 and beta-catenin. Dev Biol 193:182–194.
Carreira-Barbosa F, Concha ML, Takeuchi M, Ueno N, Wilson SW, Tada
M. 2003. Prickle 1 regulates cell movements during gastrulation
and neuronal migration in zebrafish. Development 130:4037–
Carter M, Chen X, Slowinska B, Minnerath S, Glickstein S, Shi L, Cam-
pagne F, Weinstein H, Ross ME. 2005. Crooked tail (Cd) model of
human folate-responsive neural tube defects is mutated in Wnt
coreceptor lipoprotein receptor-related protein 6. Proc Natl Acad
Sci USA 102:12843–12848.
Chapman DL, Agulnik I, Hancock S, Silver LM, Papaioannou VE. 1996.
Tbx6, a mouse T-Box gene implicated in paraxial mesoderm forma-
tion at gastrulation. Dev Biol 180:534–542.
Cole SE, Levorse JM, Tilghman SM, Vogt TF. 2002. Clock regulatory ele-
ments control cyclic expression of Lunatic fringe during somito-
genesis. Dev Cell 3:75–84.
Cox S, Smith L, Bogani D, Cheeseman M, Siggers P, Greenfield A. 2006.
Sexually dimorphic expression of secreted frizzled-related (SFRP)
genes in the developing mouse Mullerian duct. Mol Reprod Dev
Dale JK, Maroto M, Dequeant ML, Malapert P, McGrew M, Pourquie O.
2003. Periodic notch inhibition by lunatic fringe underlies the
chick segmentation clock. Nature 421:275–278.
Dequeant ML, Glynn E, Gaudenz K, Wahl M, Chen J, Mushegian A,
Pourquie O. 2006. A complex oscillating network of signaling
genes underlies the mouse segmentation clock. Science 314:
Downs MK, Davies T. 1993. Staging of gastrulating embryos by mor-
phological landmarls in the dissecting microscope. Development
Echelard Y, Epstein DJ, St-Jacques B, Shen L, Mohler J, McMahon JA,
McMahon AP. 1993. Sonic hedgehog, a member of a family of puta-
tive signaling molecules, is implicated in the regulation of CNS po-
larity. Cell 75:1417–1430.
Katayama S, Tomaru Y, Kasukawa T, Waki K, Nakanishi M, Nakamura
M, Nishida H, Yap CC, Suzuki M, Kawai J, Suzuki H, Carninci P,
Hayashizaki Y, Wells C, Frith M, Ravasi T, Pang KC, Hallinan J,
Mattick J, Hume DA, Lipovich L, Batalov S, Engstro ¨m PG, Mizuno
Y, Faghihi MA, Sandelin A, Chalk AM, Mottagui-Tabar S, Liang Z,
Lenhard B, Wahlestedt C; RIKEN Genome Exploration Research
Group; Genome Science Group (Genome Network Project Core
Group); FANTOM Consortium. 2005. Antisense transcription in
the mammalian transcriptome. Science 309:1564–1566.
Finley KR, Tennessen J, Shawlot W. 2003. The mouse secreted frizzled-
related protein 5 gene is expressed in the anterior visceral endo-
derm and foregut endoderm during early post-implantation devel-
opment. Gene Expr Patterns 3:681–684.
Furushima K, Yamamoto A, Naganom T, Shibata M, Miyachi H, Abe T,
Ohshima N, Kiyonari H, Aizawa S. 2007. Mouse homologues of
Shisa antagonistic to Wnt and Fgf signalings. Dev Biol 306:480–
Galceran J, Sustmann C, Hsu SC, Folberth S, Grosschedl R. 2004. LEF1-
mediated regulation of Delta-like1 links Wnt and Notch signaling
in somitogenesis. Genes Dev 18:2718–2723.
Galli LM, Barnes T, Cheng T, Acosta L, Anglade A, Willert K, Nusse R,
Burrus LW. 2006. Differential inhibition of Wnt-3a by Sfrp-1. Sfrp-2,
and Sfrp-3. Dev Dyn 235:681–690.
THE ROLES OF SFRPS IN EARLY TRUNK FORMATION
Goulding MD, Chalepakis G, Deutsch U, Erselius JR, Gruss P. 1991. Pax-
3, a novel murine DNA binding protein expressed during early
neurogenesis. EMBO J 10:1135–1147.
Henry CA, Hall LA, Burr Hille M, Solnica-Krezel L, Cooper MS. 2000.
Somites in zebrafish doubly mutant for knypek and trilobite form
without internal mesenchymal cells or compaction. Curr Biol
Heisenberg CP, Tada M, Rauch GJ, Saude L, Concha ML, Geisler R, Stem-
ple DL, Smith JC, Wilson SW. 2000. Silberblick/Wnt11 mediates
convergent extension movements during zebrafish gastrulation.
Hoang BH, Thomas JT, Abdul-Karim FW, Correia KM, Conlon RA,
Luyten FP, Ballock RT. 1998. Expression pattern of two Frizzled-
related genes. Frzb-1 and Sfrp-1, during mouse embryogenesis sug-
gests a role for modulating action of Wnt family members. Dev
Hofmann M, Schuster-Gossler K, Watabe-Rudolph M, Aulehla A, Herr-
mann BG, Gossler A. 2004. WNT signaling, in synergy with T/
TBX6, controls Notch signaling by regulating Dll1 expression in
the presomitic mesoderm of mouse embryos. Genes Dev
Jenny A, Darken RS, Wilson PA, Mlodzik M. 2003. Prickle and Strabis-
mus form a functional complex to generate a correct axis during
planar cell polarity signaling. EMBO J 22:4409–4420.
Jones SE, Jomary C. 2002. Secreted Frizzled-related proteins: Searching
for relationships and patterns. Bioessays 24:811–820.
Kawano Y, Kypta R. 2003. Secreted antagonists of the Wnt signaling
pathway. J Cell Sci 116:2627–2634.
Keller R. 2002. Shaping the vertebrate body plan by polarized embry-
onic cell movements. Science 298:1950–1954.
Kinder SJ, Tsang TE, Quinlan GA, Hadjantonakis AK, Nagy A, Tam PP.
1999. The orderly allocation of mesodermal cells to the
extraembryonic structures and the anteroposterior axis during
gastrulation of the mouse embryo. Development 126:4691–
Kinder SJ, Tsang TE, Wakamiya M, Sasaki H, Behringer RR, Nagy A, Tam
PP. 2001. The organizer of the mouse gastrula is composed of a
dynamic population of progenitor cells for the axial mesoderm.
Leaf I, Tennessen J, Mukhopadhyay M, Westphal H, Shawlot W. 2006.
Sfrp5 is not essential for axis formation in the mouse. Genesis
Leimeister C, Bach A, Gessler M. 1998. Developmental expression pat-
terns of mouse sFRP genes encoding members of the secreted
frizzled related protein family. Mech Dev 75:29–42.
Liu P, Wakamiya M, Shea MJ, Albrecht U, Behringer RR, Bradley A.
1999. Requirement for Wnt3 in vertebrate axis formation. Nat
Lu X, Borchers AG, Jolicoeur C, Rayburn H, Baker JC, Tessier-Lavigne
M. 2004. PTK7/CCK-4 is a novel regulator of planar cell polarity in
vertebrates. Nature 430:93–98.
MacDonald BT, Adamska M, Meisler MH. 2004. Hypomorphic expres-
sion of Dkk1 in the doubleridge mouse: Dose dependence and
compensatory interactions with Lrp6. Development 131:2543–
Mansouri A, Yokota Y, Wehr R, Copeland NG, Jenkins NA, Gruss P.
1997. Paired-related murine homeobox gene expressed in the
developing sclerotome, kidney, and nervous system. Dev Dyn 210:
Merrill BJ, Pasolli HA, Polak L, Rendl M, Garcia-Garcia MJ, Anderson KV,
Fuchs E. 2004. Tcf3: A transcriptional regulator of axis induction
in the early embryo. Development 131:263–274.
Mikels AJ, Nusse R. 2006. Purified Wnt5a protein activates or inhibits
b-catenin-TCF signaling depending on receptor context. PLoS Biol
Morales AV, Yasuda Y, Ish-Horowicz D. 2002. Periodic lunatic fringe
expression is controlled during segmentation by a cyclic tran-
scriptional enhancer responsive to notch signaling. Dev Cell
Morimoto M, Takahashi Y, Endo M, Saga Y. 2005. The Mesp2 transcrip-
tion factor establishes segmental borders by suppressing Notch ac-
tivity. Nature 435:354–459.
Mukhopadhyay M, Shtrom S, Rodriguez-Esteban C, Chen L, Tsukui T,
Gomer L, Dorward DW, Glinka A, Grinberg A, Huang SP, Niehrs C,
Belmonte JC, Westphal H. 2001. Dickkopf1 is required for embry-
onic head induction and limb morphogenesis in the mouse. Dev
Mukhopadhyay M, Teufel A, Yamashita T, Agulnick AD, Chen L, Downs
KM, Schindler A, Grinberg A, Huang SP, Dorward D, Westphal H.
2003. Functional ablation of the mouse Ldb1 gene results in
severe patterning defects during gastrulation. Development
Niehrs C. 2006. Function and biological roles of the Dickkopf family of
Wnt modulators. Oncogene 25:7469–7481.
Oishi I, Suzuki H, Onishi N, Takada R, Kani S, Ohkawara B, Koshida I,
Suzuki K, Yamada G, Schwabe GC, Mundlos S, Shibuya H, Takada
S, Minami Y. 2003. The receptor tyrosine kinase Ror2 is involved
in non-canonical Wnt5a/JNK signalling pathway. Genes Cells
Park M, Moon TR. 2002. The planar cell-polarity gene stbm regulates
cell behaviour and cell fate in vertebrate embryos. Nat Cell Biol
Prud’homme B, Lartillot N, Balavoine G, Adoutte AVervoort M. 2002.
Phylogenetic analysis of the Wnt gene family. Insights from lopho-
trochozoan members. Curr Biol 12:1395–1400.
Qian D, Jones C, Rzadzinska A, Mark S, Zhang X, Steel KP, Dai X, Chen
P. 2007. Wnt5a functions in planar cell polarity regulation in mice.
Dev Biol 306:121–33.
Robert B, Sassoon D, Jacq B, Gehring W, Buckingham M. 1989. Hox-7, a
mouse homeobox gene with a novel pattern of expression during
embryogenesis. EMBO J 8:91–100.
Satoh W, Gotoh T, Tsunematsu Y, Aizawa S, Shimono A. 2006. Sfrp1
and Sfrp2 regulate anteroposterior axis elongation and somite seg-
mentation during mouse embryogenesis. Development 133:989–
Schambony A, Wedlich D. 2007. Wnt-5A/Ror2 regulate expression of
XPAPC through an alternative noncanonical signaling pathway.
Dev Cell 12:779–792.
Seifert JR, Mlodzik M. 2007. Frizzled/PCP signalling: A conserved mech-
anism regulating cell polarity and directed motility. Nat Rev Genet
Shimono A, Behringer RR. 1999. Isolation of novel cDNAs by subtrac-
tions between the anterior mesendoderm of single mouse gastrula
stage embryos. Dev Biol 209:369–380.
Shimono A, Behringer RR. 2003. Angiomotin regulates visceral endo-
derm movements during mouse embryogenesis. Curr Biol 13:613–
Suzuki H, Watkins DN, Jair KW, Schuebel KE, Markowitz SD, Chen WD,
Pretlow TP, Yang B, Akiyama Y, Van Engeland M, Toyota M, Tokino
T, Hinoda Y, Imai K, Herman JG, Baylin SB. 2004. Epigenetic inacti-
vation of SFRP genes allows constitutive WNT signaling in colorec-
tal cancer. Nat Genet 36:417–422.
Torban E, Kor C, Gros P. 2004. Van Gogh-like2 (Strabismus) and its role
in planar cell polarity and convergent extension in vertebrates.
Trends Genet 20:570–577.
Tree DR, Shulman JM, Rousset R, Scott MP, Gubb D, Axelrod JD. 2002.
Prickle mediates feedback amplification to generate asymmetric
planar cell polarity signaling. Cell 109:371–381.
Wang J, Hamblet NS, Mark S, Dickinson ME, Brinkman BC, Segil N,
Fraser SE, Chen P, Wallingford JB, Wynshaw-Boris A. 2006. Dishev-
elled genes mediate a conserved mammalian PCP pathway to regu-
late convergent extension during neurulation. Development 133:
Widelitz R. 2005. Wnt signaling through canonical and non-canonical
pathways: recent progress. Growth Factors 23:111–116.
Wilkinson DG. 1992. Whole mount in situ hybridization of vertebrate
embryos. In: Wilkinson DG, editor. In situ hybridization. Oxford:
IRL. pp 75–83.
Wilkinson DG, Bhatt S, Herrmann BG. 1990. Expression pattern of the
mouse T gene and its role in mesoderm formation. Nature
SATOH ET AL.
Wilson V, Rashbass P, Beddington RS. 1993. Chimeric analysis of T (Bra- Download full-text
chyury) gene function. Development 117:1321–131.
Yagi T, Nada S, Watanabe N, Tamemoto H, Kohmura N, Ikawa Y, Aizawa
S. 1993. A novel negative selection for homologous recombinants
using diphtheria toxin A fragment gene. Anal Biochem 214:77–86.
Yamaguchi TP, Takada S, Yoshikawa Y, Wu N, McMahon AP. 1999a. T
(Brachyury) is a direct target of Wnt3a during paraxial mesoderm
specification. Genes Dev 13:3185–3190.
Yamaguchi TP, Bradley A, McMahon AP, Jones S. 1999b. A Wnt5a path-
way underlies outgrowth of multiple structures in the vertebrate
embryo. Development 126:1211–1223.
Yamamoto A, Kemp C, Bachiller D, Geissert D, De Robertis EM. 2000.
Mouse paraxial protocadherin is expressed in trunk mesoderm
and is not essential for mouse development. Genesis 27:49–57.
Ybot-Gonzalez P, Savery D, Gerrelli D, Signore M, Mitchell CE, Faux CH,
Greene ND, Copp AJ. 2007. Convergent extension, planar-cell-po-
larity signalling and initiation of mouse neural tube closure. Devel-
Zeng L, Fagotto F, Zhang T, Hsu W, Vasicek TJ, Perry WL III, Lee JJ,
Tilghman SM, Gumbiner BM, 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.
THE ROLES OF SFRPS IN EARLY TRUNK FORMATION