Visualization and exploration of Tcf/Lef function using a highly responsive Wnt/Β-catenin signaling-reporter transgenic zebrafish

Article (PDF Available)inDevelopmental Biology 370(1):71-85 · July 2012with42 Reads
DOI: 10.1016/j.ydbio.2012.07.016 · Source: PubMed
Abstract
Evolutionarily conserved Tcf/Lef transcription factors (Lef1, Tcf7, Tcf7l1, and Tcf7l2) mediate gene expression regulated by Wnt/β-catenin signaling, which has multiple roles in early embryogenesis, organogenesis, adult tissue homeostasis, and tissue regeneration. However, the spatiotemporal dynamics of Tcf/Lef activity during these events remain poorly understood. We generated stable transgenic zebrafish lines carrying a new Wnt/β-catenin signaling reporter, Tcf/Lef-miniP:dGFP. The reporter revealed the transcriptional activities of four Tcf/Lef members controlled by Wnt/β-catenin signaling, which were expressed in known Wnt/β-catenin signaling-active sites during embryogenesis, organ development and growth, and tissue regeneration. We used the transgenic lines to demonstrate the contribution of Tcf/Lef-mediated Wnt/β-catenin signaling to the development of the anterior lateral line, dorsal and secondary posterior lateral lines, and gill filaments. Thus, these reporter lines are highly useful tools for studying Tcf/Lef-mediated Wnt/β-catenin signaling-dependent processes.

Figures

From the Society for Developmental Biology
Visualization and exploration of Tcf/Lef function using a highly responsive
Wnt/
b
-catenin signaling-reporter transgenic zebrafish
Nobuyuki Shimizu
a
, Koichi Kawakami
b
, Tohru Ishitani
a,
n
a
Division of Cell Regulation Systems, Department of Immunobiology and Neuroscience, Medical Institute of Bioregulation, Kyushu University, 3-1-1 Maidashi,
Higashi-ku, Fukuoka 812-8582, Japan
b
Division of Molecular and Developmental Biology, National Institute of Genetics, 1111 Yata, Mishima, Shizuoka 411-8540, Japan
article info
Article history:
Received 2 February 2012
Received in revised form
3 July 2012
Accepted 10 July 2012
Available online 25 July 2012
Keywords:
Tcf/Lef
Wnt/
b
-catenin signaling
Zebrafish
abstract
Evolutionarily conserved Tcf/Lef transcription factors (Lef1, Tcf7, Tcf7l1, and Tcf7l2) mediate gene
expression regulated by Wnt/
b
-catenin signaling, which has multiple roles in early embryogenesis,
organogenesis, adult tissue homeostasis, and tissue regeneration. However, the spatiotemporal
dynamics of Tcf/Lef activity during these events remain poorly understood. We generated stable
transgenic zebrafish lines carrying a new Wnt/
b
-catenin signaling reporter, Tcf/Lef-miniP:dGFP. The
reporter revealed the transcriptional activities of four Tcf/Lef members controlled by Wnt/
b
-catenin
signaling, which were expressed in known Wnt/
b
-catenin signaling-active sites during embryogenesis,
organ development and growth, and tissue regeneration. We used the transgenic lines to demonstrate
the contribution of Tcf/Lef-mediated Wnt/
b
-catenin signaling to the development of the anterior lateral
line, dorsal and secondary posterior lateral lines, and gill filaments. Thus, these reporter lines are highly
useful tools for studying Tcf/Lef-mediated Wnt/
b
-catenin signaling-dependent processes.
& 2012 Elsevier Inc. All rights reserved.
Introduction
Wnt/
b
-catenin signaling controls cell proliferation and fate
decisions during embryogenesis and adult tissue homeostasis
(Logan and Nusse, 2004; Clevers, 2006). At the transcriptional
level, Wnt/
b
-catenin signaling is primarily mediated by the
Tcf/Lef family of transcription factors (Lef1, Tcf1/Tcf7, Tcf3/Tcf7l1,
and Tcf4/Tcf7l2) (Logan and Nusse, 2004; Clevers, 2006). Several
transgenic animals have been generated to track the in vivo
activity of Tcf/Lef-mediated Wnt/
b
-catenin signaling. These ani-
mals carry a Wnt/
b
-catenin signaling reporter, which is driven
by a promoter containing multiple Tcf/Lef binding sites and a
TATA-containing minimal promoter sequence (Riese et al., 1997;
DasGupta and Fuchs, 1999; Staal et al., 2001; Dorsky et al., 2002;
Geng et al., 2003; Maretto et al., 2003; Mohamed et al., 2004;
Nakaya et al., 2005; Denayer et al., 2006; Moro et al., 2012).
However, several studies have reported that the activities of these
reporters do not reflect Wnt/
b
-catenin signaling activity (Riese
et al., 1997; Geng et al., 2003; Dessimoz et al., 2005; Fathke et al.,
2006). For example, the BAT-gal reporter did not drive reporter
expression in Wnt/
b
-catenin signaling-active sites in the devel-
oping pancreas (Dessimoz et al., 2005). The TATA-containing
minimal promoter is considered to be the source of this problem.
All minimal promoters in previously generated Wnt/
b
-catenin
signaling reporters were derived from the promoters of natural
genes, such as heat shock protein, c-fos, thymidine kinase,andsiamois
(Riese et al., 1997; DasGupta and Fuchs, 1999; Staal et al., 2001;
Dorsky et al., 2002; Geng et al., 2003; Maretto et al., 2003; Mohamed
et al., 2004; Nakaya et al., 2005; Denayer et al., 2006; Moro et al.,
2012). They not only contained TATA but also sequences derived
from each gene promoter, which may be affected by non-specific
signals (Robertson et al., 1995; Barolo, 2006).
Zebrafish are one of the most suitable animals for live imaging
because of their optical clarity and rapid development. A trans-
genic zebrafish line carrying the Wnt/
b
-catenin signaling reporter
TOPdGFP, which contains four copies of Tcf/Lef binding sites, a
c-fos minimal promoter, and a d2EGFP reporter gene (Dorsky
et al., 2002), has proved to be a useful tool for understanding the
regulation of in vivo Wnt/
b
-catenin signaling. However, we found
that the fine reporter activity can be observed only in some Wnt/
b
-catenin signaling-active sites in living TOPdGFP-transgenic fish
using a fluorescence stereomicroscope (present study). TOPdGFP-
transgenic fish also expressed d2EGFP transcripts in a manner
that was dependent on Lef1, but not Tcf7l1a, during early
embryonic development (Dorsky et al., 2002).
In the present study, we showed that a new transgenic zebrafish
line that carried a new Wnt/
b
-catenin signaling reporter, Tcf/Lef-
miniP:dGFP, emitted clear green fluorescence at various known Wnt/
b
-catenin signaling-active sites during embryogenesis. Tcf/Lef-
miniP:dGFP reporter activity was regulated by Lef1, Tcf7, Tcf7l1a,
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/developmentalbiology
Developmental Biology
0012-1606/$ - see front matter & 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ydbio.2012.07.016
n
Corresponding author. Fax: þ 81 92 642 6790.
E-mail address: tish@bioreg.kyushu-u.ac.jp (T. Ishitani).
Developmental Biology 370 (2012) 71–85
100
and Tcf7l2. Furthermore, this new reporter was used to visualize the
spatiotemporal activity of Wnt/
b
-catenin signaling during the growth
of young adult fish and the healing of an amputated tail of adult fish.
We also used reporter transgenic zebrafish to determine new roles for
Lef1, Tcf7, and Tcf7l2. This transgenic fish will prove to be a useful
tool for investigating the function and regulation of Wnt/
b
-catenin
signaling in vivo.
Materials and methods
Reporter plasmid constructs
A pT2-dGFP plasmid was constructed by inserting the XhoI-AflII
fragment of pd2EGFP-N1 (Clontech, Mountain View, CA, USA),
including the d2EGFP gene and SV40 polyadenylation sequence,
between the Tol2 excision sites of the Tol2 donor plasmid
pT2AL200R150G (Urasaki et al., 2006). To produce the reporter
plasmid Tcf/Lef-tkP:dGFP, a 400-bp promoter/enhancer region of
TOPFLASH (Millipore, Billerica, MA, USA), which contained six copies
of consensus Tcf/Lef binding sites and a thymidine kinase minimal
promoter (tkP: CCCGCCCAGCGTCTTGTCATTGGCGAATTCGAACACG-
CAGATGCAGTCGGGGCGGCGCGGTCCGAGGTCCACTTCGCATATTAAG-
GTGACGCGTGTGGCCTCGAACACCGAGCGACCCTGCAGC), was ampli-
fied by PCR and inserted into pT2-dGFP. Tcf/Lef-miniP:dGFP was
generated by replacing tkP of Tcf/Lef-tkP:dGFP with a minimal
promoter (miniP, AGAGGGTATATAATGGAAGCTCGACTTCCAG) deri-
ved from pGL4 (Promega, Madison, WI, USA). Tcf/Lef(MT)-
miniP:dGFP was constructed by changing the Tcf/Lef binding sites
of Tcf/Lef-miniP:dGFP using the mutant Tcf/Lef binding sites derived
from FOPFLASH (Millipore).
Cell culture assay
HEK293 cells were grown in Dulbecco’s modified Eagle’s
medium (DMEM) supplemented with 10% fetal bovine serum
(FBS) and 100 U/ml penicillin–streptomycin mixed solution. Cells
were transfected with the expression plasmids and reporter
plasmids using polyethylenimine MW 25,000 (Polysciences, War-
rington, PA, USA). Reporter gene expression was observed using
an MZ16FA microscope (Leica, Wetzlar, Germany).
Zebrafish maintenance
Zebrafish (AB, tcf7 mutants, TOPdGFP, hsp70:
D
Tcf3-GFP,
hsp70:Dkk1-GFP, and Tcf/Lef-miniP:dGFP transgenic fishes) were
raised and maintained under standard conditions.
mRNA synthesis
Tol2 transposase enzyme mRNA was generated using the pCS-
TP plasmid (Kawakami et al., 2004 ) as a template. Capped mRNA
was synthesized using an SP6 mMessage mMachine kit (Ambion,
Austin, TX, USA) and purified using Micro Bio-Spin columns (Bio-
Rad, Hercules, CA, USA).
Generation of transgenic zebrafish
We coinjected 25–50 pg of reporter plasmid DNA with
30–40 pg of Tol2 transposase mRNA into one-cell stage wild-type
zebrafish (AB) embryos. A transgenic fish strongly expressing d2EGFP
was outcrossed with a wild-type fish to produce the founder line
Tg(Tcf/Lef-miniP:dGFP)
isi01
(Line-1), which was maintained by
d2EGFP-positive sibling intercrossing. To generate a transgenic line
carrying a single transgene, Line-1 was outcrossed with wild-type fish
to produce Tg(Tcf/Lef-miniP:dGFP)
isi04
(Line-2) and Tg(Tcf/Lef-miniP:
dGFP)
isi05
(Line-3). Line-2 and Line-3 were maintained as homozy-
gous transgenic fishes.
Morpholino injections
Translation-blocking morpholinos against lef1 (lef1 MO),
tcf7l1a (tcf7l1a MO), tcf7l2 (tcf7l2 MO), and p53 MO (p53 MO)
and splice-blocking morpholinos against lef1 (lef1 spl MO) and
tcf7 (tcf7 spl MO) have been described previously (Dorsky et al.,
2002, 2003; Bonner et al., 2008; Ishitani et al., 2005; Meier et al.,
2006; Robu et al., 2007). Standard control MO was obtained from
Gene Tools (Philomath, OR, USA). We injected 2–3 ng of Tcf/Lef
morpholinos with 5 ng of p53 MO at the one-cell stage, as shown
in all figures, with the exception of Fig. 6(A–D). As shown in
Fig. 6(A–D), 0.2 or 1 ng of lef1 MO was coinjected with 5 ng of p53
MO. Coinjection of Tcf/Lef MOs with p53 MO eliminated the
possibility that Tcf/Lef MOs-induced phenotypes were due to
the MO-induced artificial activation of p53 (Robu et al., 2007).
Whole-mount in situ hybridization
Whole-mount in situ hybridization was performed according
to a standard protocol. A digoxigenin-labeled antisense RNA
probe was prepared from a template encoding the C terminus of
d2EGFP (corresponding to residues 422–461 of mouse ornithine
decarboxylase).
Genomic DNA isolation and Southern blot analysis
The tail fins of adult transgenic fish were amputated using a
razor’s edge and transferred to lysis buffer containing 0.1
m
g/
m
l
Proteinase K (ProK). The sample was incubated at 55 1C overnight,
followed by the standard ethanol precipitation. Purified genomic
DNA samples were digested with EcoRl, which cuts the plasmid
reporter once. Southern blot hybridization was performed using a
digoxigenin (Roche Diagnostics, Basel, Switzerland)-labeled probe
and standard methods.
Live imaging of reporter activity in zebrafish
Reporter expression was visualized using the MZ16FA fluor-
escence microscope (Leica) with a GFP3 filter, as shown in all
figures, with the exception of Fig. 3(K, L–N, and R). An MVX10
macro zoom fluorescence microscope (Olympus, Tokyo, Japan)
was used for obtaining Fig. 3(K, L, and N). A TCS-LSI macro zoom
fluorescence microscope (Leica) was used for obtaining Fig. 3(M
and R).
Zebrafish surgery
Zebrafish fin amputations were performed as described pre-
viously (Poss et al., 2000) after which the fish were returned to
water at a temperature of 28.5 1C.
Antibody
Anti-
b
-catenin (catenin-beta 1/2 (NT), Z-Fish) and anti-
a
-
tubulin antibodies were obtained from Sigma (St. Louis, MO,
USA) and ANA SPEC (Fremont, CA, USA), respectively.
Chemical inhibitor treatment
BIO (Wako, Osaka, Japan), XAV939, and IWR-1 (Enzo Life
Sciences, Farmingdale, NY, USA) were dissolved in DMSO. Early-
stage embryos were manually dechorionated and treated with
10
m
M BIO, XAV939, IWR-1, or a 1% DMSO control in the dark at
N. Shimizu et al. / Developmental Biology 370 (2012) 71–8572
28.5 1C. Young adult fish [1 month postfertilization (mpf) fish]
were incubated in 5
m
M XAV939 and IWR-1 for 18 hpf.
Quantitative PCR (qPCR)
Zebrafish embryos were randomly collected by nonfluores-
cence microscopy. Total RNAs from 20 embryos were purified
using Trizol reagents (Invitrogen), and cDNAs were synthesized
using ReverTra Ace qPCR RT Master Mix (Toyobo, Osaka, Japan).
qPCR analysis of DNA was performed using THUNDERBIRD SYBR
qPCR Mix (Toyobo) and d2EGFP primers (FW primer: AGGAGCG-
CACCATCTTCTT, RV primer: GATGTTGTGGCGGATCTTG) on an
Applied Biosystems 7500 real-time PCR system (Applied Biosys-
tems, Carlsbad, CA, USA). qPCR was performed in triplicate. The
level of gapdh (FW primer: AATTCTGGGATACACGGAGCACCA, RV
primer: TCAGGTCACATACACGGTTGCTGT) was used as a loading
control for real-time PCR.
DASPEI staining
Zebrafish embryos were immersed in 1 mM 2-(4-(dimethylami-
no)styryl)-N-ethylpyridiniumiodide (DASPEI; Molecular Probes,
Eugene, OR, USA) in egg water for 90 min. The embryos were then
rinsed thoroughly in egg water three times. DASPEI-stained cells
were visualized using the MZ16FA fluorescence microscope (Leica)
with a DsRed filter.
Results
Generation of new Wnt/
b
-catenin signaling reporter transgenic
zebrafish
Expression of d2EGFP proteins is detectable in many Wnt-
responsive domains in fixed TOPdGFP transgenic zebrafish larvae
that have been immunostained with anti-GFP antibody (Dorsky
et al., 2002), but the fine reporter activity can be observed only in
the midbrain of living TOPdGFP transgenic fish larvae using a
fluorescence stereomicroscope (Fig. S1AandBinsupplementary
material). It has been suggested that the sensitivity of Wnt/
b
-catenin signaling using the TOPdGFP reporter is insufficient for
live imaging. Furthermore, the expression pattern of the TOPdGFP
reporter at 8 hpf was not completely consistent with that of tbx6
mRNA (Fig. S2 in supplementary material), the induction of which
completely depends on Wnt8 signaling (Szeto and Kimelman,
2004). In addition, TOPdGFP mRNA was not detected in the median
fin fold at 25 hpf (Fig. S3 in supplementary material); Wnt/
b
-catenin signaling occurs in the median fin fold (Aman and
Piotrowski, 2008; Nagayoshi et al., 2008). These results suggested
that the TOPdGFP reporter does not completely reflect Wnt/
b
-catenin signaling activation in zebrafish embryos. Therefore,
we improved the sensitivity and responsiveness of the TOPdGFP
reporter. We increased the copy number of the Tcf/Lef binding sites
from four to six to amplify reporter sensitivity (Fig. 1A). We also
improved the responsiveness by changing the minimal promoter
from c-fos to either a pGL4 vector-derived promoter (miniP) or a
thymidine kinase promoter (tkP) (Fig. 1A) because the c-fos
minimal promoter of the TOPdGFP reporter contained a cAMP-
responsive element, which may affect the TOPdGFP activity
(Barolo, 2006). miniP is an artificial synthetic minimal promoter
that includes TATA boxes but no sequences derived from any
organisms. Thus, it can respond to the activation of upstream
responsive elements but not to background signals.
We tested whether the new reporters could drive d2EGFP
expression in response to Wnt/
b
-catenin signaling activation in
mammalian cells. Wnt-1 or
b
-catenin overexpression induced the
activation of both the miniP-type reporter (Tcf/Lef-miniP:dGFP)
and the tkP-type reporter (Tcf/Lef-tkP:dGFP) in HEK293 cells
(Fig. 1B) indicating that both reporters were under the control
of the Wnt/
b
-catenin signaling pathway.
We then introduced these reporters into zebrafish embryos by
Tol2 transposase-mediated transgenesis (Kawakami et al., 2004;
Urasaki et al., 2006). We found that 30-hpf embryos injected with
Tcf/Lef-miniP:dGFP or Tcf/Lef-tkP:dGFP expressed d2EGFP in the
dorsal midbrain (Nyholm et al., 2007), otic vesicle (Riccomagno
et al., 2005), posterior lateral line primordium (prim-I/pllp)
(Aman and Piotrowski, 2008), and pectoral fin bud and median
fin fold (Nagayoshi et al., 2008)(Fig. 1C) which are known as Wnt/
b
-catenin signaling-active sites. The activity of the Tcf/Lef-
miniP:dGFP reporter in zebrafish embryos was much stronger
than that of the Tcf/Lef-tkP:dGFP reporter (Fig. 1C). A Tcf/Lef-
miniP(MT):dGFP reporter containing mutated Tcf/Lef binding
sites did not express d2EGFP in HEK293 cells (Fig. 1B) or zebrafish
embryos (Fig. 1C) indicating that the Tcf/Lef binding sites and
not the miniP sequences were responsible for the reporter’s
activity. Therefore, we generated a stable transgenic line using
only Tcf/Lef-miniP:dGFP. We then obtained a stable line (Line-1)
that carried four copies of the transgenes (Fig. 1D and E). We also
outcrossed Line-1 twice with wild-type fish and obtained two
single-copy lines, i.e., Line-2 and Line-3 (Fig. 1D and E). The
strength of the reporter activity in Line-2 was almost equivalent
to that in Line-1, but it was relatively stronger than that in Line-3
(Fig. 1E). The reporter activity in these three lines was observed in
Wnt/
b
-catenin signaling-active sites, and their expression pat-
terns were almost identical throughout their development
(Fig. 1E and data not shown), suggesting that the action of the
Tcf/Lef-miniP:dGFP transgene was independent of its insertion
positions in a chromosome. We further confirmed that Tcf/Lef-
miniP:dGFP transgenic fish embryos expressed d2EGFP mRNA in a
domain that was very similar to the tbx6-expressing ventrolateral
region at 8 hpf (Fig. S2 in supplementary material), suggesting
that Tcf/Lef-miniP:dGFP transgenic fish may have correctly
expressed the reporter in Wnt/
b
-catenin signaling-active cells.
By observing the fluorescence and quantifying the d2EGFP mRNA
level, we found that reporter expression in Tcf/Lef-miniP:dGFP
transgenic fish was much stronger than that in TOPdGFP trans-
genic fish throughout embryogenesis (Fig. S1A in supplementary
material).
Tcf/Lef-miniP:dGFP drives reporter gene expression in
a Wnt/
b
-catenin signaling-dependent manner
To confirm that the reporter activities were dependent on
Wnt/
b
-catenin signaling, we blocked Wnt/
b
-catenin signaling
activity in the reporter line using an hsp70:
D
Tcf3-GFP or
hsp70:Dkk1-GFP construct. hsp70:
D
Tcf3-GFP and hsp70:Dkk1-
GFP express a dominant inhibitor of all Tcf/Lef,
D
Tcf3-GFP, and
Dkk1, a secreted inhibitor of Wnt/
b
-catenin signaling (Glinka
et al., 1998), in response to heat shock (Lewis et al., 2004;
Stoick-Cooper et al., 2007). Heat shock dramatically reduced
the expression of d2EGFP transcripts in embryos carrying both
hsp70:
D
Tcf3-GFP and Tcf/Lef-miniP:dGFP (8 hpf: 100%, n¼26,
12 hpf: 100%, n¼37, 25 hpf: 100%, n¼ 14) and in those carrying
both hsp70:Dkk1-GFP and Tcf/Lef-miniP:dGFP (8 hpf: 100%,
n¼ 16, 12 hpf: 100%, n ¼ 27, 25 hpf: 100%, n¼ 15) but not in those
carrying only Tcf/Lef-miniP:dGFP (Fig. 2A), suggesting that Wnt/
b
-catenin signaling activities were required for d2EGFP expres-
sion in the reporter lines. We further tested whether reporter
activity reflected the activity of the Wnt/
b
-catenin signaling
pathway using chemical inhibitors and an activator. Treatment
of the embryos with BIO, which is a specific inhibitor of the
N. Shimizu et al. / Developmental Biology 370 (2012) 71–85 73
Wnt/
b
-catenin signaling negative regulator GSK-3
b
, upregulated
reporter activity (100%, n¼12; Fig. 2B). XAV939 and IWR-1
induced a reduction in the
b
-catenin protein level by promoting
the stabilization of Axin protein in mammalian cells (Huang et al.,
2009; Chen et al., 2009). Treatment with XAV939 or IWR-1
decreased both the protein level of endogenous
b
-catenin in
zebrafish embryos (Fig. 2C) and the d2EGFP expression level in
Tcf/Lef-miniP:dGFP-carrying zebrafish embryos (XAV939: 100%,
n¼ 18; IWR-1: 100%, n¼18; Fig. 2D). Overall, these data suggest
that Tcf/Lef-miniP:dGFP transgenic fish expressed d2EGFP in a
manner that was dependent on Wnt/
b
-catenin signaling.
Dynamic expression of Tcf/Lef-miniP:dGFP in embryos, larvae,
juveniles, and adults
We then tested whether the reporter was precisely dri ven
in known Wnt/
b
-catenin signaling -active sites in Tcf/Lef-
miniP:dGFP transgenic fish. At approximately 4 hpf, maternal
b
-catenin activates bozozok gene expression in the dorsal
yolk syn cyti al layer (YSL), which controls organizer formation
(Scheider et al., 1996; Kelly et al., 2000; Xiong et al., 2006). In
agreement with this, the initial expression of d2EGFP mRNA wa s
observed in a s mall group of cells on the embryon ic margin that
expressed a dorsal marker, chordin,at3.7hpf(Fig. 3A), and this
expression bec ame stronger at 4.3 hpf (Fig. 3A). From 4.7 hpf,
d2EGFP mRNA expression decreased gradually i n the dorsal YSL
but increased in the ventrolateral mesoderm (Fig. 3 A), where
the initial zygotic Wnt/
b
-catenin signali ng activity is produced
by the Wnt8 ligand ( Christian et al., 1991). During ga strula tion,
expression in the ventrolateral mesoderm was continuous (Figs.
S2 and S3 in supplementary ma terial). During early develop-
ment, we f ailed to detect the r eporter activity in living embryos
using a fluorescence stereomicroscope because the fluorescence
was faint and it overlapped with the autofluorescence of the
yolk (Movie S1 in supplementary material). After 10 hpf, we
detected fluorescence in the posterior mesoderm of living
embryos (Fig. 3B and C; Movie S1 in supplementary material).
Throughout the s omite stage , fluorescence was observed in the
Fig. 1. Generation of new Wnt/
b
-catenin signaling reporters. (A) Schematic diagrams of Wnt/
b
-catenin signaling-reporter constructs. Tcf/Lef BS: consensus sequence of the
Tcf/Lef-binding site. PolyA: SV40 polyadenylation sequence. (B) The Wnt/
b
-catenin signaling-reporter constructs drive d2EGFP expression in HEK293 cells in response to
the activation of Wnt/
b
-catenin signaling. HEK293 cells were transfected using a reporter gene with a control empty vector (Vector) or an expression plasmid-encoding
mouse Wnt-1 (Wnt-1) or
b
-catenin (
b
-catenin), as indicated. Scale bar, 200
m
m. (C) Comparison of reporter activities in zebrafish embryos transiently introduced with
various Wnt/
b
-catenin signaling-reporter constructs. Left side views of 30-hpf zebrafish embryos injected with either Tcf/Lef-miniP:dGFP, Tcf/Lef(MT)-miniP:dGFP, or Tcf/
Lef-tkP:dGFP with Tol2 transposase mRNA, where the anterior is to the left. Cells expressing d2EGFP were visualized by fluorescence microscopy (right panels). Bright-field
(BF) images are shown in the left panels. Scale bar, 200
m
m. Fluorescence was observed in embryos injected with Tcf/Lef-miniP:dGFP (55%, n ¼ 22) and Tcf/Lef-tkP:dGFP
(45%, n¼ 20). In contrast, Tcf/Lef(MT)-miniP:dGFP did not express d2EGFP (n¼ 43). dmb: dorsal midbrain, ov: otic vesicle, mff: median fin fold, pfb: pectoral fin bud.
(D) Southern blot analysis of the transgenes in Tcf/Lef-miniP:dGFP transgenic zebrafish lines. Genomic DNA was prepared from the tail fins of Line-1, Line-2, and Line-3,
and used for Southern blot analysis. (E) Comparison between Line-1, Line-2, and Line-3 reporter activities. Panels show the left side views of 30-hpf Line-1, Line-2, and
Line-3 embryos, with the anterior to the left. Cells expressing d2EGFP were visualized by fluorescence microscopy (right panels). Bright-field (BF) images are shown in the
left panels. Scale bar, 200
m
m.
N. Shimizu et al. / Developmental Biology 370 (2012) 71–8574
Fig. 2. Tcf/Lef-miniP:dGFP reporter expression reflected Wnt/
b
-catenin signaling in vivo. (A) Dkk1-GFP or
D
Tcf3-GFP expression inhibited Tcf/Lef-miniP:dGFP reporter
activation. Homozygous Tcf/Lef-miniP:dGFP transgenic zebrafish Line-2 was crossed with heterozygous hsp70:Dkk1-GFP or hsp70:
D
Tcf3-GFP transgenic zebrafish. The
collected embryos were exposed to heat shock at 37 1C from 5 to 6, 10 to 11, or 22 to 23 hpf. They were then classified as hsp70:Dkk1-GFP carriers ( þ / ), hsp70:
D
Tcf3-
GFP carriers (þ / ), or non-carriers ( / ) by determining whether Dkk1-GFP or d2EGFP expression was detected by in situ hybridization at 8, 12, or 25 hpf. To determine
d2EGFP mRNA expression, the PEST sequence-coding region of d2EGFP was used as a probe (the schematic diagram shown in the lower panel). The left and middle panels
show embryos with the ventral side to the left. The right panels show embryos with the anterior side to the left. Scale bar, 100
m
m. (B) Tcf/Lef-miniP:dGFP reporter activity
was enhanced by BIO treatment. Left-side views of Line-2 treated with 10
m
M BIO from 24 to 30 hpf, with the anterior side to the left. (C) XAV939 or IWR-1 treatment
reduced the
b
-catenin protein level. Zebrafish embryos were treated with DMSO or 10
m
M XAV939 or IWR-1 from 5 to 8 hpf. Extracts were harvested from embryos at
8 hpf and immunoblotted with anti-
b
-catenin and anti-
a
-tubulin antibodies. (D) XAV939 or IWR-1 treatment reduced Tcf/Lef-miniP:dGFP reporter activity. Panels show
lateral views of Line-2 treated with DMSO, 10
m
M XAV939, or IWR-1 from 18 to 28 hpf, with the anterior side to the left. d2EGFP-expressing cells were visualized by
fluorescence microscopy (lower panels). Bright-field (BF) images are shown in the upper panels. Scale bar, 200
m
m.
N. Shimizu et al. / Developmental Biology 370 (2012) 71–85 75
Fig. 3. The Tcf/Lef-miniP:dGFP reporter was activated in a range of zebrafish embryo and larval tissues. (A) Whole-mount in situ hybridization of d2EGFP (red) and chordin
(blue) at the indicated stage in Tcf/Lef-miniP:dGFP transgenic zebrafish embryos. The upper panels show dorsal views of embryos, with the animal side to the top. The
lower panels show animal-side views of embryos, with the dorsal side to the right. The d2EGFP mRNA-expressing dorsal and ventrolateral regions are indicated with
arrowheads and broken lines, respectively. Scale bar, 200
m
m. (B–R) Expression of d2EGFP proteins in Tcf/Lef-miniP:dGFP transgenic zebrafish embryos and larvae.
d2EGFP-expressing cells were visualized by fluorescence microscopy (C, E, G, I–R). (B, D, F, H) show bright-field (BF) images. Scale bar, 200 100
m
m. (C–H) Left-side views of
10- (B, C), 18- (D, E), and 24-hpf (F, G) embryos. tail bud (tb), midbrain–hindbrain boundary (mhb), newly formed somites (so), dorsal retina (dre), dorsal midbrain (dmb),
otic vesicle (ov), pronephric duct (pd), and median fin fold (mff). (H, I) Otic vesicle of 50-hpf embryos, with the anterior to the left and dorsal to the top. (J, K) Left-side head
(J) and dorsal trunk (K) views of 30-hpf embryos. dorsal diencephalon (dd), hypothalamus (hy), pectoral fin bud (pfb), and neural crest cells (ncc). (L) Lower left-side head
view of 40-hpf embryos. Developing pharyngeal arch (pa). (M–P) Dorsal head (M), dorsal trunk (N, O), and left-side tail (P) views of 50-hpf embryos were observed by
macro confocal microscopy (M, N, P) or using a fluorescence stereomicroscope (O). Tectum (te), retina (re), epiphysis (ep), pectoral fin fold (pff) and mesenchymal cells
(pfm), liver (lv), pancreas (pnc), and caudal fin mesenchymal cells (cfm). (Q) Heart of 58-hpf embryos. The endocardial cushion (ec) is indicated with white arrowheads.
Atrium (at), ventricle (ve). (R) Front facial view of an 80-hpf embryo, with the dorsal side to the top.
N. Shimizu et al. / Developmental Biology 370 (2012) 71–8576
posterior tissues (Movie S1 in the s upplementary material).
Fromthelatesomitestage,fluorescencewasobservedinthe
tail bud, newly formed somites, midbrain–hindbrain boundary
(MHB), and prim-I (Fig. 3D and E). Dorsal midbrain, dorsal
retina, median fin fold, otic vesicle, and pronephric ducts also
emitted fl uorescence from 24 hpf (Fig. 3F and G ). Fluor escence
was more restricted in the otic vesicle at 5 0 hpf, where it
became localized in the dorsal region ( Fig. 3HandI).Fluores-
cence was also observed in the dorsal diencephalon, pectoral fin
bud, pe ctoral fin fo ld, hypothalamus, migrating neural crest
cells, developing pharyngeal arch, growing midbrain tectum,
retina, epiphysis, liver, pancreas, endocardial cushion, and the
lip in larvae (Fig. 3 J–M and O–R; Movies S2 and S3 in supple-
mentary material). The following tissues have been reported as
Wnt/
b
-catenin signaling-active sites: zebrafish tail bud (Thorpe
et al., 2005), mouse somites (Ikeya and Takada, 1998), MHB
(Lekven et al., 2003; Buckles et al. , 2004), zebr afish retina
(Yamaguchi et al., 2005), pronephric duct (Caron et al ., 2012),
dorsal di encephalon, epiphysis (Masai et al., 199 7), pectoral fin
fold (Nagayoshi et al., 2008), hypothalamus (Lee e t al ., 2006;
Wang et al., 2009), zebrafish migrating neural crest (Dorsky
et al., 1998), pharyngeal arch (Lewis et a l., 2004), endocardial
cushion (Hurlstone et al., 2003), mouse lip (Song et al., 2009),
and liver and pancreas (Ver zi and Shivdasani, 2 008). T hus, the
Tcf/Lef-miniP:dGFP re porter was driven in kn own Wn t/
b
-cate-
nin signaling-active sites du ring embr yogenesis an d
organogenesis.
Supplementary material related to this article can be found
online at doi:10.1016/j.ydbio.2012.07.016.
We observed Tcf/Lef-miniP:dGFP-induced fluorescence in
young fish. Reporter expression was continuously detected in
the tectum, retina, liver, and fin during larval growth (Fig. 4A–D).
Furthermore, the skin and nose also expressed the reporter at
1 mpf (Fig. 4A and B). In agreement with these observations,
Wnt/
b
-catenin signaling is known to contribute to homeostasis
in the mouse olfactory epithelium (Wang et al., 2011) and skin
(Haegebarth and Clevers, 2009). Treatment of young fish with
XAV939 or IWR-1 reduced reporter expression in the aforemen-
tioned juvenile fish tissues (XAV939: 100%, n¼8; IWR-1: 100%,
n¼ 9; Fig. 4E and data not shown), confirming Wnt/
b
-catenin
signaling activity in these regions.
We failed to detect clear reporter activity in living adult
Tcf/Lef-miniP:dGFP transgenic fish. Several previous reports have
shown that the Wnt/
b
-catenin signaling pathway contributes to
tissue regeneration (Tal et al., 2010). After fin amputation, the
expression of lef1 transcripts was upregulated in a Wnt/
b
-catenin
signaling-dependent manner during epidermis regeneration (Poss
et al., 2000; Stoick-Cooper et al., 2007). However, we did not
detect TOPdGFP reporter activation in the amputated fins of living
transgenic fish (data not shown), although TOPdGFP transcripts
were detectable in the amputated fin of the fixed fish at 2 days
postamputation (dpa) by in situ hybridization (Stoick-Cooper
et al., 2007). In contrast, Tcf/Lef-miniP:dGFP activation was read-
ily detected in the amputated fin of living transgenic fish from
Fig. 4. Tcf/Lef-miniP:dGFP transgenic zebrafish expressed d2EGFP in various tissues during juvenile growth processes. (A–D) d2EGFP protein expression in 1 mpf Tcf/Lef-
miniP:dGFP transgenic zebrafish. Lateral (A), dorsal head (B), ventral head (C), and lateral liver (D) views of transgenic fish, with the anterior side to the left. d2EGFP-
expressing cells were visualized by fluorescence microscopy. Scale bar, 200
m
m. Nose (no), retina (re), tectum (te), gill, pectoral fin (pf), skin (sk), pelvic fin (pf), median fin
mesenchymal cell (mf), anal fin (af), dorsal fin (df), median fin fold (mff), and caudal fin (cf). We could not distinguish d2EGFP expression in the liver or skin of intact fish
(data not shown). Therefore, the skin was peeled away to clearly visualize expression in the liver (D). (E) XAV939 or IWR-1 treatment reduced the
b
-catenin protein level
and Tcf/Lef-miniP:dGFP reporter activity. The panels show left-side head or trunk views of 1-mpf Line-2 juveniles treated with DMSO or 10
m
M XAV939 or 10
m
M IWR-1
for 24 h, with the anterior side to the left. d2EGFP-expressing cells were visualized by fluorescence microscopy (lower panels). Bright-field (BF) images are shown in the
upper panels. Scale bar, 200
m
m.
N. Shimizu et al. / Developmental Biology 370 (2012) 71–85 77
2 dpa (Fig. 5). Thus, Tcf/Lef-miniP:dGFP transgenic fish were
useful for the detection of Wnt/
b
-catenin signaling activity during
tissue regeneration.
We continuously observed reporter expression and discovered
that the anterior lateral line primordium (prim-A/allp), dorsal
lateral line primordium (prim-D), secondary posterior lateral line
primordium (prim-II), caudal and pectoral fin mesenchymal cells,
and gills (Figs. 3G, N–P, R, and 4C; see Fig. S3 and Movies S2
and S3 in supplementary material) contained novel populations of
potential Wnt/
b
-catenin signaling-active cells.
The Tcf/Lef-miniP:dGFP reporter revealed the activity of four Tcf/Lef
members in zebrafish
Zebrafish have ve Tcf/Lef family genes, i.e., lef1, tcf7, tcf7l1a,
tcf7l1b,andtcf7l2. TOPdGFP reporter activation is dependent on
lef1 but not tcf7l1a (Dorsky et al., 2002). Therefore, we tested
which Tcf/Lef members participated in the regul ation of the Tcf/
Lef-miniP:dGFP reporter by quantifying and observing d2EGF P
expression in transgenic fish injected with translation-blocking
morpholinos against lef1 (lef1 MO), tcf7l1a (tcf7l1a MO), and
tcf7l2 (tcf7l2 MO) and splice-blocking morpholinos against lef1
(lef1 spl MO) and tcf7 (tcf7 spl MO). d2EGFP mRNA expression in
10.5-hpf embryos and d2EGFP uorescence in the posterior
tis sues o f 12-hpf embryos were severely blocked by inj ection
with 1 ng of lef1 MO (100%, n¼ 20) (Fig. 6A and B) or coinjection
of tcf7 s pl MO with 0.2 ng of lef1 MO (100%, n ¼ 20) (F ig. 6Cand
D). In contrast, in jection with tcf7 spl MO (Fig. 6A and B) or a low
dose (0.2 ng) of lef1 MO (Fig. 6C a nd D) had little effect, while
tcf7l2 MO injection had no effect (Fig. 6A and B). These results
indicate that lef1 and tcf7, but not tcf7l2, contribute to Tcf/Lef-
miniP:dGFP reporter act ivation in the posterior tissues, while
they also sugges t that Wnt/
b
-catenin signaling in these tissues
was synergistically p romoted by Lef1 and Tc f7. This w as con-
sistent with the expression of lef1 and tcf7, but not tcf7l2,inthe
posterior tissues (Fig. 6E). Tcf7l1a and Tcf7lb coopera te to induce
the anterior–posterior patterning of the zebrafish brain by
repressing the caudal gene (Dorsky et al., 2003; Ki m et al.,
2000). Consistent with this, tcf7l1a mRNA expre ssion was
observed in both the anterior and posterior tissues (Fig. 6E),
and tcf7l1a MO injection enhanced the d2EFGP expression level
(Fig. 6A and B) and expanded the wid th of the reporter-expres -
sing posterior tissues (F ig. 6F and G). These r esults indicate that
Tcf7l1a negatively regulates Tcf/Lef-miniP:dGFP reporter
expression. Previo us reports h ave shown that Lef1 and Tcf7
function redundantly during zebrafish development. Lef1 and
Tcf7 cooperatively activate the expression of a Wnt/
b
-catenin
signaling target gene, zic2a, which promotes cell proliferation in
the dorsal midbrain (Nyholm et al., 2007; Ota et al., 2012). A
mutation in the lef1 gene reduces the number of prim-I-derived
posterior lateral line (pLL) neuromasts, which is further reduced
by a Tcf7 knockdow n in the lef1 mutant (McGraw et al., 2011;
Valdivia et al., 2011). It has been reported that T cf7l2 is also
expressed in the dorsal midbrain and prim-I (Young et al., 2002;
Valdivia et al., 2 011). In agreement with these observations,
injection with lef1 MO (100%, n¼34) , tcf7 spl MO (100%, n¼32),
or tcf7l2 MO (100%, n¼ 32) reduced reporter expressio n in t he
dorsal midbrain an d p rim-I of 24-hpf embryos (Fig. 7A). W e
confirmed that the dorsal midbrain an d prim-I exp ressed tcf7l2,
lef1,andtcf7 at 24 hpf (Fig. 7B). These results suggest that Lef1,
Tcf7, and Tcf7l2 positi vely regulate Wnt/
b
-catenin signaling in
the midbrain and prim-I. We also tested whether Tcf7l2 was
involved in midbrain developme nt using a transg enic zebrafish
line carrying the HuC:Kaede reporter, whi ch expresses the
fluorescent protein Kaede in neurons under the control of a
neuron-specific HuC/elavl3 promoter (Sato et al., 2006). Similar
to lef1 MO injection, tcf7l2 MO inj ection decreased the size of the
midbrain tectum of 80 -hpf embryos (Fig. 7CandFig. S4 in
supplementary material). These results suggest that Lef1, Tcf7,
and Tcf7l2 contribute to midbrain tectum development via the
positive regulation of Wnt/
b
-catenin signaling. Lef1 is also
essential f or somitogenesis (Galceran et al., 2004)andXenopus
tail formation (Ro
¨
el et al., 2009). We found that lef1 MO injection
(100%, n ¼ 34) or coinjection of tcf7 spl MO with lef1 spl MO
(100%, n ¼ 29) drastically reduced reporter expression in the tail
bud of transgenic zebrafish e mbryos (Fig. 7D), whereas injection
with tcf7 spl MO alone (100%, n ¼ 32) caused a slight reduction
(Fig. 7D) and the posterior somites a nd tail bud expressed
lef1 and tcf7 mRNA (Fig. 7B). These results suggest that Lef1
and Tcf7 function cooperatively in the som ites and tail bud.
Thus, lef1, tcf7, tcf7l1a,andtcf7l2 participate in the regulation of
Tcf/Lef-miniP:dGFP.
Wnt/
b
-catenin signaling is involved in the development of the
anterior lateral line (aLL), secondary pLL, and dorsal pLL
Previous reports have shown that Wnt/
b
-catenin signaling
functions in prim-I and regulates primary pLL neuromast forma-
tion (Aman and Piotrowski, 2008; Ma and Raible, 2009). Inhibi-
tion of Wnt/
b
-catenin signaling narrows the distance between the
primary pLL neuromasts and reduces their number (McGraw
et al., 2011). In this study, we found that the Tcf/Lef-miniP:
dGFP reporter was expressed in prim-I, prim-A, prim-II, and prim-
D. Prim-A, prim-II, and prim-D also contribute to neuromast forma-
tion in aLL, secondary pLL, and dorsal pLL, respectively (Raible and
Kruse, 2000; Ledent, 2002; Sap
ede et al., 2002; Ghysen and Dambly-
Fig. 5. Tcf/Lef-miniP:dGFP reporter was activated in the amputated caudal fin regenerative outgrowths of adult transgenic fish. The caudal fin was amputated at the white
line indicated, as shown for 4 dpa (0–4 dpa). d2EGFP-expressing cells were visualized by fluorescence microscopy (lower panels). The fluorescence images were merged
with bright-field (BF) images (upper panels). The Tcf/Lef-miniP:dGFP reporter was upregulated in the amputated fin during regeneration. Scale bar, 200
m
m.
N. Shimizu et al. / Developmental Biology 370 (2012) 71–8578
Chaudi
ere, 2007). Therefore, we tested whether Wnt/
b
-catenin
signaling was also required for the development of aLL, secondary
pLL, and dorsal pLL. Induction of
D
Tcf3-GFP (100%, n¼ 11) or Dkk1-
GFP (100%, n¼ 24) expression from 22 to 23 hpf blocked d2EGFP
mRNA expression in prim-A of 25-hpf embryos of Tcf/Lef-
miniP:dGFP transgenic fish. Furthermore, these inductions blocked
the formation of DASPEI-labeled neuromasts in the head and trunk
at 80 hpf (
D
Tcf3-GFP: 100%, n¼ 9, Dkk1-GFP: 100%, n¼ 23), which
are derived from prim-A and prim-I, respectively (Raible and Kruse,
2000)(Fig. 8A and B). Treatment with XAV939 (100%, n¼ 10) or
IWR-1 (100%, n¼ 10) from 18 to 27 hpf and from 18 to 44 hpf also
inhibited Tcf/Lef-miniP:dGFP reporter expression in prim-A at
27 hpf and neuromast formation in the head at 80 hpf, respectively
(Fig. 8C). In addition, BIO treatment enhanced reporter expression
in prim-A (Fig. 2B, arrowheads). These results suggest that Wnt/
b
-catenin signaling is activated in the migrating prim-A and that
it regulates aLL development. However, induction of Dkk1-GFP
expression from 50 to 52 hpf blocked d2EGFP mRNA expression in
the prim-D and prim-II of 55-hpf embryos of Tcf/Lef-miniP:dGFP
transgenic fish (Fig. S5A in supplementary material, 100%, n¼ 14),
Fig. 6. Tcf/Lef-miniP:dGFP reporter activity during early embryogenesis was dependent on the Tcf/Lef family of transcription factors. (A–D) The Tcf/Lef-miniP:dGFP
reporter activity was regulated positively by Lef1, Tcf7, and Tcf7l2 and negatively by Tcf7l1a. Control MO, 1 or 0.2 ng of lef1 MO, tcf7l2 MO, tcf7l1a MO, or tcf7 spl MO was
coinjected with p53 MO into one cell-stage Line-2 embryos, as indicated. (A, C) At 10.5 hpf, 20 embryos were collected and their total RNA was purified. The relative
expression levels of d2EGFP mRNAs at 10.5 hpf were analyzed by qPCR. (B, D) Left-side views of MO-injected reporter fish embryos, as indicated, with the anterior side to
the top. Scale bar, 200
m
m. (E) Whole-mount in situ hybridization staining for lef1, tcf7, tcf7l2, and tcf7l1a in 12.5-hpf zebrafish embryos. Tcf/Lef mRNA expression in the
anterior and posterior tissues is indicated with arrows and arrowheads, respectively. Scale bar, 200
m
m. (F, G) Tcf7l1a negatively regulated Tcf/Lef-miniP:dGFP reporter
activity. Control MO (n¼ 28) or tcf7l1a MO (n¼ 38) was coinjected with p53 MO into one cell-stage Line-2 embryos as indicated. (F) Whole-mount in situ hybridization
staining for d2EGFP using 10.5-hpf MO-injected Line-2 embryos. Scale bar, 200
m
m. (G) The width of the d2EGFP mRNA-expressing area was quantified by measuring the
lengths of ‘‘a’’ and ‘‘b’’, which are indicated in (F), using Image J (NIH).
N. Shimizu et al. / Developmental Biology 370 (2012) 71–85 79
while repeated induction of Dkk1-GFP expression during 50–52 and
72–74 hpf inhibited the formation of prim-D-derived dorsal neuro-
masts and prim-II-derived neuromasts (Fig. S5A in supplementary
material, 91%, n¼ 33). In addition, treatment with XAV939 from 36
to 50 hpf also blocked d2EGFP mRNA expression in the prim-D
and prim-II of 50-hpf embryos of Tcf/Lef-miniP:dGFP transgenic fish
(Fig. S5B in supplementary material, 100%, n¼ 12) as well as the
formation of prim-D- and prim-II-derived neuromasts (Fig. S5Bin
supplementary material, 58%, n¼ 12). These results suggest that
Wnt/
b
-catenin signaling also contributes to the development of
dorsal and secondary pLLs.
Recent reports have shown that lef1 mutants lack a proportion
of prim-I-derived trunk neuromasts, while tcf7 MO injection
further reduces the trunk neuromast number (McGraw et al.,
2011; Valdivia et al., 2011 ). Therefore, we investigated whether
Lef1 and Tcf7 were also involved in aLL development. Injection
with lef1 spl MO or tcf7 spl MO had little effect on the activity of
the Tcf/Lef-miniP:dGFP reporter in prim-A, whereas it weakly
reduced the neuromast number in the head (Fig. 8D and Fig. S6A
in supplementary material). In contrast, coinjection with lef1 spl
MO and tcf7 spl MO strongly reduced both the activity of the Tcf/
Lef-miniP:dGFP reporter in prim-A and the neuromast number in
the head (Fig. 8D and Fig. S6A in supplementary material). In
addition, aLL was formed normally in a tcf7 mutant that was
generated by Tol2 transposon-mediated mutagenesis (Nagayoshi
et al., 2008), and injection of lef1 spl MO drastically reduced
the neuromast number in the head (Fig. 8E and Fig. S6Bin
supplementary material). These results suggest that Tcf7 and
Lef1 promote Wnt/
b
-catenin signaling in prim-A and that their
cooperative regulation of Wnt/
b
-catenin signaling contributes to
aLL development.
Wnt/
b
-catenin signaling is involved in gill filament growth
Gill filaments are feather-like projections on the gills. Gill
filaments bud from the pharyngeal ectoderm around 72 hpf in
zebrafish embryos (Kimmel et al., 1995; Hogan et al., 2004). We
found that the Tcf/Lef-miniP:dGFP reporter was strongly
expressed in the gill filament buds at 82 hpf (Fig. 9A). In addition,
d2EGFP expression decreased gradually with gill filament growth
and persisted in the body region of gill filaments (Fig. 9A,
100 hpf). This dynamic expression of the reporter in the gill
Fig. 7. The Tcf/Lef family of transcription factors was required for Wnt/
b
-catenin signaling activity in the midbrain, pLL, and tail bud. (A, D) Tcf/Lef-miniP:dGFP reporter
activity in the dorsal midbrain (dmb) and prim-I was dependent on Lef1, Tcf7, and Tcf7l2, and this activity in the newly formed somites (so) and tail bud (tb) was regulated
by Lef1 and Tcf7. Control MO, lef1 MO, tcf7 spl MO, tcf7l2 MO, or lef1 spl MO was coinjected with p53 MO into one cell-stage Line-2 embryos, as indicated. Left-side head
(A) and tail (D) views of MO-injected Line-2 embryos, with the anterior side to the left. Bright-field (BF) images are shown in the upper panels. d2EGFP-expressing cells
were visualized by fluorescence microscopy (lower panels). Scale bar, 200
m
m. (B) Whole-mount in situ hybridization staining for lef1, tcf7, and tcf7l2 using 24-hpf
(D) zebrafish embryos. Tcf/Lef mRNAs were detected in the dorsal midbrain (dmb), prim-I, newly formed somites (so) and tail bud (tb). Scale bar, 200
m
m. (C) Tcf7l2 was
required for midbrain development. Dorsal views of 80-hpf HuC-Kaede transgenic zebrafish embryos injected with control MO, lef1 MO, tcf7l2 MO, or p53 MO. In A, bright-
field (BF) images are shown in the left panels. Kaede-expressing cells were observed by fluorescence microscopy (right panels). Broken lines indicate the tectum or
presumptive tectal region. Scale bar, 100
m
m.
N. Shimizu et al. / Developmental Biology 370 (2012) 71–8580
filaments suggests that Wnt/
b
-catenin signaling may be involved
in gill filament development. Therefore, we tested this possibility.
Treatment with XAV939 (100%, n¼10) or IWR-1 (100%, n¼10)
from 72 hpf blocked d2EGFP expression in the gill filament buds
of 82-hpf reporter fish embryos (Fig. 9A). The gill filament length
of 100-hpf embryos treated with XAV939 or IWR-1 was shorter
than that of DMSO-treated control embryos (Fig. 9A and B),
although it did not affect the formation of gill filament buds
(Fig. 9A). In addition, induction of Dkk1-GFP expression also
reduced d2EGFP mRNA expression in the gill filament buds of
77-hpf reporter fish embryos (Fig. 9C, 100%, n¼ 14) and the gill
filament length of 102-hpf embryos (Fig. 9C and D, 100%, n¼ 11).
These results suggest that Wnt/
b
-catenin signaling is required for
gill filament growth. We also found that Lef1 was expressed in the
gill filament buds (Fig. S7A in supplementary material) and that
lef1 MO or lef1 spl MO injection decreased the gill filament length
of 102-hpf embryos (Fig. S7B and C in supplementary material),
suggesting that Lef1 is involved in gill filament growth.
Fig. 8. Tcf7/Lef1-mediated Wnt/
b
-catenin signaling was required for aLL development. (A, B) Induction of
D
Tcf3-GFP or Dkk1-GFP expression blocked Tcf/Lef-miniP:dGFP
reporter activity in prim-A and reduced the neuromast number. hsp70:
D
Tcf3-GFP- or hsp70:Dkk1-GFP-carrying Line-2 embryos were generated and classified using methods
similar to those shown in Fig. 2(A). hsp70:Dkk1-GFP carriers (þ / ), hsp70:
D
Tcf3-GFP carriers (þ / ) or non-carriers ( / ) were exposed to heat shock for 1 h at 37 1Cat
22 hpf. d2EGFP expression was detected by in situ hybridiz