Preaxial polydactyly caused by Gli3
haploinsufficiency is rescued by Zic3
loss of function in mice
Malgorzata E. Quinn, Allison Haaning and Stephanie M. Ware∗
Division of Molecular Cardiovascular Biology, Cincinnati Children’s Hospital Medical Center,
University of Cincinnati College of Medicine, Cincinnati, OH, USA
Received September 29, 2011; Revised December 28, 2011; Accepted January 4, 2012
Limb anomalies are important birth defects that are incompletely understood genetically and mechanistically.
GLI3, a mediator of hedgehog signaling, is a genetic cause of limb malformations including pre- and postaxial
polydactyly, Pallister–Hall syndrome and Greig cephalopolysyndactyly. A closely related Gli (glioma-asso-
ciated oncogene homolog)-superfamily member, ZIC3, causes X-linked heterotaxy syndrome in humans
but has not been investigated in limb development. During limb development, post-translational processing
of Gli3 from activator to repressor antagonizes and posteriorly restricts Sonic hedgehog (Shh). We demon-
strate that Zic3 and Gli3 expression overlap in developing limbs and that Zic3 converts Gli3 from repressor to
activator in vitro. In Gli3 mutant mice, Zic3 loss of function abrogates ectopic Shh expression in anterior limb
buds, limits overexpression in the zone of polarizing activity and normalizes aberrant Gli3 repressor/Gli3
activator ratios observed in Gli31/2 embryos. Zic3 null;Gli31/2 neonates show rescue of the polydactylous
phenotype seen in Gli31/2 animals. These studies identify a previously unrecognized role for Zic3 in regu-
lating limb digit number via its modifying effect on Gli3 and Shh expression levels. Together, these results
indicate that two Gli superfamily members that cause disparate human congenital malformation syndromes
interact genetically and demonstrate the importance of Zic3 in regulating Shh pathway in developing limbs.
Zic (zinc finger protein of the cerebellum) and Gli
(glioma-associated oncogene homolog) transcription factors
are members of the Gli superfamily of proteins which share
a highly conserved zinc finger domain and have critical
roles in multiple developmental processes. The murine Zic
was first identified as a zinc finger protein expressed in
granule cells throughout cerebellar development (1). The
expression pattern of the mouse Zic1-5 genes suggests their
essential roles in body pattern formation (2). In humans, muta-
tions in ZIC1-4 and GLI3 genes result in important classes of
developmental abnormalities: ZIC1 and ZIC4 mutations result
in Dandy–Walker malformation; ZIC2 causes holoprosence-
phaly; ZIC3 mutation or deletion results in heterotaxy
syndrome, a disorder characterized by disruption of left–
right axis patterning; GLI3 mutations result in complex anom-
alies of the brain and digits (Greig cephalopolysyndactyly and
Pallister–Hall syndromes) as well as isolated polydactyly
Gli3 is a downstream mediator of Sonic hedgehog (Shh)
signaling, a role which has been particularly well documented
in limb development (9–14). Anteroposterior (A/P) patterning
in the limbs of amniotes is controlled through secretion of Shh
protein by posterior limb bud mesoderm (zone of polarizing
activity—ZPA). Studies in vivo and in vitro suggest that
Gli3 negatively regulates the expression of both Shh (by
restricting it to the posterior mesoderm) and its target genes
through a repressor form (Gli3R) (11,12,15,16). In the
absence of Shh signal, Gli3 protein is phosphorylated and
cleaved constitutively to produce the transcriptional repressor
(Gli3R). Shh activity blocks Gli3 processing, yielding a full-
length activator form (Gli3A). The Gli3R is present at high
levels in the anterior developing limb bud and at low levels
in the posterior (15). The intracellular gradient of Gli3R,
which opposes the extracellular gradient of Shh, ultimately
∗To whom correspondence should be addressed at: Cincinnati Children’s Hospital Medical Center, 240 Albert Sabin Way, MLC 7020, Cincinnati,
OH 45229, USA. Tel: +1 5136369427; Fax: +1 5136365958; Email: email@example.com
# The Author 2012. Published by Oxford University Press. All rights reserved.
For Permissions, please email: firstname.lastname@example.org
Human Molecular Genetics, 2012, Vol. 21, No. 8
Advance Access published on January 10, 2012
controls A/P limb patterning in a manner not yet fully unrav-
eled (10,13,15,17). It has been proposed that the gradient of a
Gli3R/Gli3A ratio determines digit number and identity (10).
Shh expression in the developing limb bud is regulated via a
complicated network of transcription factors and growth
factors acting in concert, in which Gli3 plays its part. Others
include Hand2 (Hand—heart and neural crest derivatives
expressed transcript) and Tbx3 (Tbx—T box) providing com-
petence for Shh in posterior limb bud mesenchyme (13,18,19),
Fgf4 and Fgf8 (Fgf, fibroblast growth factor) factors from the
apical ectodermal ridge inducing and maintaining Shh expres-
sion (20,21) and Alx4 (homeobox protein aristaless-like 4) and
other factors preventing Shh expression in anterior and distal
mesenchyme (22). In this paper, we document for the first
time a potential role of Zic3 in regulating Shh expression in
the developing limb buds.
Mutations in Gli3 and/or Shh are known to affect digit
number and identity. The semidominant mouse mutation
Extra toes-J (Xt-J) generates a Gli3 null allele resulting in pre-
axial digit 1 iterations in heterozygotes (23). The limbs of
(hereafter designated Gli32/2) exhibit
severe polydactyly and loss of digit identities (23). The com-
plete loss of Shh function in the Shh2/2 mutant mouse
results in severe skeletal deficiencies distal to the stylopod–
zeugopod junction (elbow/knee joints); all zeugopod and
autopod elements are either missing, fused or lack normal
identity, except for a single digit 1 in the hindlimb (10,24).
Gli32/2 embryos express Shh ectopically at the anterior
mesoderm of developing limb buds (11), without effect on
skeletal patterningin the
Shh2/2;Gli32/2 and Gli32/2 limbs are virtually indistin-
guishable (10). The current paradigm argues that Gli3R exerts
a potent negative effect on the number of digits and the loss of
Gli3R activity is the direct cause of polydactyly in both
Gli32/2 or Shh2/2;Gli32/2 mutants.
The functional relationship of the Zic and Gli zinc finger
proteins, particularly during development, is not understood.
Gli proteins bind a consensus nonamer target DNA sequence
(GLI-BS—Gli-binding site) (25) to which Zic proteins can
also bind (1), although with lower affinity. Zic proteins signifi-
cantly enhance gene expression, most efficiently in the pres-
ence of GLI-BS, but also from promoters without GLI-BS
(26), suggesting they may function as transcriptional coactiva-
tors. Co-expression of Zic and Gli in vitro leads to synergistic
enhancement or mutual suppression of GLI-BS-mediated tran-
scription depending on the cell type (26). Zic and Gli proteins
also physically interact through their zinc finger domains and
regulate each other’s subcellular localization and transcrip-
tional activity (27).
In this paper, we attempted to shed a new light on Zic3
function by studying its modifying effect on Gli3 expression
and activity, as well as Shh expression in the developing
limb buds—a role of Zic3 not investigated previously. To
study the role of Zic3 during development, we previously
generated Zic3 null mice (Zic32/2 and Zic3-/Y) carrying a
targeted deletion of the entire Zic3 gene (28). The Zic3 null
phenotype closely resembles defects seen in patients with
X-linked heterotaxy, which include complex congenital heart
disease, disturbances of laterality, neural tube abnormalities
and vertebral defects. The null phenotype indicates that Zic3
plays an important role in axial midline development and
left–right patterning. To study Zic3 expression pattern, we
generated a novel Zic3 reporter transgenic mouse line,
artificial chromosome), expressing b-galactosidase under the
control of Zic3 regulatory sequences and we demonstrate
that Zic3 is expressed in the distal limb bud mesenchyme in
a pattern spatially and temporally overlapping with Gli3.
To gain a better understanding of Zic3 and Gli3 interactions
in vivo, we investigated whether mice carrying mutations of
both Zic3 and Gli3 would present with features distinct from
the single-gene mutants. We demonstrate that Zic3 loss of
function rescues polydactyly in Gli3+/2 neonates and abro-
gates ectopic Shh expression in anterior limb buds of
Gli32/2 embryos. At the molecular level, loss of function
of Zic3 leads to normalization of Shh expression and Gli3R/
Gli3A ratio in Gli3+/2 limb buds. Finally, we show increases
in Gli3 transcripts in limb buds of Zic3 null embryos. These
results uncover an important role for Zic3 in modulating the
balance between Gli3A and Gli3R in developing limb buds
and in regulating Shh expression level.
Zic3 and Gli3 expression overlap during limb patterning
Zic3 and Gli3 expression in developing embryos were
analyzed using whole-mount in situ hybridization (WISH) at
embryonic day 10.5 (E10.5) and E11.5 (Fig. 1), the critical
window for Shh-mediated A/P patterning. In the whole
embryo, expression of Zic3 and Gli3 overlap and are present
in somites, forebrain, midbrain, hindbrain, spinal cord, limb
buds and the developing eye (Fig. 1). Gli3 is broadly
expressed in the developing limb bud mesenchyme (Fig. 1B
and D) at these stages. Zic3 expression in limb bud distal
mesenchyme (Fig. 1F and H) overlaps with that of Gli3.
Zic3 expression in limb buds was further validated using a
novel Zic3 reporter transgenic mouse line, Zic3-LacZ-BAC
(See Materials and Methods and Supplementary Material,
Fig. S1). Eight independent Zic3-LacZ-BAC lines were ana-
lyzed and the limbs of three representative lines are shown
in Supplementary Material, Figure S2. These analyses
confirm that Zic3 is indeed expressed in limb bud distal
Zic3 synergizes with Gli3 to activate transcription
We investigated the ability of Zic3 to modulate Gli transcrip-
tional activity, using in vitro reporter assays. Zic3 and Gli
expression constructs were used in transactivation assays
with the 12Gli-RE-TKO-luciferase (12Gli-luc) reporter, con-
taining multimerized GLI-BSs (Fig. 2). Zic3 activates the re-
porter 7-fold, consistent with its known ability to bind the
canonical Gli-binding sequence (26). Transfection with Gli1
results in ?40-fold activation of transcription. Co-transfection
of Zic3 and Gli1 results in no significant increase in transcrip-
tional activation (P ¼ 0.374) compared with Gli1 alone. In
contrast, Gli3 acts as a weak repressor of the reporter.
Co-transfection of Zic3 and Gli3 results in nearly 30-fold ac-
tivation of transcription (Fig. 2A). This effect is significantly
Human Molecular Genetics, 2012, Vol. 21, No. 8 1889
greater than with Zic3 alone (P , 0.0001), indicating a syner-
gistic effect of the two proteins.
In vivo, one downstream target of Gli-mediated Shh signal
transduction is the receptor Patched. The Patched promoter
contains a single GLI-BS. We utilized a Patched promoter-
luciferase construct to test the ability of Gli3 and Zic3 to
synergize in transcriptional activation (Fig. 2B). Qualitatively
similar results were obtained. Transfected independently, Zic3
results in a 2-fold activation of the reporter, whereas Gli3
is responsible for a 5-fold repression of transcription.
Co-transfection results in a 4-fold increase in activation
(P , 0.0001). These results suggest that co-transfection with
Zic3 transforms Gli3 from a functional repressor to an
activator in vitro.
Zic3 loss of function rescues Gli31/2 polydactyly
To determine Zic3 and Gli3 interactions in vivo, we asked
whether mice carrying mutations of both Zic3 and Gli3
present with a phenotype distinct from single-gene mutants.
The polydactylous limb phenotype of Gli3 mutants both in
mice and in humans has been well documented (5,23,29).
Mice heterozygous and homozygous for the XtJdeletion in
Gli3 exhibit varying numbers of preaxial extra digits. We
analyzed the fore- and hindlimb phenotypes of WT, Zic3
null, Gli3+/2, Gli32/2, Zic3 null;Gli3+/2 and Zic3
null;Gli32/2 neonates or E19 embryos (Table 1, Fig. 3 and
Supplementary Material, Fig. S4). Limb abnormalities were
not observed in Zic3 null mice (Fig. 3D–F). Gli3+/2 limbs
show duplication of digit 1 (arrows in Fig. 3G–I). This pheno-
type is rescued by Zic3 loss of function (Fig. 3J–L)
(P ¼ 0.025 for number of digits in Zic3 null; Gli3+/2
versus Gli3+/2 animals). In some cases, a subtle phenotype
of widened first digit was identified in Zic3 null;Gli3 +/2
limbs, but no polydactyly was seen (Table 1 and Supplemen-
tary Material, Fig. S4). Gli32/2 limbs exhibit severe poly-
dactyly combined with the loss of normal A/P digit identity
(Fig. 3M–O). Zic3 loss of function does not rescue the
Figure 2. Zic3 synergizes with Gli3 to activate transcription in vitro. HeLa
cells were co-transfected with Zic3, Gli1 or Gli3 expression constructs and
Gli12luc (A) or Ptch-luc (B). Histograms show fold activation compared
with the reporter alone. Results represent average from three separate experi-
ments + standard error.
Figure 1. Zic3 and Gli3 expression overlap during embryonic stages crucial for A/P limb patterning and digit specification (E10.5–E11.5), as shown by WISH.
(A–D) Gli3 expression in whole embryos and limb buds. (E–H) Zic3 expression at stages indicated on the bottom of the figure. (B, D, F and H) High mag-
nification views of forelimbs shown in (A), (C), (E) and (G), respectively. In whole embryos, expression of both Zic3 and Gli3 is detectable in the brain, spinal
cord, somites and limb bud mesenchyme. Anterior is on top and posterior on bottom of each limb panel.
Table 1. Polydactyly phenotype information
Genotype Number of
Broad digit 1 in
the absence of
ectopic digit (%)
a50% of Gli3 +/2 embryos had broad digit 1 on the forelimb (see
Supplementary Material, Fig. S4), but 100% had ectopic digits on the hindlimb.
1890Human Molecular Genetics, 2012, Vol. 21, No. 8
Gli32/2 phenotype (Fig. 3P–R) (P ¼ 0.102 for number of
digits in Zic3 null;Gli32/2 versus Gli32/2 animals),
although a trend toward subtle partial rescue cannot be
Zic3 loss of function abrogates ectopic Shh expression
in limb buds
Proper location and level of Shh expression are requirements
for normal limb patterning. We used WISH to interrogate
the location of expression, and real-time PCR to investigate
the expression level. With regard to location of expression,
Gli32/2 embryos are known to express Shh ectopically in
the anterior mesoderm of developing limb buds (11). Shh
expression pattern is normal in Zic3 null limb buds (Fig. 4C
and D). In Gli32/2 embryos, ectopic Shh expression is
present in anterior forelimbs and hindlimbs (Fig. 4E and F,
arrows). In Zic3 null;Gli32/2 embryos, this ectopic Shh
expression is abrogated (Fig. 4G and H).
Zic3 loss of function rescues Shh overexpression
in posterior limb buds of Gli31/2 embryos
Shh is the major regulator of Gli3 processing. We next asked
whether the level of Shh expression is affected by Gli3 and
Zic3 loss of function. We examined the level of Shh transcript
in posterior hindlimbs of E11.5 WT (n ¼ 5), Zic3 null (n ¼ 3),
Gli3+/2 (n ¼ 4), Gli32/2 (n ¼ 3), Zic3 null;Gli3+/2
(n ¼ 3) and Zic3 null;Gli32/2 (n ¼ 3) embryos by quantita-
tive real-time PCR. We observed upregulation of Shh expres-
sion in posterior limb buds of Gli3+/2 and Gli32/2 E11.5
embryos (11-fold and 6-fold, respectively), as well as 2.7-fold
upregulation in Zic3 null posterior limb buds (Fig. 5). Our
finding that Shh expression level in Gli3+/2 limb buds was
higher than in Gli32/2 was unexpected, since Gli3 is
thought to downregulate Shh expression (16). In Gli3+/2
and Gli32/2 embryos lacking Zic3 function, Shh expression
in posterior limb buds is rescued to the level comparable with
WT (Fig. 5). Therefore, anatomic analysis shows rescue or
partial rescue of polydactyly (Fig. 3), and investigation of
Shh location (Fig. 4) and expression levels (Fig. 5) demon-
strate relative rescue of molecular abnormalities.
Figure 3. Loss of function of Zic3 rescues the Gli3 haploinsufficient limb
phenotype. Dorsal (A, D, G, J, M and P) and ventral (B, E, H, K, N and
Q) views and skeletal preparations (C, F, I, L, O and R) of E18-P1 mouse
forelimbs of indicated genotypes are shown. Note the presence of extra digit
in Gli3+/2 limbs (arrows in G–I) versus normal number of digits in Zic3
null;Gli3+/2 limbs (J–L). Severe polydactyly and abnormal A/P patterning
of Gli32/2 (M–O) as well as Zic3 null;Gli32/2 (P–R) limbs is seen.
Digit 1 is identified by the presence of two phalanges, whereas digits 2–5
have three. Alizarin red staining of ossification centers indicates that the
Zic3 null;Gli3+/2 forelimb shown in (L) is E17, whereas WT ossification
in (C) is indicative of E19 as per Patton and Kaufman staging (45). MC1,
first metacarpal; P1, proximal phalanx; P2, distal phalanx.
Figure 4. Zic3 loss of function abrogates ectopic Shh expression in developing
Gli32/2 limb buds. Shh expression in forelimbs and hindlimbs of E10.5 WT
(A and B), Zic3 null (C and D), Gli32/2 (E and F) and Zic3 null;Gli32/2
(G and H) embryos is shown. The ectopic Shh expression seen in anterior limb
buds of Gli32/2 embryos (E–F, arrows) is abrogated in Zic3 null;Gli32/2
limb buds (G–H).
Human Molecular Genetics, 2012, Vol. 21, No. 8 1891
Zic3 loss of function increases Gli3 expression level
in the limb buds
We examined Gli3 expression level in the developing limb
buds of WT (n ¼ 4) and Zic3 null (n ¼ 8) embryos by real-
time PCR. In Zic3 null limb buds, Gli3 expression is signifi-
cantly increased compared with WT (1.33-fold to 1.74-fold,
P ¼ 0.0014)
Zic3 loss of function restores normal Gli3A and Gli3R
levels in Gli31/2 embryos
To determine the effect of Zic3 loss of function on Gli3
protein processing in Gli3+/2 limb buds, western blot ana-
lysis was performed using a Gli3-specific antibody (gift
from B. Wang) on protein extracts from dissected limb buds
ofE11.5 WT(n ¼ 3),
null;Gli3+/2 (n ¼ 4) embryos. Anti-Gli3 antibody recog-
nizes full-length Gli3 (Gli3A; 190 kDa), Gli3 repressor
(Gli3R; 83 kDa) and truncated mutant Gli3 protein (Xt-J;
66 kDa) (Fig. 6A). Densitometric analysis of Gli3 bands nor-
(GAPDH) shows the expected graded distribution of Gli3R
along the A/P axis in limb buds of all genotypes, with a
higher level of Gli3R in the anterior domains (Fig. 6B).
Gli3R/Gli3A ratios are increased in both anterior and posterior
halves of Gli3+/2 limb buds compared with WT, with the
P-value showing a trend toward significance in the posterior
half (P ¼ 0.067) (Fig. 6C). Importantly, a normal Gli3R/
Gli3A ratio is restored in Zic3 null;Gli3+/2 limb buds, and
Gli3R/Gli3A ratios in WT compared with Zic3 null;Gli3+/2
(n ¼ 4)and
(P ¼ 0.146 forposterior
The ZIC transcription factor family plays important roles in
human development. Mutations in ZIC3 cause X-linked het-
erotaxy (HTX1 MIM 306955) and isolated congenital heart
defects as a result of its role in left–right patterning and inter-
action with the Nodal signal transduction pathway (28,30,31).
Previous analyses of Zic and Gli function have suggested that
Zic proteins may act as transcriptional co-activators in con-
junction with Gli proteins (26). Together with the expression
patterns of Zic1-3 genes (2) overlapping with regions of Shh
pathway activity, these findings suggest a role for Zic tran-
scription factors in the regulation of, or response to, Shh sig-
naling. In this study, we present evidence that Zic3 alters
Gli3 function in vitro and modulates its phenotypic effects
on limb development in vivo, thereby identifying a novel
genetic regulatory interaction.
Figure 6. Increased Gli3R/A ratio in Gli3+/2 limb buds is rescued in the
absence of Zic3. (A) Representative western blot of protein extracts from an-
terior (A) and posterior (P) limb bud halves of E11.5 embryos. (B) Levels of
Gli3A and Gli3R proteins in anterior and posterior forelimb halves (FLA and
FLP, respectively) in arbitrary units normalized to GAPDH with standard error
bars. (C) Relative Gli3R/Gli3A ratios in anterior and posterior forelimb
halves, with standard error bars. Ratios were calculated from protein levels
shown in (B). Quantitation values are means from three WT embryos, four
Gli3+/2 embryos and four Zic3 null;Gli3+/2 embryos, analyzed independ-
Figure 5. Shh overexpression in E11.5 posterior limb buds of Gli3 mutant
embryos is rescued in the absence of Zic3, as determined by real-time PCR.
Expression of Shh is significantly increased in Gli3+/2 and Gli32/2 poster-
ior limb buds compared with WT (P , 0.0001, asterisks). The numbers of
embryos studied were WT (n ¼ 5), Zic3 null (n ¼ 3), Gli3+/2 (n ¼ 4),
Gli32/2 (n ¼ 3), Zic3 null; Gli3+/2 (n ¼ 3) and Zic3 null; Gli32/2
(n ¼ 3). Shh expression was normalized to Gapdh.
1892Human Molecular Genetics, 2012, Vol. 21, No. 8
Zic3 is involved in regulating Gli3 expression (Supplemen-
tary Material, Fig. S3) and thus the rescue of phenotypic and
molecular abnormalities in Zic3 null;Gli3+/2 embryos may
stem from compensation for Gli3 haploinsufficiency by Zic3-
mediated upregulation of Gli3 transcript levels. Gli3 western
results are also consistent with this model, showing normaliza-
tion of Gli3A and Gli3R levels in Zic3 null;Gli3+/2 limb
buds. We find that Zic3 loss of function normalizes many
changes seen in the developing Gli3+/2 limbs (Figs 3, 5
and 6), as well as in Gli32/2 limb buds (Fig. 4). Figure 7
summarizes the expression analyses relevant for Zic3-
mediated rescue of preaxial polydactyly. In Gli32/2
embryos, the polydactylous phenotype persists despite the
fact that Zic3 loss of function rescues ectopic expression of
Shh in anterior limb buds and Shh overexpression in the
ZPA. These findings imply that the presence of functional
Gli3 protein is necessary for phenotypic rescue and are in
agreement with Litingtung et al. (10), who argue that Shh
has no effect on skeletal patterning in the absence of Gli3.
The increase of Gli3R/Gli3A ratio observed in Gli3+/2
neonates with polydactyly was unexpected since this pheno-
type has previously been associated with a Gli3R/Gli3A
ratio shifted in favor of Gli3A (32). Our results suggest that
the decrease in the absolute Gli3R level is more important
in the development of preaxial polydactyly than any change
in the Gli3R/Gli3A ratio.
Shh mRNA expression is also affected in limb buds of Zic3,
Gli3 and Zic3/Gli3 mutants. Previous work proposes an auto-
regulation of Shh expression in the ZPA, via modulation of
ZPA domain expansion, cell fate and cell death (33). Further-
more, only a fraction of ZPA cells capable of expressing Shh
actively produce Shh pre-mRNA at any given time, implying
that Shh expression is a dynamic process in which the tran-
script level may fluctuate within each ZPA cell (34). Thus,
the upregulation of Shh expression seen in Gli3 mutant
embryos could result from increased Shh transcript levels
within cells or from an expanded ZPA domain. Normalization
by Zic3 loss of function reflects an important role in modulat-
ing Shh expression level. We expected that complete loss of
Gli3R in Gli32/2 embryos would further upregulate Shh ex-
pression compared with Gli3+/2 embryos, but we observed
the opposite (Fig. 5). These results suggest that qualitative
upregulation is more important than the specific level, which
may depend on transient fluctuations of Shh expression. In
addition to transcript level and cell number, the length of
Shh exposure is a third important variable. Specification of
the anterior digits depends upon differential concentrations
of Shh, whereas the length of time of exposure to Shh is crit-
ical for differential patterning of the most posterior digits (35).
Mutations of Gli family members could therefore influence
digit patterning by altering the number of Shh-expressing
cells in the ZPA, changing the level of Shh expression
within cells, or affecting the duration of Shh expression.
Finally, it should be noted that the alterations identified are
specific to mRNA and protein levels require further assess-
ment. Gli proteins function within a regulatory network to co-
ordinate limb patterning. Although the mechanistic basis of
Gli3 function in the limb is not completely understood, im-
portant Gli3 interactions with other transcription factors in
the limb mesenchyme with effects on morphogenesis and
pattern, such as Plzf (promyeloytic leukemia zinc finger)
(36) and Hoxd12 (Hoxd—homeobox D) (37), have been iden-
tified. With this new identification of a Gli3–Zic3 interaction
within the limb, the relationship of Zic3 proteins to this mo-
lecular circuitry requires investigation. For example, both
Zic3 and Hoxd12 physically interact with Gli3 via its N-
terminal zinc finger domains, suggesting a potential for com-
petitive interactions that will require future study. There are
a number of similarities between Hoxd12–Gli3 and Zic3–
Gli3 interactions. Hoxd12 interacts physically and genetically
with Gli3 during limb development, and can convert the Gli3
repressor into an activator of Shh target genes. Gli3 and
Hoxd12 expression overlap in the developing limb bud, as
do Gli3 and Zic3. Both Hoxd12 and Zic3 affect the phenotype
of Gli3+/2 mutants but not Gli32/2 mutants. The presence
of functional Gli3 protein is required for Hoxd12 to exert
effects on digit morphology and on Shh expression. Likewise,
Zic3 loss of function rescues the Gli3 mutant phenotype only
in the presence of Gli3 protein (in Gli3+/2). Finally, genetic
redundancy within the limb buds may be important for both
Hoxd genes (Hoxd10-13) and Zic genes. Each of the five
murine Zic genes contains five C2H2 zinc fingers. Based on
homology within the zinc finger domain, Zic1, 2 and 3 are
more similar to each other and appear to form a subfamily
(38). The expression studies document that Zic2 and Zic3
overlap in the developing limb buds at E10.5–E11.5,
whereas Zic1 is not expressed in limb buds at that time (2).
The expression overlap and the structural similarities of Zic2
and Zic3 indicate the potential for functional compensation
in the developing limb buds, a hypothesis which requires add-
In conclusion, this study identifies Zic3 as an important
potential genetic modifier of the polydactyly phenotype and
shows that Zic3 deficiency rescues preaxial polydactyly by
Figure 7. Model of preaxial polydactyly rescue by Zic3 loss of function.
Levels of Shh, Zic3 and Gli3R are represented by dots. In Gli3+/2 limb
buds, Shh expression is upregulated, resulting in lower than WT levels of
Gli3R in anterior limb bud and polydactyly (compare A and B). In Zic3
null;Gli3+/2 limb buds, Shh expression and Gli3R normalize, resulting in
rescue of polydactyly (C). In the absence of Gli3 in Gli32/2 and Zic3
null;Gli32/2 limb buds, severe polydactyly is observed (D and E). Shh
levels are increased in Gli32/2, whereas they normalize in Zic3
null;Gli32/2 limb buds. However, the polydactylous phenotype cannot be
rescued in the absence of functional Gli3 protein.
Human Molecular Genetics, 2012, Vol. 21, No. 81893
increasing Gli3R (Fig. 7). In addition, these results identify a
potential function for Zic3 in regulating Shh pathway in vivo.
We propose that Zic3 plays a significant role in the network of
transcription factors regulating limb development by interact-
ing with Gli3 to regulate transcription, to fine-tune Gli3 and
Shh expression levels and to maintain the proper balance
between Gli3A and Gli3R in developing limb buds.
MATERIAL AND METHODS
Mouse embryos of E10.5, E11.5, E19 and P1 (postnatal day 1)
were collected for gene expression or phenotype analyses. The
Zic3 null mice have been described previously (28). These
mice are maintained on a mixed 129;C57BL/6 background
due to lethality on an inbred background. The Zic3-LacZ-BAC
transgenic mice were generated using a pBACe3.6 vector car-
rying a modified 200 kb insert of mouse genomic DNA with
Zic3 gene located at the center of the transgene (Supplemen-
tary Material, Fig. S1). The original BAC clone was purchased
from BAC PAC resources. Exon 1 of Zic3 was replaced with
b-galactosidase coding sequence at the translational start site,
using an Escherichia coli recombination system (39). Trans-
genic founders were generated by the University of Michigan
Transgenic Animal Model Core and eight independent trans-
genic lines were established on an FVB/N background.
Mice heterozygous for the Extra toes-J mutation (Gli3Xt-J)
were purchased from The Jackson Laboratory. Zic3 null and
Gli3Xt-Jheterozygous mice were intercrossed for more than
six generations to obtain double-heterozygous animals on a
mixed background (Zic3+/2;Gli3+/2), and further inter-
crosses were performed to obtain Zic3 null;Gli32/2 and
Zic3 null;Gli3+/2 embryos and neonates. All experiments
used wild-type littermate controls.
All mice were housed in the Association for Assessment
and Accreditation of Laboratory Animal Care (AALAC)-
accredited Cincinnati Children’s Hospital Research Founda-
tion Animal Facility and experiments were approved by the
Institutional Animal Care and Use Committee.
WISH, b-galactosidase staining and skeleton preparations
WISH was performed as described previously (28). Probes
were labeled using a DIG RNA Labeling Kit (Roche
Applied Science). A Zic3 probe spanning 471 bp in the 3′
region of the gene was generated by PCR using forward
reverse primer 5′-AGAAGCACTTTAACCATGAG-3′. The
Shh probe was described elsewhere (40), as was the Gli3
probe (41). Staining for b-galactosidase was performed on
Zic3-LacZ-BAC embryos using standard protocols (28). Ske-
letons of E19 embryos and neonates were prepared according
to published protocols (42).
HeLa cells were transfected using Lipofectamine Plus
Co-transfection experiments were performed with 200 ng
each of the reporter and 100 ng each of the expression con-
structs and with 20 ng of pRL-TK Renilla luciferase
(Promega) as an internal standard. Total DNA per transfection
was kept constant by adding empty vectors. Reporters
included 12Gli-luc, a construct generated by ligating 12 multi-
merized oligonucleotides corresponding to Gli consensus-
binding sequence (43), and Patched-luciferase reporter with
one Gli3-binding sequence in the promoter region (44). The
Zic3, Gli1 and Gli3 expression constructs have been described
previously (8,26). Luciferase activities were measured using
the Dual Luciferase Reporter Assay System (Promega) 48 h
after transfection. The relative fold activation is presented as
the ratio of the normalized value of reporter and expression
construct to reporter alone. All results represent three
independent transfection experiments, compared for statistical
significance by Student’s t-test.
Quantitative real-time PCR
Limb buds of E11.5 embryos were collected in RNAlaterw
(Ambion). We harvested posterior autopod halves for Shh ex-
pression analysis and whole-limb buds for Gli3 expression
analysis. RNA extractions were performed using Totally
RNATMKit (Ambion). cDNA was generated with High-
Capacity cDNA Reverse Transcription Kit (Applied Biosys-
tems). Real-time PCR was performed using ABI PRISMw
7000 Sequence Detection System (Applied Biosystems) and
Power SYBRwGreen PCR Master Mix (Applied Biosystems),
with intron-spanning primer pairs (see Supplementary Mater-
ial for sequences). Six independent reactions were run for
each biological sample. Shh and Gli3 gene expression results
were normalized to Gapdh. Histograms represent relative
expression + standard error.
Quantification of Gli3 protein by western blot
Anterior and posterior E11.5 autopod halves (Fig. 6A) were
lysed in radioimmunoprecipitation assay buffer containing
complete protease inhibitor cocktail (Roche). Samples in 1×
Laemmli buffer with 5% b-ME were boiled for 5 min and sub-
jected to electrophoresis through 8% SDS–PAGE under redu-
cing conditions and transferred onto a polyvinylidene fluoride
membrane (Millipore). Gli3 protein was detected using a
rabbit anti-Gli3 antibody (1:250; gift from B. Wang) (15)
and goat anti-rabbit horseradish peroxidase-conjugated sec-
ondary antibody (1:5000; Santa Cruz Biotechnology) followed
by the ECL Plus detection system (Amersham). Anti-GAPDH
antibody (1:2500; Abcam) was used as a loading control.
Bands were quantitated from exposed films. Images were digi-
tized using CCD technology-based Molecular Imager GelDoc
XR+ system (BioRad). Densitometry was performed using
the ImageQuant software (Molecular Dynamics). Histogram
peak background correction was applied. Levels of Gli3A
and Gli3R were normalized to GAPDH. Results were
compared for statistical significance by Student’s t-test.
Supplementary Material is available at HMG online.
1894 Human Molecular Genetics, 2012, Vol. 21, No. 8
We thank Jennifer Purnell for expert technical assistance. We
thank Dr Baolin Wang for anti-Gli3 antibody.
Conflict of Interest statement. The authors have no conflicts to
This work was funded by RO1 HL088639 from the National
Institutes of Health (S.M.W.).
1. Aruga, J., Yokota, N., Hashimoto, M., Furuichi, T., Fukuda, M. and
Mikoshiba, K. (1994) A novel zinc finger protein, zic, is involved in
neurogenesis, especially in the cell lineage of cerebellar granule cells.
J. Neurochem., 63, 1880–1890.
2. Nagai, T., Aruga, J., Takada, S., Gunther, T., Sporle, R., Schughart, K.
and Mikoshiba, K. (1997) The expression of the mouse Zic1, Zic2, and
Zic3 gene suggests an essential role for Zic genes in body pattern
formation. Dev. Biol., 182, 299–313.
3. Brown, S.A., Warburton, D., Brown, L.Y., Yu, C.Y., Roeder, E.R.,
Stengel-Rutkowski, S., Hennekam, R.C. and Muenke, M. (1998)
Holoprosencephaly due to mutations in ZIC2, a homologue of Drosophila
odd-paired. Nat. Genet., 20, 180–183.
4. Gebbia, M., Ferrero, G.B., Pilia, G., Bassi, M.T., Aylsworth, A.,
Penman-Splitt, M., Bird, L.M., Bamforth, J.S., Burn, J., Schlessinger, D.
et al. (1997) X-linked situs abnormalities result from mutations in ZIC3.
Nat. Genet., 17, 305–308.
5. Vortkamp, A., Gessler, M. and Grzeschik, K.H. (1991) GLI3 zinc-finger
gene interrupted by translocations in Greig syndrome families. Nature,
6. Grinberg, I. and Millen, K.J. (2005) The ZIC gene family in development
and disease. Clin. Genet., 67, 290–296.
7. Grinberg, I., Northrup, H., Ardinger, H., Prasad, C., Dobyns, W.B. and
Millen, K.J. (2004) Heterozygous deletion of the linked genes ZIC1 and
ZIC4 is involved in Dandy-Walker malformation. Nat. Genet., 36,
8. Ware, S.M., Peng, J., Zhu, L., Fernbach, S., Colicos, S., Casey, B.,
Towbin, J. and Belmont, J.W. (2004) Identification and functional
analysis of ZIC3 mutations in heterotaxy and related congenital heart
defects. Am. J. Hum. Genet., 74, 93–105.
9. Ahn, S. and Joyner, A.L. (2004) Dynamic changes in the response of cells
to positive hedgehog signaling during mouse limb patterning. Cell, 118,
10. Litingtung, Y., Dahn, R.D., Li, Y., Fallon, J.F. and Chiang, C. (2002) Shh
and Gli3 are dispensable for limb skeleton formation but regulate digit
number and identity. Nature, 418, 979–983.
11. Masuya, H., Sagai, T., Wakana, S., Moriwaki, K. and Shiroishi, T. (1995)
A duplicated zone of polarizing activity in polydactylous mouse mutants.
Genes Dev., 9, 1645–1653.
12. Dai, P., Akimaru, H., Tanaka, Y., Maekawa, T., Nakafuku, M. and Ishii,
S. (1999) Sonic Hedgehog-induced activation of the Gli1 promoter is
mediated by GLI3. J. Biol. Chem., 274, 8143–8152.
13. te Welscher, P., Zuniga, A., Kuijper, S., Drenth, T., Goedemans, H.J.,
Meijlink, F. and Zeller, R. (2002) Progression of vertebrate limb
development through SHH-mediated counteraction of GLI3. Science, 298,
14. Zeller, R., Lopez-Rios, J. and Zuniga, A. (2009) Vertebrate limb bud
development: moving towards integrative analysis of organogenesis. Nat.
Rev. Genet., 10, 845–858.
15. Wang, B., Fallon, J.F. and Beachy, P.A. (2000) Hedgehog-regulated
processing of Gli3 produces an anterior/posterior repressor gradient in the
developing vertebrate limb. Cell, 100, 423–434.
16. Buscher, D., Bosse, B., Heymer, J. and Ruther, U. (1997) Evidence for
genetic control of Sonic hedgehog by Gli3 in mouse limb development.
Mech. Dev., 62, 175–182.
17. Bastida, M.F., Delgado, M.D., Wang, B., Fallon, J.F., Fernandez-Teran,
M. and Ros, M.A. (2004) Levels of Gli3 repressor correlate with Bmp4
expression and apoptosis during limb development. Dev. Dyn., 231,
18. Davenport, T.G., Jerome-Majewska, L.A. and Papaioannou, V.E. (2003)
Mammary gland, limb and yolk sac defects in mice lacking Tbx3, the
gene mutated in human ulnar mammary syndrome. Development, 130,
19. Galli, A., Robay, D., Osterwalder, M., Bao, X., Benazet, J.D., Tariq, M.,
Paro, R., Mackem, S. and Zeller, R. (2010) Distinct roles of Hand2 in
initiating polarity and posterior Shh expression during the onset of mouse
limb bud development. PLoS Genet., 6, e1000901.
20. Laufer, E., Nelson, C.E., Johnson, R.L., Morgan, B.A. and Tabin, C.
(1994) Sonic hedgehog and Fgf-4 act through a signaling cascade and
feedback loop to integrate growth and patterning of the developing limb
bud. Cell, 79, 993–1003.
21. Niswander, L., Jeffrey, S., Martin, G.R. and Tickle, C. (1994) A positive
feedback loop coordinates growth and patterning in the vertebrate limb.
Nature, 371, 609–612.
22. Qu, S., Niswender, K.D., Ji, Q., van der Meer, R., Keeney, D., Magnuson,
M.A. and Wisdom, R. (1997) Polydactyly and ectopic ZPA formation in
Alx-4 mutant mice. Development, 124, 3999–4008.
23. Hui, C.C. and Joyner, A.L. (1993) A mouse model of Greig
cephalopolysyndactyly syndrome: the extra-toesJ mutation contains an
intragenic deletion of the Gli3 gene. Nat. Genet., 3, 241–246.
24. Chiang, C., Litingtung, Y., Harris, M.P., Simandl, B.K., Li, Y., Beachy,
P.A. and Fallon, J.F. (2001) Manifestation of the limb prepattern: limb
development in the absence of sonic hedgehog function. Dev. Biol., 236,
25. Kinzler, K.W. and Vogelstein, B. (1990) The GLI gene encodes a nuclear
protein which binds specific sequences in the human genome. Mol. Cell.
Biol., 10, 634–642.
26. Mizugishi, K., Aruga, J., Nakata, K. and Mikoshiba, K. (2001) Molecular
properties of Zic proteins as transcriptional regulators and their
relationship to GLI proteins. J. Biol. Chem., 276, 2180–2188.
27. Koyabu, Y., Nakata, K., Mizugishi, K., Aruga, J. and Mikoshiba, K.
(2001) Physical and functional interactions between Zic and Gli proteins.
J. Biol. Chem., 276, 6889–6892.
28. Purandare, S.M., Ware, S.M., Kwan, K.M., Gebbia, M., Bassi, M.T.,
Deng, J.M., Vogel, H., Behringer, R.R., Belmont, J.W. and Casey, B.
(2002) A complex syndrome of left-right axis, central nervous system and
axial skeleton defects in Zic3 mutant mice. Development, 129,
29. Kang, S., Graham, J.M. Jr., Olney, A.H. and Biesecker, L.G. (1997) GLI3
frameshift mutations cause autosomal dominant Pallister-Hall syndrome.
Nat. Genet., 15, 266–268.
30. Ware, S.M., Harutyunyan, K.G. and Belmont, J.W. (2006) Zic3 is critical
for early embryonic patterning during gastrulation. Dev. Dyn., 235,
31. Ware, S.M., Harutyunyan, K.G. and Belmont, J.W. (2006) Heart defects
in X-linked heterotaxy: evidence for a genetic interaction of Zic3 with the
nodal signaling pathway. Dev. Dyn., 235, 1631–1637.
32. Babbs, C., Furniss, D., Morriss-Kay, G.M. and Wilkie, A.O. (2008)
Polydactyly in the mouse mutant Doublefoot involves altered Gli3
processing and is caused by a large deletion in cis to Indian hedgehog.
Mech. Dev., 125, 517–526.
33. Sanz-Ezquerro, J.J. and Tickle, C. (2000) Autoregulation of Shh
expression and Shh induction of cell death suggest a mechanism for
modulating polarising activity during chick limb development.
Development, 127, 4811–4823.
34. Amano, T., Sagai, T., Tanabe, H., Mizushina, Y., Nakazawa, H. and
Shiroishi, T. (2009) Chromosomal dynamics at the Shh locus: limb
bud-specific differential regulation of competence and active
transcription. Dev. Cell, 16, 47–57.
35. Harfe, B.D., Scherz, P.J., Nissim, S., Tian, H., McMahon, A.P. and Tabin,
C.J. (2004) Evidence for an expansion-based temporal Shh gradient in
specifying vertebrate digit identities. Cell, 118, 517–528.
36. Barna, M., Pandolfi, P.P. and Niswander, L. (2005) Gli3 and Plzf
cooperate in proximal limb patterning at early stages of limb
development. Nature, 436, 277–281.
37. Chen, Y., Knezevic, V., Ervin, V., Hutson, R., Ward, Y. and Mackem, S.
(2004) Direct interaction with Hoxd proteins reverses Gli3-repressor
Human Molecular Genetics, 2012, Vol. 21, No. 81895
function to promote digit formation downstream of Shh. Development, Download full-text
38. Furushima, K., Murata, T., Matsuo, I. and Aizawa, S. (2000) A new
murine zinc finger gene, Opr. Mech. Dev., 98, 161–164.
39. Yu, D., Ellis, H.M., Lee, E.C., Jenkins, N.A., Copeland, N.G. and Court,
D.L. (2000) An efficient recombination system for chromosome
engineering in Escherichia coli. Proc. Natl Acad. Sci. USA, 97,
40. Echelard, Y., Epstein, D.J., St-Jacques, B., Shen, L., Mohler, J.,
McMahon, J.A. and McMahon, A.P. (1993) Sonic hedgehog, a member of
a family of putative signaling molecules, is implicated in the regulation of
CNS polarity. Cell, 75, 1417–1430.
41. Hui, C.C., Slusarski, D., Platt, K.A., Holmgren, R. and Joyner, A.L.
(1994) Expression of three mouse homologs of the Drosophila segment
polarity gene cubitus interruptus, Gli, Gli-2, and Gli-3, in ectoderm- and
mesoderm-derived tissues suggests multiple roles during postimplantation
development. Dev. Biol., 162, 402–413.
42. Kochhar, D.M. (1973) Limb development in mouse embryos. I. Analysis
of teratogenic effects of retinoic acid. Teratology, 7, 289–298.
43. Kogerman, P., Grimm, T., Kogerman, L., Krause, D., Unden, A.B.,
Sandstedt, B., Toftgard, R. and Zaphiropoulos, P.G. (1999) Mammalian
suppressor-of-fused modulates nuclear-cytoplasmic shuttling of Gli-1.
Nat. Cell. Biol., 1, 312–319.
44. Shin, S.H., Kogerman, P., Lindstrom, E., Toftgard, R. and Biesecker, L.G.
(1999) GLI3 mutations in human disorders mimic Drosophila cubitus
interruptus protein functions and localization. Proc. Natl Acad. Sci. USA,
bones, and growth rates of various long bones of the fore and hind limbs of
the prenatal and early postnatal laboratory mouse. J. Anat., 186, 175–185.
1896Human Molecular Genetics, 2012, Vol. 21, No. 8