Smad1/Smad5 signaling in limb ectoderm functions redundantly and is required for interdigital programmed cell death.
ABSTRACT Bone morphogenetic proteins (BMPs) are secreted signals that regulate apical ectodermal ridge (AER) functions and interdigital programmed cell death (PCD) of developing limb. However the identities of the intracellular mediators of these signals are unknown. To investigate the role of Smad proteins in BMP-regulated AER functions in limb development, we inactivated Smad1 and Smad5 selectively in AER and ventral ectoderm of developing limb, using Smad1 or/and Smad5 floxed alleles and an En1(Cre/+) knock-in allele. Single inactivation of either Smad1 or Smad5 did not result in limb abnormalities. However, the Smad1/Smad5 double mutants exhibited syndactyly due to a reduction in interdigital PCD and an increase in interdigital cell proliferation. Cell tracing experiments in the Smad1/Smad5 double mutants showed that ventral ectoderm became thicker and the descendents of ventral En1(Cre/+) expressing ectodermal cells were located at dorsal interdigital regions. At the molecular level, Fgf8 expression was prolonged in the interdigital ectoderm of embryonic day (E) 13 Smad1/Smad5 double mutants, suggesting that the ectopic Fgf8 expression may serve as a survival signal for interdigital epithelial and mesenchymal cells. Our result suggests that Smad1 and Smad5 are required and function redundantly as intracellular mediators for BMP signaling in the AER and ventral ectoderm. Smad1/Smad5 signaling in the AER and ventral ectoderm regulates interdigital tissue regression of developing limb. Our mutants with defects in interdigital PCD could also serve as a valuable model for investigation of PCD regulation machinery.
- SourceAvailable from: Carlos Ignacio Lorda-Diez[Show abstract] [Hide abstract]
ABSTRACT: Interdigital cell death is a physiological regression process responsible for sculpturing the digits in the embryonic vertebrate limb. Changes in the intensity of this degenerative process account for the different patterns of interdigital webbing among vertebrate species. Here, we show that Reelin is present in the extracellular matrix of the interdigital mesoderm of chick and mouse embryos during the developmental stages of digit formation. Reelin is a large extracellular glycoprotein which has important functions in the developing nervous system, including neuronal survival; however, the significance of Reelin in other systems has received very little attention. We show that reelin expression becomes intensely downregulated in both the chick and mouse interdigits preceding the establishment of the areas of interdigital cell death. Furthermore, fibroblast growth factors, which are cell survival signals for the interdigital mesoderm, intensely upregulated reelin expression, while BMPs, which are proapototic signals, downregulate its expression in the interdigit. Gene silencing experiments of reelin gene or its intracellular effector Dab-1 confirmed the implication of Reelin signaling as a survival factor for the limb undifferentiated mesoderm. We found that Reelin activates canonical survival pathways in the limb mesoderm involving protein kinase B and focal adhesion kinase. Our findings support that Reelin plays a role in interdigital cell death, and suggests that anoikis (apoptosis secondary to loss of cell adhesion) may be involved in this process.Cell Death & Disease 09/2013; 4:e800. · 5.18 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: The analysis of vertebrate limb bud development provides insight of general relevance into the signaling networks that underlie the controlled proliferative expansion of large populations of mesenchymal progenitors, cell fate determination and initiation of differentiation. In particular, extensive genetic analysis of mouse and experimental manipulation of chicken limb bud development has revealed the self-regulatory feedback signaling systems that interlink the main morphoregulatory signaling pathways including BMPs and their antagonists. It this review, we showcase the key role of BMPs and their antagonists during limb bud development. This review provides an understanding of the key morphoregulatory interactions that underlie the highly dynamic changes in BMP activity and signal transduction as limb bud development progresses from initiation and setting-up the signaling centres to determination and formation of the chondrogenic primordia for the limb skeletal elements.Seminars in Cell and Developmental Biology 04/2014; · 6.20 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Bone morphogenetic proteins (BMPs) are multifunctional growth factors that play crucial roles during embryonic development and cell fate determination. Nuclear transduction of BMP signals requires the receptor type Smad proteins, Smad1, Smad5 and Smad9. However, how these Smad proteins cooperate in vivo to regulate various developmental processes is largely unknown. In zebrafish, it was widely believed that the maternally expressed smad5 is essential for dorso-ventral (DV) patterning, and the zygotically transcribed smad1 is not required for normal DV axis establishment. In the present study, we have identified zygotically expressed smad9, which cooperates with smad1 downstream of smad5, to mediate zebrafish early DV patterning in a functional redundant manner. Although knockdown of smad1 or smad9 alone does not lead to visible dorsalization, double knockdown strongly dorsalizes zebrafish embryos, which cannot be efficiently rescued by smad5 overexpression. While the dorsalization induced by smad5 knockdown can be fully rescued by overexpression of smad1 or smad9. We have further revealed that the transcription initiation of smad1 or smad9 is repressed by each other and they are direct transcriptional targets of Smad5, and smad9 is required for myelopoiesis as smad1. In conclusion, our study uncovers that smad1 and smad9 act redundantly to each other downstream of smad5 to mediate ventral specification and to regulate embryonic myelopoiesis.Journal of Biological Chemistry 01/2014; · 4.60 Impact Factor
Smad1/Smad5 signaling in limb ectoderm functions redundantly and is required for
interdigital programmed cell death
Yuk Lau Wonga, Richard R. Behringerc, Kin Ming Kwana,b,⁎
aSchool of Life Sciences, The Chinese University of Hong Kong, Hong Kong, PR China
bCenter for Cell and Developmental Biology, The Chinese University of Hong Kong, Hong Kong, PR China
cDepartment of Genetics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA
a b s t r a c ta r t i c l ei n f o
Received for publication 6 June 2011
Revised 17 December 2011
Accepted 22 December 2011
Available online 3 January 2012
Programmed cell death
Bone morphogenetic proteins (BMPs) are secreted signals that regulate apical ectodermal ridge (AER) func-
tions and interdigital programmed cell death (PCD) of developing limb. However the identities of the intra-
cellular mediators of these signals are unknown. To investigate the role of Smad proteins in BMP-regulated
AER functions in limb development, we inactivated Smad1 and Smad5 selectively in AER and ventral ecto-
derm of developing limb, using Smad1 or/and Smad5 floxed alleles and an En1Cre/+knock-in allele. Single in-
activation of either Smad1 or Smad5 did not result in limb abnormalities. However, the Smad1/Smad5 double
mutants exhibited syndactyly due to a reduction in interdigital PCD and an increase in interdigital cell prolif-
eration. Cell tracing experiments in the Smad1/Smad5 double mutants showed that ventral ectoderm became
thicker and the descendents of ventral En1Cre/+expressing ectodermal cells were located at dorsal interdigital
regions. At the molecular level, Fgf8 expression was prolonged in the interdigital ectoderm of embryonic day
(E) 13 Smad1/Smad5 double mutants, suggesting that the ectopic Fgf8 expression may serve as a survival sig-
nal for interdigital epithelial and mesenchymal cells. Our result suggests that Smad1 and Smad5 are required
and function redundantly as intracellular mediators for BMP signaling in the AER and ventral ectoderm.
Smad1/Smad5 signaling in the AER and ventral ectoderm regulates interdigital tissue regression of develop-
ing limb. Our mutants with defects in interdigital PCD could also serve as a valuable model for investigation of
PCD regulation machinery.
© 2012 Elsevier Inc. All rights reserved.
The developing limb has easily identified discrete pattern and
structure formation and is conveniently accessible for experimental
manipulation. It has therefore long been an excellent model for inves-
tigating how different signals interact to control different cellular ac-
tivities during pattern formation (Towers and Tickle, 2009). The
mouse limbs develop from the embryonic limb buds, coming out
from either side of body wall at the appropriate positions along
the body axis. Early limb bud is made up of a mass of undifferen-
tiated mesenchymal cells covered with surface ectoderm. The
growth and patterning of the limb bud along the proximal–distal,
anterior–posterior and dorsal–ventral axes are coordinated by re-
ciprocal interactions between specialized regions known as apical
ectodermal ridge (AER), zone of polarizing activity and the non-
ridge ectoderm of the limb bud respectively (Capdevila and
Izpisua Belmonte, 2001; Martin, 1990). AER is a stratified columnar
epithelial structure located at the apex of the forelimb and hindlimb
bud starting from embryonic day (E) 10.5 and 11.5 respectively. Itis
a source of secreted factors required for limb outgrowth (Dudley et al.,
2002). Alternatively, this structure is also involved in regulating pro-
grammed cell death (PCD) in the developing limb (Maatouk et al.,
2009; Pajni-Underwood et al., 2007). PCD is an important genetically
ticellular organisms (Fadeel and Orrenius, 2005; Jacobson et al., 1997).
PCD is mainly observed in the interdigital undifferentiated mesenchy-
mal cells and the AER (Fernandez-Teran et al., 2006; Zuzarte-Luis and
Hurle, 2005). This process has to be controlled precisely so that the de-
veloping limb is sculpted into a particular shape and structure that is
particular to an organism (Zuzarte-Luis and Hurle, 2005). Interdigital
region of developing limb is of particular interest because the fate of
interdigital mesenchymal cells to undergo cell death or not or chondro-
genesis is subjected toprecise genetic controland appropriate signaling
between the ectoderm and underlying mesenchyme (Bandyopadhyay
et al., 2006; Hernández-Martínez and Covarrubias, 2011; Hurle and
Ganan, 1986). Dying cells located in the mesenchyme between the
forming digits of the developing limbs are referred as interdigital PCD
Developmental Biology 363 (2012) 247–257
⁎ Corresponding author at: School of Life Sciences, The Chinese University of Hong
Kong, Shatin, Hong Kong, PR China. Fax: +852 2603 5646.
E-mail address: firstname.lastname@example.org (K.M. Kwan).
0012-1606/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/developmentalbiology
that is responsible in restricting interdigital tissue growth and promot-
ing tissue regression in order to separate digits (Hernández-Martínez
and Covarrubias, 2011; Macias et al., 1997; Montero et al., 2001). De-
fects in patterning and interdigital PCD in the developing limb are cor-
related with various types of congenital limb malformations (Al-
Qattan et al., 2009; Goodman, 2002).
Interdigital PCD in developing limb is tightly controlled under
the effects of different signaling factors like fibroblast growth fac-
tors (FGFs) (Delgado et al., 2008; Montero et al., 2001; Pajni-
Underwood et al., 2007; Salas-Vidal et al., 2001), BMPs (Ganan et al.,
1996; Guha et al., 2002; Macias et al., 1997; Pajni-Underwood et al.,
2007; Salas-Vidal et al.,2001; Zou and Niswander, 1996),transforming
growth factor-beta (TGF-β) (Ganan et al., 1996), Msx-2 (Marazzi
et al., 1997; Salas-Vidal et al., 2001), Wnts (Grotewold and
Ruther, 2002), Shh (Buscher et al., 1997) and retinoic acid (Ali-
Khan and Hales, 2006; Galdones et al., 2006). BMP signaling in
the AER has been shown to regulate interdigital PCD but the iden-
tity of the intracellular mediator remains unclear (Maatouk et al.,
2009; Pajni-Underwood et al., 2007; Zuzarte-Luis and Hurle,
2005). One of the pathways involves receptor-regulated Smad pro-
teins (Zuzarte-Luis and Hurle, 2005; Zuzarte-Luis et al., 2004).
Binding of BMP ligand to membrane-bound serine/threonine kinase
receptors results in receptor phosphorylation and subsequent acti-
vation of receptor-regulated Smad proteins (R-Smads) Smads1/5/
8 by phosphorylation. Activated Smad1, 5 or 8 form complexes
with common partner Smad (CoSmad), Smad4 and translocate
into the nucleus where they regulate the transcription of target
genes (Itoh et al., 2000).
In this study, we investigated the identity of the Smad proteins
involved in BMP-regulated AER functions during limb develop-
ment. Knockout of either Smad1 or Smad5 results in early embryon-
ic lethality, hindering the studies of Smad function in limb
formation (Chang et al., 1999; Lechleider et al., 2001). Therefore,
Cre/loxP approach was used to inactivate Smad1 and/or Smad5 in
the developing limb ectoderm in a tissue-specific manner by the
use of Smad1 and/or Smad5 conditional null (floxed) alleles
(Huang et al., 2002; Umans et al., 2003) and an Engrailed1-Cre-
recombinase knock-in allele (En1Cre/+) (Kimmel et al., 2000).
En1Cre/+first expresses in the ventral ectoderm and the ventral
part of the developing AER of forelimb bud at E9.5 (Kwan et al.,
2004). After E11.5, En1Cre/+expressing lineage is present through-
out the dorsal–ventral extent of the AER (Kimmel et al., 2000). Our
data showed that conditional inactivation of either Smad1 or Smad5
in the limb AER and ventral ectoderm does not result in limb ab-
normalities. However, double inactivation of both Smad1/Smad5
resulted in syndactyly due to a reduction in interdigital PCD and
an increase in interdigital cell proliferation. Our data suggest that
Smad1 and Smad5 act as the intracellular mediator downstream
of the BMP receptor Ia (BmprIa) to transduce BMPs signal in the
AER. The two proteins are required and function redundantly to
regulate AER functions. The Smad1/Smad5 signaling in AER indi-
rectly regulates interdigital PCD and cell proliferation in developing
limb. Furthermore, we have investigated on the possible molecular
mechanism of the Smad1/Smad5 signaling in regulating the inter-
digital tissue regression.
Materials and methods
Mouse strain generation and genotyping
The generation and genotyping of conditional (floxed) alleles of
Smad1 (Smad1f) and Smad5 (Smad5f) have been described previously
(Huang et al., 2002; Pangas et al., 2008; Umans et al., 2003). The
En1Creknock-in allele has been described previously (Kimmel et al.,
2000; Kwan et al., 2004). To generate the AER and ventral ectoderm
conditional Smad1/Smad5 knockout mutant, we crossed En1Cre/+
mice with Smad1f/f; Smad5f/fmice. Their En1Cre/+; Smad1f/+; Smad5f/+
female offspringwere then crossed with Smad1f/f; Smad5f/fmale to pro-
duce En1Cre/+; Smad1f/+; Smad5f/fmales. To generate the En1Cre/+;
Smad1f/f; Smad5f/fmutant embryos for analysis, Smad1f/f; Smad5f/ffe-
males were then crossed with En1Cre/+; Smad1f/+; Smad5f/fmales.
Their littermates, En1Cre/+; Smad1f/+; Smad5f/f, which developed
normally, were used as controls. Since the En1Creis a knock-in allele,
we used En1Cre/+; Smad1f/+; Smad5f/fmice as the controls to elimi-
nate the possible effect of the En1 heterozygous null genotype on
the phenotypic analysis of the double Smad1/Smad5 conditional mu-
tants. All mouse strains were on a mixed genetic background. All an-
imal procedures were conducted with the approval of the Animal
Experimentation Ethics Committee of The Chinese University of
Immunofluorescence, cell death and cell proliferation analysis, section
and whole mount RNA in situ hybridization
Embryos or embryonic limbs were isolated, fixed in 4% parafor-
maldehyde (PFA) at 4 °C overnight, paraffin-embedded and sec-
tioned using standard techniques. Antibodies against Smad1
(Invitrogen) at a dilution of 4 μg/ml, Smad5 at a dilution of 1:50
(Santa Cruz), phospho-Smad1/5/8 at a dilution of 1:100 (Cell Sig-
naling Technology) and CD44 at a dilution of 1:50 (Cell Signaling
Technology) were used for immunofluorescence as described pre-
viously (Andl et al., 2004; Flanders et al., 2001; Maatouk et al.,
2009). To detect cell death, the TUNEL assay (Roche) and anti-
cleaved caspase3 antibody at a 1:150 dilution (Cell Signaling Tech-
nology) were performed according to the manufacturers' protocol.
BrdU labeling (Roche) was used to detect cell proliferation level.
For BrdU labeling, pregnant females at 13.5 dpc were injected
with BrdU (1 mg per 10 g body weight) and sacrificed after 2 h.
Embryonic limbs were isolated, fixed in 4% PFA at 4 °C overnight,
detected by antibody against BrdU (Chemicon) at dilution of
1:500. The number of cleaved caspase3 positive and BrdU positive
cells were counted from a fixed area circle within the interdigital
under the ectoderm from 7 μm sections from each embryo. Per-
centage of BrdU positive cells were calculated by dividing number
of BrdU positive cells by the number of Hoechst-stained nuclei in
fixed area circle within the interdigital area. Section in situ hybrid-
ization was performed according to the protocol previously de-
scribed (He et al., 2001; Kwan et al., 2004). Whole mount RNA in
situ hybridization was performed according to protocol previously
described (Wilkinson, 1992). Probes used to detect Bmp2, Bmp4,
Bmp7, En1, Fgf8, Gremlin, Lmx1b, Msx1, Msx2 and Shh have been de-
scribed previously (Maatouk et al., 2009; Pajni-Underwood et al.,
2007). At least three embryos for each genotype were examined
in the aforementioned experiments. Embryos from the same litter
were used for comparisons between the mutant and control
Cell tracing and skeletal preparations
To trace the limb ectodermal cells and their descendents whose
Smad1/Smad5 were inactivated by En1Cre/+, we stained for the Cre
dependent β-gal activity of the R26R reporter allele in En1Cre/+;
Smad1f/f; Smad5f/f; Rosa26R26R/+embryos as previously described
(Soriano, 1999). Stained embryos were postfixed in 4% PFA at 4 °C
overnight, paraffin-embedded and stained with eosin for histologi-
cal analysis of the limbs. Skeletal staining using Alcian blue and
Alizarin red was performed as previously described (Ovchinnikov,
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
En1Cre/+inactivation of Smad1 and Smad5 conditional alleles in the AER
and limb ventral ectoderm
To investigate the functions of Smad1 and/or Smad5 signaling in
limb ectoderm and AER, Smad1 and/or Smad5 were inactivated in
the AER and ventral ectoderm, using Smad1 and/or Smad5 conditional
null alleles and an En1Cre/+knock-in allele. Previous reports have
shown that En1Cre/+knock-in allele is active in the ventral ectoderm
and the ventral part of the developing AER in the forelimb buds as
early as E9.5 (Kimmel et al., 2000; Kwan et al., 2004). At later stages,
from E11.5, En1Cre/+expressing cells are also found throughout the
dorsal–ventral extent of the AER (Kimmel et al., 2000). To determine
whether the inactivation of Smad genes by En1Cre/+knock-in allele
followed this pattern; we performed immunofluorescence on the
forelimb buds of E10.5 Smad1/Smad5 double conditional knockout
mutant (hereafter referred as Smad1/5 mutant) embryo using anti-
Smad1 and anti-Smad5 antibodies. Smad1 and Smad5 were absent
in the ventral ectoderm and ventral AER of the Smad1/5 mutants
(Fig. 1A–D), suggesting
knock-in allele successfully
recombined Smad1 and Smad5 conditional alleles in the limb ventral
ectoderm of the mutant. Smad1 and Smad5 were present in the dor-
sal ectoderm. WhenusingantibodyagainstpSmad1/5/8forperforming
immunofluorescence, the signal was detected in the AER and through-
out ectoderm of both the control and Smad1/5 mutant suggesting that
the expression of another BMPR-Smad, Smad8, was still present in the
limb AER and ectoderm of the Smad1/5 mutants (Fig. 1E, F).
Inactivation of both Smad1/Smad5 signaling in the limb AER and ventral
ectoderm results in interdigital tissue regression defects
Ectodermal BMP receptor BmprIa and the ligands Bmp2/4 are
important for regulating interdigital PCD (Maatouk et al., 2009;
Pajni-Underwood et al., 2007). However, it remains unclear wheth-
er canonical BMP signaling through R-Smads are involved and what
are their roles in regulating the interdigital PCD. To investigate
which of the ectodermal R-Smads signaling is essential for the
interdigital PCD, single Smad1 or Smad5 and double conditional
knockout mutants were examined. Single inactivation of either
Smad1 (En1Cre/+; Smad1f/f) or Smad5 (En1Cre/+; Smad5f/f) does not
result in limb development abnormalities (Fig. 2A–D). Embryos
harboring only one functional allele of either Smad1 allele
(En1Cre/+; Smad1f/+; Smad5f/f) (Fig. 2E, G) or Smad5 allele
(En1Cre/+; Smad1f/f; Smad5f/+) (data not shown) also had normal
limbs. Thus, inactivation of three alleles of the Smad1/5 genes in
the AER and ventral ectoderm did not result in a visible limb abnor-
mality. Embryos of En1Cre/+; Smad1f/+; Smad5f/fgenotype were
used as control throughout the study. Interestingly, combined dele-
tion of both Smad1 and Smad5 in limb AER and ventral ectoderm
resulted in limb abnormalities similar to embryos lacking BmprIa
orBMP2/BMP4 in the AER(Maatouk
Underwood et al., 2007). Both the forelimbs and hindlimbs of our
Smad1/5 mutants were severely affected at E18.5. Interdigital tissue
was retained and syndactyly was observed in the mutant limbs
(Fig. 2F, H). The defect of interdigital tissue regression could be ob-
served at E13.5 when forming digits started to separate. (Fig. 2I–L).
In controls, the forming digits started to separate at this stage and
individual digits could be observed in both forelimbs and hindlimbs
(Fig. 2I, K). However, the mutant limb autopods at E13.5 remained
as a plate-like structure and the interdigital tissue was maintained
between the forming digits of both forelimbs and hindlimbs
(Fig. 2J, L). The autopods of Smad1/5 mutant were expanded slightly
along the anterior to posterior axis. This demonstrates that Smad1
and Smad5 in the ectoderm and AER function redundantly to regu-
late the interdigital mesenchymal tissue regression.
Alcian blue/alizarin red skeletal staining revealed no major dif-
ferences of limb skeletal elements along the proximal to distal
axis between the control and the Smad1/5 mutant limbs at P10
(Fig. 2M–P). Skeletal elements along the proximal to distal axis
were formed properly in Smad1/5 mutants. Inactivation of both
Smad1 and Smad5 in the AER and ventral ectoderm did not result
in defects in proximal–distal patterning of limb. Interestingly, our
Smad1/5 mutants exhibited skeletal alteration in the anterior to
posterior axis. Postaxial polydactyly was observed at E18.5 Smad
1/5 mutants (four out of four Smad1/5 mutants analyzed) (Suppl.
Fig. 1F). The ectopic rudiment attaches to the base of the distal pha-
lange of the digitus minimus.
Smad1/Smad5 signaling in the limb AER and ventral ectoderm is required
for regulating interdigital cell death and cell proliferation
To examine whether the syndactylyl phenotype in the Smad1/5
mutants is caused by a decrease in apoptosis in the interdigital re-
gions, the TUNEL assay was employed to mark the apoptotic cells
in the E13.5 limb. Prominent apoptosis was observed in the interdi-
gital regions of the control embryos (Fig. 3A). However, there was
Fig. 1. En1-Cre inactivated Smad1 and Smad5 conditional null alleles in ventral ecto-
derm and ventral AER of forelimb buds. (A–D) Immunofluorescence detecting Smad1
(A, B) and Smad5 (C, D) in E10.5 control (A, C) (Smad1f/+; Smad5f/f) and Smad1/5 mu-
tant (B, D) (En1Cre/+; Smad1f/f; Smad5f/f) forelimb buds respectively. Smad1 and Smad5
were detected in the dorsal ectoderm but not in the ventral ectoderm and ventral half
of AER of Smad1/5 mutant. (E, F) Immunofluorescence detecting pSmad1/5/8 in the
forelimb buds of E10.5 control (E) (En1Cre/+; Smad1f/+; Smad5f/f) and Smad1/5 mutant
(F) (En1Cre/+; Smad1f/f; Smad5f/f). pSmad1/5/8 signal was detected throughout AER and
ectoderm in both control and Smad1/5 mutant. The histological samples are sectioned
saggitally along the dorsal–ventral axis of the limb buds. Arrows indicate the major
axes: D, dorsal; V, ventral; Di, distal; Pr, proximal. Scale bar: 50 μm.
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
marked decrease in apoptosis in the interdigital regions of the
Smad1/5 mutants (Fig. 3B). This suggests that a reduction in interdi-
gital cell death leads to the retention of interdigital tissue. Dying
cells may activate caspase3 to execute interdigital apoptosis. To in-
vestigate whether there was reduction in caspase3 activation, im-
munofluorescence was employed to detect activated cleaved
caspase3 in limbs from the Smad1/5 mutants at E13.5. Interdigital
mesenchyme of mutant limbs showed a drastic decrease in cleaved
caspase3 positive cells compared with controls (pb0.005) (Fig. 3C–
E), suggesting that activation of caspase3 in interdigital mesen-
chyme may be compromised after both Smad1 and Smad5 are inac-
tivated in the limb ectoderm. These results demonstrate that
Smad1/5 signaling in the AER and ventral ectoderm is required to
regulate activation of caspase3 in interdigital mesenchyme and
hence interdigital PCD.
Apart from the reduction in apoptosis, it is possible that ectopic
cell proliferation in the interdigital mesenchyme can also contribute
to the syndactyly in the Smad1/5 mutants. Thus, BrdU labeling was
employed to measure cell proliferation in the limbs of the mutant
embryos at E13.5 (Fig. 3F-J). There was an increase in percentage
of BrdU positive cells in the distal interdigital mesenchymal re-
gions of the mutant limbs when comparing to the controls. This
result demonstrates that Smad1/5 signaling in the AER and ven-
tral ectoderm also regulates cell proliferation in the interdigital
Fgf8 is persisted at the interdigital distal ectoderm and serves as survival
signal for interdigital mesenchyme upon inactivation of the Smad1/
Smad5 signaling in AER and ventral ectoderm
AER is an important signaling center in the developing limb bud
which secretes various factors regulating limb development (Dudley
et al., 2002; Pajni-Underwood et al., 2007). To analyze the roles of
Smad1/5 signaling in regulating AER structure and function, the ex-
pression of Fgf8 and En1, markers for AER, was examined in the mu-
tant limbs. At E10.5, similar pattern of Fgf8 expression was observed
at the AER of the Smad1/5 mutant and the control (Fig. 4A–B). At
E11.5, similar pattern and level of Fgf8 and En1 expressions within
Fig. 2. Inactivation of Smad1 and Smad5 in the AER and ventral ectoderm by En1Cre/+knock-in allele resulted in syndactyly and retention of interdigital tissue. Inactivation of either
Smad1 (A, B) or Smad5 (C, D) in the AER and ventral ectoderm did not result in visible phenotypes in both forelimbs (FL) and hindlimbs (HL) of the individuals at adult stage. (E–H)
Fore- and hindlimb from the control (E, G) and Smad1/5 mutant fetuses (F, H) at E18.5. Black arrows indicate separated digits in the control, while white arrows indicate the
retained interdigital tissue in the mutants. (I-L) The defect in interdigital tissue regression was observed in both fore- and hindlimbs of the Smad1/5 mutants starting from
E13.5. Syndactyly (webbing) is observed in the double mutants (J, L). White arrows indicate interdigital tissue. White arrowheads indicate that the posterior end of the autopod
of Smad1/5 mutant is broadened at E13.5. (M–P) Skeletal preparations of the forelimbs from the control (M) and Smad1/5 mutants (O) at postnatal day 10. Higher magnification
view of the autopod of the controls (N) and the mutants (P). Inactivation of Smad1 and Smad5 in the AER and ventral ectoderm did not result in skeletal defects in the autopod and
along proximal–distal axis. A–D are the ventral views of the limb with anterior to the right. E–L are dorsal views of the limb plates with anterior to the right. Scale bars: M,
O=5 mm; N, P=500 μm.
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
the AER were observed in the Smad1/5 mutant and control limb buds
as shown by the section in-situ hybridization on sections along the
dorsal–ventral axis (Fig. 4C-F). This result showed that En1 expres-
sion was maintained in the absence of Smad1/5 signaling in the AER
and ventral ectoderm while the morphology of the AER was not al-
tered in the Smad1/5 mutant at E11.5. At E13, Fgf8 expression was
expressed in the AER overlying the developing digits but ceased
over the interdigital region when interdigital cell death was
initiated in the controls (Fig. 4G, I). However, we found that Fgf8 ex-
pression was maintained at the AER over the interdigital region in
the E13 Smad1/5 mutant limbs, consistent with the previous find-
ings in the BmprIa mutants (Pajni-Underwood et al., 2007). The ex-
pression of Fgf8 was punctate throughout the mutant AER (Fig. 4H,
J). The Fgf8 expression in AER over the interdigital region ceased by
E13.5 in the Smad1/5 mutant (Fig. 4L). This suggests that FGF signal-
ing in the Smad1/5 mutant limbs may be elevated. The ectopic AER-
Fig. 3. Smad1/Smad5 signaling in the AER and ventral ectoderm is required for regulating cell death and cell proliferation in interdigital regions. (A, B) TUNEL assay and (C, D) im-
munofluorescence detecting cleaved caspase3 to measure cell death in E13.5 forelimbs. Inactivation of Smad1 and Smad5 resulted in a decrease in cell death. (E) Quantification of
the number of anti-cleaved caspase3 positive cells demonstrated that there was a significantly decrease in the number of cells undergoing cell death in the Smad1/5 mutant limbs
(***pb0.005). (F, G) Smad1/5 signaling in the AER and ventral ectoderm is required for regulating cell proliferation in the underlying mesenchyme. BrdU labeling assay showed that
inactivation of Smad1 and Smad5 resulted in an increase in cell proliferation in distal interdigital mesenchymal regions. (H, I) Cleaved caspase3 and BrdU co-immunostaining in-
dicated that distal interdigital areas (white circle) of Smad1/5 mutant are abundant in BrdU positive cells. (J) Quantification of the percentage of BrdU positive cells with Hoechst
stained nuclei demonstrated that there was a significant increase in the percentage of proliferating cells in the distal interdigital areas of the Smad1/5 mutant limbs (**pb0.005).
Scale bars: A–B=100 μm.
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
specific Fgf8 expression over the interdigital region may serve as
survival signal for interdigital cells in Smad1/5 the mutants
(Macias et al., 1996; Martin, 1998; Pajni-Underwood et al., 2007;
Weatherbee et al., 2006).
Mesenchymal BMP signals are not altered in the developing autopod of
the Smad1/5 mutants
BMP signals acting on the interdigital mesenchyme are sug-
gested to be the triggering factors for interdigital PCD. Bmp2,
Bmp4 and Bmp7 are expressed in the interdigital regions at E12.5
and suggested to be the candidate ligands in these regions (Lyons
et al., 1990, 1995; Zou and Niswander, 1996). To examine if the de-
crease in interdigital PCD in our Smad1/5 mutants was related to a
decrease in mesenchymal BMP signals, expression of the BMP sig-
naling components was examined in E12.5 autopod. The expression
of the three BMP ligands, Bmp2, Bmp4 and Bmp7, in the interdigital
regions was similar between the Smad1/5 mutants and the controls
at E12.5 (Fig. 5A–F). Interestingly, there were enhanced expressions
of Bmp2 (Fig. 5B) and Bmp4 (Fig. 5D and Supp. Fig. 1B) in the AER of
the Smad1/5 mutants. Gremlin is known to be antagonist of BMP
signaling (Khokha et al., 2003; Zuniga et al., 1999). No significant
difference in interdigital Gremlin expression was found between
the Smad1/5 mutant and control (Fig. 5G, H). Msx1 and Msx2 are
downstream targets of BMP signaling in the interdigital regions
(Guha et al., 2002; Marazzi et al., 1997; Zou and Niswander,
1996). No significant decrease in the expression of Msx1 and Msx2
was detected in the Smad1/5 mutant interdigital mesenchyme.
Taken together, these results indicate that the mesenchymal BMP
signaling does not alter in our Smad1/5 mutants and the defect of
interdigital PCD in the mutants is not caused by the decrease in
mesenchymal BMPs signaling levels. Interestingly, the ectopic
Bmp2 and Bmp4 expression in the distal ectoderm in the Smad1/5
mutants suggests that there may be an auto-regulatory loop of
BMP signaling and BMP signals in the limb AER (Pajni-Underwood
et al., 2007).
Smad1/Smad5 inactivation in the limb ventral ectoderm resulted in ven-
tral ectoderm thickening and ectopic En1-expressing cells and their de-
scendents in the dorsal interdigital ectoderm
The fate of mesenchymal cells of the developing limb bud has
been shown to be regulated by signaling from the ectoderm
(Hurle and Ganan, 1986; Maatouk et al., 2009; Pajni-Underwood
et al., 2007). Interdigital mesenchyme showed a reduction in apo-
ptosis and increase in cell proliferation after Smad1/5 inactivation
in the AER and ventral ectoderm. However, the fate of the ventral
ectodermal cells after Smad1/Smad5 inactivation remains unclear.
To investigate the possible changes in the ventral ectoderm after
Smad1/5 inactivation, a R26R lacZ reporter system was employed
to trace the ventral ectodermal cells. The morphology of the ventral
ectoderm was examined in E15.5 limbs. Transverse sections of the
control limbs showed that the ventral ectoderm was a thin struc-
ture consisting of 1–2 layers of cells (Figs. 6E and 7E). However,
the ventral ectoderm of the Smad1/5 mutant became thicker and
consisted of 8–10 layers of cells (Figs. 6F and 7F). To determine if
this change occurs in the earlier stage limb buds, we examined an
ectodermal marker CD44 at E13.5. CD44 immunostaining showed
that the ectoderm of E13.5 control limb was thin and made up of
1–2 layers of cells (Fig. 6K). However, the ectoderm of the mutant
was thicker, which was made up of multiple layers of cells
(Fig. 6L). These results suggest that Smad1/5 signaling regulates
the number of cell layers formation at the ventral ectoderm. When
Smad1 and Smad5 were inactivated, the ventral limb ectoderm be-
came thickened with multiple layers of cells. Very interestingly,
some ventral En1-expressing cells and their descendents were lo-
cated at the interdigital webbing of the developing limb of the
Smad1/5 mutants at E14.5 (Fig. 6G, H) and E15.5 (Fig. 6A, B) while
the digits of the controls were separated. The border of interdigital
blue streaks extend more proximally at E15.5 compared to E14.5
Fig. 4. AER and ventral ectoderm Smad1/Smad5 signaling is required to repress expres-
sion of Fgf8 in the interdigital distal ectoderm. (A, B) Whole mount in situ hybridization
of E10.5 mutant forelimb buds showed no expansion of Fgf8 expression along dorsal–
ventral or anterior–posterior axis. (C–F) Section in situ hybridization of E11.5 mutant
forelimb buds showed that the expression of AER marker Fgf8 (C, D) and En1 (E, F)
was restricted in AER and ventral ectoderm along the dorsal–ventral axis that was
comparable with the control. The morphology of AER in the Smad1/5 mutant did not
show any abnormality. (G–J) Ectopic Fgf8 expression was observed in the forelimb
interdigital ectoderm of the Smad1/5 mutants at E13 (white arrowhead) and the Fgf8
expression was punctuate throughout the distal ectoderm of the mutants. (K, L) Fgf8
expression faded in AER over interdigit of Smad1/5 mutants at E13.5 indicated by
white arrows. Arrows indicate the major axes: A, anterior; P, posterior; D, dorsal; V,
ventral; Di, distal; Pr, proximal. Scale bars: A–B=500 μm; C–F=50 μm; G–L=1 mm.
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
(Fig. 6B, H). TransversesectionsoftheE15.5mutantlimbsshowedthat
someEn1-expressingcellsand their descendents were located ectop-
ically at the dorsal interdigital ectoderm (Fig. 6D). The boundary of
ventral ectoderm was shown to displace dorsally slightly in E13.5
mutant limb buds (Fig. 6I, J).
Inactivation of both Smad1 and Smad5 in ventral limb ectoderm and AER
does not cause defects in dorsal–ventral patterning
Previous report shows that BMP signaling in the preAER/AER is
required for dorsal–ventral patterning. Inactivation of BmprIa in
Fig. 5. Syndactyly phenotype of the Smad1/5 mutants is not caused by changes in mesenchymal BMP signaling of the developing autopod. Whole mount in situ hybridization of
E12.5 controls (A, C, E, G, I and K) and mutants (B, D, F, H, J and L) forelimb buds using the probes as indicated. All are dorsal views. Arrowheads indicated ectopic expression of
the genes indicated in the distal ectoderm. Arrows indicate the major axes: A, anterior; P, posterior; Di, distal; Pr, proximal. Scale bars = 500 μm.
Fig. 6. Smad1/5 signaling in the ventral AER and ectoderm is required for regulating the cell fate of ventral ectodermal cells. The Cre/loxP system was employed to map the ventral
ectodermal cells and their descendenst after Smad1/5 inactivation in ventral AER and ectoderm. (A–B, G–J) Whole-mount X-gal staining of forelimb plates at E15.5 (A, B), E14.5 (G,
H) and E13.5 (I, J) to identify En1-expressing cells and their descendents after Smad1/Smad5 inactivation. The border of the interdigital blue streak (arrowheads) extended more
proximally at E15.5 (B) compared to E14.5 (H). (C-F) Transverse sections of the X-gal stained E15.5 forelimb plates showing β-galactosidase activity. (D, F) Ventral ectoderm
was thickened in the Smad1/5 mutant. Blue lacZ staining was observed in the ventral ectoderm and interdigital ectoderm in the dorsal regions of the Smad1/5 mutant (arrows).
(E, F) Higher magnification views of the areas indicated by the squares in (C and D). Ventral ectoderm layers of the Smad1/5 mutants became thicker and consisted of multiple layers
of cells at E15.5 (compare E and F). (K-L) Immunostaining using ectodermal marker CD44 showed that ectoderm thickening was observed in mutant forelimb plates at E13.5. Dot-
ted lines (E and F) indicated the basement membrane of the epithelium. Arrows indicate the major axes: A, anterior; P, posterior; D, dorsal; V, ventral. Scale bars: A–B, G–
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
limb ectoderm by Brn4-Cre would result in double dorsal structures
in the limb and absence of En1 expression (Ahn et al., 2001). To in-
vestigate if the inactivation of ectodermal Smad1 and Smad5 would
affect dorsal–ventral patterning, the expressions of En1 and Lmx1b
were examined. En1 was expressed normally in the AER and ventral
ectoderm of E11.5 forelimb buds of the Smad1/5 mutants. The ex-
pression pattern of En1 in the mutants was comparable to the con-
trol limbs (Figs. 4E, F and 7A, B). Lmx1b is expressed at the dorsal
mesenchyme which promotes dorsal structure differentiation.
Lmx1b was expressed normally at the dorsal mesenchyme of the
Smad1/5 mutant limb buds at E11.5. No ectopic Lmx1b expression
was observed at the ventral mesenchyme of mutant limbs (Fig. 7C,
D). Histological sections across the limb dorsal–ventral axis at
E15.5 revealed that ventral mesenchyme structure such as tendon
were formed and located properly in the ventral positions of the
Smad1/5 mutant limbs, which is in contrast with loss of ventral
mesenchyme structure in the conditional BmprIa mutant (Ahn et
al., 2001). Also, no ectopic dorsal ectodermal structure was
observed at the ventral side of our Smad1/5 mutant limbs (Fig. 7G,
H), contrasting with the conditional BmprIa mutant.
Smad1/Smad5 inactivation in the limb AER and ventral ectoderm
resulted in postaxial polydactyly and expanded Shh expression in the
posterior limb mesenchyme
Shh is expressed at the posterior mesenchyme of the developing
limb bud to regulate anterior–posterior axis formation (Chang et al.,
1994; Riddle et al., 1993). Anterior expansion of Shh signaling within
the limb mesenchyme is suggested to be one of the mechanisms un-
derlying the polydactyly (Selever et al., 2004). Alcian blue/alizarin red
skeletal staining revealed that our Smad1/5 mutants exhibited post-
axial polydactyly (Suppl. Fig. 1F). Shh expression was expanded in
the posterior mesenchyme of the Smad1/5 mutant forelimb buds
when compared with the control at E10.5 (Suppl. Fig. 1C, D). Thus,
the lost of ventral ectodermal Smad1/5 signaling may lead to the ex-
panded Shh expression that induce the postaxial polydactyly.
Fig. 7. Smad1 and Smad5 signaling in the AER and ventral ectoderm are not required for dorsal–ventral patterning of the limb. (A, B) En1 expression in the AER (white arrowheads)
and the ventral ectoderm of the forelimb buds from control (A) and Smad1/5 mutant (B) at E11.5. (C, D) Lmx1b, which is a dorsal limb mesoderm marker, showed restricted ex-
pression in the dorsal mesoderm of the mutant forelimb bud at E11.5 (D), similar to the control (C). Black arrowheads mark the location of AER. For panel A–D, dorsal side of
limb buds towards the right. (E, F) Transverse sections through the E15.5 forelimb hand-plates are shown. The tendons, which are a ventral mesodermal structure, were formed
and located properly at the ventral side of the mutant limbs. Aberrant thickening of ventral epidermal structure is observed in the Smad1/5 mutant limb (white bars). (G, H) Sagittal
section Smad1/5 mutant limbs at E18.5 does not show ectopic nail plate on the ventral side. NP, nail plate; PNF, Proximal nail fold; Vt, ventral tendon. Scale bars: A–D=1 mm; E–
F=100 μm; G–H=200 μm.
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
Previous reports have shown that BMP signaling in the AER plays
an essential role in regulating limb patterning and interdigital PCD
(Ahn et al., 2001; Guha et al., 2002; Maatouk et al., 2009; Pajni-
Underwood et al., 2007). However, these studies do not address the
intracellular mechanism downstream of the receptor BmprIa mediat-
ing the BMP signals in the AER. In addition, complete inactivation of
Smad1 or Smad5 or heterozygous inactivation of both Smad1 and
Smad5 would lead to early embryonic lethality (Chang et al., 1999;
Lechleider et al., 2001). Thus, the functions and the interactions of ec-
todermal R-Smads in AER and ventral ectoderm were investigated by
conditional inactivation approach in this study. First, we discovered
that Smad1 and Smad5 are employed as intracellular mediators
of BMP signaling in the limb AER and ventral ectoderm. Second,
the AER and ectoderm Smad1 and Smad5 function redundantly to
regulate interdigital PCD and cell proliferation of developing limb
indirectly. Third, the interaction between the limb ectoderm and
mesenchyme is required to regulate the aforementioned process-
es. Smad1/Smad5 signaling regulates AER Fgf expression to control
cell death and survival of the interdigital mesenchyme of limb. Fi-
nally, apart from the interdigital mesechyme, the development of
ectoderm is also regulated by ectodermal Smad1/Smad5 signaling.
Functional redundancy between Smad1 and Smad5 in BMP-regulated
AER and ventral ectoderm functions
The syndactyly phenotype of our mutant embryos in which both
Smad1 and Smad5 have been inactivated is similar to the mice lacking
BmprIa or Bmp2/Bmp4 in the ectoderm of developing limb bud
(Maatouk et al., 2009; Pajni-Underwood et al., 2007). This suggests
that R-Smad1 and R-Smad5 act as the intracellular mediator down-
stream of BmprIa to transduce BMP signals in the AER and ventral ec-
toderm to regulate interdigital PCD and digit separation. Since single
inactivation of either Smad1 or Smad5 does not result in limb abnor-
mality, we suggest that there is functional redundancy between the
two R-Smads in the AER and ventral ectoderm. Smad1 and Smad5
can compensate the functions of one another to regulate digit separa-
tion and even one allele of either Smad1 or Smad5 is sufficient to
transduce BMP signals in the AER for regulating interdigital PCD. In
addition, this study provides further evidence that functional redun-
dancy between Smad1 and Smad5 is not restricted to cancer develop-
ment (Pangas et al., 2008) and bone development (Retting et al.,
2009) but also in the limb AER function and PCD. However, the ex-
pression of Smad8 alone, another member of BMP receptor-
regulated Smad, cannot compensate for the function of Smad1 and
Smad5 in AER and ventral ectoderm. This may be explained by the
early divergence of Smad1/5 and Smad8 during vertebrate evolution
(Arnold et al., 2006).
The syndactyly defect of the Smad1/5 mutant is caused by reduc-
tion of interdigital PCD as suggested by TUNEL assay. The interdigital
mesenchymal cells fail to undergo normal PCD and are maintained in
the interdigital regions of the mutant to form the webbing. This result
suggests that the ectodermal Smad1/Smad5 signaling in the AER indi-
rectly regulate interdigital mesenchymal cell death. On the other
hand, controversies remain regarding whether the dying cells acti-
vate caspase3 to execute interdigital apoptosis (Zuzarte-Luis and
Hurle, 2005). Caspase3 activation is compromised in our Smad1/5
mutant as detected by anti-cleaved caspase3 antibody. This suggests
that caspase3 activation is involved in interdigital PCD and Smad1/
Smad5 signaling in the ectoderm regulates activation of caspase3 in
interdigital mesenchyme and hence interdigital apoptosis. Interdigi-
tal mesenchyme of our Smad1/5 mutants also undergoes ectopic cell
proliferation that contributes to the formation of the webbing. Since
the AER and ventral ectodermal inactivation of both Smad1 and
Smad5 leads to the decrease in cell death and ectopic cell proliferation
in mesenchyme, this suggests that appropriate interactions between
the ectoderm and the mesenchyme are required to regulate cell
death and cell proliferation in the mesenchyme.
Molecular mechanisms of limb AER and ventral ectodermal Smad1/
Smad5 signaling to control cell death and survival in the interdigital
Previous studies suggested that removal of BMP signaling compo-
nents in the AER would result in failure in AER maturation and expan-
sion of Fgf8 expression. The prolonged expression of Fgf8 at the AER
would maintain interdigital cell survival in the Bmpr1a (Pajni-
Underwood et al., 2007) and Bmp2/4 (Maatouk et al., 2009) condi-
tional mutants. The BMPs expressed in AER control Fgf activity of
the AER to regulate interdigital cell death. In our Smad1/5 mutants,
Fgf8 was ectopically expressed in the AER over the interdigits and
was punctate throughout the AER at E13, which is consistent with
the previous findings. We propose that Smad1 and Smad5 are
employed as intracellular mediator to transduce BMPs signal acting
on the AER and ventral ectoderm to repress Fgf expression in the
AER over interdigits. Interdigital PCD is triggered by the reduction in
Fgf signaling in AER over interdigital regions and this leads to separa-
tion of digits (Macias et al., 1996; Martin, 1998). Smad1 and Smad5
function redundantly to repress the Fgf8 expression in AER over inter-
digital region. Simultaneous inactivation of both Smad1 and Smad5 in
the AER and ventral ectoderm would result in prolonged Fgf8 expres-
sion. This prolonged expression of Fgf8 in the AER over interdigital re-
gion serves as survival signal to the interdigital mesenchyme, leading
to decrease in cell death and increase in cell proliferation in this tissue
(Fig. 8). In addition, our data showing the unaltered expression of
mesenchymal BMP signaling components in our Smad1/5 mutants
further support the idea that the BMP signaling within the AER and
ectoderm does not regulate the mesenchymal BMP signaling of devel-
oping limbs (Pajni-Underwood et al., 2007). The Smad1/5 signaling in
the AER and ventral ectoderm only indirectly regulates interdigital
PCD through regulating the ectodermal FGF expression. On the
other hand, previous report showed that knockout of BMP ligands
in the mesenchyme resulted in expanded Fgf8 expression in AER of
E11.5 limb bud (Bandyopadhyay et al., 2006). The question of wheth-
er mesenchymal BMP signaling can regulate BMP signaling within
AER and ectoderm awaits to be elucidated.
Interestingly, the expression of Fgf8 in our E10.5 Smad1/5 mutants
did not expanded as previously reported for the Bmpr1a and Bmp2/4
conditional mutants, suggesting that the AER of our Smad1/5 mutants
instead undergoes maturation properly. The difference in Fgf8 expres-
sion of our Smad1/5 mutants at E10.5 could possibly be due to the dif-
ferent Cre mouse line used in our study. At E10.5, the En1-Cre knock-
in allele is only active in the ventral half of the developing AER and
the ventral ectoderm (Kwan et al., 2004) while the Msx2-Cre
employed in previous reports is active in the entire AER of the early
limb bud and the ectoderm (Barrow et al., 2003; Maatouk et al.,
2009; Pajni-Underwood et al., 2007). Our results suggest that BMP
signaling in the dorsal and ventral AER at E10.5 may be responsible
for differential functions during limb development. Dorsal ectodermal
BMP signaling may be required for the maturation of the AER and
restricting the expression of Fgf8 at this earlier stage of limb bud de-
velopment. Our experiments give insights into the possible differen-
tial functions of different parts of limb ectoderm in AER development.
The role of Smad1/Smad5 signaling in limb ectoderm structure formation
The epidermis is an important organ throughout the body that
provides protection against dehydration, injury and infection. During
development, the epidermis is a well-organized structure which con-
sists of 2–3 layers of cells at E15.5. The inner layer is the stratum ger-
minativum and contains proliferating cells that divide continuously to
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
replace the outer layer. The outer layer gives rise to periderm which is
a temporary covering of the skin and will shed continuously. The bal-
ance between proliferation of the stratum geminativum and the
shedding of the outer layer is critical for the maintenance of the struc-
ture of the epidermis and its protection function (Simpson et al.,
2011). When both Smad1 and Smad5 were inactivated in the ventral
ectoderm of the developing limb, the ectoderm lost the well-
organized structure of 2–3 layers of cells. The ectoderm of the
Smad1/Smad5 mutants became a thick structure densely packed
with multiple layers of cells. These multiple layers of cells are descen-
dents of the Smad1/Smad5-inactivated ventral ectodermal cells. Fur-
thermore, the ventral En1-expressing cells and their descendents
were found ectopically at the dorsal interdigital ectoderm. Our data
suggests that Smad1/Smad5 signaling in the ventral ectoderm is im-
portant in regulating its own structural development and hence the
functions of the epidermis. We propose that Smad1/Smad5 signaling
may also function redundantly in the ventral ectoderm to limit ecto-
dermal cells in the ventral position. When both Smad1 and Smad5 are
inactivated in the ectodermal cells at the distal end of the developing
limb, these ectodermal cells may undergo ectopic proliferation and
migrate to the dorsal position as suggested by our cell tracing exper-
iment and the thickening of ventral ectoderm in the Smad1/5
The role of ventral ectodermal Smad1/5 signaling in dorsal–ventral
Previous report showed that preAER/AER BMP signaling is re-
quired for dorsal–ventral patterning. Inactivation of the receptor
BmprIa in limb ectoderm by Brn4-Cre would result in double dorsal
defects including the absence of ventral mesenchymal structures
such as flexor digitorum profundus tendon and sesamoid process
and the loss of En1 expression (Ahn et al., 2001). Moreover, knockout
of BMP ligands Bmp2/Bmp4 in the AER would result in ectopic nail
plate on the ventral side (Maatouk et al., 2009). In contrast, expres-
sion of En1 and Lmx1b in the developing limb buds of our Smad1/5
mutants was comparable to the controls. No visible dorsal–ventral
defects were observed in the Smad1/5 mutants. Ventral mesenchyme
structures are formed and located properly in the Smad1/5 mutant
limbs, which is in contrast with loss of ventral mesenchyme structure
in the conditional BmprIa mutant (Ahn et al., 2001). No ectopic dorsal
ectodermal structure was observed at the ventral side of the mutant
limb, contrasting with the conditional Bmp2/4 mutant (Maatouk
et al., 2009). The lack of defects in the dorsal–ventral patterning
of our Smad1/5 mutant could possibly due to the different Cre
mouse line used in our study. The Brn4-Cre and Msx2-Cre show
their activities in the entire ectoderm during limb bud formation
while the En1-Cre is only active in the AER and ventral ectoderm.
The signaling through the Smad1/Smad5 in the dorsal ectodermal
region of our Smad1/5 mutant limb may be sufficient to maintain
the En1 expression and the patterning of dorsal–ventral axis. It is
also possible that the AER-expressed BMP ligands may function
through alternative pathways, other than R-Smad, upstream of
En1 to regulate dorsal–ventral patterning.
Supplementary materials related to this article can be found on-
line at doi:10.1016/j.ydbio.2011.12.037.
The authors thank Ms. Sze-Nee Lim for technical assistance, Dr
Yasuhide Furuta (University of Texas MD Anderson Cancer Center,
USA) for providing the probes for Bmp2, Bmp4 and Bmp7, and Dr
James Li (University of Connecticut Health Center, USA) for pro-
viding the probes for En1 and Lmx1b. This research was supported
by Hong Kong Research Grant Council General Research Fund
(CUHK 466708) and Focused Investments Scheme — Scheme B of
The Chinese University of Hong Kong (BL08650).
Fig. 8. Our proposed model for the roles and interactions of Smad1/Smad5 in AER and ventral ectoderm of developing limbs. (A) Smad1 and Smad5 functions redundantly as the
intracellular mediator of the BMP signaling downstream of BmprIa in the AER and ventral ectoderm (Maatouk et al., 2009; Pajni-Underwood et al., 2007). Fgf8 expression at the AER
serves as cell survival signal to the interdigital mesenchyme (arrow). Smad1/5 signaling at the AER and ventral ectoderm inhibits Fgf8 expression. The termination of Fgf8 expres-
sion initiates programmed cell death in the intedigital mesenchyme. BMP signaling within the mesenchyme was shown to regulate interdigital PCD (Bandyopadhyay et al., 2006).
(B) Inactivation of Smad1/Smad5 in the AER and ventral ectoderm would result in persisted Fgf8 expression. Fgf8 is expressed ectopically in the AER over interdigital region that
leads to inhibition of PCD and ectopic cell survival and proliferation in the interdigital mesenchyme. In addition, there is an auto-regulatory loop of BMP signaling and BMP signals
in the limb AER. In the absence of Smad1/5 signaling, BMP signals in the AER is elevated.
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257
Ahn, K., Mishina, Y., Hanks, M.C., Behringer, R.R., Crenshaw III, E.B., 2001. BMPR-IA signal-
ing is required for the formation of the apical ectodermal ridge and dorsal–ventral
patterning of the limb. Development 128, 4449–4461.
Ali-Khan, S.E., Hales, B.F., 2006. Retinoid receptor antagonists alter the pattern of apo-
ptosis in organogenesis stage mouse limbs. Toxicol. Sci. 90, 208–220.
Al-Qattan, M.M., Yang, Y., Kozin, S.H., 2009. Embryology of the upper limb. J. Hand Surg.
Am. 34, 1340–1350.
Andl, T., Ahn, K., Kairo, A., Chu, E.Y., Wine-Lee, L., Reddy, S.T., Croft, N.J., Cebra-Thomas,
J.A., Metzger, D., Chambon, P., Lyons, K.M., Mishina, Y., Seykora, J.T., Crenshaw III,
E.B., Millar, S.E., 2004. Epithelial Bmpr1a regulates differentiation and proliferation
in postnatal hair follicles and is essential for tooth development. Development 131,
Arnold, S.J., Maretto, S., Islam, A., Bikoff, E.K., Robertson, E.J., 2006. Dose-dependent
Smad1, Smad5 and Smad8 signaling in the early mouse embryo. Dev. Biol. 296,
Bandyopadhyay, A., Tsuji, K., Cox, K., Harfe, B.D., Rosen, V., Tabin, C.J., 2006. Genetic
analysis of the roles of BMP2, BMP4, and BMP7 in limb patterning and skeletogen-
esis. PLoS Genet. 2, e216.
Barrow, J.R., Thomas, K.R., Boussadia-Zahui, O., Moore, R., Kemler, R., Capecchi, M.R.,
McMahon, A.P., 2003. Ectodermal Wnt3/beta-catenin signaling is required for the
establishment and maintenance of the apical ectodermal ridge. Genes Dev. 17,
Buscher, D., Bosse, B., Heymer, J., Ruther, U., 1997. Evidence for genetic control of Sonic
hedgehog by Gli3 in mouse limb development. Mech. Dev. 62, 175–182.
Capdevila, J., Izpisua Belmonte, J.C., 2001. Patterning mechanisms controlling verte-
brate limb development. Annu. Rev. Cell Dev. Biol. 17, 87–132.
Chang, D.T., Lopez, A., von Kessler, D.P., Chiang, C., Simandl, B.K., Zhao, R., Seldin, M.F.,
Fallon, J.F., Beachy, P.A., 1994. Products, genetic linkage and limb patterning activ-
ity of a murine hedgehog gene. Development 120, 3339–3353.
Chang, H., Huylebroeck, D., Verschueren, K., Guo, Q., Matzuk, M.M., Zwijsen, A., 1999.
Smad5 knockout mice die at mid-gestation due to multiple embryonic and extra-
embryonic defects. Development 126, 1631–1642.
Delgado, I., Dominguez-Frutos, E., Schimmang, T., Ros, M.A., 2008. The incomplete inac-
tivation of Fgf8 in the limb ectoderm affects the morphogenesis of the anterior
autopod through BMP-mediated cell death. Dev. Dyn. 237, 649–658.
Dudley, A.T., Ros, M.A., Tabin, C.J., 2002. A re-examination of proximodistal patterning
during vertebrate limb development. Nature 418, 539–544.
Fadeel, B., Orrenius, S., 2005. Apoptosis: a basic biological phenomenon with wide-
ranging implications in human disease. J. Intern. Med. 258, 479–517.
Fernandez-Teran, M.A., Hinchliffe, J.R., Ros, M.A., 2006. Birth and death of cells in limb
development: a mapping study. Dev. Dyn. 235, 2521–2537.
Flanders, K.C., Kim, E.S., Roberts, A.B., 2001. Immunohistochemical expression of Smads
1–6 in the 15-day gestation mouse embryo: signaling by BMPs and TGF-βs. Dev.
Dyn. 220, 141–154.
Galdones, E., Lohnes, D., Hales, B.F., 2006. Role of retinoic acid receptors alpha1 and
gamma in the response of murine limbs to retinol in vitro. Birth Defects Res. A.
Clin. Mol. Teratol. 76, 39–45.
Ganan, Y., Macias, D., Duterque-Coquillaud, M., Ros, M.A., Hurle, J.M., 1996. Role of TGF
beta s and BMPs as signals controlling the position of the digits and the areas of
interdigital cell death in the developing chick limb autopod. Development 122,
Goodman, F.R., 2002. Limb malformations and the human HOX genes. Am. J. Med.
Genet. 112, 256–265.
Grotewold, L., Ruther, U., 2002. Bmp, Fgf and Wnt signalling in programmed cell death
and chondrogenesis during vertebrate limb development: the role of Dickkopf-1.
Int. J. Dev. Biol. 46, 943–947.
Guha, U., Gomes, W.A., Kobayashi, T., Pestell, R.G., Kessler, J.A., 2002. In vivo evidence
that BMP signaling is necessary for apoptosis in the mouse limb. Dev. Biol. 249,
He, W., Cao, T., Smith, D.A., Myers, T.E., Wang, X.J., 2001. Smads mediate signaling of the
TGFbeta superfamily in normal keratinocytes but are lost during skin chemical car-
cinogenesis. Oncogene 20, 471–483.
Hernández-Martínez, R., Covarrubias, L., 2011. Interdigital cell death function and reg-
ulation: new insights on an old programmed cell death model. Dev. Growth Differ.
Huang, S., Tang, B., Usoskin, D., Lechleider, R.J., Jamin, S.P., Li, C., Anzano, M.A., Ebendal,
T., Deng, C., Roberts, A.B., 2002. Conditional knockout of the Smad1 gene. Genesis
Hurle, J.M., Ganan, Y., 1986. Interdigital tissue chondrogenesis induced by surgical re-
moval of the ectoderm in the embryonic chick leg bud. J. Embryol. Exp. Morphol.
Itoh, S., Itoh, F., Goumans, M.J., ten Dijke, P., 2000. Signaling of transforming growth
factor-beta family members through Smad proteins. Eur. J. Biochem. 267,
Jacobson, M.D., Weil, M., Raff, M.C., 1997. Programmed cell death in animal develop-
ment. Cell 88, 347–354.
Khokha, M.K., Hsu, D., Brunet, L.J., Dionne, M.S., Harland, R.M., 2003. Gremlin is the BMP
antagonist required for maintenance of Shh and Fgf signals during limb patterning.
Nat. Genet. 34, 303–307.
Kimmel, R.A., Turnbull, D.H., Blanquet, V., Wurst, W., Loomis, C.A., Joyner, A.L., 2000.
Two lineage boundaries coordinate vertebrate apical ectodermal ridge formation.
Genes Dev. 14, 1377–1389.
Kwan, K.M., Li, A.G., Wang, X.J., Wurst, W., Behringer, R.R., 2004. Essential roles of
BMPR-IA signaling in differentiation and growth of hair follicles and in skin tumor-
igenesis. Genesis 39, 10–25.
Lechleider, R.J., Ryan, J.L., Garrett, L., Eng, C., Deng, C., Wynshaw-Boris, A., Roberts, A.B.,
2001. Targeted mutagenesis of Smad1 reveals an essential role in chorioallantoic
fusion. Dev. Biol. 240, 157–167.
Lyons, K.M., Pelton, R.W., Hogan, B.L., 1990. Organogenesis and pattern formation in
the mouse: RNA distribution patterns suggest a role for bone morphogenetic
protein-2A (BMP-2A). Development 109, 833–844.
Lyons, K.M., Hogan, B.L., Robertson, E.J., 1995. Colocalization of BMP 7 and BMP 2 RNAs
suggests that these factors cooperatively mediate tissue interactions during mu-
rine development. Mech. Dev. 50, 71–83.
Maatouk, D.M., Choi, K.S., Bouldin, C.M., Harfe, B.D., 2009. In the limb AER Bmp2 and
Bmp4 are required for dorsal–ventral patterning and interdigital cell death but
not limb outgrowth. Dev. Biol. 327, 516–523.
Macias, D., Gañan, Y., Ros, M.A., Hurle, J.M., 1996. In vivo inhibition of programmed cell
death by local administration of FGF-2 and FGF-4 in the interdigital areas of the
embryonic chick leg bud. Anat. Embryol. (Berl) 193, 533–541.
Macias, D., Gañan, Y., Sampath, T.K., Piedra, M.E., Ros, M.A., Hurle, J.M., 1997. Role of
BMP-2 and OP-1 (BMP-7) in programmed cell death and skeletogenesis during
chick limb development. Development 124, 1109–1117.
Marazzi, G., Wang, Y., Sassoon, D., 1997. Msx2 is a transcriptional regulator in the
BMP4-mediated programmed cell death pathway. Dev. Biol. 186, 127–138.
Martin, P., 1990. Tissue patterning in the developing mouse limb. Int. J. Dev. Biol. 34,
Martin, G.R., 1998. The roles of FGFs in the early development of vertebrate limbs.
Genes Dev. 12, 1571–1586.
Montero, J.A., Ganan, Y., Macias, D., Rodriguez-Leon, J., Sanz-Ezquerro, J.J., Merino, R.,
Chimal-Monroy, J., Nieto, M.A., Hurle, J.M., 2001. Role of FGFs in the control of
Ovchinnikov, D., 2009. Alcian blue/alizarin red staining of cartilage and bone in mouse.
Cold Spring Harb. Protoc. 4 (3), prot5170.
Pajni-Underwood, S., Wilson, C.P., Elder, C., Mishina, Y., Lewandoski, M., 2007. BMP sig-
nals control limb bud interdigital programmed cell death by regulating FGF signal-
ing. Development 134, 2359–2368.
Pangas, S.A., Li, X., Umans, L., Zwijsen, A., Huylebroeck, D., Gutierrez, C., Wang, D.,
Martin, J.F., Jamin, S.P., Behringer, R.R., Robertson, E.J., Matzuk, M.M., 2008. Con-
ditional deletion of Smad1 and Smad5 in somatic cells of male and female go-
nads leads to metastatic tumor development in mice. Mol. Cell. Biol. 28,
Retting, K.N., Song, B., Yoon, B.S., Lyons, K.M., 2009. BMP canonical Smad signaling
through Smad1 and Smad5 is required for endochondral bone formation. Develop-
ment 136, 1093–1104.
Riddle, R.D., Johnson, R.L., Laufer, E., Tabin, C., 1993. Sonic hedgehog mediates the po-
larizing activity of the ZPA. Cell 75, 1401–1416.
Salas-Vidal, E., Valencia, C., Covarrubias, L., 2001. Differential tissue growth and pat-
terns of cell death in mouse limb autopod morphogenesis. Dev. Dyn. 220,
Selever, J., Liu, W., Lu, M.F., Behringer, R.R., Martin, J.F., 2004. Bmp4 in limb bud meso-
derm regulates digit pattern by controlling AER development. Dev. Biol. 276,
Simpson, C.L., Patel, D.M., Green, K.J., 2011. Deconstructing the skin: cytoarchitectural
determinants of epidermal morphogenesis. Nat. Rev. Mol. Cell Biol. 129, 65–80.
Soriano, P., 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat.
Genet. 21, 70–71.
Towers, M., Tickle, C., 2009. Generation of pattern and form in the developing limb. Int.
J. Dev. Biol. 53, 805–812.
Umans, L., Vermeire, L., Francis, A., Chang, H., Huylebroeck, D., Zwijsen, A., 2003. Gen-
eration of a floxed allele of Smad5 for cre-mediated conditional knockout in the
mouse. Genesis 37, 5–11.
Weatherbee, S.D., Behringer, R.R., Rasweiler IV, J.J., Niswander, L.A., 2006. Interdigital
webbing retention in bat wings illustrates genetic changes underlying amniote
limb diversification. Proc. Natl. Acad. Sci. U. S. A. 103, 15103–15107.
Wilkinson, D.G., 1992. Whole mount in situ hybridization to vertebrate embryos. In:
Wilkinson, D.G. (Ed.), In Situ Hybridization: a Practical Approach. IRL Press, Oxford,
Zou, H., Niswander, L., 1996. Requirement for BMP signaling in interdigital apoptosis
and scale formation. Science 272, 738–741.
Zuniga, A., Haramis, A.P., McMahon, A.P., Zeller, R., 1999. Signal relay by BMP antago-
nism controls the SHH/FGF4 feedback loop in vertebrate limb buds. Nature 401,
Zuzarte-Luis, V., Hurle, J.M., 2005. Programmed cell death in the embryonic vertebrate
limb. Semin. Cell Dev. Biol. 16, 261–269.
Zuzarte-Luis, V., Montero, J.A., Rodriguez-Leon, J., Merino, R., Rodriguez-Rey, J.C.,
Hurle, J.M., 2004. A new role for BMP5 during limb development acting
through the synergic activation of Smad and MAPK pathways. Dev. Biol. 272,
Y.L. Wong et al. / Developmental Biology 363 (2012) 247–257