www.elsevier.com/locate/ydbio

Developmental Biology 272 (2004) 39–52

A new role for BMP5 during limb development acting through the

synergic activation of Smad and MAPK pathways

V. Zuzarte-Luıs,a J.A. Montero,a J. Rodriguez-Leon,b R. Merino,c

J.C. Rodrıguez-Rey,c and J.M. Hurlea,*

aDepartamento de Anatomıa y Biologıa Celular, Facultad de Medicina, Universidad de Cantabria, Santander 39011, Spainb Instituto Gulbenkian de Ciencia, Oeiras, Portugal

cDepartamento de Biologıa Molecular, Facultad de Medicina, Universidad de Cantabria, Santander 39011, Spain

Received for publication 27 February 2004, revised 20 April 2004, accepted 20 April 2004

Available online 28 May 2004

Abstract

In an attempt to identify new genes implicated in the control of programmed cell death during limb development, we have generated a

cDNA library from the regressing interdigital tissue of chicken embryos. We have analyzed 804 sequences from this library and identified 23

genes involved in apoptosis in different models. One of the genes that came up in the screening was the Bone Morphogenetic Protein family

member, Bmp5, that has not been previously involved in the control of apoptosis during limb development. In agreement with a possible role

in the control of cell death, Bmp5 exhibited a regulated pattern of expression in the interdigital tissue. Transcripts of Bmp5 and BMP5 protein

were abundant within the cytoplasm of the fragmenting apoptotic interdigital cells in a way suggesting that delivery of BMPs into the tissue is

potentiated during apoptosis. Gain-of-function experiments demonstrated that BMP5 has the same effect as other interdigital BMPs inducing

apoptosis in the undifferentiated mesoderm and growth in the prechondrogenic mesenchyme. We have characterized both Smad proteins and

MAPK p38 as intracellular effectors for the action of BMPs in the developing limb autopod. Activation of Smad signaling involves the

receptor-regulated genes Smad1 and -8, and the inhibitory Smad6, and results in both the upregulation of gene transcription and protein

phosphorylation with subsequent nuclear translocation. MAPK p38 is also quickly phosphorylated after BMP stimulation in the limb

mesoderm. Treatment with the inhibitor of p38, SB203580, revealed that there are interdigital genes induced by BMPs in a p38-dependent

manner (DKK, Snail and FGFr3), and genes induced in a p38-independent manner (BAMBI, Msx2 and Smads). Together, our results suggest

that Smad and MAPK pathways act synergistically in the BMP pathway controlling limb development.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Programmed cell death; Chondrogenesis; Smad; MAPK; p38; c-jun; BAMBI

Introduction rays in all amniota embryos and have the function of

The developing vertebrate limb provides a paradigm

model for analyzing programmed cell death. Several areas

of cell death, including the anterior (ANZ) the posterior

(PNZ), the opaque patch (OP) and the interdigital areas

(INZs) are present in the developing avian limb (see Chen

and Zhao, 1998; Zuzarte-Luis and Hurle, 2002). INZs are

observed in the mesoderm between the developing digital

0012-1606/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.ydbio.2004.04.015

* Corresponding author. Departamento de Anatomıa y Biologıa

Celular, Facultad de Medicina, Universidad de Cantabria, C/Cardenal

Herrera Oria s/n, Santander 39011, Spain. Fax: +34-942-201903.

E-mail address: hurlej@unican.es (J.M. Hurle).

sculpting the shape of the digits.

A variety of morphological and molecular approaches

have established that cell death in INZs is executed by

apoptosis through the activation of caspases (Garcia-Marti-

nez et al., 1993; Jacobson et al., 1996; Nakanishi et al.,

2001). Several genes of the apoptotic machinery, including

the proapototic factors: Apaf-1 (Cecconi et al., 1998); Bax

(Dupe et al., 1999); Bak (Lindsten et al., 2000); DIO-1

(Death Inducer-Obliterator-1; Garcia-Domingo et al.,

1999); Gas1 and Gas2 (Growth Arrest Specific; Lee et al.,

1999, 2001), have been implicated in this process. In

addition, some antiapoptotic factors, including Bcl-2, Bcl-x

and A1, are expressed in the digital rays but not in the

interdigital spaces of the mouse autopod (Carrio et al., 1996;

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5240

Novack and Korsmeyer, 1994). Moreover, the interdigital

regions before the onset of apoptosis express Bag-1 which

encodes an antiapoptotic protein that binds to Bcl-2 (Crocoll

et al., 2002). Another antiapoptotic factor, Dad-1 (Defender

Against apoptotic cell Death), has been implicated in the

control of limb programmed cell death since heterozygous

mutant mice for this gene display soft-tissue syndactyly

(Nishii et al., 1999).

It has been shown that BMP2, BMP4 and BMP7 are

triggering signals for limb programmed cell death (Ganan et

al., 1996; Guha et al., 2002; Kawakami et al., 1996; Yokouchi

et al., 1996; Zou and Niswander, 1996; Zou et al., 1997).

These secreted factors are expressed in the preapoptotic limb

mesoderm, and mesodermal apoptosis is induced and

inhibited in gain- and loss-of-function experiments, respec-

tively. It is well known that other signaling pathways,

including Wnt (Grotewold and Ruther, 2002), FGF (Montero

et al., 2001) and retinoic acid (Dupe et al., 1999; Lussier et al.,

1993; Rodriguez-Leon et al., 1999), interact with BMP

signaling in the control of apoptosis. However, the mecha-

nism by which BMPs trigger apoptosis and the intracellular

signaling pathway operating between BMPs and the activa-

tion of caspases remains largely unknown. This is particularly

important since it has been shown that caspase-independent

necrosis substitutes apoptosis in Apaf1-deficient embryos

(Chautan et al., 1999), suggesting that BMPs are able to

trigger both apoptosis and necrosis. In addition, it is not clear

whether the cell death in different areas of the limb is

triggered by the same signals. In fact, experimental evidence

suggests that this may be not the case. For example, Bmp2 is

not expressed in ANZ, while TGFh2 appears to cooperate

with BMPs in the control of INZ (Dunker et al., 2002).

Furthermore, SHH, which promotes the expression of Bmp2,

is a proapoptotic factor for PNZ while exerts an antiapoptotic

effect in ANZ (Sanz-Ezquerro and Tickle, 2000).

In an attempt to identify new genes implicated in the

control of INZ apoptosis, we have generated a cDNA

library from the regressing interdigital tissue of chicken

embryos. By exploring this library, we have detected the

expression of genes implicated in apoptotic processes in

other models. These genes included: components of the

proteosome; genes implicated in DNA repair; genes pos-

itively implicated in apoptosis; antiapoptotic genes; and

genes of potential importance in phagocytosis. One of the

genes identified in the library, Bmp5, exhibited a temporal

and spatial pattern of expression closely coincident with

INZ and was selected for functional studies. Our results

provide evidence for a role of this BMP member in the

development of limb autopodium through the activation of

Smad proteins and MAPK p38.

Material and methods

Fertilized chicken embryos (Rhode Island), ranging from

day 3 to day 9 of incubation, were employed in this study.

Interdigital mesoderm cDNA library

Six hundred interdigital spaces were carefully micro-

dissected from chicken leg buds at day 7.5 of incubation

and total RNA was extracted using TRIzol Reagent (Gib-

coBRL). mRNA was next isolated with the Dynabeads

Oligo(dT)25 kit (Dynal) and employed to construct a cDNA

library using the Creator SMART cDNA Library Construc-

tion kit (CLONTECH). The obtained cDNAwas cloned into

the pGEM-T easy vector and transformed into competent

cells. The clones were analyzed by enzymatic digestion and

fragments bigger than 600 bp were selected for sequencing.

Sequences were analyzed and blasted against both NCBI

and a chicken database (www.chickest.udel.edu) in search

for homologies.

Probes and in situ hybridization

Digoxygenin-labeled sense and antisense RNA probes

were generated for all the cloned sequences of potential

interest. Sense RNA probes were employed as controls.

Specific RNA probes for Msx2, Dkk, Snail and Fgfr3 were

also generated in addition to those obtained from the library.

Fragments of chicken Smad1 (744 bp), Smad5 (714

bp), Smad6 (702 bp) and Smad8 (972 bp) genes were

obtained by RT-PCR. First-strand cDNA was synthesized

with a random hexamers and M-MulV reverse transcrip-

tase (Fermentas) and 1 Ag of total RNA from a day-7

autopods. The following primers were used to subsequent

PCR amplification: degenerated primers for cSmad1, 5V-GARGARGARAARTGGGC-3V (sense) and 5V-RCAC-CARTGYTTNGGYTC-3V (antisense); degenerated primers

for cSmad5, 5V-GAYGARGARGARAARTGG-3V (sense)

and 5V-RCACCARTGYTTNGGYTC-3V (antisense);

cSmad6, 5V-AGTACAAGCCACTGGATCTGTCCGA-3V(sense) and 5V-CCAGCCCTTGGCGAAGCTGATGC-3V(antisense); and for cSmad8, 5V-AGGTGCCATGGAA-GAACTGG-3V (sense) and 5V-AGTTACAATTGCGGC-TCTGC-3V (antisense).

The PCR conditions were 94jC, 4 min and then 35 cycles

of 94jC, 20 s; 55jC, 30 s; 72jC, 60 s; and final extension by72jC, 10 min. PCR products were run on a 1% agarose gel,

isolated, and subcloned into pGEMT-easy (Promega). Iso-

lated clones were sequenced using dideoxy sequencing.

Digoxygenin-labeled antisense RNA probes were gener-

ated for in situ hybridization analysis.

In situ hybridization was performed in whole mount

specimens of limb buds. Samples were treated with protein-

ase K (10 Ag/ml) for 25–30min at 20jC. Hybridization of thedigoxygenin-labeled probes was performed at 68jC. Reac-tions were developed with BM purple AP substrate (Roche).

Experimental manipulation of the limb

The function of BMP5 in the control of cell death and in

chondrogenesis was studied by analyzing the effects of local

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–52 41

administration into the limb mesoderm of recombinant

BMP5 (R&D Systems) at concentrations ranging from 10

Ag/ml using as carriers heparin acrylic beads (Sigma). The

potential function of Noggin to block cell death mediated by

BMP5 was explored by the implantation of beads incubated

in BMP5 and beads incubated in Noggin at 1 mg/ml

(generously donated by Regeneron Pharmaceuticals Inc.,

Tarrytown). In both experiments, the treated limbs were next

processed to study changes in the skeleton and in the pattern

of apoptosis. To explore if the MAP kinase p38 was the

cellular effector for BMPs, the BMP beads were co-

implanted with beads (affigel blue, heparin or AG1-X2)

incubated in SB203580 (Calbiochem).

In all the experiments, PBS-soaked beads were also

implanted as a control. These treatments did not induce

changes either in the normal development of the limb or in

the expression of the analyzed genes.

Limb skeletal morphology and pattern of cell death

The morphology of the limbs treated with BMP5 beads

was studied in whole mount specimens after cartilage

staining with Alcian green as described previously (Ganan

et al., 1996). The pattern of cell death was analyzed with

neutral red vital staining (see Macias et al., 1997) and by

TUNEL assay performed in paraffin sections following the

instructions of the manufacturer (Roche). To evaluate the

rate between living and dead cells, TUNEL was combined

with DAPI staining or immunolabeling with an anti-panhi-

stone antibody (see confocal microscopy).

Confocal microscopy

Immunolabeling, in situ hybridization and Tdt-mediated

dUTP nick end labeling (TUNEL) were performed in

dissociated interdigital mesodermal cells and examined with

a BioRad MRC-1024 confocal microscope using argon and

HeNe lasers. For this purpose, small squares of the third

interdigit of limbs (of embryos ranging from 5.5 to 8 days of

incubation) fixed in 4% paraformaldehyde were isolated and

carefully squashed with a coverslide to dissociate the

mesodermal cells. To stick the cells to the glass surface,

the slides were then deep frozen over dry ice. The samples

were then processed for the different methodologies.

For immunolabeling experiments, cells were permeabi-

lized with 0.5% Triton X-100 (10 min) before incubation

with the primary antibody, and after several PBS washes,

the cells were incubated with the secondary antibody.

The following primary antibodies were employed: poly-

clonal antibody against phosphorylated Smad1 (a gift from

Dr. C. Heldin); polyclonal antibodies against Smad 1/5/8,

phosphorylated p38 and BMP5/BMP7 from Santa Cruz

Biotechnology; polyclonal antibody against lamina B

(Drouin et al., 1991); monoclonal antibody against panhi-

stone (Anti-Histon, pan, Boehringer) and a polyclonal

antibody against ubiquitin (Dako, Denmark).

TUNEL assay was employed as a specific marker to

distinguish between living and apoptotic cells. We followed

the protocol for isolated cells recommended by the manu-

facturer (Roche).

For in situ hybridization, cells were permeabilized as for

immunolabeling and incubated with the digoxygenin-la-

beled probe for 3 h at 42jC. Detection was made with

rhodamine-labeled anti-digoxygening antibody (Roche).

For double labeling purposes, we performed first the in

situ hybridization followed by the immunolabeling or the

immunolabeling followed by TUNEL assay.

Results

Identification of genes expressed in the regressing

interdigital tissue

By screening a cDNA library of the third interdigital

space of 7.5-day chicken leg buds we expected to

obtain new genes implicated either in the different

stages of the apoptotic process or in all mechanisms

related with tissue regression, including phagocytosis and

extracellular matrix degradation. Analysis of TUNEL-

and DAPI-stained sections (Figs. 1A and B) showed

that at this stage the interdigital mesoderm was com-

posed of living mesenchymal cells negative for TUNEL

(60%); apoptotic, TUNEL-positive cells (30%) and phag-

ocytes (10%). It must be remarked that living meso-

dermal cells die within the next 24–36 h since by day

9 of incubation the interdigital tissue has been totally

removed. Therefore, genes identified in our library could be

potentially implicated in all mechanisms related with tissue

regression.

Eight hundred four clones containing cDNA fragments

of sizes between 600 and 1000 bp were sequenced. From

all the analyzed sequences, 447 did not show any similar-

ity with published sequences, 117 sequences corresponded

to ribosomal and mitochondrial sequences. The remaining

240 sequences were non-ribosomal cDNAs corresponding

to 131 different known genes (see Table 1). Some of these

genes were factors implicated in the regression of the

interdigits, including Wnt5a (Chimal-Monroy et al.,

2002), Msx2 (Ferrari et al., 1998) Bmp7 (Macias et al.,

1997) and BAMBI (Grotewold et al., 2001). We also

detected factors implicated in apoptotic processes in a

variety of systems including: components of the proteo-

some (polyubiquitin; proteosome 28-kDa subunit homo-

logue; ubiquitin C ubiquitin I; ubiquitin II; ubiquitin

specific proteases 8 and 24); genes implicated in DNA

repair (Prkdc gene for DNA-dependent protein kinase;

Mori et al., 2001); genes positively implicated in apoptosis

(FLASH, a caspase 8 binding protein, Imai et al., 1999;

damage-specific DNA binding protein, Sun et al., 2002;

voltage-dependent anion channel, VDAC2, Sugiyama et

al., 2002; spermidine/spermine N1-acetyltransferase, Ha et

Fig. 1. (A, B) Section of the third interdigital space of the leg bud at 7.5 days of incubation double stained with DAPI (A) and TUNEL (B) to show that the

interdigital mesoderm at this stage contains both living cells negative for TUNEL and apoptotic TUNEL-positive cells.

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5242

al., 1997; cytochrome c oxidase subunits I, II and III,

Chandra et al., 2002; cathepsin-L, Boland and Campbell,

2004; SAP18, Schwerk et al., 2003); antiapoptotic genes

(dynein protein inhibitor of neuronal nitric oxide synthase;

Chang et al., 2000; elongation factor 1-alpha 1, Talapatra

et al., 2002; protein-disulfide isomerase, Ko et al., 2002)

and genes of potential importance in phagocytosis (apoli-

ipoprotein A1; Hamon et al., 2000). In addition, Bmp5, a

member of the bone morphogenetic signaling molecules

and Dek that codes for an oncogenic-like protein, which is

activated by BMP signaling (Hollnagel et al., 1999), were

also isolated from the library. In situ hybridizations using

riboprobes generated for each one of these genes that are

potentially implicated in apoptosis revealed that except for

dek, VDAC2 and bmp5, which had interdigital expression

domains, all the remaining genes exhibited a ubiquitous

pattern of expression. In this work, we chose BMP5 for

additional functional studies.

Expression of Bmp5 in the developing limb

Bmp5 exhibited a regulated pattern of expression

throughout the different stages of the developing limb

bud. Transcripts were detected between days 3 and 4.5 of

incubation in the distal mesoderm of the limb bud (Figs. 2A

and B). From day 5, Bmp5 was also expressed in the dorsal

and ventral muscular masses of the developing limb (Figs.

2C and D). By days 5.5 and 6, transcripts were abundant in

the proximal part of the interdigital regions (Fig. 2E). From

day 7, the interdigital domains were displaced to the most

distal part of the interdigits (Figs. 2F and G) reaching the tip

of the digit by day 8.5 of incubation (Figs. 2G and H).

Interestingly, the evolution of the interdigital pattern of

expression resembles the pattern of interdigital cell death

in the chick limb autopod (not shown). An additional

domain of Bmp5 expression was found in the developing

flexor tendons of the autopod from day 6.5 of incubation

(Figs. 2F and G). At later stages of development, transcripts

of Bmp5 were detected in the perichondrium of the pha-

lanxes (2H).

The presence of Bmp5 transcripts and protein in the

regressing interdigital mesenchymal cells was further ana-

lyzed by confocal microscopy. Using this approach, we

found that the interdigital mesoderm expresses Bmp5 tran-

scripts up to the initiation of apoptosis (Fig. 3A). Using an

antibody that cross-reacts with BMP5 and BMP7, we

found that these proteins are abundant in the degenerating

mesenchyme (Fig. 3B). Furthermore, intense labeling was

conserved within the cytoplasm of the dying cells positive

for TUNEL, suggesting that BMP release into the extra-

cellular medium may be enhanced during apoptotic cell

fragmentation (Figs. 3C–E). In an attempt to check

whether BMP5/7 in the apoptotic cells undergoes proteo-

somal degradation, we performed double labeling with

anti-BMP5/7 and anti-ubiquitin antibodies. In agreement

with our initial interpretation, we found that there is no co-

localization between BMP5/7 and ubiquitin both in healthy

(Fig. 3F) and in apoptotic cells (Fig. 3G), suggesting that

BMPs might remain biologically active in degenerating

cells.

BMP5 promotes apoptosis of undifferentiated mesoderm

and cartilage growth

To check the involvement of BMP5 in apoptosis, we

analyzed the effects of local application of BMP5 protein in

the undifferentiated mesenchyme of the developing limb.

For this purpose, heparin beads incubated in human recom-

binant BMP5 protein were implanted at different locations

within the limb bud. In these experiments, an intense area of

apoptosis was induced 10 h after the treatment, when the

beads were implanted in the undifferentiated mesoderm of

the anterior or posterior margin of the limb between days 3.5

and 5 of incubation (Figs. 4A and B), or in the interdigital

Table 1

List of the genes identified in the screening of a cDNA library of the third

interdigital space of 7.5-day chicken leg buds with correspondent

homologous specie

Actin related protein 2/3 complex subunit

3 (ARP2/3)

Homo sapiens

Acyl-CoA dehydrogenase (Long-chain) Sus scrofa

Adhesion protein RA175C Gallus gallus

Alpha-D globin Gallus gallus

Apolipoprotein AI Gallus gallus

ATP synthase, H+ transporting,

mitochondrial F0 complex, subunit c

(subunit 9)

Mus musculus

BAMBI protein Xenopus laevis

Basic cytosolic protein Bos taurus

Beta-actin Anser anser

Bithoraxoid-like protein/HSPC162 protein Homo sapiens

Bone Morphogenetic Protein-5 Gallus gallus

Bone Morphogenetic Protein-7 Gallus gallus

Bromodomain-containing protein BP75 Mus musculus

Cathepsin L Engraulis japonicus

Cct5-prov protein Xenopus laevis

CDC10 cell division cycle 10 homolog Homo sapiens

Cellular nucleic acid binding protein

(CNBP)

Gallus gallus

Centrin (CDC31 homolog) Homo sapiens

CGI-152 protein Homo sapiens

Chaperonin subunit 5 (epsilon) Mus musculus

Chromobox protein (CHCB2) Gallus gallus

Chromosome 13 open reading frame 12

(C13orf12)

Homo sapiens

Clone cerebral inh no syntase Homo sapiens

Clone WAG-69H2 Gallus gallus

Clone XXbac-97F19 Gallus gallus

Cofactor required for Sp1 transcriptional

activation

Gallus gallus

Collagen Type-1 Gallus gallus

CpG island associated sequence Homo sapiens

Cytochrome b Gallus gallus

Cytochrome c oxidase subunit I Gallus gallus

Cytochrome c oxidase subunit II Gallus gallus

Cytochrome c oxidase subunit III Gallus gallus

Damage-specific DNA binding protein 1

(DDB1)

Rattus norvegicus

DC6 protein Rattus norvegicus

DEAD-box RNA helicase Gallus gallus

Degrnerin Caenorhabditis elegans

DEK oncogene Homo sapiens

Diazepam binding protein DBIgene Gallus gallus

Dolichyl-phosphate mannosyltransferase

polypeptide 1 (DPM1)

Homo sapiens

Dynein/protein inhibitor of neuronal

nitric oxide synthase (PIN) gene

Homo sapiens

EDRK-rich factor Homo sapiens

Elongation factor 1-alpha 1 Gallus gallus

Enhancer of rudimentary homologue

(ERH)

Xenopus laevis

Equarin-S Gallus gallus

Erythropoietin 4 Mus musculus

EURL protein homolog (early undifferentiated

retina and lens)

Gallus gallus

FLASH Mus musculus

Glucosamindase Homo sapiens

Golgi complex associated protein 1 Homo sapiens

Growth-related translationally controlled tumor

protein (pCHK23)

Gallus gallus

GTP binding protein (ARL3) Homo sapiens

Heat shock protein 1 (chaperonin) Mus musculus

Heat shock protein 10 Gallus gallus

hematological and neurological expressed

sequence 1

Mus musculus

Hemoglobin alpha-A chain Gallus gallus

Hemoglobin alpha-D chain Gallus gallus

Hemoglobin beta chain Gallus gallus

Heterogeneous nuclear ribonucleoprotein

H3 isoform

Mus musculus

High mobility group 1 protein Gallus gallus

Homeobox protein MSX-2 (CHOX-8) Gallus gallus

HSPC036 protein Homo sapiens

Hypothetical prot 121 Homo sapiens

Hypothetical protein HSPC014 Homo sapiens

Hypothetical protein HSPC148 Homo sapiens

Inosine monophosphate dehydrogenase

1 isoform b

Homo sapiens

Intersectin short isoform Homo sapiens

Lactate dehydrogenase B Gallus gallus

LIM homeobox protein cofactor (CLIM-1) Homo sapiens

Maternal G10 transcript Homo sapiens

Mesoderm development candidate 1 Mus musculus

MHC class II associated invariant chain Gallus gallus

Mitogen-activated protein kinase 6 Homo sapiens

NADH dehydrogenase (ubiquinone) 1 beta

subcomplex, 9,

Homo sapiens

NADH-ubiquinone oxidoreductase 30 KDA Homo sapiens

Non-muscle caldesmon Gallus gallus

Nuclease sensitive element binding protein 1

(Y box binding protein-1)

Gallus gallus

Nucleolar prot #38 Gallus gallus

Okadin; basic helix-loop-helix transcription

factor 6

Mus musculus

Ornithine decarboxylase antizyme inhibitor Homo sapiens

Pericentriolar material-1 (PCM-1 gene) Gallus gallus

Perioedoxin-1 proliferation associated Homo sapiens

Phosphotidylinositol transfer protein, beta Homo sapiens

PNAS-110 Homo sapiens

Poly(A) binding protein, cytoplasmic 1

(PABPC1)

Homo sapiens

Polyglutamine-containing C14ORF4 gene Homo sapiens

Polyubiquitin gene Ub II Gallus gallus

Prematurely terminated mRNA decay

factor-like

Homo sapiens

Prkdc gene for DNA-dependent protein

kinase catalytic Subunit

Gallus gallus

PRO0529 protein Homo sapiens

Pro-alpha-1 (I) collagen Gallus gallus

Probable protein Medicago sativa

Propionyl CoenzymeA carboxylase, beta

polypeptide (Pccb)

Rattus norvegicus

Proteosome 28 kDa subunit homolog Gallus gallus

Protein disulfide isomerase (PDI) Gallus gallus

Putative transmembrane protein (NMA) Homo sapiens

RAC GTPase-activating protein Homo sapiens

Radixin pseudogene 2 (RDXP2) Homo sapiens

RAN protein Homo sapiens

Reelin Gallus gallus

Restricted expression proliferation

associated protein-100

Gallus gallus

Reticulocalcin Gallus gallus

RNA polymerase II Rattus norvegicus

RUN and FYVE domain-containing 1 Homo sapiens

Table 1 (continued)

(continued on next page)

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–52 43

SAP18, sin3 associated polypeptide p18 Gallus gallus

SET translocation (myeloid leukemia-associated) Homo sapiens

Small EDRK-rich factor 1A

(telomeric)

Homo sapiens

Small nuclear ribonucleoprotein

polypeptide A

Mus musculus

Smooth muscle caldesmon Gallus gallus

Spermidine/spermine N1-acetyltransferase

(SSAT)

Gallus gallus

Stathmin Gallus gallus

Tax responsive element binding protein Homo sapiens

TB2 gene Homo sapiens

T-cell leukemia virus enhancer factor Homo sapiens

THOC4 protein Homo sapiens

TIP120 protein Homo sapiens

Toll-like receptor 9 Felis catus

Translocating chain-associated membrane

protein (TRAM)

Gallus gallus

Tyrosine 3-monooxygenase/tryptophan

5-monooxygenase activation protein

Homo sapiens

Ubiquitin C Rattus norvegicus

Ubiquitin I (UbI) gene Gallus gallus

Ubiquitin specific protease 24 (USP24) Homo sapiens

Ubiquitin specific protease 8 (USP8) Homo sapiens

Ubiquitin-conjugating enzyme E2 variant 2 Homo sapiens

Ubiquitin-fusion protein Gallus gallus

Variant of TSC-22 Gallus gallus

Virus similar to variola C15Land vaccinia F11 Molluscum contagiosum

Voltage-dependent anion channel (VDAC2) Gallus gallus

Voltage-gated K channel beta subunit 4.1 Homo sapiens

Wnt-5a Gallus gallus

WSB-1 form Homo sapiens

X Fragile Chromosome 17/11 Homo sapiens

Table 1 (continued)

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5244

spaces (Figs. 4D and E) before the onset of interdigital cell

death (days 5.5 and 6 of incubation). The intensity of

apoptosis in the early limb after implantation of BMP5

beads was followed by a deficient development of the

skeleton (Fig. 4G).

Local application of the BMP antagonist Noggin into the

interdigital mesoderm has been found to cause an intense

inhibition of INZ leading to severe soft tissue syndactyly

(Merino et al., 1998). To confirm the hypothesis that BMP5

could be physiologically involved in the control of apopto-

sis, we checked the ability of Noggin to block apoptosis

induced by exogenous BMP5. For this purpose, a BMP5

bead and a Noggin bead were implanted into the mesoderm

close to each other. As expected, this double treatment

caused a strong inhibition of the area of apoptosis induced

by the BMP5 bead (Figs. 4C and F).

Previous studies have shown that those BMPs implicated

in the control of limb programmed cell death have an

opposite effect promoting growth and differentiation when

they are applied in the prechondrogenic aggregates. To test

whether this effect was also shared by BMP5, beads bearing

recombinant protein were implanted at the tip of differen-

tiating digits (days 5.5 and 6 of incubation). This treatment

caused a dramatic enlargement of the digital cartilages (Figs.

4H and I).

BMP5 upregulates genes implicated in interdigital

apoptosis

To discard the possibility that apoptosis induced by

BMP5 was mediated through the activation of Bmp2,

Bmp4 or Bmp7, we analyzed whether interdigital implanta-

tion of BMP5 beads regulated the expression of those genes.

None of them were regulated during the first 12 h after the

treatment (not shown). In contrast, the expression of the

pseudoreceptor BAMBI (BMP and activin membrane-bound

inhibitor) that is supposed to constitute a synexpression

gene group with Bmp4 (Grotewold et al., 2001) and nega-

tively modulates BMP signaling (Onichtchouk et al., 1999)

was intensely upregulated by BMP5 beads (Fig. 5A). To

further characterize the function of BMP5 in INZ, we next

explored the effect of BMP5 beads on the expression of

Msx2, Dkk, Snail and Fgfr3 (Montero et al., 2001; Grote-

wold and Ruther, 2002) which were all previously impli-

cated in interdigital tissue regression acting under the

control of BMP signaling. As shown in Figs. 5B–E, all

these genes were intensely upregulated during the first 10

h after interdigital implantation of BMP5 beads.

Interdigital function of BMP5 implicates both Smad and

MAPK signaling

Recently, it has been found that in addition to the well-

characterized Smad signaling pathway, BMPs may also

exert their functions through the activation of the MAPK

pathway (reviewed by Derynck and Zhang, 2003). Here,

we have found evidence for a role of BMP5 activating

both pathways in the interdigital mesoderm (similar experi-

ments were also performed with BMP7 with identical

results).

The cytoplasm of interdigital mesodermal cells was

intensely labeled with a polyclonal antibody against Smad1,

5 and 8 both before and during the apoptotic process (Fig.

6A). In contrast, nuclear labeling for phospho-Smad1 was

almost absent before tissue regression (day 6 of incubation

or earlier; Fig. 6C). However, phospho-Smad1 expression

was significantly increased in the following developmental

stages, concomitantly with the establishment of the apopto-

tic process (Fig. 6B), and upon interdigital application of

BMP5 beads (Fig. 6D). This increase concerned both the

intensity of nuclear labeling in each cell and the number of

labeled cells.

To study the implication of Smads at RNA level, we

cloned the chick BMP-responsive Smad genes (1, 5 and 8)

and the inhibitory Smads. Smad1 was expressed at low

levels in the interdigital mesoderm (Fig. 7A) and this

expression was not affected by BMP treatments. Expression

of Smad5 was undetectable by whole-mount in situ hybrid-

ization both in normal and BMP-treated interdigits (not

shown). In contrast, Smad8 was highly expressed through-

out the whole process of interdigital tissue regression (Fig.

7E). In addition, its expression was intensely downregulated

Fig. 2. (A–H) Expression of Bmp5 in the developing leg bud at days 3.5 (A), 4 (B), 5 (C, dorsal view; D, lateral view), 6 (E), 7.5 (F), 8 (G) and 9 (H). At early

stages, transcripts are observed in the undifferentiated distal mesenchyme (A, B). Next in development in addition of being expressed in the distal mesoderm,

transcripts are also observed in the dorsal and ventral muscle masses (arrows in C and D). As digits become identifiable, well-defined domains are

progressively observed in the interdigital regions (E–G) which become displaced towards the tip of the digits in coincidence with the elimination of the most

distal mesenchyme of the limb (H). Note the presence of additional domains of expression in the developing tendons (F, G) and in the perichondrium of the

differentiating phalanxes (arrows, H).

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–52 45

by Noggin beads (Fig. 7F) and upregulated by application of

BMP beads (Fig. 7G). The upregulation caused by BMP

beads was also observed in early limb buds (not shown)

preceding the induction of mesodermal apoptosis. Finally,

the inhibitor Smad6 was expressed in the interdigital meso-

derm (Fig. 7B), as previously described (Yamada et al.,

1999), and as Smad8, its expression was upregulated by

BMP beads (Fig. 7C) and downregulated by Noggin (not

shown).

Erk, c-jun and p38 are the major components of the

MAPK signaling pathway. The involvement of Erk kinase

in limb development has been previously studied by Kawa-

kami et al. (2003) that showed that it is implicated in

apoptosis and that its effect is dependent on FGF signaling.

Here, we analyzed whether the other MAPK pathways,

namely, c-jun and p38, are involved in the apoptotic BMP

pathway responsible for interdigital tissue regression.

Immunolabeling for phospho-c-jun was positive mainly

within the cytoplasm of the interdigital mesoderm but its

expression did not change either during physiological de-

generation or after BMP treatments (not shown). In contrast,

immunolabeling for the MAPK p38 in interdigital meso-

derm revealed that phophorylated p38 (phospho-p38) was

present within the nucleus of the interdigital mesodermal

cells. Before the onset of apoptosis, nuclear phospho-p38

was not abundant in the interdigital mesodermal cells (Fig.

8A); however, the intensity of the immunolabeling increased

in the subsequent stages, and by day 7.5 of incubation,

immunolabeling was very intense (Fig. 8B). To test whether

this expression was regulated by BMPs, we analyzed

changes in phospho-p38 upon interdigital application of

BMP at day 5.5 of incubation. As shown in Figs. 8C and

D, application of BMP dramatically increases the amount of

nuclear phospho-p38 (after 1 and 6 h, respectively).

The functional implication of p38 in interdigital tissue

regression was analyzed by local treatment with the p38

inhibitor, SB203580. Our results reveal that there are two

categories of genes, p38 dependent and p38 independent,

induced by BMPs in the interdigital mesoderm. Interdigital

BMP-mediated upregulation of Dkk, Snail and Fgfr3 was

severely impaired after the double treatments with a BMP

bead and a bead incubated in SB203580 (Figs. 5H–J), while

the expressions of BAMBI, Msx2, Smad6 and Smad8 were

not affected in these experiments (Figs. 5F, G and 7D, H).

The inhibitory effect of SB203580 beads over gene

expression was only observed during the first 10 h after

the treatment. Neither cell death nor chondrogenesis were

altered in long-term in vivo treatments, suggesting that none

of the beads tested (affigel blue, heparin and AG1-X2) were

efficient retaining and slowly delivering the drug. In accor-

Fig. 3. Confocal images of squashed interdigital mesodermal cells showing the expression of bmp5 transcripts at day 6.5 of incubation (A). (B)

Immunolabeling for BMP5 (red) combined with TUNEL (green) at day 7 of incubation. Note the intense cytoplasmic labeling both in living and in TUNEL-

positive cells. (C–E) Apoptotic mesodermal cell after TUNEL and BMP5 immunolabeling. D and E show independent channels for BMP5 and TUNEL,

respectively. (F, G) Double immunolabeling for ubiquitin (red) and BMP5 (green). Note the lack of co-localization between the two proteins both in the

rounded apoptotic cell (F) and in the healthy stellate cell (G).

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5246

dance with this interpretation, local application of SB203580

beads was unable to disturb digit chondrogenesis but addi-

tion of SB203580 to micromass cultures severely impaired

both spontaneous chondrogenesis and BMP-mediated chon-

drogenesis (not shown). However, other explanations for the

absence of long-term effects of SB203580 beads can be

proposed. Thus, it is possible that not all forms of p38

present in the interdigit were inhibited by SB203580 (Stan-

ton et al., 2003); alternatively, it is also possible that after the

first 10 h of treatment, the SMAD pathway restores normal

BMP function.

Discussion

The third interdigit of the chick leg bud was selected in

this study to obtain a cDNA library due to its potential

advantages to detect genes responsible for apoptosis. First,

the spatial arrangement of the area of interdigital cell death

is very well defined and can be isolated surgically without

risk of including nonregressing tissues. Second, apoptosis

exhibits a constant and relatively short (48 h) temporal

pattern of distribution. And third, cells destined to die

constitute a uniform population of undifferentiated meso-

derm with potential to differentiate into cartilage. At the

stage selected for the generation of the cDNA library, the

interdigital mesoderm contained both preapoptotic cells

committed to die within the next few hours and apoptotic

cells at different stages of degeneration. By exploring this

library, we identified several genes potentially implicated in

the regulation of interdigital cell death. These genes in-

cluded: components of the proteosome which is an struc-

ture responsible for protein degradation implicated in

apoptosis (see Lee and Peter, 2003); genes implicated in

DNA repair (Prkdc gene for DNA-dependent protein ki-

nase; Mori et al., 2001); genes implicated positively in

apoptosis (caspase8 binding protein FLASH, Imai et al.,

1999; damage-specific DNA binding protein, Sun et al.,

2002; voltage-dependent anion channel, VDAC2, Sugiyama

et al., 2002; spermidine/spermine N1-acetyltransferase, Ha

et al., 1997; cytochrome c oxidase subunit I, II and III,

Chandra et al., 2002; cathepsin-L, Boland and Campbell,

2004; SAP18, Schwerk et al., 2003); antiapoptotic genes

(dynein protein inhibitor of neuronal nitric oxide synthase;

Chang et al., 2000; elongation factor 1-alpha 1, Talapatra

et al., 2002) and genes of potential importance in phago-

cytosis (apolipoprotein A1; Hamon et al., 2000). By in situ

hybridization, we noted that except for VDAC2, a gene

Fig. 4. (A– I) Effects of local application of beads bearing BMP5 into the limb bud. (A, D) TUNEL-positive apoptosis induced by BMP5 beads (*) implanted in

the anterior margin mesoderm of the wing bud at day 3.5 of incubation (A) and in the interdigital mesoderm before the onset of physiological apoptosis (D). (B,

C) Experimental limbs vital stained with neutral red to show the area of cell death induced after implanting a BMP5 bead (*; B) and its inhibition by co-

implantation of a Noggin bead (white *; C). E–F show similar results obtained when a BMP5 bead alone (E) or in combination with a Noggin bead (F) is

implanted into the third interdigit. (G) Skeletal pattern of the wing bud lacking the radius (arrow) consequence of the experiment illustrated in B. (H, I) Neutral

red staining of cell death (H) and skeletal staining with Alcian green (I) of limbs after implantation of a BMP5 bead at the tip of digit 3. Note that cell death is

restricted to the most distal mesenchyme of the tip of the digit (H) while the digit cartilage exhibits intense overgrowth (*, I).

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–52 47

responding to the Bcl-2 family of apoptotic regulators

involved in the control of mitochondrial permeability and

Dek, an oncogenic-like gene, which have interdigital ex-

Fig. 5. (A–E) In situ hybridizations for BAMBI (A), Msx2 (B), Dkk (C), Snail (D)

showing upregulation of all the genes. (F–J) Effect of beads soaked in SB203580

BMP-mediated upregulation of Dkk (H), Snail (I) and Fgfr3 (J) is inhibited by the a

(F) and Msx2 (G) remains unchanged.

pression domains, the remaining genes were ubiquitously

expressed in the limb. Therefore, further studies are re-

quired to clarify their function in the apoptotic cascade of

and Fgfr3 (E) 8 h after the application of a BMP5 bead in the third interdigit

co-implanted with BMP5 beads in the expression of these genes. Note that

pplication of the p38 inhibitor SB203580, while the upregulation of BAMBI

Fig. 6. Interdigital mesodermal cells immunolabeled for Smad proteins. (A) Double labeling for non-phosphorylated Smads 1, 5, 8 (red) and panhistone (green)

used as a nuclear marker, of 7.5 days interdigital mesodermal cells. (B) Double labeling for phosphorylated Smad1 (green) and panhistone (red) at day 7.5 of

development showing co-localization of both proteins in several mesodermal cells (yellow nuclei). (C, D) Control (C) and BMP-treated (D) interdigital

mesodermal cells at day 6 of incubation. Note the increase in labeling intensity for phospho-Smad1 and in the number of labeled cells 90 min after BMP treatment.

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5248

the interdigit. However, it is remarkable the presence in our

library of lysosomal genes, such as cathepsin-L, which

might provide an alternative death pathway to caspases

Fig. 7. Expression and regulation of Smads in the interdigital mesoderm. (A) Ex

Smad6 at day 7 of incubation. The developing joints show also a tenuous expres

3 h after the treatment with a BMP bead that is not modified by co-implantation

of incubation. Note that in addition to the intense interdigital domains, a more r

expression is inhibited by Noggin beads (F) and upregulated by BMP beads (G). T

induced upregulation (H). Arrows indicate SB203580 bead.

(Boland and Campbell, 2004; Dietrich et al., 2003) explain-

ing the maintenance of interdigital degeneration in mice

deficient for Apaf1 (Chautan et al., 1999).

pression of Smad1 at 5.5 days of incubation. (B) Interdigital expression of

sion nonappreciable in the picture. (C) Upregulation of Smad6 expression

with a bead incubated in SB203580 (D). (E) Expression of Smad8 at day 6

educed expression is also observed at the tip of the digits. The interdigital

he double treatment with BMP and SB203580 bead does not inhibit BMP-

Fig. 8. Immunolabeling with phospho-p38 (green) combined with lamina B (red) that stains the nuclear envelope in control and BMP-treated interdigital

mesodermal cells. A and B are control mesodermal cells at days 5.5 and 7.5 of incubation, respectively, showing the increase of the nuclear expression domain

during interdigital tissue regression. (C, D) Interdigital mesoderm at day 5.5 of incubation, 1 h (C) and 6 h (D) after BMP treatment. Note the intense labeling in

C and D in comparison with the nontreated cells at the same stage (A).

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–52 49

In contrast with the predominant ubiquitous expression

of the genes mentioned above, Bmp5 exhibited a pattern of

expression closely coincident with interdigital apoptosis.

Previous studies revealed that BMP2, BMP4 and BMP7

are triggering signals for physiological cell death in the

developing limb (Ganan et al., 1996; Guha et al., 2002;

Kawakami et al., 1996; Macias et al., 1997; Yokouchi et al.,

1996; Zou and Niswander, 1996; Zou et al., 1997). This

conclusion was deduced from their pattern of expression

and by gain- and loss-of-function experiments (Guha et al.,

2002; Merino et al., 1998; Yokouchi et al., 1996; Zou and

Niswander, 1996). By using the same criteria, we found that

BMP5 constitutes a further member of the BMP family

implicated in the control of limb programmed cell death.

Previously, the implication of BMP5 in embryonic cell

death has been reported only in the developing nervous

system (Golden et al., 1999).

Taking into account that at least four different members

of the BMP family (BMP2, BMP4, BMP5 and BMP7) are

involved in the control of limb programmed cell death, it is

not surprising the absence of an interdigital phenotype in

mice deficient for Bmp5 (King et al., 1994; Solloway and

Robertson, 1999). It must be remarked that all these BMPs

share a high identity. BMP2 and BMP4 share 92% identity,

while BMP5 and BMP7 share 87% identity. This high

redundancy in BMPs in the regressing interdigital tissue

suggests that BMPs exert their function by forming hetero-

dymeric complexes as has been found in other biological

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5250

processes (Butler and Dodd, 2003; Katagiri et al., 1998).

Furthermore, on the basis of the phenotypes of single and

double KO mice for Bmp7, Bmp5 and taking into account

their distinct, but rather overlapping, pattern of expression in

the early embryo, it has been proposed that BMP5 might

modulate the effect of BMP7, creating zones of increased

signaling within the Bmp7 expression domain (Solloway

and Robertson, 1999). An interesting observation in our

study was the identification of intense immunolabeling for

BMP5/7 proteins within the cytoplasm of TUNEL-positive

apoptotic cells. This finding suggests that delivery of BMPs

might be potentiated during cell fragmentation associated

with apoptosis explaining why interdigital apoptosis pro-

gresses at advanced stages of degeneration when the amount

of healthy cells, able to secrete BMPs, are dramatically

reduced. A comparable release of growth factors associated

with mechanical disruption of plasma membrane has been

observed in endothelial cell cultures (McNeil et al., 1989).

An additional task of this study was to explore the

intracellular downstream effectors of BMPs in the regress-

ing interdigit. There is now increasing evidence showing

that in addition to the well-characterized Smad signaling

pathway, BMPs, and other members of the TGFh super-

family, may also activate MAP kinases (Derynck and

Zhang, 2003). Furthermore, it has been suggested that in

some cases, the cellular response induced by BMPs may be

the result of the balance between a counteracting function

between Smad and MAPK pathways (see, Massague, 2003).

However, both pathways may act in a cooperative fashion.

Thus, in mouse hybridoma MH60 cells, BMPs induce

apoptosis through p38 kinase and this cell death is inhibited

when smad6 is overexpressed (Kimura et al., 2000).

The presence of Smad1/5/8 proteins in the interdigital

mesoderm along with the progressive abundance of nuclear

phosphorylated Smad1 during regression, and its potentia-

tion by application of BMP beads, indicate that Smad

proteins exert a central role in the control of interdigital

tissue regression by BMPs. In support to such an implica-

tion of Smad signaling in interdigital tissue regression, we

have also found that Smad6 and Smad8 exhibit interdigital

expression domains regulated by BMPs.

Our findings also provide evidence for the activation of

MAPK p38 during interdigital tissue regression in response

to BMPs. Phosphorylated p38 is distributed throughout the

nucleoplasm of the interdigital mesoderm. Furthermore, as

observed for phospho-Smad1, nuclear labeling for phospho-

p38 increases during tissue regression and upon application

of BMP beads. Double treatments with BMPs and the

inhibitor of p38 indicate that this kinase mediates the

upregulation of some of the genes activated by BMPs in

the interdigit, including Dkk, Snail and Fgfr3, but not

BAMBI, Msx2, Smad6 and Smad8.

Although we cannot prove that all forms of p38 were

inhibited by SB203580 in our experiments (see review by

Stanton et al., 2003), the phosphorylation of Smad1 in the

interdigital mesenchyme upon application of BMPs, along

with the occurrence of specific p38-independent interdigital

domains of expression of Smad8 and Smad6, are in favor of

a synergic/cooperative rather then antagonistic function of

both cellular pathways in the regulation of the cellular

response to BMPs in the developing limb.

Although Bmp5 KO mice lack a limb phenotype, our

study is indicative of a function of BMP5 in the differen-

tiation of the limb skeleton. We have found that Bmp5 is

expressed in the perichondrium of the developing phalanxes

and at the tip of the digits, and application of BMP5 in the

differentiating cartilages promotes a dramatic growth. The

intense inhibition of the chondrogenic effect of BMP5 (and

of BMP7) by SB203580 in limb mesoderm micromass

cultures indicates that p38 is an important effector for the

chondrogenic effect of BMPs. This finding is in agreement

with previous studies showing the importance of p38 in

chondrogenesis (Stanton et al., 2003). Furthermore, this

observation argues against a counteracting function between

Smads and p38 regulating the different cellular responses

mediated by BMPs in the limb mesoderm.

Finally, we also show in this study that Bmp5 is expressed

in precise domains in the developing muscle masses and in

the autopodial tendons. Again in these regions, and consis-

tent with the absence of phenotypes in mice deficient for

Bmp5 (King et al., 1994), other BMPs are present controlling

tissue differentiation (Amthor et al., 2002).

Acknowledgments

Thanks are due to A. Economides from Regeneron Phar-

maceuticals for human recombinant Noggin; to C. Heldin

for the antibody against phospho-Smad1; to S. Perez-

Mantecon, M. Fernandez and A. Ibaseta for technical

assistance; and to Hannu Mansukoski for reading this

manuscript. This work was supported by a grant from the

Ministerio de Ciencia y Tecnologıa to J.M.H. (BMC2002-

02346). V.Z.L. and J.R.L. are supported by grants from the

Fundac�ao para a Ciencia e a Tecnologia, Portugal.

References

Amthor, H., Christ, B., Rashid-Doubell, F., Kemp, C.F., Lang, E., Patel, K.,

2002. Follistatin regulates bone morphogenetic protein-7 (BMP-7) ac-

tivity to stimulate embryonic muscle growth. Dev. Biol. 243, 115–127.

Boland, B., Campbell, V., 2004. Abeta-mediated activation of the apoptotic

cascade in cultured cortical neurones: a role for cathepsin-L. Neurobiol.

Aging 25, 83–91.

Butler, S.J., Dodd, J., 2003. A role for BMP heterodimers in roof plate-

mediated repulsion of commissural axons. Neuron 38, 389–401.

Carrio, R., Lopez-Hoyos, M., Jimeno, J., Benedict, M.A., Merino, R.,

Benito, A., Fernandez-Luna, J.L., Nunez, G., Garcia-Porrero, J.A., Me-

rino, J., 1996. A1 demonstrates restricted tissue distribution during

embryonic development and functions to protect against cell death.

Am. J. Pathol. 149, 2133–2142.

Cecconi, F., Alvarez-Bolado, G., Meyer, B.I., Roth, K.A., Gruss, P., 1998.

Apaf1 (CED-4 homolog) regulates programmed cell death in mamma-

lian development. Cell 94, 727–737.

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–52 51

Chandra, D., Liu, J.W., Tang, D.G., 2002. Early mitochondrial activation

and cytochrome c up-regulation during apoptosis. J. Biol. Chem. 277,

50842–50854.

Chang, Y.W., Jakobi, R., McGinty, A., Foschi, M., Dunn, M.J., Sorokin,

A., 2000. Cyclooxygenase 2 promotes cell survival by stimulation of

dynein light chain expression and inhibition of neuronal nitric oxide

synthase activity. Mol. Cell. Biol. 20, 8571–8579.

Chautan, M., Chazal, G., Cecconi, F., Gruss, P., Golstein, P., 1999. Inter-

digital cell death can occur through a necrotic and caspase-independent

pathway. Curr. Biol. 9, 967–970.

Chen, Y., Zhao, X., 1998. Shaping limbs by apoptosis. J. Exp. Zool. 282,

691–702.

Chimal-Monroy, J., Montero, J.A., Ganan, Y., Macias, D., Garcia-Porrero,

J.A., Hurle, J.M., 2002. Comparative analysis of the expression and

regulation of Wnt5a, Fz4, and Frzb1 during digit formation and in

micromass cultures. Dev. Dyn. 224, 314–320.

Crocoll, A., Herzer, U., Ghyselinck, N.B., Chambon, P., Cato, A.C.B.,

2002. Interdigital apoptosis and downregulation of BAG-1 expression

in mouse autopods. Mech. Dev. 111, 149–152.

Derynck, R., Zhang, Y.E., 2003. Smad-dependent and Smad-independent

pathways in TGF-h family signalling. Nature 425, 577–584.

Dietrich, N., Thastrup, J., Holmberg, C., Gyrd-Hansen, M., Fehrenbacher,

N., Lademann, U., Lerdrup, M., Herdegen, T., Jaattela, M., Kallunki, T.,

2003. JNK2 mediates TNF-induced cell death in mouse embryonic

fibroblasts via regulation of both caspase and cathepsin protease path-

ways. Cell Death Differ. 11, 301–313.

Drouin, R., Lemieux, N., Richer, C.L., 1991. Chromosome condensation

from prophase to late metaphase: relationship to chromosome bands and

their replication time. Cytogenet. Cell Genet. 57, 91–99.

Dunker, N., Schmitt, K., Krieglstein, K., 2002. TGF-beta is required for

programmed cell death in interdigital webs of the developing mouse

limb. Mech. Dev. 113, 111–120.

Dupe, V., Ghyselinck, N.B., Thomazy, V., Nagy, L., Davies, P.J., Chambon,

P., Mark, M., 1999. Essential roles of retinoic acid signalling in inter-

digital apoptosis and control of BMP-7 expression in mouse autopods.

Dev. Biol. 208, 30–43.

Ferrari, D., Lichtler, A.C., Pan, Z., Dealy, C.N., Upholt, W.B., Kosher,

R.A., 1998. Ectopic expression of Msx2 in posterior limb bud meso-

derm impairs limb morphogenesis while inducing Bmp4 expression,

inhibiting cell proliferation, and promoting apoptosis. Dev. Biol. 197,

12–24.

Ganan, Y., Macias, D., Duterque-Coquillaud, M., Ros, M.A., Hurle, J.M.,

1996. Role of TGFhs 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, 2349–2357.

Garcia-Domingo, D., Leonardo, E., Grandien, A., Martinez, P., Albar, J.P.,

Izpisua-Belmonte, J.C., Martinez-A, C., 1999. DIO-1 is a gene involved

in onset of apoptosis in vitro, whose misexpression disrupts limb de-

velopment. Proc. Natl. Acad. Sci. U. S. A. 96, 7992–7997.

Garcia-Martinez, V., Macias, D., Ganan, Y., Garcia-Lobo, J.M., Francia,

M.V., Fernandez-Teran, M.A., Hurle, J.M., 1993. Internucleosomal

DNA fragmentation and programmed cell death (apoptosis) in the inter-

digital tissue of the embryonic chick leg bud. J. Cell Sci. 106, 201–208.

Golden, J.A., Bracilovic, A., McFadden, K.A., Beesley, J.S., Rubenstein,

J.L., Grinspan, J.B., 1999. Ectopic bone morphogenetic proteins 5 and 4

in the chicken forebrain lead to cyclopia and holoprosencephaly. Proc.

Natl. Acad. Sci. U. S. A. 96, 2439–2444.

Grotewold, L., Plum, M., Dildrop, R., Peters, T., Ruther, U., 2001. Bambi

is coexpressed with Bmp-4 during mouse embryogenesis. Mech. Dev.

100, 327–330.

Grotewold, L., Ruther, U., 2002. The Wnt antagonist Dickkopf-1 is regu-

lated by Bmp signalling and c-Jun and modulates programmed cell

death. EMBO J. 21, 966–975.

Guha, U., Gomes, W.A., Kobayashi, T., Pestell, R.G., Kessler, J.A., 2002.

In vivo evidence that BMP signalling is necessary for apoptosis in the

mouse limb. Dev. Biol. 249, 108–120.

Ha, H.C., Woster, P.M., Yager, J.D., Casero Jr., R.A., 1997. The role of

polyamine catabolism in polyamine analogue-induced programmed cell

death. Proc. Natl. Acad. Sci. U. S. A. 94, 11557–11562.

Hamon, Y., Broccardo, C., Chambenoit, O., Luciani, M.F., Toti, F., Chas-

lin, S., Freyssinet, J.M., Devaux, P.F., McNeish, J., Marguet, D.,

Chimini, G., 2000. ABC1 promotes engulfment of apoptotic cells

and transbilayer redistribution of phosphatidylserine. Nat. Cell Biol.

2, 399–406.

Hollnagel, A., Oehlmann, V., Heymer, J., Ruther, U., Nordheim, A., 1999.

Id genes are direct targets of bone morphogenetic protein induction in

embryonic stem cells. J. Biol. Chem. 274, 19838–19845.

Imai, Y., Kimura, T., Murakami, A., Yajima, N., Sakamaki, K., Yone-

hara, S., 1999. The CED-4-homologous protein FLASH is involved

in Fas-mediated activation of caspase-8 during apoptosis. Nature 398,

777–785.

Jacobson, M.D., Weil, M., Raff, M.C., 1996. Role of Ced-3/ICE-family

proteases in staurosporine-induced programmed cell death. J. Cell Biol.

133, 1041–1051.

Katagiri, T., Boorla, S., Frendo, J.L., Hogan, B.L., Karsenty, G., 1998.

Skeletal abnormalities in doubly heterozygous Bmp4 and Bmp7 mice.

Dev. Genet. 22, 340–348.

Kawakami, Y., Ishikawa, T., Shimabara, M., Tanda, N., Enomoto-Iwamoto,

M., Iwamoto, M., Kuwana, T., Ueki, A., Noji, S., Nohno, T., 1996.

BMP signalling during bone pattern determination in the developing

limb. Development 122, 3557–3566.

Kawakami, Y., Rodriguez-Leon, J., Koth, C.M., Buscher, D., Itoh, T.,

Raya, A., Ng, J.K., Esteban, C.R., Takahashi, S., Henrique, D.,

Schwarz, M.F., Asahara, H., Izpisua Belmonte, J.C., 2003. MKP3

mediates the cellular response to FGF8 signalling in the vertebrate

limb. Nat. Cell Biol. 5, 513–519.

Kimura, N., Matsuo, R., Shibuya, H., Nakashima, K., Taga, T., 2000.

BMP2-induced apoptosis is mediated by activation of the TAK1-p38

kinase pathway that is negatively regulated by Smad6. J. Biol. Chem.

275, 17647–17652.

King, J.A., Marker, P.C., Seung, K.J., Kingsley, D.M., 1994. BMP5 and the

molecular, skeletal, and soft-tissue alterations in short ear mice. Dev.

Biol. 166, 112–122.

Ko, H.S., Uehara, T., Nomura, Y., 2002. Role of ubiquilin associated with

protein-disulfide isomerase in the endoplasmic reticulum in stress-in-

duced apoptotic cell death. J. Biol. Chem. 277, 35386–35392.

Lee, J.C., Peter, M.E., 2003. Regulation of apoptosis by ubiquitination.

Immunol. Rev. 193, 39–47.

Lee, K.K., Tang, M.K., Yew, D.T., Chow, P.H., Yee, S.P., Schneider, C.,

Brancolini, C., 1999. Gas2 is a multifunctional gene involved in the

regulation of apoptosis and chondrogenesis in the developing mouse

limb. Dev. Biol. 207, 14–25.

Lee, K.K., Leung, A.K., Tang, M.K., Cai, D.Q., Schneider, C., Brancolini,

C., Chow, P.H., 2001. Functions of the growth arrest specific 1 gene in

the development of the mouse embryo. Dev. Biol. 234, 188–203.

Lindsten, T., Ross, A.J., King, A., Zong, W.X., Rathmell, J.C., Shiels,

H.A., Ulrich, E., Waymire, K.G., Mahar, P., Frauwirth, K., Chen, Y.,

Wei, M., Eng, V.M., Adelman, D.M., Simon, M.C., Ma, A., Golden,

J.A., Evan, G., Korsmeyer, S.J., MacGregor, G.R., Thompson, C.B.,

2000. The combined functions of proapoptotic Bcl-2 family members

bak and bax are essential for normal development of multiple tissues.

Mol. Cell 6, 1389–1399.

Lussier, M., Canoun, C., Ma, C., Sank, A., Shuler, C., 1993. Interdigital

soft tissue separation induced by retinoic acid in mouse limbs cultured

in vitro. Int. J. Dev. Biol. 37, 555–564.

Macias, D., Ganan, 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.

Massague, J., 2003. Integration of Smad and MAPK pathways: a link and a

linker revisited. Genes Dev. 17, 2993–2997.

McNeil, P.L., Muthukrishnan, L., Warder, E., D’Amore, P.A., 1989.

Growth factors are released by mechanically wounded endothelial cells.

J. Cell Biol. 109, 811–822.

V. Zuzarte-Luıs et al. / Developmental Biology 272 (2004) 39–5252

Merino, R., Ganan, Y., Macias, D., Economides, A.N., Sampath, K.T.,

Hurle, J.M., 1998. Morphogenesis of digits in the avian limb is con-

trolled by FGFs, TGFbetas, and noggin through BMP signalling. Dev.

Biol. 200, 35–45.

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 programmed cell death during limb

development. Development 128, 2075–2084.

Mori, N., Matsumoto, Y., Okumoto, M., Suzuki, N., Yamate, J., 2001.

Variations in Prkdc encoding the catalytic subunit of DNA-dependent

protein kinase (DNA-PKcs) and susceptibility to radiation-induced

apoptosis and lymphomagenesis. Oncogene 20, 3609–3619.

Nakanishi, K., Maruyama, M., Shibata, T., Morishima, N., 2001. Identifi-

cation of a caspase-9 substrate and detection of its cleavage in

programmed cell death during mouse development. J. Biol. Chem.

276, 41237–41244.

Nishii, K., Tsuzuki, T., Kumai, M., Takeda, N., Koga, H., Aizawa, S.,

Nishimoto, T., Shibata, Y., 1999. Abnormalities of developmental cell

death in Dad1-deficient mice. Genes Cells 4, 243–252.

Novack, D.V., Korsmeyer, S.J., 1994. Bcl-2 protein expression during

murine development. Am. J. Pathol. 145, 61–73.

Onichtchouk, D., Chen, G., Dosh, R., Gawantka, V., Delius, H., Massague,

J., Niehrs, J.M., 1999. Silencing of TGF-h signalling by the pseudor-

eceptor BAMBI. Nature 401, 480–485.

Rodriguez-Leon, J., Merino, R., Macias, D., Ganan, Y., Santesteban, E.,

Hurle, J.M., 1999. Retinoic acid regulates programmed cell death

through BMP signalling. Nat. Cell Biol. 1, 125–126.

Sanz-Ezquerro, J., Tickle, C., 2000. Autoregulation of Shh expression and

Shh induction of cell death suggest a mechanism for modulating

polarizing activity during chick limb development. Development

127, 4811–4823.

Schwerk, C., Prasadm, J., Degenhardtm, K., Erdjument-Bromage, H.,

White, E., Tempst, P., Kidd, V.J., Manley, J.L., Lahti, J.M., Reinberg,

D., 2003. ASAP, a novel protein complex involved in RNA processing

and apoptosis. Mol. Cell. Biol. 23, 2981–2990.

Solloway, M.J., Robertson, E.J., 1999. Early embryonic lethality in

Bmp5;Bmp7 double mutant mice suggests functional redundancy with-

in the 60A subgroup. Development 126, 1753–1768.

Stanton, L.A., Underhilll, T.M., Beier, F., 2003. MAP kinases in chondro-

cyte differentiation. Dev. Biol. 263, 165–175.

Sugiyama, T., Shimizu, S., Matsuoka, Y., Yoneda, Y., Tsujimoto, Y., 2002.

Activation of mitochondrial voltage-dependent anion channel by apro-

apoptotic BH3-only protein Bim. Oncogene 21, 4944–4956.

Sun, N.K., Lu, H.P., Chao, C.C., 2002. Identification of rat DDB1, a

putative DNA repair protein, and functional correlation with its dam-

aged-D recognition activity. J. Biomed. Sci. 9, 371–380.

Talapatra, S., Wagner, J.D., Thompson, C.B., 2002. Elongation factor-1

alpha is a selective regulator of growth factor withdrawal and ER

stress-induced apoptosis. Cell Death Differ. 9, 856–861.

Yamada, M., Szendro, P.I., Prokscha, A., Schwartz, R.J., Eichele, G.,

1999. Evidence for a role of Smad6 in chick cardiac development.

Dev. Biol. 215, 48–61.

Yokouchi, Y., Sakiyama, J., Kameda, T., Iba, H., Suzuki, A., Ueno, N.,

Kuroiwa, A., 1996. BMP-2/-4 mediate programmed cell death in chick-

en limb buds. Development 122, 3725–3734.

Zou, H., Niswander, L., 1996. Requirement for BMP signalling in inter-

digital apoptosis and scale formation. Science 272, 738–741.

Zou, H., Wieser, R., Massague, J., Niswander, L., 1997. Distinct roles of

type I bone morphogenetic protein receptors in the formation and dif-

ferentiation of cartilage. Genes Dev. 11, 2191–2203.

Zuzarte-Luis, V., Hurle, J.M., 2002. Programmed cell death in the deve-

loping limb. Int. J. Dev. Biol. 46, 871–876.