RESEARCH ARTICLE
1483
Synergistic effects of different bone morphogenetic
protein type I receptors on alkaline phosphatase
induction
Hiromasa Aoki1,2, Makiko Fujii1,*, Takeshi Imamura1, Ken Yagi1, Kazuhiko Takehara2, Mitsuyasu Kato1 and
Kohei Miyazono1,‡
1Dept
of Biochemistry, The JFCR Cancer Institute, and Research for the Future Program, the Japan Society for the Promotion of Science, 1-37-1
Kami-ikebukuro, Toshima-ku, Tokyo 170-8455, Japan
2Dept of Dermatology, Kanazawa University School of Medicine, Takara-machi 13-1, Kanazawa 920-8640, Japan
*Present address: Laboratory of Cell Regulation and Carcinogenesis, Bldg 41, National Cancer Institute, Bethesda, MD 20892, USA
‡Author for correspondence (e-mail: [email protected])
Accepted 25 January 2001
Journal of Cell Science 114, 1483-1489 © The Company of Biologists Ltd
SUMMARY
Bone morphogenetic proteins (BMPs) are members of the
transforming growth factor-β superfamily, which regulate
the differentiation of osteoprogenitor cells. Here we show
that among members of the BMP family, BMP-4 and
growth/differentiation factor 5 (GDF-5) induce osteoblast
differentiation through the activation of three receptorregulated Smads (i.e. Smad1, Smad5 and Smad8). By
contrast, BMP-6 and BMP-7 induce alkaline phosphatase
activity through Smad1 and Smad5, but not through
Smad8. Consistent with these findings, BMP-4 induced
phosphorylation and nuclear translocation of Smad1,
Smad5 and Smad8, but BMP-6 activated only Smad1 and
Smad5. BMP-4 and GDF-5 are known to bind to activin
receptor-like kinase 3 (ALK-3) and/or ALK-6 (also termed
BMP type IA and type IB receptors, respectively), whereas
BMP-6 and BMP-7 preferentially bind to ALK-2.
Compared with the effects induced by only one of the type
I receptors, the combination of constitutively active forms
of ALK-2 and ALK-3 (or ALK-6) more strongly induced
alkaline phosphatase activity in C2C12 cells. Moreover,
addition of BMP-4 and BMP-6 to C2C12 cells resulted in
higher alkaline phosphatase activity than that of only one
of these BMPs. The combination of ALK-2 and ALK-3 also
induced higher transcriptional activity than either receptor
alone. Thus, ALK-2 and ALK-3 (or ALK-6) might
synergistically induce osteoblast differentiation of C2C12
cells, possibly through efficient activation of downstream
signaling pathways.
INTRODUCTION
cartilaginous tissues and induces the formation of cartilage and
tendon-like structures in vivo (Chang et al., 1994; Wolfman
et al., 1997). Mice lacking the GDF-5 gene are known as
brachypodism mice and have short limbs and reduced numbers
of digits. Mutations of the human GDF-5/CDMP-1 gene result
in autosomal recessive chondrodysplasia syndromes (Thomas
et al., 1997; Luyten et al., 2000). These findings suggest that
members of the BMP family have similar but distinct functions
in vivo.
Cytokines of the TGF-β superfamily bind to two different
types of serine/threonine kinase receptors, known as type II
and type I receptors (Heldin et al., 1997). Type I receptors act
downstream of type II receptors; therefore, the former
determine the specificity of the intracellular signals. Of seven
distinct type I receptors in mammals, activin receptor-like
kinase 3 (ALK-3) and ALK-6 (also termed BMP type IA and
type IB receptors, respectively) are structurally similar and
function as BMP type I receptors (ten Dijke et al., 1994).
ALK-2 is structurally less related to ALK-3 and ALK-6, but
induces osteoblast differentiation of osteoprogenitor cells,
such as C2C12 cells (ten Dijke et al., 1994; Macías-Silva et
al., 1998; Ebisawa et al., 1999). ALK-3 and ALK-6 are most
Bone morphogenetic proteins (BMPs) are members of the
transforming growth factor-β (TGF-β) superfamily, which
induce growth, differentiation, apoptosis and morphogenesis of
various types of cells (Reddi, 1998). To date, more than 20
BMP isoforms have been identified, and they can be divided
into several subtypes based on their structures (Kawabata et al.,
1998a; Kawabata and Miyazono, 2000). BMP-2, BMP-4 and
Drosophila Decapentaplegic (DPP) are prototypes of BMPs
that induce bone and cartilage formation when subcutaneously
implanted. Mice lacking the BMP-2 or BMP-4 gene die during
early developmental stages, indicating that these genes play
important roles during early embryogenesis (reviewed in
Goumans and Mummery, 2000). BMP-6, BMP-7 (also
termed osteogenic protein-1), and Drosophila Gbb-60A are
structurally related to each other, and induce bone formation
in vivo. Mice lacking the BMP-7 gene exhibit abnormalities
in kidneys, eyes and bone, indicating that BMP-7 plays
critical roles in both hard and soft tissues in vivo.
Growth/differentiation factor 5 (GDF-5; also termed cartilagederived morphogenetic protein 1 or CDMP-1) is produced by
Key words: ALK, BMP, Osteoblast differentiation, Serine/threonine
kinase receptor, Smad
1484
JOURNAL OF CELL SCIENCE 114 (8)
similar to Drosophila Thick veins (TKV), whereas ALK-2 is
related to Saxophone (SAX). Although ALK-3, ALK-6 and
ALK-2 bind most BMP ligands when they are overexpressed
in mammalian cells, a clear difference in their ligand-binding
properties can be observed when binding is examined in
nontransfected cells. BMP-4 binds to ALK-3 in human
foreskin fibroblasts and MC3T3-E1 osteoblasts, whereas
GDF-5 preferentially binds to ALK-6 in ROB-C26
osteoprogenitor cells (ten Dijke et al., 1994; Nishitoh et al.,
1996; Ebisawa et al., 1999). BMP-2, and probably BMP-4,
can also bind to ALK-6 in certain types of cells, possibly
depending on levels of expression of ALK-6 protein
(Nishitoh et al., 1996). By contrast, BMP-6 and BMP-7
exhibit binding properties distinct from those of BMP-4 and
GDF-5; they preferentially bind to ALK-2 in cells such as
C2C12 and MC3T3-E1, although weak binding to ALK-3
and ALK-6 can be detected in certain types of cells (ten Dijke
et al., 1994; Nishitoh et al., 1996). Thus, BMPs can be
divided into several subtypes based on the properties of their
binding to the type I receptors (e.g. BMPs that preferentially
bind to ALK-3 and/or ALK-6, and those that bind to
ALK-2).
The TGF-β superfamily proteins transduce signals through
Smad proteins (Heldin et al., 1997; Massagué and Wotton,
2000). Receptor-regulated Smads (R-Smads) are direct
substrates of type I receptors and are phosphorylated at
the C-terminal SSV/MS motif (Abdollah et al., 1997;
Souchelnytskyi et al., 1997). R-Smads then form heteromeric
complexes with Co-Smads, and translocate into the nucleus
where they regulate transcription of target genes. Smad1,
Smad5 and Smad8 are structurally highly related to each
other and are activated by BMPs (BMP-R-Smads). In
Drosophila melanogaster, Mad is the only R-Smad acting in
the BMP-like signaling pathway. The functional roles of
Smad1 and Smad5 have been studied in various systems
(Meersseman et al., 1997; Kretzschmar et al., 1997), whereas
those of Smad8 have been only poorly examined.
Interestingly, recent results revealed that Smad1 and Smad5
are activated by BMP-6, but that Smad8 is neither
phosphorylated nor translocated into the nucleus as a result
of BMP-6 stimulation (Ebisawa et al., 1999).
Several investigations have strongly suggested that
heterodimeric forms of BMPs, for example, heterodimers
composed of BMP-2 (or -4) and BMP-6 (or -7), are more
potent than homodimers of any of the BMPs (Aono et al., 1995;
Israel et al., 1996; Suzuki et al., 1997; Kusumoto et al., 1997).
The functional importance of BMP heterodimers has been
reported for Drosophila BMP-like molecules (Khalsa et al.,
1998; Chen et al., 1998). An intriguing possibility is that since
BMP-2/4 and BMP-6/7 bind to different type I receptors,
activation of two distinct subtypes of type I receptors, that is
ALK-3 (or ALK-6) and ALK-2, might be responsible for the
potent bioactivity of BMP heterodimers.
To address these issues, we studied the signaling activity
of different BMPs, including BMP-4, BMP-6, BMP-7 and
GDF-5. In addition, we tested combinations of different BMP
type I receptors and BMP ligands. Our results indicate that
BMP-4 and BMP-6 activate different sets of Smad proteins,
and that combinatorial activation of distinct type I receptors
results in more potent biological activity than with either
receptor alone.
MATERIALS AND METHODS
Cell culture and infection with recombinant adenoviruses
The C2C12 mouse muscle myoblast cell line was obtained from
American Type Culture Collection. The cells were maintained in
Dulbecco’s modified Eagle’s medium containing 15% fetal bovine
serum and 100 units/ml penicillin. Recombinant adenoviruses were
constructed and infected into C2C12 cells, as described previously
(Fujii et al., 1999). COS-7 cells were cultured as described (Yagi et
al., 1999) and DNAs were transfected using FuGENE6 (Boehringer
Mannheim).
Assays for alkaline phosphatase activity
BMP-4 was purchased from R&D systems. BMP-6, BMP-7 and GDF5 were kindly provided by T. K. Sampath (Creative BioMolecules
Inc.). Twenty-four hours after infection with adenoviruses, C2C12
cells were stimulated with BMPs for 72 hours, with a change of
medium at 48 hours. For quantitative analysis of alkaline phosphatase
activity, cells were washed and extracted with a lysis buffer as
previously described (Asahina et al., 1993; Nishitoh et al., 1996).
Alkaline phosphatase activity was determined using p-nitrophenyl
phosphate (Sigma) as a substrate. For histochemical analysis of
alkaline phosphatase activity, cells were fixed with 3.7%
formaldehyde. After washing with phosphate-buffered saline, the cells
were incubated with a mixture of 0.1 mg/ml naphtol AS-MX
phosphate (Sigma), 0.5% N,N-dimethylformamide, 2 mM MgCl2, and
0.6 mg/ml fast blue BB salt (Sigma) in 0.1 M Tris-HCl, pH 8.5.
Immunoblotting
C2C12 cells infected with adenoviruses or COS-7 cells transfected
with DNAs were washed with phosphate-buffered saline, and
solubilized in a buffer containing 20 mM Tris-HCl, pH 7.5,
150 mM NaCl, 1% Triton X-100, 1% aprotinin, and 1 mM
phenylmethylsulfonyl fluoride. Lysates were subjected to
immunoprecipitation by anti-FLAG M2 antibody (Sigma) or directly
subjected to SDS-gel electrophoresis (Kawabata et al., 1998b; Yagi et
al., 1999). Proteins were then electrotransferred to polyvinylidene
difluoride membranes, immunoblotted with anti-phospho-Smad1
antibody (courtesy of P. ten Dijke) (Korchynskyi et al., 1999) or antiFLAG M2 antibody (Sigma), and visualized using an enhanced
chemiluminescence detection system (Amersham Pharmacia
Biotech).
Immunofluorescence and confocal microscopy
Immunofluorescence staining of C2C12 cells was carried out as
described previously (Ebisawa et al., 1999; Fujii et al., 1999). Cells
were examined by confocal laser scanning microscopy (Olympus).
Luciferase assay
A BMP-responsive promoter-reporter construct, 3GC2-lux, was
previously reported (Ishida et al., 2000). After transient transfection
of DNAs (total 1 µg) into C2C12 cells in six-well tissue culture plates,
luciferase activity in the cell lysates was determined using a
luminometer. Luciferase activities were normalized to sea-pansy
luciferase activity under control of the thymidine kinase promoter.
RESULTS
Induction of alkaline phosphatase activity by
different BMP isoforms and R-Smads
To determine which R-Smads are responsible for signaling by
various BMPs, we infected C2C12 mouse myoblast cells with
adenoviruses carrying Smad1, Smad5 or Smad8. We have
previously shown that, although R-Smads alone can weakly
Synergism of BMP type I receptors
1485
ALP (nmol pNP/min/mg protein)
0
250
500
control
LacZ
Smad4 (300)
Fig. 1. Induction of alkaline
phosphatase activity by different
BMP isoforms and R-Smads.
C2C12 cells were infected with
adenoviruses carrying Smads at
multiplicity of infection (m.o.i.) of
100–300 (shown in parentheses),
and cells were stimulated (filled
boxes) or not (hatched boxes) with
different BMP isoforms. Since each
BMP isoform has distinct potency,
we added different concentrations
of BMPs; BMP-4 (200 ng/ml) (A),
BMP-6 (200 ng/ml) (B), BMP-7
(1000 ng/ml) (C), and GDF-5 (1500
ng/ml) (D). Adenovirus carrying βgalactosidase (LacZ; m.o.i. 300)
was used as a control. Alkaline
phosphatase activity (ALP) was
quantified in duplicate cultures.
pNP, p-nitrophenyl phosphate.
750
1000
0
250
500
(A) BMP-4
750
1000
(B) BMP-6
Smad1 (200)
Smad1 (300)
Smad1 (200) Smad4 (100)
Smad5 (200)
Smad5 (300)
Smad5 (200) Smad4 (100)
Smad8 (200)
Smad8 (300)
Smad8 (200) Smad4 (100)
0
control
LacZ
Smad4 (300)
250
500
750
1000
(C) BMP-7
0
200
400
600
800
(D) GDF-5
Smad1 (200)
Smad1 (300)
Smad1 (200) Smad4 (100)
Smad5 (200)
Smad5 (300)
Smad5 (200) Smad4 (100)
Smad8 (200)
Smad8 (300)
Smad8 (200) Smad4 (100)
Fig. 2. Phosphorylation of R-Smads by different BMP isoforms. (A) COS-7 cells were transfected with indicated cDNAs. Cell lysates were
immunoprecipitated (IP) with anti-FLAG antibody, followed by immunoblotting (Blot) using anti-phospho-Smad1 or anti-FLAG antibody. c.a.,
constitutively active; ALK-5, TGF-β type I receptor. (B) C2C12 cells were infected with indicated adenoviruses at m.o.i. of 300 and stimulated
with different BMP isoforms at concentrations as given in Fig. 1. Cell lysates were directly subjected to immunoblotting using anti-phosphoSmad1 or anti-FLAG antibody. B4, BMP-4; B6, BMP-6; B7, BMP-7; G5, GDF-5.
1486
JOURNAL OF CELL SCIENCE 114 (8)
induce alkaline phosphatase activity, addition of BMPs
dramatically enhances the effects of R-Smads through their
nuclear translocation (Fujii et al., 1999). Levels of expression
of R-Smads obtained by adenovirus infection depended on
multiplicity of infection (m.o.i.) of adenoviruses (data not
shown; see Fujii et al., 1999) and were comparable between
different R-Smads (see Fig. 2B). BMP-4 induced alkaline
phosphatase activity equally well in the presence of Smad1,
Smad5 and Smad8 (Fig. 1A). However, consistent with our
previous findings (Ebisawa et al., 1999), BMP-6 induced
strong alkaline phosphatase activity in the presence of Smad1
and Smad5, but Smad8 was much less potent in this respect
than Smad1 and Smad5 (Fig. 1B). BMP-7 exhibited activity
similar to BMP-6, whereas that of GDF-5 was similar to that
of BMP-4 (Fig. 1C,D). Thus, BMP-4 and
GDF-5 might transmit signals through all
three BMP-R-Smads, whereas BMP-6 and
BMP-7 do not activate Smad8 in signaling.
Phosphorylation and nuclear
translocation of R-Smads by BMPs
Phosphorylation of BMP-R-Smads was
examined using anti-phospho-Smad1 antibody
directed against a synthetic peptide containing
phospho-serine residues corresponding to the
C-terminal region of Smad1 (Korchynskyi
et al., 1999). We have confirmed that
the phospho-Smad1 antibody detects
phosphorylated Smad1, as well as Smad5 and
Smad8 induced by constitutively active ALK6 in transfected COS-7 cells (Fig. 2A, and data
not shown). Because the C-terminal SSVS
sequence is also conserved in Smad3, the
antibody crossreacted with phospho-Smad3,
which was induced by constitutively active
TGF-β type I receptor (ALK-5), but not
with phospho-Smad2, which contained the
C-terminal SSMS sequence (Fig. 2A, right
panel). Immunoblotting analysis using
adenovirus-infected C2C12 cells revealed
that BMP-4, BMP-6 and BMP-7 induced
phosphorylation of Smad1, but that GDF-5
was much less potent than the other BMPs in
phosphorylation of Smad1 (Fig. 2B). Smad5
was phosphorylated by all BMP isoforms
tested in the present study. By contrast, BMP4 efficiently induced phosphorylation of
Smad8, but BMP-6 and BMP-7 failed to do so.
GDF-5 was less potent than BMP-4 in
phosphorylation of Smad8.
Fig. 3. Nuclear translocation of R-Smads by
BMP-4, BMP-6 and GDF-5. C2C12 cells were
infected with adenoviruses carrying Smad1 (A, D,
G, J), Smad5 (B, E, H, K) or Smad8 (C, F, I, L) at
m.o.i. of 300. Cells were not stimulated (A–C) or
stimulated with BMP-4 (200 ng/ml; D–F), BMP-6
(200 ng/ml; G–I), or GDF-5 (1500 ng/ml; J–L),
and subjected to immunofluorescence staining and
observation by confocal laser scanning
microscopy.
Nuclear translocation of BMP-R-Smads was also studied
using C2C12 cells. BMP-4 induced the nuclear translocation
of all three BMP-R-Smads, whereas Smad8 failed to
translocate into the nucleus upon stimulation by BMP-6 (Fig.
3), in agreement with our previous findings (Ebisawa et al.,
1999). Similar results were obtained when Smad4 was coinfected with the BMP-R-Smads (data not shown). Although
GDF-5 phosphorylated Smad1 and Smad8 weakly compared
with BMP-4, it induced nuclear translocation of all BMP-RSmads.
ALK-2 and ALK-3/6 synergistically induce alkaline
phosphatase and transcriptional activities
Because BMP-4 and BMP-6 activate R-Smads in different
Synergism of BMP type I receptors
A
(B)
1487
ALP (nmol pNP/min/mg protein)
0
200
600
400
control
LacZ
ALK-2 (300)
ALK-2 (600)
ALK-3 (300)
ALK-3 (600)
ALK-6 (300)
ALK-6 (600)
ALK-2 (300) + ALK-3 (300)
ALK-2 (300) + ALK-6 (300)
ALK-3 (300) + ALK-6 (300)
ALP (nmol pNP/min/mg protein)
(C)
0
200
400
control
LacZ
BMP-4 (50)
BMP-6 (50)
BMP-4 (25) + BMP-6 (25)
BMP-4 (100)
BMP-6 (100)
BMP-4 (50) + BMP-6 (50)
BMP-4 (200)
BMP-6 (200)
BMP-4 (100) + BMP-6 (100)
Fig. 4. Synergistic induction of alkaline phosphatase activity by ALK-2 and ALK-3/6. (A) Alkaline phosphatase activity was histochemically
determined in C2C12 cells infected with adenoviruses that carried constitutively active type I receptors at m.o.i. of 300 to 600 (shown in
parentheses). Adenovirus carrying β-galactosidase (LacZ; m.o.i. 300) was used as a control. (B) Quantitative analysis of alkaline phosphatase
activity (ALP) was carried out in C2C12 cells treated as in (A). Alkaline phosphatase activity was determined in duplicate cultures. pNP, pnitrophenyl phosphate. (C) Synergistic induction of alkaline phosphatase activity by BMP-4 and BMP-6 in the presence of Smad5. C2C12 cells
were infected with adenovirus carrying Smad5 at m.o.i. of 300, and stimulated with indicated ligands. Concentrations of ligands (ng/ml) are
shown in parentheses.
fashions, we examined whether activation of both BMP-4 and
BMP-6 signaling pathways leads to biological activity in C2C12
cells distinct from that induced by either signaling pathway
alone. Since BMP-4 preferentially binds to ALK-3 and BMP-6
binds to ALK-2, we infected C2C12 cells with adenoviruses
carrying constitutively active forms of type I receptors, ALK3(QD) [or ALK-6(QD)] and ALK-2(QD). These type I receptors
induced alkaline phosphatase activity depending on the m.o.i. of
adenoviruses. When ALK-2(QD) and ALK-3(QD) [or ALK6(QD)] were co-infected into C2C12 cells, much stronger
alkaline phosphatase activity than with either receptor alone was
observed in both histochemical and quantitative analyses (Fig.
4A,B). By contrast, the combination of ALK-3(QD) and ALK6(QD) induced alkaline phosphatase activity nearly equal to
either ALK-3(QD) or ALK-6(QD) alone. These results indicate
that ALK-2 and ALK-3/6 synergistically induce alkaline
phosphatase activity in C2C12 cells.
We also examined whether two distinct BMP isoforms have
synergistic effects on C2C12 cells. In the presence of Smad5,
addition of BMP-4 and BMP-6 were more potent in inducing
alkaline phosphatase activity than either BMP alone, indicating
that two different subtypes of BMPs act synergistically in
osteoblast differentiation (Fig. 4C).
Synergism between two different BMP type I receptors was
also studied using a BMP-responsive promoter–reporter
construct, 3GC2-lux. Transcriptional activation of 3GC2-lux
was observed by transfection of ALK-2(QD) or ALK-3(QD),
which was enhanced by co-transfection of Smad1 (Fig. 5).
Interestingly, the combination of ALK-2(QD) and ALK-3(QD)
induced higher transcriptional activity than either receptor
alone in the absence or presence of Smad1. Thus, the
synergistic activity of two BMP type I receptors might be
induced by efficient activation of downstream signaling
pathways, including the Smad pathways.
DISCUSSION
Of the three different R-Smads activated by BMPs, Smad8 has
not been studied in detail. We previously reported that Smad8
is neither phosphorylated nor translocated into the nucleus by
BMP-6 stimulation (Ebisawa et al., 1999). The present study
revealed that certain BMPs, including BMP-4 and GDF-5,
activate Smad8, but that neither BMP-6 nor BMP-7 does.
BMP-4 and GDF-5 are known to bind preferentially to ALK3 and/or ALK-6, and BMP-6 and BMP-7 bind to ALK-2 (ten
1488
JOURNAL OF CELL SCIENCE 114 (8)
Luciferase Activity (arbitrary unit)
20
15
10
5
0
Smad1
-
+
-
+
-
+
-
+
-
+
-
+
-
ALK-2(QD)
-
0.03
-
0.015
0.1
-
0.05
ALK-3(QD)
-
-
0.03
0.015
-
0.1
0.05
+
Fig. 5. Synergy in the transcriptional activation by ALK-2 and ALK3. Effects of ALK-2(QD) and ALK-3(QD) on the transcriptional
activation of 3GC2-lux were determined in the presence and absence
of Smad1 (0.3 µg) in C2C12 cells. Amounts of ALK-2/3(QD) DNAs
(µg) are shown.
Dijke et al., 1994; Nishitoh et al., 1996; Macías-Silva et al.,
1998; Ebisawa et al., 1999). Thus, ALK-3 and ALK-6 might
transmit signals through Smad8 together with the other BMPR-Smads; on the other hand, ALK-2 does not use Smad8
for signaling. However, when ALK-2 and Smad8 were
overexpressed in mammalian cells, ALK-2 was able to activate
Smad8 (Chen et al., 1997; Nishita et al., 1999; Kawai et al.,
2000; and our unpublished data). It will be interesting to study
whether ALK-2 can activate Smad8 under certain conditions
other than overexpressed systems.
Smad8 is structurally similar to Smad1 and Smad5; however,
functional differences have not been identified between the
BMP-R-Smads. Smad2 and Smad3 are R-Smads involved in
the TGF-β/activin pathway. Smad3 has been shown to bind
DNA directly, whereas Smad2 fails to do so because of the
insertional sequence in the DNA-binding N-terminal Mad
homology 1 domain of Smad2 (Yagi et al., 1999; Dennler et
al., 1999). Thus, some functional differences exist between
Smad2 and Smad3; for example, Smad3 is more potent than
Smad2 in activating promoter–reporter construct p3TP-Lux
(Yagi et al., 1999). In addition, it has been reported that certain
transcription factors associate with Smad3 but not with Smad2
(Kurokawa et al., 1998; Yanagisawa et al., 1999). Whether
such functional differences can be observed among BMP-RSmads, especially between Smad1/5 and Smad8, remains to be
elucidated.
GDF-5 is structurally most similar to GDF-6 (also termed
CDMP-2 and BMP-13) and GDF-7 (BMP-12) (Chang et al.,
1994). GDF-5, -6, and -7 exhibit unique biological activities
different from those of other BMP ligands. GDF-5 induces
cartilage and tendon-like tissues when implanted in vivo
(Hötten et al., 1996; Wolfman et al., 1997). GDF-5
preferentially binds to ALK-6 (Nishitoh et al., 1996), and
recent studies have revealed that ALK-6-deficient mice exhibit
phenotypes similar to those of the GDF-5 null mice (Baur et
al., 2000; Yi et al., 2000). Interestingly, in Drosophila there are
no BMP-like proteins, which are structurally similar to GDF-
5, suggesting that GDF-5 and related factors have unique
functions only in vertebrates. GDF-5 phosphorylated Smad5
most efficiently, whereas Smad1 and Smad8 were only weakly
or not phosphorylated by GDF-5. However, nuclear
translocation of Smad1 and Smad8 was induced by GDF-5 in
C2C12 cells infected with adenoviruses; whether nuclear
translocation of Smad1 and Smad8 by GDF-5 also occurs
under physiological conditions remains to be determined.
We have found that two different subtypes of BMP type I
receptors, ALK-3/6 and ALK-2, have more potent biological
activity in combination than either receptor alone in inducing
alkaline phosphatase and transcriptional activation activities. It
has been reported that BMP heterodimers are more potent than
BMP homodimers in various systems, such as osteoinduction in
vivo (Israel et al., 1996; Kusumoto et al., 1997) and mesoderm
induction in Xenopus (Suzuki et al., 1997). Because BMP
heterodimer proteins are not available in our laboratory, we
cannot compare the biological activities of BMP heterodimers
with those of combinations of two different BMP homodimers.
The combination of two different BMP-like proteins, DPP
and Gbb-60A, has been shown to play important roles in
Drosophila development (Chen et al., 1998; Khalsa et al.,
1998). TKV and SAX are the type I receptors for BMP-like
ligands in Drosophila, and activation of these two receptors at
different ratios is important for the growth and patterning of
the Drosophila wing. In contrast to the type I receptors,
however, only one R-Smad, Mad, acts in the BMP-like
pathway in Drosophila. It is possible that the combination of
two different type I receptors efficiently activates the
downstream Smad pathways (Fig. 5). Another possibility is
that in addition to the Smad signaling pathways, non-Smad
signaling pathways are important for synergistic cooperation
of the two type I receptors. Although not much is known about
non-Smad pathways induced by BMPs, activation of MAP
kinase pathways by BMP-2 and GDF-5, including the p38
MAP kinase pathway, has been reported (Iwasaki et al., 1999;
Nakamura et al., 1999). Whether there are differences in the
non-Smad pathways activated by ALK-3/6 and ALK-2 remains
to be determined.
Because Smad8 is activated by BMP-4 but not by BMP-6,
the former might have a broader biological activity than the
latter. However, our preliminary results using Gene Chip
analysis revealed that certain genes are specifically induced by
BMP-6 but not by BMP-4 (our unpublished data), suggesting
that BMP-6 activates unique signals that are not shared with
BMP-4. Future studies might be directed towards the
identification of target genes that are specifically activated by
different BMP ligands/receptors or by combinations of them,
and elucidation of the involvement of Smad and non-Smad
pathways activated by these ligands and receptors.
We thank T. K. Sampath for BMPs, and P. ten Dijke for antiphospho-Smad1 antibody. We also thank Y. Yuuki, Y. Sasaki and A.
Nishitoh-Sakai for technical help. This study was supported by
Grants-in-Aid for Scientific Research from the Ministry of Education,
Science, Sports and Culture of Japan.
REFERENCES
Abdollah, S., Macías-Silva, M., Tsukazaki, T., Hayashi, H., Attisano, L.
and Wrana, J. L. (1997). TβRI phosphorylation of Smad2 on Ser465 and
Synergism of BMP type I receptors
Ser467 is required for Smad2-Smad4 complex formation and signaling. J.
Biol. Chem. 272, 27678-27685.
Aono, A., Hazama, M., Notoya, K., Taketomi, S., Yamasaki, H., Tsukuda,
R., Sasaki, S. and Fujisawa, Y. (1995). Potent ectopic bone-inducing
activity of bone morphogenetic protein-4/7 heterodimer. Biochem. Biophys.
Res. Commun. 210, 670-677.
Asahina, I., Sampath, T. K., Nishimura, I. and Hauschka, P. V. (1993).
Human osteogenic protein-1 induces both chondroblastic and osteoblastic
differentiation of osteoprogenitor cells derived from newborn rat calvaria.
J. Cell Biol. 123, 921-933.
Baur, S. T., Mai, J. J. and Dymecki, S. M. (2000). Combinatorial signaling
through BMP receptor IB and GDF5: shaping of the distal mouse limb and
the genetics of distal limb diversity. Development 127, 605-619.
Chang, S. C., Hoang, B., Thomas, J. T., Vukicevic, S., Luyten, F. P., Ryba,
N. J., Kozak, C. A., Reddi, A. H. and Moos, M., Jr (1994). Cartilagederived morphogenetic proteins: New members of the transforming growth
factor-β superfamily predominantly expressed in long bones during human
embryonic development. J. Biol. Chem. 269, 28227-28234.
Chen, Y., Bhushan, A. and Vale, W. (1997). Smad8 mediates the signaling
of the ALK-2 receptor serine kinase. Proc. Nat. Acad. Sci. USA 94, 1293812943.
Chen, Y., Riese, M. J., Killinger, M. A. and Hoffmann, F. M. (1998). A
genetic screen for modifiers of Drosophila decapentaplegic signaling
identifies mutations in punt, Mothers against dpp and the BMP-7
homologue, 60A. Development 125, 1759-1768.
Dennler, S., Huet, S. and Gauthier, J. M. (1999). A short amino-acid
sequence in MH1 domain is responsible for functional differences between
Smad2 and Smad3. Oncogene 18, 1643-1648.
Ebisawa, T., Tada, K., Kitajima, I., Tojo, K., Sampath, T. K., Kawabata,
M., Miyazono, K. and Imamura, T. (1999). Characterization of bone
morphogenetic protein-6 signaling pathways in osteoblast differentiation. J.
Cell Sci. 112, 3519-3527.
Fujii, M., Takeda, K., Imamura, T., Sampath, T. K., Enomoto, S.,
Kawabata, M., Kato, M., Ichijo, H. and Miyazono, K. (1999). Roles of
bone morphogenetic protein type I receptors and Smad proteins in osteoblast
and chondroblast differentiation. Mol. Biol. Cell 10, 3801-3813.
Goumans, M-J. and Mummery, C. (2000). Functional analysis of the TGFβ
receptor/Smad pathway through gene ablation in mice. Int. J. Dev. Biol. 44,
253-265.
Heldin, C-H., Miyazono, K. and ten Dijke, P. (1997). TGF-β signalling from
cell membrane to nucleus through SMAD proteins. Nature 390, 465-471.
Hötten, G. C., Matsumoto, T., Kimura, M., Bechtold, R. F., Kron, R.,
Ohara, T., Tanaka, H., Satoh, Y., Okazaki, M., Shirai, T. et al. (1996).
Recombinant human growth/differentiation factor 5 stimulates mesenchyme
aggregation and chondrogenesis responsible for the skeletal development of
limbs. Growth Factors 13, 65-74.
Ishida, W., Hamamoto, K., Kusanagi, K., Yagi, K., Kawabata, M.,
Takehara, K., Sampath, T. K., Kato, M. and Miyazono, K. (2000).
Smad6 is a Smad1/5-induced Smad inhibitor: Characterization of bone
morphogenetic protein-responsive element in the mouse Smad6 promoter.
J. Biol. Chem. 275, 6075-6079.
Israel, D. I., Nove, J., Kerns, K. M., Kaufman, R. J., Rosen, V., Cox, K. A.
and Wozney, J. M. (1996). Heterodimeric bone morphogenetic proteins
show enhanced activity in vitro and in vivo. Growth Factors 13, 291-300.
Iwasaki, S., Iguchi, M., Watanabe, K., Hoshino, R., Tsujimoto, M. and
Kohno, M. (1999). Specific activation of the p38 mitogen-activated protein
kinase signaling pathway and induction of neurite outgrowth in PC12 cells
by bone morphogenetic protein-2. J. Biol. Chem. 274, 26503-26510.
Kawabata, M., Imamura, T. and Miyazono, K. (1998a). Signal transduction
by bone morphogenetic proteins. Cytokine Growth Factor Rev. 9, 49-61.
Kawabata, M., Inoue, H., Hanyu, A., Imamura, T. and Miyazono, K.
(1998b). Smad proteins exist as monomers in vivo and undergo homo- and
hetero-oligomerization upon activation by serine/threonine kinase receptors.
EMBO J. 17, 4056-4065.
Kawabata, M. and Miyazono, K. (2000). Bone morphogenetic proteins. In
Skeletal Growth Factors (ed. E. Canalis). pp. 269-290, Lippincott Williams
& Wilkins, Philadelphia.
Kawai, S., Faucheu, C., Gallea, S., Spinella-Jaegle, S., Atfi, A., Baron, R.
and Roman, S. R. (2000). Mouse Smad8 phosphorylation downstream of
BMP receptors ALK-2, ALK-3, and ALK-6 induces its association with
Smad4 and transcriptional activity. Biochem. Biophys. Res. Commun. 271,
682-687.
Khalsa, O., Yoon, J. W., Torres-Schumann, S. and Wharton, K. A. (1998).
TGF-β/BMP superfamily members, Gbb-60A and Dpp, cooperate to
1489
provide pattern information and establish cell identity in the Drosophila
wing. Development 125, 2723-2734.
Korchynskyi, O., Landstrom, M., Stoika, R., Funa, K., Heldin, C-H., ten
Dijke, P. and Souchelnytskyi, S. (1999). Expression of Smad proteins in
human colorectal cancer. Int. J. Cancer 82, 197-202.
Kretzschmar, M., Liu, F., Hata, A., Doody, J. and Massagué, J. (1997). The
TGF-β family mediator Smad1 is phosphorylated directly and activated
functionally by the BMP receptor kinase. Genes Dev. 11, 984-995.
Kurokawa, M., Mitani, K., Irie, K., Matsuyama, T., Takahashi, T., Chiba,
S., Yazaki, Y., Matsumoto, K. and Hirai, H. (1998). The oncoprotein Evi1 represses TGF-β signalling by inhibiting Smad3. Nature 394, 92-96.
Kusumoto, K., Bessho, K., Fujimura, K., Akioka, J., Ogawa, Y. and
Iizuka, T. (1997). Comparison of ectopic osteoinduction in vivo by
recombinant human BMP-2 and recombinant Xenopus BMP-4/7
heterodimer. Biochem. Biophys. Res. Commun. 239, 575-579.
Luyten, F. P., Kaplan, F. S. and Shore, E. M. (2000). Clinical disorders
associated with bone morphogenetic proteins. In Skeletal Growth Factors
(ed. E. Canalis). pp. 323-334, Lippincott Williams & Wilkins, Philadelphia.
Macías-Silva, M., Hoodless, P. A., Tang, S. J., Buchwald, M. and Wrana,
J. L. (1998). Specific activation of Smad1 signaling pathways by the BMP7
type I receptor, ALK2. J. Biol. Chem. 273, 25628-25636.
Massagué, J. and Wotton, D. (2000). Transcriptional control by the TGFβ/Smad signaling system. EMBO J. 19, 1745-1754.
Meersseman, G., Verschueren, K., Nelles, L., Blumenstock, C., Kraft, H.,
Wuytens, G., Remacle, J., Kozak, C. A., Tylzanowski, P., Niehrs, C. et
al. (1997). The C-terminal domain of Mad-like signal transducers is
sufficient for biological activity in the Xenopus embryo and transcriptional
activation. Mech. Dev. 61, 127-140.
Nakamura, K., Shirai, T., Morishita, S, Uchida, S., Saeki-Miura, K. and
Makishima, F. (1999). p38 mitogen-activated protein kinase functionally
contributes to chondrogenesis induced by growth/differentiation factor-5 in
ATDC5 cells. Exp. Cell Res. 250, 351-363.
Nishita, M., Ueno, N. and Shibuya, H. (1999). Smad8B, a Smad8 splice
variant lacking the SSXS site that inhibits Smad8-mediated signalling.
Genes Cells 4, 583-591.
Nishitoh, H., Ichijo, H., Kimura, M., Matsumoto, T., Makishima, F.,
Yamaguchi, A., Yamashita, H., Enomoto, S. and Miyazono, K. (1996).
Identification of type I and type II serine/threonine kinase receptors for
growth/differentiation factor-5. J. Biol. Chem. 271, 21345-21352.
Reddi, A. H. (1998). Role of morphogenetic proteins in skeletal tissue
engineering and regeneration. Nat. Biotechnol. 16, 247-252.
Souchelnytskyi, S., Tamaki, K., Engström, U., Wernstedt, C., ten Dijke,
P. and Heldin, C-H. (1997). Phosphorylation of Ser465 and Ser467 in
the C terminus of Smad2 mediates interaction with Smad4 and is required
for transforming growth factor-β signaling. J. Biol. Chem. 272, 2810728115.
Suzuki, A., Kaneko, E., Maeda, J. and Ueno, N. (1997). Mesoderm
induction by BMP-4 and -7 heterodimers. Biochem. Biophys. Res. Commun.
232, 153-156.
ten Dijke, P., Yamashita, H., Sampath, T. K., Reddi, A. H., Estevez, M.,
Riddle, D. L., Ichijo, H., Heldin, C-H. and Miyazono, K. (1994).
Identification of type I receptors for osteogenic protein-1 and bone
morphogenetic protein-4. J. Biol. Chem. 269, 16985-16988.
Thomas, J. T., Kilpatrick, M. W., Lin, K., Erlacher, L., Lembessis, P.,
Costa, T., Tsipouras, P. and Luyten, F. P. (1997). Disruption of human
limb morphogenesis by a dominant negative mutation in CDMP1. Nat.
Genet. 17, 58-64.
Wolfman, N. M., Hattersley, G., Cox, K., Celeste, A. J., Nelson, R., Yamaji,
N., Dube, J. L., DiBlasio-Smith, E., Nove, J., Song, J. J. et al. (1997).
Ectopic induction of tendon and ligament in rats by growth and
differentiation factors 5, 6, and 7, members of the TGF-β gene family. J.
Clin. Invest. 100, 321-330.
Yagi, K., Goto, D, Hamamoto, T., Takenoshita, S., Kato, M. and Miyazono,
K. (1999). Alternatively spliced variant of Smad2 lacking exon 3.
Comparison with wild-type Smad2 and Smad3. J. Biol. Chem. 274, 703709.
Yanagisawa, J., Yanagi, Y., Masuhiro, Y., Suzawa, M., Watanabe, M.,
Kashiwagi, K., Toriyabe, T., Kawabata, M., Miyazono, K. and Kato, S.
(1999). Convergence of transforming growth factor-β and vitamin D
signaling pathways on SMAD transcriptional coactivators. Science 283,
1317-1321.
Yi, S. E., Daluiski, A., Pederson, R., Rosen, V. and Lyons, K. M. (2000).
The type I BMP receptor BMPRIB is required for chondrogenesis in the
mouse limb. Development 127, 621-630.