Transforming Growth Factor B1 1 To The Bone PDF

0163-769X/05/$20.
00/0 Endocrine Reviews 26(6):743–774

Printed in U.S.A. Copyright © 2005 by The Endocrine Society
doi: 10.1210/er.2004-0001
Transforming Growth Factor-␤1 to the Bone

Katrien Janssens, Peter ten Dijke, Sophie Janssens,* and Wim Van Hul*
Department of Medical Genetics (K.J., W.V.H.), University of Antwerp, 2610 Antwerp, Belgium; Department of Molecular
Cell Biology (P.t.D.), Leids Universitair Medisch Centrum, 2333 AL Leiden, The Netherlands; and Department of
Biochemistry (S.J.), University of Lausanne, 1066 Epalinges-Lausanne, Switzerland
Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

TGF-␤1 is a ubiquitous growth factor that is implicated in the bone environment, helping to retain the balance between the
control of proliferation, migration, differentiation, and sur- dynamic processes of bone resorption and bone formation.
vival of many different cell types. It influences such diverse Many seemingly contradictory reports have been published
processes as embryogenesis, angiogenesis, inflammation, and on the exact functioning of TGF-␤1 in the bone milieu. This
wound healing. In skeletal tissue, TGF-␤1 plays a major role review provides an overall picture of the bone-specific actions
in development and maintenance, affecting both cartilage and of TGF-␤1 and reconciles experimental discrepancies that
bone metabolism, the latter being the subject of this review. have been reported for this multifunctional cytokine. (Endo-
Because it affects both cells of the osteoblast and osteoclast crine Reviews 26: 743–774, 2005)
lineage, TGF-␤1 is one of the most important factors in the
I. Introduction A. Bone phenotypes of knockout and transgenic mouse

A. Functions of TGF-␤1 models of the TGF-␤ signaling pathway
B. Processing and storage of TGF-␤1 B. TGFB1 mutations in the pathogenesis of Camurati-En-
C. Activation mechanisms of TGF-␤1 gelmann disease (CED)
II. The TGF-␤ Signaling Pathway C. Osteolytic metastases: a role for TGF-␤1 in malignancy
A. General aspects D. Bone-related association studies
B. Smad-dependent signaling V. Therapeutic Use of TGF-␤1 as Bone-Forming Agent
C. Smad-independent signaling VI. Concluding Remarks
III. TGF-␤1 in Bone
A. Introduction
B. TGF-␤ isoforms in bone I. Introduction
C. Role of TGF-␤1 in osteoblastogenesis and bone forma-
tion in vitro
D. TGF-␤1 in osteoclast formation and bone resorption in
T GF-␤1 IS THE prototype of the TGF-␤ superfamily, an
evolutionary conserved family of structurally related
dimeric cytokines with representatives in organisms as di-
vitro verse as mammals and invertebrates. Its members share a
E. Interaction of TGF-␤1 with other growth factors and cluster of conserved cysteine residues that form a cysteine
hormones knot structure held together by intramolecular disulfide
IV. Bone Phenotypes Associated with Abnormal TGF-␤1 bonds. Moreover, they all have the same precursor structure
Signaling with a hydrophobic signal sequence, a prodomain, and a
mature C-terminal domain (1). The superfamily includes
First Published Online May 18, 2005 TGF-␤s, bone morphogenetic proteins (BMPs), growth and
* S.J. and W.V.H. contributed equally to this work as senior authors. differentiation factors, activins, inhibin, and anti-Mullerian
Abbreviations: ALK, Activin receptor-like kinase; ALP, alkaline phos- hormone. All its members play important roles in the reg-
phatase; AP-1, activator protein 1; BMD, bone mineral density; BMP, ulation of cell proliferation and differentiation and have piv-
bone morphogenetic protein; BMSC, bone marrow stromal cells; CED,
Camurati-Engelmann disease; Co-Smad, common mediator Smad; otal functions during embryogenesis.
ECM, extracellular matrix; ER, estrogen receptor; GS, glycine- and The TGF-␤ family contains three closely related mamma-
serine-rich; I-Smad, inhibitory Smad; JNK, c-Jun N-terminal kinase; lian isoforms—TGF-␤1, -␤2, and -␤3—that arose by dupli-
LAP, latency-associated peptide; LLC, large latent complex; LTBP, latent cation of a common ancestor. Similarity is most striking in the
TGF-␤ binding protein; M-CSF, macrophage-colony stimulating factor;
C-terminal domain (64 – 82%), with nine conserved cysteine
MH, Mad homology; MNC, multinucleated cell; NF, nuclear factor;
1,25-(OH)2-D3, 1,25-dihydroxyvitamin D3; OPG, osteoprotegerin; residues forming four intrachain and one interchain disulfide
PBMC, peripheral blood mononuclear cell; PDGF, platelet-derived bond. Despite this high sequence homology, analysis of the
growth factor; RANKL, receptor activator of NF-␬B ligand; R-Smad, in vivo functions of the three isoforms by gene knockouts
receptor-mediated Smad; SARA, Smad anchor for receptor activation; revealed striking differences, illustrating their nonredun-
SBE, Smad-binding element; SLC, small latent complex; SNP, single
nucleotide polymorphism; TAB, TAK1 binding protein; T␤RI, TGF-␤ dancy (see Section IV.A.). Overall, TGF-␤1 is the most abun-
type I receptor; VDR, vitamin D receptor. dant isoform with the largest sources of TGF-␤1 being plate-
Endocrine Reviews is published bimonthly by The Endocrine Society lets (20 mg/kg) (2) and bone (200 ␮g/kg) (3).
(http://www.endo-society.org), the foremost professional society serv- TGF-␤1 is a ubiquitous, multifunctional growth factor.
ing the endocrine community. TGF-␤ was first identified as a factor that synergizes with
743
744 Endocrine Reviews, October 2005, 26(6):743–774 Janssens et al. • TGF-␤1 to the Bone
TGF-␣ to induce colony formation of normal rat kidney fi- hematopoietic precursor cell proliferation (see Section III).
broblasts in soft agar, hence their name (4). Since its discov- Regarding the diversity of processes in which TGF-␤1 is
ery, numerous other functions have been attributed to this involved, it is not surprising that this cytokine is of major
cytokine, and several alternative names have been proposed, importance both during embryogenesis and in maintaining
such as cartilage-inducing factor, differentiation-inhibiting tissue homeostasis during life.
factor, and tissue-derived growth inhibitor. However, none
B. Processing and storage of TGF-␤1
of these express the multitude of functions in which TGF-␤1
is involved. TGF-␤1 is synthesized as a 390-amino acid protein (pre-
A. Functions of TGF-␤1 pro-TGF-␤1) consisting of three distinct parts: the signal pep-
tide (SP; 29 amino acids), the latency-associated peptide
TGF-␤1 regulates a broad range of biological processes, (LAP; 249 amino acids), and the mature peptide (112 amino
including cell proliferation, cell survival, cell differentiation, acids). The pre-pro-TGF-␤1 monomer is extensively pro-

cell migration, and production of extracellular matrix (ECM) cessed before its secretion (outlined in Fig. 1).
(for review, see Refs. 5– 8). The combined actions of these Unlike most other members of the TGF-␤ superfamily, the
cellular responses mediate the global effects of TGF-␤1 on mature peptide of the TGF-␤ isoforms stays noncovalently
immune responses, angiogenesis, wound healing, develop- associated with its propeptide or LAP (13). The LAP confers
ment, and bone formation (9 –12). Bone formation by TGF-␤1 latency to the mature peptide, shielding the epitope(s) that
is promoted through chemotactic attraction of osteoblasts, can interact with the TGF-␤ receptor. Moreover, this complex
enhancement of osteoblast proliferation and the early stages of mature peptide and LAP— called the small latent complex
of differentiation with production of ECM proteins, stimu- or SLC— can become associated with a latent TGF-␤ binding
lation of type II collagen expression and proteoglycan syn- protein (LTBP) to form the large latent complex (LLC). The
thesis by chondrocyte precursor cells, and suppression of LTBP is dispensable for latency (14) but has a role in the
FIG. 1. TGF-␤1 processing. Pre-pro-

TGF-␤1 undergoes extensive posttrans-
lational processing. 1, The signal peptide,
targeting the protein to the secretory
pathway, is cleaved off during transit
through the rough endoplasmatic reticu-
lum (RER). 2, Two monomers dimerize by
way of disulfide bridges between cysteine
residues at positions 223 and 225 in the
LAP and cysteine residue 356 in the ma-
ture peptide. 3, The protein is cleaved by
furin convertase at the dibasic arginine
residue at position 278. This yields the
LAP and the mature peptide. Noncova-
lent bonds between them prevent the pre-
mature activation of the mature peptide,
forming the SLC. 4, The SLC can become
covalently attached to a LTBP to form the
LLC. Binding occurs between the cys-
teine residue at position 33 of the LAP
and the third 8 Cys-repeat of the LTBP.
5, After its secretion, the LLC is directed
to the ECM and stored through binding of
the LTBP with the ECM. The SLC is more
readily available for activation. The SLC
to LLC ratio depends on the cell type.
Janssens et al. • TGF-␤1 to the Bone Endocrine Reviews, October 2005, 26(6):743–774 745
correct folding and secretion of TGF-␤ (15), as well as in GS domain. The type II receptors are characterized by their
targeting the LLC for storage through interactions with the constitutively active kinase domain. Seven type I and five
ECM (16). Currently, four different LTBPs are known of type II receptors transmitting signals from TGF-␤ superfam-
which LTBP-1, -3, and -4 have been shown to covalently bind ily members are present in mammals (27). Despite the fact
TGF-␤1 (for two recent reviews on this subject, see Refs. 17 that more than 35 members form part of the TGF-␤ super-
and 18). family, the combinations between the type I and type II
Although most cell types secrete TGF-␤ as part of the LLC, receptors occurring under normal conditions are limited.
bone cells form an exception as they efficiently secrete the For members of the TGF-␤ family, the TGF-␤ type II re-
SLC (see Section III.B). Consequently, the SLC is the predom- ceptor (T␤RII) is the sole type II receptor shown to mediate
inant form in the bone environment. The remaining TGF-␤1 signaling. This is reflected by the phenotypic identity of the
is bound to LTBP-1 or -3 (19, 20). tgfbr2 and those tgfb1 knockout mice that die in utero (see
Section IV.A). Of the type I receptors, ALK5, ALK1, and
C. Activation mechanisms of TGF-␤1

possibly ALK2 can transmit TGF-␤ signals. ALK5 (T␤RI) is
Because members of the TGF-␤ family are secreted as the most important type I receptor for TGF-␤, which is un-
latent complexes, they need to be activated to exhibit their derscored by the comparable (although not identical) phe-
biological activity. In view of the ubiquitous expression of notypes of tgfb1 and alk5 knockout mice: histological exam-
TGF-␤1 and its receptors and the multitude of processes in ination of the yolk sacs of tgfbr1⫺/⫺ embryos shows an image
which TGF-␤1 is involved, it comes as no surprise that ac- very similar to that of tgfb1⫺/⫺ embryos that die during
tivation is tightly regulated. The goal of the activation pro- embryogenesis (312). In bone cells, T␤RI seems to be the only
cess is the release of the epitope(s) on the mature peptide type I receptor involved in signaling.
responsible for interaction with the receptor. Activation of Betaglycan and endoglin are so-called type III or accessory
the LLC initiates with its release from the ECM a process receptors, which are indirectly involved in signaling through
mediated by proteases (plasmin, thrombin, leukocyte elas- the modulation of ligand-binding specificity. Betaglycan
tase, mast cell chymase) that cleave the LTBP at a protease- (T␤RIII) can bind all three TGF-␤ isoforms and is implicated
sensitive hinge region and target the cleaved complex to the in the presentation of TGF-␤ to T␤RII (28). For TGF-␤2, which
cell surface (21, 22). The truncated LLC and the SLC can be has a low intrinsic affinity for T␤RII, both in vitro and in vivo
subjected to three different mechanisms of in vivo activation: data have demonstrated signaling to be dependent on pre-
1) degradation of the LAP by proteases; 2) induction of a sentation of this isoform by betaglycan (29, 30). However, it
conformational change in the LAP by interaction with inte- must be noted that a soluble form of betaglycan, shed by cells
grins and thrombospondin, for example; and 3) rupture of upon proteolysis in the juxtamembrane region, has a role in
the noncovalent bonds between LAP and mature TGF-␤1. An sequestering mature TGF-␤, thus inhibiting signaling (31).
overview of in vivo TGF-␤ activation mechanisms has been Moreover, upon T␤RII-mediated phosphorylation of betag-
presented by Annes et al. (23). The recent development of a lycan, ␤-arrestin-dependent internalization of the T␤RII/be-
genetic screen to discover new TGF-␤ activators might enable taglycan complex serves to down-regulate TGF-␤ signaling
the identification of other relevant activating mechanisms in (32). Endoglin can bind TGF-␤1 and -␤3 in the presence of
vivo (24). T␤RII (33). Mutations in ENG, the gene encoding endoglin,
A unique activation mechanism has been proposed in lie at the basis of the human disease hereditary hemorrhagic
bone, in which resorbing osteoclasts may activate TGF-␤1 in telangiectasia, an autosomal dominant disorder character-
their acidic microenvironment (25). Acidification might ized by multisystem vascular dysplasia (34). A murine model
break the noncovalent bonds between LAP and mature TGF- of this disorder presents with a phenotype that is remarkably
␤1, thus releasing the active peptide. However, others doubt similar to that of tgfb1 and tgfbr2 knockout mice, suggesting
that the mild acidic microenvironment created by osteoclasts an in vivo requirement for endoglin in TGF-␤1 signaling (35).
is able to activate TGF-␤1 and suggest the activation to be a Bone marrow stromal cells (BMSCs) and mature osteoblasts
consequence of the release of proteases by these bone-resorb- express the two types of type III receptors (36), whereas
ing cells (26). osteoclasts seem to lack betaglycan.
In the absence of ligand, both type I and type II receptors
are present as homodimers. Upon TGF-␤1 binding to T␤RII,
II. The TGF-␤ Signaling Pathway T␤RI can be recruited into a heterotetrameric T␤RII/T␤RI
complex. Ligand-induced multimerization of the receptor
A. General aspects
complex is followed by transphosphorylation of the GS do-
Once activated, TGF-␤ can interact with its receptor to main of T␤RI by the constitutively phosphorylated T␤RII
induce signaling. All members of the TGF-␤ superfamily kinase, resulting in activation of T␤RI (37). This transphos-
signal through a dual receptor system of type I and type II phorylation is the first step in the intracellular transmission
transmembrane serine/threonine kinases. These receptors of the signal.
belong to a family of glycoproteins characterized by a
cysteine-rich extracellular region, a single transmembrane B. Smad-dependent signaling
␣-helix, and a cytoplasmic domain with a kinase domain. In
addition, the type I receptors, also termed activin receptor- Genetic studies in Drosophila melanogaster provided a
like kinases (ALKs), share a highly conserved glycine- and breakthrough in our understanding of intracellular TGF-␤
serine-rich (GS) domain adjacent to the kinase domain, the signaling through the identification of mothers against dpp
(Mad). Its protein product plays a role in mediating the func- presumably in the endocytotic compartment (47– 49). In the
tion of decapentaplegic (dpp), the D. melanogaster ortholog of basal, unphosphorylated state, the MH1 and MH2 domains
BMP-2 or BMP-4 (38). This discovery was followed by the of the R-Smads inhibit each other reciprocally. Binding of the
genetic identification of the homologous Sma genes in Cae- R-Smad to T␤RI is followed by phosphorylation of the
norhabditis elegans (39) and subsequently the Smad genes (for former at its carboxy-terminal SSXS motif by the T␤RI kinase
Sma and Mad related) in vertebrates (reviewed in Ref. 40). domain, which causes the R-Smad to dissociate from the
The Smads turned out to play a central role in the transmis- receptor complex. This induces a conformational change that
sion of signals from all receptors activated by TGF-␤ super- relieves reciprocal MH1-MH2 domain inhibition and pro-
family members to target genes in the nucleus. motes the formation of heteromeric complexes with variable
Because several comprehensive reviews on Smad signal- stoichiometry with the Co-Smad4 (50 –53). The Smad com-
ing by members of the TGF-␤ superfamily have been pub- plex translocates to the nucleus, where it can bind directly to
lished recently (37, 41– 43), this signaling pathway will be DNA or recruit other DNA binding partners (transcription

discussed only briefly. Based on their structural and func- factors). The MH1 domains of phosphorylated Smad3 and
tional properties, Smads can be classified into three groups: Smad4 can bind to Smad-binding elements (SBEs) or GC-rich
receptor mediated (R)-Smads, common mediator (Co)- regions in the promoter of TGF-␤ target genes. However, the
Smads and inhibitory (I)-Smads. R-Smads 2 and 3 are re- interaction of Smads with DNA is of both low specificity and
sponsible for transmitting most signals from TGF-␤. The only low affinity and is not sufficient to induce transcriptional
Co-Smad identified so far in mammals is Smad4, which is activity. DNA binding partners that bind to recognition se-
commonly used by all TGF-␤ superfamily members. The R- quences in close proximity of the SBE site and to the MH1 or
and Co-Smads share a similar structure with conserved MH2 domains of the Smads are required. In addition, Smads
amino- and carboxy-terminal domains, the Mad homology recruit general cofactors with activating or repressive capac-
(MH)-1 and MH2 domains, connected by a more divergent ity (mostly obtained through interaction with histone acety-
linker region. In addition, the R-Smads contain a carboxy- lases or deacetylases, respectively), which further determine
terminal phosphorylation site, the SSXS motif. Lacking any the outcome of the Smad-mediated gene transcriptional re-
recognizable enzyme activity, Smads achieve their signaling sponse. An extensive list of Smad binding partners and co-
capacity mainly through protein-protein or DNA-protein in- factors can be found in a recent review by Miyazawa et al.
teractions, exerted by the different domains. The MH1 do- (54).
main can mediate direct DNA binding, whereas the MH2
domain is implicated in receptor interaction, Smad oligomer- C. Smad-independent signaling
ization, and transcriptional activation. Both domains further
drive nuclear import and allow binding to various transcrip- Although the Smads are critical mediators in the TGF-␤
tion factors and cofactors (see below). The divergent linker signaling pathway, a substantial body of evidence illustrates
region contains multiple phosphorylation sites, allowing fine the existence of additional, Smad-independent pathways.
tuning of Smad functioning by many different signaling First, partial preservation of TGF-␤ signaling in Smad4-
pathways in the cell, which converge on phosphorylation of deficient cells is highly suggestive of Smad-independent sig-
this region. Furthermore, the linker region of R-Smads (with naling (55–57). Indeed, genome-based searches revealed that
the exception of Smad8) contains the PY motif, which directs the current sets of identified Smads represent the full com-
interaction of the Smad proteins with the E3 ubiquitin ligases plement (408), rendering the hypothesis of redundancy at the
of the Smurf or SCF families, targeting the protein for deg- Co-Smad level improbable. Although it was proposed that
radation (reviewed in Refs. 44 and 45). The class of the Smad2 or Smad3 could partially substitute for Smad4 (55,
I-Smads comprises Smad6 and -7. Smad6 is an inhibitor of 409), this effect was only seen under conditions of ectopically
BMP signaling, whereas Smad7 inhibits both TGF-␤/activin expressed Smad2 or Smad3 and has yet to be proven to play
and BMP signaling. I-Smads share the MH2 domain with the a role in vivo. Second, studies making use of a mutant T␤RI
R-Smads but show only weak similarity to the MH1 domain. defective in Smad recruitment demonstrated that TGF-␤
Via its MH2 domain, Smad7 is recruited to the receptor could still activate MAPK signaling (58, 59). In addition,
complex, thereby mechanically blocking the access of several other lines of evidence point to the involvement of
R-Smads. Moreover, it can direct T␤RI for ubiquitination and MAPK signaling pathways in transmitting TGF-␤ signals
degradation through binding of Smurfs to its PY motif (44, from receptor to nucleus. In vitro kinase assays have dem-
45). Finally, it acts as an adaptor protein in the formation of onstrated that TGF-␤ can activate all three MAPK pathways,
a protein phosphatase holoenzyme that targets T␤RI for de- leading to ERK, c-Jun N-terminal kinase (JNK), and p38
phosphorylation (46). MAPK activation (see Fig. 2; reviewed in Ref. 60) and phos-
In Fig. 2, the canonical Smad-dependent signaling path- phorylation of members of the Jun, Fos, and ATF transcrip-
way has been outlined. Upon binding of TGF-␤ to its type II tion factor families, which homo- and heterodimerize to form
receptor and formation of the heterotetrameric type II/type the activator protein (AP)-1 (see Ref. 61 and references
I receptor complex, T␤RII transphosphorylates and activates therein). Furthermore, transfection of dominant-negative
T␤RI. Through their MH2 domain, R-Smads can bind the GS forms of MAPK signaling intermediates or application of
domain of T␤RI, an interaction promoted by adaptor pro- inhibitors of these components interferes with TGF-␤ related
teins such as Smad anchor for receptor activation (SARA). processes or luciferase transcriptional activity from a TGF-
SARA interacts specifically with T␤RI and functions to re- ␤-responsive reporter construct in various cell lines, whereas
cruit Smad2 and Smad3 to the activated receptor complex, overexpression of signaling intermediates has the opposite
FIG. 2. Signaling by the TGF-␤ family

members through the Smad-dependent
and MAPK-dependent pathways. A, Af-
ter activation, TGF-␤ can bind to its type
II receptor, T␤RII. B, Binding does not
alter the phosphorylation state of T␤RII,
which is constitutively active, but in-
duces the formation of a heterotet-
rameric receptor complex of T␤RII and a
type I receptor, in most cases T␤RI
(ALK5). C, T␤RII activates T␤RI by
transphosphorylation of the GS domain.
1, Smad-dependent signaling. 1D, Acti-
vated T␤RI can, in turn, phosphorylate

one of the R-Smads at the C-terminal
SSXS domain. Presentation of R-Smads
to T␤RI can be promoted through bind-
ing with SARA. 1E, Phosphorylation of
the R-Smad relieves the reciprocal inhi-
bition of its MH1 and MH2 domains and
allows its interaction with a Co-Smad,
forming a heteromeric complex. 1F, The
R-Smad/Co-Smad complex translocates
to the nucleus. 1G, Smads 3 and 4 are
able to bind DNA through their MH1 do-
main but are unable to induce transcrip-
tion independently. Instead, they modu-
late the transcription of diverse genes
through their interaction with a set of
corepressors, coactivators, and tran-
scription factors. 2, MAPK-dependent
signaling. 2D, The factors between the
receptor complex and the MAPKKKs re-
main largely unknown. 2E, Through se-
quential phosphorylation of different
MAPKKKs, MAPKKs, and MAPKs
(ERK, JNK, and p38 MAPK), transcrip-
tion factors of the ATF, Jun, and Fos
families are activated. 2F, Transcription
factors homo- and heterodimerize into
AP-1 complexes, which can bind to AP-
1-binding sites in the DNA sequence.
KD, Kinase domain; P, phosphorylation;
ATF, activating transcription factor;
MAPKK, MAPK kinase; MAPKKK,
MAPK kinase kinase.
effect (57, 62–70). Finally, reduction or even abolition of profile of hundreds of TGF-␤-controlled genes in fibroblasts
TGF-␤ responsiveness upon mutation/deletion of AP-1 el- deficient in Smad2, Smad3, or ERK signaling, respectively,
ements in various TGF-␤-inducible promoters (e.g., Smad7, Smad3 was demonstrated to be the critical mediator for ex-
COL1A2, osteocalcin, plasminogen activator inhibitor-1, tis- pression of immediate-early target genes. Smad2 and the
sue inhibitor of metalloproteinase-1, matrix metalloprotein- ERK pathways were found to function predominantly in the
ase 1) underscores the importance of MAPK signaling transmodulation of immediate-early and intermediate gene
(71–75). regulation. It would be also interesting to investigate this
Crosstalk between Smad and MAPK pathways adds to the expression profile in p38 MAPK- and JNK-deficient cell lines.
complexity of TGF-␤ signaling. Crosstalk can be obtained Despite ample in vitro evidence in the literature for the
through physical interaction between Smad2, -3, and -4, and involvement of MAPKs in the TGF-␤ signaling cascade, data
members of the Jun, Fos and ATF families bound to their that unequivocally demonstrate the need for MAPK path-
AP-1 site in the promoter of target genes, possibly stabilized ways in in vivo TGF-␤-mediated responses are lacking. Al-
by Smad-DNA binding at an adjacent SBE site (76, 77). In though knockout and transgenic mouse models of numerous
addition, JNK (activated by TGF-␤) can phosphorylate MAPK signaling intermediates are available (81) (for mouse
Smad3, thus facilitating activation and nuclear translocation models with a bone phenotype, see Refs. 82 and 83) none of
of the latter in response to TGF-␤ (66). On the other hand, them are scored for defects in TGF-␤ signaling. However,
TGF-␤-activated c-Jun was shown to antagonize Smad sig- keratinocytes derived from MEKK1-deficient mice show no
naling by enhancing interaction of Smad2 with a corepressor migration in response to TGF-␤1 (84), and MKK3(⫺/⫺) mes-
(78, 79). Recently, a hierarchical model of gene regulation by angial cells are defective in TGF-␤1-induced vascular endo-
TGF-␤ was proposed (80). Upon investigating the expression thelial growth factor expression (85). These observations
clearly show the requirement for MAPK-dependent signal- bone formation. Bone resorption by the osteoclasts involves
ing in transmitting TGF-␤ signals. demineralization of the inorganic matrix by acidification fol-
Although it has been established that these Smad-inde- lowed by enzymatic degradation of the organic matrix by
pendent pathways, like the Smad-dependent pathways, are cathepsin K and matrix metalloproteinases (106). Osteoclasts
initiated by the ligand-induced activation of the T␤RI/T␤RII are large, multinucleated cells (MNCs) of hematopoietic or-
receptor complex, the mediators acting between the receptor igin that differentiate from monocyte/macrophage precur-
complex and the MAPKKKs have not been fully elucidated. sor cells within the bone environment. The recognition that
In vitro studies have established a role for hematopoietic osteoclast differentiation requires the presence of marrow
progenitor kinase-1 and TAK1 binding protein (TAB)-1 as stromal cells or osteoblasts led to the discovery of the two
TGF-␤-activated MAPKKKKs (70, 86) and for the X-linked osteoblast-derived factors essential and sufficient to promote
inhibitor of apoptosis protein as a cofactor in TGF-␤ signal- osteoclastogenesis: macrophage-colony stimulating factor
ing, possibly linking T␤RI and TAB1 (87, 88). Upon disrup- (M-CSF) and receptor activator of nuclear factor (NF)-␬B

tion of tab1, delayed ossification and decreased responsive- ligand (RANKL). Upon binding to their respective receptors
ness to TGF-␤ stimulation are observed, suggesting a role for on the osteoclast precursor cell surface (c-fms and RANK),
this MAPKKKK in TGF-␤-mediated bone formation (89). two prominent transcription factor complexes, the NF-␬B
In bone, examples of TGF-␤-mediated processes involving and AP-1 proteins, are activated, and signaling cascades
MAPK signaling are chemotaxis of osteoclasts (90), oste- essential for proper osteoclast differentiation, fusion, func-
oclastogenesis (91), changes in osteoblast shape (92), osteo- tion, motility, and survival are initiated. Downstream factors
blast-to-osteocyte transdifferentiation (93), Runx2 expres- that are indispensable during osteoclast differentiation and
sion in mesenchymal precursor cells (94), ␣1 and ␣2 collagen which control expression of genes that typify the osteoclast
I expression in osteosarcoma cells (95, 96), collagenase 3 lineage include PU.1 and MITF (107, 108), NFAT-c (109), and
expression in osteoblastic cells (97), suppression of osteo- the transmembrane adaptor protein DAP12 (110). Molecules
blastic osteocalcin expression (98), and inhibition of alkaline that identify the mature, functional osteoclast and are es-
phosphatase (ALP) activity and mineralization by osteo- sential for cell survival, cell motility, and bone resorption,
blasts (99). comprise c-Src, TRAP, carbonic anhydrase II, and cathepsin
In conclusion, we can state that the JNK, ERK, and p38 K (111). Osteoclastogenesis and bone resorption are modu-
MAPK pathways contribute considerably to the whole of lated by a series of growth factors, cytokines, and hormones
TGF-␤-induced responses, but further characterization is that can act directly on the osteoclast or indirectly through
needed to assess their importance in relation to the Smad- the osteoblast/stromal cell. Examples of proresorptive fac-
dependent and other TGF-␤-induced signaling pathways tors are TNF-␣, IL-1, 1,25-(OH)2D3, PTH, and PTHrP; among
(reviewed in Ref. 100). the antiresorptive factors, we find estrogens, calcitonin,
BMP-2 and -4, PDGF, calcium, interferon-␥, and IL-4, -10, -17,
and -18 (112).
III. TGF-␤1 in Bone A pivotal role in the bone-remodeling process has been
assigned to TGF-␤1 because it was proven to affect both bone
A. Introduction
resorption and formation. TGF-␤1 is secreted in a latent form
Bone is a mineralized tissue that serves many functions: by bone cells and is stored in the ECM. Active, resorbing
providing mechanical support to joints, tendons, and liga- osteoclasts are capable of activating TGF-␤1, which in turn
ments; protecting soft tissues; supporting hematopoiesis; attenuates further bone resorption by impairing osteoclas-
regulating blood calcium levels, etc. It consists largely of an togenesis and promotes bone formation through chemotactic
organic matrix of type I collagen and noncollagenous pro- attraction and stimulation of proliferation and differentiation
teins mineralized with hydroxyapatite crystals. Formation, of osteoblast precursors. Although this seems straightfor-
deposition, and mineralization of bone tissue are executed by ward, the story is much more complicated because it turned
the osteoblasts that differentiate from mesenchymal precur- out that the in vitro effects of TGF-␤1 on cells of the osteoblast
sor cells. The key transcription factor that drives the mes- and osteoclast lineage depend greatly on factors such as cell
enchymal precursor cell toward the osteoblast lineage and differentiation stage, cell density, TGF-␤1 concentration, the
controls bone formation is Runx2 (Cbfa1), which regulates presence of serum, and other culture conditions. In vivo, the
the expression of all known marker genes expressed by the presence of other growth factors in the bone environment
osteoblast (101). In addition, several other transcription fac- and the environment as such determine the exact outcome of
tors and homeobox proteins, such as Dlx5, Msx2, Bapx1, TGF-␤1 functioning.
Hoxa-2, Osx, and AP-1, affect osteoblast differentiation (102,
104). Local growth factors and cytokines regulating osteo- B. TGF-␤ isoforms in bone
blast differentiation include BMPs, fibroblast growth factors,
platelet-derived growth factor (PDGF), IGFs, and Indian All three TGF-␤ isoforms are detected in bone, but the
Hedgehog (103). Moreover, bone formation is regulated by TGF-␤1 isoform is the most abundant at the protein level
endocrine factors such as sex steroid hormones, PTH , 1,25- (113). In cartilage of mouse embryos, TGF-␤1 is highly ex-
dihydroxyvitamin D3 [1,25-(OH)2D3, the active metabolite of pressed in perichondrial cells, TGF-␤2 is expressed in chon-
vitamin D], and leptin (104, 105). drocytes, and TGF-␤3 is expressed in both (114). In embry-
Throughout life, bone tissue is continuously remodeled by onic bone, TGF-␤1 levels are high in periosteum and
the balanced processes of bone resorption and consecutive osteocytes. Little TGF-␤2 or -␤3 is detected in the periosteum,
but they are readily detected in osteocytes (114). In neonatal committed to osteogenesis, TGF-␤1 increases the pool of
and adult mice, TGF-␤1 protein is detected in bone marrow osteoprogenitors both by inducing chemotaxis (129 –131) and
cells, chondrocytes, and cartilaginous matrix (115). To our proliferation. Indeed, most studies illustrate the mitogenic
knowledge, no detailed information on protein expression of effect of TGF-␤1 on osteoprogenitors and osteoblast-
TGF-␤2 or -␤3 in the adult mouse is currently available. In enriched cell cultures (132–137), although some have reported
neonatal human bone, all isoforms can be found at sites of growth inhibition of osteoblast-like cells by this cytokine (138,
endochondral and intramembranous ossification but, again, 139). The biphasic, concentration-dependent effect of TGF-␤1
the patterns of expression differ. At sites of endochondral on osteoblast proliferation, with inhibition of DNA synthesis at
bone formation, TGF-␤1 and TGF-␤3 are detected in prolif- high concentrations, lies at the basis of this discrepancy (140).
erative and hypertrophic zone chondrocytes, and TGF-␤2 is Moreover, variables such as cell density, serum concentration,
detected in all zones of the cartilage. During intramembra- and differentiation stage were found to affect the outcome of
nous bone formation, TGF-␤1 and -␤2 colocalize with sites of TGF-␤1 treatment (138 –140). In sparse cultures, TGF-␤1 is in-

mineralization, whereas TGF-␤3 is more widely distributed hibitory at concentrations above 0.15 ng/ml, but this dose shifts
(116). Osteoclasts also express TGF-␤, mostly TGF-␤1, in high to higher levels as the cell density increases, with a peak re-
amounts (117). It is also important to note that expression of sponse at 15 ng/ml in confluent cultures. The underlying mech-
all three TGF-␤ isoforms is up-regulated during fracture anism is unknown, but it seems plausible to assume that the in
healing, suggesting that their roles are not restricted to em- vivo situation, in which osteoblasts are found in tight clusters,
bryonic bone development, but extend to adult bone remod- corresponds best to a confluent culture, implying that at phys-
eling (118). iological (low range) concentrations, TGF-␤1 stimulates osteo-
Almost all nonmalignant cells secrete TGF-␤1 as LLC in blast proliferation.
conjunction with a LTBP (see Section I.B). Bone cells, how- Differentiation of osteoblast precursors can be monitored
ever, form an exception: the SLC is the predominant form by the expression of proteins that compose the bone matrix,
and is secreted with great efficiency (19, 119, 120). The SLC e.g., type I collagen, osteopontin, and osteonectin, as well as
has not been observed in such large amounts in other cell by the expression of the osteoblast differentiation markers,
systems (121), suggesting an important function for this form ALP and, in a later stage, osteocalcin. Conflicting data con-
in bone. The SLC probably represents a pool of TGF-␤1 that cerning the effect of TGF-␤1 exist for most of these markers.
is readily available, while another part is deposited in the This is the case for the expression of type I collagen (132, 133,
bone matrix for storage through covalent binding of the LAP 135, 137, 138, 140 –143), other organic matrix components
with LTBP-1 (19) and possibly LTBP-3 (20). Moreover, such as fibronectin, plasminogen activator inhibitor-1, os-
TGF-␤1 is present at a physiologically significant level in teonectin, osteopontin, and decorin (135, 136, 138, 141–144),
plasma (122), and this source may contribute to the reservoir and ALP activity (136 –138, 140 –142, 145). Osteocalcin ex-
stored in the bone matrix. pression has been shown to be inhibited (142, 146, 147). We
believe that these ambiguous observations can again be at-
C. Role of TGF-␤1 in osteoblastogenesis and bone formation
tributed to differences in the osteoblastic cell model system
in vitro
(tumorigenic vs. nontumorigenic), culture conditions (e.g.,
serum concentration), cell density, TGF-␤1 concentration,
The cellular events involved in bone formation are che- and the presence of other growth factors. However, the most
motaxis and proliferation of osteoblast precursors, differen- important variable is the differentiation stage of the target
tiation to the mature osteoblast phenotype with synthesis of cell population with a stimulatory effect of TGF-␤1 on dif-
ECM proteins (e.g., type I collagen, osteopontin), and min- ferentiation of bone-forming cells in the early stage but an
eralization of the resulting matrix. Finally, osteoblasts either inhibitory effect in later stages.
undergo apoptosis or transdifferentiate to osteocytes or In recent years, some of the molecular mechanisms un-
bone-lining cells. All these events are under the control of derlying TGF-␤1 actions in bone formation have been elu-
both systemic hormones and local growth factors. cidated. AP-1 binding sites and/or SBEs were identified in
Data from numerous in vitro experiments have demon- the promoters of many bone matrix proteins, e.g., osteopon-
strated the role of TGF-␤1 in every stage of bone formation. tin, fibronectin, type I collagen, osteocalcin, and ALP, point-
Despite conflicting results (see below), most data support the ing to a role for both MAPK- and Smad-dependent signaling
following model. TGF-␤1 increases bone formation in vitro (71, 73). Crosstalk between the Smad and MAPK pathways
mainly by recruiting osteoblast progenitors and stimulating is also relevant in osteogenesis. Thus, despite the fact that
their proliferation, thus expanding the pool of committed TGF-␤1 inhibits ALP activity and mineralization in vitro,
osteoblasts, as well as by promoting the early stages of dif- Smad3 was shown to stimulate both processes in osteoblasts
ferentiation (bone matrix production). On the other hand, it (148). Inhibition of the ERK and JNK pathways antagonizes
blocks later phases of differentiation and mineralization (123, the inhibitory effect of TGF-␤1 on ALP activity, showing that
124). These later stages are regulated by other growth factors the MAPK pathways negatively regulate the Smad pathway
such as the BMPs (125). Apoptosis of osteoblasts is blocked (99). The observation that an anti-Smad4 antibody or expres-
by TGF-␤1 through maintenance of survival during trans- sion of dominant-negative Smad3 or Smad4 can down-reg-
differentiation into osteocytes (93, 126, 127). ulate TGF-␤1-induced AP-1 DNA binding in osteoblasts also
In contrast to the BMPs, TGF-␤1 is unable to induce os- points to a role for Smads in modulating AP-1 activity (98).
teogenesis in mesenchymal pluripotent cells, although it can In addition to acting directly, TGF-␤ signaling can also
inhibit differentiation to myogenic cells (128). However, once affect bone formation indirectly. A master factor in bone
formation is Runx2, also known as Cbfa1, a DNA-binding D. TGF-␤1 in osteoclast formation and bone resorption
transcription factor specific for cells of the osteogenic lineage in vitro
(149). Runx2 binding sites are found in the promoters of
Bone resorption involves the dissolution of bone mineral
several bone formation markers including collagen 1, ALP,
and the enzymatic degradation of the organic bone matrix by
osteopontin, RANKL, and osteocalcin (149). Runx2 is a com-
osteoclasts, giant MNCs. In vitro, spleen cells, bone marrow
mon target of TGF-␤1 and BMP2, mediating the inhibitory
effect of these factors on myogenic differentiation of C2C12 cells, peripheral blood mononuclear cells (PBMCs), and al-
pluripotent mesenchymal precursor cells (150). Induction of veolar macrophages can act as a source of osteoclast precur-
osteoblast-specific gene expression in these cells requires sors. The events of recruitment of osteoclast precursors to the
coordinated action between Runx2 and BMP2-induced bone environment, differentiation to the mature osteoclast,
Smad5 (151). In the early differentiation stage, TGF-␤1 in- bone resorption, and osteoclast apoptosis are all modulated
duces the expression of Runx2 in combination with BMPs. by TGF-␤1. Like many of the other cytokines influencing

However, in later stages of differentiation and maturation of osteoclastogenesis and/or bone resorption, TGF-␤1 does not
osteoblasts, TGF-␤1 opposes BMP2 actions (152). Smad3, solely modulate these processes by direct action on oste-
activated by TGF-␤1, physically interacts with Runx2 at oclasts and their precursors, but also acts via osteogenic cells.
Runx2-responsive elements, thus suppressing the expression Upon binding to its receptor on the osteoblast membrane,
of Runx2 and other osteogenic genes (collagen 1, ALP, os- expression of proteins involved in formation and activation
teocalcin) by an autoregulatory feedback mechanism (123). of osteoclasts is induced.
Furthermore, Smad2 overexpression decreases Runx2 The role of TGF-␤1 in osteoclastogenesis and bone resorp-
mRNA levels (153). Recently, menin was identified as an tion is very complex, and many seemingly contradictory
additional regulatory factor essential in promoting the BMP- reports have been published. Several important parameters
induced commitment of mesenchymal stem cells to osteo- must be taken into account when evaluating the studies
progenitors through physical and functional interaction with performed in the past: 1) Is the study looking at osteoclas-
Smads1/5 and Runx2 (154). It was shown that after osteo- togenesis (MNC formation) or at bone resorption by mature
blast commitment, menin turns into a repressor of osteoblast osteoclasts? 2) Is the study utilizing isolated osteoclast cul-
maturation: by binding to TGF-␤1-activated Smad3, it me- tures or a system where supporting cells (lymphocytes, stro-
diates the inhibitory effect of the latter on Runx2 activity mal cells) are present? 3) Which TGF-␤1 concentration has
(154). The inhibitory role of TGF-␤1 in late phase osteogen- been used? 4) How long and during which differentiation
esis in vitro was further confirmed through the use of a T␤RI stage has the growth factor been applied?
kinase inhibitor to suppress TGF-␤ signaling. Whereas BMP- Let us first turn our attention to MNC formation. In the last
induced osteoblast commitment was unaltered, osteoblast few years, a general model for the action of TGF-␤1 on
differentiation and matrix mineralization were stimulated osteoclastogenesis has emerged. According to this model
(124). (illustrated in Fig. 3), TGF-␤1 inhibits osteoclast formation in
We conclude that TGF-␤1 generally inhibits mineral- cocultures at high concentrations, while stimulating it in
ization of the matrix it helps to produce. However, the isolated cultures.
question is how responsive osteoblasts normally are to In the highly heterogeneous cell population of the bone
TGF-␤1 in their late differentiation stage. The answer marrow culture, the osteoblasts provide an endogenous
might lie in the flux of TGF-␤ receptors on the osteoblast source of RANKL. Several investigators reported a biphasic
membrane. A decrease in T␤RI and T␤RII expression is effect of TGF-␤1 on MNC formation in these cultures. Low
observed as human BMSCs progress from osteoprogenitor concentrations of TGF-␤1 (1–100 pg/ml) stimulated MNC
cells to maturing osteoblasts (36), confirming earlier find- formation, whereas high concentrations (0.1–10 ng/ml) were
ings in murine and rat osteoblastic cells that TGF-␤/ inhibitory (159 –162). Recently, two groups provided a pos-
receptor interactions decrease during osteoblast differen- sible explanation for this bidirectional effect. One set of ex-
tiation (155, 156). This would imply that osteoblasts in later periments examined the differential activation of different
stages are less sensitive to TGF-␤1. Moreover, TGF-␤1 TGF-␤-induced signaling pathways and found low concen-
itself has been shown to transiently or persistently (at low trations of TGF-␤1 to induce the ERK pathway in hemato-
and high concentrations, respectively) down-regulate the poietic cells isolated from adult human bone marrow; at
levels of all receptor types on the osteoblast surface, pri- higher concentrations, however, the p38 MAPK pathway
marily in late differentiation stage cells (157, 158). We was activated, suggesting that different MAPK pathways can
propose the hypothesis that receptor down-regulation evoke opposite responses (163, 164). Another group moni-
provides a way to decrease the responsiveness of the os- tored RANKL, osteoprotegerin (OPG), and M-CSF mRNA
teoblast toward TGF-␤1 to circumvent late-phase inhibi- expression in the function of the TGF-␤1 concentration in
tion by this cytokine. cocultures. From their findings, they conclude that osteoclast
From the above, it is clear that the effect of TGF-␤1 on in differentiation is stimulated at low TGF-␤1 concentrations
vitro osteogenesis is highly dependent on a broad range of because both the RANKL to OPG ratio and M-CSF levels are
experimental conditions and is the final outcome of many high. In contrast, at high TGF-␤1 concentrations, the RANKL
interacting factors. Interactions are expected to be fully elu- to OPG ratio is repressed as TGF-␤1 suppresses RANKL
cidated in the coming years and will be helpful to further expression and increases OPG expression by the osteoblast
explain the paradoxical findings reported in the past. (165). In combination with the dose-dependent inhibition by
FIG. 3. Role of TGF-␤1 in osteoclasto-

genesis. 1, Effect of TGF-␤1 on osteoclast
precursors in coculture. In the presence
of osteoblasts/stromal cells, TGF-␤1
binds to its receptor on the osteoblast

membrane (a) and modulates the expres-
sion of OPG and RANKL through Smad
(b)- and MAPK (c)- dependent pathways.
OPG expression and secretion are en-
hanced (d), whereas RANKL expression
is down-regulated (e). This balance shift
between OPG and RANKL results in a
maximal occupation of RANKL by OPG
and interferes with RANKL/RANK inter-
action (f). Consequently, osteoclastogen-
esis is impaired at high TGF-␤1 concen-
trations (g). 2, When added to isolated
hematopoietic precursor cultures to-
gether with RANKL and M-CSF, TGF-␤1
induces osteoclastogenesis. After binding
to its receptor on the osteoclast mem-
brane (a), TGF-␤1 induces expression of
RANK (b– d). The resulting increase in
RANK/RANKL interactions leads to
NF-␬B activation (e) and expression of
osteoclastogenesis-inducing genes (f).
TGF-␤1 of M-CSF expression, this results in inhibition of Others consistently observe a costimulation of RANKL/
osteoclast differentiation (166). M-CSF-induced MNC formation by TGF-␤1, both at low and
Duration of TGF-␤1 application affects the outcome of the high concentrations, in cultures of isolated M-CSF-depen-
experiment as well. Thus, a switch from inhibition to stim- dent bone marrow cells or other osteoclast precursors for
ulation of osteoclast formation was detected in a mixed cell which extreme care was taken to remove all contaminating
population (fetal long bone) subjected to high TGF-␤1 levels cells (169 –173). Direct effects of TGF-␤1 on osteoclast pre-
in the initial part of the culture period (d 1–3) or for longer cursors, such as up-regulation of RANK expression (172, 174)
time periods (d 1–7) (167). Furthermore, contamination of and induction of NF-␬B activation (169) and suppressor of
osteoclast cultures by lymphocytes (when using PBMCs as cytokine signaling expression (175), are responsible for this
starting material) was shown to influence osteoclastogenesis. positive effect. Moreover, because no OPG-expressing cells
Thus, high levels of TGF-␤1 present in the initial part of the
are present, TGF-␤1 is unable to induce OPG to counteract
culture period vastly increase RANKL-/M-CSF-induced
RANK/RANKL interaction.
MNC formation and bone resorption in a human lympho-
What about the effect of TGF-␤1 on bone resorption? Like-
cyte-rich population (91, 168). When TGF-␤1 is maintained
during the entire culture period (28 d) or applied at a later wise, this completely depends on the cellular context. When
stage of differentiation, this effect levels out, implying that the system is dependent on osteoclast recruitment from he-
the stimulatory effects of TGF-␤1 are restricted to the mono- matopoietic precursors for bone resorption, a process that is
cyte stage of the culture and shift to counteracting osteoclas- inhibited by TGF-␤1, resorption will be impaired. This is the
togenesis in pre- and mature osteoclasts (91). In a lympho- case, for example, in fetal long bones where the marrow
cyte-poor or pure monocyte population, osteoclast formation cavity is still developing and osteoclasts have not yet in-
evoked by M-CSF and RANKL was only modestly enhanced vaded. On the contrary, in calvaria or older long bones with
by TGF-␤1 (91, 168). an established marrow cavity, where mature osteoclasts are
already present, TGF-␤1 stimulates them to resorb bone at all ptosis and osteoclastogenesis (290). TGF-␤ has been shown
concentrations (162, 167, 176 –178). to modulate both processes (239, 291), but the underlying
mechanisms are thus far unknown. In fibroblasts, experi-
mental evidence points to a role for the NF-␬B and AP-1
E. Interaction of TGF-␤1 with other growth factors and
transcription factor families activated by TNF-␣ in TNF-␣/
hormones
TGF-␤ crosstalk. Upon TNF-␣ treatment, JNK-mediated c-
Upon elucidation of the Smad signaling pathway, it ap- Jun and JunB phosphorylation decreases Smad/DNA inter-
peared remarkably simple for such a complex group of cy- actions, either through formation of off-DNA Smad/Jun
tokines as the TGF-␤ superfamily: the ligand assembles a complexes or through competition of Smad and Jun for bind-
membrane receptor complex that activates the Smads, and ing to the coactivator p300 (241, 292). Moreover, a TNF-␣-
the Smads assemble complexes that regulate transcription. and NF-␬B-mediated up-regulation of Smad7 synthesis an-
However, since this discovery, an intricate web of crosstalk tagonizes TGF-␤ signaling in mouse embryonic fibroblasts

has been revealed. On the one hand, TGF-␤-induced path- (293). In a similar setup in human dermal fibroblasts, how-
ways (such as the Smad and MAPK signaling pathways) ever, Smad7 expression was not observed (292). In addition,
interact with each other (see Section II.C). On the other hand, Smad7 expression was shown to be decreased by TNF-␣ and
crosstalk occurs with pathways initiated by other local and NF-␬B in human embryonic kidney 293 cells through com-
systemic factors. The most important hormones and cyto- petition of NF-␬B with Smad3 (a stimulator of Smad7 ex-
kines are discussed in Table 1. It must be noted that some of pression) for binding to p300 (294). These examples show
the interactions have been reported in a specific culture that the effect of TNF-␣ on Smad signaling is influenced by
model and do not necessarily apply to other models as well. the cell system used. Further research needs to be performed
For a few of these factors the molecular mechanisms of syn- in bone model systems to elucidate the molecular basis of the
ergy and antagonism have been discussed below. crosstalk between these two growth factors in bone.
The active metabolite of vitamin D [1,25-(OH)2D3] is an Estrogen has powerful effects on cells both of the osteo-
important regulator of calcium homeostasis and a major blast and the osteoclast lineage (295). It stimulates osteoblast
player in the bone environment. It stimulates bone formation proliferation, differentiation, deposition of ECM, and min-
through the up-regulation of osteoblast differentiation and eralization (296). On the other hand, osteoclast maturation
ECM mineralization (reviewed in Ref. 248). Moreover, 1,25- and function are impaired, whereas osteoclast apoptosis is
(OH)2D3 inhibits PTH-induced bone resorption at physio- promoted (261, 297). Taken together, estrogen is a potent
logical doses (250), making it a useful drug for the treatment anabolic agent in bone. In vivo, this is illustrated by the bone
of osteoporosis. 1,25-(OH)2D3 signals via the nuclear vitamin loss upon estrogen depletion after the menopause. Accumu-
D receptor (VDR), which binds to vitamin D-responsive el- lating evidence points to a role for TGF-␤1 in mediating some
ements in the promoter of target genes (e.g., osteocalcin and of the effects of estrogen, e.g., promotion of murine osteoclast
osteopontin). TGF-␤ and 1,25-(OH)2D3 can both synergize or apoptosis (261). In vivo, TGF-␤1 mRNA and protein levels
antagonize each other’s functions. In recent years, some of decrease in ovariectomized rats (298, 299), whereas estrogen
the mechanisms of this reciprocal crosstalk have been elu- treatment of postmenopausal women stimulates TGF-␤1 and
cidated. 1,25-(OH)2D3 up-regulates mRNA and protein ex- TGF-␤2 mRNA and protein production (300). Information
pression of TGF-␤2 (and possibly other TGF-␤ isoforms) from the mechanism of interplay between TGF-␤ and estro-
through binding of VDR/retinoic X receptor-␣ heterodimers gen signaling pathways comes from studies in human em-
to distinct sequences in the TGFB2 promoter (287). In mouse bryonic kidney 293T and human breast cancer cells, in which
osteoblastic cells, 1,25-(OH)2D3 synergistically stimulated crosstalk is mediated by physical interaction between Smads
TGF-␤1-induced c-Jun expression, binding of AP-1 to its and estrogen receptor (ER)-␣. These studies identified ER-␣
response element, and transcriptional activity of AP-1 (288). as a transcriptional corepressor for Smad activity (301).
The effect of TGF-␤1 on 1,25-(OH)2D3 functioning seems to Smad3, on the other hand, can enhance ER-mediated tran-
diverge at the MAPK/Smad level. Thus, up-regulation of scriptional activity (301), although Smad4 behaves as a tran-
AP-1 activity by TGF-␤1 in an osteosarcoma cell line was scriptional corepressor, suppressing the Smad3-mediated
suggested to antagonize osteocalcin and osteopontin expres- ER-␣ transactivation (302).
sion through steric hindrance of 1,25-(OH)2D3-dependent From the data presented in Table 1, it can be concluded
protein/DNA interactions by AP-1 family members (255). that the effects of TGF-␤1 on bone formation and resorption,
On the other hand, synergy between 1,25-(OH)2D3 and both in vitro and in vivo, must be evaluated in view of the
TGF-␤1 in activating osteocalcin expression, as observed in presence of other cytokines and hormones, which modulate
a study using COS-1 cells, was dependent on the close prox- or are modulated by TGF-␤1 signaling in a number of ways.
imity of the vitamin D responsive elements and SBEs in the
osteocalcin promoter (254). Indeed, upon overexpression,
Smad3 can act as a coactivator of 1,25-(OH)2D3 signaling via IV. Bone Phenotypes Associated with Abnormal
physical interaction between Smad3 and VDR (289). Thus, it TGF-␤1 Signaling
appears that signals transmitted through Smad and MAPK A. Bone phenotypes of knockout and transgenic mouse
pathways evoke opposite effects on 1,25-(OH)2D3 models of the TGF-␤ signaling pathway
functioning.
TNF-␣, a proinflammatory cytokine, has profound effects Information about the function of TGF-␤ and its down-
in the bone environment, as it induces both osteoblast apo- stream signaling mediators can be gained from the study of
TABLE 1. Interaction of TGF-␤1 with other growth factors and hormones in bone
Effect on ob/bone Effect on oc/bone

Effect on TGF-␤1 Effect by TGF-␤1 Synergy with TGF-␤1?
formation resorption
Cytokines
BMP-2 ⫹ Ob differentiation ⫹ Oc differentiation, ⫹ TGF-␤1 mRNA ⫺ BMP mRNA ⫹ In vivo fracture
in uncommitted survival and expression in expression in fetal healing (185)
progenitors (128) function (182, 183) human ob-like cells rat calvarial cell ⫺ TGF-␤1 antagonizes
⫹ In vivo bone (184) cultures (142) ALP, OCN, and
formation and ⫺ Reduction of cell Dlx5 expression
fracture healing surface T␤RII (155) induced by BMP2
(179 –181) (152, 186, 187)
⫹ Runx2 expression in
early differentiation

stage (78)
⫺ BMP2 suppresses
mitogenic effect of
TGF-␤1 on fetal rat
ob (155)
⫹ Collagen synthesis
in fetal rat ob (155)
IGF ⫹ Ob proliferation and ⫹ Oc genesis (188) / ⫹ IGF-I and IGFBP3 ⫹ Ob differentiation
collagen synthesis Couples bone mRNA expression and proliferation
(188, 189) resorption to in human BMSCs (193, 194)
⫹ Bone formation in formation (188) (190) ⫹ Fracture healing in
vivo (188) ⫹ IGF-I release from vivo (195)
ob (191)
⫺ IGF-I mRNA and
protein expression
in ob (192)
bFGF ⫹ Ob replication (196) ⫹ Oc genesis (directly) / ⫹ bFGF mRNA and ⫹ Proliferation of
⫹ Fracture healing (198) protein expression bovine ob (204)
(197) ⫺ Oc genesis in ob (202) and
⫺ Bone matrix (indirectly) (199) osteosarcoma cells
collagen synthesis ⫹ Oc genesis (203)
(196) (indirectly) (200)
⫺ Bone resorption by
mature oc (200)
⫹ Bone resorption by
mature oc (directly
or indirectly) (201)
PDGF ⫹ Ob replication (205) ⫹ Oc replication and / ⫺ Expression of ⫹ Proliferation of
⫺ ECM formation bone resorption PDGF-␣ receptor primary human ob-
(205) (208) (209) derived cultures
⫹ Bone density and ⫹ Expression of (210)
fracture healing in PDGF-␣ receptor
vivo (206, 207) (210)
⫹ PDGFB mRNA
expression in ob
(211)
M-CSF ⫺ Ob formation (212) ⫹ Oc formation, / ⫹ M-CSF expression ⫹ Oc genesis (in
migration, survival and activity in Ob combination with
(213, 214) and BMSCs (215) RANKL) (112)
⫺ Bone resorption in ⫺ M-CSF expression ⫹ Oc genesis (in the
isolated oc (213) in cocultures (166) absence of RANKL)
(216)
IL-1 ⫹ Ob proliferation and ⫹ Oc differentiation ⫹ TGF-␤1 activity in ⫺ IL-1 receptor ⫺ TGF-␤1 inhibits IL-
differentiation in and function (219, calvaria (221) expression level on 1-induced bone
vitro (217) 220) ⫺ TGF-␤1 secretion by oc precursors (223) resorption in rat
⫺ Ob proliferation in ob-like cells (222) long bone cultures
vitro (218) and calvarial ob
(177, 224)
⫺ TGF-␤1 opposes
enhanced survival
of hematopoietic
precursors in
response to IL-1
(225)
TABLE 1. Continued

IL-6 ⫺ Ob differentiation ⫹ Oc differentiation / ⫹ IL-6 mRNA ⫺ TGF-␤1 opposes
in vitro (226) and function expression in enhanced survival
⫹ Ob differentiation (indirectly) (219, primary rat ob (228) of hematopoietic
in vitro (227) 220) ⫹ IL-6 mRNA and precursors in
protein expression response to IL-6
in murine BMSCs (225)
(229) ⫺ TGF-␤1 inhibits IL-
⫹ IL-6 expression by 6-induced bone
human ob (230) resorption in
calvarial ob (224)

IL-11 ⫺ Ob differentiation ⫹ Oc differentiation / ⫹ IL-11 production by ⫺ TGF-␤1 opposes
(231) and function (233) Ob (234) enhanced survival
⫹ Bone formation in of hematopoietic
vivo (232) precursors in
response to IL-11
(225)
TNF-␣ ⫹ Ob proliferation ⫹ Oc genesis from / / ⫹ TGF-␤ potentiates
(235) bone marrow TNF-␣-stimulated
⫹ Ob apoptosis (236) macrophages (220, oc formation (238)
⫺ Ob function (235) 237) ⫺ TGF-␤1 attenuates
TNF-␣-induced ob
apoptosis and bone
resorption (224,
239)
⫺ TNF-␣ opposes the
inhibitory effects of
TGF-␤1 on
proliferation of bone
marrow
hematopoietic
precursors and
TGF-␤-induced up-
regulation of
collagen (240, 241)
IFN-␥ ⫺ Bone formation in ⫺ Oc genesis and bone / / ⫺ TGF-␤1 antagonizes
vivo (242) resorption (244) IFN-␥-induced
⫺ DNA and bone ⫹ Oc genesis and bone suppression of oc
collagen synthesis resorption in vivo formation (238)
(243) (242, 245)
Hormones
1,25 ⫹ Expression of type I ⫹ Bone resorption at ⫹ TGF-␤1 secretion ⫹ VDR expression in ⫹ FN and ALP
(OH)2D3 collagen, FN, ALP, high doses and binding by mature ob cell line expression, ALP
OCN in mature ob (indirectly) (250) human ob (251, (253) activity, type I
(246, 247) ⫺ Bone resorption in 252) collagen synthesis
⫹ Mineralization of vitro at and OCN
ECM (248) physiological doses expression in ob
⫺ Ob precursor and in vivo (250) (246, 254)
proliferation (248) ⫺ TGF-␤1 antagonizes
⫹/⫺ Ob proliferation 1,25(OH)2D3-
(concentration- induced synthesis of
dependent) (249) OCN and OPN in
ob (255, 256)
⫹ IGFBP-3 production
in human ob
progenitors (257)
⫺ TGF-␤1 inhibits
1,25(OH)2D3-
induced bone
resorption in rat
long bone cells (258)
TABLE 1. Continued

Estrogen ⫹ Ob proliferation and ⫹ Apoptosis of oc and ⫹ TGF-␤1 production / /
differentiation (259) oc precursors (261) in ob (263)
⫹ Bone formation in ⫺ Oc maturation and ⫹ Estrogen prevents
vivo (260) function (262) bone loss through a
⫺ Bone resorption in TGF-␤-dependent
vivo (260) mechanism (264)
⫹ Estrogen-induced oc
apoptosis in vitro
and in vivo is
mediated by TGF-␤

(261)
Gluco- ⫹ Ob apoptosis (265) ⫹ Bone resorption in ⫹ Activate TGF-␤1 / ⫹ Oc formation at
corticoids ⫺ Ob proliferation and vitro and in vivo secreted by ob (268) low concentrations
differentiation (266) (266, 267) ⫺ Redistribute TGF- (271)
⫺ Bone formation in ␤1 binding from
vivo (267) T␤RI/T␤RII toward
T␤RIII (269)
⫺ Expression and
activity of T␤RI
(indirect) (270)
PGE2 ⫹ Bone formation ⫹ Bone resorption ⫹ TGF-␤1 expression ⫹ PGE2 production by
through replication through replication in Ob (274) ob (176)
and differentiation and differentiation ⫹ TGF-␤1 stimulates
of ob precursors of oc precursors ob proliferation
(272) (272) through PGE2-
⫹ Bone formation in dependent
vivo (273) mechanism (275)
PTH ⫹ Bone formation in ⫹ Bone resorption ⫹ TGF-␤1 activity in ⫺ PTH/PTHrP /
vitro and in vivo (indirectly) calvaria (221) receptor expression
(intermittent (prolonged ⫹ TGF-␤-induced in pre-ob (279)
administration) exposure) (276) collagen I ⫹ PTH receptor
(276) production in ob expression in
(277) osteosarcoma cells
⫹ TGF-␤1 expression (280)
in bone in vitro and
in vivo (263, 278)
Gonadal ⫹ Proliferation and ⫺ Bone resorption in ⫹ TGF-␤1 expression / /
androgens differentiation of ob vitro (285) and secretion in ob,
(281) osteosarcoma cells
⫺ Proliferation of and oc (282, 285,
osteosarcoma cells 286)
and ob cell line ⫺ TGF-␤1 mRNA
(282, 283) expression in ob cell
⫹ Maintenance of line (283)
bone mass and
integrity in vivo
(284)
/ indicates that no effect or interaction has been reported thus far. ob, Osteoblast; oc, osteoclast; OCN, osteocalcin; IGFBP, IGF-binding
protein; FN, fibronectin; OPN, osteopontin; bFGF, basic fibroblast growth factor. When not specified, the experiment has been performed in
vitro.
knockout mice, in which one intermediate of the pathway has before 4 wk of age or treated with an immunosuppressive
been eliminated by gene targeting. In Table 2, an overview drug to minimize the effects of the excessive inflammation
has been presented of the targeted deletions of the genes that they develop. Bone mineral content in the metaphyses,
encoding the ligands TGF-␤1, -␤2, and -␤3, the binding pro- width of the growth plates, and length and elasticity of the
teins LTBP-3 and -4, the receptors T␤RI and T␤RII, and the long bones were significantly decreased. The study was
intracellular mediators involved in Smad-dependent TGF-␤ broadened by the work of Atti et al. (322). Histology showed
signaling, Smad2, -3, and -4. With the exception of ltbp4 null that osteoblasts were practically absent in the trabecular
mice, all knockout mice that outlive the stage of osteogenesis bone. Growth plate thickness was reduced due to alterations
develop severe bone defects. This finding stresses the im- in chondrocyte proliferation and differentiation. Imbalance
portance of TGF-␤ signaling in the ossification process both between modeling activity (absent because of osteoblast de-
during embryonic development and postnatally. pletion) and osteoclast activity (present) resulted in thinning
Geiser et al. (306) studied in detail the bone phenotype of of the cortical bone. Bones were more fragile and had a low
the tgfb1 knockout mouse as a model for the role of TGF-␤1 mineral-to-matrix ratio. Together, these findings point to an
in postnatal bone development. Mice were either studied important role of TGF-␤1 in bone modeling and quality and
TABLE 2. Overview of knockout mouse models of the TGF-␤1 signaling pathway
Gene Mutant phenotype Bone phenotype Refs.

tgfb1 ⬃50% intrauterine death around E10.5 due to defects in No change in BMD 303–306
yolk sac vasculogenesis and hematopoiesis Decreased bone mineral content in metaphyses
⬃50% develops severe wasting syndrome 2–3 wk after Width of growth plate diminished
birth with inflammatory cell infiltration of primarily Longitudinal growth rate diminished
lymphocytes and neutrophils in many organs, leading Bone less elastic and more fragile
to death at 3–5 wk (percentages depend on genetic Osteoblasts practically absent in trabecular bone
background)
tgfb2 Perinatal mortality due to congenital cyanosis Rib, sternum, vertebral, limb and mandible defects 307
Wide range of developmental defects Reduction in bone size and cranial ossification
Incompletely penetrant defective palatogenesis
tgfb3 Delayed pulmonary development Incompletely penetrant defective palatogenesis 308, 309

Death within 24 h after birth
ltbp3 Only bone phenotype Craniofacial abnormalities from day 10 postnatally 310
due to premature ossification of the
synchondroses
Osteosclerosis and osteoarthritis; growth
retardation
ltbp4 Abnormal lung development, cardiomyopathy, colorectal No bone phenotype 311
cancer
Profound defects in elastic fiber structure
Reduced deposition of TGF-␤ in the EC space
tgfbr1 100% embryonic lethality around E10.5 due to severe / 312
defects in yolk sac vasculogenesis and anemia, but the
hematopoietic potential is not deficient
tgfbr2 100% embryonic lethality around E10.5 due to defects / 313
in yolk sac vasculogenesis and hematopoiesis
smad2 100% embryonic lethality around E6.5 due to failure to / 314, 315
form the extraembryonic portion of the egg cylinder;
abnormal gastrulation, no mesoderm induction
smad3 Disruption exon 1: Viable, but reduced size; fibroblasts – Disruption exon 1: Osteopenia due to 316 –319
and T-cells are no longer responsive to TGF-␤1 in impairment of inhibition of osteoblast
vitro differentiation and apoptosis by TGF-␤
Disruption exon 2: Viable, but reduced size; develop – Disruption exon 2: No obvious defects to account
colorectal cancer between 4 and 6 months of age for the reduced size
Disruption exon 8: Death between 1 and 8 months of – Disruption exon 8: degenerative joint disease due
age due to progressively developing immune to loss of inhibition of terminal chondrocyte
dysregulation and susceptibility to infection from the differentiation by TGF-␤
time of weaning onwards
smad4 100% embryonic lethality before E7.5 due to failure to / 320, 321
gastrulate, absence of mesoderm formation and
abnormal visceral endoderm development; embryonic
growth is retarded
Rescued embryos show severe anterior truncations
/ Indicates that the bone phenotype could not be investigated because of the embryonic lethality of the knockout. E, Embryonic day; EC,
extracellular.
shed new light on the in vitro studies. Thus, it appears that and TGF-␤3 mRNA levels are unaltered (306), which pro-
osteoblast proliferation, matrix deposition, and collagen ma- vides additional evidence for their nonredundancy.
turity are severely diminished in the absence of TGF-␤1, as A large percentage of TGF-␤ is found in association with
could be expected from earlier in vitro experiments. How- a LTBP, which facilitates folding and secretion and targets
ever, mineralization of the cortical bone was reduced, pos- the complex to the ECM (see Section I.B). Therefore, absence
sibly pointing to a positive role for TGF-␤1 in in vivo min- of one of the LTBP isoforms could influence normal TGF-␤
eralization. It must be noted that this mineralization defect processing and storage. The ltbp2 null mouse is not included
could be secondary to the delay in collagen maturation (as in Table 2 because LTBP-2 is unable to bind TGF-␤ and
collagen is important for mineral deposition) or be simply a therefore has no role in the formation of the LLC. Although
consequence of the reduced osteoblast number. TGF-␤1 LTBP-4 can bind TGF-␤, the absence of a bone phenotype in
withdrawal did not seem to affect osteoclast formation and the ltbp4 knockout mouse (311) suggests that this LTBP iso-
function in vivo. form is not involved in the folding and storage of TGF-␤ in
Tgfb2 (307) and tgfb3 (308, 309) knockouts both show ab- bone. The bone phenotype observed when knocking out ltbp3
normalities in bone development, although far more severe (310) (see Table 2) implies that this isoform is probably re-
in the former (see Table 2). The marked differences of the sponsible for the formation of a percentage of the LLC found
bone phenotypes of the three tgfb knockout mice reflect the in bone tissue. This is in accordance with earlier observations
differences in bone-specific expression of the isoforms. It is that human LTBP-3 is expressed in osteoblasts and some
also important to note that in tgfb1 knockout mice, TGF-␤2 osteosarcoma cell lines and secreted as a LLC in conjunction
with TGF-␤1 (20). Unfortunately, no knockout mouse model of transcriptional activation (e.g., Smad3 can bind directly to
has been developed for ltbp1, which is presumed to be an DNA, whereas Smad2 is unable to do so). In this context, it
important TGF-␤1-binding protein in bone as well (19, 323). is worth noting that Felici et al. (325) recently identified a new
T␤RI or T␤RII deficiency is embryonic lethal, precluding T␤RII-interacting protein, TLP1, that plays a role in regulat-
any study of the effect of their absence on bone development ing the balance between Smad2 and Smad3 signaling. The
(312, 314, 315). The severity of the phenotypes highlights the marked differences of the respective knockout phenotypes
requirement of TGF-␤ signaling during embryonic develop- confirm the functional specificity of these two R-Smads in
ment. Recently, a conditional knockout of tgfbr2, which limits vivo.
absence of the type II receptor to Col2a-expressing cells, was At present, no data are available on the bone-specific over-
developed (324). The majority of the mice did not survive expression of tgfb1. However, Erlebacher and Derynck (326)
postnatally, but bone defects were examined in 13.5-d targeted a constitutively active form of tgfb2 under the os-
through 17.5-d embryos. Bone abnormalities (size reduc- teocalcin promoter to mature osteoblasts. Heterozygous

tions) were confined to the parts of the skull that develop mice with a 16-fold increase in TGF-␤2 production showed
through endochondral ossification and the spine, demon- a dramatic, age-dependent loss of bone mass reminiscent of
strating a crucial role for T␤RII in axial skeleton develop- high-turnover osteoporosis. The long bones were normal in
ment. The appendicular skeleton remained normal until 1–12 length, but their cortices were thinner; clavicles were prac-
wk after birth, when a progressive reduction in the length of tically absent. Histologically, meta- and epiphyseal trabec-
the proximal long bones was noted. Although tgfbr2 is re- ulation were dramatically reduced. Osteocyte density and
dundant in embryonic long bone development, it appears to osteoprogenitor cell numbers were markedly increased, and
function postnatally. mineralization in the cortical bone was impaired. These ef-
As intracellular mediators of the canonical TGF-␤ signal- fects could be accounted for by the increased activities of both
ing pathway, the absence of one of the TGF-␤-induced R- osteoblasts and osteoclasts. By expressing a cytoplasmati-
Smads or the Co-Smad is anticipated to profoundly affect cally truncated T␤RII from the osteocalcin promoter, Fil-
normal development. Because Smad4 is the common medi- varoff et al. (327) inhibited TGF-␤ signaling in osteoblasts.
ator Smad for all TGF-␤ superfamily members, it is not un- The resultant bone phenotype was in many ways the oppo-
expected that its absence causes 100% embryonic lethality site of that of the tgfb2-overexpressing mice. An age-depen-
(320, 321). Likewise, Smad2 deficiency inevitably leads to dent increase in trabecular bone mass was associated with
embryonic death (314, 315). Remarkably, the phenotype ob- decreased osteoblast differentiation and osteoclastic resorp-
served in smad3 knockouts depends greatly on the strategy tion. Generation of double transgenic mice, overexpressing
used to knock out the gene. Thus, mice with a targeted both TGF-␤2 and the truncated T␤RII, allowed workers to
disruption in exon 1 or 2 are viable (317–319), whereas dis- discern osteoblast-dependent and -independent effects (328).
ruption of exon 8 causes death between 1 and 8 months of age These studies confirm the role of TGF-␤ (in this case the
(316). These discrepancies prove the need to carefully exam- TGF-␤2 isoform) in bone remodeling and in keeping the
ine expression and function of the mutant protein. Indeed, balance between bone formation and resorption. The find-
Yang et al. (316) observed that Smad3ex8/ex8 has residual ings are surprising in the light of the in vivo stimulation of
activity and is capable of suppressing TGF-␤-induced re- bone formation upon administration of TGF-␤1 (see Section
porter activity at supraphysiological levels in in vitro assays. V) or -␤2 (329, 330). Several explanations come to mind. First,
In our opinion it cannot be excluded that at physiological restriction of TGF-␤2 expression to the mature, nondividing
levels, the mutant protein still interacts with cofactors or osteoblast is not fully representative of the natural in vivo
DNA, thus inhibiting, for example, proper functioning of the situation, in which TGF-␤ is expressed at all stages of os-
other TGF-␤/activin-specific R-Smad. In bone, loss of Smad3 teoblast differentiation. It is even plausible that forced ex-
was shown to result in osteopenia or degenerative joint dis- pression in the late differentiation stage inhibits bone for-
ease when exon 1 or 8, respectively, is disrupted. In the mation, as shown in vitro for TGF-␤1 as well (see Section III.C).
former, TGF-␤-mediated proliferation is intact, but TGF-␤ is In addition, the major isoform in bone is TGF-␤1, accounting
no longer able to inhibit osteoblast differentiation, thereby for up to 90% of TGF-␤ found in the bone environment.
increasing the osteocyte fate and eventually leading to ap- Second, osteoblast activity is most perturbed in the epiphyses
optosis. In the latter, terminal differentiation of chondrocytes and diaphyses, which are the sites of highest osteocalcin
is no longer inhibited by TGF-␤. The reason for this discrep- expression, suggesting that the action of TGF-␤2 is restricted
ancy is currently unclear but might involve differences in to its sites of production. Again, this does not correspond to
genetic background and the presence of modifier loci. De- the normal in vivo situation. Third, the authors report an
spite these differences, the main conclusion is that Smad3 increased bone resorption without alterations in osteoclast
seems to lack a crucial role during embryonic development number. However, it should be taken into consideration that
but is indispensable in adult tissues. In bone, not bone for- because osteoclasts lack betaglycan, they are probably rela-
mation as such, but rather bone remodeling and maintenance tively unresponsive to TGF-␤2, as has been shown before for
are affected by loss of Smad3. Although Smad2 and Smad3 hematopoietic progenitor cells (331). High levels of TGF-␤2
have 95% sequence identity and have been used interchange- might overcome the low affinity of TGF-␤2 for T␤RII in this
ably in mediating TGF-␤ signaling in in vitro assays in the mouse model, but the direct effect of physiological levels of
past, it is now well established that they activate separate sets TGF-␤2 on osteoclasts remains to be elucidated. Therefore,
of target genes, as shown recently by Yang et al. (80) (see we feel that observations in these mouse models need further
Section II.C), as well as display differences in their mechanism experimental support. Construction of a transgenic mouse
overexpressing tgfb1 or tgfb2 under a promoter for early RANKL and M-CSF (see Section III.D). Overactivity of
osteoblast differentiation, such as collagen type I, would be TGF-␤1 is therefore expected to promote this process. In a
helpful in unraveling the in vivo effects of these isoforms in coculture with osteoblast/stromal cells, however, which
bone formation. more closely reflects the in vivo reality, the mutant protein is
likely to inhibit osteoclast formation. Once more, this is an
example of how in vitro experiments can lead to a conclusion
B. TGFB1 mutations in the pathogenesis of Camurati-
that is not in line with the in vivo reality, because the culture
Engelmann disease (CED)
conditions do not take into account the in vivo interactions
CED or progressive diaphyseal dysplasia is a rare bone taking place.
disorder with an autosomal dominant mode of inheritance. The phenotype of CED patients contrasts with the reported
Radiologically, it is characterized by hyperostosis and scle- phenotype of the tgfb2 transgenic mouse (see Section IV.A).
rosis of the diaphyses of the long bones and sclerosis at the This can be attributed to different reasons. First, knocking out

skull base. Patients suffer mainly from bone pain, muscle tgfb1 and tgfb2 has shown that the isoforms are functionally
weakness, a waddling gait, and fatigue (332). In 2000, we and nonredundant (Section IV.A), making it difficult to compare
others succeeded in identifying TGFB1 as the disease-causing their in vivo activities after overexpression. Second, the 16-
gene underlying this bone disorder (333, 334). Ten different fold overexpression of TGF-␤2 is restricted to the mature
mutations have been reported thus far (333–338) (Table 3). osteoblast, whereas in CED patients, the limited TGF-␤1
With one exception, a duplication of three Leu-residues in overactivity is present during all phases of osteoblast dif-
the signal peptide, all are missense mutations in the LAP. ferentiation and in all bone cells. Construction of a knock-in
Recently, we were able to show that the mutations can be mouse model carrying one of the CED mutations would
functionally divided into two groups (337). The first group, therefore be a valuable tool to gain information on osteoblast
represented by mutations concentrated around the cysteine and osteoclast functioning upon tgfb1 overactivity.
residues responsible for dimerization of the LAP (see Fig.1), The relatively mild phenotype of CED patients is surpris-
affects activation of the mutant protein upon overexpression: ing when taking into account the versatile processes in which
secretion is normal, but the amount of active protein is ap- TGF-␤1 is implicated during embryogenesis and adult life
proximately doubled, due to destabilization of the dimer- and is in sharp contrast to the severity of the phenotypes
ization process. This is reflected by a major overinduction of observed after knocking out the gene or overexpressing it in
Smad-dependent signaling, as measured by the phosphor- a tissue-specific manner. We suggest the following hypoth-
ylation level of Smad2 and luciferase activity evoked by a esis. In most tissues, TGF-␤1 appears as a high-molecular
TGF-␤ responsive transcriptional reporter. The second group weight latent complex (LLC) in conjunction with a LTBP and
of mutants, located at the N terminus, shows severe impair- is stored as such in the ECM (340). However, bone cells
ment of TGF-␤1 secretion. Nevertheless, the Smad-depen- produce predominantly the SLC, consisting of the LAP and
dent transcriptional response is likewise increased. The ob- the mature peptide, but excluding the LTBP (19, 119, 120).
servation of signaling in the absence of extracellular active This form is suggested to represent a pool of readily available
protein led us to formulate a hypothesis of intracrine sig- TGF-␤1, necessary in an environment in which this cytokine
naling in which the TGF-␤ receptor complex, which is con- plays such an important role throughout life (19). The effect
stantly internalized and recycled back to the plasma mem- of a particular mutation might depend on the nature of the
brane, initiates signaling when encountering active TGF-␤1 latent complex, with a role for the LTBP in neutralizing the
intracellularly. However, this hypothesis awaits further ex- conformational changes brought forth by the mutations,
perimental evidence. thereby restricting full activity of the mutant protein to the
The phenotype displayed by CED patients, impaired bone bone environment. Furthermore, it is intriguing to note that
resorption at the endosteal side in combination with overall affected cell types, osteoblasts, chondrocytes, adipocytes,
activity of the osteoblasts at the periosteal side, is in line with and myocytes, originate from the same mesenchymal stem
the presumed action of the mutant protein: the gain-of-func- cell population. Consequently, it would be interesting to
tion mutations stimulate bone formation while impairing investigate the nature of the latent complex in fat and muscle
osteoclastogenesis. However, McGowan et al. (339) reported tissue.
on the enhancement of osteoclast formation and bone re-
sorption in vitro by PBMCs of patients harboring one of the C. Osteolytic metastases: a role for TGF-␤1 in malignancy
mutations. This inconsistency can be explained by the ex-
perimental setup: in isolated cultures, TGF-␤1 has indeed Of the different tumor types known to be associated with
been found to enhance osteoclast formation in concert with osteolytic lesions, breast carcinoma is the most common: in
TABLE 3. Overview of reported CED mutations in TGF␤1
Exon 1 Exon 2 Exon 4

Mutations (DNA) 28_36dup 241T⬎C 463C⬎T 652C⬎T 653G⬎A 664C⬎G ? 667T⬎C 667T⬎G 673T⬎C
Mutations (protein) L10 –L12dup Y81H R156C R218C R218H H222D C223S C223R C223S C225R
No. of families reported 2 2 3 18 8 1 1 1 1 6
Percentage 4.7 4.7 7.0 41.9 18.6 2.3 2.3 2.3 2.3 14.0
Percentage per exon 9.3 7.0 83.7
more than 80% of patients with advanced disease, breast D. Bone-related association studies
cancer metastasizes to bone where it gives rise to osteolytic
Osteoporosis is a multifactorial bone disorder character-
lesions, both by activating osteoclastic bone resorption and
ized by low bone mass and increased bone turnover, leading
inhibiting osteoblastic bone formation. These lesions can give
to nontraumatic fractures. It is the most common bone-
rise to pain, hypercalcemia, fractures, and nerve-compres-
associated pathology and its socioeconomic impact is high.
sion syndromes. The avidity of breast tumors for the bone Therefore, the search for genetic risk factors for osteoporosis
environment is due to the high concentration of growth is intensive. Because of the key role of TGF-␤1 in bone,
factors present, possibly in combination with favorable in- affecting both bone resorption and formation, numerous
teractions between specific receptors on the bone marrow studies have investigated the effect of TGFB1 polymor-
endothelial cells and cell surface structures on the osteotropic phisms on susceptibility to osteoporosis, bone mineral den-
tumor (for review, see Ref. 341). Heavily vascularized areas sity (BMD), and bone turnover.
of the skeleton, such as the red bone marrow of the axial

The single nucleotide polymorphisms (SNPs) detected in
skeleton and the proximal ends of the long bones, ribs, and TGFB1 can be divided in three classes, according to their
vertebral column, are the most likely to be affected. position: promoter SNPs, coding SNPs, and intronic SNPs.
PTHrP turned out to be one of the key hormones in breast The promoter polymorphisms C-1348T and G-1369A possi-
cancer-mediated osteolysis. First evidence for its role was bly affect proper gene expression. The coding polymor-
provided in 1996 by Guise et al. (342), who found concen- phisms T29C and C788T can alter, respectively, the traffick-
trations of PTHrP to be increased in bone marrow plasma ing and tertiary structure of the protein. The functional
from affected bones of nude mice inoculated with human importance of intronic polymorphisms is questionable, but
breast cancer cells. PTHrP-neutralizing antibodies could in- they could influence proper splicing or be in linkage dis-
hibit new osteolysis and decrease osteolytic bone destruction equilibrium with the true functional variations.
and tumor burden in bone. This identified PTHrP as the In Table 4, an overview is presented of all bone-related
tumor product that stimulates bone resorption, thus contrib- association studies performed. This overview shows that it
uting to osteolysis. PTHrP was shown to up-regulate proves difficult to come to conclusive or consistent evidence
RANKL production by the osteoblasts, while down-regulat- for an association between a given SNP and parameters such
ing OPG production (343). as BMD, bone mass, bone turnover, and osteoporosis risk;
The role of TGF-␤1 as a promotor of tumorigenesis and this is illustrated by the coding variation Leu10Pro (T29C).
tumor invasion has long been acknowledged: it induces the The CC genotype of this SNP has been associated with in-
epithelial-to-mesenchymal transition necessary for invasion creased BMD, lower susceptibility to osteoporosis, decreased
and contributes to changes in the microenvironment favor- frequency of vertebral fractures, and/or increased response
able for tumor growth and angiogenesis (reviewed in Ref. to 1,25-(OH)2D3 treatment in postmenopausal Japanese
344). Expression of a dominant-negative T␤RII in a murine women (357, 358). Langdahl et al. (352) confirmed the asso-
model of bone metastases decreases bone destruction, tumor ciation with BMD, but they could not detect any association
growth, osteoclast number, and mortality, suggesting a role with osteoporotic fractures in their Danish population. Re-
for TGF-␤ in tumor-mediated osteolysis (345). The interac- markably, three other studies (359 –361) found BMD to be
tion between PTHrP and TGF-␤1, resulting in a vicious cycle higher with the TT genotype in their Caucasian populations,
whereas one report makes mention of an association of the
of tumor growth and osteolysis, was established not long
TC genotype with decreased BMD and increased fracture
afterward (reviewed in Ref. 346). TGF-␤1 promotes PTHrP
rate (356). Still other groups failed to detect any correlation
production by the tumor cell, which in turn stimulates bone
between the Leu10Pro variation and osteoporosis, BMD,
resorption, leading to the release of more active TGF-␤1.
bone mass, or bone quality among Korean or Caucasian
Recently, evidence was obtained that bone degradation is
women (354, 362–364). In overexpression studies, the Pro (C)
expedited through inhibition of osteoblast adhesion and dif- form of TGF-␤1 was shown to cause a 2.8-fold increase in
ferentiation (347) and induction of osteoblast apoptosis (348). secretion compared with the Leu (T) form (368), but this
A role for TGF-␤1 in these processes is suspected but must finding is not supported by in vivo data. Whereas Yamada et
be explored in further detail. al. (357) did observe an association of the CC genotype with
Recently, transcriptional profiling of subpopulations of elevated TGF-␤1 serum concentrations, a European study
MDA-MB-231 human breast cancer cells with low or en- found association with the opposite genotype (359). Two
hanced bone metastatic abilities uncovered a set of genes that other studies detected no association at all (361, 369).
mediate bone metastasis (349). When overexpressed, these There are several possible explanations for the discrepan-
genes cooperatively stimulate osteolytic bone metastasis by cies reported for this and other SNPs. Differences in genetic
promoting homing to bone, angiogenesis, invasion, or oste- background (which might alter linkage disequilibrium and
oclast recruitment. PTHrP was not identified in this screen modifier genes), in status of the population (age, gender,
because this protein is mainly regulated by TGF-␤ at postmenopausal status), in environmental influences, in study
transcriptional level. Two genes identified in this screen, the design, or in statistical analysis can all attribute to differences
angiogenic factor connective tissue growth factor and the in the reported results. To quantify the contribution of TGFB1
osteoclast differentiation factor IL-11, were found to be ac- genetic variants to the natural variation of BMD in the gen-
tivated by TGF-␤ and may thus, like PTHrP, participate in the eral population, we feel that studies should be extended to
cycle of tumor growth and osteolysis. contain larger sample sizes to circumvent the lack of infor-
TABLE 4. Bone-related association studies with TGF␤1 polymorphisms
SNP Position Effect Population Refs.

C-1348T Promoter TT: 2 BMD Japan (postmenopausal) 350
1 prevalence of osteoporosis
7 BMD Korea (postmenopausal) 351
7 prevalence of osteoporosis
TT: 1 BMD Denmark (mixed gender; women pre- and 352, 353
7 risk of osteoporotic fractures postmenopausal)
7 BMD UK (postmenopausal) 354
7 BMD Caucasian (nuclear families) 355
7 BMD China (postmenopausal) 356
G-1369A Promoter AA: 1 bone loss UK (postmenopausal) 354
T29C Exon 1 CC: 1 BMD Japan (postmenopausal) 357, 358

Leu10Pro (signal peptide) 2 susceptibility to osteoporosis
2 vertebral fracture rate
1 response to 1,25-(OH)2D3
treatment
CC: 1 BMD Denmark (mixed gender) 352
7 risk of osteoporotic fractures
TT: 1 BMD Caucasian (postmenopausal) 359 –361
7 osteoporosis Caucasian, Korea (postmenopausal) 362, 363
7 bone mass and quality Korea (postmenopausal) 364
TC: 2 BMD China (postmenopausal) 356
1 fracture rate
C788T Exon 5 7 BMD UK (postmenopausal) 354
Thr263Ile (mature peptide)
713-8delC Intron 4 delC: 2 bone mass Denmark (pre- and postmenopausal 365
1 bone turnover osteoporotic women)
1 risk of osteoporosis
delC: 1 risk of fragility fractures Italy (postmenopausal) 366
1 bone turnover
2 total hip BMD
T861-20C Intron 5 CC: 2 BMD UK (premenopausal) 367
1 risk of osteoporosis at the femoral
neck
TT: 1 bone mass Denmark (mixed gender) 352
Protection against osteoporotic
fractures
7 BMD China (postmenopausal) 356
1, Increased; 2, decreased; 7, no effect.
mation content in some of the current studies. In addition to mote fracture healing or reverse the excessive bone resorp-
this, metaanalysis, uniting all information generated in the tion seen in osteoporosis. An advantage in the use of TGF-␤1
studies performed so far, could lead to a more overall view. is the conservation of its mature peptide across species, en-
Finally, a haplotype-based study— combining data of SNPs suring biological activity of recombinant human TGF-␤1 in
distributed over the entire gene, rather than the analysis of several animal models without facing the problem of anti-
one or a few selected SNPs in the gene—is recommended to body responses. There are, however, several disadvantages.
gain full insight in the role of TGF-␤1 in the pathogenesis of First, TGF-␤1 not only modulates bone formation, but can
osteoporosis. also stimulate osteoclast formation and function under cer-
tain circumstances (see Section III.D). Consequently, treat-
ment may induce both bone formation and resorption. Sec-
V. Therapeutic Use of TGF-␤1 as ond, the half-life of TGF-␤1 is short (⬃2 min), implying the
Bone-Forming Agent need for a matrix to allow for a slow release of the growth
In bone, TGF-␤1 plays an important role in keeping the factor. Third, TGF-␤1 is implicated in diverse functions out-
balance between the two tightly regulated processes of bone side the bone environment, suggesting that its systemic ap-
resorption and subsequent bone formation (370). Moreover, plication might cause unwanted side effects.
like other growth factors (BMPs, fibroblast growth factors, A growth factor or hormone can be applied systemically
IGFs, PDGFs), TGF-␤1 is highly expressed during fracture or locally. Although systemic administration offers the ad-
healing (118), suggesting that its role is not restricted to bone vantage of simplicity, this is not an option in the case of
development and turnover, but extends to the process of TGF-␤1: due to the widespread tissue distribution of the
bone repair. Consequently, TGF-␤1 is one of the growth TGF-␤ receptors, serious side effects develop in diverse or-
factors considered for use as a bone-forming agent to pro- gans upon systemic administration of TGF-␤1 (371). Conse-
quently, most experimental settings involve a local admin- are promising for the concomitant use of these growth factors
istration of TGF-␤1, either as a single dose or continuously, in fracture healing.
in a free form, associated with a carrier or produced by Until a few years ago, treatment of osteoporosis relied
regional gene therapy. almost exclusively on the use of antiresorptive agents such
Table 5 gives an overview of studies performed over recent as estrogen, calcium, 1,25-(OH)2D3, calcitonin, and bisphos-
years. A few reports describe the successful use of a single phonates (406). Although these drugs are effective in pre-
local dose of free recombinant human TGF-␤1 (373–376), but venting further bone loss, they cannot restore the microar-
others conclude that it is not capable of promoting clinically chitectural damage already made. Therefore, new strategies
relevant osteogenesis in calvarial defects (372), although this have been developed, using drugs that stimulate bone for-
could be attributed to the low dose applied. As a biphasic mation, e.g., fluoride, PTH, GH, and recombinant growth
effect of TGF-␤1 concentration on osteoblast proliferation in factors such as IGF (reviewed in Ref. 407). The use of TGF-␤1
vitro has been detected (Section III.C), concentration might in treating osteoporosis deals with the same problems as its

also be a relevant issue in vivo. To overcome the loss of free use in fracture healing. In addition, a drug for osteoporosis
recombinant growth factor through diffusion and inactiva- should work systemically. Until now, TGF-␤1 has not been
tion, administration can be made continuous by repeated applied in the treatment of osteoporosis.
injections, the use of a pump system, association with os-
teoinductive scaffolds, or gene therapy. Dose, application
VI. Concluding Remarks
mode and site, follow-up time, and animal species are seen
to affect the final outcome of the therapy. Overall, positive The importance of TGF-␤1 in bone development and ho-
effects of TGF-␤1 on fracture healing and bone formation meostasis has been extensively illustrated both in vitro and
predominate. Some investigators report that the effect of the in vivo with strong evidence for profound effects on bone
initial wave of bone formation caused by TGF-␤1, induced by formation, bone resorption, and the interplay between these
increasing the amount of osteoblasts, not the rate of bone two processes. However, major problems have arisen in at-
formation as such, is erased by a concomitant increase in tempting to define the precise effects of TGF-␤1 on different
bone resorption (380 –382). This could explain why the fol- aspects and stages of these processes. Results obtained in
low-up time plays a role in determining the final outcome vitro are often not in line or even contradictory to in vivo
[compare, for example, studies by Tanaka et al. (376), Zhou observations. Moreover, ambiguous effects of TGF-␤1 are
et al. (382), and Tieline et al. (398)]. However, the in vivo effect observed when different in vitro experiments are compared,
of TGF-␤1 administration on osteoclasts has been investi- indicating that the results are influenced by the experimental
gated by only a few laboratories, making it difficult to con- setup. This reflects the fact that TGF-␤1 interacts with a wide
clude whether this is a general problem. range of other growth factors and hormones in bone, gen-
From the reports presented in Table 5, it can be concluded erating an intricate network of interpathway crosstalk, thus
that, in most cases, TGF-␤1 promotes fracture healing and provoking a complex response. The key feature for future in
bone formation in vivo. This contrasts with the results ob- vitro research in this field is to carefully control and minimize
tained in in vitro assays, in which the final stage of bone experimental variability.
formation, i.e., mineralization of the bone matrix, is inhibited Physiological relevance for in vitro observations can be
(see Section III.C), but corresponds with findings in tgfb1 gained from the study of in vivo models. The development of
knockout mice (Section V.A) and CED patients (Section V.B). mouse models and the occurrence of TGFB1 mutations un-
Based on the in vivo results, we propose TGF-␤1 to have only derlying a human sclerosing bone dysplasia have allowed
a minor active role in ECM mineralization under physiolog- researchers to find in vivo evidence for the role of this cyto-
ical circumstances. In our opinion, the observed inhibition of kine in bone, but there is clearly a lack of bone-specific
mineralization in vitro is accomplished mainly by the forced knockout or transgenic mice that could solve current ambi-
presence of this growth factor during a phase of bone for- guities on the functioning of TGF-␤1 in bone tissue. For
mation in which it has no major function. The observed further investigation of the complex role of TGF-␤1 in bone,
down-regulation of TGF-␤ receptors as osteoblasts mature is the analysis of inducible or cell type-specific knockout and
in line with this way of thinking. transgenic mice of various TGF-␤ signaling components, in-
Despite these positive reports, the osteoinductive capacity cluding receptors and Smads, will be highly informative.
of TGF-␤1 is rather weak when compared with that of the Several questions remain open and are subject to future
BMPs. As yet, no clinical application has been developed for research. How can we come to an “ideal” in vitro system for
TGF-␤1, whereas BMP-2 and BMP-7 have already been used the bone milieu—mimicking as perfectly as possible the in
in clinical trials and have proven their efficacy in the healing vivo situation—to avoid further discrepancies? Is the role of
of critical-sized fibular defects and tibial nonunions in hu- TGF-␤1 in vivo restricted to stimulating early stages of bone
mans (403– 405). In an in vivo experimental model of fracture formation or does it extend to later stages? If yes, does it have
healing, BMP-2 induced differentiation of mesenchymal cells an inhibitory or stimulatory effect on mineralization? How
into osteoblasts and chondrocytes during intramembranous exactly does TGF-␤1 influence osteoclast formation and func-
bone formation and early chondrogenesis (commitment tion under physiological conditions (both direct and indi-
phase), whereas TGF-␤1 expression correlated with active rect)? Also, additional research is required to elucidate the
differentiated osteoblasts and chondrocytes during chondro- precise role of the different TGF-␤ isoforms in bone. More-
genesis and endochondral ossification (maturation phase) over, further evidence must be gained on the potential role
(185). These complementary functions of BMP-2 and TGF-␤1 of TGF-␤1 as a therapeutic anabolic agent.
TABLE 5. TGF-␤1 as bone-forming agent
Mode of application Test animal Doses Follow-up time Site of application Effect Refs.
Single local application of free rhTGF-␤1 by sc injection
Rabbit 0.1 ␮g 6 wk Skull defect Defect filled with soft 372
connective tissue;
almost no bone
regeneration
Rabbit 0.4 or 2 ␮g 1–70 d Skull defect Dose-dependent 373
decrease of defect
area during first 35
d
Rabbit 1 or 5 ␮g 28, 70 or 180 d Skull defect Complete bridging of 373
defect after 28 d
Rabbit 1 ␮g 4 wk Skull defect Partial closure 374
Rat 5 or 50 ng 8d Subperiosteal layer of Induction of woven 375

parietal bone bone formation
Neonatal rat 1 ␮g 12 d Periosteum of parietal Increased bone matrix 376
bone formation
Continuous local application of free rhTGF-␤1
Injection Rat 5, 20 or 200 ng (daily 16 d Center of parietal bone Decrease in mineral 375
for 14 d) apposition rate;
inhibition of
lamellar bone
formation
Rat 40 ng (each second d 40 d Tibial fracture Increase in callus 377
for 40 d) formation and
strength
Neonatal rat 20 or 200 ng rpTGF-␤1 3 and 5 wk Subperiosteal layer of Induction of bone 378
(daily for 14 d) the femur formation via
cartilaginous
intermediate and
intramembranous
bone formation
Neonatal rat 0.2 or 1 ␮g (daily for 13 or 20 d Periosteum of parietal 13 d: stimulation of 376, 379
12 d) bone bone formation at
the site of injection
20 d: no effect of
treatment
Mouse 1 ␮g (daily for 5 d) 8 or 19 d sc over frontal and Stimulation of bone 380
parietal bones formation and
mineralization
Mouse 2.5 ␮g (daily for 5 d) 6, 10, 20 or 40 d sc over calvaria Production of new 381
bone matrix that
mineralizes within 2
wk
Mouse 5 ␮g (daily for 5 d) 6d sc over calvaria Stimulation of 381
periosteal formation
of woven bone and
bone resorption at
endosteal surface
Pump system Rabbit 200 ng (daily for 2, 4 2, 4, or 6 wk Tibia After 2 and 4 wk: 382
or 6 wk) marked difference in
trabecular bone
volume; after 6 wk:
no difference due to
active bone
resorption
Rabbit 1 or 10 ␮g (daily for 6 6 wk Midtibial osteotomy Increase in maximal 383
wk) bending strength
and callus
formation, without
affection of bone
mineral content or
cortical thickness
Continuous local application of rhTGF-␤1 associated with a carrier (osteoconductive scaffold)
Calcium carbonate Rabbit 1 ␮g 4 wk Skull defect Strong promotion of 374
healing
Rat 1, 5 or 25 ␮g 8 wk Critical size defect in Significant 384
periosteum of enhancement of
parietal bone bone formation
Gelatin sponge Rat 2, 5 or 10 ␮g 2 wk Calvarial defect No clinically relevant 385
osteogenesis
Collagen sponge Rabbit 0.1 ␮g 6 wk Skull defect Bone regeneration 372
Dog 1, 5 or 10 ␮g 4, 6 or 8 wk Mandibular defect Enhanced calcified 386
bone formation
Collagenous matrix Baboon 5, 30 or 100 ␮g/g 30 d Calvarial defect Limited chondro- 387
matrix osteogenesis
TABLE 5. Continued
Mode of application Test animal Dosis Follow-up time Site of application Effect Refs.
Tricalcium Dog 0.3 or 3 ␮g 6 wk Condyle 0.3 ␮g: 3-fold increase 388
phosphate in fixation of the
implant and
maximal stimulation
of bone growth
3 ␮g: 2-fold increase in
fixation of the
implant and
maximal stimulation
of bone volume
Dog 0.3 ␮g 6 wk Condyle Both ob and oc exhibit 389
higher activity, but
bone formation

predominates;
significantly
increased bone
volume and bone
surface
Octacalcium Rat 0.5 ␮g 2, 4, or 8 wk Defect in parietal bone 4 wk: new bone 390
phosphate formation higher
than with
octacalcium
phosphate alone
Tricalcium Dog 120 or 335 ␮g 4 wk Gap in the cancellous 3-fold increase in bone 391
phosphate ⫹ bone of humerus growth
hydroxyapatite
Demineralized bone Rabbit 0.4, 4 or 40 ␮g 4 or 8 wk Skull critical size 4 wk: higher 392
defect promotion of bone
formation in 40
␮g-group
8 wk: bone formation
equivalent in all
treated groups
Sheep 750 ␮g 12 wk Tibial diaphyseal Bridging of the defect 393
defect
Lactic acid/glycolic Rat 250 –750 ␮g/g implant 6 wk Critical calvarial No induction of new 394
acid polymer ⫹ defect bone formation
demineralized
bone
Lactic acid/glycolic Rat 10 ␮g 4 wk Midshaft fracture tibia Strong stimulation of 395
acid polymer fracture healing
Dog 10 ng 12 wk Nonunion in radial Failure to induce 396
diaphysis significant healing
Calcium sulfate Rat 780 ␮g/g implant 6 wk Critical calvarial Strong induction of 394
defect model new bone formation
Hydroxyapatite Rat 1, 33, or 1000 ng 6 wk Proximal tibia Negative correlation 397
cylinders between dose and
new bone formation
Bioabsorbable self- Rat 50 ␮g 1, 3, or 6 wk Distal femur 3 wk: increase in new 398
reinforced bone formation
polylactide pin 1 and 6 wk: no
increase
Biodegradable Rabbit 2 ␮g 6 or 12 wk Calvarial defect Retarded osseous 399
polymer ceramic regeneration due to
composite incomplete
degradation of
carrier
Gelatin microsphere Rabbit 0.5 ␮g 3 wk Skull bone gap Significant bone 400
healing and higher
BMD
Methylcellulose Baboon 1 or 10 ␮g 22 d Osseous wound in the Increased ob 401
placed in analytic tibial diaphysis proliferation; oc
bone implant number, and activity
not altered (same in
both groups)
No increase in
trabecular bone
volume
Regional gene therapy
Adenoviral transfer Rat / 1 wk Osseous tissue Significant changes in 402
epiphyseal plate
Rabbit / 16 wk Femur defect Elevated bone matrix 179
synthesis
ob, Osteoblast; oc, osteoclast; rh, recombinant human; rp, recombinant porcine. /, Not applicable.
Acknowledgments 19. Dallas SL, Park-Snyder S, Miyazono K, Twardzik D, Mundy GR,

Bonewald LF 1994 Characterization and autoregulation of latent
We thank Dr. Rutger van Bezooyen for helpful comments on this transforming growth factor ␤ (TGF ␤) complexes in osteoblast-like
manuscript. We apologize to those researchers whose work was not cell lines. Production of a latent complex lacking the latent TGF
included in the review because of space restraints. ␤-binding protein. J Biol Chem 269:6815– 6821
20. Penttinen C, Saharinen J, Weikkolainen K, Hyytiainen M, Keski-
Address all correspondence and requests for reprints to: Professor Dr. Oja J 2002 Secretion of human latent TGF-␤-binding protein-3
Wim Van Hul, Department of Medical Genetics, University of Antwerp, (LTBP-3) is dependent on co-expression of TGF-␤. J Cell Sci 115:
Campus Drie Eiken, Building T6, Universiteitsplein 1, 2610 Antwerp, 3457–3468
Belgium. E-mail: wim.vanhul@ua.ac.be 21. Taipale J, Koli K, Keski-Oja J 1992 Release of transforming growth
This work was supported by a grant (G.0404.00) from the “Fonds voor factor-␤ 1 from the pericellular matrix of cultured fibroblasts and
Wetenschappelijk onderzoek” (FWO) and an Interuniversity Attraction fibrosarcoma cells by plasmin and thrombin. J Biol Chem 267:
Pole grant (to W.V.H.) K.J. holds a postdoctoral position at the 25378 –25384
FWO. P.t.D. is supported by a grant from the Netherlands Organization 22. Taipale J, Lohi J, Saarinen J, Kovanen PT, Keski-Oja J 1995 Hu-

for Health and Development (MW 902–16-295). S.J. is a postdoctoral man mast cell chymase and leukocyte elastase release latent trans-
fellow at the European Molecular Biology Organization (EMBO). forming growth factor-␤ 1 from the extracellular matrix of cultured
human epithelial and endothelial cells. J Biol Chem 270:4689 – 4696
23. Annes JP, Munger JS, Rifkin DB 2003 Making sense of latent TGF␤
References activation. J Cell Sci 116:217–224
24. Annes J, Vassallo M, Munger JS, Rifkin DB 2004 A genetic screen
1. Massagué J 1990 The transforming growth factor-␤ family. Annu to identify latent transforming growth factor ␤ activators. Anal
Rev Cell Biol 6:597– 641 Biochem 327:45–54
2. Assoian RK, Komoriya A, Meyers CA, Miller DM, Sporn MB 1983 25. Oreffo RO, Mundy GR, Seyedin SM, Bonewald LF 1989 Activa-
Transforming growth factor-␤ in human platelets. Identification of tion of the bone-derived latent TGF ␤ complex by isolated oste-
a major storage site, purification, and characterization. J Biol Chem oclasts. Biochem Biophys Res Commun 158:817– 823
258:7155–7160 26. Oursler MJ 1994 Osteoclast synthesis and secretion and activation
3. Seyedin SM, Thomas TC, Thompson AY, Rosen DM, Piez KA of latent transforming growth factor ␤. J Bone Miner Res 9:443– 452
1985 Purification and characterization of two cartilage-inducing 27. Derynck R, Feng XH 1997 TGF-␤ receptor signaling. Biochim Bio-
factors from bovine demineralized bone. Proc Natl Acad Sci USA phys Acta 1333:F105–F150
82:2267–2271 28. Lopez-Casillas F, Wrana JL, Massague J 1993 Betaglycan presents
4. Roberts AB, Lamb LC, Newton DL, Sporn MB, De Larco JE, ligand to the TGF ␤ signaling receptor. Cell 73:1435–1444
Todaro GJ 1980 Transforming growth factors: isolation of polypep- 29. Sankar S, Mahooti-Brooks N, Centrella M, McCarthy TL, Madri
tides from virally and chemically transformed cells by acid/ethanol JA 1995 Expression of transforming growth factor type III receptor
extraction. Proc Natl Acad Sci USA 77:3494 –3498 in vascular endothelial cells increases their responsiveness to trans-
5. Moses HL, Serra R 1996 Regulation of differentiation by TGF-␤. forming growth factor ␤ 2. J Biol Chem 270:13567–13572
Curr Opin Genet Dev 6:581–586 30. Stenvers KL, Tursky ML, Harder KW, Kountouri N, Amatayakul-
6. Massague J, Blain SW, Lo RS 2000 TGF␤ signaling in growth Chantler S, Grail D, Small C, Weinberg RA, Sizeland AM, Zhu
control, cancer, and heritable disorders. Cell 103:295–309 HJ 2003 Heart and liver defects and reduced transforming growth
7. Verrecchia F, Mauviel A 2002 Transforming growth factor-␤ sig-
factor ␤2 sensitivity in transforming growth factor ␤ type III re-
naling through the Smad pathway: role in extracellular matrix gene
ceptor-deficient embryos. Mol Cell Biol 23:4371– 4385
expression and regulation. J Invest Dermatol 118:211–215
31. Lopez-Casillas F, Payne HM, Andres JL, Massague J 1994 Beta-
8. Siegel PM, Massague J 2003 Cytostatic and apoptotic actions of
glycan can act as a dual modulator of TGF-␤ access to signaling
TGF-␤ in homeostasis and cancer. Nat Rev Cancer 3:807– 821
receptors: mapping of ligand binding and GAG attachment sites.
9. Pepper MS 1997 Transforming growth factor-␤: vasculogenesis,
J Cell Biol 124:557–568
angiogenesis, and vessel wall integrity. Cytokine Growth Factor
Rev 8:21– 43 32. Chen W, Kirkbride KC, How T, Nelson CD, Mo J, Frederick JP,
10. O’Kane S, Ferguson MW 1997 Transforming growth factor ␤s and Wang XF, Lefkowitz RJ, Blobe GC 2003 ␤-Arrestin 2 mediates
wound healing. Int J Biochem Cell Biol 29:63–78 endocytosis of type III TGF-␤ receptor and down-regulation of its
11. Letterio JJ, Roberts AB 1998 Regulation of immune responses by signaling. Science 301:1394 –1397
TGF-␤. Annu Rev Immunol 16:137–161 33. Cheifetz S, Bellon T, Cales C, Vera S, Bernabeu C, Massague J,
12. Gorelik L, Flavell RA 2002 Transforming growth factor-␤ in T-cell Letarte M 1992 Endoglin is a component of the transforming
biology. Nat Rev Immunol 2:46 –53 growth factor-␤ receptor system in human endothelial cells. J Biol
13. Gentry LE, Lioubin MN, Purchio AF, Marquardt H 1988 Molec- Chem 267:19027–19030
ular events in the processing of recombinant type 1 pre-pro-trans- 34. McAllister KA, Grogg KM, Johnson DW, Gallione CJ, Baldwin
forming growth factor ␤ to the mature polypeptide. Mol Cell Biol MA, Jackson CE, Helmbold EA, Markel DS, McKinnon WC,
8:4162– 4168 Murrell J, McCormick MK, Pericak-Vance MA, Heutink P, Oos-
14. Gentry LE, Webb NR, Lim GJ, Brunner AM, Ranchalis JE, tra BA, Haitjema T, Westerman CJJ, Porteous ME, Guttmacher
Twardzik DR, Lioubin MN, Marquardt H, Purchio AF 1987 Type AE, Letarte M, Marchuk DA 1994 Endoglin, a TGF-␤ binding
1 transforming growth factor ␤: amplified expression and secretion protein of endothelial cells, is the gene for hereditary haemorrhagic
of mature and precursor polypeptides in Chinese hamster ovary telangiectasia type 1. Nat Genet 8:345–351
cells. Mol Cell Biol 7:3418 –3427 35. Bourdeau A, Dumont DJ, Letarte M 1999 A murine model of
15. Miyazono K, Olofsson A, Colosetti P, Heldin CH 1991 A role of hereditary hemorrhagic telangiectasia. J Clin Invest 104:1343–1351
the latent TGF-␤1-binding protein in the assembly and secretion of 36. Walsh S, Jefferiss C, Stewart K, Beresford JN 2003 TGF␤1 limits
TGF-␤1. EMBO J 10:1091–1101 the expansion of the osteoprogenitor fraction in cultures of human
16. Nunes I, Gleizes PE, Metz CN, Rifkin DB 1997 Latent transform- bone marrow stromal cells. Cell Tissue Res 311:187–198
ing growth factor-␤ binding protein domains involved in activation 37. Massagué J 1998 TGF-␤ signal transduction. Annu Rev Biochem
and transglutaminase-dependent cross-linking of latent transform- 67:753–791
ing growth factor-␤. J Cell Biol 136:1151–1163 38. Sekelsky JJ, Newfeld SJ, Raftery LA, Chartoff EH, Gelbart WM
17. Rifkin DB 2004 Latent TGF-␤ binding proteins: orchestrators of 1995 Genetic characterization and cloning of mothers against dpp,
TGF-␤ availability. J Biol Chem 280:7409 –7412 a gene required for decapentaplegic function in Drosophila mela-
18. Hyytiainen M, Penttinen C, Keski-Oja J 2004 Latent TGF-␤ bind- nogaster. Genetics 139:1347–1358
ing proteins: extracellular matrix association and roles in TGF-␤ 39. Savage C, Das P, Finelli AL, Townsend SR, Sun CY, Baird SE,
activation. Crit Rev Clin Lab Sci 41:233–264 Padgett RW 1996 Caenorhabditis elegans genes sma-2, sma-3, and
sma-4 define a conserved family of transforming growth factor ␤ ton requires signaling by small GTPases Cdc42 and RhoA. Mol Biol
pathway components. Proc Natl Acad Sci USA 93:790 –794 Cell 13:902–914
40. Padgett RW, Savage C, Das P 1997 Genetic and biochemical anal- 65. Atfi A, Djelloul S, Chastre E, Davis R, Gespach C 1997 Evidence
ysis of TGF ␤ signal transduction. Cytokine Growth Factor Rev for a role of Rho-like GTPases and stress-activated protein kinase/
8:1–9 c-Jun N-terminal kinase (SAPK/JNK) in transforming growth fac-
41. Wrana JL, Attisano L 2000 The Smad pathway. Cytokine Growth tor ␤-mediated signaling. J Biol Chem 272:1429 –1432
Factor Rev 11:5–13 66. Engel ME, McDonnell MA, Law BK, Moses HL 1999 Interdepen-
42. Shi Y, Massague J 2003 Mechanisms of TGF-␤ signaling from cell dent SMAD and JNK signaling in transforming growth factor-␤-
membrane to the nucleus. Cell 113:685–700 mediated transcription. J Biol Chem 274:37413–37420
43. ten Dijke P, Hill CS 2004 New insights into TGF-␤-Smad signal- 67. Rodriguez-Barbero A, Obreo J, Yuste L, Montero JC, Rodriguez-
ling. Trends Biochem Sci 29:265–273 Pena A, Pandiella A, Bernabeu C, Lopez-Novoa JM 2002 Trans-
44. Zhang F, Laiho M 2003 On and off: proteasome and TGF-␤ sig- forming growth factor-␤1 induces collagen synthesis and accumu-
naling. Exp Cell Res 291:275–281 lation via p38 mitogen-activated protein kinase (MAPK) pathway
45. Izzi L, Attisano L 2004 Regulation of the TGF␤ signalling pathway in cultured L(6)E(9) myoblasts. FEBS Lett 513:282–288

by ubiquitin-mediated degradation. Oncogene 23:2071–2078 68. Rousse S, Lallemand F, Montarras D, Pinset C, Mazars A, Prunier
46. Shi W, Sun C, He B, Xiong W, Shi X, Yao D, Cao X 2004 GADD34- C, Atfi A, Dubois C 2001 Transforming growth factor-␤ inhibition
PP1c recruited by Smad7 dephosphorylates TGF␤ type I receptor. of insulin-like growth factor-binding protein-5 synthesis in skeletal
J Cell Biol 164:291–300 muscle cells involves a c-Jun N-terminal kinase-dependent path-
47. Tsukazaki T, Chiang TA, Davison AF, Attisano L, Wrana JL 1998 way. J Biol Chem 276:46961– 46967
SARA, a FYVE domain protein that recruits Smad2 to the TGF␤ 69. Chin BY, Petrache I, Choi AM, Choi ME 1999 Transforming
receptor. Cell 95:779 –791 growth factor ␤1 rescues serum deprivation-induced apoptosis via
48. Wu G, Chen YG, Ozdamar B, Gyuricza CA, Chong PA, Wrana JL, the mitogen-activated protein kinase (MAPK) pathway in macro-
Massague J, Shi Y 2000 Structural basis of Smad2 recognition by phages. J Biol Chem 274:11362–11368
the Smad anchor for receptor activation. Science 287:92–97 70. Zhou G, Lee SC, Yao Z, Tan TH 1999 Hematopoietic progenitor
49. Di Guglielmo GM, Le Roy C, Goodfellow AF, Wrana JL 2003 kinase 1 is a component of transforming growth factor ␤-induced
Distinct endocytic pathways regulate TGF-␤ receptor signalling c-Jun N-terminal kinase signaling cascade. J Biol Chem 274:13133–
and turnover. Nat Cell Biol 5:410 – 421 13138
50. Wu JW, Fairman R, Penry J, Shi Y 2001 Formation of a stable 71. Chung KY, Agarwal A, Uitto J, Mauviel A 1996 An AP-1 binding
heterodimer between Smad2 and Smad4. J Biol Chem 276:20688 – sequence is essential for regulation of the human ␣2(I) collagen
20694 (COL1A2) promoter activity by transforming growth factor-␤.
51. Qin BY, Lam SS, Correia JJ, Lin K 2002 Smad3 allostery links J Biol Chem 271:3272–3278
TGF-␤ receptor kinase activation to transcriptional control. Genes 72. Brodin G, Ahgren A, ten Dijke P, Heldin CH, Heuchel R 2000
Dev 16:1950 –1963 Efficient TGF-␤ induction of the Smad7 gene requires cooperation
between AP-1, Sp1, and Smad proteins on the mouse Smad7 pro-
52. Inman GJ, Hill CS 2002 Stoichiometry of active smad-transcription
moter. J Biol Chem 275:29023–29030
factor complexes on DNA. J Biol Chem 277:51008 –51116
73. Banerjee C, Stein JL, Van Wijnen AJ, Frenkel B, Lian JB, Stein GS
53. Chacko BM, Qin BY, Tiwari A, Shi G, Lam S, Hayward LJ, De
1996 Transforming growth factor-␤ 1 responsiveness of the rat
Caestecker M, Lin K 2004 Structural basis of heteromeric smad
osteocalcin gene is mediated by an activator protein-1 binding site.
protein assembly in TGF-␤ signaling. Mol Cell 15:813– 823
Endocrinology 137:1991–2000
54. Miyazawa K, Shinozaki M, Hara T, Furuya T, Miyazono K 2002
74. Keeton MR, Curriden SA, van Zonneveld AJ, Loskutoff DJ 1991
Two major Smad pathways in TGF-␤ superfamily signalling. Genes
Identification of regulatory sequences in the type 1 plasminogen
Cells 7:1191–1204 activator inhibitor gene responsive to transforming growth factor
55. Sirard C, Kim S, Mirtsos C, Tadich P, Hoodless PA, Itie A, Max- ␤. J Biol Chem 266:23048 –23052
son R, Wrana JL, Mak TW 2000 Targeted disruption in murine cells 75. Hall MC, Young DA, Waters JG, Rowan AD, Chantry A, Edwards
reveals variable requirement for Smad4 in transforming growth DR, Clark IM 2003 The comparative role of activator protein 1 and
factor ␤-related signaling. J Biol Chem 275:2063–2070 Smad factors in the regulation of Timp-1 and MMP-1 gene expres-
56. Dai JL, Schutte M, Bansal RK, Wilentz RE, Sugar AY, Kern SE sion by transforming growth factor-␤ 1. J Biol Chem 278:10304 –
1999 Transforming growth factor-␤ responsiveness in DPC4/ 10313
SMAD4-null cancer cells. Mol Carcinog 26:37– 43 76. Liberati NT, Datto MB, Frederick JP, Shen X, Wong C, Rougier-
57. Hocevar BA, Brown TL, Howe PH 1999 TGF-␤ induces fibronectin Chapman EM, Wang XF 1999 Smads bind directly to the Jun family
synthesis through a c-Jun N-terminal kinase-dependent, Smad4- of AP-1 transcription factors. Proc Natl Acad Sci USA 96:4844 – 4849
independent pathway. EMBO J 18:1345–1356 77. Zhang Y, Feng XH, Derynck R 1998 Smad3 and Smad4 cooperate
58. Yu L, Hebert MC, Zhang YE 2002 TGF-␤ receptor-activated p38 with c-Jun/c-Fos to mediate TGF-␤-induced transcription. Nature
MAP kinase mediates Smad-independent TGF-␤ responses. EMBO 394:909 –913
J 21:3749 –3759 78. Pessah M, Prunier C, Marais J, Ferrand N, Mazars A, Lallemand
59. Itoh S, Thorikay M, Kowanetz M, Moustakas A, Itoh F, Heldin F, Gauthier JM, Atfi A 2001 c-Jun interacts with the corepressor
CH, ten Dijke P 2003 Elucidation of Smad requirement in trans- TG-interacting factor (TGIF) to suppress Smad2 transcriptional
forming growth factor-␤ type I receptor-induced responses. J Biol activity. Proc Natl Acad Sci USA 98:6198 – 6203
Chem 278:3751–3761 79. Pessah M, Marais J, Prunier C, Ferrand N, Lallemand F, Mauviel
60. Mulder KM 2000 Role of Ras and Mapks in TGF␤ signaling. Cy- A, Atfi A 2002 c-Jun associates with the oncoprotein Ski and sup-
tokine Growth Factor Rev 11:23–35 presses Smad2 transcriptional activity. J Biol Chem 277:29094 –
61. Johnson GL, Lapadat R 2002 Mitogen-activated protein kinase 29100
pathways mediated by ERK, JNK, and p38 protein kinases. Science 80. Yang YC, Piek E, Zavadil J, Liang D, Xie D, Heyer J, Pavlidis P,
298:1911–1912 Kucherlapati R, Roberts AB, Bottinger EP 2003 Hierarchical model
62. Yamamura Y, Hua X, Bergelson S, Lodish HF 2000 Critical role of of gene regulation by transforming growth factor ␤. Proc Natl Acad
Smads and AP-1 complex in transforming growth factor-␤-depen- Sci USA 100:10269 –10274
dent apoptosis. J Biol Chem 275:36295–36302 81. Wada T, Penninger JM 2004 Mitogen-activated protein kinases in
63. Sano Y, Harada J, Tashiro S, Gotoh-Mandeville R, Maekawa T, apoptosis regulation. Oncogene 23:2838 –2849
Ishii S 1999 ATF-2 is a common nuclear target of Smad and TAK1 82. Jochum W, Passegue E, Wagner EF 2001 AP-1 in mouse develop-
pathways in transforming growth factor-␤ signaling. J Biol Chem ment and tumorigenesis. Oncogene 20:2401–2412
274:8949 – 8957 83. Wagner EF 2002 Functions of AP1 (Fos/Jun) in bone development.
64. Edlund S, Landstrom M, Heldin CH, Aspenstrom P 2002 Trans- Ann Rheum Dis 61(Suppl 2):ii40-ii42
forming growth factor-␤-induced mobilization of actin cytoskele- 84. Zhang L, Wang W, Hayashi Y, Jester JV, Birk DE, Gao M, Liu CY,
Kao WW, Karin M, Xia Y 2003 A role for MEK kinase 1 in TGF- Cbfa1: a transcriptional activator of osteoblast differentiation. Cell
␤/activin-induced epithelium movement and embryonic eyelid 89:747–754
closure. EMBO J 22:4443– 4454 102. Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer
85. Wang L, Kwak JH, Kim SI, He Y, Choi ME 2004 Transforming RR, de Crombrugghe B 2002 The novel zinc finger-containing
growth factor-␤1 stimulates vascular endothelial growth factor 164 transcription factor osterix is required for osteoblast differentiation
via mitogen-activated protein kinase kinase 3-p38␣ and p38␦ mi- and bone formation. Cell 108:17–29
togen-activated protein kinase-dependent pathway in murine mes- 103. Karsenty G, Wagner EF 2002 Reaching a genetic and molecular
angial cells. J Biol Chem 279:33213–33219 understanding of skeletal development. Dev Cell 2:389 – 406
86. Wang W, Zhou G, Hu MC, Yao Z, Tan TH 1997 Activation of the 104. Ducy P, Schinke T, Karsenty G 2000 The osteoblast: a sophisticated
hematopoietic progenitor kinase-1 (HPK1)-dependent, stress-acti- fibroblast under central surveillance. Science 289:1501–1504
vated c-Jun N-terminal kinase (JNK) pathway by transforming 105. Harada S, Rodan GA 2003 Control of osteoblast function and
growth factor ␤ (TGF-␤)-activated kinase (TAK1), a kinase medi- regulation of bone mass. Nature 423:349 –355
ator of TGF ␤ signal transduction. J Biol Chem 272:22771–22775 106. Teitelbaum SL, Ross FP 2003 Genetic regulation of osteoclast de-
87. Yamaguchi K, Nagai S, Ninomiya-Tsuji J, Nishita M, Tamai K, velopment and function. Nat Rev Genet 4:638 – 649

Irie K, Ueno N, Nishida E, Shibuya H, Matsumoto K 1999 XIAP, 107. Luchin A, Purdom G, Murphy K, Clark MY, Angel N, Cassady AI,
a cellular member of the inhibitor of apoptosis protein family, links Hume DA, Ostrowski MC 2000 The microphthalmia transcription
the receptors to TAB1-TAK1 in the BMP signaling pathway. EMBO factor regulates expression of the tartrate-resistant acid phospha-
J 18:179 –187 tase gene during terminal differentiation of osteoclasts. J Bone
88. Birkey Reffey S, Wurthner JU, Parks WT, Roberts AB, Duckett CS Miner Res 15:451– 460
2001 X-linked inhibitor of apoptosis protein functions as a cofactor 108. Luchin A, Suchting S, Merson T, Rosol TJ, Hume DA, Cassady
in transforming growth factor-␤ signaling. J Biol Chem 276:26542– AI, Ostrowski MC 2001 Genetic and physical interactions between
26549 Microphthalmia transcription factor and PU. 1 are necessary for
89. Komatsu Y, Shibuya H, Takeda N, Ninomiya-Tsuji J, Yasui T, osteoclast gene expression and differentiation. J Biol Chem 276:
Miyado K, Sekimoto T, Ueno N, Matsumoto K, Yamada G 2002 36703–36710
Targeted disruption of the Tab1 gene causes embryonic lethality 109. Takayanagi H, Kim S, Koga T, Nishina H, Isshiki M, Yoshida H,
and defects in cardiovascular and lung morphogenesis. Mech Dev Saiura A, Isobe M, Yokochi T, Inoue J, Wagner EF, Mak TW,
119:239 –249 Kodama T, Taniguchi T 2002 Induction and activation of the tran-
90. Pilkington MF, Sims SM, Dixon SJ 2001 Transforming growth scription factor NFATc1 (NFAT2) integrate RANKL signaling in
factor-␤ induces osteoclast ruffling and chemotaxis: potential role terminal differentiation of osteoclasts. Dev Cell 3:889 –901
in osteoclast recruitment. J Bone Miner Res 16:1237–1247 110. Kaifu T, Nakahara J, Inui M, Mishima K, Momiyama T, Kaji M,
91. Karsdal MA, Hjorth P, Henriksen K, Kirkegaard T, Nielsen KL, Sugahara A, Koito H, Ujike-Asai A, Nakamura A, Kanazawa K,
Lou H, Delaisse JM, Foged NT 2003 Transforming growth factor-␤ Tan-Takeuchi K, Iwasaki K, Yokoyama WM, Kudo A, Fujiwara
controls human osteoclastogenesis through the p38 MAPK and M, Asou H, Takai T 2003 Osteopetrosis and thalamic hypomyeli-
nosis with synaptic degeneration in DAP12-deficient mice. J Clin
regulation of RANK expression. J Biol Chem 278:44975– 44987
Invest 111:323–332
92. Karsdal MA, Fjording MS, Foged NT, Delaisse JM, Lochter A
111. Wagner EF, Karsenty G 2001 Genetic control of skeletal develop-
2001 Transforming growth factor-␤-induced osteoblast elongation
ment. Curr Opin Genet Dev 11:527–532
regulates osteoclastic bone resorption through a p38 mitogen-ac-
112. Boyle WJ, Simonet WS, Lacey DL 2003 Osteoclast differentiation
tivated protein kinase- and matrix metalloproteinase-dependent
and activation. Nature 423:337–342
pathway. J Biol Chem 276:39350 –39358
113. Hering S, Isken E, Knabbe C, Janott J, Jost C, Pommer A, Muhr
93. Karsdal MA, Larsen L, Engsig MT, Lou H, Ferreras M, Lochter A,
G, Schatz H, Pfeiffer AF 2001 TGF␤1 and TGF␤2 mRNA and
Delaisse JM, Foged NT 2002 Matrix metalloproteinase-dependent protein expression in human bone samples. Exp Clin Endocrinol
activation of latent transforming growth factor-␤ controls the con- Diabetes 109:217–226
version of osteoblasts into osteocytes by blocking osteoblast apo- 114. Pelton RW, Saxena B, Jones M, Moses HL, Gold LI 1991 Immu-
ptosis. J Biol Chem 277:44061– 44067 nohistochemical localization of TGF␤1, TGF ␤2, and TGF␤3 in the
94. Lee KS, Hong SH, Bae SC 2002 Both the Smad and p38 MAPK mouse embryo: expression patterns suggest multiple roles during
pathways play a crucial role in Runx2 expression following induc- embryonic development. J Cell Biol 115:1091–1105
tion by transforming growth factor-␤ and bone morphogenetic 115. Thompson NL, Flanders KC, Smith JM, Ellingsworth LR, Roberts
protein. Oncogene 21:7156 –7163 AB, Sporn MB 1989 Expression of transforming growth factor-␤ 1
95. Palcy S, Bolivar I, Goltzman D 2000 Role of activator protein 1 in specific cells and tissues of adult and neonatal mice. J Cell Biol
transcriptional activity in the regulation of gene expression by 108:661– 669
transforming growth factor ␤1 and bone morphogenetic protein 2 116. Horner A, Kemp P, Summers C, Bord S, Bishop NJ, Kelsall AW,
in ROS 17/2.8 osteoblast-like cells. J Bone Miner Res 15:2352–2361 Coleman N, Compston JE 1998 Expression and distribution of
96. Palcy S, Goltzman D 1999 Protein kinase signalling pathways transforming growth factor-␤ isoforms and their signaling recep-
involved in the up-regulation of the rat ␣1(I) collagen gene by tors in growing human bone. Bone 23:95–102
transforming growth factor ␤1 and bone morphogenetic protein 2 117. Sandberg M, Vuorio T, Hirvonen H, Alitalo K, Vuorio E 1988
in osteoblastic cells. Biochem J 343:21–27 Enhanced expression of TGF-␤ and c-fos mRNAs in the growth
97. Selvamurugan N, Kwok S, Alliston T, Reiss M, Partridge NC 2004 plates of developing human long bones. Development 102:461– 470
Transforming growth factor-␤ 1 regulation of collagenase-3 ex- 118. Cho TJ, Gerstenfeld LC, Einhorn TA 2002 Differential temporal
pression in osteoblastic cells by cross-talk between the Smad and expression of members of the transforming growth factor ␤ su-
MAPK signaling pathways and their components, Smad2 and perfamily during murine fracture healing. J Bone Miner Res 17:
Runx2. J Biol Chem 279:19327–19334 513–520
98. Lai CF, Cheng SL 2002 Signal transductions induced by bone 119. Bonewald LF, Wakefield L, Oreffo RO, Escobedo A, Twardzik
morphogenetic protein-2 and transforming growth factor-␤ in nor- DR, Mundy GR 1991 Latent forms of transforming growth factor-␤
mal human osteoblastic cells. J Biol Chem 277:15514 –15522 (TGF ␤) derived from bone cultures: identification of a naturally
99. Sowa H, Kaji H, Yamaguchi T, Sugimoto T, Chihara K 2002 occurring 100-kDa complex with similarity to recombinant latent
Activations of ERK1/2 and JNK by transforming growth factor ␤ TGF ␤. Mol Endocrinol 5:741–751
negatively regulate Smad3-induced alkaline phosphatase activity 120. Pedrozo HA, Schwartz Z, Robinson M, Gomes R, Dean DD,
and mineralization in mouse osteoblastic cells. J Biol Chem 277: Bonewald LF, Boyan BD 1999 Potential mechanisms for the plas-
36024 –36031 min-mediated release and activation of latent transforming growth
100. Derynck R, Zhang YE 2003 Smad-dependent and Smad-indepen- factor-␤1 from the extracellular matrix of growth plate chondro-
dent pathways in TGF-␤ family signalling. Nature 425:577–584 cytes. Endocrinology 140:5806 –5816
101. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G 1997 Osf2/ 121. Olofsson A, Miyazono K, Kanzaki T, Colosetti P, Engstrom U,
Heldin CH 1992 Transforming growth factor-␤1, -␤2, and -␤3 se- 141. Wrana JL, Maeno M, Hawrylyshyn B, Yao KL, Domenicucci C,
creted by a human glioblastoma cell line. Identification of small and Sodek J 1988 Differential effects of transforming growth factor-␤ on
different forms of large latent complexes. J Biol Chem 267:19482– the synthesis of extracellular matrix proteins by normal fetal rat
19488 calvarial bone cell populations. J Cell Biol 106:915–924
122. Wakefield LM, Letterio JJ, Chen T, Danielpour D, Allison RS, Pai 142. Harris SE, Bonewald LF, Harris MA, Sabatini M, Dallas S, Feng
LH, Denicoff AM, Noone MH, Cowan KH, O’Shaughnessy JA, JQ, Ghosh-Choudhury N, Wozney J, Mundy GR 1994 Effects of
Sporn MB 1995 Transforming growth factor-␤1 circulates in nor- transforming growth factor ␤ on bone nodule formation and ex-
mal human plasma and is unchanged in advanced metastatic breast pression of bone morphogenetic protein 2, osteocalcin, osteopontin,
cancer. Clin Cancer Res 1:129 –136 alkaline phosphatase, and type I collagen mRNA in long-term
123. Alliston T, Choy L, Ducy P, Karsenty G, Derynck R 2001 TGF- cultures of fetal rat calvarial osteoblasts. J Bone Miner Res 9:855–
␤-induced repression of CBFA1 by Smad3 decreases cbfa1 and 863
osteocalcin expression and inhibits osteoblast differentiation. 143. Breen EC, Ignotz RA, McCabe L, Stein JL, Stein GS, Lian JB 1994
EMBO J 20:2254 –2272 TGF ␤ alters growth and differentiation related gene expression in
124. Maeda S, Hayashi M, Komiya S, Imamura T, Miyazono K 2004 proliferating osteoblasts in vitro, preventing development of the
Endogenous TGF-␤ signaling suppresses maturation of osteoblas-

mature bone phenotype. J Cell Physiol 160:323–335
tic mesenchymal cells. EMBO J 23:552–563 144. Noda M, Yoon K, Prince CW, Butler WT, Rodan GA 1988 Tran-
125. Canalis E, Economides AN, Gazzerro E 2003 Bone morphogenetic scriptional regulation of osteopontin production in rat osteosar-
proteins, their antagonists, and the skeleton. Endocr Rev 24:218 – coma cells by type ␤ transforming growth factor. J Biol Chem
235 263:13916 –13921
126. Karsdal MA, Andersen TA, Bonewald L, Christiansen C 2004 145. Rosen DM, Stempien SA, Thompson AY, Seyedin SM 1988
Matrix metalloproteinases (MMPs) safeguard osteoblasts from ap- Transforming growth factor-␤ modulates the expression of osteo-
optosis during transdifferentiation into osteocytes: MT1-MMP blast and chondroblast phenotypes in vitro. J Cell Physiol 134:337–
maintains osteocyte viability. DNA Cell Biol 23:155–165 346
127. Jilka RL, Weinstein RS, Bellido T, Parfitt AM, Manolagas SC 1998 146. Noda M 1989 Transcriptional regulation of osteocalcin production
Osteoblast programmed cell death (apoptosis): modulation by by transforming growth factor-␤ in rat osteoblast-like cells. Endo-
growth factors and cytokines. J Bone Miner Res 13:793– 802 crinology 124:612– 617
128. Katagiri T, Yamaguchi A, Komaki M, Abe E, Takahashi N, Ikeda 147. Harris SE, Sabatini M, Harris MA, Feng JQ, Wozney J, Mundy GR
T, Rosen V, Wozney JM, Fujisawa-Sehara A, Suda T 1994 Bone 1994 Expression of bone morphogenetic protein messenger RNA in
morphogenetic protein-2 converts the differentiation pathway of prolonged cultures of fetal rat calvarial cells. J Bone Miner Res
C2C12 myoblasts into the osteoblast lineage. J Cell Biol 127:1755– 9:389 –394
1766 148. Sowa H, Kaji H, Yamaguchi T, Sugimoto T, Chihara K 2002
129. Lucas PA 1989 Chemotactic response of osteoblast-like cells to Smad3 promotes alkaline phosphatase activity and mineralization
transforming growth factor ␤. Bone 10:459 – 463 of osteoblastic MC3T3–E1 cells. J Bone Miner Res 17:1190 –1199
130. Pfeilschifter J, Wolf O, Naumann A, Minne HW, Mundy GR, 149. Stein GS, Lian JB, van Wijnen AJ, Stein JL, Montecino M, Javed
Ziegler R 1990 Chemotactic response of osteoblastlike cells to trans- A, Zaidi SK, Young DW, Choi JY, Pockwinse SM 2004 Runx2
forming growth factor ␤. J Bone Miner Res 5:825– 830 control of organization, assembly and activity of the regulatory
131. Hughes FJ, Aubin JE, Heersche JN 1992 Differential chemotactic machinery for skeletal gene expression. Oncogene 23:4315– 4329
responses of different populations of fetal rat calvaria cells to plate- 150. Lee MH, Javed A, Kim HJ, Shin HI, Gutierrez S, Choi JY, Rosen
let-derived growth factor and transforming growth factor ␤. Bone V, Stein JL, van Wijnen AJ, Stein GS, Lian JB, Ryoo HM 1999
Miner 19:63–74 Transient upregulation of CBFA1 in response to bone morphoge-
132. Centrella M, Massague J, Canalis E 1986 Human platelet-derived netic protein-2 and transforming growth factor ␤1 in C2C12 myo-
transforming growth factor-␤ stimulates parameters of bone genic cells coincides with suppression of the myogenic phenotype
growth in fetal rat calvariae. Endocrinology 119:2306 –2312 but is not sufficient for osteoblast differentiation. J Cell Biochem
133. Robey PG, Young MF, Flanders KC, Roche NS, Kondaiah P, 73:114 –125
Reddi AH, Termine JD, Sporn MB, Roberts AB 1987 Osteoblasts 151. Lee KS, Kim HJ, Li QL, Chi XZ, Ueta C, Komori T, Wozney JM,
synthesize and respond to transforming growth factor-type ␤ Kim EG, Choi JY, Ryoo HM, Bae SC 2000 Runx2 is a common
(TGF-␤) in vitro. J Cell Biol 105:457– 463 target of transforming growth factor ␤1 and bone morphogenetic
134. Hock JM, Canalis E, Centrella M 1990 Transforming growth fac- protein 2, and cooperation between Runx2 and Smad5 induces
tor-␤ stimulates bone matrix apposition and bone cell replication osteoblast-specific gene expression in the pluripotent mesenchy-
in cultured fetal rat calvariae. Endocrinology 126:421– 426 mal precursor cell line C2C12. Mol Cell Biol 20:8783– 8792
135. Chen TL, Bates RL 1993 Recombinant human transforming growth 152. Spinella-Jaegle S, Roman-Roman S, Faucheu C, Dunn FW, Kawai
factor ␤ 1 modulates bone remodeling in a mineralizing bone organ S, Gallea S, Stiot V, Blanchet AM, Courtois B, Baron R, Rawadi
culture. J Bone Miner Res 8:423– 434 G 2001 Opposite effects of bone morphogenetic protein-2 and trans-
136. Yamada T, Kamiya N, Harada D, Takagi M 1999 Effects of transforming growth factor-␤1 on osteoblast differentiation. Bone 29:
forming growth factor-␤1 on the gene expression of decorin, bi- 323–330
glycan, and alkaline phosphatase in osteoblast precursor cells and 153. Li J, Tsuji K, Komori T, Miyazono K, Wrana JL, Ito Y, Nifuji A,
more differentiated osteoblast cells. Histochem J 31:687– 694 Noda M 1998 Smad2 overexpression enhances Smad4 gene ex-
137. Kassem M, Kveiborg M, Eriksen EF 2000 Production and action pression and suppresses CBFA1 gene expression in osteoblastic
of transforming growth factor-␤ in human osteoblast cultures: de- osteosarcoma ROS17/2.8 cells and primary rat calvaria cells. J Biol
pendence on cell differentiation and modulation by calcitriol. Eur Chem 273:31009 –31015
J Clin Invest 30:429 – 437 154. Sowa H, Kaji H, Hendy GN, Canaff L, Komori T, Sugimoto T,
138. Noda M, Rodan GA 1987 Type ␤ transforming growth factor (TGF Chihara K 2004 Menin is required for BMP-2-and TGF-␤-regulated
␤) regulation of alkaline phosphatase expression and other phe- osteoblastic differentiation through interaction with Smads and
notype-related mRNAs in osteoblastic rat osteosarcoma cells. J Cell Runx2. J Biol Chem 279:40267– 40275
Physiol 133:426 – 437 155. Centrella M, Casinghino S, Kim J, Pham T, Rosen V, Wozney J,
139. Antosz ME, Bellows CG, Aubin JE 1989 Effects of transforming McCarthy TL 1995 Independent changes in type I and type II
growth factor ␤ and epidermal growth factor on cell proliferation receptors for transforming growth factor ␤ induced by bone mor-
and the formation of bone nodules in isolated fetal rat calvaria cells. phogenetic protein 2 parallel expression of the osteoblast pheno-
J Cell Physiol 140:386 –395 type. Mol Cell Biol 15:3273–3281
140. Centrella M, McCarthy TL, Canalis E 1987 Transforming growth 156. Takeuchi Y, Nakayama K, Matsumoto T 1996 Differentiation and
factor ␤ is a bifunctional regulator of replication and collagen cell surface expression of transforming growth factor-␤ receptors
synthesis in osteoblast-enriched cell cultures from fetal rat bone. are regulated by interaction with matrix collagen in murine osteo-
J Biol Chem 262:2869 –2874 blastic cells. J Biol Chem 271:3938 –3944
157. Centrella M, Ji C, Casinghino S, McCarthy TL 1996 Rapid flux in forming growth factors stimulate prostaglandin production and
transforming growth factor-␤ receptors on bone cells. J Biol Chem bone resorption in cultured mouse calvaria. Proc Natl Acad Sci
271:18616 –18622 USA 82:4535– 4538
158. Gebken J, Feydt A, Brinckmann J, Notbohm H, Muller PK, Batge 177. Pfeilschifter J, Seyedin SM, Mundy GR 1988 Transforming
B 1999 Ligand-induced downregulation of receptors for TGF-␤ in growth factor ␤ inhibits bone resorption in fetal rat long bone
human osteoblast-like cells from adult donors. J Endocrinol 161: cultures. J Clin Invest 82:680 – 685
503–510 178. Lerner UH 1996 Transforming growth factor-␤ stimulates bone
159. Chénu C, Pfeilschifter J, Mundy GR, Roodman GD 1988 Trans- resorption in neonatal mouse calvariae by a prostaglandin-unre-
forming growth factor ␤ inhibits formation of osteoclast-like cells lated but cell proliferation-dependent pathway. J Bone Miner Res
in long-term human marrow cultures. Proc Natl Acad Sci USA 11:1628 –1639
85:5683–5687 179. Baltzer AW, Lattermann C, Whalen JD, Wooley P, Weiss K,
160. Shinar DM, Rodan GA 1990 Biphasic effects of transforming Grimm M, Ghivizzani SC, Robbins PD, Evans CH 2000 Genetic
growth factor-␤ on the production of osteoclast-like cells in mouse enhancement of fracture repair: healing of an experimental seg-
bone marrow cultures: the role of prostaglandins in the generation mental defect by adenoviral transfer of the BMP-2 gene. Gene Ther

of these cells. Endocrinology 126:3153–3158 7:734 –739
161. Yamaguchi M, Kishi S 1995 Differential effects of transforming 180. Park J, Ries J, Gelse K, Kloss F, von der Mark K, Wiltfang J,
growth factor-␤ on osteoclast-like cell formation in mouse marrow Neukam FW, Schneider H 2003 Bone regeneration in critical size
culture: relation to the effect of zinc-chelating dipeptides. Peptides defects by cell-mediated BMP-2 gene transfer: a comparison of
16:1483–1488 adenoviral vectors and liposomes. Gene Ther 10:1089 –1098
162. Hattersley G, Chambers TJ 1991 Effects of transforming growth 181. Khan SN, Lane JM 2004 The use of recombinant human bone
factor ␤ 1 on the regulation of osteoclastic development and func- morphogenetic protein-2 (rhBMP-2) in orthopaedic applications.
tion. J Bone Miner Res 6:165–172 Expert Opin Biol Ther 4:741–748
163. Kale VP 2004 Differential activation of MAPK signaling pathways 182. Kanatani M, Sugimoto T, Kaji H, Kobayashi T, Nishiyama K,
by TGF-␤1 forms the molecular mechanism behind its dose-de- Fukase M, Kumegawa M, Chihara K 1995 Stimulatory effect of
pendent bidirectional effects on hematopoiesis. Stem Cells Dev bone morphogenetic protein-2 on osteoclast-like cell formation and
13:27–38 bone-resorbing activity. J Bone Miner Res 10:1681–1690
164. Kale VP, Vaidya AA 2004 Molecular mechanisms behind the dose- 183. Itoh K, Udagawa N, Katagiri T, Iemura S, Ueno N, Yasuda H,
dependent differential activation of MAPK pathways induced by Higashio K, Quinn JM, Gillespie MT, Martin TJ, Suda T, Taka-
transforming growth factor-␤1 in hematopoietic cells. Stem Cells hashi N 2001 Bone morphogenetic protein 2 stimulates osteoclast
Dev 13:536 –547 differentiation and survival supported by receptor activator of
165. Takai H, Kanematsu M, Yano K, Tsuda E, Higashio K, Ikeda K, nuclear factor-␬B ligand. Endocrinology 142:3656 –3662
Watanabe K, Yamada Y 1998 Transforming growth factor-␤ stim- 184. Zheng MH, Wood DJ, Wysocki S, Papadimitriou JM, Wang EA
ulates the production of osteoprotegerin/osteoclastogenesis inhib- 1994 Recombinant human bone morphogenetic protein-2 enhances
itory factor by bone marrow stromal cells. J Biol Chem 273:27091– expression of interleukin-6 and transforming growth factor-␤ 1
27096 genes in normal human osteoblast-like cells. J Cell Physiol 159:
166. Karst M, Gorny G, Galvin RJ, Oursler MJ 2004 Roles of stromal 76 – 82
cell RANKL, OPG, and M-CSF expression in biphasic TGF-␤ reg- 185. Si X, Jin Y, Yang L, Tipoe GL, White FH 1997 Expression of BMP-2
ulation of osteoclast differentiation. J Cell Physiol 200:99 –106 and TGF-␤ 1 mRNA during healing of the rabbit mandible. Eur
167. Dieudonné SC, Foo P, van Zoelen EJ, Burger EH 1991 Inhibiting J Oral Sci 105:325–330
and stimulating effects of TGF-␤ 1 on osteoclastic bone resorption 186. Lee MH, Kim YJ, Kim HJ, Park HD, Kang AR, Kyung HM, Sung
in fetal mouse bone organ cultures. J Bone Miner Res 6:479 – 487 JH, Wozney JM, Ryoo HM 2003 BMP-2-induced Runx2 expression
168. Massey HM, Scopes J, Horton MA, Flanagan AM 2001 Trans- is mediated by Dlx5, and TGF-␤ 1 opposes the BMP-2-induced
forming growth factor-␤1 (TGF-␤) stimulates the osteoclast-form- osteoblast differentiation by suppression of Dlx5 expression. J Biol
ing potential of peripheral blood hematopoietic precursors in a Chem 278:34387–34394
lymphocyte-rich microenvironment. Bone 28:577–582 187. Hanada K, Solchaga LA, Caplan AI, Hering TM, Goldberg VM,
169. Kaneda T, Nojima T, Nakagawa M, Ogasawara A, Kaneko H, Sato Yoo JU, Johnstone B 2001 BMP-2 induction and TGF-␤ 1 modu-
T, Mano H, Kumegawa M, Hakeda Y 2000 Endogenous produc- lation of rat periosteal cell chondrogenesis. J Cell Biochem 81:284 –
tion of TGF-␤ is essential for osteoclastogenesis induced by a com- 294
bination of receptor activator of NF-␬ B ligand and macrophage- 188. Hayden JM, Mohan S, Baylink DJ 1995 The insulin-like growth
colony-stimulating factor. J Immunol 165:4254 – 4263 factor system and the coupling of formation to resorption. Bone
170. Koseki T, Gao Y, Okahashi N, Murase Y, Tsujisawa T, Sato T, 17(Suppl 2):93S–98S
Yamato K, Nishihara T 2002 Role of TGF-␤ family in osteoclasto- 189. Conover CA 2000 In vitro studies of insulin-like growth factor I and
genesis induced by RANKL. Cell Signal 14:31–36 bone. Growth Horm IGF Res 10(Suppl B):S107–S110
171. Chin SL, Johnson SA, Quinn J, Mirosavljevic D, Price JT, Dudley 190. Kveiborg M, Flyvbjerg A, Eriksen EF, Kassem M 2001 Trans-
AC, Thomas DM 2003 A role for ␣V integrin subunit in TGF-␤- forming growth factor-␤1 stimulates the production of insulin-like
stimulated osteoclastogenesis. Biochem Biophys Res Commun 307: growth factor-I and insulin-like growth factor-binding protein-3 in
1051–1058 human bone marrow stromal osteoblast progenitors. J Endocrinol
172. Fuller K, Lean JM, Bayley KE, Wani MR, Chambers TJ 2000 A role 169:549 –561
for TGF␤(1) in osteoclast differentiation and survival. J Cell Sci 191. Elford PR, Lamberts SW 1990 Contrasting modulation by trans-
113:2445–2453 forming growth factor-␤-1 of insulin-like growth factor-I produc-
173. Sells Galvin RJ, Gatlin CL, Horn JW, Fuson TR 1999 TGF-␤ tion in osteoblasts and chondrocytes. Endocrinology 127:1635–1639
enhances osteoclast differentiation in hematopoietic cell cultures 192. Canalis E, Pash J, Gabbitas B, Rydziel S, Varghese S 1993 Growth
stimulated with RANKL and M-CSF. Biochem Biophys Res Com- factors regulate the synthesis of insulin-like growth factor-I in bone
mun 265:233–239 cell cultures. Endocrinology 133:33–38
174. Yan T, Riggs BL, Boyle WJ, Khosla S 2001 Regulation of oste- 193. Giannobile WV, Whitson SW, Lynch SE 1997 Non-coordinate
oclastogenesis and RANK expression by TGF-␤1. J Cell Biochem control of bone formation displayed by growth factor combinations
83:320 –325 with IGF-I. J Dent Res 76:1569 –1578
175. Fox SW, Haque SJ, Lovibond AC, Chambers TJ 2003 The possible 194. Fukumoto T, Sperling JW, Sanyal A, Fitzsimmons JS, Reinholz
role of TGF-␤-induced suppressors of cytokine signaling expres- GG, Conover CA, O’Driscoll SW 2003 Combined effects of insulin-
sion in osteoclast/macrophage lineage commitment in vitro. J Im- like growth factor-1 and transforming growth factor-␤1 on peri-
munol 170:3679 –3687 osteal mesenchymal cells during chondrogenesis in vitro. Osteo-
176. Tashjian Jr AH, Voelkel EF, Lazzaro M, Singer FR, Roberts AB, arthritis Cartilage 11:55– 64
Derynck R, Winkler ME, Levine L 1985 ␣ And ␤ human trans- 195. Schmidmaier G, Wildemann B, Gabelein T, Heeger J, Kandziora
F, Haas NP, Raschke M 2003 Synergistic effect of IGF-I and TGF-␤1 induces osteoclast formation in the absence of RANKL. Bone 34:
on fracture healing in rats: single versus combined application of 57– 64
IGF-I and TGF-␤1. Acta Orthop Scand 74:604 – 610 217. Ikeda E, Kusaka M, Hakeda Y, Yokota K, Kumegawa M,
196. Canalis E, Centrella M, McCarthy T 1988 Effects of basic fibroblast Yamamoto S 1988 Effect of interleukin 1 ␤ on osteoblastic clone
growth factor on bone formation in vitro. J Clin Invest 81:1572–1577 MC3T3–E1 cells. Calcif Tissue Int 43:162–166
197. Nakamura T, Hara Y, Tagawa M, Tamura M, Yuge T, Fukuda H, 218. Hanazawa S, Ohmori Y, Amano S, Hirose K, Miyoshi T, Ku-
Nigi H 1998 Recombinant human basic fibroblast growth factor megawa M, Kitano S 1986 Human purified interleukin-1 inhibits
accelerates fracture healing by enhancing callus remodeling in ex- DNA synthesis and cell growth of osteoblastic cell line (MC3T3–
perimental dog tibial fracture. J Bone Miner Res 13:942–949 E1), but enhances alkaline phosphatase activity in the cells. FEBS
198. Hurley MM, Lee SK, Raisz LG, Bernecker P, Lorenzo J 1998 Basic Lett 203:279 –284
fibroblast growth factor induces osteoclast formation in murine 219. Rifa L 1999 Bone and cytokines: beyond IL-1, IL-6 and TNF-␣.
bone marrow cultures. Bone 22:309 –316 Calcif Tissue Int 64:1–7
199. Jimi E, Shuto T, Ikebe T, Jingushi S, Hirata M, Koga T 1996 Basic 220. Kwan Tat S, Padrines M, Theoleyre S, Heymann D, Fortun Y 2004
fibroblast growth factor inhibits osteoclast-like cell formation. J Cell IL-6, RANKL, TNF-␣/IL-1: interrelations in bone resorption patho-

Physiol 168:395– 402 physiology. Cytokine Growth Factor Rev 15:49 – 60
200. Zuo J, Jiang J, Dolce C, Holliday LS 2004 Effects of basic fibroblast 221. Pfeilschifter J, Mundy GR 1987 Modulation of type ␤ transforming
growth factor on osteoclasts and osteoclast-like cells. Biochem Bio- growth factor activity in bone cultures by osteotropic hormones.
phys Res Commun 318:162–167 Proc Natl Acad Sci USA 84:2024 –2028
201. Chikazu D, Hakeda Y, Ogata N, Nemoto K, Itabashi A, Takato T, 222. Tsai JA, Rong H, Torring O, Matsushita H, Bucht E 2000 Inter-
Kumegawa M, Nakamura K, Kawaguchi H 2000 Fibroblast leukin-1␤ upregulates PTHrP-mRNA expression and protein pro-
growth factor (FGF)-2 directly stimulates mature osteoclast func- duction and decreases TGF-␤ in normal human osteoblast-like
tion through activation of FGF receptor 1 and p42/p44 MAP kinase. cells. Calcif Tissue Int 66:363–369
J Biol Chem 275:31444 –31450 223. Dubois CM, Ruscetti FW, Palaszynski EW, Falk LA, Oppenheim
202. Hurley MM, Abreu C, Gronowicz G, Kawaguchi H, Lorenzo J JJ, Keller JR 1990 Transforming growth factor ␤ is a potent inhibitor
1994 Expression and regulation of basic fibroblast growth factor of interleukin 1 (IL-1) receptor expression: proposed mechanism of
mRNA levels in mouse osteoblastic MC3T3–E1 cells. J Biol Chem inhibition of IL-1 action. J Exp Med 172:737–744
269:9392–9396 224. Park YG, Kang SK, Kim WJ, Lee YC, Kim CH 2004 Effects of
203. Sobue T, Gravely T, Hand A, Min YK, Pilbeam C, Raisz LG, TGF-␤, TNF-␣, IL-␤ and IL-6 alone or in combination, and tyrosine
Zhang X, Larocca D, Florkiewicz R, Hurley MM 2002 Regulation kinase inhibitor on cyclooxygenase expression, prostaglandin E2
of fibroblast growth factor 2 and fibroblast growth factor receptors production and bone resorption in mouse calvarial bone cells. Int
by transforming growth factor ␤ in human osteoblastic MG-63 cells. J Biochem Cell Biol 36:2270 –2280
J Bone Miner Res 17:502–512 225. Jacobsen FW, Stokke T, Jacobsen SE 1995 Transforming growth
204. Globus RK, Patterson-Buckendahl P, Gospodarowicz D 1988 factor-␤ potently inhibits the viability-promoting activity of stem
Regulation of bovine bone cell proliferation by fibroblast growth cell factor and other cytokines and induces apoptosis of primitive
factor and transforming growth factor ␤. Endocrinology 123:98 – murine hematopoietic progenitor cells. Blood 86:2957–2966
105 226. Hughes FJ, Howells GL 1993 Interleukin-6 inhibits bone formation
205. Hock JM, Canalis E 1994 Platelet-derived growth factor enhances in vitro. Bone Miner 21:21–28
bone cell replication, but not differentiated function of osteoblasts. 227. Taguchi Y, Yamamoto M, Yamate T, Lin SC, Mocharla H,
Endocrinology 134:1423–1428 DeTogni P, Nakayama N, Boyce BF, Abe E, Manolagas SC 1998
206. Mitlak BH, Finkelman RD, Hill EL, Li J, Martin B, Smith T, Interleukin-6-type cytokines stimulate mesenchymal progenitor
D’Andrea M, Antoniades HN, Lynch SE 1996 The effect of sys- differentiation toward the osteoblastic lineage. Proc Assoc Am
temically administered PDGF-BB on the rodent skeleton. J Bone Physicians 110:559 –574
Miner Res 11:238 –247 228. Franchimont N, Rydziel S, Canalis E 2000 Transforming growth
207. Nash TJ, Howlett CR, Martin C, Steele J, Johnson KA, Hicklin DJ factor-␤ increases interleukin-6 transcripts in osteoblasts. Bone 26:
1994 Effect of platelet-derived growth factor on tibial osteotomies 249 –253
in rabbits. Bone 15:203–208 229. Gimble JM, Hudson J, Henthorn J, Hua XX, Burstein SA 1991
208. Zhang Z, Chen J, Jin D 1998 Platelet-derived growth factor Regulation of interleukin 6 expression in murine bone marrow
(PDGF)-BB stimulates osteoclastic bone resorption directly: the role stromal cells. Exp Hematol 19:1055–1060
of receptor ␤. Biochem Biophys Res Commun 251:190 –194 230. Hierl T, Borcsok I, Sommer U, Ziegler R, Kasperk C 1998 Reg-
209. Yeh YL, Kang YM, Chaibi MS, Xie JF, Graves DT 1993 IL-1 and ulation of interleukin-6 expression in human osteoblastic cells in
transforming growth factor-␤ inhibit platelet-derived growth fac- vitro. Exp Clin Endocrinol Diabetes 106:324 –333
tor-AA binding to osteoblastic cells by reducing platelet-derived 231. Hughes FJ, Howells GL 1993 Interleukin-11 inhibits bone forma-
growth factor-␣ receptor expression. J Immunol 150:5625–5632 tion in vitro. Calcif Tissue Int 53:362–364
210. Kells AF, Coats SR, Schwartz HS, Hoover RL 1995 TGF-␤ and 232. Takeuchi Y, Watanabe S, Ishii G, Takeda S, Nakayama K, Fu-
PDGF act synergistically in affecting the growth of human osteo- kumoto S, Kaneta Y, Inoue D, Matsumoto T, Harigaya K, Fujita
blast-enriched cultures. Connect Tissue Res 31:117–124 T 2002 Interleukin-11 as a stimulatory factor for bone formation
211. Rydziel S, Canalis E 1996 Expression and growth factor regulation prevents bone loss with advancing age in mice. J Biol Chem 277:
of platelet-derived growth factor B transcripts in primary osteo- 49011– 49018
blast cell cultures. Endocrinology 137:4115– 4119 233. Girasole G, Passeri G, Jilka RL, Manolagas SC 1994 Interleukin-
212. Gyda M, Corisdeo S, Zaidi M, Troen BR 2001 Macrophage colony- 11: a new cytokine critical for osteoclast development. J Clin Invest
stimulating factor suppresses osteoblast formation. Biochem Bio- 93:1516 –1524
phys Res Commun 285:328 –334 234. Morinaga Y, Fujita N, Ohishi K, Tsuruo T 1997 Stimulation of
213. Teitelbaum SL 2000 Bone resorption by osteoclasts. Science 289: interleukin-11 production from osteoblast-like cells by transform-
1504 –1508 ing growth factor-␤ and tumor cell factors. Int J Cancer 71:422– 428
214. Owens J, Chambers TJ 1993 Macrophage colony-stimulating factor 235. Canalis E 1987 Effects of tumor necrosis factor on bone formation
(M-CSF) induces migration in osteoclasts in vitro. Biochem Biophys in vitro. Endocrinology 121:1596 –1604
Res Commun 195:1401–1407 236. Hill PA, Tumber A, Meikle MC 1997 Multiple extracellular signals
215. Takaishi T, Matsui T, Tsukamoto T, Ito M, Taniguchi T, Fukase promote osteoblast survival and apoptosis. Endocrinology 138:
M, Chihara K 1994 TGF-␤-induced macrophage colony-stimulat- 3849 –3858
ing factor gene expression in various mesenchymal cell lines. Am J 237. Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Ko-
Physiol 267:C25–C31 take S, Nakagawa N, Kinosaki M, Yamaguchi K, Shima N, Ya-
216. Itonaga I, Sabokbar A, Sun SG, Kudo O, Danks L, Ferguson D, suda H, Morinaga T, Higashio K, Martin TJ, Suda T 2000 Tumor
Fujikawa Y, Athanasou NA 2004 Transforming growth factor-␤ necrosis factor ␣ stimulates osteoclast differentiation by a mecha-
nism independent of the ODF/RANKL-RANK interaction. J Exp Shalhoub V, Aronin N, Wright KL, van Wijnen AJ, Stein JL,
Med 191:275–286 Curran T, Lian JB, Stein GS 1990 Coordinate occupancy of AP-1
238. Fox SW, Fuller K, Bayley KE, Lean JM, Chambers TJ 2000 TGF-␤ sites in the vitamin D-responsive and CCAAT box elements by
1 and IFN-␥ direct macrophage activation by TNF-␣ to osteoclastic Fos-Jun in the osteocalcin gene: model for phenotype suppression
or cytocidal phenotype. J Immunol 165:4957– 4963 of transcription. Proc Natl Acad Sci USA 87:9990 –9994
239. Chua CC, Chua BH, Chen Z, Landy C, Hamdy RC 2002 TGF-␤1 257. Kveiborg M, Flyvbjerg A, Kassem M 2002 Synergistic effects of
inhibits multiple caspases induced by TNF-␣ in murine osteoblastic 1,25-dihydroxyvitamin D3 and TGF-␤1 on the production of insu-
MC3T3–E1 cells. Biochim Biophys Acta 1593:1– 8 lin-like growth factor binding protein 3 in human bone marrow
240. Snoeck HW, Weekx S, Moulijn A, Lardon F, Lenjou M, Nys G, stromal cell cultures. APMIS 110:410 – 414
Van Ranst PC, Van Bockstaele DR, Berneman ZN 1996 Tumor 258. Kim CH, Kim YH, Kim YK, Kang BS, Lee TK, Moon SH, Park YG
necrosis factor ␣ is a potent synergistic factor for the proliferation 2003 IL-1␤ induces and TGF-␤ reduces vitamin D3-induced bone
of primitive human hematopoietic progenitor cells and induces resorption in mouse calvarial bone cells. Immunol Invest 32:171–
resistance to transforming growth factor ␤ but not to interferon ␥. 186
J Exp Med 183:705–710 259. Qu Q, Perala-Heape M, Kapanen A, Dahllund J, Salo J, Vaananen

241. Verrecchia F, Tacheau C, Wagner EF, Mauviel A 2003 A central HK, Harkonen P 1998 Estrogen enhances differentiation of osteo-
role for the JNK pathway in mediating the antagonistic activity of blasts in mouse bone marrow culture. Bone 22:201–209
pro-inflammatory cytokines against transforming growth factor- 260. Chow JW, Lean JM, Chambers TJ 1992 17 ␤-Estradiol stimulates
␤-driven SMAD3/4-specific gene expression. J Biol Chem cancellous bone formation in female rats. Endocrinology 130:3025–
278:1585–1593 3032
242. Mann GN, Jacobs TW, Buchinsky FJ, Armstrong EC, Li M, Ke HZ, 261. Hughes DE, Dai A, Tiffee JC, Li HH, Mundy GR, Boyce BF 1996
Ma YF, Jee WS, Epstein S 1994 Interferon-␥ causes loss of bone Estrogen promotes apoptosis of murine osteoclasts mediated by
volume in vivo and fails to ameliorate cyclosporin A-induced os- TGF-␤. Nat Med 2:1132–1136
teopenia. Endocrinology 135:1077–1083 262. Shevde NK, Bendixen AC, Dienger KM, Pike JW 2000 Estrogens
243. Smith DD, Gowen M, Mundy GR 1987 Effects of interferon-␥ and suppress RANK ligand-induced osteoclast differentiation via a
other cytokines on collagen synthesis in fetal rat bone cultures. stromal cell independent mechanism involving c-Jun repression.
Endocrinology 120:2494 –2499 Proc Natl Acad Sci USA 97:7829 –7834
244. Takahashi N, Mundy GR, Roodman GD 1986 Recombinant hu- 263. Oursler MJ, Cortese C, Keeting P, Anderson MA, Bonde SK,
man interferon-␥ inhibits formation of human osteoclast-like cells. Riggs BL, Spelsberg TC 1991 Modulation of transforming growth
J Immunol 137:3544 –3549 factor-␤ production in normal human osteoblast-like cells by 17␤-
245. Vignery A, Niven-Fairchild T, Shepard MH 1990 Recombinant estradiol and parathyroid hormone. Endocrinology 129:3313–3320
murine interferon-␥ inhibits the fusion of mouse alveolar macro- 264. Gao Y, Qian WP, Dark K, Toraldo G, Lin AS, Guldberg RE,
phages in vitro but stimulates the formation of osteoclastlike cells Flavell RA, Weitzmann MN, Pacifici R 2004 Estrogen prevents
on implanted syngeneic bone particles in mice in vivo. J Bone Miner bone loss through transforming growth factor ␤ signaling in T cells.
Res 5:637– 644 Proc Natl Acad Sci USA 101:16618 –16623
246. Heberden C, Denis I, Pointillart A, Mercier T 1998 TGF-␤ and 265. Cooper MS, Hewison M, Stewart PM 1999 Glucocorticoid activity,
calcitriol. Gen Pharmacol 30:145–151 inactivity and the osteoblast. J Endocrinol 163:159 –164
247. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by in- 266. Lukert BP, Raisz LG 1990 Glucocorticoid-induced osteoporosis:
teractions between transforming growth factor-␤ and 1␣,25-dihy- pathogenesis and management. Ann Intern Med 112:352–364
droxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299 –317 267. Canalis E, Delany AM 2002 Mechanisms of glucocorticoid action
248. van Leeuwen JP, van Driel M, van den Bemd GJ, Pols HA 2001 in bone. Ann NY Acad Sci 966:73– 81
Vitamin D control of osteoblast function and bone extracellular 268. Oursler MJ, Riggs BL, Spelsberg TC 1993 Glucocorticoid-induced
matrix mineralization. Crit Rev Eukaryot Gene Expr 11:199 –226 activation of latent transforming growth factor-␤ by normal human
249. van Driel M, Pols HA, van Leeuwen JP 2004 Osteoblast differen- osteoblast-like cells. Endocrinology 133:2187–2196
tiation and control by vitamin D and vitamin D metabolites. Curr 269. Centrella M, McCarthy TL, Canalis E 1991 Glucocorticoid regu-
Pharm Des 10:2535–2555 lation of transforming growth factor ␤ 1 activity and binding in
250. Suda T, Ueno Y, Fujii K, Shinki T 2003 Vitamin D and bone. J Cell osteoblast-enriched cultures from fetal rat bone. Mol Cell Biol 11:
Biochem 88:259 –266 4490 – 4496
251. Finkelman RD, Linkhart TA, Mohan S, Lau KH, Baylink DJ, Bell 270. Chang DJ, Ji C, Kim KK, Casinghino S, McCarthy TL, Centrella
NH 1991 Vitamin D deficiency causes a selective reduction in M 1998 Reduction in transforming growth factor ␤ receptor I ex-
deposition of transforming growth factor ␤ in rat bone: possible pression and transcription factor CBFa1 on bone cells by glucocor-
mechanism for impaired osteoinduction. Proc Natl Acad Sci USA ticoid. J Biol Chem 273:4892– 4896
88:3657–3660 271. Takuma A, Kaneda T, Sato T, Ninomiya S, Kumegawa M, Hakeda
252. Nagel D, Kumar R 2002 1␣,25-Dihydroxyvitamin D3 increases Y 2003 Dexamethasone enhances osteoclast formation synergisti-
TGF␤1 binding to human osteoblasts. Biochem Biophys Res Com- cally with transforming growth factor-␤ by stimulating the priming
mun 290:1558 –1563 of osteoclast progenitors for differentiation into osteoclasts. J Biol
253. Staal A, Birkenhager JC, Pols HA, Buurman CJ, Vink-van Wi- Chem 278:44667– 44674
jngaarden T, Kleinekoort WM, van den Bemd GJ, van Leeuwen 272. Kawaguchi H, Pilbeam CC, Harrison JR, Raisz LG 1995 The role
JP 1994 Transforming growth factor ␤-induced dissociation be- of prostaglandins in the regulation of bone metabolism. Clin Or-
tween vitamin D receptor level and 1,25-dihydroxyvitamin D3 ac- thop Relat Res 313:36 – 46
tion in osteoblast-like cells. Bone Miner 26:27– 42 273. Vrotsos Y, Miller SC, Marks Jr SC 2003 Prostaglandin E—a pow-
254. Subramaniam N, Leong GM, Cock TA, Flanagan JL, Fong C, erful anabolic agent for generalized or site-specific bone formation.
Eisman JA, Kouzmenko AP 2001 Cross-talk between 1,25-dihy- Crit Rev Eukaryot Gene Expr 13:255–263
droxyvitamin D3 and transforming growth factor-␤ signaling re- 274. Klein-Nulend J, Semeins CM, Burger EH 1996 Prostaglandin me-
quires binding of VDR and Smad3 proteins to their cognate DNA diated modulation of transforming growth factor-␤ metabolism in
recognition elements. J Biol Chem 276:15741–15746 primary mouse osteoblastic cells in vitro. J Cell Physiol 168:1–7
255. Staal A, Van Wijnen AJ, Desai RK, Pols HA, Birkenhager JC, 275. Ghayor C, Rey A, Caverzasio J 2005 Prostaglandin-dependent
Deluca HF, Denhardt DT, Stein JL, Van Leeuwen JP, Stein GS, activation of ERK mediates cell proliferation induced by trans-
Lian JB 1996 Antagonistic effects of transforming growth factor-␤ forming growth factor ␤ in mouse osteoblastic cells. Bone 36:93–100
on vitamin D3 enhancement of osteocalcin and osteopontin tran- 276. Qin L, Raggatt LJ, Partridge NC 2004 Parathyroid hormone: a
scription: reduced interactions of vitamin D receptor/retinoid X double-edged sword for bone metabolism. Trends Endocrinol
receptor complexes with vitamin E response elements. Endocri- Metab 15:60 – 65
nology 137:2001–2011 277. Sowa H, Kaji H, Iu MF, Tsukamoto T, Sugimoto T, Chihara K
256. Owen TA, Bortell R, Yocum SA, Smock SL, Zhang M, Abate C, 2003 Parathyroid hormone-Smad3 axis exerts anti-apoptotic action
and augments anabolic action of transforming growth factor ␤ in deficiency-associated bone loss. Proc Natl Acad Sci USA 89:12190 –
osteoblasts. J Biol Chem 278:52240 –52252 12193
278. Pfeilschifter J, Laukhuf F, Muller-Beckmann B, Blum WF, Pfister 299. Ikeda T, Shigeno C, Kasai R, Kohno H, Ohta S, Okumura H,
T, Ziegler R 1995 Parathyroid hormone increases the concentration Konishi J, Yamamuro T 1993 Ovariectomy decreases the mRNA
of insulin-like growth factor-I and transforming growth factor ␤ 1 levels of transforming growth factor-␤ 1 and increases the mRNA
in rat bone. J Clin Invest 96:767–774 levels of osteocalcin in rat bone in vivo. Biochem Biophys Res
279. McCauley LK, Koh AJ, Beecher CA, Cui Y, Decker JD, Franceschi Commun 194:1228 –1233
RT 1995 Effects of differentiation and transforming growth factor 300. Bord S, Beavan S, Ireland D, Horner A, Compston JE 2001 Mech-
␤ 1 on PTH/PTHrP receptor mRNA levels in MC3T3–E1 cells. anisms by which high-dose estrogen therapy produces anabolic
J Bone Miner Res 10:1243–1255 skeletal effects in postmenopausal women: role of locally produced
280. Seitz PK, Zhu BT, Cooper CW 1992 Effect of transforming growth growth factors. Bone 29:216 –222
factor ␤ on parathyroid hormone receptor binding and cAMP for- 301. Matsuda T, Yamamoto T, Muraguchi A, Saatcioglu F 2001 Cross-
mation in rat osteosarcoma cells. J Bone Miner Res 7:541–546 talk between transforming growth factor-␤ and estrogen receptor
281. Kasperk CH, Wergedal JE, Farley JR, Linkhart TA, Turner RT, signaling through Smad3. J Biol Chem 276:42908 – 42914

Baylink DJ 1989 Androgens directly stimulate proliferation of bone 302. Wu L, Wu Y, Gathings B, Wan M, Li X, Grizzle W, Liu Z, Lu C,
cells in vitro. Endocrinology 124:1576 –1578 Mao Z, Cao X 2003 Smad4 as a transcription corepressor for es-
282. Benz DJ, Haussler MR, Thomas MA, Speelman B, Komm BS 1991 trogen receptor ␣. J Biol Chem 278:15192–15200
High-affinity androgen binding and androgenic regulation of ␣ 303. Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin M,
1(I)-procollagen and transforming growth factor-␤ steady state Allen R, Sidman C, Proetzel G, Calvin D, Annunziata N,
messenger ribonucleic acid levels in human osteoblast-like osteo- Doetschman T 1992 Targeted disruption of the mouse transform-
sarcoma cells. Endocrinology 128:2723–2730 ing growth factor-␤ 1 gene results in multifocal inflammatory dis-
283. Hofbauer LC, Hicok KC, Khosla S 1998 Effects of gonadal and ease. Nature 359:693– 699
adrenal androgens in a novel androgen-responsive human osteo- 304. Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, Flanders
blastic cell line. J Cell Biochem 71:96 –108 KC, Roberts AB, Sporn MB, Ward JM, Karlsson S 1993 Trans-
284. Vanderschueren D, Vandenput L, Boonen S, Lindberg MK, forming growth factor ␤ 1 null mutation in mice causes excessive
Bouillon R, Ohlsson C 2004 Androgens and bone. Endocr Rev inflammatory response and early death. Proc Natl Acad Sci USA
25:389 – 425 90:770 –774
285. Pederson L, Kremer M, Judd J, Pascoe D, Spelsberg TC, Riggs BL, 305. Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson S,
Oursler MJ 1999 Androgens regulate bone resorption activity of Akhurst RJ 1995 Defective haematopoiesis and vasculogenesis in
isolated osteoclasts in vitro. Proc Natl Acad Sci USA 96:505–510 transforming growth factor-␤ 1 knock out mice. Development 121:
286. Kasperk C, Fitzsimmons R, Strong D, Mohan S, Jennings J, 1845–1854
Wergedal J, Baylink D 1990 Studies of the mechanism by which 306. Geiser AG, Zeng QQ, Sato M, Helvering LM, Hirano T, Turner
androgens enhance mitogenesis and differentiation in bone cells. CH 1998 Decreased bone mass and bone elasticity in mice lacking
J Clin Endocrinol Metab 71:1322–1329 the transforming growth factor-␤1 gene. Bone 23:87–93
287. Wu Y, Craig TA, Lutz WH, Kumar R 1999 Identification of 1 307. Sanford LP, Ormsby I, Gittenberger-de Groot AC, Sariola H,
␣,25-dihydroxyvitamin D3 response elements in the human trans- Friedman R, Boivin GP, Cardell EL, Doetschman T 1997 TGF␤2
forming growth factor ␤ 2 gene. Biochemistry 38:2654 –2660 knockout mice have multiple developmental defects that are non-
288. Takeshita A, Imai K, Kato S, Kitano S, Hanazawa S 1998 1␣,25- overlapping with other TGF␤ knockout phenotypes. Development
Dihydroxyvitamin D3 synergism toward transforming growth fac- 124:2659 –2670
tor-␤1-induced AP-1 transcriptional activity in mouse osteoblastic 308. Proetzel G, Pawlowski SA, Wiles MV, Yin M, Boivin GP, Howles
cells via its nuclear receptor. J Biol Chem 273:14738 –14744 PN, Ding J, Ferguson MW, Doetschman T 1995 Transforming
289. Yanagisawa J, Yanagi Y, Masuhiro Y, Suzawa M, Watanabe M, growth factor-␤ 3 is required for secondary palate fusion. Nat Genet
Kashiwagi K, Toriyabe T, Kawabata M, Miyazono K, Kato S 1999 11:409 – 414
Convergence of transforming growth factor-␤ and vitamin D sig- 309. Kaartinen V, Voncken JW, Shuler C, Warburton D, Bu D, Heis-
naling pathways on SMAD transcriptional coactivators. Science terkamp N, Groffen J 1995 Abnormal lung development and cleft
283:1317–1321 palate in mice lacking TGF-␤ 3 indicates defects of epithelial-mes-
290. Nanes MS 2003 Tumor necrosis factor-␣: molecular and cellular enchymal interaction. Nat Genet 11:415– 421
mechanisms in skeletal pathology. Gene 321:1–15 310. Dabovic B, Chen Y, Colarossi C, Obata H, Zambuto L, Perle MA,
291. Lovibond AC, Haque SJ, Chambers TJ, Fox SW 2003 TGF-␤- Rifkin DB 2002 Bone abnormalities in latent TGF-[beta] binding
induced SOCS3 expression augments TNF-␣-induced osteoclast protein (Ltbp)-3-null mice indicate a role for Ltbp-3 in modulating
formation. Biochem Biophys Res Commun 309:762–767 TGF-[beta] bioavailability. J Cell Biol 156:227–232
292. Verrecchia F, Pessah M, Atfi A, Mauviel A 2000 Tumor necrosis 311. Sterner-Kock A, Thorey IS, Koli K, Wempe F, Otte J, Bangsow T,
factor-␣ inhibits transforming growth factor-␤ /Smad signaling in Kuhlmeier K, Kirchner T, Jin S, Keski-Oja J, von Melchner H 2002
human dermal fibroblasts via AP-1 activation. J Biol Chem 275: Disruption of the gene encoding the latent transforming growth
30226 –30231 factor-␤ binding protein 4 (LTBP-4) causes abnormal lung devel-
293. Bitzer M, von Gersdorff G, Liang D, Dominguez-Rosales A, Beg opment, cardiomyopathy, and colorectal cancer. Genes Dev 16:
AA, Rojkind M, Bottinger EP 2000 A mechanism of suppression 2264 –2273
of TGF-␤/SMAD signaling by NF-␬ B/RelA. Genes Dev 14:187–197 312. Larsson J, Goumans MJ, Sjostrand LJ, van Rooijen MA, Ward D,
294. Nagarajan RP, Chen F, Li W, Vig E, Harrington MA, Nakshatri Leveen P, Xu X, ten Dijke P, Mummery CL, Karlsson S 2001
H, Chen Y 2000 Repression of transforming-growth-factor-␤-me- Abnormal angiogenesis but intact hematopoietic potential in
diated transcription by nuclear factor ␬B. Biochem J 348:591–596 TGF-␤ type I receptor-deficient mice. EMBO J 20:1663–1673
295. Oursler MJ 1998 Estrogen regulation of gene expression in osteo- 313. Oshima M, Oshima H, Taketo MM 1996 TGF-␤ receptor type II
blasts and osteoclasts. Crit Rev Eukaryot Gene Expr 8:125–140 deficiency results in defects of yolk sac hematopoiesis and vascu-
296. Robinson JA, Harris SA, Riggs BL, Spelsberg TC 1997 Estrogen logenesis. Dev Biol 179:297–302
regulation of human osteoblastic cell proliferation and differenti- 314. Nomura M, Li E 1998 Smad2 role in mesoderm formation, left-right
ation. Endocrinology 138:2919 –2927 patterning and craniofacial development. Nature 393:786 –790
297. Qu Q, Harkonen PL, Monkkonen J, Vaananen HK 1999 Condi- 315. Weinstein M, Yang X, Li C, Xu X, Gotay J, Deng CX 1998 Failure
tioned medium of estrogen-treated osteoblasts inhibits osteoclast of egg cylinder elongation and mesoderm induction in mouse
maturation and function in vitro. Bone 25:211–215 embryos lacking the tumor suppressor smad2. Proc Natl Acad Sci
298. Finkelman RD, Bell NH, Strong DD, Demers LM, Baylink DJ USA 95:9378 –9383
1992 Ovariectomy selectively reduces the concentration of trans- 316. Yang X, Letterio JJ, Lechleider RJ, Chen L, Hayman R, Gu H,
forming growth factor ␤ in rat bone: implications for estrogen Roberts AB, Deng C 1999 Targeted disruption of SMAD3 results
in impaired mucosal immunity and diminished T cell responsive- mutations in TGFB1 result in Camurati-Engelmann disease. Nat
ness to TGF-␤. EMBO J 18:1280 –1291 Genet 26:19 –20
317. Borton AJ, Frederick JP, Datto MB, Wang XF, Weinstein RS 2001 335. Hecht JT, Blanton SH, Broussard S, Scott A, Rhoades Hall C,
The loss of Smad3 results in a lower rate of bone formation and Milunsky JM 2001 Evidence for locus heterogeneity in the Camu-
osteopenia through dysregulation of osteoblast differentiation and rati-Engelmann (DPD1) Syndrome. Clin Genet 59:198 –200
apoptosis. J Bone Miner Res 16:1754 –1764 336. Mumm SR, Obrecht S, Podgornik MN, Whyte MP 2001 Camurati-
318. Datto MB, Frederick JP, Pan L, Borton AJ, Zhuang Y, Wang XF Engelmann disease: new mutations in the latency-associated pep-
1999 Targeted disruption of Smad3 reveals an essential role in tide of the transforming growth factor ␤-1 gene. J Bone Miner Res
transforming growth factor ␤-mediated signal transduction. Mol 16(Suppl 1):S223
Cell Biol 19:2495–2504 337. Janssens K, ten Dijke P, Ralston SH, Bergmann C, Van Hul W
319. Zhu Y, Richardson JA, Parada LF, Graff JM 1998 Smad3 mutant 2003 Transforming growth factor-␤ 1 mutations in Camurati-En-
mice develop metastatic colorectal cancer. Cell 94:703–714 gelmann disease lead to increased signaling by altering either ac-
320. Sirard C, de la Pompa JL, Elia A, Itie A, Mirtsos C, Cheung A, tivation or secretion of the mutant protein. J Biol Chem 278:7718 –
Hahn S, Wakeham A, Schwartz L, Kern SE, Rossant J, Mak TW 7724

1998 The tumor suppressor gene Smad4/Dpc4 is required for gas- 338. Kinoshita A, Fukumaki Y, Shirahama S, Miyahara A, Nishimura
trulation and later for anterior development of the mouse embryo. G, Haga N, Namba A, Ueda H, Hayashi H, Ikegawa S, Seidel J,
Genes Dev 12:107–119 Niikawa N, Yoshiura K 2004 TGFB1 mutations in four new families
321. Yang X, Li C, Xu X, Deng C 1998 The tumor suppressor SMAD4/ with Camurati-Engelmann disease: confirmation of independently
DPC4 is essential for epiblast proliferation and mesoderm induc- arising LAP-domain-specific mutations. Am J Med Genet 127A:
tion in mice. Proc Natl Acad Sci USA 95:3667–3672 104 –107
322. Atti E, Gomez S, Wahl SM, Mendelsohn R, Paschalis E, Boskey 339. McGowan NW, MacPherson H, Janssens K, Van Hul W, Frith JC,
AL 2002 Effects of transforming growth factor-␤ deficiency on bone Fraser WD, Ralston SH, Helfrich MH 2003 A mutation affecting
development: a Fourier transform-infrared imaging analysis. Bone the latency-associated peptide of TGF␤1 in Camurati-Engelmann
31:675– 684 disease enhances osteoclast formation in vitro. J Clin Endocrinol
323. Dallas SL, Rosser JL, Mundy GR, Bonewald LF 2002 Proteolysis Metab 88:3321–3326
of latent transforming growth factor-␤ (TGF-␤)-binding protein-1 340. Oklü R, Hesketh R 2000 The latent transforming growth factor ␤
by osteoclasts. A cellular mechanism for release of TGF-␤ from binding protein (LTBP) family. Biochem J 352:601– 610
bone matrix. J Biol Chem 277:21352–21360 341. van der Pluijm G, Lowik C, Papapoulos S 2000 Tumour progres-
324. Baffi MO, Slattery E, Sohn P, Moses HL, Chytil A, Serra R 2004 sion and angiogenesis in bone metastasis from breast cancer: new
Conditional deletion of the TGF-␤ type II receptor in Col2a ex- approaches to an old problem. Cancer Treat Rev 26:11–27
pressing cells results in defects in the axial skeleton without alter- 342. Guise TA, Yin JJ, Taylor SD, Kumagai Y, Dallas M, Boyce BF,
ations in chondrocyte differentiation or embryonic development of Yoneda T, Mundy GR 1996 Evidence for a causal role of parathy-
long bones. Dev Biol 276:124 –142 roid hormone-related protein in the pathogenesis of human breast
cancer-mediated osteolysis. J Clin Invest 98:1544 –1549
325. Felici A, Wurthner JU, Parks WT, Giam LR, Reiss M, Karpova TS,
343. Guise TA 2000 Molecular mechanisms of osteolytic bone metas-
McNally JG, Roberts AB 2003 TLP, a novel modulator of TGF-␤
tases. Cancer 88:2892–2898
signaling, has opposite effects on Smad2- and Smad3-dependent
344. Derynck R, Akhurst RJ, Balmain A 2001 TGF-␤ signaling in tumor
signaling. EMBO J 22:4465– 4477
suppression and cancer progression. Nat Genet 29:117–129
326. Erlebacher A, Derynck R 1996 Increased expression of TGF-␤ 2 in
345. Yin JJ, Selander K, Chirgwin JM, Dallas M, Grubbs BG, Wieser
osteoblasts results in an osteoporosis-like phenotype. J Cell Biol
R, Massague J, Mundy GR, Guise TA 1999 TGF-␤ signaling block-
132:195–210
ade inhibits PTHrP secretion by breast cancer cells and bone me-
327. Filvaroff E, Erlebacher A, Ye J, Gitelman SE, Lotz J, Heillman M, tastases development. J Clin Invest 103:197–206
Derynck R 1999 Inhibition of TGF-␤ receptor signaling in osteo- 346. Guise TA, Chirgwin JM 2003 Transforming growth factor-␤ in
blasts leads to decreased bone remodeling and increased trabecular osteolytic breast cancer bone metastases. Clin Orthop Relat Res
bone mass. Development 126:4267– 4279 (Suppl 415):S32–S38
328. Erlebacher A, Filvaroff EH, Ye JQ, Derynck R 1998 Osteoblastic 347. Mercer RR, Miyasaka C, Mastro AM 2004 Metastatic breast cancer
responses to TGF-␤ during bone remodeling. Mol Biol Cell 9:1903– cells suppress osteoblast adhesion and differentiation. Clin Exp
1918 Metastasis 21:427– 435
329. Rosen D, Miller SC, DeLeon E, Thompson AY, Bentz H, Mathews 348. Mastro AM, Gay CV, Welch DR, Donahue HJ, Jewell J, Mercer
M, Adams S 1994 Systemic administration of recombinant trans- R, DiGirolamo D, Chislock EM, Guttridge K 2004 Breast cancer
forming growth factor ␤ 2 (rTGF-␤ 2) stimulates parameters of cells induce osteoblast apoptosis: a possible contributor to bone
cancellous bone formation in juvenile and adult rats. Bone 15:355– degradation. J Cell Biochem 91:265–276
359 349. Kang Y, Siegel PM, Shu W, Drobnjak M, Kakonen SM, Cordon-
330. Sumner DR, Turner TM, Urban RM, Leven RM, Hawkins M, Cardo C, Guise TA, Massague J 2003 A multigenic program me-
Nichols EH, McPherson JM, Galante JO 2001 Locally delivered diating breast cancer metastasis to bone. Cancer Cell 3:537–549
rhTGF-␤2 enhances bone ingrowth and bone regeneration at local 350. Yamada Y, Miyauchi A, Takagi Y, Tanaka M, Mizuno M, Harada
and remote sites of skeletal injury. J Orthop Res 19:85–94 A 2001 Association of the C-5093 T polymorphism, alone or in
331. Ohta M, Greenberger JS, Anklesaria P, Bassols A, Massague J combination with the T8693 C polymorphism, of the transforming
1987 Two forms of transforming growth factor-␤ distinguished by growth factor-␤1 gene with bone mineral density and genetic sus-
multipotential haematopoietic progenitor cells. Nature 329:539 – ceptibility to osteoporosis in Japanese women. J Mol Med 79:149 –
541 156
332. Sparkes RS, Graham CB 1972 Camurati-Engelmann disease. Ge- 351. Park BL, Han IK, Lee HS, Kim LH, Kim SJ, Shin HD 2003 Iden-
netics and clinical manifestations with a review of the literature. tification of novel variants in transforming growth factor-␤1
J Med Genet 9:73– 85 (TGFB1) gene and association analysis with bone mineral density.
333. Janssens K, Gershoni-Baruch R, Guanabens N, Migone N, Ral- Hum Mutat 22:257–258
ston S, Bonduelle M, Lissens W, Van Maldergem L, Vanhoe- 352. Langdahl BL, Carstens M, Stenkjaer L, Eriksen EF 2003 Polymor-
nacker F, Verbruggen L, Van Hul W 2000 Mutations in the gene phisms in the transforming growth factor ␤ 1 gene and osteopo-
encoding the latency-associated peptide of TGF-␤ 1 cause Camu- rosis. Bone 32:297–310
rati-Engelmann disease. Nat Genet 26:273–275 353. Grainger DJ, Heathcote K, Chiano M, Snieder H, Kemp PR,
334. Kinoshita A, Saito T, Tomita H, Makita Y, Yoshida K, Ghadami Metcalfe JC, Carter ND, Spector TD 1999 Genetic control of the
M, Yamada K, Kondo S, Ikegawa S, Nishimura G, Fukushima Y, circulating concentration of transforming growth factor type ␤1.
Nakagomi T, Saito H, Sugimoto T, Kamegaya M, Hisa K, Murray Hum Mol Genet 8:93–97
JC, Taniguchi N, Niikawa N, Yoshiura K 2000 Domain-specific 354. Wells FA, Reid DM, Ralston SH 2001 Polymorphisms in the trans-
forming growth factor-1 and bone mass and bone loss. Bone 370. Mundy GR 1991 The effects of TGF-␤ on bone. Ciba Found Symp
28(Suppl 1):S131 157:137–143; discussion 143–151
355. Long JR, Liu PY, Lu Y, Dvornyk V, Xiong DH, Zhao LJ, Deng HW 371. Terrell TG, Working PK, Chow CP, Green JD 1993 Pathology of
2004 Tests of linkage and/or association of TGF-␤1 and COL1A1 recombinant human transforming growth factor-␤1 in rats and
genes with bone mass. Osteoporos Int 16:86 –92 rabbits. Int Rev Exp Pathol 34(Pt B):43– 67
356. Lau HH, Ho AY, Luk KD, Kung AW 2004 Transforming growth 372. Ueda H, Hong L, Yamamoto M, Shigeno K, Inoue M, Toba T,
factor-␤1 gene polymorphisms and bone turnover, bone mineral Yoshitani M, Nakamura T, Tabata Y, Shimizu Y 2002 Use of
density and fracture risk in southern Chinese women. Calcif Tissue collagen sponge incorporating transforming growth factor-␤1 to
Int 74:516 –521 promote bone repair in skull defects in rabbits. Biomaterials 23:
357. Yamada Y, Miyauchi A, Goto J, Takagi Y, Okuizumi H, Kane- 1003–1010
matsu M, Hase M, Takai H, Harada A, Ikeda K 1998 Association 373. Beck LS, Amento EP, Xu Y, Deguzman L, Lee WP, Nguyen T,
of a polymorphism of the transforming growth factor-␤1 gene with Gillett NA 1993 TGF-␤ 1 induces bone closure of skull defects:
genetic susceptibility to osteoporosis in postmenopausal Japanese temporal dynamics of bone formation in defects exposed to
women. J Bone Miner Res 13:1569 –1576 rhTGF-␤ 1. J Bone Miner Res 8:753–761

358. Yamada Y, Harada A, Hosoi T, Miyauchi A, Ikeda K, Ohta H, 374. Arnaud E, Morieux C, Wybier M, de Vernejoul MC 1994 Poten-
Shiraki M 2000 Association of transforming growth factor ␤1 ge- tiation of transforming growth factor (TGF-␤ 1) by natural coral and
notype with therapeutic response to active vitamin D for post- fibrin in a rabbit cranioplasty model. Calcif Tissue Int 54:493– 498
menopausal osteoporosis. J Bone Miner Res 15:415– 420 375. Fujimoto R, Tanizawa T, Nishida S, Yamamoto N, Soshi S, Endo
359. Hinke V, Seck T, Clanget C, Scheidt-Nave C, Ziegler R, N, Takahashi HE 1999 Local effects of transforming growth fac-
Pfeilschifter J 2001 Association of transforming growth factor-␤1 tor-␤1 on rat calvaria: changes depending on the dose and the
(TGF␤1) T29 3 C gene polymorphism with bone mineral density injection site. J Bone Miner Metab 17:11–17
(BMD), changes in BMD, and serum concentrations of TGF-␤1 in 376. Tanaka T, Taniguchi Y, Gotoh K, Satoh R, Inazu M, Ozawa H 1993
a population-based sample of postmenopausal German women. Morphological study of recombinant human transforming growth
Calcif Tissue Int 69:315–320 factor ␤ 1-induced intramembranous ossification in neonatal rat
360. Prince RL, Dick IM, Devine A, Dhaliwal SS, Li S, Wilton S 2002 parietal bone. Bone 14:117–123
Effect of the TGF-␤ T869C and the CYP119 TTTA repeat polymor- 377. Nielsen HM, Andreassen TT, Ledet T, Oxlund H 1994 Local
phisms on bone mineral density and prevalent fracture in elderly injection of TGF-␤ increases the strength of tibial fractures in the rat.
women. J Bone Miner Res 17(Suppl 1):S422 Acta Orthop Scand 65:37– 41
361. Dick IM, Devine A, Li S, Dhaliwal SS, Prince RL 2003 The T869C 378. Joyce ME, Terek RM, Jingushi S, Bolander ME 1990 Role of
TGF ␤ polymorphism is associated with fracture, bone mineral transforming growth factor-␤ in fracture repair. Ann NY Acad Sci
density, and calcaneal quantitative ultrasound in elderly women. 593:107–123
Bone 33:335–341 379. Noda M, Camilliere JJ 1989 In vivo stimulation of bone formation
362. Kim S, Kim H, Hong J, Park J, Kim G 2000 T/C polymorphism of by transforming growth factor-␤. Endocrinology 124:2991–2994
the TGF-␤1 gene is not associated with quantitative ultrasound
380. Mackie EJ, Trechsel U 1990 Stimulation of bone formation in vivo
values of calcaneus in Korean postmenopausal women in
by transforming growth factor-␤: remodeling of woven bone and
Chung-Up district. J Bone Miner Res 15(Suppl 1):S362
lack of inhibition by indomethacin. Bone 11:295–300
363. Ziv E, Kahn A, Cauley J, Morin P, Saiz R, Browner W 2003 No
381. Marcelli C, Yates AJ, Mundy GR 1990 In vivo effects of human
association between the TGF-␤1 Leu10Pro polymorphism and os-
recombinant transforming growth factor ␤ on bone turnover in
teoporosis among white women in the United States. Am J Med
normal mice. J Bone Miner Res 5:1087–1096
114:227–231
382. Zhou H, Choong PC, Chou ST, Kartsogiannis V, Martin TJ, Ng
364. Koh JM, Nam-Goong IS, Hong JS, Kim HK, Kim JS, Kim SY, Kim
GS 2004 Oestrogen receptor ␣ genotype, and interactions between KW 1995 Transforming growth factor ␤ 1 stimulates bone forma-
vitamin D receptor and transforming growth factor-␤1 genotypes tion and resorption in an in-vivo model in rabbits. Bone 17(Suppl
are associated with quantitative calcaneal ultrasound in postmeno- 4):443S– 448S
pausal women. Clin Endocrinol (Oxf) 60:232–240 383. Lind M, Schumacker B, Soballe K, Keller J, Melsen F, Bunger C
365. Langdahl BL, Knudsen JY, Jensen HK, Gregersen N, Eriksen EF 1993 Transforming growth factor-␤ enhances fracture healing in
1997 A sequence variation: 713– 8delC in the transforming growth rabbit tibiae. Acta Orthop Scand 64:553–556
factor-␤ 1 gene has higher prevalence in osteoporotic women than 384. Vuola J, Bohling T, Goransson H, Puolakkainen P 2002 Trans-
in normal women and is associated with very low bone mass in forming growth factor ␤ released from natural coral implant en-
osteoporotic women and increased bone turnover in both osteo- hances bone growth at calvarium of mature rat. J Biomed Mater Res
porotic and normal women. Bone 20:289 –294 59:152–159
366. Bertoldo F, D’Agruma L, Furlan F, Colapietro F, Lorenzi MT, 385. Bosch C, Melsen B, Gibbons R, Vargervik K 1996 Human recom-
Maiorano N, Iolascon A, Zelante L, Locascio V, Gasparini P 2000 binant transforming growth factor-␤ 1 in healing of calvarial bone
Transforming growth factor-␤1 gene polymorphism, bone turn- defects. J Craniofac Surg 7:300 –310
over, and bone mass in Italian postmenopausal women. J Bone 386. Shigeno K, Nakamura T, Inoue M, Ueda H, Kobayashi E, Naka-
Miner Res 15:634 – 639 hara T, Lynn AK, Toba T, Yoshitani M, Fukuda S, Kawanami R,
367. Keen RW, Snieder H, Molloy H, Daniels J, Chiano M, Gibson F, Shimizu Y 2002 Regenerative repair of the mandible using a col-
Fairbairn L, Smith P, MacGregor AJ, Gewert D, Spector TD 2001 lagen sponge containing TGF-␤1. Int J Artif Organs 25:1095–1102
Evidence of association and linkage disequilibrium between a 387. Ripamonti U, Bosch C, van den Heever B, Duneas N, Melsen B,
novel polymorphism in the transforming growth factor ␤1 gene Ebner R 1996 Limited chondro-osteogenesis by recombinant hu-
and hip bone mineral density: a study of female twins. Rheuma- man transforming growth factor-␤ 1 in calvarial defects of adult
tology (Oxf) 40:48 –54 baboons (Papio ursinus). J Bone Miner Res 11:938 –945
368. Dunning AM, Ellis PD, McBride S, Kirschenlohr HL, Healey CS, 388. Lind M, Overgaard S, Soballe K, Nguyen T, Ongpipattanakul B,
Kemp PR, Luben RN, Chang-Claude J, Mannermaa A, Kataja V, Bunger C 1996 Transforming growth factor-␤ 1 enhances bone
Pharoah PD, Easton DF, Ponder BA, Metcalfe JC 2003 A trans- healing to unloaded tricalcium phosphate coated implants: an ex-
forming growth factor␤1 signal peptide variant increases secretion perimental study in dogs. J Orthop Res 14:343–350
in vitro and is associated with increased incidence of invasive 389. Lind M, Overgaard S, Glerup H, Soballe K, Bunger C 2001 Trans-
breast cancer. Cancer Res 63:2610 –2615 forming growth factor-␤1 adsorbed to tricalciumphosphate coated
369. Koh J-M, Kim S-W, Chung Y-E, Hong JS, Kim S-Y, Kim GS 2003 implants increases peri-implant bone remodeling. Biomaterials 22:
Lack of association of transforming growth factor ␤ 1 T293 C gene 189 –193
polymorphism with calcaneal quantitative ultrasound parameters 390. Kamakura S, Sasano Y, Nakajo S, Shimizu T, Suzuki O, Katou F,
and plasma TGF-␤1 levels in postmenopausal Korean women. Kagayama M, Motegi K 2001 Implantation of octacalcium phos-
Calcif Tissue Int 72(Suppl 1):S418 phate combined with transforming growth factor-␤1 enhances
bone repair as well as resorption of the implant in rat skull defects. bone with biodegradable microspheres containing transforming
J Biomed Mater Res 57:175–182 growth factor-␤1. Tissue Eng 6:331–340
391. Sumner DR, Turner TM, Purchio AF, Gombotz WR, Urban RM, 401. Aufdemorte TB, Fox WC, Holt GR, McGuff HS, Ammann AJ,
Galante JO 1995 Enhancement of bone ingrowth by transforming Beck LS 1992 An intraosseous device for studies of bone-healing.
growth factor-␤. J Bone Joint Surg Am 77:1135–1147 The effect of transforming growth-factor ␤. J Bone Joint Surg Am
392. McKinney L, Hollinger JO 1996 A bone regeneration study: trans- 74:1153–1161
forming growth factor-␤1 and its delivery. J Craniofac Surg 7:36 – 45 402. Mehrara BJ, Saadeh PB, Steinbrech DS, Dudziak M, Spector JA,
393. Moxham JP, Kibblewhite DJ, Dvorak M, Perey B, Tencer AF, Greenwald JA, Gittes GK, Longaker MT 1999 Adenovirus-medi-
Bruce AG, Strong DM 1996 TGF-␤ 1 forms functionally normal ated gene therapy of osteoblasts in vitro and in vivo. J Bone Miner
bone in a segmental sheep tibial diaphyseal defect. J Otolaryngol Res 14:1290 –1301
25:388 –392 403. Geesink RG, Hoefnagels NH, Bulstra SK 1999 Osteogenic activity
394. Gombotz WR, Pankey SC, Bouchard LS, Phan DH, Puolakkainen of OP-1 bone morphogenetic protein (BMP-7) in a human fibular
PA 1994 Stimulation of bone healing by transforming growth fac- defect. J Bone Joint Surg Br 81:710 –718
tor-␤1 released from polymeric or ceramic implants. J Appl Bio- 404. Friedlaender GE, Perry CR, Cole JD, Cook SD, Cierny G, Mus-

mater 5:141–150 chler GF, Zych GA, Calhoun JH, LaForte AJ, Yin S 2001 Osteo-
395. Schmidmaier G, Wildemann B, Heeger J, Gabelein T, Flyvbjerg genic protein-1 (bone morphogenetic protein-7) in the treatment of
A, Bail HJ, Raschke M 2002 Improvement of fracture healing by tibial nonunions. J Bone Joint Surg Am 83-A (Suppl 1):S151–S158
systemic administration of growth hormone and local application 405. Govender S, Csimma C, Genant HK, Valentin-Opran A, Amit Y,
Arbel R, Aro H, Atar D, Bishay M, Borner MG, Chiron P, Choong
of insulin-like growth factor-1 and transforming growth factor-␤1.
P, Cinats J, Courtenay B, Feibel R, Geulette B, Gravel C, Haas N,
Bone 31:165–172
Raschke M, Hammacher E, van der Velde D, Hardy P, Holt M,
396. Heckman JD, Ehler W, Brooks BP, Aufdemorte TB, Lohmann CH,
Josten C, Ketterl RL, Lindeque B, Lob G, Mathevon H, McCoy G,
Morgan T, Boyan BD 1999 Bone morphogenetic protein but not
Marsh D, Miller R, Munting E, Oevre S, Nordsletten L, Patel A,
transforming growth factor-␤ enhances bone formation in canine Pohl A, Rennie W, Reynders P, Rommens PM, Rondia J, Rossouw
diaphyseal nonunions implanted with a biodegradable composite WC, Daneel PJ, Ruff S, Ruter A, Santavirta S, Schildhauer TA,
polymer. J Bone Joint Surg Am 81:1717–1729 Gekle C, Schnettler R, Segal D, Seiler H, Snowdowne RB, Stapert
397. Aspenberg P, Jeppsson C, Wang JS, Bostrom M 1996 Transform- J, Taglang G, Verdonk R, Vogels L, Weckbach A, Wentzensen A,
ing growth factor ␤ and bone morphogenetic protein 2 for bone Wisniewski T 2002 Recombinant human bone morphogenetic pro-
ingrowth: a comparison using bone chambers in rats. Bone 19:499 – tein-2 for treatment of open tibial fractures: a prospective, con-
503 trolled, randomized study of four hundred and fifty patients. J Bone
398. Tieline L, Puolakkainen P, Pohjonen T, Rautavuori J, Tormala P, Joint Surg Am 84-A:2123–2134
Rokkanen P 2002 The effect of transforming growth factor-␤1, 406. Raisz LG 1999 Osteoporosis: current approaches and future pros-
released from a bioabsorbable self-reinforced polylactide pin, on a pects in diagnosis, pathogenesis, and management. J Bone Miner
bone defect. Biomaterials 23:3817–3823 Metab 17:79 – 89
399. Schmitt JM, Buck D, Bennett S, Skalla W, Christoforou C, 407. Rosen CJ, Bilezikian JP 2001 Clinical review 123: Anabolic therapy
Buechter D, Gruskin E, Hollinger J 1998 Assessment of an exper- for osteoporosis. J Clin Endocrinol Metab 86:957–964
imental bone wax polymer plus TGF-␤ 1 implanted into calvarial 408. Attisano L, Lee-Hoeflich ST 2001 The Smads. Genome Biology
defects. J Biomed Mater Res 41:584 –592 2:REVIEWS3010
400. Hong L, Tabata Y, Miyamoto S, Yamada K, Aoyama I, Tamura M, 409. Yeo CY, Chen X, Whitman M 1997 The role of FAST-1 and Smads
Hashimoto N, Ikada Y 2000 Promoted bone healing at a rabbit skull in transcriptional regulation by activin during early Xenopus em-
gap between autologous bone fragment and the surrounding intact bryogenesis. J Biol Chem 274:26584 –26590
Endocrine Reviews is published bimonthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society
serving the endocrine community.

Transforming Growth Factor B1 1 To The Bone PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Transforming Growth Factor B1 1 To The Bone PDF

Uploaded by

Copyright:

Available Formats

0163-769X/05/$20.

00/0 Endocrine Reviews 26(6):743–774

Transforming Growth Factor-␤1 to the Bone

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

I. Introduction A. Bone phenotypes of knockout and transgenic mouse

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

FIG. 1. TGF-␤1 processing. Pre-pro-

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

FIG. 2. Signaling by the TGF-␤ family

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

FIG. 3. Role of TGF-␤1 in osteoclasto-

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Effect on ob/bone Effect on oc/bone

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Effect on ob/bone Effect on oc/bone

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Effect on ob/bone Effect on oc/bone

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

TABLE 2. Overview of knockout mouse models of the TGF-␤1 signaling pathway

Gene Mutant phenotype Bone phenotype Refs.

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

TABLE 3. Overview of reported CED mutations in TGF␤1

Exon 1 Exon 2 Exon 4

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

TABLE 4. Bone-related association studies with TGF␤1 polymorphisms

SNP Position Effect Population Refs.

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

TABLE 5. TGF-␤1 as bone-forming agent

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Acknowledgments 19. Dallas SL, Park-Snyder S, Miyazono K, Twardzik D, Mundy GR,

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

Downloaded from https://academic.oup.com/edrv/article-abstract/26/6/743/2355157 by guest on 17 October 2018

You might also like