0013-7227/01/$03.00/0
Endocrinology
Copyright © 2001 by The Endocrine Society
Vol. 142, No. 7
Printed in U.S.A.
Minireview: Transcriptional Control of Osteoblast
Differentiation
GERARD KARSENTY
Baylor College of Medicine, Houston, Texas 77030
ABSTRACT
The last 5 yr have witnessed major progress in skeleton biology.
One of these areas of progress has been the partial elucidation of the
transcriptional mechanisms of osteoblast differentiation and their
conservation between mouse and human. Cbfal, or runx2, a homolog
of the Drosophila Runt protein serves both as the earliest transcriptional regulator of osteoblast differentiation and as controller of bone
T
HE MOLECULAR UNDERSTANDING of the embryonic development and of the physiology of the skeleton
has been completely renewed in the last 5 yr. Up until the
early mid 1990s the only class of regulatory proteins that
were known to affect skeleton biology were the bone morphogenetic proteins, and the main structural proteins to
which functions were assigned were collagens. This has completely changed through the progress of mouse genetics and
to a large extent of human genetics. We know now of many
more genes encoding either circulating molecules or transcription factors that control patterning of the skeleton, differentiation of the three specific cell types of the skeleton
and/or the functions of these cells. These three specific cell
types are the chondrocytes in cartilage, the osteoblasts or
bone forming cells, and the osteoclast or bone-resorbing cells
in bone. This review will focus on only one cell type the
osteoblast, a bone forming cell, and only transcriptional regulation of cell differentiation and function. However, as the
field develops it has become clear that at least one transcription factor controls cell differentiation in both osteoblast and
chondrocyte lineages and therefore we will touch upon chondrocyte differentiation as well.
Interaction between Chondrocyte and Osteoblast
Differentiation during Endochondral Ossification
Except for the clavicles and some bones of the skull that
develop without a cartilage template through a process
called intramembranous ossification, all the other bones develop through a process called endochondral ossification,
which does make use of a cartilage template (1). Beyond
differentiation of mesenchymal cells into proliferating chondrocytes in each mesenchymal condensation prefiguring the
future skeleton, there are three key events occurring late
during endochondral ossification. There is first hypertrophy
Received March 26, 2001.
Address all correspondence and requests for reprints to: Gerard
Karsenty, Ph.D., M.D., Baylor College of Medicine, Department of Human and Molecular Genetics, Room 5930, One Baylor Plaza, Houston,
Texas 77030.
formation by already differentiated osteoblast. Moreover, Cbfal is also
a hypertrophic chondrocyte differentiation factor in several skeletal
elements. Although other transcription factors are likely to be involved in osteoblast differentiation, Cbfal, by the multiplicity of the
role that it plays, can be viewed as the central regulator of intramembranous and endochondral ossification. (Endocrinology 142: 2731–
2733, 2001)
of chondrocytes, a process finely controlled by several
growth factors (2– 4). Hypertrophic chondrocytes are surrounded by a calcified extracellular matrix rich in type X
collagen that, through unknown mechanisms, favors the second key event of endochondral ossification, which is vascular invasion from the perichondrium or bone collar (5, 6). This
vascular invasion brings in mesenchymal cells from the bone
collar that will differentiate into osteoblasts secreting a type
I collagen rich extracellular matrix. This is the last key cellular
event during endochondral ossification. This sequence of
events illustrates best the interdependence between chondrocyte hypertrophy and osteoblast differentiation for bone
formation to occur. The transcriptional control of these two
cellular events has been partly elucidated in the recent years.
Transcriptional Control of Osteoblast Differentiation
In theory a transcriptional activator of osteoblast differentiation should fulfill three criteria: it should be expressed in
osteoblast progenitors, it should control transcription of many
genes expressed in osteoblasts and it should be necessary for
osteoblast differentiation in vitro and in vivo. To date there is
only one transcription factor identified that fulfills all these
criteria. This factor is called Cbfa1 or Runx2. For the sake of
clarity we will use the name of Cbfa1 in this review because it
is the name most commonly used. Although Cbfa1 plays a
critical role during osteoblast differentiation and this will be the
focus of this review, it should be emphasized that most likely
other transcription factors fulfilling these three criteria will be
identified. Cbfa1 is one of the mouse homologs of the drosophila runt protein (7). Runt itself is the founding member of a small
family of important transcription/differentiation factors that
are conserved from C. elegans to human (8). They are all characterized by a DNA binding domain called the runt domain,
which is 128 amino acids long (9). Outside the Runt domain,
there is little homology between runt and lozenge, the two
drosophila members of this family, and their mammalian homologues which were all cloned in the beginning of the 1990s
(10).
Although Cbfa1 was initially thought to be expressed in
2731
2732
OSTEOBLAST DIFFERENTIATION
the thymus (7), its biologic importance lies elsewhere. From a
molecular biology perspective, Cbfa1 was identified as one of
the two key regulators of the osteoblast-specific expression of
osteocalcin (11, 12), the most if not the only osteoblast specific
gene. Cbfa1 binds to a cis-acting element present in the osteocalcin promoter in mouse, rat, and human that is necessary and
sufficient to confer osteoblast-specific expression to osteocalcin
or other genes. Cbfa1 during development is initially expressed
at embryonic day 12 (E12) in the mesenchymal cells present in
every mesenchymal condensations prefiguring the future skeletal elements whether they form through endochondral or intramembranous ossification (12). This expression of Cbfa1 in
these cells identifies them as chondro-osteoprogenitor. Starting
at E14 Cbfa1 expression becomes progressively stronger in cells
of the osteoblast lineage and fades away in cells of the chondrocyte lineage as will be discussed below. To date Cbfa1 is the
most specific and the earliest marker of the osteoblast lineage
and certainly fulfills the first condition to be a transcriptional
activator of osteoblast differentiation (12). An interesting aspect
of Cbfa1 biology that has not been fully understood yet is the
fact that Cbfa1 expression precedes osteogenesis from 2–3 days.
This could be explained by two mechanisms: either Cbfa1 controls the expression of other transcription factors or the function
of Cbfa1 is prevented during development by posttranslational
processes or binding to an inhibitory co-factor for example.
The early pattern of expression of Cbfa1 contrasts with the
fact that its role in regulating Osteocalcin, a late marker of the
osteoblast lineage. This in fact reflects that Cbfa1 is expressed
at high levels in differentiated osteoblasts in postnatal life
where it does play a role (see below). More in line with the
early expression of Cbfa1 during skeleton development is
the observation that Cbfa1 binding sites are present in all the
genes expressed in an osteoblast and whose gene products
contribute to form a bone extracellular matrix. Those are the
type I collagen genes, bone sialoprotein (Bsp), and Osteopontin
among others (12, 13). Cbfa1 expression is not only cellspecific but it is also induced by the bone morphogenetic
proteins (12), a family of secreted proteins that play an important role at the onset of endochondral ossification (14).
The function of Cbfa1 fulfills all the promises made by its
pattern of expression. For instance, Cbfa1 does act as a differentiation factor in vitro and in vivo. Indeed, transfection of
Cbfa1 expression vector in mouse primary fibroblasts or in
myoblast leads to the expression of the cells of molecular
markers of the osteoblast lineage such as Osteocalcin and
Bone sialo protein (BSP) (12). The differentiation function of
Cbfa1 is better demonstrated in mice. Inactivation of Cbfa1
in mice prevents the appearance of osteoblast (15, 16). As a
result the animals are left with a purely cartilaginous skeleton, although as will be explained below chondrocyte differentiation is not completely normal in these mutant mice.
Moreover, Otto et al. (1997) uncovered that mutant mice
heterozygote for the Cbfa1 deletion had a phenotype similar
to a classical mouse mutant called cleidocranial dysplasia
(CCD) (17). Cbfa1 maps at the same location as CCD and in
fact the two mutations are allelic. Human genetic studies also
contributed to our understanding of Cbfa1 importance during skeleton development (18, 19). Indeed, as it is the case for
mice, individuals heterozygous for deletion, missense mutation, and substitution mutation in the DNA binding do-
Endo • 2001
Vol. 142 • No. 7
main of Cbfa1 develop cleidocranial dysplasia a skeletal dysplasia marked primarily by a delay in the suture of the
fontanelle and a virtual absence of clavicles. Most of the
patients with CCD have a normal bone mass suggesting that
a 50% decrease in Cbfa1 expression is not sufficient to cause
bone loss.
Role of Cbfa1 during chondrocyte hypertrophy
As mentioned above Cbfa1 is expressed early during development (E12.5) in a cell type with apparently the potential
to become either an osteoblast or a chondrocyte. From E12 to
birth, in mice, Cbfa1 expression in cartilage is restricted to
prehypertrophic and hypertrophic chondrocytes (20). The
expression of Cbfa1 seems to be higher in prehypertrophic
chondrocytes although this could be explained by the fact
that there are more cells in the hypertrophic zone than in the
hypertrophic zone. In any case, this temporal and spatial
pattern of restriction is consistent with a role for Cbfa1 as an
inducer of chondrocyte hypertrophy since it starts to be
expressed in cells of the chondrocytic lineage before chondrocyte hypertrophy does take place. This suspicion is reinforced by the observation that in Cbfa1-deficient mice hypertrophic chondrocytes are absent in femur and humerus
(21, 22). These two skeletal elements develop before the ulna
and tibia for instance that do contain hypertrophic chondrocytes in Cbfa1-deficient mice. This indicates that the absence
of these cells in the humeri and femur cannot be ascribed to
a mere delay in cell differentiation but rather reflect a true
function of Cbfa1.
In agreement with the hypothesis that Cbfa1 could act as
an inducer of chondrocyte hypertrophy, in transgenic mice
expressing Cbfa1 in nonhypertrophic chondrocytes under
control of the chondrocyte targeting promoter, col X,
throughout development there is hypertrophic chondrocyte
differentiation in skeletal elements where it normally never
occurs (20). This ectopic chondrocyte hypertrophy takes
place for instance in the chondrocostal cartilage and the
intervertebral discs. In turn, this ectopic chondrocyte hypertrophy led to ectopic bone formation leaving unanswered the
question of whether Cbfa1 is inducing transdifferentiation of
proliferating chondrocytes into osteoblasts or if instead
Cbfa1 was a hypertrophic chondrocyte differentiation factor.
This question was answered when the Cbfa1 transgene under the control of the col X promoter, was transferred onto
a Cbfa1-deficient genetic background. Indeed, on this background the transgene restored chondrocyte hypertrophy and
vascular invasion in the bones of the mutant mice but did not
induce osteoblast differentiation (20), thus establishing that
Cbfa1 is also a hypertrophic chondrocyte differentiation factor. This experiment also established that there is a common
regulation of osteoblast and chondrocyte differentiation and
that Cbfa1 is one of the genes involved in this regulation.
Transcriptional control of osteoblast function
The main function of the osteoblast throughout life is to
produce a bone extracellular matrix. This process, also called
bone formation, is obviously important for longitudinal
growth but also to maintain a constant bone mass through
bone remodeling, the process by which bone constantly re-
OSTEOBLAST DIFFERENTIATION
news itself during adulthood. Cbfa1 is also involved in this
cellular physiology aspect of bone biology. The DNA binding
domain of Cbfa1 has no transactivation function but has a
higher affinity for DNA than Cbfa1 itself, a feature seen in
other members of the runt-related family of transcription
factors (23–25). In transgenic mice overexpression of this
transcriptionally inactive form of Cbfa1 only in differentiated
osteoblasts and only after birth led to a bone loss phenotype
characterized by its post development and differentiation in
appearance and its coexistence with a normal number of
osteoblasts (24). The postnatal expression of this dominant
negative of Cbfa1 led to a down-regulation of Cbfa1 expression itself and of the expression of the genes required to
produce a bone extracellular matrix (24). This function of
Cbfa1 could not be uncovered by studying mice or patients
heterozygote for a Cbfa1 deletion. Thus, Cbfa1 is a transcription factor for all seasons in the life of an osteoblast. It is
necessary for osteoblast differentiation and it also acts as a
regulator of osteoblast function.
Before, Beyond, and Besides Cbfa1
As mentioned at the beginning of this review, it is likely
that several gene products acting either positively or negatively are involved in the control of osteoblast differentiation
and function. Cbfa1 expression must be induced and must be
regulated, Cbfa1 itself may control the expression of other
osteoblast-specific transcription factors and it is possible that
other transcription factors affect osteoblast differentiation in
a Cbfa1 independent manner. Given that there are several
hundred skeletal elements in the body it is likely that Cbfa1
expression will be regulated by different genes in different
parts of the skeleton. Several transcription factors have been
suspected, based on genetic arguments, to act upstream of
Cbfa1 in one linear cascade controlling osteoblast differentiation. For instance, the deletion of Msx2, a homeobox containing protein, causes severe skeletal abnormalities associated with a decrease in Cbfa1 expression (26) suggesting that
it could act upstream of Cbfa1. Likewise, Bpx, another homeobox containing protein may also be involved in regulating Cbfa1 expression however this time in the prospective
vertebral column (27).
As mentioned above Cbfa1 expression precedes osteogenesis by several days. On one hand, there is to date no report
of a gene whose expression is controlled by Cbfa1 and that
would mediate some of its action. This does not mean that
such factors do not exist. On the other hand, the observation
that HoxA2 deficiency leads to ectopic Cbfa1 expression and
bone formation (28) suggests the existence of a negative
regulation of osteogenesis by factors that either inhibit Cbfa1
gene expression or directly interact with the Cbfa1 protein to
prevent its transcriptional activity and/or stability. The first
challenge that the field faces in the future is to identify all the
players in the genetic cascade that include Cbfa1. The second
challenge will be to determine if there is any Cbfa1-independent genetic pathway(s) involved in osteoblast differentiation and/or function.
References
1. Horton WA 1993 Morphology of connective tissue: Cartilage. In: Connective
Tissue and Its Heritable Disorders. Wiley-Liss Inc., New York, pp 73– 84
2733
2. Karaplis AC, Luz A, Glowacki J, Bronson RT, Tybulewicz VL, Kronenberg
HM, Mulligan RC 1994 Lethal skeletal dysplasia from targeted disruption of
the parathyroid hormone-related peptide gene. Genes Dev 8:277–289
3. St-Jacques B, Hammerschmidt M, McMahon AP 1999 Indian hedgehog signaling regulates proliferation and differentiation of chondrocytes and is essential for bone formation. Genes Dev 1:2072–2086
4. Vortkamp A, Lee K, Lanske B, Segre GV, Kronenberg HM, Tabin CJ 1996
Regulation of rate of cartilage differentiation by Indian hedgehog and PTHrelated protein. Science 273:613– 622
5. Linsenmayer TF, Chen QA, Gibney E, Gordon MK, Marchant JK, Mayne R,
Schmid TM 1991 Collagen type IX and X in the developing chick tibiotarsus:
analysis of mRNA and proteins. Development 111: 191–196
6. Poole AR 1991 Cartilage: molecular aspects. In: Hall BK, Newman SA (eds) The
Growth Plate: Cellular Physiology, Cartilage Assembly and Mineralization. CRC
Press, Boca Raton, FL
7. Satake M, Nomura S, Yamaguchi-Iwai Y, Takahara Y, Hashimoto Y, Niki M,
Kitamura Y, Ito Y 1995 Expression of the Runt domain-encoding PEBP2 ␣
genes in T cells during thymic development. Mol Cell Biol 15:1662–1670
8. Wheeler JC, Shigesada K, Gergen JP, Ito Y 2000 Mechanisms of transcriptional
regulation of Runt domain proteins. Semin Cell Dev Biol 11:369 –375
9. Kagoshima H, Shigesada K, Satake M, Ito Y, Miyoshi H, Ohki M, Pepling
M, Gergen P 1993 The Runt domain identifies a new family of heteromeric
transcriptional regulators. Trends Genet 9:338 –341
10. Ito Y, Bae, S-C 1997 The Runt domain transcription factor PEBP2/CBF, and
its involvement in human leukemia. In: Yaniv M, Ghysdael J (eds) Leukemia,
vol 2. Basel, Switzerland
11. Ducy P, Karsenty G 1995 Two distinct osteoblast-specific cis-acting elements
control expression of a mouse osteocalcin gene. Mol Cell Biol 15:1858 –1869
12. Ducy P, Zhang R, Geoffroy V, Rydall AL, Karsenty G 1997 Osf2/Cbfa1, a
transcriptional regulator of osteoblast differentiation. Cell 8:747–754
13. Kern B, Shen J, Starbuck M, Karsenty G, Cbfa1 contributes to the osteoblastspecific expression of type I collagen genes. J Biol Chem 276:7101–7107
14. Wozney JM, Rosen V 1998 Bone morphogenetic protein and bone morphogenetic protein gene family in bone formation and repair. Clin Orthop
346:26 –37
15. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu
Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki
S, Kishimoto T 1997 Targeted disruption of Cbfa1 results in a complete lack
of bone formation owing to maturational arrest of osteoblasts. Cell 89:755–764
16. Otto F, Thronelkl AP, Crompton T, Denzel A, Gilmour KC, Rosewell IR,
Stamp GWH, Beddington RSP, Nundlos S, Olsen BR, Selby PB, Owen MJ
1997 Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential
for osteoblast differentiation and bone development. Cell 89:765–771
17. Sillence DO, Ritchie HE, Selby PB 1987 Animal model: skeletal anomalies in
mice with cleidocranial dysplasia. Am J Med Genet 27:75– 85
18. Lee B, Thirunavukkarasu K, Zhou L, Pastore L, Baldini A, Hecht J, Geoffroy
V, Ducy P, Karsenty G 1997 Missense mutations abolishing DNA binding of
the osteoblast-specific transcription factor OSF2/CBFA1 in cleidocranial dysplasia. Nat Genet 16:307–310
19. Mundlos S, Otto F, Mundlos C, Mulliken JB, Aylsworth AS, Albright S,
Lindhout D, Cole WG, Henn W, Knoll JH, Owen MJ, Mertelsmann R, Zabel
BU, Olsen BR 1997 Mutations involving the transcription factor CBFA1 cause
cleidocranial dysplasia. Cell 89:773–779
20. Takeda S, Bonnamy JP, Owen MJ, Ducy P, Karsenty G 2001 Continuous
expression of Cbfa1 in non-hypertrophic chondrocytes uncovers its ability to
induce hypertrophic chondrocytes differentiation and partially rescues Cbfa1deficient mice. Genes Dev 15:467– 480
21. Inada M, Yasui T, Nomura S, Miyake S, Deguchi K, Himeno M, Sato M,
Yamagiwa H, Kimura T, Yasui N, Ochi T, Endo N, Kitamura Y, Kishimoto
T, Komori T 1999 Maturational disturbance of chondrocytes in Cbfa1-deficient
mice. Dev Dyn 214:279 –290
22. Kim IS, Otto F, Abel B, Mundlos S 1999 Regulation of chondrocyte differentiation by Cbfa1. Mech Dev 80:159 –170
23. Aronson BD, Fisher AL, Blechman K, Caudy M, Gergen JP 1997 Grouchodependnt and -independent repression activities of Runt domain proteins. Mol
Cell Biol 17:5581–5587
24. Ducy P, Starbuck M, Priemel M, Shen J, Pinero G, Geoffroy V, Amling M,
Karsenty G 1999 A Cbfa1-dependent genetic pathway controls bone formation
beyond embryonic development. Genes Dev 13:1025–1036
25. Tracey WD, Pepling ME, Horb ME, Thomsen GH, Gergen JP 1998 A Xenopus
homologue of aml-1 reveals unexpected patterning mechanisms leading to the
formation of embryonic blood. Development 125:1371–1380
26. Satokata I, Ma L, Ohshima H, Bei M, Woo I, Nishizawa K, Maeda T, Takano
Y, Uchiyama M, Heaney S, Peteres H, Tang Z, Maxson R, Maas R 2000 Msx2
deficiency in mice causes pleotropic defects in bone growth and ectodermal
organ formation. Nat Genet 24: 391–395
27. Tribioli C, Lufkin T 1999 The murine Bapx1 homeobox gene plays a critical
role in embryonic development of the axial skeleton and spleen. Development
126:5699 –5711
28. Kanzler B, Kuschert SJ, Liu Y-H, Mallo M 1998 Hoxa-2 restricts the chondrogenic domain and inhibits bone formation during development of the
branchial area. Development 125:2587–2597