| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655; and Genetics Institute (V.R.), Cambridge, Massachusetts 02140
Address all correspondence and requests for reprints to: Dr. Jane B. Lian, Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655. E-mail: jane.lian{at}umassmed.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A family of
transcription factors includes three distinct genes that encode
proteins with crucial roles in the regulation of cell fate decision and
transcriptional control of critical genes for cellular differentiation
and development (reviewed in Refs. 1 and 2).
Several key studies have established that Cbfa1 is required
for in vivo bone formation (3, 4, 5) as well as
maturation of hypertrophic chondrocytes (6, 7) and
osteoblast differentiation (3, 8, 9). There is a complete
lack of intramembranous and endochondral bone formation in
Cbfa1 null mice (3), and haploinsufficiency of
this gene results in cleidocranial dysplasia (CCD), a dominantly
inherited developmental disorder of bone (4, 5). The loss
of bone formation is attributed to maturational arrest of the
osteoblast differentiation process (3, 8, 9).
Cbfa1 is expressed in mesenchymal condensations of
developing bones during embryogenesis (3, 5), and the mRNA
has been shown to increase in osteogenic tissues
(8, 9, 10, 11). In the past few years several isoforms for each of the three Cbfa transcription factors have been identified. A shared property of the Cbfa genes is that expression is regulated by at least two distinct promoters that generate two N-terminal isoforms (11, 12, 13). Additional isoforms arise as a result of alternative splicing, exon skipping, as well as deletions and frameshift mutations in the N-terminal, C-terminal, and internal regions of the gene (11, 14). Cbfa1 isoform structure and expression have been studied in rat, mouse, and human (9, 11, 13, 14) and in the context of CCD phenotype (15, 16) where patient genotypes reveal perturbation of the Cbfa1 gene structure. The first Cbfa1 isoform identified, PEBP2aA1 (17) and recently described as the type I isoform (18), is a 513-amino acid protein (designated p56/type I) that initiates in exon 2 at the sequence MRIPV. It was initially shown to be expressed in T cells and Ha-ras-transformed fibroblasts (17, 19) and thymus (20) and was detected in other nonosseous (21), chondrogenic (7, 22), and osteoblast cell lines (8, 11, 18). The second major isoform, til-1 (14) (designated p57 or type II isoform), initiates in exon 1 at the sequence MASNS (11, 14, 23) and is only 15 amino acids longer than the p56/type I isoform. Forced expression of these isoforms modulates transcription of skeletal genes (18, 24, 25, 26), indicating that both proteins are functionally active in osteoblasts and hypertrophic chondrocytes. While attention has focused on the functional activities of these isoforms, their expression in relation to osteoblast maturation and differentiation under control of osteo-inductive factors remains to be addressed. Members of the family of bone morphogenetic proteins, BMP-2 and BMP-4/-7, mediate the commitment of undifferentiated mesenchymal progenitor cells to the skeletal lineage (27, 28, 29) and are potent inducers of Cbfa1 transcription (9, 18, 21, 30). Thus, the early events of osteogenesis, regulated by BMP-2, are closely linked to Cbfa1 induction of the genes involved in bone formation and osteoblast differentiation.
To increase understanding of the role of Cbfa1 in regulating osteogenesis, we examined the expression of Cbfa1 isoforms at different stages of osteoblast development. We also determined the BMP-2 responsiveness of the two major N-terminal isoforms in primary rat calvarial osteoblasts, mouse MC3T3 preosteoblasts, and cells that represent earlier stages of osteoprogenitors. Here we show that the type I transcript is constitutively expressed in nonosseous mesenchymal tissues and during osteoblast differentiation. However, expression of the type II transcript is regulated during osteoblast differentiation and is induced by BMP-2. Our studies provide novel insights into the regulation of Cbfa1 activity in relation to the development of skeletal lineage cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MEM supplemented with 10%
FBS (Atlanta Biologicals, Norcross, GA). MLB13MYC clone 14 and
MLB13MYC clone 17 cell lines were maintained in DMEM (Life Technologies, Inc., Gaithersburg, MD) supplemented with 10% FBS
(32). Cells at passage 12 were changed to DMEM with 1%
FBS when treated with 100 ng/ml BMP-2 (Genetics Institute,
Cambridge, MA). C2C12 cells were maintained in DMEM (33)
supplemented with 5% FBS and treated with 300 ng/ml BMP-2 when
required. Charcoal-stripped serum was prepared for BMP-2-treated cells
by the addition of 5% activated charcoal to FBS at 4 C overnight,
followed by filter sterilization.
Plasmids and analysis of promoter activity
Several promoter-reporter constructs were used for transient
transfection assays as previously described (25). Rat
osteosarcoma ROS 17/2.8 cells and nonosseous HeLa cells plated on
six-well plates were transiently transfected with 2.5 µg/well of
either the 0.6-kb chick bone sialoprotein
promoter-chloramphenicol acetyltransferase (CAT; a gift from
Dr. L. Gerstenfeld, Musculoskeletal Research Laboratory, Boston
University Medical Center, Boston, MA) (34), the 1.6-kb
TGFß type I receptor (TGF-ßRI) promoter-Luc (gift from
Dr. M. Centrella, Department of Surgery, Yale University School of
Medicine, New Haven, CT) (35), or the 1.1-kb rat
osteocalcin (OC) (36) promoter-CAT. The
reporter constructs were cotransfected with 750 ng/well expression
plasmid containing cDNAs of either PEBP2A1 (type I; gift from Dr.
Yoshiaki Ito, Department of Viral Oncology, Kyoto University, Kyoto,
Japan) (17) or til-1(type II; gift from Dr.
James Neil, Department of Veterinary Pathology, University of Glasgow,
Glasgow, UK) (14) using Superfect transfection reagent
(Life Technologies, Inc.). CAT activity was measured
24–36 h after transfection, and transfection efficiency was normalized
by the luciferase activity of the internal control plasmid Rous sarcoma
virus-luciferase (100 ng/well). Representative results of
three independent studies are shown. Data shown are the mean ±
SD (n = 9).
Protein-DNA interaction analysis
Nuclear extracts were prepared from proliferating (d 2 or 3),
differentiated (d 14) and mineralized (d 20 or 23) primary rat
osteoblasts as previously described (37) using 0.45
M KCl for extraction. For mouse MC3T3-E1 nuclear extracts,
cells were collected on d 7, 10, 16, and 22. MLB13MYC clone 14 and
MLB13MYC clone 17 cells [control and 48-h BMP-2-treated (100 ng/ml)]
were collected on d 7, C2C12 cells [control and 48 h
BMP-2-treated (300 ng/ml)] were collected on d 5. EMSAs were performed
using conditions previously described (8). Four micrograms
of nuclear extracts were incubated with 1 µg of the nonspecific
competitor poly(dI-dC)(dI-dC) (Pharmacia Biotech,
Piscataway, NJ) and 10 fmol 32P end-labeled,
double stranded Cbfa binding consensus oligonucleotide
(5'-CGAGTATTGTGGTTAATACG-3'). Protein-DNA complexes were resolved in
4% nondenaturing polyacrylamide gels using Tris-glycine-EDTA buffer.
Antibody supershift experiments contained polyclonal antiserum raised
against a C-terminal peptide of Cbfa1 (38) or preimmune
serum (control). Gels were dried and exposed to Kodak
films (Eastman Kodak Co., New Haven, CT) at -70 C for
6–12 h.
Western blot analysis
Nuclear extracts (30 µg/lane) were resolved in 10% SDS-PAGE
and electroblotted (using a semidry electroblotter; Owl Scientific
Plastics, Cambridge, MA) onto nitrocellulose membranes (0.2 µm,
Protran, Schleicher & Schuell, Inc., Keene, NH) according
to the manufacturer’s specifications. Western blot analyses were
performed as previously described (8). Based on our data
and those of others (35, 39, 40, 41), it is not possible to
electrophoretically separate Runx2/Cbfa1 isoforms, type I (p56) and
type II (p57), nor has it been possible to generate antibodies that can
discriminate between these two proteins. Membranes were incubated at a
1:100 or 1:150 dilution of antibody in Tris-buffered saline containing
1% BSA. IgG-fractionated rabbit polyclonal antibody specific for Cbfa1
(38) or a mouse monoclonal antibody (16) was
used in these studies. Membranes were incubated with secondary antibody
for 45 min, followed by chemiluminescent detection using the enhanced
chemiluminescence kit (Amersham Pharmacia Biotech,
Arlington Heights, IL) according to the manufacturer’s specifications.
Membranes were exposed for 10 sec to 5 min to Amersham Pharmacia Biotech Hyperfilm for detection of signals. Lamin B antibody was
purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz,
CA).
Northern blot analysis
Total cellular RNA was isolated using TRIzol (Life Technologies, Inc., Gaithersburg, MD) according to the
manufacturer’s specifications. Polyadenylated
[poly(A)+] RNA was isolated using an mRNA
isolation kit (Roche, Indianapolis, IN). Ten micrograms of
total RNA or 2 µg poly(A)+ RNA/lane were
separated in a 1% agarose-formaldehyde gel, transferred onto Zetaprobe
membrane (Bio-Rad Laboratories, Inc., Hercules, CA), and
hybridized to probes specific for Cbfa1 exon 1 (which detects the
N-terminal region of type II isoform; GenBank accession no. AF155361),
a full-length cDNA, or a 266-bp BamHI-NcoI
fragment of PEBP2aA1 (GenBank accession no. D14636) that is common to
all isoforms. Hybridization was performed as previously described
(42) in the presence of buffer containing 50% formamide
at 42 C, and the blots were washed extensively in buffer
containing 1 x SSC (standard saline citrate) and 0.1% SDS at
55 C. Data were analyzed after overnight exposure using a Storm
840 PhosphorImager (Molecular Dynamics, Inc., Sunnyvale,
CA). Ethidium bromide staining of the gels was used to assess equal
loading of samples.
RT and PCR
RT was performed on total RNA using Moloney murine leukemia
virus reverse transcriptase (Roche Molecular Biochemicals)
as specified by the manufacturer. Total RNA (1–2 µg) was incubated
at 37 C for 60 min in the presence of deoxy-NTPs, ribonuclease
inhibitor (Promega Corp., Madison, WI),
oligo(deoxythymidine)15 primers (Promega Corp.), and reverse transcriptase. For PCR, Taq
polymerase (Promega Corp.) was used in reactions
containing cDNA from the RT reactions, deoxy-NTPs, 1
µM each of forward and reverse primers, and 1
mM MgCl2. The forward
primer for the type I isoform (p56/PEBP2A1) corresponds to the
N-terminal end of the published cDNA sequence of this isoform (GenBank
accession no. D14636). The forward primer for the type II (p57/til-1)
isoform corresponds to sequences upstream of the translational start
site of til-1. The reverse primer is common to both isoforms. PCR
amplification primers for Cbfa1 and all other genes are listed in Table 1
.
|
A1) cDNA. Hybridized blots were
washed at 65 C with 1 x SSC and 0.1% SDS. Data were analyzed
using a PhosphorImager after exposure for 2–24 h.
Immunofluorescence
Osteoblasts were grown on glass coverslips (Fisher Scientific, Springfield, NJ) at 37 C in growth medium to 40%
confluence for 24 h. Cells were fixed in 4% paraformaldehyde in
PBS or subjected to in situ extraction of cytoskeletal and
soluble chromatin proteins to reveal the nuclear matrix-intermediate
filament scaffold (see below). Coverslips were processed using
protocols previously described by our laboratory (25).
Antibody staining was performed using an affinity-purified Cbfa1 primary antibody (38) at a dilution of 1:200 and was incubated for 1–1.5 h at 37 C. Coverslips were then incubated with a fluorescein isothiocyanate (FITC)-conjugated goat antirabbit secondary antibody (1:500; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) and incubated for 1 h at 37 C to detect Cbfa1. In situ nuclear matrexes were prepared as previously described (43). Briefly, cells on coverslips were washed in PBS and extracted twice in cytoskeletal (CSK) buffer for 15 min each. CSK buffer contains 10 mM PIPES (pH 6.8), 100 mM NaCl, 300 mM sucrose, 3 mM MgCl2, 1 mM EGTA, 0.5% Triton X-100, 1.2 mM phenylmethylsulfonylfluoride, and 1% vanadyl ribonucleoside complex. Deoxyribonuclease I digestion was performed twice in digestion buffer (CSK buffer with 50 mM NaCl) containing 100 µg/ml deoxyribonuclease I for 30 min, followed by an extraction in digestion buffer containing 0.25 M (NH4)2SO4 for 10 min. DNA content was evaluated by staining with 4',6-diamidino-2-phenylindole (DAPI; 5 µg/ml in PBS containing BSA and 0.05% Triton X-100). Cells were mounted in Vectashield H-1000 (Vector Laboratories, Inc., Burlingame, CA). Images were obtained using a CCD camera interfaced with a digital microscope system (Carl Zeiss, Thornwood, NY). Images were analyzed by Metamorph software (Universal Imaging Corp., West Chester, PA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) is regulated by modulation of protein
levels, we performed Western blot analysis (Fig. 1B) with nuclear
extracts isolated during growth and differentiation of primary rat
osteoblasts. The Western analysis showed low Cbfa1 protein levels on d
2 and significant levels in postproliferative d 14 cells. Cbfa1
protein level was thereafter moderately increased on d 20 (Fig. 1B).
These results were confirmed using different antibodies [described in
Ref. 16 and obtained from Oncogene Research Products
(Boston, MA)] and with different time courses. It is noteworthy that
an additional Cbfa1 minor band is present in very heavily mineralized
osteoblast cultures, perhaps indicating partial degradation (data not
shown). At late stages of differentiation, secreted osteocalcin levels
are still in the peak range (Fig. 1C), consistent with increased Cbfa
DNA binding activity (Fig. 1A). Thus, the greatest increases in Cbfa1
cellular protein occur between proliferating (d 2) and differentiated
cells (d 14). However, the changes in Cbfa1 protein between d 14 and
d 20 or d 23 are less dramatic than the increase observed in
DNA binding activity. This discordance may reflect posttranslational
modifications in the osteoblast-specific complex (OBSC) that
enhance Cbfa1 DNA binding activity, which can be observed in native
EMSA gels, in contrast to the denaturing conditions of the Western
analysis.
|
, A and B). Together,
these results establish that all the Cbfa1 protein detected is present
in the nuclei. In proliferating osteoblasts, Cbfa1 is completely
localized in subnuclear domains; thus, the low protein levels and DNA
binding activity are not related to partitioning between the nucleus
and the cytoplasm.
|
). We hybridized
blots of ROB total RNA (from d 2, 7, 11, 18, and 22; Fig. 3, A–C) with
a full-length cDNA probe with the potential to recognize all the Cbfa1
mRNAs (Fig. 3A), as well as with an exon 1 probe (Fig. 3B), which
should specifically detect mRNAs encoding type II isoforms (illustrated
in Fig. 4A). Our results show that
multiple mRNA transcripts ranging in size from 0.6–6 kb are detected
with the full-length probe (Fig. 3A), and there is a developmental
increase in the major transcript from growth to differentiation stages
of rat osteoblasts. Using the exon 1-specific probe, the increase in
the major transcript above the 28S marker becomes more evident (Fig. 3B). Northern analysis of poly A+ RNA using full-length
Cbfa1 cDNA probe (Fig. 3D) confirmed the presence of multiple
transcripts and their increased expression during osteoblast
differentiation.
|
|
, E–G). The MC3T3-E1 cell line has been well characterized
with respect to the stages of osteoblast maturation (44).
Using the same exon 1-specific probe, we observed that type II Cbfa1
RNA levels increased from proliferating (d 4–7) to postproliferative
(d 10) and mineralizing (d 16–22) MC3T3-E1 cultures. However,
consistent with other time courses, we observed a modest decline after
d 22 when the matrix was heavily mineralized in the mouse cell
cultures. Other mRNAs (e.g. osteocalcin) also declined.
Northern blot analysis of poly(A)+ RNA with a
full-length cDNA probe (Fig. 3G) confirmed our findings that multiple
Cbfa1 transcripts are expressed, similar to total cellular RNA in mouse
osteoblasts (Fig. 3E). Thus, the cellular levels of Cbfa1 transcripts
that are derived from exon 1 increase in expression and are
dramatically up-regulated during the onset of differentiation in both
calvarial rat and mouse osteoblasts (Fig. 3, H and I, summary of data
from several time courses).
To further confirm the specific development patterns of expression of
the Cbfa type I and type II transcripts during osteoblast
differentiation, we performed RT-PCR analysis with primers (shown in
Table 1) specific for exon 1 or exon 2 (Fig. 4). Semiquantitative
RT-PCR of mRNA from primary rat osteoblasts (Fig. 4B) and mouse MC3T3
cells (Fig. 4C) indicated that transcripts of both Cbfa1 isoforms were
detected throughout the courses of osteoblast differentiation. Although
the mRNA encoding type I was constitutively expressed from growth to
mineralization stages, the type II encoded transcript increased
postproliferatively from d 2 to 14/d 23 in rat osteoblasts (Fig. 4, B and C) and from d 7 to d 10/d 16 in MC3T3E1 cells (Fig. 4D). Notably, we observed a decline in representation of the type II
isoform in heavily mineralized mouse osteoblasts (d 22; Fig. 4, D and
E), consistent with Northern blot analyses (Fig. 3, E and G). We
confirmed the specificity of the rat osteoblast RT-PCR products
(visualized in agarose gels) by Southern hybridization with a
Cbfa1-specific cDNA probe (described in Materials and
Methods; Fig. 4B, right panel). For comparison,
osteoblast phenotypic markers alkaline phosphatase (ALP) and OC are
shown to reflect osteoblast differentiation, and as expected, the
housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is
constitutive. Taken together, these results show constitutive
expression of type I and developmental and preferential expression of
type II in final stages of osteoblast differentiation Thus, it appears
that the two isoforms together account for Cbfa1 protein levels by
EMSA, Western, and immunofluorescence analyses.
To ascertain the tissue specificity of Cbfa1 type I and II isoforms, we
examined their expression in nonosseous organs and primary cultures of
cells from soft tissues of the newborn mouse. We examined total
cellular RNA from four organs, liver, lung, muscle, and skin, and the
primary cells cultured to confluence (Fig. 5). The type I isoform, but not the type
II isoform, was present in primary cell cultures as well as in RNA from
the tissues. The RT-PCR product for the type I isoform was consistently
detected at very low levels in liver tissue, but at significant levels
in the mesenchymal tissues, particularly skin. Thus, the Cbfa1 type I
isoform, which was shown to be present in thymus and spleen (1, 20), appears to be more ubiquitously expressed than
previously appreciated.
|
) regulated by distinct promoters and the N-terminus of the type II
isoform is 15 amino acids longer. However, the functional consequences
of these molecular differences have not been unequivocally established.
To assess the transcriptional activities of the type II and type I
isoforms, we performed transient cotransfection studies with these two
isoforms and a panel of bone-related Cbfa-responsive target genes
(e.g. OC, TGFß-RI, and bone sialoprotein). Transfections
were performed with osteoblastic ROS 17/2.8 cells (Fig. 6B) and HeLa cells that do not contain
Cbfa1 (Fig. 6A). In both cell types we observed that type I and type II
isoforms activated the OC and TGF-ßRI promoters by 4- to 8-fold and
repressed bone sialoprotein promoter activity by 4- to 8-fold. No
statistically significant differences were observed in the activation
or repression potential of the two isoforms, and both transfected
proteins were expressed at similar levels (Fig. 6C). The similarities
in transcriptional results for both isoforms suggest that the 15-amino
acid extension of the type II isoform does not contribute to activation
or repression of transcription. Our results demonstrate that the type I
and type II isoforms have nearly equivalent transcriptional
activities.
|
, A and B). Thus, Cbfa1 proteins are
uniformly and highly expressed in these endochondral skeletal cells,
which are not yet committed to the osteoblast lineage.
|
) showed that both type
I and type II transcripts were detected at low levels in the clone 14
and clone 17 cells before BMP-2 treatment, when cell phenotype-specific
markers are not expressed (e.g. mALP and mOC). BMP-2
treatment for 48 h significantly enhanced the expression of the
type II transcript, but not the type I transcript, in both cell lines.
In parallel, expression of bone phenotype transcripts mOC and mALP
increased with BMP-2 treatment, indicating differentiation of the
skeletal progenitor cells, whereas expression of the housekeeping gene
mouse GAPDH remained constant. Hence, our data clearly show selective
up-regulation of the type II isoform along with onset of expression of
bone phenotypic markers.
|
). Treatment of cells with BMP-2 substantially increased the
Cbfa1 protein-DNA complex. This result parallels the observed increase
in the type II transcript in differentiated rat and mouse osteoblasts
detected by RT-PCR (Fig. 4). Thus, these biochemical studies reveal
that Cbfa1 proteins are expressed in very early stage mesenchymal
progenitors, before their differentiation into the osteoblast
phenotype. The enhanced expression of the type II transcript in
response to BMP-2 correlates with increased DNA binding activity that
parallels induction of osteoblast differentiation by BMP-2.
BMP-2 induces selective expression of the type II Cbfa1 isoform
during osteogenic differentiation of myogenic progenitor C2C12
cells
The presence of Cbfa1 type II isoform in endochondral progenitor
cells suggests that expression of the type II isoform precedes cellular
commitment to osteoblasts and/or is a marker for cells destined to
differentiate into bone cells. To address this question, we examined
the C2C12 cell model, a premyoblastic cell line of mesenchymal origin,
because muscle cells lack expression of Cbfa1 type II isoform (Fig. 5),
and the C2C12 line is capable of trans-differentiating into
the osteoblast lineage in response to BMP-2 (33). We used
RT-PCR to assess expression of the Cbfa1 type I and type II isoforms in
the myoblast stage and during BMP-2-induced osteoblast differentiation
(Fig. 9). These studies were carried out
in 5% serum, which supports myogenesis, and 10% serum, which favors
osteogenic differentiation. In parallel, cells were also analyzed in
the presence of charcoal treated FBS (stripped serum). Figure 9
demonstrates that expression of the type II isoform was not detected in
control C2C12 cells cultured in either 5% or 10% serum (regular or
stripped). Interestingly, after 48 h of treatment with BMP-2, we
observed a significant increase in expression of the type II transcript
in cultures grown in either 5% or 10% serum, although the increase
was greater in 10% serum. Notably, the type I transcript was expressed
in the C2C12 cell line before treatment with BMP-2 (see control lanes).
Expression of type I did not significantly differ between control and
BMP-2-treated cells in either 10% or 5% serum in repeated studies;
however, in stripped serum, BMP-2 stimulated the type I RT-PCR product.
This finding indicates that factors present in regular serum may
influence the expression of this isoform. As expected, the expression
of myogenic MyoD was suppressed by BMP-2 in cultures supplemented with
either 5% or 10% serum. This down-regulation of MyoD with BMP-2
treatment occurred concomitant with induction of the osteoblast marker
OC, whereas Smad 1, a downstream target of the BMP-2 signaling pathway,
showed no appreciable change in expression, as expected
(45). Together, these results demonstrate that the
expression of Cbfa1 type II isoform is induced with onset of the
osteoblast phenotype (that is, expressing OC), whereas the type I
isoform is expressed in mesenchymal progenitors and is not restricted
to skeletal tissues.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The intricate regulatory pattern of Cbfa1 isoform expression is generated at least in part by utilization of alternative ATG start codons located in different exons and transcribed from two distinct promoters (11, 14, 19). We detected at least two major long transcripts and several minor shorter transcripts by Northern analysis using the full-length Cbfa1 cDNA as probe. Different mRNA variants, which represent utilization of the two promoters as well as alternative splicing and exon skipping (11, 14, 19), have been documented for other Cbfa factors (12, 46). Cbfa2, a key determinant for hemopoiesis (47), is also encoded by multiple transcripts produced through alternate promoter usage and exon skipping (46). Interestingly, alternatively spliced smaller transcripts of the Cbfa1 type II form are predominant in the testis (48). Differential regulation by two alternative promoters has also been observed for other genes. The human c-src gene (49) and the rat bone/liver/kidney/placenta ALP gene (50) each contain two alternative promoters and associated exons that splice to a common downstream exon, resulting in identical coding regions with different 5'-ends. Also, tissue-specific expression of the c-myc gene in Xenopus is controlled by two differentially regulated promoters (51). The expression of Cbfa1 variants from distinct promoters represents a versatile mechanism for temporal and cell-type specific control of Cbfa1 activity during development and for responsiveness to osteogenic factors during bone formation.
The results from our studies as well as those of others have indicated that there is considerable complexity in the expression, molecular characteristics, and biological roles of Cbfa proteins (1, 2, 11, 18, 24). Our studies show that Cbfa-dependent gene regulation involves transcriptional control of Cbfa1, cellular levels of the protein, and posttranslational formation of multiple Cbfa1-containing protein-DNA complexes. Northern blot and RT-PCR analyses of total cellular RNA from primary rat osteoblasts and mouse MC3T3E1 cells revealed increased type II RNA during progressive development of the osteoblast phenotype. The developmental increase in type II mRNA is paralleled by a similar increase in Cbfa1 protein and DNA binding activity of the osteoblast-specific Cbfa1-containing complex during early stages of mature osteoblast differentiation. However, there is a decline in type II expression in heavily mineralized stages of mouse MC3T3 osteoblasts. This decrease may be due to autoregulation of Cbfa1 (13) or may result from apoptosis of cells in vitro. A decline in Cbfa1 mRNA levels has been observed during cellular aging of human trabecular osteoblasts (52).
Although it is not possible to correlate specific transcripts with proteins identified by antibodies (by Western or gel shift analyses), our data indicate that transcription of Cbfa1 is rate limiting for cellular protein and DNA binding activity from the immature osteoblast stage (proliferating) to postproliferative differentiated cells. However, we demonstrate a discordance between expression and protein levels relative to DNA binding in mature osteoblasts only after d 14/d 16. We suggest that Cbfa1 cellular protein is rate limiting for formation of the DNA binding complex in early stage osteoblasts (d 2–14), whereas in late stage osteoblasts, protein-protein interactions or the known posttranslational modifications of Cbfa1 may contribute to enhanced DNA binding activity. Indeed, studies have shown that phosphorylation by MAPK pathways plays a role in regulation of Cbfa1 transcriptional activity (53). Alternatively, the more significant increase in DNA binding occurring in mature osteoblasts may involve Cbfa1 interactions with numerous coregulatory proteins (2, 25). Posttranscriptional regulation or protein-protein interactions with cell type-specific cofactors may also play a major role in tissue-specific expression and activity of the Cbfa1 isoforms. Thus, the significant increase in Cbfa1 DNA binding activity in mature osteoblasts indicates that Cbfa1-interacting proteins may contribute to the regulation of a broad spectrum of genes in osteoblasts and other cell types (25, 26).
Previous in vivo studies have documented that during embryonic development and fracture repair, Cbfa1 is expressed in mesenchymal cell condensations of the early developing mouse (embryonic day 10) skeleton (3, 4, 5, 6, 7, 9, 10, 54). Cbfa1 is present in abundant levels in thymus and T and B cells (17, 55) and is also detected in stromal populations derived from human and mouse bone marrow (56), representing both fully differentiated and intermediate preosteoblastic cells (57), as well as in nonosteogenic clonal colonies (58). The significance of our findings relative to those of previous studies is that we have clearly distinguished the expression patterns of Cbfa1 type I and type II transcripts during commitment and development of the osteoblast phenotype. We demonstrate the presence of the Cbfa1 type I isoform in nonosseous cells of mesenchymal origin and constitutive expression in uninduced skeletal progenitors and early stage proliferating osteoblasts that do not yet express genes reflecting mature osteoblasts (e.g. bone sialoprotein, ALP, and OC) (31, 32, 44). Thus, our studies show that the type I isoform is expressed ubiquitously among cells of mesenchymal origin. We suggest that the Cbfa1 type I isoform appears in early stages of fetal development to provide pluripotent stem cells with the option for commitment to the mesenchymal lineage.
We have provided novel evidence to support the concept that p57/type II isoform of Cbfa1 is specifically related to osteogenic commitment and differentiation, because the type II transcript is selectively up-regulated by BMP-2 and during development of the osteoblast phenotype. Interestingly, we have examined 3 kb of the Cbfa1 type II promoter (13) for BMP-2 responsiveness, and under the conditions of these experiments in which endogenous mRNA and protein are increased, BMP-2 did not mediate a change in the activity of this segment of the promoter (unpublished observations). Our finding that the Cbfa1 type II isoform is selectively induced during BMP-2-mediated osteogenesis in skeletal progenitor cells and nonosseous mesenchymal C2C12 cells clarifies the differences reported in other studies. Gori et al. (29) showed up-regulation of Cbfa1 in response to BMP-2 in immortalized marrow stromal osteoprogenitor cells that already expressed osteoblast phenotypic markers (59). Another study showed that Cbfa1 mRNA is unaffected by BMP-2 treatment in both skeletal and nonskeletal cells, whereas BMP-4/-7 treatment up-regulates Cbfa1 expression in the same cells (21). Thus, future studies must consider interpretation of results within the context of a particular BMP, the cell type, and the specific Cbfa1 isoform.
In summary, our studies demonstrate expression of the p56/type I isoform of Cbfa1 in a spectrum of nonosseous, pluripotent, and committed osteoprogenitor cells before expression of the p57 type II isoform. We observe the regulated expression of the p57/type II isoform in committed osteoprogenitor cells and in osteoblasts as well as after BMP-2-mediated osteoblastic induction of stromal cells or trans-differentiation of C2C12 cells. The data indicate that BMP2-mediated induction of the Cbfa1 type II isoform through utilization of promoter 1 is critical for differentiation to osteoblasts. Our comparative expression studies indicate that although the two N-terminal isoforms are differentially expressed, they have functionally equivalent transcriptional activity on promoters in both osseous and nonosseous cell lines. Recent studies in vivo support this concept (60). Thus, the timing of expression of each Cbfa1 isoform, perhaps in conjunction with responses to BMPs in different subpopulations of cells during specific stages of bone development, must be an important component of the mechanism by which Cbfa1 regulates osteogenesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
B), RUNX2 (AML3/CBFA1/PEBP2A),
and RUNX3 (AML2/CBFA3/PEBP2C). Abbreviations: ALP, Alkaline phosphatase; BMP, bone morphogenetic protein; CCD, cleidocranial dysplasia; CSK, cytoskeletal; DAPI, 4',6-diamidino-2-phenylindole; FITC, fluorescein isothiocyanate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; mALP, mouse alkaline phosphatase; OBSC, osteoblast-specific complex; OC, osteocalcin; poly(A)+, polyadenylated; ROB, rat osteoblasts; TGFß-RI, TGFß type I receptor.
Received October 9, 2000.
Accepted for publication May 10, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A
(CBFA1): evidence for a new myc collaborating oncogene. Proc Natl Acad
Sci USA 94:8646–8651 A/CBFA1 mutation displaying impaired transactivation and
Smad interaction in cleidocranial dysplasia. Proc Natl Acad Sci USA 97:10549–10554 genes in T cells during
thymic development. Mol Cell Biol 15:1662–1670[Abstract]
A/AML3/CBFA1 gene is regulated by BMP4/7 heterodimer and its
overexpression suppresses type I collagen and osteocalcin gene
expression in osteoblastic and nonosteoblastic mesenchymal cells. Bone 22:87–92[Medline]
/AML/PEBP2) dependent activation of
tissue-specific gene transcription. J Cell Sci 113:2221–2231[Abstract]
transcription factors. Proc Natl Acad Sci USA 94:6746–6751. J Biol Chem 275:37604–37611This article has been cited by other articles:
 |
|
 |
|
  S.-L. Cheng, J.-S. Shao, J. Cai, O. L. Sierra, and D. A. Towler Msx2 Exerts Bone Anabolism via Canonical Wnt Signaling J. Biol. Chem., July 18, 2008; 283(29): 20505 - 20522. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chen, J. Gluhak-Heinrich, M. Martinez, T. Li, Y. Wu, H.-H. Chuang, L. Chen, J. Dong, I. Gay, and M. MacDougall Bone Morphogenetic Protein 2 Mediates Dentin Sialophosphoprotein Expression and Odontoblast Differentiation via NF-Y Signaling J. Biol. Chem., July 11, 2008; 283(28): 19359 - 19370. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Lamour, C. Detry, C. Sanchez, Y. Henrotin, V. Castronovo, and A. Bellahcene Runx2- and Histone Deacetylase 3-mediated Repression Is Relieved in Differentiating Human Osteoblast Cells to Allow High Bone Sialoprotein Expression J. Biol. Chem., December 14, 2007; 282(50): 36240 - 36249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Simionescu, D. T. Simionescu, and N. R. Vyavahare Osteogenic Responses in Fibroblasts Activated by Elastin Degradation Products and Transforming Growth Factor-{beta}1: Role of Myofibroblasts in Vascular Calcification Am. J. Pathol., July 1, 2007; 171(1): 116 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Q. Hassan, R. S. Tare, S. H. Lee, M. Mandeville, M. I. Morasso, A. Javed, A. J. van Wijnen, J. L. Stein, G. S. Stein, and J. B. Lian BMP2 Commitment to the Osteogenic Lineage Involves Activation of Runx2 by DLX3 and a Homeodomain Transcriptional Network J. Biol. Chem., December 29, 2006; 281(52): 40515 - 40526. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Handschin, M. Egermann, O. Trentz, G. A. Wanner, H.-J. Kock, G. Zund, and O. A. Trentz Cbfa-1 (Runx-2) and Osteocalcin Expression by Human Osteoblasts in Heparin Osteoporosis in Vitro Clinical and Applied Thrombosis/Hemostasis, October 1, 2006; 12(4): 465 - 472. [Abstract] [PDF]
|
|
|
|
 |
|
 J. E. Phillips, C. A. Gersbach, A. M. Wojtowicz, and A. J. Garcia Glucocorticoid-induced osteogenesis is negatively regulated by Runx2/Cbfa1 serine phosphorylation J. Cell Sci., February 1, 2006; 119(3): 581 - 591. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Seeherman, R. Li, M. Bouxsein, H. Kim, X. J. Li, E. A. Smith-Adaline, M. Aiolova, and J. M. Wozney rhBMP-2/Calcium Phosphate Matrix Accelerates Osteotomy-Site Healing in a Nonhuman Primate Model at Multiple Treatment Times and Concentrations J. Bone Joint Surg. Am., January 1, 2006; 88(1): 144 - 160. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-H. Lee, Y.-J. Kim, W.-J. Yoon, J.-I. Kim, B.-G. Kim, Y.-S. Hwang, J. M. Wozney, X.-Z. Chi, S.-C. Bae, K.-Y. Choi, et al. Dlx5 Specifically Regulates Runx2 Type II Expression by Binding to Homeodomain-response Elements in the Runx2 Distal Promoter J. Biol. Chem., October 21, 2005; 280(42): 35579 - 35587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chen, S. Rani, Y. Wu, A. Unterbrink, T. T. Gu, J. Gluhak-Heinrich, H.-H. Chuang, and M. MacDougall Differential Regulation of Dentin Sialophosphoprotein Expression by Runx2 during Odontoblast Cytodifferentiation J. Biol. Chem., August 19, 2005; 280(33): 29717 - 29727. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Galindo, J. Pratap, D. W. Young, H. Hovhannisyan, H.-J. Im, J.-Y. Choi, J. B. Lian, J. L. Stein, G. S. Stein, and A. J. van Wijnen The Bone-specific Expression of Runx2 Oscillates during the Cell Cycle to Support a G1-related Antiproliferative Function in Osteoblasts J. Biol. Chem., May 27, 2005; 280(21): 20274 - 20285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kojima and T. Uemura Strong and Rapid Induction of Osteoblast Differentiation by Cbfa1/Til-1 Overexpression for Bone Regeneration J. Biol. Chem., January 28, 2005; 280(4): 2944 - 2953. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Selvamurugan, S. Kwok, and N. C. Partridge Smad3 Interacts with JunB and Cbfa1/Runx2 for Transforming Growth Factor-{beta}1-stimulated Collagenase-3 Expression in Human Breast Cancer Cells J. Biol. Chem., June 25, 2004; 279(26): 27764 - 27773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z.-S. Xiao, A. B. Hjelmeland, and L. D. Quarles Selective Deficiency of the "Bone-related" Runx2-II Unexpectedly Preserves Osteoblast-mediated Skeletogenesis J. Biol. Chem., May 7, 2004; 279(19): 20307 - 20313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Komaki, A. Asakura, M. A. Rudnicki, J. Sodek, and S. Cheifetz MyoD enhances BMP7-induced osteogenic differentiation of myogenic cell cultures J. Cell Sci., March 15, 2004; 117(8): 1457 - 1468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Pratap, M. Galindo, S. K. Zaidi, D. Vradii, B. M. Bhat, J. A. Robinson, J.-Y. Choi, T. Komori, J. L. Stein, J. B. Lian, et al. Cell Growth Regulatory Role of Runx2 during Proliferative Expansion of Preosteoblasts Cancer Res., September 1, 2003; 63(17): 5357 - 5362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Canalis, A. N. Economides, and E. Gazzerro Bone Morphogenetic Proteins, Their Antagonists, and the Skeleton Endocr. Rev., April 1, 2003; 24(2): 218 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Willis, A. P. Loewy, N. Charlton-Kachigian, J.-S. Shao, D. M. Ornitz, and D. A. Towler Regulation of Osteocalcin Gene Expression by a Novel Ku Antigen Transcription Factor Complex J. Biol. Chem., September 27, 2002; 277(40): 37280 - 37291. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Silva and L. J. Sandell What's New in Orthopaedic Research J. Bone Joint Surg. Am., August 12, 2002; 84(8): 1490 - 1496. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Stewart, N. MacKay, E. R. Cameron, and J. C. Neil The Common Retroviral Insertion Locus Dsi1 Maps 30 Kilobases Upstream of the P1 Promoter of the Murine Runx3/Cbfa3/Aml2 Gene J. Virol., March 27, 2002; 76(9): 4364 - 4369. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Gilbert, X. He, P. Farmer, J. Rubin, H. Drissi, A. J. van Wijnen, J. B. Lian, G. S. Stein, and M. S. Nanes Expression of the Osteoblast Differentiation Factor RUNX2 (Cbfa1/AML3/Pebp2alpha A) Is Inhibited by Tumor Necrosis Factor-alpha J. Biol. Chem., January 18, 2002; 277(4): 2695 - 2701. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
