1. Top of page
  2. Abstract
  7. Acknowledgements

We investigated the regulation of Sox9, a transcription factor known to play a role in chondrogenesis, by bone morphogenetic protein-2 (BMP-2) and hedgehog proteins in order to better understand their signaling function in endochondral bone formation. The mesenchymal progenitor cell line C3H10T1/2 was stimulated with BMP-2. Sox9 expression levels were measured by quantitative reverse transcriptase-polymerase chain reaction and Northern analysis. We found that Sox9 was up-regulated by BMP-2 in a dose-dependent manner. The expression of Col2a1, a downstream response gene of Sox9, was also significantly increased upon BMP-2 addition. We also monitored Sox9 expression after the addition of BMP-2 to osteosarcoma cell lines; BMP-2 treatment increased Sox9 mRNA levels in MG63, considered to be early osteoblast-like, but not in human osteogenic sarcoma (HOS) cells, which are thought to be more advanced in the osteoblastic lineage. This response seems to be influenced by differences in BMP receptor expression; MG63 cells express BMP receptor IA (BMPR-IA), whereas HOS cells express BMPR-IA and BMPR-IB. We also saw an increase in Sox9 mRNA levels in BMP-2–treated primary human bone cells (HBCs) derived from femoral heads. We found that in addition to BMP-2, Sonic and Indian hedgehog can increase Sox9 expression in C3H10T1/2 and primary HBCs. Time course studies with C3H10T1/2 cells after BMP-2 stimulation showed increasing expression of cartilage markers, decrease of collagen I mRNA, and a late induction of osteocalcin expression. Moreover, the treatment of C3H10T1/2 cells with Sox9 antisense oligonucleotides revealed that Sox9 is a downstream mediator of BMP-2 affecting the expression of chondrocyte and osteoblast marker genes. Our data show that Sox9 is an important downstream mediator of the BMP-2 and hedgehog signaling pathways in osteogenic cells.


  1. Top of page
  2. Abstract
  7. Acknowledgements

The SOX factors comprise a group of ∼20 proteins characterized by the presence of a 79 amino acid HMG-type DNA-binding domain (SRY box).(1) The expression patterns of some Sox family members throughout development correlate with early cell fate decisions(2) and in vitro studies have demonstrated that several Sox proteins, including Sox9, activate transcription.(3–5)

Sox9 expression is high in mouse chondroprogenitor cells and in fully differentiated chondrocytes during embryonic development.(6) The identification of Sox9 mutations in campomelic dysplasia, a severe dwarfism syndrome characterized by abnormalities in all skeletal elements derived from cartilage, also suggests a role for Sox9 in endochondral bone development.(7,8) There is already evidence that Sox9 regulates the transcription of type II collagen, a chondrocyte specific gene.(9,10)

Bone morphogenetic proteins (BMPs) were originally detected in and purified from demineralized bone using an in vivo cartilage and bone induction assay.(11) BMPs are regulators of chondrocyte proliferation and differentiation(12,13) and induce ectopic cartilage and bone when implanted intramuscularly in rats and mice.(14,15) In mouse embryos, BMP-2 mRNA is expressed in areas of developing cartilage and bone.(16,17) BMP-2 has been shown to cause osteoblastic and chondroblastic maturation of the mesenchymal stem cell line C3H10T1/2 in vitro.(18–21)

BMP-2 is secreted in the mesoderm and apical ectodermal ridge in response to Sonic hedgehog (SHH) during limb development.(22) SHH is one of three vertebrate homologs (Sonic, Indian, and Desert Hedgehog) of the Drosophila segment polarity gene hedgehog.(23) Recent results indicate that SHH induces osteoblast differentiation and ectopic bone formation.(24) Indian Hedgehog (IHH) has been reported to regulate chondrocyte differentiation in chicken embryos during skeletogenesis.(25–27)

We investigated the regulation of Sox9 expression by BMP-2 and the downstream signaling of BMP-2 by Sox9 in order to better understand their function in endochondral bone formation. Our study identifies for the first time Sox9 as a downstream mediator of BMP-2 and hedgehog signaling.


  1. Top of page
  2. Abstract
  7. Acknowledgements

All enzymes, media, and reagents were from Boehringer Mannheim (Roche Molecular Biochemicals, Mannheim, Germany) except where noted differently.

Cell culture

Cells were routinely cultured at 37°C in a 5% CO2 atmosphere. Mouse fibroblast C3H10T1/2 cells (ATCC CCL 226) were grown in monolayers in Dulbecco's modified Eagle's medium with 10% fetal calf serum (FCS). MG63 human osteosarcoma cells (ATCC CRL 1427) were maintained in Earle's modified essential medium with 5% FCS. Human osteogenic sarcoma (HOS) cells (TE 85, clone F-5, ATCC CRL 1543) were grown in Earle's modified essential medium containing 10% FCS. For primary culture, bone cells were obtained from the trabecular bone of femoral heads of patients undergoing orthopaedic surgery for coxarthrosis. The bone samples were dissected into small pieces, washed twice in phosphate-buffered saline, and digested with collagenase A five times. The bone cells obtained from digestion steps three to five were pooled, maintained in Dulbecco's modified Eagle's medium containing 5% FCS and used in no higher passage than four.

BMP-2 was provided by W. Sebald, Würzburg, Germany, and used in 200 ng/ml concentration. Human hedgehog proteins were derived from baculovirus-mediated expression in insect cells. cDNA clones for IHH and SHH were obtained from Ontogeny, Inc. (Cambridge, MA, U.S.A.), and the genes were subcloned into Baculovirus expression vector pBlueBac4 (Invitrogen, San Diego, CA, U.S.A.). The baculovirus supernatants containing ∼20 μg/ml hedgehog protein were diluted 1:40 (v/v) in the assay. Confluent cells were adapted to serum-free media 24 h before either BMP-2 or hedgehog protein was added. The experiment was stopped with lysis buffer (4.5 M guanidin hydrochloride, 50 mM Tris-HCl, 30% TritonX-100 [w/v], pH 6.6) after 72 h. The cell lysate was stored at –80°C until RNA extraction. For antisense experiments, the serum-free cell cultures were treated 6 h prior to BMP-2 addition with 2 μM phosphorothioate oligonucletides (sox9-antisense 5′ agg aga ttc ata cgc ggg cc; sox9-sense 5′ ggc ccg cgt atg aat ctc ct), which had been designed to bind to the translation initiation site.(28)

RNA isolation

RNA was isolated using the Boehringer Mannheim High Pure RNA Isolation Kit. RNA quantity was determined by absorption at 260 nm, and quality was checked in an agarose gel.

Quantitative reverse transcriptase-polymerase chain reaction

Sox9 and actin mRNAs were quantitated via competitive reverse transcriptase-polymerase chain reaction (RT-PCR) using a multigene standard.(29) The expression of the BMP type I and type II receptors was examined by RT-PCR. The following primers for standard and target amplification were used. Each primer combination was designed for and tested on human and murine sequences.

Sox9-F 5′ atc tga aga agg aga gcg ag

Sox9-R 5′ tca gaa gtc tcc aga gct tg

actin-F 5′ aca cct tct aca atg agc tgc g

actin-R 5′ cgc tcg gtg agg atc ttc atg

Alk3-F 5′ gcg aac tat tgc caa aca g

Alk3-R 5′ gag gtg gca cag acc aca ag

Alk6-F 5′ gac act ccc att cct cat c

Alk6-R 5′ gct ata gtc ctt tgg acc ag

BMPRII-F 5′ tca aga acg gct gtg tgc at

BMPRII-R 5′ cgc tca tcc aag gaa cct tt

ActRII-F 5′ tct ctt gct ctt cag gtg ct

ActRII-R 5′ gaa caa gta cag gag ggt ag

ActRIIB-F 5′ ctg ctg gct aga tga ctt c

ActRIIB-R 5′ ctt cca cgt gat gat gtt cc

After linearization, the standard vector was transcribed with T7 RNA polymerase and quantitated at 260 nm after DNAse I treatment. Cellular RNA, 0.5 μg, and different concentrations of standard cRNA were combined and reverse transcribed. PCR was performed using ExpandTM High Fidelity PCR System in a GeneAmp 9600 thermocycler (Perkin-Elmer, Norwalk, CT, U.S.A.). The regime for PCR was one cycle: 94°C 3 minutes; 58°C 1 minutes; 72°C 2 minutes followed by 45 cycles of: 94°C 20 s; 58°C 20 s; 72°C 1 minute and completed with 5 minutes 72°C. PCR products were analyzed by ethidium bromide staining after gel electrophoresis.

To quantitate the PCR products, resulting from competitive RT-PCR, the intensity of the bands in the gel was measured with the gel imaging system E.A.S.Y. (Herolab GmbH, Wiesloch, Germany). The relative expression level of Sox9 was calculated as follows:

[(standard intensity, control) × (Sox9 intensity, induced)]/[(standard intensity, induced) × (Sox9 intensity control)]

The relative expression level of actin was used to normalize the Sox9 level in each sample. Relative expression levels of four independent cell culture experiments were used for means ± SD.

Northern blot analysis

Poly(A)-tailed RNA was isolated from total RNA samples by immobilization of mRNA on streptavidin magnetic particles. For each sample, 1 μg of mRNA per lane was fractionated on a 1% agarose-formaldehyde gel. Subsequent to electrophoresis, RNA was transferred onto a positively charged nylon membrane and hybridized with a Digoxigenin-labeled Sox9 or Aggrecan (aggrecan-F 5′ tcc cca aat ccc tca tac tc; aggrecan-R 5′ atc acc aca cag tcc tct cc) PCR product. For Col2a1, ColX, ColI, and osteocalcin detection a Digoxigenin-labeled RNA probe transcribed from pBSK-m-collagenII, pBSKS-m-collagenI(30) pMT7T3-f1-m-osteocalcin, or pBS-m-ColX,(31) kindly provided by K. von der Mark, was used.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We analyzed the involvement of Sox9 in BMP-2 signaling by measuring its expression in femur-derived human bone cells (HBCs) as well as in MG63, HOS, and C3H10T1/2 cells. A synthetic vector containing multiple primer binding sites was used as an internal standard for monitoring Sox9 and actin expression by quantitative RT-PCR after BMP-2 treatment. After separation of the standard and Sox9 mRNA signal by agarose gel electrophoresis, Sox9 expression with reference to the internal PCR control can be seen in the cell samples without and with BMP-2 treatment (Fig. 1A). Expression levels were quantitated by densitometric measurement of the PCR products and normalization against PCR standard and actin. Sox9 expression increased nearly 4-fold in C3H10T1/2 cells after treatment with 200 ng/ml BMP-2 (Fig. 1B), this up-regulation is dose-dependent (data not shown). In MG63 cells, considered to be early osteoblast-like,(32) and in HBCs, Sox9 expression was increased nearly 3-fold by BMP-2. In contrast to this, HOS cells, which are thought to be more advanced in the osteoblastic lineage(32) did not respond with an increase in Sox9 mRNA to BMP-2 treatment (Fig. 1B).

thumbnail image

Figure FIG. 1. Sox9 expression in C3H10T1/2 cells, in two osteosarcoma cell lines (MG63 and HOS) and in human bone cells (HBCs) after BMP-2 induction. (A) Three percent agarose gels of competitive RT-PCR products after 72 h stimulation without (a) and with (b) 200 ng/ml BMP-2. mRNA signal of Sox9 expression at 263 bp, vector signal of internal standard at 150 bp. (B) Relative expression levels of Sox9 in the BMP-2–induced cell lines C3H10T1/2, MG63, HOS and in human bone cells (HBCs). The relative expression levels were calculated (see Materials and Methods) after densitometric measurement of the PCR products, representing the standard and the Sox9 mRNA. For normalization, the expression level of actin was used. Data are the means ± SD for four independent cell culture experiments, whereas HBCs from four different donors were used. The relative expression levels refer to a noninduced expression level of one.

Download figure to PowerPoint

To verify the results obtained by quantitative RT-PCR, Sox9 expression was also examined in the different cell lines by Northern blot using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as control. As expected, the Northern blot showed increased expression of Sox9 in C3H10T1/2 and MG63 cells after BMP-2 treatment, whereas no change in expression could be detected in HOS cells (Fig. 2).

thumbnail image

Figure FIG. 2. Sox9 expression in three different cell lines after BMP-2 induction for 72 h. Northern blot detection of Sox9 mRNA in C3H10T1/2, MG63, and HOS cells after (b) induction with 200 ng/ml BMP-2 compared with the (a) serum-free control. GAPDH expression is shown to indicate loading variations.

Download figure to PowerPoint

The results obtained with MG63 and HOS cells indicate that the ability of cells to respond to BMP-2 treatment with Sox9 expression might be dependent upon their differentiation status. One reason for this difference in sensitivity toward BMP-2 could be the presence or absence of different BMP receptors. To test this assumption, we examined the expression of the two known BMP type I receptors (Alk3 and Alk6), which are responsible for signal transduction(33) and of three type II receptors (BMPR-II, ActR-II, and ActR-IIB) known to bind BMP-2.(34) Each type II receptor was expressed in all three cell lines, but only in C3H10T1/2 and MG63 cells this expression seemed to be up-regulated by BMP-2 (Fig. 3). Interestingly, Alk3 mRNA could be detected in all three cell types, whereas Alk6 mRNA was present only in HOS cells (Fig. 3), indicating that this receptor type might have a role in the different response to BMP-2. Following expression analysis of Sox9 as a downstream target of BMP-2, we intended to test the response of Sox9 expression to one or more factors which are supposed to be upstream of BMP-2. Hedgehog proteins are thought to be such upstream regulators of BMP-2,(22) and therefore we examined their influence on Sox9 expression. Baculovirus supernatants containing SHH or IHH were added to C3H10T1/2 and to HBCs, obtained from four different donors. Both hedgehog proteins were able to increase Sox9 mRNA levels in both types of cells (Fig. 4), but the induction was only about 2-fold, whereas the BMP-2 stimulation of Sox9 mRNA had been nearly 3-fold in HBC and nearly 4-fold in C3H10T1/2 cells.

thumbnail image

Figure FIG. 3. Expression of BMP type I and type II receptors. RT-PCR products of Alk3 (BMPR-IA), Alk6 (BMPR-IB), BMPR-II, ActR-II, and ActR-IIB in three different cell lines: C3H10T1/2, MG63, and HOS. (a) RT-PCR product of noninduced cells and (b) of the cell culture sample induced with 200 ng/ml BMP-2 for 72 h. Actin was amplified as a control for total RNA concentration used in the RT-PCR.

Download figure to PowerPoint

thumbnail image

Figure FIG. 4. Sox9 expression in C3H10T1/2 cells and in HBCs after induction with SHH, IHH, and BMP-2. (A) Three percent agarose gel of competitive RT-PCR products after 72 h of stimulation without inductor (a), with SHH (b), IHH (c), and 200 ng/ml BMP-2 (d). mRNA signal of Sox9 at 263 bp, vector signal of internal standard at 150 bp. Hedgehog proteins were expressed baculovirus-mediated in insect cells, the supernatant was used. (B) Relative expression levels of Sox9 in C3H10T1/2 cells and in human bone cells induced by SHH (open bars), IHH (hatched bars) and BMP-2 (filled bars). After 72 h of treatment, Sox9 expression was measured by quantitative RT-PCR (see Material and Methods); the expression level of the noninduced sample is set to one. Data are the means ± SD for four independent cell culture experiments, whereas HBCs were originating from four different donors.

Download figure to PowerPoint

After identification of Sox9 as a BMP-2 and hedgehog response factor, we wanted to examine more closely downstream signaling of Sox9. Since Sox9 is known to be a regulator of the chondrocyte matrix protein Col2a1, we analyzed the expression of Col2a1 by Northern analysis. The experiment revealed that C3H10T1/2 cells express a Col2a1 transcript of the expected size of 5.4 kb, whereas the two osteosarcoma cells express only an 800-bp mRNA which corresponds in size to chondrocalcin mRNA (Fig. 5). The Col2a1 probe used was complementary to mouse and human pro-alpha (II) collagen and can also detect the 800 bp chondrocalcin mRNA, whose sequence is included within the type II collagen gene.(35) A significant up-regulation of Col2a1 mRNA by BMP-2 was detected in C3H10T1/2 as expected, whereas both osteosarcoma cell lines showed no Col2a1 signal and a decrease of the chondrocalcin transcript.

thumbnail image

Figure FIG. 5. Northern blot analysis of Col2a1 expression in three different cell lines after an induction period of 72 h. GAPDH expression is shown to indicate loading variations. Serum-free control (a); induced (b) with 200 ng/ml BMP-2.

Download figure to PowerPoint

A time course study in C3H10T1/2 cells allowed us to investigate more closely BMP-2 signaling by monitoring cartilage and osteoblast marker expression. Northern analysis showed the increase of Col2a1 expression after 24, 48, and 72 h of BMP-2 stimulation. Aggrecan and Col X were only slightly expressed after 24 h; aggrecan reached its highest level after 72 h and Col X after 48 h. Collagen I mRNA was decreased by BMP-2 at all induction times monitored, whereas osteocalcin expression started after 48 h and was clearly visible after 72 h of BMP-2 stimulation (Fig. 6). Northern analysis to test if MG63 cells express other chondrocyte markers besides Sox9 did not detect aggrecan and Col X mRNA (data not shown).

thumbnail image

Figure FIG. 6. Time course of the expression of cartilage (Col 2a1, Aggrecan, Col X) and osteoblast markers (Col I, osteocalcin) in noninduced (a) and in BMP-2 induced (b) C3H10T1/2 cells. The expression of the marker genes was monitored by Northern blotting after 24, 48, and 72 h. GAPDH expression is shown to indicate loading variations.

Download figure to PowerPoint

To unravel further the role of Sox9 in BMP-2 signaling, cells were treated with antisense Sox9 oligonucleotides to inhibit Sox9 mRNA translation. As expected, Col2a1 up-regulation by BMP-2 was decreased. Moreover, this experiment revealed that aggrecan and Col X expression is also influenced by Sox9. Both mRNAs are increased by BMP-2 stimulation, whereas Sox9 antisense treatment diminishes this effect. While the expression of osteocalcin seemed not to be influenced, the decrease in collagen I expression upon BMP-2 stimulation was reduced by Sox9 antisense treatment (Fig. 7).

thumbnail image

Figure FIG. 7. Influence of antisense-Sox9-oligonucleotide on the expression of cartilage (Col2a1, Aggrecan, Col X) and osteoblast markers (Col I, osteocalcin) in BMP-2–treated C3H10T1/2 cells. Expression of the marker genes was measured by Northern blotting in noninduced (a), in BMP-2–treated (b), in Sox9-sense-oligonucleotide and BMP-2–treated (c) and in Sox9-antisense-oligonucleotide and BMP-2–treated (d) C3H10T1/2 cells after 48 h treatment. Oligonucleotides were added 6 h prior to BMP-2. GAPDH expression is shown to indicate loading variations.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  7. Acknowledgements

We investigated the involvement of Sox9 in BMP-2 signaling by testing different cells related to the osteoblastic lineage. BMP-2 treatment of the cell lines C3H10T1/2 and MG63 results in an increase of Sox9 expression, whereas HOS cells do not respond. The differentiation status of each cell line could be a reason for these different responses.

C3H10T1/2 cells are considered as pluripotent mesenchymal progenitors because of their potential to differentiate into myoblasts, adipocytes, chondrocytes, or osteoblasts in vitro.(20) Since C3H10T1/2 cells are known to differentiate toward the osteoblastic and chondroblastic lineage after BMP-2 stimulation,(18–21) we consider this cell line as a model for cells early in the endochondral lineage. The human osteosarcoma cell line MG63 shows a number of features typical for undifferentiated committed osteoblasts(32) (e.g., low basal alkaline phosphatase expression), whereas HOS cells are considered to be more advanced in the osteoblastic lineage.(32)

A possible explanation for why the two less-differentiated cell lines increase their Sox9 expression after BMP-2 treatment, whereas the cells more advanced in the osteoblastic lineage do not respond, could be different expression of BMP receptors. Signaling of BMPs involves two types of receptors, type I and type II, with serine/threonine kinase activity. Type I receptors transduce the signal after they are phosphorylated by type II receptors.(33) Two type I BMP receptors have been identified in vertebrates, BMPR-IA (Alk3) and BMPR-IB (Alk6). BMP-2, BMP-4, and BMP-7 bind to a combination of either Alk3 or Alk6 with a type II receptor.(36) Alk3 mRNA is nearly ubiquitously expressed in mouse development, including mesenchymal condensations in the developing limb bud. The expression of Alk6 mRNA is more localized in all cartilage condensations, and bone tissues express more Alk6 than Alk3.(37,38) Alk6 is induced by Alk3, suggesting a hierachy of receptor action.(39) The type II receptor for BMPs (BMPR-II) is expressed in many different tissues.(34) In addition to BMPR-II, which is specific for BMPs, type II receptors for activin (ActR-II and ActR-IIB) bind certain members of the BMP family, e.g., BMP-2.(34) Using RT-PCR, we tested if the cells used differ in their BMP-receptor expression. Our results demonstrate that only HOS cells express type IB BMP receptor (Alk6) mRNA, whereas all other BMP receptors were expressed in all cell lines tested. This BMP receptor might therefore be indicative for a differentiation status without the potential for up-regulation of chondrocyte-specific factors, like Sox9, in response to BMP-2. It has already been described for TGF-β that the ratio of type II to type I receptors could influence how signaling is perceived in bone cells.(40)

Alk6 might therefore also have a major role in the commitment of osteoblasts, but the relationship between this type I BMP receptor and Sox9 expression remains unclear and has to be analyzed further.

Besides BMP receptors, a difference in expression of (a) cofactor(s) could also be an explanation for changing responses to BMP-2 upon maturation of the cells. Correspondingly, several signaling molecules such as Wnt5a, Wnt7a, and the fibroblast growth factors have been proposed as modifiers of BMP function,(41) but their involvement in Sox9 signaling is still under investigation.

Since immortal cell lines often do not represent cells in vivo, we also wanted to test the inducibility of human osteoblasts directly derived from femoral bone of hip transplantation patients. These experiments were only performed by quantitative RT-PCR since the relatively large RNA quantity necessary for Northern blotting would have been available only after prolonged culture with the risk of losing the primary characteristics of these cells. We found that Sox9 expression is also induced in these primary HBCs by BMP-2. Hence, it seems that this regulatory effect could also be functional in the adult human organism. It could be argued that chondrocytes are present after establishing the primary human bone culture, which could be a reason why Sox9 expression and response is found. However, it is unlikely that cells other than osteoblasts were cultured since only bone spongiosa was dissected, and prior to cell collection the bone pieces were washed two times and digested twice with collagenase A.

It has been reported that hedgehog proteins, as well as BMP-2, induce alkaline phosphatase expression in C3H10T1/2, indicating a role in osteoblastic differentiation.(42,43) The expression of the BMP-2 homolog dpp is also regulated by the segment polarity gene hedgehog in imaginal discs of Drosophila.(41) The mammalian transcripts of hedgehog are often found in the embryo adjacent to BMP expression, and in the chick limb bud ectopic SHH locally induces BMP-2 and BMP-4.(44) Since hedgehog is an upstream regulator of BMP-2, we investigated the influence of hedgehog proteins on the expression of Sox9. SHH, as well as IHH, up-regulated Sox9 expression in C3H10T1/2 and HBCs. This increase was not as high as with BMP-2, which could be due to a lower concentration of active hedgehog protein in the baculovirus supernatant.

BMP-2 is expressed during mesenchyme condensation in the limb,(45) also suggesting a role in cartilage initiation. In an ectopic bone formation assay, BMP-2 has been shown to first induce cartilage, which is then transformed into bone.(12,15) Although various effects of BMPs on chondrocytes and osteoblasts have been reported,(12–13,46,47) it remains unclear how they regulate development and function in these cells. As shown recently, Sox9 expression correlates with expression of Col2a1 in chondrocytes,(10) and the minimal Col2a1 enhancer is a direct target for Sox9.(9) Type II collagen is the most abundant protein in the extracellular matrix of chondrocytes. Since we already knew that BMP-2 up-regulates Sox9 transcription, we tested the downstream response by examining Col2a1 expression. We saw the expected up-regulation of the 5.4 kb Col2a1 mRNA in C3H10T1/2 cells after BMP-2 treatment. In contrast, MG63 and HOS cells did not express the Col2a1 transcript, demonstrating their osteoblastic phenotype. However, we could detect chondrocalcin mRNA, which is encoded within the type II collagen gene.(30) Chondrocalcin is the C-propeptide of type II procollagen(48) and is thought to be associated with cartilage calcification in endochondral bone formation.(35) This demonstrates the ability of BMP-2 to up-regulate the chondrogenic marker Col2a1 in progenitor cells, whereas in osteoblastic cells it causes down-regulation of chondrocalcin, which is normally found in calcifying chondrocytes. This correlates well with a report that BMP-2 causes expression of cartilage markers in skeletal progenitor cells derived from mouse limb buds, whereas these markers are down-regulated by BMP-2 in cells already committed to the endochondral pathway.(49)

Since MG63 cells showed inducibility of Sox9 expression, we tested whether these cells express further chondrocyte markers after 72 h of BMP-2 treatment. Via Northern blotting, neither aggrecan nor Col X mRNA could be detected confirming the osteoblastic phenotype of this cell line. Interestingly, Sox9 but not Col2a1 mRNA was up-regulated in MG63 cells. Recent results indicate a need for other chondrocyte-specific nuclear protein(s) in addition to Sox9 for Col2a1 transcription, since the enhancer of the Col2a1 gene contains multiple elements essential for chondrocyte-specific expression.(50) The absence of these coactivators in the osteoblastic MG63 cells could be a reason for not allowing Col2a1 up-regulation by Sox9 in these cells. Moreover, the differential expression of Sox9 and Col2a1 in nonchondrogenic tissues(51) activation of Col2a1 in some but not all areas of ectopic Sox9 expression(5) and the requirement of other HMG proteins for tissue-specific cofactors(52) strongly support the necessity of cooperating factors only present in chondrocytes, and in the C3H10T1/2 cells used in this study, for Col2a1 transcription.

Time course experiments showed that first Col2a1 is up-regulated by BMP-2 in C3H10T1/2 cells, followed by aggrecan and Col X. While an osteocalcin signal was already present at 48 h, significant up-regulation could only be detected after 72 h, whereas Col I was down-regulated by BMP-2 at each time tested. As in our experiments, in limb development type II collagen mRNA is also the first marker gene detected in the endochondral lineage, already expressed during the condensation of mesenchymal prechondrogenic cells.(53) The mRNA level of aggrecan, the most abundant proteoglycan in cartilage,(54) was increasing with each day of BMP-2 treatment. Col X, a marker for hypertrophic chondrocytes preceding cartilage calcification, was highly transcribed after 48 h, whereas its concentration seemed to decline at 72 h. Col I is known to be expressed in osteoblasts and fibroblasts but not in chondrocytes.(55) Therefore, the down-regulation of Col I expression is demonstrating the chondrocytic phenotype of C3H10T1/2, supported by the expression of Col2a1, Col X, and aggrecan. Osteocalcin, a noncollagenous bone matrix protein, is expressed in mature osteoblasts but its expression also has been reported in hypertrophic chondrocytes prior to mineralization.(56)

In conclusion, C3H10T1/2 cells in our system express a phenotype of hypertrophic chondrocytes switching to calcification. It would be interesting to monitor even longer periods of BMP-2 stimulation to find out if differentiation toward osteoblasts will occur, but C3H10T1/2 cells can only be maintained serum-free for 72 h. However, already after 72 h, the decline of Col X, the expression of osteocalcin, and the reduced Col I down-regulation suggest beginning osteogenesis.

In C3H10T1/2 cells we could show that inhibiting Sox9 mRNA translation by antisense oligonucleotides interferes with the up-regulation of Col2a1 expression as expected. Whereas clear effects could be seen after antisense Sox9 oligonucleotide treatment, a complete inhibition of the target mRNA could not be expected with this method.

Besides Col2a1, the increase in aggrecan and Col X expression was also reduced by Sox9 antisense oligonucleotides. Above that, Sox9 mRNA turned out not only to be responsible for the expression of chondrocyte markers but also to influence transcription of a known osteoblast marker, Col I. Whereas Col I expression was down-regulated by BMP-2 treatment, blocking Sox9 mRNA inhibited this effect. Our results strongly suggest that Sox9 is mediating BMP-2 up-regulation of Col X and aggrecan in addition to Col2a1 expression.

We demonstrate, for the first time, that Sox9 is not only controlling Col2a1 but also Col X and aggrecan transcription in cells with chondrocyte properties. It is already known from in situ studies in mouse embryos that Sox9 overlaps Col2a1 expression in stages from mesenchymal stem cell condensations to chondrocytic cells, but both Col2a1 and Sox9 transcripts are undetectable in hypertrophic cartilage when chondrogenesis is complete and osteoblasts appear.(6,10) In contrast to this, we detected Sox9 expression in osteosarcoma cell lines as well as in bone cells directly isolated from human tissue and saw a regulatory impact of Sox9 on the osteoblast marker Col I. Moreover, our findings that BMP-2 enhances Sox9 expression in the early osteoblastic cell lines and in HBCs suggest that the transcription factor Sox9 might also play a role in later stages of endochondral bone formation.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We thank U. Leser, S. Lingke, and C. Groenlinger for cell culture support, G. Proetzel for reading the manuscript, and all our colleagues in the bone group for helpful discussion.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • 1
    Pevny LH, Lovell-Badge R 1997 Sox genes find their feet Curr Opin Genet Dev 7:338344.
  • 2
    Prior HM, Walter MA 1996 Sox genes: Architects of development Mol Med 2:405412.
  • 3
    Sudbeck P, Schmitz ML, Baeuerle PA, Scherer G 1996 Sex reversal by loss of the C-terminal transactivation domain of human Sox9 Nat Genet 13:230232.
  • 4
    Wotton D, Lake RA, Farr CJ, Owen MJ 1995 The high mobility group transcription factor, Sox4, transactivates the human CD2 enhancer J Biol Chem 270:75157522.
  • 5
    Bell DM, Leung KK, Wheatley SC, Ng LJ, Zhou S, Ling KW, Sham MH, Koopman P, Tam PP, Cheah KS 1997 Sox9 directly regulates the type-II collagen gene Nat Genet 16:174178.
  • 6
    Wright E, Hargrave MR, Christiansen J, Cooper L, Kun J, Evans T, Gangadharan U, Greenfield A, Koopman P 1995 The Sry-related gene Sox9 is expressed during chondrogenesis in mouse embryos Nat Genet 9:1520.
  • 7
    Wagner T, Wirth J, Meyer J, Zabel B, Held M, Zimmer J, Pasantes J, Bricarelli FD, Keutel J, Hustert E, Wolf U, Tommerup N, Schempp W, Scherer G 1994 Autosomal sex reversal and campomelic dysplasia are caused by mutations in and around the SRY-related gene Sox9 Cell 79:11111120.
  • 8
    Cameron FJ, Hageman RM, Cooke-Yarborough C, Kwok C, Goodwin LL, Sillence DO, Sinclair AH 1996 A novel germ line mutation in Sox9 causes familial campomelic dysplasia and sex reversal Hum Mol Genet 5:16251630.
  • 9
    Lefebvre V, Huang W, Harley VR, Goodfellow PN, De Crombrugghe B 1997 Sox9 is a potent activator of the chondrocyte-specific enhancer of the pro alpha1 (II) collagen gene Mol Cell Biol 17:23362346.
  • 10
    Zhao Q, Eberspaecher H, Lefebvre V, De Crombrugghe B 1997 Parallel expression of Sox9 and Col2a1 in cells undergoing chondrogenesis Dev Dyn 209:377386.
  • 11
    Wang EA, Rosen V, Cordes P, Hewick RM, Kriz MJ, Luxenberg DP, Sibley BS, Wozney JM 1988 Purification and characterization of other distinct bone-inducing factors Proc Natl Acad Sci USA 85:94849488.
  • 12
    Duprez DM, Coltey M, Amthor H, Brickell PM, Tickle C 1996 Bone morphogenetic protein-2 (BMP-2) inhibits muscle development and promotes cartilage formation in chick limb bud cultures Dev Biol 174:448452.
  • 13
    Enomoto-Iwamoto M, Iwamoto M, Mukudai Y, Kawakami Y, Nohno T, Higuchi Y, Takemoto S, Ohuchi H, Noji S, Kurisu K 1998 Bone morphogenetic protein signaling is required for maintenance of differentiated phenotype, control of proliferation, and hypertrophy in chondrocytes J Cell Biol 140:409418.
  • 14
    Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW, Hewick RM, Wang EA 1988 Novel regulators of bone formation: Molecular clones and activities Science 242:15281534.
  • 15
    Wang EA, Rosen V, D'Alessandro JS, Bauduy M, Cordes P, Harada T, Israel DI, Hewick RM, Kerns KM, LaPan P, Luxenberg DP, McQuaid D, Moutsatsos IK, Nove J, Wozney JM 1990 Recombinant human bone morphogenetic protein induces bone formation Proc Natl Acad Sci USA 87:22202224.
  • 16
    Lyons KM, Pelton RW, Hogan BL 1990 Organogenesis and pattern formation in the mouse: RNA distribution patterns suggest a role for bone morphogenetic protein-2a (BMP-2a) Development 109:833844.
  • 17
    Lyons KM, Pelton RW, Hogan BL 1989 Patterns of expression of murine Vgr-1 and BMP-2a mRNA suggest that transforming growth factor-β-like genes coordinately regulate aspects of embryonic development Genes Dev 3:16571658.
  • 18
    Wang EA, Israel DI, Kelly S, Luxenberg DP 1993 Bone morphogenetic protein-2 causes commitment and differentiation in C3H10T1/2 and 3T3 cells Growth Factors 9:5771.
  • 19
    Yamaguchi A, Katagiri T, Ikeda T, Wozney JM, Rosen V, Wang EA, Kahn AJ, Suda T, Yoshiki S 1991 Recombinant human bone morphogenetic protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation in vitro J Cell Biol 113:681687.
  • 20
    Katagiri T, Yamaguchi A, Ikeda T, Yoshiki S, Wozney JM, Rosen V, Wang EA, Tanaka H, Omura S, Suda T 1990 The non-osteogenic mouse pluripotent cell line, C3H10T1/2, is induced to differentiate into osteoblastic cells by recombinant human bone morphogenetic protein-2 Biochem Biophys Res Commun 172:295299.
  • 21
    Atkinson BL, Fantle KS, Benedict JJ, Huffer WE, Gutierrez-Hartmann A 1997 Combination of osteoinductive bone proteins differentiates mesenchymal C3H10T1/2 cells specifically to the cartilage lineage J Cell Biochem 65:325339.
  • 22
    Laufer E, Nelson CE, Johnson RL, Margan BA, Tabin C 1994 Sonic hedgehog and Fgf-4 act through a signaling cascade and feedback loop to integrate growth and patterning of the developing limb bud Cell 79:9931003.
  • 23
    Fietz MJ, Concordet JP, Barbosa R, Johnson R, Krauss S, McMahon AP, Tabin C, Ingham PW 1994 The hedgehog gene family in Drosophila and vertebrate development. Dev Suppl 4351.
  • 24
    Kinto N, Iwamoto M, Enomoto-Iwamoto M, Noji S, Ohuchi H, Yoshioka H, Kataoka H, Wada Y, Yuhao G, Takahashi HE, Yoshiki S, Yamaguchi A 1997 Fibroblasts expressing Sonic hedgehog induce osteoblast differentiation and ectopic bone formation FEBS Lett 404:319323.
  • 25
    Vortkamp A, Lee K, Lanske B, Segre GV, Kronenberg HM, Tabin CJ 1996 Regulation of rate of cartilage differentiation by Indian hedgehog and PTH-related protein Science 273:613622.
  • 26
    Lanske B, Karaplis AC, Lee K, Luz A, Vortkamp A, Pirro A, Karperien M, Defize LHK, Ho C, Mulligan RC, Abou-Samra AB, Jüppner H, Segre GV, Kronenberg HM 1996 PTH/PTHrP receptor in early development and Indian hedgehog-regulated bone growth Science 273:663666.
  • 27
    Kronenberg HM, Lee K, Lanske B, Segre GV 1997 Parathyroid hormone-related protein and Indian hedgehog control the pace of cartilage differentiation J Endocrinol 154:S39S45.
  • 28
    Brysch W, Schlingensiepen KH 1994 Design and application of antisense oligonucleotides in cell culture, in vivo, and as therapeutic agents Cell Mol Neurobiol 14:557568.
  • 29
    Gilliland G, Perrin S, Blanchard K, Bunn HF 1990 Analysis of cytokine mRNA and DNA: Detection and quantitation by competitive polymerase chain reaction Proc Natl Acad Sci USA 87:27252729.
  • 30
    Metsaranta M, Toman D, De Crombrugghe B, Vuorio E 1991 Specific hybridization probes for mouse type I, II III and IX collagen mRNAs Biochim Biophys Acta 1089:241243.
  • 31
    Apte SS, Seldin MF, Hayashi M, Olsen BR 1992 Cloning of the human and mouse type X collagen genes and mapping of the mouse type X collagen gene to chromosome 10 Eur J Biochem 206:217224.
  • 32
    Clover J, Gowen M 1994 Are MG-63 and HOS TE85 human osteosarcoma cell lines representative models of the osteoblastic phenotype? Bone 15:585591.
  • 33
    Reddi AH 1998 Role of morphogenetic proteins in skeletal tissue engineering and regeneration Nat Biotechnol 16:247252.
  • 34
    Kawabata M, Imamura T, Miyazono K 1998 Signal transduction by bone morphogenetic proteins Cytokine Growth Factor Rev 9:4961.
  • 35
    Poole AR, Pidoux I, Reiner A, Choi H, Rosenberg LC 1984 Association of an extracellular protein (chondrocalcin) with the calcification of cartilage in endochondral bone formation J Cell Biol 98:5465.
  • 36
    Massague J, Attisano L, Wrana JL 1994 The TGF-β family and its composite receptors Trends Cell Biol 4:172178.
  • 37
    Ishidou Y, Kitajima I, Obama H, Maruyama I, Murata F, Imamura T, Yamada N, Ten Dijke P, Miyazono K, Sakou T 1995 Enhanced expression of type I receptors for Bone morphogenetic proteins during bone formation J Bone Miner Res 10:16511659.
  • 38
    Dewulf N, Verschueren K, Lonnoy O, Moren A, Grimsby S, Vande Spiegle K, Miyazono K, Huylebroeck D, Ten Dijke P 1995 Distinct spatial and temporal expression patterns of two type I receptors for bone morphogenetic proteins during mouse embryogenesis Endocrinology 136:26522663.
  • 39
    Zou H, Wieser R, Massague J, Niswander L 1997 Distinct roles of type I bone morphogenetic protein receptors in the formation and differentiation of cartilage Genes Dev 11:21912203.
  • 40
    Centrella M, Casinghino S, Kim J, Pham T, Rosen V, Wozney J, McCarthy TL 1995 Independent changes in Type I and Type II receptors for transforming growth factor β induced by bone morphogenetic protein 2 parallel expression of the osteolast phenotype Mol Cell Biol 15:32733281.
  • 41
    Hogan BL 1996 Bone morphogenetic proteins: Multifunctional regulators of vertebrate development Genes Dev 10:15801594.
  • 42
    Nakamura T, Aikawa T, Iwamoto-Enomoto M, Iwamoto M, Higuchi Y, Pacifici M, Kinto N, Yamaguchi A, Noji S, Kurisu K, Matsuya T 1997 Induction of osteogenic differentiation by hedgehog proteins Biochem Biophys Res Commun 237:465469.
  • 43
    Takuwa Y, Ohse C, Wang EA, Wozney JM, Yamashita K 1991 Bone morphogenetic protein-2 stimulates alkaline phosphatase activity and collagen synthesis in cultured osteoblastic cells, MC3T3–E1 Biochem Biophys Res Commun 174:96101.
  • 44
    Francis PH, Richardson MK, Brickell PM, Tickle C 1994 Bone morphogenetic proteins and a signalling pathway that controls patterning in the developing chick limb Development 120:209218.
  • 45
    Kingsley DM 1994 What do BMPs do in mammals? Clues from the mouse short-ear mutation Trends Genet 10:1621.
  • 46
    Reddi AH 1994 Bone and cartilage differentiation Curr Opin Genet Dev 4:737744.
  • 47
    Chen P, Carrington JL, Hammonds RG, Reddi AH 1991 Stimulation of chondrogenesis in limb bud mesoderm cells by recombinant human bone morphogenetic protein 2B (BMP-2B) and modulation by transforming growth factor beta 1 and beta 2 Exp Cell Res 195:509515.
  • 48
    van der Rest M, Rosenberg LC, Olsen BR, Poole AR 1986 Chondrocalcin is identical with the C-propeptide of type II procollagen Biochem J 237:923925.
  • 49
    Rosen V, Nove J, Song JJ, Thies S, Cox K, Wozney JM 1994 Responsiveness of clonal limb bud cell lines to bone morphogenetic protein 2 reveals a sequential relationship between cartilage and bone cell phenotypes J Bone Miner 9:17591768.
  • 50
    Zhou G, Lefebvre V, Zhang Z, Eberspaecher H, de Crombrugghe B 1998 Three high mobility group-like sequences within a 48-base pair enhancer of the Col2a1 gene are required for cartilage-specific expression in vivo J Biol Chem 273:1498914997.
  • 51
    Ng LJ, Wheatley S, Muscat GE, Conway-Campbell J, Bowles J, Wright E, Bell DM, Tam PP, Cheah KS, Koopman P 1997 Sox9 binds DNA, activates transcription, and coexpresses with type II collagen during chondrogenesis in the mouse Dev Biol 183:108121.
  • 52
    Yuan H, Corbi N, Basilico C, Dailey L 1995 Developmental-specific activity of the FGF-4 enhancer requires synergistic action of Sox2 and October-3 Genes Dev 9:26352645.
  • 53
    Kosher RA, Kulyk WM, Gay SW 1986 Collagen gene expression during limb cartilage differentiation J Cell Biol 102:11511156.
  • 54
    Hardingham TE, Fosang AJ, Dudhia J 1994 The structure, function and turnover of aggrecan, the large aggregating proteoglycan from cartilage Eur J Clin Chem Clin Biochem 32:249257.
  • 55
    Sandberg M, Vuorio E 1987 Localization of types I, II, and III collagen mRNAs in developing human skeletal tissues by in situ hybridization J Cell Biol 104:10771084.
  • 56
    Neugebauer BM, Moore MA, Broess M, Gerstenfeld LC, Hauschka PV 1995 Characterization of structural sequences in the chicken osteocalcin gene: Expression of osteocalcin by maturing osteoblasts and by hypertrophic chondrocytes in vitro J Bone Miner Res 10:157163.