Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (27)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kolettas, E.
Right arrow Articles by Hardingham, T. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kolettas, E.
Right arrow Articles by Hardingham, T. E.
Related Collections
Right arrow Experimental Arthritis
Right arrow Immunogenetics
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Rheumatology 2001; 40: 1146-1156
© 2001 British Society for Rheumatology

Original Papers

Chondrocyte phenotype and cell survival are regulated by culture conditions and by specific cytokines through the expression of Sox-9 transcription factor

E. Kolettas, H. I. Muir, J. C. Barrett1 and T. E. Hardingham

Wellcome Trust Centre for Cell-Matrix Research, School of Biological Sciences, University of Manchester, Manchester M13 9PT, UK and
1 Laboratory of Molecular Carcinogenesis, National Institute of Environmental Health Sciences, NIH, PO Box 12233, Research Triangle Park, NC 27709, USA


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Objective. To investigate the effects of culture conditions, serum and specific cytokines such as insulin-like growth factor (IGF) 1 and interleukin (IL) 1{alpha} on phenotype and cell survival in cultures of Syrian hamster embryonic chondrocyte-like cells (DES4+.2).

Methods. Proteins and RNA extracted from subconfluent and confluent early- and late-passage DES4+.2 cells cultured in the presence or absence of serum and IL-1{alpha} or IGF-1 or both cytokines together were analysed for the expression of chondrocyte-specific genes and for the chondrogenic transcription factor Sox-9 by Western and Northern blotting. Apoptosis was assessed by agarose gel electrophoresis of labelled low-molecular weight DNA extracted from DES4+.2 cells and another Syrian hamster embryonic chondrocyte-like cell line, 10W+.1, cultured under the different conditions and treatments.

Results. Early passage DES4+.2 cells expressed chondrocyte-specific molecules such as collagen types {alpha}1(II) and {alpha}1(IX), aggrecan, biglycan and link protein and collagen types {alpha}1(I) and {alpha}1(X) mRNAs, suggesting a prehypertrophic chondrocyte-like phenotype. The expression of all genes investigated was cell density- and serum-dependent and was low to undetectable in cell populations from later passages. Early-passage DES4+.2 and 10W+.1 cells survived when cultured at low cell density, but died by apoptosis when cultured at high cell density in the absence of serum or IGF-1. IGF-1 and IL-1{alpha} had opposite and antagonistic effects on the chondrocyte phenotype and survival. Whereas IL-1{alpha} acting alone suppressed cartilage-specific gene expression without significantly affecting cell survival, IGF-1 increased the steady-state mRNA levels and relieved the IL-1{alpha}-induced suppression of all the chondrocyte-specific genes investigated; it also enhanced chondrocyte survival. Suppression of the chondrocyte phenotype by the inflammatory cytokine IL-1{alpha} correlated with marked down-regulation of the transcription factor Sox-9, which was relieved by IGF-1. The expression of the Sox9 gene was closely correlated with the expression of the chondrocyte-specific genes under all conditions and treatments.

Conclusions. The results suggest that the effects of cartilage anabolic and catabolic cytokines IGF-1 and IL-1{alpha} on the expression of the chondrocyte phenotype are mediated by Sox-9. As Sox-9 appears to be essential for matrix production, the potent effect of IL-1{alpha} in suppressing Sox-9 expression may limit the ability of cartilage to repair during inflammatory joint diseases.

KEY WORDS: Chondrocytes, Gene expression, Cytokines, Apoptosis, Sox9.


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
During embryogenesis, selected mesenchymal cells condense and differentiate into chondrocytes, which express cell-specific products. Those cells within the growth plate subsequently undergo a programme of hypertrophy, calcification and cell death [1–3]. Various treatments of embryo cells have been shown to induce the differentiation of mesenchymal cells into chondrocytes [4, 5] or to select out mesenchymal cells that express chondrocyte-specific markers from mixed embryo populations [6–8]. These cell lines enable aspects of the regulation of chondrocyte gene expression and cell survival to be studied in cell culture.

Chondrocytes synthesize an elaborate extracellular matrix rich in collagen type II, aggrecan and link protein and smaller amounts of collagen types IX and XI. However, during the process of endochondral ossification, chondrocytes in the growth plate pass through a number of stages of differentiation before they cease to proliferate, become hypertrophic and start to synthesize type X collagen [9]. In culture, isolated primary mammalian chondrocytes steadily lose the expression of the cartilage-specific genes such as those encoding collagens and proteoglycans, although the slow progressive rate of decline suggests that cell shape alone is not a factor that determines the chondrocyte phenotype [10–16]. However, the loss of chondrocyte differentiated characteristics in vitro has been linked to the reduced expression of chondrocyte-specific transcription factors, principally Sox-9 [16, 17].

Sox-9 is a transcription factor with a high mobility group DNA-binding domain similar to SRY, the testis-determining factor [18], and is a member of a larger family of Sox genes. Sox-9 expression accompanies chondrocyte differentiation [17]. It is coexpressed with collagen type II (Col2{alpha}1) during chondrogenesis in the mouse and in cultured chondrocytes [15–21]. Ectopic expression of Sox-9 transactivates both a Col2{alpha}1-driven reporter gene and the endogenous Col2{alpha}1 gene in transgenic mice [20] and in cultured chondrocytes and embryonic cells [15, 16]. Recently Sox-9 was shown to enhance aggrecan gene promoter/enhancer activity [22]. In humans, heterozygous mutations in the Sox9 gene cause campomelic dysplasia, a malformation syndrome associated with sex-reversal and abnormal skeletal development, suggesting an essential role for Sox-9 in skeleton formation [23–26]. In mouse embryo chimaeras, Sox9-/- mutant cells were blocked in their differentiation to become chondrocytes and persisted as mesenchymal cells that were unable to express chondrocyte-specific markers, such as collagen types II, IX and XI and aggrecan, which shows that Sox-9 is an essential factor for chondrocyte differentiation [27].

Numerous studies have shown that the regulation of the synthesis of cartilage-specific molecules, such as collagens and proteoglycans in cultured chondrocytes and cartilage explants [28–33] or in undifferentiated mesodermal progenitors [5], is influenced by a variety of growth factors and cytokines. Chondrocytes are capable of producing and responding to a large number of peptide growth factors and cytokines, including insulin-like growth factor 1 (IGF-1) and interleukin 1 (IL-1) [28–33], and these results suggest that an imbalance of anabolic and catabolic growth factors and cytokines is likely to effect the chondrogenic phenotype. There is also evidence that growth factors and cytokines are required for chondrocyte survival, as in their absence chondrocytes undergo morphological changes and die by apoptosis [34–38].

Despite these studies, little is yet known of how anabolic and catabolic cytokines such as IGF-1 and IL-1 affect chondrocyte survival or the mechanism(s) whereby they influence the expression of the chondrocyte phenotype. In the present work, these questions were investigated using a chondrocyte-like cell line produced by carcinogen-immortalization of normal Syrian hamster embryo cells [6–8]. Culture of these cells in monolayer showed that at early passage these cells expressed chondrocyte-specific genes in a cell density- and serum-dependent manner [8, 12]. The present results show that, as with primary chondrocytes, the expression of these extracellular matrix genes is differentially modulated by specific cytokines. IGF-1 and IL-1{alpha} had opposite and antagonistic effects on gene expression and cell survival. While IL-1{alpha} was a potent inhibitor of the chondrocyte phenotype and down-regulated the expression of the chondrogenic transcription factor Sox-9, IGF-1 up-regulated Sox-9 expression and relieved the IL-1{alpha}-induced inhibition of the chondrocyte phenotype. Thus, we suggest that IGF-1 and IL-1{alpha}, both of which may be important in degenerative diseases of cartilage, both modulate chondrocyte survival and differentiation through changes in the expression of the Sox9 gene.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Cell culture
DES4 supB+ clone 2 (referred to as DES4+.2 hereafter) [6] and 10W supB+ clone 1 (referred to as 10W+.1 hereafter) [7] are two carcinogen-immortalized, non-tumorigenic Syrian hamster embryonic chondrocyte-like cell lines derived from early-passage diploid Syrian hamster embryo cells following treatment with diethylstilboestrol and asbestos respectively. The cells were cultured in Minimal Essential Medium supplemented with 10% fetal calf serum (FCS), 1% (v/v) non-essential amino acids, 1.4 mM L-glutamine, 100 units/ml penicillin and 100 µg/ml streptomycin (referred to as complete growth medium) in tissue culture dishes at 37°C under 5% carbon dioxide.

Treatment of chondrocyte cell lines with IL-1{alpha}, IGF-1 and 5-azacytidine
Early-passage DES4+.2 and 10W+.1 cells were split at 1 : 16 and cultured in complete growth medium for 3 days to reach 50–60% subconfluency or for 4–5 days to become confluent. The medium was changed from the subconfluent and confluent cells to either fresh complete growth medium or serum-free growth medium with or without 5 ng/ml human recombinant IL-1{alpha} or 50 ng/ml human recombinant IGF-1 or both cytokines together (Gibco-BRL Life Technologies, Paisley, UK), at the given concentrations, for 1 or 14 h before extraction of RNA or low molecular weight DNA, as indicated in the figure legends.

Serum-stimulated or serum-free subconfluent or confluent cultures of early-passage DES4+.2 cells were treated with 4.10 µM 5-azacytidine (5-aza-CR) (Sigma, Poole, Dorset, UK) for 14 h. The following day the medium was replaced with fresh medium containing 4.10 µM 5-aza-CR. Twenty-four hours later, floating cells were harvested and low molecular weight DNA was isolated.

Isolation of RNA and Northern blot analysis
Total cellular RNA was prepared using the LiCl–urea method [39] from subconfluent or confluent monolayers of early passage DES4+.2 cells in the presence or absence of 10% FCS and IL-1{alpha} or IGF-1, or both cytokines together, as indicated in the figure legends. Total RNA (30 µg) was subjected to Northern blot hybridization analysis using 32P-labelled specific probes as described previously [13]. The probes used were a 1.65-kilobase (kb) EcoRI human procollagen {alpha}1(I) cDNA fragment from HF-677 [40], a 0.6-kb EcoRI–HindIII fragment carrying exon 1 of the mouse procollagen {alpha}1(II) gene from pELIII [41], a 1.55-kb EcoRI–HindIII fragment carrying exon 2 of the human procollagen {alpha}1(II) gene from pTII (a kind gift of Drs D. Prockop and L. Ala-Kokko, Thomas Jefferson University, USA), a 0.75-kb EcoRI rat collagen {alpha}1(IX) cDNA fragment from pKT1643 [42], a 0.7-kb HindIII cDNA fragment encoding most of the C-terminal non-collagenous domain and part of the 3' untranslated region of the human collagen type X cDNA from pUChColX [43], a 2.1-kb EcoRI human aggrecan cDNA fragment from pKS.H4 [44], a 1.3-kb human versican cDNA fragment from PG-350 encompassing the 3' region of human versican (Bioquote, York, UK), a 1.8 kb EcoRI human link protein cDNA fragment from pKS8.1D3 [45], a 1.3-kb EcoRI human biglycan cDNA fragment from P16 [46] and a 255-base pair (bp) EcoRI–HindIII mouse Sox9 cDNA fragment from pKSmSox9 [16].

Reverse transcriptase–polymerase chain reaction (RT–PCR)
Total RNA isolated from early- (pdl 4) and late (pdl 12)-passage confluent DES4+.2 cells grown as monolayers was reverse-transcribed using Super RT grade reverse transcriptase (HT Biotechnology, Cambridge, UK). The reaction was performed in a 20-µl volume containing 2 µg of total RNA, 10 mM dithiothreitol, 1 mM dNTPs, 10 ng random primers pd(N)6 in 1xRT reaction buffer (50 mM Tris–HCl, pH 8.3, 40 mM KCl, 6 mM MgCl2) and 10 units of reverse transcriptase. The tubes were incubated on a Hybaid Combi (TR2) thermal reactor (Thermo Hybaid, Ashford, UK) at 37°C for 1 h. After heat-inactivation for 5-min at 95°C, the cDNAs were amplified in a reaction volume of 100 µl. The reaction contained 2 µl of cDNA, 1 mM MgCl2, 200 µM dNTPs, 0.5 µl of Taq polymerase (Promega, Southampton, UK), 1 µM of each forward and reverse primer to collagen type II or aggrecan in 1xreaction buffer.

The primer to collagen type II was 5'-GTACCGATCACAGAAGACCTCC-3' and 5'-TGCAGATCCTGGAGTGACTG-3' and the primer to aggrecan was 5'-GGTCCCATACTCCCACAGTTGGC-3' and 5'-GGAATCCCACTAACATATCCCCC-3'.

PCR reactions were carried out on a Hybaid Combi (TR2) thermal reactor as follows: one cycle of 95°C (2 min) and 72°C (2 min), 35 cycles of 95°C (1 min), 55°C (1 min) and 72°C (1 min) followed by one cycle of 72°C (5 min) and 30°C (1 min). The PCR reaction products were analysed on 1% agarose gels stained with ethidium bromide. The markers used were a 50–2000 bp DNA ladder (Amersham International, Little Chalfont, UK).

Isolation of chondrocyte-specific molecules and Western blot analysis
The culture media from early- and late-passage DES4+.2 cells cultured as monolayers were collected and processed for the isolation of collagens, proteoglycans and link protein and for immunoblot analysis as described previously [13]. Briefly, collagens or link protein were resolved on 8% sodium dodecyl sulphate (SDS)–polyacrylamide gels, blotted onto nitrocellulose and probed with a mouse monoclonal anti-human collagen type II antibody (Chemicon International Ltd, Harrow, UK) and the mouse monoclonal antibody 8-A-4, raised against rat chondrosarcoma link protein, respectively. Proteoglycans were separated on a polyacrylamide composite gel under dissociative conditions and blotted onto Immobilon nylon membrane (Millipore Ltd, Watford, UK). The filter was incubated for 1 h at room temperature with 0.05 units/ml ABC chondroitinase (Sigma, Dorset, UK), washed with phosphate-buffered saline (PBS), blocked with 3% low-fat dried milk/PBS and probed with an anti-human aggrecan hyaluronic acid binding region antiserum diluted at 1 : 1000. Collagen type II, aggrecan and link protein were visualized using an ECL Western blot detection kit (Amersham International, Little Chalfont, UK).

Isolation of low molecular weight DNA and analysis of DNA fragmentation
Early-passage DES4+.2 and 10W+.1 cells were cultured as monolayers under low or high cell density in the presence or absence of 10% FCS with or without cytokines, or after treatment with 5-aza-CR, as indicated in Fig. 5Go. The culture medium containing floating cells was collected from four 90-mm dishes after 24 h and placed on ice. The monolayers were washed with 5 ml calcium- and magnesium-free PBS (referred to as PBS hereafter) and the washes were collected and combined with the culture medium. The floating cells were pelleted by centrifugation at 4°C for 5 min at 1500 r.p.m. The cell pellets were resuspended in 1.5 ml of PBS containing 1 mM ethylenediamine tetraacetic acid (EDTA) (pH 8.0) and centrifuged at 4°C for 5 min at 6000 r.p.m. After a second wash as above, the cell pellet was resuspended in 400 µl of 0.6% SDS/1 mM EDTA (pH 8.0) and the lysate was mixed briefly by inversion. After standing for 10 min on ice, the lysate was mixed briefly and 100 µl of 5 M NaCl was added to give a final concentration of 1 M NaCl. The solution was mixed by inversion and left overnight at 4°C. After centrifugation at 4°C for 20 min at 6500 r.p.m., the supernatant was transferred to a clean tube and the low molecular weight DNA was precipitated by the addition of two volumes of ethanol at -20°C overnight. The DNA was pelleted by centrifugation at 4°C for 30 min at 13000 r.p.m., washed with 70% ethanol, dried under vacuum and dissolved in sterile distilled water. For in vitro labelling of the DNA, 3 µg of DNA in 20 µl was mixed with 3 µl of 100 mM Tris–HCl (pH 7.5), 50 mM MgCl2, 0.1 µCi [32P]dCTP and 5 U Klenow polymerase in a final reaction volume of 30 µl and incubated at room temperature for 25 min. The reaction was terminated by the addition of 0.6 µl of 0.5 M EDTA (pH 8.0) [47]. The DNA was analysed by electrophoresis on agarose gel containing 1.8% ethidium bromide for 3 h at 100 V. The gel containing labelled DNA was then dried under vacuum without heat and exposed to autoradiography.



View larger version (71K):
[in this window]
[in a new window]
  		FIG. 5. Effects of cell density, serum, 5-aza-CR and specific cytokines on the survival of chondrocytes. Low molecular weight DNA was extracted from floating DES4+.2 (lanes 1–10) and 10W+.1 (lanes 11–15) cells cultured under the different conditions, radiolabelled with [32P]dCTP, fractionated on a 1.8% agarose gel, dried and exposed to autoradiography. DES4+.2 cells were cultured at low cell density and stimulated with FCS for 1 h (lane 1) or with 10% FCS for 14 h (lane 2). DES4+.2 cells were cultured at high cell density and stimulated with 10% FCS for 14 h (lane 3). DES4+.2 and 10W+.1 cells were cultured at high cell density in the absence of serum (lanes 4 and 12 respectively) and in the presence of IL-1{alpha} (lanes 5 and 13 respectively) or in the presence of IGF-1 (lanes 6 and 14, respectively). DES4+.2 and 10W+.1 cells (lanes 7 and 15 respectively) were cultured at high cell density in serum-free medium and in the presence of IL-1{alpha} and IGF-1. DES4+.2 cells were cultured at high cell density and stimulated with FCS for 1 h (lane 8) or 14 h (lane 9) or incubated in the absence of serum (lane 10) and then treated with 5-aza-CR for 24 h. 10W+.1 cells were cultured at low cell density in the absence of serum (lane 11). Exposure was for 10 min at room temperature.

 

   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Characterization of Syrian hamster embryonic chondrocyte-like cells
The two preneoplastic Syrian hamster embryonic cell lines, DES4+.2 and 10W+.1 [6, 7], have been shown previously to express chondrocyte markers, such as collagen type II and type IX mRNA, suggesting that carcinogen-induced immortalization selected for chondrocyte-like cell lines from a mixed embryo cell population [8, 12]. To further characterize one of these cell lines, DES4+.2, proteins were extracted from confluent early-passage (pdl 4) and late-passage (pdl 12) cells grown as monolayers and subjected to immunoblot analysis using antibodies specific for the chondrocyte markers collagen type II, aggrecan and link protein (Fig. 1AGo). Whereas in early passage the DES4+.2 cells synthesized collagen type II, aggrecan and link protein, in late passage they failed to synthesize collagen type II, while aggrecan synthesis was decreased and link protein synthesis remained unaffected (Fig. 1AGo). The chondrocyte-like phenotype exhibited by early-passage cells was stable during 10–12 population doublings, but at later passage collagen II and aggrecan were detectable only by RT–PCR (Fig. 1BGo).



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 1. Expression of chondrocyte-specific markers in DES4+.2 cells. (A) Western blot analysis of collagen type II, aggrecan and link protein isolated from culture media of early (pdl 4) and late (pdl 12) DES4+.2 cells, using specific antibodies. The positions of the chondrocyte-specific molecules are indicated by arrows. Link protein (LP) appears as a doublet of 48 kDa and 45 kDa, corresponding to LP1 and LP2 respectively, which differ in their degree of glycosylation. (B) RT-PCR analysis of total RNA extracted from early-passage (pdl 4) and late-passage (pdl 12) DES4+.2 cells for collagen type II and aggrecan.

 
Total RNA was extracted from confluent early- and late-passage cells grown as monolayers and subjected to Northern blot hybridization analysis using a variety of chondrocyte markers as probes. Early-passage, but not late-passage, DES4+.2 cells were found to express collagen types II and IX, collagen {alpha}1(I) and collagen X, aggrecan, biglycan and link protein (Figs 2Go and 3Go), suggesting that these embryonic cells have a chondrogenic potential that declines with passage [9]. The simultaneous expression of collagen types II and IX and collagen type X, a marker of cellular hypertrophy [9], suggests that these cells may express a prehypertrophic chondrocyte-like phenotype.



View larger version (85K):
[in this window]
[in a new window]
  		FIG. 2. Effects of cell density, serum and specific cytokines on the expression of cartilage collagens. Total RNA isolated from early-passage DES4+.2 cells cultured at low cell density and stimulated with FCS for 1 h (lane 1) or 14 h (lane 2) at high cell density and stimulated with FCS for 1 h (lane 3) or 14 h (lane 4) and in the absence of serum at low (lane 5) or high (lane 6) cell density and at high cell density in the absence of serum and in the presence of IL-1{alpha} (lane 7), IGF-1 (lane 8) or both cytokines together (lane 9) was fractionated, blotted and hybridized with specific probes to (A) collagen type {alpha}1(II), (B) collagen type {alpha}1(IX), (C) collagen type {alpha}1(I), (D) collagen type {alpha}1(X) and human ß-actin (bottom panel). The arrows indicate the sizes of the transcripts obtained. Exposure times at -80°C: A–C, 16 h; D, 48 h.

 


View larger version (92K):
[in this window]
[in a new window]
  		FIG. 3. Effects of cell density, serum and specific cytokines on the expression of cartilage proteoglycans and link protein. Total RNA isolated from early-passage DES4+.2 cells cultured at low cell density and stimulated with FCS for 1 h (lane 1) or 14 h (lane 2), at high cell density and stimulated with FCS for 1 h (lane 3) or 14 h (lane 4) and in the absence of serum at low (lane 5) or high (lane 6) cell density and at high cell density in the absence of serum and in the presence of IL-1{alpha} (lane 7), IGF-1 (lane 8) or both cytokines together (lane 9) was fractionated, blotted and hybridized with specific probes to (A) aggrecan, (B) biglycan, (C) link protein and human ß-actin (bottom panel). The arrows indicate the sizes of the transcripts obtained. Exposure times at -80°C: A, 16 h; B, 36 h; C, 24 h.

 

Expression of chondrocyte-specific collagens depends on cell density and serum
To study how the expression of chondrocyte-specific collagen genes such as collagen types II and IX was modulated by different culture conditions and specific cytokines, early-passage DES4+.2 cells were cultured as monolayers at low or high cell density in the presence or absence of 10% FCS. Northern blot analysis of the expression of collagen type {alpha}1(II) (Fig. 2AGo) and collagen type {alpha}1(IX) (Fig. 2BGo) showed that it was dependent on cell density and on the presence of serum. Stimulation of subconfluent or confluent cultures with serum for 1 h increased (Fig. 2AGo and BGo, lanes 1 and 3) the expression of collagen types II and IX compared with subconfluent or confluent cultures incubated in serum-containing medium for 14 h (Fig. 2AGo and BGo, lanes 2 and 4). The expression of both cartilage-specific collagens was much less in serum-containing subconfluent (Fig. 2AGo and BGo, lanes 1 and 2) than in the corresponding confluent cultures (Fig. 2AGo and BGo, lanes 3 and 4). Similarly, the mRNA levels of collagen types II and IX were lower in the subconfluent (Fig. 2AGo and B, lane 5) than in the confluent (Fig. 2AGo and BGo, lane 6) cells cultured in the absence of serum. Furthermore, densitometric analysis showed that the expression of collagen types II and IX in cells cultured in the presence of serum was higher than in cells cultured in the absence of serum (Fig. 2AGo and BGo; compare lanes 1 and 2 with lane 5 and lanes 3 and 4 with lane 6).

In the confluent DES4+.2 cells cultured in the presence of serum, collagen type {alpha}1(IX) was detected in two transcripts of about 4.5 and 4.2 kb (Fig. 2BGo, lanes 3 and 4), whereas in serum-stimulated subconfluent cells (Fig. 2BGo, lanes 1 and 2) and in serum-free cells (Fig. 2BGo, lanes 5 and 6) only the 4.5-kb mRNA was present. Two collagen type {alpha}1(IX) transcripts have been detected previously in several Syrian hamster chondrocyte-like cell lines [8] and also in chicken embryonic sternal cartilage [48, 49], but the significance of their absence in subconfluent and serum-free cells in the present work remains to be determined.

Hybridization of the Northern blot with a human collagen {alpha}1(I) cDNA fragment (Fig. 2CGo) produced two transcripts, of 7.5 and 5.9 kb. The pattern of expression of collagen {alpha}1(I) was serum- and density-dependent. The expression of collagen type I was higher in the serum-stimulated, confluent cells (Fig. 2CGo, lanes 3 and 4) than in the subconfluent DES4+.2 cells (Fig. 2CGo, lanes 1 and 2) and was very low or undetectable in the serum-free, subconfluent and confluent cells (Fig. 2CGo, lanes 5 and 6). The expression of type X collagen, as a marker of chondrocyte hypertrophy, appeared to be low (Fig. 2DGo), but it was cell density- and serum-dependent, as was that of the other collagens.

Expression of chondrocyte-specific aggrecan and link protein is cell density- and serum-dependent
The expression of cartilage aggrecan and link protein genes was studied to assess if they followed the same pattern of expression as cartilage collagens. The DES4+.2 cells expressed an 8.5-kb aggrecan mRNA under all culture conditions tested (Fig. 3AGo, lanes 1–6). Like the expression of cartilage collagens (Fig. 2Go), the expression of aggrecan was serum- and cell density-dependent. Aggrecan expression was higher in the serum-stimulated, confluent cells (Fig. 3AGo, lanes 3 and 4) than in the subconfluent cells (Fig. 3AGo, lanes 1 and 2), and also higher than in cells cultured in serum-free media (Fig. 3AGo, lanes 5 and 6). Versican [13, 50] was also expressed by DES4+.2 cells, but much longer exposures were required to detect it (not shown).

Chondrocytes also express small proteoglycans, such as biglycan and decorin [51–53]. The DES4+.2 cells expressed a 2.7-kb biglycan mRNA (Fig. 2BGo) in a cell density- and serum-dependent manner, but the changes in its expression were less profound than those in the expression of aggrecan (Fig. 2BGo, lanes 1–6).

Northern blot analysis using a human link protein cDNA fragment as probe (Fig. 3CGo) showed three link protein transcripts, as reported previously in mouse [11], rat [54] and human [13] chondrocytes. The pattern of expression of link protein in low-density (Fig. 3CGo, lanes 1 and 2) and high-density (Fig. 3CGo, lanes 3 and 4) cultures in response to serum was similar to that of aggrecan (Fig. 3AGo) and the proportions of transcripts of different sizes did not appear to change.

Modulation of cartilage-specific gene expression by IL-1{alpha} and IGF-1
The rapid induction of cartilage-specific gene expression by serum prompted us to investigate the effects of the cytokines IGF-1 and IL-1{alpha}, both of which have been reported previously to play a role in chondrocyte matrix production and in growth and differentiation. Confluent monolayers of early-passage DES4+.2 cells cultured with IL-1{alpha} or IGF-1 were analysed by Northern blot hybridization for the expression of cartilage-specific genes (Figs 2Go and 3Go). Addition of IL-1{alpha} reduced dramatically the mRNA levels of collagen types II, IX and X (Fig. 2AGo, BGo and DGo, lane 7), aggrecan and link protein (Fig. 3BGo and CGo, lane 7) compared with serum-free control cultures (Figs 2Go and 3Go, lane 6). In contrast, the expression of collagen {alpha}1(I) increased in response to IL-1{alpha} (Fig. 2CGo, lane 7) and the expression of biglycan was unaffected (Fig. 3BGo, lane 7) compared with controls (Figs 2CGo and 3BGo, lane 6).

Whereas treatment of confluent, serum-free cultures of DES4+.2 with IGF-1 resulted in a significant increase in the mRNA levels of collagen types II (Fig. 2AGo, lane 8) and IX (Fig. 2BGo, lane 8), aggrecan (Fig. 3AGo, lane 8), biglycan (Fig. 3BGo, lane 8) and link protein (Fig. 3CGo, lane 8), it had no effect on the expression of collagen {alpha}1(I) (Fig. 2CGo, lane 8) or collagen type X (Fig. 2DGo, lane 7) compared with controls (Figs 2Go and 3Go, lane 6).

However, when both cytokines (IL-1{alpha} and IGF-1) were present, IGF-1 relieved the IL-1{alpha}-induced suppression of chondrocyte-specific gene expression (Figs 2Go and 3Go, lane 9). In the presence of both cytokines, at the concentrations used there was a significant increase in the mRNA levels of collagen type II (Fig. 2AGo, lane 9), collagen type IX (Fig. 2BGo, lane 9), aggrecan (Fig. 3AGo, lane 9), link protein (Fig. 3CGo, lane 9) and, to a lesser degree, biglycan (Fig. 3BGo, lane 9) compared with cultures treated with IL-1{alpha} alone (lane 7 in Fig. 2AGo and B and Fig. 3AGo–CGo). In contrast, the expression of collagen {alpha}1(I) (Fig. 2CGo, lane 9) in the presence of IL-1{alpha} and IGF-1 was less than in cells treated with IL-1{alpha} alone (Fig. 2CGo, lane 7). Together in serum-free confluent cultures, IL-1{alpha} and IGF-1 did not appear to increase significantly the expression of collagen {alpha}1(X) (Fig. 2DGo, lane 9). Hence, IL-1{alpha} and IGF-1 had opposite and antagonistic effects, IL-1{alpha} acting as an inhibitor and IGF-1 acting as an inducer of chondrocyte-specific gene expression.

Inhibition of Sox9 expression by IL-1{alpha} is relieved by IGF-1
In order to gain some insight of the mechanism(s) whereby IL-1{alpha} and IGF-1 affect the expression of the chondrocyte phenotype, we investigated the expression of the chondrogenic transcription factor gene Sox9 in response to these cytokines acting alone or together. Expression of Sox9 was cell-density dependent (Fig. 4Go, lanes 2 and 3) but serum appeared to have little effect on confluent cells (Fig. 4Go, lanes 1 and 3). IL-1{alpha}-treatment of DES4+.2 cells markedly inhibited the expression of Sox9 (Fig. 4Go, lane 4), whereas IGF-1 up-regulated its expression (Fig. 4Go, lane 5) and partially relieved the IL-1{alpha}-induced down-regulation of Sox9 (Fig. 4Go, lane 6). Densitometric analysis using Kodak DC120 zoom digital camera and a gel analyser software (Biosure Ltd, Athens, Greece) showed a four-fold increase in the expression of Sox9 when IGF-1 and IL-1{alpha} were added together (Fig. 4Go, lane 6), compared with that in the cells treated with IL-1{alpha} alone (Fig. 4Go, lane 4). Interestingly, the expression of Sox9 was absent in late-passage DES4+.2 cells (Fig. 4Go, lane 7) and IGF-1-treatment was not sufficient to induce it in these cells (Fig. 4Go, lane 8). These changes in Sox9 expression thus paralleled the changes in the expression of the chondrocyte-specific markers such as collagen types II and IX (Fig. 2Go), aggrecan and link protein (Fig. 3Go).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 4. Effects of IL-1 and IGF-1 on Sox9 gene expression in DES4+.2 cells. Total RNA isolated from early-passage DES4+.2 cells cultured at high cell density and stimulated with FCS for 14 h (lane 1), at low cell density in the absence of serum (lane 2), at high cell density in the absence of serum (lane 3) and in the presence of IL-1{alpha} (lane 4), IGF-1 (lane 5) or both cytokines together (lane 6) and from late-passage cells cultured at high cell density in the absence of serum (lane 7) and in the presence of IGF-1 (lane 8) was fractionated, blotted and hybridized with a cDNA specific probe to Sox9 (upper panel) or GAPDH (lower panel).

 

Induction of apoptosis in cultured chondrocytes and the effect of IL-1{alpha}, IGF-1 and 5-aza-CR on cell survival
In order to detect apoptosis, DNA was isolated from floating DES4+.2 and 10W+.1 cells growing under the different culture conditions. After radioactive labelling and agarose gel electrophoresis, the characteristic DNA ladders of cells undergoing apoptosis were detected (Fig. 5Go). Whereas subconfluent (Fig. 5Go, lanes 1 and 2) and confluent (Fig. 5Go, lane 3) cells cultured as monolayers in the presence of serum did not undergo apoptosis, cells cultured under serum-free conditions at high cell density (Fig. 5Go, lanes 4 and 12), but not at low cell density (Fig. 5Go, lane 11), showed extensive DNA fragmentation, with 180-bp nucleosome-size fragments characteristic of cells undergoing apoptosis.

Whereas addition of IL-1{alpha} to serum-free, confluent DES4+.2 and 10W+.1 cells cultured as monolayers did not increase apoptosis (Fig. 5Go, lanes 5 and 13 respectively), addition of IGF-1 (Fig. 5Go, lanes 6 and 14 respectively) reduced the extent of DNA fragmentation compared with the untreated controls (Fig. 5Go, lanes 4 and 12 respectively). The addition of both IL-1{alpha} and IGF-1 to serum-free confluent monolayer cultures of DES4+.2 cells (Fig. 5Go, lane 7) and 10W+.1 cells (Fig. 5Go, lane 15) showed that IL-1{alpha}, which when acting alone had no effect on apoptosis, opposed slightly the IGF-1-mediated positive effect on cell survival. This was noticeable in 10W+.1 cells (Fig. 5Go, compare lane 14 with 15) but not in DES4+.2 (Fig. 5Go, compare lane 6 with 7). Thus IGF-1, even in the presence of IL-1{alpha}, was capable of increasing cell survival compared with serum-free, confluent, untreated (Fig. 5Go, lanes 4 and 12) and IL-1{alpha}-treated cells (Fig. 5Go, lanes 5 and 13).

Several factors have been reported to induce apoptosis in cultured chondrocytes [35–38, 55–59]. Treatment of confluent DES4+.2 cells with 5-aza-CR, a DNA-demethylating agent, induced apoptosis irrespective of the presence or absence of serum (Fig. 5Go, lanes 8–10). However, the extent of DNA fragmentation appeared to be higher in the cells stimulated with serum for 1 h (Fig. 5Go, lane 8) than in those stimulated with serum for 14 h (Fig. 5Go, lane 9) and those cultured in serum-free medium (Fig. 5Go, lane 10).


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
A Syrian hamster embryonic chondrocyte-like cell line, DES4+.2 [6, 8, 12, 55, 56], was used to investigate the effects of culture conditions, serum, cell density and specific cytokines (IGF-1 and IL-1{alpha}) on chondrocyte-specific gene expression and cell survival. The main novel findings were that chondrocyte-specific gene expression was cell density- and serum-dependent and that two specific cytokines, IL-1{alpha} and IGF-1, had opposite and antagonistic effects on the expression of chondrocyte markers and cell survival. The effects of these two cytokines on the chondrocyte phenotype correlated with changes in the expression of Sox9, a key chondrocyte-regulatory transcription factor gene.

Early- but not late-passage DES4+.2 cells cultured as monolayers expressed chondrocyte-specific markers, such as collagen types II, IX and X (Fig. 2Go), aggrecan, biglycan and link protein (Fig. 3Go) in a cell density- and serum-dependent manner. In later-passage cells, synthesis of collagen type II, aggrecan and link protein was absent, reduced and remained unaffected respectively. Failure of the ability of late-passage cells to express chondrocyte differentiation markers was linked to the loss of their ability to express tumour suppressor gene function, which may be involved in the regulation of chondrocyte differentiation [8, 12]. Collagen type {alpha}1(II) mRNA was present as a 5.2-kb transcript in cells cultured under all the different conditions (Fig. 2AGo). However, type II procollagen mRNA can be alternatively spliced to give two mRNA species [60, 61]. The type IIA mRNA species contains exon 2, encoding a cysteine-rich domain in the amino propeptide, and is expressed in several human [62] and mouse [63] tissues, whereas type IIB, which lacks exon 2, is the major transcript in cartilage [60–63]. Although the cells used in the present study were of embryonic origin, no expression of collagen type IIA was detectable by Northern blot hybridization analysis using a human exon 2-specific probe. This showed that DES4+.2 cells predominantly expressed type IIB, the cartilage-specific procollagen mRNA. In addition, DES4+.2 cells expressed collagen {alpha}1(I) and low levels of collagen type X mRNA, a marker of chondrocyte hypertrophy, suggesting that the cells may be in a prehypertrophic state of maturation. Furthermore, the expression of collagen type X, and hence progression towards cellular hypertrophy, could be minimized or prevented by culturing the cells at low cell density and in the absence of serum, which is consistent with previous reports [64, 65]. Evidence for the synthesis of the protein products of the major chondrocyte-specific genes, such as collagen type II, aggrecan and link protein, by early-passage DES4+.2 cells suggests that these cells grown as monolayers retained their differentiation characteristics at least for several passages. Furthermore, the cell density- and serum-dependent expression of all the cartilage-specific genes shows that culture conditions have a major affect on the transcription of these genes. The results presented here show that these effects correlate with the expression of Sox9 (Fig. 5Go). This is consistent with studies showing that transfection of Sox9 enhances the expression of collagen types II and IX [15, 16, 19, 20] and aggrecan [22] and that Sox9 expression is essential for chondrogenesis [17, 27, 66, 67]. Thus, DES4+.2 cells required growth factors for the expression of their differentiated characteristics, which may be provided in serum by other cell types or may be produced by chondrocytes themselves [30, 31]. Moreover, brief stimulation of the cells with serum was sufficient to enhance the expression of all the genes investigated, showing that a rapid response was involved.

The major action of serum on the expression of chondrocyte-specific genes may be due to its IGF-1 content. Investigation of the effects of two specific cytokines, IGF-1 and IL-1{alpha}, which have been implicated in chondrocyte growth and differentiation [30, 31] as well as in pathological conditions of cartilage [28], showed that they had opposite and antagonistic effects on chondrocyte-specific gene expression. Whereas IL-1{alpha} suppressed the basal mRNA levels, IGF-1 up-regulated the expression of all the chondrocyte-specific genes investigated. Similar results were also obtained with IL-1ß, at least on collagen type II and aggrecan gene expression (not shown). Furthermore, IGF-1 relieved the IL-1{alpha}-induced inhibition of chondrocyte-specific gene expression at the concentration used. The opposite and antagonistic effects of IL-1{alpha} and IGF-1 on the expression of the chondrocyte phenotype paralleled changes in the expression of the Sox9 gene. Whereas IGF-1 has been shown previously to induce the transcription and synthesis of chondrocyte-specific markers, IL-1 has been shown to decrease proteoglycan and collagen type II synthesis in chondrocytes and in explant cultures of articular cartilage [28–33, 68]. In particular, IL-1{alpha} or IL-1ß has been shown to suppress the transcription of human collagen type IX [69] and collagen type II in cultured human [69, 70], rat, rabbit [71] and mouse [72] chondrocytes, an effect shown previously to be mediated by a transcriptional mechanism [71]. Recent studies have reported that IL-1ß inhibits the expression of the chondrocyte phenotype by markedly down-regulating the expression of Sox9 and that transfection of a Sox9 expression vector in chondrocytes overcomes the inhibitory effect of IL-1ß on the activity of a Col2{alpha}1 enhancer construct [72]. This is consistent with the present studies, which show that IGF-1 overcomes the inhibitory effect of IL-1{alpha} on Sox9 expression, thus maintaining the chondrocyte phenotype. Late-passage cells that did not express the chondrocyte phenotype failed to express Sox9, and IGF-1 was unable to induce Sox9 expression in these cells. Given the recent genetic evidence for the role of Sox-9 in chondrocyte differentiation [17, 27, 66, 67] and the fact that IL-1{alpha} and IGF-1 modulate the expression of a number of chondrocyte-specific matrix molecules, the pronounced effects of these cytokines on the expression of the Sox9 gene strongly suggest that their principal effects on the chondrocyte phenotype is mediated by their regulation of Sox9 expression. Cloning of the regulatory sequences of the Sox9 gene may elucidate the precise mechanism(s) by which IL-1 and IGF-1 exert this control.

Several studies regarding chondrocyte survival in culture produced apparently conflicting results [34–38, 55–59], perhaps because cells from different sources and at different stages of maturation were used. Whereas DES4+.2 and 10W+.1 cells, which appear to express a prehypertrophic chondrocyte-like phenotype when cultured as monolayers, survived in low cell density cultures, they died rapidly by apoptosis when cultured at high cell density in the absence of serum or IGF-1. Addition of IL-1{alpha} to serum-free, confluent cultures of DES4+.2 and 10W+.1 cells did not further stimulate apoptosis, but appeared to partially suppress the IGF-1-mediated positive effect on cell survival. Hence, IL-1{alpha} and IGF-1 had opposite and antagonistic effects on cell survival, IGF-1 being a cell survival factor. Several studies demonstrated that IGF-1 or insulin (at high enough concentrations to activate IGF-1 receptors) were effective in promoting the survival of cultured chondrocytes [34, 36, 38]. Thus, chondrocytes require exogenous signals to survive in culture, just like several other cell types [73, 74], but it appears that the survival of prehypertrophic chondrocytes, unlike that of nearly all cell types investigated to date [74], is independent of cell-to-cell communication. This agrees with recent findings with embryonic chick sternal chondrocytes [37], which depended on the presence of serum or IGF-1.

Treatment of DES4+.2 cells with 5-aza-CR induced DNA fragmentation irrespective of the presence of serum. One mechanism of DNA fragmentation involves the activation of endogenous endonucleases, hence 5-aza-CR may activate these DNA-degrading enzymes, leading to the activation of apoptosis of cultured chondrocytes.

In this study, we provide evidence that loss of chondrocyte phenotype by in vitro passage and after IL-1{alpha} treatment is linked to down-regulation of the expression of Sox-9, a factor that is required for chondrocyte differentiation and that IGF-1 relieves the IL-1{alpha}-induced inhibition of the expression of Sox-9, and hence promotes the chondrocyte phenotype. Furthermore, our results demonstrate that IL-1{alpha} and IGF-1 affect chondrocyte apoptosis, IGF-1 being a chondrocyte survival factor. Given that IL-1 is present at elevated levels in inflammatory joint diseases, such as arthritis, and that it antagonizes the effects of IGF-1, which is available systemically or produced locally [28, 30, 31, 38], the control of the expression of the Sox9 gene and the influence on cell survival by these two factors could limit the repair capacity of cartilage in inflammatory joint diseases.


   Acknowledgments
 
We thank Drs R. Boot-Handford (University of Manchester, UK), K. Cheah (University of Hong-Kong), D. Prockop and L. Ala-Kokko (Thomas Jefferson University, Philadelphia, USA), B. Olsen (Harvard Medical School, Boston, USA), V. Lefebvre (University of Texas MD Anderson Cancer Center, USA), J. Dudhia (Royal Veterinary College, London, UK), L. Fisher and M. Young (National Institute of Dental Research, Bethesda, MD, USA) for providing probes. We are indebted to Professor C. W. Archer (University of Wales at Cardiff, UK) for critical reading of the manuscript. EK was supported by a Postdoctoral Research Fellowship from the Arthritis and Rheumatism Council of Great Britain and a Travel grant from the Wellcome Trust. This work was supported by grants from the Arthritis and Rheumatism Council of Great Britain to HIM.


   Notes
 
Correspondence to: E. Kolettas, Cell and Molecular Physiology Unit, Laboratory of Physiology, University of Ioannina Medical School, 45110 Ioannina, Greece. Back


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Erlebacher A, Filvaroff EH, Gitelman SE, Derynck R. Toward a molecular understanding of skeletal development. Cell1998;80:371–8.
  2. Hall BK, Miyake T. Divide, accumulate, differentiate: cell condensation in skeletal development revisited. Int J Dev Biol1995;39:881–93.[Medline]
  3. Mundlos S, Olsen BR. Heritable diseases of the skeleton. Part I: Molecular insights into skeletal development—transcription factors and signalling pathways. FASEB J1997;11:125–32.[Abstract]
  4. Poliard A, Nifuji A, Lamblin D, Plee E, Forest C, Kellermann O. Controlled conversion of an immortalized mesodermal progenitor cell towards osteogenic, chondrogenic or adipogenic pathways. J Cell Biol1995;130:1461–72.[Abstract/Free Full Text]
  5. Atkinson BL, Fantle KS, Benedict JJ, Huffer WE, Gutierrez-Hartmann A. Combination of osteoinductive bone proteins differentiates mesenchymal C3H/10T1/2 cells specifically to the cartilage lineage. J Cell Biochem1997;65:325–39.[ISI][Medline]
  6. McLachlan JA, Wong A, Degen GH, Barrett JC. Morphological and neoplastic transformation of Syrian hamster embryo fibroblasts by diethylstilbestrol and its analogs. Cancer Res1982;42:3040–5.[Abstract/Free Full Text]
  7. Koi M, Barrett JC. Loss of tumour suppressor function during chemically induced neoplastic progression of Syrian hamster embryo cells. Proc Natl Acad Sci USA1986;83:5992–6.[Abstract/Free Full Text]
  8. Cizdziel PE, Hosoi J, Montgomery JC, Wiseman RW, Barrett JC. Loss of tumour suppressor gene function is correlated with downregulation of chondrocyte-specific collagen expression in Syrian hamster embryo cells. Mol Carcinog1991;4:14–24.[Medline]
  9. Muir HI. The chondrocyte, architect of cartilage: Biomechanics, structure, function and molecular biology of cartilage matrix macromolecules. Bioessays1995;17:1039–48.[ISI][Medline]
  10. Horton WE, Cleveland J, Rapp U et al. An established rat cell line expressing chondrocyte properties. Exp Cell Res1988;178:457–68.[Medline]
  11. Mallein-Gerin F, Olsen BR. Expression of simian virus 40 large T (tumour) oncogene in mouse chondrocytes induces cell proliferation without loss of the differentiated phenotype. Proc Natl Acad Sci USA1993;90:3289–93.[Abstract/Free Full Text]
  12. Owen RD, Hosoi J, Montgomery JC, Wiseman RW, Barrett JC. Coordinate regulation of collagen II (alpha 1) and H19 expression in immortalised hamster cells. Cell Growth Differ1993;4:1013–21.[Abstract]
  13. Kolettas E, Buluwela L, Bayliss MT, Muir HI. Expression of cartilage-specific molecules is retained on long-term culture of human articular chondrocytes. J Cell Sci1995;108:1991–9.[Abstract]
  14. Mataga N, Tamura M, Yanai N et al. Establishment of a novel chondrocyte-like cell line derived from transgenic mice harboring the temperature sensitive simian virus 40 large T-antigen gene. J Bone Miner Res1996;11:1646–54.[Medline]
  15. Lefebvre V, Huang W, Harley VR, Goodfellow PN, de Crombrugghe B. Sox9 is a potent activator of the chondrocyte-specific enhancer of the pro{alpha}1(II) collagen gene. Mol Cell Biol1997;17:2336–46.[Abstract]
  16. Lefebvre V, Li P, de Crombrugghe B. A new long form of Sox5 (L-Sox5), Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively activate the type II collagen gene. EMBO J1998;17:5718–33.[ISI][Medline]
  17. Lefebvre V, de Crombrugghe B. Toward understanding Sox9 function in chondrocyte differentiation. Matrix Biol1998;16:529–40.[ISI][Medline]
  18. Wright E, Hargrave MR, Christiansen J et al. The Sry-related gene Sox9 is expressed during chondrogenesis in mouse embryos. Nat Genet1995;9:15–20.[ISI][Medline]
  19. Ng LJ, Wheatley SC, Muscast GEO et al. Sox9 binds DNA, activates transcription and coexpresses with type II collagen during chondrogenesis in the mouse. Dev Biol1997;183:108–21.[ISI][Medline]
  20. Bell DM, Leung KKH, Wheatley SC et al. Sox9 directly regulates the type-II collagen gene. Nat Genet1997;16:174–8.[ISI][Medline]
  21. Zhao Q, Eberspaecher H, Lefebvre V, de Crombrugghe B. Parallel expression of Sox9 and Col2{alpha}1 in cells undergoing chondrogenesis. Dev Dyn1997;209:377–86.[ISI][Medline]
  22. Sekiya I, Tsuji K, Koopman P et al. Sox9 enhances aggrecan gene promoter/enhancer activity and is up-regulated by retinoic acid in a cartilage-derived cell line, TC6. J Biol Chem2000;275:10738–44.[Abstract/Free Full Text]
  23. Wagner T, Wirth J, Meyer J et al. Autosomal sex reversal and campomelic dysplasia are caused by mutations in and around the Sry-related gene Sox9. Cell1994;79:1111–20.[ISI][Medline]
  24. Foster JW, Dominguez-Steglich MA, Guioli S et al. Campomelic dysplasia and autosomal sex reversal caused by mutations in an Sry-related gene. Nature1994;372:525–30.[Medline]
  25. Cameron FJ, Hageman RM, Cook-Yarborough C et al. A novel germ line mutation in Sox9 causes familial campomelic dysplasia and sex reversal. Hum Mol Genet1996;5:1625–30.[Abstract/Free Full Text]
  26. Südbeck P, Schmitz L, Baeuerle PA, Scherer G. Sex reversal by loss of the C-terminal transactivation domain of human Sox9. Nat Genet1996;13:230–2.[ISI][Medline]
  27. Bi W, Deng JM, Zhang Z, Behringer RR, de Crombrugghe B. Sox9 is required for cartilage formation. Nat Genet1999;22:85–9.[ISI][Medline]
  28. Tyler JA. Cartilage degradation. In: Hall BK, Newman SA, eds. Cartilage: molecular aspects. Boca Raton (FL): CRC Press, 1991:213–56.
  29. Fosang A, Tyler JA, Hardingham TE. Effect of interleukin-1 and insulin-like growth factor-1 on the release of proteoglycan components and hyaluronan from pig articular cartilage in explant culture. Matrix1991;11:17–24.[ISI][Medline]
  30. Seydin SM, Rosen DM. Cartilage growth and differentiation factors. In: Hall BK, Newman SA, eds. Cartilage: molecular aspects. Boca Raton (FL): CRC Press, 1991:131–52.
  31. Hill DJ, Logan A. Peptide growth factors and their interactions during chondrogenesis. Prog Growth Factor Res1992;4:45–68.[Medline]
  32. Smith P, Shuler FD, Georgescu HI et al. Genetic enhancement of matrix synthesis of articular chondrocytes: comparison of different growth factor genes in the presence and absence of interleukin-1. Arthritis Rheum2000;43:1156–64.[ISI][Medline]
  33. Hui W, Rowan AD, Cawston T. Insulin-like growth factor-1 blocks collagen release and downregulates matrix metalloproteinase-1, -3, -8 and -13 mRNA expression in bovine nasal cartilage stimulated with oncostatin M in combination with interleukin 1alpha. Ann Rheum Dis2001;60:254–61.[Abstract/Free Full Text]
  34. Bruckner P, Horler I, Mendler M et al. Induction and prevention of chondrocyte hypertrophy in culture. J Cell Biol1989;109:2537–45.[Abstract/Free Full Text]
  35. Tshan T, Horler I, Houze Y, Winterhalter KH, Richter C, Bruckner P. Resting chondrocytes in culture survive without growth factors but are sensitive to toxic oxygen metabolites. J Cell Biol1990;111:257–60.[Abstract/Free Full Text]
  36. Ishizaki Y, Burne J, Raff MC. Autocrine signals enable chondrocytes to survive in culture. J Cell Biol1994;126:1069–77.[Abstract/Free Full Text]
  37. Gibson G, Lin D-L, Roque M. Apoptosis of terminally differentiated chondrocytes in culture. Exp Cell Res1997;233:372–82.[ISI][Medline]
  38. Loester RF, Shanker G, Autocrine stimulation by insulin-like growth factor 1 and insulin-like growth factor 2 mediates chondrocyte survival in vitro. Arthritis Rheum2000;43:1552–9.[ISI][Medline]
  39. Auffray C, Rougeon F. Purification of mouse immunoglobulin heavy chain messenger RNAs from total myeloma tumour RNA. Eur J Biochem1980; 107:303–14.[ISI][Medline]
  40. Chu M-L, Myers JC, Bernard MP, Ding JF, Ramirez F. Cloning and characterization of five overlapping cDNAs for the human pro alpha 1(I) collagen chain. Nucleic Acids Res1982;10:5925–34.[Abstract/Free Full Text]
  41. Cheah KSE, Lau ET, Au PK, Tam PP. Expression of the mouse alpha1 (II) collagen gene is not restricted to cartilage during development. Development1994;111:945–53.[Abstract/Free Full Text]
  42. Kimura T, Mattei M-G, Stevens JW, Goldring MB, Ninomiya Y, Olsen BR. Molecular cloning of rat and human type (IX) collagen cDNA and localization of the {alpha}1(IX) gene on the human chromosome 6. Eur J Biochem1989;179:71–8.[Medline]
  43. Thomas JT, Cresswell CJ, Rash B et al. The human collagen X gene. Complete primary translated sequence and chromosomal localisation. Biochem J1991;280:617–23.
  44. Bolton MC, Dudhia J, Bayliss MT. Quantification of aggrecan and link protein mRNA in human articular cartilage of different ages by competitive reverse transcriptase–PCR. Biochem J1996;319:489–98.
  45. Dudhia J, Hardingham TE. The primary structure of human cartilage link protein. Nucleic Acids Res1990;18:2214.
  46. Fisher LW, Termine JD, Young MF. Deduced protein sequence of bone small proteoglycan I (biglycan) shows homology with proteoglycan II (decorin) and several non-connective tissue proteins in a variety of species. J Biol Chem1989;264:4571–6.[Abstract/Free Full Text]
  47. Rösl FA. Single and rapid method for the detection of apoptosis in human cells. Nucleic Acids Res1992; 20:5243.
  48. Nishimura I, Muragaki Y, Olsen BR. Tissue specific forms of type IX collagen-proteoglycan arise from the use of two widely separated promoters. J Biol Chem1989;264:20033–41.[Abstract/Free Full Text]
  49. Muragaki Y, Nishimura I, Henney A, Ninomiya Y, Olsen BR. The {alpha}1(IX) collagen gene gives rise to two different transcripts in both mouse embryonic and human fetal RNA. Proc Natl Acad Sci USA1990;87:2400–4.[Abstract/Free Full Text]
  50. Grover J, Roughley PJ. Versican gene expression in human articular cartilage and comparison of mRNA splicing variation with aggrecan. Biochem J1993; 291:361–7.
  51. Sawhney RS, Hering TM, Sandell LJ. Biosynthesis of small proteoglycan II (decorin) by chondrocytes and evidence for a procore protein. J Biol Chem1991;266:9231–40.[Abstract/Free Full Text]
  52. Poole AR, Rosenberg LC, Reiner A, Ionescu M, Bogoch E, Roughley PJ. Contents and distributions of the proteoglycans decorin and biglycan in normal and osteoarthritic human articular cartilage. J Orthop Res1996;14:681–9.[ISI][Medline]
  53. Kokenyesi R, Tan L, Robbins SR, Goldring MB. Proteoglycan production by immortalized human chondrocyte cell lines cultured under conditions that promote expression of the differentiated phenotype. Arch Biochem Biophys2000;383:79–80.[Medline]
  54. Doege K, Hassell JR, Caterson B, Yamada Y. Link protein cDNA sequence reveals tandemly repeated protein structure. Proc Natl Acad Sci USA1986;83:3761–5.[Abstract/Free Full Text]
  55. Preston GA, Lang JE, Maronpot RR, Barrett JC. Regulation of apoptosis by low serum in cells of different stages of neoplastic progression: Enhanced susceptibility after loss of a senescence gene and decreased susceptibility after loss of a tumour suppressor gene. Cancer Res1994;54:4214–23.[Abstract/Free Full Text]
  56. Preston GA, Barrett JC, Biermann JA, Murphy E. Effects of alterations in calcium homeostasis on apoptosis during neoplastic progression. Cancer Res1997; 57:537–42.[Abstract/Free Full Text]
  57. Feng L, Precht P, Balakir R, Horton WE Jr. Evidence of a direct role for bcl-2 in the regulation of articular chondrocyte apoptosis under the conditions of serum withdrawal and retinoic acid treatment. J Cell Biochem1998;71:302–9.[ISI][Medline]
  58. Hashimoto S, Ochs RL, Rosen F et al. Chondrocyte-derived apoptotic bodies and calcification of articular cartilage. Proc Natl Acad Sci USA1998;95:3094–9.[Abstract/Free Full Text]
  59. Cao L, Yang BB. Chondrocyte apoptosis induced by aggrecan G1 domain as a result of decreased adhesion. Exp Cell Res1999;246:527–37.[Medline]