The mandibular condylar growth center: Separation and characterization of the cellular elements

The developing mandibular condylar growth center consists of a number of histologically distinct cell types. There is an increase in cell volume that takes place from the condylar surface layer through the center of ossification, resulting in a disorganized, irregular cellular pattern. Consequently, the isolation and separation of the different cells from this tissue is difficult using standard methodologies. Countercurrent centrifugal elutriation, whereby cells are separated on the basis of size, was applied to bovine mandibular condylar growth center cells. The cell volume, alkaline phosphatase content, proteoglycan synthesis, and type X collagen synthesis all showed a positive correlation with increasing cell size. The largest cells had characteristics that are consistent with hypertrophic chondrocytes; the smallest cells, on the other hand, had many fibroblastic characteristics.     

Electrolytes of isolated epiphyseal chondrocytes,matrix vesicles,and extracellular fluid

Summary Chondrocyte, matrix vesicle, and membrane fractions, as well as interstitial fluid samples from the proliferating and hypertrophic zones of chicken epiphyseal cartilage were analyzed for electrolyte content. Intracellular Ca levels were 1.4–2.1 mM, over 90% of which was nondiffusible. Isolated hypertrophic chondrocytes had higher intracellular Na and lower K than proliferating cells. Matrix vesicles contained 25 to 50 times higher concentrations of Ca than the adjacent cells. Vesicles from the zone of hypertrophy contained twice as much Ca as did those from the proliferating area. Ca/P₁ molar ratios of matrix vesicles were much higher than those of cells or of later mineral deposits. These findings indicate that Ca is concentrated in matrix vesicles during formation, but acuumulation of Ca and P₁ must continue in the matrix. X-ray diffraction of freeze-dried vesicle and membrane fractions failed to detect crystalline apatite, suggesting that crystals seen in electron micrographs of matrix vesicles may be artifacts. Interstitial fluid expressed from epiphyseal cartilage was higher in K, P_i, Mg and nucleotides, and lower in Na and Cl, than blood plasma. Fluid from the hypertrophic zone was higher in K and nucleotides, but not P_i or Mg, than that from the proliferating layer. These data suggest that selective leakage or extrusion of these constituents, which are normally intracellular, must occur, especially in the hypertrophic zone. More of the Ca and Mg, and less of the P_i, was protein-bound in cartilage fluid than in blood plasma. There was more binding of the divalent cations in fluid from proliferating than from hypertrophic cartilage. The presence of greater amounts of ultrafilterable peptides in fluid from hypertrophic than from proliferating cartilage or blood plasma, suggests that proteolytic activity may release bound divalent cations during mineralization.     

c-Myc protein in the rabbit growth plate. Changes in immunolocalisation with age and possible roles from proliferation to apoptosis.

Growth plates taken from five- to 20-week-old Japanese white rabbits were immunostained for c-Myc protein. This was localised both in the proliferating zone and upper hypertrophic zone at five weeks, whereas after ten weeks it was found mostly in the lower hypertrophic zone. The proliferating chondrocytes tended to show nuclear staining and the hypertrophic cells cytoplasmic staining, although the terminal hypertrophic chondrocytes sometimes expressed the protein in their nuclei. In the younger rabbits, c-Myc co-localised with proliferating cell nuclear antigen, whereas in the hypertrophic zone of older rabbits, it was present in some chondrocytes the nuclei of which also contained DNA breaks. Our study suggests that, in the rabbit growth plate, c-Myc is associated with different cellular processes, depending on the age and the developmental stage of the chondrocytes.     

Physiological death of hypertrophic chondrocytes

OBJECTIVE: Post-proliferative chondrocytes in growth cartilage are present in two forms, light and dark cells. These cells undergo hypertrophy and die by a mechanism that is morphologically distinct from apoptosis, but has not been characterized. The aims of the current study were to document the ultrastructural appearance of dying hypertrophic chondrocytes, and to establish a culture system in which the mechanism of their death can be examined. DESIGN: Growth cartilage from fetal and growing postnatal horses was examined by electron microscopy. Chondrocytes were isolated from epiphyseal cartilage from fetal horses and grown in pellet culture, then examined by light and electron microscopy, and quantitative polymerase chain reaction. RESULTS: In tissue specimens, it was observed that dying dark chondrocytes underwent progressive extrusion of cytoplasm into the extracellular space, whereas light chondrocytes appeared to disintegrate within the cellular membrane. Pellets cultured in 0.1% fetal calf serum (FCS) contained dying light and dark chondrocytes similar to those seen in vivo. Transforming growth factor-beta1 or 10% FCS increased the proportion of dark cells and induced cell death. Triiodothyronine increased the differentiation of dark and light cells and induced their death. Dark cells were associated with higher levels of matrix metalloproteinase-13 expression than light cells, and light cells were associated with higher levels of type II collagen expression. CONCLUSIONS: Light and dark hypertrophic chondrocytes each undergo a distinctive series of non-apoptotic morphological changes as they die. Pellet culture can be used as a model of the two forms of physiological death of hypertrophic chondrocytes.     

Activation of transforming growth factor beta in chondrocytes undergoing endochondral ossification.

Transforming growth factor beta (TGF-beta) has well-documented roles in chondrocyte maturation and endochondral ossification, but the mechanisms of TGF-beta activation during these processes remain unclear. In this study, we analyzed TGF-beta activation in chick embryo resting, proliferating, and hypertrophic chondrocytes in culture. We found that both levels and activation of TGF-beta increased substantially with maturation. The majority of TGF-beta produced by resting cells over culture time remained latent, but a larger portion produced by proliferating and hypertrophic cells was activated with increasing maturation. Zymography of gelatin gels revealed that matrix metalloprotease 2 (MMP-2) and MMP-9 were expressed by each population and that MMP-13 characterized hypertrophic chondrocytes and to a lesser extent proliferating chondrocytes in late cultures. Treatment with pharmacologic agents revealed that both MMPs and serine proteases are involved in activation. However, because inhibition of MMPs almost completely prevented TGF-beta activation, MMPs appear crucial for activation. During culture, inclusion of the tetracycline-derived, collagenase/gelatinase inhibitor chemically modified nonantimicrobial tetracycline (CMT-8) at concentrations specific for MMP-13 inhibition resulted in complete inhibition of TGF-beta activation by proliferating and hypertrophic chondrocytes. These results show that TGF-beta production, release, and activation are regulated developmentally in chondrocytes. Our findings point to a strict mode of regulation of this potent factor to elicit diverse and highly specific effects during chondrocyte maturation and ossification.     

Connective tissue growth factor mRNA expression pattern in cartilages is associated with their type I collagen expression

Connective tissue growth factor (CTGF) has been identified as a secretory protein encoded by an immediate early gene and is a member of the CCN family. In vitro CTGF directly regulates the proliferation and differentiation of chondrocytes; however, a previous study showed that it was localized only in the hypertrophic chondrocytes in the costal cartilages of E 18 mouse embryos. We described the expression of CTGF mRNA and protein in chondrocytes of different types of cartilages, including femoral growth plate cartilage, costal cartilage, femoral articular cartilage, mandibular condylar cartilage, and cartilage formed during the healing of mandibular ramus fractures revealed by in situ hybridization and immunohistochemistry. To characterize the CTGF-expressing cells, we also analyzed the distribution of the type I, type II, and type X collagen mRNA expression. Among these different types of cartilages we found distinct patterns of CTGF mRNA and protein expression. Growth plate cartilage and the costal cartilage showed localization of CTGF mRNA and protein in the hypertrophic chondrocytes that expressed type X collagen mRNA with less expression in proliferating chondrocytes that expressed type II collagen mRNA, whereas it was also expressed in the proliferating chondrocytes that expressed type I collagen mRNA in the condylar cartilage, the articular cartilage, and the cartilage appearing during fracture healing. In contrast, the growth plate cartilages or the costal cartilages were negative for type I collagen and showed sparse expression of CTGF mRNA in the proliferating chondrocytes. We found for the first time that CTGF mRNA could be differentially expressed in five different types of cartilage associated with those expressing type I collagen. Moreover, the spatial distribution of CTGF mRNA in the cartilages with type I collagen mRNA suggested its roles in the early differentiation, as well as in the proliferation and the terminal differentiation, of those cartilages.     

Growth and development of the human cricoid cartilage: an immunohistochemical analysis of the maturation sequence of the chondrocytes and surrounding cartilage matrix.

OBJECTIVES: The goal was to determine maturational changes in the human cricoid cartilage. Study Design: The study involved immunohistochemical staining of collagen II (a marker of proliferating chondrocytes), matrilin-1 (a marker of post-proliferative chondrocytes), and collagen X (a marker of hypertrophic chondrocytes). Specimens included uninjured human cricoid cartilages at 18 and 41 weeks' gestation and 1, 4, and 13 years postpartum. RESULTS: This study demonstrated that type II collagen peaks in concentration at approximately 41 weeks' gestation. Matrilin-1 is present in progressively lower concentration in the central core of the cricoid ring, but the peripheries of the ring contain the protein in relatively high concentration. Type X collagen is not expressed in the age groups tested. CONCLUSIONS: These biochemical markers lend further support to a chondrocyte proliferative phase that slows between 1 and 4 years of age. Chondrocytes then enter a phase histologically similar to the hypertrophic phase but are biochemically different than hypertrophic chondrocytes destined for endochondral ossification.     

Parathyroid hormone [PTH(1-34)] and parathyroid hormone-related protein [PTHrP(1-34)] promote reversion of hypertrophic chondrocytes to a prehypertrophic proliferating phenotype and prevent terminal differentiation of osteoblast-like cells.

The effects of parathyroid hormone/parathyroid hormone-related protein (PTH/PTHrP) on late events in chondrocyte differentiation were investigated by a dual in vitro model where conditions of suspension versus adhesion culturing are permissive either for apoptosis or for the further differentiation of hypertrophic chondrocytes to osteoblast- like cells. Chick embryo hypertrophic chondrocytes maintained in suspension synthesized type II and type X collagen and organized their extracellular matrix, forming a tissue highly reminiscent of true cartilage, which eventually mineralized. The formation of mineralized cartilage was associated with the expression of alkaline phosphatase (ALP), arrest of cell growth, and apoptosis, as observed in growth plates in vivo. In this system, PTH/PTHrP was found to repress type X collagen synthesis, ALP expression, and cartilage matrix mineralization. Cell proliferation was resumed, whereas apoptosis was blocked. Hypertrophic chondrocytes cultured in adherent conditions in the presence of retinoic acid underwent further differentiation to osteoblast-like cells (i.e., they resumed cell proliferation, switched to type I collagen synthesis, and produced a mineralizing bone-like matrix). In this system, PTH addition to culture completely inhibited the expression of ALP and matrix mineralization, whereas cell proliferation and expression of type I collagen were not affected. These data indicate that PTH/PTHrP inhibit both the mineralization of a cartilage-like matrix and apoptosis (mimicked in the suspension culture) and the production of a mineralizing bone-like matrix, characterizing further differentiation of hypertrophic chondrocytes to osteoblasts like cells (mimicked in adhesion culture). Treatment of chondrocyte cultures with PTH/PTHrP reverts cultured cells in states of differentiation earlier than hypertrophic chondrocytes (suspension), or earlier than mineralizing osteoblast-like cells (adhesion). However, withdrawal of hormonal stimulation redirects cells toward their distinct, microenvironment-dependent, terminal differentiation and fate.     

Cellular basis of decreased rate of longitudinal growth of bone in pseudoachondroplastic dogs.

Regulation of growth of long bones occurs in cartilage growth plates, where proliferation of chondrocytes, matrix synthesis, and an increase in vertical height in the direction of growth all contribute to the final length of a bone. In this study, we tested the hypothesis that an increase in chondrocytic vertical height is a major variable that accounts for the decreased rate of growth of long bones in Scottish deerhound dogs that had pseudoachondroplasia. The diagnosis of pseudoachondroplasia is based, primarily, on the demonstration of alternating electron-dense and electron-lucent lamellae with a periodicity of 100 to 150 nanometers in dilated rough endoplasmic reticulum. These ultrastructural changes are similar to those seen in humans who have pseudoachondroplasia. In Scottish deerhounds that have the disease, growth of bone is approximately 65 per cent of that in normal animals. There were striking differences in the diameters of proliferating and hypertrophic chondrocytes in pseudoachondroplastic animals compared with normal animals. Specifically, the horizontal diameter of proliferating chondrocytes was 22.7 micrometers in normal animals and 11.3 micrometers in pseudoachondroplastic animals. The vertical diameter of proliferating chondrocytes was 4.8 and 7.6 micrometers in normal and pseudoachondroplastic animals. In the distal 100 micrometers of the hypertrophic zone, the mean horizontal diameter of hypertrophic chondrocytes was 29.6 and 19.1 micrometers and the mean vertical diameter was 22.8 and 18.6 micrometers in normal and pseudoachondroplastic animals. All these differences were statistically significant. The changes in vertical height resulted in a significant difference in the incremental difference in vertical height between chondrocytes from the proliferative and hypertrophic zones in normal animals (18.0 micrometers per chondrocyte) and pseudoachondroplastic animals (11.0 micrometers per chondrocyte). Each chondrocyte in the abnormal plates achieved only 61 per cent of the incremental difference of chondrocytes in normal plates. The mean cellular volume of chondrocytes in the hypertrophic zone was 13,050 cubic micrometers in the normal animals and 10,740 cubic micrometers in the pseudoachondroplastic animals. This difference was not statistically significant. These results are discussed in relation to current theories of the role of the shape and change in volume of chondrocytes in the regulation of longitudinal growth of bone.(ABSTRACT TRUNCATED AT 400 WORDS)     

The role of hepatocyte growth factor in growth plate cartilage

To investigate the physiological role of hepatocyte growth factor (HGF) in endochondral bone formation, we examined the expression of HGF and its receptor c-met and the effects of HGF on growth plate chondrocytes. HGF was highly expressed in the prehypertrophic zone and hypertrophic zone in rat costal growth plate cartilage. The expression of HGF increased in rabbit chondrocytes as they matured in culture. Conversely, c-met expression was down regulated along maturation of growth plate chondrocytes. HGF had weak stimulatory effects on DNA and proteoglycan synthesis of growth plate chondrocytes. However, HGF strongly inhibited expression of terminal differentiation-related phenotypes, such as type X collagen and alkaline phosphatase (APase) synthesis and cartilage matrix mineralization. When HGF was removed from the cultures, cells quickly expressed type X collagen and APase. Once chondrocytes differentiated to mature chondrocytes, HGF did not inhibit further differentiation of these cells. These results suggested that HGF is a negative regulator of terminal differentiation of growth plate chondrocytes.. Received: Feb. 12, 1998 / Accepted: March 12, 1998     

Transforming growth factor-β1 and fibroblast growth factors in rat growth plate

Chondrocytes in the growth plate progress in an orderly fashion from resting through proliferating to hypertrophic cells. In the region of hypertrophic chondrocytes, the cartilage is invaded by capillary loops and endochondral ossification is initiated. It is currently believed that growth factors may regulate the proliferation and maturation of chondrocytes and the synthesis of extracellular matrix in the growth plate. The ordered sequence of proliferation and differentiation observed in the growth plate provides a unique opportunity to study the role of acidic fibroblast growth factor, basic fibroblast growth factor, and transforming growth factor-β1 in the regulation of these processes. In this study, expression of the mRNA of these growth factors was examined using total RNA extracted from the physis and epiphysis of rat tibias. Transforming growth factor-β1 mRNA was detected by Northern hybridization. Expression of the genes encoding acidic and basic fibroblast growth factors was demonstrated by polymerase chain reaction amplification. In addition, using polyclonal antibodies against these growth factors, we localized them by immunohistochemical analysis. Strong intracellular staining with a predominantly nuclear pattern was observed in chondrocytes from the proliferating and upper hypertrophic zones. In contrast, chondrocytes in the resting zone stained only faintly for the presence of these growth factors. Some chondrocytes in the resting zone adjacent to the proliferating zone stained with these antibodies, and the antibodies also stained cells in the zone of Ranvier, which regulates latitudinal bone growth. Lastly, the location of transforming growth factor-β1 was examined further with use of a polyclonal antipeptide antibody specific for its extracellular epitope. Interestingly, extracellular staining for transforming growth factor-β1 was observed only around chondrocytes in the hypertrophic zone. These results suggest a role for these growth factors in the regulation of proliferation and maturation of chondrocytes and in endochondral ossification.     

Separation of allostimulatory and natural suppressor/stem cell functions of murine bone marrow--implications for bone marrow transplantation.

Natural suppressor (NS) activity is defined by the ability of a null cell population to inhibit antigenic and mitogenic activation of lymphoid cells. We had previously demonstrated that murine and human bone marrow was a rich source of NS activity and could be enriched in a population of large, low-density cells after counterflow centrifugal elutriation (CCE). In this report we confirm previous findings in the mixed lymphocyte reaction (MLR) that NS activity is an endogenous function of BM and is not present in peripheral immune tissues. Furthermore, suppression by BM NS cells in the MLR is not MHC restricted, which distinguishes NS activity from veto activity also associated with BM. To enrich for NS activity, BM cells were separated into 3 fractions according to size and density by CCE. Fraction (F) 1 contained small, high-density cells; F2 contained cells of intermediate size and density; and F3 comprised large, low-density cells. Addition of CCE-enriched fractions to the MLR revealed potent NS activity associated with F3, consistent with our previous findings in the Ab response. CCE-enriched fractions of BM cells were also used as stimulators in an MLR to determine which fraction, if any, could support alloantigen-induced proliferation. The use of unfractionated C57Bl/6 (H-2b) BM cells as stimulators for BALB/c (H-2d) splenic responders resulted in little detectable proliferation as compared with that induced by C57Bl/6 splenic stimulators. However, when elutriated fractions were used as stimulators, F1 BM cells induced a significant proliferative response, albeit to a lesser magnitude than spleen cells. In order to determine which fraction contained hematopoietic progenitors, CCE-separated cells were assayed for granulocyte/macrophage colony formation (CFU-GM). F3 BM was enriched five-fold for CFU-GM progenitors as compared with unfractionated BM, while progenitors were virtually absent in F1 and F2 BM. The results suggest that CCE may represent an effective way of removing potential alloantigen-presenting cells (F1) from BM, while retaining hematopoietic progenitors and NS activity (F3). The implications of the findings as they relate to allogeneic BM transplantation are discussed.     

Beneficial effect of basic fibroblast growth factor on the repair of full-thickness defects in rabbit articular cartilage

The effects of exogenous basic fibroblast growth factor (bFGF) on the repair of full-thickness cartilage defects were examined. Four-millimeter diameter, cylindrical defects were made in rabbit articular cartilage and were filled with human recombinant bFGF. The addition of bFGF to the defect induced the formation of a thick cartilage layer composed of chondrocytes and a metachromatic-stained matrix after 6 weeks. The score of the bFGF-treated tissue, as evaluated by a semiquantitative histological scale, was significantly higher than that of the untreated tissue. At 24 weeks, the cartilage-like matrix that contained the proteoglycans and type II collagen was thicker in the bFGF-treated tissue than in the untreated tissue. Immunohistochemical analysis of the tissues at 6–12 weeks with an anti-bFGF monoclonal antibody suggested that a single application of bFGF increased the number of differentiating chondrocytes that synthesized bFGF at a high level. In contrast, immunostaining of the tissues at 6–12 weeks with a monoclonal antibody against proliferating cell nuclear antigen showed that the number of proliferating cells in the bFGF-treated tissue was fewer than in the untreated tissue. These findings suggest that administration of bFGF into cartilagenous defects promotes the differentiation of chondrocytes and their matrix synthesis, and that this growth factor is useful for improving cartilage repair. Received: 18 May 1998     

Thyroid hormones regulate hypertrophic chondrocyte differentiation and expression of parathyroid hormone-related peptide and its receptor during endochondral bone formation.

Hypothyroidism in children causes developmental abnormalities in bone and growth arrest, while thyrotoxicosis accelerates growth rate and advances bone age. To determine the effects of thyroid hormones on endochondral bone formation, we examined epiphyseal growth plates in control, hypothyroid, thyrotoxic, and hypothyroid-thyroxine (hypo-T4)-treated rats. Hypothyroid growth plates were grossly disorganized, contained an abnormal matrix rich in heparan sulfate, and hypertrophic chondrocyte differentiation failed to progress. These effects correlated with the absence of collagen X expression and increased parathyroid hormone-related protein (PTHrP) messenger RNA (mRNA) expression. In thyrotoxic growth plates, histology essentially was normal but PTHrP receptor (PTHrP-R) mRNA was undetectable. PTHrP is a potent inhibitor of hypertrophic chondrocyte differentiation that acts in a negative feedback loop with the secreted factor Indian hedgehog (Ihh) to regulate endochondral bone formation. Thyroid hormone receptor alpha1(TRalpha1), TRalpha2, and TRbeta1 proteins were localized to reserve zone progenitor cells and proliferating chondrocytes in euthyroid rat cartilage; regions in which PTHrP and PTHrP-R expression were affected by thyroid status. Thus, dysregulated Ihh/PTHrP feedback loop activity may be a key mechanism that underlies growth disorders in childhood thyroid disease.     

Development of Avian Tibial Dyschondroplasia: Gene Expression and Protein Synthesis

Age-dependent gene expression and protein synthesis associated with chondrocyte differentiation were evaluated in the epiphyseal growth plates of normal and tibial dyschondroplasia (TD)-afflicted chickens. In the normal growth plate, collagen type II gene is expressed mainly by chondrocytes at the upper zone of the growth plate and by the chondrocytes in the articular cartilage. Collagen type X and osteopontin (OPN) genes are expressed in the lower zone of the growth plate and in the zone of cartilage-to-bone transition. No age-dependent changes in the pattern of OPN and collagen type II or X gene expression were observed up to 20 days of age. In the TD-afflicted growth plates, the lesion is enlarged with age, and chondrocytes expressing the collagen type II gene were observed in the hypertrophic zone as early as 8 days posthatching. Abnormal expression of OPN and collagen type X genes was also observed starting at 13 days of age. At day 20, the entire TD lesion—which was significantly enlarged—was surrounded by collagen type II, collagen type X, and OPN expressing cells. The level of OPN in TD was reduced with increasing age, and at 20 days almost no OPN could be detected in either the upper or the lower hypertrophic zones. The level of bone sialoprotein (BSP) also diminished with increasing age in the TD growth plates. In contrast to OPN, the age-dependent reduction in BSP levels was mainly in the lower hypertrophic zone (LHZ), and at 20 days of age, BSP was barely detected in the LHZ, whereas in the upper hypertrophic zone, the levels of BSP were similar to those in normal growth plate. In summary, our results suggest that the primary event of the TD lesion occurs in cells of proliferative phenotype within the hypertrophic zone. These cells divide and form the TD lesion, which consists of cells that do not express the genes associated with hypertrophy. Received: 11 June 1997 / Accepted: 11 May 1998     

Retinoid signaling regulates CTGF expression in hypertrophic chondrocytes with differential involvement of MAP kinases.

Retinoids are important for growth plate chondrocyte maturation, but their downstream effectors remain unclear. Recently, CTGF (CCN2) was found to regulate chondrocyte function, particularly in the hypertrophic zone. The goal of the study was to determine whether CTGF is a retinoid signaling effector molecule, how it is regulated, and how it acts. INTRODUCTION: Using a combination of in vivo and in vitro approaches, we carried out a series of studies at the cellular, biochemical, and molecular level to determine whether and how retinoid signaling is related to expression and function of connective tissue growth factor (CTGF) in chondrocyte maturation and endochondral ossification. MATERIALS AND METHODS: Limbs of chick embryos in ovo were implanted with retinoic pan-antagonist RO 41-5253-filled beads, and phenotypic changes were assessed by in situ hybridization. CTGF gene expression and roles were tested in primary cultures of immature and hypertrophic chondrocytes. Cross-talk between retinoid signaling and other pathways was tested by determining endogenous levels of active ERK1/2 and p38 MAP kinases and phenotypic modulations exerted by specific antagonists of mitogen-activated protein (MAP) kinases and BMP signaling (Noggin). RESULTS: Interference with retinoid signaling blocked expression of CTGF and other posthypertrophic markers in long bone anlagen in vivo and hypertrophic chondrocyte cultures, whereas all-trans-retinoic acid (RA) boosted CTGF expression and even induced it in immature proliferating cultures. Exogenous recombinant CTGF stimulated chondrocyte maturation, but failed to do so in presence of retinoid antagonists. Immunoblots showed that hypertrophic chondrocytes contained sizable levels of phosphorylated ERK1/2 and p38 MAP kinases that were dose- and time-dependently increased by RA treatment. Experimental ERK1/2 inhibition led to a severe drop in baseline and RA-stimulated CTGF expression, whereas p38 inhibition increased it markedly. These responses were gene-specific, because the opposite was seen with other hypertrophic chondrocyte genes such as collagen X and RA receptor gamma (RARgamma). Tests with Noggin showed that RA induction of CTGF expression was negatively influenced by BMP signaling, whereas induction of collagen X expression was BMP-dependent. CONCLUSIONS: Retinoids appear to have a preeminent role in controlling expression and function of CTGF in hypertrophic and posthypertrophic chondrocytes and do so with differential cooperation and intervention of MAP kinases and BMP signaling.