---

Discussion

• Abstract
• Introduction
• References

Endochondral ossification consists of a highly complex sequence of proliferation, differentiation, and tissue remodeling events and includes several distinct steps of late cartilage differentiation. Each of the steps is subject to positive and negative control by environmental signals, which interfere at distinct checkpoints. Several anatomically distinct hyaline cartilage tissues can undergo late differentiation and can sequentially express markers for each stage. However, this inherent capacity may not become manifest under normal conditions, because it is suppressed by a powerful environmental control mechanism.

Under pathological conditions, however, this arrest may be released. For example, osteoarthritis may well result from inappropriate hypertrophy of articular cartilage (von der Mark et al., 1992).

The consequence is degeneration and loss or remodeling of the tissue that would be normal in other circumstances, such as in endochondral ossification of callus tissue during the repair of bone fractures. Our results clearly show that hypertrophic development takes place in costal cartilage well before the tissue is mineralized and ossified. This implies a novel checkpoint of negative control downstream of the expression of the overtly hyper-trophic cartilage
phenotype.
Collagen X is a well-recognized marker of terminal chondrocyte differentiation.
The protein has been found in many tissues undergoing endochondral ossification, including rib growth plates (Remington et al., 1983, 1984; Grant et al., 1985). In costal cartilage, however, the in situ occurrence of collagen X has escaped previous detection by immunohistology (Claassen et al., 1996), even though this tissue undergoes ossification very slowly. However, we have previously found evidence for the protein made in vitro by costal cartilage cells derived from funnel-chest patients (Erdmann et al., 1996).

As shown here, the protein was recovered from rib cartilage of very young children and was identified by immunoblotting and immunohistochemistry.

Its amounts increased to maximal levels well before puberty, i.e. before the onset of alkaline pbosphatase expression and slow mineralization and ossification. This clear separation in time of the expression of the two hypertrophy markers has not been found previously in any system undergoing late chondrocyte differentiation.

This necessitates the existence of independent control mechanisms for the activation of the genes for collagen X and alkaline phosphatase.

The function of collagen X still is not entirely understood. The protein is a prominent component of hypertrophic cartilage and is specific for this tissue, but its suprastructural organization is still incompletely understood. Collagen VIII, a structural homo-logue of collagen X, associates into hexagonal networks in corneal Descemet’s membrane (Sawada et al,, 1990), and collagen X can form similar suprastruc-tures in vitro (Kwan et al., 1991). In addition, collagen X may associate with cartilage fibrils, thus modulating the suprastructure of hypertrophic cartilage matrix (Chen et al., 1992).

Studies on gene defects further support the role of collagen X as a structural component. Mutations in the human collagen X-gene can cause Schmid metaphyseal chondrodysplasia (Dhar-mavaram et al., 1994; Mclntosh et al., 1994; Wallis et al., 1994) and mice with mutated or absent collagen X show disturbed matrix compartmentalization during endochondral ossification.

This results in abnormal formation of marrow cavities in long bones, which, in turn, impairs ematopoiesis (Kwan et al., 1997; Chan and Jacenko, 1998).
In addition, collagen X may regulate mineral deposition in hypertrophic cartilage. The protein reportedly modulates the calcium flux in matrix vesicles through binding to extracellular annexin V (Kirsch et al., 1997).

Our observations, however, imply that collagen X expression is not sufficient for cartilage mineralization and ossification. Only after the appearance of alkaline phosphatase activity can mineral deposition take place.
The studies on costal cartilage presented here have also afforded new insights into environmental
signals controlling late cartilage differentiation.
PTH may positively or negatively control chondrocyte hypertrophy, depending on the stage at which it interferes. The aminoterminal fragment, comprising residues 1-34 of the hormone, achieves negative signaling through cell-surface PTH/PTHrP receptors that also recognize the homologous region PTH-related protein (PTHrP) (Iwamoto et al., 1995; O’Keefe et al., 1997;
Fitzpatrick and Bilezikian, 1999; Karaplis and Goltzman, 1999).

Hedgehog-induced, perichondrially derived PTHrP prevents progression of growth-plate chondrocytes from a late proliferative stage towards a hypertrophycompetent stage (Vortkamp et al., 1996). As shown here, costal chondrocytes have advanced in situ to post-proliferative stages of terminal  if ferentiation and respond inversely towards this signal. Under the direction of the aminoterminal portion of PTH, the cells increased expression of collagen X, but not alkaline phosphatase. By contrast, the carboxyterminal portion of the hormone, acting through receptoKs) distinct from the PTH/PTHrP receptors, stimulated the production of both hypertrophy markers.

A comparable response has been reported previously for osteocarcinoma cells in culture
(Murray et al., 1991). Studies on the distinct responses to the PTH segments and their intracellular signaling cascade are in progress.

Fig. 5. Mo st co stal cho ndro c ytes in culture are p a st their proliferative p ha se. P ha se co ntra st-micro grap hs are shown of represe ntative fields revisited thro ugho ut the entire c ulture p eriod , as indic ated at the top of the figure. The med ia were sup plemented by spe cific signals or FBS, as indica ted on the le ft ma rgin. Panels A-D were d erived from a serum-free co ntrol culture (DMEM).

Note: isolated cells exhibit proliferation capa city in re spo nse to FBS (arrows in p anels Q-R). The ce lls uniformly incre ase their size thro ugho ut t h e cultures und er the d ire ction of IGF -1 or 10% FBS (arrowhe ads in panels E -H and Q-T).

Fig. 6. Hypertrophy of costal chondrocytes in agarose culture, as revealed by collagen X synthesis, is stimulated by parathyroid hormone. Costal chondrocytes were cultured in agarose gels at a density of 1.5 ´ 106 cells / ml for 2 weeks. The medium was DMEM and, during the last 48 h, contained 10 -8 M PTH (1-34) (lane 2), 10 -8 M PTH (53-84) (lane 3), or 10% FBS (lane 4). Serum-free control, lane 1. Collagens were extracted by limited digestion with pepsin and were subjected to electrophoresis on a 4.5-15% polyacrylamide gradient gel. Immunoblots with a polyclona! serum specific for human collagen X are shown.

Fig. 7. Hypertrophy of costal chondrocytes in agarose culture, as revealed by alkaline phosphatase activity, is stimulated by the carboxyterminal region of parathyroid hormone. Chondrocytes were cultured for the times indicated and alkaline phosphatase activity in culture media was determined by hydrolysis of p-nitrophenyl phosphate (arbitrary units).

The culture conditions were: full circles, 10% of fetal bovine serum; full squares, 10 -8 M PTH (53 84); and open symbols, 10 -8 M PTH (1-34), 100 ng / ml of IGF-1, 50 ng / ml of thyroxine, or serum-free control.

Leave a Reply

You must be logged in to post a comment.