Annexin-mediated Ca2+ Influx Regulates Growth Plate Chondrocyte Maturation and Apoptosis*

Maturation of epiphyseal growth plate chondrocytes plays an important role in endochondral bone formation. Previously, we demonstrated that retinoic acid (RA) treatment stimulated annexin-mediated Ca2+ influx into growth plate chondrocytes leading to a significant increase in cytosolic Ca2+, whereas K-201, a specific annexin Ca2+channel blocker, inhibited this increase markedly. The present study addressed the hypothesis that annexin-mediated Ca2+ influx into growth plate chondrocytes is a major regulator of terminal differentiation, mineralization, and apoptosis of these cells. We found that K-201 significantly reduced up-regulation of expression of terminal differentiation marker genes, such as cbfa1, alkaline phosphatase (APase), osteocalcin, and type I collagen in RA-treated cultures. Furthermore, K-201 inhibited up-regulation of annexin II, V, and VI gene expression in these cells. RA-treated chondrocytes released mineralization-competent matrix vesicles, which contained significantly higher amounts of annexins II, V, and VI as well as APase activity than vesicles isolated from untreated or RA/K-201-treated cultures. Consistently, only RA-treated cultures showed significant mineralization. RA treatment stimulated the whole sequence of terminal differentiation events, including apoptosis as the final event. After a 6-day treatment gene expression ofbcl-2, an anti-apoptotic protein, was down-regulated, whereas caspase-3 activity and the percentage of TUNEL-positive cells were significantly increased in RA-treated cultures compared with untreated cultures. Interestingly, the cytosolic calcium chelator BAPTA-AM and K-201 protected RA-treated chondrocytes from undergoing apoptotic changes, as indicated by higher bcl-2 gene expression, reduced caspase-3 activity, and the percentage of TUNEL-positive cells. In conclusion, annexin-mediated Ca2+influx into growth plate chondrocytes is a positive regulator of terminal differentiation, mineralization, and apoptosis events in growth plate chondrocytes.

Maturation of epiphyseal growth plate chondrocytes, which plays an important role during endochondral ossification, is accompanied by major changes of chondrocyte morphology, biosynthetic activities, and energy metabolism. These processes involve an ordered progression of various cell differentiation stages, including proliferation, hypertrophic differentiation, terminal differentiation, and ultimately programmed cell death (apoptosis) (1,2). During normal development these sequential events are under the strict control of local and systematic factors such as hormones and growth factors. If these processes, however, occur during pathological conditions, they can result in serious cartilage or bone defects. Evidence of endochondral ossification is also seen during osteophyte formation in osteoarthritic cartilage (3,4). Terminal differentiation of growth plate chondrocytes is an essential process, which primes the cartilage skeleton for its subsequent invasion by osteoblasts and its replacement by a bone matrix. Despite the obvious importance of these terminal differentiation events still little is known about mechanisms regulating these processes.
cbfa1, a member of the runt domain family of transcription factors, was originally discovered as a key transcription factor, which controls osteoblast differentiation. In cbfa1-null mice no endochondral and intramembranous bone formation occurs due to an arrest in osteoblast differentiation (5)(6)(7)(8). Recent studies have indicated that cbfa1 also plays an important regulative role in terminal chondrocyte maturation. Transgenic mice, which overexpress cbfa1 in non-hypertrophic chondrocytes, display an acceleration of endochondral ossification. Overexpression of cbfa1 in chondrocytes of cbfa1-null mice partially rescued the abnormalities of cbfa1-null mutant mice. In particular, it rescued hypertrophic chondrocyte differentiation in the humerus and femur (9). Thus, cbfa1 seems to play dual functions in endochondral bone formation; it plays a key role in osteoblast differentiation from mesenchymal precursor cells, and it has the ability to stimulate hypertrophic and terminal chondrocyte differentiation.
Chondrocyte hypertrophy and terminal differentiation are accompanied by an increase in cytosolic calcium, [Ca 2ϩ ] i (10 -12). Calcium is recognized as an important regulator of many cellular processes, and it controls a diverse range of cell functions, including adhesion, motility, gene expression, cell differentiation, and proliferation. For example, the amplitude and duration of calcium signals control differential activation of different transcription factors in B lymphocytes (13). Calcium has been shown to play several roles in vesiculation and the formation of vesicles. For example, Iannotti et al. (14) have shown a correlation between increasing [Ca 2ϩ ] i and the release of matrix vesicles (14). Matrix vesicles are small membraneenclosed particles, which are released from the plasma membrane of growth plate chondrocytes and which initiate the mineralization process (15). We have previously shown that RA, 1 which stimulates terminal differentiation and mineraliza-tion of hypertrophic chondrocytes, induces Ca 2ϩ influx into growth plate chondrocytes causing the release of mineralization-competent matrix vesicles (16).
Annexins II, V, and VI, which are highly expressed in hypertrophic and mineralizing growth plate cartilage, are major components of matrix vesicles (17)(18)(19). Annexins II, V, and VI belong to a family of Ca 2ϩ -and phospholipid-binding proteins. In addition, annexins II, V, and VI have been shown to form Ca 2ϩ channels in phospholipid bilayers or in liposomes (20). They also form Ca 2ϩ channels in matrix vesicles enabling Ca 2ϩ influx into these particles as a possible initial step for the formation of the first mineral phase within the vesicle lumen (21).
Careful studies have provided evidence that apoptosis is the final fate of terminally differentiated growth plate chondrocytes. Chondrocyte apoptosis in the growth plate is centered at the site of the transition of cartilage to bone (2,22). These apoptotic chondrocytes show characteristic hallmarks of apoptosis, including condensed nuclei, DNA fragmentation, activation of caspase cascade, and phosphatidylserine externalization. Previous studies have indicated that elevation of [Ca 2ϩ ] i is involved in the induction of apoptosis. For example, apoptosis of cultured human endothelial cells was inhibited by chelating extracellular calcium with EGTA or by inhibiting the calcium influx by calcium channel blockers. It has been suggested that elevated [Ca 2ϩ ] i leads to activation of proteases, lipases, and nucleases. All these actions can contribute to cell death (23)(24)(25).
We have provided evidence that RA promotes annexin channel formation in growth plate chondrocytes and that annexinmediated Ca 2ϩ influx into these cells controls Ca 2ϩ homeostasis (16). To determine the role of annexin-mediated alteration of Ca 2ϩ homeostasis in terminal differentiation, mineralization, and apoptosis of growth plate chondrocytes, we cotreated growth plate chondrocytes isolated from the hypertrophic zone of day 19 embryonic chicken growth plate cartilage with RA and K-201, a specific annexin channel blocker, or RA and BAPTA-AM, a cytosolic Ca 2ϩ chelator, and analyzed the rate of terminal differentiation, mineralization, and apoptosis in these cells.
Isolation of Matrix Vesicles-Matrix vesicles were isolated from chondrocyte cultures after a 6-day treatment by enzymatic digestion and ultracentrifugation as described previously (19).
Measurement of APase Activity and Protein Content-APase activity was measured using p-nitrophenyl phosphate (Sigma-Aldrich) as a substrate as described previously (19). Protein content was analyzed by the BCA protein assay from Pierce.
SDS-PAGE and Immunoblotting-To determine the amounts of annexin II, V, and VI in matrix vesicles, vesicle fractions (total protein of 30 g) were subjected to SDS-PAGE and immunoblotted with primary antibodies specific for annexins II, V, and VI. Samples were dissolved in 4ϫ NuPAGE SDS sample buffer (Invitrogen). Prior to electrophoresis, the reducing reagent was added to the sample solution, denatured at 70°C for 10 min, and analyzed by electrophoresis in 10% Bis-Tris gels following the NuPAGE electrophoresis protocols. Samples were electroblotted onto nitrocellulose filters after electrophoresis. After blocking with a solution of low fat milk protein, blotted proteins were immunostained with primary antibodies followed by peroxidase-conjugated secondary antibody, and the signal was detected by enhanced chemiluminescene (Pierce).
Alizarin Red S Staining-To determine the degree of mineralization chondrocyte cultures were stained with alizarin red S after a 6-day treatment as described previously (16). Briefly, chondrocyte cultures were fixed with 70% ethanol and then stained with 0.5% alizarin red S solution, pH 4.0, for 5 min at room temperature. To quantify the intensity of alizarin red S staining, alizarin red S-stained cultures were incubated with 100 mM cetylpyridinium chloride for 1 h to solubilize and release calcium-bound alizarin red S into solution (26). The absorbance of the released alizarin red S staining was measured at 570 nm using a spectrophotometer. Data were expressed as units of alizarin red S released per mg of protein in each culture.
Caspase-3 Activity Assay-Caspase-3 activity was determined using the ApoAlert caspase fluorescent assay kit (Clontech) following the manufacturer's protocol. Briefly, after a 6-day treatment chondrocyte cultures were washed twice with ice-cold phosphate-buffered saline (PBS), scraped into tubes, and centrifuged at 1500 rpm for 10 min. Cell pellets were washed one more time with ice-cold PBS and centrifuged again. Then air-dried cell pellets were resuspended in 60 l of chilled cell lysis buffer and incubated on ice for 10 min. Cellular debris was removed by centrifugation, and 50 l of 2 ϫ reaction buffer/dithiothreitol mixture and 5 l of 1 mM caspase-3 substrate (DEVD-7-amino-4trifluoromethylcoumarin) were added to 50 l of each sample and incubated for 1 h at 37°C. Caspase-3 activity was measured in a fluorimeter (Photon Technology Instruments) using the excitation wavelength of 400 nm and the emission wavelength of 505 nm. Caspase-3 activity was quantitated using 7-amino-4-trifluoromethylcoumarin standard and normalized to the protein content in each culture.
In Situ Detection of Apoptotic Chondrocytes by TUNEL Labeling-Apoptotic chondrocytes in day 6 chondrocyte cultures were detected using ApopTag in situ apoptosis detection kit to label apoptotic cells by modifying genomic DNA utilizing terminal deoxynucleotidyltransferase (TdT) (Intergen Co., Purchase, NY). Briefly, chondrocytes were washed twice with PBS and fixed with 1% paraformaldehyde/PBS solution (pH 7.4) for 10 min. Then fixed chondrocytes were incubated with 1% Triton/ PBS solution, followed by incubation with a proteinase K solution (20 g/ml) for 10 min at room temperature. Samples were then incubated in 3% hydrogen peroxide/PBS for 5 min at room temperature to quench endogenous peroxidases, followed by rinsing with PBS and incubation with equilibration buffer. Samples were incubated with a reaction mixture containing terminal deoxynucleotidyltransferase enzyme and digoxigenin-labeled dNTPs at 37°C in a humidified chamber. After 1 h, the reaction was stopped, and digoxigenin-labeled nucleotides were detected by peroxidase-conjugated anti-digoxigenin antibodies in a humidified chamber for 30 min at room temperature. The signal was detected using 3,3Ј-diaminobenzidine as a color substrate. Sections were counterstained with methylene green, mounted, and viewed under an Olympus microscope. To gain insights into the extent of apoptosis in the various chondrocyte cultures, the percentage of stained cells was determined. 500 chondrocytes were counted in 10 randomly chosen areas of three different cultures. Data were expressed as the mean Ϯ S.D. of the percentage of total cells that show TUNEL staining.
Statistical Analysis-Numerical data are presented as mean Ϯ S.D. (n Ͼ 4), and statistical significance between groups was identified using the two-tailed Student's t test (p values are reported in the figure legends).

RESULTS
Treatment of hypertrophic growth plate chondrocytes with RA induced terminal differentiation of these cells, as indicated by up-regulation of terminal differentiation marker genes, including cbfa1 (Fig. 1A), APase (Fig. 1B), and osteocalcin ( Fig.   1C), compared with the expression levels in untreated cells. An approximately 9-fold increase in cbfa1 gene expression was detected after a 3-day treatment, whereas APase gene expression was up-regulated by ϳ16-fold after a 5-day treatment, and osteocalcin gene expression increased ϳ16-fold after a 3-day treatment and ϳ14-fold after a 5-day treatment. Furthermore, alizarin red S staining revealed that RA-treated cultures were heavily mineralized after a 6-day treatment, whereas untreated cultures showed only little signs of mineralization (Fig.  2). Previously, we have shown that RA treatment led to a 3-fold increase in [Ca 2ϩ ] i of growth plate chondrocytes compared with the concentration of untreated cells. In addition, we provided evidence that most of this increase was mediated by Ca 2ϩ influx through annexin channels (16). Thus, it is possible that annexin-mediated Ca 2ϩ influx into growth plate chondrocytes regulates terminal differentiation events of these cells. To test this hypothesis, we cotreated cells with RA and the annexinspecific Ca 2ϩ channel blocker K-201 or antibodies specific for annexin V. Blocking annexin channel activities with K-201 led to a significant reduction of cbfa1, APase, and osteocalcin gene expression (Fig. 1), and mineralization of RA-treated chondrocyte cultures (Fig. 2). We have previously shown that antibodies specific for annexin V blocked its Ca 2ϩ channel activities (26). Cotreatment of RA-treated cultures with antibodies specific for annexin V also significantly reduced the rate of mineralization compared with the degree of mineralization in RAtreated cultures (Fig. 2).
RA treatment also up-regulated annexin II, V, and VI gene expression. K-201 significantly reduced the up-regulation of annexin II, V, and VI gene expression in RA-treated cultures to levels similar to untreated cultures (Fig. 3, A, B, and C). Type I collagen gene expression was up-regulated in RA-treated cultures (Fig. 4A), whereas type II collagen gene expression was down-regulated in these cultures (Fig. 4B). Cultures cotreated with RA and K-201 showed levels of type I and II collagen gene expression similar to levels in untreated cultures (Fig. 4, A and B). Type X collagen gene expression was not affected by RA or RA/K-201 treatment (Fig. 4C).
We have demonstrated that RA treatment led to the release of mineralization-competent APase and annexin II-, V-, and VI-containing matrix vesicles, whereas untreated chondrocytes released vesicles that contain no or little APase activity and annexins II, V, and VI (16; see also Fig. 5). Cotreatment of chondrocytes with RA and K-201 significantly reduced the amounts of APase activity (Fig. 5A) and annexins II, V, and VI (Fig. 5B) in matrix vesicle fractions. Thus, annexin-mediated Ca 2ϩ influx into growth plate chondrocytes regulates expression of terminal differentiation marker genes, the release of APase-and annexin II-, V-, and VI-containing matrix vesicles, and subsequent mineralization.
It is now well established that the final fate of terminally differentiated chondrocytes is apoptosis (2). Therefore, we addressed the question of whether RA triggers the whole cascade of terminal differentiation events including apoptosis and whether annexin-mediated Ca 2ϩ influx into chondrocytes is also involved in the regulation of apoptotic changes. To determine the degree of apoptosis in the various treated chondrocyte cultures we measured bcl-2 gene expression and caspase-3 activity and performed TUNEL labeling. A 5-day treatment with RA led to a significant decrease in gene expression of bcl-2, an anti-apoptotic protein (27) (Fig. 6). In contrast, caspase-3 activity, an active cell death protease involved in the execution phase of apoptosis (28), was more than 5-fold elevated in cultures treated for 6 days with RA compared with untreated cells (Fig. 7). Cotreatment of cultures with RA and the cytosolic Ca 2ϩ chelator BAPTA-AM abolished the decrease in bcl-2 gene expression (Fig. 6) and the increase in caspase-3 activity (Fig. 7), suggesting that cytosolic calcium is directly involved in the regulation of apoptotic events. Interestingly, bcl-2 gene expression was also higher in RA/K-201-treated cells than in RA-treated cells (Fig. 6), whereas caspase-3 activity was lower (Fig. 7). In addition, TUNEL labeling revealed that in RA-treated cultures more than 10% of cells were TUNELpositive, whereas only ϳ2% were TUNEL-positive in untreated and RA/BAPTA-treated cells, and ϳ4% of cells were TUNELpositive in RA/K-201-treated cultures (Fig. 8). DISCUSSION In this study we show that RA triggers a whole series of terminal differentiation events, including up-regulation of terminal differentiation marker genes (APase, cbfa1, osteocalcin), release of mineralization-competent matrix vesicles, subsequent mineralization, and finally apoptosis. RA also induces annexin-mediated Ca 2ϩ influx into growth plate chondrocytes. Blocking annexin Ca 2ϩ channel activities inhibited the whole series of terminal differentiation events, including up-regulation of terminal differentiation marker gene expression, release of mineralization-competent matrix vesicles, extracellular matrix mineralization, and apoptosis. These findings clearly establish the prominent regulatory function of annexin-mediated Ca 2ϩ influx into growth plate chondrocytes in terminal differentiation and apoptosis of these cells.
RA is known to regulate transcription after binding to the retinoic acid receptor complex. Three RA receptors have been identified, retinoic acid receptors ␣, ␤, and ␥. RA binds to one of these receptors, and this receptor complex then dimerizes with another receptor, retinoid X receptor (RXR). These receptor complexes then directly activate gene expression of transcription factors and other genes (29). Our study, however, indicates that RA also uses other mechanisms to regulate cell differentiation. As shown in this and previous studies, RA induces annexin channel formation in the plasma membrane of growth plate chondrocytes leading to an increased cytosolic calcium concentration (16). This increased cytosolic calcium concentration leads to further up-regulation of annexin and other terminal differentiation marker gene expression and causes the re-lease of mineralization-competent matrix vesicles and the induction of apoptotic events.
It is not clear how RA induces annexin channel formation. However, several possibilities can be envisioned. RA may bind to membrane receptors yet to be discovered. This binding activates an initial increase in cytosolic calcium concentration. Annexins require a certain Ca 2ϩ concentration to bind to phospholipids (20). The initial increase in cytosolic calcium may then lead to a relocation of annexins from the cytoplasma to the plasma membrane and channel formation causing a further boost in cytosolic calcium. Alternatively, RA might have similar effects on the membrane as vitamin D, which has been shown to increase plasma membrane fluidity (30). Increased membrane fluidity might favor annexin channel formation. A third possibility is that annexins bind directly to RA, and this binding favors annexin channel formation. This possibility is supported by recent findings showing that annexin II binds directly to vitamin D and acts as an alternative vitamin D receptor (31).
RA up-regulates gene expression of cbfa1, APase, and osteo- calcin. These genes are considered terminal differentiation markers and are expressed by late hypertrophic chondrocytes in the growth plate. cbfa1, a member of the runt-domain family of transcription factors, is expressed in osteoblasts and hypertrophic chondrocytes (6 -9). In cbfa1-deficient mice no endochondral and intramembranous ossification occurs because of an arrest of osteoblast differentiation (7,8). A disturbance of chondrocyte differentiation, especially terminal differentiation, was also observed in these mice (8). Continuous expression of cbfa1 in non-hypertrophic chondrocytes induced hypertrophic differentiation and endochondral ossification (9). Furthermore, overexpression of cbfa1 in chick immature chondrocytes induced type X collagen and MMP-13 expression, APase activity, and extensive matrix mineralization (32). These findings indicate that cbfa1 is an important regulatory factor in chondrocyte terminal differentiation. Our results show that annexin-mediated Ca 2ϩ influx into hypertrophic chondrocytes sequentially activates gene expression of cbfa1, APase, and osteocalcin. Interestingly, cbfa1 directly regulates gene expression of osteocalcin and other genes (33). Thus, it is possible that annexinmediated alteration of Ca 2ϩ homeostasis might control terminal differentiation and apoptotic events through the regulation of cbfa1 gene expression.
Terminal differentiation is also accompanied by extracellular matrix remodeling and alteration of collagen gene expres-sion. When chondrocytes undergo hypertrophic changes they turn on type X collagen gene expression and down-regulate type II collagen synthesis (34). Furthermore, previous studies in vivo and in vitro have demonstrated that terminal differentiated mineralizing chicken growth plate chondrocytes express type I collagen and other bone-related proteins, including osteocalcin and osteopontin (35,36). Our study showing that RA treatment leads to down-regulation of type II collagen and up-regulation of type I collagen gene expression confirms these previous findings and indicates that alteration of Ca 2ϩ homeostasis is involved in regulating collagen gene expression during growth plate development. In addition, our findings reveal that RA or RA/K-201 treatment did not affect type X collagen gene expression, indicating that type X collagen expression in growth plate chondrocytes used in this study was already at a high level. These results are consistent with previous findings demonstrating that type X collagen synthesis was greatly upregulated in immature growth plate chondrocytes after RA treatment but remained unchanged in RA-treated mature hypertrophic growth plate chondrocytes (37,38).
Annexin-mediated alteration of Ca 2ϩ homeostasis up-regulates annexin II, V, and VI and APase gene expression. The up-regulation of annexin and APase gene expression might be required for the release of mineralization-competent matrix vesicles. These vesicles contain large amounts of annexins II, V, and VI and APase activity. Annexins II, V, and VI also form Ca 2ϩ channels in matrix vesicles (19,21). Thus, annexin channel formation seems to play multiple functions in terminal differentiation of growth plate chondrocytes. Firstly, channel formation alters Ca 2ϩ homeostasis, which controls terminal differentiation events, and secondly annexin channel formation in matrix vesicles allows Ca 2ϩ influx into these particles as a possible initial step of mineral formation.
RA does not only induce mineralization of growth plate chondrocytes, but it triggers the whole cascade of terminal differentiation events, including apoptosis. Apoptosis, or programmed cell death, has been shown to be the final event of chondrocyte differentiation (2,22). Induction of apoptosis by RA has also been demonstrated in other cell types, including leukemia cells, thymocytes, neuroblastoma cell lines, and articular chondrocytes (39 -42). We show that RA down-regulates bcl-2 gene expression and stimulates caspase-3 activity. In addition, the percentage of TUNEL-positive cells was significantly higher in RA-treated cells compared with the number of apoptotic cells in untreated cultures. bcl-2 belongs to a rapidly expanding family of genes implicated in the control of apoptosis. Up-regulation of bcl-2 by PTHrP delays maturation of growth plate chondrocytes toward hypertrophy and subsequent apoptosis (27). In contrast, caspase-3 is an active cell death protease involved in the execution phase of apoptosis (28). Interestingly, apoptotic changes were significantly reduced in cultures cotreated with RA and BAPTA or RA and K-201, suggesting that annexin-mediated Ca 2ϩ influx into growth plate chondrocytes is involved in the regulation of the complete terminal differentiation program, including apoptosis. Thus, RA and RA-mediated annexin Ca 2ϩ channel formation appear to stimulate a similar sequence of events as observed in growth plate cartilage. A recent study has demonstrated high amounts of retinoids in the perichondrium and that implantation of beads containing RA antagonist near the humeral anlagen drastically decreased chondrocyte hypertrophy and terminal differentiation. In addition, retinoic acid receptor ␥ is expressed in hypertrophic and terminally differentiated chondrocytes (43). Thus, it is likely that RA and the resulting annexin Ca 2ϩ channel formation play important regulatory roles in the regulation of terminal differentiation events in the growth plate during endochondral ossification. Annexin V has also been shown to mediate Ca 2ϩ influx induced by hydrogen peroxide into B-lymphocytes (44). Thus, annexins not only form Ca 2ϩ channels in chondrocytes but also other cell types. In addition, B-lymphocytes lacking annexin V are resistant to apoptosis (45). Ca 2ϩ is known to be required for apoptosis, and it is known that Ca 2ϩ can cause apoptosis by itself under conditions of Ca 2ϩ overload. Our study demonstrates that the cytosolic calcium chelator BAPTA-AM or the annexin channel blocker K-201 significantly inhibits apoptosis of growth plate chondrocytes. These findings suggest that annexin-mediated Ca 2ϩ influx or related annexin functions in various cell types can lead to a Ca 2ϩ overload and apoptosis resulting from this overload.
Hypertrophic and terminally differentiated growth plate chondrocytes express three annexins (annexins II, V, and VI). Previous studies have shown that all three annexins can form Ca 2ϩ channels and that K-201 inhibits Ca 2ϩ channel activities of all three annexins (21). However, it is not clear, whether all three annexins form Ca 2ϩ channels in growth plate chondrocytes independently and mediate Ca 2ϩ influx into these cells. Our previous findings show that each antibody fraction specific for annexin II, V, or VI partially inhibited increases in [Ca 2ϩ ] i in growth plate chondrocytes, suggesting that all three annexins contributed to Ca 2ϩ influx into these cells (16). However, antibodies specific for annexin V inhibited mineralization of growth plate chondrocytes to a degree similar to K-201 (see Fig.  2). Other studies have shown that peroxide-mediated Ca 2ϩ influx was altered only in B cells lacking annexin V but not in cells lacking annexin II (44). In addition, only B cells lacking annexin V but not cells lacking annexin II were resistant to apoptosis (45). Future studies have to establish whether only annexin V modulates Ca 2ϩ homeostasis of growth plate chondrocytes and is involved in the regulation of chondrocyte terminal differentiation, whether all three annexins form Ca 2ϩ channels independently and regulate Ca 2ϩ influx into growth plate chondrocytes, or whether the interactions between the three annexins are required for optimal annexin V channel activities in growth plate chondrocytes.
Previous studies from our and other laboratories have shown the expression of hypertrophic and terminal differentiation markers, including annexins II, V, and VI, APase, osteopontin, osteocalcin, and type X collagen in osteoarthritic cartilage (17, 18, 46 -49). In addition, mineralization and apoptosis were detected in osteoarthritic cartilage (17, 50 -52). Thus, it is possible that up-regulation of annexin gene expression in osteoarthritic cartilage might lead to annexin-mediated Ca 2ϩ influx into articular chondrocytes and subsequent stimulation of terminal differentiation events in these cells. Terminal differentiation events are required for endochondral bone formation during normal development; however, if these events occur under pathological conditions, such as osteoarthritis, they will lead to cartilage destruction. If these annexins are as essential for terminal differentiation in osteoarthritic cartilage as they are in growth plate cartilage during endochondral ossification, they could be promising targets for therapies.