Bone Morphogenetic Proteins and bFGF Exert ... - Wiley Online Library

Download PDF
20 downloads 982 Views 527KB Size Report
Comment
The BCA Protein Assay (Pierce Company, Rockford, IL,. U.S.A.) was performed to ..... Indian hedgehog and PTH-related protein. Science 273:613–. 622. 8. ... Sakano S, Murata Y, Miura T, Iwata H, Sato K, Matsui N, Seo. H 1993 Collagen and ...
JOURNAL OF BONE AND MINERAL RESEARCH Volume 13, Number 6, 1998 Blackwell Science, Inc. © 1998 American Society for Bone and Mineral Research

Bone Morphogenetic Proteins and bFGF Exert Opposing Regulatory Effects on PTHrP Expression and Inorganic Pyrophosphate Elaboration in Immortalized Murine Endochondral Hypertrophic Chondrocytes (MCT Cells)* ROBERT A. TERKELTAUB, KRISTEN JOHNSON, DAVID ROHNOW, RANDY GOOMER, DOUG BURTON, and LEONARD J. DEFTOS

ABSTRACT A fundamental question in endochondral development is why the expression of parathyroid hormone–related protein (PTHrP), which inhibits chondrocyte maturation and mineralization, becomes attenuated at the stage of chondrocyte hypertrophy. To address this question, we used clonal, phenotypically stable SV40-immortalized murine endochondral chondrocytes that express a growth-arrested hypertrophic phenotype in culture (MCT cells). Addition of individual cytokines to the medium of MCT cells revealed that bone morphogenetic protein (BMP)-6, which commits chondrocytes to hypertrophy, markedly inhibited PTHrP production. This activity was shared by three other osteogenic bone morphogenetic proteins (BMP-2, BMP-4, and BMP-7) and by transforming growth factor b (TGF-b), which all inhibited the level of PTHrP mRNA. In contrast, basic fibroblast growth factor (bFGF), an inhibitor of chondrocyte maturation to hypertrophy, induced PTHrP in MCT cells and antagonized the effects of BMP-2, BMP-4, BMP-6, and BMP-7 and TGF-b on PTHrP expression. Opposing effects of bFGF and BMPs also were exerted on the elaboration of inorganic pyrophosphatase (PPi), which regulates the ability of hypertrophic chondrocytes to mineralize the matrix. Specifically, BMP-2 and BMP-4, but not BMP-6 and BMP-7, shared the ability of TGF-b to induce PPi release, and this activity was inhibited by bFGF in MCT cells. Our results suggest that effects on PTHrP expression could contribute to the ability of BMP-6 to promote chondrocyte maturation. BMPs and bFGF exert opposing effects on more than one function in immortalized hypertrophic chondrocytes. Thus, the normal decrease in bFGF responsiveness that accompanies chondrocyte hypertrophy may function in part by removing the potential for bFGF to induce PTHrP expression and to oppose the effects of BMPs. MCT cells may be useful in further understanding the mechanisms regulating the differentiation and function of hypertrophic chondrocytes. (J Bone Miner Res 1998;13:931–941)

INTRODUCTION

T

HE CENTRAL PROCESS in endochondral bone formation is the progressive differentiation of proliferating chondrocytes to growth-arrested hypertrophic cells that express type

*Presented in part at the 19th Annual Meeting of the American Society for Bone and Mineral Research, Cincinnati, OH, U.S.A., September, 1997.

X collagen and direct the mineralization of the cartilage matrix and its subsequent replacement by bone.(1,2) However, efforts to better understand differentiation, gene expression, and function in mammalian hypertrophic chondrocytes are generally hampered by limitations in yield and phenotypic consistency and stability of these cells in primary culture.(3) To address some of these limitations, Lefebvre et al. immortalized primary mouse rib chondrocytes with the thermolabile SV40 large T antigen, establishing

VA Medical Center, University of California at San Diego, San Diego, California, U.S.A.

931

932 cells (termed the MCT line) that proliferate at temperatures permissive for large T (32–33°C) but undergo hypertrophy and fail to progress through the cell cycle at 37– 39°C.(3) The transition of MCT cells to growth-arrested conditions is associated with diminished expression of type II collagen mRNA and increased expression of type X collagen, osteocalcin, and osteopontin mRNA.(3) A similar pattern of progressively altered gene expression, including the appearance of mRNAs for osteoblastoid proteins, is consistent with the progressive differentiation to a terminal hypertrophic chondrocyte phenotype.(1,2) We used MCT cells to explore mechanisms regulating the expression of parathyroid hormone–related protein (PTHrP) in late endochondral development.(4 – 8) PTHrP is normally expressed in perichondrium and within the resting and proliferative zones of the cartilage growth plate.(9) Furthermore, PTHrP also localizes (in conjunction with PTH/PTHrP receptors) at the junction of the proliferative and hypertrophic zones of endochondral cartilages.(4) However, a fundamental question in endochondral development is the basis for the marked down-regulation of PTHrP expression when endochondral chondrocytes become hypertrophic.(9 –11) PTHrP expressed by prehypertrophic chondrocytes inhibits further differentiation to hypertrophy.(6) In addition, PTHrP modulates cartilage matrix synthesis(5) and inhibits chondrocyte apoptosis, functions that are directly linked to ultimate mineralization of the matrix.(12,13) The critical effects of PTHrP on the organization of endochondral development have been demonstrated by the production of cartilaginous deficiency of PTHrP (or the PTH/PTHrP receptor) in mice, which results in accelerated endochondral mineralization.(8,9,11) Conversely, PTHrP overexpression caused delayed endochondral differentiation and impaired long bone growth and ossification.(6,10) In this study, by adding individual cytokines to the media of MCT cells, we focused on the possibility that bone morphogenetic proteins (BMPs) and basic fibroblast growth factor (bFGF) could regulate PTHrP expression. BMPs can directly stimulate endochondral chondrocyte development and osteogenesis in vivo, in part by recruiting the precursors of chondrocytes and osteoblasts.(2,14,15) BMPs also can modulate chondrocyte gene expression, matrix synthesis, and other differentiated functions at early and more advanced stages of the endochondral development cascade.(2,14,16 –21) Expression of BMP-2 and certain other BMPs perseveres in perichondrial cells of the cartilage growth plate.(14 –17) Furthermore, BMP-2 can promote chondrocyte maturation to hypertrophy and affect the function of calcifying posthypertrophic chondrocytes. (14,16,22) However, changes in the level of chondrocyte expression of most BMPs during progression of endochondral bone morphogenesis appear functionally significant.(2,14) De novo expression of BMP-6 is relatively restricted to endochondral prehypertrophic chondrocytes,(2,14,23) where it is believed to be critical in committing chondrocytes to hypertrophic differentiation, including the expression of osteocalcin and the enhanced formation of bone.(14,22,24) Significantly, PTHrP inhibits BMP-6 expression by chondrocytes.(24) De novo chondro-

TERKELTAUB ET AL. cyte expression of BMP-7, which is highly homologous to BMP-6,(14) appears at the stage of final calcification.(25) In contrast to the promotion of endochondral development and mineralization by BMPs,(2,14) bFGF inhibits chondrocyte differentiation from prehypertrophy to hypertrophy, and bFGF suppresses cartilage-matrix calcification in cultured rabbit growth plate chondrocytes in vitro.(26 –28) Ectopic expression of bFGF in transgenic animals reproduces a syndrome phenotypically identical to achondroplasia.(29) However, the downstream mediators by which bFGF produces these effects on endochondral bone growth are not completely understood.(30) Because hypertophic chondrocytes direct cartilage matrix mineralization,(1,2,31) we also examined the potential for bFGF and BMPs to regulate the extracellular release of inorganic pyrophosphate (PPi) by MCT cells. In this regard, the modulated elaboration of substantial amounts of PPi is a relatively unique function of cells that mineralize (chondrocytes and osteoblasts).(32–35) The major role of PPi in mineralization is believed to be as a critical and potent inhibitor of calcium phosphate (hydroxyapatite) crystal deposition.(32,35,36) Specifically, the removal of PPi by pyrophosphatase activity, or the denial of access of PPi at the site of mineralization, is needed for the active phase of crystal deposition to proceed.(37,38) An excess of free extracellular PPi to inorganic phosphate (Pi) also suppresses hydroxyapatite crystal propagation.(32) Significantly, the adsorption of PPi to existing bone mineral is believed to serve an additional regulatory function by inhibiting hydroxyapatite crystal dissolution.(32,39 – 41) Thus, the ability of growth and differentiation factors to actively regulate ambient PPi appears to be one of the means to control the rate, extent, and stability of mineralization.(32–35) Our results, detailed below, reveal that BMPs and bFGF exert opposing effects on not only PTHrP expression but also PPi release in MCT cells. The potential significance of these findings is discussed further below.

MATERIALS AND METHODS Reagents Recombinant human BMP-2, BMP-4, BMP-6, and BMP-7 were a generous gift of the Genetics Institute (Cambridge, MA, U.S.A.). Recombinant human bFGF, aFGF, transforming growth factor b1 (TGF-b1), interleukin1b(IL-1b), and insulin-like growth factor I were obtained from R&D Systems (Minneapolis, MN, U.S.A.). 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) was a generous gift of Dr. M. Uskokovic (Hoffman-Laroche, Nutley, NJ, U.S.A.). Retinoic acid was obtained from Sigma (St. Louis, MO, U.S.A.).

Cells and cell culture Immortalized murine endochondral chondrocytes (MCT cells),(3) a generous gift of Drs. Veronique Lefebvre and Benoit de Crombrugghe (University of Texas, Houston, TX, U.S.A.), were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 1.0 g/l glucose and sup-

bFGF and BMPs REGULATE CHONDROCYTE PTHrP AND PPi RELEASE plemented with 10% heat-inactivated fetal calf serum, 1% penicillin-streptomycin, and 2 mM glutamine (Life Technologies, Gaithersburg, MD, U.S.A.) under permissive growth conditions at 32°C. We confirmed(3) that MCT cells expressed type X collagen, using reverse-transcriptase polymerase chain reaction (RT-PCR; by the method described). Where indicated, MCT cells at 70% confluence (700,000 cells in a 60-mm tissue culture dish in 1 vol of 5.0 ml) were incubated with the indicated agonists at 32°C or 37°C (nonpermissive growth-arrested conditions) for up to 72 h. Articular cartilage chondrocytes were isolated by methods previously outlined in detail(42) from normal human knee cartilage at autopsy, following institutional approval. Cartilage was collected only from noncalcified areas, and precautions to avoid chondrocyte culture contamination by other articular tissues were exercised as previously described.(42) Cartilage was minced with a scalpel and treated with trypsin (10% v/v) for 15 minutes in a 37°C waterbath. The samples were transferred to DMEM containing 5% fetal bovine serum, penicillin-streptomycin-fungizone, and 2 mg/ml clostridial collagenase type IV (Sigma) and digested for 3 h. Cells were trypsinized from 175 cm2 flasks and plated in 60-mm dishes in DMEM supplemented with 2 mM L-glutamine, 4.5 g/l glucose, 10% fetal calf serum, and 1% Pen-Strep (first passage chondrocytes). When 70% confluency was reached, the cells (700,000 cells in a 60-mm tissue culture dish in a volume of 5.0 ml) were incubated at 32°C or 37°C for 72 h as described below.

PTHrP immunoassays PTHrP was measured in conditioned media using three specific immunoassays, via previously described methods.(43) In brief, tyrosinated hPTHrP peptide fragments 1–34 (or 38 – 64 where indicated), were radioiodinated using chloramine-T, and rabbit antisera specific for each peptide were employed in a 3-day nonequilibrium immunoassay format. All samples were assayed in triplicate and in multiple dilutions, which paralleled the corresponding PTHrP standard curve. Intra-assay and interassay variations were between 7 and 12%, respectively.

Assays of PPi and enzymes active in PPi metabolism PPi concentrations were determined radiometrically via differential adsorption on activated charcoal of UDP-D-[63 H] glucose (Amersham, Chicago, IL, U.S.A.) from its PPi-catalyzed reaction product 6-phospho[6-3H]gluconate, as previously described.(42) PPi was equalized for the DNA concentration in each well, determined chromogenically following precipitation in perchlorate, as previously described.(42) DNA concentrations were confirmed to vary less than 5% in MCT cells with or without each agonist studied under nonpermissive conditions. To determine specific activity of cellular NTPPPH, the MCT cells were lysed in a buffer containing 1% Triton X-100 in 0.2 M Tris base with 1.6 mM MgCl2, pH 8.1.(44) The BCA Protein Assay (Pierce Company, Rockford, IL, U.S.A.) was performed to determine protein concentrations in cell lysates. Five micrograms of cell lysate protein was

933

used for each replicate in both the alkaline 59 nucleotide phosphodiesterase I assay for NTPPPH, which used 1 mM p-nitrophenyl-thymidine 59 monophosphate (Sigma) as substrate, and the alkaline phosphatase (ALP) assay, which used 4-nitrophenyl phosphate (7.6 mM) as the substrate.(44) One unit of enzyme activity was defined as equivalent to 1 mmol of substrate hydrolyzed per hour per 5 mg sample.

Analysis of PTHrP mRNA For semiquantitative RT-PCR, total RNA was isolated from the cells using 1.0 ml RNAZOL (Tel-Test Inc., Friendswood, TX, U.S.A.) per plate, with RNA extracted in chloroform and then precipitated in isopropanol for 18 h at –20°C. One microgram of each cellular total RNA sample was reverse transcribed using AMV Reverse Transcriptase (10 U/ml) and random primers in the Invitrogen (Carlsbad, CA, U.S.A.) cDNA Cycle kit. Phenol-extracted cDNA products were amplified for 30 – 40 cycles using characterized primers and reactions for the ribosomal protein L30 (the “housekeeping gene” control for equal loading(44) which amplified a 196 bp product, for PTHrP using the primers (sense 59 CTG GTT CAG CAG TGG AGC GTC and antisense 59 CCT GAG TTA GGT ATC TGC CC), that amplified a 306 base pair product,(4) and for the a1 chain of homotrimeric Type X Collagen(3) (using the sense primer 59 TAG GAG CTA AAG GAG TGC CTG GAC, and the antisense primer 59 GAC TAC CTG TTA CCC CGT GGT TAG), which amplified a 322 base pair product. The RT-PCR gel bands were quantitatively analyzed with a digital imaging system (Alpha Innotech Corp., San Leandro, CA, U.S.A.). The intensities of the bands were assigned integrated density values, which represent the sum of all the pixel values in the box. All bands were quantitated in equal area boxes and autobackground subtract was used to control for background signal. The band intensities were determined to be below saturation by a false color palette in the image analysis software.

Statistical analysis Where indicated, error bars represent SD. Statistical analysis was performed using the Student’s t-test (paired two-sample testing for means), applied on Microsoft Excel 5.0 for the Macintosh computer.

RESULTS bFGF is a potent inducer of PTHrP expression by immortalized endochondral chondrocytes The mechanism by which bFGF inhibits terminal hypertrophic chondrocyte differentiation is unknown.(26) Because PTHrP profoundly inhibits terminal chondrocyte differentiation,(6) we determined whether bFGF influenced PTHrP expression in MCT cells. We observed that bFGF (10 ng/ml) markedly induced PTHrP release (measured by accumulation of PTHrP(1–34)) by MCT cells cultured for 72 h under both growth-arrested (37°C) and proliferative conditions (32°C) (Fig. 1). In contrast, IL-1b had no signif-

934

FIG. 1. bFGF induces PTHrP release from immortalized endochondral chondrocytes. MCT cells (murine chondrocytes immortalized with thermolabile SV40 large T) were plated at 70% confluence (;700,000 cells in a 60 mm plate in 5.0 ml) as described in the Materials and Methods, and incubated with the indicated agonists (each at 10 ng/ml) for 72 h at 32°C (permissive, proliferating) or 37°C (nonpermissive, growth arrested). PTHrP(1–34) was quantitated by immunoassay in the conditioned media at 72 h as described in the Materials and Methods (n 5 3).

icant effect on PTHrP release, insulin-like growth factor I had a modest stimulatory effect, and TGF-b inhibited PTHrP release by 40 –50% (Fig. 1). 1,25(OH)2D3 (10 nM) had no significant effect on PTHrP expression (data not shown). Because PTHrP was induced more strongly under growth-arrested as compared with proliferative conditions (up to 15-fold versus up to 5-fold) in MCT cells (Fig. 1), we focused the remaining studies on PTHrP expression in cells under growth-arrested conditions. aFGF also induced PTHrP release by MCT cells, but bFGF consistently induced greater and more rapid PTHrP release than did aFGF (Fig. 2). The potency of bFGF as an inducer of PTHrP was further underscored by the observation, in dose–response experiments, that at bFGF concentrations of only 0.1 and 1.0 ng/ml, the amount of PTHrP released reached at least 60% of the amount released in response to 10 ng/ml bFGF (data not shown). In contrast, aFGF at concentrations below 10 ng/ml did not consistently induce PTHrP in immortalized chondrocytes (data not shown).

BMPs suppress PTHrP release in immortalized endochondral chondrocytes We observed that BMP-2, BMP-4, BMP-6, and BMP-7 (at 10 ng/ml) each inhibited PTHrP release (measured by accumulation of PTHrP(1–34) in conditioned media at 72 h) by ;50% (Fig. 3A). Dose–response studies (using BMPs at 0.1–30.0 ng/ml) indicated that BMP-4 was the

TERKELTAUB ET AL.

FIG. 2. Kinetics of PTHrP release by immortalized endochondral chondrocytes in response to bFGF and aFGF. MCT cells were incubated at 37°C, as described above, with the indicated agonists (at 10 ng/ml), and PTHrP release into the conditioned media was measured by immunoassay (for the 1–34 domain) at the indicated time points (n 5 3). most potent inhibitor of PTHrP release, but that the inhibitory activity of each BMP reached a peak at the concentration of 10 ng/ml (data not shown). Murine PTHrP is a 139 amino acid polypeptide that can be proteolytically processed.(4) However, the suppression by BMP-2, BMP-4, BMP-6, and BMP-7 of PTHrP(1–34) accumulation in the conditioned media of MCT cells was not an artefact of catalysis of PTHrP, because BMP-2, BMP-4, BMP-6, and BMP-7 also suppressed PTHrP(38 – 64) release (Fig. 3B). Certain BMPs can colocalize with bFGF at the resting, proliferative, and terminal stages of endochondral chondrocyte development.(2) We observed that BMP-2, BMP-4, BMP-6, and BMP-7 (10 ng/ml) each prevented bFGF (10 ng/ml) from stimulating PTHrP(1–34) and PTHrP(38 – 64) release (Figs. 3A and 3B). MCT cells constitutively expressed PTHrP mRNA (Figs. 4A and 4B). BMP-2, BMP-4, BMP-6, and BMP-7 markedly suppressed the level of PTHrP mRNA (Figs. 4A and 4C). In contrast, bFGF induced PTHrP mRNA (Figs. 4A– 4D). Because the inhibitory effects of BMPs on bFGF-induced PTHrP(1–34) release were most potent for BMP-2, we evaluated the effects of BMP-2 on bFGF-induced PTHrP mRNA expression. We observed that BMP-2 blocked the ability of bFGF to up-regulate the level of PTHrP mRNA (Figs. 4A and 4C). BMPs are members of the TGF-b superfamily.(14) We observed that TGF-b inhibited not only basal PTHrP release (Fig. 1) but also bFGF-induced PTHrP release (data not shown). TGF-b also suppressed basal and bFGF-stimulated PTHrP mRNA expression (Figs. 4B and 4D) in a manner comparable to the inhibition by BMPs. Thus, the abilities of BMPs, TGF-b, and bFGF to regulate PTHrP release appeared to be mediated, at least in part, by effects on PTHrP mRNA expression.

bFGF and BMPs REGULATE CHONDROCYTE PTHrP AND PPi RELEASE

935

FIG. 3. BMPs inhibit PTHrP release and prevent bFGF from inducing PTHrP release in MCT cells. MCT cells were incubated at 37°C, as described above, with the indicated agonists (at 10 ng/ml). PTHrP(1–34) release (A) and PTHrP (38 – 64) release (B) into the conditioned media at 72 h was measured by immunoassay in triplicate as described in the Materials and Methods. Labels on this figure (and the other figures) are of the same type for agonists of each class, and the data for each agonist are presented in the same order as listed outside the bar graph. PTHrP(1–34) and (38 – 64) release in the control samples incubated in buffer alone were 250 6 84 pg/ml and 345 6 61 pg/ml, respectively (data pooled from five separate experiments). Changes in PTHrP release relative to controls were significant (p , 0.05) for each BMP alone and for bFGF alone.

MCT cells elaborate PPi To study another regulated function of differentiated chondrocytes that controls mineralization, we assessed PPi elaboration in MCT cells.(32– 41) We verified that both proliferating (32°C) and growth-arrested (37°C) MCT cells shared the ability of articular chondrocytes (studied at the same temperatures) to elaborate PPi, and we confirmed(33,42) that TGF-b stimulated this activity in a differentiation-independent and temperature-independent manner in MCT cells (Fig. 5). Growth-arrested MCT cells elaborated more PPi than proliferating MCT cells (Fig. 5). The increase in PPi elaboration with growth arrest appeared only partially attributable to the increased temperature in culture, as revealed by comparison with articular

chondrocytes cultured at the same temperatures (Fig. 5). More extensive studies were subsequently carried out in growth-arrested MCT cells to allow comparison with the characterized effects of BMPs and bFGF on PTHrP expression.

BMP-2 and BMP-4 increase PPi release BMP-2 stimulated an increase (of more than 50%) in PPi release by growth-arrested MCT cells (Table 1). The increase in PPi release stimulated by BMP-2 was less than that induced by retinoic acid(45) but comparable to that induced by both 1,25(OH)2D3 (Table 1) and TGF-b (Fig. 5). BMP-2 and TGF-b did not synergistically enhance

936

TERKELTAUB ET AL.

FIG. 5. Proliferative (32°C) and growth-arrested (37°C) MCT cells share the ability of articular chondrocytes to elaborate PPi in response to TGF-b. Human articular chondrocytes were isolated and studied in first passage in monolayer culture as described in the Materials and Methods. The MCT cells, and the articular chondrocytes were incubated at 32°C or 37°C for 72 h, and with or without TGF-b (10 ng/ml) present, as indicated. PPi concentrations in the conditioned media, and DNA concentrations, were determined as described in the Materials and Methods. Data are expressed as percentage increase in PPi relative to respective controls (n 5 6). Mean PPi concentrations were: 101 pmol and 176 pmol/mg DNA for articular chondrocytes at 32°C and 37°C, respectively, and 91 pmol and 242 pmol/mg DNA for MCT cells at 32°C and 37°C, respectively. p , 0.01 for the increase in PPi elaboration at 37°C versus 32°C in unstimulated cells, and for the TGF-b–induced increase in PPi release for both articular chondrocytes and MCT cells.

FIG. 4. bFGF induces PTHrP mRNA expression, but this activity is attenuated by BMP-2 (A and B) and TGF-b (C and D). MCT cells were incubated at 37°C for 24 h, as described above, in the presence of the agonists indicated (each at 10 ng/ml). The figure indicates representative results of semiquantitative RT-PCR analysis of mRNA levels, in which total RNA from each sample was reverse-transcribed and amplified for 30 cycles by PCR using previously validated primer sets specific for PTHrP, and as a “housekeeping gene” the ribosomal protein L30, yielding products of the correct sizes (A and B), as detailed in the Materials and Methods. The differences between PTHrP expression were quantitatively analyzed by densitometry (C and D) as described in the Materials and Methods. Similar stimulation of PTHrP by bFGF, and inhibition of PTHrP expression by BMP-2 (A and C) and TGF-b (B and D) remained present when the cDNAs were amplified for 29, 33, and 35 cycles (data not shown). (A) Lane 1, unstimulated control; lane 2, BMP-2; lane 3, BMP-4; lane 4, BMP-6; lane 5, BMP-7; lane 6, BMP-2 and bFGF; lane 7, bFGF. (B) Lane 1, unstimulated control; lane 2, TGFb; lane 3, bFGF; lane 4, bFGF and TGFb.

PPi release (Fig. 6), suggesting that they acted through similar basic mechanisms. In this context, BMP-2, like TGF-b (Fig. 6), as well as 1,25(OH)2D3 and retinoic acid (Table 1), induced a significant increase in the cell-associated specific activity of the PPi-generating enzyme NTPPPH,(25,34,44) without altering the specific activity of the PPi-degrading enzyme ALP. IL-1b suppressed PPi release(42) and diminished the ability of BMP-2 to induce NTPPPH and to promote PPi release in MCT cells (Table 1). As expected,(42) IL-1b exerted comparable inhibitory effects on TGF-b–stimulated NTPPPH expression and PPi release on MCT cells (data not shown). These observations suggested the possibility that induction of NTPPPH might be critical for BMPinduced PPi release in immortalized endochondral chondrocytes. However, BMP-2, BMP-4, BMP-6, and BMP-7 each induced an increase in NTPPPH-specific activity in MCT cells, but only BMP-2 and BMP-4 induced a significant increase in PPi release (Fig. 7). Thus, BMP-2 and BMP-4 induced chondrocyte PPi elaboration by a mechanism that required events beyond the induction of increased NTPPPH activity. Though the ligand binding affinities and functional responses mediated by BMP receptors can be

bFGF and BMPs REGULATE CHONDROCYTE PTHrP AND PPi RELEASE TABLE 1. BMP-2 STIMULATES PPi RELEASE

Agonist Control BMP-2 IL-1b BMP-2 1 IL-1b Retinoic acid 1,25(OH)2D3

AND

937

INDUCES INCREASES NTPPH ACTIVITY

IN

GROWTH-ARRESTED MCT CELLS

PPi (pmoles per ug DNA) (% of control)

NTPPPH (U/ml) (% of control)

Alkaline phosphatase (U/ml) (% of control)

178 6 11 (100) 274 6 22*(154) 76 6 18* (43) 222 6 29 (125) 303 6 11*(170) 250 6 36*(140)

3.72 6 0.2 (100) 5.39 6 0.6*(145) 2.13 6 0.1* (57) 3.39 6 0.1* (91) 6.89 6 0.3*(185) 6.29 6 0.4*(169)

1.29 6 0.0(100) 1.26 6 0.0 (98) 1.29 6 0.1(100) 1.27 6 0.0 (98) 1.27 6 0.0 (98) 1.27 6 0.0 (98)

*p , 0.05 vs. control. MCT cells were incubated at 37°C for 72h. Where indicated, the added agonists were human recombinant cytokines (at 10 ng/ml), retinoic acid (1 uM), or 1,25 dihydroxy vitamin D3 (1,25(OH)2D3) (10 nM). PPi release into the media, and the specific activities of NTPPPH and alkaline phosphatase in cell lysates (expressed as Units, with 1 U 5 1 micromole of substrate hydrolyzed/hour/ml), were determined as described in the Methods. Data was pooled from 4 experiments done in triplicate.

DISCUSSION

FIG. 6. BMP-2– and TGF-b–induced PPi release are comparable in MCT cells. MCT cells were incubated in triplicate at 37°C for 72 h with no agonist (control), BMP-2 (10 ng/ml), TGF-b (10 ng/ml), and both BMP-2 and TGF-b (at 10 ng/ml each), as indicated, and PPi release into the conditioned medium, and the specific activity of NTPPPH and ALP in cell lysates collected at 72 h were measured as described above. The data are from one experiment representative of five with comparable results. Mean values for the control samples incubated with buffer alone were: PPi 172 pmol/mg DNA, NTPPPH 3.4 U/ml, ALP 1.5 U/ml.

regulated by the formation of certain BMP heterodimers,(14) we did not observe any synergy in the abilities of BMP-2, BMP-4, BMP-6, and BMP-7 to affect PPi metabolism in MCT cells (data not shown). bFGF alone exerted small but significant suppressive effects on PPi release and NTPPPH activity, and bFGF antagonized the ability of BMP-2 and BMP-4 to induce PPi elaboration and to increase NTPPPH activity (Fig. 7). Thus, BMPs and bFGF exerted opposing effects on several activities in MCT cells.

Chondrocytes immortalized in a hypertrophic state of differentiation have been proposed(3) to be useful for investigating mechanisms regulating chondrocyte maturation(2) and mineralizing activity.(1,3,31) A fundamental question has been the basis for the attenuation in hypertrophic chondrocytes of the expression of PTHrP.(8 –11) In this study, MCT cells, which express features of chondrocyte hypertrophy in culture, including growth arrest and progressively increased type X collagen and osteocalcin expression,(3) were demonstrated to express PTHrP. By adding individual cytokines to the medium of MCT cells, we discovered that BMP-6,(14,22) which commits chondrocytes to hypertrophy,(23) markedly attenuated PTHrP release by MCT cells. This activity was shared by three other osteogenic BMPs (BMP-2, BMP-4, and BMP-7), and by TGF-b, which all inhibited the level of PTHrP mRNA (Figs. 3, 4A, and 4C). PTHrP is a critical mediator of the proliferation of chondrocytes in early endochondral development.(9,11) Thus, BMP-2 is believed to act in part by inducing the expression of chondrocyte PTH/PTHrP receptors.(14) However, in normal endochondral development, both PTHrP expression and PTH/PTHrP receptor expression are attenuated when chondrocytes become hypertrophic.(8 –11) Significantly, PTHrP was recently observed to inhibit endochondral chondrocyte expression of BMP-6,(14,24) which is produced most abundantly by prehypertrophic chondrocytes.(23,24) The capacity of BMP-6 to markedly suppress PTHrP expression by MCT cells (Figs. 3 and 4) suggests not only a potential mechanism whereby prehypertrophic chondrocytes might contribute to down-regulation of PTHrP but also a potential negative feedback loop between BMP-6 and PTHrP. TGF-b also suppressed PTHrP production in MCT cells (Figs. 1, 4B, and 4D).(1,40,41) The level of TGF-b expression decreases in situ in transitional prehypertrophic chick endochondral chondrocytes,(46) a cartilage zone in which PTHrP localizes.(6,11) However, TGF-b is an abundant product of cultured chick hypertrophic chondrocytes.(47)

938

TERKELTAUB ET AL. FIG. 7. bFGF and BMPs exert antagonistic effects on PPi release by MCT cells. MCT cells were incubated in triplicate at 37°C for 72 h with no agonist (control), BMP-2, BMP-4, BMP-6, and BMP-7 (10 ng/ml) and/or bFGF (10 ng/ ml) and PPi release into the conditioned medium, and the specific activity of NTPPPH and ALP in cell lysates collected at 72 h were measured as described above. Data are pooled from four experiments. Mean values for the control samples incubated with buffer alone were: PPi 205 pmol/mg DNA, NTPPPH 3.1 U/ml, ALP 1.4 U/ml. The changes in PPi relative to controls were significant (p , 0.05) for BMP-2 and BMP-4 alone and bFGF alone, and changes in NTPPPH significant for BMP-2, BMP-4, BMP-6, BMP-7, and bFGF.

Thus, it will be of interest to test directly the relationship between TGF-b and PTHrP expression in endochondral chondrocytes in situ. We observed that the effects of BMPs on growth-arrested MCT cells extended beyond regulation of PTHrP expression. BMP-2 and BMP-4 induced a marked increase of PPi release by MCT cells, and this activity was shared by not only TGF-b but also by retinoic acid and 1,25(OH)2D3 (Table 1 and Figs. 6 and 7), which promote endochondral mineralization.(1,35,45,48) In contrast, bFGF,(28 –30) and to a greater degree IL-1 (which is expressed by endochondral chondrocytes but has an unclear role in endochondral mineralization),(49,50) suppressed PPi elaboration by MCT cells (Table 1 and Fig. 7). Active regulation of PPi production and catabolism are fundamental events in endochondral development,(38) and the ability of these mediators to raise or lower ambient PPi could help control the rate, extent, and stability of hydroxyapatite crystal deposition.(32– 41) The increase in ambient PPi induced by BMP-2 and BMP-4 (and TGF-b) may underestimate the actual amount of PPi generated because of not only potential adsorption of PPi to matrix constituents but also the catabolism of PPi.(32) In this context, we observed that BMP-2 and BMP-4 and TGF-b induced PPi-generating nucleotide pyrophosphohydrolase (NTPPPH) activity(32,34,35,44,46) without a change in PPi-degrading ALP activity (Table 1 and Figs. 6 and 7). However, effects of BMP-2 and BMP-4 on factors other than NTPPPH activity must play a role in PPi elaboration, because BMP-6 and BMP-7 induced NTPPPH activity but not PPi release (Fig. 7), and because bFGF (Fig. 7) and IL-1 (Table 1) differed in the extent of their effects on BMP-induced NTPPPH activity. In this study, we have demonstrated that bFGF, an inhibitor of chondrocyte maturation to hypertrophy,(26) potently induced PTHrP under both proliferative and growtharrested conditions in MCT cells (Fig. 1). aFGF, which is less abundant than bFGF in growth cartilage,(30) also induced PTHrP in endochondral chondrocytes, but did so less potently and rapidly than bFGF (Fig. 2).

Our results suggest that the normal decrease in the responsiveness to bFGF that accompanies terminal chondrocyte hypertrophic differentiation(27) might function in part by removing the potential for bFGF to induce PTHrP expression. The potential for bFGF to induce expression of PTHrP in chondrocytes (Figs. 1– 4) also may be relevant to the linkage of “gain of function” mutations of certain FGF receptors with disorders of endochondral bone growth.(30) These include mutations of the transmembrane domain of FGFR-3, which account for most cases of achondroplasia.(30,51,52) FGFR-3 is abundant in resting endochondral chondrocytes in vivo.(30,51,52) Because FGFR-3 has been directly observed to mediate effects of bFGF on proliferation, hypertrophic differentiation, and cartilage matrix calcification,(51) FGFR-3 is a potential mediator of bFGFinduced PTHrP expression. Targeted disruption of the murine FGFR-3 gene has demonstrated that this receptor restrains expansion of proliferating and hypertrophic growth plate chondrocytes and promotes bone growth.(28) Similarly, targeted disruption of the PTHrP gene accelerates terminal hypertrophic chondrocyte differentiation and expands the ossification of perichondrium.(11,28) However, deficient expansion of the proliferating chondrocyte zone and marked distortions of endochondral chondrocyte columns are unique to PTHrPdeficient animals.(11,28) Our studies add weight to the supposition that PTHrP is not the only significant downstream effector for bFGF in endochondral chondrocytes. Specifically, bFGF antagonized not only BMP-induced stimulatory effects on PTHrP expression (Fig. 4) but also PPi release in MCT cells (Fig. 7). In addition, the ability of bFGF to attenuate BMP-2– and BMP-4 –induced PPi elaboration does not appear to be PTHrP-dependent in MCT cells, which demonstrated no significant reversal of bFGF-induced inhibition of PPi elaboration when incubated with neutralizing antibodies for PTHrP (R. Terkeltaub, unpublished observations). The expression of bFGF can coincide with the timed appearances of individual BMPs during endochondral de-

bFGF and BMPs REGULATE CHONDROCYTE PTHrP AND PPi RELEASE velopment.(2,14,26) Thus, the ability of BMP-6 to antagonize several activities of bFGF in MCT cells (Figs. 3, 4, and 7) suggests the possibility that timed expression of BMP-6 at the prehypertrophic stage of endochondral development may function in part by lifting the limits on endochondral mineralization imposed by bFGF.(27) In contrast to the opposing effects of BMP-2, BMP-4, BMP-6, BMP-7, and bFGF identified here in immortalized hypertrophic chondrocytes, the effects of bFGF and BMPs on chondrocytes may be synergistic at the earliest and latest stages of endochondral development.(2,25) Significantly, endochondral chondrocytes express three classes of FGF receptors (FGFR-1, FGFR-2, and FGFR-3).(30) Each of these FGF receptors has a unique pattern of expression during endochondral development and can be expressed as alternative splice variants with differing tissue specificities and affinities for aFGF and bFGF.(30) The antagonism by bFGF of multiple functional activities induced by TGF-b and BMPs in MCT cells suggests that distinct, intersecting signaling pathways for members of the FGF and TGF-b superfamilies may appear during chondrocyte hypertrophy. Chondrocytes have the potential to express more than one of the heterotetrameric receptor complexes for TGF-b superfamily proteins,(19,53,54) and functional transitions in BMP receptor signaling occur during endochondral differentiation.(55) This study illustrates the potential value of PTHrP expression and PPi release as probes for distinct mechanisms of BMP signaling. Specifically, BMP-2, BMP-4, BMP-6, and BMP-7 suppressed PTHrP expression (Figs. 3 and 4), but BMP-6 and BMP-7, which do not bind strongly to certain receptors shared by BMP-2 and BMP4,(54,56) were insufficient to induce PPi elaboration (Fig. 7). It will be of interest to determine if selective induction of PPi release by (TGF-b, BMP-2, and BMP-4) reflects known specificities within the TGF-b receptor family for the downstream signal transducing molecules of the Mad/Smad family.(53,54,57,58)

CONCLUSION Our results suggest that effects on PTHrP expression could contribute to the ability of BMP-6 to modify hypertrophic chondrocyte maturation. Furthermore, the normal decrease in bFGF responsiveness that accompanies chondrocyte hypertrophy may function in part by removing the potential for bFGF to induce PTHrP expression. Opposing effects of BMPs and bFGF on immortalized hypertrophic chondrocyte function are not limited to PTHrP expression, because they also are exerted at the level of the regulation of PPi elaboration. MCT cells may be useful in further understanding the mechanisms regulating the differentiation and mineralization-related functions of hypertrophic chondrocytes. Further study is needed to delineate the direct relationships in vivo between the control of mineralization, endochondral chondrocyte PTHrP expression, and the activities of bFGF and specific BMPs.

939

ACKNOWLEDGMENTS We acknowledge the technical assistance of Marjie Acosta Garcia, Peter Sears, Robert Santiago, Cheryl Chalberg, and Kathy Smith in these studies. This work was supported by Merit Review Awards from the Veterans Affairs Medical Service, and National Institutes of Health grants PO1AG07996 and R01CA71347, and a Biomedical Sciences Research Award (to R.T.) from the Arthritis Foundation.

REFERENCES 1. Gerstenfeld LC, Shapiro FD 1996 Expression of bone-specific genes by hypertrophic chondrocytes: Implication of the complex functions of the hypertrophic chondrocyte during endochondral bone development. J Cell Biochem 62:1–9. 2. Cancedda R, Descalzi-Cancedda F, Castagnola P 1995 Chondrocyte differentiation. Int Rev Cytol 159:265–358. 3. Lefebvre V, Garofalo S, de Crombrugghe B 1995 Type X collagen gene expression in mouse chondrocytes immortalized by a temperature-sensitive simian virus 40 large tumor antigen. J Cell Biol 128:239 –245. 4. Yang KH, Stewart AF 1996 Parathyroid hormone-related protein: The gene, its mRNA species, and protein products. In: Bilezikian JP, Raisz LG, Rodan GA (eds.) Principles of Bone Biology. Academic Press, New York, NY, U.S.A., pp. 347–362. 5. Loveys LS, Gelb D, Hurwitz SR, Puzas JE, Rosier RN 1993 Effects of parathyroid hormone-related peptide on chick growth plate chondrocytes. J Orthop Res 11:884 – 891. 6. Lee K, Lanske B, Karaplis AC, Deeds JD, Kohno H, Nissenson RA, Kronenberg HM, Segre GV 1996 Parathyroid hormonerelated peptide delays terminal differentiation of chondrocytes during endochondral bone development. Endocrinology 137: 5109 –5118. 7. Vortkamp A, Lee K, Lanske B, Segre GV, Kroneberg M, Tabin CJ 1996 Regulation of rate of cartilage differentiation by Indian hedgehog and PTH-related protein. Science 273:613– 622. 8. Amizuka N, Karaplis AC, Henderson JE, Warshawsky H, Lipman ML, Matsuki Y, Ejiri S, Tanaka M, Izumi N, Ozawa H, Goltzman D 1996 Haploinsufficiency of parathyroid hormonerelated peptide (PTHrP) results in abnormal postnatal bone development. Dev Biol 175:166 –176. 9. Amizuka N, Henderson JE, Hoshi K, Warshawsky H, Ozawa H, Goltzman D, Karaplis AC 1996 Programmed cell death of chondrocytes and aberrant chondrogenesis in mice homozygous for parathyroid hormone-related peptide gene deletion. Endocrinology 137:5055–5067. 10. Weir EC, Philbrick WM, Amling M, Neff LA, Baron R, Broadus AE 1996 Targeted overexpression of parathyroid hormone-related peptide causes chondrodysplasia and delayed endochondral bone formation. Dev Biol 93:10240 –10245. 11. Karaplis AC, Luz A, Glowacki J, Bronson RT, Tybulewicz VLJ, Kronenberg HM, Mulligan R 1994 Lethal skeletal dysplasia from targeted disruption of the parathyroid hormonerelated peptide gene. Genes Dev 8:277–289. 12. Henderson JE, Amizuka N, Warhawsky H, Biasotto D, Lanske BM, Goltzman D, Karaplis AC 1995 Nucleolar localization of parathyroid hormone-related peptide enhances survival of chondrocytes under conditions that promote apoptotic cell death. Mol Cell Biol 15:4064 – 4075. 13. Amling M, Neff L, Tanaka S, Inoue D, Kuida K, Weir E, Philbrick WM, Broadus AE, Baron R 1997 Bcl-2 lies downstream of parathyroid hormone-related peptide in a signaling pathway that regulates chondrocyte maturation during skeletal development. J Cell Biol 136:205–213. 14. Rosen V, Cox K, Hattersley G 1996 Bone morphogenetic proteins. In: Bilezikian JP, Raisz LG, Rodan GA (eds.) Prin-

940

15. 16.

17.

18.

19.

20.

21.

22.

23. 24.

25. 26. 27.

28. 29.

30.

31. 32.

TERKELTAUB ET AL. ciples of Bone Biology. Academic Press, New York, NY, U.S.A., pp. 661– 671. Reddi AH, Cunningham NS 1993 Initiation and promotion of bone differentiation by bone morphogenetic proteins. J Bone Miner Res 8 (Suppl 2):S499 –S502. Rosen V, Nove J, Song JJ, Thies RS, Cox K, Wozney JM 1994 Responsiveness of clonal limb bud cell lines to Bone Morphogenetic Protein 2 reveals a sequential relationship between cartilage and bone cell phenotypes. J Bone Miner Res 9:1759 – 1768. Hiraki Y, Inoue H, Shigeno C, Sanma Y, Bentz H, Rosen DM, Asada A, Suzuki F 1991 Bone morphogenetic proteins (BMP-2 and BMP-3) promote growth and expression of the differentiated phenotype of rabbit chondrocytes and osteoblastic MC3T3-E1 cells in vitro. J Bone Miner Res 6:1373–1385. Komaki M, Katagiri T, Suda T 1996 Bone morphogenetic protein-2 does not alter the differentiation pathway of committed progenitors of osteoblasts and chondroblasts. Cell Tissue Res 284:9 –17. Mayer H, Scutt AM, Ankenbauer T 1996 Subtle differences in the mitogenic effects of recombinant human bone morphogenetic proteins-2 to -7 on DNA synthesis on primary boneforming cells and identification of BMP-2/4 receptor. Calcif Tissue Int 58:249 –255. Duprez D, Bell EJ, Richardson MK, Archer CW, Wolpert L, Brickell PM, Francis-West PH 1996 Overexpression of BMP-2 and BMP-4 alters the size and shape of developing skeletal elements in the chick limb. Mech Dev 57:145–157. Sakano S, Murata Y, Miura T, Iwata H, Sato K, Matsui N, Seo H 1993 Collagen and alkaline phosphatase gene expression during bone morphogenetic protein (BMP)-induced cartilage and bone differentiation. Clin Orthop 292:337–344. Boden SD, McCuaig K, Hair G, Racine M, Titus L, Wozney JM, Nanes MS 1996 Differential effects and glucocorticoid potentiation of bone morphogenetic protein action during rat osteoblast differentiation in vitro. Endocrinology 137:3401– 3407. Carey DE, Liu X 1995 Expression of bone morphogenetic protein-6 messenger RNA in bovine growth plate chondrocytes of different size. J Bone Miner Res 10:401– 405. Grimsrud CD, Romano PR, DeSouza MJ, Reynolds PR, Puzas JE, Rosier RN, O’Keefe RJ 1997 BMP-6 commits chondrocytes to the hypertrophic phenotype. J Bone Miner Res 12 (Suppl 1):S161. Houston B, Thorp BH, Burt DW 1994 Molecular cloning and expression of bone morphogenetic protein-7 in the chick epiphyseal growth plate J Mol Endocrinol 13:289 –301. Kato Y, Iwamoto M 1990 Fibroblast growth factor is an inhibitor of chondrocyte terminal differentiation. J Biol Chem 265: 5903–5909. Iwamoto M, Shimazu F, Nakashima K, Suzuki F, Kato Y 1991 Reduction in basic fibroblast growth factor receptor is coupled with terminal differentiation of chondrocytes. J Biol Chem 266:461– 467. Deng C, Wynshaw-Boris A, Zhou F, Kuo A, Leder P 1996 Fibroblast growth factor receptor 3 is a negative regulator of bone growth. Cell 84:911–921. Coffin JD, Florkiewicz RZ, Neumann J, Mort-Hopkins T, Dorn GW III, Lightfoot P, German R, Howles P, Kier A, O’Toole BA, Sasse J, Gonzalez AM, Baird A, Doeschman TC 1995 Abnormal bone growth and selective translational regulation in basic fibroblast growth factor (FGF-2) transgenic mice. Mol Cell Biol 6:1861–1873. Hurley MM, Florkiewicz RZ 1996 Fibroblast growth factor and vascular endothelial cell growth factor families. In: Bilezikian JP, Raisz LG, Rodan GA (eds.) Principles of Bone Biology. Academic Press, New York, NY, U.S.A., pp. 627– 645. Kirsch T, Nah HD, Shapiro IM, Pacifici M 1997 Regulated production of mineralization-competent matrix vesicles in hypertrophic chondrocytes. J Cell Biol 137:1149 –1160. Ryan LM, McCarty DJ 1997 Calcium pyrophosphate crystal deposition disease, pseudogout, and articular chondrocalcinosis. In: Koopman WJ (ed.) Arthritis and Allied Conditions,

33.

34.

35.

36. 37. 38. 39. 40.

41. 42.

43.

44.

45. 46.

47.

48.

49.

50.

51.

13th Ed. Williams and Wilkins, Baltimore, MD, U.S.A., pp. 2103–2126. Rosenthal AK, McCarty BA, Cheung HS, Ryan LM 1993 A comparison of the effect of transforming growth factor beta 1 on pyrophosphate elaboration from various articular tissues. Arthritis Rheum 36:539 –542. Oyajobi BO, Caswell AM, Russell RG 1994 Transforming growth factor beta increases ecto-nucleoside triphosphate pyrophosphatase activity of human bone-derived cells. J Bone Miner Res 9:99 –109. Oyajobi BO, Russell RGR, Caswell AM 1994 Modulation of ecto-nucleoside triphosphate pyrophosphatase activity of human osteoblast-like bone cells by 1a,25-dihydroxyvitamin D3, 24R,25-dihydroxyvitamin D3, parathyroid hormone, and dexamethasone. J Bone Miner Res 9:1259 –1266. Anderson HC 1988 Mechanisms of pathologic calcification. Rheum Dis Clin North Am 14:303–319. Meyer JL 1984 Can biological calcification occur in the presence of pyrophosphate? Arch Biochem Biophys 231:1– 8. Meyer JL, Reddi AH 1985 Changes in pyrophosphatase activity during the de novo mineralization associated with cartilage and bone formation. Arch Biochem Biophys 242:532–539. Jung A, Bisaz S, Fleisch H 1973 The binding of pyrophosphate and two diphosphonates by hydroxyapatite crystals. Calcif Tissue Res 11:269 –280. McGaughey C 1983 Binding of polyphosphates and phosphonates to hydroxyapatite, susbsequent hydrolysis, phosphate exchange and effects on demineralization, mineralization, and microcrystal aggregation. Caries Res 17:229 –241. Fleisch HA 1997 Bisphosphonates: Preclinical aspects and use in osteoporosis. Ann Med 29:55– 62. Lotz M, Rosen F, McCabe G, Quach J, Blanco F, Dudler J, Solan J, Goding J, Seegmiller JE, Terkeltaub R 1995 IL-1b suppresses TGFb-induced inorganic pyrophosphate (PPi) production and expression of the PPi-generating enzyme PC-1 in human chondrocytes. Proc Natl Acad Sci USA 92:10364 – 10368. Ditmer LS, Burton DW, Deftos LJ 1996 Elimination of the carboxy-terminal sequences of parathyroid hormone-related 1–173 increases production and secretion of the truncated forms. Endocrinology 137:1608 –1617. Solan J, Deftos LJ, Goding JW, Terkeltaub RA 1996 Expression of the nucleoside Triphosphate pyrophosphohydrolase PC-1 is induced by basic fibroblast growth factor (bFGF) and modulated by activation of the protein kinase A and C pathways in osteoblast-like osteosarcoma cells. J Bone Miner Res 11:183–192. Rosenthal AK, Henry LA 1996 Retinoic acid stimulates pyrophosphate elaboration by cartilage and chondrocytes. Calcif Tissue Int 59:128 –133. Loveridge N, Farquharson C, Hesketh JE, Jakowlew SB, Whitehead CC, Thorp BH 1993 The control of chondrocyte differentiation during endochondral bone growth in vivo: Changes in TGF-beta and the proto-oncogene c-myc. J Cell Sci 105:949 –956. D9Angelo M, Pacifici M 1997 Articular chondrocytes produce factors that inhibit maturation of sternal chondrocytes in serum-free agarose cultures: A TGF-b independent process. J Bone Miner Res 12:1368 –1377. Berry JL, Farquharson C, Whitehead CC, Mawer EB 1996 Growth plate chondrocyte vitamin D receptor number and affinity are reduced in avian tibial dyschondroplastic lesions. Bone 19:197–203. Horan J, Dean DD, Kieswetter K, Schwartz Z, Boyan BD 1996 Evidence that interleukin-1, but not interleukin-6, affects costochondral chondrocyte proliferation, differentiation, and matrix synthesis through an autocrine pathway. J Bone Miner Res 11:1119 –1129. Vargas SJ, Naprta A, Glaccum M, Lee SK, Kalinowski J, Lorenzo JA 1996 Interleukin-6 expression and histomorphometry of bones from mice deficient in receptors for interleukin-1 or tumor necrosis factor. J Bone Miner Res 11:1736 –1744. Rousseau F, Bonaventure J, Legeal-Mallet L, Pelet A, Rozet

bFGF and BMPs REGULATE CHONDROCYTE PTHrP AND PPi RELEASE

52.

53. 54. 55. 56.

JM, Maroteaux P, Le Merrer M, Munnich A 1994 Mutations in the gene encoding fibroblast growth factor receptor-3 in achondroplasia. Nature 371:252–254. Shiang R, Thompson LM, Zhy YZ, Church DM, Fielder TJ, Bocian M, Winokur ST, Wasmuth JJ 1994 Mutations in the transmembrane domain of FGFR3 cause the most common genetic form of dwarfism, achondroplasia. Cell 78:335–342. Massague J 1996 TGFb signaling: Receptors, transducers, and Mad proteins. Cell 85:947–950. Yamashita H, Ten Dijke P, Heldin C-H, Miyazono K 1996 Bone morphogenetic protein receptors. Bone 19:569 –574. Zou H, Wieser R, Massague J, Niswander L 1997 Distinct roles of type I bone morphogenetic protein receptors in the formation and differentiation of cartilage. Genes Dev 11:2191–2203. Yamaji N, Celeste AJ, Thies RS, Song JJ, Bernier SM, Goltzman D, Lyons KM, Nove J, Rosen V, Wozney JM 1994 A mammalian serine/threonine kinase receptor specifically binds

941

BMP-2 and BMP-4. Biochem Biophys Res Commun 205: 1944 –1951. 57. Baker JC, Harland RM 1997 From receptor to nucleus: The Smad pathway. Curr Opin Genet Dev 7:467– 473. 58. Hoodless Haerry T, Abdollah S, Stapleton M, O’Connor MB, Attisano L, Wrana JL 1996 MADR1, a MAD-related protein that functions in BMP2 signaling pathways. Cell 85:489 –500.

Address reprint requests to: Robert Terkeltaub, M.D. VA Medical Center 3350 La Jolla Village Drive San Diego, CA 92161 U.S.A. Received in original form October 20, 1997; in revised form December 18, 1997; accepted February 17, 1998.