Identification of an Autocrine Chondrocyte Colony-Stimulating Factor: Chondromodulin-I Stimulates the Colony Formation of Growth Plate Chondrocytes in Agarose Culture

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS 241, 395–400 (1997)ARTICLE NO. RC977820

Identification of an Autocrine ChondrocyteColony-Stimulating Factor: Chondromodulin-IStimulates the Colony Formation of GrowthPlate Chondrocytes in Agarose Culture

Hiroyuki Inoue,* Jun Kondo,† Tatsuya Koike,‡ Chisa Shukunami,‡ and Yuji Hiraki‡,1

‡Department of Biochemistry and *Department of Orthodontics, Osaka University Faculty of Dentistry, Suita, Osaka 565,Japan; and †Research Center, Mitsubishi Chemical Corporation, Yokohama, Kanagawa 227, Japan

Received October 28, 1997

matrix are still poorly defined and their functions areChondrocytes are unique among non-transformed largely unknown.

cells in that they are capable of anchorage-indepen- Due to their growth potential and matrix synthesis,dent growth in soft agar. Fibroblast growth factor chondrocytes are capable of anchorage-independent(FGF) is known as a potent colony-stimulating factor growth resulting in the formation of colonies in softfor chondrocytes. However, cartilage extracts contain agar (5). The proliferation of chondrocytes in monolayera potent colony-stimulating activity which is not expli- culture is stimulated by a variety of growth factorscable only by contaminating FGF. We previously iso-

which include FGF, TGF-b, epidermal growth factorlated the 25 kDa cartilage-specific glycoprotein chon-(EGF), platelet-derived growth factor (PDGF) and insu-dromodulin-I (ChM-I) which stimulates the growth oflin-like growth factor-I (IGF-I) (6-8). However, amongchondrocytes. In the present study, we observed thatthese factors, only a few agents are capable of stimulat-ChM-I stimulates the colony formation of rabbiting colony formation of chondrocytes (9). To date, FGF-growth plate chondrocytes in agarose culture. ChM-I2 is the only fundamental factor stimulating the colonyalone weakly stimulated the formation of chondrocyteformation of these cells (5). TGF-b1 is far less active,colonies, but it markedly stimulated colony formationbut it markedly potentiates the colony-stimulating ac-synergistically in the presence of an optimal dose oftion of FGF-2 (9).FGF-2. This effect was dependent on the dose of ChM-

In cartilage, chondrocytes are sparsely located andI. These results suggest that ChM-I participates in anseparated with each other by a large amount of extra-autocrine signaling mechanism for the anchorage-in-cellular matrix without the direct cell-cell contacts. Itdependent growth of chondrocytes in vitro. q 1997

has been assumed that a growth response of chondro-Academic Press

cytes in three-dimensional culture such as agarose cul-ture more naturally reflects the in vivo behavior of thecells, compared to that of cells in monolayer culture.

The growth of cartilage plays a vital role during endo- Therefore, a search for the chondrocyte colony-stimu-chondral bone development in the embryonic stage and lating factors is important for understanding thein the longitudinal growth of bone in young animals. growth regulation of cartilage in vivo. Kato and hisGrowth plate chondrocytes proliferate rapidly in vitro, coworkers demonstrated the presence of a novel colony-maintaining an active synthesis of matrix proteins stimulating activity in cartilage (10), the molecular ba-

sis of which has not been identified. In the presentsuch as type II collagen and proteoglycans (1). The ex-study, we characterized a colony-stimulating activity intracellular matrix filling the large intercellular spacecartilage extracts, and found that the cartilage-specificin cartilage functions as a reservoir of growth and dif-growth-modulating factor chondromodulin-I (ChM-I)ferentiation factors such as fibroblast growth factormarkedly potentiated the action of FGF on the colony(FGF) and transforming growth factor-b (TGF-b) (2-formation of rabbit growth plate chondrocytes in agar-4). However, the growth-regulating components in theose culture (11).

MATERIALS AND METHODS1 To whom correspondence should be addressed at Department ofBiochemistry, Osaka University Faculty of Dentistry, Suita, Osaka Materials. ChM-I was purified from fetal bovine epiphyseal carti-

lage, as previously described (11). In brief, fetal bovine epiphyseal565, Japan. Fax: 81 6 879 2889. E-mail: [email protected].

0006-291X/97 $25.00Copyright q 1997 by Academic PressAll rights of reproduction in any form reserved.

395

AID BBRC 7820 / 6942$$$601 11-21-97 08:46:50 bbrcg AP: BBRC

Vol. 241, No. 2, 1997 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

cartilage was homogenized and extracted with 1 M guanidinium chlo-ride, and then fractionated with 45-60% acetone. The precipitateswere subfractionated by ultrafiltration into the 10-50 kDa fractionin 4 M guanidinium chloride (pH 7.4), and then applied to a Sepha-cryl S-200 column. The bioactive fractions were pooled and applied toa heparin-Toyopearl affinity column equilibrated with 20 mM sodiumphosphate (pH 7.4) containing 0.15 M NaCl and 0.03% CHAPS. Thematerials eluted with from 0.5 M to 1.2 M NaCl were pooled(Hep.01.2M fraction). Finally, ChM-I was purified to homogeneityby high-performance liquid chromatography (HPLC) on a reverse-phase column of YMC C4 equilibrated with 25% acetonitrile/2-propa-nol (3/7, v/v) in 0.1% trifluoroacetic acid. The sequence data of ChM-I are available from GenBank under the accession number M65081.Recombinant human FGF-2 and bovine bone-derived TGF-b2 weregifts from Dr. T. Kurokawa (Takeda Chemical Industries, Osaka,Japan) and Dr. D. M. Rosen (Collagen Corporation, Palo Alto, CA),respectively. Hydrocortisone and transferrin were purchased fromWako Pure Chemicals (Osaka). Alcian blue 8 GX, 3-(4,5-dimethylthi-azol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) and type VIIagarose were from Sigma Chemicals (St. Louis, MO).

FIG. 1. Stimulation of proteoglycan synthesis in rabbit chondro-Isolation of chondrocytes. Chondrocytes were isolated from the cytes by ChM-I. Rabbit growth plate chondrocytes were grown to

growth plate cartilage of the ribs of young male New Zealand rabbits confluency, and preincubated in DMEM containing 0.3% FBS for 24by collagenase digestion, as described (12). The isolated cells (11 104

hr. The medium was then replaced by DMEM containing 0.3% FBScells/well) were plated in 48-miltiwell plates in Dulbecco’s modified and test samples, and incubated for another 20 hr. The cultures wereEagle’s medium (DMEM) containing 10% fetal bovine serum (FBS). labeled with [35S]sulfate for the last 17 hr. Radioactivity incorporatedWhen the culture reached Ç50% confluency on day 3, the medium into proteoglycans was determined. Purified ChM-I (h) and Hep.-was replaced by DMEM containing 0.5% FBS and test samples. The 1.2M fraction (j) were added at the indicated doses. [35S]Sulfatecultures were maintained in 5% CO2 in air at 377C for another 3 incorporation in control wells was 6333 { 534 dpm/well (100 { 8%).days until the cultures reached confluency. The cells were then fixed Values are the means { S.D. of triplicate wells.with methanol for 10 min, and stained with 0.1% alcian blue in 0.1M HCl for 2 hr.

Proteoglycan synthesis in chondrocytes. Proteoglycan synthesisthe extracts was purified from the heparin-bindingwas assayed as previously described (13). Rabbit growth plate chon-

drocytes (1 1 104 cells/well) were plated in 96-multiwell plates, and fraction eluted with from 0.5 M to 1.2 M NaCl (Hep.-grown to confluency in DMEM containing 10% FBS. The cells were 1.2M fraction) and named ChM-I (11). ChM-I is a glyco-then preincubated in DMEM containing 0.3% FBS for 24 hr. The protein with an average molecular mass of Ç25 kDamedium was then replaced by the same medium containing test

on sodium dodecyl sulfate-polyacrylamide gel electro-samples and 0.3% FBS, and incubated for another 20 hr. The cultureswere labeled with 5 mCi/ml [35S]sulfate for the last 17 hr. [35S]Sulfate phoresis (SDS-PAGE). In the present study, the growthincorporated into proteoglycans was determined after pronase E di- of chondrocytes was synergistically stimulated bygestion and precipitation by 1% cetylpyridinium chloride in the pres- ChM-I in the presence of FGF-2. As shown in Fig. 1,ence of chondroitin sulfate.

the majority of the proteoglycan synthesis-stimulatingColony formation in agarose culture. The colony formation assay activity in cartilage extracts was also recovered in the

was performed by the reported method with some modifications (5).Hep.-1.2M fraction (11, 16). Reverse-phase HPLC indi-A 0.5-ml of base layer of 0.72% agarose in Ham’s F-12 medium wascated that the proteoglycan synthesis and DNA synthe-added to each well of 12-multiwell plates, and the basal layer was

allowed to solidify. Rabbit growth plate chondrocytes (5 1 103 cells/ sis stimulating activities present in Hep.-1.2M fractionwell) were suspended in 0.5-ml of 0.41% agarose in F-12 medium were solely attributed to ChM-I (11). ChM-I stimulatedsupplemented with 5% FBS, 2 1 1007 M hydrocortisone, and 60 mg/ proteoglycan synthesis in the chondrocytes in a dose-ml transferrin, and used as an overlay. The cells were incubated

dependent manner (Fig. 1). Maximal stimulation wasovernight, and test samples were then added evenly on the top layer.observed at a dose range of 200-600 ng/ml.Growth factors were added only once to the culture on day 1, and

the cultures were maintained for ten days without further supple- TGF-b has an effect similar to that of ChM-I onmentation of factors. Peptide growth factors were dissolved in 1-5 ml monolayer cultures of chondrocytes (7): TGF-b wasof F-12 medium containing 0.1% fatty acid-free bovine serum albu- observed to stimulate proteoglycan synthesis andmin (BSA). The colonies were counted over a glass plate with 2-

synergistically stimulated DNA synthesis in chon-mm grids under a phase-contrast microscope. In some experiments,drocytes in the presence of FGF-2. However, unlikecolonies of chondrocytes were stained by incubation with 5 mg/ml

MTT dissolved in phosphate-buffered saline (PBS) for 6 hr (14). ChM-I, TGF-b transformed the cellular morphologyof chondrocytes into a dedifferentiated spindle-likeshape when the cells were treated in the presence ofRESULTS AND DISCUSSIONFGF-2 (13, 17). TGF-b2 and TGF-b1 each mobilizedproteoglycans produced by chondrocytes to releaseAs we previously reported, the guanidine extracts of

cartilage stimulate DNA synthesis and proteoglycan them into the culture medium, although they stimu-lated the net synthesis of proteoglycans. Thus, TGF-synthesis in the confluent monolayer culture of rabbit

chondrocytes (15). The growth-promoting activity in b reduced the accumulation of alcian blue-stainable

396



the extracts (Hep.-1.2M fraction) potentiated the col-ony-stimulating action of FGF-2 (Fig. 3C). More than500 colonies were formed by the Hep.-1.2M fraction (2mg/ml) in the presence of FGF-2 after 8 days of incuba-tion.

ChM-I alone stimulated colony formation in agaroseculture. Approximately 50-70 colonies were formedafter 8 days of incubation (Fig. 4A). The colony-stimu-lating action of ChM-I was evidently weaker than thatof FGF-2. However, when ChM-I was added in the pres-ence of FGF-2, it markedly stimulated colony formationin a dose-dependent manner (Fig. 4A). The maximaleffect was observed at a ChM-I dose of 200-600 ng/ml.Depending on the experiments, 800-1,000 cells formedcolonies among 5,000 cells upon treatment with 600ng/ml ChM-I and 1 ng/ml FGF-2 after 8 days (Fig. 4A).As shown in Fig. 3D, individual cells in the coloniestook on a rounded and matured morphology. When cellbodies in the colonies were visualized by MTT staining,the rounded cells were clearly separated by the abun-dant extracellular matrix in colonies cultured in thepresence of ChM-I and FGF-2 compared to FGF-2 aloneFIG. 2. Effects of growth factors on the accumulation of cartilage

matrix in monolayer cultures of growing chondrocytes. Cells (1 1 104 (Figs. 3E and 3F). The number of large colonies (¢ 0.2cells/well) were plated in 48-multiwell plates, and grown to Ç50% mm in diameter) was increased in the cultures treatedconfluency in the presence of 10% FBS. On day 3, the medium was with ChM-I and FGF-2, to over ten-fold that of thereplaced by DMEM containing FGF-2 (1 ng/ml), ChM-I (200 ng/ml),

cultures treated with FGF-2 alone (Fig. 4B). Thus, theor TGF-b2 (3 ng/ml) in the presence of 0.5% FBS. The control cultureswere treated with 0.5% FBS alone. Cultures were maintained to actions of Hep.-1.2M fraction could be substituted bybecome confluent for another 3 days. Cells were stained with 0.1% purified ChM-I.alcian blue in 0.1 M HCl. As reported earlier (9), TGF-b1 potentiated the col-

ony-stimulating action of FGF-2, although TGF-b1

alone showed only a weak action on colony formation.TGF-b2 also stimulated the colony formation of rabbitproteoglycans on the cell surface when it was added

to the monolayer culture of chondrocytes at the grow- chondrocytes (Fig. 4C), but the effect of TGF-b2 alonewas weaker than that of ChM-I. TGF-b2 similarly po-ing stage in the present study (Fig. 2). The reduction

of proteoglycan accumulation induced by TGF-b was tentiated the FGF-2 action on the colony formation ofrabbit chondrocytes (Fig. 4C). Our histological analysismore pronounced when the growth of chondrocytes

was synergistically stimulated in the presence of of sections of the colonies indicated that the architec-ture of the chondrocyte colonies was fundamentally dif-FGF-2. However, ChM-I only marginally affected

proteoglycan accumulation (Fig. 2). Thus, ChM-I af- ferent from that of transformed fibroblasts. Well-differ-entiated chondrocytes were buried in a mass of matrixfected the matrix metabolism of chondrocytes in a

manner similar to but distinct from TGF-b. in chondrocyte colonies. Along with their growth, chon-drocytes form colonies by producing and accumulatingAs shown in Fig. 3A, the rabbit growth plate chondro-

cytes did not form colonies even in the presence of 5% a large amount of cartilage matrix around the cells.For the potentiation of the colony-stimulating action ofFBS when they were cultured in agarose. During 10

days of incubation, the number of colonies remained FGF-2, it is probably necessary that the agent is capa-ble of stimulating matrix synthesis in chondrocytes asessentially at zero in the presence of 5% FBS alone

(Fig. 4). The MTT staining indicated that viability of well as synergistically stimulating DNA synthesis inthe presence of FGF-2. ChM-I and TGF-b appeared tochondrocytes was maintained at over 90% of the cells

in culture (data not shown). In contrast, the colony fulfill these requirements. It should be noted here thatPDGF and IGF-I also stimulate both DNA synthesisformation was augmented by the supplementation of

FGF-2 (1 ng/ml) in culture (Fig. 3B). After 8 days of and matrix synthesis in chondrocytes (8, 18, 19). How-ever, they failed to stimulate colony formation of chon-incubation, approximately 200 cells formed colonies

among the 5,000 cells that were seeded in the culture drocytes in agar (5, 9). Thus, it is not known at presentwhy only ChM-I and TGF-b could potentiate the col-(Fig. 4). In agreement with the previous report (10),

the extracts from fetal bovine epiphyseal cartilage ony-stimulating action of FGF-2. A search for the spe-cific cell-surface receptor for ChM-I is now underway.markedly stimulated the colony formation of chondro-

cytes in agarose culture. The heparin-bound fraction of Seyedin reported that TGF-b purified from bone

397



FIG. 3. Colony formation of rabbit chondrocytes in agarose. Growth plate chondrocytes (5 1 103 cells/well) were suspended in 0.41%agarose in F-12 medium containing 5% FBS, 2 1 1007 M hydrocortisone, and 60 mg/ml transferrin. The cell suspensions were overlaid ona base agarose layer. After incubation overnight, test samples were added. On day 5, colony formation was examined under a phase-contrastmicroscope. Cells were treated with 0.1% BSA alone (A), 1 ng/ml FGF-2 (B), 2 mg/ml Hep.-1.2M fraction in the presence of 1 ng/ml FGF-2(C), and 200 ng/ml ChM-I in the presence of 1 ng/ml FGF-2 (D). Panels E and F show the MTT staining of colonies in the cultures treatedwith 1 ng/ml FGF-2 alone (E) and 600 ng/ml ChM-I in the presence of 1 ng/ml FGF-2 (F). Bars represent 200 mm.

stimulated the chondrogenic differentiation and colony logical conditions. Thus, it is unlikely that activatedTGF-b can be a physiological colony-stimulating factorformation of chondrocytes in agarose culture (20-22).

Cartilage contains TGF-b subtypes (2, 23). As pre- for chondrocytes in vivo. Moreover, the architecture ofthe chondrocyte colonies in the TGF-b treated culturesviously suggested (24), the latent forms of TGF-b in

cartilage are activated by metalloproteinases and plas- was distinct from that in the ChM-I treated cultures:the cells transformed into flat cells were loosely con-minogen activator in association with remodeling of

cartilage matrix during cellular hypertrophy and min- nected with matrix by TGF-b.In an earlier report, Kato and his coworkers de-eralization (the late-phase differentiation). However,

there is no evidence suggesting that the latent forms scribed chondrocyte colony-stimulating activity in car-tilage extracts which was distinct from that induced byof TGF-b were activated in cartilage during the growth

and maturation stages of chondrocytes under physio- TGF-b (10). They estimated the molecular weight of

398



FIG. 4. Time course of colony formation in agarose. Cells were grown in agarose in the presence of 5% FBS, hydrocortisone (2 1 1007

M), and transferrin (60 mg/ml), as described in the legend to Fig. 3. Colonies were counted over a glass plate with 2-mm grids under aphase-contrast microscope. In (A), cells were treated with 0.1% BSA alone (h), 600 ng/ml ChM-I (L), 1 ng/ml FGF-2 (s), 60 ng/ml ChM-Iand 1 ng/ml FGF-2 (n), 200 ng/ml ChM-I and 1 ng/ml FGF-2 (m), or 600 ng/ml ChM-I and 1 ng/ml FGF-2 (j). In (B), the numbers of largecolonies (¢ 0.2 mm in diameter) were counted. Cells were treated with 0.1% BAS alone (h), 1 ng/ml FGF-2 (s), or 600 ng/ml ChM-I and1 ng/ml FGF-2 (j). In (C), cells were treated with 0.1% BSA alone (h), 3 ng/ml TGF-b2 (n), 1 ng/ml FGF-2 (s), or 3 ng/ml TGF-b2 and 1ng/ml FGF-2 (m). Values are the means of triplicate wells. The S.D. values were within 10% of the means.

this factor as 24-27 kDa. A crude extract exhibited a ritorial space of extracellular matrix in the avascularzones of cartilage, i.e. the resting, proliferating andpotent stimulatory activity, but the colony-forming ef-

ficiency was evidently reduced upon stimulation even early hypertrophic zones (26). However, it was absentin the late hypertrophic and calcifying zones of carti-at a maximal dose after purification by serum-protein

affinity chromatography (10). Their observations sug- lage which allow vascular invasion during endochon-dral bone formation (26). Mature ChM-I was tightlygested that the growth factors of the FGF family were

eliminated by the affinity step. The potent colony-stim- bound to cartilage matrix so that it does not diffuseout to the neighboring vascularized connective tissue.ulating activity in the crude extracts may arise from a

colony-stimulating factor which potentiates the action Distribution of ChM-I mRNA was completely over-lapped to the sites of protein localization in avascularof contaminating FGF. As previously reported (11), pu-

rified ChM-I migrated as a broad band of 25 kDa on cartilage, as shown by in situ hybridization (26). Theseresults indicate that ChM-I may play an unique roleSDS-PAGE. Moreover, ChM-I specifically interacts and

forms a complex with several serum proteins including for the rapid growth of cartilage in vivo, while it confersan anti-angiogenic property on cartilage.albumin (J. Kondo, unpublished data). The physico-

chemical properties of ChM-I are compatible with thoseof the colony-stimulating factor in cartilage described ACKNOWLEDGMENTSearlier by Kato (10). These results suggest that ChM-I is a novel colony-stimulating factor for chondrocytes. This paper is dedicated to Professor Emeritus Fujio Suzuki on

the occasion of his retirement from Osaka University. We gratefullyAs shown in Fig. 4, chondrocytes requires a high doseacknowledge Dr. T. Kurokawa (Takeda Chemical Industries, Osaka)of ChM-I (ú 200 ng/ml) for colony formation, comparedand Dr. D. M. Rosen (Collagen Corporation, Palo Alto, CA) for theto TGF-b. In compatible with this observation, the generous gifts of human recombinant FGF-2 and TGF-b, respec-

ChM-I content in epiphyseal cartilage was very high. tively. Chisa Shukunami is a recipient of a Research Fellowship ofNeame and his coworkers estimated the content of the Japan Society for the Promotion of Science for Young Scientists.

This work was supported in part by Special Coordination Funds forChM-I in cartilage to be ú 1 mg/g tissue which meetsPromoting Science and Technology from the Science and Technologythe high dose-requirement of ChM-I for action in vivoAgency of Japan and Grants-in-Aid from the Ministry of Education,(25). Unlike TGF-b, ChM-I is specifically expressed in Science and Culture of Japan (#8672363 to HI; #7672014 to YH).

cartilage, and secreted out in a mature active form afterproteolytic cleavage from its precursor protein (11, 26).

REFERENCESAs we reported (26), ChM-I protein inhibited angiogen-esis due to its growth inhibitory action on vascular en- 1. Suzuki, F. (1994) in Bone and Mineral Research (Heersche,dothelial cells. In contrast, TGF-b stimulates angiogen- J. N. M., and Kanis, J. A., Eds.), Vol. 8, pp. 115–142, Elsevier

Science Publishers, New York.esis in vivo (27). Immunohistochemistry indicated thatChM-I protein was specifically localized to the interter- 2. Ellingsworth, L. R., Brennan, J. E., Fok, K., Rosen, D. M., Bentz,

399



H., Piez, K. A., and Seyedin, S. M. (1986) J. Biol. Chem. 261, 14. Montesano, R., Schaller, G., and Orci, L. (1991) Cell 66, 697–711.12362–12367.

15. Suzuki, F., Hiraki, Y., and Kato, Y. (1987) in Methods in Enzy-3. Twal, W. O., Vasilatos Younken, R., Gay, C. V., and Leach, R. M.,mology (Barnes, D., and Sirbasku, D. A., Eds.), Vol. 146, pp. 313–Jr. (1994) J. Bone Miner. Res. 9, 1737–1744.320, Academic Press, San Diego, CA.4. Chang, S. C., Hoang, B., Thomas, J. T., Vukicevic, S., Luyten,

16. Hiraki, Y., Inoue, H., Kondo, J., Kamizono, A., Yoshitake, Y.,F. P., Ryba, N. J. P., Kozak, C. A., Reddi, A. H., and Moos, M. J.Shukunami, C., and Suzuki, F. (1996) J. Biol. Chem. 271, 22657–(1994) J. Biol. Chem. 269, 28227–28234.22662.

5. Kato, Y., Iwamoto, M., and Koike, T. (1987) J. Cell. Physiol. 133,17. Inoue, H., Kato, Y., Iwamoto, M., Hiraki, Y., Sakuda, M., and491–498.

Suzuki, F. (1989) J. Cell. Physiol. 138, 329–337.6. Hiraki, Y., Kato, Y., Inoue, H., and Suzuki, F. (1986) Eur. J.

18. Kato, Y., Iwamoto, M., Koike, T., Suzuki, F., and Takano, Y.Biochem. 158, 333–337.(1988) Proc. Natl. Acad. Sci. USA 85, 9552–9556.

7. Hiraki, Y., Inoue, H., Hirai, R., Kato, Y., and Suzuki, F. (1988) 19. Inoue, H., Hiraki, Y., Ohta, Y., and Suzuki, F. (1995) Dentist.Biochim. Biophys. Acta. 969, 91–99. Japan 32, 41–45.

8. Wroblewski, J., and Edwall, C. (1992) Cell Biol. Int. Rep. 16, 20. Thompson, A. Y., Piez, K. A., and Seyedin, S. M. (1985) Exp. Cell133–144. Res. 157, 483–494.

9. Iwamoto, M., Sato, K., Nakashima, K., Fuchihata, H., Suzuki, 21. Seyedin, S. M., Thompson, A. Y., Bentz, H., Rosen, D. M., Mc-F., and Kato, Y. (1989) Biochem. Biophys. Res. Commun. 159, Pherson, J. M., Conti, A., Siegel, N. R., Galluppi, G. R., and Piez,1006–1011. K. A. (1986) J. Biol. Chem. 261, 5693–5695.

10. Kato, Y., Nakashima, K., Sato, K., Yan, W., Iwamoto, M., and 22. Seyedin, S. M., Segarini, P. R., Rosen, D. M., Thompson, A. Y.,Suzuki, F. (1991) in Methods in Enzymology (Barnes, D., Bentz, H., and Graycar, J. (1987) J. Biol. Chem. 262, 1946–1949.Mather, J. P., and Sato, G. H., Eds.), Vol. 198, pp. 416–424, Aca- 23. Carrington, J. L., Roberts, A. B., Flanders, K. C., Roche, N. S.,demic Press, San Diego. and Reddi, A. H. (1988) J. Cell Biol. 107, 1969–1975.

11. Hiraki, Y., Tanaka, H., Inoue, H., Kondo, J., Kamizono, A., and 24. Dean, D. D., Boyan, B. D., Muniz, O. E., Howell, D. S., andSuzuki, F. (1991) Biochem. Biophys. Res. Commun. 175, 971– Schwartz, Z. (1996) Calcif. Tissue. Int. 59, 109–116.977. 25. Neame, P. J., Treep, J. T., and Young, C. N. (1990) J. Biol. Chem.

265, 9628–9633.12. Shimomura, Y., Yoneda, T., and Suzuki, F. (1975) Calcif. TissueRes. 19, 179–187. 26. Hiraki, Y., Inoue, H., Iyama, K.-i., Kamizono, A., Ochiai, M.,

Shukunami, C., Iijima, S., Suzuki, F., and Kondo, J. (1997) J.13. Hiraki, Y., Inoue, H., Shigeno, C., Sanma, Y., Bentz, H., Rosen,Biol. Chem., in press.D. M., Asada, A., and Suzuki, F. (1991) J. Bone Miner. Res. 6,

1373–1385. 27. Yang, E. Y., and Moses, H. L. (1990) J. Cell Biol. 111, 731–741.

400


Documents

Identification of an Autocrine Chondrocyte Colony-Stimulating Factor: Chondromodulin-I Stimulates the Colony Formation of Growth Plate Chondrocytes in Agarose Culture