[CANCER RESEARCH 47, 4961-4966, September 15, 1987|

Characterization of a Human Osteosarcoma Cell Line (Saos-2)

with Osteoblastic PropertiesSevgi B. Rodan, Yasuo Imai, Mark A. Thiede, Gregg Wesolowski, David Thompson, Zvi Bar-Shavit, Susan Shull,Kenneth Mann, and Gideon A. Rodan1

Department of Bone Biology and Osteoporosis, Merck Sharp & Dohme Research Laboratories, West Point, Pennsylvania 19486 [S. B. R., Y. I., M. A. T., G. W., D. T.,G. A. K.J; The Hebrew University, Hadassah Medical School, Jerusalem, Israel [Z. B-S.J; and Department of Biochemistry, University of Vermont, Burlington,Vermont 05405 fS. S., K. M.¡

ABSTRACT

This study examines the osteoblastic properties of the establishedhuman osteosarcoma cell line Saos-2. Saos-2 cells inoculated into diffusion chambers, which were implanted i.p. into nude mice, producedmineralized matrix in 4 of 6 chambers at 8 weeks. In 5 of 6 chambersthere was a strong positive alkaline phosphatase reaction. In culture thealkaline phosphatase levels increased with time and cell density, reachingvery high levels at confluence: 4-7 /imol/mg protein/min. The cells showa sensitive adenylate cyclase response to parathyroid hormone, 50%effective dose = 2.8 UM,which increases with cell density and is furtherraised by dexamethasone treatment. They also exhibit typical binding of1-25-dihydroxyvitamin I), to 3.2S receptor protein with an apparent K¿of 0.21 IIM;the numbers of sites per cell were 3,300 at 50,000 cells/cm2and 1,800 at 280,000 cells/cm2. The presence of osteonectin was visual

ized with a monoclonal antibody which revealed a reticular pattern onthe cell surface. Osteonectin was also detected in the medium by Westernblots, migrating at around M, 40,000 in nonreduced gels and M, 44,000in reduced gels. The Saos-2 cells thus possess several osteoblasticfeatures and could be useful as a permanent line of human osteoblast-like cells and as a source of bone-related molecules.

INTRODUCTION

Malignant cells often express differentiated features of thetissue of origin (1) along with cellular immortality. This combination of properties makes it possible to establish cell linesin which tissue-specific products and phenotype-related cellularfunctions can be studied. The rat osteosarcoma-derived celllines UMR and ROS have been used in this fashion (2, 3). Theobject of this study was to characterize a human cell line withosteoblastic properties, since cellular products (peptides/pro-teins) as well as cellular responses (4) are sometimes speciesspecific. Studies of isolated cells from embryonic calvaría,osteosarcoma, and other types of bone, along with histológica!and histochemical studies, have helped to establish a set ofproperties believed to be associated with the osteoblastic phe-notype (3). These include elevated alkaline phosphatase, parathyroid hormone-stimulatable adenylate cyclase, synthesis andsecretion of type I collagen, and production of mineralizedmatrix in Millipore filter chambers. Other features include theexpression of bone 7-carboxyglutamic acid-containing protein,fibronectin, osteonectin, sialoprotein, proteoglycans, collagen-ase and others, and the modulation of the above properties by1,25(OH)2Dj,2 PTH, glucocorticoids, prostanoids, growth factors, and so forth. Saos-2 is an established human osteosarcoma

Received 3/6/87; revised 6/5/87; accepted 6/10/87.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at the Department of

Bone Biology and Osteoporosis, Merck Sharp & Dohme Research Laboratories,West Point, PA 19486.

'The abbreviations used are: 1,25(OH)2D3, 1-25-dihydroxyvitamin D3; PTH,

parathyroid hormone; FBS, fetal bovine serum; BSA, bovine serum albumin;PBS, phosphate-buffered saline; KTED buffer, 300 mM KCI, 10 mw Tris-HCl,1.5 mM EOT A, 5 mM dithiothreitol, pH 7.4; SDS, sodium dodecyl sulfate.

cell line which has been partially characterized (5, 6). We showin this investigation that several of the osteoblastic featureslisted above are present in this human cell line.

MATERIALS AND METHODS

Materials. RPMI powder, FBS, and kanamycin were from GrandIsland Biological Co. (Grand Island, NY). Cells were grown in tissueculture dishes purchased from Costar (Cambridge, MA). [3H]Adenine(17 Ci/mmol) was from New England Nuclear (Boston, MA). [3H]-

1,25(OH)2D3 (180 Ci/mmol) was obtained from Amersham Corp.(Arlington Heights, IL). Dexamethasone, isobutylmethylxanthine, adenine, cAMP, and BSA were from Sigma Chemical Co. (St. Louis,MO). Synthetic human PTH was purchased from Bachern Biochemicals(Terranee, CA). Fluorescein isothiocyanate-conjugated goat anti-rabbitIgG and peroxidase-conjugated goat antimouse IgG were purchasedfrom Cappel Laboratories (Cochranville, PA). The radioimmunoassaykit for osteocalcin was purchased from Immunonuclear Corporation(Stillwater, MN). l,25(OH)2D3-treated medium from human bone cellswas generously provided by Dr. B. Sacktor (National Institute of Aging,Baltimore, MD). 1,25(OH)2D3 was a gift of Dr. M. Uskokovic (Hoffmann-La Roche, Nutley, NJ). Monoclonal anti-bovine osteonectinantibody was developed and characterized in the laboratory of Dr.Kenneth Mann. Saos-2 cells were obtained from Dr. H. Harris of theUniversity of Pennsylvania. This cell line as well as Saos-1 wereestablished in 1973 (7). Labtek culture chamber/slides were purchasedfrom Miles Scientific (Naperville, IL). Lucite rings with holes andthread were from Millipore Corp. (Bedford, MA).

Cell Culture. Saos-2 cells were cultured in RPMI medium supplemented with 10% FBS, 2 mM L-glutamine, and 0.01% kanamycin in ahumidified atmosphere of 95% air-5% CO2 at 37°C.The cells were

passaged using 0.01% trypsin-1 mM EDTA every 10 days. For cAMPmeasurement, cells were seeded into 6 x 35-mm-well culture dishes.

Chromosomal analysis carried out by Dr. Warren Nichols (MerckSharp and Dohme Research Laboratories, West Point, PA) showedthat this is a heteroploid line compatible with human cells.

Adenylate Cyclase Assay. cAMP was measured in whole cells aspreviously described (8) by the conversion of [3H]ATP to pHjcAMP.Briefly, the cells were incubated with 1 uCi [3H]adenine for 2 h at 37'C.

Cells were washed, preincubated with I mM isobutylmethylxanthinefor 10 min, and incubated with hormones for 5 min. [3H]cAMP was

isolated by the method of Salomon et al. (9).Alkaline Phosphatase Assay. Alkaline phosphatase activity was mea

sured as previously described (10). The assay mixture contained 0.1 M2-amino-2-methyl-l-propanol (pH 10.5), 2 mM MgCI2, and 2 mM 4-nitrophenylphosphate in 0.2 ml.

Protein Determination. Protein was measured by the method ofSpector (11) using BSA as standard.

Immunofluorescence Microscopy. Cells were plated onto Labtek culture chamber/slides at a density of 50,000/cm2 in RPMI medium

supplemented with 10% FBS and 2 mM glutamine. After 24 h ofincubation, the medium was removed and serum-free medium wasadded to the cells for another 24 h. All of the staining procedures weredone at room temperature. Cells were washed twice with calciummagnesium-free Hanks' balanced salt solution, fixed with 1% paratomi

aldehyde in PBS for 20 min, and permeabilized with 0.01% NonidetP-40 for 20 min. At each subsequent step the cells were washed withPBS. Monoclonal antibody against osteonectin (1:100 dilution in PBS)was added to each slide for 1 h. Cells were incubated with a 1:200

4961

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

OSTEOBLASTIC HUMAN OSTEOSARCOMA CELL LINE

dilution of goat anti-mouse IgG conjugated with fluorescein isothio-cyanate (which had been preabsorbed to the Saos-2 cells for 1 h) for 30min. An inverted glasslip with a drop of n-propyl gallate in a 1:1mixture of PBS and glycerol was put on the slide and observed with aNikon microscope.

Preparation of Cell Cytosol and | 'I I| 1,25«)I I);L><Binding Assay. Cells

were incubated with 0.01% trypsin-1 HIMEDTA, collected by centrif-ugation at 800 x g, washed twice in cold Dulbecco's PBS containing

Ca and Mg (137 mM NaCl, 2.68 HIMKC1, 8 mM Na2HPO4, 2.68 mMKH2PO4) containing 10 mg/ml BSA, then rinsed 3 times in cold PBS,resuspended in 300 mM KC1, 10 HIMTris-HCl, 1.5 mM EDTA, 5 mMdithiothreitol, pH 7.4 (KTED buffer), homogenized, and kept on icefor 30 min. The homogenates were centrifugée!at 40,000 x g for 15min, the supernatant was kept on ice, and the pellet was taken up inKTED buffer for an additional 30 min on ice. After additional centrif-ugation, the 2 supernatants were pooled and aliquots (220 n\ containingcytosol derived from 5 x IO6 cells) were combined and assayed for[3H]l,25(OH)2Dj binding at 4°Cfor 18 h. Parallel incubations contained [3H]1,25(OH)2D3 plus 100-fold excess nonradioactive hormone

to assess nonspecific binding. Bound ligand was recovered using thehydroxylapatite method (12).

Sucrose Gradient Analysis of [•'!I]1,25(OI !),!>, Binding Moiety. Cellextracts were incubated with 1 nM [3H]1,25(OH)2D3 ±100 nM nonradioactive hormone, layered on linear gradients of 5-20% sucrose inKTED buffer, and centrifuged at 4°Cfor 20 h in a Beckman L3-50

ultracentrifuge (SW 50.1 rotor) at 50,000 rpm (13). Fractions of 160i/l were collected and counted. Sedimentation coefficients were estimated by comparison with protein markers (BSA, 6.0 S; ovalbumin,3.7 S; and chymotrypsinogen, 2.54 S) that were centrifuged in the sameexperiment.

Preparation and Implantation of Diffusion Chambers. Lucite ringswith holes and thread were glued to 0.45-Mm pore size membrane filtersand polyethylene tubing to make chambers with a volume of 200 ,«1.The chambers were sterilized by ethylene oxide. Cell suspensions (about1 x IO6 cells) were injected into the chambers, which were then

implanted i.p. in athymic nu/nu (14) mice, 2 per animal. Some of theanimals were implanted with 1 chamber containing Saos-2 and 1containing ROS 17/2.8 cells as positive controls. The chambers wereharvested after 8 weeks.

Processing for Light Microscopic Examination. The chambers wereremoved from the animals immediately at sacrifice and placed intact in70% ethanol for the fixation of cells and matrix. The chambers wereimpregnated for 48 h and embedded in glycol methacrylate undervacuum. Sections, 4 urn thick, were removed from the chambers in atransverse plane. Multiple levels were sectioned in each chamber. Thesections were stained with von Kossa for mineral deposits and counter-stained with toluidine blue and for alkaline phosphatase (Sigma). Allsections were examined under bright light microscopy.

Preparation of Cell-conditioned Media for Western Blot Analysis.Saos-2 cells were grown to confluence and maintained in serum-freemedium for the last 24 h prior to harvesting.

The medium was centrifuged to remove cellular debris and concentrated 100-fold by centrifugation at 400C x g in a Centricon-10 tube.One half of the concentrated medium received an equal volume ofdouble-concentration SDS cocktail (0.125 M Tris-HCl, pH 6.8, 4%SDS, and 20% glycerol) containing 1.46 M .<-inert-aptoeiliaiiol; theother half received the same solution without /3-mercaptoethanol. Samples were then fractionated on SDS-12.5% polyacrylamide gels according to Laemmli (15). Protein bands were electrophoretically transferredto nitrocellulose membranes according to Towbin et al. (16) for 14 h at4"C' and 0.2 amperes. Following transfer, the filter was blocked in a

solution of PBS (10 mM sodium phosphate, 150 mM NaCl, pH 7.4)containing 1% BSA for 1 h at room temperature. The paper was thenincubated for 2 h in 6 ml of PBS-1% BSA containing 250 Mgof anti-bovine osteonectin monoclonal antibody. The blot was then washed 3times with PBS and incubated for 1 h in 10 ml of PBS-1% BSAcontaining 300 vii of peroxidase-conjugated goat antimouse IgG andwashed 3 times with PBS. The blot was developed for 10 min in amixture of 3 ml of 3 mg/ml 4-di Ioro I-nap litho I in methanol and 15ml of 50 mM Tris-HCl, pH 7.5, 200 mM NaCl, and 0.001% H2O2.

RESULTS

Growth Rate. Fig. 1 shows the growth rate of Saos-2 cells in10% fetal bovine serum. Cells seeded at 5000 cells/cm2 exhib

ited logarithmic growth for 7 days with a doubling time of 37h. Cells reached the density of 300,000 cells/cm2.

Alkaline Phosphatase. The specific activity of alkaline phosphatase was about 4 /¿mol/mgprotein/min 24 h after plating.It then dropped to 1.8 ^ mol/mg protein/min at 4 days of cultureand rose to 6.5 /¿mol/mgprotein/min at confluence (Fig. 2).

Parathyroid Hormone-sensitive Adenylate Cyclase. Fig. 3shows the dose-response curve for PTH (1-34) stimulation ofin situ adenylate cyclase measured by the conversion of endog-enously labeled [3H]ATP to [3H]cAMP. Statistically significantstimulation (about 50%) is seen at 0.3 nM hormone. Half-

Co

o'S

u

50 100 150 200 250

Time in Culture (hrs)

300

Fig. 1. Growth curve of Saos-2 cells. Saos-2 cells were plated at 5000 cells/cm2 in 35-mm 6-multiwell dishes in a medium containing 10% FBS. At the times

shown, cells were trypsinized and counted in a Coulter counter (Coulter Electronics, Hialeah, FL). Values are the means of determinations from triplicate wellsfrom 1 of 2 experiments.

o. eO>

I 5JE

I «3^

S 3"o

in5 2EO)

50 10O 150 200 250 300 350 400

Time in Culture (hrs)

Fig. 2. Time course of alkaline phosphatase activity changes in Saos-2 cells.Saos-2 cells were plated at 5000 cells/cm2 in 35-mm 6-multiwell dishes in a

medium containing 10% FBS. At the times shown, cultures were assayed foralkaline phosphatase activity as described in "Methods." values are the mean ±

SD of triplicate determinations from 1 of 2 experiments.

4962

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

OSTEOBLASTIC HUMAN OSTEOSARCOMA CELL LINE

8000 r

10"

PTH (M)Fig. 3. PTH stimulation of adenylate cyclase in Saos-2 cells. Adenylate cyclase

(cpm/106 cells/5 min) was estimated in whole cells ( 1.65 x 10*/well) as describedin "Methods." Values are the mean ±SD from triplicate wells.

Table PTH stimulation of adenylate cyclase as a function of cell density inSaos-2 cells

Adenylate cyclase activity was estimated in whole cells as described in "Materials and Methods." Cells were incubated for 5 min with PTH or vehicle (0.001 %

acetic acid and 0.001% BSA) in the presence of 0.2 HIMisobutylmethylxanthine.Results are the mean ±SD of triplicate cultures.

3H-cAMP (cpm/106 cells/5 min)

PTH (nin)0

110

1000.5

x10"3825

±1353371±3437638±5388423±3090.7

x10"1656

±202036±1743434±613624±2851.4

xIO5"842

±621260±1503534±1733605±1723.0

x10"564

±151679±244455±504644±50

* Number of cells/cm2.

15,000 r

g 10,000

5,000 -

IO

Time (min)

Fig. 4. Time course of cAMP accumulation in the presence of PTH indexamethasone-treated and untreated cells. Saos-2 cells were exposed to vehicle(0.0025% ethanol) or dexamethasone (30 nvt) for 3 days. Adenylate cyclaseactivity (cpm/106 cells/5 min) was estimated in whole cells, as described in"Methods." Control (•)and dexamethasone-treated (O) cells were incubated with

10 nM PTH for the times indicated. Values are the means of determinations fromtriplicate wells from 1 of 2 experiments.

maximal stimulation occurred around 2.2 nM. This dose-response curve of the PTH-stimulatable adenylate cyclase wasobtained at 1.65 x IO6 cells/well. Table 1 shows the respon

siveness to PTH as a function of cell number. The baselineadenylate cyclase activity decreased and the sensitivity to PTHincreased with cell density.

8o

o

&

[3H]l,25(OH)2D3(nM)

Fig. 5. Analysis of I.2SIOIlh|'II|l>< binding in Saos-2 cells. Saturation plotshowing total (•)and nonspecific (O) binding in the presence of 100-fold excessradioinert I.25(OII);1),. ln\rt. Scatchard analysis of specific binding data.1,25(OH)2D3 binding to Saos-2 cell cytosol was described in "Methods." Values

are the mean of triplicate determinations from 1 of 3 experiments.

2000

1600

o 1200

800

400

10 15 20 25 30

Fraction NumberFig. 6. Sucrose density gradient analysis of [3H]1,25(OH)2D3 binding in Saos-

2 cytosol. Cytosol was prepared as described in "Methods." Aliquots of cytosolwere preincubated with 1 nM [3H]1,25(OH)2D3 in the absence (•)or presence (O)of a 100-fold excess of radioinert 1,25(OH)2D3. Fractions of 160 fil were collectedand counted after sucrose density gradient centrifugation as described in "Methods."

Dexamethasone was shown to enhance PTH stimulation ofadenylate cyclase in osteoblastic calvaríaand ROS 17/2.8 cells(17, 18). Fig. 4 shows the effect of dexamethasone on the timecourse of PTH ( 10 nM)-dependent cAMP accumulation in Saos-2 cells. Maximal cAMP levels in dexamethasone-treated and-untreated cells were seen at about 7-10 min, but the extent ofPTH stimulation was enhanced about 2.5-fold in dexamethasone-treated cells.

1,25(OH)2D3 Receptors. 1,25(OH)2D3 receptors were quan-titated in 40,000 x g supernatants prepared from Saos-2 cellswhen the cells reached a density of 50,000 cells/cm2. Fig. 5

4963

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

OSTEOBLASTIC HUMAN OSTEOSARCOMA CELL LINE

1 2

Fig. 7. Immunocytochemical demonstration of osteonectin in Saos-2 cells.Cells exposed to antiosteonectin antibody: I. permeabilized with 0.01% NonidetP40; B, unpermeabilized. C, cells exposed to preimmune serum.

presents a binding curve of [3H]1,25(OH)2D3 as a function of

hormone concentration. Scatchard analysis of the data (Fig. 5,inset) indicates an apparent Kd of 0.21 nM and a receptorabundance of 3,300 binding sites per cell. At a higher densityof 280,000 cells/cm2, the number of sites dropped to 1,800/

cells, with no change in the apparent Ad. Fig. 6 shows thesedimentation properties of Saos-2 cell cytosol that had beenlabeled with 1.0 nM [3H]1,25(OH)2D3 in the presence and

absence of 100 nM unlabeled 1,25(OH)2D3. Receptors forl,25(OH)2Dj sediment at 3.2 S on sucrose gradients.

Immunofluorescence and Western Blot Analysis of Osteonectin. Immunocytochemical visualization of osteonectin in permeabilized cells revealed a reticular pattern. The staining intensity was highest in the perinuclear region, possibly correspond-

92.51

66.21

45*

311

21.5*

4964

Fig. 8. Western Blot analysis of osteonectin. Samples of concentrated mediawere electrophoresed and electroblotted to nitrocellulose as described in "Methods." Sample in Lane 2 was electrophoresed under reducing conditions.

ing to the Golgi apparatus (Fig. 7). This staining was not seenin nonpermeabilized cells.

Since osteonectin is a secreted protein (19), the presence ofosteonectin in cell media was examined by Western blot analysis. As seen in Fig. 8, the media contained an immunoreactiveband which migrated on SDS-polyacrylamide gel electropho-resis at Mr 40,000 (lane 1) and was retarded by |8-mercaptoeth-anol reduction (lane 2).

Histológica! Analysis of Saos-2 Cells in Millipore ChambersImplanted in Nude Mice. Eight weeks after i.p. implantation ofdiffusion chambers in nude mice, 4 of 6 chambers examinedshowed positive staining for calcified matrix with the von Kossamethod. The calcified matrix was invariably localized within 1mm of the filters (Fig. 9B). No cells or calcified matrix wasfound in the center of the chambers.

Five of 6 chambers and all 4 chambers which containedcalcified matrix stained positively for alkaline phosphataseactivity (Fig. 9A).

DISCUSSION

This study describes a human osteosarcoma cell line withosteoblastic properties. The production of mineralized matrix

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

OSTEOBLASTIC HUMAN OSTEOSARCOMA CELL LINE

Fig. 9. Photomicrographs of sections (4i/m thick) within (/) and immediately outside(/:') the Saos-2-containing chamber separatedby the filter (/•").A. sections stained for alkaline

phosphatase. Inside the chamber the sectionindicated high levels of alkaline phosphataseas indicated by staining intensity while theadherent cells outside the chamber showed little alkaline phosphatase activity. B, Sectionsstained by the von Kossa method. Saos-2 cellswithin the chamber produced matrix whichmineralized as indicated by intensely stainedblack regions. Adherent cells to the outside ofthe filter did not produce a mineralized matrixafter 8 weeks.

is the major phenotypic characteristic of osteoblastic cells.Similar to bone marrow cells (20), calvaríacells (21), andosteoblastic rat osteosarcoma cell lines 17/2 and 17/2.8 (14),the Saos-2 cells form a calcifying matrix in diffusion chambers.Light microscopy examination revealed production of a mineralized matrix typical of woven bone, similar to that producedby ROS 17/2.8 cells. Cell growth, matrix production, andmineralization were less extensive than for 17/2.8 cells, whichwere strongly positive in 6 of 6 chambers in this series. Thereare several possible reasons for this difference. Unlike the 17/2.8 cells, the Saos-2 cells do not form tumors when injected s.c.in nu/nu mice (data not shown) and may grow less well in thediffusion chambers. Some murine growth or differentiationfactors may be species specific and act better on rat than onhuman cells. Nonetheless, a majority of the implanted Saos-2cells scored positively in this assay, a strong indication of theirosteoblastic nature.

Biological mineralization has long been associated with elevated alkaline phosphatase levels (22). The Saos-2 cells reachvery high levels of alkaline phosphatase, 4-6 j/mol/mg protein/min, about 5- to 10-fold higher levels than calvaríacells (2) orosteoblastic rat osteosarcoma cells (2, 23). Similar to ROS 17/2.8 cells the alkaline phosphatase-specific activity in Saos-2cultured cells decreased after seeding at low density and increased thereafter, peaking at confluence. The density- and/orgrowth-dependent expression of this phenotypic property in the2 different cell lines may be a feature of osteoblastic maturation.Unlike in ROS 17/2.8 cells (24), dexamethasone had only avery small effect on alkaline phosphatase activity in Saos-2cells, increasing its level by about 30% after 4 days in culture(data not shown), possibly due to the very high level of alkalinephosphatase expression in these cells.

Another osteoblastic feature of Saos-2 cells is the presenceof PTH receptors coupled to adenylate cyclase. This propertyhas been observed in osteoblasts in situ and in vitro (25-27).PTH stimulation of adenylate cyclase in Saos-2 cells has beenreported by Boland et al. (6) along with vasoactive intestinalpeptide and prostaglandin E stimulation. The present studyprovides a detailed dose-response curve for PTH-stimulatableadenylate cyclase and shows that the Saos-2 cells become in

creasingly sensitive to lower doses of PTH with time in culture,resulting from larger stimulation without changes in the apparent PTH affinity (2.8 nM ±1.0, N = 6). At the end of logarithmic growth significant stimulation is seen at 0.3 nM PTH. Theincreased sensitivity to PTH is also partly due to the fall inbasal activity with time in culture (Table 1). Maximal PTHstimulation of adenylate cyclase is 6- to 10-fold, as comparedto 100-fold in ROS 17/2.8 cells, primarily a reflection of thehigh basal adenylate cyclase activity of Saos-2 cells. PTH sen

sitivity was further enhanced by dexamethasone treatment, a

feature shared with the ROS 17/2.8 rat osteosarcoma cells ( 18).The Saos-2 cells also possess 1,25(OH)2D3 receptors, which

are found in fetal bone cells (28), mouse bone cells (29), osteosarcoma cells (30, 31), and many other cells. The 1,25(OH)2D3Ka (0.21 nM) was comparable to that reported in rat osteosarcoma and bone cells, and the number of sites per cell was almostidentical to that found in the human osteosarcoma cell lineMG-63 (32). A sedimentation coefficient of 3.2 S for1,2S(OH)2D3 receptors is similar to that reported in bone cells(4, 33) and other cells. Receptor abundance decreased with celldensity. There were 3,300 sites/cell when the cells were harvested at 50,000/cm2 and 1,800/cell when the cells were harvested at 240,000 and 280,000/cm2, with no change in KA.

These results are very similar to those observed in mouse bonecells (33) and MG-63 cells (32). The physiological role of1,25(OH)2D3 in osteoblastic cells remains to be determined.One of the effects of this hormone, documented both in vivoand in vitro, is the enhanced production of osteocalcin (34, 35).This protein may be bone specific, since only bone-derived cellshave been shown to synthesize it. Using immunocytochemistryand radioimmunoassay of l,25(OH)2D3-treated cells, we foundno positive reaction for osteocalcin in Saos-2 cells in culture. Ifthis observation is upheld and Saos-2 cells also fail to makeosteocalcin in vivo when they produce mineralized matrix,osteocalcin may not be a requisite marker for osteoblastic cells.

Another bone-abundant protein, osteonectin, is present inSaos-2 cells. Osteonectin was visualized by immunocytochemistry and was also shown to be secreted into the medium byWestern blot analysis. Osteonectin was first isolated from bonematrix and was shown to be produced by fetal bovine (36),human (37), and porcine (19) bone cells. The SDS-polyacryl-amide gel electrophoresis migration of osteonectin ¡mmunopre-cipitated from the medium of human or porcine cells wasaround M, 40,000-45,000, similar to that identified by theWestern blot of Saos-2 cell media.

Among the human osteosarcoma cell lines reported so far(38-40) none has been examined for all of the propertiesreported here. MG-63 (32) and Saos-1 cells (41) have very lowlevels of alkaline phosphatase activity (2.5 nmol/min/mg protein) compared to Saos-2 cells (4-6 ¿/mol/min/mg protein). InSaos-1 cells, alkaline phosphatase levels did not vary with celldensity; however, 1,25(OH)2D3 and hydrocortisone increasedalkaline phosphatase activity 3- to 5-fold in these cells. Evenafter stimulation, alkaline phosphatase levels were less than 1%of these in Saos-2. MG-63 cells have PTH-unresponsive adenylate cyclase (our unpublished observation). As pointed out,the Saos-2 cells share many properties with the rat osteosarcoma ROS 17/2.8 cell line. Since the mechanism for malignanttransformation in those 2 tumors from different species is likelyto be different, their common properties are probably related

4965

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

OSTEOBLASTIC HUMAN OSTEOSARCOMA CELL LINE

to their bone origin. This cell line could thus offer a usefulexperimental model for studying osteoblastic properties andosteoblast-produced molecules in an established human cellline.

ACKNOWLEDGMENTS

We thank Dr. Harry Harris for providing the SaoS-2 cells and forstimulating discussions on alkaline phosphatase expression in thesecells. Dr. Warren Nichols for the chromosomal analysis, and Dr.Bertram Sacktor for media from human bone cells.

REFERENCES

1. Sato, G. Functionally differentiated cell lines. New York: Alan R. Liss, 1981.2. Partridge, N. C, Alcorn, D., Michelangeli. V. P., Ryan, G., and Martin, T.

J. Morphological and biochemical characterization of four clonal osteogenicsarcoma cell lines of rat origin. Cancer Res., 43:4308-4314, 1983.

3. Rodan. G. A., and Rodan, S. B. Expression of the osteoblastic phenotype In:W. A. Peck (ed.), Advances in Bone and Mineral Research, Annual II, pp.244-285. Amsterdam: Excerpta Medica, 1984.

4. Chen, T. L., Cone, C. M., Morey-Holton, E., and Feldman, D. 1,25-Dihy-droxyvitamin I>,receptors in cultured rat osteoblast-like cells. J. Biol. Chem.,25«:4350-4355, 1983.

5. Shupnik, M. A., and Tashjian, A. H., Jr. Epidermal growth factor andphorbol ester actions on human osteosarcoma cells. J. Biol. Chem., 257:12161-12164, 1982.

6. Boland, C. J., Fried, R. M., and Tashjian. A. H., Jr. Measurement of cytosolicfree Ca2* concentrations in human and rat osteosarcoma cells: actions ofbone resorption-stimulating hormones. Endocrinology, //*: 980-989, 1986.

7. Fogh, J., and Trempe, G. New human tumor cell lines: In: J. Fogh (ed.),Human Tumor Cell Lines in vitro, pp. 115-159, New York: Plenum Press,1975.

8. Rodan. S. B., Insogna. K. L., Vignery, A. M. -C., Stewart, A. F., Broadus,A. E., D'Souza, S. M., Bertolini, D. R., Mundy, G. R., and Rodan, G. A.

Factors associated with humoral hypercalcemia of malignancy stimulateadenylale cyclase in osteoblastic cells. J. Clin. Invest., 72:1511-1515, 1983.

9. Salomon, Y., Londos, C., and Rodbell, M. A highly sensitive adenylatecyclase assay. Anal. Biochem., SS: 541-548. 1974.

10. Majeska, R. J., and Rodan, G. A. Alkaline phosphatase inhibition by parathyroid hormone and isoproterenol in a clonal rat osteosarcoma cell line:possible mediation by cAMP. Calcif. Tissue Int., 34: 59-66, 1982.

11. Spector, T. Refinement of the Coomassie blue method of protein quantita-tion. Anal. Biochem., 86: 142-146, 1978.

12. Wecksler, W. R., and Norman, A. W. An hydroxylapatite batch assay for thequantitation of la,25-dihydroxyvitamin D3 receptor complexes. Anal.Biochem., 92: 314-323, 1979.

13. Brumbaugh, P. F., Hughes. M. R., and Haussler, M. R. Cytoplasmic andnuclear binding components for ln,25-dihydroxyvitamin D3 in chick parathyroid glands. Proc. Nati. Acad. Sei. U. S. A., 72:4871-4875, 1975.

14. Shteyer, A., Gazit, D., Passi-Even, L., Bab, I., Majeska, R., Gronowicz, G.,Lurie, A., and Rodan, G. A. Formation of calcifying matrix by osteosarcomacells in diffusion chambers HI vivo. Calcif. Tissue Int., 39: 59-54, 1986.

15. Laemmli, U. K. Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature (Lond.), 227:680-685, 1970.

16. Towbin, H., Staehelin, T., and Gordon, J. Electrophoretic transfer of proteinsfrom polyethylene glycol to nitrocellulose sheets: procedure and some applications. Proc. Nati. Acad. Sei. U. S. A. 76: 4350-4354, 1979.

17. Chen, T. L., and Feldman, D. Glucocorticoid potentiation of the adenosine3', 5'-monophosphate response to parathyroid hormone in cultured rat bonecells. Endocrinology, 102: 589-596, 1978.

18. Rodan, S. B., Fischer, M. K., Egan, J. J., Epstein, P. M., and Rodan, G. A.The effect of dexamethasone on parathyroid hormone stimulation of adenyl

ate cyclase in ROS 17/2.8 cells. Endocrinology, US: 951-958, 1984.19. Otsuka, K., Yao, K.-L., Wasi, S., Tung, P. S., Aubin, J.-E., Sodek, J., and

Termine, J. D. Biosynthesis of osteonectin by fetal porcine calvaría!cells invitro. J. Biol. Chem., 259: 9805-9812, 1984.

20. Ashton, B. A., Allen, T. D., Howlett, C. R., Eaglesom, C. C., Hattori, A.,and Owen, M. E. Formation of bone and cartilage by marrow stromal cellsin diffusion chambers in vivo. Clin. Orthop. Relat. Res., 151: 294-307,1980.

21. Simmons, D. J., Kent, G. N., Jilka, R. L., Scott, D. M., Fallón, M., andCohn, D. V. Formation of bone by isolated cultured osteoblasts in milliporediffusion chambers. Calcif. Tissue Int., 34: 291-294, 1982.

22. Parlili. A. M. The actions of parathyroid hormone on bone; relation to boneremodeling and turnover, calcium homeostasis, and metabolic bone disease.II. PTH and bone cells: bone turnover and plasma calcium regulation.Metabolism, 25:909-955, 1976.

23. Majeska, R. J., Rodan, S. B., and Rodan, G. A. Parathyroid hormone-responsive clonal cell lines from rat osteosarcoma. Endocrinology, ¡07:1494-1503, 1980.

24. Majeska, R. J., Nair, B. C., and Rodan. G. A. Glucocorticoid regulation ofalkaline phosphatase in the osteoblastic osteosarcoma cell line ROS I7/2.8.Endocrinology, 116: 170-179, 1985.

25. Silve, C. M., Hradek, G. T., Jones, A. L., and Arnaud, C. D. Parathyroidhormone receptor in intact embryonic chicken bone: characterization andcellular localization. J. Cell Biol., 94: 379-386, 1982.

26. Luben, R. A., Wong, G. L., and Cohn, D. V. Biochemical characterizationwith parathormone and calcitonin of isolated bone cells: provisional identification of osteoclasts and osteoblasts. Endocrinology, 99: 526-534, 1976.

27. Peck, W. A., Burks, J. K., Wilkins, J., Rodan, S. B., and Rodan, G. A.Evidence for preferential effects of parathyroid hormone, calcitonin andadenosine on bone and periosteum. Endocrinology, 100: 1357-1364,1977.

28. Kream, B. E., Jose, M., Yamada, S., and Deluca, H. F. A specific highaffinity macromolecule for 1.:>«H11.1>, in bone fetal bone. Science (Wash.DC), 197: 1086-1088, 1977.

29. Chen, T. L., Hirst, M. A., and Feldman, D. A receptor-like binding macro-molecule for 1a,25-dihydroxycholecalciferol in cultured mouse bone cells. J.Biol. Chem., 254: 7491-7494, 1979.

30. Manolagas, S. C., Haussler, M. R., and Deftos, L. J. 1,25-DihydroxyvitaminI>i receptor-like macromolecule in rat osteogenic sarcoma cell lines. J. Biol.Chem., 255:4414-4417, 1980.

31. Partridge, N. C., l rampimi. R. J., Eisman, J. A., Michelangeli, V. P., Elms,E., Bradley, T. R., and Martin, T. J. Receptors for l,2S(OH)vitamin D3enriched in cloned osteoblast-like rat osteogenic sarcoma cells. FEBS Lett.115: 139-142, 1980.

32. Franchesi, R. T.. James, W. M., and Zerlauth, G. l o,25-DihydroxyvitaminI), specific regulation of growth, morphology and fibronectin in a humanosteosarcoma cell line. J. Cell. Physiol., 123:401-409, 1985.

33. Chen, T. L., and Feldman, D. Regulation of 1,25 dihydroxyvitamin D3receptors in cultured mouse bone cells. J. Biol. Chem., 256: 5561-5566,1981.

34. Price, P., and Baukol, S. A. 1,25-Dihydroxyvitamin I), increases serum levelsof the vitamin K-dependent bone protein. Biochem. Biophys. Res. Commun.,99:928-935, 1981.

35. Price, P. A., and Baukol, S. A. 1,25-Dihydroxyvitamin D3 increases synthesisof the vitamin K-dependent bone protein by osteosarcoma cells. J. Biol.Chem., 255: 11660-11663, 1980.

36. Whitson, S. W., Harrison, W., Dunlap, M. K. Bowers D. E., Jr., Fisher, L.W., Robey, P. G., and Termine, J. D. Fetal bovine bone cells synthesizebone-specific matrix proteins. J. Cell Biol., 99:607-614, 1984.

37. Robey, P. G., and Termine, J. D. Human bone cells in vitro. Calcif. TissueInt., 3 7: 453-460, 1985.

38. Heremans, H., Billiau, A., Cassiman, J. J., Mulier. J. D., and De Somer, P.In vitro cultivation of human tumor tissues II. Morphological and virologicalcharacterization of three cell lines. Oncology, 35: 246-252, 1978.

39. Tsang, K. Y., Pai, G. S., and Fidenberg, H. H. Characterization of newlyestablished human osteosarcoma cell line, IMI. In Vitro (Rockville), 17:308-314, 1981.

40. Yamane, T. Establishment and characterization of cell lines derived from ahuman osteosarcoma. Clin. Orthop. Relat. Res., /99: 261-271, 1985.

41. Mulkins, M. A., Manolagas, S. L., Deftos, L. J., and Sussman, H. H. 1,25-Dihydroxyvitamin I), increases bone alkaline phosphatase isoenzyme levelsin human osteogenic sarcoma cells. J. Biol. Chem., 25«:6219-6225, 1983.

4966

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from

1987;47:4961-4966. Cancer Res   Sevgi B. Rodan, Yasuo Imai, Mark A. Thiede, et al.   with Osteoblastic PropertiesCharacterization of a Human Osteosarcoma Cell Line (Saos-2)

  Updated version

  http://cancerres.aacrjournals.org/content/47/18/4961

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/47/18/4961To request permission to re-use all or part of this article, use this link

Research. on December 11, 2020. © 1987 American Association for Cancercancerres.aacrjournals.org Downloaded from