[CANCER RESEARCH 50, 6902-6907. November 1, 1990]

Human Prostatic Cancer Cells, PC3, Elaborate Mitogenic Activity WhichSelectively Stimulates Human Bone Cells1

Victor S. Perkel,2 Subburaman Mohan, Sandra J. Herring, David J. Baylink, and Thomas A. Linkhart

Departments of Medicine [V. S. P., S. M., S. J. H., D. J. B.J, Biochemistry [S. M., D. J. B..T.A. LJ, Physiology fS. M.], and Pediatria. [T. A. L.], Loma Linda Universityand Jerry L. Pettis Veterans Administration Hospital, Loma Linda, California 92357

ABSTRACT

Prostatic cancer typically produces osteoblastic métastaseswhich arenot attended by marrow fÃ-brosis(i.e., osteoblast but not stromal flbroblastproliferation). In the present study we sought to test the hypothesis thatprostatic cancer cells produce factor(s) which act selectively on humanosteoblasts. Such a paracrine mechanism would explain the observedincrease in osteoblasts, unaccompanied by an increase in marrow fibro-blasts. To test this hypothesis we investigated the mitogenic activityreleased by the human prostatic tumor cell line, PC3. PC3 cells havebeen reported previously to produce mitogenic activity for cells that wasrelatively specific for rat osteoblasts compared to rat fibroblasts. However, the effects of this activity on human cells has not been examinedpreviously. PCS-conditioned medium (CM) (5-50 Mg CM protein/ml)stimulated human osteoblast proliferation by 200-950% yet did notstimulate human fibroblast proliferation (| 'I I|tli>midine incorporation).

PC3 CM also increased cell numbers in human osteoblast but notfibroblast cell cultures. To determine whether the osteoblast-specificmitogenic activity could be attributed to known bone growth factors,specific assays for these growth factors were performed. PC3 CM contained 10 pg insulin-like growth factor (IGF) I,

PROSTATIC HUMAN BONE CELL SPECIFIC MITOGEN

MM). ['HJThymidine (50 Ci/mol) was purchased from ICN (Irvine,

CA).

Collection and Preparation of PC3 Cell CM

The PC3 cell line (American Type Culture Collection) was maintained in a 1:1 mixture of DMEM and Ham's F-12 media with 10%

calf serum in 10-cm dishes. For collection of CM, cells were plated in850 cnr roller bottles, grown to confluence, and then rinsed once withPBS, pH 7.4, and changed to serum-free medium (DMEM/F-12). Thefirst 24-h CM was discarded and replaced with fresh medium. Subsequent CM was collected and replaced every day. The CM was centri-fuged at 10,000 x g for 20 min to remove cell debris and frozen at-20°Cuntil used. Several batches of CM were pooled, concentrated inan Amicon stirred cell with a YM-5 filter (nominal molecular weightcutoff of 5,000). Concentrated CM proteins were dialyzed against PBSbefore testing for effects on cell proliferation.

Gel Filtration on Sephadex G-75

PC3 CM was dialyzed against l M acetic acid and applied to aSephadex G-75 gel filtration column (1 x 100 cm) in the same solvent.Fractions (2 ml) were collected and assayed for mitogenic activity inhuman osteoblasts and for IGF I and II (as described below). To reducethe number of tubes for assay, fractions were pooled.

Heparin Sepharose Chromatograph)

PC3 CM was dialyzed against 10 mMTris acetate, pH 7.4, subjectedto FPLC with a heparin Sepharose affinity column, and eluted with alinear eluting gradient of 0-3 M NaCl in 10 mM Tris acetate (pH 7.4).Fractions (2 ml) were assayed for osteoblast mitogenic activity usingMC3T3-E1 cells.

Ion-Exchange Chromatograph) on Mono Q and Mono S

PC3 CM was dialyzed against 10 m.\i Tris acetate, pH 7.4, containing100 mM NaCl, and then fractionated either on a Mono Q anion-exchange or a Mono S cation-exchange chromatography column usinga linear eluting gradient of 100 mvi to l M NaCl. Fractions (2 ml) wereassayed for osteoblast mitogenic activity using MC3T3-E1 cells.

Assays

|'H|Thymidine Incorporation Assay. Human osteoblast cells wereisolated from femoral head bone samples (24-26) or from mandible (agenerous gift of Dr. C. Kasperk, Heidelberg, West Germany) andhuman skin fibroblast cultures were established from foreskin specimens or adult skin biopsies. The cells were maintained for up to 10population doublings in DMEM + 10% calf serum and plated at 5000cells/well in 5% calf serum (0.5 ml/well). After 24 h, cultures werechanged to serum-free medium (DMEM) and test agents were added 2h later. After 18 h, cells were labeled with 0.75 /jCi ['Hjthymidine/well.

Labeling was terminated 6 h later by rinsing the cultures with PBS.The plates were frozen at —20°Cand used for determination of trichlo-

roacetic acid-insoluble radioactivity (21). For some experimentsMC3T3-E1 mouse osteoblasts were used to determine effects of PC3CM on DNA synthesis as described before (27). These cells maintainmany osteoblast characteristics (28, 29). The effect of reducing agentsor low pH on mitogenic activity of PC3 cell CM was tested as describedpreviously (22).

Growth Factor Assays. IGF I was determined by an RIA (23, 30).IGF II was quantitated by a radioreceptor assay using competitivedisplacement of '25I-labeled human IGF II binding to monolayer cultures of H-35 rat hepatoma cells (23, 30). TGF0 activity was determinedby its ability to inhibit ['Hjthymidine incorporation into mink lung

epithelial cells (American Type Culture Collection) (19). bFGF wasassayed by an RIA using polyclonal rabbit antiserum to a (1-24) bFGFfragment (Andrew Baird, Salk Institute) (31).

Protein Determination. All protein concentrations were determinedusing the Bradford dye-binding assay (32), with bovine -y-globulin as a

standard.

Statistical Analyses. All statistical comparisons were made usingStudent's paired t test analysis. A P value

PROSTATIC HUMAN BONE CELL SPECIFIC MITOGEN

by osteoblasts, they are present in bone matrix, and they stimulate osteoblast proliferation). To determine whether theseknown osteoblast mitogens were released by PC3 cells, serum-free PC3 CM was initially assayed for their presence, usingspecific assays. Concentrated CM was found to contain 10 pgIGF I,

PROSTATIC HUMAN BONE CELL SPECIFIC M1TOGEN

El mouse osteoblast cell line due to limited availability ofhuman osteoblasts. DNA synthesis in these mouse cells wasstimulated over the same concentration range of PC3 CMprotein that stimulated human osteoblasts. Because sensitivityof the known bone cell growth factors to reducing agents andto acid conditions has been well documented (22), we tested theeffects of reducing agents and low pH on PC3 CM. Thestimulation of ['Hjthymidine incorporation into MC3T3-E1

cells by PC3 CM was not inhibited by pretreatment for 2 hwith 0.1 M HC1 and was inhibited only 27% by 2 m\i DTT(Table 4). In contrast, pretreatment of 1% PCS with 2 miviDTT for 2 h induced a marked inhibition of [3H]thymidine

incorporation in these cells. The only other bone mitogenknown to be resistant to both reducingagentsand lowpH isaFGF.

The ability to bind heparin is a known characteristic of certainbone growth factors [i.e., aFGF and bFGF bind firmly toheparin; PDGF, the IGFs, and epidermal growth factor bindweakly; while TGF/3 does not appear to bind (23, 33)]. Themitogenic activity present in PC3 CM when tested withMC3T3-E1 cells did not bind to heparin (data not shown). Incontrast, IGF I was detected by specific RIA, both in the weaklybound fractions that eluted with 0.5-1 M NaCl and tightlybound fractions that eluted with 2.5-3.0 M NaCl. When 1 /¿gbovine bFGF in gelatin was chromatographed on heparin >95%of the mitogenic activity eluted in the tightly bound fraction.

PC3 CM was chromatographed on anión- (Mono Q) andcation-exchange (Mono S) FPLC columns. On Mono Q aniónexchange, all the mitogenic activity was found in the unboundfractions (data not shown). On Mono S cation exchange, themajority of the mitogenic activity bound to Mono S and elutedwith 200-400 HIMNaCl (Fig. 4). There was no immunoreactive

.2800-"oo&C>

600-82

.EC^o400-^

200-x"

o1 ' n -ÊiliÈìj_iw\

,, J.,,

A B C D E FPooled G75 column fractions

Fig. 3. The gel filtration profile of PC3 CM mitogen on a Sephadex G-75column. Concentrated PC3 cell CM was dialyzed against 1 M acetic acid, appliedto a Sephadex G-75 column, and eluted with the same buffer. Pooled fractionswere diluted 1:10 with PBS and tested with normal human femoral osteoblasts.The molecular weight range of the pooled samples was as follows: A, 38,000-45,000; B, 26,000-30,000; C, 14.000-22.000; D, 8.000-12,500; E, 4.500-7.500;F, 1,750-3,100. Columns, % of unstimulated control [3H]thymidine incorporation(6-h pulse initiated 18 h after treating the cells with pooled fractions); bars, SD.Maximal stimulation by 1% FCS in the same experiment was 420%.

Table 4 Effect of the reducing agent 2 mM dithiothreilol on the stimulationoff'HJthymidine incorporation in MC3T3-E1 cells by PC3 CM and 1%fetal

calf serumThe effect of reducing agents or low pH on mitogenic activity of PC3 cell CM

and 1% FCS was tested as described previously (21). Results are expressed as %of unstimulated control [3H]thymidine incorporation (control. 323 cpm) ±SD(6-h pulse initiated 18 h after adding the factors at the concentrations indicated).

Control1% fetal calf serumPC3 CM (5 Kg/ml)No

agents100

±353163 ±2231297± 119+DTT133

±371432 ±333°950 ±365*°

P< 0.001 vi. no agents.*/>< 0.01 >'s.no agents.

600-,Cog8-^cT

400O

i•-

°300•

—"o-à i 20°'>,£

100-

fO;

• •PC3CM1., — opticaldensityi

i1I!!

Ai i/\i/\i •\i! /10008000600

^^>400

E200

!n10 20 40 50 60

Fraction number

Fig. 4. Elution profile of the PC3 CM mitogen on a FPLC Mono S column.Concentrated PC3 CM was subjected to Mono S cation-exchange chromatogra-phy using a linear gradient of NaCl from 100 ITIMto l M (dotted line, withoutsymbols). Alternate fractions were tested in MC3T3 El mouse osteoblastic cells.Results are expressed as % of unstimulated control [3H]thymidine incorporation

(1400 cpm). Cells were treated for 20 h with unconcentrated alternate fractions(0.025 ml/well) and pulse labeled for 3 h. Maximal stimulation by 1% fetal calfserum in the same experiment was 1300% of control. IGF I determined by aspecific RIA (see "Materials and Methods") was detected in fractions 2-4. PurifiedbFGF mitogenic activity (run separately) eluted mainly in fractions 39-51. Theabsorbance of each fraction was read at 280 nm.

IGF I in the active mitogenic fractions. Purified bovine bFGFwhen chromatographed on Mono S, using the same NaCl-eluting gradient, exhibited stronger binding characteristics thanthe PC3 mitogen (i.e., bFGF mitogenic activity required concentrations >600 mM NaCl for elution).

DISCUSSION

These studies were initiated to verify the ability of PC3 CMto specifically stimulate ['Hjthymidine incorporation (as a

marker of DNA synthesis), in human bone cells without concomitant human fibroblast stimulation. Koutsilieris et al. (15-18) had previously shown that benign prostatic hypertrophiecells, malignant prostatic adenocarcinomas, and PC3 cells released a growth factor with higher specificity for rat osteoblaststhan fibroblasts. In contrast to the M, 26,000-30,000 humanosteoblast-specific mitogen described in the present study,Koutsilieris et al. described an M, —10,000 rat osteoblast-specific and an M, —13,000nonspecific rat osteoblast mitogen.Simpson et al. (6) found that PC3 cells synthesize a mRNA,—1,800 bases in length, which codes for an M, -20,000 ratosteoblast-stimulating protein. This would suggest that themitogen characterized by Koutsilieris et al. might be an activefragment of a larger molecular weight mitogen or the highermolecular weight activity we observed was the result of aggregation of smaller proteins.

The presence of other bone mitogens in PC3 CM mightsuggest a potential pathophysiological mitogenic role for thesegrowth factors. However, on close examination of the propertiesof these other mitogens (Table 5), it is unlikely that any ofthem are mediators of the observed tissue-specific PC3 CM-induced mitogenic effects in human osteoblasts and fibroblasts(22, 33-40). That these mitogens are present in PC3 CM, yetdo not exert their full mitogenic effect, suggests that there couldbe inhibitors to these mitogens in the CM. In support of this,we have recently isolated and purified an M, 25,000 IGF-binding protein from PC3 CM that inhibits IGF mitogenicactivity (14). Although bFGF was present in PC3 CM inapparently high enough concentrations to stimulate fibroblasts,bFGF was assayed in an FGF RIA using an antibody directedagainst a synthetic NH2-terminal fragment. This assay may

6905

on June 16, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PROSTATIC HUMAN BONE CELL SPECIFIC MITOGEN

Table 5 Properties of crude PC3 CM-derived mitogen compared to known hone cell growth factors

Data for this table was compiled from experiments performed in the Mineral Metabolism Laboratory at the Jerry L. Pettis Veterans Hospital and from Refs. 22,and 33-40.

PropertyM,

(thousands)HeparinbindingCation-exchangebindingAnion-exchangebinding2

mMDTTpH2.0Action

on humanosteoblastsActionon humanfibroblastsActionon MC3T3 cellsPC326-30—++———TÎΗTÎÎIGFI7.6+++iÎÎÎÎÎIGF

II7.5+++1ÎÎÎÎÎÎbFGF18-20++++++——iÎÎÎÎÎÎÎÎÎaFGF15-16+++—+++——ÎÎTHTîiiTGF/j25—++NDÕ—mîmPDGF28-35+++N

Di—ÎTÎÎîEGF/TGF«6+NONOiîÎÎND

1ND. not determined.

have overestimated the amount of bFGF present. The amountassayed could have been greater than the amount of biologicallyactive FGF, that was present.

Although growth factor tissue specificity is unusual sincemost growth factors act on many different tissues, there are afew precedents for tissue specificity. A recently described kera-tinocyte growth factor, structurally related to the FGF familyof growth factors, demonstrates specificity for epithelial cells(41). A recently described breast cell inhibitor, mammastatin,acts specifically on mammary cells (42). Nerve growth factor isrelatively specific for nerve cells (43). However, an alternativepossibility that the observed prostatic cancer cell-induced boneformation could be a result of the combined action of two ormore known growth factors cannot be overlooked. We havetested combinations of known growth factors in an attempt toobtain a tissue-specific effect (data not shown), but our resultsdo not make this possibility compelling. It is also theoreticallypossible that prostatic tumor cells, rather than producing agrowth factor, could produce a regulatory factor that stimulatesosteoblasts (or other cells in bone) to produce abnormally highamounts of one or more of the constitutively produced growthfactors. Such a regulatory factor might appear to have mitogenicactivity when added to osteoblast target cells.

The PC3 cell line was obtained from a 62-year-old man withprostatic carcinoma metastatic to bone and has been characterized as being representative of prostatic carcinoma (44, 45).Although primary samples of prostatic carcinoma have not asyet been evaluated by us for human osteoblast-specific mitogenic activity, we have preliminary evidence that another prostatic cell line LnCaP (46) may also produce similar activity.This information, taken together with the data of Koutsilieriset al. (15-17), suggesting that human primary prostatic carcinoma cells and benign prostatic hypertrophie tissue contain anosteoblast-specific mitogen (for rat cells), would implicate alocal paracrine, rather than an endocrine, effect of a mitogennormally present in prostate.

In summary, our results suggest that the PC3 cell line, whichis an in vitro model for prostatic carcinoma metastatic to bone,produces a human osteoblast growth factor, with characteristicswhich distinguish it from known bone mitogens. This growthfactor is an M, 26,000-30,000 basic protein which is resistantto low pH and reducing agents. This mitogen is capable ofstimulating human osteoblast proliferation in the absence ofhuman fibroblast stimulation. The PC3 mitogen seems to stimulate human osteoblast proliferation via a mechanism independent of other bone mitogens and may play an etiologicalrole in the osteoblastic métastasesseen in prostatic carcinoma.

ACKNOWLEDGMENTS

We are grateful to Drs. J. C. Jennings and K. H. W. Lau for usefuldiscussions and for the expert technical assistance of Maria Keffer andDaniel MacCharles.

REFERENCES

1. Hüben.R. P., and Murphy. G. P. Prostate cancer: an update. CA-C'ancer J.Clin., 36: 274-292. 1986.

2. Paterson. A. H. G. Bone métastasesin breast cancer, prostate cancer, andmyeloma. Bone (NY), «(Suppl. l):S17-22, 1987.

3. Jacobs, S. C. Spread of cancer to bone. Urology, 21: 337-344, 1983.4. Burki, F., Coindre. J-M., and Mauriac, L. Sclerotic bone métastasesof

prostatic origin and osteomalacia. Importance of a histomorphometric study.Prog. Clin. Biol. Res., 243A: 569-571, 1987.

5. Willis. R. A. Secondary tumors of bone. In: R. A. Willis (ed.). The Spread ofTumors in the Human Body, Ed. 3. pp. 229-250. London: Butterworth andCo., 1973.

6. Simpson, E., Harrod, J., Eilon, G.. Jacobs. J. W., and Mundy, G. R.Identification of a messenger ribonucleic acid fraction in human prostaticcancer cells coding for a novel osteoblast stimulating factor. Endocrinology117: 1615-1620. 1985.

7. Jacobs, S. C.. Pikna, D., and Lawson. K. L. Prostatic osteoblast factor.Invest. Urol., 17: 195-198, 1979.

8. Jacobs. S. C., Story, M. T., Sasse, J.. and Lawson. R. K. Characterization ofgrowth factors derived from rat ventral prostate. J. Urol.. 139: 1106-1110.1988.

9. Lawson. R. K. Growth factors and benign prostatic hyperplasia. World J.Urol.. 6: 190-193. 1989.

10. Story. M. T., Sasse. J., Jacobs, S. C., and Lawson. R. K. Prostatic growthfactor: purification and structural relationship to basic fibroblast growthfactor. Biochemistry, 26: 3843-3849. 1987.

11. Hierowski, M. T.. McDonald, M. W., Dunn. L.. and Sullivan. J. W. Thepartial dependency of human prostatic growth factor on steroid hormones instimulating thymidine incorporation into DNA. J. Urol., 138: 909-912.1987.

12. Ikeda. T., Liouban, M. N., and Marquart, H. Human transforming growthfactor type t!2°-production by a prostatic adenocarcinoma cell line, purificationand initial characterization. Biochemistry, 26: 2406-2410, 1987.

13. Sitaras. N. M.. Sariban. S., Bravo. M.. Pantazis, P., and Antoniades, H. N.Constitutive production of platelet derived growth factor-like proteins byhuman prostatic carcinoma cell lines. Cancer Res., 48: 1930-1935. 1988.

14. Perkel. V. S., Mohan. S.. Baylink. D. J.. and Linkhart. T. A. An inhibitoryinsulin-like growth factor binding protein (ln-IGFBP) from human prostaticcell conditioned medium reveals N-terminal sequence identity with bone-derived In-IGFBP. J. Clin. Endocrinol. Metab.. 71: 533-535. 1990.

15. Koutsilieris, M.. Rabbani, S. A., and Goltzman, D. Effects of human prostaticmitogens on rat bone cells and fibroblasts. Endocrinology, 115: 447-454,1986.

16. Koutsilieris. M.. Rabbani. S. A.. Bennet. H. P., and Goltzman, D. Characteristics of prostate derived growth factors for cells of the osteoblasticphcnotype. J. Clin. Invest.. 80: 941-946. 1987.

17. Koutsilieris. M., Rabbani. S. A., and Goltzman. D. Selective osteoblastmitogens can be extracted from prostatic tissue. Prostate, 9: 109-115, 1986.

18. Rabbani, S. A., Desjardins, J., and Goltzman, D. A human prostatic cancercell line releases mitogenic activity selective for cells of the osteoblast phe-notype. Annual Meeting of the American Cancer Society, New York City,NY, Nov. 8-12, 1988. Abstract 955, 1988.

19. Jennings. J. C.. Mohan, S., Linkhart, T. A., Widstrom, R.. and Baylink. D.J. Comparison of the biological activities of TGFrf, and TGF/i2: differentialactivity in endothelial cells. J. Cell. Physiol., 137: 167-172. 1988.

20. Wergedal. J. E.. Mohan, S., Lundy, M., and Baylink. D. J. Skeletal growthfactor and other growth factors known to be present in bone matrix stimulateproliferation and protein synthesis in human bone cells. J. Bone Miner. Res.,J: 179-186, 1990.

21. Mohan, S.. Jennings, J. C.. Linkhart. T. A., and Baylink, D. J. Isolation andpurification of a low-molecular-weight skeletal growth factor from humanbones. Biochim. Biophys. Acta. 884: 234-242, 1986.

22. Mohan, S., Linkhart. T. A., Jennings. J. C.. and Baylink, D. J. Chemicaland biological characterization of a low-molecular-weight skeletal growthfactor from human bones. Biochim. Biophys. Acta, 884: 243-250. 1986.

23. Mohan, S., Jennings, J. C., Linkhart, T. A., and Baylink, D. J. Primarystructure of human skeletal growth factor: homology with human insulin like

6906

on June 16, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PROSTATIC HUMAN BONE CELL SPECIFIC MITOGEN

growth factor II. Biochim. Biophys. Acta. 996: 44-55. 1988.24. Wergedal. J. E.. and II.i\ link. D. J. Characterization of cells isolated and

cultured from human bone. Proc. Soc. Exp. Biol. Med.. 176: 27-31. 1984.25. Wergedal. J. E.. Mohan. S.. Taylor. A. K.. and Baylink. D. J. Human skeletal

growth factor is produced by human osteoblast-like cells in culture. Biochim.Biophys. Acta. 889: 163-170, 1986.

26. Mohan. S.. Wergedal, J. E., Taylor, A. K.. and Baylink. D. J. IGT I and IIare produced by human osteoblast cells in culture. FASEB J.. 2: A842. 1988.

27. Linkhart. S.. Mohan. S., Linkhart. T.. Kumegawa, M., and Baylink, D. J.Human skeletal growth factor stimulates collagen synthesis and inhibitsproliferation in a clonal osteoblast cell line (MC3T3-E1). J. Cell. Physiol.,128: 307-312, 1986.

28. Sudo, H., Kodama. H., Amagai, Y., Yamamoto, S., and Kasai. S. In vitrodifferentiation and calcification in a new clonal osteogenic cell line derivedfrom newborn mouse calvarÃ-a.J. Cell Biol.. 96: 191-198. 1983.

29. Kurihara. N., Ishizuka. S., Kiyoki. M., Haketa, V., Ikeda. K.. and Kumegawa.M. Effects of 1.25-dihydroxyvitamin D, on osteoblastic MC3T3-E1 cells.Endocrinology. 118: 940-947. 1986.

30. Mohan, S.. Bautista, C. M.. Herring. S. J.. Linkhart. T. A., and Baylink. D.J. Development of valid methods to measure insulin-like growth factors 1and II in bone cell conditioned medium. Endocrinology, 126: 2534-2542.1990.

31. Baird. A.. Mormede. P., and Bohlen. P. Immunoreactive fibroblast growthfactor in cells of peritoneal exúdate suggests its identity with macrophage-derived growth factor. Biochem. Biophys. Res. Commun.. 128: 358-364.1985.

32. Bradford, M. M. A rapid and sensitive method for the quantitation ofmicrogram amounts of protein utilizing the principal of protein-dye binding.Anal. Biochem.. 72:248-255. 1976.

33. Hauschka, P. V., Mavrakos, A. E., lafrati. M. D., Doleman, S. E., andKlagsbrun, M. Growth factors in bone matrix. Isolation of multiple types byaffinity chromatography on heparin Sepharose. J. Biol. Chem.. 261: 12665-12674, 1986.

34. Frolik. C. A.. Dart, L. L.. Meyers, C. A.. Smith. D. M., and Sporn, M. D.Purification and initial characterization of a type rf transforming growth

factor from human placenta. Proc. Nati. Acad. Sci. USA. 80: 3676-3680,1983.

35. Taylor. J. M.. Cohen, S.. and William. M. M. Epidermal growth factor: highand low molecular weight forms. Proc. Nati. Acad. Sci. USA, 67: 164-171,1970.

36. Taylor, J. M.. William. M. M.. and Cohen. S. Epidermal growth factor:physical and chemical properties. J. Biol. Chem.. 247: 5928-5934, 1972.

37. Thomas. K. A.. Rios-Candelore, M.. and Fitzpatrick. S. Purification andcharacterization of acidic fibroblast growth factor from bovine brain. Proc.Nati. Acad. Sci. USA. 81: 357-361. 1984.

38. Gospadarowicz, D.. Bialecki. H., and Greenburg. G. Purification of fibroblastgrowth factor from bovine brain. J. Biol. Chem., 253: 3736-3743, 1978.

39. Raines. E. W'.. and Ross. R. Platelet derived growth factor. High yieldpurification and evidence for multiple forms. J. Biol. Chem.. 257: 5154-5160. 1982.

40. Stroobant. P.. and Watcrfield. M. D. Purification and properties of porcineplatelet-derived growth factor. EMBO J.. 3: 2963-2967. 1984.

41. Finch, P. W., Rubin, J. S., Miki, T., Ron, D., and Aaronson. S. A. HumanKGF is FGF-relatcd with properties of a paracrine effector of epithelial cellgrowth. Science (Washington DC), 245: 752-755, 1989.

42. Ervin, P. R.. Kaminski. M. S., Cody, R. L., and Wicha, M. S. Production ofmammastatin. a t issue-specific growth inhibitor, by normal human mammarycells. Science (Washington, DC), 244: 1585-1587. 1989.

43. Matsuda. H., Coughlin, M. D., Bienenstock. J.. and Dcnburg, J. A. Nervegrowth factor promotes human hemopoietic growth and differentiation. Proc.Nati. Acad. Sci. USA, 85: 6508-6512, 1988.

44. Kaighn. M. E., Lechner, J. F., Narayan, K. S., and Jones, L. W. Prostatecarcinoma: tissue culture cell lines. Nati. Cancer Inst. Monogr., 49: 17-21,1978.

45. Kaighn, M. E.. Narayan. K. S., Ohnuki, Y., Lechner, J. F., and Jones, L. W.Establishment and characterization of a human prostatic cell line (PC3).Invest. Virol., 17: 16-23, 1979.

46. Horoszewicz. J. S.. Leong, S. S.. Kawinski. E., Karr, J. P., Rosenthal. H..Chu. T. M.. Mu.ni.! E. A., and Murphy, G. P. LnCaP model of humanprostatic carcinoma. Cancer Res.. 43: 1809-1818. 1983.

6907

on June 16, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

1990;50:6902-6907. Cancer Res Victor S. Perkel, Subburaman Mohan, Sandra J. Herring, et al. Activity Which Selectively Stimulates Human Bone CellsHuman Prostatic Cancer Cells, PC3, Elaborate Mitogenic

Updated version

http://cancerres.aacrjournals.org/content/50/21/6902

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/50/21/6902To request permission to re-use all or part of this article, use this link

on June 16, 2021. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/content/50/21/6902http://cancerres.aacrjournals.org/cgi/alertsmailto:pubs@aacr.orghttp://cancerres.aacrjournals.org/content/50/21/6902http://cancerres.aacrjournals.org/