Human osteosarcoma-derived cell lines produce soluble factor(s) that induces differentiation of blood monocytes to osteoclast-like cells

Human osteosarcoma-derived cell lines produce soluble factor(s) that induces differentiation of blood monocytes to osteoclast-like cells

International Immunopharmacology 2 Ž2002. 25–38 www.elsevier.comrlocaterintimp Human osteosarcoma-derived cell lines produce soluble factor žs/ that ...

554KB Sizes 2 Downloads 68 Views

International Immunopharmacology 2 Ž2002. 25–38 www.elsevier.comrlocaterintimp

Human osteosarcoma-derived cell lines produce soluble factor žs/ that induces differentiation of blood monocytes to osteoclast-like cells q Noriki Miyamoto a,b, Yasumitsu Higuchi a,b, Kouki Mori a,b, Morihiro Ito a , Masato Tsurudome a , Machiko Nishio a , Hiroyuki Yamada b, Akihiro Sudo b, Ko Kato b, Atsumasa Uchida b, Yasuhiko Ito a,) a b

Department of Microbiology, Mie UniÕersity School of Medicine, 2-174, Edobashi, Tsu, Mie Prefecture 514, Japan Department of Orthopaedics, Mie UniÕersity School of Medicine, 2-174, Edobashi, Tsu, Mie Prefecture 514, Japan Received 7 May 2001

Abstract When monocytes were cocultured with human osteosarcoma-derived cells ŽHOS cells., multinucleated giant cell formation of monocytes was induced. Intriguingly, even when a filter was interposed between monocytes and HOS cells, polykaryocytes also appeared. The multinucleated giant cells have characters similar to osteoclast-like cells. These findings indicate that soluble factorŽs. secreted from HOS cells play an important role in polykaryocyte formation from monocytes. Twelve cloned cells were established from HSOS-1 cells and their capacities of inducing osteoclasts were investigated. Three cloned cells inducing nos. 4 and 9 had an ability of inducing osteoclasts Žmultinucleated giant cells, TRAPq, calcitonin receptor and c-src mRNAs, osteoresorbing activity., and three cells, including nos. 1 and 5, did not show the ability. HOS cells and the cloned cells expressed several cytokine mRNAs. M-CSF was detected in the culture fluids of HOS cells, which also expressed RANK and RANKrODFrOPGL mRNAs. Intriguingly, HOS cells secreting a soluble osteoclast inducing factorsŽs. expressed TNF-a converting enzyme mRNA. Furthermore, OCIFrOPG inhibited HOS cell-induced osteoclastogenesis and soluble RANKL could be detected in the culture fluids of HOS cells expressing TACE, suggesting that one of soluble osteoclast-inducing factorŽs. is soluble RANKL. When blood monocytes were indirectly cocultured with HSOS-1 cells or cloned no. 9 cells in the presence of OCIF for 14 days, HOS cell-mediated osteoclastogenesis was suppressed, indicating that RANK–RANKL system is involved in the HOS cell-mediated osteoclastogenesis. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Osteosarcoma; Osteoclastogenesis; RANKL; TACE

q

This study was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan and by the Mie Medical Research Fund. ) Corresponding author. Tel.rfax: q81-592-31-5008. E-mail address: [email protected] ŽY. Ito..

1. Introduction Osteosarcoma is defined as a malignant tumor of mesenchymal cells, characterized by the direct formation of malignant osteoid andror woven bone by

1567-5769r02r$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S 1 5 6 7 - 5 7 6 9 Ž 0 1 . 0 0 1 3 4 - 5

26

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

the tumor. Some are composed largely of fibroblastic cells, others have abundant bone formation, some show chondroid differentiation and still others are highly vascular. Osteosarcoma lesion is also characterized by bone destruction with high biological virulence, appearing to be caused by active osteoclasts. Rare cases of giant cell-rich osteosarcoma have been reported w1x. The established cell lines derived from osteosarcomas consist of mixed-cell populations containing fibroblastic cells osteoblastic cells and multinucleated giant cells. We have been interested in the mechanismŽs. by which multinucleated giant cells are induced in osteosarcoma. One of the possibilities is that formation of multinucleated giant cells is inducedrsupported by a constituent cellŽs. of osteosarcoma. Recently, osteoclast differentiation factor ŽODF. was identified and was found to be identical to TRANCErRANKL w2,3x. The ODF was also reported to play a critical role in osteoclast activity, bone resorption w4,5x. Suda et al. w6x and Udagawa et al. w7x previously showed that the cell-to-cell interaction between osteoblastsrstromal cells and osteoclast progenitors is considered to be essential for in vitro osteoclastogenesis. RANKL is expressed on the membrane of osteoblastsrstromal cells and mediates of osteoblastsrstromal cell and mediates an essential signal to osteoclast progenitors for their differentiation into osteoclasts w2–5x. In this study, we found that direct or indirect cocultures of human blood monocytes with human osteosarcoma-derived cell lines resulted in formation of multinucleated giant cells. Furthermore, these polykaryocytes exhibited the properties of osteoclasts.

such as alkaline phosphatase and osteocalcin. Tumors induced in nude mice by HSOS-1 cell inoculation revealed typical histological features of osteoblastic osteosarcoma producing prominent osteoid matrix with calcification. OSRb cells were established from a human osteosarcoma developed in a 17-year-old female patient in our laboratory w9x. The morphology of OSRb cells is not characteristic: polygonal in shape, with round or oval nuclei. Multinucleated cells often appear in OSRb cells. G292 cells were established from human osteosarcoma and were positive for osteocalcin and alkaline phosphatase w10x. KHOS cells are the Kirsten mouse sarcoma virus-transformed nonproducer cell clone of human osteosarcoma w11x. HSOS-1, OSRb and G292 cells are heterologous cell lines. These cells were cultured with Dulbecco’s minimum essential medium ŽDMEM. fortified with 5% fetal calf serum. 2.2. Other cells HeLa and CV-1 cells, which were grown in Eagle’s MEM supplemented with 5% fetal calf serum, were used as control cells. 2.3. Antibody and reagents Anti-human soluble RANKL rabbit serum and recombinant human soluble RANKL were purchased from Pepro Tech EC ŽLondon, England.. Recombinant human osteoclastogenesis inhibitory factor ŽOCIF.rosteoprotegerin ŽOPG. was bought from R & D Systems ŽMinneapolis, USA.. Recombinant MCSF, MG-CSF and IFN-g were purchased from Chemicon International ŽTemecula, CA., Genzyme ŽCambridge, MA. and Seikagaku ŽTokyo, Japan., respectively.

2. Materials and methods 2.4. Cell cloning 2.1. Cell lines deriÕed human osteosarcomas (HOS cells) Four cell lines, HSOS-1, OSRb, G292 and KHOS, were used in this study. HSOS-1 cells were established from an osteoblastic tumor arising in the left humerus of an 11-year-old girl by Sonobe et al. w8x The cell line was not cloned. Their cytoplasm is strongly positive for specific markers of osteoblasts,

Cell cloning was carried out by limiting dilution method. 2.5. Isolation of monocytes Peripheral blood monocular cells ŽPBMC. were isolated from healthy human volunteers’ heparinized whole blood by Ficoll–Hypaque density gradient

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

centrifugation. PBMC at the interface were collected and suspended in RPMI 1640 with 10% fetal calf serum ŽFCS.. To remove contaminating platelets, PBMC were washed with balanced salt solution w150 mM NaCl, 10 mM EDTA, 10 mM Tris–HCl ŽpH 7.5.x layered onto FCS and then centrifuged. Adherent monocytes were obtained from PBMC by their attachment to tissue culture dishes. The purity of monocytes ŽCD14q cells. isolated by Ficoll–Hypaque and adherence was approximately 90–95%. The viability of the adherent cells after removal with a cell scraper ŽNippon Becton Dickinson, Tokyo, Japan. instead of EDTA is more than 98%, as measured by trypan blue exclusion. The monocytes were cultured in RPMI 1640 supplemented with 10% FCS instead of human serum because in previously reported studies in which human serum was used, spontaneous formation of small-sized giant cells of monocytes was found w12x Spontaneous giant cell formation was rarely detected during 4 weeks of monocyte cultivation in RPMI 1640 medium supplemented with 10% FCS. 2.6. Cocultures The purified monocytes were cocultured with the cell lines derived from human osteosarcomas for 14 days. We used two methods Ždirect and indirect methods. for cocultivation of monocytes and HOS cells: one Ždirect method. was that monocytes Ž1–2 = 10 5 cells. were cocultured with monolayers of the cell line Ž2–3 = 10 4 cells. in 24-well dishes, and the other Žindirect method. was that monocytes Ž4–5 = 10 5 cells. were cultured in 24-well dishes and the cell lines Ž2–3 = 10 4 cells. were cultured in the cell-culture insert Žpore size 0.45 m m. of the same well ŽFalcon.. 2.7. Staining with Giemsa solution The cells were dried, were fixed with methanol for 5 min and then were stained with 5% Giemsa solution for 30 min. 2.8. Tartrate-resistant acid phosphatase (TRAP) assays Staining for TRAP was performed by incubating the fixed cells for 30 min at 37 8C in an acute buffer

27

containing a-naphthol phosphate at pH 4 in the presence of 500 mM tartrate. Fast violet was then added to make the product visible ŽHisto-chemical Kit 386-A: Sigma, St. Louis, MO.. 2.9. Isotopic labeling, RIPA and SDS-PAGE Isotopic labeling of cells, radioimmuno-precipitation assay ŽRIPA. and sodium dodecyl sulfate-polyacrylamide gel electrophoresis ŽSDS-PAGE. were done as described elsewhere w13x. 2.10. RNA isolation and first-strand cDNA synthesis Total cellular RNA was extracted from 10 5 –10 6 cells using guanidine isothiocyanate–cesium chloride method, as described previously. PolyŽA.q RNA was purified by oligoŽdT.-cellulose chromatography ŽPharmacia Biotech.. RNA primed with specific primers was reverse transcribed using cloned Moloney murine leukemia virus reverse transcriptase Ž2.5 U., 1 mM of each deoxy-NTP, 0.2 m g specific primer, 1 unit RNase inhibitor in a final volume of 15 m l. The reaction was run at 37 8C for 60 min to complete the extension reaction. The reaction mixture was heated at 90 8C for 5 min to denature the RNA–cDNA hybrids and quick-chilled on ice. The reverse transcriptase used in this study is Thermoscript RT ŽGibco BRL, Life Technologies, Rockville, MD, USA.. 2.11. ReÕerse transcription-polymerase chain reaction assays (RT-PCR) The first-strand cDNA was submitted to PCR amplification using gene-specific PCR primers Žsee Table 1. as follows: b-actin Žpositive control., calcitonin receptor ŽCTR., c-src, RANK Žreceptor activator of NF-k B., RANKL ŽRANK ligand.rODF Žosteoclast differentiating factor.rOPGL Žosteoprotegerin ligand., M-CSF Žmacrophage colony-stimulating factor., interleukin ŽIL.-6, CD14 Žnegative control., IL-3, IL-4, interferon-g , GM-CSF Žgranulocyte-macrophage colony-stimulating factor., osteoprotegerinrosteoclastogenesis-inhibitory factor ŽOCIF., TNF-a converting enzyme ŽTACE, ADAM 17., tissue inhibitor of metaloproteinases ŽTIMP.-3,

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

28

Table 1 List of primers used for RT-PCRa Target mRNA

Primer sequences

b-actin

5 -TGACGGGGTCACCCACACTGTGCCCATCTA-3 X X 5 -CTAGAAGCATTTGCGGTGGACGATGGAGGG-3 X X 5 -ACTGCTGGCTGAGTGTGGAAA-3 X X 5 -GAGCAGTAGATGGTCGCAAC-3 X X 5 -CGCCGAGCCCAAGCTGTTCGGAGGCTTCA-3 X X 5 -CAGCCTGGATGGAGTCGGAGGGCGCCACGTA-3 X X 5 -TTTGCAGATCGCTCCTCCAT-3 X X 5 -AGGCATCAGAGAAGTAGCCT-3 X X 5 -CCTGAGACTCCATGAAAACGC-3 X X 5 -TAACCCTTAGTTTTCCGTTGC-3 X X 5 -ATGCGCTTCAGAGATAACAC-3 X X 5 -CAAGAACTGCAACAACAGCT-3 X X 5 -CTGAGAAAGGAGACATGTAA-3 X X 5 -GAACCAGTGGCTGCAGGACA-3 X X 5 -ACTCCCTCAATCTGTCGTTCGCTG-3 X X 5 -CTGAAGCCAAGGCAGTTTGAGTCC-3 X X 5 -ATGAGCCGCCTGCCCGTCCTG-3 X X 5 -GCGAGGCTCAAAGTCGTCTGTTG-3 X X 5 -ATGGGTCTCACCTCCCAACTGCT-3 X X 5 -CGAACACTTTGAATATTTCTCTCTCAT-3 X X 5 -ATGAAATATACAAGTTATATCTTGGCTTT-3 X X 5 -GATGCTCTTCGACCTCGAAACAGCAT-3 X X 5 -ACACTGCTGAGATGAATGAAACAGTAG-3 X X 5 -TGGACTGGCTCCCAGCAGTCAAAGGGGATG-3 X X 5 -CCTGACCACTACTACACAGACA-3 X X 5 -GTTAGCAGGAGACCAAAGACACTGCA-3 X X 5 -TTTCAAGAACGCAGCAATAAAGTTTGTGGG-3 X X 5 -ACAGTGTCATCTTCAGCATTTCCTGGAG-3 X X 5 -CCTTTGGCACACTGGTCTACA-3 X X 5 -AAAGAGAAGGAGTAGGGTGGG-3 X X 5 -AATATCCTCTGGGGTTTGGCCTA-3 X X 5 -TCCATCCTCATACCTGCACCTTT-3 X X 5 -GAGAGCATGGTGGTAGTAGCAACC-3 X X 5 -CCCTGCCAAGCACACCCAGTAGTC-3 X X 5 -GAGTGACAAGCCTGTAGCCCATGTTGTAGCA-3 X X 5 -GCAATGATCCCAAAGTAGACCTGCCCAGACT-3

CTR c-src RANK RANKLrODFrOPGL M-CSF IL-6 CD14 IL-3 IL-4 IFN-g GM-CSF OCIFrOPG TACErADAM17 TIMP-3 OsCa IL-1 a TNF-a a

X

X

Top sequence is sense primers and buttom sequence is antisense primers.

osteocalcin, IL-1 a and TNF-a ,. PCR was performed in 50-m l reactions containing 10 mM Tris–HCl ŽpH 8.3., 50 mM KCl, 1.5 mM MgCl 2 , 0.2 mM of each dNTP, 0.5 m M of each of the sense and antisense PCR primers and 2.5 units of Taq DNA polymerase ŽPerkin-ElmerrCetus, Norwalk, CT.. The reaction mixture was then subjected to 30 cycles of amplification in a DNA thermal cycler. Each cycle consisted of a heat-denaturation step at 94 8C for 1 min, annealing of primers at 50–60 8C Žoptimized for

each primer pair. for 1 min, and an extension step at 72 8C for 1 min. Following completion of 30 PCR cycles, the reactions were incubated at 72 8C for 5 min. The PCR products were separated by electrophoresis on a 1.5% agarose gel and visualized by ethidium bromide staining with ultraviolet light illumination. To confirm that the PCR products were derived from target mRNA, the products were cloned using a TA cloning kit ŽInvitrogen, San Diego, CA. and the nucleotide sequences were analyzed.

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

2.12. Bone resorption assays A piece of human cortical bone was obtained during operations after informed consent, bleached in air for several days and sterilized with ethylene oxide for experiment. A small piece of cortical bone together with blood monocytes was placed in soft agar in 16-mm wells and partially submerged to mask the cross-cut surfaces and concentrate the cells on the center of cortical bones. HOS cells were cultured in the cell-culture insert Žpore size 0.45 m m. of the same well. One milliliter of medium was added to each well, then the cultures were incubated for 14 days in a humid atmosphere of 15% CO 2 at 37 8C. As a control, the culture was incubated with monocytes or HOS cells alone under the same conditions. The specimens were fixed with a mixture of 2.5% glutaraldehyde and 2% formaldehyde in 0.1 M sodium cacodylate buffer ŽpH 7.2. and osmificated. They were then dehydrated in graded concentrations of ethanol, critical point dried using liquid CO 2 and sputter-coated with platinum. All specimens were examined with a scanning electron microscope. 2.13. Fusion index To estimate the degree of cell fusion, numbers of cells or nuclei within multinucleated giant cells ŽP 3 nucleircell. were counted. Fusion index was calculated using the formula: Fusion index Ž % . total number of nuclei within giant cells s total number of nuclei counted

= 100

The result were compared between each group and the significant difference of each mean value was analyzed statistically with the Student’s t-test.

3. Results 3.1. Osteoclast-like cell formation by cocultiÕation of monocytes with cell lines deriÕed from human osteosarcomas When blood monocytes were cocultured with the monolayer of cell lines derived from human os-

29

teosarcomas ŽHOS cells., cell aggregation was formed within 1 day, and multinucleated giant cells began to appear at approximately 3–4 days Ždata not shown.. The polykaryocytes increased in size until 7 days of incubation and the number of multinucleated giant cells continued to increase for 14 days ŽFig. 1I.. On the other hand, when human blood monocytes were cocultured without or with CV-1 or HeLa cells as control cells, small clusters of monocytes appeared within a few hours, the size of these aggregates did not became bigger thereafter and these aggregates were abolished by 24 h Ždata not shown.. No multinucleated giant cell was formed in these cultures ŽFig. 1I and Table 2.. Intriguingly, even when a Millipore filter Žpore size 0.45 m m. was interposed between monocytes and HOS cells Žindirect coculture., cell aggregation and multinucleated giant cell formation of monocytes were also induced ŽFig. 1I., indicating that soluble factorŽs. secreted from HOS cells plays an important role in polykaryocyte formation from monocytes. These multinucleated giant cells exhibited tartrate-resistant acid phosphatase ŽTRAP. positive and expressed calcitonin receptor and c-src mRNAs ŽFigs. 1II and 2.. Specificities of the mRNAs were confirmed by the nucleotide sequence Ždata not shown.. Bone resorption activity of the multinucleated giant cells was also examined. Monocytes were indirectly cocultured on human cortical bone slices for 14 days with HSOS-1 cells. The multinucleated giant cells formed resorption pits on the surfaces of human cortical bone slices ŽFig. 1II.. These findings show that the multinucleated giant cells have characters as osteoclast-like cells. Collectively, all four cell lines derived from human osteosarcomas were found to have the capacity of inducing osteoclast-like cells from blood monocytes directly or via soluble factorŽs. ŽFig. 1, Table 2.. 3.2. Osteoclast-inducing actiÕity of the cloned cells deriÕed from HSOS-1 cell line Twelve cloned cells were established from HSOS1 cells and their capacities of inducing osteoclasts were investigated ŽFig. 2.. Human blood monocytes Ž4–5 = 10 5 cells. were cultured in 24-well dishes and the cloned cells Ž2–3 = 10 4 cells. were cultured

30

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

Fig. 1. Osteoclast-like cell formation by cocultivation of monocytes with human osteosarcoma-derived cells ŽHOS cells.. ŽI. Human blood monocytes were directly cocultured with CV-1 cells ŽA. or HSOS-1 cells ŽB., and in the next experiment, a Millipore filter Žsize 0.45 m m. was interposed between monocytes and HeLa cells ŽC. or HOS cells wHSOS-1 ŽD, E., OSRb ŽF., G292 ŽG., KHOS ŽH.x. ŽC. and ŽD. show unfixed monocytes at 1 day post incubation. Other monocytes were cultured for 14 days, fixed and then stained with Giemsa solution for enumerating the number of nuclei. The magnifications of ŽA., ŽB., ŽE., ŽF., ŽG. and ŽH. are the same, and those of ŽC. and ŽD. are also the same. ŽII. TRAP staining and bone resorption activity of osteoclast-like cells. Human blood monocytes were cocultured with HSOS-1 ŽA., OSRb ŽB., G292 ŽC. or KHOS ŽD., and a millipore filter Žsize 0.45 m m. was interposed between monocytes and HOS cells. These cells were incubated for 14 days, fixed and then stained with TRAP staining solution. Subsequently, monocytes were indirectly cocultured with CV-1 ŽE. or HSOS-1 cells ŽF, G. on human cortical bone slices for 14 days, and then bone resorption was examined. All specimens were examined with a scanning electron microscope. The magnifications of ŽA., ŽB., ŽC. and ŽD. are the same. Bars indicate 50 m m.

in the cell-culture insert Žpore size 0.45 m m. of the same well. Three cloned cells, including nos. 4 and 9, had an ability of inducing osteoclasts Žmultinucleated giant cells, TRAPq, calcitonin receptor and c-src mRNAs, osteoresorbing activity., and three cells, including nos. 1 and 5, did not show the

ability. We could not obtain a clear result from the remaining six cells. Furthermore, the culture fluid from clone cells no. 4 or 9 were capable of inducing differentiation of monocytes to osteoclasts, while the culture fluid from clone cells no. 1 or 5 showed no activity Ždata not shown.. When clone cells no. 5

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

31

Table 2 Phenotypes of HSOS-1 and the cloned cells HSOS-1

RT-PCR b-actin TNF-a IL-1 a RANK RANKLrODFrOPGL M-CSF IL-6 CD14 IL-3 IL-4 IFN-g GM y CSF OCIFrOPG TACErADAM17 TIMP-3 Fusion inducing ability c

qa q q q q q q y q q q q q q q 64.3 " 12.3

Clone cells Žno.. 1

4

5

9

q y y q q q q y q q y y y y y -2

q y y q q q q y q q q y y q q 45.8 " 5.7

q y y q q q q y q q y y y y y -2

q y y q q q q y q q q q y q q 67.7 " 2.4

HeLa

CV-1

Monocytes

q q q q yb q q y y y y y y y y -2

q q y q y q q y y y y y q y y -2

q q q q y q y q q y q y y q y r

a

Žq. Positive reaction. Žy. Negative reaction. c Monocytes Ž4–5 = 10 5 cells. were cultured in 24-well dishes and HOS or control cells Ž2–3 = 10 5 cells. were cultured in the cell-culture insert Žpore size: 0.45 m m. of the same well for 14 days. To estimate the degree of cell fusion, the number of cells or nuclei within multinucleated giant cells of monocytes was counted. Fusion inducing ability was expressed as fusion index Žmeans" S.D... b

were cocultured with blood monocytes without Millipore filter, the clone cells induced multinucleated giant cell formation monocytes ŽFig. 2III., indicating that a factorŽs. showing osteoclast-inducing ability is expressed on the cell surfaces of the cloned cells. 3.3. Cytokine gene expressions in the cloned cells deriÕed from HSOS-1 Expression of various cytokine genes in the cloned cells were studied by using RT-PCR ŽFig. 3I, Table 2.. Expression of M-CSF, IL-3, IL-4 and IL-6 mRNAs was detected in the all cloned cells irrespective of showing an osteoclast-inducing activity. IL-1 a and TNF-a mRNAs were found in HSOS-1 cells, while the clone cells tested expressed neither IL-1 a nor TNF-a mRNA. GM-CSF mRNA was detected in clone 9 and IFNg mRNA was detected in clones nos. 4 and 9, suggesting that GM-CSF and IFNg may be related to an osteoclast-inducing activity via a soluble factorŽs.. However, when blood monocytes were indirectly cocultured with cloned cells no. 1 or 5 in the presence of GM-CSF andror IFNg , multin-

ucleated giant cell formation and TRAP positive cells were not found Ždata not shown., indicating that GM-CSF and IFN-g are not involved in osteoclastogenesis in this system. In addition, we assayed various cytokines in culture media derived from HSOS-1 clones 5 and 9 ŽTable 3.. Neither IL-1 a , IL-1 b , IL-1r a , IL-2, IL-4, IL-8, IL-10, IL-18, TNF-a nor G-CSF could be detected in both culture media from HSOS-1 clones 5 and 9. A very small amount of IL-6 or IFN-g was found in one lot of clone 5 or 9 culture medium, respectively. Interestingly, a significant amount of M-CSF could be detected in both the media, and M-CSF in culture medium of clone 5 was 10–15 times as much as that of clone 9. 3.4. Expression of RANK, RANKL, TACE, OCIF and TIMP-3 mRNAs in HOS cells We tried to detect the expression of RANK and RANKL ŽFig. 3II, Table 2., because the RANKLr ODFrOPGrTRANCE has been recently reported to play a critical role in osteoclastogenesis w2–4x. All the cells, HSOS-1 and cloned cells nos. 1, 4, 5 and 9,

32

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

Fig. 2. Formation of osteoclast-like cells by clone cells from HSOS-1 cells. ŽI. Morphology of clone cells from HSOS-1 cells wŽA. clone cells no. 1; ŽB. no. 4; ŽC. no. 5; ŽD. no. 9x. Human blood monocytes were cocultured with clone cell no. 1 ŽE., no. 4 ŽF., no. 5 ŽG. or no. 9 ŽH., and a Millipore filter Žsize 0.45 m m. was interposed between monocytes and the clone cells. These cells were incubated for 14 days, fixed and then stained with TRAP staining solution. Subsequently, monocytes were indirectly cocultured with clone cells no. 1 ŽI., no. 4 ŽJ., no. 5 ŽK. or no. 9 ŽL. on human cortical bone slices for 14 days, and then bone resorption was examined. All specimens were examined with a scanning electron microscope. White bars indicate 50 m m. ŽII. Detection of calcitonin receptor and c-src mRNAs. Human blood monocytes were indirectly cocultured with HOS cells for 14 days and the mRNAs were isolated from the monocytes. Expression of calcitonin receptor and c-src mRNAs was analyzed by RT-PCR. Lane 1, clone 1; lane 2, clone 4; lane 3, clone 5; lane 4, clone 9; lane 5, HSOS-1 cells. Ž™.: positive PCR product. ŽIII. Induction of polykaryocytes of monocytes by direct cocultivation with clone 5 cells. Human blood monocytes were directly cocultured with clone 5 cells for 14 days, fixed and then stained with Giemsa’ solution.

expressed both RANK and RANKL mRNAs. Expression of TACE which has been reported to cleave RANKL, resulting in release to soluble RANKL w14x was found in clones nos. 4 and 9, but was not in

clones 1 and 5 ŽFig. 3II, Table 2.. OCIF mRNA was detected in HSOS-1, but not in clone cells nos. 1, 4, 5 and 9, showing that disability of clone cells nos. 1 and 5 in differentiation of monocytes to osteoclast-

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

like cells by indirect coculture is not due to production OCIF. In next experiment, expression of the tissue inhibitor of metalloproteinase-3 ŽTIMP-3. was analyzed. TIMP-3 mRNA was detected in HOS cells expressing TACE mRNA ŽFig. 3II, Table 2..

33

3.5. Blocking of HOS cell-mediated osteoclastogenesis by OCIF When blood monocytes were indirectly cocultured with HSOS-1 cells or cloned no. 9 cells in the presence of OCIF for 14 days, HOS cell-mediated osteoclastogenesis was suppressed ŽFig. 4., indicating that RANK–RANKL system is involved in the HOS cell-mediated osteoclastogenesis. 3.6. Detection of soluble RANKL in the culture of HOS cells We tried to detect soluble RANKL in the culture fluids of HOS cells. HeLa, HSOS-1, clones 5 and 9 cells were labeled with w35 Sx-methionine, and then culture fluids were analyzed by immunoprecipitation using anti-soluble RANKL antibody and SDS-PAGE. As shown in Fig. 3III, soluble RANKL is detected in the culture fluids of HSOS-1 and clone 9, but not in those of clone 5 and HeLa cells, showing that one of soluble osteoclast-inducing factorŽs., which was capable of inducing differentiation of blood monocytes to osteoclast-like cells RANKLq M-CSF for 7 days, soluble RANKL induced polykaryocyte formation of monocytes with aide of M-CSF under our experimental conditions Ždata not shown..

Fig. 3. ŽI. Expression of various cytokines mRNAs in HOS cells and control cells. Expression of b-actin Žpositive control mRNA., TNF-a , IL-1 a , M-CSF, GM-CSF, IL-3, IL-4, IL-6, IFN-g and CD14 Žnegative control mRNA. in HSOS-1 Žlane 1., clone cells no. 1 Žlane 2., no. 4 Žlane 3., no. 5 Žlane 4., no. 9 Žlane 5., HeLa Žlane 6., CV-1 cells Žlane 7. and freshly isolated monocytes Žlane 8. was analysed by RT-PCR. ŽII. Expression of RANK, RANKL, TACE, OCIF and TIMP-3 mRNAS in HOS cells and control cells. Expression of b-actin Žpositive control mRNA., RANK, RANKLrODFrOPGL. TACE, OCIF, TIMP-3 mRNAs and CD14 Žnegative control mRNA. in HSOS-1 Žlane 1., clone cells no. 1 Žlane 2., no. 4 Žlane 3., no. 5 Žlane 4., no. 9 Žlane 5., HeLa Žlane 6., CV-1 cells Žlane 7. and freshly isolated monocytes Žlane 8. was analyzed by RT-PCR. ŽIII. Detection of soluble RANKL in the culture fluid of HOS cells: HeLa cells, HSOS-1 cells, clones 5 and 9 were labeled with w35 Sx-methionine for 48 h, and then culture fluids were analyzed by immunoprecipitation using antisoluble RANKL antibody and SDS-PAGE. Lane 1, HeLa cells; lane 2, HSOS-1 cells; lane 3, clone 5; lane 4, clone 9.

34

Culture fluid derived from

Sample no.

HSOS-1 clone 5

1 2 1 2

HSOS-1 clone 9

Concentrations of various cytokines IL-1 a IL-1 b IL-1r a IL-2 Žpgrml. Žpgrml. Žpgrml. Žpgrml.

IL-4 Žpgrml.

IL-6 Žpgrml.

IL-8 Žpgrml.

IL-10 Žpgrml.

IL-18 Žpgrml.

TNF-a Žpgrml.

M-CSF Žpgrml.

G-CSF Žpgrml.

IFN-g Žpgrml.

ND b ND ND ND

ND ND ND ND

ND 2 ND ND

ND ND ND ND

ND ND ND ND

ND ND ND ND

ND ND ND ND

1416 1441 97 151

ND ND ND ND

ND ND ND 11

ND ND ND ND

ND ND ND ND

ND ND ND ND

a Monolayers of HSOS-1 cell clones 5 and 9 were washed with MEM, further incubated with MEM containing no FCS for 2 days and then culture fluids obtained. Concentrations of various cytokines were assayed by ELISA. b Not detectable.

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

Table 3 Cytokine titration in the culture media derived from HSOS-1 clones 5 and 9 a

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

35

Fig. 4. Blocking of HOS cell-mediated osteoclastogenesis by OCIF. ŽI. Human blood monocytes were indirectly cocultured with HSOS-1 cells ŽA, B. or cloned no. 9 cells ŽC, D. without ŽA, C. or with OCIF ŽB, D; 100 ng. for 14 days, fixed and then stained with Giemsa solution. ŽII. Quantitative analysis of multinucleated giant cell formation. For quantitative analysis of multinucleated giant cell formation, fusion index was calculated. The results were compared between each group and the significant difference of each mean value was analyzed statistically with the Student’s t-test.

36

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

4. Discussion Osteosarcoma is a malignant tumor that originates within bone, permeates and rapidly destroyed the cancellous and cortial bones. The destruction of bones is thought to be caused by osteoclasts. Giant cell-rich osteosarcoma was first described by Bathurst and Sanerskin w1x and its incidence was reported to make up 3% of all esteosarcoma cases. The established cell lines derived from osteosarcomas consist of mixedcell populations containing multinucleated giant cells. The mechanismŽs. by which multinucleated giant cells are induced in osteosarcoma is interesting, and regulation of osteoclast formation in osteosarcoma may control tumor growth and bone destruction. In this study, we examined whether osteoclast-like cells were induced by cocultivation of blood monocytes with HOS cells. When blood monocytes were cocultured with HOS cells for about 3 days, multinucleated giant cell formation of monocytes was observed. When these multinucleated giant cells were further incubated together with HOS cells, these cells showed the properties of osteoclast cells. Therefore, it is inferred that HOS cells support osteoclast formation in osteosarcoma. In addition, HOS cells were found to secrete a soluble factorŽs., which was capable of inducing differentiation of blood monocytes to osteoclast-like cells. All the clone cells from HSOS-1 cells are found to express several cytokines, M-CSF, IL-3, IL-4 and IL-6 mRNAs, all of which were osteoclast differentiation-inducing and activating factors for monocyter macrophage lineage cells w15,16x. On the other hand, GM-CSF and IFN-g mRNAs were expressed only in HOS cell lines secreting a soluble osteoclast-inducing factor. However, GM-CSF and IFNg are not involved in osteoclastogenesis in this system. Suda et al. w6x and Udagawa et al. w7x previously showed that the cell-to-cell interaction between osteoblastsrstromal cells and osteoclast progenitors was essential for in vitro osteoclastogenesis, and they proposed a hypothesis that osteoblastrstromal cells express a common factor, osteoclast differentiation factorsŽs. ŽODF., in response to other osteotropic factors w17x. They speculated that ODF appeared a member-associated factor, since the cell-to-cell contact between osteoclast precursor and osteoblastsr stromal cells was prerequisite for osteoclast forma-

tion w17x. Recently, Yasuda et al. w3x and Lacey et al. w2x have found that osteoclast differentiation factor is a ligand for osteoprotegerinrosteoclastogenesis-inhibitory factor and is identical to TRANCErRANKL. Therefore, ODF is considered to be the mediator of cell interaction between osteoblastsrstromal cells and osteoclast progenitors. Furthermore, M-CSF is indispensable for both the proliferative phase and differentiation phase of osteoclast development w2–5x. A genetically prepared soluble form of ODF can induce differentiation of peripheral blood mononuclear cells to osteoclasts with aide of M-CSF w2–5x. In addition, soluble RANKLq M-CSF also induced polykaryocyte formation of human blood monocytes under our experimental conditions. However, the soluble RANKL was not found under physiological conditions. It has recently reported that TNF-a stimulates osteoclast differentiation with the aid of IL-1 ar IL-1 b by a mechanism independent of the ODFr RANKL–RANK interaction w18,19x. All the cloned HOS cells express neither TNF-a nor IL-1 a . Giant cell tumor-derived stromal cells ŽGCTSC. secreted a soluble factorŽs. which was capable of differentiating of blood monocytes to osteoclast-like cells w12x. GCTSC was also found to express several cytokines, M-CSF, IL-6 and RANKLrODFrOPGL w12x. Since GCTSC express RANKL and M-SCF and blood monocytes express RANK, one of possible mechanisms is that the soluble RANKL is related from GCTSC and the soluble factor interacts with RANK expressed on monocytes, resulting in osteoclast-like cell formation in cooperation with M-CSF secreted from GCTSC. Miyamoto et al. w12x discussed as follows: although the soluble RANKL has not been found under physiological conditions, it can not be excluded that RANKL is cleaved by a cellular proteaseŽs. and truncated RANKL is released from cells. Surprisingly, all the cells analyzed, HSOS-1 and the clone cells nos. 1, 4, 5 and 9, expressed RANK and RANKL mRNAs. The finding that the clone cells no. 5 express RANKL mRNA is not inconsistent with that the cells induced differentiation of monocytes to osteoclast-like cells via direct cell-tocell interaction. TNF-a converting enzyme ŽTACE; ADAM-17. is a membrane-bound disintegrin metalloproteinase w14x. Since TACE has been reported to be able to cleave RANKL, resulting in release of

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

soluble RANKL w14x, expression of TACE mRNA was investigated. The HOS cells secreting a soluble osteoclast-inducing factor expressed TACE mRNA, but the other HOS cells which did not secrete such factor expressed no TACE mRNA. Soluble RANKL was detected in the culture fluids of HSOS-1 and clone 9 but in those of HeLa cells and clone 5 cells, indicating that one of soluble osteoclast-inducing factorŽs. is soluble RANKL. TACE was found to be will inhibited by tissue inhibitors of metalloproteinase-3 ŽTIMP-3. but not by TIMP-1, -2 and -4 w20x. The TIMP-3 mRNA was found in the cells expressing TACE mRNA, strongly suggesting that TACE and TIMP-3 are coregulated in the same cells. It is interesting that TACE and its inhibitor, TIMP-3, are coexpressed. It is supposed that release of RANKL is dependent on amount of RANKL expressed on the cell surfaces and on balance between TACE and TIMP-3. One of possibilities is that the clone showing weak osteoclast forming activity express low amounts of RANKL or low rate of TACErTIMP-3. In addition, rate of TACErTIMP-3 may vary in the clones showing no constant positive osteoclast forming activity every culture. The osteoclastogenesis inhibitory factor ŽOCIF.r osteoprotegerin ŽOPG. is a member of the TNF receptor family, but it does not have a transmembrane domain w21,22x. The OCIFrOGF ligand is identical to ODF ŽTRANCErRANKL. and OCIFr OPG functions as a decoy receptor for ODFr RANKL, resulting in blocking RANKL activity w21,22x. Intriguingly, HOS cell-mediated osteoclastogenesis was suppressed by OCIF, indicating that RANK–RANKL system is involved in HOS cellmediated osteoclastogenesis. However, the possibility that some fusion factors other than RANKL were also produced in this experiment could not be completely excluded. In summary, the cell lines derived from human osteosarcomas were found to have the capacity of inducing osteoclast-like cells from blood monocytes directly or via soluble factorŽs.. Taken together, RANKL is cleaved by TACE and truncated RANKL is released from some HOS cells. The soluble RANKL interacts with RANK expressed on monocytes, resulting in osteoclast-like cell formation in cooperation with M-CSF secreted from HOS cells in this system.

37

Acknowledgements We acknowledge Drs. Yasuichi Ohmoto and Kaori Murata ŽOtsuka Pharmaceutical, Tokushima, Japan. for the assay of cytokines. We also acknowledge Mr. Kasuyoshi Nanba in the Department of Microbiology for his excellent technical assistance.

References w1x Bathurst N, Sanerskin N. Osteoclast-rich osteosarcoma. Br J Radiol 1986;59:667–73. w2x Lacey DL, Timms E, Tan HL, Kelly MJ, Dunstan CR, et al. Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 1998;93:165–76. w3x Yasuda H, Shima N, Nakagawa N, et al. Osteoclast differentiation factor is a ligand for osteoprotegerinrosteoclastogenesis-inhibitory factor and is identical to TRANCErRANKL. Proc Natl Acad Sci U S A 1998;95:3597–602. w4x Jimi E, Akiyama S, Tsurukai T, Okahashi N, Kobayashi K, et al. A combination of osteoclast differentiation factor and macrophage-colony stimulating factor is sufficient for both human and mouse osteoclast formation. Endocrinology 1998;139:4424–7. w5x Quinn JM, Elliot J, Gillespie MT, Martin TJ. A combination of osteoclast differentiation factor and macrophage-colony stimulating factor is sufficient for both human and mouse osteoclast formation. Endocrinology 1998;139:4424–7. w6x Suda T, Takahashi N, Martin TJ. Modulation of osteoclast differentiation. Endocr Rev 1992;13:66–80. w7x Udagawa N, Takahashi N, Akatsu T, Sasaki T, Yamagutchi A, et al. The bone marrow-derived stromal cell lines MC3T3-G2rPA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 1989;125:1805–13. w8x Sonobe H, Mizobutchi H, Manabe Y, Furihata M, Iwata J, et al. Morphological characterization of a newly established human osteosarcoma cell line, HS-Os-1, revealing its distinct osteoblastic nature. Virchows Arch B: Cell Pathol Include Mol Pathol 1991;60:181–7. w9x Yamazaki T. Molecular characterization of a second primary osteosarcoma cell line ŽOsrbrN-M. established from a patient cured of bilateral retinoblastoma. Mie Miecal J 1995; 45:101–11. w10x Peebles PT, Trisch T, Papageorge AG. Isolation of four unusual pediatric solid tumor cell lines. Pediatr Res 1978; 12:485. w11x Rhim JS, Cho HY, Huebner RJ. Non producer human cells induced by murine sarcoma virus. Int J Cancer 1975;15:23–9. w12x Miyamoto N, Higuchi Y, Tajima M, Ito M, Tsurudome M, et al. Spindle-shaped cells derived from giant cell tumor of bone support differentiation of blood monocytes to osteoclast-like cells. J Orthop Res 2000;18:647–54.

38

N. Miyamoto et al.r International Immunopharmacology 2 (2002) 25–38

w13x Ito Y, Komada H, Kusagawa S, Tsurudome M, Matsumura H, et al. Isolation and characterization of monoclonal antibodies which enhance giant polykaryocyte formation in Newcastle disease virus-infected cell lines of human origin. J Virol 1992;66:5999–6007. w14x Lum L, Wong BR, Josien R, Becherer JD, ErdjumentBromage H, et al. Evidence for a role of a tumor necrosis factor-a ŽTNF-a .-converting enzyme-like protease in shedding of TRANCE, a TNF family member involved in osteoclastogenesis and dendric cell survival. J Biol Chem 1999; 274:13613–8. w15x Tamura T, Udagawa N, Takahashi N, Miyaura C, Tanaka S, et al. Soluble interleukin-6 receptor triggers osteoclast formation by interleukin 6. Proc Natl Acad Sci U S A 1993;90: 11924–8. w16x Yoshida H, Hayashi S, Kunisada T, Ogawa M, Nishikawa ST, et al. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature ŽLondon. 1990;345:442–4. w17x K. Matsuzaki, N. Udagawa, N. Takahashi, K. Yamaguchi, H. Yasuda, et al., Osteoclast differentiation factor ŽODF. in-

w18x

w19x

w20x

w21x

w22x

duces osteoclast-like cell formation in human peripheral blood mononuclear cell cultures. Biochem. Biophys. Res. Azuma Y, Kaji K, Katogi R, Takeshita S, Kudo Y, et al. Tumor necrosis factor-alpha induces differentiation of and bone resorption by osteoclasts. J Biol Chem 2000;275: 4858–64. Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, et al. Tumor necrosis factor alpha stimulates osteoclast differentiation by mechanism independent of the ODFr RANKL–RANK interaction. J Exp Med 2000;191:275–86. Amour A, Slocombe PM, Webster A, Butler M, Knight CG, et al. TNF-a converting enzyme ŽTACE. is inhibited by TIMP-3. FEBS 1998;435:39–44. Simonet WS, Lacey DL, Dunstan CR, Kelly M, Chang MS, et al. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 1997;89:309–19. Tsuda E, Goto M, Mochizuki S, Yano K, Kobayashi F. Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem Biophys Res Commun 1997;234:137–42.