- Email: [email protected]
SARG: A new human osteosarcoma cell line. Expression of bone markers and of major histocompatibility antigens
Annals of Oncology 3 (Suppl. 2): S29-S31, 1992. © 1992 Kluwer Academic Publishers. Printed in the Netherlands.
Original article SARG: A new human osteosarcoma cell line. Expression of bone markers and of major histocompatibility antigens K. Scotlandi,1 M. Serra,1 L. Landuzzi2 & N. Baldini1 1
Laboratorio di Ricerca Oncologica, Istituto Ortopedico Rizzoli, Bologna, Italy; 2Istituto di Cancerologia, Universitd di Bologna, Bologna, Italy
gamma (IFN-y), whereas interferon alpha (IFN-a) and tumor necrosis factor alpha (TNF-a) increased the expression of class I antigens, but not of class II antigens. SARG was tumorigenic after subcutaneos injection in nude mice. Experimental metastases were never detected.
Introduction
vated Fetal Calf Serum (Biological Industries, Kibbutz Beth Haemek, Israel). Cells were maintained at 37 °C in a humidified 5% CO2 atmosphere. Doubling time and saturation density were determined by daily harvesting of cells after seeding of 20,000 cells/cm2. Cloning efficiency was evaluated 7 days after seeding of 8—512 cells/cm2. Cloning efficiency in semi-solid medium was determined 14 days after seeding 400-4,000 cells/cm2 in 60 mm dishes. Cells were found to be mycoplasma-free by fluorescence staining with Hoechst 33258 [9]. The analysis of DNA content was performed with a static cytofluorometer (Nikon) after staining with DAPI.
Several cell lines have been established from human OSA [1-7], but the phenotypic pattern and the ability to differentiate along the osteoblastic pathway of many OSA cell lines have not been sufficiently characterized [1, 3-5]. Furthermore, most of them have been maintained in vitro for many years; progressive changes have likely occurred during long-term maintenance in vitro, therefore some of human OSA cell lines may have lost the original features. We report a new cell line (SARG) derived from a human radiation-induced OSA. Its recent origin makes it a good model for the study of the effects of cytokines with potential therapeutic use on OSA cells. As IFNs have been suggested to influence OSA differentiation [8], we have investigated the ability of IFNs and TNF-a to modulate the osteoblastic differentiation in vitro and the HLA expression of SARG cells.
Key words: osteosarcoma, MHC antigens, interferons, tumor necrosis factor-alpha
Cytokine treatment Cells were treated in vitro with IFN-a (10-10,000 U/ml), IFN-y (5-1000 U/ml), TNF-a (100-10,000 U/ml). Recombinant HuIFNy and recombinant HuTNF-a were purchased from Boehringer Mannheim Italia SPA, Milan, Italy. Recombinant HuIFN-a was purchased from Hoffman-La Roche, Basel, Switzerland. Cytokines were added to cultures 24 h after seeding; cells were harvested 144 h later. Differentiation markers and HLA expression
Osteoblastic differentiation markers were analyzed on cytospins fixed with a methanol-aceton solution (3:7) at - 2 0 °C for 10 min. The alkaline phosphatase intracellular content was determined by a Materials and methods cytochemical method (Sigma Diagnostic, St. Louis, MO, USA). The expression of other markers was evaluated by indirect immunofluorescence using polyclonal antibodies respectively specific for type I Establishment of the SARG ceil line and III collagen (Chemicon International, Inc. Temecula, CA, USA) This cell line was established from a human radiation-induced and osteonectin (bON-II, kindly supplied by L. W. Fisher, N.I.H., osteosarcoma in a 24-year-old male with histiocytosis X. The Bethesda, MD, USA). The percentage of positive cells was calcutumour was metastatic at the clinical onset and the patient had no lated out of at least 300 cells. received chemotherapy. HLA expression was determined by flow cytometry (FACS SCAN IV, Becton Dickinson, San Jose, CA, USA) after indirect immunofluorescence with the following monoclonal antibodies: In vitro procedure W6/32 (anti-HLA A,B,C Sera-Lab, Crawley Down, UK); OKDR Cells were routinely cultured in Iscove's Modified Dulbecco's (anti-HLA-DR, Ortho Diagnostic Systems, Milan, Italy); TECMedium (IMDM) supplemented with 100 U/ml penicillin, 100 u.g/ antiDC-1 (anti-HLA-DQ, Technogenetics, Milan, Italy); B7/21 ml streptomycin (GIBCO, Paisley, Scotland) and 10% heat inacti- (anti-HLA-DP, Becton Dickinson, San Jose, CA, USA).
Downloaded from http://annonc.oxfordjournals.org/ at University of Windsor on July 8, 2014
Summary. A new cell line (SARG) was established from a human radiation-induced osteosarcoma (OSA). It showed an epithelial-like morphology with polimorphous and sometimes bizarre nuclei. SARG had an osteoblastic differentiation pattern: almost 100% of the cells were positive for alkaline phosphatase, type I and III collagens and osteonectin. The expression of class I HLA antigens was detectable even after 40 in vitro passages. The expression of MHC antigens was greatly increased after in vitro treatment with interferon
30 In vivo studies Male athimic 4-5 weeks old Crl: nu/nu(CD-l)BR mice (Charles River Italia, Como, Italy) were used. Tumorigenicity was determined after intraperitoneal (i.p.) and subcutaneous (s.c.) injection of 27 x 106 cells. Tumor growth was assessed twice weekly. Mice were sacrified 6 months after injection. Experimental metastatic ability was evaluated after injection of 0.5 x 106 cells in a lateral tail vein (i.v.) of untreated mice and of mice treated with a single intraperitoneal injection of cyclophosphamide (CY) (200 mg/Kg, Endoxan Asta, FRG) 4 days before cell injection in order to decrease NK activity |10] and enhance experimental metastasis [11|. Mice were sacrified 2 months after injection. For the assessment of pulmonary metastases, lungs were stained with black India ink.
Results
Effect of cytokines. The effects of IFN-a, IFN-y and TNF-a on cell growth are shown in Fig. 2. IFN-y induced a greater inhibition of cell growth than IFN-a; TNF-a was unable to inhibit cell proliferation of SARG cells. Fig. 3 shows the effects of the three cytokines on the expression of class I and II HLA antigens. All the cytokines enhanced the expression of class I antigens, but only IFN-y was able to increase class II antigens (HLA-DR). HLA-DP and HLA-DQ were not induced. The ability of TNF-a to increase the expression of class I HLA antigens indicates that these cells have receptors for TNF-a. The osteoblastic differentiation markers (alkaline phosphatase, collagens and osteonectin) were not modulated by IFN-y or TNF-a.
5 IFN
gamma
500
1 ,000
10 IFN
Differentiation pattern. SARG showed the classical markers of the osteoblastic lineage: 100% of the cells were positive for alkaline phosphatase, 75% for type I collagen, 65% for type III collagen and 61% for osteonectin. HLA A,B,C and HLA-DR antigens were detectable respectively in 84% and 68% of cells,
50
alpha
100
1 ,000
10 ,000 TNF
alpha
100
10 ,00C
0
20
40
60
80
100
% INHIBITION OF PROLIFERATION Fig. 2. Inibition of SARG cell growth by IFN-y, IFN-a and TNF-a after a 144 h in vitro treatment.
Table 1. In vitro patterns of SARG cell line. Doubling time (hours)
Saturation density (cells/cm2)
Cloning efficiency
Cloning efficiency in semisolid medium
58 ±4
2.5 x l O 5
2.7 ± 0.3%
0.5 ±0.1%
HLA-A,B,C
HLA-DR
Fig. 1. In vitro morphology of SARG cell line.
Fig. .?. Modulation of A.B.C HLA and HLA-DR antigens expression in SARG cells after a 144 h in vitro treatment with IFN-a. IFN-y and TNF-a. • = Negative control; g9 = SARG cell line: H =lFN-a 10.01)0 U/ml; 0 = TNF-a 10.000 U/ml; B = IFN-y 500 U/ml.
Downloaded from http://annonc.oxfordjournals.org/ at University of Windsor on July 8, 2014
In vitro characterization. All the classical parameters of the in vitro growth were studied between the 20th and the 40th culture passage after 1 year continuous in vitro culture (Table 1). Morphologically, SARG showed a predominance of epithelial-like mononuclear cells, in addition to giant cells with multilobated nuclei (Fig. 1). The cellular diameter was 21.9 ±0.4 \im with a range from 9.24 to 34.65 um. The analysis of DNA content showed a hyperdiploid (near tetraploid) pattern.
whereas no expression of HLA-DP and HLA-DQ antigens was found.
31
In vivo growth and metastasis. SARG was able to grow after subcutaneous injection (27 x 106 cells/mice) even if with a low tumorigenicity and a long latent period (2-4 months) and did not grow in nude mice after intraperitoneal injection (27 x 106 cells/mice). Macroscopic lung metastasis were never observed in untreated mice, whereas a low experimental metastatic ability were found after CY-pretreatment (Table 2).
Table 2. Tumorigenicity and metastatic ability of SARG cells in nude mice. Site of injection
Pretreatment of mice
Cell dose
Mice with tumor/ mice treated
s.c. i.p. i.v. i.v.
None None None CY
27xlO 6 27xlO 6
2/10 0/5
0.5x10" 0.5 xio 6
Incidence of lung metastasis
0/5 2/5
Discussion References
Acknowledgements
This work was supported by grants from Ministero della Sanita-Special project "Identificazione dei parametri biologici di importanza prognostica nei tumori ossei, con particolare riferimento alFosteosarcoma" and from Associazione Italiana per la Ricerca sul Cancro. L. Landuzzi is in receipt of a fellowship from Associazione Italiana per la Ricerca sul Cancro.
1. Ponten J, Saksela E. Two established in vitro cell lines from human mesenchymal tumours. Int J Cancer 1967; 2: 434-7. 2. Fogh J, Trempe G. New human tumour cell lines, chapter 5. In Fogh J (ed): Human tumour cells in vitro. Plenum Press: New York and London 1975; pp. 115-59. 3. McAllister RM, Gardner MB, Greene AE et al. Cultivation in vitro of cells derived from a human osteosarcoma. Cancer 1971; 27: 397-401. 4. Tsang KY, Poi GS, Finderberger HH. Characterization of newly established human osteosarcoma cell line, LM-1. In vitro 1981; 17:308-14. 5. Yamane T. Establishment and characterization of cell lines derived from a human osteosarcoma. Clin Orthop 1985; 199: 261-71. 6. Bogenmann E, Moghadam H, DeClerck YA et al. c-myc amplification and expression in newly established human osteosarcoma cell lines. Cancer Res 1987; 47: 3808-14. 7. Kawai A. A newly established human osteosarcoma cell line with osteoblastic properties. Clin Orthop 1990; 259: 256-67. 8. Brosjo O, Bauer HCF, Brostrom LA et al. Growth inhibition of human osteosarcomas in nude mice by human interferon-a: significance of dose and tumor differentiation. Cancer Res 1987; 47: 258-62. 9. Chen TR. In situ detection of mycoplasma contamination in cell cultures by fluorescent Hoechst 33258 stain. Exp Cell Res 1977; 104: 255-61. 10. Hanna N, Burton RC. Definitive evidence that natural killer (NK) cells inhibit experimental tumor metastasis in vivo. J Immunoll981; 104:255-61. 11. Nanni P, De Giovanni C, Nicoletti G et al. Human rhabdomyosarcoma cells in nude mice as a model for metastasis and differentiation. Invasion Metastasis 1989; 9: 231-41. 12. Skjodt H, Hughes D, Pauline RM et al. Constitutive and inducible expression of HLA class II determinants by human osteoblast-like cells in vitro. J Clin Invest 1990; 85:1421-6. Correspondence to: N. Baldini, M.D. Laboratorio di Ricerca Oncologica Instituto Ortopedico Rizzoli Via di Barbiano 1/10 40136 Bologna, Italy
Downloaded from http://annonc.oxfordjournals.org/ at University of Windsor on July 8, 2014
SARG is a recent cell line established from a human radiation-induced OSA. It showed the typical markers of the osteoblastic lineage (alkaline phosphatase activity, osteonectin, type I and III collagen). In addition to a high expression of HLA A,B,C antigens, SARG showed low but reproducible levels of HLA-DR antigens expression. The presence of class II HLA antigens in normal osteoblasts has been recently reported [12]. The in vitro treatment of SARG with IFN-a and IFN-y induced a marked inhibition of cell growth, whereas TNF-a did not have any effect on cell proliferation. Although differentiation markers (alkaline phosphatase, collagens, and osteonectin) were not modulated by IFN-y and TNF-a in vitro, all three cytokines modulated the expression of class I HLA antigens. Only LFN-y was able to increase HLA class II (HLA-DR) expression. Our data indicate that, in addition to its marked cytostatic effect, IFN-y is able to enhance HLA class I and HLA II antigens expression of OSA cells in vitro. Therefore, it may be considered as an interesting candidate for differentiation therapy of OSA.
Original article SARG: A new human osteosarcoma cell line. Expression of bone markers and of major histocompatibility antigens K. Scotlandi,1 M. Serra,1 L. Landuzzi2 & N. Baldini1 1
Laboratorio di Ricerca Oncologica, Istituto Ortopedico Rizzoli, Bologna, Italy; 2Istituto di Cancerologia, Universitd di Bologna, Bologna, Italy
gamma (IFN-y), whereas interferon alpha (IFN-a) and tumor necrosis factor alpha (TNF-a) increased the expression of class I antigens, but not of class II antigens. SARG was tumorigenic after subcutaneos injection in nude mice. Experimental metastases were never detected.
Introduction
vated Fetal Calf Serum (Biological Industries, Kibbutz Beth Haemek, Israel). Cells were maintained at 37 °C in a humidified 5% CO2 atmosphere. Doubling time and saturation density were determined by daily harvesting of cells after seeding of 20,000 cells/cm2. Cloning efficiency was evaluated 7 days after seeding of 8—512 cells/cm2. Cloning efficiency in semi-solid medium was determined 14 days after seeding 400-4,000 cells/cm2 in 60 mm dishes. Cells were found to be mycoplasma-free by fluorescence staining with Hoechst 33258 [9]. The analysis of DNA content was performed with a static cytofluorometer (Nikon) after staining with DAPI.
Several cell lines have been established from human OSA [1-7], but the phenotypic pattern and the ability to differentiate along the osteoblastic pathway of many OSA cell lines have not been sufficiently characterized [1, 3-5]. Furthermore, most of them have been maintained in vitro for many years; progressive changes have likely occurred during long-term maintenance in vitro, therefore some of human OSA cell lines may have lost the original features. We report a new cell line (SARG) derived from a human radiation-induced OSA. Its recent origin makes it a good model for the study of the effects of cytokines with potential therapeutic use on OSA cells. As IFNs have been suggested to influence OSA differentiation [8], we have investigated the ability of IFNs and TNF-a to modulate the osteoblastic differentiation in vitro and the HLA expression of SARG cells.
Key words: osteosarcoma, MHC antigens, interferons, tumor necrosis factor-alpha
Cytokine treatment Cells were treated in vitro with IFN-a (10-10,000 U/ml), IFN-y (5-1000 U/ml), TNF-a (100-10,000 U/ml). Recombinant HuIFNy and recombinant HuTNF-a were purchased from Boehringer Mannheim Italia SPA, Milan, Italy. Recombinant HuIFN-a was purchased from Hoffman-La Roche, Basel, Switzerland. Cytokines were added to cultures 24 h after seeding; cells were harvested 144 h later. Differentiation markers and HLA expression
Osteoblastic differentiation markers were analyzed on cytospins fixed with a methanol-aceton solution (3:7) at - 2 0 °C for 10 min. The alkaline phosphatase intracellular content was determined by a Materials and methods cytochemical method (Sigma Diagnostic, St. Louis, MO, USA). The expression of other markers was evaluated by indirect immunofluorescence using polyclonal antibodies respectively specific for type I Establishment of the SARG ceil line and III collagen (Chemicon International, Inc. Temecula, CA, USA) This cell line was established from a human radiation-induced and osteonectin (bON-II, kindly supplied by L. W. Fisher, N.I.H., osteosarcoma in a 24-year-old male with histiocytosis X. The Bethesda, MD, USA). The percentage of positive cells was calcutumour was metastatic at the clinical onset and the patient had no lated out of at least 300 cells. received chemotherapy. HLA expression was determined by flow cytometry (FACS SCAN IV, Becton Dickinson, San Jose, CA, USA) after indirect immunofluorescence with the following monoclonal antibodies: In vitro procedure W6/32 (anti-HLA A,B,C Sera-Lab, Crawley Down, UK); OKDR Cells were routinely cultured in Iscove's Modified Dulbecco's (anti-HLA-DR, Ortho Diagnostic Systems, Milan, Italy); TECMedium (IMDM) supplemented with 100 U/ml penicillin, 100 u.g/ antiDC-1 (anti-HLA-DQ, Technogenetics, Milan, Italy); B7/21 ml streptomycin (GIBCO, Paisley, Scotland) and 10% heat inacti- (anti-HLA-DP, Becton Dickinson, San Jose, CA, USA).
Downloaded from http://annonc.oxfordjournals.org/ at University of Windsor on July 8, 2014
Summary. A new cell line (SARG) was established from a human radiation-induced osteosarcoma (OSA). It showed an epithelial-like morphology with polimorphous and sometimes bizarre nuclei. SARG had an osteoblastic differentiation pattern: almost 100% of the cells were positive for alkaline phosphatase, type I and III collagens and osteonectin. The expression of class I HLA antigens was detectable even after 40 in vitro passages. The expression of MHC antigens was greatly increased after in vitro treatment with interferon
30 In vivo studies Male athimic 4-5 weeks old Crl: nu/nu(CD-l)BR mice (Charles River Italia, Como, Italy) were used. Tumorigenicity was determined after intraperitoneal (i.p.) and subcutaneous (s.c.) injection of 27 x 106 cells. Tumor growth was assessed twice weekly. Mice were sacrified 6 months after injection. Experimental metastatic ability was evaluated after injection of 0.5 x 106 cells in a lateral tail vein (i.v.) of untreated mice and of mice treated with a single intraperitoneal injection of cyclophosphamide (CY) (200 mg/Kg, Endoxan Asta, FRG) 4 days before cell injection in order to decrease NK activity |10] and enhance experimental metastasis [11|. Mice were sacrified 2 months after injection. For the assessment of pulmonary metastases, lungs were stained with black India ink.
Results
Effect of cytokines. The effects of IFN-a, IFN-y and TNF-a on cell growth are shown in Fig. 2. IFN-y induced a greater inhibition of cell growth than IFN-a; TNF-a was unable to inhibit cell proliferation of SARG cells. Fig. 3 shows the effects of the three cytokines on the expression of class I and II HLA antigens. All the cytokines enhanced the expression of class I antigens, but only IFN-y was able to increase class II antigens (HLA-DR). HLA-DP and HLA-DQ were not induced. The ability of TNF-a to increase the expression of class I HLA antigens indicates that these cells have receptors for TNF-a. The osteoblastic differentiation markers (alkaline phosphatase, collagens and osteonectin) were not modulated by IFN-y or TNF-a.
5 IFN
gamma
500
1 ,000
10 IFN
Differentiation pattern. SARG showed the classical markers of the osteoblastic lineage: 100% of the cells were positive for alkaline phosphatase, 75% for type I collagen, 65% for type III collagen and 61% for osteonectin. HLA A,B,C and HLA-DR antigens were detectable respectively in 84% and 68% of cells,
50
alpha
100
1 ,000
10 ,000 TNF
alpha
100
10 ,00C
0
20
40
60
80
100
% INHIBITION OF PROLIFERATION Fig. 2. Inibition of SARG cell growth by IFN-y, IFN-a and TNF-a after a 144 h in vitro treatment.
Table 1. In vitro patterns of SARG cell line. Doubling time (hours)
Saturation density (cells/cm2)
Cloning efficiency
Cloning efficiency in semisolid medium
58 ±4
2.5 x l O 5
2.7 ± 0.3%
0.5 ±0.1%
HLA-A,B,C
HLA-DR
Fig. 1. In vitro morphology of SARG cell line.
Fig. .?. Modulation of A.B.C HLA and HLA-DR antigens expression in SARG cells after a 144 h in vitro treatment with IFN-a. IFN-y and TNF-a. • = Negative control; g9 = SARG cell line: H =lFN-a 10.01)0 U/ml; 0 = TNF-a 10.000 U/ml; B = IFN-y 500 U/ml.
Downloaded from http://annonc.oxfordjournals.org/ at University of Windsor on July 8, 2014
In vitro characterization. All the classical parameters of the in vitro growth were studied between the 20th and the 40th culture passage after 1 year continuous in vitro culture (Table 1). Morphologically, SARG showed a predominance of epithelial-like mononuclear cells, in addition to giant cells with multilobated nuclei (Fig. 1). The cellular diameter was 21.9 ±0.4 \im with a range from 9.24 to 34.65 um. The analysis of DNA content showed a hyperdiploid (near tetraploid) pattern.
whereas no expression of HLA-DP and HLA-DQ antigens was found.
31
In vivo growth and metastasis. SARG was able to grow after subcutaneous injection (27 x 106 cells/mice) even if with a low tumorigenicity and a long latent period (2-4 months) and did not grow in nude mice after intraperitoneal injection (27 x 106 cells/mice). Macroscopic lung metastasis were never observed in untreated mice, whereas a low experimental metastatic ability were found after CY-pretreatment (Table 2).
Table 2. Tumorigenicity and metastatic ability of SARG cells in nude mice. Site of injection
Pretreatment of mice
Cell dose
Mice with tumor/ mice treated
s.c. i.p. i.v. i.v.
None None None CY
27xlO 6 27xlO 6
2/10 0/5
0.5x10" 0.5 xio 6
Incidence of lung metastasis
0/5 2/5
Discussion References
Acknowledgements
This work was supported by grants from Ministero della Sanita-Special project "Identificazione dei parametri biologici di importanza prognostica nei tumori ossei, con particolare riferimento alFosteosarcoma" and from Associazione Italiana per la Ricerca sul Cancro. L. Landuzzi is in receipt of a fellowship from Associazione Italiana per la Ricerca sul Cancro.
1. Ponten J, Saksela E. Two established in vitro cell lines from human mesenchymal tumours. Int J Cancer 1967; 2: 434-7. 2. Fogh J, Trempe G. New human tumour cell lines, chapter 5. In Fogh J (ed): Human tumour cells in vitro. Plenum Press: New York and London 1975; pp. 115-59. 3. McAllister RM, Gardner MB, Greene AE et al. Cultivation in vitro of cells derived from a human osteosarcoma. Cancer 1971; 27: 397-401. 4. Tsang KY, Poi GS, Finderberger HH. Characterization of newly established human osteosarcoma cell line, LM-1. In vitro 1981; 17:308-14. 5. Yamane T. Establishment and characterization of cell lines derived from a human osteosarcoma. Clin Orthop 1985; 199: 261-71. 6. Bogenmann E, Moghadam H, DeClerck YA et al. c-myc amplification and expression in newly established human osteosarcoma cell lines. Cancer Res 1987; 47: 3808-14. 7. Kawai A. A newly established human osteosarcoma cell line with osteoblastic properties. Clin Orthop 1990; 259: 256-67. 8. Brosjo O, Bauer HCF, Brostrom LA et al. Growth inhibition of human osteosarcomas in nude mice by human interferon-a: significance of dose and tumor differentiation. Cancer Res 1987; 47: 258-62. 9. Chen TR. In situ detection of mycoplasma contamination in cell cultures by fluorescent Hoechst 33258 stain. Exp Cell Res 1977; 104: 255-61. 10. Hanna N, Burton RC. Definitive evidence that natural killer (NK) cells inhibit experimental tumor metastasis in vivo. J Immunoll981; 104:255-61. 11. Nanni P, De Giovanni C, Nicoletti G et al. Human rhabdomyosarcoma cells in nude mice as a model for metastasis and differentiation. Invasion Metastasis 1989; 9: 231-41. 12. Skjodt H, Hughes D, Pauline RM et al. Constitutive and inducible expression of HLA class II determinants by human osteoblast-like cells in vitro. J Clin Invest 1990; 85:1421-6. Correspondence to: N. Baldini, M.D. Laboratorio di Ricerca Oncologica Instituto Ortopedico Rizzoli Via di Barbiano 1/10 40136 Bologna, Italy
Downloaded from http://annonc.oxfordjournals.org/ at University of Windsor on July 8, 2014
SARG is a recent cell line established from a human radiation-induced OSA. It showed the typical markers of the osteoblastic lineage (alkaline phosphatase activity, osteonectin, type I and III collagen). In addition to a high expression of HLA A,B,C antigens, SARG showed low but reproducible levels of HLA-DR antigens expression. The presence of class II HLA antigens in normal osteoblasts has been recently reported [12]. The in vitro treatment of SARG with IFN-a and IFN-y induced a marked inhibition of cell growth, whereas TNF-a did not have any effect on cell proliferation. Although differentiation markers (alkaline phosphatase, collagens, and osteonectin) were not modulated by IFN-y and TNF-a in vitro, all three cytokines modulated the expression of class I HLA antigens. Only LFN-y was able to increase HLA class II (HLA-DR) expression. Our data indicate that, in addition to its marked cytostatic effect, IFN-y is able to enhance HLA class I and HLA II antigens expression of OSA cells in vitro. Therefore, it may be considered as an interesting candidate for differentiation therapy of OSA.