Portal de Periódicos da CAPES

Effects of Tumor Necrosis Factor-α on Antimitogenicity and Cell Cycle-related Proteins in MCF-7 Cells

1995; Elsevier BV; Volume: 270; Issue: 31 Linguagem: Inglês

10.1074/jbc.270.31.18367

1083-351X

Dooil Jeoung, Baiqing Tang, Martin Sonenberg,

Cytokine Signaling Pathways and Interactions

Tumor necrosis factor-α (TNF-α) demonstrated antimitogenic activity in MCF-7 cells (estrogen receptor-positive human breast cancer cells) in a dose- and time-dependent manner (EC-50 of 2.5 ng/ml). This antimitogenic effect of TNF-α was accompanied by a decreased number of cells in S phase in a dose- and time-dependent manner. Based on growth arrest experiments using aphidicolin, it is apparent that TNF-α acted in early G1 phase. It did not show antimitogenic effects once cells reentered the S phase based on [3H]thymidine incorporation into DNA and cell cycle analysis. Specificity of TNF-α was established by using monoclonal anti-human TNF-α antibody. On the basis of Western immunoblot analysis of Rb, p53 and cell cycle inhibitory protein (Cip1) (p21) proteins, TNF-α decreased Rb protein expression in a dose- and time-dependent manner whereas it increased the expression level of tumor suppressor p53 protein. TNF-α also increased the expression level of Cip1 (p21) protein in a dose-dependent manner. This induction of Cip1 (p21) protein was preceded by the induction of p53 protein in MCF-7 cells. Cip1 (p21) protein associated with cyclin D was also increased. Tumor suppressor Rb protein expression was increased during G1 to S phase progression. Cyclin D protein expression levels were not changed in response to TNF-α treatment, although serine/threonine kinase inhibitors such as H7 and the protein kinase C inhibitor staurosporine decreased cyclin D expression levels in MCF-7 cells. Based on experiments with staurosporine, it appears that TNF-α does not utilize a protein kinase C pathway in MCF-7 cells. Other cell cycle-related proteins such as Cdk2, Cdc2, and Cdk4 did not show any change in response to TNF-α. TNF-α did not affect complexes between cyclin D and Cdk2, Cdk4, and Rb proteins in MCF-7 cells. Taken together these results suggest that Rb, p53, and Cip1 (p21) proteins mediate TNF-α antimitogenic activity, and TNF-α induces growth arrest in the G1 phase in MCF-7 cells. Tumor necrosis factor-α (TNF-α) demonstrated antimitogenic activity in MCF-7 cells (estrogen receptor-positive human breast cancer cells) in a dose- and time-dependent manner (EC-50 of 2.5 ng/ml). This antimitogenic effect of TNF-α was accompanied by a decreased number of cells in S phase in a dose- and time-dependent manner. Based on growth arrest experiments using aphidicolin, it is apparent that TNF-α acted in early G1 phase. It did not show antimitogenic effects once cells reentered the S phase based on [3H]thymidine incorporation into DNA and cell cycle analysis. Specificity of TNF-α was established by using monoclonal anti-human TNF-α antibody. On the basis of Western immunoblot analysis of Rb, p53 and cell cycle inhibitory protein (Cip1) (p21) proteins, TNF-α decreased Rb protein expression in a dose- and time-dependent manner whereas it increased the expression level of tumor suppressor p53 protein. TNF-α also increased the expression level of Cip1 (p21) protein in a dose-dependent manner. This induction of Cip1 (p21) protein was preceded by the induction of p53 protein in MCF-7 cells. Cip1 (p21) protein associated with cyclin D was also increased. Tumor suppressor Rb protein expression was increased during G1 to S phase progression. Cyclin D protein expression levels were not changed in response to TNF-α treatment, although serine/threonine kinase inhibitors such as H7 and the protein kinase C inhibitor staurosporine decreased cyclin D expression levels in MCF-7 cells. Based on experiments with staurosporine, it appears that TNF-α does not utilize a protein kinase C pathway in MCF-7 cells. Other cell cycle-related proteins such as Cdk2, Cdc2, and Cdk4 did not show any change in response to TNF-α. TNF-α did not affect complexes between cyclin D and Cdk2, Cdk4, and Rb proteins in MCF-7 cells. Taken together these results suggest that Rb, p53, and Cip1 (p21) proteins mediate TNF-α antimitogenic activity, and TNF-α induces growth arrest in the G1 phase in MCF-7 cells. INTRODUCTIONTumor necrosis factor-α (TNF-α) ( 1The abbreviations used are: TNF-αtumor necrosis factor-αCip1cell cycle inhibitory proteinDMEMDulbecco's modified Eagle's mediumFACSfluorescence-activated cell sorterFCSfetal calf serumH71-(5-isoquinolinylsulfonyl)-2-methylpiperazineH8N-(2-methylaminoethyl)-5-isoquinolinesulfoneamideHBSSHank's balanced salt solutionPAGEpolyacrylamide gel electrophoresisTBSTris-buffered salineTGF-βtransforming growth factor-βSFMserum-free medium. )is a macrophage-derived multi-functional cytokine that acts as a cytostatic or cytotoxic agent in many transformed and normal cells (1Danforth Jr., D.N. Sgagias M.K. J. Endocrinol. 1993; 138: 517-528Crossref PubMed Scopus (24) Google Scholar, 2Rosen O.M. Krebs E.G. Cell Prolif. 1981; 8: 715-808Google Scholar, 3Ruggiero V. Latham K. Baglioni C. J. Immunol. 1987; 138: 2711-2717PubMed Google Scholar, 4Sgagias M.K. Kasid A. Danforth Jr., D.N. Mol. Endocrinol. 1991; 5: 1740-1747Crossref PubMed Scopus (41) Google Scholar, 5Sugarman B.J. Aggarwal B.B. Hass P.E. Figari I.S. Palladino M.A. Shepard H.M. Science (Wash. D. C.). 1985; 230: 943-945Crossref PubMed Scopus (1359) Google Scholar). There have been reports of functional roles of TNF-α such as phosphorylation of epidermal growth factor receptor (6Donato N.J. Gallick G.E. Steck P.A. Rosenblum M.G. J. Biol. Chem. 1989; 264: 20474-20481Abstract Full Text PDF PubMed Google Scholar, 7Donato N.J. Rosenblum M. Steck P.A. Cell Growth Diff. 1992; 3: 259-268PubMed Google Scholar) and cellular molecules, notably hsp27 and hsp70 protein(8Alberts A.S. Thorburn A.M. Shenolikar S. Mumby M.C. Feramisco J.R. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 388-392Crossref PubMed Scopus (141) Google Scholar, 9Arrigo A.-P. Mol. Cell. Biol. 1990; 10: 1276-1280Crossref PubMed Scopus (105) Google Scholar, 10Bird T.A. Saklatvala J. J. Biol. Chem. 1990; 265: 235-240Abstract Full Text PDF PubMed Google Scholar, 11Hepburn A. Demolle D. Boeynaems J.-M. Fiers W. Dumont J.E. FEBS Lett. 1988; 227: 175-178Crossref PubMed Scopus (40) Google Scholar, 12Jaattela M. Wissing D. Bauer P.A. Li G. EMBO J. 1993; 11: 3507-3512Crossref Scopus (354) Google Scholar, 13Minshull J. Golsteyn R. Hill C. Hunt T. EMBO J. 1990; 9: 2865-2875Crossref PubMed Scopus (255) Google Scholar, 14Kaur P. Saklatvala J. FEBS Lett. 1988; 241: 6-10Crossref PubMed Scopus (59) Google Scholar, 15Schutze S. Scheurich P. Pfizenmaier K. Kronke M. J. Biol. Chem. 1989; 264: 3562-3567Abstract Full Text PDF PubMed Google Scholar, 16Marino M.W. Pfeffer L.M. Guidon Jr., P.T. Donner D.B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 8417-8421Crossref PubMed Scopus (42) Google Scholar). Overexpression of hsp27 and hsp70 proteins protect cells against the cytotoxic effects of TNF-α(12Jaattela M. Wissing D. Bauer P.A. Li G. EMBO J. 1993; 11: 3507-3512Crossref Scopus (354) Google Scholar, 17Chretien P. Landry J. J. Cell. Physiol. 1988; 137: 157-166Crossref PubMed Scopus (109) Google Scholar, 18Landry J.P. Chretien P. Lambert H. Hickey E. Weber L.A. J. Cell Biol. 1989; 109: 7-15Crossref PubMed Scopus (579) Google Scholar). It also induces phosphorylation of NF-KB (19Duh E.J. Maury W.J. Folks T.M. Farci A.S. Robson A.B. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 5974-5978Crossref PubMed Scopus (631) Google Scholar) and the eukaryotic initiation factor EIF 4E(20Marino M.W. Fels L.J. Jaffe E.A. Pfeffer L.M. Han H.-M. Donner D.B. J. Biol. Chem. 1991; 266: 2685-2688Abstract Full Text PDF PubMed Google Scholar). The cellular responses to TNF-α are necrotic or apoptotic killing, DNA fragmentation, changes in arachidonic acid metabolism, lipid peroxidation, inhibition of mitochondrial electron transfer(21Lancaster Jr., J.R. Lancaster S.M. Gooding L.R. FEBS Lett. 1989; 248: 169-174Crossref PubMed Scopus (131) Google Scholar), and increased production of prostaglandin E(22Baud L. Perez J. Friedlander G. Ardailloud R. FEBS Lett. 1989; 239: 50-54Crossref Scopus (88) Google Scholar). Although TNF-α is a well known antimitogenic agent, the mechanisms explaining these pleiotropic effects have not been well understood.Rb protein is known to be cell cycle-regulated(23Buchkovich K. Duffy L.A. Harlow E. Cell. 1989; 58: 1097-1105Abstract Full Text PDF PubMed Scopus (795) Google Scholar, 24Chen P.-L. Scully D.P. Shew J.-Y. Wang J.Y. Lee W.-H. Cell. 1989; 58: 1193-1198Abstract Full Text PDF PubMed Scopus (790) Google Scholar, 25DeCaprio J.A. Ludlow J.W. Lynch D.F. Furukawa Y. Griffin J. Piwnica-Worms H. Huang C.-M. Livingston D.M. Cell. 1989; 58: 1085-1095Abstract Full Text PDF PubMed Scopus (688) Google Scholar, 26Mihara K. Cao X.-R. Yen A. Chandler S. Driscoll B. Murphree A.L. Tang A. Fung Y.K. Science. 1989; 246: 1300-1303Crossref PubMed Scopus (429) Google Scholar, 27Weinberg R. Science. 1991; 254: 1138-1146Crossref PubMed Scopus (1352) Google Scholar, 28Green M.R. Cell. 1989; 56: 1-3Abstract Full Text PDF PubMed Scopus (142) Google Scholar). This protein has tumor suppressing activity when the wild type Rb gene is transfected into transformed mammalian cells, which do not express functional Rb protein. In MCF-7 cells, as well as in osteocarcomas, Rb protein was shown to be abnormally expressed(29Eva Y.-H. Lee P. To H. Shew J.-Y. Bookstein R. Scully P. Lee W.-H. Science. 1988; 241: 218-221Crossref PubMed Scopus (496) Google Scholar, 30Friend S.H. Horowitz J.M. Gerber M.R. Fang X.-F. Bogenmann E. Li F.P. Weinberg R.A. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 9059-9063Crossref PubMed Scopus (319) Google Scholar). The hypophosphorylated form of Rb protein acts as a functional tumor suppressor(30Friend S.H. Horowitz J.M. Gerber M.R. Fang X.-F. Bogenmann E. Li F.P. Weinberg R.A. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 9059-9063Crossref PubMed Scopus (319) Google Scholar). This hypophosphorylated Rb protein is associated with oncoproteins such as SV40 T antigen and adenovirus E1A protein, indicating their involvement in cellular transformation(31Ludlow J.W. DeCaprio J.A. Huang C.M. Lee W.H. Paucha E. Livingston D. Cell. 1989; 56: 57-65Abstract Full Text PDF PubMed Scopus (368) Google Scholar). Phosphorylation of Rb protein is dependent on the growth state of cells (23Buchkovich K. Duffy L.A. Harlow E. Cell. 1989; 58: 1097-1105Abstract Full Text PDF PubMed Scopus (795) Google Scholar, 26Mihara K. Cao X.-R. Yen A. Chandler S. Driscoll B. Murphree A.L. Tang A. Fung Y.K. Science. 1989; 246: 1300-1303Crossref PubMed Scopus (429) Google Scholar), and antimitogenic agents or mitogenic agents exert their effects by changing the phosphorylation state of Rb protein(32Kumar R. Atlas I. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6599-6603Crossref PubMed Scopus (92) Google Scholar).As with Rb protein, cyclins are known to be cell cycle-related (33Ewen M.E. Sluss H.K. Sherr C.J. Matsushime H. Kato J.Y. Livingston D.M. Cell. 1993; 73: 487-497Abstract Full Text PDF PubMed Scopus (915) Google Scholar, 34Kato J.-Y. Matsushime H. Hiebert S.W. Ewen M.E. Sherr C.J. Genes & Dev. 1993; 7: 331-342Crossref PubMed Scopus (1086) Google Scholar, 35Matsushime H. Roussel M.F. Ashmun R.A. Sherr C.J. Cell. 1991; 65: 701-713Abstract Full Text PDF PubMed Scopus (988) Google Scholar, 36Motokura T. Bloom T. Kim H.G. Juppner H. Luderman J.V. Kronenberg H.M. Arnold A. Nature. 1991; 350: 512-515Crossref PubMed Scopus (1153) Google Scholar, 37Pines J. Hunter T. Nature. 1990; 34: 760-763Crossref Scopus (528) Google Scholar, 38Pines J. Hunter T. J. Cell Biol. 1991; 115: 1-17Crossref PubMed Scopus (673) Google Scholar, 39Won K.-A. Xiong Y. Beach D. Gilman M. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 9910-9914Crossref PubMed Scopus (267) Google Scholar). Levels of cyclin-associated histone H1 kinase activities are also cell cycle-regulated(35Matsushime H. Roussel M.F. Ashmun R.A. Sherr C.J. Cell. 1991; 65: 701-713Abstract Full Text PDF PubMed Scopus (988) Google Scholar, 37Pines J. Hunter T. Nature. 1990; 34: 760-763Crossref Scopus (528) Google Scholar, 39Won K.-A. Xiong Y. Beach D. Gilman M. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 9910-9914Crossref PubMed Scopus (267) Google Scholar, 40Baldin V.J. Likas M.J. Marcote M.P. Bartek J. Draetta G. Genes & Dev. 1993; 7: 812-821Crossref PubMed Scopus (1423) Google Scholar). Some of these cyclins are known to have oncogenic activities associated with Rb activity, and a shortened G1 phase(41Quelle D.E. Ashmun R.A. Shurtleff S.A. Kato J.-Y. Bar-Sagi D. Roussel M.F. Sherr C.J. Genes & Dev. 1993; 7: 1559-1571Crossref PubMed Scopus (979) Google Scholar). These cyclins are associated with other cell cycle-regulated proteins such as Cdc2 and Cdk2 in many mammalian cells(42Tam S.W. Theodoras A.M. Shay J.W. Draetta G.F. Pagano M. Oncogene. 1994; 9: 2663-2674PubMed Google Scholar). Complexes between Rb and these cyclins are assumed to regulate Rb protein functions in many mammalian cells(33Ewen M.E. Sluss H.K. Sherr C.J. Matsushime H. Kato J.Y. Livingston D.M. Cell. 1993; 73: 487-497Abstract Full Text PDF PubMed Scopus (915) Google Scholar, 43Akiyama T. Ohuchi T. Sumida S. Matsumoto K. Toyoshima K. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 7900-7904Crossref PubMed Scopus (194) Google Scholar, 44Dowdy S.F. Hinds P.W. Louis K. Reed S.I. Arnold A. Weinberg R.A. Cell. 1993; 73: 499-511Abstract Full Text PDF PubMed Scopus (688) Google Scholar, 45Hall F.L. Williams R.T. Wu L. Wu F. Carbonaro-Hall D.A. Harper J.W. Warburton D. Oncogene. 1993; 8: 1377-1384PubMed Google Scholar, 46Weintraub S.J. Prater C.A. Dean D.C. Nature. 1992; 358: 259-261Crossref PubMed Scopus (558) Google Scholar).Cdc2 and Cdk2 proteins are known to phosphorylate Rb protein both in vitro and in vivo(47Hu Q.-J. Lees J.A. Buchkovich K.J. Harlow E. Mol. Cell. Biol. 1992; 2: 971-980Crossref Scopus (80) Google Scholar, 48Lees J.A. Buchkovich K.J. Marshak D.R. Anderson C.W. Harlow E. EMBO J. 1991; 10: 4279-4290Crossref PubMed Scopus (268) Google Scholar). This phosphorylation of Rb protein is believed to be achieved by physical association of these proteins and cyclins.Tumor suppressor cell cycle inhibitory protein (Cip1) (p21) was found to be a potent inhibitor of cyclin-dependent kinases(50Kastan M.B. Onyekwere O. Didransky D. Vogelstein B. Craig R.W. Cancer Res. 1991; 51: 6304-6311PubMed Google Scholar). It is known to be associated with various cyclin-Cdk complexes. The main role of Cip1 (p21) protein is to inhibit cyclin-dependent kinase activities. Cip1 (p21) protein is known to be induced in mammalian cells undergoing G1 arrest or apoptosis(51El-Deiry W.S. Tokino T. Velculescu V.E. Levy D.B. Parsons R. Trent J.M. Lin D. Mercer W.E. Kinzler K.W. Vogelstein B. Cell. 1993; 75: 817-825Abstract Full Text PDF PubMed Scopus (7890) Google Scholar). G1 arrest is linked to p53 gene induction, and it is believed that induction of p53 in turn induces expression of Cip1 (p21) protein. The fact that oncogenes such as SV40 T antigen inhibits induction of Cip1 (p21) protein suggests that Cip1 (p21) protein might act as an anti-oncogene.We investigated the antimitogenic effects of TNF-α in MCF-7 cells as determined by expression levels of cell cycle-related proteins such as Rb, Cip1 (p21), p53, and cyclins to extend our understanding of the mechanisms of TNF-α action in MCF-7 cells. We found by cell cycle analysis that TNF-α acted during the G1 phase of the cell cycle. This was associated with a decrease in the expression of Rb and increased expressions of Cip1 (p21) and p53. There was no effect on cyclin D1, Cdk2, or Cdk4 proteins.MATERIALS AND METHODSCell Culture and ReagentsMCF-7 (estrogen receptor-positive human breast cancer) cells were cultured in a humidified air atmosphere (5% CO2) and Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FCS. Human recombinant TNF-α was purchased from Sigma. Monoclonal antibodies to human cyclin A, cyclin B, Rb protein, and TNF-α were purchased from Upstate Biotechnology, Inc. (UBI, Lake Placid, NY). Polyclonal antibodies to human Cdc2, Cdk2, Cdk4, and cyclin D were also purchased from UBI. Polyclonal rabbit anti-human Cip1 (p21) Ab was purchased from Santa Cruz Biotech (Santa Cruz, CA). Monoclonal anti-human Rb antibody was purchased from Pharmingen (San Diego, CA). Aphidicolin and propidium iodide were purchased from Sigma and RNase A (Bovine) was purchased from Boehringer Mannheim. ECL (enhanced chemiluminescence) kit was purchased from Amersham Corp. [3H]Thymidine was purchased from DuPont NEN. H7 and H8 were purchased from Sigma. Staurosporine was purchased from UBI.[3H]Thymidine Incorporation AssayMCF-7 cells (2 × 104/well) were incubated with TNF-α for various time intervals. [3H]Thymidine (20 μCi/ml) was added to each well for 1 h, washed with 1 × HBSS (Hanks' buffered salt solution) three times, and cell lysates were prepared with 0.5% (w/v) SDS solution. To the latter was added 20% trichloroacetic acid (v/v) solution and incubation continued for 30 min on ice. Trichloroacetic acid-precipitated samples were washed with 20% trichloroacetic acid (v/v), 10% trichloroacetic acid (v/v), and 90% ethanol (v/v) successively, and filters were dried. Radioactivity was determined in a scintillation counter. [3H]Thymidine incorporation data were normalized with cell counts and protein concentration of each sample.Cell Cycle AnalysisMCF-7 cells treated with various concentrations of TNF-α were trypsinized. Cell pellets were resuspended in 1 ml of HBSS buffer, and were fixed by HBSS buffer containing 70% (v/v) ethanol for 1 h at 4°C. Cells were centrifuged at 1,000 rpm for 10 min. Cell pellets were resuspended in HBSS buffer containing 50 μg/ml RNase A and 50 μg/ml propidium iodide. Incubation continued for 1 h at room temperature. Cells were filtered through nylon mesh (41 μm) and DNA content was measured in an Epics II flow cytometer. Propidium iodide staining does not distinguish G0 cells from G1.Western Blot AnalysisCell lysates were prepared from MCF-7 cells treated with TNF-α by extraction buffer (20 mM Hepes (pH 7.2), 1% Triton X-100 (v/v), 10% glycerol (v/v), 2 mM sodium fluoride, 1 mM sodium orthovanadate, 50 μg/ml leupeptin, and 0.5 mM phenylmethylsulfonyl fluoride). Cell lysates were cleared by centrifugation at 15,000 rpm for 30 min in a microcentrifuge and were loaded for SDS-PAGE. Samples were transferred to nitrocellulose filters by electroblot transfer for 2 h at 200 mA. The nitrocellulose filter was incubated with blocking buffer (1 × TBS (Tris-buffered saline), Fraction V 3% BSA (w/v), 0.2% (v/v) Tween 20) for 1 h at room temperature and washed with 1 × TBS buffer for 20 min. The nitrocellulose filter was incubated with primary antibody and dissolved in blocking buffer overnight at 4°C (anti-human Rb Ab was used at a concentration of 1 μg/ml). On the following day the nitrocellulose filter was washed with 1 × TBS for 20 min and incubated with either anti-mouse or anti-rabbit horseradish peroxidase conjugated with IgG for 1 h at room temperature. Secondary antibody was at a concentration of 1:1000 dilution. After washing with 1 × TBS for 20 min at room temperature, detection of proteins of interest was carried out by the ECL method.For immunoprecipitation, confluent MCF-7 cells (106/100-mm dish) were lysed and cell lysates were immunoprecipitated with polyclonal anti-human cyclin D Ab conjugated with protein A-Sepharose. Reaction was carried out at 4°C for 4 h on a rotary shaker. Immune complexes were washed three times with lysis buffer, and 2 × sample buffer was added to the beads. Boiled samples were loaded for SDS-PAGE. Western immunoblot analyses were followed according to standard procedures.RESULTSAntimitogenicity of TNF-α in MCF-7 CellsTo determine whether TNF-α has antimitogenic effects on MCF-7 cells, various concentrations of TNF-α were added to MCF-7 cells for 24 h at 37°C and [3H]thymidine incorporation into DNA was carried out. The antimitogenic effect of TNF-α was dose-related as shown in Fig. 1. TNF-α did not show any effect at low concentrations but showed clear antimitogenic effects at concentrations exceeding 0.5 ng/ml. TNF-α did not show mitogenic activity at any concentration. The antimitogenic effect was also seen in MCF-7 cells grown under SFM conditions, suggesting that the antimitogenic effect is due not to serum factors but to direct antimitogenic effects of TNF-α on MCF-7 cells. To determine the specificity of TNF-α, a blocking experiment using monoclonal anti-human TNF-α Ab was carried out. Briefly, TNF-α (10 ng/ml) was preincubated with various concentrations of monoclonal anti-human TNF-α Ab at 37°C for 1 h before addition to each well. The next day [3H]thymidine incorporation into DNA was carried out. As shown in Table 1, the antimitogenic effect of TNF-α was efficiently blocked by an excess concentration of monoclonal anti-human TNF-α Ab while antibody itself did not show any effect on MCF-7 cell growth. Table 2 shows that this inhibition of TNF-α action is consistent with the cell cycle analysis of MCF-7 cells treated with TNF-α (10 ng/ml) alone or in combination with anti-TNF-α Ab (500 ng/ml).Tabled 1 Open table in a new tab Tabled 1 Open table in a new tab To determine whether antimitogenicity of TNF-α is cell cycle-related, FACS (fluorescence-activated cell sorter) analysis was carried out. As shown in Fig. 2A, TNF-α changed the percentage of cells in S phase. The decrease in the fraction of cells in S phase was dose-dependent, as evidenced by larger changes in response to higher concentrations of TNF-α. Changes in cell cycle phases were also time-dependent (data not shown). These results suggest that TNF-α might prevent cells from entering S phase. Based on these results, it is likely that TNF-α acts at the G1/S phase boundary. To prove this point, MCF-7 cells were growth-arrested at the G1/S boundary by aphidicolin treatment (5 μg/ml) for 24 h, washed with 1 × HBSS buffer, and refed with 10% FCS/DMEM. At each interval after release from aphidicolin growth arrest, TNF-α (10 ng/ml) was added. MCF-7 cells were harvested for cell cycle analysis, and cell lysates for [3H]thymidine incorporation into DNA. As shown in Fig. 2B, TNF-α added at 0, 3, 6, and 12 h after release from aphidicolin treatment showed antimitogenic effects, but not at 20 h. This result suggests that TNF-α acts in the G1 phase but not after cells reenter the S phase. This suggests that TNF-α does not offset the mitogenic stimulation of serum. The antimitogenic effect of TNF-α was more apparent in a synchronized population of MCF-7 cells. To determine whether the blocking effect of TNF-α by antibody is also accompanied by changes in cell cycle phases, growth arrest experiments using aphidicolin were carried out as before. As shown in Table 1, monoclonal anti-human TNF-α Ab itself did not affect cellular growth, suggesting that endogenous TNF-α did not play a role in MCF-7 cell growth. TNF-α added at 0, 3, 6, and 10 h but not 20 h showed antimitogenic effects. When TNF-α was added along with monoclonal TNF-α Ab, the latter efficiently inhibited the antimitogenic effect of TNF-α.Figure 2:A, temporal profile of TNF-α action in MCF-7 cells. MCF-7 cells (106 cells/100-mm dish) were treated with various concentrations of TNF-α for 16 h. Cell cycle analysis was carried out according to standard procedures. Columns a-f refer to observations made at different concentrations of TNF-α of 0, 1, 5, 10, 50, and 100 ng/ml. B, temporal profile of TNF-α action in MCF-7 cells. MCF-7 cells were growth-arrested by aphidicolin treatment (5 μg/ml) for 24 h, washed with 1 × HBSS buffer extensively, and refed with 10% FCS/DMEM. TNF-α (10 ng/ml) was added at the indicated times. FACS analysis was carried out according to standard procedures.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Effects of TNF-α on Cell Cycle-related ProteinsBased on the previous results of the antimitogenic effects of TNF-α, it was necessary to determine whether TNF-α affects expression levels of cell cycle-regulated proteins. For this purpose, Western blot analysis using monoclonal anti-human Rb Ab was carried out. MCF-7 cells were treated with aphidicolin for 16 h, washed with 1 × HBSS buffer, and incubated with 10% FCS/DMEM. TNF-α (10 ng/ml) was added at each time point, and cell lysates were prepared according to standard procedures. As shown in Fig. 3, TNF-α added at 0, 3, 6, 12, and 20 h after release showed decreased expression levels of Rb proteins. This result suggests that the decreased DNA synthesis seen in Fig. 1 is consistent with decreased expression levels of Rb protein in response to TNF-α. This result also suggests that Rb protein is important for G1/S transition in the cell cycle. Based on these results, it is plausible that TNF-α acts, in part, by decreasing Rb tumor suppressor protein in MCF-7 cells.Figure 3:TNF-α acts at G1 phase to decrease Rb expression. Normally growing MCF-7 cells (106/100-mm dish) were growth-arrested by incubation with aphidicolin (5 μg/ml) for 16 h, washed with 1 × HBSS buffer three times, and refed with 10% FCS/DMEM. TNF-α was subsequently added at the indicated times after aphidicolin release and harvesting at 24 h. Thus cells were exposed to TNF-α for 24, 21, 18, 12, and 4 h, respectively. Western blot using monoclonal anti-human Rb Ab (1 μg/ml) was carried out according to standard procedures.View Large Image Figure ViewerDownload Hi-res image Download (PPT)TGF-β, a well known antimitogenic agent in many other mammalian cells, did not manifest any antimitogenic activity or induce decreased Rb protein levels in MCF-7 cells (data not shown). Therefore, it is likely that TNF-α activity is not mediated by TGF-β, as was the case with interferon-α in human Burkitt lymphoma Daudi cells(32Kumar R. Atlas I. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6599-6603Crossref PubMed Scopus (92) Google Scholar). G1 cyclins such as cyclin D did not show any change in expression levels in response to TNF α in MCF-7 cells. Cdk4 protein did not respond to TNF-α under these conditions. To determine whether cyclin D protein reflects the growth state of the MCF-7 cells, normally proliferating MCF-7 cells were treated with TNF-α (10 and 25 ng/ml), Bt2cAMP (1 mM), or staurosporine (1 nM) for 24 h and Western immunoblot analysis using polyclonal anti-human cyclin D Ab (1 μg/ml) was employed. As seen in Fig. 4, there was no significant reduction in the expression level of cyclin D in response to TNF-α. However, in response to H7 as well as staurosporine, cyclin D expression levels were significantly reduced. H7 and staurosporine were found to be strong antimitogenic agents in MCF-7 cells (data not shown). Cdk4 protein levels remained constant regardless of the treatments.Figure 4:Growth-regulated expression of cyclin D1 in MCF-7 cells. Normally proliferating MCF-7 cells (106/100-mm dish) were treated with various concentrations of TNF-α (10, 25 ng/ml), Bt2cAMP (1 mM), and staurosporine (1 nM) for 24 h, and cell lysates were prepared for Western immunoblot analysis. Polyclonal anti-human cyclin D Ab was used at a concentration of 1 μg/ml.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig. 5 shows that TNF-α-mediated regulation of Rb protein expression is specific in nature, as preincubation of TNF-α with monoclonal anti-human TNF-α Ab prevented TNF-α antimitogenicity in MCF-7 cells (lanes 3, 5, 7, 9, and 11). This result suggests that endogenous TNF-α does not regulate Rb protein expression. This figure also shows that TNF-α did not have an effect on Rb protein expression once cells enter S phase (lane 13), suggesting an important role of Rb protein in G1/S transition in the cell cycle of MCF-7 cells.Figure 5:Specificity of TNF-α activity in MCF-7 cells. Normally proliferating MCF-7 (106/100-mm dish) cells were treated with aphidicolin (5 μg/ml) for 16 h, washed with 1 × HBSS buffer, and refed with 10% FCS/DMEM. TNF-α was added at the indicated times. For the blocking experiment, TNF-α (10 ng/ml) was preincubated with monoclonal anti-human TNF-α Ab (500 ng/ml) for 1 h at 37°C before adding to cells. Western immunoblot analysis using monoclonal anti-human Rb Ab (1 μg/ml) was carried out according to standard procedures. Detection of Rb protein was done by the ECL method. Columns 1-13 correspond to the experimental conditions indicated below the numbers.View Large Image Figure ViewerDownload Hi-res image Download (PPT)To determine which cellular molecules are important for cell cycle progression, cell synchronization studies were carried out. Normally proliferating MCF-7 cells (106/100-mm dish) were treated with aphidicolin (5 μg/ml) for 12 h, washed with 1 × HBSS buffer extensively and refed with 10% FCS/DMEM. At each time after release from growth arrest by aphidicolin, cells and cell lysates were harvested for Western blot and FACS analysis. As shown in Fig. 6, tumor suppressor Rb protein showed induction at late G1 phase as expected. This induction of Rb protein remained through S phase and then declined to normal levels. Cyclin D protein was induced in late G1 phase and remained at high expression levels through the S phase. Cdk4 protein was similarl