OBJECT: Tamoxifen (TAM) has been found to be effective in inhibiting proliferation of glioblastoma cells in vitro, but clinical studies have been disappointing. The purpose of this study was to determine whether insulin-like growth factor I (IGF-I), a potential autocrine/paracrine mitogen produced by glioblastomas, interferes with the antimitogenic actions of TAM. METHODS: Human glioblastoma cells were treated with or without TAM and/or IGF-I in vitro and evaluated for: viability by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenol tetrazolium bromide cleavage assay; apoptosis by histochemical analysis of nuclear morphology and 3'-OH DNA fragments; and expression of the IGF-I receptor, and the bcl-2, bcl-xL, and bax proteins by immunoblot analysis. In addition, p53 status was determined by DNA sequencing and by transient transfection with luciferase reporter plasmids containing wild-type or mutant p53. Results indicated that after 72 hours of exposure to 2 mg/ml TAM in vitro, 56.3% of WITG3 and 43.8% of U87-MG glioblastoma cells contained apoptotic nuclei (p < 0.01 compared with untreated cells). Apoptosis was independent of the presence of p53 because the WITG3 cells, in contrast to the U87-MG cells, expressed a mutant, nonfunctional p53. The WITG3 cells expressed IGF-I receptor proteins and demonstrated IGF-I binding. Exogenous IGF-I stimulated WITG3 cell proliferation and significantly (p < 0.05) antagonized the cytotoxic effects of TAM in a dose-dependent fashion; IGF-I, but not TAM, enhanced expression of bcl-2 and bcl-xL proteins; however, bax protein expression was unchanged by either treatment. CONCLUSIONS: Because many gliomas secrete large amounts of IGF-I in autocrine/paracrine growth pathways, these data may, in part, explain the failure of TAM to achieve clinical results as dramatic as those in vitro.
OBJECT: Tamoxifen (TAM) has been found to be effective in inhibiting proliferation of glioblastoma cells in vitro, but clinical studies have been disappointing. The purpose of this study was to determine whether insulin-like growth factor I (IGF-I), a potential autocrine/paracrine mitogen produced by glioblastomas, interferes with the antimitogenic actions of TAM. METHODS:Humanglioblastoma cells were treated with or without TAM and/or IGF-I in vitro and evaluated for: viability by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenol tetrazolium bromide cleavage assay; apoptosis by histochemical analysis of nuclear morphology and 3'-OH DNA fragments; and expression of the IGF-I receptor, and the bcl-2, bcl-xL, and bax proteins by immunoblot analysis. In addition, p53 status was determined by DNA sequencing and by transient transfection with luciferase reporter plasmids containing wild-type or mutant p53. Results indicated that after 72 hours of exposure to 2 mg/ml TAM in vitro, 56.3% of WITG3 and 43.8% of U87-MG glioblastoma cells contained apoptotic nuclei (p < 0.01 compared with untreated cells). Apoptosis was independent of the presence of p53 because the WITG3 cells, in contrast to the U87-MG cells, expressed a mutant, nonfunctional p53. The WITG3 cells expressed IGF-I receptor proteins and demonstrated IGF-I binding. Exogenous IGF-I stimulated WITG3 cell proliferation and significantly (p < 0.05) antagonized the cytotoxic effects of TAM in a dose-dependent fashion; IGF-I, but not TAM, enhanced expression of bcl-2 and bcl-xL proteins; however, bax protein expression was unchanged by either treatment. CONCLUSIONS: Because many gliomas secrete large amounts of IGF-I in autocrine/paracrine growth pathways, these data may, in part, explain the failure of TAM to achieve clinical results as dramatic as those in vitro.