Our previous studies showed that the translational control of gene expression is a component of the cellular response to ionizing radiation (IR). However, the mechanisms mediating this process and its specific role in radiosensitivity have not been defined. To begin to address these issues, investigations performed in the last year have focused on eIF4E, a critical component of the eIF4F translation initiation complex. eIF4E, an mRNA 5 cap-binding protein, is a critical protein involved in regulating gene translation in response to a variety of environmental signals. Moreover, it is downstream from mTOR;we had previously shown that the mTOR inhibitor rapamycin enhances the radiosensitivity of some tumor cell lines. To determine the significance of eIF4E as a regulator of radiosensitivity, tumor and normal cell lines were treated with siRNA to eIF4E and radiation survival curves were generated based on a colony forming assay. Knockdown of eIF4E in the normal cell lines MRC9 (fibroblasts) and HMEC (mammary epithelial cells) had no effect on radiation-induced cell death. In contrast, knockdown of eIF4E in each of the 3 tumor cell lines (MDA-MB-231 breast carcinoma, DU145 prostate carcinoma and A549 lung carcinoma) resulted in an increase in radiosensitivity with DEFs (dose enhancement factors at 0.1 surviving fraction) ranging from 1.24 to 1.44. These data suggest that eIF4E serves as a tumor specific target for radiosensitization. To investigate the mechanism through which eIF4E influences the radiosensitivity of tumor cells, initial studies focused on the MDA-MB-231 cell line. A critical parameter in determining radiosensitivity is cell cycle phase distribution with S-phase cells typically the most resistant. Therefore, flow cytometry was used to determine the cell cycle phase distribution of MDA-MB-231 cells before and after eIF4E knockdown. As compared to cells exposed to non-targeted control siRNA and to untreated cells, cell cycle distribution was not significantly altered with eIF4E knockdown. These results indicate that a loss of S-phase cells does not account for the observed radiosensitization. To determine whether apoptosis was involved in the radiosensitization induced by eIF4E knockdown, annexin-V analysis was performed at 24 and 48h after irradiation (6 Gy). Knockdown of eIF4E alone did not induce apoptotic death. Radiation alone did not significantly increase apoptosis in MDA-MB-231 cultures;the combination of eIF4E knockdown and radiation also did not induce apoptosis. These data indicated that apoptosis is not involved in the radiosensitization. Finally, loss of the G2/M phase checkpoint is a frequently invoked mechanism for radiosensitization. Based on phospho-H3 labeling of mitotic cells, eIF4E knockdown also did not inhibit the radiation-induced activation of the G2/M checkpoint. Radiation-induced death of solid tumor cells typically occurs via mitotic catastrophe. Thus, this mode of cell death was defined for MDA-MB-231 cells exposed to the combination of eIF4E knockdown and radiation. As compared to cells transfected with control siRNA or untreated cells, fthe percentage of cells undergoing mitotic catastrophe after exposure to 2 Gy was significantly increased in cells treated with siRNA to eIF4E. These results indicate that the radiosensitization induced by eIF4E knockdown is mediated through mitotic catastrophe, which suggests that DNA damage or its repair may be involved. To investigate this possibility, we used gammaH2AX foci as a measure of radiation-induced DNA double strand breaks in MDA-MB-231 cells. Knockdown of eIF4E had no effect on the level of gammaH2AX foci in unirradiated MDA-MB-231 cells nor did it influence the number of foci detected at 1 hour after irradiation with 2 Gy. However, the number of gammaH2AX foci detected at 6 and 24h after irradiation was significantly elevated in cells transfected with the siRNA to eIF4E. These data suggest that eIF4E knock down has no effect on the initial level of radiation-induced double strand breaks but does inhibit their repair. Thus, data generated in the last year suggest that eIF4E is a tumor specific target for radiosensitization, a process that involves the inhibition of DNA double strand break repair. The mechanism mediating this repair inhibition will be addressed in the coming year. Whereas the studies described above clearly suggest eIF4E as a target for tumor radiosensitization, the use of siRNA is a difficult strategy to apply to in vivo studies, let alone to a human treatment situation. Therefore, to extend these studies to a more clinically relevant approach we investigated the effects of ribavirin on the radiosensitivity of MDA-MB-231 cells. Ribavirin is a well established anti-viral agent;more recently it was shown to act as a mRNA 5 cap mimetic and bind to eIF4E resulting in an inhibition of cap-dependent translation. Our initial studies showed that exposure of MDA-MB-231 cells to ribavirin (50 micromolar) for 24h before irradiation enhanced the level of radiation-induced cell death with a DEF of 1.4, similar to that induced by siRNA to eIF4E. In the coming year the radiosensitizing potential of ribavirin will be investigated in xenografts models.
