Biological response to ionizing radiation depends not only on the type of radiation and dose, but also on the duration and dose rate of treatment. For a given radiation dose, the biological response may differ based on duration and dose rate. We studied the properties of two human cell lines, M059K glioma and U2OS osteosarcoma, continuously exposed to γ rays for long time periods of more than five months. Growth inhibition in both cell lines was dependent on total dose when exposed to acute radiation over several minutes, whereas prolonged growth inhibition was dependent on dose rate after continuous irradiation over several months. The minimum dose rate for growth inhibition was 53.6 mGy/h. Cell cycle analysis showed G 1 phase accumulation in cell populations continuously exposed to γ rays, and G 2 phase accumulation in cells acutely exposed to high-dose-rate γ rays. Cells continuously exposed to γ rays continued to exhibit delayed growth suppression even after one month in an environment of background radiation, and maintained a high-level expression of c-Jun and its phosphorylation forms, as well as resistance to apoptosis induced by staurosporine and chemotherapeutic agents. These delayed effects were not observed in cells acutely exposed to 5 Gy of radiation. These results suggest that optimization of the irradiation schedule is crucial for risk estimation, protection and therapeutic utilization of ionizing radiation.

While lifespan studies provide basic information for estimating the risk of ionizing radiation, findings on the effect of low-dose/low-dose-rate irradiation on the lifespan of mammals are controversial. Here we evaluate the effect of continuous exposure to low-dose-rate γ radiation on the lifespan of mice with accelerated aging caused by mutation of the klotho gene. While control mice died within 80 days after birth, more than 10% of mice exposed continuously to 0.35 or 0.7 or mGy/h γ radiation from 40 days after birth survived for more than 80 days. Two of 50 mice survived for more than 100 days. Low-dose-rate irradiation significantly increased plasma calcium concentration in mutant mice, and concomitantly increased hepatic catalase activity. Although hepatic activity of superoxide dismutase in mutant mice decreased significantly compared to wild-type mice, continuous γ irradiation decreased the activity in mutant mice significantly. These results suggest that low-dose-rate ionizing radiation can prolong the lifespan of mice in certain settings.

Tissue stem cells have self-renewal capability throughout their whole life, which is high enough to lead to the accumulation of DNA damage in a stem cell pool. Whether radiation-induced damage accumulates in tissue stem cells remains unknown, but could be investigated if the fate of tissue stem cells could be followed after irradiation. To realize this goal, we used an Lgr5-dependent lineage tracing system that allows the conditional in vivo labeling of Lgr5 + intestinal stem cells and their progeny. We found that radiation induced loss of Lgr5 + stem cells in the colon, but not in the duodenum. Interestingly, the loss of colonic Lgr5 + cells was compensated by de novo production of Lgr5 + cells, which increased after irradiation. These findings show that ionizing radiation effectively stimulates the turnover of colonic Lgr5 + stem cells, implying that radiation-induced damage does not accumulate in the colonic Lgr5 + stem cells by this mechanism.

Irradiation time and dose rate are important factors in the evaluation of radiation risk for human health. We previously proposed a novel dose-rate effect model, the modified exponential (MOE) model, which predicts that radiation risks decline exponentially as the dose rate decreases. Here we show that, during the early phase of exposure, up to 1000 h, the proliferation of cells continuously exposed to γ rays at a constant dose rate is gradually suppressed, even as the total dose increases. This trend holds for a number of cell lines including tumor cells, nontransformed fibroblasts and leukocytes. The accumulation of total dose by longer exposure times does not increase this suppressive effect even in cells with a defective DNA repair system, suggesting that risk is determined solely by dose rate in the later phase. The dose-rate effect in the early phase follows the MOE model in DNA repair-proficient cell lines, while cells with impaired DNA-PK or ATM show no dose-rate effect. In the later phase, however, a certain dose-rate effect is observed even in mutant cell lines, and suppression of cell proliferation no longer follows the MOE model. Our results suggest that a distinct mechanism that can operate in the absence of intact DNA-PK or ATM influences the dose-rate effect in the later phase of continuous radiation exposure.

It has been proposed that the development of diabetic nephropathy is caused in large part by oxidative stress. We previously showed that continuous exposure of mice to low-dose-rate γ radiation enhances antioxidant activity. Here, we studied the ameliorative effect of continuous whole-body irradiation with low-dose-rate γ rays on diabetic nephropathy. Ten-week-old female db / db mice, an experimental model for type II diabetes, were irradiated with low-dose-rate γ rays from 10 weeks of age throughout their lives. Nephropathy was studied by histological observation and biochemical analysis of serum and urine. Antioxidant activities in kidneys were determined biochemically. Continuous low-dose-rate γ radiation significantly increases life span in db / db mice. Three of 24 irradiated mice were free of glucosuria after 80 weeks of irradiation. Histological studies of kidney suggest that low-dose irradiation increases the number of normal capillaries in glomeruli. Antioxidant activities of superoxide dismutase, catalase and glutathione are significantly increased in kidneys of irradiated db / db mice. Continuous low-dose-rate γ irradiation ameliorates diabetic nephropathy and increases life span in db / db mice through the activation of renal antioxidants. These findings have noteworthy implications for radiation risk estimation of non-cancer diseases as well as for the clinical application of low-dose-rate γ radiation for diabetes treatment.

Ogura, K., Magae, J., Kawakami, Y. and Koana, T. Reduction in Mutation Frequency by Very Low-Dose Gamma Irradiation of Drosophila melanogaster Germ Cells. Radiat. Res. 171, 1–8 (2009). To determine whether the linear no-threshold (LNT) model for stochastic effects of ionizing radiation is applicable to very low-dose radiation at a low dose rate, we irradiated immature male germ cells of the fruit fly, Drosophila melanogaster , with several doses of 60 Co γ rays at a dose rate of 22.4 mGy/h. Thereafter, we performed the sex-linked recessive lethal mutation assay by mating the irradiated males with nonirradiated females. The mutation frequency in the group irradiated with 500 μGy was found to be significantly lower than that in the control group ( P < 0.01), whereas in the group subjected to 10 Gy irradiation, the mutation frequency was significantly higher than that in the control group ( P < 0.03). A J-shaped dose–response relationship was evident. Molecular experiments using DNA microarray and quantitative reverse transcription PCR indicated that several genes known to be expressed in response to heat or chemical stress and grim , a positive regulator of apoptosis, were up-regulated immediately after irradiation with 500 μGy. The involvement of an apoptosis function in the non-linear dose–response relationship was suggested.

Tsuruga, M., Taki, K., Ishii, G., Sasaki, Y., Furukawa, C., Sugihara, T., Nomura, T., Ochiai, A., and Magae, J. Amelioration of Type II Diabetes in db / db Mice by Continuous Low-Dose-Rate γ Irradiation. Radiat. Res. 167, 592–599 (2007). Low-dose-rate radiation modulates various biological responses including carcinogenesis, immunological responses and diabetes. We found that continuous irradiation with low-dose-rate γ rays ameliorated type II diabetes in db / db mice, diabetic mice that lack leptin receptors. Whole-body exposure of db / db mice to low dose-rate γ radiation improved glucose clearance without affecting the response to insulin. Histological studies suggested that degeneration of pancreatic islets was significantly suppressed by the radiation. Insulin secretion in response to glucose loading was increased significantly in the irradiated mice. These results suggest that low-dose-rate γ radiation ameliorates type II diabetes by maintaining insulin secretion, which gradually decreases during the progression of diabetes due to degeneration of pancreatic islets. We also inferred that protection from oxidative damage is involved in the anti-diabetic effect of low-dose-rate γ rays because expression and activity of pancreatic superoxide dismutase were significantly elevated by low-dose-rate γ radiation.

Sugihara, T., Magae, J., Wadhwa, R., Kaul, S. C., Kawakami, Y., Matsumoto, T. and Tanaka, K. Dose and Dose-Rate Effects of Low-Dose Ionizing Radiation on Activation of Trp53 in Immortalized Murine Cells. Radiat. Res. 162, 296– 307 (2004). A derivative of immortalized murine NIH/PG13Luc cells stably transfected with a Trp53-dependent luciferase reporter plasmid was used to study the transcriptional activity of Trp53 in response to radiation. The cell line was sensitive enough to detect the response of Trp53 to 0.2 cGy of 60 Co γ radiation. To examine the biological effects of low-dose-rate 60 Co γ radiation (from 0.1–10 cGy/h), we have analyzed the cell cycle, Trp53 transcriptional activity, and gene expression profiles of control and treated cells. Microarray analysis revealed up-regulation of six Trp53-mediated genes (Cdkn1a/ p21, Mdm2, Sip27, Ccng1/cyclin G1, Ei24/Pig8 and Dinb/ Polk) after exposure of cells to low-dose-rate radiation for 72 h. Using real-time PCR, a significant elevation in the expression of Ccng1/cyclin G1, Mdm2 and Cdkn1A/p21 was observed with low-dose-rate irradiation at dose rates over 5 cGy/ h. A dose-rate dependence was also observed for these three Trp53-mediated genes. The expression of Ccng1/cyclin G1 at high dose rates of γ rays was higher than that for low dose rate. However, the expression of Mdm2 for low-dose-rate γ rays was higher than for the high dose rate. Cells irradiated at low dose rates of 0.1 cGy/h and 1 cGy/h underwent G 1 -phase arrest. Furthermore, G 2 -phase growth arrest was observed in cells irradiated at the low dose rates of 5 cGy/h and 10 cGy/h, which correlated with Trp53-mediated Ccng1/cyclin G1 up-regulation. These results show that cellular response to radiation depended on the dose rate used; i.e., the responses seen at dose rates from 0.1–1 cGy/h were different from those observed at dose rates over 5 cGy/h.

Magae, J., Hoshi, Y., Furukawa, C., Kawakami, Y. and Ogata, H. Quantitative Analysis of Biological Responses to Ionizing Radiation, Including Dose, Irradiation Time, and Dose Rate. Radiat. Res. 160, 543–548 (2003). Because biological responses to radiation are complex processes that depend on both irradiation time and total dose, consideration of both dose and dose rate is necessary to predict the risk from long-term irradiations at low dose rates. Here we mathematically and statistically analyzed the quantitative relationships between dose, dose rate and irradiation time using micronucleus formation and inhibition of proliferation of human osteosarcoma cells as indicators of biological response. While the dose–response curves did not change with exposure times of less than 20 h, at a given dose, both biological responses clearly were reduced as exposure time increased to more than 8 days. These responses became dependent on dose rate rather than on total dose when cells were irradiated for 20 to 27 days. Mathematical analysis demonstrates that the relationship between effective dose and dose rate is well described by an exponential function when the logarithm of effective dose is plotted as a function of the logarithm of dose rate. These results suggest that our model, the modified exponential (ME) model, can be applied to predict the risk from exposure to low-dose/low-dose-rate radiation.