Abstract
Background/Aim: Ultra-high dose rate irradiation (uHDR) (>40 Gy/s), commonly referred to as FLASH, has garnered attention in radiation therapy research due to its potential to mitigate damage to normal tissues while maintaining tumoricidal effects. Research on FLASH therapy using electron beams, X-rays, and proton beams has preceded studies using carbon ion beams. However, the clinical potential of FLASH carbon ion irradiation is increasingly being recognized, similar to other radiation modalities. This study aimed to evaluate the cell-sparing effect of carbon ion beams under normoxic conditions – a phenomenon that has not been previously reported.
Materials and Methods: Human salivary gland cell line (HSGc-c5), human dermal fibroblast (HDF) and human lung bronchial epithelial cell line (Nuli-1) were employed. In this study, we compared two types of linear energy transfer (19 and 50 keV/μm) and two oxygen concentrations (4% and 21%) to thoroughly investigate the cell-sparing effect, with cell death as the endpoint.
Results: A significant cell-sparing effect was observed with carbon ion beam uHDR irradiation under normoxic conditions. Linear energy transfer (LET) influenced the manifestation of the sparing effect, with higher LET (50 keV/μm) demonstrating a stronger protective effect compared to lower LET (19 keV/μm). DNA damage, as indicated by γH2AX foci, was significantly reduced under uHDR compared to conventional dose rates.
Conclusion: Carbon ion uHDR irradiation induces a cell-sparing effect under normoxic conditions, which is influenced by LET and oxygen concentration. These findings provide essential insights into the mechanisms underlying the FLASH effect and pave the way for advancing the clinical application of uHDR carbon ion therapy.
Introduction
Irradiation at ultra-high dose rates (uHDR) (>40 Gy/s) has garnered attention in the field of radiation therapy research due to its potential to mitigate damage to normal tissues while maintaining tumoricidal effects. This irradiation method was proposed by Favaudon et al. in 2014, and the observed reduction in radiation-induced damage through uHDR irradiation is referred to as the “FLASH effect” (1). Research on FLASH therapy using electron beams, X-rays, and proton beams has preceded studies using carbon ion beams. Although experimental reports on uHDR irradiation with carbon ion beams are comparatively limited, they are gradually increasing, and clinical applications of uHDR irradiation similar to other radiation modalities are anticipated (2-7).
A decade since the initial report, FLASH research has predominantly focused on animal studies. This trend stems from the fact that the protective effects observed in cellular experiments sometimes extend to tumor cells as well as normal cells (8-11), which deviates from the original definition of the “FLASH effect”. Currently, it is becoming common practice to distinguish the protective effects observed in cellular experiments as the “Sparing effect”, whereas those observed in animal experiments or clinical studies are termed the “FLASH effect”.
However, the current state of uHDR irradiation experiments shows that each facility is examining different experimental conditions; and parameters such as the dose rate, total dose, and fractionation method required to reliably induce the FLASH effect remain under discussion. Additionally, introducing FLASH radiation therapy into clinical practice necessitates consideration of the α/β ratio concept. Therefore, we believe that elucidating the mechanisms underlying the FLASH effect requires the use of simplified experimental systems, such as cellular experiments, to explore the conditions under which the cell-sparing (FLASH) effect can be observed.
Our team aims to compare three radiation sources – electron beams, proton beams, and carbon ion beams – using a uniform uHDR irradiation methodology. We have modified a therapeutic carbon ion beam irradiation system (12, 13) to establish a uHDR irradiation system. In our previous research, we reported the sparing effect under normoxic conditions using proton beam uHDR irradiation (14). In this study, we report the observation of the sparing effect under normoxic conditions using a carbon ion beam irradiation system, which has not been previously reported.
Materials and Methods
Cell lines. The Human Salivary Gland cell line (HSGc-c5) was purchased from RIKEN (Tsukuba, Ibaraki, Japan) and Human dermal fibroblast (HDF) and Human lung bronchial epithelial cell line (Nuli-1) were purchased from ATCC (Manassas, VA, USA). They were cultured in Dulbecco’s Modified Eagle Medium (DMEM) medium supplemented with 10% fetal bovine serum, 1% penicillin, streptomycin, and L-glutamine (Thermo Fisher Scientific, Waltham, MA, USA) at 37°C in humidified atmosphere containing 5% CO2.
Hypoxic condition. To conduct cell irradiation experiments under hypoxic conditions, an airtight container (Sugiyamagen, Tokyo, Japan) was sealed with an oxygen absorber (AnaeroPack, Mitsubishi Gas Chemical Company, Inc. Tokyo, Japan), an air oxygen meter (OXY-1, Jikco, Tokyo, Japan), and 24 well plate. All wells were filled with 1.4 ml of culture medium, and one of them had a non-invasive oxygen sensor (SP-PSt3-YAU-D5-YOP, PreSens, Regensburg, Germany) attached at the bottom of the well to measure oxygen concentration. Fiber optic oxygen transmitters (Fibox 4, PreSens) were inserted through a laser coaster (Coaster for Shake Flasks CFG, PreSens) from the lower outside of the airtight container. The oxygen concentration in the culture medium was measured by irradiating the laser beam of fiber optic oxygen transmitters. Fiber optic oxygen transmitters is temperature-dependent, but the measured value was corrected in real time by a temperature sensor.
Irradiation. Monoenergetic carbon ions were accelerated by synchrotron at Osaka Heavy Ion Medical Accelerator in Kansai (Osaka HIMAK) (15). The linear energy transfer (LET) values were 19.0 or 50.0 keV/μm. Irradiation dose rates were 1.6 Gy/s or 100 Gy/s for conventional dose rate (Conv) or uHDR, respectively.
Colony formation assay. Immediately after irradiation, cells were washed in PBS and trypsinized. The cells were added to 60-mm diameter dishes, cultured for 2 or 3 weeks, fixed with formalin, and stained with 0.25% crystal violet. The colonies comprising >50 cells were scored as survivors, and the survival fraction was calculated. Survival curves were fitted to a Linear-Quadratic (LQ) model (16). At least three independent experiments for each irradiation condition were performed, and the results are expressed as the average and standard deviation. The significance of differences between variables was evaluated using a student t-test, and p<0.05 indicates a significant difference.
γH2AX immunofluorescent staining. One hour after irradiation, cells were fixed using 4% Paraformaldehyde (Sigma Chemicals, Kewdale, WA, USA) for 30 min, and permeabilized with 0.5% Triton X-100 (Wako, Osaka, Japan) for 30 min. After blocking with 5% bovine serum albumin (BSA) (Sigma Aldrich, St. Louis, MO, USA) in PBS (−) for an hour, rabbit γH2AX antibody (Cell Signaling Technology, Danvers, MA, USA) diluted 1:500 in 5% BSA was added to the cells. Then, cells were incubated with the secondary antibody Alexa Fluor 488 anti-rabbit immunoglobulin G (1:1,000 dilution; Cell Signaling Technology). After 4′, 6-diamidino-2-phenylindole dihydrochloride (DAPI; Thermo Fisher Scientific) staining, γH2AX formation was measured using a fluorescence microscope, BZ-X800 (KEYENCE, Osaka, Japan).
Results
The uHDR irradiation with low LET suppressed the cell killing effect on normal cell lines in 4% O2. The pO2 was measured real time (Figure 1A). The cells were irradiated with carbon ion beams (Conv or uHDR) with 19 keV/μm at a time when the oxygen concentration of the solution reached 4%. Survival of the irradiated cells was assayed using the colony formation assay. The HSGc-C5 cells irradiated at uHDR showed a trend toward higher cell survival than those were irradiated at Conv, but statistical tests showed that the difference was not significant (Figure 1B). When HDF or Nuli-1 cells were irradiated with 2.7, 5.4, and 6.5 Gy by carbon ion beams, significantly higher cell viability was observed for cells irradiated with uHDR than for cells irradiated with the same dose of Conv from 5.4 Gy on HDF and 6.5 Gy on Nili-1, respectively (Figure 1C and D).
Comparison of cell survival curves following conventional and ultra-high dose rate irradiation at linear energy transfer (LET) = 19 keV/μm and 4% oxygen concentration. (A) Time-dependent changes in oxygen concentration within the cell culture medium, measured using a non-invasive oxygen concentration sensor. (B) Survival curves for human salivary gland cell line (HSGc-c5) cells, (C) human dermal fibroblast (HDF) cells, and (D) human lung bronchial epithelial cell line (Nuli-1) cells. Asterisks indicate a statistically significant difference (p<0.05) in cell survival rates between conventional and ultra-high dose rate at the same physical dose.
Carbon ion beam irradiation at ultra-high dose rate induces a sparing effect even under normoxic conditions. Next, we evaluated cell survival under normoxia without changing LET. The results of the colony formation assay using HSGc-C5 cells showed no significant difference in cell survival rates between Conv and uHDR irradiation (Figure 2A). Similar to the experiments performed under hypoxic conditions in HDF and Nuli-1, the cell survival rate was increased in the uHDR irradiation experiments compared to the Conv (Figure 2B and C). To ensure the accuracy of the experimental results, double-strand breaks in DNA were evaluated 1 h after irradiation using fluorescent immunostaining for γH2AX on HDF and Nuli-1 cell lines (Figure 2D and E). The γH2AX foci numbers were significantly reduced under uHDR compared to Conv in both cell lines (Figure 2F).
Comparison of cell survival curves following conventional and ultra-high dose rate irradiation at linear energy transfer (LET) = 19 keV/μm and 21% oxygen concentration. (A) Survival curves for human salivary gland cell line (HSGc-c5) cells, (B) human dermal fibroblast (HDF) cells, and (C) human lung bronchial epithelial cell line (Nuli-1) cells. Asterisks indicate a statistically significant difference (p<0.05) in cell survival rates between conventional and ultra-high dose rate at the same physical dose. (D) Expression of γ-H2AX, a marker for DNA double-strand breaks, in HDF cells 1 h after 6.5 Gy irradiation. (E) The corresponding results for Nuli-1 cells. (F) A graph based on the results from (D) and (E), illustrating the number of γ-H2AX foci per cell. Asterisks indicate statistically significant differences (p<0.05) as determined using Student’s t-test.
LET was a modifier of the sparing effect. A distinguishing feature of carbon ion beams is their variable LET. In carbon ion beam therapy, it is well known that a mixture of carbon ions with different energies is used near the tumor to obtain a spread-out Bragg peak (SOBP) (17). In the present experiment, we utilized a monoenergetic beam with a LET of 50 keV/μm, representative of the therapeutic LET near the tumor, to observe the differences in cellular response to Conv and uHDR irradiation. In the results for HSGc-C5 cells, the cell-sparing effect, which was not observed under the 19 keV/μm irradiation condition, was detected at 6.5 Gy under 4% O2 conditions (Figure 3B). The sparing effect was not observed under 21% O2 conditions (Figure 3A). Additionally, for both HDF and Nuli-1 cells, the cell-sparing effect was observed at a physical dose of 5.4 Gy, which is lower than 19 keV/μm, under normoxic conditions (Figure 3C and E). And, under 4% O2 conditions, the cell-sparing effect was observed in both normal cell lines, similar to the experiments conducted using 19 keV/μm (Figure 3D and F). These findings suggested that the cell-sparing effect is more readily observed at 50 keV/μm compared to 19 keV/μm. In other words, the results indicated that an increase in LET may be a contributing factor to the manifestation of the cell-sparing effect.
Comparison of cell survival curves following conventional and ultra-high dose rate irradiation at linear energy transfer (LET) = 50 keV/μm and 4 or 21% oxygen concentration. (A, B) Survival curves for human salivary gland cell line (HSGc-c5) cells, (C, D) human dermal fibroblast (HDF) cells, and (E, F) human lung bronchial epithelial cell line (Nuli-1) cells. Asterisks indicate a statistically significant difference (p<0.05) in cell survival rates between conventional and ultra-high dose rate at the same physical dose.
Discussion
In this study, we observed the protective effect on normal human cells under normoxic conditions following irradiation, marking the first such observation worldwide. Tinganelli et al. reported on the protective effect observed under hypoxic conditions in cellular experiments using uHDR irradiation with carbon ion beams. This study used Chinese hamster ovary cells (CHO-K1), and the dose used was 7.5 Gy, which is close to the dose at which the sparing effect was observed in our irradiation system (2). The key differences in our study include the use of a human-derived cell line and the observation of the protective effect under atmospheric conditions. Additionally, in our experimental setup, we conducted similar experiments at a 4% oxygen concentration, demonstrating that the protective effect was more pronounced compared to atmospheric conditions. Tashiro et al. were the first to conduct uHDR irradiation experiments with carbon ion beams using human cancer and normal cells. Although the doses used were up to 3 Gy, two different LET values (13 keV/μm and 50 keV/μm) were employed, corresponding to the skin and tumor regions of carbon ion beams (3). However, no sparing effect was observed in their study. Similarly, in our experiments, a dose of 3 Gy did not show a sparing effect, suggesting that the minimum dose required to observe the protective effect with carbon ion beams may be approximately 5-6 Gy. When comparing these results with our previous findings using proton beams, a protective effect was observed at 15 Gy for protons and 5.4 Gy for carbon ions in human dermal fibroblasts (HDF). This indicates a potential difference in the physical dose threshold for observing the sparing effect based on radiation sources. However, given that the outcome remains consistent – namely, the variation in DNA strand breaks depends on differences in dose rate, regardless of the radiation source – it may be appropriate to focus on DNA damage as a fundamental aspect of radiation biology to elucidate the mechanisms underlying the sparing effect of ultra-high dose rate irradiation.
This study has limitations, primarily due to the use of cell death as the endpoint, omitting assays related to cellular responses other than cell death. In contemporary radiobiology, examining immune responses following radiation exposure has become common. In separate studies, we investigated immune-related factors in cancer using pancreatic cancer cells and osteosarcoma cells in mice. Our findings indicated that high-dose rate irradiation reduced PD-L1 expression and increased MHC-1 expression in pancreatic cancer cells compared to standard dose rate irradiation (7). This suggests that differences in dose rate may lead to distinct immune responses, potentially favoring uHDR irradiation over standard dose rate irradiation in terms of immune environment. In contrast, a German study using animal experiments with LM8 cells examined lymphocyte expression but found no significant differences in the number of cytotoxic T-cells, regulatory T-cells, and helper T-cells between standard and uHDR irradiations. Our cellular experiments with LM8 cells also showed no differences in cancer immune-related factors based on dose rate, suggesting that tumor type may influence responses to uHDR irradiation (18). Considering the comprehensive cellular responses post-irradiation, beyond just cell death, further investigations with a variety of tumor types are necessary to determine which cancers are most effectively treated with FLASH therapy.
Conclusion
We successfully observed the cell-sparing effect under normoxic conditions in cellular experiments using carbon ion beams, following our previous studies with proton beams. The successful observation of the sparing effect in a simple experimental setup provides crucial insights for the continued exploration of the mechanisms underlying the FLASH effect.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number 22H03025 and in part by JSPS KAKENHI Grant Numbers 22K07770 and 22K07695. The Authors acknowledge and thank the Osaka Heavy Ion Therapy Center staff for their help in conducting uHDR measurements as well as the staff in Osaka Heavy Ion Administration Company and Hitachi Ltd. for their help in operating the accelerator.
Footnotes
Authors’ Contributions
Conceptualization; MY, KM, and SS. Data curation; KM, MY, KF, KN, RH, NH and TT. Formal analysis; KM, MY, KF, KN and RH, Funding; KM, MY, JF, SS and KO. Acquisition; KM, MY, JF, SS and KO. Investigation; KM, MY, KF, KN, RH, NH, TT and SS. Methodology; KM, MY, SS. Project administration; MU, TN, JF, SS and KO. Resources; KM, MY, NH, TT, MT, MU, TN, MS, YK, JF and KO. Software; MY, NH, TT, MT, MU, TN, MS and YK. Supervision; JF, SS and KO. Validation; KM, MY, KF, KN, RH, NH and TT. Visualization; KM, MY, KF, KN and RH. Writing – original draft; KM and MY. Writing – review & editing; KM, MY, KF, KN, RH, JF, SS and KO.
Conflicts of Interest
The Authors, Masumi Umezawa, Takuya Nomura, Masaki Shimizu, and Yoshiaki Kuwana are employees of Hitachi High-Tech Corporation.
- Received January 8, 2025.
- Revision received January 25, 2025.
- Accepted January 27, 2025.
- Copyright © 2025 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).