Evaluation of DNA damage induced by proton and lithium beams in murine melanoma cells

Evaluation of DNA damage induced by proton and lithium beams in murine melanoma cells IL Ibañez *1,2, C Bracalente 1, M Edreira 1, MA Palmieri 3, B Molinari 1,2, L Policastro 1,2, A Kreiner 1,2,4, A Burlón 1, A Valda 4, MF García Morillo 4, J Davidson 2, M Davidson 2, M Vázquez 1, M Ozafrán 1 and H Durán 1,2,4 1 Comisión Nacional de Energía Atómica, Av. Gral. Paz 1499, (165) San Martín, Argentina 2 CONICET, Argentina 3 Facultad de Ciencias Exactas y Naturales, UBA, Argentina 4 Escuela de Ciencia y Tecnología, Universidad Nacional de San Martín, Argentina Abstract. The aim of this study was to evaluate the response of melanoma cells to low and high linear energy transfer (LET) radiation. For this purpose, DNA damage and survival were determined as a function of dose in mouse melanoma cells (B16-F). Irradiations were performed with monoenergetic proton (14 MeV, 3.4 kev/µm and 2.9 MeV, 14 kev/µm) and lithium (14 MeV, 135 kev/µm) beams generated by the TANDAR accelerator (Buenos Aires, Argentina) and with a 137 Cs γ source. Survival curves were obtained by clonogenic assay and fitted to the linear quadratic model. DNA single strand breaks (SSBs) were evaluated by the alkaline comet assay and double strand breaks (DSBs) by the detection of phosphorylated histone H2AX (γh2ax) foci. Results showed an increase in the α parameter and a decrease in the β parameter of the survival curves as a function of LET. The comet assay showed an increase in SSBs as a function of dose for all the radiations evaluated. γh2ax labeling demonstrated that the number of foci increases as a function of dose. A decrease in the number of foci was found for 1 and 2 Gy of γ rays and 14 MeV protons 6 h post-irradiation, revealing the repair capacity of DSBs for low LET radiation. The comparison of γh2ax labeling results for low and high LET radiations demonstrated an increase in the number of foci as a function of LET. Moreover, there was an increase in the size of γh2ax foci in cells irradiated with lithium beams, which could be attributed to the clusters of DSBs induced by high LET radiation. Concluding, our results showed a good correlation between the evaluation of γh2ax and survival, which is consistent with the fact that DSBs are difficult to repair and may be correlated with cell death. KEYWORDS: ionizing radiation, LET, DNA damage, melanoma cells, comet assay, γh2ax. 1. Introduction The incidence of melanoma has substantially increased worldwide over the last 4 years. Although melanoma accounts for only 1% of skin cancer, it is responsible for at least 8% of skin cancer death. Most advanced melanomas respond poorly to radiotherapy and chemotherapy and no effective therapy exists to inhibit the metastatic spread of this cancer [1]. Protons and lithium, and more generally charged hadrons, have a well-defined penetration depth, known as range, leaving towards the end of that range a maximum energy per traversed unit length. This generates a peak in the delivered dose, known as the Bragg peak, the dose being very small beyond that region. This behavior is completely different to that of photons (and neutrons), which involves an exponential attenuation in intensity from the surface towards the interior of tissue. Within this context, the utilization of hadron beams has important comparative advantages as far as dose localization is concerned [2-4]. Thus hadron therapy is becoming increasingly important for the treatment of tumors localized in the vicinity of vital and radiosensitive organs. In particular, for a certain number of locations, such as choroidal melanomas and certain tumors of the base of the skull, hadron therapy is recognized as the best alternative [5]. Moreover, high LET charged particles therapy is an interesting approach for the treatment of melanoma because of its increased potential to kill low LET radioresistant cells. Ionizing radiation produces a broad spectrum of molecular lesions to DNA, including single strand breaks (SSB), double strand breaks (DSB) and a great variety of base damages. There are many methods to measure SSB and DSB. Particularly, to evaluate the response of heterogeneous cells such as tumor cells, some methods have been developed to detect DNA strand breaks in individual cells. In 198s, an alkaline electrophoretic technique capable of detecting DNA damage in individual cells was * Presenting author, E-mail: irenuliz@yahoo.com 1

developed [6, 7]. This technique, known as the comet assay or single cell gel electroforesis assay, is one of the most sensitive and accurate assays to measure DNA damage [8]. It allows measurements of total DNA content per cell and the amount of DNA migration which is proportional to the number of strand breaks per cell [9, 1]. In the last two decades, the comet assay has been extensively applied to study DNA damage induced by chemicals and radiations in different cell types [11]. More recently, the phosphorylation of the histone protein H2AX (γh2ax) has been used as a measure of the formation and rejoining of DSB induced by different types of chemicals and radiations [12, 13]. This approach is based on one of the early steps in the response of mammalian cells to DSBs. H2AX is a histone variant of the H2A family, which is phosphorylated at serine 139 within its conserved COOH-terminal region in response to the presence of DSBs. The induction of γh2ax by radiation is reported to be mediated mainly by ATM protein, occurs at the sites of DSBs in the nuclear DNA and precedes the induction of DNA repair enzymes [14]. The number of γh2ax foci formed in this way has been shown to be directly proportional to the number of DSBs and their dephosphorylation has been correlated with repair of DSBs [15]. The local formation of γh2ax allows microscopic detection of distinct foci that most likely represent a single DSB [15]. The potential to detect a single focus within the nucleus makes this the most sensitive method currently available for detecting DSBs [16]. DSBs are the most toxic form of DNA damage, because a single unrepaired DSB can lead to abnormal mitosis with losses of large fragments of DNA [17]. It has been shown that the few DSBs that are produced by ionizing radiation correlate closely to the amount of induced cell death, whereas the 1-fold more abundant SSBs and base damage are of minor importance in relation to the cytotoxic effect. The aim of this study was to evaluate the response of a melanoma cell line to proton and lithium irradiations by determining DNA damage and survival curves. For this purpose the formation and repair of DSBs was determined by the detection of γh2ax foci at different times post-irradiation, the formation of SSBs was determined by the alkaline comet assay and survival curves were obtained by clonogenic assay. 2. Materials and Methods 2.1. Cell Line and Cell Culture The mouse melanoma cell line B16-F was used. Cells were grown in RPMI 164 medium (Invitrogen Argentina SA) supplemented with 1 % fetal bovine serum (FBS) (Natocor, Argentina), 5 U /ml penicillin and 5 µg /ml streptomycin, at 37 C in a 5% CO 2 humidified atmosphere. 2.2. Irradiation experiments The proton and lithium beams were produced by an electrostatic 2 MV vertical tandem accelerator, the TANDAR accelerator of the National Atomic Energy Commission in Buenos Aires [18]. The energies of the particles were 14.4 ±.2 and 2.9 ±.2 MeV for irradiations with protons in the plateau region of the Bragg curve and in the Bragg peak respectively, and 14 ± 1 MeV for lithium irradiations. A 137 Cs source was used for gamma irradiations (IBL 437C H-type Irradiator, CIS Bio International, Schering S.A.). Cells were irradiated with -7 Gy for survival curve experiments and -4 Gy for determinations of DNA damage. 2.3. Survival curves For clonogenic assays, cells from mid-log growing cultures were plated in 6 mm diameter cell culture dishes and incubated at 37 ºC in a 5% CO 2 humidified atmosphere to allow cell adhesion and spreading prior to treatment. At the moment of irradiation, 6 h after plating, the cell medium was removed and the dish covers were replaced by new 4 µm Mylar foils. Immediately after irradiation, cells were supplemented with complete fresh medium. Control cells were incubated in the same conditions as irradiated cells. Cells were incubated for 1 days at 37 ºC in a 5% CO 2 humidified atmosphere. Colonies were fixed with methanol/acetic acid 3:1, stained with violet crystal and the fraction of clonogenic cells for each dose was determined by scoring colonies containing at least 5 2

cells. Each experiment was performed at least twice with triplicates for each condition. Survival curves were fitted to the linear quadratic model (equation 1) or to the exponential model for curves without shoulder (equation 2) S = exp (αd+βd 2 ) (1) S = exp (αd) (2) 2.4. Comet assay The single cell gel electroforesis assay (comet assay) was performed under alkaline conditions as previously described [19] with some modifications. Immediately after irradiation, cells were harvested by tripsinization and 4 x 1 4 cells were suspended in 5 µl of phosphate-buffered saline (PBS) and mixed with 45 µl of.5% low melting-point agarose (Invitrogen Argentina) at 37 C. This mix was spread on a microscope slide precoated with two layers of 5 µl of 1% agarose and covered with a coverslip. After gelling at C for 1 min, the coverslip was gently peeled off from the agarose layer. The cells were lysed by dipping the slides in a lysis solution (1 mm Na-EDTA, 1% sodium sarcosinate, 2.5 mm NaCl, 1% Triton X-1, ph 1) for 1 h at 4 C. The slides were rinsed in an alkali buffer (1 mm Na-EDTA, 3 mm NaOH, ph 13), subsequently immersed in a horizontal gelelectrophoresis cell with fresh alkali buffer without power for 2 min to allow DNA denaturation and unwinding and reveal the alkali-labile damage. Then, a weak electric field was applied (3 mm,.6 V/cm for 25 min at 25 C) under indirect light to prevent additional DNA damage. After electrophoresis, the slides were washed twice for 5 min in a neutralizing buffer (.4 M Tris, ph 7.5) and stained with 1 µl of ethidium bromide (2 g/ml). Observations were performed using an Olympus BX51 epifluorescence microscope coupled to an intensified solid-state charge-coupled device (CCD) camera (Olympus DP7) and image analysis system. Slides were illuminated with green light (excitation filter of 515 545 nm), and emission barrier filter of 59 nm. The parameters determined for each comet were: head and tail intensity, head and tail area, tail length and tail moment. The tail moment was defined as the product of the percentage of fluorescence intensity in the comet tail and the tail length. At least 3 cells randomly selected were measured in each slide. Each experiment was performed twice with two slides for each experimental condition. 2.5. Immunofluorescence and quantification of γh2ax B16-F cells were grown to 7 % confluence in 6 mm-dishes. At the moment of irradiation cell medium was removed and the dish covers were replaced by new 4 µm Mylar foils. Control cells were incubated in the same conditions as irradiated cells. Immediately after irradiation, cells were supplemented with complete fresh medium and maintained at 37 ºC in a 5% CO 2 humidified atmosphere until 4 % paraformaldehyde fixation at either 3 min or 6 h post-irradiation. Cells were fixed for 15 min at room temperature and washed three times with PBS, chilled on ice and permeabilized with.5 % Triton X-1 in PBS during 15 min, washed thrice with PBS and blocked with 5 % FBS in PBS during 3 min at room temperature. After blocking, cells were incubated overnight at 4 ºC with a monoclonal anti-γh2ax antibody (Upstate, Lake Placid, NY), 1:5 dilution in PBS, washed three times and incubated with FITC-labeled secondary antibody (Sigma) in the dark for 1 h at room temperature. Cells were then washed thrice with PBS and counterstained and mounted with a DAPI-antifade solution (1 µg/ml 4,6-diamidino-2-phenylindole dihydrochloride dihydrate (Sigma) in an antifade solution) in the dark. Cells were examined in an Olympus BX51 epifluorescence microscope utilizing immersion oil with a 1X (UPlanApo 1 X/1.35 oil) objective lens. For each treatment condition, FITC and DAPI images were serially captured by a CCD camera (Olympus DP7) and more than 5 fields containing approximately 2 cells each were stored. γh2ax foci were counted by eye. More than 1 cells were randomly screened per each experimental condition from the stored images. The area of each γh2ax foci was quantified by using the NIH Image J software. Two independent experiments were performed. 3

2.6. Statistical analysis Experiments were performed at least twice and the results are presented as mean ± SD and in the specific case of the γh2ax analysis data are shown as mean ± variance, since the experiment is characterized by a Poisson distribution. Mean ± SE was utilized when appropriate. Significant changes were assessed using unpaired t-test, one-way analysis of variance and nonparametric Kruskal-Wallis test followed by Tukey s or Dunn s multiple comparison tests to determine significant differences between group means. P values of less than.5 were considered significant for all tests. 3. Results 3.1. Survival curves To characterize the radiation response of B16-F cells to proton and lithium beams, survival curves were obtained by the clonogenic assay and analysed by fitting to the linear quadratic model or the exponential model as described in M&M. Figure 1 shows the survival curves for protons (in the plateau region of the Bragg curve and in the Bragg peak) and lithium irradiations. The survival curves obtained after low LET proton irradiation (plateau) showed a survival curve with shoulder similar to the response to γ irradiation (not shown). The survival curves for higher LET protons (Bragg peak) and lithium irradiations had no shoulder. Table 1 shows the parameters of the survival curves fitted to the linear-quadratic model for proton irradiations in the plateau region of the Bragg curve and to the exponential model for protons in the Bragg peak and lithium irradiations. An increase of the α parameter value as a function of LET and the loss of the quadratic component of the survival curves (β = ) at intermediate and high LET were found. Figure 1: Survival curves of B16-F melanoma cells after irradiation with proton and lithium beams. Clonogenic assays were performed as described in M&M. ( ) Plateau protons, ( ) Bragg peak protons and ( ) lithium. Data represent means ± SD of a representative experiment. 1 Survival.1,1.1,1 1 2 3 4 5 6 7 8 Dose (Gy) Table 1: Parameters of the fitted survival curves Mean Energy [MeV] Survival curve parameters Mean LET [kev/µm] α β Plateau 14.4 ±.2 3.4 ±.1.73 ±.11.49 ±.22 Protons Bragg peak 2.9 ±.2 14 ± 1.61 ±.21 Lithium 14 ± 1 135 ± 7 1.24 ±.16 4

3.2. Detection of SSBs SSBs were determined by the alkaline comet assay. All the parameters evaluated increased as a function of dose, both for low LET protons (plateau) and for intermediate LET protons (Bragg peak). Figure 2 shows representative photographs of comets and the results obtained for tail intensity and tail moment. Table 2 shows all the parameters evaluated for 2 Gy of gamma and proton irradiations. Significant differences vs. controls were observed in the induction of SSBs both for gamma and proton irradiations. No significant differences were observed between gamma and plateau or Bragg peak protons. Figure 2: Detection of SSBs by alkaline comet assay. (a) Demonstrative photographs of comets of B16-F cells irradiated with 2 Gy of proton beams (plateau and Bragg peak) or unirradiated (control). (b) Representative comet parameters: tail moment (top) and tail intensity (bottom) vs. dose for cells irradiated with ( ) plateau protons and ( ) Bragg peak protons. Data represent means ± SE. *p <.5 vs. control. (a) control (b) Tail moment (arbitrary units) 15 1 5 * 1 2 (plateau) (Bragg peak) Tail intensity (%) 8 7 6 5 Dose (Gy) * 1 2 Dose (Gy) Table 2: Comet parameters for cells irradiated with 2 Gy of gamma rays and proton beams Mean Energy (MeV) Mean LET (kev/µm) Tail Intensity (%) Tail Area (%) Tail Length (arbitrary units) Tail Moment (arbitrary units) Controls 56.82 ± 3.89 73.7 ± 3.83 146.9 ± 2.4 8.62 ± 1.55 Gamma.667.2 63.14 ± 3.24 76.82 ± 2.22 29.1 ± 16.5 (a) 13.68 ± 1.56 (a) Protons Plateau Protons Bragg peak 14.4 ±.2 3.4 ±.1 68.45 ± 2.13 (b) 8.65 ± 1.74 21.1 ± 16. (a) 14.95 ± 1.49 (a) 2.9 ±.2 14 ± 1 72. ± 1.72 (a) 86.5 ±.99 (a) 191.4 ± 7.4 (a) 14.2 ±.8 (b) Data represent means ± SE (a) p <.1 vs. control (b) p <.5 vs. control 5

3.3. Detection of γh2ax To evaluate the formation and repair of DSBs, nuclear γh2ax foci were determined by immunocytochemistry as described in M&M. Figure 3 shows photographs of a representative experiment of cells irradiated with 1 Gy of gamma rays and proton and lithium beams. Figure 4 shows the quantification of the number of foci per nucleus for all the radiations evaluated. DSBs increased as a function of dose at 3 min post-irradiation for all the radiations evaluated and significantly decreased (p <.1) at 6 h post-irradiation only for 1-2 Gy of gamma rays and plateau protons (low LET radiation) (Figure 4a), which may be explained by the repair capacity. The frequency histograms show the increase of the number of cells with less foci after 6 h of incubation (Figure 4b), confirming the repair of DSBs after 1 and 2 Gy of low LET radiation. Conversely, for high LET radiation, the frequency histogram after 6 h is similar to that obtained immediately after irradiation, revealing that the repair capacity of densely damaged DNA is significantly lower. Figure 5 shows the increase in the foci size for high LET as compared to low LET radiation, which could be related with the highly concentrated DSBs in the clustered DNA damage induced by high LET radiation. Figure 3: Detection of nuclear γ-h2ax foci 3 min and 6 h post-irradiation. B16-F melanoma cells exposed to 1 Gy of low and high LET radiation, i.e. gamma rays and proton (plateau and Bragg peak) and lithium beams, or unirradiated (control). Cells were fixed at the specified times for immunocytochemical detection of nuclear γh2ax foci as described in M&M. DAPI: staining of nuclear DNA. FITC: staining of γh2ax foci. 3 min DAPI 5 µm Control Gamma rays (plateau) (Bragg peak) Lithium beams 6 6h FITC DAPI FITC

Figure 4: Quantification of nuclear γh2ax foci 3 min and 6 h post-irradiation. B16-F melanoma cells were exposed to 1-4 Gy of gamma rays and proton and lithium beams or unirradiated. Cells were fixed at ( ) 3 min and ( ) 6 h post-irradiation and immunocytochemical detection of γh2ax was performed as described in M&M. (a) Average number of γh2ax foci per nucleus vs. dose. Data represent means ± variance. (b) Frequency histograms of percentage of cells presenting to more than 6 γh2ax foci per nucleus for the different doses and types of radiation utilized. Note that the bars on the far right correspond to > 6 foci/ nucleus. Different scales were used to optimize visualization. Gamma rays Plateau Bragg Peak Lithium beams (a) Foci / Nucleus (Average) 1 5 1 2 3 4 1 5 1 2 3 4 1 5 1 2 3 4 1 5 1 2 3 4 Dose (Gy) Dose (Gy) Dose (Gy) Dose (Gy) (b) 2 3 2 2 Gy 1 15 1 1 2 4 6 2 4 6 2 4 6 2 4 6 4 3 24 4 1 Gy 2 15 12 2 2 Gy Frequency (Percentage) 2 4 6 5 25 2 4 6 2 4 6 3 15 2 4 6 2 4 6 5 25 2 4 6 2 4 6 7 35 2 4 6 6 8 8 8 3 Gy 3 4 4 4 2 4 6 2 4 6 2 4 6 2 4 6 1 1 1 1 4 Gy 5 5 5 5 2 4 6 Foci / Nucleus 2 4 6 Foci / Nucleus 2 4 6 Foci / Nucleus 2 4 6 Foci / Nucleus 7

Figure 5: Determination of γh2ax foci size. (a) Representative images of nuclear γh2ax foci of B16- F melanoma cells exposed to 1-4 Gy of low and high LET radiation, i.e. plateau proton and lithium beams 3 min post-irradiation, or unirradiated (control). DAPI: staining of nuclear DNA. FITC: staining of γh2ax foci. (b) γh2ax foci size (top) and nuclear size (bottom) vs. dose for cells irradiated with ( ) plateau proton and ( ) lithium beams. Data represent means ± SD. *p <.5 vs. foci size of cells irradiated with the same dose of plateau proton beams. (a) control (Plateau) Lithium beams DAPI FITC DAPI FITC 17.85µm (b) Foci size (arbitrary units) 1.5 1,5 1.5,5 * * * 1 Gy 2 Gy 4 Gy Nuclear size (arbitrary units) 15 1 5 1 2 4 Dose (Gy) 1 2 4 Dose (Gy) 4. Discussion In this study we evaluated the response of B16-F melanoma cells to different quality radiations, gamma and proton and lithium beams. We demonstrated: a) the increase in the number of γh2ax foci as a function of dose for all the radiations evaluated, b) a marked decrease in the repair capacity as the LET of the radiation increased, c) an increase in the size of the γh2ax foci for high LET radiation. Moreover, we confirmed in this cell line the greater effectiveness of intermediate or high LET radiation, revealed by the loss of the shoulder of the survival curve for Bragg peak protons and lithium irradiations with an increase in the α parameter and the disappearance of the β parameter of the fitted survival curves. Regarding the induction of SSBs a dose dependent increase was found for all the radiations evaluated, but no differences were detected between different quality radiations. Our results of γh2ax labeling in untreated melanoma cells demonstrated a high number of foci per nucleus (mean range =17-25). These values are consistent with those reported by Warters et al. [2] who found higher levels of γh2ax foci in human metastatic melanoma cells as compared to primary melanocytes and Wasco et al. [21] who described the overexpression of γh2ax in human malignant melanocytic lesions. Loss of control of genome stability is an important aspect of the carcinogenesis process. In particular, melanoma cells express genome and chromosomal instability. The chromosome rearrangements observed in melanoma cells likely require the induction of DSBs as intermediates [2]. These authors demonstrated the co-localization of γh2ax foci with DNA repair complex proteins, such as phosphorylated ATM and 53BP1 proteins and the Nbs1 protein, suggesting that the γh2ax foci detected in melanoma cells are associated either with DNA DSBs or chromatin alterations recognized as DNA DSBs [2]. 8

It was reported that γh2ax is detectable within 3 min after irradiation and increases until a plateau is reached at 1-3 min post-irradiation [22] and the kinetics of the loss of γh2ax could be associated with DSB repair process. In this study we determined the induction of γh2ax foci at 3 min and the persistent foci at 6 h post-irradiation for different quality radiations. We found a dose dependent increase in the number of γh2ax foci. However, no significant increase in the number of foci was found for increasing LET at 3 min post-irradiation. Conversely, at 6 h post-irradiation the number of foci for low LET radiation was significantly lower than for intermediate or high LET, revealing the differential repair capacity for the DSBs induced for the different quality radiations. These results reveal the greater complexity of DSB induced by high LET radiation, which potentially leads to increased mutagenicity and decreased repairability of the damaged site. Our results are consistent with those reported by Leatherbarrow et al. [23], who showed that variations in the rate of dephosphorylation are dependent on radiation quality. Assuming that one DSB generates one focus, most of the reports have considered the number of foci per nucleus independently of the size of the foci. However, to evaluate γh2ax foci in cells exposed to charged particles, the size of the foci should be considered. Each high LET radiation track induces multiple DSBs along its core and such clusters of DSBs cannot be resolved by optical microscopy and thus appear as a single focus [24]. This could explain the lack of increase in the number of foci that we described herein for lithium irradiations. Thus, we measured the size of the γh2ax foci and we demonstrated a significant increase of foci size induced by lithium beams as compared to low LET protons (plateau region). In agreement with our results, Costes et al. reported that the average γh2ax focus size doubles in normal human fibroblasts 2 h after nitrogen beam irradiations as compared to X ray exposure [24]. It has been previously demonstrated that mitotic cell death induced by ionizing radiation or other DNA-damaging agents correlate with the induction and repair capacity of DSBs [25]. The amount of SSBs increases linearly with the increment of dose both for low and high LET radiation [26]. The relationship between the amount of DSBs vs. dose is quadratic or linear-quadratic for low LET radiation and linear for high LET radiation [26]. These relationships are closely related with the behaviour of survival curves. Moreover, previous reports showed that the relationship between nonrepairable DNA damage and dose correlates with survival curves [25]. The analysis of results presented herein reveals that non-repairable or slowly rejoined DSBs would be the most accurate mitotic cell death indicator, in agreement with Banáth et al. [27] who reported that although the induction of γh2ax foci in response to radiation in cervical cancer cell lines showed no obvious relationship with radiosensitivity, the residual level of γh2ax 24 h after irradiation correlated with the surviving fraction at 2 Gy. In summary, these data provide an insight into the response of melanoma cells to charged particle radiations. The relevance of these studies lies in the fact that high LET radiation is suitable for treatment of melanoma because of its increased potential to kill low LET radioresistant cells. Acknowledgements This research was supported by the National Research Council (PIP-6134) and by the Medical Physics and Biomedical Applications Research Programme of the University of San Martín. The authors acknowledge CEBIRSA for the use of the 137 Cs gamma source. The authors wish to thank to R. Frank and M.E. Ibalo for their assistance during particle irradiations, D. Navalesi for her technical assistance and F. Grings for his valuable help in statistical analysis. References [1] Ivanov, V.N., Zhou, H., Hei, T. Sequential treatment by ionizing radiation and sodium arsenite dramatically accelerates TRAIL-mediated apoptosis of human melanoma cells. Cancer Res. 67: 5397 (27). [2] Chen, G.T.Y. et al. Treatment planning for heavy ion radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 5: 189 (1979). 9

[3] Bonnett, D.E. Current developments in proton therapy: a review. Phys. Med. Biol. 38: 1371 (1993). [4] Oozer, R. et al. A model for the lateral penumbra in water of a 2-MeV proton beam devoted to clinical applications. Med. Phys. 24: 1599 (1997). [5] Habrand, J.L. et al. Protontherapy. Clinical features. Bull. Cancer Radiother. 83: 247 (1996). [6] Östling, O., Johanson, K.J. Microelectrophoretic study of radiation-induced DNA damages in individual mammalian cells. Biochem. Biophys. Res. Commun. 123: 291 (1984). [7] Singh, N.P. et al. A simple technique for quantitation of low levels of DNA damage in individual cells. Exp. Cell Res. 175: 184 (1988). [8] Collins, A.R. Investigating oxidative DNA damage and its repair using the comet assay. Mutat. Res. (in press) doi:1.116/j.mrrev.27.1.2 (27). [9] Olive, P.L., Banáth, J.P. Induction and rejoining of radiation-induced DNA single-strand breaks: tail moment as a function of position in the cell cycle. Mutat. Res. 294: 275 (1993). [1] Olive, P.L., Wlodek, D., Banáth, J.P. DNA double-strand breaks measured in individual cells subjected to gel electrophoresis. Cancer Res. 51: 4671 (1991). [11] Olive, P.L. Impact of the comet assay in radiobiology. Mutat. Res. (in press) doi:1.116/j.mrrev.27.11.1 (27). [12] Rogakou, E.P. et al. Megabase chromatin domains involved in DNA double-strand breaks in vivo. J. Cell Biol. 146: 95 (1999). [13] Tanaka, T. et al. Cytometry of ATM activation and histone H2AX phosphorylation to estimate extent of DNA damage induced by exogenous agents. Cytometry A. 71: 648 (27). [14] Burma, S. et al. ATM phosphorylates histone H2AX in response to DNA double-strand breaks. J. Biol. Chem. 276: 42462 (21). [15] Rothkamm, K., Löbrich, M. Evidence for a lack of DNA double-strand break repair in human cells exposed to very low x-ray doses. Proc. Natl. Acad. Sci. USA. 1: 557 (23). [16] Ismail, I.H., Wadhra, T.I., Hammarsten O. An optimized method for detecting gamma-h2ax in blood cells reveals a significant interindividual variation in the gamma-h2ax response among humans. Nucleic Acids Res. 35: e36 (27). [17] Van Gent, D.C., Hoeijmakers, J.H., Kanaar, R. Chromosomal stability and the DNA doublestranded break connection. Nat Rev. Genet. 2: 196 (21). [18] Schuff, J. et al. Relative biological effectiveness measurements of low energy protons and lithium beams on tumor cells. Nucl. Instr. & Meth. B 187: 345 (22). [19] Olive, P.L., Banáth, J.P. Induction and rejoining of radiation-induced DNA single-strand breaks: tail moment as a function of position in the cell cycle. Mutat. Res. 294: 275 (1993). [2] Warters, R.L. et al. Melanoma cells express elevated levels of phosphorylated histone H2AX foci. J. Invest. Dermatol. 124: 87 (25). [21] Wasco, M.J. et al. Expression of gamma-h2ax in melanocytic lesions. Hum. Pathol. (in press) doi:1.116/j.humpath.28.3.7 (28). [22] Pilch, D.R. et al. Characteristics of γ-h2ax foci at DNA double-strand breaks sites. Biochem. Cell Biol. 81: 123 (23). [23] Leatherbarrow, E.L. et al. Induction and quantification of γ-h2ax foci following low and high LET-irradiation. Int. J. Radiat. Biol. 82: 111 (26). [24] Costes, S.V. et al. Imaging features that discriminate between foci induced by high- and low- LET radiation in human fibroblasts. Radiat. Res. 165: 55 (26). [25] Dikomey, E. et al. Molecular mechanisms of individual radiosensitivity studied in normal diploid human fibroblasts. Toxicology 193: 125 (23). [26] Kanko, I. et al. DNA-lesion and cell death by alpha-particles and nitrogen ions. Adv. Space Res. 9: 73 (1989). [27] Banáth, J.P., MacPhail, S.H., Olive, P.L. Radiation sensitivity, H2AX phosphorylation, and kinetics of repair of DNA strand breaks in irradiated cervical cancer cell lines. Cancer Res. 64: 7144 (24). 1