The potential value of the neutral comet assay and γH2AX foci assay in assessing the radiosensitivity of carbon beam in human tumor cell lines

Background Carbon ions (12C6+) are high linear energy transfer (LET) radiation characterized by higher relative biological effectiveness than low LET radiation. The assessment of tumour radiosensitivity would be particularly useful in optimizing the radiation dose during radiotherapy. The aim of the current study was to evaluate the potential value of the neutral comet assay and γH2AX foci assay in assessing 12C6+ radiosensitivity of tumour cells. Materials and methods The doses of 12C6+ and X-rays used in the present study were 2 and 4 Gy. The survival fraction, DNA double-strand breaks (DSB) and repair kinetics of DSB were assayed with clonogenic survival, neutral comet assay and γH2AX foci assay in human cervical carcinoma HeLa cells, hepatoma HepG2 cells, and mucoepidermoid carcinoma MEC-1 cells at the time points of 0.5, 4, 16 and 24 h after 12C6+ and X-rays irradiation. Results The survival fraction for 12C6+ irradiation was much more inhibited than for X-rays (p < 0.05) in all three tumour cell lines tested. Substantial amounts of residual damage, assessed by the neutral comet assay, were present after irradiation (p < 0.05). The highest residual damage was observed at 0.5 or 4 h, both for 12C6+ and X-ray irradiation. However, the residual damage in HeLa and MEC-1 cells was higher for 12C6+ than X-rays (p < 0.05). The strongest induction of γH2AX foci was observed after 30 min, for all three tumour cell lines (p < 0.01). The franction of γH2AX foci persisted for at least 24 h after 12C6+ irradiation; in HeLa cells and MEC-1 was higher than after X-ray irradiation (p < 0.05). The correlation coefficients between the clonogenic survival, neutral comet assay and γH2AX foci assay were not statistically significant, except for some tumour cells at individual irradiation doses and types. Conclusions Our study demonstrated that the neutral comet assay and γ-H2AX foci assay could be used to assess the radiosensitivity of 12C6+ in human tumour cells.


Introduction
Due to their physical and radiobiological properties, high linear energy transfer (LET) radiation qualities are of special interest for tumour therapy. Carbon ions radiotherapy has the potential to broaden the spectrum of primary radiotherapy, as first reports on favourable results for 'radio-resistant tumours' like primary renal cell carcinoma have become available and particle irradiation was shown to suppress metastatic potential of cancer cells. 1,2 In order to obtain more insight into this promising strategy some features of the underlying molecular mechanisms inducing the cellular response to carbon ions radiation are currently analyzed. Ionizing radiation causes various DNA lesions such as base damage, DNA-protein cross-links, DNA single-strand breaks and DNA doublestrand breaks (DSB). An important prerequisite for a better understanding of such high-LET radiation on the DNA is the mechanistic description of the processing of DSB. It is well-known that the extremely large localized energy deposition and energy ions can lead to particularly complex types of DSB. 3 Although these effects can lead to cell death, mutations, genomic instability, or carcinogenesis, problems associated with the repair of the high-LET induced DSB are not fully understood. An estimation of the DNA damage would seem a likely surrogate marker for radiosensitivity because DNA is considered as a primary target of ionizing radiation. The comet assay, which is widely used to quantify DNA damage after exposure to various genotoxic agents, can be a fast and reliable assay system for determining cellular radiosensitivity. Many reporters also have shown that there was a close correlation between γH2AX residual foci and radiosensitivity in some tumour cell lines and γH2AX foci could be used as a sensitive detector of low dose ionizing radiation. The clonogenic assay is the gold standard technique in determining the cellular radiosensitivity. Attempts have been made in the past to correlate DNA damage with the clonogenic survival of the tumour cells using the comet assay and γH2AX assay. 4,5 However, this correlation was not profound, observed in tumour cells with carbon ions ( 12 C 6+ ) irradiation, suggesting for more studies in this direction for the potential application of the comet assay and γH2AX foci assay in determining the 12 C 6+ radiosensitivity of tumour cells.
In the present study, the neutral comet assay and γH2AX foci assay were evaluated for their usefulness in assessing radiosensitivity and the correlation between clonogenic survival, neutral comet assay and γH2AX foci assay in human cervical carcinoma Hela cells, hepatoma HepG2 cells and mucoepidermoid carcinoma MEC-1 after irradiation with 12 C 6+ as compared to X-rays. Our studies emphasize the usefulness of the neutral comet assay and γH2AX foci assay in the assessment of the 12 C 6+ radiosensitivity of tumour cells.

Cell lines and treatments
Human cervical carcinoma HeLa cell and human hepatoma HepG2 cell were purchased from shanghai institute of biochemistry and cell biology. Human mucoepidermoid carcinoma MEC-1 cell was purchased from School of Stomatology Fourth Military Medical University. The cells were routinely subcultured in Dulbecco's Modified Eagle Medium (DMEM) (GIBCO, USA), containing 10% newborn calf serum, 100U/mL penicillin, 125 ug/mL streptomycin, and 0.03% glutamine. Exponentially growing cells were seeded at 2×10 4 cells/100 mm dish and were exposed to different dose of carbon ions ( 12 C 6+ ) or X-rays. Immediately following irradiation medium was quickly removed and cells were then incubated for various time intervals at 37˚C to allow repair.

Irradiation using carbon ions beam and X-rays
Carbon ions beam was supplied by the Heavy Ion Research Facility in Lanzhou (HIRFL) at the Institute of Modern Physics, Chinese Academy of Sciences (IMP-CAS). Cell exposures were conducted at the therapy terminal of the HIRFL, which had a vertical beam line. Due to the energy degradation by the vacuum window, air gap, Petri dish cover and medium, the energy of the ion beam on cell samples was calculated to be 300 Mev/u, corresponding to a LET of 15 keV/um and the dose rate was adjusted to be about 0.4 Gy/min. The ion beam was calibrated by absolute ionization chamber. The tumour cells were irradiated by plateau of carbon ions LET curve and the dose of scatter off the walls of the plate has been calculated and incorporated into the total dose. The acquisition of the data (preset numbers converted to absorbed dose of particle radiation) was automatically obtained using a microcomputer during irradiation.
Low-LET irradiations were performed using Faxitron Cabinet X-ray System, (Model RX-650, 130kVp, 5mA, USA) operated at 100 kVp. The dose rate was approximately 1.38 Gy/min and the dose used for each irradiation ( 12 C 6+ or X-rays) was 0,2 and 4 Gy. All irradiations were performed once for each dish of cells at room temperature on the same day.

Clonogenic survival assays
The cells were cultured for 0.5, 4, 16 and 24 h after irradiation and then were washed with phosphatebuffered saline, trypsinized, and counted using a Coulter counter, replated at a density of 5×10 2 ~ 3×10 4 cells in duplicate using 100 mm dishes for cell-survival assays. Plates were stained and colonies were counted two weeks later. Counts from the two plates were averaged, and surviving fraction was calculated as the ratio of the plating efficiency of the treated cells divided by the plating efficiency of the control cells. Experiments were repeated 3-4 times. 6 The survival fraction was calculated using the following formula:

Double-strand break detection using the neutral comet assay
The neutral comet assay was used as described previously. 7 After the trypsin treatment to produce a single cell suspension, approximately 1.5×10 4 cells were embedded in 0.75% low-gelling-temperature agarose and rapidly pipetted onto a precoated microscope slide. Samples were lysed for 4 h at 50˚C in 0.5% SDS, 30 mM EDTA, pH 8.0. After rinsing overnight at room temperature in Tris/borate/ EDTA buffer, pH 8.0, samples were electrophoresed for 25 min at 0.6 V/cm, then stained with propidium iodide. Slides were viewed using a fluorescence microscope with a CCD camera, and 150 individual comet images were analyzed from each sample for tail moment, DNA content, and percentage DNA in tail using CASP software (www. casplab.com). CASP is a tool to image analysis in comet assay. An unlimited number of images can be marked, CASP could load them successively into an "image view" window. Only comets oriented from left (head) to right (tail) can be analyzed correctly. In addition to such parameter as head radius, tail length etc., the program calculates the tail moment (TM) and the Olive tail moment (OTM). Three independent experiments were performed.

Immunofluorescence microscopy for γH2AX
Immunofluorescent microscopy was conducted essentially as described earlier with some modifications. 8 Briefly, 2×10 4 cells were seeded into 35 mm dishes containing a glass cover slip in each well. After irradiation, slides were air-dried, and fixed for 30 min in 2% paraformaldehyde in TBS. Cells were rinsed in TBS, placed in -20˚C methanol for 1 min, rinsed, then placed for 20 min in TBS plus 1% bovine serum albumin and 0.2% Tween-20 (TTN) and finally incubated for 2 h with diluted anti-phospho-histone H2AX (Ser-139) mAb (Upstate, Lake Placid, NY) diluted 1:500 in TTN. Slides were washed and incubated with FITCconjugated anti-mouse goat F (ab') 2 fragment (DAKO, Carpinteria,CA) diluted 1:200 in TTN for 1 h at room temperature. Slides were rinsed and then immersed in 0.05 mg/mL DAPI for 15 min, rinsed and mounted with cover slips using 10 uL Fluorogard (Bio-Rad) as the antifade mounting medium, and sealed. To prevent bias in selection of cells that display foci, over 800 randomly selected cells were counted. Cells with three or more foci of any size were classified as positive. Experiments were repeated at least three times.

Statistical analysis
SPSS version 13.0 software program was employed for the statistical analysis. Data were expressed as mean ± standard error (SD). A two-tailed Student's t-test was also performed to compare the differences between 2 groups. The significance of the correlation coefficient was also calculated. A value of P < 0.05 was considered statistically significant.

Growth dynamics of colony survival assay
Colony assay experiments were performed to compare the differences in terms of the clonogenic growth dynamics of 12 C 6+ and X-rays ( Figure 1). Clonogenics cells were inactivated immediatedly, but in turn, significantly increased during 24 h after 12 C 6+ and X-rays irradiation (p < 0.05). At each time point after irradiation, the increase in the survival for 12 C 6+ was much more inhibited rather than for X-rays (p < 0.05), except for HepG2 cell at 4 Gy irradiation (p > 0.05, 24 h). For Hela cells after 2 Gy X-rays irradiation, the survival fraction almost reached the control level at 24 h (p > 0.05), however, the survival fraction for 12 C 6+ was still lower (p < 0.01).

A B
C D E F

Neutral comet assay
The neutral comet assays were applied to tumour cell lines exposed to 12 C 6+ and X-rays and allowed 0.5, 4, 16 and 24 h for double strand break rejoining and then examined for DSB. Typical comet images were shown at Figure 2. The fraction of residual damage is given in Figure 3. Substantial amounts of residual damage were present for tumour cells after irradiation (p < 0.05) and the highest residual damage was almost all shown at 0.5 or 4 h either for 12 C 6+ or X-rays. After 0.5 or 4 h of irradiation, the residual damages, contained in HeLa cells and MEC-1, was higher for 12 C 6+ than for X-rays (p < 0.05).

Immunofluorescence staining of phosphorylated H2AX foci
Since DSB are the primary toxic lesions induced by radiation, the number of DSB was scored by examining the foci of phosphorylated H2AX in irradiated cells. γH2AX foci were observed with anti-γH2AX antibodies (green) and the nuclei were stained with DAPI (blue). Typical images of 12 C 6+ and X-rays induced γH2AX foci are shown in Figure 4. The formation of γH2AX foci in untreated cells was kept at lower levels and no significantly difference was seen between cell lines and culture times (data not shown). After 30 min of radiation, γH2AX foci, visualized as bright spots, were present in all the cells. The time-dependent induction of γH2AX foci by 12 C 6+ and X-rays were counted in all tumour cell lines. It was noted the strongest inductions of γH2AX foci at 0.5 h for all three tumour cell lines (p < 0.01) and then decreased over time. A fraction of γH2AX foci persisted for at least 24 h after 12 C 6+ radiation in HeLa cells (2 and 4 Gy) and MEC-1 (4 Gy), in contrast to foci induced by X-rays that had a lower level after 24 h (p < 0.05, Figure 5). The correlation between the clonogenic survival, neutral comet assay and γH2AX foci in 12 C 6+ and X-rays irradiated tumour cells A negative correlation was observed between the clonogenic survival and tail moment assessed by neutral comet assay. The negative correlation was also shown between the clonogenic survival and γH2AX foci. However, a positive correlation was shown between the tail moment and γH2AX foci. However, the correlation coefficients for the most parameters we used, such as different doses and irradiated types, were not statistically significant (P > 0.05, Tables 1, 2 and 3).

Discussion
In general, it is believed that the DNA damage induced by high-LET heavy ions radiation is more complex than that by X-or gamma rays and leads to more severe biological consequences including cell death, mutation and transformation. 9,10 This complex DNA damage is also referred to as clus-tered damage or multiply damaged sites, which was originally introduced by Ward. [11][12][13] These damages could be generated during the processing period of the initially induced DNA lesions. Ionizing radiation even at low doses such as 1 Gy could induce this kind of damage, and with high-LET radiation, the degree of complexity may increase depending on the LET of the particular radiation. 14 The clustered DNA damage is thought to be more difficult to repair when compared with simple or isolated DNA damage, probably because of retarded enzymatic activities leading to serious biological consequences. 15 A substantial number of studies on DSB and its repair in cells exposed to high-LET heavy ions have been reported. 16,17 In general, DSB repair is inhibited as a function of LET (up to 200 keV/μm), and the degree of rejoining correlates with the cell survival. If the rejoining is inefficient, a high number of remaining DSB persist after irradiation, leading to a lower cell survival.
In the present study, radiosensitivities of different tumour cell lines to 12 C 6+ and X-rays were established using the clonogenic assay and this was correlated with DSB and repair assessed by the  neutral comet assay and γH2AX foci. We selected three tumour cell lines which were of different tissue origins. The different cell types were used to ensure that the assay was able to distinguish the radiosensitivity across different tumour types. In the clonogenic assay, a significantly survival inhibition was shown in 12 C 6+ and X-rays irradiation over time (Figure 1). It, therefore, seemed reasonable to conclude that an early significant increase in the survival fraction within 24 h occurred after X-rays irradiation, whereas the increase in survival was inhibited after 12 C 6+ irradiation. These results obviously showed that 12 C 6+ was more cytotoxic than X-rays.

Hela
The comet assay can be used when combined with repair enzymes as damage probes to detect DSB for high-LET. [18][19][20] This method has been shown to correlate radiosensitivity with residual damage, measured at longer times after irradiation. 21 The assay requires little specialized equipment other than a microscope and under neutral conditions assesses primarily the repair of DSB. From the DNA damage repair analysis, which was performed by measuring the DSB at different time intervals (0.5, 4, 16 and 24 h) and irradiation types, the differences of DSB depended on repair time and irradiation types. For example, the most severe damage was observed after 0.5 or 4 h repair and the damage decreased with time of repair. At 0.5 or 4 h, the residual damages contained in HeLa and MEC-1 cells were much higher for 12 C 6+ compared to X-rays ( Figure 3).
The results of the neutral comet assay were also compared with those of clonogenic assay in determining the radiosensitivity of the tumour cell lines for different irradiation types. For the three cell lines, the DNA repair kinetics after 12 C 6+ and X-rays irradiation, as measured using the neutral comet assay, failed to strongly correlate with the radiosensitivity of clonogenicity, which is in agreement with some of the earlier reports but not with other studies. 4,22 Though DNA repair capacity is one major determinant of the radiosensitivity of the cell, a comet assay may not be sensitive enough to resolve the smaller differences in the residual DNA damage between the cell lines and irradiation types. Our results also showed that the tail moments of the comets had mostly returned to control by 24 h, whereas the survival fraction had not. The reason was that the higher doses (> 10 Gy) were usually used in comet studies, which was significantly higher than the dose range used for the cell survival. 23,24 For example, pulsed-field gel electrophoresis experiments have established that high LET radiation induces persistent DNA double-strand breaks for doses in the range of 25-80 Gy. 25,26 In the present study the doses of 2 and 4 Gy were used for both clonogenic assay and comet assay. The smaller doses usually produced lesser DSB results in the residual damages after 24h contained in cells almost could not be detected by the comet assay.
In 1998, Bonner's group identified a site in histone molecules, namely H2AX, which is phosphorylated after DSB is induced. 27 The ''so-called'' γH2AX assay was introduced to detect DSB by simply counting foci in cells using an antibody of γH2AX. 27-30 γH2AX foci along with other radiation induced foci (RIF) were also used to observe the spatial distribution of the radiation tracks from high-LET heavy ion particles. [31][32][33][34][35] Asaithamby et al. indicated that high-LET iron ions induced more clustered DNA damage than low-LET radiation; more colocalized foci from three markers were observed with iron ions. 36 They also showed that high-LET iron ions led to a significantly higher number of chromosome aberrations per mitotic cell than low-LET radiation. 36 Although many reporters have shown γH2AX foci could be used as a sensitive detector of low dose ionizing radiation, Tomohiro's research showed that there was not a close correlation between residual foci and radiosensitivity in some tumour cell lines which showed high expression of endogenous γH2AX foci. 37 In the present study, we firstly compared the background values of γH2AX in three tumour cell lines. The expression of endogenous γH2AX foci was lower and there was not a significant difference between the three tumour cell lines we used (p > 0.05). We, then, measured foci frequency for up to 24 h and found that a fraction of γH2AX foci persisted for at least 24 h after high LET carbon ions radiation in Hela and MEC-1 cells, in contrast to foci induced by X-rays that had a lower level after 24 h ( Figure 5). This confirms the earlier studies that these persistent γH2AX foci as evidence of persistent DSB. Karlsson and Stenerlöw found persistent γH2AX foci for up to 24 h after 1 Gy of high LET nitrogen-ion irradiation in non-proliferating normal human fibroblasts. 38 Takahashi et al. showed a slower decrease in the γH2AX intensity using flow cytometry after exposure to 500 MeV/ amu iron ions compared to X-rays irradiation in exponentially growing human AG01522 fibroblasts. 39 Desai et al. also showed a LET dependency for the disappearance of γH2AX clusters in confluent normal human fibroblasts between iron (176 keV/mm) and silicon (54 keV/mm) ions. 40 A recent study using γH2AX fluorescence indicated that high-LET carbon irradiation contributed significantly to the slow component of γH2AX loss kinetics than X-irradiation (80% after carbon irradiation vs. 47% after X-irradiation) confirming the complex nature of DSB by high-LET radiation. 41 Some results showed the DNA damage measured right after X-rays exposure could be a better reflection of the radiosensitivity than the residual damage 5 , however, for 12 C 6+ irradiation, the results were not clear. For both 12 C 6+ and X-rays irradiation, although our results clearly indicated that the γH2AX assay could predict the radiosensitivity of different tumour cells and irradiation types, which was not in line with the results of the clonogenic survival assay and tail moment of comet. This problem can be attributed to that for three tumour cells we used, the highest level of DSB were at 0.5 and 4 h for the comet assay and the strongest inductions of γH2AX foci were at 0.5 h for γH2AX assay, however, the lowest survival fractions should be right after the radiation exposure.

Conclusions
Our studies showed that the neutral comet assay and γH2AX assay could be used in assessing the radiosensitivity of tumour cells in 12 C 6+ irradiation. However, to gain greater confidence in use of these techniques in determining the 12 C 6+ radiosensitivity of tumour cells, a larger number of cell lines and validation in a clinical scenario using biopsy samples may be needed.