A nucleoside anticancer drug, 1-(3-C-ethynyl-β-D-ribo-pentofuranosyl)cytosine (TAS106), sensitizes cells to radiation by suppressing BRCA2 expression

Background A novel anticancer drug 1-(3-C-ethynyl-β-D-ribo-pentofuranosyl)cytosine (ECyd, TAS106) has been shown to radiosensitize tumor cells and to improve the therapeutic efficiency of X-irradiation. However, the effect of TAS106 on cellular DNA repair capacity has not been elucidated. Our aim in this study was to examine whether TAS106 modified the repair capacity of DNA double-strand breaks (DSBs) in tumor cells. Methods Various cultured cell lines treated with TAS106 were irradiated and then survival fraction was examined by the clonogenic survival assays. Repair of sublethal damage (SLD), which indicates DSBs repair capacity, was measured as an increase of surviving cells after split dose irradiation with an interval of incubation. To assess the effect of TAS106 on the DSBs repair activity, the time courses of γ-H2AX and 53BP1 foci formation were examined by using immunocytochemistry. The expression of DNA-repair-related proteins was also examined by Western blot analysis and semi-quantitative RT-PCR analysis. Results In clonogenic survival assays, pretreatment of TAS106 showed radiosensitizing effects in various cell lines. TAS106 inhibited SLD repair and delayed the disappearance of γ-H2AX and 53BP1 foci, suggesting that DSB repair occurred in A549 cells. Western blot analysis demonstrated that TAS106 down-regulated the expression of BRCA2 and Rad51, which are known as keys among DNA repair proteins in the homologous recombination (HR) pathway. Although a significant radiosensitizing effect of TAS106 was observed in the parental V79 cells, pretreatment with TAS106 did not induce any radiosensitizing effects in BRCA2-deficient V-C8 cells. Conclusions Our results indicate that TAS106 induces the down-regulation of BRCA2 and the subsequent abrogation of the HR pathway, leading to a radiosensitizing effect. Therefore, this study suggests that inhibition of the HR pathway may be useful to improve the therapeutic efficiency of radiotherapy for solid tumors.


Background
Radiation is one of the effective treatments for cancer therapy. Double-strand breaks (DSBs) in tumor cells exposed to ionizing radiation are believed to cause apoptosis, mitotic catastrophe and reproductive cell death [1,2]. However, because DSBs are immediately repaired by DNA repair mechanisms, the cellular DNA repair capacity seems to be closely associated with the outcome of radiotherapy [3]. Therefore, targeting DNA DSB repair pathways can be a potential therapeutic strategy to enhance the antitumor effect of radiation.
In repair mechanisms for DNA DSBs, there are two major pathways, non-homologous end joining (NHEJ) and homologous recombination (HR). In the NHEJ pathway, which is active during all phases of the cell cycle, DNA ends are joined with little or no base deletion at the end-joining site. In contrast, the HR pathway employs the sister chromatid after DNA replication, which results in error-free repair. Therefore, HR is most active in the late S and G2 phases [4]. In the HR pathway, a large number of proteins are involved, including Mre11-Rad50-NBS1 (MRN) complex, RPA, Rad51, BRCA1, and BRCA2. In response to DSBs, Rad51 forms nucleoprotein filaments on single-strand DNA (ssDNA) and causes strand exchanges between ssDNA and homologous double-strand DNA [5]. Therefore, Rad51 acts as a central player in HR and its cellular expression level affects radiosensitivity and chemosensitivity [6]. BRCA2 phospholylated at Ser3291 directly interacts with Rad51 through BRC repeats, facilitating the formation of Rad51 filaments [7,8]. Accordingly, BRCA2 is a key protein to promote Rad51 recombinase function after DNA damage. In fact, cells lacking functional BRCA2 exhibit genomic instability and sensitivity to DNA-damaging agents such as etopside, bleomycin and X-rays [9,10].
The ribonucleoside anticancer drug, 1-(3-C-ethynyl-β-D-ribo-pentofuranosyl)cytosine (ECyd, TAS106) inhibits RNA synthesis through competitive inhibition of RNA polymerase ( Figure 1) [11]. TAS106 rapidly undergoes phosphorylation to a 5'-triphosphate form (ECTP) after its uptake into cells, and ECTP strongly inhibits RNA polymerase to cause RNA synthesis inhibition [12,13]. Furthermore, Naito et al. have demonstrated that TAS106 strongly induces JNK-dependent apoptosis through activation of an RNase L-mediated RNA degradation pathway [14]. In the phosphorylation of TAS106, uridine/cytidine kinase (UCK) is responsible for the first phosphorylation of TAS106 to the 5'-monophosphate form. The UCK activity in tumor cells is higher than that in non-tumor cells, thereby causing the accumulation of TAS106 preferentially in tumor cells [13][14][15][16]. We have previously reported that a sublethal dose of TAS106 strongly suppresses the expression of anti-apoptotic proteins and G2/M checkpoint-related proteins, and enhances radiation-induced cell death and growth delay in gastric tumor cell lines MKN45, MKN28 and murine rectum adenocarcinoma cell line Colon26 in vitro [17] and in vivo [18]. Furthermore, this radiosensitizing effect is also observed in radioresistant hypoxic cells through the inhibition of hypoxia inducible factor 1α (HIF-1α) expression [19]. However, the precise mechanism underlying TAS106-induced radiosensitization remains elusive.
In this study, to further examine the mechanism of TAS106-induced radiosensitization, we investigated whether TAS106 could modify the repair capacity of DNA DSBs. We demonstrate that TAS106 decreases cellular DNA DSB repair capacity and radiosensitizes human lung carcinoma A549 cells. In addition, we show that this radiosensitizing effect is mainly due to abrogation of the HR pathway through the suppression of BRCA2 expression.

TAS106 enhances radiosensitivity in tumor and immortalized cells
To determine whether pretreatment with TAS106 enhanced the radiosensitivity of tumor and immortalized cells, we performed a clonogenic survival assay. Figure  2A shows the X-ray dose-response curves for cell survival in A549 cells pretreated with TAS106 at various concentrations. Pretreatment with TAS106 suppressed the clonogenic cell survival in a concentration-dependent manner and the 10% lethal dose (D 10 ) of the surviving fraction was reduced from 7.88 Gy in the control to 5.24 Gy by the treatment with 1 μM TAS106. The sensitizer enhancement ratio (SER) judged by the D 10 was 1.50, indicating the increase of sensitivity to X-irradiation induced by TAS106. In addition, HEp-2 cells and V79 cells pretreated with TAS106 also exhibited sensitization to X-irradiation as shown in Figures 2B  and 2C. SER values for HEp-2 and V79 cells were 1.59 and 1.28, respectively. Furthermore, TAS106 increased the α values in all cell lines tested (Table 1). These results indicated that pretreatment with TAS106 enhanced radiosensitivity in various cell lines.

TAS106 suppresses cellular DSB repair capacity
To investigate whether TAS106 radiosensitized tumor cells by inhibiting DNA DSB repair, we measured the sublethal damage (SLD) repair in A549 cells. The SLD assay is based on evidence that cell survival increases with extended interval times between two split doses of X-rays, provided that the cells can repair the initial DSBs prior to the second irradiation [20]. Therefore, SLD repair is considered to reflect the cellular DSB repair capacity. As shown in Figure 3A, the survival ratio in A549 cells exposed to fractionated irradiation increased as the interval time extended up to 6 h and then decreased at 12 h. The maximum increase of the survival ratio was 1.4 at the 6 h interval time. In contrast, there was little increase in the survival ratio of A549 cells pretreated with TAS106. This result suggested that the pretreatment with TAS106 suppressed DNA DSB repair in A549 cells.
To further support these data, we analyzed DSBs in Xirradiated cells by γ-H2AX and 53BP1 foci formation assay. Histone H2AX is known to be phosphorylated at serine 139 (γ-H2AX) immediately after DSB induction, and then γ-H2AX forms nuclear foci in the region of the DSBs, and subsequently undergoes dephosphorylation after the repair of DNA strand breaks. p53 binding protein 1 (53BP1) also localizes at the DSBs region. Therefore, the numbers of γ-H2AX and 53BP1 foci are used as a measure of the relative amount of DSBs and repair kinetics [21,22]. When cells were irradiated at 1 Gy, the average number of γ-H2AX foci per cell peaked at 30 min after irradiation and decreased with time for both cells with or without TAS106 ( Figure 3B). There were no significant differences in the numbers of γ-H2AX foci between cells without and with TAS106 30 min after irradiation. However, at 1 h, 2 h and 6 h after irradiation, the numbers of foci in cells pretreated with TAS106 were higher than in cells without TAS106. In addition, the average number of 53BP1 foci in cells treated with TAS106 was significantly greater than that in cells without TAS106 at 6 h after irradiation. These results suggested that TAS106 inhibited DSB repair.  The data were fitted using the linear-quadratic model, SF = exp(-αD-βD 2 ). The 10% lethal dose (D 10 ) was calculated from the α and β values. The sensitizer enhancement ratio (SER) was calculated from the each values with or without TAS106.

TAS106 suppresses the expression of DSB repair proteins
To explore the mechanism of DSB inhibition by TAS106, we examined the effect of TAS106 on the expression levels of DSB repair-related proteins in A549 cells. As shown in Figure 4A, the treatment with 1 μM TAS106 for 24 h clearly suppressed the expression levels of BRCA2 and Rad51, which are the key proteins of the HR pathway. The expression levels of Mre11 and NBS1, which are constituent proteins of the MRN complex, were slightly reduced by TAS106. On the other hand, there was no obvious change in the expression levels of DNA-PKcs and Ku70, which are involved in the NHEJ pathway. It has been reported that TAS106 inhibits RNA synthesis and regulates the expression of its target proteins at the mRNA level [12,13]. Therefore, we examined the effect of TAS106 on the mRNA expression levels of DSB repair-related proteins using semiquantitative RT-PCR. Figure 4B shows that the treatment with 1 μM TAS106 suppressed the mRNA levels of BRCA2, Rad51, Mre11 and NBS1 in a timedependent manner. However, the mRNA levels of DNA-PKcs and Ku70 were not changed. These tendencies had good correlation with the results of Western blot analysis. These results suggested that TAS106 suppressed HR-related proteins but not NHEJ-related proteins.
Down-regulation of BRCA2 and Rad51 was maintained for 12 h after the removal of TAS106 For TAS106-induced down-regulation of BRCA2 and Rad51 to result in tumor radiosensitization, it needs to be maintained during the period of initial DSB repair after X-irradiation. Therefore, we assessed how long this inhibitory effect was prolonged after the removal of TAS106. A549 cells were treated with 1 μM TAS106 or vehicle for 24 h and replaced with fresh medium. After exposure to X-rays at 10 Gy, cells were collected at each time point. In cells exposed to X-irradiation alone there was no change in the expression levels of BRCA2 and Rad51 ( Figure 5A). In contrast, TAS106 reduced the expression levels of BRCA2 and Rad51 and this persisted up to 12 h after X-irradiation ( Figure 5B). These results suggested that TAS106 reduced the expression levels of HR-related proteins for a period long enough to inhibit the repair of radiation-induced DSBs after its removal.
Down-regulation of BRCA2 is responsible for the radiosensitizing effect of TAS106 Based on the results described above, we hypothesized that the radiosensitizing effect of TAS106 was mainly due to abrogation of the HR pathway through the down-regulation of BRCA2. To test this hypothesis, we next compared the radiosensitizing effect of TAS106 in BRCA2-deficient V-C8 cells with that in parental V79 cells by clonogenic survival assay. It has been shown that the HR-pathway-mediated DNA damage repair is impaired in BRCA2-deficient V-C8 cells [23]. While a significant radiosensitizing effect by TAS106 was observed in V79 cells, the treatment with 1 μM TAS106 did not induce any radiosensitizing effect in V-C8 cells ( Figure 6). In V79 cells, the pretreatment with 1 μM TAS106 reduced the D 10 of the surviving fraction from 9.78 Gy to 7.67 Gy and its SER was 1.28. In V-C8 cells, the D 10 of the surviving fraction was reduced from 5.09 Gy to 4.51 Gy and SER was 1.13. In addition, SER calculated from α value in V79 cells was also higher than that in V-C8 cells (Table 1). Therefore, this result suggested that down-regulation of BRCA2 by TAS106 suppressed the HR pathway of DSB repair, leading to the radiosensitizing effect.

Discussion
The DNA repair pathway is a promising target for cancer radiotherapy because intrinsic DNA repair pathway enables tumor cells to survive by repairing radiationinduced DNA lesions. Therefore, there have been several studies aimed at radiosensitizing tumor cells by modulating DNA repair-related molecules. For example, the ATM inhibitor KU-55933 shows high specificity and induces radiosensitizing effects in tumor cells [24]. Gemcitabine and Gimeracil, which disrupt the metabolic pathway for nucleic acids, were also reported to have radiosensitizing effects through the inhibition of HRmediated DSB repair [25,26].
We have previously reported that TAS106 enhances X-irradiation-induced apoptosis and reproductive cell death regardless of p53 status in tumor cells in vitro and in vivo [17,18]. The down-regulation of survivin, a key protein regulating apoptosis, and the abrogation of arrest at the G2/M phase by TAS106 are partly responsible for the enhancement of cell death [16]. Furthermore, TAS106 enhances X-irradiation-induced apoptosis even under hypoxic conditions through the down-regulation of HIF-1α [18]. Although TAS106 exhibits a clear radiosensitizing effect, its effect on DNA DSBs, which are the most lethal DNA lesions caused by X-irradiation, has not been evaluated yet. Therefore, we assessed the effect of TAS106 on the repair of radiation-induced DSBs.
In the present study, pretreatment with TAS106 enhanced radiation-induced cell death in tumor cell lines A549 and HEp-2 as well as the immortalized cell line V79 (Figure 2). To determine whether this radiosensitizing effect by TAS106 could be explained by the suppression of DNA repair capacity, we next investigated the effect of TAS106 on SLD repair. It has been reported that SLD repair reflects the cellular DSBs repair capacity mediated by the HR pathway [27,28]. As shown Figure 3A, TAS106 suppressed the SLD repair in A549 cells exposed to fractionated irradiation. In addition, the average numbers of radiation-induced γ-H2AX and 53BP1 foci in TAS106-pretreated cells were higher than in control cells up to 6 h after X-irradiation, as shown in Figure 3B and 3C. These results suggested that TAS106 suppressed DSB repair through inhibition of the HR pathway.
To clarify the molecular mechanisms of DSB repair inhibition by TAS106, we tested the effect of TAS106 on the expression levels of DSB repair-related proteins using Western blot analysis. The expression of NHEJrelated proteins, DNA-PKcs and Ku70, was not affected by TAS106. On the other hand, TAS106 suppressed the expression of the HR-related proteins BRCA2 and Rad51 ( Figure 4A). Therefore, our results suggested that down-regulation of Rad51 and BRCA2 by TAS106 inhibited the HR pathway, leading to the radiosensitizing effect, as shown in Figure 2. RAD51 is a crucial component of the HR pathway and its down-regulation enhances radiosensitivity in tumor cells [29,30]. It has also been reported that deficiency or down-regulation of BRCA2 results in high sensitivity to X-irradiation due to  insufficient DSB repair [10,31]. Recent studies reported that loss of HR capacity in BRCA2-deficient cells was restored by overexpression of wild-type Rad51 [32,33]. Therefore, although TAS106 reduced the expression levels of BRCA2 and Rad51, it might be possible that the down-regulation of Rad51 by TAS106 is more influential than that of BRCA2.
Down-regulation of BRCA2 and Rad51 was maintained up to 12 h after the removal of TAS106 and the following X-irradiation ( Figure 5B). As shown in Figure  3B, DSBs were mostly repaired within 6 h after irradiation in control cells. Therefore, the maintenance of this down-regulation likely contributed to the inhibition of DSB repair by TAS106. Additionally, TAS106 reduced the mRNA levels of HR-related proteins in consistent with the result of Western blot analysis in A549 cells ( Figure 4). These results suggested that TAS106 downregulated these proteins transcriptionally, and supported the results that TAS106 inhibited RNA transcription by suppressing RNA polymerase II in previous studies [34][35][36].
To further support our findings, we compared the radiosensitizing effect of TAS106 in BRCA2-deficient V-C8 cells with that in parental V79 cells. TAS106 sensitized V79 cells, but not V-C8 cells, to X-irradiation (Figure 6). In addition, we examined the effect of TAS106 on the expression level of Rad51 in V79 and V-C8 cells. TAS106 did not change it in both cells (data not shown). These results suggested that the down-regulation of BRCA2, rather than Rad51, was primarily attributable to the suppression of cellular DNA repair capacity and the radiosensitizing effect in TAS106-treated cells.

Conclusions
We have demonstrated that TAS106 suppresses the repair of radiation-induced DSBs and enhances radiosensitivity, which is mainly associated with the inhibition of HR pathway through the down-regulation of BRCA2. DSBs are the main target of cancer radiotherapy, and targeting inhibition of DNA repair is one approach to improve the efficiency of it. Therefore, TAS106 could be a good molecular candidate to achieve it, and the combination of TAS106 and X-irradiation may be an effective strategy for enhancing tumor cell death.
Cell culture, X-irradiation and drug treatment Human lung carcinoma cell line A549, human larynx squamous carcinoma cell line HEp-2 and Chinese hamster fibroblast cell line V79 were grown in RPMI 1640, DMEM and α-MEM medium containing 10% fetal bovine serum at 37°C in 5% CO 2 , respectively. Chinese hamster fibroblast cell line V-C8, derived from V79, was grown in Ham's F-10 medium containing 10% fetal bovine serum. X-irradiation was performed with a Shimadzu PANTAK HF-350 X-ray generator (1.0 mm Al filter, 200 kVp, 20 mA, Shimadzu, Kyoto, Japan). Cells were treated with TAS106 for 6 h (V79 and V-C8) or 24 h (A549 and HEp-2) and subsequent X-irradiation was performed in the absence of TAS106.

Clonogenic survival assay
Cells were seeded on 6-cm dishes and treated with TAS106 at the indicated concentrations for 6 h or 24 h. Then they were washed twice with PBS and replaced with fresh medium. Immediately after replacement, cells were exposed to X-rays and incubated for 7-14 days. Following this they were then fixed with methanol and stained with Giemsa solution (Sigma-Aldrich). Colonies containing more than 50 cells were scored as surviving cells. In A549 cells, 64, 48 and 29% of cells were alive at the concentration of 0.5, 0.75 and 1 μM TAS106, respectively. In addition, cell survivals after TAS106 treatment were 25, 98, and 35% at 0.1 μM for HEp-2, 1 μM for V79 and 1 μM for V-C8, respectively. Each surviving fractions were corrected using these cell survivals. The survival curves were fitted to a linear-quadratic model by data analysis software Origin Pro 7 (OriginLab Co. Northampton, MA).
Sublethal damage (SLD) repair capacity was measured as the increase of surviving cells after irradiation with a split dose at the indicated interval. Cells treated with 1 μM TAS106 for 24 h were washed twice with PBS and replaced with fresh medium. Immediately after replacement, cells were exposed to X-rays (2.5 Gy) and incubated for 0-12 h for the repair of SLD. At the indicated times, the cells were exposed to X-rays (2.5 Gy) again and incubated for 10 days.

Immunofluorescent staining for g-H2AX and 53BP1
At the indicated times after TAS106 treatment and Xirradiation, cells attached on glass coverslips were fixed in 4% paraformaldehyde/PBS for 30 min at room temperature. After being permeabilized with PBS containing 0.5% Triton X-100 for 5 min at 4°C, cells were treated with PBS containing 6% goat serum for 30 min at room temperature. Then they were incubated with the anti-g-H2AX antibody at a 1:500 dilution or the anti-53BP1 antibody at a 1:1000 dilution in 3% goat serum overnight at 4°C. Cells were then incubated in the dark with the Alexa Fluor ® 488-conjugated anti-mouse or antirabbit secondary antibody at a 1:500 dilution for 1.5 h. After incubation, they were counterstained with 300 nM 4',6'-diamidino-2-phenylindole (Invitrogen) for 5 min at room temperature. Coverslips were mounted with Prolong Gold antifade reagent (Invitrogen). Fluorescent microscopic analysis was performed using an Olympus BX50 microscope with reflected light fluorescence and foci were counted manually.