Fang Wang · Yi Xiao · Jun-Fang Yan · Guo-Min Huang · Jin-Hua Zhang ·Cui-Xia Di · Jing Si· Hong Zhang· Ai-Hong Mao

Abstract Carbon ion radiotherapy has the advantages of better therapeutic effect and fewer side effects compared with those of X-rays in many kinds of tumors, including prostate cancer, and thus is an attractive treatment approach for prostate cancer. However, the biological effects and underlying mechanisms of carbon ion irradiation in prostate cancer are not yet fully understood.Therefore, this study systematically compared the effects of carbon ion irradiation with those of X-ray irradiation on DNA damage response and found that carbon ion irradiation was more effective than X-ray irradiation. Carbon ion irradiation can induce a high level of DNA double-strand break damage, reflected by the number of γ-H2A histone family member X foci, as well as by the foci lasting time and size.Moreover,carbon ion irradiation exhibited strong and long-lasting inhibitory effect on cell survival capability, induced prolonged cell cycle arrest, and increased apoptosis in PC-3 cells. As an underlying mechanism, we speculated that carbon ion irradiation-induced DNA damage evokes cell cycle arrest and apoptosis via the pRb/E2F1/c-Myc signaling pathway to enhance the radiosensitivity of p53-deficient prostate cancer PC-3 cells. Collectively,the present study suggests that carbon ion irradiation is more efficient than X-ray irradiation and may help to understand the effects of different radiation qualities on the survival potential of p53-deficient prostate cancer cells.

Keywordst Carbon ion irradiation · DNA damage · Cell cycle arrest · Apoptosis · C-Myc · Prostate cancer

1 Introduction

Carbon ion radiotherapy (CIRT) is considered the most advanced radiotherapy(RT)technology in the world today and has gained increased interest in the field of high-precision cancer therapy[1].Compared with conventional RT with low-linear energy transfer (LET) photons, CIRT with high-LET carbon-ion beams has several advantageous properties; for example, the localization of radiation dose to the tumor is more accurate,the probability of damage to surrounding healthy tissue is lower, the relative biological effectiveness is higher, and the ability to repair radiation injury is lower [2–6]. In addition, it can effectively kill hypoxia-or RT-resistant tumor cells and has a killing effect on tumor cells in each cell cycle. Based on these advantages, some new heavy-ion therapy facilities are under development or investigation worldwide,which will lead to an increasing number of cancer patients receiving CIRT in the future. Thus, it is of great significance to identify the biological effects and underlying mechanism of CIRT, as well as its advantages compared with conventional RT.

CIRT has shown remarkable efficacy [7–9] in some tumors, including prostate cancer [10]. Prostate cancer is the second most common cancer in men worldwide, with an annual incidence of ～1,100,000 cases, which resulted in 307,000 mortalities in 2012 [11]. Currently, prostate cancer is the most common type of cancer treated with carbon ion technology. Compared with conventional RT,CIRT can improve the 5-year tumor-free survival rate of patients with prostate cancer by 15 to 20% [12–14].Therefore, CIRT is expected to become an attractive and ideal treatment modality for patients with prostate cancer.

Previous studies have reported that heavy ions can induce clustered DNA double-strand breaks (DSB), which are difficult to repair [15, 16], and it has been shown that DNA damage repair patterns induced by high-LET irradiation and low-LET irradiation are different [17, 18].Therefore,DNA damage response pattern could have a key role in carbon ion irradiation (CIR) and lead to different biological effects in different irradiation types. However,the biological effects and underlying mechanisms of CIR in prostate cancer are not yet fully understood.c-Myc was the first proto-oncogene to be identified. Overexpression of c-Myc is associated with various human tumors [19]. In prostate cancer, approximately 30% of patients show c-Myc overexpression, which is closely related to high grade and poor prognosis of prostate cancer [20–22].c-Myc not only participates in the initiation and progression of a variety of human malignant tumors but also influences the effects of radiotherapy and chemotherapy[23–25]. However, the potential role and mechanism of c-Myc in irradiation-induced DNA damage response remain unclear. Thus, this study systematically compared the effects of CIR with those of X-ray irradiation(XRR)on DNA damage response in the p53-deficient prostate cancer cell line PC-3, which is not sensitive to conventional RT.Results indicated that CIR induced more severe DSB damage compared with that induced by X-rays, which caused an enhanced and durable inhibition in cell survival,stronger and longer-lasting cell cycle delays, and higher rates of apoptosis. The underlying mechanism may be that CIR-induced DNA damage evokes cell cycle arrest and apoptosis via the pRb/E2F1/c-Myc signaling pathway to enhance the radiosensitivity of PC-3 cells. Thus, this research explored the biological effects and mechanism underlying CIR in prostate cancer cells and presented preliminary data that could facilitate the application of CIR in the clinical treatment of prostate cancer.

2 Experimental section

2.1 Cell line

The human prostate cancer cell line PC-3 was obtained from the Committee on Preservation of Typical Cultures of the Chinese Academy of Sciences. Cells were cultured in Minimum Essential Medium (Gibco, #11095080) supplemented with 10%fetal bovine serum(Gibco,#10099141C)at 37 °C in a humidified 5% CO2incubator.

2.2 Irradiation procedure

Cells were plated in a 35-mm culture dish and subsequently irradiated, and XRR was carried out using the Faxitron RX-650 biological irradiation system (Faxitron Bioptics,USA).The irradiation conditions were as follows:Energy and dose rate were 100 kVp and 0.52 Gy/min,respectively. CIR (12C6+) was performed at the Heavy Ion Research Facility in Lanzhou, China. The irradiation conditions were as follows: Energy was approximately 81 MeV/u, LET was 32.54 keV/μm, and dose rate was approximately 2 Gy/min. In both cases, the beam was orthogonal to the cell layer and passed through a 1-mmthick plate and 2-mm high liquid surface before reaching the cells. The irradiation field was a circle with a diameter of 5 cm.

2.3 Cell viability assay

Cell viability was determined as previously described[26]. Cells were irradiated with the indicated doses of X-rays and carbon ion beams and then seeded at 5 × 103cells per well into 96-well plates with six replicates in each group. The viability of cells was estimated 24 h after irradiation using a microplate reader(Tecan Infinite M 200,Switzerland).

2.4 Clonogenic survival experiment

Cells were plated in Φ60 dishes at 1000 cells per dish in triplicate following irradiation. After incubating at 37 °C for 13 days, colonies were fixed and stained as previously described [26]. The colonies of each dish were counted visually, and the survival curve was calculated using a linear quadratic model.

2.5 Cell cycle assay

PC-3 cells were first irradiated with the indicated doses of X-rays and carbon ion beams. The next day, detached and fixed cells were prechilled with 75% ethanol prepared in PBS for ＞24 h at - 20 °C.On the day of detection,the samples were centrifuged at 800 × g for 5 min to discard 75%ethanol and rehydrated with 2–5-mL PBS for 15 min.The cells were then resuspended in 200-μL DNA staining buffer (MultiSciences, #CCS012) and incubated for 30 min. Data were collected using a FlowSight flow cytometer (Merck, Germany) and analyzed with FlowJo-V10.

2.6 Apoptosis detection

For cell apoptosis, irradiation was as described for the cell cycle assay. After incubation for 48 h, the cells were detached and stained as previously described [26]. Cell apoptosis data were obtained using Flowsight (Merck,Germany) and analyzed using IDEAS (Merck, Germany).

2.7 Hoechst 33258 staining

PC-3 cells were plated into Φ35 dishes with 2 × 105cells per dish in triplicate and then irradiated. After irradiation, confluent cells were detached and plated into 6-well plates with coverslips at a density of 104–105. The cells were cultured in a 5% CO2incubator for 48 h,washed, and fixed for 30 min at 20–25 °C. Subsequently,0.5 μg/mL Hoechst 33258 dye solution (prepared by PBS)was evenly dropped onto the coverslips and dyed in the dark at 20–25 °C for 15 min. The coverslips were washed three times with PBS and sealed with glycerin and PBS(1:9) mixture on dry coverslips. The prepared slides were observed using a fluorescence microscope (Nikon, Japan)to identify apoptotic cells.

2.8 Cell transfection

Small interfering RNA(siRNA)targeting c-Myc and the negative control were obtained from Guangzhou RiboBio.C-Myc siRNA sequences were: si–c-Myc-1 (S1), 5′-GC UUCACCAACAGGAACUATT-3′/3′-TTCGAAGUGGU UGUCCUUGAU-5′; si–c-Myc-2 (S2), 5′-GGAAACGAC GAGAACAGUUTT-3′/3′-TTCCUUUGCUGCUC UUGU CAA-5′; si–c-Myc-3 (S3), 5′-CCACACAUCAGCACAA CUATT-3′/3′-TTGG UGUGUAGUCGUGUUGAU-5′;and siRNA negative control (NC), 5′-UUCUCCGAACGU GUCACGUTT-3′/3′-TTAAGAGGCUUGCACAGUGCA-5′. The transfection concentration was 50 nM. Cells were seeded at 50–60% confluence ～24 h before transfection,and transfection was proceeded with Lipofectamine?3000(Thermo Fisher Scientific, #L3000015).

2.9 Western blot

PC-3 cells were irradiated and harvested after incubation at 37 °C for 48 h. Cells were lysed using RIPA buffer supplemented with PMSF (1%) (Solarbio, #R0010-100 mL). Protein concentration was quantified using the protein BCA assay kit(BOSTER,#AR0146).After routine denaturation, proteins were fractionated in 4–15% SDSPAGE and transferred to PVDF membrane pre-activated with methanol (GE Healthcare, USA). The membrane was blocked with 5% BSA for 1 h at 20–25 °C. Further, antibodies against cyclin B1 (1:1000 dilution, #4138), Rb(1:1000 dilution, #9313),p-Rb (S807/811)(1:500 dilution,#8516), cytochrome C (1:1000 dilution, #4280), E2F1(1:1000 dilution, #3742), c-Myc (1:1000 dilution, #9402),caspase 3 (1:500 dilution, #9662; Cell Signaling Technology, USA), Bcl-2 (1:250 dilution, #sc-7382), and GAPDH(1:500 dilution, #sc-47724)(Santa Cruz Biotechnology,USA) were used and incubated overnight at 4 °C. The membrane was then removed and washed, and anti-mouse(1:10,000 dilution, #7076) or anti-rabbit IgG secondary antibodies (1:10,000 dilution, #7074; Cell Signaling Technology, USA) was added and incubated for 1 h at 20–25 °C. The signals were detected by chemiluminescence using ImmobilonTMWestern Chemilum HRP Substrate (Millipore, #WBKLS0500) with a super-sensitive multifunctional imager (AI680;GE Healthcare,Germany).

2.10 Immunofluorescence

Cells plated on coverslips were irradiated and fixed at different times. The samples were then washed and permeabilized for 10 min with 0.1% Triton X-100. After permeabilization, the cells were blocked with 5% goat serum for 1 h at 20–25 °C.Subsequently,γH2AX antibody(1:250 dilution, #ab11174; Abcam, Cambridge, UK) was added, followed by incubation at 4 °C overnight. The next day, AlexaFluor488 labeled secondary antibody (1:500 dilution,#ab150077)(Abcam,UK)was added.Cells were then washed with PBS (0.01 M), and nuclei were counterstained using DAPI(Thermo Fisher Scientific,#D1306).The prepared cells were observed using a confocal laser microscope (Carl Zeiss, Germany).

2.11 Statistical analysis

Results were obtained from at least three independent replicates and presented as mean ± standard deviation(SD). Two-tailed Student’s t test and analysis of variance(ANOVA) completed with Tukey’s post hoc test were applied to calculate statistical significance using GraphPad Prism 7.0. P value ＜0.05 was considered as statistically significant.*P ＜0.05;**P ＜0.01.

3 Results

3.1 CIR induces high levels of DNA DSB damage

To explore the effects of CIR and XRR on DNA damage response, the time response of the DSB marker γ-H2A histone family member X(γH2AX)was measured.Results indicated that γH2AX induction reached a maximum level at 3 h after irradiation for both irradiation types and that CIR induced more γH2AX foci at different time points than those induced by XRR,with CIR-induced foci lasting for a longer duration (Fig. 1a and b). In addition, carbon ions induced larger foci compared with those induced by X-rays when assessed using 3D images (Fig. 1c). These results suggest that CIR can induce higher levels of DSB damage than those induced by XRR, which was reflected not only by γH2AX foci numbers but also by the foci lasting time and size.

3.2 CIR exhibits strong and long-lasting inhibitory effect on cell survival capability

To further evaluate the effects of CIR and XRR on cell viability, we performed cell cytotoxicity and colony formation experiments.The results of the cell cytotoxicity test showed that both types of irradiation inhibited cell viability in a time-and dose-dependent manner,and CIR exhibited a stronger inhibitory effect than that of XRR(Fig. 2a and b).Moreover, carbon-ion-induced cell viability inhibition lasted for ≤72 h after irradiation, while the inhibition of X-rays lasted for only 48 h (Fig. 2a).

Similarly, colony formation experiments indicated that survival fraction decreased as dosage increased in PC-3 cells when exposed to both types of radiation (Fig. 2c).According to the comparison of survival curves, the sensitivity of PC-3 cells to CIR was higher than that to XRR(Fig. 2d). Thus, CIR exhibited stronger and longer-lasting inhibitory effect on cell survival than that of XRR.

3.3 CIR evokes a more pronounced shift of the cell cycle distribution

Cell cycle distribution is closely related to radiation response. Therefore, the effects of CIR and XRR on cell cycle distribution were examined. We found that the proportion of CIR cells in the G2/M phase significantly increased compared with that in XRR cells(Fig. 3a).Thus,CIR evoked a more pronounced shift in cell cycle distribution from G1to G2phase,and G2/M arrest exhibited a dose-dependent effect (Fig. 3b and c).

3.4 CIR leads to high rates of apoptosis

The effects of CIR and XRR on apoptosis were investigated. First, apoptosis was assessed via flow cytometry,and we found that along with stronger DSB damage and cell cycle arrest,CIR resulted in higher apoptotic rates than those caused by XRR. The rates of apoptosis were 27.34 and 37.93% after 2 and 4 Gy CIR, respectively (versus 14.1 and 23.59%following 2 and 4 Gy XRR,respectively;Fig. 4a and b).

Cell apoptosis was further verified by Hoechst 33,258 staining, which revealed the nuclear morphological changes in the cells (Fig. 4c and d). Results showed that apoptotic cells could form pyknotic apoptotic bodies with condensed chromatin (indicated by arrow), and there were more apoptotic bodies in carbon-ion-irradiated cells.

3.5 CIR regulates the radiosensitivity of PC-3 cells through pRb/E2F1/c-Myc signaling pathway

The above results indicate that CIR-induced DNA damage is more likely to evoke cell cycle arrest and apoptosis in PC-3 cells. Subsequently, we attempted to understand why CIR is more effective than XRR in PC-3 cells. PC-3 is a p53-deficient prostate cancer cell line, and accumulating evidence shows that E2F1 may have an important role in the CIR response of p53-deficient cells;thus, the present study investigated whether E2F1 is involved in the CIR response.We found that compared with XRR,CIR significantly suppressed the expression of E2F1,and pRb, which is a key regulator of E2F1, was also suppressed by CIR (Fig. 5a and b). In addition, some critical proteins involved in G2/M arrest and apoptosis were detected, indicating that CIR significantly affected the expressions of proteins related to G2/M transition and apoptosis (Fig. 5a and b). Interestingly, we observed that CIR dramatically suppressed the expression of c-Myc.(Fig. 5c and d). E2F1 can act as a transcription factor to regulate the expression of many genes; thus, we wondered whether E2F1 can also regulate the expression of c-Myc.Using the transcription factor prediction database JASPAR[27],we predicted the sequence logo of E2F1(Fig. 5e)and found that there were several E2F1 binding sites in the c-Myc promoter region(Fig. 5f),suggesting that E2F1 may regulate the expression of c-Myc.To further verify this,the GTRD database was used for further analysis,and we found that there was indeed one E2F1enriched region in the c-Myc promoter (Fig. 5g) [28]. Our previous study showed that c-Myc plays an important role in ionizing radiation [29];thus, we downregulated the expression of c-Myc using siRNA to detect its role in ionizing radiation in PC-3 cells(Fig. 6a and b). Downregulation of c-Myc enhanced the radiosensitivity of PC-3 cells (Fig. 6c and d). Thus, we speculated that CIR killed prostate cancer cells more efficiently than XRR partly via the pRb/E2F1/c-Myc signaling pathway to enhance the radiosensitivity of PC-3 cells.

Fig. 1 CIR induces higher levels of DSB damage than those of XRR. a Microscopic evaluation of γH2AX induction at the indicated time points after 2 Gy CIR or XRR in PC-3 cells.The scale bar means 20 μm.b The γH2AX foci numbers per cell at the indicated time points after 2 Gy CIR or XRR in PC-3 cells. c CIR induces larger γH2AX foci than those induced by XRR. Microscopic evaluation of the sizes of the γH2AX foci at 6 h following 2 Gy CIR or XRR in PC-3 cells.The scale bar means 20 μm.*P ＜0.05; **P ＜0.01

Fig. 2 CIR exhibited stronger and longer-lasting inhibitory effect on cell survival capability than that of XRR.a Cell viability of PC-3 cells treated at the indicated times after 2 Gy CIR or XRR.b Cell viability of PC-3 cells treated with the indicated doses of CIR or XRR at 24 h after IR. (c, d) Colony formation by PC-3 cells treated with the indicated doses of CIR or XRR. SF(%) = αD + βD2. *P ＜0.05;**P ＜0.01

4 Discussion

Prostate cancer was the second most common cancer in men worldwide in 2012,and radiotherapy is a conventional therapy approach for localized prostate cancer patients[29, 30]. However, radioresistance severely limits the efficacy of conventional RT [31, 32]. Thus, enhancing the efficacy of RT is becoming increasingly important in prostate cancer. Charged particle irradiation, such as CIR,can produce a highly lethal effect on radioresistant tumors[33].Furthermore,prostate cancer patients receiving CIRT have good local tumor control and survival rates, particularly for high-risk patients [13, 14]. However, the biological effects and underlying mechanisms of CIR in prostate cancer are not yet fully understood. This study focused on the DNA damage response of CIR versus that of XRR in prostate cancer cells. We found that CIR produced higher levels of DSB damage and induced larger and longerlasting γH2AX foci than those of XRR. Costes et al. [34]revealed that high-LET irradiation (130 keV/mm nitrogen ions) can induce large γH2AX foci, and the size and frequency of irradiation-induced foci can be used as evaluation indices of irradiation quality; therefore, large γH2AX foci may be a characteristic of high-LET irradiation. In addition, it has been reported that high-LET irradiation can induce clustered DSB damage, which is difficult to repair [35], which may be related to the stronger,longer-lasting,and larger γH2AX foci observed in our study. Moreover, these effects may be consistent with changes in cell viability, cell cycle, and apoptosis following CIR.

Fig.3 CIR evoked a more pronounced shift in cell cycle distribution than that evoked by XRR.a Cell cycle was investigated via flow cytometry.Cell cycle distribution at 24 h following 2 Gy b or 4 Gy c CIR or XRR. *P ＜0.05; **P ＜0.01

CIR-induced DNA damage induces cell cycle arrest at the G2/M phase to prevent replication and segregation of the damaged genome and initiate DNA repair responses[36–38]. Numerous pathways participate in G2/M arrest;however, the specific molecular mechanism is still unknown. Previous studies have suggested that cyclin B1-CDC2 complexes play an important role in G2/M conversion [39, 40]. The present results indicate that ionizing irradiation may significantly reduce the expression of cyclin B1, especially CIR. Moreover, another important cell cycle regulatory protein, Rb, has been shown to participate in G2/M arrest [39]. Rb is an important substrate of cyclin-dependent kinases, and once Rb is phosphorylated, the E2F1 protein is released to activate downstream gene expression[41].The present study found that c-Myc, which is a potential target of E2F1, was dramatically suppressed by CIR.As one of the most important oncogenes, activated c-Myc is an important molecular marker of many cancers[42].Targeted inhibition of c-Myc expression may be an effective strategy for the treatment of human cancer[43];thus,the strong inhibition of c-Myc by CIR may be a reason for its high biological effects.On the contrary, as a transcription factor, c-Myc regulates 15% of human gene expression and participates in various cellular processes [44]. Therefore, inhibiting the expression of c-Myc by CIR can further regulate cell proliferation, cell cycle, and apoptosis, resulting in strong lethal effects of CIR. Moreover, both mentioned previous studies and our present study have revealed that the downregulation of c-Myc can enhance the radiosensitivity of cancer cells[29],suggesting that c-Myc is not only involved in regulating the initiation and progression of various tumors but also plays an important role in the radiation response of cancer cells.In conclusion,as a transcription factor that plays important roles in tumor development and radiation response, the significant decrease of c-Myc in CIR rather than in traditional XRR is likely to have a great influence on the biological effects of CIR.In addition,we hypothesized that the CIR response may be partly through the pRb/E2F1/c-Myc signaling pathway; however, we observed that the expression of c-Myc significantly changed after CIR. Therefore,other factors may be involved in the regulation of c-Myc expression, and we hope to further investigate this regulatory network to better clarify the underlying mechanisms.

Fig. 4 CIR led to high rates of apoptosis. a Apoptosis was investigated via flow cytometry.b The percent of apoptotic cells measured with flow cytometry at 48 h after 2 Gy or 4 Gy CIR or XRR. c Apoptosis was assessed via Hoechst 33,258 staining. d The percent of apoptotic cells assessed with Hoechst 33,258 staining at 48 h after 2 Gy or 4 Gy CIR or XRR.*P ＜0.05; **P ＜0.01

In summary, carbon ion radiotherapy has advantageous properties and is a viable therapeutic option for patients with cancer.In the present study,we mainly focused on the biological effects and underlying mechanisms of CIR in prostate cancer. However, many questions remain unanswered and require further inquiry. A new heavy-ion therapy facility, Wuwei Heavy-ion Therapy Hospital(Gansu, China), has recently begun to be used for clinical treatment. Subsequently, we hope to explore the effects of CIR in patients with prostate cancer treated in Wuwei Heavy-ion Therapy Hospital to further understand the role of heavy ions in prostate cancer in vivo.

Fig. 5 c-Myc is a potential target of E2F1. a Western blot and b grayscale analysis of the expression levels of E2F1, p-Rb, Rb,cyclin B1, Bcl-2, cytochrome C, and caspase 3 after 2 Gy CIR or XRR. c Western blot and d grayscale analysis of c-Myc expression after 2 Gy CIR or XRR. e The sequence logo of E2F1 predicted through the transcription factor prediction database JASPAR. f The E2F1 binding sites in the region of c-Myc promoter predicted through the transcription factor prediction database JASPAR.g E2F1-enriched region in the c-Myc promoter using the GTRD database.*P ＜0.05;**P ＜0.01

Fig.6 Downregulation of c-Myc enhances the radiosensitivity of PC-3 cells. a Western blot and b grayscale analysis of c-Myc expression after transfection with siRNA-c-Myc or negative control(NC).c Cell viability and d colony formation by c-Myc-knockdown (si–c-Myc)cells treated with the indicated doses of XRR.*P ＜0.05; **P ＜0.01

5 Conclusion

The present study systematically compared the effects of CIR with those of XRR on DNA damage response and found that CIR was more effective than XRR.

CIR can induce a high level of DNA DSB damage and exhibit strong and long-lasting inhibitory effect on cell survival capability, induce prolonged cell cycle arrest, and increase apoptosis in human prostate cancer cells PC-3.The underlying mechanism may be partly through the pRb/E2F1/c-Myc signaling pathway.

Collectively,the present study suggests that CIR is more efficient than XRR and may help to understand the effects of different radiation qualities on the survival potential of p53-deficient prostate cancer cells.