Screen identifies fasudil as a radioprotector on human fibroblasts

Abstract

Background

Radioprotectors safeguard biological system exposed to ionizing radiation (IR) by protecting normal cells from radiation damage during radiotherapy. Due to the toxicity and limited clinical utility of the present radioprotectors, it prompts us to identify novel radioprotectors that could alleviate IR-induced cytotoxicity of normal tissues.

Aims and Methods

To identify new radioprotectors, we screened a chemical molecular library comprising 253 compounds in normal human fibroblasts (HFs) or 16HBE cells upon IR by CCK-8 assays and clonogenic survival assays. Fasudil was identified as a potential effective radioprotector.

Results

The results indicated that Fasudil exerts radioprotective effects on HFs against IR-induced DNA double-strand breaks (DSBs) through the regulation of DSB repair. Fasudil increased homologous recombination (HR) repair by 45.24% and decreased non-homologous end-joining (NHEJ) by 63.88% compared with untreated cells, without affecting changes to cell cycle profile. We further found that fasudil significantly facilitated the expression and foci formation of HR core proteins such as Rad51 and BRCA1 upon IR, and decreased the expression of NHEJ-associated proteins such as DNA-PKcs at 24 h post-IR.

Conclusion

Our study identified fasudil as a novel radioprotector that exert radioprotective effects on normal cells through regulation of DSB repair by promoting HR repair.

Introduction

Radiotherapy is a highly effective, targeted therapy for the treatment of malignant tumors.^1–3 However, it is sometimes limited in its operation and capacity to achieve optimal tumor destruction due to the ionizing radiation (IR)-induced normal tissue toxicity.^4–6Thus, locating suitable radioprotectors that can attenuate the IR-induced cytotoxicity of normal tissue has garnered much interest.^7,8

Currently, amifostine (Ethyol from Schering-Plough) and palifermin (Kepivance from Amgen) are the two radioprotectors approved by the US Food and Drug Administration. However, adverse effects from the toxicity and high cost of these approved therapeutic agents may have limited their wider clinical usage, especially in resource-constraint countries.^4,9–14 A number of novel polymeric nanomaterials and radioprotective polymers are being explored as suitable alternatives, while some are yet to be ready for human clinical trial, and none are yet approved clinically.^10,15,16 The search continues for low-cost, efficacious radioprotectors.

Through screening a repository of chemical molecules, we identified fasudil as a potential radioprotector that effects DNA repair in normal cells exposed to IR through the regulation of homologous recombination (HR) and nonhomologous end-joining (NHEJ). Fasudil (Eril by Asahi Kasei), a Rho-kinase inhibitor (ROCK inhibitor), is approved in Japan (since 1995) and China for the improvement of cerebral vasospasm and associated cerebral ischemic symptoms and has been demonstrated to be an effective vasodilator.¹⁷ Furthermore, fasudil has been recently reported to exhibit neuroprotective effect in vivo and to stimulate neural stem cells regeneration in vitro.¹⁸ Fasudil has also been shown to increase tumor cells sensitivity to chemotherapy as an adjunct treatment.^19,20 Thus, our findings support the further exploration of fausdil as a promising pharmaceutical adjunct to cancer treatment.

Materials and methods

Cell culture and cell lines

Human fibroblasts (HFs), 16HBE normal cell lines, and human breast cancer (MCF7) cell lines were obtained from the American Type Culture Collection (Manassas, VA, United States). MCF7 cells stably expressing the pDR-GFP substrate construct were established by our group.²¹ U2OS cells expressing the EJ5-GFP reporter were obtained from Maria Jasin’s Lab (Developmental Biology Program, Memorial Sloan-Kettering Cancer Centre, New York, NY, United States).²² The cells were cultured in Dulbecco’s modified eagle medium (Gibco) culture medium containing 10% fetal bovine serum (Gibco) and 1% penicillin–streptomycin (V900929, Sigma) in a 5% CO₂ incubator at 37 °C.

Chemical library screening

The chemical library, comprised of 253 unique molecules, was purchased from Selleckchem. The experiment flowchart of chemical library screening is shown in Fig. 1A.

Figure 1.

Chemical molecule library screen identified fasudil as a radioprotective effector. A) The schematic illustrates the procedure of chemical library screening. The upper panel indicates chemical library screening using CCK-8 assay and the lower panel shows the screening using clonogenic survival assay. B) The process of chemical molecule library screening, exclusion, and inclusion for the identification of the target compounds, fasudil, is shown in the flow chart. C) A total of 253 compounds were preliminarily screened and the figure shows the cell viability of the 253 compounds-treated cells after irradiation. Molecule compounds were diluted to 10 μM. 16HBE cells were seeded in 96-well plates with 1 × 10³ cells per well. After incubation overnight, cells were pretreated with the compounds for 24 h and then irradiated with 6 Gy. After 72 or 96 h (cell density occupied 90% of each well in the control group), cell viability was determined by CCK-8 assay. D) Totally 30 (out of 51) compounds were screened out by the secondary CCK-8 assay based on the cell viability after IR (>88%). E) Among the above 30 compounds, representative images of 13 clones and quantification are shown after treatment. The cloning results derived from dexamethasone and dexamethasone acetate treatment served as positive controls. 16HBE cells were seeded on 6-well plates at densities of 1.5 × 10³ cells/well. After the treatment with the compounds individually, the cells were cultured for about 2 weeks and stained with crystal violet. F) Under the condition of compounds combined with 6-Gy IR treatment, cells were processed for crystal violet staining as described in E). Representative images of 13 clones and quantification are shown. At least three independent experiments were performed. P-values were calculated by unpaired Student’s t-test. ^*P < 0.05, ^**P < 0.01 versus control group; ^#P < 0.05, ^##P < 0.01 versus IR group.

C‌CK-8 assay

Treated cells were incubated with 110 μL of fresh cell culture medium containing 10 μL of CCK-8 reagent (B34304, Selleck) for 2 h. Absorbance was measured at 450 nm by microplate reader (Infinite M200 PRO, TECAN).

Fasudil treatment and IR

Fasudil (S1573, Selleck) was diluted with the cell culture medium to 10 μM. Cells were pretreated with cell culture medium containing 10 μM of fasudil for the indicated time period prior to IR treatment. Cells were irradiated using the X-ray Irradiator (Precision X-Ray, X-RAD225 OptiMAX, Pxi Inc, United States) following the manufacturer’s instructions. The 6 Gy irradiation was delivered once at a rate of 2.08 Gy/min.

Clonogenic survival assay

Cells were seeded onto 6-well plates/60 mm cell culture dishes, and 3 parallel samples were set in each group. The 1 × 10³–2 × 10³ cells were planted in each well and were grown overnight. After pretreatment with 10 μM fasudil and the subsequent 6Gy irradiation treatment, the cells were cultured in the cell incubator for 2 weeks and were then subjected to crystal violet staining. The number of cell colonies in each dish were quantified. Survival fractions were calculated as the plating efficiency of treated cells relative to the plating efficiency of untreated control cells.

Comet assay

DNA damage in cells was measured by neutral comet assay and alkaline comet assay. The level of intracellular DSBs was detected by neutral comet assay. The length of the comet’s tail is proportional to the extent of DNA damage. Alkaline comet assay is more sensitive than neutral comet assay in detecting DNA single-stranded and double-stranded DNA breaks.²³ The neutral and alkaline comet assays were used to detect damaged DNA in cells after X-ray irradiation following in accordance with the manufacturer instruction (Trevigen). The specific experimental protocols have been previously described.²² Olive tail moment was analyzed using Comet score software (TriTek, Sumerduck, VA).

Western blotting analysis

HFs were pretreated with fasudil and subjected to irradiation, the total protein of the cells was extracted, and cells were washed with phosphate buffer and lysed for 10 min on ice in RIPA lysate (G2002, Servicebio) containing protease inhibitor cocktail (G2006, Servicebio). The BCA protein assay kit (Thermo) was used to determine protein concentration. SDS-polyacrylamide gel was used to separate the protein samples, which were transferred to PVDF membranes (Thermo). The membranes were incubated for 1 h at room temperature with 6% skim milk dissolved in TBST (10 M Tris, 150 mM NaCl and 0.05% Tween 20 [pH 8.3]). The membranes were incubated overnight at 4 °C with primary antibody to each protein. The primary antibodies included BRCA1 (1:200, SC-6954, Santa Cruz), Rad51 (1:1000, PC130, Calbiochem), DNA-PKcs (1:5000, ab32566, Abcam), Ku70 (1:200, SC-5309, Santa Cruz), Ku80 (1:200, SC-5280, Santa Cruz), β-actin (1:2000, TA-09, ZSGB-BIO), phospho-DNA-PKcs (S2056) (ab18192, Abcam), and GAPDH (1:2000, TA-08, ZSGB-BIO). The membranes were washed for 5 min for 3 times in TBST and were incubated with secondary antibodies for 1 h at room temperature. The secondary antibodies were goat antimouse IgG (H + L) secondary antibody (1:5000, 31430, Thermo) and goat antirabbit IgG (H + L) secondary antibody (1:5000, 31460, Thermo). Chemiluminescence was monitored with a luminometer (Tanon Science & Technology, Shanghai, China).

Immunofluorescence

The protocol was described in our previous publication.²⁴ Primary antibodies included: γH2AX (1:500, 05–636, EMD Millipore), Rad51(1:300, PC130, Calbiochem), BRCA1 (1:100, sc-6954, Santa Cruz), 53BP1 (1:1,000, NB100-304, Gene Tex), and phospho-DNA-PKcs (S2056) (ab18192, Abcam). Secondary antibodies consisted of Alexa Fluor594 goat antimouse lgG(H + L) (1:300, A11032, Molecular probes) and Alexa Fluor488 chicken antirabbit lgG(H + L) (1:300, A21441, Molecular probes). Nuclei were counterstained with DAPI. Images were acquired with a fluorescence inverted microscope (Axio observer Z1, ZEISS) and were processed using Adobe PhotoShop software.

HR assay and NHEJ assay

MCF7 cells stably expressing the pDR-GFP construct were transfected with the I-SceI plasmid. The introduction of I-SceI nuclease created a double-strand DNA break that was repaired by HR.^21,25 Cells were subsequently treated with 10 μM of fasudil and were dissociated. GFP signal was detected and quantified for HR repair efficiency by flow cytometry at 24 h after transfection.

U2OS cells stably expressing the EJ5-GFP constructs were transfected with the I-SceI plamid.²¹ I-SceI nuclease was able to create a double-strand DNA break, which was repaired by non-HR. Cells were dissociated and subjected to flow cytometry for the measurement of NHEJ repair efficiency at 24 h after transfection.

Cell cycle analysis

Specific methods are described in detail elsewhere.^26,27 Briefly, the cells were planted onto 60-mm cell culture dishes and were cultured overnight, which were then treated with a fresh cell culture medium containing 10 μM of Fasudil for 24 h. The cells were then treated with BrdU (Sigma) and were cultured for 30 min before fixation using 70% cold ethanol overnight. Subsequently, the cells were digested with trypsin, incubated with BrdU primary antibody (1:20, 347580, BD), and then secondary antibody: Alexa Fluor594 goat antimouse lgG(H + L) (1:200, A11032, Molecular probes). Finally, the cells were stained with DAPI and were subjected to flow cytometry analysis (FACS Aria III, BD).

Statistical analysis

Data are presented as the means ± standard deviations. All statistical analyses were performed using Student’s t-test on SPSS Statistics for Windows, version 23.0 (Armonk, NY: IBM Corp; licensed to Shandong University). P < 0.05 was considered as the statistically significant difference.

Results

Screen of chemical molecule library identifies fasudil to have radioprotective effect

A customized chemical molecule library consisting of 253 compounds were preliminarily screened for potential radioprotectors in 16HBE cells using the CCK-8 assay (Fig. 1A, upper panel, and B). The cell viability of cells pretreated with the respective 253 compounds after IR treatment are shown in Fig. 1C. We then performed a secondary screen using 51 compounds with cell viability >80% after IR treatment by CCK-8 assay (the control group was considered as 100%, and the IR alone group was 51.87%; Fig. 1A, upper panel). We next applied a more stringent cutoff to exclude compounds with cell viability <88% after IR treatment, and 30 compounds were selected (Fig. 1B and D).

Since CCK-8 assay is a short-term acute experiment, and in order to examine longer term cell survivability after treatment with the remaining 30 potential compounds, we next employed the clonogenic survival assay and validated the effects of these 30 compounds on 16HBE cells (Fig. 1A, lower panel). We observed 13 compounds to have cell proliferation significantly greater than that of the IR-only group (Fig. 1B). Of these 13 compounds, dexamethasone and dexamethasone acetate were included; these compounds were previously reported to exert potential radioprotective effects on Chinese hamster V-79 cells and normal bone marrow cells,^28,29 thus affirming the validity and reliability of our screening study (Fig. 1E, right, and F, right). We further noticed that the cloning efficiency of the fasudil-treated group was higher than that of other experimental compounds (P < 0.01, Fig. 1E, left). In response to IR treatment, the cloning efficiency of the fasudil-treated was significantly higher than that of other experimental compounds (P < 0.01, Fig. 1F, left) and almost comparable to the dexamethasone. The above findings suggest that fasudil might be a potential radioprotector of normal cells.

Fasudil exerts radioprotective effects on HFs

Since HFs is considered to be hypersensitive to IR,³⁰ we next detected whether fasudil may radioprotect HFs against IR. We first examined the toxicity of fasudil in HFs using CCK-8 assay in vitro. The results indicated that fasudil exhibited cytotoxic effect at 120 μM (P < 0.01) but not at lower concentrations of 5–60 μΜ (Fig. 2A). For the next series of experiments, we opted for 10 μM fasudil which is a lower and safe dose with no noticeable cytotoxicity to the cells and is also recommended by the supplier of the chemical library.

Figure 2.

Evaluation of cytotoxicities and radioprotective effects of fasudil on HFs and MCF7 cells. A) In vitro cytotoxicities of fasudil on HFs were measured by CCK-8 assay. Cells (1× 10³ cells per well) were seeded in 96-well plates, then incubated with fasudil at the concentrations of 0, 5, 10, 30, 60, and 120 μM. Cell viability was determined at 48 h after fasudil treatment. B) Effect of 10 μM fasudil on cell viability after irradiation. HFs were pretreated with fasudil for 24 h and then subjected to 6 Gy. Cell viability was determined at 48 h post-IR. C) Effect of fasudil on cell viability after irradiation with different dosages. HFs were pretreated with fasudil for 24 h, then irradiated with different dosages (0, 2, 4, 6, 8, and 10 Gy) and incubated for 72 h before subjecting to CCK-8 assay. D and E) The effect of fasudil on survivability of irradiated cells. D) HFs or E) 16HBE cells were pretreated with fasudil for 0, 6, and 24 h before subjecting to 2, 4, and 6 Gy. Cells were cultured for about 2 weeks and stained with crystal violet. F) The cell viabilities of fasudil-treated cells with different incubation time were measured. MCF7 cells were seeded in 96-well plates and incubated with fasudil for 0, 1, 2, 3, 4, and 5 days before subjecting to CCK-8 assays. G) Effect of fasudil on survivability of MCF7 cells irradiated with different doses. After pretreatment with fasudil for 24 h and the subsequent irradiation, MCF7 cells were cultured for about 2 weeks before subjecting to crystal violet staining. A minimum of three independent experiments were performed. The percentage of survival fractions was quantified and shown. P-values were calculated by unpaired Student’s t-test. ^**P < 0.01 versus control group; ^#P < 0.05, ^##P < 0.01 versus IR group; ns, not significant.

We found the cell viabilities of HFs in IR-treated alone group significantly decreased by 42.53% compared with the control group (P < 0.01, Fig. 2B). When the HFs were pretreated with fasudil for 24 h before 6 Gy treatment, the cell viability of HFs in the combined group significantly increased compared with the IR-alone group at 48 h post-IR (P < 0.01, Fig. 2B), suggesting that fasudil exerts radioprotective effects on HFs.

To further verify the radioprotective effect of fasudil on HFs, we measured cell viability at 72 h after IR treatment (2, 4, 6, 8, and 10 Gy) using the CCK-8 assay. As shown in Fig. 2C, the cell proliferation rates of the fasudil-treated groups were significantly higher than that of the groups treated with IR-alone (P < 0.01), indicating that fasudil exerts strong radioprotective effects on HFs. Furthermore, the pretreatment of fasudil promotes the survival of HFs (Fig. 2D) and 16HBE (Fig. 2E) in response to IR by clonogenic survival assay, supporting the radioprotective effects of fasudil on HFs.

We observed that fasudil alone did not noticeably affect the cell viability by CCK-8 assay (Fig. 2F) and colony formation of cancer cell line MCF7 cells after IR (Fig. 2G), thus suggesting that fasudil may exert radioprotective effects preferentially only in normal cells.

Fasudil promotes IR-induced DSB repair

DSBs are the major form of lethal DNA damage. To investigate whether fasudil may exert the observed radioprotective effects by regulating IR-induced DSB repair in the HFs, we next examined the accumulation of IR-induced DSBs in the cells pretreated with fasudil by comet assay. The data derived from neutral comet assay showed that fasudil did not significantly reduce the olive tail moment at 0 and 30 min after IR in HFs (Fig. 3A). Notably, we found that the olive tails in the combined IR and fasudil treatment groups were shorter compared with the IR-only group at 2, 6, and 24 h post-IR (P < 0.01, Fig. 3A). The dynamic change of DNA damage in the cells treated with the combination of fasudil and IR were also validated by the alkaline comet assay (Fig. 3B).

Figure 3.

Fasudil promotes IR-induced DSB repair. A) HFs were pretreated with 10 μM of fasudil for 24 h and then subjected to 6 Gy IR. The DSBs levels were detected by neutral comet assay at 0, .5, 2, 6, and 24 h post-IR. The olive tail moments were quantified using comet score software. B) HFs were treated with fasudil as described in A) and subjected to alkaline comet assay. C) 53BP1 and γH2AX foci were measured in fasudil-pretreated HFs after IR treatment. Briefly, HFs were pretreated with 10 μM fasudil for 24 h and were subjected to immunofluorescence at 6 and 24 h post-6 Gy irradiation. The representative images of 53BP1 and γH2AX foci in the HFs were shown and the percentage of cells with 53BP1 and γH2AX foci were quantified (the cell with ≥10 foci was considered positive and counted). DAPI was used for nuclear staining. D) The protein levels of 53BP1 and γH2AX in C) were examined by western blotting. At least three independent experiments were performed. P-values were calculated by unpaired Student’s t-test. ^**P < 0.01 versus control group; ^##P < 0.01 versus IR group.

We next tested whether fasudil affects the level of DSBs by measuring DSB markers γH2AX and 53BP1 using immunofluorescence staining. The number of cells with ≥10 γH2AX foci formation was counted. The percentage of cells with γH2AX and 53BP1 foci in the IR-only group were significantly increased than those in the control group at both 6 and 24 h post-IR. Interestingly, the combination of IR and fasudil resulted in decreased percentage of cells with γH2AX and 53BP1 foci at both 6 and 24 h post-IR (P < 0.01, Fig. 3C). In line with the changes of γH2AX and 53BP1 foci with fasudil pretreatment, similar decreased protein levels of γH2AX and 53BP1 in the combined treatment group were also observed in a time-dependent manner (Fig. 3D).

Together, through the observation of the dynamic changes of key DSB markers, the findings suggest that fasudil could facilitate IR-induced DSB repair.

Fasudil regulates IR-induced DSB repair by preferentially promoting HR repair

To examine the repair pathways involved in fasudil-mediated reduction of IR-induced DSBs, we measured the protein levels and foci formation of core DSB repair-associated proteins in HFs. We first detected the foci formation and protein levels of Rad51 and BRCA1, which are core proteins in HR repair pathway. Compared with the control, the percentage of cells with Rad51 foci and BRCA1 foci in the IR-only group increased compared with control without IR treatment (P < 0.01, ≥10 foci/cell counted as positive cell), and fasudil treatment further increased the percentage of cells with Rad51 and BRCA1 foci following the combination treatment compared with the IR-only group at both 6 and 24 h post-IR (Fig. 4A), suggesting the increased accumulation of key HR factors at DSBs. Consistently, the protein levels of Rad51 and BRCA1 in the combined group were further increased compared with the IR-only group at both 6 and 24 h post-IR (Fig. 4B), indicating the robust promotion of HR repair in the presence of fasudil in response to IR.

Figure 4.

The effect of fasudil on HR and NHEJ core proteins. A) Rad51 and BRCA1 foci were measured in fasudil-pretreated HFs after IR treatment. HFs were pretreated with fasudil for 24 h and subjected to immunofluorescence at 6 and 24 h post-6 Gy. The percentage of cells with Rad51 and BRCA1 foci were quantified (the cell with ≥10 foci was considered positive and counted). Nuclei were counterstained with DAPI. B) The protein levels of Rad51 and BRCA1 were examined by western blotting in the cells mentioned in A). C) The protein levels of NHEJ core proteins, including Ku70, Ku80, DNA-PKcs, and phospho-DNA-PKcs, at S2056 (pDNA-PKcs) were examined by western blotting in the cells described in A). GAPDH was used as loading control. D) pDNA-PKcs foci were measured in the cells described in A). Nuclei were counterstained with DAPI. The percentage of cells with pDNA-PKcs foci was quantified. Zoomed images show the details of the colocalization of pDNA-PKcs foci and γH2AX foci denoted with white square. E) The schematic depicts HR reporter gene pDR-GFP. MCF7 cells stably expressing pDR-GFP plasmid were transfected with I-SceI or GFP plasmids and then treated with fasudil for 24 h before subjecting to flow cytometry analysis. GFP was used as a positive control for indicating transfection efficiency. The images and the quantification indicate the relative HR frequency in fasudil-treated cells. F) The schematic depicts NHEJ reporter gene EJ5-GFP. U2OS cells stably expressing EJ5-GFP were transfected with the plasmid of I-SceI or GFP and treated with fasudil for 24 h before flow cytometry analysis. G–I) Cell cycle analysis of G) fasudil-treated HFs, H) MCF7, and I) 16HBE cells by flow cytometry. At least three independent experiments were performed. P-values were calculated by unpaired Student’s t-test. ^*P < 0.05, ^**P < 0.01 versus control group; ^##P < 0.01 versus IR group; ns, not significant.

Next, we measured whether fasudil can affect NHEJ repair pathway through detecting the major NHEJ-associated proteins such as Ku80, Ku70, and DNA-PKcs. Although we observed dissimilar change of the protein levels of Ku70, Ku80, DNA-PKcs, and autophosphorylation of DNA-PKcs S2056 residue (pDNA-PKcs) in fasudil-pretreated cells without IR or at 6 h post-IR, we did find the protein expression of the above NHEJ factors in fasudil-pretreated cells were significantly decreased compared with IR-only groups at 24 h post-IR by western blotting analysis (Fig. 4C), suggesting that fasudil may suppress NHEJ repair by inhibiting the protein expression of the above key NHEJ regulators. Additionally, the foci formation of Ku70 and Ku80 was not observed, but we found significantly decreased percentage of cells with pDNA-PKcs foci formation at both 2 and 6 h in the combination group treated with fasudil and IR (P < 0.01, Fig. 4D). DNA-PKcs S2056 phosphorylation is required for NHEJ repair, which promotes the DNA-PK disassembly from DSBs, allows DNA end ligation, and inhibits HR repair.^31–33 The depressed percentage of cells with pDNA-PKcs foci formation, as shown in Fig. 4D, indicates the decreased NHEJ repair in fasudil-treated cells upon IR treatment. Consistent with the data from Fig. 3C, the cells with γH2AX foci formation at 6 h post-IR were noticeably decreased (Fig. 4D). Most of pDNA-PKcs foci colocalized with γH2AX foci in the cells treated with IR alone; for the combination group, despite the percentage of cell with pDNA-PKcs and γH2AX foci formation decreased as compared with IR alone (Figs. 3C and 4D, left), the cells with the protein foci formation were still presented, and only the foci colocalization of both proteins was hardly observed in the cells (Fig. 4D, left zoom), indicating that fasudil can inhibit the autophosphorylation of DNA-PKcs S2056 to participate in DNA repair.

To confirm the effect of fasudil on DSBs repair pathways, we next measured the efficiency of HR and NHEJ in the presence of fasudil. We used MCF7 cells stably expressing the pDR-GFP substrate constructs and U2OS cells expressing the EJ5-GFP reporter to measure the repair efficiency of HR and NHEJ, respectively (Fig. 4E and F). It was found that cells pretreated with fasudil for 24 h dramatically increased the HR repair efficiency by 45.24% (P < 0.05, Fig. 4E) and decreased the NHEJ repair efficiency by 63.88% compared with the cells without fasudil treatment (P < 0.01, Fig. 4F). Additionally, fasudil did not noticeably affect cell cycle profiling in HFs (Fig. 4G), MCF7 cells (Fig. 4H), and 16HBE (Fig. 4I). The observations suggest that fasudil-promoted HR efficiency is independent of the cell cycle.

Together, that data suggest that fasudil induces robust high expression of HR factors to facilitate HR repair and may promote the switch from NHEJ to HR for its radioprotective effects.

Discussion

Fasudil, a novel radioprotector of normal cells

In this study, we screened a customized chemical molecular library designed to isolate new compounds that have radiation protection by measuring the cell survival and identified fasudil as a new potential radioprotector. As a ROCK inhibitor, fasudil has been reported to demonstrate neuroprotective effect in vivo and to stimulate neural stem cells regeneration in vitro.¹⁸ Furthermore, fasudil was demonstrated to be protective of PC12 cells from oxidative stress injury.³⁴ Our study extends the current knowledge of fasudil by discovering that 10 μM of fasudil not only increases the survival rate of HFs after IR treatment but also has no cytotoxicity at normal therapeutic dose, uncovering that fasudil might be a new and safe radioprotector of normal cells.

Furthermore, it has been reported that fasudil exerted antitumor effects on several types of cancer.²⁰ Treatment of MDA-MB 231 cells with a lower dose (10 μM) of fasudil for 24 h has been shown to reduce cell migration without affecting cell viability.³⁵ In this study, the fasudil treatment did not noticeably affect the cell viability and has no radiosensitization effect on MCF7 cells (Fig. 2F and G), suggesting that fasudil has no effect on the tumor cells in our working system.

The mechanism of fasudil-induced radioprotective effects is correlated with elevating HR function

To explore the mechanism of fasudil-induced radioprotective effects, we examined the effects of fasudil on DNA repair. The initial events of DNA damage response are ATM-mediated H2AX phosphorylation at Ser139 (γH2AX), which is a marker of DNA DSBs, and γH2AX then coordinates DNA repair by recruiting DNA repair factors, including Rad51, BRCA1, and Ku complex, to the surrounding of DSB sites.^36–39 We found that the protein expression and foci formation of HR core proteins Rad51 and BRCA1 were elevated, while the core NHEJ repair proteins (Ku80, Ku70, DNA-PKcs, and pDNA-PKcs) were downregulated at 24 h in fasudil-treated cells upon IR (Fig. 4A–D). Fasudil treatment results in the increased frequency of HR repair and decreased the frequency of NHEJ repair, respectively (Fig. 4E and F). These data suggest that fasudil may regulate DNA repair by preferentially promoting HR repair and suppressing NHEJ repair pathway. In addition to the measurement of γH2AX and 53BP1 foci formation, we also performed comet assay, another well-validated biomonitoring tool for DNA damage.⁴⁰ We found that fasudil markedly decreased IR-induced DSBs in HFs (Fig. 3A and B), indicating that DNA repair activity was increased in fasudil-treated cells.

Since fasudil is a ROCK inhibitor, it would rise a question whether Rho-kinase family involves in the regulation of DNA repair pathway. Currently, it was suggested that Rho-kinase is activated in response to DNA damage and functions as a transducer of DNA damage signaling.⁴¹ The inhibition of Rho-kinase promotes DNA repair in neuroblastoma cells following cisplatin treatment.⁴² Additionally, pharmacological inhibition of Rho kinase promotes the repair of IR-induced DSBs in both primary human lung cells and lung tissue,⁴³ potentially suggesting the involvement of Rho-kinase in DNA damage repair. Thus, it would be necessary to further explore whether Rho-kinase family works through regulating the HR and NHEJ activity in fasudil-induced radioprotection for normal cells.

The specific regulation of fasudil to HR may depend on 53BP1

Most DNA DSBs are repaired by either NHEJ or HR depending in part on the precise chemistry of the DNA ends and the stage of the cell cycle.^44,45 HR maintains the integrity of genome, while the DSBs repaired by NHEJ is prone to error, resulting in chromosome instability.⁴⁶ The efficiency of HR is competed by other DSB repair pathways, including NHEJ.^47,48 It has been reported that 53BP1 is also a key regulator of the choice between NHEJ and HR, which promotes NHEJ by suppressing DNA end resection and limits HR in part by blocking DNA end resection as well as inhibiting BRCA1 recruitment to DSB sites.^49–51 Thus, it would be important to uncover a compound that would favor HR repair and limit NHEJ through the inhibition of 53BP1.⁵² Interestingly, we found reduced 53BP1 foci and protein level after irradiation following fasudil treatment in a time-dependent manner (Fig. 3C and D). But, the recruitment of HR-associated Rad51 and BRCA1 were significantly promoted at IR-induced DSBs after fasudil treatment, and the protein levels of Rad51 and BRCA1 were also increased under the same conditions in a time-dependent manner (Fig. 4A and B), indicating HR repair pathway was significantly promoted in fasudil-induced radioprotection for normal cells. Instead, the protein levels of NHEJ core proteins (Ku70, Ku80, DNA-PKcs, and pDNA-PKcs) were significantly decreased at 24 h post-IR (Fig. 4C) and the foci formation of autophosphorylation of DNA-PKcs S2056 was dramatically decreased at DSBs at early time points (2 and 6 h) post-IR in fasudil-treated cells (Fig. 4D), indicating NHEJ repair pathway was suppressed in fasudil-induced radioprotection for normal cells by inhibiting either the expression or phosphorylation of the indicated NHEJ regulators upon IR. These data suggest fasudil-induced bidirectional effect of DNA repair pathways may inhibit the activity of 53BP1 and coordinate productive repair of IR-induced DSBs by modulating the interplay of BRCA1-53BP1 to suppress the NHEJ repair pathway and to enhance HR repair. However, further research is also needed to elucidate the exact mechanism that fasudil drives the switch from NHEJ to HR repair.

In this study, we uncovered that the activity of the DNA repair pathway mediated by fasudil is the key mechanism for its radioprotective effects on HFs independent of cell cycle. Fasudil, at the concentration of 10 μM, significantly reduced IR-induced DSBs by modulating the activities of core proteins involved in either the HR or NHEJ repair pathway, thus promoting HR repair and suppressing the NHEJ pathway. Ideally, the assays on the repair efficiencies of HR and NHEJ should be measured by normal cell lines stably expressing the pDR-GFP and EJ5-GFP constructs. Since we found that 10 μM of fasudil has no effect on the survivability of MCF7 cells, we sought to measure the repair efficiency of HR and NHEJ in cancer cell lines stably expressing the pDR-GFP and EJ5-GFP constructs, which, at least in part, reflects the effects of fasudil on the DNA repair of HFs after irradiation. However, the detailed mechanism of Fasudil coordinating the choice of HR and NHEJ repair pathways remains elusive. The potential use of Fasudil as a radioprotector still need further research.

Conclusion

Radioprotective effects of fasudil is related to the enhanced expression of HR-associated regulators (BRCA1 and Rad51) and higher level of HR repair. Moreover, the higher expression of HR-associated regulators may prefer HR repair by enabling switching from NHEJ to HR.

Authors’ contributions

Y.Y., Z.F., and Z.C. designed the research; Y.Y. analyzed the data; Y.Y., C.C., and C.D. performed the research; Y.Y., C.D., D.L., and Z.F. wrote the paper.

Conflict of interest statement

None declared.