Isoorientin triggers apoptosis of hepatoblastoma by inducing DNA double- strand breaks and suppressing homologous recombination repair

Dehong Huanga,1, Lei jinb,1, Zhengkang Li c,1, Ji Wuc, Ni Zhangc, Dianrong Zhouc, Xiaorong Nid,⁎⁎, Tieying Houc,⁎

Keywords:Isoorientin,ATM,MRN complex,DSBs,HR

ABSTRACT
Hepatoblastoma (HB) is the most common malignant liver tumor in children. DNA and DNA-associated processes are one of the most important targets of chemotherapeutic agents. Isoorientin (Iso), a natural ﬂavonoid com- pound, can be extracted from several plant species. The eﬀects of Iso and its molecular mechanisms on hepatic malignancies remain unclear. Herein, the anti-tumor eﬀects of Iso in HB and its underlying mechanisms were explored. We found that Iso signiﬁcantly inhibited the proliferation of HB cells both in vitro and in vivo. Mechanistic studies showed that Iso triggered cell apoptosis by inducing DNA double-stranded breaks and blocking the initiation process of homologous recombination repair, which was related to the attenuation of ataxia telangiectasia mutated (ATM) activation and inhibiting the binding of phosphorylated ataxia tel- angiectasia mutated (pATM) and the MRE11-RAD50-NBS1 (MRN) complex. Furthermore, Iso markedly sensi- tized HB cells to the anti-proliferative eﬀects of the poly ADP-ribose polymerase (PARP) inhibitor olaparib both in vivo and in vitro. Taken together, our study ﬁrst showed that Iso was a DNA-damage agent, and the combi- nation of Iso with a PARP inhibitor might be a promising strategy for treating HB patients.

1.Introduction
Hepatoblastoma (HB) is the most common type of malignant liver tumor in children. It is usually diagnosed during the ﬁrst 5 years of life [1]. The average risk of a child developing HB is about 1 in 1,000,000 in the US [2]. Approximately 70% of children with this disease are treated successfully with surgery and chemotherapy. The survival rate is > 90% for early-stage HB [3, 4]. However, the prognosis in advanced stages remains poor, and the typical chemotherapeutic agents used for treatment are still limited by signiﬁcant toxicity [5]. Consequently, the current therapeutic strategy for HB is mainly focused on reducing chemotherapy-related toxicity by decreasing dose intensity and im- proving the clinical outcome of patients with metastatic disease by intensifying chemotherapy in combination with new drugs. To further improve the outcomes of high-risk children, in particular those with metastatic HB, new therapeutic approaches to increasing the eﬃciency of chemotherapy and avoiding excessive toxicity are urgently needed.Natural compounds are regarded as a fertile source of potential cancer chemotherapeutic and chemopreventive agents that have re- ceived increasing attention in recent years because of their safety in chemotherapy and advantages in reducing the risk of mutagenicity in normal cells [6,7]. Isoorientin (3′,4′,5,7-tetrahydroxy-6-C-glucopyr- anosyl ﬂavone, Iso), a chemical ﬂavonoid-like compound, can be ex- tracted from many plant species, such as ﬂax straw [8], aqueous leaf [9], Gypsophila elegans [10], Phyllostachys pubescens [11], Patrinia [12], and Drosophyllum lusitanicum [13]. It has been reported that Iso has a variety of pharmacological properties, including anti-oxidant, anti-in- ﬂammatory and anti-nociceptive activities [14,15]. Recent investiga- tions have found that Iso also induces apoptosis through an increase in the generation of reactive oxidation state (ROS) in human hepatocel- lular carcinoma [16,17]. However, the eﬀects of Iso on hepatic malig- nancies, especially on HB, and the molecular mechanisms underlying apoptosis induced by Iso in HB cells remain unclear. DNA double-stranded breaks (DSBs), one of the most deleterious

Fig. 1. Iso inhibited proliferation and induced apoptosis of HB cells in vitro.
A. HB cells (HepG2 and HuH6) and normal hepatocyte (L-02) were treated with increasing concentrations of Iso and the cell viability was assessed via CCK8 assay. B–C. Colony formation assays of inhibition roles of Iso in HepG2 (A) and HuH6 (B) cells. Columns (right panel) represent the colony numbers of HepG2 cells (B) and HuH6 cells (C) from three independent experiments, which are shown as the mean ± SD. D–E. Flow cytometry assays of cell apoptosis of HepG2 (D) and HuH6 (E) cells treated with indicated concentrations of Iso. Columns (right panel) represent the average percent of apoptosis of HepG2 cells (B) and HuH6 cells (C) from three independent experiments, which are shown as the mean ± SD. **p < 0.01, compared to the control group.forms of DNA damage, may cause genomic instability, apoptosis and cancer if not properly repaired [18]. The maintenance of genomic in- tegrity needs a conserved and intricate signaling pathway in response to DSB damage [19]. DSBs are mainly repaired by homologous re- combination (HR) or non-homologous end joining (NHEJ) [20]. HR is an error-free repair pathway that uses the sister chromatid as a template for the correct replacement of the DNA sequence [21]. Factors, such as the MRN complex, ATM, 53BP1, BRCA1, BRCA2, and Rad51, are in- volved in the HR pathway [22]. The MRN complex has been reported to be essential for the regulation of the cellular response to DSB, which is involved in multiple processes of the DNA damage response (DDR), such as initial DSB detection, signal transduction, and the promotion of DSB repair [23]. The ATM kinase activates a network of checkpoint and DNA repair proteins in response to DNA damage, thus being another keystone in maintaining genomic stability [24].In the present study, we investigated the antiproliferative and proapoptotic eﬀects of Iso on HB and explored its mechanism of action. Furthermore, the putative potentiation of the anti-tumor eﬀect on HB as a result of combining Iso with olaparib, the ﬁrst-in-class PARP inhibitor approved by the Food and Drug Administration (FDA) for patients with BRCA-mutated ovarian cancer, was also assessed here. 2.Materials and methods
2.1.Cell lines and cell culture
The HB cell line HepG2 and the normal liver cell line L-02 were purchased from the American Type Culture Collection (Manassas, VA, USA). The HB cell line HuH6 was purchased from the Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). The HB cell lines (HepG2 and HuH6) were maintained in logarithmic growth in Dulbecco’s Modiﬁed Eagle Medium (DMEM) (Gibco, Carlsbad, CA, USA) with addition of 1% MEM Non-Essential Amino Acids Solution (Gibco, USA), 10% fetal bovine serum (FBS, Gemini, USA), 100 U/ml penicillin, and 100 U/ml streptomycin. The normal liver cell line L-02 was cultured in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) with 10% FBS, 1% penicillin, and 1% streptomycin. All cell lines were incubated at 37。C in an atmosphere of 5% CO2.

2.2.Reagents and antibodies
Iso (purity = 98%) was purchased from Sigma Chemical (St Louis, MO, USA). olaparib (AZD2281) was purchased from Selleck Chemical (Houston, TX, USA). Both drugs were dissolved in dimethyl sulfoxide (DMSO) and diluted to diﬀerent concentrations. When the cells reached 70%–90% conﬂuence, they were then treated with Iso or olaparib and then prepared for subsequent experiments. The amount of DMSO added to the cell culture was less than 0.1% in all cases.

Fig. 2. Iso inhibited the growth of implanted HuH6 cells in a xenograft mouse model.
HuH6-bearing nude mice were treated with indicated concentrations of Iso or DMSO (control). The tumors were then collected after the mice were sacriﬁced. A. Statistical analysis of tumor volumes (mm3) during the chemotherapeutic experiment. B. Statistical analysis of tumor weights (g) at the terminal time point. C. Statistical analysis of body weights (g) of nude mice during the chemotherapeutic experiment. Data were represented as means ± SD. Error bars, SD. *p < 0.05 and **p < 0.01, compared to the control group antibodies used in this study were as follows: γH2AX (Millipore, Watford, UK, #05-636-I), γH2AX (Abcam, Cambridge, MA, USA, ab2893), RNF8 (Abcam, ab105362), MDC1 (Abcam, ab11169), RNF168 (Santa Cruz, sc-101125), BRCA1 (Santa Cruz, sc-642), NBS1 (Abcam, ab32074), ATM (Abcam, ab78), phospho-ATM (pATM) (pS1681) (Abcam, ab36810), chk2 (Abcam, ab47433), pchk2 (pT68) (Abcam, ab85743), Tubulin (Abcam, ab7291), 53BP1 (Abcam, ab175933), Rad51(Abcam, ab88572), MRE11 (Abcam, ab214), and Rad50 (Abcam, ab87918). 2.3. Anti-tumor activity analysis in vitro
To explore the potential roles of Iso in tumor inhibition, cell counting kit-8 (CCK8) assays and colony formation assays were per- formed according to our previous report [25,26]. The IC50 values of Iso in HB cells and normal liver cells were evaluated using SPSS v22.0 (IBM, Armonk, NY, USA). The proapoptotic eﬀect of Iso on hepatoma cells was explored using the double-staining method of the Annexin V/ propidium iodide (PI) apoptosis detection kit (BDPharmingen, Franklin Lakes, NJ, USA). HepG2 or HuH6 cells were treated with diﬀerent doses of Iso, and then the cells were stained with Annexin V-FITC/ PI and analyzed by ﬂow cytometry following manufacturer’s instructions. Both the early apoptosis (Annexin V+/PI-) and late apoptosis (Annexin V +/PI+) were counted.

2.4. Antitumor activity analysis in vivo
To evaluate the potential eﬀects of Iso on tumor AT7519 inhibition, in vivo chemotherapeutic tumor-bearing nude mice assays were performed.Male mice aged 10 weeks were purchased from Shanghai Super-B&K Laboratory Animal Corp. (Shanghai, China). The HuH6 cells (6 × 106) were injected subcutaneously into the mice. After tumors were mea- surable, the mice were randomly divided into four groups, each with ﬁve mice. Then, the mice were intraperitoneally injected with 0.1% DMSO or 10 mg/kg Iso, 20 mg/kg Iso, or 30 mg/kg Iso three times per week. The tumors volumes and mice weights were measured once a week. The tumors volumes was calculated using the formula TV = (L × W2)/2. After eight weeks, the mice were humanely killed by cervical dislocation after injection with 1% pentobarbital sodium and the tumors were dissected, imaged and weighed. For the in vivo che- mosensitivity analysis, HuH6 cells (9 × 106) were injected sub- cutaneously into the mice. Then, the HuH6-bearing nude mice were randomly assigned to four groups, and each group was received an intraperitoneal injection of either DMSO (as a control) or Iso (20 mg/ kg) with or without olaparib (15 mg/kg) for three times per week. 18 day later, the mice were sacriﬁced. All animal-related procedures were approved by the Institutional Review Board of Guangdong General Hospital, Guangdong Academy of Medical Sciences.

2.5. Neutral comet assay
The neutral comet assay was performed to detect DNA damage in response to Iso treatment. HepG2 and HuH6 cells were treated with diﬀerent doses of Iso for 24 h. Then, the cells were trypsinized, washed, and suspended in 0.5% low-melting point agarose and then the sus- pension (150 μl) was added into a precoated glass slide (1% agarose). Subsequently, the slides were solidiﬁed at 4 °C for 20 min in the dark and then immersed in iced neutral lysis buﬀer (2.5 M NaCl, 100 mM

Fig. 3. Iso induced DNA double-strand breaks and reduced homologous recombination repair in HB cells.
A-B. HB cells (HepG2 and HuH6) were treated with increasing concentrations of Iso and levels of DNA damage, especially DSBs induced by Iso were examined via neutral comet assays. The size of the comet tail corresponded to the severity of DNA damage. Columns (right panel) represent the olive tail moment of HepG2 cells (A) and HuH6 cells (B) from three independent experiments, which are shown as the mean ± SD. C. Analysis of HR or NHEJ eﬃciency after HepG2 and HuH6 cells treated with diﬀerent concentrations of Iso. D. HB cells (HepG2 and HuH6) exposure to agents that could induceDSBs, including MMC 0.1 mM, HU 5 mM, CPT 3 mM, UV 10 J/m2, IR 10 Gy, were examined after 24 h growth in the presence of increasing doses of Iso, and then the cell viability was evaluated using CCK8 assays. Data from three independent experiments were represented as means ± SD. Error bars, SD. *p < 0.05 and **p < 0.01, compared to the control group.EDTA, 10 mM TRIS, 10% DMSO, and 1% Triton X-100; pH 10) over- night. The next day, the slides were immersed in ice-cold electrophor- esis buﬀer (300 mM NaOH and 1 mM EDTA) for 30 min and electro- phoresed at 20 V for 30 min. Finally, the slides were washed using neutralization buﬀer (0.4M TRIS, pH 7.5), stained with SYBR Green I, and imaged with an inverted ﬂuorescence microscope (Leica). 2.6. HR and NHEJ reporter assays
Using a GFP-based HR and NHEJ reporter system [27,28], we transfected HepG2 and HuH6 cells with DR-GFP (negative control), GFP (positive control), DR-GFP + Isce-I (HR) or DR-GFP + Isce-I(NHEJ). Transfection was performed after 24 h,and diﬀerent concentrations of Iso were co-cultured with HB cells for 24 h. Then, the cells were har- vested to detect the GFP percentage (GFP%). Finally, the repair eﬃ- ciency of HR or NHEJ in each sample was calculated using the formula: (GFP% of HR/NHEJ-GFP% of negative control)/GFP% of positive control.

2.7. Confocal immunoﬂuorescence
HepG2 and HuH6 cells were seeded into 24-well plates with cov- erslips and treated with Iso in combination with or without olaparib the next day. Afterwards, the slides were washed with ice-cold phosphate buﬀered saline (PBS), ﬁxed with 4% paraformaldehyde, permeabilized with 0.5% TritonX-100, and blocked with goat serum. The cells were then incubated with primary antibodies overnight at 4 °C. Afterwards, the slides were washed twice with PBS and stained with Dylight 488- conjugated Aﬃnipure Donkey Anti-Rabbit IgG (H + L) (Jackson, 1:2000) for 1 h at room temperature in the dark. Finally, the slides were sealed with ProLong Gold Antifade Reagent and 4, 6-diamidino-2- phenylindole (DAPI). The ﬂuorescence was detected using confocal ﬂuorescence microscopy (Carl Zeiss).

2.8. Immunoprecipitation
HepG2 and HuH6 cells were ﬁrst lysed in RIPA buﬀer (50 mM Tris-HCL, pH 7.5, 150 mM NaCL2, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate and a protease inhibitor cocktail). 500 μg-1 mg proteins were then incubated with primary an- tibodies and dynabeads Protein G (Invitrogen, Carlsbad, CA, USA) overnight rocking at 4 °C. The next day, the mixture was centrifuged and eluted in lysis buﬀer for ﬁve times. Then, the bead-bound proteins were separated in an SDS-PAGE gel, electro-transferred to poly- vinylidene ﬂuoride membranes (Millipore, Billerica, MA, USA), and incubated with primary antibodies overnight at 4 °C. Finally, the membranes were incubated with ﬂuorescence-conjugated secondary antibodies and the protein signals were detected by the Odyssey two-

Fig. 4. Iso enhanced the antitumor eﬀects of olaparib in vitro.
A-B. Eﬀects of Iso in combination with olaparib on the proliferation of HepG2 (A) and HuH6 (B) cells were assessed via the clone formation assay. C-D. Immunoﬂuorescence analysis of foci formation of γH2AX in HepG2 (C) and HuH6(D) cells treated with diﬀerent concentrations of Iso in the presence or absence of olaparib.

2.9. Subcellular fractionation
For isolation of the nuclear soluble and chromatin fractions, a total of 3 × 106 cells were ﬁrst washed with ice-cold PBS andresuspended in 200 μL of solution A (10 mM HEPES at pH7.9, 10 mM KCL,1.5 mM NaCL2, 0.34 mM sucrose, 10% glycerol, 1 mM DTT, 10 mM NaF, 1 mM Na2VO3, and 1 × protease inhibitor cocktail). Triton X-100 (0.1%) was then added into the mixture. After left on ice for 5 min, the cells were subjected to a low speed centrifugation (1300 × g for 4 min) to remove the nuclei from cytoplasmic proteins. Isolated nuclei were then washed once with solution A and lysed in solution B (3 mM EDTA, 0.2 mM EGTA, and 1 mM DTT). After left on ice for 10 min, the suspension was centrifugated at 1700 × g for 4 min to separate soluble nuclear proteins from the chromatin. Isolated chromatin was then washed once with solution B and harvested by centrifugation (10,000 × g for 1 min). Finally, chromatin was dissolved in SDS sample buﬀer and sheared by sonication (Fisher Scientiﬁc Sonic Dismembrator Model 150) on ice.

2.10. Statistics
Data were shown as mean ± standard deviation (SD) from three separate experiments.Statistical signiﬁcance was determined by Student’s t-test or one-way ANOVA for multiple comparisons with SPSS v22.0. *p < 0.05 and **p < 0.01 were considered statistically sig- niﬁcant. 3.Results
3.1. Iso inhibited HB cells both in vitro and in vivo
We ﬁrst investigated the anti-proliferative eﬀect of Iso on HB cell lines (HepG2 and HuH6) and normal hepatocytes (L-02) via CCK8 as- says. We observed that Iso dramatically inhibited the proliferation of HepG2 and HuH6 cells in a concentration-dependent manner (Fig. 1A). To further compare the cytotoxicity of Iso in hepatoma cells and normal hepatocytes, the IC50 was then calculated. The IC50 values of Iso in HepG2 and HuH6 cells were 122.3 μM and 74.92 μM, respectively. Iso had much lower cytotoxicity in normal liver cells, with an IC50 of 268.2 μM for L02. Moreover, results from colony formation assays further conﬁrmed that Iso could suppress the growth of HepG2 and HuH6 cells in a concentration-dependent manner (Fig. 1B and C). Then, the proapoptotic eﬀect of Iso in hepatoma cells was explored using ﬂow cytometry assays. When HepG2 or HuH6 cells were treated with dif- ferent doses of Iso, a signiﬁcant increase in both early apoptosis and late apoptosis was observed (Fig. 1D andE). Altogether, our ﬁndings in vitro supported that Iso might be a novel anti-tumor agent for the treatment of HB.
To further investigate the anti-tumor eﬀect of Iso in vivo, HuH6 cells were used to establish subcutaneous xenograft models. Diﬀerent doses of Iso (0, 10, 20, and 30 mg/kg) were intraperitoneally injected into male nude mice bearing human HB xenografts. Results from xenograft tumor-bearing experiments showed that the tumor volumes and weights of the Iso-treated group were statistically smaller than those of the DMSO-treated group (Fig. 2A and B). Additionally, there were no signiﬁcant diﬀerences in mice body weight between the Iso-treated

Fig. 5. Iso enhanced the antitumor eﬀects of olaparib in vivo.
HuH6-bearing nude mice were treated with of DMSO (control), olaparib, Iso or Iso in combination with olaparib. The tumors were then collected after the mice were sacriﬁced. A. Statistical analysis of tumor volumes (mm3) during the experiment. B. Statistical analysis of tumor weights (g) at the terminal time point. C. Statistical analysis of body weights (g) of nude mice at end of the experiment. Data were represented as mean ± SD. Error bars, SD. *p < 0.05 and **p < 0.01, compared to the control group. Fig. 6. Iso disrupted DDR -related protein recruited to the DNA damage site.
A-C. HuH6 cells were pre-incubated with Iso for 24 h and then exposed to 10 Gy IR. One hour after incubation, the cells were ﬁxed and processed for immunoﬂuorescence with the indicated antibodies.group and the control group (Fig. 2C). No toxicity to the kidney and liver in these nude mice was also observed (data not shown).

3.2. Iso reduced HR repair in hepatoma cells
Studies have shown that Iso inhibits the proliferation of hepatoma cells through increasing intracellular ROS levels [17]. Additionally, cellular physiological stress, ROS accumulation and DNA damage may occur simultaneously [29]. To further explore the underlying anti- tumor mechanisms of Iso in HB, the role of Iso-induced DSBs and DDR was assessed. HepG2 and HuH6 cells were treated with increasing concentrations of Iso.Then, the DSBs induced by Iso were examined via

Fig. 7. Iso attenuated ATM activation and blocked the interaction between pATM and the MRN complex.
A. HuH6 cells pre-incubated with or without Iso for 24 h were exposed to IR for 0, 30, or 60 min. After that, cells were harvested and western blot analysis was performed using the indicated antibodies. Quantiﬁcation of relative proteins expression (speciﬁc protein gray values/Tubulin gray values). B-C. HuH6 cells treated with or without Iso were exposed to 10 Gy IR for 0 or 30 min. Subsequently, cells were harvested. Finally, whole cell lysates were immunoprecipitated using an anti-phospho-ATM (B) or anti-Nbs1 (C) antibody and western blotting analysis was performed using the indicated antibodies. Normal IgG was used as a negative control for the immunoprecipitation. D. Soluble nuclear or chromatin fractions were isolated and then western blot analysis was performed using the indicated antibodies. Quantiﬁcation of relative proteins expression (speciﬁc protein gray values/Tubulin gray values). Data were represented as mean ± SD. Error bars, SD. *p < 0.05 and **p < 0.01, compared to the control group neutral comet assays. We observed that Iso triggered a signiﬁcant in- crease in endogenous DSBs in both HB cell lines (Fig. 3A and B), which was indicated by a dose-dependent increase in comet tail length. To protect genomic stability and avoid apoptosis, cancer cells usually in- itiate DSB repair pathways, mainly including HR and NHEJ, in response to DNA damage. To explore if Iso reduced DSB repair via HR or NHEJ in HB cells, we created a GFP-based HR and NHEJ reporter system by which a functional GFP gene was reconstituted following an HR or NHEJ event. As indicated in Fig. 3C, the HR eﬃciency of HepG2 and HuH6 cells treated with diﬀerent doses of Iso was signiﬁcantly de- creased while the NHEJ eﬃciency was not aﬀected in either cell line. Inhibitors targeting DNA double strand damage repair pathways have been shown to be capable of sensitizing cancer cells to DNA-damaging agents [30]. Here, HepG2 and HuH6 cell exposure to agents that could induce DSBs, including ionizing radiation (IR), hydroxyurea (HU), mi- tomycin C (MMC), camptothecin (CPT) and ultraviolet (UV), were treated with increasing doses of Iso, and then the cell viability was evaluated by CCK8 assays. We observed that Iso signiﬁcantly enhanced the cytotoxic eﬀects of various agents that caused DSBs (Fig. 3D), suggesting a role of Iso in the inhibition of DDR pathways. 3.3. Iso enhanced the antitumor efects of olaparib in HB cells both in vitro and in vivo
Our results illustrated that Iso might induce DNA damage by re- ducing the eﬃciency of HR. We next assessed the putative potentiation of anti-tumor eﬀects as a result of combining Iso with additional drugs acting through a similar mechanism of action. The PARP-1 enzyme is critical in the detection of DNA damage and repair. PARP inhibitors have been demonstrated to be capital of sensitizing tumors to DNA- damaging agents and selectively killing HR-defective cancers. Olaparib is the ﬁrst-in class PARP inhibitor approved by the FDA for patients with BRCA-mutated ovarian cancer [31]. We hypothesized that the combination of Iso with the PARP-inhibitor olaparib, could induce massive apoptosis due to the accumulation of DNA damage in cells. Then, the clone formation assay was used to assess the eﬀects of Iso in combination of olaparib with on HB. HepG2 and HuH6 cells were treated with an increasing doses of Iso in the absence or presence of diﬀerent doses of olaparib for 24 h, and then cell colonies were visua- lized and counted. Our results showed that Iso signiﬁcantly enhanced the olaparib-induced decrease in cell viability (Fig. 4A and B). Ad- ditionally, HuH6 cells were used to establish subcutaneous xenograft animal models, and the antitumor eﬀects of Iso in combination with olaparib were further investigated in vivo. Male nude mice bearing human HB xenografts were randomized to receive Iso, olaparib monotreatment or combination treatment. Results from xenograft tumor-bearing experiments showed that the tumor volumes and weights of combination treatment were signiﬁcantly decreased com- pared with those of the olaparib, Iso or DMSO-treated groups (Fig. 5A and B). Additionally, no diﬀerences in the body weights of nude mice were observed between the combination treatment and single treat- ment (Fig. 5C). Here, our ﬁndings showed that Iso in combination with olaparib signiﬁcantly increased the antitumor eﬀect without an in- crease in drug toxicity, both in vitro and in vivo, indicating that the combination therapy of Iso andolaparib might be a promising approach for cancer treatment. Olaparib increases the accumulation of single strand breaks and then converts them to irreparable DSBs during re- plication [32]. In order to detect the degree of DNA damage upon combination treatment, an immunoﬂuorescence assay was then pre- pared to evaluate DNA damage through γH2AX foci formation (Fig. 4C and D). Surprisingly, the formation of γH2AX foci was gradually re- duced with the increase of Iso doses. This showed that Iso might weaken DNA repair by reducing the activation of DDR-related proteins, suggesting that Iso might be an upstream regulator of H2AX.

3.4. Iso inhibited ATM-mediated DDR signaling pathways
In response to DSBs, activated ATM phosphorylates H2AX at sites of DNA damage, after which several DDR-related proteins such as 53BP1, RNF8 and RAD51 are recruited to the DSBs [33]. These recruited pro- teins appear as discrete nuclear foci which serve as indicators of the DNA repair response. The eﬀect of Iso on DNA repair eﬃciency was investigated in our study via observing the nuclear foci formed after DSBs. HuH6 cells were treated with IR for 1 h in the presence or absence of Iso, and the foci of DDR-related proteins (γH2AX, 53BP1, RAD51, and RNF8) were observed by confocal microscopy. Here, we found that the foci of γH2AX, 53BP1, RAD51 and RNF8 were blocked (Fig. 6A and B). Then, the foci of the MRN complexes and pATM, the initial response to DNA damage, were assessed. We found that pATM foci were inter- rupted, while MRN protein foci (including NBS1, MRE11 and RAD50) were not aﬀected in the presence of Iso (Fig. 6C). Next, the phosphor- ylation levels of ATM (S1981), H2AX (Ser139) and chk2 (PT68) were examined by western blotting (WB) in HuH6 cells. We found that all the phosphorylated proteins noted above were observed at a time point 30 and 60 min after IR treatment in controls, but the phosphorylation le- vels of these proteins were decreased signiﬁcantly without obvious changes in the total protein content after IR treatment in the Iso-treated group (Fig. 7A). Then, to examine if Iso decreased the recruitment of pATM to the MRN complex upon DSB response, an immunoprecipita- tion assay was prepared using anti-pATM and anti-Nbs1 antibodies. We observed that the binding of Nbs1 and ATM was gradually decreased or even disappeared in the Iso-treated group compared with the control under radiation-induced DNA damage (Fig. 7B and C). Additionally, we found that the enrichment of pATM members in the chromatin fraction after DSB induction was attenuated in the Iso-treated group (Fig. 7D). In conclusion, our data suggested that Iso inhibited ATM signaling, re- sulting in the failure of ATM phosphorylation and blocking the inter- action of pATM and the MRN complex under IR treatment, eventually preventing pATM and downstream proteins such as γH2AX, 53BP1, Rad51 and RNF8 from being recruited to the sites of DSBs.

4.Discussion
In the current study, we investigated the antitumor eﬀect of Iso on HB and explored the underlying mechanism. We found that Iso reduced cell viability in HepG2 and HuH6 cells in a concentration-dependent manner and the induction of apoptosis was the major cause of Iso-in- duced cytotoxicity. Additionally, results from xenograft tumor-bearing experiments supported that Iso was a novel antitumor agent. We sug- gested for the ﬁrst time that Iso-induced apoptosis correlated with DSBs. Using neutral comet assays and GFP-based HR and NHEJ reporter systems, we found that Iso generated DNA damage in a dose-dependent manner and attenuated HR repair. Additionally, the combination of Iso and olaparib resulted in greater cytotoxicity than single treatment, suggesting synergistic eﬀects of the two agents both in vivo and in vitro. To explore the potential mechanism of the involvement of Iso in DDR, immunoﬂuorescence and immunoprecipitation assays were performed to detect DDR-related proteins. These ﬁndings suggested that only the MRN complex was recruited to the DSB, while pATM and its down- stream substrate were not. Importantly, both pATM and the MRN complex were blocked by Iso. These results showed that Iso triggered cell apoptosis by inducing DSBs and blocked the initiation process of HR through attenuation of ATM activation, while inhibiting the binding of pATM and the MRN complex.
Previous studies have demonstrated that Iso is lethal to various tumors. The anti-tumor eﬀects of Iso were found to be associated with the inhibition of proliferation and the promotion of apoptosis of cancer cells. Studies have reported that an increase of ROS by interfering with mitochondrial function and the MAPK or PI3K/Akt signaling pathways are involved in Iso-induced apoptosis [10,16,17,34,35]. Here, we found that Iso inhibited proliferation and triggered apoptosis of HB cells in a concentration- and time-dependent manner, which was consistent with previous studies to some extent. Above all, we discovered for the ﬁrst time that inducing DNA damage and reducing HR in DDR contributed to the proapoptotic eﬀect of Iso, suggesting that Iso might act as a DNA damage agent and DNA repair inhibitor.

Traditional radiotherapy and chemotherapy are limited by severe side eﬀects or susceptibility to drug resistance. Hence, molecular tar- geted therapy or combination therapy has been eﬀective strategies to overcome drug resistance and avoid side eﬀects. Previous study re- ported that zinc oxide nanoparticles enhanced the sensitivity of liver cancer cells to Iso [36]. Olaparib, a PARP inhibitor, kills cancer cells that harbor HR defects. However, olaparib has a limited eﬀect on tu- mors with high HR eﬃciency. Recently, enhanced anticancer eﬀects of combination therapy consisting of olaparib and other drugs, including DNA damaging agents and agents that suppressed HR have been re- ported [37-39]. In our research, we ﬁrst found that Iso could attenuate HR in DDR and it could signiﬁcantly enhance the cytotoxic eﬀects of olaparib both in vitro and in vivo, thus expanding the application scope of olaparib.Several studies have shown that the MRN complex is essential for the detection of DSBs and it also acts as a primary damage sensor in- volved in the early steps of HR [40,41]. ATM can be activated by DNA damage and then plays a central role in cellular response to DSBs [42,43]. Previous studies have found that the MRN complex is indis- pensable in facilitating ATM activation and subsequent activity on other substrates, such as CHK2 and histone H2AX [44,45]. This suggests that the MRN complex may not only be a downstream eﬀector of ATM but may also act as an upstream or a cofactor for ATM to initiate the phosphorylation of cellular substrates. In our study, the levels of phosphorylation of ATM and its downstream proteins, including CHK2 and H2AX, were partially decreased under Iso treatment, indicating that Iso could inhibit ATM activation and its downstream proteins to some extent.

However, a little activation of ATM was also observed here, which might be due to rapid autophosphorylation of ATM on serine 1981 induced by IR. Previous studies have shown that a fraction of ATM is initially activated due to autophosphorylation, which facil- itates a direct interaction between the pATM and MRN complexes, eventually resulting in the recruitment of pATM to DSBs [46]. However, we found that Iso blocked the interaction Perinatally HIV infected children of pATM and the MRN complex, thus inhibiting ATM recruitment to chromatin and substrate phosphorylation. ATM is a master regulator of the DDR, and its acti- vation can be modulated by many factors. Previous studies have con- ﬁrmed that the MRN complex can activate ATM via redox-dependent mechanisms and that ATM Timed Up and Go is associated with a diverse set of signaling pathways involved in the regulation of cancer [46,47]. Here, our re- search described a mechanism for the regulation of ATM activation by Iso, providing a theoretical basis for future research.In summary, our study showed that Iso acted as a DNA damaging agent, subsequently attenuating ATM phosphorylation and blocking its binding to the MRN complex to reduce HR eﬃciency, leading to DSB- triggered cell apoptosis in HB cells. Signiﬁcantly, combination treat- ment with Iso and olaparib might enhance the apoptosis of HB cells, thus providing an attractive strategy for the treatment of HB patients.