Time course changes of anti- and pro-apoptotic proteins in apigenin-induced genotoxicity

Background Apigenin (4′,5,7-trihydroxyflavone, AP), an active component of many medicinal Chinese herbs, exhibits anticancer properties in vitro and in vivo. This study aims to investigate the genotoxic, cytostatic, and cytotoxic effects of AP and time course changes in the levels of anti- and pro-apoptotic proteins involved in the DNA damage response in HepG2 cells. Methods The genotoxic potential of AP was determined by sister chromatid exchanges (SCEs) and chromosomal aberrations (CAs) analysis. The levels of cytostaticity and cytotoxicity were evaluated by the proliferation rate and mitotic indices, respectively. MTT was used to study cytotoxicity, while the induction of apoptosis and the expression of apoptosis-related proteins were determined by ELISA. Results At concentrations greater than 10 μM, AP decreased cell survival in a dose- (48 h: 10 vs. 20 μΜ, P < 0.001 and 20 vs. 50 μΜ, P = 0.005; 72 h: 10 vs. 20 μΜ, P < 0.001 and 20 vs. 50 μΜ, P = 0.001) and time-dependent manner (20 μΜ: 24 vs. 48 h, P < 0.001 and 48 vs. 72 h, P = 0.003; 50 μΜ: 24 vs. 48 h, P < 0.001 and 48 vs. 72 h, P < 0.001; 100 μΜ: 24 vs. 48 h, P < 0.001 and 48 vs. 72 h, P < 0.001). SCEs rates, cell proliferation, and mitotic divisions were also affected in a dose-dependent manner (P < 0.001). There was no change in the frequency of aberrant cells (1 μΜ ΑP: P = 0.554; 10 μM AP: P = 0.337; 20 μΜ AP: P = 0.239). Bcl-2 levels were reduced 3 h after AP administration (P = 0.003) and remained reduced throughout the 48 h observation period (6 h, P = 0.044; 12 h, P = 0.001; 24 h, P = 0.042; 48 h, P = 0.012). Bax and soluble Fas exhibited a transient upregulation 24 h after AP treatment. The Bax/Bcl-2 ratio was also increased at 12 h and remained increased throughout the 48 h observation period. Conclusion AP exhibited dose-dependent genotoxic potential in HepG2 cells. The protein levels of sFas, Bcl-2, and Bax were affected by AP to promote cell survival and cell death, respectively.

Although the properties of AP against various prooxidant and clastogenic agents have been studied [11,[34][35][36], there is little information on the genotoxic potential of this particular flavonoid. AP was highly clastogenic in Chinese hamster V79 cells and induced micronuclei formation in human peripheral lymphocytes in a dose-dependent manner [37,38]. Other reports mentioned that AP could intercalate into both calf thymus DNA and RNA [39,40]. The generation of DNA single-strand (SSBs) and double-strand breaks (DSBs) by DNA-crosslinking agents [41,42], could lead to sister chromatid exchanges (SCEs) or chromosomal aberrations (CAs) [43]. An in vivo and in vitro study demonstrated that AP can remodel chromatin by inhibiting class I histone deacetylases. This affects regulation, expression, and activation of various DNA damage response genes, which results in cell cycle arrest, and apoptosis. These affected genes include ATM and ATR, which participate in DSBs repair via homologous recombination [44,45].
The SCEs assay is a sensitive, simple, and rapid method to detect DNA damage and repair at low concentrations of potential genotoxic or anti-genotoxic agents [43,[46][47][48][49]. SCEs represent a useful tool in monitoring and improving chemotherapeutic strategies in vitro and in vivo [48,[50][51][52][53][54]. The efficacy of potential antitumor agents in inducing SCEs formation in vitro and in vivo correlates positively with the in vivo tumor's response to these agents [55,56]. CAs analysis is another genotoxic endpoint [43,46]. A high frequency of CAs can lead to cell death, and it has been associated with increased overall cancer risk [43,46,57,58].
AP's ability to intercalate into DNA, remodel chromatin, and upregulate p53 and p21 proteins [25,39,40,44,[59][60][61] directed us to study the genotoxic potential of this flavonoid in HepG2 cells. We also investigated the proliferation rate index (PRI) and the mitotic index (MI), markers of the cytostatic and cytotoxic properties of chemical and physical agents, respectively [49]. The time course changes in the levels of anti-and pro-apoptotic proteins involved in the DNA damage response were also investigated.

Cell cultures
HepG2 cells were maintained in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin solution, in a 37°C humidified incubator under an atmosphere of 5% CO 2 . On attaining 75-80% confluency the cells were subcultured by trypsinization and then seeded in appropriate cell numbers depending on the type of the experiments. All experiments took place 24 h after seeding.

Cytotoxicity assay
The cytotoxic potential of AP was evaluated at 24, 48, and 72 h by the MTT method. HepG2 cells were seeded in 96well plates at a density of 10 4 cells per well in 100 μL of complete culture medium. Cells were incubated with 0.1, 1, 5, 10, 20, 50, and 100 μΜ of AP or 0.1% DMSO (vehicle control). AP stock solution was prepared in DMSO and diluted in complete culture medium to the desired concentrations (0.1, 1, 5, 10, 20, 50, and 100 μΜ). At each time point, eight replicate cultures for each concentration were studied in three independent experiments. At the end of the specified incubation period (24, 48, and 72 h), the medium was discarded and each well received 200 μL of fresh medium containing 20 μL of MTT solution (5 mg/ml in phosphate buffered saline) for 4 h. MTT crystals were dissolved by adding 100 μL 0.04 M HCL/isopropanol, for fifteen minutes at 37°C. Absorbance was determined at 570 nm by an ExpertPlus microplate reader (ASYS Hitech GmbH, Austria). Absorbance was normalized to vehicletreated control cultures (equivalent to 100% cell viability).

SCEs and CAs analysis
For SCEs and CAs determination, 2 × 10 5 cells were treated with 1, 10, 20, and 50 μΜ of AP and 5 μg/mL of 5-bromodeoxyuridine for 72 h. Colcemid was added to all cultures 24 h before metaphase harvesting. At the end of the incubation period, the cells were scraped and centrifuged (Z300, Hermle Labortechnik GmbH, Germany) at 200 × g for 10 min. Pelleted cells were then treated with 5 mL of hypotonic KCl solution (0.075 M), at 37°C for 25 min. Fixative solution (3:1, methanol:acetic acid) was added to the cell suspension and was followed by another centrifugation at 200 × g for 10 min. The pellet was washed three times in fixative solution and stored at −20°C until further assayed. All treatments were examined in three independent experiments.

Fluorescence plus Giemsa
SCEs were visualized by a modified fluorescence plus Giemsa (FPG) technique [62]. Metaphase spreads were incubated in bisbenzimide H33258 solution (0.1 mg/mL) for 20 min at room temperature. A few drops of McIlivaine's buffer (pH 8; 0.1 M citric acid and 0.2 M disodium phosphate) were applied to each slide and overlaid with a coverslip. After that, the slides were exposed to UV light for 90 min and stained with 7% Giemsa solution in Gurr buffer (pH 6.8).
Since the number of chromosomes in HepG2 cells varies from 50 to 60 (modal number: 55), the SCEs/ chromosome rate was estimated. SCEs were evaluated in more than 60 well-spread second-division metaphases for each treatment. CAs were evaluated in 300 first division metaphases for each treatment. The criteria to classify different types of aberrations were in accordance with the recommendations of IPCS guidelines [43]. CAs were classified as chromatid gaps (chtg), chromatid breaks (chtb), chromosome gaps (chrg), chromosome breaks (chrb), ring (r), and dicentric chromosomes (d). Gaps were not included in the determination of total aberrant cells per treatment. Scoring was performed in a blind fashion.

PRI and MI assessment
For PRI and MI determinations, more than 450 cells and 4,500 nuclei were scored for each treatment, respectively. The proportion of cells in the first, second, third, or subsequent mitotic division was evaluated to determine the PRI. PRI was established according to the following formula: where ΣM 1 is the sum of cells in the first mitotic division, ΣM 2 in the second and ΣM 3+ in the third or subsequent mitotic divisions, while N is the total number of cells scored [63]. MI was expressed as the number of cells at metaphase per 1,000 analyzed nuclei (‰).

Assessment of apoptosis
As an index of apoptosis, cytoplasmic histone-associated DNA fragments were photometrically determined by the Cell Death Detection ΕLISA Plus kit according to the manufacturer's instructions. Ten thousand cells per well were seeded in 96-well plates and treated with AP (20 μΜ) for 24 h (six replicate cultures). Cells were lysed and centrifuged at 130 × g. Supernatants were transferred into a streptavidin-coated microplate and simultaneously incubated with a monoclonal mouse biotinylated anti-histone antibody (clone H11-4) and a monoclonal mouse peroxidase-conjugated anti-DNA antibody (clone MCA-33) at room temperature for 2 h. DNA-histone complex was used as a positive control. Absorbance was determined at 405 nm by an ExpertPlus microplate reader. Background values were subtracted from the measurements.

Statistical analysis
Data were expressed as mean ± standard deviation (SD). SCEs and CAs values were logarithmically transformed before further analysis because data were not normally distributed. Statistical analysis was performed by Student's t-test for individual comparisons between control and apigenin-treated cultures. Multiple comparisons among various AP treatments were carried out by oneway ANOVA followed by Bonferroni's post hoc test. Linear regression analysis was used to determine doseresponse relationships. Pearson's correlation coefficient was also determined. All statistical analyses were performed by SPPS version 16 (IBM, USA). All tests were two-tailed and P values less than 0.05 were considered statistically significant. For Student's t-test, significance levels were adjusted to 0.01 to reduce the overall Type I error.  Figure 1A). Similar results were reported by Chiang et al. [25] and Choi et al. [27], who studied the antiproliferative effect of AP in HepG2 cells. In agreement with Khan et al. [26], AP induced apoptosis at 24 h, eliciting an 11-fold increase in cytoplasmic histone-associated DNA fragments (P < 0.001) ( Figure 1B). AP (100 μΜ) induced DNA damage in Chinese hamster V79 cells and increased the frequency of micronuclei and CAs (concentrations higher than 37 μM) in human peripheral lymphocytes [34,[36][37][38]64,65]. In our study, we observed no effect on the frequency of aberrant cells (Table 1) but we noticed an increase in the frequency of SCEs at much lower concentrations (10 and 20 μΜ). However, 50 μΜ of AP was highly cytotoxic, making SCEs evaluation impossible (Table 2). This could be attributed to the high sensitivity of the SCEs method for detecting DNA damage and repair at doses that have little or no effect on CAs frequency [46][47][48][49]66]. Moreover, the independent mechanisms leading to SCEs and CAs formation [46,[67][68][69] and the use of different cell systems could account for the differences. HepG2 cells have a polymorphic genetic profile with a variety of structural and numerical chromosomal abnormalities [70][71][72]. Sixty-seven breakpoints were identified in liver cancer cell lines including HepG2 cells [72]. Zimonjic et al. [71] performed comparative genomic hybridization analysis in 18 liver cancer cell lines and reported that regions exhibiting gain or loss, ranged from whole chromosome arms to a medium band of a 400-band  ideogram. These previous findings could justify the high frequency of aberrant cells found in the control groups in the present study. Linear regression analysis revealed a dose-response relationship between AP and SCEs frequencies, cell proliferation, and mitotic divisions (R = 0.798, R 2 = 0.636, P < 0.001; R = −0.883, R 2 = 0.781, P < 0.001; R = −0.820, R 2 = 0.672, P < 0.001, respectively). AP concentration was positively correlated with the first and second mitotic division metaphases (R = 0.837, R 2 = 0.700, P < 0.001; R = 0.768, R 2 = 0.589, P < 0.001, respectively), while the third and subsequent mitotic division metaphases were negatively correlated (R = −0.867, R 2 = 0.751, P < 0.001) with AP ( Table 3). The flavonoid's genotoxic potential was correlated with increased cytostaticity (SCEs vs. PRI: R = −0.582, R 2 = 0.339, P = 0.018) and cytotoxicity (SCEs vs. MI: R = −0.573, R 2 = 0.329, P = 0.032).

Results and discussion
DSBs are repaired by homologous recombination, in which SCEs play an important role [73][74][75]. Iijima et al. [76] reported that NBS1, a protein involved in cellular responses to DSBs [77], regulated Bax activation in DNA damage-induced apoptosis. Furthermore, pro-apoptotic BAX could comprise a p53 downstream target gene  through the direct binding of p53 to cofactors ASPP1 and ASPP2 [78,79]. Bcl-2 could protect cells against cell death induced by ionizing radiation, alkylating agents, and various chemotherapeutic drugs [80][81][82][83]. Formation of DSBs by severe DNA damage triggered Bcl-2 decline and activated caspase-9 and caspase-3 [84]. In our study, AP treatment affected both Bcl-2 and Bax protein levels. Bcl-2 expression was downregulated at 3-48 h ( Figure 2B). Bax levels were significantly lower in apigenin-treated cells at 1 and 3 h (P = 0.003 and P < 0.001, respectively) ( Figure 2C). At 24 h, Bax expression was significantly upregulated compared with the corresponding vehicletreated cultures (P = 0.005) and the respective 12 h of treatment (P = 0.001). Bax's upregulation was transient and significantly reduced to the corresponding control cultures levels after 48 h. The Bax/Bcl-2 ratio, which is indicative of the mitochondrial induced apoptotic potential, exhibited a noticeable increase from 12-48 h of treatment ( Figure 2D). Similar changes in the Bax/Bcl-2 ratio were observed in human lung A549 cancer cells and human prostate carcinoma DU145 cells [32,33]. Genotoxic agents could lead to cell death through the Fas/FasL mediated apoptotic pathway [85,86]. In the present study, mFasL and sFasL were undetectable at all time points in all cultures. Nevertheless, sFas increased in a time-dependent manner in untreated and treated cultures (Control cultures: 6 vs. 12 h, P = 0.005; 12 vs. 24 h, P = 0.001. Apigenin treatment: 3 vs. 6h, P = 0.021; 6 vs. 12 h, P = 0.001; 12 vs. 24 h, P = 0.001; 24 vs. 48 h, P = 0.001) (Figure 2A). AP increased sFas levels at 1 and 24 h, compared with the corresponding control cultures (P = 0.002 and P = 0.001, respectively). Upregulation of sFas at 24 h, was transient and significantly reduced at 48 h, compared with the respective 24 h treatment (P = 0.001) and the corresponding control cultures (P = 0.001). There are no previous reports on the effect of AP on alternatively spliced FAS. Fas-mediated signaling is not limited to inducing cell death, and its expression in various cell types does not always correlate with susceptibility to the Fas-mediated apoptotic pathway [86,87]. Alternatively, spliced FAS variants, which encode soluble forms of the receptor, could inhibit apoptosis [88][89][90]. Because of the sharp Bcl-2 downregulation at 12 h, sFas upregulation might represent a rescuing mechanism as a means to prevail over cell death signals. Filippov et al. [91] reported that cells in response to exogenous stress, such as the effect of a genotoxic agent, regulated the expression of specific splicing factors, altering the splicing profile of target genes such as CD44 and FAS. Since ROS induce alternative splicing, it is possible that free oxygen radical generation by AP could justify the induction of sFas expression [27,92,93]. AP's intercalation into DNA/RNA might have also contributed to these changes in the expression of FAS.

Conclusion
AP exhibited dose-dependent genotoxic potential that led to changes in sFas, Bcl-2, and Bax protein levels in HepG2 cells.

Competing interests
The authors declare that they have no competing interests.
Authors' contributions CS, EC and AKT designed and supervised the study. FP wrote the manuscript and conducted the experiments. AtP, IK, NK, and AlP assisted experimentally and drafted parts of the manuscript. GT and FP performed the statistical analysis. All authors read and approved the final version of the manuscript.