Ellipticine cytotoxicity to cancer cell lines — a comparative study

Ellipticine is a potent antineoplastic agent exhibiting multiple mechanisms of action. This anticancer agent should be considered a pro-drug, whose pharmacological efficiency and/or genotoxic side effects are dependent on its cytochrome P450 (CYP)- and/or peroxidase-mediated activation to species forming covalent DNA adducts. Ellipticine can also act as an inhibitor or inducer of biotransformation enzymes, thereby modulating its own metabolism leading to its genotoxic and pharmacological effects. Here, a comparison of the toxicity of ellipticine to human breast adenocarcinoma MCF-7 cells, leukemia HL-60 and CCRF-CEM cells, neuroblastoma IMR-32, UKF-NB-3 and UKF-NB-4 cells and U87MG glioblastoma cells and mechanisms of its action to these cells were evaluated. Treatment of all cells tested with ellipticine resulted in inhibition of cell growth and proliferation. This effect was associated with formation of two covalent ellipticine-derived DNA adducts, identical to those formed by 13-hydroxy- and 12-hydroxyellipticine, the ellipticine metabolites generated by CYP and peroxidase enzymes, in MCF-7, HL-60, CCRF-CEM, UKF-NB-3, UKF-NB-4 and U87MG cells, but not in neuroblastoma UKF-NB-3 cells. Therefore, DNA adduct formation in most cancer cell lines tested in this comparative study might be the predominant cause of their sensitivity to ellipticine treatment, whereas other mechanisms of ellipticine action also contribute to its cytotoxicity to neuroblastoma UKF-NB-3 cells.

Neurospora crassa, and mammalian cells and induce prophage lambda in Escherichia coli (for an overview see Stiborová et al., 2001).
Ellipticine has been reported to arrest cell cycle progression by regulating the expression of cyclin B1 and Cdc2 as well as phosphorylation of Cdc2 (Kuo et al., 2005a; to induce apoptotic cell death by generation of cytotoxic free radicals, activation of Fas/Fas ligand system, regulation of Bcl-2 family proteins (Kuo et al., 2005a;, increase of wild-type p53, rescue of mutant p53 activity and initiation of the mitochondrial apoptosis pathway (Garbett and Graves., 2004;Kuo et al., 2005a;. Ellipticine also activates the p53 pathway in glioblastoma cells; its impact on these cancer cells depends on the p53 status. In a U87MG glioblastoma cell line expressing p53wt, ellipticine provoked an early G0/G1 cell cycle arrest, whereas in a U373 cell line expressing p53mt it caused arrest in S and G2/M phase (Martínková et al., 2010).
Ellipticine and 9-hydroxyellipticine also cause selective inhibition of p53 protein phosphorylation in several Introduction Ellipticine (5,carbazole, Figure 1), an alkaloid isolated from Apocyanaceae plants, exhibits significant antitumor and anti-HIV activities (for a summary see Stiborová et al., 2001). The main reason for the interest in ellipticine and its derivatives for clinical purposes is their high efficiency against several types of cancer, their rather limited toxic side effects, and their complete lack of hematological toxicity (Auclair, 1987). Nevertheless, ellipticine is a potent mutagen. Most ellipticine derivatives are mutagenic to Salmonella typhimurium Ames tester strains, bacteriophage T4, human cancer cell lines (Ohashi et al., 1995;Sugikawa et al., 1999) and this correlates with their cytotoxic activity. However, the precise molecular mechanism responsible for these effects has not been explained yet. Chemotherapyinduced cell cycle arrest was shown to result from DNA damage caused by a variety of chemotherapeutics. In the case of ellipticine, it has been suggested that the prevalent DNA-mediated mechanisms of its antitumor, mutagenic and cytotoxic activities are (i) intercalation into DNA and (ii) inhibition of DNA topoisomerase II activity (Auclair, 1987;Garbett & Graves, 2004;Stiborová et al., 2006;2011).
In order to evaluate contributions of DNA adduct formation by ellipticine to its toxicity to cancer cells, a comparison of cytotoxicity and of DNA adduct formation by 0.1-10 μM ellipticine in different cancer cell lines was performed. The 32 P-postlabeling method was used to determine DNA adduct formation by ellipticine and cytotoxicity of ellipticine was determined with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazoliumbromide (MTT) assay (Cinatl et al., 1997). Human cancer cell lines sensitive to ellipticine such as human breast adenocarcinoma, leukemia, neuroblastoma and glioblastoma cancer cells were utilized for this study.

Chemicals
Ellipticine was obtained from Sigma (St. Louis, MO, USA). All other chemicals used in the experiments were of analytical purity or better.

MTT assay
The cytotoxicity of ellipticine was determined by MTT test. For a dose-response curve, solution of ellipticine in dimethyl sulfoxide (DMSO ) (1 mM) was dissolved in culture medium to final concentrations of 0-10 μM. Cells in exponential growth were seeded at 1 × 10 4 per well in a 96-well microplate. After incubation (48 hours) at 37 °C in 5% CO 2 saturated atmosphere the MTT solution (2 mg/ml PBS) was added, the microplates were incubated for 4 hours and cells lysed in 50% N,N-dimethylformamide containing 20% of sodium dodecyl sulfate (SDS), pH 4.5. The absorbance at 570 nm was measured for each well by multiwell ELISA reader Versamax (Molecular devices, CA, USA). The mean absorbance of medium controls was subtracted as a background. The viability of control cells was taken as 100% and the values of treated cells were calculated as a percentage of control. The IC 50 values were calculated from at least 3 independent experiments using linear regression of the dose-log response curves by SOFTmaxPro.

Treatment of cancer cell lines with ellipticine for DNA adduct analyses
Cell lines were seeded 24 hr prior to treatment at a density of 1 × 10 5 cells/ml in two 75 cm 3 culture flasks in a total volume of 20 ml of IMDM. Ellipticine was dissolved in 20 μl of DMSO, the final concentration was 0, 0.1, 1, 5 or 10 μM. After 48 h the cells were harvested after trypsinizing by centrifugation at 2 000 × g for 3 min and two washing steps with 5 ml of PBS yielded a cell pellet, which was stored at -20 °C until DNA isolation. DNA was isolated and labeled as described in the next section.

HPLC analysis of 32 P-labeled DNA adducts
HPLC analysis was performed essentially as described previously (Stiborová et al., 2003a;. Individual spots detected by 32 P-postlabeling were excised from the thin layer and extracted (Stiborová et al., 2003a;. Cut-outs were extracted with two 800 μl portions of 6 M ammonium hydroxide/isopropanol (1:1) for 40 min. The eluent was evaporated in a Speed-Vac centrifuge.
The dried extracts were dissolved in 100 μl of methanol/ phosphate buffer (pH 3.5) 1:1 (v/v). Aliquots (50 μl) were analyzed on a phenyl-modified reversed-phase column (250 mm × 4.6 mm, 5 μm Zorbax Phenyl; Säulentechnik Knauer, Berlin, Germany) with a linear gradient of methanol (from 40 to 80% in 45 min) in aqueous 0.5 M sodium phosphate and 0.5 M phosphoric acid (pH 3.5) at a flow rate of 0.9 ml/min. Radioactivity eluting from the column was measured by monitoring Cerenkov radiation with a Berthold LB 506 C-I flow-through radioactivity monitor (500 μl cell, dwell time 6 s).

cells (L), leukemia HL-60 (M) and CCRF-CEM cells (N), neuroblastoma UK-NB-4 cells (O)
and glioblastoma U87MG cells (P) treated with ellipticine. Cancer cells lines were treated with 10 μM ellipticine except of HL-60 cells that were treated with 5 μM ellipticine. Adduct spots 1-7 correspond to the ellipticine-derived DNA adducts. Besides adduct 2 formed by 12-hydroxyellipticine, another strong adduct (spot X in panel G) was generated, which was not found in any other activation systems or in vivo. Experimental conditions for panels A-J are described in our previous studies 2007a) and those for panels L-P in Materials and Methods. Experimental conditions for panel K were as follows: female Wistar rats bearing the N-methyl-N-nitrosourea induced mammary adenocarcinoma (McCormick et al., 1981) were i.p. treated with 4 mg ellipticine per kilogram body weight. Ellipticine was administered dissolved in 1% acetic acid at a concentration of 2.5 mg/ml. One day after ellipticine treatment, DNA from tumor and normal breast tissues (McCormick et al., 1981) was isolated and analyzed for formation of DNA adducts using the nuclease P1 version of the 32 P-postlabeling assay as described (Stiborová et al., 2001). All experiments with animal models were conducted in accordance with the Regulations for the Care and Use of Laboratory Animals (311/1997, Ministry of Agriculture, Czech Republic), which is in compliance with the Declaration of Helsinki.  Table 1). When sensitivity of additional cells to ellipticine was compared, cytotoxicity of this agent to human breast adenocarcinoma MCF-7 cells and a glioblastoma U87MG cell line was found to be comparable (the IC 50 values were around 1 μM), while leukemia CCRF-CEM cells were less sensitive. The IC 50 value for ellipticine was almost 4-times higher in these leukemia cells than in MCF-7 and U87MG cells ( Table 1).

Determination of DNA adduct formation by ellipticine in human breast adenocarcinoma MCF-7, leukemia HL-60 and CCRF-CEM cells, neuroblastoma IMR-32, UKF-NB-3 and UKF-NB-4 cells, and glioblastoma U87MG cells
The cell lines shown to be sensitive to ellipticine (Table 1) were treated with increasing concentrations of ellipticine (0.1-10 μM) for 48 h and DNA was isolated from these cells. Using the nuclease P1 version of 32 P-postlabeling assay, which was found to be suitable to detect and quantify DNA adducts formed by ellipticine (Stiborová et al., 2001;2003a;2007a;b, 2008;2011), ellipticinederived adducts were detected in the DNA of these cells ( Figure 2L-P, Table 2). Two major ellipticine-DNA adducts (spots 1 and 2 in Figure 2) were formed in all cells (Table 2). No adducts were detected in DNA of control cells treated with solvent only. Chromatographic properties of the two major adduct spots on PEI-cellulose TLC plates (spots 1 and 2) were similar to those of ellipticine-derived DNA adducts found previously after in vitro incubation of calf thymus DNA with ellipticine and isolated CYPs (Stiborová et al., 2001;2003b;, or peroxidases (Stiborová et al., 2007a) or in vivo ( Figure 2D), in several organs of rats (Stiborová et al., 2003a;2007b) and mice  exposed to this agent. Both these adducts were found to be generated from 13-hydroxy-and 12-hydroxyellipticine ( Figure 2E,F), as confirmed by cochromatographic analysis using TLC and HPLC (data not shown). Both adducts were identified as deoxyguanosine adducts in DNA (Stiborová et al., 2003b;Moserová et al., 2008)). Besides these adducts, additional two minor adducts (spots 6 and 7 in Figure 5C,D) were detected in DNA of MCF-7 (spot 6) and UKF-NB-4 (spots 6 and 7) cells treated with 10 μM ellipticine (Table 2, Figure 2). Both these minor adducts are known to be generated in vitro mainly by peroxidase-catalyzed oxidation (Poljaková et al., 2006;Stiborová et al., 2007a). The low levels of these adducts prevented HPLC co-chromatographic analysis or their further characterization.

Discussion
The results of this study show a comparison of ellipticine cytotoxicity to several human cancer cell lines (breast adenocarcinoma MCF-7, leukemia HL-60 and CCRF-CEM cells, neuroblastoma IMR-32, UKF-NB-3, UKF-NB-4 lines and glioblastoma U87MG cells). In addition, the mechanism of ellipticine cytotoxicity to these cells was evaluated. The mode of antitumor, cytotoxic and mutagenic action of ellipticine is considered to be based mainly on DNA damage, such as intercalation into DNA, inhibition of topoisomerase II, and formation of covalent DNA adducts mediated by CYPs and peroxidases (Auclair, 1987, Stiborová et al., 20012011;Garbett & Graves, 2004). Intercalation of ellipticine into DNA and inhibition of topoisomerase II occur in all cell types irrespective of their metabolic capacity, because of the general chemical properties of this drug and its affinity to DNA and topoisomerase II protein (Auclair, 1987). However, the formation of ellipticine-DNA adducts, which is dependent on ellipticine activation by CYPs and peroxidases, has not yet been proven as a general mechanism. we found this ellipticine action unambiguously in vitro, using several CYP and peroxidase enzymes for ellipticine activation (Stiborová et al., 2001;2003a;2007a) and in vivo in rats and mice (Stiborová et al., 2003a;2007b;. In our former studies Poljaková et al., 2007;Martínková et al., 2009) and in the present work, ellipticine-DNA adducts were detected also in several human cancer cell lines. Here we evaluated a contribution of this mechanism to ellipticine toxicity to these cancer cells.
Toxic effects of ellipticine to leukemia HL-60 and CCRF-CEM cells (expressed as IC 50 values) correspond to levels of ellipticine-DNA adducts formed in these cells. This finding indicates that covalent modification of DNA In the case of two of neuroblastoma cell lines tested, IMR-32 and UKF-NB-4, toxic effects of ellipticine to these cells also correspond to levels of ellipticine-DNA adducts formed in these cells. The cytotoxic activity of ellipticine to IMR-32 and UKF-NB-4 neuroblastoma cell lines was also previously found to be a consequence of the formation of ellipticine-DNA adducts (Poljaková et al., 2009). In addition, the role of ellipticine-DNA adduct formation in cytotoxicity of this drug to neuroblastoma cells was further supported by the finding that a decrease in the levels of these adducts in IMR-32 and UKF-NB-4 cells under hypoxic conditions correlated with a decrease in toxicity of ellipticine under these conditions (Poljaková et al., 2009). This is, however, not the case of the UKF-NB-3 cell line; lower levels of DNA adducts were found in these cells than in UKF-NB-4 cells, although both neuroblastoma cells exhibited similar sensitivity to ellipticine. All these findings suggest that DNA adduct formation by ellipticine might be the predominant mechanism responsible for ellipticine cytotoxicity to most of cancer cell lines tested in this work, except UKF-NB-3 neuroblastoma cells. Thus the DNA adduct formation by ellipticine is probably not the major mechanism responsible for ellipticine cytotoxicity to UKF-NB-3 neuroblastoma cells. Other mechanisms such as intercalation into DNA (Auclair, 1987;Singh et al., 1994) and inhibition of DNA topoisomerase II activity (Auclair, 1987;Monnot et al., 1991;Fossé et al., 1992;Froelich-Ammon et al., 1995) that were found to be additional DNA-mediated mechanisms of ellipticine antitumor, mutagenic and cytotoxic activities [for a summary see (Stiborová et al., 2001;2011)] seem to contribute to ellipticine cytotoxicity to these neuroblastoma cells. In order to resolve the real contribution of covalent DNA adduct formation by ellipticine in these cells, additional DNA damage such as intercalation of ellipticine into DNA in these and other cancer cells should be assessed.
Recently, we used square wave voltametry to evaluate intercalation of ellipticine into DNA in vitro and covalent modification of DNA by ellipticine in a UKF-NB-3 neuroblastoma cell line and found participation of both these mechanisms in DNA damage caused by this drug in this neuroblastoma cell line (Huska et al., 2010a;Huska et al., 2010b). Nevertheless, further detailed studies have to be performed which would examine differences in electrochemical signals of DNA with intercalated ellipticine and DNA covalently modified by this antitumor agent. In our former studies, toxic effects of ellipticine to several cancer cells were found to be dependent on expression of CYP1A1, 1B1, 3A4 and peroxidases LPO, COX and MPO in these cells Poljaková et al., 2007;Martínková et al., 2009). Moreover, expression of CYP1A1, 1B1 and 3A4 enzymes was found to be induced by treating glioblastoma U87MG, adenocarcinoma MCF-7 and neuroblastoma UKF-NB-4 cells with ellipticine (Martínková et al., 2009;Stiborová et al., unpublished results). Likewise, CYP1A1 was found to be induced by ellipticine in rats in vivo . Hence, the sensitivity of individual cancer cells to ellipticine caused by covalent modification of DNA might be dependent on the capability of ellipticine to induce the CYP enzymes oxidizing ellipticine to 13-hydroxy-and 12-hydroxyellipticine, the metabolites generating DNA adducts. This suggestion, however, needs to be confirmed by further investigations including in vivo studies. This feature is one of the aims of investigations in our laboratory.