Triazoloacridone C-1305 is a novel inhibitor of DNA topoisomerase II, which exhibits potent antitumor activity toward solid tumors. In this study, antiproliferative action of C-1305 and its close analog C-1533 was investigated in nontransformed mouse fibroblasts and two mutant cell lines in which the PARP-1 gene was specifically disrupted. Unexpectedly, C-1305 very strongly affected proliferation of cells lacking poly(ADP-ribose) polymerase-1 (PARP-1), whereas the action of less active compound C-1533 toward normal and PARP-1-negative cells was comparable. The IC50 concentration of C-1305 determined for PARP-1 knockout cells was ∼150-fold lower than that determined for cells with functional PARP-1. Both studied triazoloacridones exhibited very low direct cytotoxicity as evidenced by accumulation of 7-amino-actinomycin D, and only low levels of apoptosis were observed after a 24-h exposure to studied drugs. Analysis of DNA damage induced by C-1305 by the Comet assay showed that this drug induced very low levels of DNA strand breaks. C-1305 strongly affected cell cycle progression in normal and PARP-1 mutant cells and arrested both cell types in G2-M phase. However, the G2-M arrest induced by C-1305 was greatly prolonged in PARP-1-deficient cells as compared with normal fibroblasts. Together, these results show that mouse cells lacking PARP-1 are extremely sensitive to C-1305, a new topoisomerase II inhibitor. This is in striking contrast with previous reports in which PARP-1-deficient cells were shown to be resistant to classical topoisomerase II inhibitors. Our data also suggest that the PARP-1 status might be essential for the maintenance of the G2 arrest induced by C-1305.

Type II DNA topoisomerase is a molecular target of many antitumor drugs used in the treatment of human cancers. DNA topoisomerase II is a nuclear enzyme, which regulates the topology of DNA during replication and transcription (reviewed in Ref. 1). This enzyme plays also an important role in chromatin condensation and segregation of chromosomes during mitosis. In mammalian cells, two topoisomerase II isoforms are present, topoisomerase α and β, but despite a high homology of the primary structure (2), both isoforms are differently expressed during the cell cycle and show different subcellular localization. In contrast to isoform β, the levels of which are relatively constant over the cell cycle, the isoform α is expressed in a cell cycle- and proliferation-dependent fashion (3).

Triazoloacridones are a group of antitumor acridines synthesized and developed at the Gdańsk University of Technology. These compounds exhibit very potent cytotoxic activity toward malignant cells in vitro and a strong cytostatic effect against solid tumors in nude mice (4, 5). The mechanism of action of triazoloacridones is currently under active investigation and recent studies have shown that these compounds inhibit the catalytic activity of DNA topoisomerase II in vitro and in tumor cells.4 Other studies showed that biologically active triazoloacridones undergo metabolic activation in tumor cells and covalently bind to DNA (6). Triazoloacridones have been shown to induce cell cycle arrest in G2 phase and cell death by apoptosis in several different human tumor cell lines (7, 8); however, the mechanism of the G2 arrest and apoptosis induced by triazoloacridones has not yet been identified.

Poly(ADP-ribose) polymerase-1 (PARP-1), the first described member of a growing gene family (9), is a sensitive sensor of DNA damage (10, 11). This enzyme, which is strongly activated by DNA strand breaks, has important functions in DNA repair as well as in the maintenance of genome stability. PARP-1 also plays a key role in the initiation of cell death induced by different stimuli (12). The important role of PARP-1 in base excision repair and cell death led to the development of new highly specific strategies to inactivate PARP-1, which made it possible to test the effect of PARP-1 inhibition on the therapeutic effects of anticancer drugs. Inhibition of PARP-1 activity has been shown to potentiate the cytoxic effect of ionizing radiation, DNA methylating agents, oxidative damage, and topoisomerase I inhibitors (13, 14). In contrast to hypersensitivity of cells with decreased PARP-1 activity to drugs targeting topoisomerase I, these cells are resistant to topoisomerase II inhibitors such as etoposide or doxorubicin (11, 13, 15). These observations prompted us to investigate the effect of PARP-1 inactivation on the cytotoxic action of triazoloacridone C-1305, the new topoisomerase II inhibitor with unusual activity toward solid tumors.

In this study, we evaluated the biological effects induced by two close analogs of triazoloacridone with different biological activities in normal mouse fibroblasts and two cell sublines in which PARP-1 gene was specifically disrupted by homologous recombination (16). We showed that cells lacking functional PARP-1 are extremely sensitive to the biologically active triazoloacridone compound, C-1305. The increased sensitivity of PARP-1 knockout (KO) cells to C-1305 is specific for this derivative, because another triazoloacridone, compound C-1533, with low biological activity as well as the classical DNA topoisomerase II inhibitor, amsacrine (m-AMSA), affected mouse cells independent of their intrinsic PARP-1 status.

Cells.

Mice lacking PARP-1 were generated by homologous recombination (16). Immortalized mouse embryo fibroblasts (MEF) were obtained from PARP-1 +/+ (A-19) and from PARP-1 −/− (A-11 and A-12) mice. Cells were grown in DMEM supplemented with 10% FCS at 37°C in an atmosphere of 8% CO2. Human osteosarcoma Saos-2 cells, human leukemia HL-60 cells, and human breast carcinoma MCF-cells were used as positive controls and were maintained as described earlier (17).

Drugs and Chemicals.

Two structurally related triazoloacridone compounds, C-1305 and C-1533, and the reference compound, m-AMSA, were examined in this study. Triazoloacridones were synthesized at the Department of Pharmaceutical Technology and Biochemistry (Gdańsk University of Technology) by Dr. Barbara Horowska. m-AMSA, hydroxyurea, and nocodazole were from Sigma Co. (St. Louis, MO), and etoposide (VP-16) was from Calbiochem-Novabiochem (La Jolla, CA). Stock solutions of triazoloacridones (free bases) were prepared in 0.2% lactic acid, and m-AMSA, VP-16, and nocodazole were dissolved as a stock solution in DMSO. All of the drugs were stored at −20°C until use.

Antibodies.

We used the following antibodies: monoclonal anti-p53 antibodies PAb421 (Ab-1) directed against an epitope within the COOH terminus of the mouse protein, monoclonal anti-PARP-1 antibodies (C-2–10), and monoclonal anti-MCM7 (clone DCS141.2) were from Oncogene Research Products (Cambridge, MA). Polyclonal antibodies against phospho-cdc-2 (Thr161), phospho-cdk-2 (Thr160), and monoclonal anti-phospho-cdc25C phosphatase (Ser216) antibodies (clone 9D1) were from Cell Signaling Technology Inc. (Beverly, MA). Polyclonal antibodies against phospho-cdc-2 (Thr14/Tyr15) were from Sigma Co. Antibodies against cdc-2 and cdc25C phosphatase (clone TC-15) were from Upstate Biotechnology (Lake Placid, NY). Monoclonal anti-cdk-2 antibodies (2B6 + 8D4) were from NeoMarkers Inc. (Fremont, CA). Anti-p16 (clone G175–1239) and anti-p27Kip1 (clone G-173–524) were obtained from BD PharMingen (San Diego, CA). Anti-cyclin E antibodies were from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Monoclonal M30-CytoDEATH antibodies coupled to fluorescein and recognizing the caspase-3 cleaved cytokeratin 18 were from Roche (Vienna, Austria). Monoclonal anti-actin (clone C4) antibodies were from ICN Biochemicals (Aurora, OH). Appropriate secondary antibodies linked to horseradish peroxidase or fluorochromes (Cy-2 and Cy-3) were from Amersham International (Little Chalfont, Buckinghamshire, United Kingdom). The secondary antibodies, especially those used for double immunostaining, were prepared from monospecific antiserum by immunoaffinity chromatography followed by multiple solid-phase absorptions to eliminate possible cross-reactivity.

Drug Cytotoxicity in Vitro.

The effect of studied compounds on cell proliferation was determined using a CellTiter-Glo Luminescent Cell Viability assay (Promega Corporation, Madison, WI), which is based on quantification of the ATP level and was described recently in more detail (18). Luminescent signal was measured in the Wallac 1420 Victor, a multilabel, multitask plate counter. Each point represents the mean ± SD (bars) of seven values from one representative experiment. The IC50 values were assessed as a mean from six independent experiments.

To discriminate between initial cell killing and inhibition of cell proliferation, we measured additionally cell killing at the end of drug exposure. Furthermore, the direct cytotoxicity of studied drugs was assessed by dye exclusion test (17). Cells were continuously treated by studied compounds at final concentrations of 1 μm and 5 μm for 12 h and 24 h. At the end of drug treatment, cellular morphology and the adherence of cells were evaluated under phase contrast microscopy (Eclipse TE300 inverted microscope; Nikon Corporation, Tokyo, Japan), and then cells were collected. The adherent cells were detached by trypsin or by accutase treatment (PAA Laboratories, GmbH, Coelbe, Germany), and all of the cells were washed with PBS. Trypan blue dye or the vital dye 7-amino-actinomycin D (7-AAD; BD Biosciences, San Diego, CA) appropriately diluted with PBS were added, and after 10 min the accumulation of dyes was evaluated under microscopy. Additionally, an accumulation of the 7-AAD fluorescent dye was quantified by flow cytometry using a fluorescence activated cell sorter FACScan cytometer (Becton Dickinson).

Incorporation of [3H]Thymidine.

To assess the effect of drugs on DNA synthesis in cells progressing through the S phase, incorporation of [3H]thymidine was determined as described previously (18).

Analysis of DNA Content in Cells by Flow Cytometry.

The measurement of DNA content in propidium iodide-stained nuclei was performed by flow cytometry as described previously (19). The stained cells were analyzed using a FACScan cytometer (Becton Dickinson).

Comet Assay.

The generation of DNA strand breaks was assessed by the single cell gel electrophoresis assay performed under alkaline conditions (20). The experiments were carried out according to the guidelines published by Tice et al. (20) as described previously (21). Untreated control mouse fibroblasts and cells exposed to 1 μm and 5 μm of each triazoloacridone for 24 h were harvested, washed in PBS, and the viability of cells was determined using the trypan blue exclusion test. Only samples exhibiting a viability of at least 80% were used for additional analysis. Cell suspensions (1 × 105 cells) were mixed with LMP agarose and spread on agarose-precoated slides. After electrophoresis, slides were neutralized and stained with propidium iodide (20 μg/ml). For each experimental point at least three different cultures were analyzed, and 50 cells were evaluated from each culture. Comet tail length (μm) and tail moment were measured under a fluorescence microscope (Nikon model 027012) using an automated image analysis system based on a public domain NIH image program (22).

Detection of Apoptotic Changes in Individual Cells.

Cells grown in 35-mm Petri dishes were treated with triazoloacridone compounds or with VP-16 for the indicated time periods, washed three times in PBS, and immediately fixed in ice-cold methanol for 20 min and washed in PBS, saturated in 5% BSA in PBS for 1 h. Cell samples were subsequently incubated for 1 h with fluorescein-conjugated M30-CytoDEATH monoclonal antibodies (Roche), which recognize only caspase-3 cleaved cytokeratin 18, and extensively washed in PBS (23). Air-dried preparations were covered with the mounting medium containing 4′,6-diamidino-2-phenylindole and inspected under the fluorescence microscope (Eclipse TE300 inverted microscope; Nikon Corporation). The activity of caspases 3/7 in culture medium and in cells was quantified using the APO-ONE homogenous caspase 3/7 Assay (Promega) as described previously (24). This test uses the profluorescent caspase-3/7 substrate rhodamine 110, bis-(N-CBZ-l-aspartyl-l-glutamyl-l-valyl-l-aspartic acid amide; Z-DEVD-R110), which upon cleavage by caspases 3/7 and excitation at 499 nm, becomes intensely fluorescent.

For evaluation of nuclear morphology, cells were fixed in 3.7% paraformaldehyde in PBS and stained with Hoechst 33258 dissolved in PBS at a final concentration of 1.5 μg/ml. Preparations were then washed 4 times, air-dried, and analyzed by fluorescence microscopy.

Electrophoretic Separation of Proteins and Immunoblotting.

Total cellular proteins or proteins of the distinct subcellular fractions dissolved in SDS sample buffer were separated on 10% or 15% SDS slab gels and transferred electrophoretically onto polyvinylidene difluoride membrane (Amersham International). Protein transfer and equal protein loading was confirmed by Ponceau S staining. Blots were incubated with specific primary antibodies at appropriate final dilutions, and the immune complexes were detected using corresponding peroxidase-conjugated secondary antibodies and enhanced chemiluminescent detection reagent ECL+ (Amersham International) as described (17, 18).

Strong Inhibition of Cell Proliferation by C-1305.

First, we evaluated the cell proliferation of different cell lines in the absence of drugs. We observed that within 24 h the number of normal cells as well as of PARP-1-deficient cells increased ∼2-fold. The number of normal cells increased by a factor of 1.9 and PARP-1 KO cells by a factor of 2.2 thereby indicating that under the experimental conditions used the kinetics of cell proliferation for all of the studied cell lines is comparable.

We then determined the inhibitory effect of studied drugs toward nontransformed mouse fibroblasts by cell proliferation test using CellTiter-Glo Assay. The proliferation of mouse fibroblasts was strongly inhibited by C-1305 and by m-AMSA, whereas C-1533, a compound closely related to C-1305, exhibited a markedly weaker effect (Fig. 1,A). Remarkably, comparison of the IC50 values revealed that PARP-1-deficient cells were extremely sensitive to the antiproliferative action of C-1305 (Fig. 1 B), and a 150-fold lower concentration was required to inhibit proliferation of PARP-1 KO cells than that needed to inhibit proliferation of cells with functional PARP-1 gene. Another acridine drug and classical topoisomerase II inhibitor, m-AMSA, was only ∼10-fold more cytotoxic toward mouse cells lacking PARP-1 than toward the normal counterparts. The IC50 values for mutant and normal mouse cells were at 0.1 μm and 1.2 μm, respectively.

To determine the initial cell killing, we additionally measured the number of viable cells after a 24-h drug treatment by CellTiter-Glo Assay. As illustrated in Fig. 1 C, the exposure of normal mouse cells to C-1305 for 24 h at final concentrations of up to 2 μm did not appreciably diminish the number of viable cells. However, the exposure of PARP-1 KO cells to the drug at a final concentration of 2 μm reduced the numbers of living cells by ∼30%.

Low Direct Cytotoxicity of the Triazoloacridone Agent C-1305.

It is widely known that some topoisomerase poisons, such as m-AMSA, are cytotoxic. Therefore, we performed the dye exclusion tests to determine the direct cytotoxicity of the studied agents on mouse fibroblasts. Viable cells with intact membranes exclude vital dyes such as trypan blue or 7-AAD, whereas the membranes of heavily damaged or dead cells become permeable to these dyes, which can be quantified by flow cytometry. In untreated control cultures, the fluorescent dye 7-AAD was accumulated in ∼5% cells (Fig. 2). Surprisingly, the viability of untreated A-11 mutant cells was reduced and, therefore, it was difficult to evaluate the cytotoxic effects of studied drugs. However, A-11 mutant cells tightly adhered to the substratum, and longer trypsin treatment was necessary to detach these cells compared with the other two cell lines, i.e., parental cells and A-12 mutant. Therefore, in the next series of experiments all of the cells were harvested by accutase. As illustrated in Fig. 2 (bottom), the cell membrane remained intact in ∼98% of untreated mutant cells harvested by accutase. Exposure of PARP-1-deficient mouse cells to 1 and 5 μm C-1305 for 12 h and 24 h only slightly increased the number of 7-AAD-positive cells (Fig. 2), and no marked difference was observed to the viability of all cells exposed to C-1533. The viability of normal mouse cells harvested by accutase action was more affected by C-1305.

The reference compound, m-AMSA, at a final concentration of 5 μm showed stronger direct cytotoxic effect and impaired the integrity of plasma membrane of about 15–20% of the parental cells. A-12 mutant cells were more sensitive to m-AMSA, and in some experiments up to 25% of cells accumulated 7-AAD after 24-h treatment with the same drug concentration. Together, these results indicate that the potent antiproliferative action of C-1305 on PARP-1-deficient mouse fibroblasts is not attributable to its high direct cytotoxicity.

Weak Proapoptotic Action of C-1305 on Mouse Cells.

In the next series of experiments, we examined whether the mouse cells exposed to C-1305 die by apoptosis. First, we inspected under phase contrast microscopy the morphology of MEFs treated with 1 μm and 5 μm C-1305 for 24 h and compared with cells exposed to 1 μmm-AMSA for 24 h. MEF cells exposed to both concentrations of C-1305 for 24 h did not detach, and cell morphology remained apparently unchanged. Solely the reduction of the cell density in Petri dishes exposed to the drug became evident due to the strong antiproliferative effect of this compound. In contrast, a lot of cells treated with m-AMSA shrunk and detached from the substratum. Additional microscopy analysis of cell cultures stained with Hoechst fluorochrome revealed an occurrence of a number of mitotic cells as well as occasional apoptotic cells with fragmented and condensed nuclei (<1% of all cells) in untreated controls (Fig. 3). The exposure of MEFs to C-1305 at concentrations 5–25-fold exceeding the IC50 values induced apoptosis only in a low number of cells (Fig. 3). Moreover, the analysis of DNA histograms obtained after treatment with C-1305 for 24 h revealed the lack of a sub-G1 cell population characteristic for apoptotic cells with fractional DNA content. This additionally confirms the microscopic observations about the low incidence of apoptosis and indicates that the strong reduction of the number of proliferating PARP-1 KO cells incubated with C-1305 could not be attributed to the induction of apoptosis. In contrast, analysis of cell cultures exposed to m-AMSA showed a number of cells with morphological features characteristic for apoptosis such as condensed or fragmented chromatin. No substantial difference in the apoptosis rate was detected between mutant and wild-type (wt) cells treated with all of the studied drugs. We then determined the total caspase-3/7 activity in the culture medium as well as in the total cell lysates from cell cultures exposed to studied drugs. At the end of drug treatment, microtiter plates were gently centrifuged to spun down all of the floating cells, and then the culture medium was collected. The basal activity of caspase-3/7 detected in culture medium (data not shown) as well as in cell lysates (Fig. 4) did not increase after 24-h exposure of cells to C-1305. However, the treatment of mouse fibroblasts with 1 μm of m-AMSA for 24 h resulted in an increase of caspase-3/7 activity in cells lacking PARP-1. These quantitative results complete and further substantiate our observations that C-1305 even at a final concentration of 5 μm induced only low incidence of apoptosis in drug-treated cells. These observations were consistent with the determination of caspase-3-mediated cytokeratin 18 cleavage by CytoDEATH antibody (data not shown).

Low DNA Damaging Activity of C-1305.

To assess the DNA damaging activity of C-1305, we performed a single cell electrophoretic analysis (the Comet assay). The levels of DNA damage were quantified by determination of tail length (in μm) and tail moment according to international recommendations (20, 22). Mouse cells were exposed to three studied acridines for 24 h at final concentrations of 1 μm and 5 μm. C-1533 did not produce Comets as compared with untreated controls (Fig. 5) thereby additionally substantiating the low toxicity of this agent. C-1305 at a final concentration of 1 μm resulted in low levels of DNA damage in all of the studied cells, and the ∼2-fold increase of tail length was not statistically significant as compared with untreated controls. After exposure of MEFs to a 5-fold higher concentration of C-1305, the values of tail length remained similar (A-19 and A-11 cells) or slightly increased (A-12 cells), indicating that the higher dose of C-1305 did not markedly increase the number of DNA strand breaks (Fig. 5). Comparison of the levels of DNA damage induced by C-1305 and two reference compounds, m-AMSA and VP-16, showed that C-1305 induced lower levels of DNA strand breaks than the latter agents. Remarkably, the level of DNA strand breaks generated in normal mouse cells by C-1305 and VP-16 was clearly higher than in PARP-1-deficient cells; however, this difference was not statistically significant. In contrast, treatment of mouse cells with m-AMSA generated a strong DNA damage in both types of mouse cells. As shown in Fig. 5, m-AMSA-treated mouse cells generated Comets and tail length, and tail moment after a 24-h incubation at 1 μmm-AMSA increased ∼7-fold as compared with untreated controls. Statistical analysis performed with the one-way ANOVA nonparametric Bonferroni test revealed that this increase was highly statistically significant. These results unequivocally show that m-AMSA, but not C-1305, generated in MEFs high levels of DNA damage, even at low drug concentrations. Taken together, the above results evidence that strong antiproliferative action of C-1305 toward PARP-1-deficient cells could not be attributed to its higher DNA damaging activity in mutant cells.

Inhibition of the Cell Cycle Progression by C-1305.

Cell cycle analysis revealed different effects of studied drugs on the cell cycle progression in mouse fibroblasts exposed to triazoloacridones for 12 and 24 h. As illustrated in Fig. 6, cells treated with C-1305 arrested in G2-M phase of the cell cycle, and an effective accumulation of G2-M cells became evident after continuous exposure of the cells to the drug for 12 h. Comparison of the effect of the C-1305 on the population of cells in S and G2 phases revealed differences in the kinetics of the changes of these two populations between normal and mutant cells. In response to the drug, the G2 to S ratio increased rapidly in both mutant and normal cells and reached similar levels after 12 h. However, after exposure of the cells to the drug for an additional 12 h, this ratio remained elevated in normal mouse cells, whereas in PARP-1 KO cells the ratio additionally increased several fold. This observation indicates that normal mouse cells were transiently arrested in G2-M phase, but a substantial fraction of cells remained in S phase, whereas the mutant cells were blocked permanently in G2-M, and the population of S-phase cells was barely detectable. Considering the fact that the determination of the DNA content by flow cytometric measurement of propidium iodide-stained cells does not allow for discrimination between G2-arrested and mitotic cells, we examined by fluorescence microscopy cells stained by Hoechst fluorochrome. After exposure for 24 h to 1 μm C-1305, the PARP-1 KO cells became very heterogeneous in size (Fig. 3), and the majority of cells showed markedly enlarged nuclei. Only a low number of apoptotic cells was detected, and very few mitoses were observed in mutant cell preparations treated with C-1305. However, with increasing time of drug treatment (up to 24 h) the number of mitotic cells clearly increased in wt mouse cells (Fig. 3), and cells at different stages of mitosis were observed, including inappropriately dividing cells (Fig. 3, arrows). This indicates that after transient accumulation in G2 of drug-treated normal mouse cells at least part of the cell population progressed to mitosis. In striking contrast, C-1305 completely blocked cells lacking PARP-1 at the G2-M checkpoint.

In additional experiments, we assessed the cell cycle distribution of cells treated with C-1305 and postincubated in drug-free medium. Interestingly, normal mouse cells after postincubation for 24 h in drug-free medium entered the cell cycle, and ∼50% of cells were in S phase, whereas the G2-M population was largely reduced. Moreover, a substantial fraction of drug-treated cells become aneuploid and tetraploid. In contrast, cells lacking PARP-1 were still blocked in G2-M (∼80% cells) even after additional incubation without C-1305 for 24 h (Fig. 7). These results are consistent with the data of the cell proliferation assay and thymidine incorporation and explain the mechanism of the strong antiproliferative effect of C-1305 toward PARP-1 null cells. In contrast to the action of C-1305, exposure of cells to the related compound C-1533 resulted in cell cycle arrest in the G1 phase of the cell cycle.

Selective Induction of Wt p53 Protein in Normal MEFs.

It is well established that various types of DNA damage induce wt p53 that can mediate cell cycle arrest. Therefore, we examined the effect of triazoloacridone compounds on the expression of wt p53 in mouse cells. As illustrated in Fig. 8, C-1305 markedly increased the basal level of wt p53 protein in normal MEFs after treatment for 6 h. A similar effect was observed for two other topoisomerase II inhibitors. In contrast, the exposure of PARP-1 KO cells to C-1305, C-1533, and VP-16 did not induce p53 response for either short-time exposure (6 h) or prolonged drug treatment (24 h). Previous reports showed a direct relationship between the expression of p53 and DNA topoisomerase IIα (25), as well as the fact that the intrinsic level of topoisomerase IIα and IIβ may be essential for the action of their inhibitors (26). Therefore, we determined the basal expression levels of both topoisomerase II isoforms in nontreated normal and in PARP-1-deficient cells. Consistently with our earlier observations (15), the inactivation of PARP-1 did not change the basal level of both topoisomerase II isoforms (Fig. 8). Remarkably, the expression of topoisomerase IIα did not change in the course of treatment with the triazoloacridones despite a marked increase of the p53 level in normal mouse cells (data not shown). Moreover, no differences in the expression of topoisomerase IIα were observed between mutant and normal cells despite a strong up-regulation of p53 protein in the latter (data not shown).

C-1305 Mediated Inhibition of cdc-2/Cyclin B in PARP-1 KO Cells.

In the next step we examined the effect of the drugs on the activity of distinct cell cycle regulators. We focused our attention on regulatory complexes critical for the G2-M progression. Substantial differences in the activation of cyclin-dependent kinase (cdk)-2/cyclin A and cdc-2/cyclin B complexes occurred between normal and mutant cells (Fig. 9). The activating phosphorylation of cdk-2 at Thr160 was barely induced by C-1305 at low concentration. At higher concentration of C-1305, a strong phosphorylation of cdk-2 at Thr160 occurred only in normal cells, but not in PARP-1 KO cells (Fig. 9). This coincided with the accumulation of cyclin A in normal cells (data not shown) indicating that higher concentrations of C-1305 induced accumulation of normal mouse cells in the S phase. This observation is consistent with a transient accumulation of drug-treated normal cells in the S-phase cells as determined by flow cytometry (Fig. 6).

The examination of site-specific phosphorylation of cdc-2 kinase reflecting its activation in mitotic cells revealed substantial differences between normal and mutant cells. C-1305 strongly induced the phosphorylation of cdc-2 at Thr 161 in normal but not in mutant cells (Fig. 9). Furthermore, the dephosphorylation of Thr14/Tyr15 of cdc-2, which is necessary for activation of cdc-2/cyclin B complexes, occurred only in C-1305-treated normal cells but not in mutant cells (Fig. 9). This was consistent with the lack of activation of cdc25C phosphatase in mutant cells (data not shown). The increased phosphorylation of Thr 161 accompanied by dephosphorylation of Thr14/Tyr15 was indicative that the population of normal mouse cells, which initially accumulated in G2 after treatment with C-1305, progressed from G2 to mitosis. These results combined with DNA histograms indicate that a fraction of drug-treated normal mouse cells passed the G2 checkpoint and entered mitosis.

The aim of our study was to investigate the antiproliferative action of two related triazoloacridones, C-1305 and C-1533, toward nontransformed mouse fibroblasts differing in PARP-1 status. Previous studies have shown that PARP-1-depleted hamster cells (27) and PARP-1 null mouse fibroblasts (15) exhibited reduced susceptibility to classical topoisomerase II poisons such as etoposide (27) or doxorubicin (15). Therefore, it was reasonable to explore the action of the newly developed topoisomerase II inhibitor C-1305 on cells lacking PARP-1. We examined the cytotoxicity of two triazoloacridone compounds, C-1305 and C-1533, as well as their effect on the cell cycle progression and compared it with the action of m-AMSA, a known classical topoisomerase II inhibitor. Despite a close similarity of the chemical structure between the two triazoloacridone compounds, C-1533 exhibited only weak cytotoxic and biological activity. In contrast, C-1305 showed a high cell growth-inhibitory potential and much more strongly affected proliferation of cells lacking PARP-1 than their normal counterparts. The IC50 concentration of C-1305 determined for PARP-1 KO cells was ∼150-fold lower than that determined for cells with functional PARP-1 enzyme. The antitumor and cytotoxic action of C-1305 on tumor cells was described previously (8, 9). It has also been reported that C-1305 blocked the cell cycle progression and induced apoptosis in tumor cells (8, 9); however, the detailed mechanism of its action in tumor cells remains unknown. C-1305 has been shown recently to inhibit the catalytic activity of topoisomerase II.

One might propose at least three different mechanisms of this unusual sensitivity of PARP-1 mutant cells to C-1305, such as the effect of this drug on cell proliferation, direct cell killing, or induction of cell death by apoptosis. Two of these mechanisms could be excluded, because no activation of caspases 3/7 was detected after a 24-h exposure of mouse fibroblasts to increasing concentrations of C-1305, neither in culture medium nor in whole cell lysates. These results are consistent with the lack of hypodiploid cell population in DNA histograms obtained for cells treated with C-1305 for 24 h. Moreover, cells with apoptotic morphology were only occasionally detected in drug-treated cell populations. Secondly, both studied triazoloacridones exhibited very low direct cytotoxicity as evidenced by accumulation of vital dye 7-AAD. It became evident that trypsin, which was initially used for cell harvesting, affected much more strongly the integrity of plasma membrane than a 24-h exposure of cells to both triazoloacridone compounds. These results suggest that higher sensitivity of mutant cells could be attributable to the inhibition of cell proliferation but not to direct cytotoxicity or rapid induction of apoptosis.

It was also possible that higher sensitivity of PARP-1 mutant cells could result from the changed levels in topoisomerase II content in mutant cells. It has been reported previously that cells expressing lower intracellular levels of topoisomerase II are resistant to topoisomerase II inhibitors such as etoposide (26, 27). We observed no changes in basal levels of both topoisomerase IIα and β in cells lacking PARP-1, and no major changes of topoisomerase IIα expression occurred in the course of treatment with C-1305. This observation is consistent with our previous report (15) and excludes the possibility that elevated levels of topoisomerase II in mutant cells might sensitize them to the action of C-1305.

We then wanted to clarify whether the increased cytotoxicity of C-1305 could be attributed to high levels of DNA damage induced by this compound. To this end, we used the Comet assay, which is a sensitive method, for detection of DNA strand breaks induced by different DNA damaging agents or result from incomplete DNA excision repair. This assay can also be used to detect and quantify DNA damage induced by topoisomerase inhibitors and to monitor DNA fragmentation occurring during apoptosis (28). Our results showed unequivocally that C-1305 generated only low levels of DNA strand breaks in all of the studied cell types, and even slightly higher DNA injury was observed in normal mouse cells, which are the most resistant to the drug. These results evidence that high susceptibility of mutant cells to antiproliferative action of C-1305 was not attributable to increased DNA damage or elevated apoptosis rate in these cells.

In the next step, we examined the effect of C-1305 on the cell cycle progression in normal and PARP-1 mutant cells. We found that this drug inhibited wt as well as mutant cells in G2-M phase in a dose- and time-dependent manner. However, the kinetics of the cell cycle arrest differed between normal and mutant cells. In normal cells the G2 to S ratio increased by only 50% between 12 and 24 h of drug treatment, whereas in mutant cells this value increased 4–8-fold. This indicates that the G2-M arrest was prolonged in PARP-1-deficient cells treated with C-1305. Interestingly, both studied triazoloacridones induced wt p53 in normal mouse fibroblasts. A similar effect was observed in human breast carcinoma MCF-7 cells.5 P53 plays an important role in the regulation of the cell cycle, and the question appeared whether the antiproliferative action of triazoloacridones on the cell cycle, i.e., induction of G2-M arrest, is associated with up-regulation of p53 levels. Two lines of evidence seem to argue against this notion. First, C-1305 induced the G2-M arrest in both normal and in PARP-1 KO cells, despite the lack of detectable levels of p53 protein in two PARP-1 mutants. Secondly, C-1305 induced also the G2-M arrest in human leukemia HL-60 cells5 in which the p53 gene is deleted. Furthermore, the expression of p53 protein was elevated in normal mouse cells by both studied triazoloacridones, but these agents affected different phases of the cell cycle, i.e., C-1305 blocked cells at the G2-M border and C-1533 led to cell growth arrest in G1 phase. These observations suggest that p53 protein is not necessary for the induction of the G2-M arrest by C-1305. However, upon more detailed analysis we cannot exclude that the induced p53 was essential for fine regulation of the activity of distinct complexes mediating the transition from G2 to mitosis, which would lead to re-entry of wt cells to the normal cell cycle. During G2 arrest induced by C-1305, the cdc-2/cyclin B complex was kept inactive only in PARP-1 KO cells by phosphorylation on threonine 14 and tyrosine 15 of cdc-2, which is known to be catalyzed by Myt1 and Wee1 kinases, respectively. Normal mouse cells were capable of the progression into the mitosis, because both of these residues were dephosphorylated by the activated phosphatase cdc25C.

The question remains about the relationship among p53, PARP-1, and topoisomerase II inhibition in cells treated with C-1305. It is well documented that PARP-1 is involved in the regulation of the stability of wt p53 protein (29), and reduced stability of wt p53 in PARP-1-deficient cells is associated with its enhanced nuclear export (30, 31). Our detailed studies have shown recently that the COOH-terminal domain of p53, which harbors the nuclear export signal, is involved in the formation of p53-PARP-1 complex (30, 31). It could be speculated that PARP-1 may mask the nuclear export signal of p53 protein, which could lead to the nuclear retention of p53 as a result of its decreased nuclear export and subsequent degradation. More importantly, the interaction between p53 and PARP-1 seems to be essential for the control of centrosome duplication (32, 33, 34, 35). Results obtained by different groups showed that p53 protein and PARP-1 are localized to centrosomes (32, 33), major microtubule organizing centers that play a pivotal role in the assembly of bipolar mitotic spindles. Other studies reported the important role of PARP-1 and poly(ADP-ribose)ylation in the regulation of centromere function (34, 35). In unstressed cells, PARP-1 has been shown to interact with different centromere proteins, including centrosomal p53, the constitutive centromere proteins CENB-A and CENB-B, and the spindle checkpoint protein BUB3 (35). Upon induction of DNA damage, these proteins become poly(ADP-ribose)ylated (35), and PARP-1 and PARP-1-mediated poly(ADP-ribose)ylation of p53 and other specific centrosomal proteins seems to regulate the proper centrosome duplication. Therefore, loss of PARP-1 activity causes the uncoupling of the centrosome and DNA duplication, resulting in the centrosome hyperamplification. We observed accumulation of cells with aberrant mitoses in C-1305-treated cell populations, and this effect could be associated with dysfunctional centrosomes. Finally, recent studies have shown that PARP is able to reactivate stalled DNA topoisomerase I in covalent complexes, which are stabilized by topoisomerase I inhibitors (36). Given that similar effect is induced for C-1305-specific DNA damage, which is associated with the inhibition of topoisomerase II by this drug, inactivation of PARP-1 would lead to greatly impaired repair of enzyme-associated DNA strand breaks and higher cytotoxic effect of C-1305 in cells with inactivated PARP-1. Additional studies are required to clarify which of these different mechanisms is responsible for the unusual oversensitivity of PARP-1 null cells to triazoloacridone C-1305.

Together, we show here that inactivation of PARP-1 renders mouse cells more sensitive to the action of a new topoisomerase II inhibitor, compound C-1305, compared with cells with functional PARP-1. This is in striking contrast to previous reports in which cells lacking PARP-1 were resistant to classical topoisomerase II inhibitors. Additionally, our data suggest that the PARP-1 status might be essential for the control of the duration of G2 arrest induced by C-1305 and possibly other DNA damaging agents. On the basis of our study, one may propose that the combination of C-1305 with selective inhibitors of PARP-1 could greatly potentiate the cytotoxic and antitumor activity of this compound, and we are currently actively pursuing this issue in our laboratory.

Grant support: Grant no. 10364 from the Jubiläumsfonds of the Oesterreichische Nationalbank and the State Committee for Scientific Research (KBN) Poland, Grant no. 3P05A 12 623.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Józefa Węsierska-Gądek, Cell Cycle Regulation Group, Institute of Cancer Research, Medical University of Vienna, Borschkegasse 8 a, A-1090 Vienna, Austria. Phone: 43-1-4277-65247; Fax: 43-1-4277-65194; E-mail: [email protected]

4

K. Lemke, V. Poinessous, A. K. Larsen, A. Składanowski. The antitumor triazoloacridone C-1305 is a topoisomerase II poison with unusual properties, submitted for publication.

5

Unpublished observations.

We thank Maria Eisenbauer for cultivation of cells, Christian Balcarek for technical assistance, Paul Breit for preparation of photomicrographs, and Drs. Slavica Tudzarova-Trajkovska and Jacek Wojciechowski for preparation of graphics.

1
Berger JM. Type II DNA topoisomerases.
Curr Opin Struct Biol
,
8
:
26
-32,  
1998
.
2
Austin CA, Sng JH, Patel S, Fisher JM. Novel HeLa topoisomerase II is the II beta isoforms: complete coding sequence and homology with the other type topoisomerases.
Biochim Biophys Acta
,
1172
:
283
-91,  
1993
.
3
Larsen AK, Skladanowski A, Bojanowski K. The roles of topoisomerase II during the cell cycle Meijer L Guidet S Vogel S eds. .
Progress in Cell Cycle Research
,
Vol. 2
:
p. 229
-39, Plenum Press New York  
1996
.
4
Cholody WM, Martelli S, Konopa J. 8-Substituted 5-[(aminoalkyl)amino]-6H-v-triazolo[4,5,1-de]acridin-6-ones as potential antineoplastic agents. Synthesis and biological activity.
J Med Chem
,
33
:
2852
-56,  
1990
.
5
Kusnierczyk H, Cholody WM, Paradziej-Lukowicz J, Radzikowski C, Konopa J. Experimental antitumor activity and toxicity of the selected triazolo- and imidazoacridinones.
Arch Immunol Ther Exp (Warsz)
,
42
:
415
-23,  
1994
.
6
Koba M, Konopa J. Interaction of antitumor triazoloacridones with DNA.
Proc Am Assoc Cancer Res
,
43
:
959
2001
.
7
Augustin E, Konopa J. Cell cycle and apoptosis induced by antitumor triazoloacridinones in tumor cells.
Folia Histochem Cytobiol
,
37 (Suppl. 1)
:
41
1999
.
8
Augustin E, Plocka E, Konopa J. Induction of cell death (apoptosis) by antitumor triazoloacridinones in tumor cells.
Drugs Metab Rev
,
32 (Suppl. 1)
:
33
2000
.
9
D’Amours D, Desnoyers S, D’Silvia I, Poirier GG. Poly(ADP-ribosyl)ation reactions in the regulations of nuclear functions.
Biochem J.
,
342
:
249
-68,  
1999
.
10
De Murcia G, Menissier-de Murcia J. Poly(ADP-ribose) polymerase: a molecular nick-sensor.
Trends Biochem Sci
,
19
:
172
-6,  
1994
.
11
Berger NA. Poly(ADP-ribose) in the cellular response to DNA damage.
Radiat Res
,
101
:
4
-15,  
1985
.
12
Herceg Z, Wang Z-Q. Failure of poly(ADP-ribose) polymerase cleavage by caspases leads to induction of necrosis and enhanced apoptosis.
Mol Cell Biol
,
19
:
5124
-33,  
1999
.
13
Bowman KJ, Newell DR, Calvert AH, Curtin NJ. Differential effects of the poly(ADP-ribose) polymerase (PARP) inhibitor NU1025 on topoisomerase I and II inhibitor cytotoxicity in L1210 cells in vitro.
Br J Cancer
,
84
:
106
-12,  
2001
.
14
Griffin RJ, Curtin NJ, Newell DR, Golding BT, Durkacz BW, Calvert AH. The role of inhibitors of poly(ADP-ribose) polymerase as resistance-modifying agents in cancer therapy.
Biochemie
,
77
:
408
-22,  
1995
.
15
Wurzer G, Herceg Z, Wesierska-Gadek J. Increased resistance to anticancer therapy of mouse cells lacking the poly(ADP-ribose) polymerase attributable to up-regulation of the multidrug resistance gene product P-glycoprotein.
Cancer Res
,
60
:
4238
-44,  
2000
.
16
Wang Z-Q, Auer B, Sting lL, et al Mice lacking ADPRT and poly(ADP-ribosyl)ation develop normally but are susceptible to skin disease.
Genes Dev
,
9
:
509
-20,  
1995
.
17
Wojciechowski J, Horky M, Gueorguieva M, Wesierska-Gadek J. Rapid onset of nucleolar disintegration preceding cell cycle arrest in roscovitine-induced apoptosis of human MCF-7 breast cancer cells.
Int J Cancer
,
106
:
486
-95,  
2003
.
18
Lövborg H, Wojciechowski J, Larsson R, Wesierska-Gadek J. Action of a novel anticancer agent, CHS 828, on mouse fibroblasts: increased sensitivity of cells lacking poly(ADP-ribose) polymerase-1.
Cancer Res
,
62
:
4206
-11,  
2002
.
19
Wesierska-Gadek J, Schmid G. Overexpressed poly(ADP-ribose) polymerase delays the release of rat cells from p53 mediated G1 checkpoint.
J Cell Biochem
,
80
:
85
-103,  
2000
.
20
Tice RR, Agurell E, Anderson D, et al Single cell gel/Comet assay: guidelines for in vitro and in vivo genetic toxicology testing.
Environ Mol Mutagen
,
35
:
206
-21,  
2000
.
21
Wesierska-Gadek J, Gueorguieva M, Schloffer D, Uhl M, Wojciechowski J. Non-apoptogenic killing of HeLa cervical carcinoma cells after short exposure to the alkylating agent N-methyl-N′-Nitro-N-Nitrosoguanidine (MNNG).
J Cell Biochem
,
89
:
1222
-34,  
2003
.
22
Helma C, Uhl M. A public domain image analysis program for the single cell gel electrophoresis (Comet) assay.
Mutat Res
,
466
:
9
-15,  
2000
.
23
Wesierska-Gadek J, Schloffer D, Kotala V, Horky M. Escape of p53 protein from E-6 mediated degradation of HeLa cells after cisplatin therapy.
Int J Cancer
,
101
:
128
-36,  
2002
.
24
Tudzarova-Trajkovska S, Wesierska-Gadek J. Strong induction of p73 protein in vivo coincides with the onset of apoptosis in rat liver after treatment with the hepatocarcinogen N-nitrosomorpholine (NNM).
J Cell Biochem
,
90
:
837
-55,  
2003
.
25
Scheltema JM, Romijn JC, Steenbrugge GJ, Beck WT, Schroder FJ, Mickisch GH. Decrease levels of topoisomerase IIα in human renal cell carcinoma lines resistant to etoposide.
J Cancer Res Clin Oncol
,
123
:
546
-54,  
1997
.
26
Zhou Z, Zwelling LA, Kawakami Y, et al Adenovirus-mediated human topoisomerase IIα gene transfer increases the sensitivity of etoposide-resistant human breast cancer cells.
Cancer Res
,
59
:
4618
-24,  
1999
.
27
Chatterjee S, Cheng M-F, Berger RB, Berger SJ, Berger NA. Effect of inhibitors of poly(ADP-ribose) polymerase on the induction of grp78 and subsequent development of resistance to etoposide.
Cancer Res
,
55
:
868
-73,  
1995
.
28
Godard T, Deslandes E, Sichel F, Poul J-M, Gauduchon P. Detection of topoisomerase inhibitor-induced DNA strand breaks and apoptosis by the alkaline comet assay.
Mutat Res
,
520
:
47
-56,  
2002
.
29
Wesierska-Gadek J, Wang Z-Q, Schmid G. Reduced stability of regularly spliced but not alternatively spliced p53 protein in PARP-deficient mouse fibroblasts.
Cancer Res
,
59
:
28
-34,  
1999
.
30
Wesierska-Gadek J, Wojciechowski J, Schmid G. Central and carboxy-terminal regions of human p53 protein are essential for interaction and complex formation with PARP-1.
J Cell Biochem
,
89
:
220
-32,  
2003
.
31
Wesierska-Gadek J, Wojciechowski J, Schmid G. Phosphorylation regulates the interaction and complex formation between wt p53 protein and PARP-1.
J Cell Biochem
,
89
:
1260
-84,  
2003
.
32
Blair-Zajdel ME, Blair GE. The intracellular distribution of the transformation-associated protein p53 in adenovirus-transformed rodent cells.
Oncogene
,
2
:
579
-84,  
1988
.
33
Tarapore P, Horn HF, Tokuyama Y, Fukasawa K. Direct regulation of the centrosome duplication cycle by the p53–p21Waf/Cip1 pathway.
Oncogene
,
20
:
3173
-84,  
2001
.
34
Kanai M, Tong W-M, Sugihara E, Wang Z-Q, Fukasawa K, Miwa M. Involvement of poly(ADP-ribose) polymerase-1 and poly(ADP-ribosyl)ation in regulation of centrosome function.
Mol Cell Biol
,
23
:
2451
-62,  
2003
.
35
Saxena A, Saffery R, Wong LH, Kalitsis P, Choo KHA. Centromere proteins Cenpa, Cenpb, and Bub3 interact with poly(ADP-ribose) polymerase-1 protein and are poly(ADP-ribosylated).
J Biol Chem
,
277
:
26921
-26,  
2002
.
36
Malanga M, Althaus FR. Poly(ADP-ribose) reactivates stalled DNA topoisomerase I and Induces DNA strand break resealing.
J Biol Chem
,
279
:
5244
-8,  
2004
.