点击显示 收起
Group of Molecular and Clinical Cancer Therapeutics, Institut National de la Santee et de la Recherche Meedicale U673, Hpital Saint-Antoine, Paris, France (A.S., M.-G.C., A.E.E., A.K.L.)
Laboratory of Molecular and Cellular Pharmacology, Department of Pharmaceutical Technology and Biochemistry, Gdansk University of Technology, Gdansk, Poland (A.S., M.S.)
Abstract
Microtubule binders are cell cycle-specific agents with preferential cytotoxicity toward mitotic cells. We have characterized vincristine-selected human leukemia cells to establish whether development of vincristine resistance was accompanied by changes in cell cycle kinetics and distribution. Our results indicate that vincristine resistance is accompanied by delayed G2 transit and prolonged early mitosis in both the absence and the presence of the microtubule binder nocodazole. The altered G2/M regulation is accompanied by resistance to short-term (12 h) but not continuous nocodazole exposure in agreement with the transient nature of the observed cell cycle alterations. Western blot analysis indicates that vincristine-selection is accompanied by down-regulation of topoisomerase II without detectable alterations of the other mitotic regulators studied, including Cdk1, p21, 14-3-3, and 14-3-3. This was associated with at least 7-fold less chromosome-associated topoisomerase II, decreased catalytic activity, and cross-resistance to topoisomerase II inhibitors. Characterization of isogenic cell lines expressing different levels of topoisomerase II proteins shows that cellular levels of topoisomerase II, but not the closely related topoisomerase II, directly influence the cell cycle kinetics in G2 and early mitosis as well as the resistance to nocodazole. These results underline the importance of topoisomerase II in late G2 and early M phases and provide evidence for an as-yet-unsuspected interaction between topoisomerase II and microtubule-directed agents.
Microtubule interacting agents (also called tubulin binding agents or spindle poisons), such as vinca alkaloids and taxanes, comprise an important class of antineoplastic agents that act by interfering with tubulin dynamics during mitosis (for review, see Jordan and Wilson, 2004). Despite the efficacy of these agents toward both solid and hematological cancers, drug resistance to tubulin binding agents remains a serious clinical problem.
The different classes of microtubule-interacting agents share an important number of properties. They are natural products or semisynthetic compounds that bind to various sites on the -tubulin subunit of the microtubules. Although tubulin binders act on microtubules throughout the entire cell cycle, cells are particularly vulnerable to this class of agents during the metaphase-to-anaphase transition, where a functional mitotic spindle is absolutely required for the proper positioning, separation, and migration of the mitotic chromosomes. Tubulin binders are able to induce at least three different types of cell cycle checkpoints: the prophase checkpoint (which delays chromosome condensation and metaphase entry), the mitotic spindle assembly/stability checkpoint (which delays the metaphase-to-anaphase transition), and a p53- and Rb-dependent G1 tetraploidy checkpoint (which arrests polyploid cells that have exited mitosis without cytokinesis) (Trielli et al., 1996; Scolnick and Halazonetis, 2000; Meraldi et al., 2004).
One of the most widely reported mechanisms of resistance to tubulin binders is associated with overexpression of different ATP-binding cassette transporters such as P-glycoprotein/Mdr1 (ABCB1) and different MRP (ABCC) family members (for review, see Gottesman et al., 2002). Drug resistance has also been associated with microtubule-associated alterations, such as differential expression and/or mutations of -tubulin isoforms or of microtubule-binding agents such as MAP4 (for reviews, see Drukman and Kavallaris, 2002; Orr et al., 2003). In contrast, the contribution of the different cell cycle checkpoints remains unclear, because loss of these checkpoints has been reported to provide either resistance (Wahl et al., 1996; Kasai et al., 2002) or increased sensitivity to microtubule inhibitors (Trielli et al., 1996; Scolnick and Halazonetis, 2000).
Gene expression analysis of human lung carcinoma cells showed that paclitaxel (Taxol) influenced the expression of several cell cycle regulators. In particular, after 24-h drug treatment, the cyclin-dependent kinase inhibitor p21waf1/cip1/sdi1 (CDKN1A) was up-regulated, as was another Cdk1 modulator, 14-3-3. In contrast, Cdk1 (also known as cdc2 or p34cdc2) and topoisomerase II were down-regulated (Chen et al., 2003). Similar findings were reported when the effect of paclitaxel was studied in vivo on ovarian carcinoma xenografts. It is interesting that topoisomerase II was consistently down-regulated, whereas p21 was up-regulated in the paclitaxel-responsive tumors (Bani et al., 2004). Proteome analysis of vinca alkaloid-sensitive and -resistant leukemia cells identified cell cycle regulators altered in both the response and in the resistance to vinca alkaloids (Verrills et al., 2003). These included 14-3-3, a protein reported to modulate the cellular effects of both Cdc25C phosphatase (Kumagai et al., 1998) and topoisomerase II (Kurz et al., 2000). It is noteworthy that most of these genes are known to play important roles in maintaining G2 arrest in response to under-replicated or damaged DNA (Chan et al., 2000; Larsen et al., 2003). We speculated that the repression of proteins needed for the G2-to-M transition could be a distinct mechanism of maintaining G2 arrest, thereby preventing cells from entering mitosis where microtubule-interacting agents are most cytotoxic.
We report herein that down-regulation of topoisomerase II in both vincristine-resistant human leukemia cells and in untreated isogenic cell lines expressing different levels of topoisomerase II proteins is accompanied by delayed cell cycle progression during G2 and early mitosis and increased resistance to the tubulin-interacting agent nocodazole. These results underline the importance of topoisomerase II in late G2 and early M phases and provide evidence for an as-yet-unsuspected interaction between topoisomerase II and microtubule-directed agents.
Materials and Methods
Chemicals. Vincristine was obtained from Laboratoire Pierre Fabre (Castres, France) and nocodazole was purchased from Sigma (Saint-Quentin Fallavier, France). Symadex (compound C-1311) and amsacrine were synthesized by Dr. Marek Konieczny (Gdansk University of Technology, Gdansk, Poland). Mitoxantrone was a gift from Laboratoire Lederle (Rungis, France) and ICRF-187 (Cardioxane) was purchased from Chiron BV (Amsterdam, The Netherlands). [3H]Thymidine (90 Ci/mmol), [14C]thymidine (52 mCi/mmol), and [14C]leucine (300 mCi/mmol) were obtained from Amersham Biosciences AB (Uppsala, Sweden).
DNA Substrates. Highly catenated kinetoplast DNA (form I) was purified from Trypanosoma leishmania (kindly provided by Prof. Jean-Pierre Dedet, Universitee de Montpellier I, Montpellier, France) and purified by cesium chloride/ethidium bromide density centrifugation.
Antibodies. Monoclonal antibodies directed toward topoisomerase II were obtained from Calbiochem (La Jolla, CA), monoclonal antibodies directed toward topoisomerase II were purchased from BioTrend Chemikalien GmbH (Kln, Germany), whereas monoclonal mouse anti-Cdk1/Cdc2, polyclonal rabbit anti-p21, polyclonal goat anti-14-3-3, polyclonal goat anti-14-3-3, and polyclonal goat anti-actin antibodies were provided by Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Antibodies directed toward phosphotyrosine were obtained from Upstate Biotechnology (Lake Placid, NY). All secondary horseradish-conjugated antibodies were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA) and FITC-conjugated anti-mouse IgG antibodies were provided by Amersham Biosciences (Little Chalfont, Buckinghamshire, UK).
Cell Lines. Parental HL-60 cells and vincristine-selected multidrug-resistant HL-60 cells (HL60/Vinc) (McGrath et al., 1989) were a kind gift from Dr. Melvin Center (Kansas State University, Manhattan, KA). CEM and vincristine-selected CEM/VCR1000 cells (Kimmig et al., 1990) were generously provided by Dr. Johann Hofmann (University of Innsbruck, Innsbruck, Austria). For all experiments, HL-60/Vinc and CEM/VCR cells were reselected with 1 e vincristine for 3 days followed by incubation in drug-free media for 1 to 4 weeks. This dose of vincristine is nontoxic to the resistant cell lines but lethal to the parental ones. All cell lines were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 units/ml penicillin, and 100 e/ml streptomycin at 37°C in a 5% CO2/95% air atmosphere.
The DC-3F fibrosarcoma cell line, its 9-hydroxyellipticine-selected subline DC-3F/9-OH-E, and DC-3F/9-OH-E cells transfected with human topoisomerase II or human topoisomerase II have been extensively characterized (Larsen and Jacquemin-Sablon, 1989; Dereuddre et al., 1997; Khelifa et al., 1999). The cells were maintained in minimal essential medium supplemented with 8% fetal calf serum, 100 units/ml penicillin, and 100 e/ml streptomycin and grown at 37°C in a 5% CO2/95% air atmosphere. All cells were screened routinely for Mycoplasma spp. by the DNA hybridization method (Gen-Probe, Inc., San Diego, CA).
Cell Cycle Analysis. All measurements were made using an epics Profile II flow cytometer (Beckman Coulter, Inc., Fullerton, CA) equipped with an argon laser to give 488 nm light. For cell cycle distribution studies, cells were fixed in 70% ethanol, rehydrated in PBS, and stained in PBS containing propidium iodide (20 e/ml) and ribonuclease A (100 e/ml) for 30 min at room temperature. Data from 104 cells were collected and analyzed by Multicycle software (Phoenix Flow Systems, San Diego, CA). Values are given as the mean of four independent experiments, each done in duplicate.
Mitotic Index. Exponentially growing nontreated cells (2 x 106) were collected, swelled in 75 mM KCl for 10 min at 4°C, and centrifuged for 5 min at 1000 rpm. Cells were fixed three times in 5 ml of freshly prepared Carnoy's fixative [methanol/acetic acid, 3:1 (v/v)] for 15 min at room temperature, spotted onto microscopic slides and stained with 0.1 e/ml DAPI. At least 500 cells in 10 different microscopic fields were counted under 300x magnification using an epifluorescence microscope (BX60; Olympus, Tokyo, Japan). Mitotic cells were photographed with a digital camera (DP50; Olympus). Values are given as the means of two independent experiments.
MPM-2 Staining. For indirect immunofluorescence staining, cells were fixed with 70% ethanol at -20°C overnight, rehydrated in ice-cold PBS for 10 min, and permeabilized in 0.25% Triton X-100/PBS for 5 min on ice. Cells were then washed with PBS and blocked in 1% BSA/PBS for 15 min at room temperature. After washing with PBS, cells were incubated for 1 h at room temperature with monoclonal mouse anti-MPM-2 (DakoCytomation Denmark A/S, Glostrup, Denmark) antibodies diluted at 1:100 in 0.5% BSA/PBS. Samples were washed with 0.2% Tween 20 and 0.5% BSA/PBS and incubated with anti-mouse IgG-FITC (Amersham Biosciences) for 1 h at room temperature, counterstained with 5 e/ml propidium iodide and analyzed by flow cytometry as described previously (Skladanowski and Larsen, 1997).
Quantitation of DNA Fragmentation by the Filter-Binding Assay. Cells were labeled with [14C]thymidine (0.05 e藽i/ml, 48 h) and chased in radiolabel-free medium for 6 h. After drug exposure, approximately 5 x 105 cells suspended in 5 ml of ice-cold PBS were applied onto Matricel filters (Gelman Science, Ann Arbor, MI) and washed twice with 5 ml of PBS. DNA fragmentation was determined by the filter binding assay.
Viability and Growth Inhibition. The viability of nocodazole-treated human leukemia cells was determined by the propidium iodide exclusion assay. In brief, propidium iodide was added directly to the medium (5 e/ml final concentration) of drug-treated cells and the cell suspension was immediately analyzed by flow cytometry. Otherwise, cells were exposed to nocodazole for 12 h and postincubated in drug-free medium for an additional 60 h, and viability was determined by the MTT assay.
Exponentially growing leukemia cells were exposed to different drug concentrations for 1 h followed by postincubation for an additional 72 h in drug-free media. After postincubation, cells were counted with a Coulter counter (Beckman Coulter, Fullerton, CA), and the growth inhibition was determined as described earlier (Skladanowski et al., 1996). Values are given as the mean of three independent experiments, each done in duplicate.
Survival of the DC-3F parental and ellipticine-resistant fibrosarcoma cell lines was determined after 24-h drug exposure followed by a colony formation assay as described previously (Dereuddre et al., 1997). In brief, exponentially growing cells were plated into 60-mm Petri dishes (approximately 250 cells per dish) for 18 to 20 h at 37°C before the drug treatment. After 24-h drug exposure (each dose in triplicate), cells were washed with drug-free medium and postincubated for 5 to 7 days. Colonies were washed with 0.9% saline, fixed with ethanol, and stained with Giemsa stain. Colonies containing at least 50 cells were counted, and growth curves were plotted as survival fractions (number of colonies in drug-treated samples to controls) versus drug dose.
Western Blot Analysis. Exponentially growing cells (5 x 106 cells) were washed twice with PBS, resuspended in 200 e of 0.9% saline with protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 10 e/ml aprotinin, 10 e/ml leupeptin, and 10 e/ml pepstatin A), and mixed with 200 e of double-strength Laemmli buffer (100 mM Tris-HCl, pH 6.8, 200 mM dithiothreitol, 4% SDS, 0.2% bromphenol blue, and 20% glycerol). Cell lysates were immediately denatured for 10 min at 95°C. Cell lysates (25 e) were loaded onto 7.5% polyacrylamide-SDS gels, the proteins were separated by polyacrylamide gel-SDS electrophoresis and transferred to polyvinylidene difluoride filters in 0.192 M glycine, 0.025 M Tris, and 20% methanol for 4 h at 4°C. Membranes were saturated overnight at 4°C in TBS, pH 7.4, containing 5% nonfat milk and probed for 3 h with primary antibodies directed toward the topoisomerase II isoform (1:200 dilution) or the topoisomerase II isoform (1:2000 dilution) at room temperature. After three 15-min washes in TBS/0.05% Tween 20, the blots were probed for 1 h at room temperature with a 1:10,000 dilution of peroxidase-conjugated secondary antibodies. After three washes in TBS/Tween 20, antibody binding was detected by enhanced chemiluminescence (Amersham Life Sciences, Amersham, UK) and quantitated by densitometric analysis using MicroImage Gel Pro software (Media Cybernetics, Silver Spring, MD). All experiments were done with at least three different cellular lysates. Equal loading of proteins was verified by probing dehybridized membranes for 1 h with polyclonal anti-actin antibodies (at 1:1000) followed by donkey anti-goat IgG (Jackson ImmunoResearch Labs) and detection by enhanced chemiluminescence.
Immunolocalization of DNA Topoisomerase II on Mitotic Chromosomes. The fraction of mitotic cells was enriched by nocodazole exposure (25 ng/ml) for 12 h. Approximately 5 x 106 cells were collected, swelled in 75 mM KCl for 10 min at 4°C, and centrifuged for 5 min at 1000 rpm. Cells were fixed three times in 5 ml of freshly prepared Carnoy's fixative [methanol/acetic acid, 3:1 (v/v)] for 15 min at room temperature, spotted onto microscopic slides, and air-dried. Chromosome spreads were then rehydrated in blocking solution (1% BSA in PBS) for 30 min at room temperature, and incubated for 1 h with anti-topoisomerase II-directed antibodies (Ab-1; Calbiochem, San Diego, CA) diluted in 0.5% BSA/PBS at 1:50. After three washes in PBS, chromosomes were incubated with anti-mouse FITC-conjugated antibodies (Amersham) diluted at 1:50. After three additional washes in PBS, chromosome preparations were air-dried and mounted in VectaShield solution (Vector Laboratories, Burlington, CA). All preparations were examined with an Axiovert 100M epifluorescence microscope equipped with appropriate filters using PlanApochromat 100x objective (Zeiss, Jena, Germany) and all data acquisition and image analysis were carried out by a LSM510 laser confocal scanning system (Zeiss).
Decatenation Assay. Nuclei were isolated from exponentially growing cells as described above, and the nuclei were extracted with 0.35 M NaCl for 30 min at 4°C (Lelievre et al., 1995). The nuclei were centrifuged at 11,000 rpm, and the supernatants were collected and frozen immediately after. Nuclear extracts were stored at -80°C and analyzed within 4 days. For decatenation assays, the reaction mixture contained 50 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 100 mM KCl, 0.5 mM dithiothreitol, 10 mM MgCl2, 1 mM ATP, and 200 ng of kinetoplast DNA as described previously (Escargueil et al., 2001). The reaction was initiated by the addition of different concentrations of nuclear extract and was allowed to proceed at 37°C for the indicated times. Reactions were stopped by addition of 1% SDS, 0.5% bromphenol blue, and 30% glycerol. The samples were subjected to electrophoresis in 1.2% agarose gels at 5 V/cm for 6 h in Tris/borate/EDTA buffer, pH 8.3. Liberated minicircles were quantitated by densitometric scanning using a Chromoscan 3 densitometer (Joyce-Loebl and Co., Gateshead, UK). Decatenation assays were carried out with three different nuclear extract preparations.
Statistical Analysis. Comparisons of the cell cycle distribution between parental and drug-resistant cell lines were calculated by the Student's t test using the SigmaPlot program.
Results
Vincristine Resistance Is Accompanied by Altered Cell Cycle Distribution in G2 and M. Cell cycle kinetics and distribution was determined for two different vincristine-resistant human leukemia cell lines, HL-60/Vinc and CEM/VCR1000, as well as for the corresponding parental cell lines. Both resistant cell lines are multidrug-resistant and have been reported to overexpress functional P-glycoprotein (McGrath et al., 1989; Kimmig et al., 1990). Comparison of untreated parental and resistant cell lines revealed no differences with respect to the cell doubling times, which corresponded to 23 h for HL-60 and HL-60/Vinc cells and 25 h for CEM and CEM/VCR1000 cells (data not shown). Likewise, the fraction of cells in the S phase of the cell cycle was comparable between parental and vincristine-resistant cells, with 44.9 ± 3.2% and 44.8 ± 3.8% for HL-60 and HL-60/Vinc, respectively, and 49.5 ± 4.7% and 46.7 ± 5.7% for CEM and CEM/VCR1000, respectively (Fig. 1A).
In contrast, the percentage of cells in the G2/M phase of the cell cycle was enhanced in both vincristine-resistant cell lines with 9.0 ± 1.2% compared with 11.3 ± 1.2% for HL-60 and HL-60/Vinc, respectively, and 8.3 ± 0.7% versus 11.9 ± 1.2% for CEM and CEM/VCR1000, respectively. To establish whether the increased G2/M fraction was due to an increased number of mitotic cells, the mitotic index (which principally represents metaphase cells but does not include prophase cells) was determined (Fig. 1B). We were surprised to find that the mitotic indices of both resistant cell lines were only approximately half that observed for the sensitive parental cells; the mitotic index was 4.5 ± 1.4% for HL-60 cells compared with 2.2 ± 0.6% for HL-60/Vinc cells and 4.0 ± 0.8% for CEM cells versus 2.2 ± 0.9% for CEM/VCR1000 cells. These differences were statistically significant (p < 0.01). Because the cell doubling times were the same for parental and vincristine-selected cells, these results suggested that vincristine resistance was accompanied by an increased fraction of cells in G2 and/or early mitosis.
Parental and Vincristine-Resistant Human Leukemia Cells Differ in the Temporal Response to Nocodazole. We next wished to establish whether the differences in cell cycle distribution observed for untreated parental and vincristine-resistant cells were maintained when the cells were treated with microtubule-interacting agents. The vincristine resistance of both HL-60/Vinc and CEM/VCR cells is associated with overexpression of P-glycoprotein/Mdr1 (ABCB1), a membrane-associated drug efflux pump able to transport most microtubule-interacting agents except for nocodazole. We therefore selected nocodazole for all following studies to exclude any influence of drug uptake and intracellular drug distribution on the results. As illustrated for HL-60 and HL-60/Vinc cells, continuous exposure to nocodazole (25 ng/ml) was accompanied by accumulation of more than 60% of the cell population in the G2/M phase of the cell cycle (Fig. 2A). The fraction of parental HL-60 cells in G2/M was maximal after 12-h drug exposure and then started to decrease. In comparison, the fraction of HL-60/Vinc cells with a G2/M content remained elevated for an additional 8 h and started to decrease only after 20-h nocodazole exposure. To further confirm these findings, the fraction of MPM-2-positive cells was determined (Fig. 2B). The MPM-2 antibody specifically recognizes phosphoepitopes present in mitotic cells, particularly during the early phases of mitosis (Escargueil et al., 2000). In the presence of nocodazole, cellular levels of the MPM-2 epitope diminished after 9-h drug exposure in the parental HL-60 cells but were maintained for at least 16 h in the vincristine-resistant HL-60/Vinc cells. The differences in the fraction of MPM-2 positive cells between the two cell lines are statistically different (p < 0.001) by 12, 16, and 20 h.
Transient cell cycle arrest during G2/M can be followed by either resumption of the cell cycle or by induction of apoptotic cell death. The presence of cells with apoptotic features suggested that the decrease of G2/M cells was accompanied by apoptosis (results not shown). Apoptotic DNA fragmentation was assessed by a DNA filter binding assay (Fig. 2C). The results show that after 12-h nocodazole exposure, an increasing fraction of the total cellular DNA of the parental HL-60 cells is present as small fragments to reach 92.4 ± 4.6% of the total cellular DNA by 24 h. The fragmentation of cellular DNA was much slower for the vincristine-resistant HL-60/Vinc cells with 60.4 ± 5.6% DNA fragmentation by 24 h. For both cell lines, DNA fragmentation was followed by loss of membrane integrity and cell death as shown by a propidium iodide permeability assay (Fig. 2D). It is noteworthy that the occurrence of cell death was systematically delayed in the vincristine-resistant cells compared with the parental cells and was statistically significantly different (p < 0.001) for all time points from 12 h onward.
Prolonged G2/M Transit in Vincristine-Resistant Cells Is Accompanied by Nocodazole Resistance. Transient inhibition of cell cycle progression in G2/M is believed to provide additional time for repair of damaged DNA and other cellular lesions before chromosome separation and cell division. To determine whether the prolonged G2/M transit in vincristine-resistant cells provided these cells with a survival advantage, parental and resistant cells were treated with different concentrations of nocodazole for 12 h, followed by postincubation in drug-free medium for 72 h and determination of cell viability by the MTT assay. The results (Fig. 3, A and B) show marked differences between parental and vincristine-resistant cells for all doses associated with at least 20% cell death. For example, at 5 e/ml nocodazole, the viability of HL-60 cells was 16.4 ± 3.4% compared with 69.4 ± 2.7% for HL-60/Vinc cells, whereas the fraction of viable CEM cells was 9.8 ± 2.9% compared with 55.7 ± 3.2% for CEM/VCR cells. These differences are statistically significantly different (p < 0.001). In marked contrast, parental and vincristine-resistant cells were equally sensitive to continuous nocodazole exposure, with IC50 values of 9.6 ± 1.2 ng/ml for HL-60 compared with 9.4 ± 1.8 ng/ml for HL-60/Vinc and 21.9 ± 2.4 ng/ml for CEM compared with 20.5 ± 2.8 ng/ml for CEM/VCR. Therefore, the prolonged G2 transit provided the vincristine-resistant cell lines with a time-dependent survival advantage in the presence of nocodazole in agreement with the transient nature of the observed cell cycle alterations.
Topoisomerase II Is Down-Regulated in Vincristine-Resistant Cells. To elucidate the molecular basis for the prolonged G2 transit, the expression of important mitotic regulators was determined by Western blot analysis of parental and vincristine-resistant cells (Fig. 4). The proteins examined included p34cdc2, the catalytic subunit of Cdk1, the cyclin kinase inhibitor p21cip1/waf1/sdi1 and DNA topoisomerase II. In addition, we determined the expression of 14-3-3 and 14-3-3, which are able to inhibit the activities of topoisomerase II and the Cdc25C protein phosphatase, respectively. The results show no clear differences for the expression of Cdk1, p21, or the two 14-3-3 proteins (Fig. 4). In addition, the relative distribution between the fast-migrating hypophosphorylated, catalytically active form of Cdk1 and the slow-migrating, catalytically inactive tyrosine-phosphorylated form of Cdk1 kinase was comparable for parental and vincristine-selected cell lines. We were surprised to find that topoisomerase II was down-regulated to approximately 33% of the initial levels in both vincristine-resistant cell lines. This was not the case for the closely related isoform of topoisomerase II that has no known mitotic functions.
The Association of Topoisomerase II with Mitotic Chromosomes Is Decreased in Vincristine-Resistant Cells. We next wanted to establish whether the decreased expression of topoisomerase II in the vincristine-resistant cells influenced the association of the enzyme with mitotic chromosomes. Chromosome spreads of sensitive and vincristine-resistant cells were immunolabeled with a topoisomerase II-directed antibody and examined by confocal fluorescence microscopy (Fig. 5). Topoisomerase II localized to the chromosome arms of parental cells (Fig. 5A) with a clear concentration in centromeric regions apparent in certain images (Fig. 5C). The association of topoisomerase II with mitotic chromosomes of the vincristine-resistant cells was strongly reduced (Fig. 5B), although the overall labeling pattern appeared similar to that of the parental HL-60 cells. Scanning of fluorescence intensities of comparably sized chromosomes (Fig. 5, top, indicated with a line) on nonmodified images showed at least 7-fold difference in the levels of chromosome-associated topoisomerase II between sensitive and resistant cells (Fig. 5, bottom).
The Catalytic Activity of Topoisomerase II Is Reduced in Vincristine-Resistant Cells. The decatenation activities of nuclear extracts from parental and vincristine-resistant cells were determined to establish whether the different expression of topoisomerase II protein was accompanied by differences in the catalytic activity (Fig. 6). The amount of nuclear extracts needed to decatenate 50% of the kinetoplast DNA corresponded to 23 e for the vincristine-resistant cells and 8 e for the parental cells (Fig. 6A). The differences in catalytic activity were further confirmed by kinetic analysis that revealed that extracts from parental cells are 3- to 4-fold more active than extracts from vincristine-resistant cells (Fig. 6B). Thus, the down-regulation of topoisomerase II in vincristine-resistant cells was accompanied by decreased catalytic activity.
Vincristine-Resistant Cells Are Cross-Resistant to Topoisomerase Ii Inhibitors. The cytotoxic activity of different topoisomerase II inhibitors was determined to establish whether the reduced levels of topoisomerase II also influenced the sensitivity to topoisomerase II inhibitors. Because most topoisomerase II inhibitors are recognized by the P-glycoprotein, we selected topoisomerase II inhibitors that are either poor substrates for the P-glycoprotein (amsacrine, mitoxantrone) or, in the case of Symadex (C-1311), not recognized at all (Skladanowski et al., 1996). The results showed that vincristine-resistant cells were 4- to 20-fold cross-resistant to the tested topoisomerase II inhibitors compared with the parental cells (Table 1). It is interesting that the vincristine-resistant cells were also cross-resistant to ICRF-187, a topoisomerase II inhibitor that stabilizes noncovalent, rather than covalent complexes between DNA and topoisomerase II (Roca et al., 1994). Therefore, down-regulation of topoisomerase II in the vincristine-selected cells was accompanied by cross-resistance to different functional classes of topoisomerase II inhibitors.
Exponentially growing cells were exposed to different drug concentrations for 1 h, washed and resuspended in drug -free medium. After 72-h postincubation, cells were counted with a Coulter counter, and the growth inhibition was determined. Values are given as the mean of three independent experiments, each done in duplicate. The growth inhibitory effects of ICRF 187 were determined after 72-h continuous drug exposure. IC50 indicates the concentration inhibiting the growth of sensitive and resistant HL-60 cells by 50% compared to untreated control cells. -Fold resistance is the ratio between the IC50 values obtained for resistant (HL-60/Vinc) and sensitive cells (HL-60)
Topoisomerase II Modulates the Cell Cycle Distribution in G2 and M. Our studies revealed that vincristine-selection of human leukemia cells was accompanied by altered cell cycle progression in G2 and M and by decreased expression of topoisomerase II. To establish whether cellular topoisomerase II levels have a direct influence on the cell cycle distribution, the cell cycle distribution was studied in isogenic cell lines expressing different levels of topoisomerase II protein. Previous results show that ellipticine selection of mammalian DC-3F cells was accompanied by different phenotypic changes including 4- to 5-fold down-regulation of topoisomerase II and complete loss of topoisomerase II (Larsen and Jacquemin-Sablon, 1989; Khelifa et al., 1999). Transfection of the ellipticine-resistant cells with either topoisomerase II or II was accompanied by expression of the corresponding proteins and partial restoration of the sensitivity to topoisomerase II inhibitors (Dereuddre et al., 1997; Khelifa et al., 1999). For comparison, we selected transfected clones that expressed topoisomerase II or II at levels comparable with the parental DC-3F cells (Dereuddre et al., 1997; Khelifa et al., 1999).
The results showed that ellipticine resistance was accompanied by an increased fraction of cells in G2/M (Fig. 7A) and a reduced mitotic index (Fig. 7B). Transfection of the DC-3F/9-OH cells with topoisomerase II was accompanied by a decreased G2/M fraction and an increased mitotic index. In contrast, transfection with topoisomerase II had only minor influence on the cell cycle distribution and the fraction of mitotic cells.
The mitotic index is principally a reflection of the number of metaphase cells and does not include prophase cells according to classic criteria. Prophase and metaphase cells can be distinguished by morphological features such as chromosome orientation, the degree of chromatin condensation, and the presence of a nucleolar remnant as illustrated in Fig. 8A. Determination of prophase and metaphase cells (Fig. 8B) showed a very modest fraction of prophase cells (0.32 ± 0.08%) in the parental cells compared with the fraction of metaphase cells (5.03 ± 0.7%). In contrast, the number of prophase cells (1.25 ± 0.3%) was much higher in the ellipticine-resistant cells, both in relative and absolute terms, because the fraction of metaphase cells was only 1.67 ± 0.26%. Transfection of the ellipticine-resistant cells with topoisomerase II had marginal influence on the proportion of prophase and metaphase cells and corresponded to 1.32 ± 0.25% and 1.94 ± 0.3%, respectively. In marked contrast, transfection with topoisomerase II completely altered the proportion of prophase and metaphase cells to 0.26 ± 0.07% and 4.85 ± 0.6%, respectively. These values were comparable with those observed for the sensitive parental cells. Taken together, these experiments strongly suggest that the expression of topoisomerase II, but not topoisomerase II, directly influences the cell cycle progression in G2 and early mitosis.
To establish whether topoisomerase II protein levels directly influence the cellular sensitivity to microtubule interacting agents, the isogenic ellipticine-resistant cells were exposed to nocodazole for 24 h followed by postincubation in drug-free media and the cytotoxicity was determined by colony formation (Fig. 9). The results show that the expression of topoisomerase II plays an important role in the sensitivity to nocodazole with IC50 values corresponding to 0.09 e/ml for the topoisomerase II transfectants compared with 0.30 e/ml for the ellipticine-resistant parental cells. These changes are statistically significantly different (p < 0.001). In contrast, no significant differences were observed between the topoisomerase II transfectants and the ellipticine-resistant parental cells.
Discussion
Cell cycle arrest in the G2 phase of the cell cycle can be triggered by a wide range of different stress stimuli known to influence DNA and nuclear functions. More recently, it has been reported that microtubule-interacting agents can also influence the expression of several important regulators of the G2-to-M transition, such as Cdk1 kinase and topoisomerase II, which are down-regulated, and the cyclin-dependent kinase inhibitor p21 and 14-3-3 proteins that are upregulated (Chen et al., 2003; Verrills et al., 2003; Bani et al., 2004). These findings suggest that tubulin interacting-agents might be able to influence the G2-to-M progression, thereby retaining cells in a part of the cell cycle where they are less sensitive to the cytotoxic effects of such agents.
To explore this hypothesis, we characterized the cell cycle distribution in two independently selected human leukemia cell lines. The results showed no clear differences between vincristine-selected and parental cells with respect to the proportion of S phase cells or generation doubling times. In contrast, the G2/M fraction was significantly enhanced in both resistant cell lines, whereas the fraction of metaphase cells was reduced. Therefore, vincristine selection resulted in enrichment of cells present in G2 and/or early mitosis.
Western blot analysis was performed to compare cellular levels of major G2/M regulators in parental and vincristine-resistant cells. Because the same proteins often are implicated in both the response and the resistance to cytotoxic agents, we focused on genes previously shown to be influenced by exposure to paclitaxel, epothilone B, or vincristine (Chen et al., 2003; Verrills et al., 2003; Bani et al., 2004). We were surprised to find no detectable alterations for Cdk1, p21, 14-3-3, or 14-3-3. In clear contrast, topoisomerase II was down-regulated to approximately one-third the initial levels in both vincristine-resistant cell lines, accompanied by a 3- to 4-fold decrease in the catalytic activity and cross-resistance to topoisomerase II inhibitors.
Topoisomerase II is a dynamic molecule that can be either associated with DNA or present in the nucleoplasm (Tavormina et al., 2002). We therefore determined the association of topoisomerase II with mitotic chromosomes in parental and vincristine-resistant cells by immunoblotting followed by image analysis. The results showed that the association of topoisomerase II with mitotic chromosomes was reduced at least 7-fold in the vincristine-resistant cells compared with the sensitive parental cells. These findings suggested that the altered cell cycle progression of vincristine-selected cells might be directly related, at least in part, to the down-regulation of topoisomerase II.
Topoisomerase II is already known to play several essential roles during late G2 and early mitosis (for reviews, see Larsen et al., 2003; Swedlow and Hirano, 2003). First, topoisomerase II is required for premitotic chromosome individualization, the process by which different chromosomes become untangled and converted into individual units (Gimenez-Abian et al., 2000). The protein also plays a direct role in early chromosome condensation by an as-yet poorly understood mechanism that seems to rely, at least in part, on protein-protein and DNA-protein interactions (Adachi et al., 1991). In addition, topoisomerase II is needed for recruitment of Cdk1 to the chromatin of mitotic chromosomes (Escargueil et al., 2001) and thus for mitotic phosphorylation of other chromatin-associated proteins such as the condensins. Finally, topoisomerase II is required for the separation of intertwined sister chromatids in prometaphase (Gimenez-Abian et al., 1995; Losada et al., 2002). A potential influence of topoisomerase II protein levels on the fraction of G2 and prophase cells would therefore be consistent with the known biological functions of topoisomerase II during G2 and early mitosis.
To further explore this hypothesis, we selected a panel of isogenic cell lines in which the topoisomerase II levels have been extensively characterized (Larsen and Jacquemin-Sablon, 1989; Dereuddre et al., 1997; Khelifa et al., 1999). Previous results showed that prolonged selection of parental DC-3F cells with the topoisomerase II inhibitor 9-hydroxyellipticine is accompanied by 4- to 5-fold decreased expression of the topoisomerase II protein and total loss of topoisomerase II. We now report that the decreased topoisomerase II levels were accompanied by a significantly increased G2/M fraction. To establish the individual roles of the two topoisomerase II isoforms, 9-hydroxyellipticine-resistant cells were transfected with either topoisomerase II or II and two transfected clones selected where the catalytic activity had been restored to the same levels as in the sensitive parental cell line. It is interesting that transfection of the ellipticine-resistant cells with topoisomerase II resulted in a clear reduction of G2 cells, whereas transfection with topoisomerase II had no detectable influence on the G2/M distribution. Morphological studies showed that the increased G2/M fraction in cells with diminished topoisomerase II is a mixture of cells with classic G2 morphology and of prophase cells. In contrast, the fraction of metaphase cells was diminished.
To further explore the influence of topoisomerase II on the sensitivity to microtubule-interacting agents, the different ellipticine-resistant cell lines were exposed to nocodazole for 24 h followed by postincubation in drug-free media. The results reveal that topoisomerase II, but not topoisomerase II, directly influences the cellular sensitivity to microtubule-interacting agents.
Recent results have characterized the influence of small interfering RNA coding for topoisomerase II or II on the cell cycle progression in untreated HeLa cells (Sakaguchi and Kikuchi, 2004). The results show that decreased topoisomerase II levels are associated with an increased G2 fraction as well as an increased proportion of cells in the early phases of mitosis. In contrast, knockdown of topoisomerase II had no apparent effect on the cell cycle. Thus, topoisomerase II was shown to directly influence G2/early mitosis in two different genetic models (Sakaguchi and Kikuchi, 2004; this study). Furthermore, a similar phenomenon was observed in the current study for two clinically relevant models of acquired vincristine resistance. These results have important pharmacological implications because the topoisomerase II-mediated cell cycle modifications were accompanied by increased resistance to the microtubule-interacting agent nocodazole.
In the present studies, we consistently worked with resistant cell lines that had been reselected with vincristine at least 8 days before use. Therefore, the down-regulation of topoisomerase II would qualify as an adaptation process rather than a checkpoint mechanism. Although long-lasting, this is a dynamic response, because no differences in topoisomerase II levels and activities were observed when more than 4 weeks had passed because the selection process ended. Another major difference between our findings and classic G2 and prophase checkpoints is that the modified cell cycle progression in G2 and early M phases can be observed even for untreated cells.
It is interesting that down-regulation of topoisomerase II shows several similarities to the effects mediated by the prophase checkpoint protein Chfr. The Chfr ubiquitin ligase is activated upon mitotic stress and is able to modify key mitotic regulators, such as Plk1 and Aurora A, and to delay the activation of Cdk1/Cyclin B (Kang et al., 2002; Summers et al., 2005). Activation of Chfr is accompanied by prolonged chromosome condensation and a delay of approximately 6 h in metaphase entry (Scolnick and Halazonetis, 2000). It is noteworthy that transfection of Chfr into Chfr-deficient cells was accompanied by prolonged prophase, delayed metaphase entry, and resistance to both nocodazole and paclitaxel compared with the Chfr-deficient parental cells (Scolnick and Halazonetis, 2000). These results suggest that prophase prolongation and delayed metaphase entry are involved in both the checkpoint response and in the adaptation to microtubule poisons.
The down-regulation of topoisomerase II could be due to down-regulation of its promoter. The topoisomerase II promoter is regulated in a cell cycle-dependent manner, with maximal activity at the G2/M interphase (Woessner et al., 1991), and contains five inverted CCAAT boxes (ICBs), two GC boxes, and an ATF site (Hochhauser et al., 1992). It has previously been reported that vinblastine is able to down-regulate the topoisomerase II promoter, at least in p53-proficient cells (Joshi et al., 2003). This is likely to also to be the case for p53-deficient cells, because microtubule inhibitors may lead to DNA damage as a result of abnormal mitosis, producing chromosomal breakage and other chromosomal damage (Chen et al., 2003).
In conclusion, we report here that down-regulation of topoisomerase II in vincristine-selected human leukemia cells and in isogenic cell lines expressing different levels of topoisomerase II proteins is accompanied by delayed cell cycle progression during G2 and early mitosis and increased resistance to the tubulin-interacting agent nocodazole. Topoisomerase II is already known to be down-regulated after DNA damage and heat shock and is monitored by the G2 catenation checkpoint. The observation that microtubule-interacting agents may also influence topoisomerase II expression makes it one of the most versatile proteins implicated in the control of the G2-to-M transition.
Acknowledgements
We are grateful to Arlette Verwish (Service Cytomeetrie, Villejuif, France) for help with flow cytometry analysis and Raymond Hellio (Institut Pasteur, Paris, France) for expert assistance with confocal microscopy. We thank Prof. Jean-Pierre Dedet for kindly providing us with kinetoplast DNA from Trypanosoma leishmania.
A.S. and M.G.C. contributed equally to this work.
References
Adachi Y, Luke M, and Laemmli UK (1991) Chromosome assembly in vitro: topoisomerase II is required for condensation. Cell 64: 137-148.
Bani MR, Nicoletti MI, Alkharouf NW, Ghilardi C, Petersen D, Erba E, Sausville EA, Liu ET, and Giavazzi R (2004) Gene expression correlating with response to paclitaxel in ovarian carcinoma xenografts. Mol Cancer Ther 3: 111-121.
Chan TA, Hwang PM, Hermeking H, Kinzler KW, and Vogelstein B (2000) Cooperative effects of genes controlling the G2/M checkpoint. Genes Dev 14: 1584-1588.
Chen JG, Yang CP, Cammer M, and Horwitz SB (2003) Gene expression and mitotic exit induced by microtubule-stabilizing drugs. Cancer Res 63: 7891-7899.
Dereuddre S, Delaporte C, and Jacquemin-Sablon A (1997) Role of topoisomerase II beta in the resistance of 9-OH-ellipticine-resistant Chinese hamster fibroblasts to topoisomerase II inhibitors. Cancer Res 57: 4301-4308.
Drukman S and Kavallaris M (2002) Microtubule alterations and resistance to tubulin-binding agents. Int J Oncol 21: 621-628.
Escargueil AE, Plisov SY, Skladanowski A, Borgne A, Meijer L, Gorbsky GJ, and Larsen AK (2001) Recruitment of cdc2 kinase by DNA topoisomerase II is coupled to chromatin remodeling. FASEB J 15: 2288-2290.
Escargueil AE, Plisov SY, Filhol O, Cochet C, and Larsen AK (2000) Mitotic phosphorylation of DNA topoisomerase II alpha by protein kinase CK2 creates the MPM-2 phosphoepitope on Ser-1469. J Biol Chem 275: 34710-34718.
Gimenez-Abian JF, Clarke DJ, Devlin J, Gimenez-Abian MI, De la Torre C, Johnson RT, Mullinger AM, and Downes CS (2000) Premitotic chromosome individualization in mammalian cells depends on topoisomerase II activity. Chromosoma 109: 235-244.
Gimenez-Abian JF, Clarke DJ, Mullinger AM, Downes CS, and Johnson RT (1995) A postprophase topoisomerase II-dependent chromatid core separation step in the formation of metaphase chromosomes. J Cell Biol 131: 7-17.
Gottesman MM, Fojo T, and Bates SE (2002) Multidrug resistance in cancer: role of ATP-dependent transporters. Nat Rev Cancer 2: 48-58.
Hochhauser D, Stanway CA, Harris AL, and Hickson ID (1992) Cloning and characterization of the 5'-flanking region of the human topoisomerase II alpha gene. J Biol Chem 267: 18961-18965.
Jordan MA and Wilson L (2004) Microtubules as a target for anticancer drugs. Nat Rev Cancer 4: 253-265.
Joshi AA, Wu Z, Reed RF, and Suttle DP (2003) Nuclear factor-Y binding to the topoisomerase II promoter is inhibited by both the p53 tumor suppressor and anticancer drugs. Mol Pharmacol 63: 359-367.
Kang D, Chen J, Wong J, and Fang G (2002) The checkpoint protein Chfr is a ligase that ubiquitinates Plk1 and inhibits Cdc2 at the G2 to M transition. J Cell Biol 156: 249-259.
Kasai T, Iwanaga Y, Iha H, and Jeang KT (2002) Prevalent loss of mitotic spindle checkpoint in adult T-cell leukemia confers resistance to microtubule inhibitors. J Biol Chem 277: 5187-5193.
Khelifa T, Rene B, Le Mee S, Lambert B, Saucier JM, Markovits J, Jacquemin-Sablon H, and Jacquemin-Sablon A (1999) Transfection of 9-hydroxyellipticine-resistant Chinese hamster fibroblasts with human topoisomerase II cDNA: selective restoration of the sensitivity to DNA religation inhibitors. Cancer Res 59: 4927-4936.
Kimmig A, Gekeler V, Neumann M, Frese G, Handgretinger R, Kardos G, Diddens H, and Niethammer D (1990) Susceptibility of multidrug-resistant human leukemia cell lines to human interleukin 2-activated killer cells. Cancer Res 50: 6793-6799.
Kumagai A, Yakowec PS, and Dunphy WG (1998) 14-3-3 proteins act as negative regulators of the mitotic inducer Cdc25 in Xenopus egg extracts. Mol Biol Cell 9: 345-354.
Kurz EU, Leader KB, Kroll DJ, Clark M, and Gieseler F (2000) Modulation of human DNA topoisomerase II function by interaction with 14-3-3. J Biol Chem 275: 13948-13954.
Larsen AK, Escargueil AE, and Skladanowski A (2003) From DNA damage to G2 arrest: the many roles of topoisomerase II. Prog Cell Cycle Res 5: 295-300.
Larsen AK and Jacquemin-Sablon A (1989) Multiple resistance mechanisms in Chinese hamster cells resistant to 9-hydroxyellipticine. Cancer Res 49: 7115-7119.
Leblond CP and El-Alfy M (1998) The eleven stages of the cell cycle, with emphasis on the changes in chromosomes and nucleoli during interphase and mitosis. Anat Rec 252: 426-443.
Lelievre S, Benchokroun Y, and Larsen AK (1995) Altered topoisomerase I and II activities in suramin-resistant lung fibrosarcoma cells. Mol Pharmacol 47: 898-906.
Losada A, Hirano M, and Hirano T (2002) Cohesin release is required for sister chromatid resolution, but not for condensin-mediated compaction, at the onset of mitosis. Genes Dev 16: 3004-3016.
McGrath T, Latoud C, Arnold ST, Safa AR, Felsted RL, and Center MS (1989) Mechanisms of multidrug resistance in HL60 cells. Analysis of resistance associated membrane proteins and levels of mdr gene expression. Biochem Pharmacol 38: 3611-3619.
Meraldi P, Draviam VM, and Sorger PK (2004) Timing and checkpoints in the regulation of mitotic progression. Dev Cell 7: 45-60.
Orr GA, Verdier-Pinard P, McDaid H, and Horwitz SB (2003) Mechanisms of Taxol resistance related to microtubules. Oncogene 22: 7280-7295.
Roca J, Ishida R, Berger JM, Andoh T, and Wang JC (1994) Antitumor bisdiox-opiperazines inhibit yeast DNA topoisomerase II by trapping the enzyme in the form of a closed protein clamp. Proc Natl Acad Sci USA 91: 1781-1785.
Sakaguchi A and Kikuchi A (2004) Functional compatibility between isoform alpha and beta of type II DNA topoisomerase. J Cell Sci 117: 1047-1054.
Scolnick DM and Halazonetis TD (2000) Chfr defines a mitotic stress checkpoint that delays entry into metaphase. Nature (Lond) 406: 430-435.
Skladanowski A and Larsen AK (1997) Expression of wild-type p53 increases etoposide cytotoxicity in M1 myeloid leukemia cells by facilitated G2 to M transition: implications for gene therapy. Cancer Res 57: 818-823.
Skladanowski A, Plisov SY, Konopa J, and Larsen AK (1996) Inhibition of DNA topoisomerase II by imidazoacridinones, new antineoplastic agents with strong activity against solid tumors. Mol Pharmacol 49: 772-780.
Summers MK, Bothos J, and Halazonetis TD (2005) The CHFR mitotic checkpoint protein delays cell cycle progression by excluding cyclin B1 from the nucleus. Oncogene 24: 2589-2598.
Swedlow JR and Hirano T (2003) The making of the mitotic chromosome: modern insights into classical questions. Mol Cell 11: 557-569.
Tavormina PA, Cme M-G, Hudson JR, Mo YY, Beck WT, and Gorbsky GJ (2002) Rapid exchange of mammalian topoisomerase II alpha at kinetochores and chromosome arms in mitosis. J Cell Biol 158: 23-29.
Trielli MO, Andreassen PR, Lacroix FB, and Margolis RL (1996) Differential Taxol-dependent arrest of transformed and nontransformed cells in the G1 phase of the cell cycle and specific-related mortality of transformed cells. J Cell Biol 135: 689-700.
Verrills NM, Walsh BJ, Cobon GS, Hains PG, and Kavallaris M (2003) Proteome analysis of vinca alkaloid response and resistance in acute lymphoblastic leukemia reveals novel cytoskeletal alterations. J Biol Chem 278: 45082-45093.
Wahl AF, Donaldson KL, Fairchild C, Lee FY, Foster SA, Demers GW, and Galloway DA (1996) Loss of normal p53 function confers sensitization to Taxol by increasing G2/M arrest and apoptosis. Nat Med 2: 72-79.
Woessner RD, Mattern MR, Mirabelli CK, Johnson RK, and Drake FH (1991) Proliferation- and cell cycle-dependent differences in expression of the 170 kilodalton and 180 kilodalton forms of topoisomerase II in NIH-3T3 cells. Cell Growth Differ 2: 209-214.