Potent inhibition of tumor survival in vivo by β-lapachone plus taxol: Combining drugs imposes different artificial checkpoints

Abstract

Ablation of tumor colonies was seen in a wide spectrum of human carcinoma cells in culture after treatment with the combination of β-lapachone and taxol, two low molecular mass compounds. They synergistically induced death of cultured ovarian, breast, prostate, melanoma, lung, colon, and pancreatic cancer cells. This synergism is schedule dependent; namely, taxol must be added either simultaneously or after β-lapachone. This combination therapy has unusually potent antitumor activity against human ovarian and prostate tumor prexenografted in mice. There is little host toxicity. Cells can commit to apoptosis at cell-cycle checkpoints, a mechanism that eliminates defective cells to ensure the integrity of the genome. We hypothesize that when cells are treated simultaneously with drugs activating more than one different cell-cycle checkpoint, the production of conflicting regulatory signaling molecules induces apoptosis in cancer cells. β-Lapachone causes cell-cycle delays in late G₁ and S phase, and taxol arrests cells at G₂/M. Cells treated with both drugs were delayed at multiple checkpoints before committing to apoptosis. Our findings suggest an avenue for developing anticancer therapy by exploiting apoptosis-prone “collisions” at cell-cycle checkpoints.

A variety of cancer chemotherapeutic drugs has been discovered, but treatment of most human solid tumors remains largely palliative. One way to improve the efficacy of anticancer therapy is to develop optimal combination regimens of chemotherapeutic drugs. In this study, we demonstrate a treatment synergism by combining β-lapachone and taxol. This combination resulted in simultaneous cell-cycle checkpoint delays at the G₁/S and G₂/M transitions, setting up for apoptosis-prone “collisions” (1). We observed very extensive death of various human tumor cells in culture. Importantly, growth of human ovarian tumor cells implanted into immunosuppressed mice was decreased dramatically, without signs of deleterious effects to the mice. Our preliminary experiments also produced dramatic antitumor activity against xenografted human prostate cancer. This study suggests a combination chemotherapy of very great effectiveness as well as a potential avenue for developing anticancer therapies.

Cell-cycle checkpoints can be targeted for cancer therapy either by activating checkpoint-mediated apoptosis pathways or by exploiting chemical sensitivity because of loss of checkpoint function (2, 3). Cells arrest in G₂/M after treatment with DNA-damaging agents, such as chemotherapeutic agents and x-rays (2, 4). Taxol is an anticancer agent with a wide spectrum of activity that blocks cell proliferation in mitosis presumably by interfering with microtubule function (5). Taxol induces a p53-independent G₂/M arrest (prophase) that triggers the rapid onset of apoptosis (5–8).

Drugs that induce G₁/S checkpoint delays are rarely used. β-Lapachone (3,4-dihydro-2,2-dimethyl-2H-naphthol[1,2-b] pyran-5,6-dione) is a plant product (9, 10). It induces a cell-cycle delay in G₁ and/or S phase before inducing either apoptotic or necrotic cell death in a variety of human carcinoma cells, including ovary, colon, lung, prostate, and breast (11–12), suggesting a wide spectrum of anticancer activity of β-lapachone in vitro (10). Both apoptotic and necrotic cell death induced by β-lapachone are preceded by a rapid release of cytochrome c, followed by activation of caspase 3 in apoptotic cell death but not in necrotic cell death (11, 13).

Materials and Methods

Chemicals.

β-Lapachone was kindly provided by A. Matter (CIBA–Geigy) or semisynthesized by us. It was dissolved at a concentration of 20 mM in DMSO, divided into aliquots, and stored at −20°C for cell-culture use. For animal experiments, β-lapachone was formulated in Lipiodol, kindly provided by H. Noguchi (Sumitomo Pharmaceuticals, Osaka), and freshly prepared for each injection. Taxol was purchased from Sigma and was dissolved in DMSO for use with cell cultures. For animal experiments, taxol that formulated in Cremophor was purchased from Bristol-Myers Squibb and diluted with Lipiodol before injection.

Cell Cultures.

All cell lines used in this study were obtained from the American Type Culture Collection unless specified otherwise. Cells were maintained at 37°C in 5% CO₂ in complete humidity. Human breast cancer cell lines MCF-7, 21 MT, 21 PT, and 21 NT (kindly provided by R. Sager, Dana–Farber Cancer Institute) were cultured in MEM-α (Life Technologies, Grand Island, NY), supplemented with 10% (vol/vol) FCS, 2 mM l-glutamine, and 1 mg/ml insulin. Human ovary carcinoma cell lines AD 2780s and AD2780DDP, a generous gift from K. J. Scanlon (City of Hope Medical Center, Duarte, CA); human colon adenocarcinoma cell lines SW1116, HT-29, and DLD; human lung carcinoma cell line G480; human melanoma cell line Skmel-28, kindly provided by G. Dranoff, (Dana–Farber Cancer Institute); and human prostate tumor cell lines PC-3, DU145, and LNCaP were cultured in DMEM (Life Technologies) supplemented with 10% (vol/vol) FCS and 2 mM l-glutamine. Human pancreatic cancer cell line ASPC-1 was cultured in RPMI medium 1640 supplemented with 20% (vol/vol) FCS.

Colony Formation Assay.

Exponentially growing cells were seeded at 1,000 cells per well in six-well plates and allowed to attach for 48 h. Drugs were added directly to dishes in less than 5 μl of concentrated solution (corresponding to a final DMSO concentration of less than 0.1%). Control plates received the same volume of DMSO alone. After 1–4 h, cells were rinsed, and fresh medium was added. Cultures were observed daily for 10–20 days and then were fixed and stained with modified Wright-Giemsa stain (Sigma). Colonies of greater than 30 cells were scored as survivors.

Cell Death Assay.

Cell death was determined by the MTT (Thiazolyl blue) assay or by trypan blue exclusion as indicated. Briefly, cells were plated in a 96-well plate at 10,000 cells per well, cultured for 48 h in complete growth medium, then treated with β-lapachone for 4 h, and cultured with drug-free medium for 24 h. MTT solution was added to the culture medium, and after 2 h, optical density was read with an ELISA reader. For the trypan blue exclusion assay, cells were cultured and treated in the same way. They were harvested, and trypan blue dye solution was added to the cell suspension. Total cell counts and viable cell numbers were determined with a hemocytometer.

Apoptosis Assays.

Apoptosis was determined by three independent assays. One determined the sub-G₁ fraction of propidium iodide-stained nuclei as described (11, 14, 15). The annexin assay measured the membrane changes determined by the externalization of phosphatidylserine (16). The third assay, analysis of DNA laddering, was carried out as described (11).

Human Ovarian Cancer Xenograft Model (17).

Athymic female nude mice (Ncr) were irradiated (300 R), and 24 h later, 10 × 10⁶ human ovarian cancer cells (36M2) were inoculated by i.p. injection. In general, metastatic foci formed 1 week after inoculation, and tumor nodules on the peritoneum and malignant ascites developed in 4–5 weeks. In three independent therapeutic experiments, with a total of 18 mice per group, administration of drugs was initiated 10 days after tumor inoculation. The control group was treated with vehicle alone. In each typical treatment cycle, the β-lapachone alone group was treated with 25–50 mg/kg i.p., and the taxol alone group was treated with 1 mg/kg i.p, followed 24 h later by i.p. injection of vehicle. In the combination group, mice were treated with β-lapachone at 25–50 mg/kg, followed 24 h later by taxol at 1 mg/kg. There was a 1-day break between each cycle. Mice were treated for a total of 10 cycles and were killed to assess the antitumor activity 2 weeks (on day 50) after discontinuation of drug treatment. The host toxicity was evaluated by general appearance and body weight.

Cell-Cycle Analysis and cdc Kinase Assay.

Cell-cycle analysis was carried out by flow cytometry after staining cells with propidium iodide; nuclear extract preparation and kinase assays were performed as described (14). For each assay, 50 μg of cell extract was immunoprecipitated with 1 μg of purified polyclonal antibody against human cyclin E, A, and B1. Histone H1 was used as the substrate for kinase assays.

Western Blot Analysis.

Nuclear extract was prepared and resolved by SDS/PAGE as described (15). The enhanced chemiluminescence assay system (Amersham Pharmacia) was used to determine the p21 (waf1) level.

Northern Blot Analysis.

Total RNA was prepared and fractionated by agarose gel electrophoresis. The mRNA of p21 was determined by Northern blot analysis with a randomly labeled cDNA as a probe, as described (18).

Results

Synergism of the Drug Combination.

Colony formation of DU145 cells in the control dish (Fig. 1, well 1) was abolished when both taxol and β-lapachone were applied. It was decreased only partly when taxol alone (well 2) or β-lapachone alone (well 3) were applied. To determine whether the order of drug addition affects this observed powerful synergism of cell killing, we varied the treatment schedule. A similar synergism was observed when cells were treated with taxol and β-lapachone simultaneously (well 6) or with β-lapachone followed by taxol (well 4). Synergism was not observed if taxol was added before β-lapachone treatment (well 5). The quantitative data are shown in Fig. 1B. This schedule dependency was observed in all the cell lines. These results suggest that the order of artificial checkpoint imposition is important for the synergism mechanism.

Figure 1.

Synergistic induction of cell death by β-lapachone and taxol. Colony formation was carried out as described in Materials and Methods. (A). In a typical experiment, control DU145 cells in well 1 were treated with solvent on days 1 and 2. Cells in well 2 were treated with β-lapachone at 4 μM on day 1 for 4 h, incubated in drug-free medium for 20 h, and then treated with solvent control on day 2. Taxol-alone well 3 was treated with solvent control for 4 h on day 1 and with taxol at 0.02 μM for 4 h on day 2. Cells in well 4 were treated with β-lapachone on day 1 and with taxol on day 2. In well 5, cells were treated with taxol on day 1 and with β-lapachone on day 2. In well 6, cells were treated with β-lapachone and taxol on day 2. (B) Average of results of three independent experiments.

Ablation of in Vitro Colonies in a Wide Spectrum of Human Carcinoma Cells by the Combination of β-Lapachone and Taxol.

Human carcinoma cell lines of different histotypes were used to determine cell survival in the colony formation assay (Table 1). The combination of β-lapachone and taxol dramatically reduced cell survival in a variety of human cancer cells, including ovarian, breast, prostate, melanoma, lung, and pancreatic cancer cell lines. β-Lapachone or taxol alone at the concentrations used were much less effective in decreasing cancer cell colony formation. This decreased cell survival was achieved by induction of cell death as determined by the MTT (Thiazolyl blue) and trypan blue assays. Cell death was by apoptosis as determined by DNA laddering formation (Fig. 2) and by annexin staining (data not shown). Taxol was at least 10-fold more potent in the presence of β-lapachone, as measured at IC₅₀ (data not shown). The exact mechanism required to produce the synergy needs to be explored.

Table 1.

Inhibition of cancer cell survival by β-lapachone and taxol

Cell line	Tissue origins	Colonies, percentage of control
		β-Lapachone	Taxol	β-Lapachone + taxol
A2780DDP	Ovary	77  (1.1)	39  (0.8)	0
MCF-7	Breast	46  (1.4)	45  (0.3)	0
21MT	Breast	56  (5.0)	63  (7.0)	0
Skmel-28	Melanoma	56  (1.4)	44  (5.1)	0
HT-29	Colon	42  (1.4)	64  (2.5)	0
ASPC-1	Pancreas	45  (1.9)	71  (0.8)	0
G480	Lung	32  (0.3)	39  (2.6)	2  (0.1)
DU145	Prostate	50  (2.2)	30  (0.9)	0

Cells were treated for 4 h with β-lapachone and/or taxol at the following concentrations: A2780DDP, β-lapachone at 2 μM and/or taxol at 0.2 μM; MCF-7 and 21-MT, β-lapachone at 4 μM and/or taxol at 0.1 μM; Skmel-28, β-lapachone at 4 μM and/or taxol at 0.1 μM; HT-29, β-lapachone at 4 μM and/or taxol at 0.1 μM; ASPC-1, β-lapachone at 4 μM and/or taxol at 0.2 μM; G480, β-lapachone at 4 μM and/or taxol at 0.2 μM; DU145, β-lapachone at 4 μM and/or taxol at 0.2 μM. The number of colonies in control well was taken as 100% survival. Treated wells are presented as percentage of control. Data are given as average (+SEM) from three independent experiments. 

Figure 2.

Induction of DNA laddering in cells treated with both β-lapachone and taxol. DU145 cells were treated for 4 h with solvent (lane 1), β-lapachone (lane 2), taxol (lane 3), or both (lane 4). Cells were incubated further in drug-free medium for an additional 20 h. Genomic DNA was extracted and subjected to agarose gel electrophoresis (14).

Potent Inhibition of in Vivo Tumor Growth by β-Lapachone plus Taxol.

To determine whether the synergism observed in vitro holds true in vivo, we tested the taxol plus β-lapachone combination in a xenograft model of human ovarian cancer in nude mice. These human ovarian cancer cells were isolated originally from malignant ascites (17). They form carcinomatosis foci 7–10 days after i.p. inoculation of 10 × 10⁶ cells. In control mice (Fig. 3A), there were numerous large tumor nodules (≈100) at 2 months after cell inoculation, in addition to formation of significant malignant ascites. To increase the stringency of these tests, administration of drugs was delayed until tumor foci were formed, unlike in most tumor therapeutic experiments. The decrease in tumor number was quite pronounced (75%) with β-lapachone treatment alone (50 mg/kg, i.p.; Fig. 3B). Mice treated with taxol alone (1.0 mg/kg, i.p.) showed a smaller effect of a ≈60% reduction in the total number of tumor nodules (Fig. 3C). Moreover, in both cases, there was considerable reduction in the size of the tumor nodules and the amount of ascites. With the combination of β-lapachone and taxol, no malignant ascites were seen on laparotomy, and the peritoneum was clean except for zero to three tiny foci per mouse (Fig. 3D), These foci were counted as tumor nodules, although they may represent fibrotic scarring. The data are summarized in Fig. 4A. Importantly, mice treated with the combination regimen seem healthy without reduction in body weight (Fig. 4B). There were no gross abnormalities in internal organs on autopsy.

Figure 3.

Potent inhibition of ovarian tumor growth in vivo by β-lapachone and taxol. Athymic female nude mice (Ncr) were irradiated (300 R) before tumor inoculation, and 24 h later, 10 × 10⁶ human ovarian cancer cells (36M2) were inoculated by i.p. injection. Administration of drugs was initiated 10 days after tumor inoculation. The control group (A) was treated with vehicle alone. The β-lapachone alone group (B) was treated with 50 mg/kg i.p., and the taxol alone group (C) was treated with 1 mg/kg i.p, followed 24 h later by i.p. injection of vehicle. In the combination group (D), mice were treated with β-lapachone alone, followed 24 h later by taxol at 1 mg/kg. There was a 1-day break between each cycle. Mice were treated for a total of 10 cycles and were killed for assessment of the antitumor activity 2 weeks after discontinuation of drug treatment.

Figure 4.

The data described in Fig. 3 are summarized in A. (B) Toxicity to mice was evaluated by body weight gain and general appearance.

In a preliminary experiment with four mice per group, androgen-independent DU145 prostate cancer cells were xenografted into immunocompromised mice (Fig. 5A). Again, to increase stringency and unlike most antitumor experiments, treatment was delayed until the tumors reached ≈0.5 cm in diameter. Either β-lapachone or taxol alone showed moderate inhibitions of tumor growth (data not shown). β-Lapachone plus taxol showed dramatic antitumor activity (Fig. 5B). Furthermore, tumors in the treated mice did not grow back as of the follow-up 6 weeks after treatment.

Figure 5.

Potent inhibition of prostate tumor growth in vivo by β-lapachone and taxol. Male SCID (ICR) mice were inoculated with androgen-independent human prostate cancer cells (DU145; 8 × 10⁶ s.c.). Administration of drugs was initiated when tumor nodules reached ≈0.5 cm in diameter. Four mice were used in this preliminary experiment. The control group (A) was treated with vehicle alone. The β-lapachone alone group was treated with 50 mg/kg i.p., and the taxol alone group was treated with 1 mg/kg i.p, followed 24 h later by i.p. injection of vehicle. In the combination group (B), mice were treated with β-lapachone alone, followed 24 h later by taxol at 1 mg/kg. There was a 1-day break between each cycle. Mice were treated for a total of six cycles. Pictures were taken 3 weeks after six cycles of treatment.

Simultaneous Treatment with β-Lapachone and Taxol Imposed Multiple Artificial Checkpoints on Cells.

To determine the cycle phases at which cells arrest before death, cells were stained with propidium iodide and subjected to cell-cycle analysis. At 8 h after treatment with β-lapachone, the cells accumulated predominantly in S (at 4 μM) or in G₁ phases (at 8 μM). Correspondingly, the fraction of cells in the G₂/M phase was reduced significantly. This pattern was observed in four independent experiments (not shown). At 24 h after β-lapachone treatment, this cell-cycle checkpoint delay persisted (Fig. 6B). These results suggest that β-lapachone induces delays in the G₁ or S phase checkpoint before apoptotic cell death. These results with β-lapachone contrast those obtained with most DNA-damaging drugs that induce G₂/M arrest of cancer cells (4). However, if cells were treated with taxol, without (Fig. 6C) or after β-lapachone treatment, we observed predominantly a G₂/M delay and not apoptosis. As shown in Fig. 6D, simultaneous treatment with β-lapachone and taxol resulted in a combined pattern of checkpoint delays at G₁, S, and G₂/M, with significant accumulation in S phase. This pattern held true if cells were treated with β-lapachone and then taxol, which was followed by synergistic induction of apoptosis. These results suggest a correlation between synergistic induction of apoptosis and cell-cycle delay at multiple checkpoints.

Figure 6.

Flow cytometric analysis of induction of multiple cell-cycle checkpoint delays by the combination of β-lapachone and taxol. DU145 cells were treated with vehicle control for 24 h (A); treated with β-lapachone at 4 μM for 4 h and incubated in drug-free medium for 20 h (B); treated with taxol at 0.2 μM for 4 h and incubated in drug-free medium for 20 h (C); and treated with β-lapachone at 4 μM plus taxol at 0.2 μM for 4 h and incubated in drug-free medium for 20 h (D). Data represent one of three independent experiments.

These Checkpoints Are Associated with the Induction of p21 and Inhibition of Cyclin E-, Cyclin A-, and Cyclin B-Associated cdc2 Kinases.

We next investigated a possible molecular mechanism for this synergistic lethality. Cell-cycle checkpoints and apoptosis are regulated by cdc2 kinases and their inhibitors (19–21), and therefore, establishment of these checkpoints should be reflected in changes in these signaling molecules. We initially chose p21, which is a critical kinase inhibitor for controlling G₁ and S phase checkpoints (19) and also is involved in the regulation of the G₂/M checkpoint (3). Treatment with β-lapachone alone induced expression of p21 at both protein and mRNA levels (Fig. 7A). This induction of p21 is independent of wild-type p53, because the DU145 cell line we used is p53-null and because we observed no changes in the level of p53 expression (15). Combination of β-lapachone with taxol did not alter the level of p21 induction further (data not shown). These results suggest that checkpoint delay in G₁ or S phases caused by treatment with β-lapachone alone or with taxol is accompanied by the induction of p21.

Figure 7.

Changes of regulatory molecules. (A) Induction of p21 by β-lapachone. Human DU145 prostate cancer cells were treated with various concentrations of β-lapachone for 4 h and further incubated with drug-free medium for 20 h. Total RNA was extracted, and Northern blotting was performed as described (15). p21 protein levels were determined by Western blotting (19). Lane 1, control; lane 2, 0.2 μM; lane 3, 1 μM; lane 4, 5 μM β-lapachone. (B) Changes of cyclins. Human prostate cancer cells (PC-3) were treated with β-lapachone and/or taxol as described above. Nuclear extracts were prepared. Cyclin-associated kinase activities were determined by immunoprecipitation as described (14). Lane 1, control cells; lane 2, 4 μM β-lapachone; lane 3, 0.02 μM taxol; lane 4, 4 μM β-lapachone plus 0.02 μM taxol; lane 5, control antibody.

We then determined changes of the cyclin/cdk2 kinases that regulate progression through G₁, S, and G₂/M phases (19–21). As shown in Fig. 7B, treatment of human prostate PC-3 cancer cells with β-lapachone alone (lane 2) resulted in decreased cyclin E-, cyclin A-, and cyclin B-associated cdc2 kinases, consistent with cell-cycle delay at G₁ and S phases. In contrast, treatment with taxol, alone or plus β-lapachone (lanes 3 and 4), resulted in enhanced cyclin A and B kinase activities. These results suggest that the G₂/M delay is caused by taxol’s interruption of microtubule function (6–8), and the enhanced cyclin A and B/cdc2 kinase activities result from accumulation of cells in G₂/M and not from their direct inhibition by taxol.

Discussion

The efficacy of anticancer therapy depends, to a large extent, on finding the right combination regimen. We demonstrate that the combination of β-lapachone followed by taxol has an unusually strong activity against a wide spectrum of human cancer cells in vitro. Strikingly, it virtually eliminated established human ovarian and prostate tumors implanted into immunosuppressed mice. Tumor cells derived from other tissue types also were sensitive in vitro to this drug combination and thus might be sensitive in vivo as well. These dramatic antitumor activities of β-lapachone in combination with taxol, at doses well tolerated by mice, suggest the potential for translating these results into clinical therapies.

Notably, we found that β-lapachone alone was as effective as taxol alone at the doses used in antitumor activity. β-Lapachone, a DNA topoisomerase I inhibitor, induces cell-cycle delay at G₁ or S phase, unlike most DNA-damaging agents, which arrest cells at the G₂/M transition (3). This artificially imposed G₁ and S checkpoint delay by β-lapachone precedes apoptosis in a p53-independent manner (15). These properties suggest the potential of β-lapachone as an anticancer agent and as an addition to combination regimens.

The combination of taxol and β-lapachone was designed according to our hypothesis that a synergistic induction of apoptosis may occur if an unsynchronized population of cells are treated with drugs imposing different artificial checkpoints. This approach represents a strategy for developing cancer therapies by exploiting cell checkpoint functions. Multiple checkpoints are built into the machinery of the cell proliferation cycle to protect chromosomal integrity (19–20). The ≈10¹⁶ cell multiplications in the human life time, together with inevitable errors in DNA replication and exposure to ultraviolet rays and mutagens, underscores the requirement for checkpoint functions (22). Major checkpoints occur at G₁/S phase and at the G₂/M phase transitions (2, 3, 23–25) where cells make a commitment to repair DNA or to undergo apoptosis (19). Cells are generally thought to undergo apoptosis when DNA damage is irreparable (26). Identification of therapeutic agents modulating the checkpoint control may improve cancer treatment (2, 4).

One approach to target checkpoints is to develop drugs that abrogate G₂/M checkpoints to enhance the sensitivity of cancer cells to DNA-damaging agents, a property first observed with caffeine and its analogues (2, 3). In this study, we attempted to develop anticancer therapy by combing drugs imposing different artificial checkpoints. In theory, such treatment may halt the cell cycle at multiple checkpoints simultaneously, setting the stage for enhanced checkpoint-mediated apoptosis. Cells that managed to escape from the checkpoint delays at one checkpoint can be trapped at the next one.

We seem to have created an apoptotic condition by imposing conditions that alter several signaling molecules including p21 and cyclins E, A, and B (19) and thereby simultaneously produce conflicting signals regarding cell-cycle progression vs. arrest, related to different checkpoints. In this fashion, cancer cells with uncontrolled proliferation signals and genetic abnormalities are blocked at multiple checkpoints, creating “collisions” that promote apoptosis (1). Normal cells with well controlled proliferation signals should be delayed at these checkpoints in a regulated fashion, resulting in no apoptosis-prone collisions. In preliminary experiments, we have observed that normal human mammary epithelial cells are less sensitive to the combination of β-lapachone and taxol as compared with human breast cancer cells (data not shown). Further testing of combinations of various drugs inducing different cell-cycle checkpoints and changes in signaling molecules should help us to understand the mechanism of this synergism as well as provide anticancer therapies.