Addition of 5-fluorouracil to doxorubicin-paclitaxel sequence increases caspase-dependent apoptosis in breast cancer cell lines

bcr1274 BCJ Research article Addition of 5-fluorouracil to doxorubicin-paclitaxel sequence increases caspase-dependent apoptosis in breast cancer cell lines Zoli Wainer w.zoli@ausl.fo.it Ulivi Paola Tesei Anna anna.tesei@ausl.fo.it Fabbri Francesco Rosetti Marco Maltoni Roberta Giunchi Casadei Donata Ricotti Luca Brigliadori Giovanni Vannini Ivan Amadori Dino res.onco@ausl.fo.it

Division of Oncology and Diagnostics, Morgagni Pierantoni Hospital, Forlì, Italy

Istituto Oncologico Romagnolo, Forlì, Italy

Breast Cancer Research 1465-5411 2005 7 5 R681 R689 http://breast-cancer-research.com/content/7/5/R681 1616811310.1186/bcr1274 9 3 2005 26 5 2005 22 6 2005 2005 Zoli et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Introduction

The aim of the study was to evaluate the activity of a combination of doxorubicin (Dox), paclitaxel (Pacl) and 5-fluorouracil (5-FU), to define the most effective schedule, and to investigate the mechanisms of action in human breast cancer cells.

Methods

The study was performed on MCF-7 and BRC-230 cell lines. The cytotoxic activity was evaluated by sulphorhodamine B assay and the type of drug interaction was assessed by the median effect principle. Cell cycle perturbation and apoptosis were evaluated by flow cytometry, and apoptosis-related marker (p53, bcl-2, bax, p21), caspase and thymidylate synthase (TS) expression were assessed by western blot.

Results

5-FU, used as a single agent, exerted a low cytotoxic activity in both cell lines. The Dox→Pacl sequence produced a synergistic cytocidal effect and enhanced the efficacy of subsequent exposure to 5-FU in both cell lines. Specifically, the Dox→Pacl sequence blocked cells in the G2-M phase, and the addition of 5-FU forced the cells to progress through the cell cycle or killed them. Furthermore, Dox→Pacl pretreatment produced a significant reduction in basal TS expression in both cell lines, probably favoring the increase in 5-FU activity. The sequence Dox→Pacl→48-h washout→5-FU produced a synergistic and highly schedule-dependent interaction (combination index < 1), resulting in an induction of apoptosis in both experimental models regardless of hormonal, p53, bcl-2 or bax status. Apoptosis in MCF-7 cells was induced through caspase-9 activation and anti-apoptosis-inducing factor hyperexpression. In the BRC-230 cell line, the apoptotic process was triggered only by a caspase-dependent mechanism. In particular, at the end of the three-drug treatment, caspase-8 activation triggered downstream executioner caspase-3 and, to a lesser degree, caspase-7.

Conclusion

In our experimental models, characterized by different biomolecular profiles representing the different biology of human breast cancers, the schedule Dox→Pacl→48-h washout→5-FU was highly active and schedule-dependent and has recently been used to plan a phase I/II clinical protocol.

Introduction

Breast cancer is still a leading cause of cancer death in women in the United States and Europe 1. Adjuvant chemotherapy has been shown to provide disease-free and overall survival benefits for patients with node-positive breast cancer in large meta-analyses conducted by the Early Breast Cancer Trialists' Collaborative Group 2. A study by Weiss et al. 3, however, showed that the overwhelming majority (80%) of node-positive patients relapse and die within 26 years of diagnosis despite the use of cyclophasphamide-methotrexate-5-fluorouracil (CMF)-based adjuvant chemotherapy. The development of drugs and strategies to improve relapse-free and overall survival therefore remains a high priority.

Anthracyclines are among the most active chemotherapeutic drugs for the treatment of breast cancer and anthracycline-containing regimens have an impact, albeit modest, on patient survival 24.

Paclitaxel (Pacl), belonging to the chemical class of taxanes, is capable of inducing in vitro apoptosis, independently of p53 status, through its microtubule-stabilizing activity and, as recently published, by inducing the release of cathepsin B from lysosomes 5. From a clinical point of view, taxanes appear to be promising, although their real impact on the natural history of breast cancer has yet to be defined. A similar impact of Pacl on clinical response has been reported, however, for patients with wild-type or p53-mutated cancers 6789, the latter representing 15% of in situ breast carcinomas and 50% of invasive disease 10. Moreover, the loss of normal p53 function sensitizes in vitro cells to drug activity 111213, hence the interest to use the taxane combination with other drugs as clinical therapy. The association of a taxane with an anthracycline is based on evidence that the two drugs have different mechanisms of action and toxicities and are not cross-resistant. Moreover, in vitro studies 14151617 on human cell lines and primary breast cancer cultures have shown a schedule-dependent interaction of the two drugs. Phase I and II studies with the doxorubicin (Dox)-Pacl sequence defined by preclinical studies have reported objective responses ranging from 40% to more than 90% 1819, and a recent prospective phase III study on advanced breast cancer reported an impressive response rate (up to 94%) for the sequence, which was shown to prolong overall survival compared to the fluorouracil, epirubicin, cyclophosphamide (FEC) combination 20.

5-Fluorouracil (5-FU), one of the most important agents for the treatment of colorectal, head and neck, pancreatic and breast carcinomas, is a pro-drug that requires conversion to 5-fluoro-deoxyuridine-monophosphate (5FdUMP) and 5-fluoro-deoxyuridine-triphosphate (5FUTP) in cancer cells. Several in vitro studies on human solid tumor cell lines have demonstrated the positive and schedule-dependent interaction of Pacl and 5-FU 212223. A synergistic effect was obtained only when tumor cells were exposed to Pacl followed by antimetabolites. Conversely, simultaneous exposure to the two drugs or pretreatment with 5-FU reduced overall cell killing compared to Pacl alone. Specifically, it has been demonstrated that a short pretreatment with Dox increases the activity of Pacl and of the Pacl→gemcitabine sequence 24. In the light of these preclinical and clinical observations, the present study aimed to investigate the cytotoxic effect produced by the combination of Dox, Pacl and 5-FU in human breast cancer cell lines and to define, at the preclinical level, the most effective treatment scheme.

Materials and methods

Cell lines

The study was performed on two human breast cancer cell lines: a commercial line (MCF-7; 40-h doubling time, obtained from the American Type Culture Collection (Rockville, MD), and a cell line established in our laboratory (BRC-230; 30-h doubling time) 25. MCF-7 cells express estrogen receptors and bcl-2, harbor a wild-type p53 gene and lack pro-caspase-3. BRC-230 cells have no estrogen receptors, do not express bcl-2, contain a p53 gene mutation, and express pro-caspase-3. Culture medium was composed of DMEM/HAM F12 (1:1) supplemented with FCS (10% v/v), glutamine (2 mM) (Mascia Brunelli SpA, Milan, Italy) and insulin (10 μg/ml) (Sigma Aldrich, Milan, Italy). Cells were used in the exponential growth phase for all experiments.

In vitro chemosensitivity assay

The sulforhodamine B (SRB) assay was used according to the method of Skehan et al. 26. Briefly, 10⁴cells in 100 μl of medium/well were plated in 96-well flat-bottomed microtiter plates and 18 to 24 h after plating (an adequate time for exponential growth recovery) culture medium was replaced with fresh medium either containing or not containing the drugs. At the end of drug exposure, cells were fixed for 1 h and stained with 0.4% SRB (Sigma Aldrich) dissolved in 1% acetic acid for 30 minutes. The plates were then washed four times with 1% acetic acid to remove unbound stain, air-dried and solubilized in 100 μl of 10 mM unbuffered Tris base (tris(hydroxymethyl)aminomethane) solution. The optical density of treated cells was detected at 490 nm. Each sample was run in octuplet, and each experiment was repeated three times. Therefore, each experimental value in the graphs represents the median of 24 samples.

Drugs and chemicals

Dox (Pharmacia Italia SpA, Milan, Italy) and 5-FU (Roche, Milan, Italy) were diluted in sterile saline solution, and Pacl (Bristol Meyers Squibb, Rome, Italy) was diluted in 95% ethanol. The drugs were then divided into aliquots and stored at -70°C. Drug stocks were freshly diluted in culture medium before each experiment. Z-LEHD-FMK (caspase-9 inhibitor) and Z-DEVD-FMK (caspase-3 inhibitor) (BD Biosciences Pharmingen, Milan, Italy) were solubilized in dimethylsulfoxide (DMSO) (Sigma Aldrich) and freshly diluted in culture medium at a concentration of 100 μM before each experiment. The final DMSO concentration never exceeded 1% and this condition was used as control in each experiment.

Drug exposure

The drugs were tested singly at scalar concentrations of 0.001, 0.01, 0.1 and 1 μg/ml for Pacl and Dox, and 0.01, 0.1, 1 and 10 μg/ml for 5-FU. The exposure time to each single agent (4 h for Dox and 24 h for Pacl or 5-FU) was chosen from the dose inhibition rate curves and represented the time that produced the maximum effect. Control samples were processed as treated samples but in drug-free medium, and evaluation of the cytotoxic effect was performed immediately after the end of drug exposure. Each experiment testing the activity of the drugs singly and in combination was run in octuplet and repeated three times.

Drug combinations

Dox/Pacl

Based on our previous results, in sequence experiments, cells were exposed to Dox for 4 h, after which the drug-containing medium was removed and cells were incubated for 24 h with fresh medium containing Pacl. In all the combination experiments, each drug was tested at the four different concentrations used for single drug exposure (0.001, 0.01, 0.1 and 1 μg/ml) at a 1:1 ratio.

Dox/Pacl/5-FU

Dox and Pacl were used as described above. After exposure, the medium was removed and cells were cultured in drug-free medium for 48 h, after which they were treated with 5-FU for 24 h. This scheme, derived from previous experimental studies, was considered as the most effective. In these combination experiments, the three drugs were tested at the four different concentrations used for single drug exposure (0.001, 0.01 and 0.1 and 1 μg/ml for Dox and Pacl and 0.01, 0.1, 1 and 10 μg/ml for 5-FU) at a 1:1:10 ratio.

The cytotoxic effect was evaluated immediately after the end of the three-drug exposure. Controls were processed as treated samples but in drug-free medium.

Data analysis

The type of drug interaction was determined by the median effect principle according to the method of Chou and Talalay 27 and was applied for the three-drug treatment 28. On the basis of this approach, the interaction between the three drugs was quantified by determining a combination index (CI) at increasing levels of cell kill. CI values lower than, equal to, or higher than 1 indicated synergy, additivity, or antagonism, respectively.

Cell cycle perturbations

Cells (2 × 10⁵) were cultured in medium either containing or not containing (control) the cytotoxic drugs in sequence at a concentration of 0.1 μg/ml for Dox and Pacl and 1 μg/ml for 5-FU. After drug exposure, cells were harvested and stained in a solution containing RNAase (10 Ku/ml; Sigma-Aldrich), Nonidet P40 (0.01%; Sigma-Aldrich) and propidium iodide (1 μg/ml; Sigma-Aldrich). Samples were analyzed 30 to 60 minutes later by flow cytometry (Becton Dickinson Italia SpA, Milan, Italy). Data acquisition (10,000 events for each sample) was performed using CELLQuest software (Becton Dickinson Italia). Data were elaborated using Modfit (DNA Modelling System) software (Verity Software House Inc., Topsham, ME, USA) and expressed as fractions of cells in the different cell cycle phases. Samples were run in triplicate and each experiment was repeated three times.

Apoptosis

Apoptosis was evaluated by flow cytometric analysis according to the previously described TUNEL assay procedure 29. Briefly, the cells, treated as outlined above, were trypsinized, fixed, exposed to TUNEL reaction mixture and counterstained with propidium iodide before FACS analysis.

Trypsinized cells were treated with anti-Fas antibody (DakoCytomation Denmark A/S, Glostrup, Denmark) diluted 1:500 for 1 h at 4°C to evaluate Fas receptor expression by flow cytometry. Cells were then incubated with a goat-mouse FITC-conjugated secondary antibody (DakoCytomation). Flow cytometric data acquisition and analysis were performed using CELLQuest software (Becton Dickinson Italia). For each sample, 15,000 events were recorded.

Western blot analysis

After the three-drug exposure, cells were treated according to the previously described western blot procedure 29. The monoclonal antibodies used were anti-p53 (PAb 1801, Bioptica, Milan, Italy) 1:400, anti-bcl-2 (clone 124, Dako Corporation, Santa Barbara, CA, USA) 1:50, anti-p21 (clone DCS-60.2, Bioptica) 1:100, antithymidylate synthase (clone TS 106, Bioptica) 1:100 and anti-caspase-8 (clone 12F5, Alexis Biochemicals Corporation, Lausanne, Switzerland) 1:500. We used polyclonal antibodies for anti-caspases -3, -6, -7 and -9 (Cell Signaling Technology, Inc., Beverly, MA, USA), all diluted at 1:500, anti-apoptosis-inducing factor (AIF) (Chemicon International, Inc., Temecula, CA, USA) 1:500 and anti-bax (Pharmingen, San Diego, CA, USA) 1:1000. Actin polyclonal antibody (Sigma-Aldrich) was used as control to demonstrate equal loading. The samples were analyzed using QuantiScan software (Biosoft, Cambridge, UK).

Results

Cytotoxic activity

The cytotoxic activity of 5-FU, of the 4-h Dox→24-h Pacl sequence, and of the 4-h Dox→24-h Pacl→48-h wash-out→24-h 5-FU exposure is reported in Fig. 1. 5-FU showed a moderate cytotoxic activity in both cell lines. The Dox→Pacl sequence produced a strong cytocidal effect and an important synergism in MCF-7 and BRC-230 cell lines, as shown by Chou-Talalay analysis (Table 1). The synergistic effect dramatically increased, starting from the lowest doses, when cells were treated with 5-FU after the sequence anthracycline→taxane→48-h washout (Fig. 1).

Figure 1

Effect of different drug treatments on the survival of MCF-7 and BRC-230 breast cancer cell lines

Effect of different drug treatments on the survival of MCF-7 and BRC-230 breast cancer cell lines. (a) 5-FU, (b) Dox→Pacl sequence and (c) Dox→Pacl→5-FU treatment. D1, D2, D3 and D4 represent the doses of the three drugs used in the sequence (see Materials and methods). Each data point is the average of at least three independent experiments performed in octuplet. The standard deviation never exceeded 5%.

Table 1

Combination index values induced by sequential treatments

MCF-7

BRC-230

CI value at inhibition of

Drug combination

Combination ratio

75%

90%

95%

75%

90%

95%

Dox→Pacl

1:1

0.3

0.4

0.1

0.2

Dox→Pacl→48-h washout→5-FU

1:1:10

0.0005

0.002

0.005

0.002

0.007

0.02

5-FU, 5-fluorouracil; CI, combination index; Dox, doxorubicin; Pacl, paclitaxel.

Cell cycle perturbations and apoptosis

A 4-h treatment with Dox or a 24-h exposure to 5-FU did not induce biologically relevant cell cycle perturbations (full recovery was observed after a 24-h culture in drug-free medium) or significant apoptosis (data not shown). Conversely, a 24-h treatment with Pacl produced a considerable increase of cells in G2-M phases, with a decrease in G1 phase, which further increased 24 h after drug removal and began to recover after 48 h. Moreover, about 10% and 15% of apoptotic cells were detected in BRC 230 and MCF-7 cell lines, respectively (data not shown). The Dox→Pacl sequence caused a dramatic block of cells in G2-M and a decrease in G1-S phases, which persisted up to 48 h after drug removal. In parallel, the fraction of apoptotic cells (Fig. 2) rose from 15% to 22% as washout time increased. This finding was similar in both cell lines (Table 2).

Figure 2

BRC-230 cells after Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout→5-FU (1 μg/ml) treatment

BRC-230 cells after Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout→5-FU (1 μg/ml) treatment. Apoptotic nuclei stained with DAPI show intense fluorescence corresponding to chromatin condensation (arrow heads) and fragmentation (arrows).

Table 2

Distribution of cells in the different cell cycle phases (%) and apoptotic cells (%) after different treatments

MCF-7

BRC-230

Treatment

G2-M

Apoptosis

G2-M

Apoptosis

Untreated cells

Dox→Pacl

6^a

14^a

80^a

22^a

10^a

27^a

63^a

15^a

Dox→Pacl→48-h washout

3^a

16^a

81^a

17^a

5^a

11^a

84^a

15^a

Dox→Pacl→48-h washout→5-FU

15^a

36^a

40^a

20^a

42^a

38^a

37^a

^aP < 0.05 by t-test. 5-FU, 5-fluorouracil; Dox, doxorubicin; Pacl, paclitaxel.

Cell cycle perturbations consisting of higher and lower fractions of cells in G2 and G1/S phases, respectively, persisted after exposure to the Dox→Pacl→5-FU sequence (Table 2), albeit to a lesser extent than that observed after the Dox→Pacl sequence. A statistically significant increase of up to 40% was observed in apoptotic cells.

The presence of caspase-9 inhibitor during exposure to the Dox→Pacl→5-FU sequence induced a reduction from 40% to 21% in the apoptotic cell fraction in the MCF-7 line (Fig. 3a). Similarly, the presence of caspase-3 inhibitor during the three-drug treatment caused a reduction in the apoptotic fraction from 37% to about 6% in BRC-230 cells (Fig. 3b).

Figure 3

Inhibition of apoptosis

Inhibition of apoptosis. Results from TUNEL assay showing inhibition induced by Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout→5-FU (1 μg/ml) treatment in (a) BRC-230 cells in the presence of caspase-3 (casp-3) inhibitor and (b) MCF-7cells in the presence of caspase-9 (casp-9) inhibitor.

Apoptotic-related markers

In MCF-7 cells basally expressing bcl-2 and harboring wild-type p53, Dox or 5-FU did not have any effect on apoptotic-related markers, whereas Pacl caused bcl-2 phosphorylation as well as p53 overexpression (data not shown). Bcl-2 phosphorylation further increased after the Dox→Pacl sequence. An increase in pro-apoptotic bax, p21 and p53 expression was observed at the end of the two-drug sequence and was even more evident after the three-drug treatment (Fig. 4).

Figure 4

Protein levels of bcl-2, bax, p53 and p21 following different drug exposures

Protein levels of bcl-2, bax, p53 and p21 following different drug exposures. Protein (50 μg) was loaded for the controls and treated samples: (a) untreated cells; (b) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml); (c) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout; (d) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout→5-FU (1 μg/ml).

The BRC-230 cell line, lacking bcl-2 and with mutated p53, showed no induction of bax or p53 overexpression after any treatment. Conversely, Dox→Pacl treatment induced an increase in p21 expression that persisted 48 h after drug removal and was still evident at the end of the three-drug exposure (Fig. 4).

With regard to the caspase cascade, caspase-9 was activated in the MCF-7 cell line after the Dox→Pacl sequence and further increased at the end of the three-drug treatment. Similar behavior was observed for caspase-8, but its active 18 kDa fragment was not detected. Caspase-7 activation was induced after the three-drug sequence only (Fig. 5), whereas caspase-6 cleavage (data not shown) was not induced by any of the treatments. In parallel, a hyperexpression of AIF was observed at the end of both two- and three-drug treatments (Fig. 5).

Figure 5

Western blot analysis of caspases and apoptosis-inducing factor (AIF) proteins following different treatments

Western blot analysis of caspases and apoptosis-inducing factor (AIF) proteins following different treatments. Protein (50 μg) was loaded for the controls and treated samples: (a) untreated cells; (b) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml); (c) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout; (d) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout→5-FU (1 μg/ml).

In BRC-230 cells, an induction of caspase-8, with its 18 kDa fragment, was observed at the end of the Dox→Pacl exposure and was still present after the three-drug treatment. Caspases -3 and -7 were activated 48 h after the end of Dox→Pacl treatment and persisted after the three-drug treatment. Conversely, none of the treatments induced caspase-9 (Fig. 5) or -6 expression (data not shown) or altered Fas expression (data not shown).

Thymidylate synthase

Thymidylate synthase (TS) was basally expressed in both cell lines, but at higher levels in MCF-7 than in BRC-230, as seen by a clear western blotting band at 36 kDa (Fig. 6). 5-FU exposure induced a more evident increase in TS expression in the BRC-230 than in the MCF-7 cell line, as demonstrated by the appearance of an additional band of approximately 38 kDa.

Figure 6

Western blot analysis of thymidylate synthase and ternary complex

Western blot analysis of thymidylate synthase and ternary complex. (a) Untreated cells; (b) 5-FU (1 μg/ml); (c) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml); (d) Dox (0.1 μg/ml)→Pacl (0.1 μg/ml)→48-h washout→5-FU (1 μg/ml). 50 μg of protein were loaded for the controls and treated samples.

Dox-Pacl treatment induced a significant decrease in TS expression in MCF-7 cells characterized by higher basal TS levels, and a slight reduction in low basal TS levels in the BRC-230 cell line. Exposure to 5-FU after Dox→Pacl treatment induced an increase in TS protein expression that was, however, lower than that induced by treatment with 5-FU alone, in both cell lines. Moreover, in BRC-230 cells, the three-drug treatment did not induce a detectable TS ternary complex (Fig. 6).

Discussion

To date the clinical design of polychemotherapeutic protocols has mainly taken into account information derived from experimental studies on mechanisms of action of different agents and has favored combinations of drugs with complementary toxicities, but it has been clearly demonstrated that sequencing and timing of administration are important to optimize drug activity.

Our results confirmed the synergistic interaction of the Dox→Pacl treatment 141516 and showed that this sequence enhances the efficacy of subsequent exposure to 5-FU, as observed in our previous studies using the antimetabolite gemcitabine 1824. In particular, we observed that cells, as expected, were trapped in the G2-M phase after the Dox→Pacl sequence, and the subsequent addition of 5-FU forced cells to progress through the cell cycle or killed them. In fact, the three-drug treatment doubled the percentage of apoptotic cells produced by the Dox→Pacl sequence in both experimental cell lines.

In MCF-7 cells, phosphorylation of the anti-apoptotic gene and persistent upregulation of pro-apoptotic markers was induced by the Dox→Pacl sequence and further enhanced by exposure to 5-FU. It can be hypothesized that the ensuing mitochondrial instability led to apoptosis through both the cytochrome-c-mediated activation of caspases -9 and -7 and the release of AIF. The important role played by AIF in the apoptosis process in MCF-7, which is lacking in caspase-3 30, was demonstrated by the partial cell death inhibition observed in the presence of caspase-9 inhibitor during the three-drug treatment.

In the BRC-230 cell line, apoptosis was triggered by the activation of caspases -8 and -3 and, albeit to a lesser degree, caspase-7. In particular, the pivotal role of caspase-3 was highlighted by the almost complete suppression of apoptosis in the presence of caspase-3 inhibitor during the three-drug exposure.

TS protein was also modulated after the different treatments in both cell lines. TS, which converts dUMP to dTMP, is the primary target of fluoropyrimidine activity 31. It represents the rate-limiting nucleotide in DNA synthesis and its overexpression has been shown to be associated with resistance to 5-FU-based treatments. Our results showed, as already reported in colon cancer cell lines 32, an increase in TS free protein and the formation of TS ternary complex following 5-FU exposure. This increase was more evident in the p53-mutated BRC-230 line than in wild-type p53 MCF-7 cells. Following the Dox→Pacl treatment, basal TS expression was significantly reduced in MCF-7 and also, albeit to a lesser degree, in BRC-230 cells. Subsequent exposure to 5-FU increased free TS expression without, however, reaching the levels observed after treatment with 5-FU alone. In particular, the increase involved only free TS in BRC-230 and was associated with the formation of the ternary complex, albeit fourfold lower than that induced by 5-FU alone, in MCF-7 cells.

In conclusion, apoptosis was observed after the Dox→Pacl sequence and was even more evident when followed by 5-FU, which induced an important apoptosis in cell lines characterized by different estrogen receptor status and apoptosis-related markers, both representative of the heterogeneous biology of clinical breast cancers. The efficacy of the antimetabolite was favored by an increase in TS levels following exposure to anthracycline and taxane.

Conclusion

The present preclinical study permitted us to define the most effective three-drug sequence, providing the basis for the rationale of an ongoing phase I/II breast cancer clinical protocol in which the oral formulations of Dox and 5-FU are used to reduce toxicity and increase safety of treatment schemes.

Abbreviations

5FdUMP = 5-fluoro-deoxyuridine-monophosphate; 5-FU = 5-fluorouracil; 5FUTP = 5-fluoro-deoxyuridine-triphosphate; AIF = apoptosis-inducing factor; CI = combination index; CMF = cyclophasphamide-methotrexate-5-fluorouracil; DMEM = Dulbecco's modified Eagle's medium; DMSO = dimethylsulfoxide; Dox = doxorubicin; FCS = fetal calf serum; FEC = fluorouracil, epirubicin, cyclophosphamide; Pacl = paclitaxel; SRB = sulforhodamine B; TS = thymidylate synthase.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

All the authors contributed equally to this study.

Acknowledgements

The authors would like to thank Prof. Rosella Silvestrini for her invaluable scientific contribution, Dr Ivan Vannini for his technical assistance and Gráinne Tierney for editing the manuscript. Supported by Istituto Oncologico Romagnolo, Forlì, Italy.

Clinical experience of capecitabine in metastatic breast cancer O'Shaughnessy J Eur J Cancer 2002 38 Suppl 2 10 14 11841930 10.1016/S0959-8049(01)00416-6 Polychemotherapy for early breast cancer: an overview of the randomized trials Early Breast Cancer Trialists' Collaborative Group Lancet 1998 352 930 942 10.1016/S0140-6736(98)03301-7 9752815 Natural history of more than 20 years of node-positive primary breast carcinoma treated with cyclophosphamide, methotrexate, and fluorouracil-based adjuvant chemotherapy: a study by the Cancer and Leukemia Group B Weiss RB Woolf SH Demakos E Holland JF Berry DA Falkson G Cirrincione CT Robbins A Bothun S Henderson IC J Clin Oncol 2003 21 1825 1835 10.1200/JCO.2003.09.006 12721260 The role of taxanes in adjuvant treatment of early stage breast cancer Piccart M Breast Cancer Res Treat 2003 79 Suppl 1 S25 S34 10.1023/A:1024393926965 12868803 Cathepsin B mediates caspase-independent cell death induced by microtubule stabilizing agents in non-small cell lung cancer cells Broker LE Huisman C Span SW Rodriguez JA Kruyt FA Giaccone G Cancer Res 2004 64 27 30 14729603 p53 mutations do not predict response to paclitaxel/radiation for nonsmall cell lung carcinoma Safran H King T Choy H Gollerkery A Kwakwa H Lopez F Cole B Myers J Tarpey J Rosmarin A Cancer 1996 78 1203 1210 10.1002/(SICI)1097-0142(19960915)78:6<1203::AID-CNCR6>3.0.CO;2-A 8826941 The European experience with docetaxel in the treatment of early-stage breast cancer Di Leo A Clin Breast Cancer 2002 3 Suppl 2 S59 S62 12435289 A prospective randomized pilot study to evaluate predictors of response in serial core biopsies to single agent neoadjuvant doxorubicin or paclitaxel for patients with locally advanced breast cancer Stearns V Singh B Tsangaris T Crawford JG Novielli A Ellis MJ Isaacs C Pennanen M Tibery C Farhad A Clin Cancer Res 2003 9 124 133 12538460 A study of the value of p53, HER2, and Bcl-2 in the prediction of response to doxorubicin and paclitaxel as single agents in metastatic breast cancer: a companion study to EORTC 10923 Hamilton A Larsimont D Paridaens R Drijkoningen M van de Vijver M Bruning P Hanby A Houston S Treilleux I Guastalla JP Clin Breast Cancer 2000 1 233 240 discussion 241–242 11899648 Prognostic value of p53 for local failure in mastectomy-treated breast cancer patients Zellars RC Hilsenbeck SG Clark GM Allred DC Herman TS Chamness GC Elledge RM J Clin Oncol 2000 18 1906 1913 10784631 p53 status does not affect sensitivity of human ovarian cancer cell lines to paclitaxel Debernardis D Sire EG De Feudis P Vikhanskaya F Valenti M Russo P Parodi S D'Incalci M Broggini M Cancer Res 1997 57 870 874 9041188 Ovarian cancer cisplatin-resistant cell lines: multiple changes including collateral sensitivity to Taxol Perego P Romanelli S Carenini N Magnani I Leone R Bonetti A Paolicchi A Zunino F Ann Oncol 1998 9 423 430 10.1023/A:1008265012435 9636834 Induction of apoptosis by taxol and cisplatin and effect on cell cycle-related proteins in cisplatin-sensitive and -resistant human ovarian cells Zaffaroni N Silvestrini R Orlandi L Bearzatto A Gornati D Villa R Br J Cancer 1998 77 1378 1385 9652752 A phase I/II study of sequential doxorubicin and paclitaxel in the treatment of advanced breast cancer Amadori D Frassineti GL Zoli W Milandri C Tienghi A Ravaioli A Gentile A Salzano E Semin Oncol 1996 23 5 Suppl 11 16 22 8893894 Doxorubicin and paclitaxel (sequential combination) in the treatment of advanced breast cancer Amadori D Frassineti GL Zoli W Milandri C Serra P Tienghi A Ravaioli A Gentile A Salzano E Oncology (Huntingt) 1997 11 4 Suppl 3 30 33 9144688 Paclitaxel plus doxorubicin in breast cancer: an Italian experience Frassineti GL Zoli W Silvestro L Serra P Milandri C Tienghi A Gianni L Gentile A Salzano E Amadori D Semin Oncol 1997 24 Suppl 17 S17 S19 S17-S25 Pharmacokinetic optimisation of treatment schedules for anthracyclines and paclitaxel in patients with cancer Danesi R Conte PF Del Tacca M Clin Pharmacokinet 1999 37 195 211 10511918 Doxorubicin, paclitaxel and gemcitabine: a Phase I study of a new sequential treatment in stage III B – IV breast cancer Ibrahim T Fabbri M Frassineti GL Zoli W Monti M Ricotti L Amadori D J Chemother 2003 15 488 494 14598942 Sequence-dependent alteration of doxorubicin pharmacokinetics by paclitaxel in a phase I study of paclitaxel and doxorubicin in patients with metastatic breast cancer Holmes FA Madden T Newman RA Valero V Theriault RL Fraschini G Walters RS Booser DJ Buzdar AU Willey J Hortobagyi GN J Clin Oncol 1996 14 2713 2721 8874332 Gemcitabine, anthracycline, and taxane combinations for advanced breast cancer Zielinski CC Oncology (Huntingt) 2003 17 12 Suppl 14 36 40 14768404 Schedule-dependent interaction between paclitaxel and 5-fluorouracil in human carcinoma cell lines in vitro Kano Y Akutsu M Tsunoda S Ando J Matsui J Suzuki K Tkeda T Inoue Y Adaki K Br J Cancer 1996 74 704 710 8795571 Combination chemotherapy of taxanes and antimetabolites: its use and limitations Smorenburg CH Sparreboom A Bontelbal M Verweij J Eur J Cancer 2001 37 2310 2323 10.1016/S0959-8049(01)00309-4 11720823 Reduced expression of p53 and p21 WAF1/CIP1 sensitizes human breast cancer cells to paclitaxel and its combination with 5-fluorouracil Johnson KR Fan W Anticancer Res 2002 22 3197 3204 12530065 Schedule-dependent interaction of doxorubicin, paclitaxel and gemcitabine in human breast cancer cell lines Zoli W Ricotti L Barzanti F Dal Susino M Frassineti GL Milandri C Casadei Giunchi D Amadori D Int J Cancer 1999 80 413 416 10.1002/(SICI)1097-0215(19990129)80:3<413::AID-IJC13>3.0.CO;2-I 9935183 Establishment and characterization of a new cell line from primary human breast carcinoma Amadori D Bertoni L Flamigni A Savini S De Giovanni C Casanova S De Paola F Amadori A Giulotto E Zoli W Breast Cancer Res Treat 1993 28 251 260 10.1007/BF00666586 8018954 New colorimetric cytotoxic assay for anticancer drug screening Skehan P Storeng R Scudiero D Monks A McMahon J Vistica D Warren JT Bokesch H Kenney S Boyd MR J Natl Cancer Inst 1990 82 1107 1112 2359136 Quantitative analysis of dose-effect relationship: the combined effects of multiple drugs or enzyme inhibitors Chou TC Talalay P Adv Enzyme Regul 1984 22 27 55 10.1016/0065-2571(84)90007-4 6382953 Computerized quantitation of synergism and antagonism of taxol, topotecan, and cisplatin against human teratocarcinoma cell growth: a rational approach to clinical protocol design Chou TC Motzer RJ Tong Y Bosl GJ J Natl Cancer Inst 1994 86 1517 1524 7932806 Schedule-dependent cytotoxic interaction between epidoxorubicin and gemcitabine in human bladder cancer cells in vitro Zoli W Ricotti L Tesei A Ulivi P Gasperi Campani A Fabbri F Gunelli R Frassineti GL Amadori D Clin Cancer Res 2004 10 1500 1507 14977854 Styrylpyrone derivative (SPD) induces apoptosis in a caspase-7-dependent manner in the human breast cancer cell line MCF-7 Lee ATC Azimahtol HLP Tan AN Cancer Cell Int 2003 3 16 23 14580263 10.1186/1475-2867-3-16 239923 Thymidylate synthase inhibition triggers apoptosis via caspase-8 and -9 in both wild-type and mutant p53 colon cancer cell lines Backus HHJ Wouters D Ferreira CG van Houten VMN Brakenhoff RH Pinedo HM Peters GJ Eur J Cancer 2003 39 1310 1317 10.1016/S0959-8049(03)00204-1 12763222 Molecular downstream events and induction of thymidylate synthase in mutant and wild-type p53 colon cancer cell line after treatment with 5-fluorouracil and the thymidylate synthase inhibitor raltitrexed Peters GJ van Triest B Backus HHJ Kuiper CM van der Wilt CL Pinedo HM Eur J Cancer 2000 36 916 924 10.1016/S0959-8049(00)00026-5 10785598