- Email: [email protected]
Caffeine diminishes cytotoxic effects of paclitaxel on a human lung adenocarcinoma cell line
Cancer Letters 191 (2003) 101–107 www.elsevier.com/locate/canlet
Caffeine diminishes cytotoxic effects of paclitaxel on a human lung adenocarcinoma cell line Yoshizumi Kitamotoa,*, Hideyuki Sakuraia, Norio Mitsuhashib, Tetsuo Akimotoa, Takashi Nakanoa a
Department of Radiology and Radiation Oncology, Gunma University School of Medicine, 3-39-22, Showa-machi, Maebashi, Gunma 371-8511, Japan b Department of Radiology, Tokyo Women’s Medical University, 8-1, Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan Received 6 August 2002; received in revised form 30 September 2002; accepted 7 October 2002
Abstract This study was performed to investigate how caffeine modifies the cytotoxic effects of paclitaxel on a human lung carcinoma cell line. Caffeine doses up to 5 mM had less effect on clonogenic survival. The cell killing effect, due to paclitaxel, increased in a dose-dependent manner up to 50 nM. For combined treatment with caffeine and paclitaxel, added caffeine reduced the cytotoxic effect of paclitaxel not only in dose – response but also in time – response curves. Caffeine combined with paclitaxel clearly suppressed cell proliferation in a dose-dependent manner. In the cell cycle analysis, caffeine alone caused early G1 accumulation, whereas paclitaxel alone caused an early increase in G2-M and a decrease in G1. As for the effect of caffeine on paclitaxel, caffeine suppressed the effect of paclitaxel on cell cycle distribution, where a dose-dependent early increase in G2-M and a decrease in G1 were not clear. We suggest that cell cycle modifying agents, such as caffeine, potentially diminish the cytotoxic activity of paclitaxel, and one should be careful when combining such agents. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Paclitaxel; Caffeine; Cell cycle; Human lung carcinoma
1. Introduction Paclitaxel is a, naturally occurring, chemotherapeutic agent isolated from the bark of the Western yew [1]. The cytotoxic mechanism of paclitaxel has been found to be a potent microtubule stabilizing agent and a promoter of microtubule assembly, and to form microtubule bundles within cells [1 – 3]. It * Corresponding author. Tel.: þ81-27-220-8383; fax: þ 81-27220-8397. E-mail address: [email protected] (Y. Kitamoto).
inhibits normal mitosis by binding to the N-terminal 31 amino acid of the b-tubulin subunit of microtubules [4] and induces a G2 – M arrest in cycling cells. Paclitaxel has, recently, gained clinical attention as a chemotherapeutic agent for some advanced neoplasms e.g. lung cancer, breast and ovarian cancers, and paclitaxel alone or combination chemotherapy including paclitaxel have been tested in many clinical trials [5 – 8]. Methylxanthine, for instance, caffeine or pentoxifylline has a multiplicity of effects on cells in culture. Caffeine causes G1 phase accumulation [9,10] due to
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. doi: 1 0 . 1 0 1 6 / S 0 3 0 4 - 3 8 3 5 ( 0 2 ) 0 0 5 9 1 - 8
102
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
inhibition of enzymes required for DNA synthesis [11,12]. Caffeine is also known as a modifying agent in combination with many cytotoxic drugs. It has been reported that caffeine enhances the cell killing of many DNA-damaging agents, presumably by shortening the time of cell cycle arrest normally caused by such agents [13 –16], that is, the caffeine-cytotoxic agent sequence leads successively to G2 arrest, followed by G2 exit, and finally apoptosis. Caffeine is believed to cause this progression by stimulating p34cdc2 kinase [14], a cell cycle regulatory protein, that triggers the transition from G2 to M, and which may be involved in induction of apoptosis [17]. For combined effect of paclitaxel and caffeine, Saunders et al. reported, that caffeine enhanced development of apoptosis is induced by paclitaxel in a human breast cancer cell line, MCF-7 [15]. In this paper, the cellular response to simultaneously combined paclitaxel and caffeine is reported and an effort is made to clarify the mechanism of the modification of this simultaneous insult.
2. Materials and methods 2.1. Cell line and culture A human lung adenocarcinoma cell line, A549, was obtained from Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University (Miyagi, Japan). In our laboratory, polymerase chain reaction single-strand conformation polymorphism (PCR –SSCP) analysis revealed that A549 has a wild-type p53. The cells were maintained in RPMI-1640 medium supplemented with 10% fetal calf serum, antibiotics and 0.3% HEPES (Immuno Biological Laboratories Ltd, Gunma Japan). The cells have an apparent doubling time of 24 h when growing exponentially over the concentration range of 1 – 5 £ 105 cells/ml. Cell suspensions were prepared from stock culture flasks and incubated in 25 cm2 cell culture flasks at 5 £ 105 cells in 5 ml complete medium. Cells in the flask were incubated at 378C, in a humidified atmosphere of 5% CO2 and 95% air, for 48 h prior to treatment. In this condition, growth was exponential for up to 4 days after the inoculation.
2.2. Reagents Paclitaxel was purchased from Wako Pure Chemical Industries, Ltd (Osaka, Japan), dissolved in dimethyl sulfoxide (DMSO) at a concentration of 500 mM and stocked at 2 208C. Caffeine (1,3,7trimethylxanthine) was obtained from Wako Pure Chemical Industries, Ltd (Osaka, Japan). The agent was dissolved in complete medium with heating at 708C for 30 min immediately before the treatment. The solution was dissolved in an appropriate concentration with complete medium. For combined treatment, the cells were exposed to 1 –100 nM paclitaxel simultaneously combined with 1 –20 mM caffeine. To estimate the dose response, the cells were incubated in these solutions at 378C for 24 h. In a time course experiment, cells were exposed to 20 nM of paclitaxel ^ 1– 20 mM caffeine at 378C for up to 48 h. 2.3. Clonogenic survival assay After the drug treatment, a single cell suspension was obtained by means of trypsin/ethylenediaminetetraaceticacid (EDTA) detachment. The cells were pelleted by centrifugation at 1000 rpm for 5 min, resuspended in complete medium, counted with a hemocytometer, serially diluted and plated in dishes in proportion to the expected survival. Three dishes with each cell inoculum were used for each treatment point. Dishes were kept at 378C in an incubator for approximately 14 days to allow colony formation. Colonies were fixed with methanol, stained with crystal violet solution and those of over 50 cells were counted as survivors. The surviving fraction was determined by the proportion of seeded cells surviving relative to untreated cells, after the treatments to form colonies. Plating efficiency of the cells was 88.5% ^ 0.014 (^ SE: n ¼ 31). Each data point was derived from the results of at least three independent experiments. 2.4. Measurement of cell growth during treatments The cells were harvested with trypsin/EDTA and re-suspended in an appropriate volume of the complete medium. The numbers of cells were counted with a hematocytometer 0, 12, 24 and 48 h after the start of treatment.
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
103
2.5. Cell cycle analysis The cells were processed in the same way as for clonogenic survival assay. After reaching a specified time, the cells were harvested with trypsin/EDTA, washed twice in phosphate-buffered saline (PBS) and finally re-suspended in 50 mg/ml of propidium iodide and 50 mg/ml of RNace. Flow cytometry was done with FACScan (Becton Dickinson, San Jose, CA, USA) and cell cycle distribution analysis with ModFit software on FACScan. The proportion of cells in the G1, S and G2 –M phases of the cell cycle were recorded immediately before and 12, 24 and 48 h after the treatments.
3. Results 3.1. Clonogenic survival assay Fig. 1 shows the effect of different doses of caffeine on A549 cells. Caffeine up to 5 mM had only a weak effect on survival for 48 h, but 20 mM caused a decrease in clonogenic survival in a time-course experiment. Fig. 2 shows the dose responses of paclitaxel alone and paclitaxel plus several doses of caffeine for 24 h. The data points in combined treatment were normalized by the surviving fractions due to caffeine alone. For paclitaxel alone, the cell killing effect increased in a dose-dependent manner up to 50 nM, but the
Fig. 1. Time– response to several doses of caffeine in A549 cells. A: caffeine 1.0 mM, B: caffeine 5.0 mM, D: caffeine 20 mM. Bar shows ^SE, where these exceed the size of the symbol.
Fig. 2. Dose–response of A549 to paclitaxel alone, paclitaxel with 1, 5 , 20 mM of caffeine for 24 h. W, paclitaxel alone, X, paclitaxel þ caffeine 1.0 mM, A, paclitaxel þ caffeine 5.0 mM, B, paclitaxel þ caffeine 20 mM. Bar shows ^SE, where these exceed the size of the symbol.
surviving fraction at 100 nM was almost the same as that at 50 nM. For combined treatments, although the added caffeine dose at 1 mM made no difference from paclitaxel alone, 5 mM of caffeine clearly reduced the cytotoxic activity of paclitaxel, and the effect of paclitaxel was remarkably suppressed when 20 mM of caffeine was added. Fig. 3 shows the time responses of the cells exposed to a 20 nM dose of paclitaxel ^ caffeine at 1– 20 mM for up to 48 h. Paclitaxel reduced the survival of cells up to 24 h, but no additional cell
Fig. 3. Time–response of A549 to 20 nM of paclitaxel alone and 20 nM of paclitaxel with 1, 5 and 20 mM of caffeine. X, paclitaxel 20 nM alone, A, paclitaxel 20 nM þ caffeine 1.0 mM, B, paclitaxel 20 nM þ caffeine 5.0 mM, D, paclitaxel 20 nM þ caffeine 20 mM. Bar shows ^SE, where these exceed the size of the symbol.
104
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
Fig. 4. Average of the relative number of cells which were treated with 20 nM of paclitaxel alone and 20 nM of paclitaxel with 1, 5 and 20 mM of caffeine plotted against time at 0, 12, 24 and 48 h. Each number is normalized by each pretreatment cell count at 0 h.
killing effect was observed at 48 h. An added caffeine dose at 1 mM had no effect, but a caffeine dose at 5– 20 mM also suppressed the cytotoxic effect of paclitaxel in a dose-dependent manner in time-course curves. 3.2. Number of cells The relative numbers of cells exposed to several doses of caffeine alone, and 20 nM of paclitaxel with or without caffeine for 12, 24 and 48 h are summarized in Fig. 4. The cells treated with 1 mM caffeine grew exponentially, as well, as the cells in the control. Growth suppression was seen in the cells treated with 5 mM caffeine, and with the caffeine dose at 20 mM, cell growth was completely suppressed. The cells exposed to paclitaxel alone grew exponentially up to 24 h, but cell growth was suppressed at 48 h. For the combined effect of caffeine and paclitaxel, the 1 mM added caffeine had no effect on cell growth. But, a caffeine dose over 5 mM clearly
suppressed the cell proliferation in dose-dependent manner. 3.3. Cell cycle Cell cycle distribution results during the treatment with paclitaxel and caffeine are shown in Fig. 5. Pretreatment fractions of G0/G1, S and G2 – M phase were 54.4, 33.7 and 12.0%, respectively. In control cells, a persistent increase in G1 and a decrease in S and G2 – M were observed. For the effect of caffeine alone, a caffeine dose at 1– 5 mM caused early G1 accumulation in a dosedependent manner, which was accompanied by a parallel decrease in the G2, S population. High dose caffeine at 20 mM resulted in G2 – M accumulation with no G1 accumulation. Exposure to 20 nM of paclitaxel alone at 12 h caused G2 – M accumulation and a slight increase in the S phase, where the fractions of G1, S and G2 – M phase were 20.7, 40.7 and 38.5%, respectively. After 48 h exposure, these
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
105
Fig. 5. The cell cycle distribution of each treatment plotted against time at 0, 12, 24 and 48 h. X: G0/G1, O: S, A: G2 þ M.
fractions almost returned to the pretreatment fractions. As a result, the effect of paclitaxel on cell cycle distribution was characterized as an early increase in G2 – M and a decrease in G1. As for the effect of caffeine on paclitaxel, caffeine suppressed the effect of paclitaxel on cell cycle distribution, where a dosedependent early increase in G2 –M and a decrease in G1 were not clear. In a high dose combination of paclitaxel and caffeine, no early cell cycle change was seen, whereas delayed G2 – M accumulation was slightly observed.
4. Discussion Caffeine is known as a modifying agent for cytotoxic chemotherapeutic drugs. Caffeine is reported to increase the cytotoxic effect of DNA damaging agents, for instance, the action of cisplatin on EL4 lymphoma cells [13] and that of etoposide on Hela cervical carcinoma cells [14] were enhanced by caffeine. The cytotoxicity of caffeine in these systems
was related to its ability to circumvent the G2 –M arrest caused by these cytotoxic cancer drugs [13 – 15], that is, the caffeine – cytotoxic agent sequence leads successively to G2 arrest, followed by G2 exit, and finally apoptosis. Caffeine is believed to cause this progression by stimulating p34cdc2 kinase [14], a cell cycle regulatory protein that triggers the transition from G2 to M and that may be involved in the induction of apoptosis [17]. For paclitaxel, Saunders et al. reported that caffeine enhanced paclitaxel-induced apoptosis in human breast cancer cells, MCF-7 [15]. They considered that there is a similar mechanism for its enhanced effect. In recent studies, the status of p53 is reported to be an important factor in sensitivity, both in caffeine and in paclitaxel. For the mechanism of radio sensitization caused by caffeine, the enhanced effect was more emphasized in p53-deficient cells, which fail to be arrested at the G1 checkpoint, rather than in the cells containing functional p53 [18 – 22]. Yao et al. demonstrated that the inhibition of G2 arrest by caffeinemediated activation of p34cdc2 kinase resulted in the
106
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
selective sensitization of p53 deficient cells to radiation-induced apoptosis [23]. In relation to paclitaxel and p53, Cassinelli et al. reported that the acquisition of cisplatin resistance in an ovarian carcinoma cell line, IGROV-1, was associated with the mutation of p53 and collateral sensitivity to paclitaxel [24]. They suggested that loss of the p53dependent post-mitotic checkpoint resulted in a different time course of paclitaxel-induced cell death. Das et al. demonstrated that the presence or absence of the wild-type p53 does not make a statistically significant difference in the level of apoptotic cell death, but the maximum is attained at a lower drug concentration in the presence of p53 [25]. At the beginning of this study, we hypothesized that caffeine could enhance the cytotoxic activity of paclitaxel, because cells, treated with paclitaxel, accumulated in the G2 –M phase and caffeine had the ability to override the cell cycle checkpoint, in particular G2 – M. However, the present data indicate that more than 5 mM of caffeine reduces the cytotoxic activity of paclitaxel, not only in a dose-dependent, but also in a time-dependent manner. The growth suppression and cell cycle arrest were apparent in the cells treated with a caffeine dose over 5 mM. These may explain the reduction in paclitaxel-induced cytotoxicity, because there are reports indicating that paclitaxel-induced cytotoxicity is diminished by cell cycle arrest in different phases. Russel et al. showed that caffeine did not inhibit a G1 phase arrest after irradiation, but blocked cells in G1 phase and Qi et al. reported that more than 5 mM of caffeine could induce the G1 phase arrest on A549 cells [21,26]. 5Fluorouracil has been reported to inhibit both paclitaxel-induced mitotic arrest and apoptotic death by arresting cells in the G1 and S phases of the cell cycle [27], and cisplatin, also, blocked paclitaxelinduced mitotic arrest and apoptosis by arresting cells in the late S to early G2 phase [28]. Qi et al. demonstrated that the G1 phase arrest by caffeine appeared to be due to suppression of CDK2 activity and suggested that caffeine might inhibit CDK2 activity at multiple levels, because caffeine caused a reduction in the CDK2 protein level [26]. Although it is still unclear how caffeine brings about the effect, they consider that other cyclin-dependent kinase inhibitors, such as IFI27 (also known as p27) or
CDKN1C (also known as p57) might be involved [26]. In addition, caffeine has been shown to diminish the cytotoxic and cytostatic activity of aromatic, DNA-intercalating and DNA topoisomerase II inhibitors, such as adriamycin, ellipticine and mitoxantrone, because of the formation of complexes between caffeine and these aromatic molecules [29]. Tragonos et al. also suggested that formation of the complexes between caffeine and paclitaxel potentially reduced the pharmacological activity of paclitaxel, because paclitaxel has aromatic structures as substituents [29]. On this point, more investigation is required in order to confirm complex formation in solution. In conclusion, this study indicates that more than 5 mM of caffeine reduces the cytotoxic activity of paclitaxel, not only in a dose-dependent, but also in a time-dependent manner. We suggest the that cell cycle modifying agents such as caffeine potentially diminish the cytotoxic activity of paclitaxel, so we should be very careful in using such combination therapy.
Acknowledgements This work was supported in part by Grants-in-Aid for Encouragement of Young Scientist, for scientific research and for scientific research on priority areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
References [1] P.B. Schiff, J. Fant, S.B. Horwitz, Promotions of microtubules assembly in vitro by Taxol, Nature 277 (1979) 665– 667. [2] J.J. Manfredi, S.B. Horwitz, Taxol: an animitotic agent with a unique mechanism of action, Pharmacol. Ther. 25 (1984) 83. [3] S.B. Horwitz, Taxol (paclitaxel): mechanism of action, Ann. Oncol. 5 (suppl. 6) (1994) S3–S6. [4] E.K. Rowinsky, R.C. Donehower, Paclitaxel (Taxol), N. Eng. J. Med. 332 (1995) 1004–1014. [5] O.P. Isokangas, H. Joensuu, M. Halme, A. Jekunen, K. Mattson, Paclitaxel (Taxol) and carboplatin followed by concomitant paclitaxel, cisplatin and radiotherapy for inoperable stage III NSCLC, Lung Cancer 20 (1998) 127 –133. [6] R.K. Gregory, M.E. Hill, J. Moore, R.P. A’Hern, S.R. Johnston, P. Blake, J. Shephard, D. Barton, M.E. Gore, Combining platinum, paclitaxel and anthracycline in patients
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
[7]
[8]
[9]
[10]
[11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
with advanced gynaecological malignancy, Eur. J. Cancer 36 (2000) 503–507. J.A. Kramer, D. Curran, M. Piccart, J.C. de Haes, P.F. Bruning, J.G. Klijin, M. Bontenbal, C. van Pottelsberghe, M. Groenvold, R. Paridaens, Randmised trial of paclitaxel versus doxorubicin as first-line chemotherapy for advanced breast cancer: quality of life evaluation using the EORTC QLQ-C30 and the Rotterdam symptom checklist, Eur. J. Cancer 36 (2000) 1488–1497. D. Mavroudis, E. Papadakis, M. Veslemes, X. Tsiafaki, J. Stavrakakis, C. Kouroussis, S. Kakolyris, E. Bania, J. Jordanoglou, M. Agelidou, J. Vlachonicolis, V. Georgoulias, Greek Lung Cancer Cooperative Group, A multicenter randomized clinical trial comparing paclitaxel–cisplatin– etoposide versus cisplatin–etoposide as first-line treatment in patients with small-cell lung cancer, Ann. Oncol. 12 (2001) 463–470. F.J. O’Neil, Differential effects of cytochalasin B and caffeine on control of DNA synthesis in normal and transformed cell, J. Cell Physiol. 101 (1979) 201 –218. G. Iliakis, M. Nusse, Effects of X-irradiated synchronous, asynchronous and plateau phase mouse ascites cells: the importance of progression through the cell cycle for caffeine enhancement of killing, Int. J. Radiat. Biol. 43 (1983) 649–663. L.J. Tolmach, R.W. Jones, P.M. Busse, The action of caffeine on X-irradiated HeLa cells, Radiat. Res. 71 (1977) 653 –665. C.P. Selby, A. Sancar, Molecular mechanisms of DNA repair inhibition by caffeine, Proc. Natl. Acad. Sci. USA 87 (1990) 3522–3525. N. Shinomiya, M. Shinomiya, H. Wakiyama, Y. Katsura, M. Rokutanda, Enhancement of CDDP cytotoxicity by caffeine is characterized by apoptotic cell death, Exp. Cell Res. 210 (1994) 236–242. R.B. Lock, O.V. Galperina, R.C. Feldhoff, L.J. Rhodes, Concentration-dependent differences in the mechanism by which caffeine potentiates etoposide cytotoxicity in Hela cells, Cancer Res. 54 (1994) 4933–4939. D.E. Saunders, W.D. Lawrence, C. Christensen, N.L. Wappler, H. Ruan, G. Deppe, Paclitaxel-induced apoptosis in MCF-7 breast-cancer cells, Int. J. Cancer 70 (1997) 214–220. J.J. Robert, Mechanism of potentiation by caffeine of genotoxic damage induced by physical and chemical agents, in: A. Collins, C.S. Downes, R.T. Johnson (Eds.), DNA Repair and its Inhibitors, IRL Press Ltd, Oxford, 1984, pp. 193–216. L. Shi, W. Nishioka, J. Thng, E. Bradbury, D. Litchfield, A. Greenberg, Premature p34cdc2 activation required for apoptosis, Science 263 (1994) 1143–1145. T. Akimoto, N. Mitsuhashi, H. Matsumoto, H. Sakurai, K. Maebayashi, K. Higuchi, M. Nozaki, H. Niibe, Potentially lethal damage repair and its inhibitory effect of caffeine in two yolk sac tumor cell lines with different radiosensitivities, Cancer Lett. 147 (1999) 199–206.
107
[19] K. Higuchi, N. Mitsuhashi, Jun-ichi Saitoh, K. Maebayashi, H. Sakurai, T. Akimoto, H. Niibe, Caffeine enhanced radiosensitivity of rat tumor cells with a mutant-type p53 by inducing apoptosis in a p53-independent manner, Cancer Lett. 152 (2000) 157–162. [20] S.N. Powell, J.S. DeFrank, P. Connell, M. Eogan, F. Preffer, D. Dombkowski, W. Tang, S. Friend, Differential sensitivity of p53(2 ) and p53(þ ) cells to caffeine-induced radiosensitization and override of G2 delay, Cancer Res. 55 (1995) 1643–1648. [21] K.J. Russel, L.W. Wiens, G.W. Demers, D.A. Galloway, S.E. Plon, M. Groudine, Abrogation of the G2 checkpoint results in differential radiosensitization of G1 checkpoint-deficient and G1 checkpoint-competent cells, Cancer Res. 55 (1995) 1639–1642. [22] K.J. Russel, L.W. Wiens, G.W. Demers, D.A. Galloway, T. Le, G.C. Rice, J.A. Bianco, J.W. Singer, M. Groudine, Preferential radiosensitization of G1 checkpoint-deficient cells by methylxanthines, Int. J. Radiat. Oncol. Biol. Phys. 36 (1996) 1099–1106. [23] S.L. Yao, A.J. Akhtar, K.A. McKenna, G.C. Bedi, D. Sidransky, M. Mabry, R. Ravi, M.I. Collector, R.J. Jones, S.J. Sharkis, E.J. Fuchs, A. Bedi, Selective radiosensitization of p53-deficient cells by caffeine-mediated activation of p34cdc2 kinase, Nat. Med. 2 (1996) 1140–1143. [24] G. Cassinelli, R. Supino, P. Perego, D. Polizzi, C. Lanzi, G. Pratesi, F. Zunino, A role for loss of p53 function in sensitivity of ovarian carcinoma cells to taxanes, Int. J. Cancer 92 (2001) 738 –747. [25] G.C. Das, D. Holiday, R. Gallardo, C. Haas, Taxol-induced cell cycle arrest and apoptosis: dose–response relationship in lung cancer cells of different wild-type p53 status and under isogenic condition, Cancer Lett. 165 (2001) 147– 153. [26] W. Qi, D. Qiao, J.D. Martinez, Caffeine induces TP53independent G1-phase arrest and apoptosis in human lung tumor cells in a dose-dependent manner, Radiat. Res. 157 (2002) 166 –174. [27] K.R. Johnson, L.M. Wang, M.C. Miller, M.C. Willingham, W.M. Fan, 5-Fluorouracil interferes with paclitaxel cytotoxicity against human solid cancer cells, Clin. Cancer Res. 3 (1997) 1729–1745. [28] P.L. Judson, J.M. Watson, P.A. Gehrig, C. Fowler Jr., J.S Haskill, Cisplatin inhibits paclitaxel-induced apoptosis in cisplatin-resistant ovarian cancer cell lines: possible explanation for failure of combination therapy, Cancer Res. 59 (1999) 2425–2432. [29] F. Tragonas, J. Kapuscinski, Z. Darzynkiewicz, Caffeine modulates the effects of DNA-intercalating drugs in vitro: a flow cytometric and spectrophotometric analysis of caffeine interaction with Novantrone, doxorubicin, ellipticine, and the doxorubicin analogue AD198, Cancer Res. 51 (1991) 3682–3689.
Caffeine diminishes cytotoxic effects of paclitaxel on a human lung adenocarcinoma cell line Yoshizumi Kitamotoa,*, Hideyuki Sakuraia, Norio Mitsuhashib, Tetsuo Akimotoa, Takashi Nakanoa a
Department of Radiology and Radiation Oncology, Gunma University School of Medicine, 3-39-22, Showa-machi, Maebashi, Gunma 371-8511, Japan b Department of Radiology, Tokyo Women’s Medical University, 8-1, Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan Received 6 August 2002; received in revised form 30 September 2002; accepted 7 October 2002
Abstract This study was performed to investigate how caffeine modifies the cytotoxic effects of paclitaxel on a human lung carcinoma cell line. Caffeine doses up to 5 mM had less effect on clonogenic survival. The cell killing effect, due to paclitaxel, increased in a dose-dependent manner up to 50 nM. For combined treatment with caffeine and paclitaxel, added caffeine reduced the cytotoxic effect of paclitaxel not only in dose – response but also in time – response curves. Caffeine combined with paclitaxel clearly suppressed cell proliferation in a dose-dependent manner. In the cell cycle analysis, caffeine alone caused early G1 accumulation, whereas paclitaxel alone caused an early increase in G2-M and a decrease in G1. As for the effect of caffeine on paclitaxel, caffeine suppressed the effect of paclitaxel on cell cycle distribution, where a dose-dependent early increase in G2-M and a decrease in G1 were not clear. We suggest that cell cycle modifying agents, such as caffeine, potentially diminish the cytotoxic activity of paclitaxel, and one should be careful when combining such agents. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Paclitaxel; Caffeine; Cell cycle; Human lung carcinoma
1. Introduction Paclitaxel is a, naturally occurring, chemotherapeutic agent isolated from the bark of the Western yew [1]. The cytotoxic mechanism of paclitaxel has been found to be a potent microtubule stabilizing agent and a promoter of microtubule assembly, and to form microtubule bundles within cells [1 – 3]. It * Corresponding author. Tel.: þ81-27-220-8383; fax: þ 81-27220-8397. E-mail address: [email protected] (Y. Kitamoto).
inhibits normal mitosis by binding to the N-terminal 31 amino acid of the b-tubulin subunit of microtubules [4] and induces a G2 – M arrest in cycling cells. Paclitaxel has, recently, gained clinical attention as a chemotherapeutic agent for some advanced neoplasms e.g. lung cancer, breast and ovarian cancers, and paclitaxel alone or combination chemotherapy including paclitaxel have been tested in many clinical trials [5 – 8]. Methylxanthine, for instance, caffeine or pentoxifylline has a multiplicity of effects on cells in culture. Caffeine causes G1 phase accumulation [9,10] due to
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. doi: 1 0 . 1 0 1 6 / S 0 3 0 4 - 3 8 3 5 ( 0 2 ) 0 0 5 9 1 - 8
102
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
inhibition of enzymes required for DNA synthesis [11,12]. Caffeine is also known as a modifying agent in combination with many cytotoxic drugs. It has been reported that caffeine enhances the cell killing of many DNA-damaging agents, presumably by shortening the time of cell cycle arrest normally caused by such agents [13 –16], that is, the caffeine-cytotoxic agent sequence leads successively to G2 arrest, followed by G2 exit, and finally apoptosis. Caffeine is believed to cause this progression by stimulating p34cdc2 kinase [14], a cell cycle regulatory protein, that triggers the transition from G2 to M, and which may be involved in induction of apoptosis [17]. For combined effect of paclitaxel and caffeine, Saunders et al. reported, that caffeine enhanced development of apoptosis is induced by paclitaxel in a human breast cancer cell line, MCF-7 [15]. In this paper, the cellular response to simultaneously combined paclitaxel and caffeine is reported and an effort is made to clarify the mechanism of the modification of this simultaneous insult.
2. Materials and methods 2.1. Cell line and culture A human lung adenocarcinoma cell line, A549, was obtained from Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University (Miyagi, Japan). In our laboratory, polymerase chain reaction single-strand conformation polymorphism (PCR –SSCP) analysis revealed that A549 has a wild-type p53. The cells were maintained in RPMI-1640 medium supplemented with 10% fetal calf serum, antibiotics and 0.3% HEPES (Immuno Biological Laboratories Ltd, Gunma Japan). The cells have an apparent doubling time of 24 h when growing exponentially over the concentration range of 1 – 5 £ 105 cells/ml. Cell suspensions were prepared from stock culture flasks and incubated in 25 cm2 cell culture flasks at 5 £ 105 cells in 5 ml complete medium. Cells in the flask were incubated at 378C, in a humidified atmosphere of 5% CO2 and 95% air, for 48 h prior to treatment. In this condition, growth was exponential for up to 4 days after the inoculation.
2.2. Reagents Paclitaxel was purchased from Wako Pure Chemical Industries, Ltd (Osaka, Japan), dissolved in dimethyl sulfoxide (DMSO) at a concentration of 500 mM and stocked at 2 208C. Caffeine (1,3,7trimethylxanthine) was obtained from Wako Pure Chemical Industries, Ltd (Osaka, Japan). The agent was dissolved in complete medium with heating at 708C for 30 min immediately before the treatment. The solution was dissolved in an appropriate concentration with complete medium. For combined treatment, the cells were exposed to 1 –100 nM paclitaxel simultaneously combined with 1 –20 mM caffeine. To estimate the dose response, the cells were incubated in these solutions at 378C for 24 h. In a time course experiment, cells were exposed to 20 nM of paclitaxel ^ 1– 20 mM caffeine at 378C for up to 48 h. 2.3. Clonogenic survival assay After the drug treatment, a single cell suspension was obtained by means of trypsin/ethylenediaminetetraaceticacid (EDTA) detachment. The cells were pelleted by centrifugation at 1000 rpm for 5 min, resuspended in complete medium, counted with a hemocytometer, serially diluted and plated in dishes in proportion to the expected survival. Three dishes with each cell inoculum were used for each treatment point. Dishes were kept at 378C in an incubator for approximately 14 days to allow colony formation. Colonies were fixed with methanol, stained with crystal violet solution and those of over 50 cells were counted as survivors. The surviving fraction was determined by the proportion of seeded cells surviving relative to untreated cells, after the treatments to form colonies. Plating efficiency of the cells was 88.5% ^ 0.014 (^ SE: n ¼ 31). Each data point was derived from the results of at least three independent experiments. 2.4. Measurement of cell growth during treatments The cells were harvested with trypsin/EDTA and re-suspended in an appropriate volume of the complete medium. The numbers of cells were counted with a hematocytometer 0, 12, 24 and 48 h after the start of treatment.
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
103
2.5. Cell cycle analysis The cells were processed in the same way as for clonogenic survival assay. After reaching a specified time, the cells were harvested with trypsin/EDTA, washed twice in phosphate-buffered saline (PBS) and finally re-suspended in 50 mg/ml of propidium iodide and 50 mg/ml of RNace. Flow cytometry was done with FACScan (Becton Dickinson, San Jose, CA, USA) and cell cycle distribution analysis with ModFit software on FACScan. The proportion of cells in the G1, S and G2 –M phases of the cell cycle were recorded immediately before and 12, 24 and 48 h after the treatments.
3. Results 3.1. Clonogenic survival assay Fig. 1 shows the effect of different doses of caffeine on A549 cells. Caffeine up to 5 mM had only a weak effect on survival for 48 h, but 20 mM caused a decrease in clonogenic survival in a time-course experiment. Fig. 2 shows the dose responses of paclitaxel alone and paclitaxel plus several doses of caffeine for 24 h. The data points in combined treatment were normalized by the surviving fractions due to caffeine alone. For paclitaxel alone, the cell killing effect increased in a dose-dependent manner up to 50 nM, but the
Fig. 1. Time– response to several doses of caffeine in A549 cells. A: caffeine 1.0 mM, B: caffeine 5.0 mM, D: caffeine 20 mM. Bar shows ^SE, where these exceed the size of the symbol.
Fig. 2. Dose–response of A549 to paclitaxel alone, paclitaxel with 1, 5 , 20 mM of caffeine for 24 h. W, paclitaxel alone, X, paclitaxel þ caffeine 1.0 mM, A, paclitaxel þ caffeine 5.0 mM, B, paclitaxel þ caffeine 20 mM. Bar shows ^SE, where these exceed the size of the symbol.
surviving fraction at 100 nM was almost the same as that at 50 nM. For combined treatments, although the added caffeine dose at 1 mM made no difference from paclitaxel alone, 5 mM of caffeine clearly reduced the cytotoxic activity of paclitaxel, and the effect of paclitaxel was remarkably suppressed when 20 mM of caffeine was added. Fig. 3 shows the time responses of the cells exposed to a 20 nM dose of paclitaxel ^ caffeine at 1– 20 mM for up to 48 h. Paclitaxel reduced the survival of cells up to 24 h, but no additional cell
Fig. 3. Time–response of A549 to 20 nM of paclitaxel alone and 20 nM of paclitaxel with 1, 5 and 20 mM of caffeine. X, paclitaxel 20 nM alone, A, paclitaxel 20 nM þ caffeine 1.0 mM, B, paclitaxel 20 nM þ caffeine 5.0 mM, D, paclitaxel 20 nM þ caffeine 20 mM. Bar shows ^SE, where these exceed the size of the symbol.
104
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
Fig. 4. Average of the relative number of cells which were treated with 20 nM of paclitaxel alone and 20 nM of paclitaxel with 1, 5 and 20 mM of caffeine plotted against time at 0, 12, 24 and 48 h. Each number is normalized by each pretreatment cell count at 0 h.
killing effect was observed at 48 h. An added caffeine dose at 1 mM had no effect, but a caffeine dose at 5– 20 mM also suppressed the cytotoxic effect of paclitaxel in a dose-dependent manner in time-course curves. 3.2. Number of cells The relative numbers of cells exposed to several doses of caffeine alone, and 20 nM of paclitaxel with or without caffeine for 12, 24 and 48 h are summarized in Fig. 4. The cells treated with 1 mM caffeine grew exponentially, as well, as the cells in the control. Growth suppression was seen in the cells treated with 5 mM caffeine, and with the caffeine dose at 20 mM, cell growth was completely suppressed. The cells exposed to paclitaxel alone grew exponentially up to 24 h, but cell growth was suppressed at 48 h. For the combined effect of caffeine and paclitaxel, the 1 mM added caffeine had no effect on cell growth. But, a caffeine dose over 5 mM clearly
suppressed the cell proliferation in dose-dependent manner. 3.3. Cell cycle Cell cycle distribution results during the treatment with paclitaxel and caffeine are shown in Fig. 5. Pretreatment fractions of G0/G1, S and G2 – M phase were 54.4, 33.7 and 12.0%, respectively. In control cells, a persistent increase in G1 and a decrease in S and G2 – M were observed. For the effect of caffeine alone, a caffeine dose at 1– 5 mM caused early G1 accumulation in a dosedependent manner, which was accompanied by a parallel decrease in the G2, S population. High dose caffeine at 20 mM resulted in G2 – M accumulation with no G1 accumulation. Exposure to 20 nM of paclitaxel alone at 12 h caused G2 – M accumulation and a slight increase in the S phase, where the fractions of G1, S and G2 – M phase were 20.7, 40.7 and 38.5%, respectively. After 48 h exposure, these
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
105
Fig. 5. The cell cycle distribution of each treatment plotted against time at 0, 12, 24 and 48 h. X: G0/G1, O: S, A: G2 þ M.
fractions almost returned to the pretreatment fractions. As a result, the effect of paclitaxel on cell cycle distribution was characterized as an early increase in G2 – M and a decrease in G1. As for the effect of caffeine on paclitaxel, caffeine suppressed the effect of paclitaxel on cell cycle distribution, where a dosedependent early increase in G2 –M and a decrease in G1 were not clear. In a high dose combination of paclitaxel and caffeine, no early cell cycle change was seen, whereas delayed G2 – M accumulation was slightly observed.
4. Discussion Caffeine is known as a modifying agent for cytotoxic chemotherapeutic drugs. Caffeine is reported to increase the cytotoxic effect of DNA damaging agents, for instance, the action of cisplatin on EL4 lymphoma cells [13] and that of etoposide on Hela cervical carcinoma cells [14] were enhanced by caffeine. The cytotoxicity of caffeine in these systems
was related to its ability to circumvent the G2 –M arrest caused by these cytotoxic cancer drugs [13 – 15], that is, the caffeine – cytotoxic agent sequence leads successively to G2 arrest, followed by G2 exit, and finally apoptosis. Caffeine is believed to cause this progression by stimulating p34cdc2 kinase [14], a cell cycle regulatory protein that triggers the transition from G2 to M and that may be involved in the induction of apoptosis [17]. For paclitaxel, Saunders et al. reported that caffeine enhanced paclitaxel-induced apoptosis in human breast cancer cells, MCF-7 [15]. They considered that there is a similar mechanism for its enhanced effect. In recent studies, the status of p53 is reported to be an important factor in sensitivity, both in caffeine and in paclitaxel. For the mechanism of radio sensitization caused by caffeine, the enhanced effect was more emphasized in p53-deficient cells, which fail to be arrested at the G1 checkpoint, rather than in the cells containing functional p53 [18 – 22]. Yao et al. demonstrated that the inhibition of G2 arrest by caffeinemediated activation of p34cdc2 kinase resulted in the
106
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
selective sensitization of p53 deficient cells to radiation-induced apoptosis [23]. In relation to paclitaxel and p53, Cassinelli et al. reported that the acquisition of cisplatin resistance in an ovarian carcinoma cell line, IGROV-1, was associated with the mutation of p53 and collateral sensitivity to paclitaxel [24]. They suggested that loss of the p53dependent post-mitotic checkpoint resulted in a different time course of paclitaxel-induced cell death. Das et al. demonstrated that the presence or absence of the wild-type p53 does not make a statistically significant difference in the level of apoptotic cell death, but the maximum is attained at a lower drug concentration in the presence of p53 [25]. At the beginning of this study, we hypothesized that caffeine could enhance the cytotoxic activity of paclitaxel, because cells, treated with paclitaxel, accumulated in the G2 –M phase and caffeine had the ability to override the cell cycle checkpoint, in particular G2 – M. However, the present data indicate that more than 5 mM of caffeine reduces the cytotoxic activity of paclitaxel, not only in a dose-dependent, but also in a time-dependent manner. The growth suppression and cell cycle arrest were apparent in the cells treated with a caffeine dose over 5 mM. These may explain the reduction in paclitaxel-induced cytotoxicity, because there are reports indicating that paclitaxel-induced cytotoxicity is diminished by cell cycle arrest in different phases. Russel et al. showed that caffeine did not inhibit a G1 phase arrest after irradiation, but blocked cells in G1 phase and Qi et al. reported that more than 5 mM of caffeine could induce the G1 phase arrest on A549 cells [21,26]. 5Fluorouracil has been reported to inhibit both paclitaxel-induced mitotic arrest and apoptotic death by arresting cells in the G1 and S phases of the cell cycle [27], and cisplatin, also, blocked paclitaxelinduced mitotic arrest and apoptosis by arresting cells in the late S to early G2 phase [28]. Qi et al. demonstrated that the G1 phase arrest by caffeine appeared to be due to suppression of CDK2 activity and suggested that caffeine might inhibit CDK2 activity at multiple levels, because caffeine caused a reduction in the CDK2 protein level [26]. Although it is still unclear how caffeine brings about the effect, they consider that other cyclin-dependent kinase inhibitors, such as IFI27 (also known as p27) or
CDKN1C (also known as p57) might be involved [26]. In addition, caffeine has been shown to diminish the cytotoxic and cytostatic activity of aromatic, DNA-intercalating and DNA topoisomerase II inhibitors, such as adriamycin, ellipticine and mitoxantrone, because of the formation of complexes between caffeine and these aromatic molecules [29]. Tragonos et al. also suggested that formation of the complexes between caffeine and paclitaxel potentially reduced the pharmacological activity of paclitaxel, because paclitaxel has aromatic structures as substituents [29]. On this point, more investigation is required in order to confirm complex formation in solution. In conclusion, this study indicates that more than 5 mM of caffeine reduces the cytotoxic activity of paclitaxel, not only in a dose-dependent, but also in a time-dependent manner. We suggest the that cell cycle modifying agents such as caffeine potentially diminish the cytotoxic activity of paclitaxel, so we should be very careful in using such combination therapy.
Acknowledgements This work was supported in part by Grants-in-Aid for Encouragement of Young Scientist, for scientific research and for scientific research on priority areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
References [1] P.B. Schiff, J. Fant, S.B. Horwitz, Promotions of microtubules assembly in vitro by Taxol, Nature 277 (1979) 665– 667. [2] J.J. Manfredi, S.B. Horwitz, Taxol: an animitotic agent with a unique mechanism of action, Pharmacol. Ther. 25 (1984) 83. [3] S.B. Horwitz, Taxol (paclitaxel): mechanism of action, Ann. Oncol. 5 (suppl. 6) (1994) S3–S6. [4] E.K. Rowinsky, R.C. Donehower, Paclitaxel (Taxol), N. Eng. J. Med. 332 (1995) 1004–1014. [5] O.P. Isokangas, H. Joensuu, M. Halme, A. Jekunen, K. Mattson, Paclitaxel (Taxol) and carboplatin followed by concomitant paclitaxel, cisplatin and radiotherapy for inoperable stage III NSCLC, Lung Cancer 20 (1998) 127 –133. [6] R.K. Gregory, M.E. Hill, J. Moore, R.P. A’Hern, S.R. Johnston, P. Blake, J. Shephard, D. Barton, M.E. Gore, Combining platinum, paclitaxel and anthracycline in patients
Y. Kitamoto et al. / Cancer Letters 191 (2003) 101–107
[7]
[8]
[9]
[10]
[11] [12]
[13]
[14]
[15]
[16]
[17]
[18]
with advanced gynaecological malignancy, Eur. J. Cancer 36 (2000) 503–507. J.A. Kramer, D. Curran, M. Piccart, J.C. de Haes, P.F. Bruning, J.G. Klijin, M. Bontenbal, C. van Pottelsberghe, M. Groenvold, R. Paridaens, Randmised trial of paclitaxel versus doxorubicin as first-line chemotherapy for advanced breast cancer: quality of life evaluation using the EORTC QLQ-C30 and the Rotterdam symptom checklist, Eur. J. Cancer 36 (2000) 1488–1497. D. Mavroudis, E. Papadakis, M. Veslemes, X. Tsiafaki, J. Stavrakakis, C. Kouroussis, S. Kakolyris, E. Bania, J. Jordanoglou, M. Agelidou, J. Vlachonicolis, V. Georgoulias, Greek Lung Cancer Cooperative Group, A multicenter randomized clinical trial comparing paclitaxel–cisplatin– etoposide versus cisplatin–etoposide as first-line treatment in patients with small-cell lung cancer, Ann. Oncol. 12 (2001) 463–470. F.J. O’Neil, Differential effects of cytochalasin B and caffeine on control of DNA synthesis in normal and transformed cell, J. Cell Physiol. 101 (1979) 201 –218. G. Iliakis, M. Nusse, Effects of X-irradiated synchronous, asynchronous and plateau phase mouse ascites cells: the importance of progression through the cell cycle for caffeine enhancement of killing, Int. J. Radiat. Biol. 43 (1983) 649–663. L.J. Tolmach, R.W. Jones, P.M. Busse, The action of caffeine on X-irradiated HeLa cells, Radiat. Res. 71 (1977) 653 –665. C.P. Selby, A. Sancar, Molecular mechanisms of DNA repair inhibition by caffeine, Proc. Natl. Acad. Sci. USA 87 (1990) 3522–3525. N. Shinomiya, M. Shinomiya, H. Wakiyama, Y. Katsura, M. Rokutanda, Enhancement of CDDP cytotoxicity by caffeine is characterized by apoptotic cell death, Exp. Cell Res. 210 (1994) 236–242. R.B. Lock, O.V. Galperina, R.C. Feldhoff, L.J. Rhodes, Concentration-dependent differences in the mechanism by which caffeine potentiates etoposide cytotoxicity in Hela cells, Cancer Res. 54 (1994) 4933–4939. D.E. Saunders, W.D. Lawrence, C. Christensen, N.L. Wappler, H. Ruan, G. Deppe, Paclitaxel-induced apoptosis in MCF-7 breast-cancer cells, Int. J. Cancer 70 (1997) 214–220. J.J. Robert, Mechanism of potentiation by caffeine of genotoxic damage induced by physical and chemical agents, in: A. Collins, C.S. Downes, R.T. Johnson (Eds.), DNA Repair and its Inhibitors, IRL Press Ltd, Oxford, 1984, pp. 193–216. L. Shi, W. Nishioka, J. Thng, E. Bradbury, D. Litchfield, A. Greenberg, Premature p34cdc2 activation required for apoptosis, Science 263 (1994) 1143–1145. T. Akimoto, N. Mitsuhashi, H. Matsumoto, H. Sakurai, K. Maebayashi, K. Higuchi, M. Nozaki, H. Niibe, Potentially lethal damage repair and its inhibitory effect of caffeine in two yolk sac tumor cell lines with different radiosensitivities, Cancer Lett. 147 (1999) 199–206.
107
[19] K. Higuchi, N. Mitsuhashi, Jun-ichi Saitoh, K. Maebayashi, H. Sakurai, T. Akimoto, H. Niibe, Caffeine enhanced radiosensitivity of rat tumor cells with a mutant-type p53 by inducing apoptosis in a p53-independent manner, Cancer Lett. 152 (2000) 157–162. [20] S.N. Powell, J.S. DeFrank, P. Connell, M. Eogan, F. Preffer, D. Dombkowski, W. Tang, S. Friend, Differential sensitivity of p53(2 ) and p53(þ ) cells to caffeine-induced radiosensitization and override of G2 delay, Cancer Res. 55 (1995) 1643–1648. [21] K.J. Russel, L.W. Wiens, G.W. Demers, D.A. Galloway, S.E. Plon, M. Groudine, Abrogation of the G2 checkpoint results in differential radiosensitization of G1 checkpoint-deficient and G1 checkpoint-competent cells, Cancer Res. 55 (1995) 1639–1642. [22] K.J. Russel, L.W. Wiens, G.W. Demers, D.A. Galloway, T. Le, G.C. Rice, J.A. Bianco, J.W. Singer, M. Groudine, Preferential radiosensitization of G1 checkpoint-deficient cells by methylxanthines, Int. J. Radiat. Oncol. Biol. Phys. 36 (1996) 1099–1106. [23] S.L. Yao, A.J. Akhtar, K.A. McKenna, G.C. Bedi, D. Sidransky, M. Mabry, R. Ravi, M.I. Collector, R.J. Jones, S.J. Sharkis, E.J. Fuchs, A. Bedi, Selective radiosensitization of p53-deficient cells by caffeine-mediated activation of p34cdc2 kinase, Nat. Med. 2 (1996) 1140–1143. [24] G. Cassinelli, R. Supino, P. Perego, D. Polizzi, C. Lanzi, G. Pratesi, F. Zunino, A role for loss of p53 function in sensitivity of ovarian carcinoma cells to taxanes, Int. J. Cancer 92 (2001) 738 –747. [25] G.C. Das, D. Holiday, R. Gallardo, C. Haas, Taxol-induced cell cycle arrest and apoptosis: dose–response relationship in lung cancer cells of different wild-type p53 status and under isogenic condition, Cancer Lett. 165 (2001) 147– 153. [26] W. Qi, D. Qiao, J.D. Martinez, Caffeine induces TP53independent G1-phase arrest and apoptosis in human lung tumor cells in a dose-dependent manner, Radiat. Res. 157 (2002) 166 –174. [27] K.R. Johnson, L.M. Wang, M.C. Miller, M.C. Willingham, W.M. Fan, 5-Fluorouracil interferes with paclitaxel cytotoxicity against human solid cancer cells, Clin. Cancer Res. 3 (1997) 1729–1745. [28] P.L. Judson, J.M. Watson, P.A. Gehrig, C. Fowler Jr., J.S Haskill, Cisplatin inhibits paclitaxel-induced apoptosis in cisplatin-resistant ovarian cancer cell lines: possible explanation for failure of combination therapy, Cancer Res. 59 (1999) 2425–2432. [29] F. Tragonas, J. Kapuscinski, Z. Darzynkiewicz, Caffeine modulates the effects of DNA-intercalating drugs in vitro: a flow cytometric and spectrophotometric analysis of caffeine interaction with Novantrone, doxorubicin, ellipticine, and the doxorubicin analogue AD198, Cancer Res. 51 (1991) 3682–3689.