due to a Combination-specific Mechanism in HL60 ...

Simultaneous Treatment with 1-β-d-Arabinofuranosylcytosine and Daunorubicin Induces Cross-Resistance to Both Drugs due to a Combination-specific Mechanism in HL60 Cells Haruyuki Takemura, Yoshimasa Urasaki, Akira Yoshida, et al. Cancer Res 2001;61:172-177.

Updated version

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/61/1/172

Cited Articles

This article cites by 39 articles, 20 of which you can access for free at: http://cancerres.aacrjournals.org/content/61/1/172.full.html#ref-list-1

Citing articles

This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/61/1/172.full.html#related-urls

E-mail alerts Reprints and Subscriptions Permissions

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].

Downloaded from cancerres.aacrjournals.org on June 12, 2013. © 2001 American Association for Cancer Research.

[CANCER RESEARCH 61, 172–177, January 1, 2001]

Simultaneous Treatment with 1-␤-D-Arabinofuranosylcytosine and Daunorubicin Induces Cross-Resistance to Both Drugs due to a Combination-specific Mechanism in HL60 Cells1 Haruyuki Takemura, Yoshimasa Urasaki, Akira Yoshida, Toshihiro Fukushima, and Takanori Ueda2 First Department of Internal Medicine, Fukui Medical University, Fukui 910-1193, Japan

ABSTRACT We have established a human myelogenous leukemia cell line (HL60/ AD) that is 10-fold cross-resistant to both 1-␤-D-arabinofuranosylcytosine (ara-C) and daunorubicin; the cell line was isolated from HL60 by simultaneous treatment with these two agents at low drug concentrations attainable in clinical trials. HL60/AD was found to have multiple resistance mechanisms. With regard to ara-C, HL60/AD cells showed decreased deoxycytidine kinase activity but did not show elevation of cytidine deaminase activity or a decrease in ara-C influx. With regard to daunorubicin, a decrease in topoisomerase II activity was found. A decrease in intracellular accumulation of daunorubicin was also found. P-glycoprotein was not detected, but the multidrug resistance-associated protein was expressed. Furthermore, an increase of total cellular glutathione (GSH) content was found. Interestingly, the resistance of HL60/AD cells not only to daunorubicin but also to ara-C was markedly reversed by treatment with L-buthionine-(S,R)-sulfoximine (BSO), a potent inhibitor of GSH synthesis. After exposure of HL60/AD to ara-C, mitochondrial membrane potential and reactive oxygen intermediates showed no significant change, but a considerable loss of mitochondrial membrane potential and an increase in reactive oxygen intermediate generation were caused by preincubation with BSO. Neither elevation of GSH nor reversal of resistance by BSO was found in ara-C-resistant HL60 cells that were selected only with ara-C. These findings suggest that in addition to the summation of the mechanisms of resistance to each agent reported previously, an increased level of GSH plays an important role in the cross-resistance induced in HL60/AD cells by simultaneous exposure to both drugs.

To date, investigators have reported on the mechanisms of action of ara-C (3) and daunorubicin (4) and elucidated the mechanisms of resistance to each agent, but cross-resistant cells induced by simultaneous treatment with ara-C and daunorubicin have not been studied until now. Furthermore, in many reports, the mechanisms of resistance were studied mainly in cell lines resistant to high drug levels, which do not reflect the conditions in most clinical trials (5). To elucidate the mechanisms of drug resistance under conditions reflecting those in clinical trials, it would be more relevant to establish cell lines resistant to low drug levels selected with drug concentrations attainable in vivo and study the mechanism of the resistance. In the present study, we established such a human myelogenous leukemia cell line (HL60/AD) resistant to both ara-C and daunorubicin, which was selected by simultaneous treatment with these drugs at low concentrations, and we examined the mechanisms of cross-resistance to these drugs. In particular, we focused on whether HL60/AD cells would acquire a new mechanism of resistance that differed from the mechanisms seen in HL60 cells selected with each single agent. MATERIALS AND METHODS

Drugs and Chemicals. Daunorubicin was kindly supplied by Meiji Seika Co. (Tokyo, Japan). Ara-C was obtained from Nippon Shinyaku Co. (Kyoto, Japan). [3H]Ara-C and NCS were purchased from Amersham Japan (Tokyo, Japan). [3H]Daunorubicin was obtained from Daiichi Pure Chemicals (Tokyo, Japan). 1-Chloro-2,4-dinitrobenzene, 5,5⬘-dithiobis-(2-nitrobenzoic acid), and BSO were obtained from Nacalai Tesque (Kyoto, Japan). GSH reductase from INTRODUCTION yeast, and DiOC6(3) were obtained from Sigma Chemical Co. (St. Louis, MO). MRPm6 was obtained from Kamiya Biomedical Co. (Seattle, WA). UIC2 was 3 Recently, remission induction rates of AML have gradually in- obtained from Immunotec (Marseille, France). NADPH and GSH (reduced form) creased, but curative treatment of AML is still a serious problem. One were obtained from Kohjin Co. (Tokyo, Japan). kDNA was obtained from Wako of the reasons for this is the development of multidrug resistance, Co. (Osaka, Japan). All other chemicals were usually obtained from commercial which is a major obstacle to successful treatment. Ara-C and dauno- sources. Cells and Drug Sensitivity. HL60/AD cells were isolated from a human rubicin are the most commonly used drugs for treatment of AML (1, 2). These two agents are commonly applied in combination to treat myelogenous leukemia cell line (HL60) by a series of stepwise selections with AML patients in clinical trials and have proved effective. However, simultaneous treatment of ara-C and daunorubicin. Resistant cells were cloned by the limiting dilution method. Drug sensitivity was tested by the trypan blue more than half of the patients treated with both drugs are primary dye exclusion method. The IC50 is defined as the drug concentration that refractory or suffer relapse, suggesting that the leukemic cells of such resulted in a 50% reduction in cell number at 72 h relative to untreated control. patients acquire resistance to both drugs. Furthermore, the exact The degree of resistance was calculated by dividing the IC value of the 50 mechanism by which leukemic cells develop cross-resistance to these resistant cells by that of the parental cells. In some experiments, a colonyagents in vivo has not been clearly elucidated. forming assay was performed concurrently to compare survival with growth inhibition. All studies were carried out with exponentially growing cells. Measurement of Ara-C Influx. Ara-C influx was determined by the Received 11/15/99; accepted 11/01/00. method of Wiley et al. (6). An aliquot of 5 ⫻ 106 cells was suspended in The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with medium and preincubated at 20°C for 5 min. The cells were incubated in the 18 U.S.C. Section 1734 solely to indicate this fact. presence of 0.2 ␮M [3H]ara-C for 0 – 60 s, after which they were separated by 1 Supported in part by a Grant in Aid from the Ministry of Education, Japan. 2 centrifugation at 12,000 ⫻ g for 6 s. A 2-ml volume of NCS tissue solubilizer To whom requests for reprints should be addressed, at First Department of Internal was added to the pellet, and then the mixture was kept at room temperature Medicine, Fukui Medical University, 23-3, Shimoaizuki, Matsuoka, Fukui 910-1193, Japan. Phone: 81-776-61-8343; Fax: 81-776-61-8109. overnight to allow the cells to lyse. The amount of radioactivity in the mixture 3 The abbreviations used are: AML, acute myelogenous leukemia; ara-C, 1-␤-Dwas determined in scintillation fluid consisting of toluene and Triton X-100 arabinofuranosylcytosine; MRP, multidrug resistance-associated protein; DCK, deoxycy(2:1, v:v), PPO (4.0 grams/liter), and 2,2⬘-p-phenylene bis(5-phenyl oxazole) tidine kinase; CDD, cytidine deaminase; topo II, topoisomerase II; TKM, 50 mM Tris HCl, (0.1 gram/liter). 25 mM KCl, 5 mM MgCl2; TKE, 50 mM Tris HCl, 25 mM KCl, 1 mM EDTA; kDNA, kinetoplast DNA; calcein-AM, calcein acetoxymethyl ester; GSH, glutathione; GST, CDD Assay. The CDD assay was performed by the method of Steuart and glutathione S-transferase; BSO, L-buthionine-(S,R)-sulfoximine; Moab, monoclonal antiBurke (7), with a slight modification. Crude enzyme was obtained by sonicabody; Mit-MP, mitochondrial membrane potential; ROI, reactive oxygen intermediate; tion of cells suspended in 200 ␮M [3H]ara-C (specific activity, 25 ␮Ci/␮mol) DiOC6(3), 3,3⬘-dihexyloxacarbocyanine iodide; DCFH-DA, 6-carboboxy-2⬘,7⬘-dichloand 50 mM Tris-HCl (pH 8.0) containing 1 mM EDTA and 10 mM 2-mercaprodihydrofluorescein diacetate di(acetoxymethyl ester). 172


CROSS-RESISTANCE BETWEEN ARA-C AND DAUNORUBICIN

toethanol. The enzyme was incubated in a reaction buffer containing 50 mM Expression of bcl-2 and bcl-XL. Cell lysates were separated by 12.5% Tris-HCl (pH 8.0), 2 mM 2-mercaptoethanol, and 1 mM EDTA. The acid- SDS-PAGE. Gels were transferred to nitrocellulose, blocked with 5% blocking agent, and incubated with mouse anti-bcl-2 or anti-bcl-xL antiserum. Blots soluble fraction was subjected to TLC. DCK Assay. DCK was assayed by the method of Ives and Durhum (8). were developed with a chemiluminescent substrate. Analysis of Data. To ascertain the significance of the differences observed Briefly, crude enzyme was obtained by sonication of cells suspended in 50 mM Tris-HCl (pH 8.0) containing 2 mM DTT and then clarified by centrifugation between HL60 and HL60/AD cells subjected to various experimental condiat 100,000 ⫻ g for 60 min at 4°C. The enzyme was incubated in a reaction tions, statistical analysis of the data was performed using Student’s t test. buffer containing 20 ␮M [3H]ara-C (specific activity, 500 ␮Ci/␮mol), 40 mM Tris-HCl (pH 8.0), 10 mM MgCl2, 10 mM ATP, 10 mM 2-mercaptoethanol, and RESULTS 1 mM tetrahydrouridine. topo II Assay. The nuclear extract was prepared by the method of Kunkel Isolation of the Resistant Cell Line HL60/AD. After simultaneet al. (9). topo II activity was assayed by decatenation of kDNA into mi- ous stepwise increases of the ara-C and daunorubicin concentrations nicircles. Briefly, an aliquot of 1 ⫻ 108 cells was prepared and incubated in in the culture medium up to 0.4 ␮M ara-C and 0.1 ␮M daunorubicin, TKM buffer containing 0.25 M sucrose and 0.25% Triton X-100 at 4°C for 5 respectively, we isolated a resistant cell line by the limiting dilution min. Nuclei were pelleted by centrifugation at 1,600 ⫻ g for 5 min and washed method and named it HL60/AD. It showed about 10-fold resistance to with TKM buffer containing 0.25 M sucrose. Nuclear protein was extracted at both ara-C (IC50 ⫽ 0.40 ⫾ 0.06 ␮M in HL60/AD and 0.04 ⫾ 0.007 4°C for 60 min in TKE buffer containing 0.5 M NaCl, 1 mM DTT, and 1 mM ␮M in HL60) and daunorubicin (IC50 ⫽ 0.1 ⫾ 0.02 ␮M in HL60/AD phenylmethylsulfonyl fluoride. Nuclear extracts were obtained by centrifugaand 0.01 ⫾ 0.002 ␮M in HL60), as compared with the parental cells tion (100,000 ⫻ g, 60 min, 4°C). They were incubated for 30 min at 30°C with kDNA in reaction buffer containing 50 mM Tris-HCl (pH 7.5), 85 mM KCl,10 in the cell growth inhibition study, respectively. A colony-forming assay also revealed about 8.5-fold resistance to ara-C and 10-fold mM MgCl2, 5 mM DTT, 0.5 mM EDTA, 1 mM ATP, and 0.01% BSA. Measurement of Intracellular Daunorubicin Accumulation. Intracellu- resistance to daunorubicin. Ara-C Influx. Ara-C influx was linear with incubation time up to lar daunorubicin accumulation was determined by the method of Fukushima et al. (10). An aliquot of 2.0 ⫻ 106 cells was suspended and incubated in the 60 s and reached a steady state. No significant differences were found presence of 0.1 mM [3H]daunorubicin for a designated time at 37°C, and then between HL60/AD and parental cells in uptake (1.17 and 1.16 pmol/ cells were separated from the medium as described above. 5 ⫻ 106 cells/min, respectively) or in the steady-state concentration Detection of P-glycoprotein and MRP. P-glycoprotein was analyzed by (24.4 and 23.1 pmol/h, respectively). the method of Chaudhary et al. (11). An aliquot of 1 ⫻ 106 cells was incubated DCK and CDD Activities. DCK and CDD enzymatic activities with the P-glycoprotein Moab UIC2 for 30 min on ice and then washed twice. were determined using ara-C as substrate. The DCK activity was MRP was analyzed by the method of Flens et al. (12). Cells (1 ⫻ 106) were 530.4 ⫾ 52.7 pmol/mg/h in HL60/AD cells and 944.1 ⫾ 76.4 pmol/ incubated with MRP Moab m6 for 1 h on ice, washed twice, and analyzed mg/h in the parental cells (P ⬍ 0.01). The DCK activity was about using a FACScan (Becton Dickinson). 50% lower in HL60/AD cells than in HL60 cells. Total Cellular GSH Content. Total cellular GSH content was measured In contrast, there was no significant difference in CDD activities by the method of Tietze (13), with a minor modification. Briefly, 5 ⫻ 106 cells were prepared and washed twice with cold PBS and suspended in 125 mM (218.4 ⫾ 36.0 pmol/mg protein/h in HL60/AD and 217.3 ⫾ 22.5 sodium phosphate buffer containing 6.3 mM EDTA (pH 7.5; sodium phosphate pmol/mg protein/h in HL60). topo II Activity. Daunorubicin is known to be a topo II inhibitor buffer). After sonication, 12% 5-sulfsulicylic acid was added to the lysates, and the mixture was allowed to precipitate for 2 h at 4°C. After centrifugation at (16). Therefore, we examined the activity of topo II in the cells by 10,000 ⫻ g for 15 min, protein-free lysates were obtained. The reaction means of a kDNA decatenation assay. As shown in Fig. 1, the topo II mixture for determination of GSH content consisted of lysates, 0.3 mM activity of HL60/AD cells was substantially decreased as compared NADPH, 6 mM 5,5⬘-dithiobis-(2-nitrobenzoic acid), and 0.5 unit of GSH with that of the parental cells. Furthermore, the expression of topo II␣ reductase. The absorbance at 412 nm was monitored for 6 min using a plate was decreased as compared with that in the parental cells (data not reader (SPECTRA Max250; Molecular Devices). The content of GSH was shown). calculated from the change in the rate of absorbance on the basis of a standard Daunorubicin Accumulation. Fig. 2 shows that the intracellular curve. accumulation of daunorubicin in HL60/AD cells was substantially GST Activity. GST activity was measured as described by Habig et al. slower and lower than that in HL60 cells. 7 (14), with a slight modification. An aliquot of 1 ⫻ 10 cells was harvested, Expression of P-glycoprotein and MRP. To determine the cause washed twice with cold PBS, and resuspended in 10 mM Tris-HCl (pH 7.8) containing 0.2 M NaCl. After sonication, the lysates were centrifuged at of the decrease in intracellular daunorubicin accumulation, the ex10,000 ⫻ g for 30 min at 4°C. The supernatant was assayed for GST activity using 1-chloro-2,4-dinitrobenzene in a spectrophotometric assay. The change of absorbance at 340 nm was monitored for 5 min. The enzymatic activity was expressed in nmol/min/mg protein. Calcein Efflux Assay. Calcein efflux assay was performed by using the method of Feller et al. (15), with a slight modification. Briefly, 1 ⫻ 106 cells were incubated with 0.1 ␮M calcein-AM with or without 2 mM probenecid for 15 min at 37°C. After centrifugation, cells were resuspended in fresh medium, and the efflux of calcein was allowed for 90 min at 37°C. Intracellular calcein accumulation was measured by collecting the fluorescent calcein through a 530 nm bandpass filter with excitation at 488 nm. Detection of Mit-MP. Cells were incubated with various concentrations of ara-C for 6 h. They were resuspended in PBS and immediately incubated with 40 nM DiOC6(3) for 15 min at 37°C in a dark room. Cells were washed twice with PBS and then analyzed by flow cytometry as described above. ROI Generation. ROI generation was measured by using a fluorescent probe, DCFH-DA. Cells were exposed to1 ␮M ara-C for 24 h and then Fig. 1. DNA topo II activity from HL60 and HL60/AD cells. Lane N, negative control; incubated with 20 ␮M DCFH-DA for 12 min at 37°C. The cells were then Lanes 1 and 2, HL60; Lanes 3 and 4, HL60/AD; 0.30 ␮g of protein, Lanes 1 and 3; 0.15 washed twice with PBS and analyzed by flow cytometry as described above. ␮g of protein, Lanes 2 and 4. 173



Fig. 2. Intracellular daunorubicin accumulation. HL60 and HL60/AD cells were incubated with 1.0 ␮M [3H]daunorubicin at 37°C for the designated times. Drug accumulation is expressed as the amount of drug in 2 ⫻ 106 cells. Points are the means of duplicate determinations. E, HL60; F, HL60/AD.

Fig. 3. Analysis of MRP phenotype. MRPm6 is an anti-MRP Moab. MS-IgG1 (mouse polyclonal antibody) was used as the negative control. HL60, A; HL60/AD, B. Thin solid line, MS-IgG1; thick solid line, MRPm6.

pression of P-glycoprotein was examined, but it was not detected in either HL60/AD cells or the parental cells (data not shown). Furthermore, we performed a rhodamine 123 efflux assay, which is a functional assay of P-glycoprotein (17), but we found no difference between the two cell lines (data not shown). Recently, a second energy-dependent efflux pump, MRP, has been identified in mammalian cells (18). MRP can confer resistance to a wide spectrum of natural product drugs (19). As shown in Fig. 3, expression of MRP was found in HL60/AD cells, but not in the parental cells. MRP Function in HL60/AD Cells. To examine the function of MRP, we performed a calcein efflux assay. As shown in Fig. 4A, intracellular calcein accumulation was significantly lower in HL60/AD cells than in HL60 cells. However, this reduction was partially reversed in the presence of probenecid, which is known to be an inhibitor of MRP synthesis, in HL60/AD cells. On the other hand, it has been reported that the function of MRP was associated with GSH and GST (20, 21). The GSH content was therefore measured in the HL60/AD cells and parental cells. The level of GSH in the cells was normalized to the amount of total cellular protein. As shown in Table 1, the cellular GSH content was about 1.5-fold higher in HL60/AD cells than in HL60 cells (P ⬍ 0.001). In addition, we examined GST activity because GST is the enzyme family catalyzing conjugation with GSH. As for GST activity, HL60/AD cells contained 101.3 ⫾ 2.9 nmol/min/mg protein, as compared with 80.6 ⫾ 3.3 nmol/min/mg protein in the parental cells (P ⬍ 0.01). To examine the role of GSH, cells were preincubated with and without a sublethal concentration of BSO (50 ␮M), a potent inhibitor of GSH synthesis,

for 24 h. Then they were incubated with and without ara-C for 72 h at 37°C and subjected to the calcein efflux assay. The accumulation of calcein in HL60/AD cells was substantially increased by pretreatment with BSO (Fig. 4B). Reversal of Daunorubicin Resistance by BSO. In an attempt to clarify whether GSH was actually correlated with the mechanism of daunorubicin resistance, cells were preincubated with or without a sublethal concentration of BSO for 24 h. They were then incubated with or without daunorubicin for 72 h at 37°C. The GSH level was reduced by less than 5% after 16 h of incubation in the presence of BSO. BSO alone did not affect the survival in any cell line. After pretreatment with BSO, there was a 3.3-fold increase in drug sensitivity as compared with the untreated cells. In contrast, the sensitivity to HL60 was not changed (Fig. 5A). These results in HL60/AD cells were confirmed by a colony-forming assay. The IC50 showed a 3.5-fold increase as compared with the untreated cells. MRP-GSH System in Ara-C Resistance Mechanism. The same experiments shown in Fig. 5A were performed using ara-C instead of daunorubicin. Interestingly, the sensitivity to ara-C was also remarkably recovered (Fig. 5B). This indicates that inhibition of GSH synthesis with BSO modifies not only daunorubicin cytotoxicity but also ara-C cytotoxicity in HL60/AD cells. Comparison of HL60/AD Cells with Cells Selected by a Single Agent. To confirm the possibility that simultaneous treatment with ara-C and daunorubicin induced an elevation of GSH, we established an ara-C-resistant HL60 cell line (10-fold resistance) selected with ara-C alone (HL60/ara-C). HL60/ara-C showed low DCK activity and almost the same CDD activity and ara-C influx as seen in HL60 cells. The activity of topo II was not decreased in HL60/ara-C. GSH did not increase compared with that in the parental cells (Table 1). In addition, the IC50 was 0.4 ␮M with BSO and 0.5 ␮M without BSO, respectively. Thus, reversal of ara-C resistance with BSO was not found in this cell line. As a model for the singly selected daunorubicin-resistant cells, we already established and characterized K562/ D1-9, which is 28.0-fold more resistant to daunorubicin than the parental cells (22). Briefly, compared with the wild type, K562/D1-9 showed decreased intracellular daunorubicin retention, decreased topo II activity, and positive expression of P-glycoprotein but did not express MRP or have an increased level of GSH. Compared with the wild type, the activities of DCK were not decreased (423.4 versus 447.4 pmol/mg/h), and CDD activities were not increased (443.8 versus 456.1 pmol/mg/h, respectively), In addition, the ara-C influx in K562/D1-9 was similar to that in parental cells. The IC50 with and without BSO was 5.6 and 5.2 ␮M, respectively, showing no reversal effect of BSO. Mit-MP and ROI Generation. Using DiOC6(3) as a fluorescent probe to assess Mit-MP, two distinct subpopulations appeared after exposure to ara-C. Cells with low fluorescence intensity have a loss of

Fig. 4. Flow cytometric histograms showing functional incorporation of calcein-AM. A, HL60 and HL60/AD with or without probenecid. Black-shaded histogram, control (without calcein-AM); thick black line, HL60 without probenecid; thin black line, HL60/AD without probenecid; dotted gray line, HL60/AD with probenecid. B, HL60/AD treated or not treated with BSO. Thin black line, HL60/AD not treated with BSO; thick black line, HL60/AD treated with BSO.

174



Fig. 5. Sensitization with BSO in both cell types. After treatment with 50 ␮M BSO for 24 h, cells were incubated with ara-C (A) or daunorubicin (B) at various concentrations for 72 h. E, HL60; F, HL60 ⫹ BSO; 䡺, HL60/AD; f, HL60/ AD ⫹ BSO.

Mit-MP (23, 24). As shown in Fig. 6, the percentage of cells with high Mit-MP decreased after treatment with ara-C in the parental cells but did not obviously decrease in HL60/AD cells. ROI generation in both cell types is shown in Fig. 7. HL60 cells showed an increase in ROI generation after ara-C exposure. In contrast, in HL60/AD cells, ROI generation did not increase after ara-C exposure but increased significantly after preincubation with BSO. There were no substantial changes of Mit-MP and ROI after exposure to ara-C ⫹ daunorubicin and daunorubicin alone with BSO pretreatment in HL60/AD cells (data not shown). DNA Fragmentation and Expression of bcl-2 and bcl-XL. After preincubation with BSO, HL60/AD cells were exposed to 10 ␮M ara-C for 4 h. Ara-C-induced internucleosomal DNA fragmentation was not inhibited (Fig. 8A). HL60 is known to express bcl-2 protein, and no difference in bcl-2 expression was found between the two cell lines (Fig. 8B). In addition, bcl-XL, which is reported to inhibit drug-induced apoptosis, could not be detected in either cell line (Fig. 8C).

between DNA and topo II. This stabilized cleavable complex seems to be the trigger for cell death (27). Previous studies have reported that cells resistant to these agents displayed quantitative or qualitative alterations of topo II (28, 29), resulting in a decrease of activity. Decreased intracellular accumulation was also found in HL60/AD cells. This result suggests the existence of a drug efflux system in the cells. P-glycoprotein is an energy-dependent efflux pump and related to drug resistance by decreasing intracellular drug accumulation (30). However, HL60/AD cells showed no expression of P-glycoprotein, but they showed the expression of MRP (Fig. 3). Recently, it has been suggested that MRP, a member of ATP-binding cassette transporter superfamily, can confer resistance to a broad range of natural product drugs by extruding the drugs from the cells (31, 32). Furthermore, it is known that resistant cells overexpressing MRP often show alterations in topo II. Some studies reported that MRP expression had an impact on clinical outcome in AML (33), and some did not (34). In

Table 1 Comparison of activity of DCK and CDD and GSH content in HL60, HL60/AD, and HL60/ara-C

DISCUSSION Resistance to ara-C is generally considered to arise from reduced influx of ara-C across cell membranes (6), decrement of 1-␤-Darabinofuranosylcytosine 5⬘-triphosphate, which is an active metabolite of ara-C, caused by decreased activities in DCK (4), and increased conversion to1-␤-D-arabinofuranosyluracil, which is an inactive metabolite of ara-C, by high level of CDD (7) and so on. On the other hand, the resistance to anthracycline, including daunorubicin and doxorubicin, is known to be correlated with reduced accumulation of drugs by an overexpression of P-glycoprotein (22), MRP (25), and other ATP-binding cassette transporter and quantitative or qualitative alterations of topo II, and so forth. It is also well known that resistance to antimetabolites, including ara-C, usually does not cross to resistance to anthracyclines. In the present study, HL60/AD cells selected with simultaneous treatment with both ara-C and daunorubicin presented multiple mechanisms of resistance to these agents. Because they were selected with drug concentrations attainable in vivo, they showed low level resistance as compared with the parental cells. As for resistance to ara-C, DCK activities were decreased in HL60/AD cell line, but differences of CDD activities were not found between the cell lines. Furthermore, there were no differences of ara-C incorporation into cells between the cell lines. Therefore, the change of DCK activitiy is one of the mechanisms of resistance to ara-C. With regard to daunorubicin, HL60/AD cells showed a decreased activity of DNA topo II (Ref. 26; Fig. 1). topo II inhibitors, such as anthracyclines, etoposide, or amsacrine, stabilize a cleavable complex

DCK activity

CDD activity

GSH content

Cell line

pmol/mg protein/h

pmol/mg protein/h

nmol/min/mg protein

HL60 HL60/AD HL60/ara-C

944.1 ⫾ 76.4 530.4 ⫾ 52.7a 467.8 ⫾ 77.4a

217.3 ⫾ 22.5 218.4 ⫾ 36.0 179.4 ⫾ 13.3

35.6 ⫾ 5.6 55.6 ⫾ 5.2a 32.6 ⫾ 6.1

a

Significantly different (P ⬍ 0.01) compared with parental cells.

Fig. 6. Comparison of the changes of Mit-MP after ara-C exposure in parental and HL60/AD cells with or without BSO. Cells were incubated with various concentrations of ara-C for 6 h. Mit-MP was determined by flow cytometry using the fluorescent probe DiOC6(3). Cells with high Mit-MP constitute a subpopulation whose Mit-MP has not been lost. The loss of Mit-MP is shown as a decrease in the percentage of cells with high Mit-MP. E, HL60; F, HL60 with BSO; 䡺, HL60/AD; f, HL60/AD with BSO.

175



Fig. 7. ROI generation after ara-C exposure in parental and HL60/AD cells. One ␮M ara-C was added to HL60 (A) and HL60/AD cells (B) for 24 h with or without BSO. ROI was determined using 20 ␮M DCFH-DA. Gray dotted line, control (without ara-C); thin black line, 10 ␮M ara-C without BSO; thick black line, 10 ␮M ara-C with BSO.

this regard, a correlation between MRP and GSH has been reported recently (20, 21). Versantvoort et al. (31) indicated that transport of drugs in MRP-overexpressing but not in P-glycoprotein-overexpressing multidrug resistance cells can be regulated by cellular GSH levels. Therefore, MRP function may be dependent on intracellular GSH level. To elucidate the correlation between MRP and GSH, we studied whether the function of MRP was affected by the level of GSH, using a fluorescent probe, calcein-AM. It was reported that the calcein efflux assay could explore MRP activity (35). In brief, cells exposed to calcein-AM become fluorescent after the cleavage of calcein-AM by cellular esterase that produces a fluorescent derivative calcein. A fluorescent calcein can be actively extruded by MRP. On the other hand, probenecid, which is a specific and effective inhibitor of MRP function, can modulate the efflux of calcein (36). As shown in Fig. 5, calcein efflux was prevented in the presence of probenecid. Furthermore, it was prevented by preincubation with BSO, which is a potent inhibitor of GSH synthesis. If MRP plays a role in transporting daunorubicin together with GSH, daunorubicin efflux would also be inhibited by lowering GSH levels with BSO. As shown in Fig. 4A, the resistance of HL60/AD cells to daunorubicin was remarkably reversed by BSO pretreatment. Accordingly, these results suggest that MRP in HL60/AD cells may function so as to decrease the intracellular accumulation of drugs, depending on the level of GSH. In addition, we assayed total GST catalytic activity in each cell line. The GSTs constitute an enzyme family catalyzing with GSH, and four classes of GSTs are known. GST␲ is the form most commonly overexpressed in many organs and is known to be associated with drug resistance (37). Its mechanism of action remains obscure, but GSH and GSTs are closely related to each other. Although it is still unclear whether daunorubicin itself is transported by MRP (38), our results suggest that the expression of MRP is correlated with an elevated level of GSH, probably in combination with increased activity of GST. Interestingly, the resistance to ara-C was also remarkably reversed by BSO pretreatment in HL60/AD cells. This evidence suggests that the elevation in cellular GSH content takes part in the resistance to ara-C in HL60/AD cells. However, GSH is not usually considered to be associated with resistance to antimetabolite drugs, such as ara-C and methotrexate. In this respect, it has been reported that AML blast stem cells can be protected against ara-C lethality by N-acetyl cysteine (39). N-Acetyl cysteine is a free radical scavenger and is known to protect cells against damage by free radicals. Fig. 6 shows the changes of Mit-MP after 6 h of treatment with ara-C. The decrease in Mit-MP observed in parental cells after treatment with ara-C was not remarkable in HL60/AD cells. The generation of ROI was increased after 24 h of treatment with ara-C in parental cells, but not in HL60/AD cells (Fig. 7). However, HL60/AD cells showed a significant loss of Mit-MP and a high level of ROI generation when they were coincubated with BSO before being exposed to ara-C. These observations suggest that GSH can protect cells against damage by oxidative stress.

Whereas the source of ROI is uncertain, the generation of ROI is known to occur after ara-C incorporation into DNA (29). As a hypothesis, it might be possible that the inner mitochondrial membrane changes and the mitochondrial permeability transition can occur due to changes in the pore of the structure (23). The permeability transition produces the loss of Mit-MP due to oxidative stress and hence the uncoupling of oxidative phosphorylation. Finally, the loss of a feedback inhibition of respiratory chain activity occurs due to the depletion of cellular ATP, thereby causing a further increase of ROI. Our results suggested that oxidative stress was a critical part of the distal events in AML blasts treated with ara-C (29) and was, by itself, injurious to cells. GSH was thought to act as an antioxidant defense mechanism against oxidative stress (40). Although this mechanism does not work in typical ara-C-resistant cells selected by exposure to ara-C, it could be induced in particular situations, such as exposure to a combination of other drugs. However, it is not clear whether GSH plays an independent role in antioxidant defense. In this respect, bcl-2 is considered to have antioxidant activity (41), and it is reported that bcl-2-expressing cells lose the suppression of apoptosis by depleting cellular thiols (42). As shown in Fig. 8, DNA fragmentation was not inhibited in HL60/AD cells when they were exposed to ara-C after preincubation with BSO. These pieces of evidence suggested that resistance to ara-C was not caused by the prevention of cell death by a downstream apoptotic pathway. Considering these observations, the role of GSH may be the control of the redox state in HL60/AD cells. In conclusion, we established HL60/AD cell line resistant to both ara-C and daunorubicin, which was selected with simultaneous exposure to these two agents. The characteristics of HL60/AD cells were as follows: (a) resistance to ara-C arising from decreased DCK activity; and (b) resistance to daunorubicin related to alteration of topo II activity and the expression of MRP. Interestingly, the elevation of GSH was shown as a common resistant mechanism to both drugs. These findings suggest that the mechanism of cross-resistance to ara-C and anthracycline may not be the summation of each drug but that a new mechanism could be induced in such a case. Because it is unclear whether the MRP-GSH mechanism would be induced in other dual-selected cell lines, it would be necessary to establish and extensively characterize such resistant cell lines including a cell line transfected with MRP. In conclusion, we suggest that a new modality of circumvention of resistance should be established in patients with refractory or relapsed leukemia after combination chemotherapy.

Fig. 8. A, DNA fragmentation. Cells were exposed to 10 ␮M ara-C for 4 h after preincubation with or without BSO. Lane 1, HL60 without BSO; Lane 2, HL60/AD without BSO; Lane 3, HL60/AD with BSO. B, expression of bcl-2. C, detection of bcl-XL. Lane C, positive control (bcl-XL protein); Lane 1, HL60; Lane 2, HL60/AD.

176



ACKNOWLEDGMENTS We thank R. Nishi and A. Murai for technical and secretarial assistance. We also thank Meiji Seika Co. for the generous gift of daunorubicin.

REFERENCES 1. Clarkson, B. D., Dowling, M. D., Gee, T. S., Cunningham, I. B., and Burchenal, J. H. Treatment of acute leukemia in adults. Cancer (Phila.), 36: 775–795, 1975. 2. Carter, S. K. Adriamycin: a review. J. Natl. Cancer Inst. (Bethesda), 55: 1265–1274, 1975. 3. Kufe, D., Spriggs, D., Egan, E. M., and Munroe, D. Relationships among Ara-CTP pools, formation of (Ara-C)DNA, and cytotoxicity of human leukemic cells. Blood, 64: 54 –58, 1984. 4. Bhalla, K., Hindenburg, A., Taub, R. N., and Grant, S. Isolation and characterization of an anthracycline-resistant human leukemic cell line. Cancer Res., 45: 3657–3662, 1985. 5. Mazzoni, A., Trave, F., Russo, P., Nicolin, A., and Rustum, Y. M. Generation and characterization of a low-degree drug-resistant human tumor cell line. Oncology (Basel), 47: 488 – 494, 1990. 6. Wiley, J. S., Jones, S. P., Sawyer, W. H., and Paterson, A. R. 1-␤-D-Arabinofuranosylcytosine influx and nucleoside transport sites in acute leukemia. J. Clin. Investig., 69: 479 – 489, 1982. 7. Steuart, C. D., and Burke, P. J. Cytidine deaminase and the development of resistance to 1-␤-D-arabinofuranoxylcytosine. Nat. New Biol., 233: 109 –110, 1971. 8. Ives, D. H., and Durham, J. P. Deoxycytidine kinase. 3. Kinetics and allosteric regulation of the calf thymus enzyme. J. Biol. Chem., 245: 2285–2294, 1970. 9. Kunkel, L. M., Smith, K. D., Boyer, S. H., Borgaonkar, D. S., Wachtel, S. S., Miller, O. J., Breg, W. R., Jones, H. J., and Rary, J. M. Analysis of human Y-chromosomespecific reiterated DNA in chromosome variants. Proc. Natl. Acad. Sci. USA, 74: 1245–1249, 1977. 10. Fukushima, T., Kawai, Y., Nakayama, T., Yamauchi, T., Yoshida, A., Urasaki, Y., Imamura, S., Kamiya, K., Tsutani, H., Ueda, T., and Nakamura, T. Superior cytotoxic potency of mitoxantrone in interaction with DNA: comparison with that of daunorubicin. Oncol. Res., 8: 95–100, 1996. 11. Chaudhary, P. M., Mechetner, E. B., and Roninson, I. B. Expression and activity of the multidrug resistance P-glycoprotein in human peripheral blood lymphocytes. Blood, 80: 2735–2739, 1992. 12. Flens, M. J., Izquierdo, M. A., Scheffer, G. L., Fritz, J. M., Meijer, C. J., Scheper, R. J., and Zaman, G. J. Immunochemical detection of the multidrug resistanceassociated protein MRP in human multidrug-resistant tumor cells by monoclonal antibodies. Cancer Res., 54: 4557– 4563, 1994. 13. Tietze, F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal. Biochem., 27: 502–522, 1969. 14. Habig, W. H., Pabst, M. J., and Jakoby, W. B. Glutathione S-transferases. The first enzymatic step in mercapturic acid formation. J. Biol. Chem., 249: 7130 –7139, 1974. 15. Feller, N., Kuiper, C. M., Lankelma, J., Ruhdal, J. K., Scheper, R. J., Pinedo, H. M., and Broxterman, H. J. Functional detection of MDR1/P170 and MRP/P190-mediated multidrug resistance in tumour cells by flow cytometry. Br. J. Cancer, 72: 543–549, 1995. 16. Liu, L. F. DNA topoisomerase poisons as antitumor drugs. Annu. Rev. Biochem., 58: 351–375, 1989. 17. Webb, M., Raphael, C. L., Asbahr, H., Erber, W. N., and Meyer, B. F. The detection of rhodamine 123 efflux at low levels of drug resistance. Br. J. Haematol., 93: 650 – 655, 1996. 18. Cole, S. P., Bhardwaj, G., Gerlach, J. H., Mackie, J. E., Grant, C. E., Almquist, K. C., Stewart, A. J., Kurz, E. U., Duncan, A. M., and Deeley, R. G. Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science (Washington DC), 258: 1650 –1654, 1992. 19. Zaman, G. J., Versantvoort, C. H., Smit, J. J., Eijdems, E. W., de Hass, M., Smith, A. J., Broxterman, H. J., Mulder, N. H., de Vries, E. G., Baas, F., et al. Analysis of the expression of MRP, the gene for a new putative transmembrane drug transporter, in human multidrug-resistant lung cancer cell lines. Cancer Res., 53: 1747–1750, 1993. 20. Zaman, G. J., Lankelma, J., van Tellingen, O., Beijnen, J., Dekker, H., Paulusma, C., Oude, E. R., Baas, F., and Borst, P. Role of glutathione in the export of compounds from cells by the multidrug-resistance-associated protein. Proc. Natl. Acad. Sci. USA, 92: 7690 –7694, 1995.

21. Lutzky, J., Astor, M. B., Taub, R. N., Baker, M. A., Bhalla, K., Gervasoni, J. J., Rosado, M., Stewart, V., Krishna, S., and Hindenburg, A. A. Role of glutathione and dependent enzymes in anthracycline-resistant HL60/AR cells. Cancer Res., 49: 4120 – 4125, 1989. 22. Urasaki, Y., Ueda, T., Yoshida, A., Fukushima, T., Takeuchi, N., Tsuruo, T., and Nakamura, T. Establishment of a daunorubicin-resistant cell line which shows multidrug resistance by multifactorial mechanisms. Anticancer Res., 16: 709 –714, 1996. 23. Backway, K. L., McCulloch, E. A., Chow, S., and Hedley, D. W. Relationships between the mitochondrial permeability transition and oxidative stress during ara-C toxicity. Cancer Res., 57: 2446 –2451, 1997. 24. Salvioli, S., Ardizzoni, A., Franceschi, C., and Cossarizza, A. JC-1, but not DiOC6(3) or rhodamine 123, is a reliable fluorescent probe to assess ⌬ psi changes in intact cells: implications for studies on mitochondrial functionality during apoptosis. FEBS Lett., 411: 77– 82, 1997. 25. Heijn, M., Hooijberg, J. H., Scheffer, G. L., Szabo, G., Westerhoff, H. V., and Lankelma, J. Anthracyclines modulate multidrug resistance protein (MRP) mediated organic anion transport. Biochim. Biophys. Acta, 1326: 12–22, 1997. 26. DiNardo, S., Voelkel, K., and Sternglanz, R. DNA topoisomerase II mutant of Saccharomyces cerevisiae: topoisomerase II is required for segregation of daughter molecules at the termination of DNA replication. Proc. Natl. Acad. Sci. USA, 81: 2616 –2620, 1984. 27. Bruno, S., Del Bino, G., Lassota, P., Giaretti, W., and Darzynkiewicz, Z. Inhibitors of proteases prevent endonucleolysis accompanying apoptotic death of HL-60 leukemic cells and normal thymocytes. Leukemia (Baltimore), 6: 1113–1120, 1992. 28. Webb, C. D., Latham, M. D., Lock, R. B., and Sullivan, D. M. Attenuated topoisomerase II content directly correlates with a low level of drug resistance in a Chinese hamster ovary cell line. Cancer Res., 51: 6543– 6549, 1991. 29. Hedley, D. W., and McCulloch, E. A. Generation of reactive oxygen intermediates after treatment of blasts of acute myeloblastic leukemia with 1-␤-D-arabinofuranosylcytosine: role of bcl-2. Leukemia (Baltimore), 10: 1143–1149, 1996. 30. Juliano, R. L., and Ling, V. A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim. Biophys. Acta, 455: 152–162, 1976. 31. Versantvoort, C. H., Broxterman, H. J., Bagrij, T., Scheper, R. J., and Twentyman, P. R. Regulation by glutathione of drug transport in multidrug-resistant human lung tumour cell lines overexpressing multidrug resistance-associated protein. Br. J. Cancer, 72: 82– 89, 1995. 32. Grant, C. E., Valdimarsson, G., Hipfner, D. R., Almquist, K. C., Cole, S. P., and Deeley, R. G. Overexpression of multidrug resistance-associated protein (MRP) increases resistance to natural product drugs. Cancer Res., 54: 357–361, 1994. 33. Schneider, E., Cowan, K. H., Bader, H., Toomey, S., Schwartz, G. N., Karp, J. E., Burke, P. J., and Kaufmann, S. H. Increased expression of the multidrug resistanceassociated protein gene in relapsed acute leukemia. Blood, 85: 186 –193, 1995. 34. Filipits, M., Suchomel, R. W., Zochbauer, S., Brunner, R., Lechner, K., and Pirker, R. Multidrug resistance-associated protein in acute myeloid leukemia: no impact on treatment outcome. Clin. Cancer Res., 3: 1419 –1425, 1997. 35. Legrand, O., Simonin, G., Perrot, J. Y., Zittoun, R., and Marie, J. P. Pgp and MRP activities using calcein-AM are prognostic factors in adult acute myeloid leukemia patients. Blood, 91: 4480 – 4488, 1998. 36. Gollapudi, S., Kim, C. H., Tran, B. N., Sangha, S., and Gupta, S. Probenecid reverses multidrug resistance in multidrug resistance-associated protein-overexpressing HL60/AR and H69/AR cells but not in P-glycoprotein-overexpressing HL60/Tax and P388/ADR cells. Cancer Chemother. Pharmacol., 40: 150 –158, 1997. 37. Kramer, R. A., Zakher, J., and Kim, G. Role of the glutathione redox cycle in acquired and de novo multidrug resistance. Science (Washington DC), 241: 694 – 697, 1988. 38. Broxterman, H. J., Heijn, M., and Lankelma, J. Re: How does the MRP/GS-X pump export doxorubicin? J. Natl. Cancer Inst. (Bethesda), 88: 466 – 468, 1996. 39. Hu, Z. B., Yang, G. S., Li, M., Miyamoto, N., Minden, M. D., and McCulloch, E. A. Mechanism of 1-␤-D-arabinofuranosylcytosine toxicity to the blast cells of acute myeloblastic leukemia: involvement of free radicals. Leukemia (Baltimore), 9: 789 – 798, 1995. 40. Salnikow, K., Gao, M., Voitkun, V., Huang, X., and Costa, M. Altered oxidative stress responses in nickel-resistant mammalian cells. Cancer Res., 54: 6407– 6412, 1994. 41. Hockenbery, D. M., Oltvai, Z. N., Yin, X. M., Milliman, C. L., and Korsmeyer, S. J. Bcl-2 functions in an antioxidant pathway to prevent apoptosis. Cell, 75: 241–251, 1993. 42. Mirkovic, N., Voehringer, D. W., Story, M. D., McConkey, D. J., McDonnell, T. J., and Meyn, R. E. Resistance to radiation-induced apoptosis in Bcl-2-expressing cells is reversed by depleting cellular thiols. Oncogene, 15: 1461–1470, 1997.

177
