ANTICANCER RESEARCH 25: 991-998 (2005) Arsenic Trioxide Circumvents Multidrug Resistance Based on Different Mechanisms in Human Leukemia Cell Lines TAMAMI SEO1, YOSHIMASA URASAKI2, HARUYUKI TAKEMURA1 and TAKANORI UEDA1 1First Department of Internal Medicine and 2Division of Transfusion Medicine, Faculty of Medical Sciences, University of Fukui, Fukui, 910-1193, Japan Abstract. To determine the antitumor effect of arsenic trioxide cells (7-9). As2O3 affects numerous intracellular signal (As2O3) on multidrug-resistant cells, we applied 3 human transduction pathways and causes many alternations in leukemia cell lines: daunorubicin (DNR)-resistant cell line cellular function. These actions of As2O3 may cause the K562/D1-9, which overexpresses p-glycoprotein (Pgp); DNR induction of apoptosis, the inhibition of growth and and 1-‚-D-arabinofuranosylcytosine (Ara-C) double-resistant angiogenesis and the promotion of differentiation. cell line HL60/AD, which overexpresses multidrug resistance- The clinical efficiency of anticancer drugs is frequently associated protein (MRP1); and Bcl-2-transfected pre-B lineage limited by the emergence of a variety of mechanisms of leukemia cell line 697/Bcl-2. Interestingly, K562/D1-9 showed resistance in tumor cells. One of these mechanisms is collateral sensitivity. Only HL60/AD showed small cross related to the increased expression of efflux pumps, resistance, but 697/Bcl-2 had no resistance to As2O3. An including P-glycoprotein (Pgp) and multidrug resistance- intracellular content of glutathione (GSH) played a critical role associated protein (MRP1). Both of them are members of in sensitivity to As2O3. Buthionine-sulfoximine (BSO), which the ABC superfamily of transporters and Zhou et al. reduces the GSH content, not only increased the As2O3 reported both of these gene expressions could play an sensitivity but also conquered the MRP1- related cross important role in the outcome of chemotherapy of acute resistance in HL60/AD. In conclusion, As2O3 was effective in leukemia (10). We examined the effect of As2O3 on all 3 cell lines, suggesting that As2O3 may be a promising agent multidrug-resistant cell lines which exhibit over-expression for the treatment of multidrug-resistant leukemia. of Pgp or MRP1 genes. We also examined the sensitivity to As2O3 in the apoptosis-resistant leukemia cell line which Arsenic trioxide (As2O3), a trivalent arsenical salt, has been carries the transfected Bcl-2 gene. used for many decades in traditional Chinese medicine for the treatment of various human diseases including tumors. Materials and Methods During the last decade, the efficacy of As O in both newly 2 3 Cell culture, cell treatment and materials. Human leukemia cell lines diagnosed and relapsed patients with acute promyelocytic K562 and its daunorubicin (DNR)-resistant cell line overexpressing leukemia (APL) has been reported (1-3). Even at low Pgp, K562/D1-9 (11), human myelogenous leukemia cell line HL60 concentrations, As2O3 induced a high rate of clinical and its DNR and 1-‚-D-arabinofuranosylcytocine (Ara-C) double- remission, causing few adverse effects and only minimal resistant cell line which overexpresses MRP1, HL60/AD (12), and human pre-B lineage leukemia cell line 697/Neo and its Bcl-2- bone marrow suppression. In patients with APL, As2O3 has been shown to cause degradation of the aberrant PML- transfected cells (697/Bcl-2), which is dexamethasone resistance,(13) were cultured and passaged in RPMI 1640 (Sigma retinoic acid receptor · fusion protein, resulting in complete Chemical Co, St. Louis, MO, USA) supplemented with 10% fetal remission (4-6). Recently, experimental studies have bovine serum(FBS, Sigma-Aldrich. Co.Ltd., St. Louis, MO, USA) provided further evidence of the antiproliferative properties in a 5% CO2 incubator at 37ÆC. of As2O3 towards malignant lymphoid cells and solid tumor Reagents. As2O3 was obtained from Sigma; DNR from Meiji Seika Co.(Tokyo, Japan); etoposide (VP-16), from Nippon Kayaku Co.(Tokyo, Japan); vincristine (VCR), from Shionogi Seiyaku Co. Correspondence to: Yoshimasa Urasaki, Division of Transfusion Ltd.(Osaka, Japan); mitoxantrone (MIT), from Takeda Co. Medicine, Faculty of Medical Sciences, University of Fukui, 23-3 (Osaka, Japan); Ara-C from Nippon Shinyaku Co.(Kyoto, Japan); Shimoaizuki, Matuoka-cho, Yoshida-gun, Fukui 910-1193, Japan. MK571 from ALEXIS (San Diego, CA, USA); buthionine sulfoximine (BSO, a selective inhibitor of Á-glutamyl cysteine Key Words: Arsenic trioxide, glutathione, multidrug-resistance, synthetase) from Nacalai Tesque (Kyoto, Japan). Other chemicals buthionine-sulfoximine, reactive oxygen species. were obtained from commercial sources. As2O3 was dissolved in a 0250-7005/2005 $2.00+.40 991 ANTICANCER RESEARCH 25: 991-998 (2005) Table I. Growth inhibition effect of As2O3 and other antileukemic agents in 3 leukemic cell lines. Cell line Drugs K562 K562/D1-9 HL60 HL60/AD 697/Neo 697/Bcl-2 As2O3 2.37±0.14 0.84±0.12* 1.40±0.34 4.54±0.10** 1.53±0.18 2.57±0.75 DNR 0.084±0.013 8.63±1.63** 0.030±0.014 0.32±0.13* ND*** ND VP-16 0.85±0.13 97.5±17.1** 0.11±0.047 6.33±1.53** ND ND VCR 0.047±0.13 1.73±0.41** ND ND ND ND MIT 0.053±0.02 3.13±1.96* 0.010±0.003 0.28±0.10* ND ND Ara-C 0.021±0.007 0.025±0.011 0.047±0.021 0.42±0.07* ND ND 4 IC50 values (mM) were determined using the MTT assay. Cells (1x10 ) were seeded and exposed to As2O3 and other antileukemic agents for 72 h. Values are means±SD of at least three independent experiments performed in triplicate. Significantly different from the parental cells (*p<0.05, **p<0.01 by Student’s t-test,*** Not determined). DNR, daunorubicin; VP-16, etoposide; VCR, vincristine; MIT, mitoxantrone small amount of 1.0N NaOH and then diluted to 1 mM with Morphological assessment. As2O3-treated or -untreated cells were phosphate-buffered saline (PBS) as the stock solution. centrifuged onto slides using Cytospin and observed using May- Giemsa staining under a light microscope. Cytotoxicity. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay was used to determine drug sensitivity. In brief, Annexin V analysis. Annexin-V assay was performed according to 1x104 cells were seeded as a suspension (100 Ìl/well) in 96-well the instructions provided in the Annexin-V-FLOUS Staining Kit microtiter plates. The cells were incubated at 37ÆC in the continuous (Roche, Penzberg, Germany). Briefly, cells (1x106) were collected presence of drug for 72 h. Cell viability was assayed by adding 50 Ìg by centrifugation at 200 xg for 5 min and washed with ice-cold PBS of MTT dye (in PBS). After a 4-h incubation, during which activated at 200 xg for 5 min at 4ÆC. The obtained cells were suspended with cells reduced the yellow MTT salt to purple formazan, the stain was 100 Ìl incubation buffer containing 2 Ìl annexin-V and incubated eluted into the medium by the addition of 100 Ìl 2-propanol for 15 min at room temperature. Fluorescent intensities were (containing 0.04N HCl). Optical densities were measured at 595 nm. determined on a FACScan. Determinations for all experiments were made in triplicate. Results were expressed as means and SD. ROS analysis. After a 120-min incubation with or without BSO, cells were incubated with dihydrorhodamine 123 Measurement of total intracellular GSH level. Total intracellular hydrochloride(DHR)and As2O3 for 60 min at 37ÆC. ROSs in the GSH content was assayed using the total Glutathione cells were determined on a FACScan as fluorescent intensities. Quantification Kit according to the manufacturer's instructions (Dojindo Molecular Technologies, Inc., Kumamoto, Japan). Results Briefly, cells (1x106) were washed with cold PBS, suspended in 10 mmol/l HCl, and two freeze-thaw cycles were performed. Five As O induced growth inhibition. Table I shows the IC for percent 5-sulfosalicylic acid was added and centrifuged at 8000 xg 2 3 50 for 10 min at 4ÆC. The supernatant was assayed for GSH content As2O3 and other anticancer agents in the parental and according to the manufacturer's instructions. The intracellular resistant cell lines. K562/D1-9, which over-expresses Pgp, GSH content was expressed in nmol/1x106 cells. showed about 100-fold resistance to DNR and cross resistance to VCR, VP-16 and MIT. Interestingly, K562/D1-9 Flow cytometric analysis of cell-cycle status. Cells (1x106) were showed about 3-fold collateral sensitivity to As2O3 which collected by centrifugation at 200 xg for 5 min, and washed twice was significantly different from the parental cell K562. with ice-cold PBS at 200 xg for 5 min at 4ÆC. The obtained cells HL60/AD, which showed MRP1 over-expression and were suspended in 1 ml PBS and 2.5 ml of 99% ethanol was added while mixing. The cells were incubated at 4ÆC for more than 30 min, 10-fold resistance to both DNR and Ara-C, was significantly washed with PBS, suspended in 200 Ìl PBS, and 100 Ìl of RNase (5 about 3-fold more resistant to As2O3 compared with the mg/ml) was added. Incubation was continued at 37ÆC for 20 min, parental cell HL60. 697/Bcl-2 did not show resistance to followed by centrifugation at 200 xg for 5 min at 4ÆC, and washing As2O3 compared with 697/Neo. with ice-cold PBS at 200 xg for 5 min at 4ÆC. The cellular DNA was then stained by applying 500 Ìl of propidium iodide (50 Ìg/ml). The Role of GSH on collateral sensitivity to As O in K562/D1-9. stained cells were analyzed on a FACScan (Becton Dickinson, 2 3 Franklin Lakes, NJ, USA). Cells with DNA content less than the We examined the relationship between the As2O3-induced cells in the G1-phase (sub-G1) were considered as apoptotic cells growth inhibition and the intracellular level of GSH.