An enhanced active efflux of CPT-11 and SN-38 in cisplatin-resistant human KB carcinoma cells

Cancer Letters 138 (1999) 13±22

An enhanced active ef¯ux of CPT-11 and SN-38 in cisplatinresistant human KB carcinoma cells Zhe-Sheng Chen a, Tomoyuki Sumizawa a, Tatsuhiko Furukawa a, Kenji Ono b, Ayako Tani a, Masaharu Komatsu a, Shin-ichi Akiyama a,* a

Department of Cancer Chemotherapy, Institute for Cancer Research, Faculty of Medicine, Kagoshima University, 8-35-1 Sakuragaoka, Kagoshima 890, Japan b Experimental Technology Research Center, Daiichi Pharmaceutical Co. Ltd., 16-13, Kita-Kasai 1-Chome Edogawa-ku, Tokyo 134, Japan Received 14 September 1998; received in revised form 10 November 1998; accepted 17 November 1998

Abstract Cisplatin-resistant KCP-4 cells were 12.4- and 31.6-fold more resistant to CPT-11 and SN-38 than parental KB-3-1 cells, respectively. We studied the mechanism of cross-resistance to CPT-11 and SN-38. Our previous study showed that multidrug resistance protein (MRP), canalicular multispeci®c organic anion transporter (cMOAT) and P-glycoprotein (P-gp) were not expressed in KCP-4 cells (Chen, Z.-S. et al., Exp. Cell Res., 240 (1998) 312±320, and Chuman, Y. et al., Biochem. Biophys. Res. Commun., 226 (1996) 158±165). The accumulation of both CPT-11 and SN-38 in KCP-4 cells was lower than that in KB3-1 cells. The ATP-dependent ef¯ux of CPT-11 and SN-38 from KCP-4 cells was enhanced compared with that from KB-3-1 cells. DNA topoisomerase (topo) I expression, topo I activity, topo I-mediated cleavable complex, and the sensitivity to SN-38 of DNA topo I in KCP-4 were similar to those in KB-3-1 cells. Furthermore, the conversion of CPT-11 to SN-38 in the two cell lines was also similar. The transport of LTC4 in KCP-4 membrane vesicles was competitively inhibited by bis-(glutathionato)platinum (II) (GS-Pt), CPT-11 and SN-38. These ®ndings suggested that an unknown transporter distinct from P-gp, MRP or cMOAT is expressed in KCP-4 cells and transports CPT-11 and SN-38. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: CPT-11; SN-38; Cisplatin-resistance; cMOAT; MRP

1. Introduction It is very important to elucidate the mechanism for resistance to camptothecin (CPT) and its analog, CPT11, since the resistance to CPT and CPT-11 in tumor cells reduces the success of chemotherapy. Many cell lines resistant to CPT analogs have been isolated in * Corresponding author. Tel.: 181-99-2755488; fax: 181-992659687. E-mail address: [email protected] (S. Akiyama)

vitro and some mechanisms of resistance to CPT analogs have been determined. These mechanisms of resistance include decreased conversion of CPT11 to SN-38 [1,2], reduced sensitivity of topoisomerase (topo) I to CPT-11 [3], and decreased expression of topo I and/or topo II [4±6]. Cells selected for resistance to cisplatin [1], melphalan [7], mAMSA [8], or mitoxantrone [9] developed cross-resistance to CPT analogs. In the present study, we found that cisplatin-resistant KCP-4 cells derived from human epidermoid carcinoma KB-3-1 cells are cross-resistant to CPT-

0304-3835/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(98)00367-X

14

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

11 and SN-38. The mechanism for the cross-resistance was examined. Our results suggest that an active ef¯ux system for CPT-11 and SN-38 exists in KCP-4 cells. 2. Materials and methods 2.1. Chemicals CPT-11 and SN-38 were obtained from Daiichi Seiyaku (Tokyo, Japan). Minimal essential medium (MEM) was purchased from Nissui Seiyaku Co. (Tokyo, Japan). Newborn calf serum was from Cell Culture Laboratories (Cleveland, OH). 14,15,19,203 H[N]Leukotriene C4 ([ 3H]LTC4) (110.5 Ci/mmol) was obtained from Du Pont NEN (Boston, MA). PAK-104P was obtained from Nissan Chemical Industries (Chiba, Japan). Cisplatin, BSO and other agents were obtained from Sigma Chemical Co. (St. Louis, MO). 2.2. Cell culture and cell lines. The human epidermoid KB carcinoma cells were obtained from Dr. M.M. Gottesman (National Cancer Institute, Bethesda, MD). KB cells were subcloned twice, and a single recloned line, KB-3-1, was used as the parental cell line for the present study [10]. KB3-1 cells were cultured in MEM containing 10% newborn calf serum, 1 mg/ml bactopeptone, 0.292 mg glutamine/ml and 100 units penicillin/ml (MEM medium). Cisplatin-resistant KCP-4 cells were isolated by culturing the KB-3-1 cells with increasing concentrations of cisplatin following ethyl methanesulfonate induced mutagenesis, then incubated in a selection medium with 7 mg/ml cisplatin [11]. An incomplete revertant, KCP-4R, was isolated from KCP-4 by culturing KCP-4 cells in non-selective MEM medium for 5 months [11]. 2.3. Cell survival by MTT assay The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay was used to assess the sensitivity of the cells to agents in vitro as described [12]. Exponentially growing cells were trypsinized and harvested, and equal numbers of cells (2000 for KB-3-1 or KCP-4R cells and 10 000

for KCP-4 cells) were inoculated into each well with 180 ml of culture medium. After an overnight incubation, 20 ml of compound solution were added and the cells were incubated for 4 days. Thereafter, 50 ml of MTT (1 mg/ml PBS) were added to each well and incubated for a further 4 h. The resulting formazan was dissolved with 100 ml of dimethyl sulfoxide after aspiration of the culture medium. Plates were placed on a plate shaker for 5 min and read immediately at 570 nm using a microplate reader, MRP-A4i (Tosoh, Tokyo). To examine the effects of BSO, PAK-104P, MK571 or piperine on drug resistance, cells were preincubated with or without 100 mM BSO for 24 h, 10 mM PAK-104P, 20 mM MK571 or 100 mM piperine for 30 min and incubated with various concentrations of drugs. 2.4. CPT-11 and SN-38 accumulation To measure drug accumulation, con¯uent KB-3-1 and KCP-4 cells in 150-mm plastic dishes were incubated overnight in MEM medium then incubated with CPT-11 or SN-38 at the indicated concentrations for 2 h at 378C. Cells were washed three times with cold PBS and immediately harvested with a rubber scraper. The harvested cells were again washed three times with cold PBS. Cells were counted with a hemocytometer before the last wash. After the addition of methanol (1 ml/10 6 cells), the cells were suspended and centrifuged at 3000 rev./min for 10 min. The supernatants were evaporated by a concentrator. A modi®ed reversed phase HPLC method reported by Kaneda and Yokokura [13] was used to analyze the CPT-11, SN-38 and SN38 glucuronide (SN38-G) content. This method permits the simultaneous assay of CPT-11 and SN38-G without the treatment of enzyme digestion. Brie¯y, the samples diluted with 0.01 M HCl were processed using a solid-phase ( 18C) extraction step. The extracts for CPT-11 and SN38-G were chromatographed on an 18C reversed phase column with a mobile phase consisting of methanol/0.1 M phosphate buffer (55:45, v/v, pH 4.0) containing 3 mM sodium 1-heptanesulfonate using ¯uorescence detection (excitation, 370 nm: emission, 430 nm), while the mobile phase for quanti®cation of SN-38 in the extracts was a mixture of acetonitrile/water (1:2, v/v). The detection was monitored by ¯uorescence with excitation and emission

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

wavelength set at 380 nm and 556 nm, respectively. Retention times for SN38-G, CPT-11 and SN-38 were 3.0, 6.0 and 5.0 min, respectively. 2.5. Ef¯ux of CPT-11 and its product SN-38 KB-3-1 and KCP-4 cells were incubated in MEM medium with 160 mM CPT-11 for 1 h at 378C. For depletion of ATP, cells were preincubated in ATPdepletion medium (i.e. glucose-free MEM medium containing 0.3 mM 2,4-dinitrophenol, 15 mM sodium azide and 50 mM 2-deoxyglucose) for 15 min at 378C, then CPT-11 was added to the medium and incubated for a further 1 h at 378C. Cells were washed three times with a total of 20 ml of 378C PBS. The cells were further incubated in the medium without CPT-11 at 378C for the indicated times. The medium was collected for measuring the ef¯uxed CPT-11, SN-38 and SN38-G. The cells were harvested after washing three times with cold PBS. Levels of CPT-11, SN-38 and SN38-G in the medium and the cells were determined as described above. 2.6. Membrane vesicle preparation Membrane vesicles were prepared as described [14] from KB-3-1 and KCP-4 cells grown in 24 £ 24 cm 2 dishes (Nunc) under standard growth conditions [10]. Protein concentration in the extract was determined according to the method of Bradford [15]. 2.7. H] LTC4 uptake in membrane vesicles LTC4 uptake in KCP-4 membrane vesicles was measured by ®ltration essentially as described by Ishikawa et al. [16,17]. The KCP-4 membrane vesicles (50 mg of protein) were incubated with 50 nM [ 3H] LTC4 at 378C for 10 min in 100 ml of the incubation medium containing 0.25 M sucrose, 10 mM Tris±HCl, pH 7.4, 10 mM MgCl2, 10 mM phosphocreatine, 100 mg/ml creatine phosphokinase and the indicated concentration of CPT-11, SN-38 or GS-Pt in the presence or absence of 1 mM ATP. The vesicles were collected and washed on millipore ®lters (GV, 0.22 mm pore size) and the level of radioactivity was measured. ATP-dependent accumulation was calculated from the difference in the radioactivity incorporated into the vesicles in the presence or absence of ATP.

15

2.8. Preparation of nuclear extracts. Crude nuclear extracts were prepared as described [18]. Following trypsinization, exponentially growing cells (one 135-mm dish) were washed in ice-cold PBS±EDTA, resuspended in 0.5 ml lysis buffer (20 mM Tris±HCl, pH 7.2; 25 mM KCl; 5 mM MgCl2; 1 mM EGTA; 250 mM sucrose; 0.5% Nonidet P-40) and kept on ice for 10 min. After 2 min of centrifugation in a microfuge, nuclei were resuspended in 0.25 ml nuclei buffer (20 mM Tris±HCl, pH 7.2, 400 mM NaCl, 20 mM EDTA, 20 mM 2-mercaptoethanol, 1 mM phenylmethylsulfonyl ¯uoride), and nuclear protein was extracted for 30 min on ice, then centrifuged for 15 min in a microfuge. The protein concentration in the extract was determined according to the method of Bradford [15]. The same volume of glycerol was added to the supernatant, and then kept at 2208C. 2.9. Topo I activity assay. Topo I activity was determined using the supercoiled Escherichia coli DNA (plasmid pBR322) relaxation assay [19]. For measurement of total activity of topo I in the three cell lines, the reaction mixtures consisted of 100 mM KCl, 10 mM MgCl2, 1 mM dithiothreitol, 0.1 mM EDTA, 10% glycerol, 50 mM Tris±HCl (pH 7.4), 1 mg of pBR322 and nuclear extracts (0.01, 0.05, 0.25 and 1.25 mg of protein). To examine the inhibition of topo I activity by CPT-11 or SN-38, different amounts of the agents (0.2, 1.0, 5.0, 25.0 mM of SN-38 or 2, 10, 50, 250 mM of CPT-11) were added to 1.25 mg of protein in the reaction. The reaction mixtures were incubated at 378C for 15 min, and the reactions terminated by the addition of 5 ml of dye solution consisting of 2.5% SDS, 0.01% bromophenol blue, and 50% glycerol. Relaxed and supercoiled DNA were separated in a 1% agarose gel by electrophoresis and visualized by staining with 2 mM ethidium bromide. 2.10. Cleavage complex formation assay The assay for precipitation of DNA±protein complexes was determined as described previously [20]. pUC18 DNA was digested with Hind III and Sca I, then the 3 0 end was labeled with [a- 32P]

16

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

Table 1 Cross-resistance to CPT-11 and SN-38 in KB cell lines IC50 (mM) a Agent

KB-3-1

Cisplatin CPT CPT-11 SN-38

0.6 0.5 3.5 0.008

^ 0.1 b (1) c ^ 0.1 (1) ^ 0.5 (1) ^ 0.001 (1)

KCP-4 30.8 10.1 43.4 0.253

KCP-4R ^ 3.6 (51.3) ^ 1.5 (20.2) ^ 5.2 (12.4) ^ 0.022 (31.6)

2.6 1.1 8.1 0.023

^ 0.2 (4.4) ^ 0.2 (2.1) ^ 0.7 (2.3) ^ 0.002 (2.9)

a

Cell survival was determined by MTT assay. Data are means ^ SD of three determinations obtained from triplicate cultures. c Relative resistance was determined by dividing the IC50 value of the drug for KCP-4 and KCP-4R cells by that for KB-3-1 cells. b

dCTP. Unincorporated nucleotides were removed by three sequential precipitations with ethanol-ammonium acetate. Cleavage assay was done in a reaction mixture containing 10 mM Tris±HCl (pH 7.0), 1 mM MgCl2, 0.5 mM EDTA, 10 mg/ml bovine serum albumin, nuclear extracts (50 mg/ml), about 5 £ 104 cpm 32P-labeled DNA, and various doses of SN-38 in a ®nal volume of 50 ml at 378C for 5 min. The reaction was terminated by the addition of 100 ml of a stop solution containing 2% SDS, 2 mM EDTA and 0.5 mg/ml of salmon sperm DNA, and incubated at 658C for 10 min. Precipitation of the topoisomerase±DNA complex was achieved by the addition of 50 ml of 0.25 M KCl and the mixture was incubated on ice for 10 min. The precipitate was collected by centrifugation at 15 000 rev./min for 15 min in the cold room. The supernatant was aspirated and the pellet was washed once with 200 ml of a solution containing 10 mM Tris±HCl (pH 8.0), 100 mM KCl, 1 mM EDTA, and 100 mg/ml of carrier salmon sperm DNA at 658C for 10 min. After cooling on ice and recentrifugation, the pellet was resuspended in 200 ml of H2O and then heated at 658C for 10 min. The suspension was transferred to a vial containing 5 ml of scintillation ¯uid and counted. 2.11. Statistical analysis Differences between groups were tested by one way ANOVA or Student's t-test. Signi®cance levels given are those for the two-tailed Student's paired ttest. Data are presented as means ^ SD. Differences were considered signi®cant when P , 0:05.

3. Results 3.1. Cross-resistance to CPT-11 and SN-38 in KB sublines As shown in Table 1, IC50 values for CPT-11 of KB-3-1 and KCP-4 cells were 3.5 and 43.4 mM, respectively. KCP-4 cells were 12.4-fold more resistant to CPT-11 than the parental KB-3-1 cells. IC50 values for SN-38 of KB-3-1 and KCP-4 cells were 0.008 and 0.253 mM, respectively. KCP-4 cells were 31.6-fold more resistant to SN-38 than the parental KB-3-1 cells. An incomplete revertant, KCP-4R, isolated from KCP-4 cells was 4.4-, 2.3- and 2.9fold more resistant to cisplatin, CPT-11 and SN-38 than KB-3-1 cells, respectively. 3.2. Accumulation of CPT-11 and SN-38 in KB sublines The accumulation of CPT-11 in KCP-4 cells was lower than that in KB-3-1 cells when the cells were incubated in the medium with 20±160 mM of CPT-11. When the cells were incubated with 160 mM CPT-11, the accumulation of CPT-11 in KCP-4 cells was 48% of that (276.29 nmol/10 6 cells) in KB-3-1 cells (Fig. 1A). The accumulation of SN-38 in KCP-4 cells was also lower than that in KB-3-1 cells. When the cells were incubated in the medium with 400 nM SN-38, the accumulation of SN-38 in KCP-4 was 35% of that (1.38 nmol/10 6 cells) in KB-3-1 cells (Fig. 1B). SN38G was not detected in the cells incubated with CPT-11 or SN-38 (data not shown).

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

17

11 for a further 1 h, 20% of CPT-11 was retained in the KCP-4 cells. On the other hand, 62% of CPT-11 was retained in KB-3-1 cells over the same period (Fig. 2A). When intracellular ATP was depleted, the CPT-11 retained in the two cell lines was similar to that in KB-3-1 cells untreated with ATP-depleting agents. The ATP-dependent ef¯ux of SN-38 from KCP-4 cells was also considerably enhanced compared with that from KB-3-1 cells. When the cells were incubated with 160 mM CPT-11 for 1 h at 378C (ef¯ux time 0), then without CPT-11 for a further 1 h, the active metabolite of CPT-11, SN-38,

Fig. 1. Accumulation of CPT-11 and SN-38. KB-3-1 and KCP-4 cells were treated with CPT-11 (panel A) and SN-38 (panel B) as indicated. The intracellular level of CPT-11 and SN-38 was determined by HPLC as indicated in Section 2. Columns represent means of triplicate determinations, bars SD.

3.3. Ef¯ux of CPT-11 and its active metabolite, SN-38 from KB sublines The ef¯ux of CPT-11 and SN-38 from KB-3-1 cells in the presence of ATP was only slightly enhanced compared with that in the absence of ATP. The ATPdependent ef¯ux of CPT-11 from KCP-4 cells was considerably enhanced compared with that from KB-3-1 cells. When the cells were incubated with 160 mM CPT-11 for 1 h at 378C, then without CPT-

Fig. 2. Ef¯ux of CPT-11 and SN-38. CPT-11 (panel A) and its active product, SN-38 (panel B), retained in KB-3-1 (circle) and KCP-4 cells (square) in the absence (open symbols) or presence (solid symbols) of ATP-depleting agents is shown. Points represent means of triplicate determinations, bars SD.

18

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

Table 2 The kinetics of CPT-11 in KB cell lines Cells

Ef¯ux time (min)

CPT-11 ef¯ux to medium nmol/10 7 cells (a)

Intracellular CPT-11 nmol/10 7 cells (b)

Ratio (a/b) (%)

Kb-31-1

0 20 60 0 20 60

± 112.69 236.77 ± 301.10 425.84

2338.93 ^ 182.15 ± ± 701.82 ^ 55.01 ± ±

± 4.8 10.25 ± 42.90 60.68

KCP-4

^ 10.12 ^ 9.64 ^ 33.17 ^ 19.74

in KCP-4 and KB-3-1 cells was decreased by 28 and 63% of those at ef¯ux time 0, respectively (Fig. 2B). SN38-G was not detected in any of the samples tested (data not shown). 3.4. CPT-11 ef¯uxed in the medium CPT-11 is converted into SN-38 by de-esteri®cation and SN-38 is conjugated with glucuronide to form SN38-G in the liver and excreted into the bile duct [21,22]. However, less than 1/1000 of the CPT11 accumulated in the KB cells was converted to SN38. We, therefore, examined whether CPT-11 in the KB cell lines was directly excreted into the medium or SN38-G was ef¯uxed from the cells. When the cells were incubated with 160 mM CPT11 for 1 h at 378C, then without CPT-11 for 1 h, 10.3, and 60.7% of the accumulated CPT-11 in KB-3-1 and KCP-4 cells at ef¯ux time 0, respectively, was

Fig. 3. Conversion of CPT-11 to SN-38. KB-3-1 and KCP-4 cells were incubated with 300 mM CPT-11 and the concentrations of CPT-11 (panel A) and SN-38 (panel B) were measured by HPLC. Columns represent means of triplicate determinations, bars SD.

ef¯uxed and detected in the medium (Table 2). On the other hand, none of the cells or media contained detectable SN38-G (data not shown). 3.5. Conversion of CPT-11 to the active metabolite SN-38 To examine the conversion of CPT-11 to SN-38, we incubated the cells in the medium with 300 mM CPT11 for 3 h, then measured the intracellular concentration of CPT-11, SN-38 and SN38-G. The level of

Fig. 4. Effect of various agents on ATP-dependent [ 3H] LTC4 uptake in KCP-4 membrane vesicles. KCP-4 membrane vesicles (50 mg of protein) were incubated with 1.34 nM [ 3H] LTC4 at 378C in 100 ml of 0.25 M sucrose containing 10 mM Tris±HCl (pH 7.4), 10 mM MgCl2, 10 mM phosphocreatine, 100 mg/ml creatine phosphokinase and the indicated concentrations of each agent in the presence or absence of 1 mM ATP. ATP-dependent [ 3H] LTC4 uptake was obtained from the difference in the radioactivities incorporated into the vesicles in the presence and absence of ATP. Values are the mean ^ SD of triplicate experiments; *P , 0:05, **P , 0:01 compared to the control.

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

19

CPT-11 in KCP-4 cells was 15% of that (688.94 nmol/ 10 6 cells) in KB-3-1 cells and the level of SN-38 in KCP-4 cells was 18% of that (0.15 nmol/10 6 cells) in KB-3-1 cells (Fig. 3). The number of SN-38 molecules in KB-3-1 and KCP-4 cells was 0.022% and 0.034% of that of CPT-11 molecules, respectively. These results indicated that the conversion of CPT11 to the active metabolite SN-38 in KCP-4 cells was not decreased compared with that in KB-3-1 cells. SN38-G was also not detected in any of the cells (data not shown). 3.6. Effect of GS-Pt, CPT-11 and SN-38 on [ 3H] LTC4 uptake To investigate whether CPT-11 and SN-38 are transported by the same transporter that transports GS±platinum complex and LTC4, we examined the effect of CPT-11 and SN-38 as well as GS-Pt on ATP-dependent [ 3H] LTC4 uptake in KCP-4 vesicles (Fig. 4). GS-Pt, CPT-11 and SN-38 at 100 mM reduced the [ 3H] LTC4 uptake in KCP-4 vesicles by 69, 52 and 59%, respectively.

Fig. 5. Comparison of topo I activity by nuclear extracts from KB3-1 and KCP-4 cells. Relaxation assays were carried out as described in Section 2. Reaction mixtures were incubated in the absence or presence of various concentrations of nuclear extracts (panel A) or in the presence of various concentrations of SN-38 (panel B).

Fig. 6. Effect of SN-38 on DNA cleavage by topo I in nuclear extracts from KB-3-1 and KCP-4 cells. 3 0 End-labeled 32P-pUC18 DNA was incubated with nuclear extracts in the presence of various concentrations of SN-38. Points represent means of triplicate determinations, bars SD.

3.7. Topo I levels in KB cell lines Decreased expression of topo I or decreased sensitivity of topo I to the topo I inhibitors may play an important role in cellular resistance to CPT-11 [23]. We, therefore, examined topo I levels and the sensitivity of topo I to CPT-11 and SN-38 in the two cell lines. There was no signi®cant difference in the expression level of topo I in the two cell lines (data no shown). We next examined the sensitivity of topo I in KB-3-1 and KCP-4 cells to CPT-11 and SN-38. We measured the total cellular activity of topo I in KB-3-1 and KCP-4 cells. In this experiment, the relaxation of supercoiled DNA by the catalytic action of topo I was monitored by gel electrophoresis. Fig. 5A shows the relaxation of pBR322 DNA with different amounts of cell nuclear extracts. Relaxed forms were observed in the presence of more than 0.25 mg of nuclear extract from KB-3-1 and KCP-4 cells. DNA topo I activity in these cell lines was similar. The effect of CPT-11 and SN-38 on the catalytic activity of topo I from the two cell lines was then examined. SN-38 similarly inhibited the topo I activity in the two cell lines and completely inhibited it at 5 mM (Fig. 5B). In contrast, CPT-11 had no effect on the activity of topo I isolated from the two cell lines even at 250 mM CPT-11 (data not shown).

20

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

3.8. Effect of SN-38 on DNA cleavage by topo I To evaluate the effect of SN-38 on the ability of topo I in KB-3-1 and KCP-4 cells to cleave DNA, we quanti®ed the protein±DNA complexes stabilized by SN-38. 3 0 -Labeled DNA was used to estimate topo I. There appeared to be no signi®cant changes in SN-38mediated cleavable complex formation with nuclear extracts from KCP-4 cells compared with parental KB-3-1 cells (Fig. 6).

4. Discussion Cancer cells treated with a certain anticancer agent acquire cross-resistance to other structurally unrelated anticancer agents. We have established and characterized a cisplatin resistant KCP-4 cell line [11,24,25]. The KCP-4 cells were highly resistant to cisplatin and cross-resistant to carboplatin, 254-S, melphalan and methotrexate, but not resistant to adriamycin (ADM) and vincristine (VCR) [11]. In the present study, KCP-4 cells were cross-resistant to CPT-11 and SN38, and we studied the mechanism of this cross-resistance. The decreased conversion of CPT-11 to SN-38 was reportedly involved in the resistance to CPT-11 in the cisplatin-resistant human ovarian cancer cell line HAC2/0.1 [1]. In our present study, the conversion ef®ciency of CPT-11 to SN-38 in KCP-4 cells seemed not to be lower than that in KB-3-1 cells since SN-38/ CPT-11 ratio in KCP-4 cells was not signi®cantly different from that in KB-3-1 cells (Fig. 3). The most common mechanisms for the resistance to camptothecin and its analogs were decreased topo I level and/or activity in the resistant cells [4±6] as well as reduced sensitivity of topo I to the inhibitors [3]. KCP-4 cells showed no changes in the level (data not shown) and activity of topo I (Fig. 5A), the sensitivity of topo I to SN-38 (Fig. 5B), and the formation of a topo I-mediated cleavable complex (Fig. 6). These results suggest that quantitative or qualitative changes of topo I are not involved in the resistance to CPT-11 and SN-38 in KCP-4 cells. The accumulation of platinum and antimony potassium tartrate in cisplatin-resistant KCP-4 cells [11,24] was signi®cantly less than that in the parental cells and the decreased accumulation played an important

role in the acquisition of resistance. In this study, we found that the accumulation of CPT-11 and SN-38 in KCP-4 cells was also lower than that in KB-3-1 cells. The active ef¯ux of CPT-11 and SN-38 from KCP-4 cells was enhanced compared with that from KB-3-1 cells. Niimi et al. [1] found that the accumulation of SN-38 in their cisplatin-resistant HAC2/0.1 cells was also decreased, but decreased accumulation of CPT11 in these cells was not observed. The ef¯ux of CPT11 and SN-38 from HAC2/0.1 cells was not investigated. The cisplatin-resistant KCP-4 cells seemed to have an active ef¯ux pump for CPT-11 and SN-38. We previously found that an active ef¯ux system for cisplatin and antimony potassium tartrate existed in KCP-4 cells [11,24]. In addition, the membrane vesicles from KCP-4 cells actively transported LTC4, the transport of [ 3H] LTC4 was inhibited by GS±platinum complex [25,26]. Our present study showed that 100 mM of GS-Pt, CPT-11 and SN-38 similarly reduced the [ 3H] LTC4 uptake in KCP-4 vesicles (Fig. 4). Furthermore, the resistance to CPT-11, SN-38 and cisplatin in KCP-4 cells was simultaneously lowered in the revertant KCP-4R cells (Table 1). These results suggest that the transporter for cisplatin and LTC4 might also transport CPT-11 and SN-38. LTC4 is the best substrate for MRP. MRP was overexpressed in cisplatin-resistant human leukemia cell line HL60/R-CP, and an increased GSH synthesis in combination with raised MRP level was argued to be involved in cisplatin resistance [27]. However, MDR reversing agents that directly interact with MRP, PAK-104P and MK-571, considerably reversed the VCR resistance in MRP expressing cells [28,29], but these agents could not reverse the resistance to CPT-11 and SN-38 in KCP-4 cells (data not shown). Furthermore, previous studies showed that P-gp, MRP and cMOAT were not expressed in KCP-4 cells [24,26]. Moreover, there was no signi®cant difference in either GST-p activity or expression between KB-31 and KCP-4 (Chen, Z-S. and Akiyama, S., unpublished observations). P-gp, MRP, cMOAT and GST-p expression thus seem not to be involved in the resistance to CPT-11 and SN-38 in KCP-4 cells. cMOAT was reportedly responsible for the biliary excretion of carboxylate forms of CPT-11, SN-38 and SN38-G and the lactone form of SN38-G [30]. A transporter other than cMOAT was suggested to

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

exist, and this unknown transporter might also contribute to the ATP-dependent transport of the lactone and carboxylate forms of SN-38 by canalicular membrane vesicles [31]. A primary active transport system for transporting several organic anions including the carboxylate form of CPT-11 and both the carboxylate and lactone forms of SN38-G, existed in Eisai hyperbilirubinemic rats (EHBRs) [31]. Although neither cMOAT nor MRP was expressed in KCP-4 cells, KCP-4 cells had an active ef¯ux pump for cisplatin, antimony, CPT-11 and SN-38. An unknown ef¯ux pump for these agents seems to be expressed in KCP-4 cells. Further studies are needed to elucidate whether the pump expressed in KCP-4 cells is identical to a primary active transport pump, expressed in EHBRs. A mitoxantrone-resistant human MCF-7 breast carcinoma cell line, MCF7/MX, was 3932-fold more resistant to mitoxantrone [18], and 56-fold and 101-fold more resistant to CPT-11 and SN-38 [9], respectively, than the parental MCF7 cells. MCF7/MX cells exhibited reduced accumulation of topotecan as well as mitoxantrone in the absence of P-gp [9,18]. An unknown active ef¯ux pump for these anticancer agents distinct from P-gp seems to be expressed in MCF7/MX cells. However, the pump expressed in MCF7/MX cells may not be identical to that expressed in KCP-4 cells, since MCF7/MX cells were not cross-resistant to cisplatin [18], and KCP-4 cells were not cross-resistant to mitoxantrone (Chen, Z-S. and Akiyama, S., unpublished observations). CPT-11 as well as camptothecin is clinically used and it is important to elucidate the mechanism of resistance to CPT-11 and its active metabolite SN-38. In this study, we have shown that an unknown transporter is involved in the resistance to CPT-11 and SN-38 in KCP-4 cells and we are now trying to isolate and analyze the gene for this ef¯ux pump for CPT-11 and SN-38 as well as heavy metals expressed in cisplatin-resistant KCP-4 cells.

Acknowledgements This work was supported by grants from the Ministry of Education, Science and Culture, the Ministry of Health and Welfare, and Sasakawa Health Science Foundation, Japan. We thank Ms. Etuko

21

Sudou for technical assistance and Ms. Hiromi Kakura for secretarial assistance. References [1] S. Niimi, K. Nakagawa, Y. Sugimoto, K. Nishio, Y. Fujiwara, S. Yokoyama, Y. Terashima, N. Saijo, Mechanism of crossresistance to a camptothecin analogue (CPT-11) in a human ovarian cancer cell selected by cisplatin, Cancer Res. 52 (1992) 328±333. [2] H. Ogasawara, K. Nishio, F. Kanzaki, Y.-S. Lee, Y. Funayama, T. Ohira, Y. Kuraishi, Y. Isogai, N. Saijo, Intracellular carboxyl esterase activity is a determinant of cellular sensitivity to the antineoplastic agent KW-2189 in cell lines resistant to cisplatin and CPT-11, Jpn. J. Cancer Res. 86 (1995) 124±129. [3] A. Tanizawa, R. Bertrand, G. Kohlhagen, A. Tabuchi, J. Jenkins, Y. Pommier, Cloning of Chinese hamster DNA topoisomerase I cDNA and identi®cation of a single point mutation responsible for camptothecin resistance, J. Biol. Chem. 268 (1993) 25463±25468. [4] J.-Y. Chang, L.A. Dethlefsen, L.R. Barley, B.-S. Zhou, Y.-C. Cheng, Characterization of camptothecin-resistant Chinese hamster lung cells, Biochem. Pharmacol. 43 (1992) 2443± 2452. [5] R.D. Woessner, W.-K. Eng, G.A. Hofmann, D.J. Rieman, F.L. McCabe, R.P. Hertzberg, M.R. Mattern, K.B. Tan, R. Johnson, Camptothecin hyper-resistant P388 cells: drugdependent reduction in topoisomerase I content, Oncol. Res. 4 (1992) 481±488. [6] Y. Sugimoto, S. Tsukahara, T. Oh-hara, T. Isoe, T. Tsuruo, Decreased expression of DNA topoisomerase I in camptothecin-resistant tumor cell lines as determined by a monoclonal antibody, Cancer Res. 50 (1990) 6925±6930. [7] H.S. Friedman, M.E. Dolan, S.H. Kaufmann, O.M. Colvin, O.W. Grif®th, R.C. Moschel, S.C. Schold, D.D. Bigner, F. Ali-Osman, Elevated DNA polymerase a, DNA polymerase b, and DNA topoisomerase II in a melphalan-resistant rhabdomyosarcoma xenograft that is cross-resistant to nitrosourea and topotecan, Cancer Res. 54 (1994) 3487±3493. [8] S. Prost, G. Riou, A human small cell lung carcinoma cell line, resistant to 4 0 -(9-acridinylamino) methanesulfon-m-anisidide and cross-resistant to camptothecin with a high level of topoisomerase I, Biochem. Pharmacol. 48 (1994) 975±984. [9] C.-H.J. Yang, J.K. Horton, K.H. Cowan, E. Schneider, Crossresistance to camptothecin analogues in a mitoxantrone-resistant human breast carcinoma cell line is not due to DNA topoisomerase I, Cancer Res. 55 (1995) 4004±4009. [10] S. Akiyama, J.A. Fojo, J.A. Hanover, I.H. Pastan, M.M. Gottesman, Isolation and genetic characterization of human KB cell lines resistant to multiple drugs, Somat. Cell Mol. Genet. 11 (1985) 117±126. [11] R. Fujii, M. Mutoh, K. Niwa, K. Yamada, T. Aikou, M. Nagagawa, M. Kuwano, S. Akiyama, Active ef¯ux system for cisplatin in cisplatin-resistant human KB cells, Jpn. J. Cancer Res. 85 (1994) 426±433.

22

Z.-S. Chen et al. / Cancer Letters 138 (1999) 13±22

[12] J. Carmichael, W.G. DeGraff, A.F. Gazdar, J.D. Minna, J.B. Mitchell, Evaluation of a tetrazolium-based semiautomated colorimetric assay: Assessment of chemosensitivity testing, Cancer Res. 47 (1987) 936±942. [13] N. Kaneda, T. Yokokura, Nonlinear pharmacokinetics of CPT-11 in rats, Cancer Res. 50 (1990) 1721±1725. [14] M.M. Cornmell, M.M. Gottesman, I.H. Pastan, Increased vincristine binding to membrane vesicles from multidrugresistant KB cells, J. Biol. Chem. 261 (1986) 7921±7928. [15] M.M.A. Bradford, Rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248± 254. [16] T. Ishikawa, F. Ali-Osman, Glutathione-associated cisdiamminedichloroplatinum (II) metabolism and ATP-dependent ef¯ux from leukemia cells. Molecular characterization of glutathione-platinum complex and its biological signi®cance, J. Biol. Chem. 268 (1993) 20116±20125. [17] T. Ishikawa, ATP/Mg 21-dependent cardiac transport system for glutathione S-conjugates. A study using rat heart sarcolemma vesicles, J. Biol. Chem. 264 (1989) 17347±17348. [18] M. Nakagawa, E. Schneider, K.H. Dixon, J. Horton, K. Kelley, C. Morrow, K.H. Cowan, Reduced intracellular drug accumulation in the absence of P-glycoprotein (mdr1) overexpression in mitoxantrone-resistant human MCF-7 breast cancer cells, Cancer Res. 52 (1992) 6175±6181. [19] L.F. Liu, K.G. Miller, Eukaryotic DNA topoisomerases; two forms of type I DNA topoisomerases nuclei, Proc. Natl. Acad. Sci. USA 78 (1981) 3487±3491. [20] H. Takano, K. Kohno, M. Ono, Y. Uchida, M. Kuwano, Increased phosphorylation of DNA topoisomerase II in etoposide-resistant mutants of human cancer KB cells, Cancer Res. 51 (1991) 3951±3957. [21] N. Kaneda, H. Nagata, T. Furuta, T. Yakohura, Metabolism and pharmacokinetics of the camptothecin analogue CPT-11 in the mouse, Cancer Res. 50 (1990) 1715±1720. [22] Y. Kawato, M. Aonuma, Y. Hirota, H. Kuga, K. Sato, Intracellular roles of SN-38, a metabolite of the camptothecin derivative CPT-11, in the antitumor effect of CPT-11, Cancer Res. 51 (1991) 4187±4191. [23] F. Kanzawa, Y. Sugimoto, K. Minato, K. Kasahara, M. Bungo,

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

K. Nakagawa, Y. Fujiwara, L.F. Liu, N. Saijo, Establishment of a camptothecin analogue (CPT-11)-resistant cell line of human non-small cell lung cancer: Characterization and mechanism of resistance, Cancer Res. 50 (1990) 5919± 5924. Z.-S. Chen, M. Mutoh, T. Sumizawa, T. Furukawa, M. Haraguchi, A. Tani, N. Saijo, T. Kondo, S. Akiyama, An active ef¯ux system for heavy metals in cisplatin-resistant human KB carcinoma cells, Exp. Cell Res. 240 (1998) 312±320. R. Fujii, M. Mutoh, T. Sumizawa, Z.-S. Chen, A. Yoshimura, S. Akiyama, Adenosine triphosphate-dependent transport of leukotriene C4 by membrane vesicles prepared from cisplatinresistant human epidermoid carcinoma tumor cells, J. Natl. Cancer Inst. 86 (1994) 1781±1784. Y. Chuman, Z.-S. Chen, T. Sumizawa, T. Furukawa, M. Haraguchi, Y. Takebayashi, K. Niwa, K. Yamada, T. Aikou, S. Akiyama, Characterization of the ATP-dependent LTC4 transporter in cisplatin-resistant human KB cells, Biochem. Biophys. Res. Commun. 226 (1996) 158±165. T. Ishikawa, J.-J. Bao, Y. Yamane, K. Akimaru, K. Frindrich, C.D. Wright, M. T. Kuo, Coordinated induction of MRP/GSX pump and g-glutamylcysteine synthetase by heavy metals in human leukemia cells, J. Biol. Chem. 271 (1996) 14981± 14988. T. Sumizawa, Z.-S. Chen, Y. Chuman, K. Seto, T. Furukawa, M. Haraguchi, A. Tani, N. Shudo, S. Akiyama, Reversal of multidrug resistance-associated protein-mediated drug resistance by the pyridine analog PAK-104P, Mol. Pharmacol. 51 (1997) 399±405. V. Gekeler, W. Ise, H. Sanders, W.-R. Ulrich, J. Beck, The leukotriene LTD4 receptor antagonist MK571 speci®cally modulates MRP associated multidrug resistance, Biochem. Biophys. Res. Commun. 208 (1995) 345±352. X.-Y. Chu, Y. Kato, Y. Sugiyama, Multiplicity of biliary excretion mechanisms for irinotecan, CPT-11, and its metabolites in rats, Cancer Res. 57 (1997) 1934±1938. X.-Y. Chu, Y. Kato, K. Niinuma, K.-I. Sudo, H. Hakusui, Y. Sugiyama, Multispeci®c organic anion transporter (cMOAT) is responsible for biliary excretion of the camptothecin derivative irinotecan, CPT-11, and its metabolites in rats, J. Pharmacol. Exp. Ther. 281 (1997) 304±314.