Prognostic Significance of Multidrug Resistance Protein in Adult T-Cell Leukemia1

3120 Vol. 7, 3120–3126, October 2001 Clinical Cancer Research

Prognostic Significance of Multidrug Resistance Protein in Adult T-cell Leukemia1

Nobuhito Ohno, Ayako Tani, Zhe-Sheng Chen, prognosis in ATL. Multidrug resistance-reversing agents, Kimiharu Uozumi, Shuichi Hanada, such as ONO-1078 and MK571, that directly interact and inhibit the transporting activity of MRP1 may be useful for Suminori Akiba, Xiao-Qin Ren, treating ATL patients. Tatsuhiko Furukawa, Tomoyuki Sumizawa, Terukatsu Arima, and Shin-ichi Akiyama2 Department of Cancer Chemotherapy, Institute for Cancer Research INTRODUCTION 3 [N. O., A. T., Z-S. C., X. R., T. F., T. S., S-i. A.], Second Department ATL is thought to be caused by HTLV-I; however, the of Internal Medicine [N. O., K. U., S. H., T. A.], and Department of mechanisms of leukemogenesis are not fully understood. ATL is Public Health, Faculty of Medicine [S. A.], Kagoshima University, divided into four clinical subtypes (smoldering, chronic, lym- Kagoshima 890-8520, Japan phoma, and acute ATL; Ref. 1). Although smoldering ATL and chronic ATL have a mild clinical course without intensive ABSTRACT chemotherapy, the outcome of chemotherapy for acute and The response of adult T-cell leukemia (ATL) to chem- lymphoma ATL remains poor even when regimens of conven- otherapy is poor, and a major obstacle to successful treat- tional chemotherapy for malignant lymphoma (2) or regimens of ment is intrinsic or acquired drug resistance. To determine weekly exchanged, non-cross-resistant drugs (3) are performed. the clinical significance of multidrug resistance protein The poor response to chemotherapy may be attributed mainly to (MRP) 1 in ATL, we studied MRP1 expression and its the resistance of ATL cells to chemotherapeutic drugs. Leuke- association with clinical outcome. The expression of MRP1 mia cells are often refractory to chemotherapy at initial presen- mRNA in leukemia cells from 48 ATL patients was studied tation and acquire resistance when ATL patients are at their by slot blot analysis. The expression level of MRP1 mRNA in terminal or relapsed stages (4). One of the mechanisms for drug chronic-type ATL was significantly higher than that in resistance is overexpression of a membrane glycoprotein termed There was no correlation P-gp (4). Overexpression of P-gp in ATL cells was observed in .(0.033 ؍ lymphoma-type ATL (P between MRP1 expression and age, gender, WBC count, 8 of 20 ATL patients at initial presentation and in all ATL LDH, hypercalcemia, blood urea nitrogen, or performance patients at relapse (4). Enhanced MDR1 mRNA expression was status. However, the expression of MRP1 mRNA correlated observed in 9 of 10 HTLV-I-infected subjects (5). However, only with peripheral blood abnormal lymphocyte counts Ikeda et al. (6) could not detect MDR1 mRNA expression in 28 .The transporting activity of MRP1 was assessed ATL patients by semiquantitative reverse transcription-PCR .(0.008 ؍ P) using membrane vesicles. Membrane vesicles prepared from Another factor involved in MDR is MRP1. MRP1 cDNA ATL cells with high expression of MRP1 mRNA showed a was cloned by Cole et al. (7) from an Adriamycin-resistant lung cancer cell line (H69) and sequenced. MRP1 is a Mr 190,000 higher ATP-dependent leukotriene C4 uptake than did those with low expression of MRP1 mRNA. This uptake was al- membrane protein belonging to the ATP-binding cassette (ABC) transporter superfamily. It transports some glutathione, most completely inhibited by LTD4 antagonists ONO-1078 and MK571. In acute- and lymphoma-type ATL, high ex- glucuronate, and sulfate conjugates (8) and confers MDR (9). pression of MRP1 mRNA at diagnosis correlated with PBMNCs from acute- and chronic-type ATL patients reportedly shorter survival, and Cox regression analysis revealed that expressed significantly higher levels of MRP1 and lung resis- MRP1 expression was an independent prognostic factor. tance-associated protein mRNA than did normal PBMNCs (6). These findings suggest that functionally active MRP1 is In this study, we show that high expression of MRP1 was expressed in some ATL cells and that it is involved in drug associated with unfavorable clinical outcome in ATL. MRP1 resistance and has a possible causal relationship with poor expressed in ATL cells has transporting activity that is inhibited by some agents. These agents may be useful in reversing MRP1- mediated MDR in ATL cells.

Received 11/07/00; revised 6/19/01; accepted 6/27/01. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked 3 The abbreviations used are: ATL, adult T-cell leukemia; MRP, mul- advertisement in accordance with 18 U.S.C. Section 1734 solely to tidrug resistance protein; HTLV-I, human T-cell lymphotrophic virus indicate this fact. type I; P-gp, P-glycoprotein; PBMNC, peripheral blood mononuclear 1 Supported by grants from the Ministry of Education, Science, and cell; MDR, multidrug resistance; PS, performance status; cMOAT, Culture and the Ministry of Health and Welfare, Japan. canalicular multispecific organic anion transporter; LT, leukotriene; 2 To whom requests for reprints should be addressed, at Institute for G3PDH, glyceraldehyde-3-phosphate dehydrogenase; PVDF, polyvi- Cancer Research, Faculty of Medicine, Kagoshima University, Sakura- nylidene difluoride; VCR, vincristine; AML, acute myelogenous leuke- gaoka 8-35-1, Kagoshima 890-8520, Japan. Phone: 81-99-275-5488; mia; LDH, lactate dehydrogenase; BUN, blood urea nitrogen; DIG, Fax: 81-99-265-9687; E-mail: [email protected]. digoxigenin.

Table 1 Patient characteristics at Ϫ110°C. We also obtained PBMNCs from healthy non- No. of patients 48 HTLV-I carriers as healthy controls. Sex Cell Lines and Cell Culture. Human epidermoid KB Male 26 carcinoma cells were grown as monolayers in MEM (Nissui Female 22 Seiyaku, Tokyo, Japan) containing 10% newborn calf serum Age (yrs) (Sera-Lab, Crawley Down, United Kingdom), 1 mg/ml bac- Median 65.5 Range 41–86 topeptone (Difco, Detroit, MI), 0.292 mg/ml glutamine, and 100 Subtype IU/ml penicillin. KB cells were subcloned twice, and a single Acute 32 recloned line, KB-3-1 (11), was used as a drug-sensitive paren- Chronic 9 tal cell line. CV60 cells (12) overexpress MRP1 and were Lymphoma 7 ␮ WBC (/␮l) maintained in MEM with 1 g/ml cepharanthine and 60 ng/ml Median 18,800 VCR. KB-C2 cells that overexpress P-gp were originally iso- Range 3,600–199,700 lated from KB-3-1 cells with increasing concentrations of col- Abnormal lymphocyte (/␮l) chicine and maintained in medium containing 2 ␮g/ml colchi- Median 6,641 cine. KB/MRP1 (13) was obtained from Dr. K. Ueda (Kyoto Range 0–171,742 LDH (IU/liter) University, Kyoto, Japan). LLC-PK1/cMOAT cells, pig kidney Median 1,191 epithelial LLC-PK1 cells transfected with human cMOAT Range 210–20,444 cDNA, were obtained from Dr. M. Kuwano (Kyushu Univer- Hypercalcemia sity, Fukuoka, Japan). KUT-1 (14) is a HTLV-I-infected cell No 38 Yes 10 line established from the PBMNCs of an ATL patient. BUNa Chemicals. 4-Oxo-8-[p-(4-phenylbutyloxy)benzoylamino]- Normal 30 2-(-tetrazol-5-yl)-4H-1-benzopyran hemihydrate (ONO-1078) High 8 was obtained from Ono Pharmaceutical Co. Ltd, and 3-([{3-(2- PS [7-chloro-2-quinolinyl]ethenyl)phenyl}-{(3-dimethylamino-3- 011 119oxopropyl)-thio}-methyl]thio)propanoic acid (MK571; Ref. 15) 27was kindly provided by Dr. A. W. Ford-Hutchinson (Merck- 32Frosst Center for Therapeutic Research, Pointe Claire-Dorval, b 4 9 Canada). QCRL-1 and QCRL-3 (16), monoclonal antibodies a Data were not available from all patients. against murine MRP1, were obtained from Dr. Susan P. C. Cole. b Caused by hypercalcemia. 3 [ H]LTC4 (150 Ci/mmol) was from DuPont New England Nu- clear (Boston, MA). All other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO). Probe Preparation. To prepare RNA probes for North- MATERIALS AND METHODS ern and slot blotting, the 3Ј-end PstI fragment (660 bp) of Patients. Between July 1989 and July 1998, we studied human MRP1 cDNA and a 982-bp of the human G3PDH cDNA 48 ATL patients (27 males and 21 females) with a mean age of fragment corresponding to G3PDH nucleotides 6 through 987 62.5 years (age range, 41–86 years). According to previously obtained by reverse transcription-PCR and a 832-bp fragment reported diagnostic criteria (1), 32 of these patients had acute- (nucleotides 1885 through 2716) of the human MDR1 cDNA type ATL, 9 had chronic-type ATL, and 7 had lymphoma-type were inserted into the pT7T3 18U multifunctional Phagemid ATL. Ten patients had hypercalcemia (corrected Ca level with and Bluescript II SKϩ (Pharmacia, Uppsala, Sweden) vectors. serum albumin, Ͼ5.5 meq/liter). PS was based on the five-grade After the vectors were linearized, single-strand RNA probes scale of the WHO. During this study, we treated acute- and were prepared using a DIG RNA labeling kit (Boehringer lymphoma-type ATL patients using combined chemotherapy Mannheim, Mannheim, Germany). regimens such as the response-oriented multidrug protocol (3), Slot and Northern Blot Analyses. Total cellular RNA CHOP (cyclophosphamide, doxorubicin, VCR, and prednisone) was extracted in a single step using Trizol reagent (Life Tech- protocol, or LSG15 protocol (10). There were no significant nologies, Inc., Rockville, MD). The integrity of isolated RNA survival differences among these protocols in this study (data was monitored by ethidium bromide staining after gel electro- not shown). Response was evaluated according to previously phoresis. For slot blotting, 1 and 5 ␮g of RNA were applied to reported criteria (10). All chronic-type ATL patients did not a nylon membrane (Hybond Nϩ) under a vacuum, and the require treatment with intensive chemotherapy. After informed membrane was dried at room temperature and then fixed with consent, 43 samples were obtained at initial presentation, and 5 UV irradiation. The membrane was prehybridized for 1 h at samples were obtained at relapse. 68°C in 50% formamide, 5ϫ SSC [1ϫ SSC ϭ 0.15 M NaCl/15 Leukemia Cells. PBMNCs separated by Ficoll-Conray mM sodium citrate (pH 7.0)], 1% SDS, and 0.05 mg/ml yeast density gradient centrifugation were collected from 26 patients tRNA. The membrane was then hybridized for 16 h at 68°C in with acute-type ATL and 9 patients with chronic-type ATL, and 50% formamide, 1% SDS, 5ϫ SSC, 0.05 mg/ml tRNA, and 100 lymph node samples were taken from 6 patients with acute-type ng/ml DIG-labeled RNA probe. ATL and 7 patients with lymphoma-type ATL. The character- After hybridization, the membrane was washed twice with istics of the patients are summarized in Table 1. All samples 2ϫ SSC and 0.1% SDS for 5 min at room temperature and then contained Ͼ80% abnormal lymphocytes, and they were stored washed twice with 0.1ϫ SSC and 0.1% SDS for 1 h at 68°C.

Fig. 1 Slot blots. Total RNA (5 ␮g and 1 ␮g) was blotted on a membrane and hybridized with MRP1 or G3PDH RNA probe.

Fig. 2 Northern blots. One ␮g of mRNA was applied to each well and hybridized with MRP1 RNA probe. The position of the 6.5-kb MRP1 The blot was detected using alkaline phosphatase-labeled anti- mRNA is indicated. A G3PDH RNA probe was used as the control probe. DIG antibody and chemiluminescent substrate according to the manufacturer’s instructions (Boehringer Mannheim, Mannheim, Germany). The density of the MRP1 mRNA signal on the autoradiograms was determined using the GS-525 Molecular mM Tris-HCl (pH 8.0), and 0.05% Tween 20] for1hatroom Imager System (Bio-Rad, Hercules, CA), and the density of the temperature and then incubated for 4 h with 40-fold diluted signals for 5 ␮g of total RNA from KB-3-1 cells was arbitrarily monoclonal antibody against MRP1, 10-fold diluted mono- assigned a value of 1 unit. The data are shown relative to MRP1 clonal antibody against cMOAT, or 1 ␮g/ml C219 in buffer C mRNA in KB-3-1 cells. containing 3% skimmed milk. The membrane was washed three For Northern blotting, polyadenylated RNA was extracted times with buffer C and then incubated for 1 h with 1000-fold from total RNA with a BioMag mRNA Purification Kit (Per- diluted antimouse immunoglobulin whole antibody labeled with septive Biosystem, Framingham, MA). Polyadenylated RNA (1 horseradish peroxidase (Amersham, Buckinghamshire, United ␮g) was resolved by electrophoresis on a formaldehyde-agarose Kingdom). The PVDF membrane was rinsed once for 15 min denaturing gel. After blotting onto a nylon membrane (Hybond and four times for 5 min with buffer C and then evenly coated Nϩ), hybridization was carried out with the same protocol used using the ECL Western blotting detection system (Amersham) for slot blot analysis. for 1 min. The membrane was immediately exposed to Fuji Membrane Vesicle Preparation. Membrane vesicles Rx-U medical X-ray film at room temperature for various pe- were prepared from KUT-1, LLC-PK1/cMOAT, and ATL cells riods in a film cassette. 3 as described previously (17). Cells were washed once and Transport of [ H]LTC4 into Membrane Vesicles. scraped into PBS containing 1% aprotinin (Sigma Chemical LTC4 uptake by the vesicles was measured by filtration as Co.). The cells were washed in PBS, centrifuged (4,000 ϫ g for described by Ishikawa (19). The standard incubation medium 10 min), and then suspended in buffer A [1 mM Tris-HCl (pH contained membrane vesicles (25 ␮g of protein), 0.25 M sucrose,

7.5), 0.025 M sucrose, and 0.2 mM CaCl2] and equilibrated at 10 mM Tris-HCl (pH 7.4), 10 mM MgCl2,1mM ATP or AMP, 2 4°C under a nitrogen pressure of 30 kg/cm for 15 min. The 10 mM creatine phosphate, and 100 mg/ml creatine kinase in a mixture was then diluted twice with buffer B [10 mM Tris-HCl final volume of 50 ␮l. The reaction was carried out at 37°C and (pH 7.5) and 0.25 M sucrose] and centrifuged at 1,000 ϫ g for stopped with 1 ml of ice-cold stop solution [0.25 M sucrose, 100 10 min to remove nuclei and unlysed cells. EDTA was added to mM NaCl, and 10 mM Tris-HCl (pH 7.4)]. The diluted samples the lysed cell suspension to a final concentration of 1 mM. The were passed through Millipore filters (GVWP, 0.22-mm pore supernatant was layered onto a 35% sucrose cushion containing size) under light suction. The filters were washed, and radioac- 0.01 M Tris-HCl (pH 7.5) and 1 mM EDTA and centrifuged for tivity was measured by liquid scintillation counting. 30 min at 16,000 ϫ g. The interface was collected and centri- Drug Accumulation. Intracellular VCR accumulation in fuged for 45 min at 100,000 ϫ g. The vesicle pellet was KUT-1 and ATL cells was measured as described previously 3 resuspended in buffer B using a 23-gauge needle. Vesicles were (20). Cells were incubated with 1 mM [ H]VCR in RPMI 1640 stored at Ϫ80°C. with or without ONO-1078 (100 ␮M) and MK571 (20 ␮M) for Immunoblot Analysis. QCRL-1 was used to detect 60 min at 37°C. After washing twice with ice-cold PBS, cells MRP1. A monoclonal antibody generated against amino acids were solubilized in 1% Triton X-100 and 0.2% SDS in 10 mM 1340–1541 of rat cMOAT protein (18) was used to detect phosphate buffer (pH 7.4), and the radioactivity was determined cMOAT. A monoclonal antibody against P-gp, C219, was used Statistical Analysis. Differences between groups were to detect P-gp. Membrane vesicles (50 ␮g of protein) were analyzed by one-way ANOVA or Student’s t test. P Ͻ 0.05 was mixed with an equal volume of SDS sample buffer consisting of considered significant. We used the Kaplan-Meier method to 125 mM Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, and 0.005% estimate survival rates and the log-rank test to compare the two bromphenol blue; resolved by electrophoresis on SDS-7.5% groups for difference in survival rates. Survival duration was (w/v) polyacrylamide minigels; and transferred to PVDF mem- measured from the time of chemotherapy to the time of death. branes (Immobilon-P; Millipore, Bedford, MA) using a Trans- The Cox proportional hazards model was used in the univariate blot SD apparatus (Bio-Rad). Thereafter, the membranes were and multivariate survival analyses. Maximum likelihood param- blocked with 3% skimmed milk in buffer C [0.35 M NaCl, 10 eter estimates and likelihood ratio statistics were obtained. We

Fig. 3 MRP1 mRNA expression. MRP1 mRNA expression is shown (A) in all ATL and healthy control and (B) in ATL subtypes. C, the correlation between MRP1 mRNA expression and absolute abnormal lymphocyte cell counts in peripheral blood. The expression levels were examined by slot blotting. Boxes correspond to the interquartile range. Lines in the boxes represent median values. The vertical lines represents the 5th and 95th percentiles, and the open circles represent the outliers.

3 ␮ Fig. 4 A, ATP-dependent [ H]LTC4 transport into membrane vesicles (25 g of protein) prepared from ATL samples 4 and 26 and KUT-1 and LLC-PK1/cMOAT cells. Cells were incubated at 37°Cin50␮l of incubation medium containing 0.01 M Tris-HCl (pH 7.5), 0.25 M sucrose, 10 mM ␮ 3 creatine phosphate, 100 g/ml creatine kinase, and [ H]LTC4 in the presence or absence of 1 mM ATP for 10 min. ATP-dependent uptakes of 3 f 3 ` o 3 [ H]LTC4 without ( ) or with 100 mM ONO-1078 ( ), 100 mM MK571( ), or QCRL-3 ( ) are shown. ATP-dependent [ H]LTC4 uptake was determined from the difference in the radioactivities incorporated into the vesicles in the presence and absence of ATP. Values are the mean Ϯ SE of triplicate experiments. Values of LLC-PK1/cMOAT cells were the mean of duplicate experiments. B, immunoblots for MRP1, cMOAT, and P-gp. Membrane vesicles (50 ␮g of protein) were resolved by SDS-PAGE and transferred to a PVDF membrane. The transferred proteins were immunoblotted with each antibody.

calculated Wald-type confidence intervals. All Ps presented ranged from 0.10–8.75 unit(s) (median, 1.53 units). The expres- were two-sided. sion levels were not significantly different from those in healthy controls (Fig. 3A). The median expression level (Ϯ SD) of RESULTS acute, chronic, and lymphoma ATL was 1.53 Ϯ 2.12, 2.70 Ϯ MRP1 mRNA Expression in ATL Cells. The levels of 2.14, and 1.06 Ϯ 0.62, respectively. The expression levels MRP1 mRNA in 48 ATL samples, in KB cell lines, and in in chronic-type ATL were significantly higher than those in KUT-1 were examined by slot blotting. The results of typical lymphoma-type ATL (P ϭ 0.033; Fig. 3B). slot blots are shown in Fig. 1. The density of the signal for 5 ␮g Next we studied whether MRP1 expression was correlated of total RNA from KB-3-1 cells was arbitrarily assigned a value with clinical and laboratory parameters. The expression of of 1 unit. ATL cells from cases 1, 13, 24, 27, 38, 42, 44, and 46 MRP1 mRNA was unrelated to age, gender, WBC counts, LDH, expressed 0.24-, 1.38-, 3.70-, 8.75-, 2.71-, 0.01-, 0.24-, and hypercalcemia, BUN, and PS (data not shown). However, 1.08-fold higher levels of MRP1 mRNA, respectively, than did MRP1 expression was significantly correlated with peripheral KB-3-1 cells. A healthy control was 0.87 (Fig. 1). blood abnormal lymphocyte counts (P ϭ 0.008; Fig. 3C). The The specificity of the probe was confirmed by Northern blots. expression of MRP1 mRNA was higher in patients with periph- A 6.5-kb MRP1 mRNA was detected in the examined samples eral blood abnormal lymphocyte counts of Ͼ4000/␮l than it was (samples 14, 25, and 46) and in CV60 and KB-3-1 cells (Fig. 2). in the rest of the patients. 3 The expression levels of MRP1 mRNA in all ATL samples [ H]LTC4 Uptake in Membrane Vesicles. To investi- were quantified by slot blot analysis. The expression levels gate whether the expressed MRP1 was functional in ATL cells,

Fig. 5 [3H]VCR accumulation in ATL samples 24 and 25 and in KUT-1 cells. The intracellular levels of VCR in the cells after incubation without (f) or with 100 ␮M ONO-1078 (Ⅺ)for1hat37°C were ,ء .determined. Values are the means of triplicate experiments; bars, SE P Ͻ 0.05. N.S., not significant.

3 we examined ATP-dependent [ H]LTC4 uptake in membrane vesicles from KUT-1 and PBMNCs from two ATL patients (patients 4 and 25; Fig. 4A). The number of ATL cells from other patients was not sufficient to prepare membrane vesicles. MRP1 mRNA expression in patient 4, patient 25, and KUT-1 cells was 0.54 unit, 5.29 units, and 0.53 unit, respectively. 3 Fig. 6 Overall survival. Kaplan-Meier survival estimates stratified by ATP-dependent [ H]LTC4 uptake by membrane vesicles from (A) subtype and (B) MRP1 expression. Survival of the MRP1-high the ATL sample of patient 25 was significantly higher than that patients was shorter than that of the MRP1-low patients. from the ATL sample of patient 4 and that from KUT-1 cells (P ϭ 0.023 and 0.017, respectively). cMOAT also transports

LTC4. Although MRP1 but not cMOAT was detected by im- mumoblot analysis in the membrane vesicles from the patient 25 according to their MRP1 mRNA expression at diagnosis and ATL sample (Fig. 4B), we examined LTC transport in the 4 compared their survival periods (Fig. 6B). We designated the presence of a specific antibody against MRP1, QCRL-3, to specimens as MRP1-high when MRP1 mRNA expression was confirm that LTC4 was transported by MRP1 and not by cMOAT in the ATL membrane vesicles. QCRL-3 (0.148 ␮g higher than twice the median expression level of healthy con- QCRL-3/10 mg membrane vesicles) reduced LTC uptake by trols and as MRP1-low when it was lower. The survival period 4 of MRP1-high patients was compared with that of MRP1-low 90.0% in the patient 25 ATL sample, but ATP-dependent LTC4 uptake by membrane vesicles from LLC-PK1/cMOAT cells that patients, and the prognosis of MRP1-high patients was worse overexpress cMOAT was not inhibited by QCRL-3 (Fig. 4A). than that of the MRP1-low patients (median, 4.4 versus 6.1 ϭ ONO-1078 and MK571, two MDR-reversing agents that inhibit months; P 0.019; Fig. 6B). We also studied the expression the transporting activity of both MRP1 and cMOAT, reduced of MDR1 mRNA in 26 ATL patients (19 acute-type and 7 lymphoma-type ATL patients) and 3 healthy controls by slot the ATP-dependent LTC4 uptake into membrane vesicles from both the patient 25 ATL sample (P ϭ 0.016 and 0.016, respec- blot analysis and designated the specimens as MDR1-high and tively) and LLC-PK1/cMOAT cells. MDR1-low by following the same criteria as described for Accumulation of [3H]VCR. We examined the accumu- MRP1 mRNA expression. The survival periods of these two lation of VCR in cells from two ATL patients and in KUT-1 groups were not significantly different (data not shown). High cells with or without the MDR-reversing agents. In the MRP1- abnormal lymphocyte counts were not associated with short positive ATL samples (samples 23 and 24; MRP1 mRNA ex- survival (data not shown). pression was 4.49 and 3.70), VCR accumulation was increased Next, we investigated whether MRP1 positivity is an in- by 100 ␮M ONO-1078 (P Ͻ 0.05), but it was not enhanced by dependent prognostic variable. A Cox proportional hazards ONO-1078 in KUT-1 cells (Fig. 5). model was constructed using established prognostic factors (21), Prognostic Relevance. We studied the correlation of MDR1 expression and MRP1 expression. Univariate analysis MRP expression and survival period. The survival curves of showed that BUN (P ϭ 0.028), PS (P ϭ 0.006), and MRP1 patients with acute-type ATL and patients with lymphoma-type (P ϭ 0.024) were significant prognostic factors (Table 2). Next, ATL were not significantly different (median, 5.5 versus 5.5 we performed a multivariate analysis that included all parame- months; P ϭ 0.444; Fig. 6A). Patients with chronic-type ATL ters that were significantly associated with outcome in the reportedly had a longer survival period than those with acute- univariative analysis. Multivariate survival analysis showed that type and lymphoma-type ATL (1). Therefore, we divided the 27 MRP1 (P ϭ 0.005) alone was a significant prognostic factor acute-type and 7 lymphoma-type ATL patients into two groups (Table 2).

Table 2 Result of survival analysis Univariate Multivariate No. of patients HRa 95% CI P HR 95% CI P Age (yrs) Ͻ65 15 1.00 Reference Ն65 19 1.18 0.576–2.403 0.655 LDH Ͻ1500 17 1.00 Reference Ն1500 17 1.50 0.735–3.065 0.264 BUNb Normal 20 1.00 Reference 1.00 Reference High 7 2.90 1.123–7.508 0.028 3.06 0.819–11.429 0.096 Hypercalcemia No 24 1.00 Reference Yes 10 1.49 0.699–3.160 0.304 PS 0 and 1 19 1.00 Reference 1.00 Reference 2–4c 15 3.09 1.392–6.858 0.006 1.46 0.438–4.839 0.541 MDR1 expressionb Low 20 1.00 Reference High 6 1.07 0.422–2.732 0.881 MRP1 expression Low 26 1.00 Reference 1.00 Reference High 8 2.72 1.139–6.471 0.024 5.78 1.691–19.772 0.005 a HR, hazard ratio; CI, confidence interval. Hazard ratios and 95% confidence intervals were obtained using the Cox proportional hazards model. b Data were not available from all patients. c Caused by hypercalcemia.

DISCUSSION sitive. The MRP1 gene is located at chromosome 16p13.13, and In our study, we demonstrated the clinical significance of the inversion results in loss of the MRP1 gene in some inv(16) MRP1 expression and that high MRP1 expression correlated AML patients (25). Legrand et al. (26) reported that functional with short survival in patients with acute-type and lymphoma- testing of MRP1 activity using calcein acetoxymethyl ester and type ATL. Abnormal lymphocyte counts were correlated with cyclosporin A revealed a significant prognostic value of MRP1 in the level of MRP1 expression but were not related to overall AML patients, suggesting that MRP1 is involved in drug resistance survival, and the survival difference between MRP1-high and in AML. MRP1-low patients cannot be explained by high abnormal lym- MRP1 in the plasma membrane (27) is thought to transport phocyte counts in patients with high MRP1. The survival period anticancer agents outside the cells. To assess the function of of patients with lymphoma-type ATL was not significantly MRP1, we used membrane vesicles from ATL cells to directly longer than that of patients with acute-type ATL in this study investigate the transporting activity of MRP1 in ATL cells.

(Fig. 6A). In chronic-type ATL, the samples showed the highest LTC4 is a good substrate for MRP1 and is actively transported mean expression level of MRP1 among the three subtypes. into MRP1-expressing membrane vesicles (28). cMOAT also

Because chronic-type ATL patients did not require treatment with transports LTC4 (29). LTC4 was transported more actively into intensive chemotherapy in this study, it is unclear whether overex- MRP1-high ATL membrane vesicles than into MRP1-low ATL pression of MRP1 is involved in drug resistance in these patients. membrane vesicles. A monoclonal antibody against MRP1,

In univariate and multivariate analyses, overexpression of QCRL-3, inhibited ATP-dependent LTC4 uptake by membrane MRP1 was a poor prognostic factor in this study. These findings vesicles prepared from MRP1-positive ATL cells but did not strongly suggest a possible causal relationship between MRP1 inhibit LTC4 uptake by membrane vesicles prepared from LLC- expression and poor survival of patients with acute-type and PK1/cMOAT cells. In the ATL sample, MRP1, but not cMOAT lymphoma-type ATL. Additional clinical trials with reversing or P-gp, was detected by immunoblotting. These findings indi- agents are required to clarify the role of MRP1. cate that the MRP1 expressed in ATL cells can actively trans-

The expression of MRP1 mRNA in leukemia cells from port LTC4. AML, acute lymphocytic leukemia, and chronic lymphocytic ONO-1078 is an antagonist of the LTD4 receptor and is leukemia patients was also reported (22–26). In de novo AML, used clinically for treatment of asthma patients. By directly expression of MRP1 was reportedly variable, and MRP1 was interacting with MRP1, ONO-1078 completely inhibited ATP- commonly overexpressed in leukemic cells from relapsed, sec- dependent LTC4 uptake into membrane vesicles prepared from ondary, or refractory AML and relapsed acute lymphocytic CV60 cells that overexpress MRP1 (28). MK571 is another leukemia patients (22, 23). In chronic lymphocytic leukemia, LTD4 receptor antagonist with a structure that differs from that 71% of patients had high expression levels of the MRP1 gene of ONO-1078, and it also completely reversed VCR resistance (24). AML associated with the inversion chromosome 16 in MRP1-mediated MDR cells (30). Both ONO-1078 and

[inv(16)(p13;q22)] has a favorable prognosis and is chemosen- MK571 almost completely inhibited ATP-dependent LTC4 up-

take by membrane vesicles prepared from MRP1-positive ATL 13. Taguchi, Y., Yoshida, A., Takada, Y., Komano, T., and Ueda, K. cells. Moreover, VCR accumulation in MRP1-positive ATL Anti-cancer drugs and glutathione stimulate vanadate-induced trapping cells from two cases was enhanced by ONO-1078. ONO-1078 of nucleotide in multidrug resistance-associated protein (MRP). FEBS Lett., 401: 11–14, 1997. or MK571 in combination with anticancer agents may be useful 14. Hanada, S., Tsubai, F., and Namba, Y. The characteristics of T-cell for treating patients with MRP1-positive ATL cells. lines derived from peripheral blood of patients with adult T-cell leuke- In conclusion, MRP1 may be one of the factors that causes mia-lymphoma. Recent Adv. RES (Reticuloendothel. Syst.) Res., 25: drug resistance in ATL patients and affects the response to 124–133, 1985. chemotherapy in ATL. Treatment with MRP1 modulators 15. Jones, T. R., Zamboni, R., Belley, M., Champion, E., Charette, L., Ford-Hutchinson, A. W., Frenette, R., Gauthier, J. Y., Leger, S., Mas- may be useful in MRP1-mediated drug resistance in patients son, P., McFarlance, S., Piechutta, H., Rokach, J., Williams, H., Young, with ATL. R. M., DeHaven, R. N., and Pong, S. S. Pharmacology of L-660,711 (MK 571): a novel potent and selective leukotriene D4 receptor antag- ACKNOWLEDGMENTS onist. Can. J. Physiol. Pharmacol., 67: 17–28, 1989. 16. Hipfner, D. R., Gauldie, S. D., Deeley, R G., and Cole, S. P. C. We thank Dr. M. Kuwano for LLC-PK1/cMOAT cells and Dr. K. Detection of the Mr 190,000 multidrug resistance protein, MRP, with Ueda for KB/MRP1 cells. We also thank Dr. A. W. Ford-Hutchinson for monoclonal antibodies. Cancer Res., 54: 5788–5792, 1994. the gift of MK571, Drs. M. Kool and P. Borst for monoclonal antibodies 17. Cornwell, M. M., Gottesman, M. M., and Pastan, I. H. Increased against cMOAT, and Dr. S. P. C. Cole for monoclonal antibodies vinblastine binding to membrane vesicles from multidrug-resistant KB against MRP1. cells. J. Biol. Chem., 261: 7921–7928, 1986. 18. Paulusma, C. C., Bosma, P. J., Zaman, G. J. R., Bakker, C. T. M., REFERENCES Otter, M., Scheffer, G. L., Scheper, R. J., Borst, P., and Oude Elferink, 1. Shimoyama, M., and members of the Lymphoma Study Group R. P. J. Congenital jaundice in rats with a mutation in a multidrug resistance-associated protein gene. Science (Wash. DC), 271: 1126– (1984–1987). Diagnostic criteria and classification of clinical subtypes 1128, 1996. of adult T-cell leukemia-lymphoma. A report from the Lymphoma 2ϩ Study Group. Br. J. Haematol., 79: 428–437, 1991. 19. Ishikawa, T. ATP/Mg -dependent cardiac transporter system for glutathione S-conjugates. J. Biol. Chem., 264: 17343–17348, 1989. 2. Lymphoma Study Group. Final results of cooperative study of VEPA (vincristine, cyclophosphamide (Endoxan), predonisolone and Adriamy- 20. Sumizawa, T., Chen, Z. S., Chuman, Y., Seto, K., Furukawa, T., cin) therapy in advanced non-Hodgkin’s lymphoma: relation between T Haraguchi, M., Tani, A., Shudo, N., and Akiyama, S. Reversal of or B cell phenotype response. Jpn. J. Clin. Oncol., 12: 227–238, 1982. multidrug resistance-associated protein-mediated drug resistance by the 3. Uozumi, K., Hanada, S., Ohno, N., Ishitsuka, K., Shimotakahara, S., pyridine analog PAK-104P. Mol. Pharmacol., 51: 399–405, 1997. Otsuka, M., Chyman, Y., Nakahara, K., Takeshita, T., Kuwazuru, Y., 21. Lymphoma Study Group (1984–1987). Major prognostic factors of Saito, T., Makino, T., Iwahashi, M., Utsunomiya, A., and Arima, T. patients with adult T-cell leukemia-lymphoma: a cooperative study. Combination chemotherapy (RCM protocol: response-oriented cyclic Leuk. Res., 15: 81–90, 1991. multidrug protocol) for the acute or lymphoma type adult T-cell leuke- 22. Shneider, E., Cowan, K. H., Barder, H., Toomey, S., Schwartz, mia. Leuk. Lymphoma, 18: 317–323, 1995. G. N., Karp, J. E., and Kaufmann, S. H. Increased expression of the 4. Kuwazuru, Y., Hanada, S., Furukawa, T., Yoshimura, A., Sumizawa, multidrug resistance-associated protein gene in relapsed acute leukemia. T., Utsunomiya, A., and Arima, T. Expression of P-glycoprotein in adult Blood, 85: 186–193, 1995. T-cell leukemia cells. Blood, 76: 2065–2071, 1990. 23. Hart, S. M., Ganeshaguru, K., Hoffbrand, A., Prentice, H. G., and 5. Lau, A., Nightingale, S., Taylor, G. P., Gant, T. W., and Cann, A. Mehta, A. B. Expression of the multidrug resistance-associated protein Enhanced MDR1 gene expression in human T-cell leukemia virus- (MRP) in acute leukemia. Leukemia (Baltimore), 8: 2163–2168, 1994. infected patients offers new prospects for therapy. Blood, 91: 2467– 24. Bureger, H., Nooter, K., Zaman, G. J. R., Sonneveld, P., Winger- 2474, 1998. den, K. E., Oostrum, R., and Stoter, G. Expression of the multidrug 6. Ikeda, K., Oka, M., Yamada, Y., Soda, H., Fukuda, M., Kinoshita, resistance-associated protein (MRP) in acute and chronic leukemias. A., Tsukamoto, K., Noguchi, Y., Isomoto, H., Takashima, F., Murase, Leukemia (Baltimore), 8: 990–997, 1994. K., Kamihira, S., Tomonaga, M., and Kohno, S. Adult T-cell leukemia 25. Kuss, B. J., Deeley, R. G., Cole, S. P. C., Willman, C. L., Kopecky, cells over-express the multidrug-resistance-protein (MRP) and lung- K. J., Wolman, S. R., Eyre, H. J., Lane, S. A., Nancarrow, J. K., resistance-protein (LRP) genes. Int. J. Cancer, 82: 599–604, 1999. Whitmore, S. A., and Callen, D. F. Deletion of gene for multidrug 7. Cole, S. P., Bhardwaj, G., Gerlach, J. H., Mackie, J. E., Grant, C. E., resistance in acute myeloid leukemia with inversion in chromosome 16: Almquist, K. C., Stewart, A. J., Kurze, U., Duncan, A. M. V., and prognostic implications. Lancet, 343: 1531–1534, 1994. Deeley, R. G. Overexpression of a transporter gene in a multidrug- 26. Legrand, O., Simonin, G., Perrot, J. Y., Zittoun, R., and Marie, J. P. resistant human lung cancer cell line. Science (Wash. DC), 258: 1650– Pgp and MRP activities using calcein-AM are prognostic factors in adult 1654, 1992. acute myeloid leukemia patients. Blood, 91: 4480–4488, 1998. 8. Jedlitschky, G., Leier, I., Buchholz, U., Center, M., and Keppler, D. 27. Hipfner, D. R., Almquist, K. C., Leslie, E. M., Gerlach, J. H., Grant, ATP-dependent transport of glutathione S-conjugates by the multidrug C. E., Deeley, R. G., and Cole, S. P. C. Membrane topology of the resistance-associated protein. Cancer Res., 54: 4833–4836, 1994. multidrug resistance protein (MRP). J. Biol. Chem., 272: 23623–23630, 9. Grant, C. E., Valdimarsson, G., Hipfner, D. R., Almquist, K. C., 1997. Cole, S. P. C., and Deeley, R. G. Overexpression of multidrug-resist- 28. Nagayama, S., Chen, Z. S., Kitazono, M., Takebayashi, Y., Niwa, ance-associated protein (MRP1) increases resistance to natural product K., Yamada, K., Tani, A., Haraguchi, M., Sumizawa, T., Furukawa, T., drugs. Cancer Res., 54: 357–361, 1994. Aikou, T., and Akiyama, S. Increased sensitivity to vincristine of MDR 10. Hanada, S., Utsunomiya, A., Suzuki, S., Uozumi, K., Makino, T., cells by leukotriene D4 receptor antagonist, ONO-1078. Cancer Lett., and Arima, T. Treatment of adult T-cell leukemia. Cancer Chemother. 130: 175–182, 1998. Pharmacol., 40: S40–S50, 1997. 29. Evers, R., Kool, M., Deemter, L. V., Janssen, H., Calafat, J., Oomen, 11. Akiyama, S., Fojo, A. T., Hanover, J. A., Pastan, I., and Gottesman, L. C. J. M., Paulusma, C. C., Oude Elferink, R. P. J., Baas, F., Schinkel, M. M. Isolation and genetic characterization of human KB cell lines A. H., and Borst, P. Drug export activity of the human multispecific organic resistant to multiple drugs. Somatic Cell Mol. Genet., 11: 117–126, 1985. anion transporter in polarized kidney MDCK cells expressing cMOAT 12. Sumizawa, T., Chuman, Y., Sakamoto, H., Iemura, K., Almquist, (MRP2) cDNA. J. Clin. Investig., 101: 1310–1319, 1998. K. C., Deeley, R. G., Cole, S. P. C., and Akiyama, S. Non-P-glycopro- 30. Gekeler, V., Ise, W., Sanders, K. H., Ulrich, W., and Beck, J. The tein-mediated multidrug-resistant human KB cells selected in medium leukotriene D4 receptor antagonist MK571 specifically modulates MRP containing Adriamycin, cepharanthine and mezerein. Somatic Cell Mol. associated multidrug resistance. Biochem. Biophys. Res. Commun., Genet., 20: 423–435, 1994. 208: 345–352, 1995.

Nobuhito Ohno, Ayako Tani, Zhe-Sheng Chen, et al.

Clin Cancer Res 2001;7:3120-3126.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/7/10/3120

Cited articles This article cites 30 articles, 13 of which you can access for free at: http://clincancerres.aacrjournals.org/content/7/10/3120.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/7/10/3120.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/7/10/3120. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.