The Role of Breast Cancer Resistance Protein in Acute Lymphoblastic Leukemia

Vol. 9, 5171–5177, November 1, 2003 Clinical Cancer Research 5171

Featured Article The Role of Breast Cancer Resistance Protein in Acute Lymphoblastic Leukemia

Sabine L. A. Plasschaert, 0.013). The influence of FTC on mitoxantrone accumulation > P ;0.52 ؍ Dorina M. van der Kolk, Eveline S. J. M. de Bont, correlated with ABCG2 protein expression (r ؍ Willem A. Kamps, Kuniaki Morisaki, 0.001; n 43). The increase in mitoxantrone accumulation, when FTC was added to cells treated with both PSC 833 and Susan E. Bates, George L. Scheffer, MK-571, correlated with the ABCG2 expression in B-lineage Rik J. Scheper, Edo Vellenga, and ALL but not in T-lineage ALL. Sequencing the ABCG2 gene Elisabeth G. E. de Vries1 revealed no ABCG2 mutation at position 482 in patients who Divisions of Pediatric Oncology and Hematology [S. L. A. P., accumulated more rhodamine after FTC. E. S. J. M. d. B., W. A. K.], Hematology [D. M. v. d. K., E. V.], and Conclusions: This study shows that ABCG2 is ex- Medical Oncology [E. G. E. d. V.], University Hospital Groningen, pressed higher and functionally more active in B-lineage Groningen 9713 GZ, the Netherlands; Division of Pathology, Free than in T-lineage ALL. University Medical Center, Amsterdam, the Netherlands [G. L. S., R. J. S.]; and National Cancer Institute, Medicine Branch, NIH, Bethesda, Maryland 20892 [S. E. B.] Introduction The prognosis of ALL2 has improved over the last decades, in children even more than in adults. Still, many ALL patients Abstract do not achieve a complete remission or develop a relapse (1, 2). Purpose: Overexpression of the transporter ABCG2, Resistance to chemotherapeutic agents plays an important role. also known as breast cancer resistance protein and mitox- MDR is defined as resistance to multiple, structurally and func- antrone resistance protein, can confer resistance to a variety tionally unrelated natural product drugs. One mechanism of of cytostatic drugs, such as mitoxantrone, topotecan, doxo- MDR is the overexpression of members of the ABC superfamily rubicin, and daunorubicin. This study analyzes the ABCG2 of membrane transporters. This family of transporters consists expression and activity in 46 human de novo acute lympho- of a number of proteins, including P-gp and the MRPs (MRP1 blastic leukemia B- and T-lineage (ALL) samples. and MRP2). Contradictory results have been observed with Experimental Design: ABCG2 expression was measured respect to the prognostic impact of P-gp expression and func- flow cytometrically with the BXP-34 monoclonal antibody. tional activity (3–8). For MRP1 protein and mRNA expression, ABCG2 functional activity was determined flow cytometri- neither a difference has been found between initial and relapse cally by measuring mitoxantrone accumulation in combina- samples, nor a correlation with complete response or survival tion with the ABCG2 inhibitor fumitremorgin C (FTC). To (3). We have studied previously the functional activity of MRP1 determine a possible effect of the transporters P-glycopro- and MRP2 in children and adults with ALL, and observed no tein and multidrug resistance-associated protein (MRP1 and correlation between functional activity and clinical outcome (9). MRP2) on mitoxantrone accumulation, the accumulation Another ATP-binding cassette transporter has been identi- was investigated in the presence of the P-glycoprotein inhib- fied in a MDR human breast cancer cell line MCF-7/AdrVp (10, itor PSC 833 and MRP inhibitor MK-571. The ABCG2 gene 11). This cell line shows an ATP-dependent reduction in the was sequenced to investigate the amino acid at position 482. intracellular accumulation of anthracycline anticancer drugs, in ؍ Results: In B-lineage ALL (n 23), the median BXP- the absence of overexpression of known MDR transporters such 34:IgG1 ratio was higher, namely 2.4 (range, 1.7–3.7), than as P-gp or MRP. This transporter protein was named ABCG2, ؍ ؍ in T-lineage ALL (n 23; 1.9; range, 1.2–6.6; P 0.003). and is also known as breast cancer resistance protein, placental The addition of FTC to mitoxantrone treatment caused a ABC transporter, or mitoxantrone resistance protein. Several median increase in mitoxantrone accumulation of 21% human cancer cell lines selected for resistance to mitoxantrone (range, 0–140%) in B-lineage ALL. In T-lineage ALL, this or topotecan have shown a marked ABCG2 overexpression. In ؍ FTC effect was less pronounced (5%; range, 0–256%; P addition, enforced expression of the full-length ABCG2 cDNA in MCF-7 breast cancer cells confers resistance to mitoxantrone, bisantrene, topotecan, and doxorubicin or daunorubicin in the presence of verapamil (11–14). The physiological function of Received 2/7/03; revised 6/25/03; accepted 7/2/03. ABCG2 may be similar to P-gp, because it is also located in the 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 18 U.S.C. Section 1734 solely to indicate this fact. This work was supported by a grant of the Foundation of Pediatric 2 The abbreviations used are: ALL acute lymphoblastic leukemia; MDR, Oncology Groningen (SKOG 99–01). multidrug resistance; MRP, multidrug-resistant protein; P-gp, P-glyco- 1 To whom requests for reprints should be addressed, at Department of protein; FTC, fumitremorgin C; AML, acute myelogenous leukemia; Internal Medicine, Division of Medical Oncology, University Hospital FACS, fluorescence-activated cell sorter; PE, phycoerythrin; MFI, me- Groningen, Hanzeplein 1, 9713 GZ Groningen, the Netherlands. Phone: dian fluorescence intensity; Rh123, rhodamine 123; CF, carboxyfluo- 31-503612821; Fax: 31-503614862; E-mail: [email protected]. rescein. Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2003 American Association for Cancer Research. 5172 ABCG2 Expression and Activity in ALL

blood-brain barrier, the placenta, the liver, and intestinal tract sone, daunorubicin, doxorubicin, methotrexate, and vincristine. (15). The spectrum of ABCG2 substrates shows an overlap with The adult patients received a regimen including the MDR drugs that of P-gp and MRP. All three of the transporters can extrude prednisone, doxorubicin, vincristine, methotrexate, and etopo- the anthracyclines, such as daunorubicin, doxorubicin, and epi- side (26). rubicin. P-gp (16), ABCG2, and probably MRP1 (17, 18) can Mononuclear cells from bone marrow or peripheral blood also transport mitoxantrone. There are indications that the spec- were isolated on a Ficoll-Isopaque (Nycomed, Oslo, Norway) trum of drugs transported by ABCG2 might differ in cell lines density gradient by centrifugation for 20 min at 800 ϫ g. The because of a mutation in the ABCG2 gene at amino acid cells were cryopreserved in RPMI 1640 supplemented with 10% position 482 (19). When a mutation is present in the ABCG2 FCS and 10% DMSO (Merck, Amsterdam, the Netherlands) and gene with threonine or glycine situated at position 482 instead of stored at Ϫ196°C. Upon analysis cells were thawed, centrifuged arginine, ABCG2 is able to transport rhodamine and doxorubi- in newborn calf serum (Life Technologies, Inc., Breda, the cin additionally (19) but not methotrexate (20). Netherlands), treated with DNase (Boehringer Mannheim, Mitoxantrone, methotrexate, daunorubicin, and doxorubi- Mannheim, Germany), and washed with RPMI 1640. Immuno- cin are used frequently during the induction or consolidation phenotyping showed that all of the samples contained Ͼ90% of 6 phase in ALL, indicating that ABCG2 might have clinical leukemic cells. For immunophenotyping, 0.5 ϫ 10 cells were impact in this patient group. The present study demonstrates that incubated with 5 ␮l of FITC- or phycoerythrin-labeled mouse the ABCG2 protein is expressed in ALL. The functional activity monoclonal antibodies to CD117, CD3, CD7 (all IQ Products, was measured by mitoxantrone accumulation in the absence or Groningen, the Netherlands), CD2, CD10, CD19, CD20, CD33, presence of the ABCG2 inhibitor FTC (21), and FTC in com- CD34, or IgG isotype-matched controls (all Becton Dickinson, bination with the inhibitors of the other transporters P-gp and Woerden, the Netherlands) for 30 min at 4°C. Subsequently, MRP1, which might obscure the ABCG2 function (12, 22). The cells were washed with RPMI 1640 and analyzed with a FAC- ABCG2 gene was sequenced in patients who accumulated more Star flow cytometer (Becton Dickinson Medical Systems, Sha- rhodamine after FTC, to investigate the amino acid at position ron, MA). The immature markers CD34 and CD117, expressed 482. Because ABCG2 expression is correlated with an immature only on progenitor cells, were analyzed in both B-lineage and phenotype in normal human hematopoietic cells (15, 23, 24) and T-lineage ALL. For B-lineage ALL samples, the B-lineage AML blasts (21), we also determined the immunophenotype of markers CD10, CD19, and CD20 were used. The T-lineage the patient samples. markers CD3 and CD7 were investigated in T-lineage ALL samples. Because the myeloid marker CD33 correlated with P-gp functional activity in an earlier study (9), CD33 expression Materials and Methods was also measured. Cell Lines. The drug-sensitive human breast cancer cell Flow Cytometric Detection of ABCG2 Protein Expres- line MCF-7 and its mitoxantrone-resistant counterpart MCF-7 sion. ABCG2 protein expression was measured with the MR, which overexpresses ABCG2 but not P-gp or MRP-1, were monoclonal antibody BXP-34 (27), which recognizes an internal cultured in RPMI 1640 (Bio Whittaker, Brussels, Belgium) epitope of the ABCG2 protein. Cells (0.5 ϫ 106) were incubated supplemented with 10% FCS (Hyclone, Logan, UT). The with FACS lysing solution (Becton Dickinson) to permeabilize MCF-7 MR cell line was cultured in the presence of 80 nM the cell membranes for 10 min at room temperature. Then the mitoxantrone. The human small cell lung cancer cell line GLC4, cells were incubated for 15 min at room temperature, with 1% the in vitro doxorubicin-selected MRP1-overexpressing subline goat serum in PBS [PBS: 140 mM NaCl, 9.0 mM

GLC4/ADR, and the MDR-1 transfected cell line GLC4/P-gp Na2HPO42H2O, and 1.3 mM NaH2PO42H2O (pH 7.4)] contain- were also cultured in RPMI 1640 with 10% FCS. The doxoru- ing 2% BSA. Subsequently, the cells were incubated for 60 min bicin-selected cell line was cultured in the presence of doxoru- at room temperature with 4 ␮l (1.0 ␮g) of BXP-34 mouse bicin (2 ␮M; Pharmacia and Upjohn, Woerden, the Netherlands) antibody or with 20 ␮l (1.0 ␮g) IgG1 isotype control. Cells were and the MDR-1 transfected cell line in the presence of 50 nM washed with PBS and incubated for 20 min with PE-conjugated Ј vincristine (Teva Pharma BV, Mijdrecht, the Netherlands). Cells rabbit-antimouse F(ab )2 fragments. PE fluorescence was meas- were cultured in drug-free medium at least 7 days before the ured on a FACSCalibur flow cytometer and expressed as MFI. analyses. The cell lines served as controls for the ABCG2 ABCG2 protein expression was expressed as adjusted for IgG1 expression and functional assay. control, i.e., the ratio of BXP-34:IgG1 antibody MFI. Control Patient Samples. From all of the patients newly diag- cell lines used to standardize the ABCG2 protein expression nosed with ALL at the University Hospital Groningen during assay included the MCF-7 cell line, expressing low levels of the period of 1989–2001, bone marrow and peripheral blood ABCG2, and the mitoxantrone-resistant subline MCF-7 MR, samples were collected for diagnostic evaluation and cryopre- expressing high levels of ABCG2. served. Patients were included in the study if a sufficient number Flow Cytometric Detection of Mitoxantrone Accumula- of preserved bone marrow or peripheral WBCs (Ͼ10 ϫ 106) tion. The capacity of ALL blasts to extrude mitoxantrone in was available. The Dutch Childhood Leukemia Study Group the absence or presence of the ABCG2 inhibitor FTC (12), the performed the immunophenotyping and cytomorphology of pe- P-gp inhibitor PSC 833 (provided by Novartis Pharma Inc., diatric samples. Adult patient samples were immunophenotyped Basel, Switzerland), the MRP inhibitor MK-571 (Merck Sharp, by the University Hospital Groningen. The children were treated Kirkland, PG, Quebec, Canada; Ref. 28), or combinations of according to the Dutch Childhood Leukemia Study Group pro- these inhibitors, was measured in a flow cytometric assay (21). tocols ALL-8 or -9 (25), containing the MDR drugs dexametha- The cell lines MCF-7 and MCF-7 MR served as controls. To

Table 1 Characteristics of ALL patients Unfavorable cytogenetics include the translocations t(9;22), t(4;11) and 11q23 rearrangements. Gender Agea Unfavorable Number (male/female) (years) WBCa,b (ϫ109/l) cytogenetics B-lineage ALL Total 23 11/12 14 (1–73) 49.6 (1.1–446.0) 4 (8 unknown) Children 12 6/6 9 (1–14) 49.0 (1.1–446.0) 1 (5 unknown) Adults 11 5/6 33 (19–73) 52.0 (8.3–332.2) 3 (3 unknown) T-lineage ALL Total 23 12/11 22 (3–75) 71.5 (3.5–590.0) 0 (3 unknown) Children 10 7/3 7 (3–15) 156.0 (34.1–590.0) 0 (1 unknown) Adults 13 5/8 28 (17–75) 17.6 (3.5–206.0) 0 (2 unknown) a The results represent median values with range. b WBC, white blood cell count at diagnosis.

exclude a possible inhibitory effect of FTC on P-gp or MRP, RNA Isolation, cDNA, and Genomic DNA Sequencing. mitoxantrone accumulation in combination with FTC was stud- The ABCG2 gene was sequenced to investigate the amino acid ied in the P-gp and MRP overexpressing cell lines GLC4/ADR at position 482 (arginine, threonine, or glycine; Ref. 19). Total and GLC4/P-gp. Cells (0.5 ϫ 106) were preincubated with the cellular RNA was extracted from leukemic cells using 1 ml of

inhibitors for 20 min at 37°C, 5% CO2, in the following com- TRIzol reagent (Life Technologies). RNA was extracted, pre- binations: RPMI 1640 alone (0.5 ml), RPMI 1640 plus FTC (10 cipitated, and washed according to the manufacturer’s protocol. ␮M), the maximum nontoxic doses of PSC 833 (2 ␮g/ml) and Subsequently, RNA was reverse transcribed in 20 ␮l reverse ␮ MK-571 (10 M), FTC plus PSC 833, FTC plus MK-571, PSC transcriptase buffer containing 10 mM DTT, 0.5 mM each of 833 plus MK-571, or FTC plus PSC 833 plus MK-571. There- dATP, dGTP, dCTP, and dTTP, 200 units of Moloney murine after, mitoxantrone (10 ␮M) was added, and the cells were leukemia virus reverse transcriptase, 5 units of RNase inhibitor, incubated for 60 min at 37°C, 5% CO2. Cells were washed with and 10 ng/␮l pd(N)6 random primers. Sequencing was per- ice-cold RPMI 1640. Mitoxantrone fluorescence was analyzed formed using the ABI PRISM Rhodamine Terminator Cycle with a FACSCalibur flow cytometer equipped with an argon Sequencing Ready Reaction kit with an ABI PRISM 377 DNA laser. The blast population was gated by forward and side scatter sequencer (Applied Biosystems, Foster City, CA). characteristics. The mitoxantrone fluorescence of 5000 gated Statistical Analysis. The Mann-Whitney U and Wil- events was logarithmically measured at a laser excitation wave- coxon nonparametric tests were performed to test for differences length of 635 nm through a 670-nm band-pass filter. Before in variables between children and adults, B-lineage ALL and assessing the mitoxantrone accumulation, the flow cytometer T-lineage ALL, and effects of combination of inhibitors on was calibrated for the specific wavelength using sphero particles mitoxantrone accumulation. The Spearman rank correlation was (Spherotech Inc., Libertyville, IL). Mitoxantrone accumulation used to determine correlation among ABCG2, P-gp, and MRP was expressed as MFI. The effects of the inhibitors were cal- activities and continuous variables. Ps Ͻ0.05 were considered culated as the ratio of the shift of MFI of the mitoxantrone significant. accumulation divided by the mitoxantrone accumulation without the inhibitors, multiplied by 100%. When an inhibitor caused a decrease in mitoxantrone accumulation, the effect of Results the inhibitor was equated to zero. The mitoxantrone accumula- Patient Characteristics. Samples from 46 patients with tion was determined in normal B and T lymphocytes from de novo ALL were studied, including 23 patients with B-lineage healthy donors in combination with CD2-FITC and CD19-PE and 23 patients with T-lineage ALL. The patient characteristics monoclonal antibodies. are summarized in Table 1. Flow Cytometric Detection of P-gp and MRP Func- ABCG2 Protein Expression. To standardize the tional Activity. Functional activity of the P-gp and MRP ABCG2 protein expression assay, the ABCG2 expression was transporter proteins was demonstrated as described previously studied in the ABCG2 low-expressing cell line MCF-7 and the (29). Shortly, P-gp activity was analyzed with Rh123 (Sigma ABCG2 overexpressing cell line MCF-7 MR. The monoclonal Chemical, Bornem, Belgium) in combination with PSC 833. antibody BXP-34 was used in combination with the IgG1 iso- Because a mutation in the ABCG2 gene can render the protein type control in a flow cytometric assay. A median BXP-34:IgG1 capable of transporting Rh123, the effect of FTC on Rh123 ratio of 8.3 (range, 5.4–11.1) was observed in the MCF-7 cell accumulation was also studied (19). To determine MRP (MRP1 line versus a high median expression of 86.4 (range, 59.8– and homologue MRP2) activity the compound CF (Sigma 112.7) in the MCF-7 MR cell line. Subsequently, the ABCG2 Chemical Co.) was used, in combination with MK-571. Rh123 protein expression was investigated in ALL blasts (Table 2). In and CF fluorescence of 5000 events was measured with a B-lineage ALL samples, the median BXP-34:IgG1 ratio was 2.4 FACSCalibur flow cytometer. The effects of the inhibitors were (range, 1.7–3.7). The T-lineage ALL samples showed a median expressed as a ratio of the shift of MFI divided by the MFI in expression of 1.9 (range, 1.2–6.6). The BXP-34:IgG1 ratios in unblocked cells, multiplied by 100%. all of the ALL samples were lower than the observed ratios in

Table 2 ABCG2 protein expression and mitoxantrone accumulation in ALL patients ABCG2 protein expression calculated as the BXP-34:IgG1 ratio. Shifts calculated as percentage of accumulation without inhibitors. The results represent median values with range. B-lineage ALL T-lineage ALL ABCG2 protein expression 2.4 (1.7–3.7) 1.9 (1.2–6.6) Mito accumulation (MFI) 148 (77–555) 108 (65–281) FTC shift 21 (0–140) 5 (0–256) PSC 833 shift 22 (6–70) 24 (0–81) MK-571 shift 16 (0–173) 18 (0–141) Rh123 accumulation PSC 833 shift 0 (0–197) 14 (0–362) CF accumulation MK-571 shift 192 (23–639) 239 (68–500)

the MCF-7 parent cell line. In B-lineage ALL, ABCG2 expression was higher compared with T-lineage ALL (P ϭ 0.003). The Fig. 1 Correlation between ABCG2 protein expression and effect of ABCG2 expression in B-lineage ALL showed also a trend FTC on mitoxantrone accumulation. ABCG2 expression was measured toward a higher expression in children than adults (P ϭ 0.071). with the BXP-34 monoclonal antibody; the effect of FTC was measured after 60 min of incubation with mitoxantrone with and without FTC. In T-lineage ALL, no difference was observed between the two age groups (P ϭ 0.574). Mitoxantrone Accumulation. The functional activity of ABCG2 was examined by measuring the intracellular mitox- The FTC effect on mitoxantrone accumulation correlated antrone accumulation, expressed as MFI, in the absence or positively with ABCG2 expression (r ϭ 0.52; P Ͻ 0.001; n ϭ presence of the ABCG2 inhibitor FTC. 43; Fig. 1). Even when the outlier is removed, the correlation In the MCF-7 MR cell line, the mean mitoxantrone accu- remains present (r ϭ 0.49; P ϭ 0.001; n ϭ 42). No difference mulation was lower than in the MCF-7 cell line (465 versus was observed between children and adults in mitoxantrone 1004), which was expected, because the overexpression of accumulation or in the effect of FTC (B-lineage P ϭ 0.268 and ABCG2 causes a decrease in mitoxantrone accumulation. P ϭ 0.566; T-lineage ALL P ϭ 0.352 and P ϭ 0.226). Next, the mitoxantrone accumulation was determined in P-gp Functional Activity. Mitoxantrone can also be ALL blasts. In the B-lineage ALL samples, the median mitox- transported by P-gp and MRP. To get an impression of the antrone accumulation was 148 (range, 77–555 MFI). The me- impact of ABCG2 on mitoxantrone transport, P-gp and MRP dian mitoxantrone accumulation in T-lineage ALL was 108 activity, and the effect of their blockers on mitoxantrone accu- (range, 65–281 MFI; Table 2). mulation were studied as well. In B-lineage ALL, there was no To investigate whether the efflux of mitoxantrone was median effect of PSC 833 on Rh123 accumulation (range, ABCG2-mediated, the ABCG2 inhibitor FTC was added. In the 0–197%; Table 2), whereas PSC 833 gave a median increase of MCF-7 cell line, FTC caused an increase in mitoxantrone ac- 14% in T-lineage ALL (range, 0–362%; P ϭ 0.001; Table 2). In cumulation of 18% (range, 15–38%). In the ABCG2 overex- brief, P-gp activity was observed only in T-lineage ALL. pressing cell line MCF-7 MR an increase of 168% was observed To study the specific transport of mitoxantrone by P-gp, the (range, 141–227%). FTC did not affect mitoxantrone accumu- effect of PSC 833 on mitoxantrone was studied. Surprisingly, lation in P-gp- and MRP-overexpressing cell lines (data not PSC 833 increased the mitoxantrone accumulation in all of the shown). B-lineage and most T-lineage ALL patients (Table 2). The addition of the ABCG2 inhibitor FTC caused a higher MRP Functional Activity. The functional activity of median increase in mitoxantrone accumulation in B-lineage MRP was determined by assessing the effect of MK-571 on CF ALL, namely 21% (range, 0–140%), than in T-lineage ALL accumulation. In B-lineage ALL, MK-571 increased the CF (5%; range, 0–256%; P ϭ 0.013). From the ALL samples, 27 accumulation with 192% (range, 23–639%) and in T-lineage (59%) patient samples showed a smaller effect of FTC on ALL with 239% (range, 68–500%), indicating that MRP is mitoxantrone accumulation than the MCF-7 cell line. Eighteen functionally active in both B- and T-lineage ALL (Table 2). To (39%) patient samples showed a FTC effect between the MCF-7 investigate the role of MRP in mitoxantrone transport in ALL, and MCF-7 MR cell lines, and 1 (2%) patient sample showed a mitoxantrone accumulation was studied in the presence of MK- larger increase in mitoxantrone accumulation by FTC than the 571. MK-571 increased the mitoxantrone accumulation in 20 of MCF-7 MR cell line. These results indicate that ABCG2 is 21 B-lineage samples and 18 of 22 T-lineage samples (Table 2). functionally active on a low but significant level in B-lineage Mutation Analysis of the ABCG2 Gene and Substrate ALL. Spectrum. The ability of ABCG2 to transport Rh123 was In normal B- and T-lymphocytes from healthy donors, the studied by measuring the effect of FTC on Rh123 accumulation. effect of FTC on mitoxantrone accumulation was also studied. In three of the 46 samples (1 B-lineage and 2 T-lineage ALL) No effect of FTC could be observed in both B and T lympho- the addition of FTC showed an increase in Rh123 accumulation. cytes. These three samples, as well as 5 control samples, were se-

Table 3 Effect of the combination of inhibitors on mitoxantrone Table 4 Wilcoxon rank test calculating the effect of the addition of accumulation in ALL patients one inhibitor to a combination of inhibitors on mitoxantrone Shifts calculated as percentage of mitoxantrone accumulation with- accumulation in ALL patients out inhibitors. The results represent median values with range. Shifts calculated as percentage of mitoxantrone accumulation without inhibitors. The results represent median values. B-lineage ALL T-lineage ALL Mito accumulation (MFI) 148 (77–555) 108 (65–281) Combination of inhibitors Shift P FTC shift 21 (0–140) 5 (0–256) B-lineage ALL PSC 833 shift 22 (6–70) 24 (0–81) FTCϩPSC833 vs. PSC833 27 vs. 22 0.002 MK-571 shift 16 (0–173) 18 (0–141) FTCϩMK571 vs. MK571 39 vs. 16 0.002 FTCϩPSC 833 shift 27 (0–156) 43 (0–308) FTCϩPSC833ϩMK571 vs. PSC833ϩMK571 44 vs. 32 0.351 FTCϩMK-571 shift 39 (0–226) 22 (0–281) T-lineage ALL PSC 833ϩMK-571 shift 32 (8–166) 50 (0–225) FTCϩPSC833 vs. PSC833 43 vs. 24 0.016 FTCϩPSC 833ϩMK-571 shift 44 (6–217) 46 (0–340) FTCϩMK571 vs. MK571 22 vs. 18 0.170 FTCϩPSC833ϩMK571 vs. PSC833ϩMK571 46 vs. 50 0.408

quenced for mutation at amino acid 482 in the ABCG2 gene. All 8 of the samples showed the wild-type variant. In summary, a number of correlations were found between Combination of Inhibitors on Mitoxantrone Accumula- more mature markers in B-lineage ALL and ABCG2. No rela- tion. Finally, the inhibitors were combined to analyze whether tion was observed between the protein expression and functional their effects are the result of the activity of one transporter or a activity of ABCG2, and an immature phenotype in our patient combined action of transporters. First, the addition of FTC to samples. PSC 833 on mitoxantrone accumulation was studied, to explain the effect of PSC 833 in samples without P-gp activity, as Discussion assessed by the Rh123 efflux assay. Our results demonstrated an This is the first study, which addresses the ABCG2 expres- increased effect of the combination of FTC and PSC 833 on sion and its functional activity in ALL in children and adults. mitoxantrone accumulation in comparison with just PSC 833 or ABCG2 expression and functional activity was low but higher FTC, in both B-lineage and T-lineage ALL samples (Tables 3 in B-lineage ALL than in T-lineage ALL. The functional and 4). ABCG2 activity correlated well with the ABCG2 expression. By adding FTC to MK-571, the role of ABCG2 in mitox- Sauerbrey et al. (30) have measured ABCG2 expression in ALL antrone transport in patients with MRP activity was studied. The only at the mRNA level, and also observed higher mRNA addition of FTC to MK-571 caused an increase in mitoxantrone expression in B-lineage ALL than in T-lineage ALL. In contrast, accumulation in only B-lineage ALL (Tables 3 and 4). the present study showed that in normal B- and T-lymphocytes The addition of FTC to the combination of PSC 833 and no ABCG2 functional activity could be determined, and conse- MK-571 gives a reflection of the ABCG2 activity not biased by quently no difference between B- and T-lymphocytes was P-gp and MRP functional activity. In the B- and T-lineage ALL observed. samples, this did not cause an overall increase in mitoxantrone In normal hematopoietic cells, it was shown that ABCG2 is accumulation (B-lineage P ϭ 0.351 respectively. T-lineage expressed in pluripotent stem cells and sharply down-regulated ALL P ϭ 0.408; Tables 3 and 4). However, in B-lineage ALL at the point of commitment to lineage-specific development (15, samples in which the addition of FTC caused an increase in 23, 24). Also in AML, an immature phenotype, as determined mitoxantrone accumulation, this increase correlated with the by the expression of CD34, correlated with ABCG2 expression ABCG2 protein expression (r ϭ 0.57; P ϭ 0.009; n ϭ 20). (21). In ALL, we did not observe an association between the Immunophenotype. ABCG2 expression has been asso- immature markers CD117 or CD34 and ABCG2. The expres- ciated with an immature phenotype in normal hematopoietic sion of other transporters, such as P-gp and MRP, is also cells (15, 23, 24) and AML blasts (21). Therefore, the immu- associated with an immature phenotype in normal hematopoietic nophenotype of the patient blasts was determined and correlated cells (31, 32) and AML blasts (33–36) but not in ALL blasts (6, with the maturation stage. 8, 37). Apparently, the connection between transporters and In B-lineage ALL, CD10 (“common” ALL antigen) ex- maturation stage is differently regulated in ALL than normal pression correlated with ABCG2 protein expression and hematopoietic cells and AML blasts. ABCG2 functional activity (r ϭ 0.51; P ϭ 0.021; n ϭ 20 resp. A mutation in the ABCG2 gene can alter substrate speci- r ϭ 0.52; P ϭ 0.018; n ϭ 20). The CD19 expression was also ficity in drug-resistant cell lines (19). Expression of the mutated associated with ABCG2 activity (r ϭ 0.48; P ϭ 0.039; n ϭ 19). form, with threonine or glycine, can confer resistance to a No correlation was observed between the markers CD20 and broader spectrum of drugs, including mitoxantrone, Rh123, and CD33, the immature markers CD34 or CD117 (c-kit), and the doxorubicin. The three patient samples, in which ABCG2 was ABCG2 protein expression and ABCG2 functional activity (i.e., suggested to transport Rh123, did not show the mutation. This the effect of FTC on mitoxantrone accumulation). In T-lineage finding underscores the limited role of this mutation in the ALL, the expression of the CD markers CD34, CD117, CD3, ABCG2 gene in clinical samples. CD7, and CD33 did not correlate with ABCG2 expression and Although no gold standard exists for the detection of functional activity. Of all of the investigated markers, CD33 and ABCG2 functional activity, Minderman et al. (38) showed that CD117 were scarcely expressed on ALL blasts. flow cytometry using mitoxantrone in combination with

ABCG2 inhibitor FTC can be used to detect functional activity. overexpressing only one of the transporters (12, 21). In cell lines Evidently, other inhibitors could have been used, such as the overexpressing P-gp or MRP, we and others showed that FTC FTC-analogue Ko143, but we have no reason to expect them to did not reverse resistance (12, 21). These observations in cell be more suitable than FTC. Moreover, FTC is used most often lines imply that FTC acts as a specific ABCG2 inhibitor. The in functional studies, and the application of a standardized specificity of PSC833 was described in the same study (21) in detection method is recommended (39). which PSC833 showed little effect on the ABCG2-overexpress- Mitoxantrone is a substrate for several other transporters, ing cell line. However, in contrast to cell lines overexpressing including P-gp and MRP. Therefore, the functional activity of specific transporters, clinical samples could express other (un- these transporters was investigated as well as their role in known) transporters, which can also be inhibited by FTC or mitoxantrone transport. The contribution to mitoxantrone trans- PSC833. The discrepancies in our data could be explained by port of the individual transporters is reflected in Table 3. the presence of these transporters, which are capable of efflux- ABCG2-mediated mitoxantrone transport in B-lineage ALL is ing mitoxantrone as well. comparable with that of P-gp and MRP. In T-lineage ALL, the In summary, this study shows that ABCG2 is expressed mitoxantrone transport by ABCG2 is negligible compared with and functional in B-lineage ALL and at a lower level in T- P-gp- and MRP-mediated mitoxantrone transport. lineage ALL. The next step will be to determine the clinical Remarkable was the effect of PSC 833 on mitoxantrone impact of this transporter in a larger prospective trial. Because accumulation in patients who did not show P-gp activity, as ABCG2 is mainly functionally active in B-lineage ALL, the trial assessed with the Rh123 efflux assay. This could possibly be will only be useful in this ALL subtype. Hopefully, it will explained by PSC 833 interfering with another transporter than elucidate whether these patients with ALL could benefit from P-gp, which is also capable of transporting mitoxantrone. In treatment adaptations based on this knowledge. general, the lack of specificity of P-gp and ABCG2 inhibitors could explain the inconclusive data, because new transporters are still being identified that may be inhibited by P-gp and/or References ABCG2 inhibitors. Also the occurrence of polymorphisms in the 1. Chessells, J. M., Hall, E., Prentice, H. G., Durrant, J., Bailey, C. C., and ABCG2 or P-gp encoding genes, altering the substrate specific- Richards, S. M. The impact of age on outcome in lymphoblastic leukaemia; ity, may be of influence. MRC UKALL X and XA compared: a report from the MRC Paediatric and In patients with functionally active P-gp, FTC showed an Adult Working Parties. Leukemia (Baltimore), 12: 463–473, 1998. increase in mitoxantrone accumulation when P-gp was inhibited. 2. Mauer, A. M. Adult and childhood acute lymphocytic leukemia: are This suggests that even in patients in which mitoxantrone is trans- they different diseases? Am. J. Hematol., 42: 127–131, 1993. ported by P-gp, ABCG2 contributes to mitoxantrone resistance. 3. den Boer, M. L., Pieters, R., Kazemier, K. M., Rottier, M. M., Zwaan, C. M., Kaspers, G. J., Janka-Schaub, G., Henze, G., Creutzig, The ability of MRP to transport mitoxantrone is not very U., Scheper, R. J., and Veerman, A. J. Relationship between major vault clear. We observed an effect of MK-571 in mitoxantrone trans- protein/lung resistance protein, multidrug resistance-associated protein, port in most of the ALL cases. The MRP-overexpressing cell P-glycoprotein expression, and drug resistance in childhood leukemia. line GLC4/ADR also showed an increase in mitoxantrone ac- Blood, 91: 2092–2098, 1998. cumulation when MK-571 was added. However, there are stud- 4. Dhooge, C., De Moerloose, B., Laureys, G., Kint, J., Ferster, A., De ies suggesting that mitoxantrone is not a substrate of MRP (18, Bacquer, D., Philippe, J., and Benoit, Y. P-glycoprotein is an independ- 40). The conflicting reports could also be the result of a func- ent prognostic factor predicting relapse in childhood acute lymphoblastic leukaemia: results of a 6-year prospective study. Br. J. Haematol., tional polymorphism that has not been discovered yet. 105: 676–683, 1999. When P-gp and MRP were inhibited, FTC did not cause an 5. Goasguen, J. E., Dossot, J. M., Fardel, O., Le Mee, F., Le Gall, E., overall increase in mitoxantrone accumulation in B- and T-lineage Leblay, R., LePrise, P. Y., Chaperon, J., and Fauchet, R. Expression of ALL. In B-lineage ALL, an effect of FTC could be observed in half the multidrug resistance-associated P-glycoprotein (P- 170) in 59 cases of the patients, and the increase in mitoxantrone accumulation of de novo acute lymphoblastic leukemia: prognostic implications. correlated with the ABCG2 protein expression. In patient samples Blood, 81: 2394–2398, 1993. in which PSC 833 and MK-571 showed a significant effect on 6. Ivy, S. P., Olshefski, R. S., Taylor, B. J., Patel, K. M., and Reaman, G. H. Correlation of P-glycoprotein expression and function in child- mitoxantrone accumulation, the addition of FTC did not cause an hood acute leukemia: a children’s cancer group study. Blood, 88: extra increase. Consequently, in these samples the effect of FTC is 309–318, 1996. most likely masked by the effect of PSC 833 and MK-571. The 7. Wattel, E., Lepelley, P., Merlat, A., Sartiaux, C., Bauters, F., Jouet, effect of inhibitors was expressed as a percentage, and conse- J. P., and Fenaux, P. Expression of the multidrug resistance P glyco- quently, if two inhibitors have a large effect, the effect of the protein in newly diagnosed adult acute lymphoblastic leukemia: absence addition of a third inhibitor is likely to be underestimated. of correlation with response to treatment. Leukemia (Baltimore), 9: 1870–1874, 1995. The distribution of the functional activity of transporters 8. Wuchter, C., Leonid, K., Ruppert, V., Schrappe, M., Buchner, T., within patients was determined by calculating the correlation Schoch, C., Haferlach, T., Harbott, J., Ratei, R., Dorken, B., and between the effect of FTC and P-gp and/or MRP functional Ludwig, W. D. Clinical significance of P-glycoprotein expression and activity. No correlation was found between the three transport- function for response to induction chemotherapy, relapse rate and over- ers in B- and T-lineage ALL. Surprisingly, the effect of FTC on all survival in acute leukemia. Haematologica, 85: 711–721, 2000. mitoxantrone accumulation correlated with the effect of only 9. Plasschaert, S. L. A., Vellenga, E., de Bont, E. S. J. M., van der Kolk, D. M., Veerman, A. J., Sluiter, W. J., Daenen, S. M. G., de Vries, PSC 833 or MK-571 on mitoxantrone accumulation. Cross- E. G. E., and Kamps, W. A. High functional P-glycoprotein activity is resistance of inhibitors could account for this phenomenon, more often present in T-cell acute lymphoblastic leukemic cells in adults although this has not been described in studies using cell lines than in children. Leuk. Lymphoma, 44: 85–95, 2002.

10. Allikmets, R., Schriml, L. M., Hutchinson, A., Romano-Spica, V., Dutch Childhood Leukemia Study Group (DCLSG) protocols for chil- and Dean, M. A human placenta-specific ATP-binding cassette gene dren with acute lymphoblastic leukemia, 1984–1991. Leukemia (Balti- (ABCP) on chromosome 4q22 that is involved in multidrug resistance. more), 14: 2240–2246, 2000. Cancer Res., 58: 5337–5339, 1998. 26. Daenen, S., van Imhoff, G. W., van den Berg, E., de Kam, P. J., 11. Doyle, L. A., Yang, W., Abruzzo, L. V., Krogmann, T., Gao, Y., Haaxma-Reiche, H., Vellenga, E., Smit, J. W., and Halie, R. M. Im- Rishi, A. K., and Ross, D. D. A multidrug resistance transporter from proved outcome of adult acute lymphoblastic leukaemia by moderately human MCF-7 breast cancer cells. Proc. Natl. Acad. Sci. USA, 95: intensified chemotherapy which includes a ’preinduction’ course for 15665–15670, 1998. rapid tumour reduction: preliminary results on 66 patients. Br. J. 12. Rabindran, S. K., He, H., Singh, M., Brown, E., Collins, K. I., Haematol., 100: 273–282, 1998. Annable, T., and Greenberger, L. M. Reversal of a novel multidrug 27. Scheffer, G. L., Maliepaard, M., Pijnenborg, A. C., van Gastelen, resistance mechanism in human colon carcinoma cells by fumitremorgin M. A., de Jong, M. C., Schroeijers, A. B., van der Kolk, D. M., Allen, C. Cancer Res., 58: 5850–5858, 1998. J. D., Ross, D. D., van der Valk, P., Dalton, W. S., Schellens, J. H., and 13. Maliepaard, M., van Gastelen, M. A., Tohgo, A., Hausheer, F. H., van Scheper, R. J. Breast cancer resistance protein is localized at the plasma Waardenburg, R. C., de Jong, L. A., Pluim, D., Beijnen, J. H., and Schel- membrane in mitoxantrone- and topotecan-resistant cell lines. Cancer lens, J. H. Circumvention of breast cancer resistance protein (BCRP)- Res., 60: 2589–2593, 2000. mediated resistance to camptothecins in vitro using non-substrate drugs or 28. Jones, T. R., Zamboni, R., Belley, M., Champion, E., Charette, L., the BCRP inhibitor GF120918. Clin. Cancer Res., 7: 935–941, 2001. Ford-Hutchinson, A. W., Frenette, R., Gauthier, J. Y., Leger, S., Mas- 14. Litman, T., Brangi, M., Hudson, E., Fetsch, P., Abati, A., Ross, son, P., et al. Pharmacology of L-660, 711 (MK-571): a novel potent D. D., Miyake, K., Resau, J. H., and Bates, S. E. The multidrug-resistant and selective leukotriene D4 receptor antagonist. Cancer J. Physiol. phenotype associated with overexpression of the new ABC half-trans- Pharmacol., 67: 17–28, 1989. porter, MXR (ABCG2). J. Cell Sci., 113: 2011–2021, 2000. 29. van der Kolk, D. M., de Vries, E. G., Koning, J. A., van den Berg, 15. Maliepaard, M., Scheffer, G. L., Faneyte, I. F., van Gastelen, M. A., E., Mu¨ller, M., and Vellenga, E. Activity and expression of the multi- Pijnenborg, A. C., Schinkel, A. H., van De Vijver, M. J., Scheper, R. J., drug resistance proteins MRP1 and MRP2 in acute myeloid leukemia and Schellens, J. H. Subcellular localization and distribution of the cells, tumor cell lines, and normal hematopoietic CD34ϩ peripheral breast cancer resistance protein transporter in normal human tissues. blood cells. Clin. Cancer Res., 4: 1727–1736, 1998. Cancer Res., 61: 3458–3464, 2001. 30. Sauerbrey, A., Sell, W., Steinbach, D., Voigt, A., and Zintl, F. 16. Consoli, U., Van, N. T., Neamati, N., Mahadevia, R., Beran, M., Expression of the BCRP gene (ABCG2/MXR/ABCP) in childhood Zhao, S., and Andreeff, M. Cellular pharmacology of mitoxantrone in acute lymphoblastic leukaemia. Br. J. Haematol., 118: 147–150, 2002. p-glycoprotein-positive and - negative human myeloid leukemic cell 31. Chaudhary, P. M., and Roninson, I. B. Expression and activity of lines. Leukemia (Baltimore), 11: 2066–2074, 1997. P-glycoprotein, a multidrug efflux pump, in human hematopoietic stem 17. Schneider, E., Horton, J. K., Yang, C. H., Nakagawa, M., and cells. Cell, 66: 85–94, 1991. Cowan, K. H. Multidrug resistance-associated protein gene overexpres- 32. Bunting, K. D. ABC transporters as phenotypic markers and func- sion and reduced drug sensitivity of topoisomerase II in a human breast tional regulators of stem cells. Stem Cells, 20: 11–20, 2002. carcinoma MCF7 cell line selected for etoposide resistance. Cancer 33. Laupeze, B., Amiot, L., Drenou, B., Bernard, M., Branger, B., Res., 54: 152–158, 1994. Grosset, J. M., Lamy, T., Fauchet, R., and Fardel, O. High multidrug 18. Diah, S. K., Smitherman, P. K., Aldridge, J., Volk, E. L., Schneider, resistance protein activity in acute myeloid leukaemias is associated E., Townsend, A. J., and Morrow, C. S. Resistance to mitoxantrone in with poor response to chemotherapy and reduced patient survival. Br. J. multidrug-resistant MCF7 breast cancer cells: evaluation of mitox- Haematol., 116: 834–838, 2002. antrone transport and the role of multidrug resistance protein family 34. Legrand, O., Simonin, G., Perrot, J. Y., Zittoun, R., and Marie, J. P. proteins. Cancer Res., 61: 5461–5467, 2001. Pgp and MRP activities using calcein-AM are prognostic factors in adult 19. Honjo, Y., Hrycyna, C. A., Yan, Q. W., Medina-Perez, W. Y., acute myeloid leukemia patients. Blood, 91: 4480–4488, 1998. Robey, R. W., van de, L. A., Litman, T., Dean, M., and Bates, S. E. 35. te Boekhorst, P. A., de Leeuw, K., Schoester, M., Wittebol, S., Acquired mutations in the MXR/BCRP/ABCP gene alter substrate spec- Nooter, K., Hagemeijer, A., Lo¨wenberg, B., and Sonneveld, P. Predom- ificity in MXR/BCRP/ABCP-overexpressing cells. Cancer Res., 61: inance of functional multidrug resistance (MDR-1) phenotype in 6635–6639, 2001. CD34ϩ acute myeloid leukemia cells. Blood, 82: 3157–3162, 1993. 20. Volk, E. L., Farley, K. M., Wu, Y., Li, F., Robey, R. W., and 36. van der Kolk, D. M., de Vries, E. G., Noordhoek, L., van den Berg, Schneider, E. Overexpression of wild-type breast cancer resistance protein E., van der Pol, M. A., Mu¨ller, M., and Vellenga, E. Activity and mediates methotrexate resistance. Cancer Res., 62: 5035–5040, 2002. expression of the multidrug resistance proteins P- glycoprotein, MRP1, 21. van der Kolk, D. M., Vellenga, E., Scheffer, G. L., Mu¨ller, M., MRP2, MRP3 and MRP5 in de novo and relapsed acute myeloid Bates, S. E., Scheper, R. J., and de Vries, E. G. Expression and activity leukemia. Leukemia (Baltimore), 15: 1544–1553, 2001. of breast cancer resistance protein (BCRP) in de novo and relapsed acute 37. Cascavilla, N., Musto, P., D’Arena, G., Ladogana, S., Matera, R., myeloid leukemia. Blood, 99: 3763–3770, 2002. and Carotenuto, M. Adult and childhood acute lymphoblastic leukemia: 22. Robey, R. W., Honjo, Y., van de Laar, A., Miyake, K., Regis, J. T., clinico-biological differences based on CD34 antigen expression. Litman, T., and Bates, S. E. A functional assay for detection of the Haematologica, 82: 31–37, 1997. mitoxantrone resistance protein, MXR (ABCG2). Biochim. Biophys. 38. Minderman, H., Suvannasankha, A., O’Loughlin, K. L., Scheffer, Acta, 1512: 171–182, 2001. G. L., Scheper, R. J., Robey, R. W., and Baer, M. R. Flow cytometric 23. Scharenberg, C. W., Harkey, M. A., and Torok-Storb, B. The analysis of breast cancer resistance protein expression and function. ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is Cytometry, 48: 59–65, 2002. preferentially expressed by immature human hematopoietic progenitors. 39. Broxterman, H. J., Sonneveld, P., Feller, N., Ossenkoppele, G. J., Blood, 99: 507–512, 2002. Wahrer, D. C., Eekman, C. A., Schoester, M., Lankelma, J., Pinedo, H. M., 24. Zhou, S., Schuetz, J. D., Bunting, K. D., Colapietro, A. M., Sam- Lowenberg, B., and Schuurhuis, G. J. Quality control of multidrug resist- path, J., Morris, J. J., Lagutina, I., Grosveld, G. C., Osawa, M., Nakau- ance assays in adult acute leukemia: correlation between assays for P- chi, H., and Sorrentino, B. P. The ABC transporter Bcrp1/ABCG2 is glycoprotein expression and activity. Blood, 87: 4809–4816, 1996. expressed in a wide variety of stem cells and is a molecular determinant 40. Cole, S. P., Sparks, K. E., Fraser, K., Loe, D. W., Grant, C. E., of the side-population phenotype. Nat. Med., 7: 1028–1034, 2001. Wilson, G. M., and Deeley, R. G. Pharmacological characterization of 25. Kamps, W. A., Veerman, A. J., van Wering, E. R., van Weerden, multidrug resistant MRP-transfected human tumor cells. Cancer Res., J. F., Slater, R., and Does-van den Berg, A. Long-term follow-up of 54: 5902–5910, 1994.

Sabine L. A. Plasschaert, Dorina M. van der Kolk, Eveline S. J. M. de Bont, et al.

Clin Cancer Res 2003;9:5171-5177.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/9/14/5171

Cited articles This article cites 40 articles, 23 of which you can access for free at: http://clincancerres.aacrjournals.org/content/9/14/5171.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/9/14/5171.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/9/14/5171. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.