Cancer Chemotherapy and Pharmacology

, Volume 53, Issue 5, pp 363–369

Broad-spectrum modulation of ATP-binding cassette transport proteins by the taxane derivatives ortataxel (IDN-5109, BAY 59-8862) and tRA96023

Authors

  • Hans Minderman
    • Leukemia Section, Department of MedicineRoswell Park Cancer Institute
  • Tracy A. Brooks
    • Department of Pharmacology and TherapeuticsRoswell Park Cancer Institute
  • Kieran L. O’Loughlin
    • Leukemia Section, Department of MedicineRoswell Park Cancer Institute
  • Iwao Ojima
    • Department of ChemistryState University of New York at Stony Brook
  • Ralph J. Bernacki
    • Department of Pharmacology and TherapeuticsRoswell Park Cancer Institute
    • Leukemia Section, Department of MedicineRoswell Park Cancer Institute
    • Department of Pharmacology and TherapeuticsRoswell Park Cancer Institute
Original Article

DOI: 10.1007/s00280-003-0745-2

Cite this article as:
Minderman, H., Brooks, T.A., O’Loughlin, K.L. et al. Cancer Chemother Pharmacol (2004) 53: 363. doi:10.1007/s00280-003-0745-2
  • 88 Views

Abstract

Purpose

The taxanes paclitaxel and docetaxel are substrates for P-glycoprotein (Pgp), an ATP-binding cassette (ABC) transport protein associated with multidrug resistance (MDR). In contrast, the synthetic taxane ortataxel (BAY 59-8862, IDN-5109) is effective against Pgp-expressing cells by virtue of modulation of Pgp-mediated transport. The synthetic taxane tRA96023 also modulates Pgp and is noncytotoxic due to removal of the tubulin-binding side chain at the C-13 position of the taxane backbone. We studied the effects of ortataxel and tRA96023 on the other MDR-associated ABC transport proteins, multidrug resistance protein (MRP-1) and breast cancer resistance protein (BCRP, MXR, ABCG2).

Methods

Modulation of mitoxantrone, daunorubicin and doxorubicin retention and cytotoxicity by ortataxel and tRA96023 was studied in established cell lines overexpressing Pgp, MRP-1 and wild type (BCRPR482) and mutant (BCRPR482T) BCRP, and was compared with modulation by the established Pgp-, MRP-1- and BCRP-specific modulators PSC-833, probenecid and fumitremorgin C, respectively.

Results

Ortataxel effectively modulated drug retention and cytotoxicity in cell lines overexpressing MRP-1 and BCRPR482, in addition to Pgp. tRA96023 modulated drug retention and cytotoxicity in cell lines overexpressing BCRPR482 and Pgp, but not those overexpressing MRP-1. Neither ortataxel nor tRA96023 modulated BCRPR482T.

Conclusions

The synthetic taxane derivatives ortataxel and tRA96023 are broad-spectrum ABC protein modulators. Further studies will seek to identify a noncytotoxic synthetic taxane that modulates Pgp, MRP-1 and BCRP.

Keywords

TaxanesP-glycoproteinMultidrug resistance proteinBreast cancer resistance proteinModulation

Introduction

Tumor cells frequently exhibit multidrug resistance (MDR) mediated by the ATP-binding cassette (ABC) membrane proteins P-glycoprotein (Pgp), multidrug resistance protein (MRP-1) and breast cancer resistance protein (BCRP) [10]. Noncytotoxic molecules, termed MDR modulators, inhibit drug efflux mediated by these proteins, restoring cellular drug sensitivity. PSC-833 [2], probenecid [9] and fumitremorgin C [33] modulate Pgp, MRP-1 and BCRP, respectively; cyclosporin A (CsA) has activity against MRP-1 and BCRP, as well as Pgp [26]. Structural properties responsible for broad-spectrum, versus ABC protein-specific, modulation are unknown.

The taxanes paclitaxel and docetaxel, semisynthetic chemotherapy drugs that inhibit microtubule depolymerization [34, 38], are substrates for Pgp [3, 16, 39]. Limited availability of natural resources led to the development of synthetic taxanes [5], and active efflux of natural product taxanes by Pgp prompted attempts to synthesize analogs that are not susceptible to Pgp-mediated drug resistance [31]. Ortataxel (formerly BAY 59-8862 and IDN 5109) [31] is a cytotoxic taxane [4] that modulates Pgp, blocking its own efflux from Pgp-overexpressing cells [17, 28, 29, 30, 31, 32, 45]. tRA96023 also modulates Pgp [12], but is noncytotoxic due to removal of the side chain at the C-13 position of the taxane backbone, which is required for tubulin binding [32]. Ortataxel is in clinical trials [6], and tRA96023 is in preclinical development.

To determine whether ortataxel and tRA96023 might be broad-spectrum MDR modulators, we studied their effects on drug retention and cytotoxicity in resistant cell lines overexpressing MRP-1 and BCRP, as well as Pgp. The drug retention assay measures short-term effects on drug transport, while the cytotoxicity assay measures long-term effects on cell survival.

Materials and methods

Cell lines

Established resistant cell lines overexpressing Pgp (8226/Dox6, MCF7/R), MRP-1 (HL60/Adr) or wild type (8226/MR20) or mutant (MCF7/AdVp3000) BCRP were studied [8, 14, 24, 25]. Wild type (BCRPR482) and mutant (BCRPR482T) BCRP, with arginine and threonine in the amino acid 482 position, respectively, both efflux mitoxantrone, but only BCRPR482T effluxes anthracyclines [15]. Wild type HL60 cells served as negative controls for all three proteins [25] since the parental 8226 and MCF7 cell lines express low levels of BCRPR482 [25, 27].

ABC protein expression

Expression of Pgp, MRP-1 and BCRP was confirmed by staining with the MRK-16 (Kamiya Biomedical Company, Tukwila, Wash.), MRPm6 (Kamiya), BXP-21 and BXP-34 (from Dr. R.J. Scheper, Free University, The Netherlands, as previously described [25]) antibodies, respectively, followed by flow cytometric detection (Fig. 1). The demonstrated expression of the MDR-associated ABC proteins in Fig. 1 is consistent with previous reports in the literature [1, 8, 14, 25].
Fig. 1

ABC protein expression in cell lines, measured by flow cytometry. Cells were stained with the MRK-16, MRPm6 and BXP-21 antibodies (solid lines), which are specific for Pgp, MRP-1 and BCRP, respectively, and with isotype controls (broken lines). Staining with antibody was compared with staining with isotype control by the Kolmogorov-Smirnov statistic, expressed as a D-value ranging from 0 to 1. D-values indicating positive staining are in bold italics. Representative comparisons are shown from triplicate experiments. Staining with BXP-34, a second BCRP-specific antibody, was similar to staining with BXP-21 (data not shown)

Drugs

Drugs used in retention and cytotoxicity assays included mitoxantrone, daunorubicin and doxorubicin (Sigma, St. Louis, Mo.). All are substrates for Pgp [22, 25], MRP-1 [22, 25] and BCRPR482T [15, 22, 25], but only mitoxantrone is also a substrate for BCRPR482 [15].

Modulators

Ortataxel (Indena, Milan, Italy) and tRA96023 (SB-RA-310124, synthesized as described previously [30]), were compared with PSC-833 (Novartis, East Hanover, N.J.), p-(dipropylsulfamoyl)benzoic acid (Probenecid, Sigma) and fumitremorgin C (FTC, from Dr. Susan Bates, National Cancer Institute, Bethesda, Md.) at concentrations of 2.5 μM, 1 mM and 10 μM, respectively, which were previously described to be most effective for modulation without cytotoxicity [1, 9, 35]. Ortataxel was minimally fluorescent at high concentrations; tRA96023 was not fluorescent (Fig. 2).
Fig. 2

Measurement of intrinsic fluorescence of ortataxel and tRA96023. HL60 cells were incubated for 30 min with ortataxel (left) and with tRA96023 (right) at the concentrations shown. Fluorescence of cells incubated with ortataxel or tRA96023 was compared with that of cells incubated without modulator by the Kolmogorov-Smirnov statistic, expressed as D-values ranging from 0 to 1. Experiments were performed in triplicate; a representative experiment is shown. Ortataxel was minimally fluorescent at high concentrations; tRA96023 was not fluorescent

Drug retention studies

Drug retention was studied as previously described [25]. Briefly, drug uptake was achieved by incubating cells with drug (3 μM) for 30 min. Drug efflux was then studied by comparing drug retention following a 90-min incubation in drug-free medium in the presence and absence of modulator. All experiments were performed in triplicate. Cellular drug content was measured by flow cytometry and analyzed using the WinList program (Verity Software House, Topsham, Me.). In experiments with ortataxel, cellular fluorescence was compensated to account for that associated with the modulator.

Flow cytometry data analysis

Labeling with specific antibody and with isotype control was compared by the Kolmogorov-Smirnov (KS) statistic, expressed as a D-value ranging from 0 to 1 [46]. Antibody positivity was defined by D-values ≥0.1 and ≥0.2 for staining of unfixed (MRK-16 [1]) and fixed (other antibodies [25]) cells, respectively. Drug retention in the presence and absence of modulator was also compared by the KS statistic; D-values ≥0.2 were considered to be indicative of modulation [25].

Cytotoxicity

Cells were cultured for 96 h in 96-well plates at densities of 10,000 (suspension) or 1,500 (adherent) cells per well in RPMI 1640 medium with 10% FBS, 2 mM l-glutamine, 20 U/ml penicillin, 20 μg/ml streptomycin and drugs at a range of doses in the absence and presence of modulators, in triplicate. Cell growth was assessed by the WST-1 assay (Roche Diagnostics, Mannheim, Germany) for suspension cells [36, 42] and the sulforhodamine-B (SRB) assay for adherent cells [40]. IC50 values, or drug concentrations inhibiting cell growth by 50% compared to untreated cells, were determined using curve-fitting software, as previously described [11]. Relative resistance (RR) was calculated as the ratio (IC50 resistant cells/IC50 wild type cells). The resistance modifying factor (RMF) was calculated as the ratio (IC50 drug/IC50 drug+modulator).

Results

Effects of ortataxel and tRA96023 on drug efflux mediated by Pgp, MRP-1 and BCRP

Mitoxantrone retention was studied in the absence and presence of increasing concentrations of ortataxel and tRA96023 in cells overexpressing Pgp (8226/Dox6), MRP-1 (HL60/Adr) and BCRP (8226/MR20) (Fig. 3). Ortataxel modulated retention in all three cell lines, but tRA96023 had no effect in HL60/Adr cells. Both modulators increased mitoxantrone retention in a concentration-dependent manner, effectively blocking efflux at 10 μM. The effects of ortataxel and tRA96023 on mitoxantrone retention were compared with those of specific modulators of Pgp, MRP-1 and BCRP (Fig. 4). The taxane derivatives had a broader spectrum of activity but were slightly less potent MRP-1 and BCRP modulators than probenecid and FTC, as evidenced by lower D-values.
Fig. 3

Concentration-dependent effects of ortataxel (left) and tRA96023 (right) on efflux of mitoxantrone in the 8226/Dox6, HL60/Adr and 8226/MR20 cell lines, overexpressing Pgp, MRP-1 and wildtype BCRP (BCRPR482), respectively. Cells were incubated with 3 μM mitoxantrone for 30 min (uptake), washed and resuspended in drug-free medium, then incubated for 90 min without modulator (efflux) and with ortataxel or tRA96023 at the concentrations shown. Efflux in the presence of ortataxel or tRA96023 at each concentration was compared with efflux in the absence of modulator by the Kolmogorov-Smirnov statistic, expressed as D-values ranging from 0 to 1. A representative example from triplicate experiments is shown. Ortataxel modulated all cell lines, while tRA96023 had no effect in HL60/Adr cells. Both modulators increased mitoxantrone retention in a concentration-dependent manner, effectively blocking efflux at 10 μM

Fig. 4

Modulation of mitoxantrone efflux by ortataxel (10 μM), tRA96023 (10 μM), PSC-833 (2.5 μM), probenecid (1 mM) and fumitremorgin C (10 μM) in cell lines expressing different MDR-associated transport proteins: HL60 (none), 8226/Dox6 and MCF7/R (Pgp), HL60/Adr (MRP-1), 8226/MR20 (BCRPR482) and MCF7/AdVp3000 (BCRPR482T). Cells were incubated with 3 μM mitoxantrone for 30 min (uptake), washed and resuspended in drug-free medium, then incubated for 90 min with and without modulator (efflux). Retention in the presence of each modulator was compared with retention in the absence of modulator by the Kolmogorov-Smirnov statistic, expressed as a D-value ranging from 0 to 1. D-values ≥0.20 (dashed lines) were considered to indicate modulation. The graphs show the mean values from triplicate experiments, with standard errors. Ortataxel and tRA96023 had a broader spectrum of activity than PSC-833, probenecid and fumitremorgin C, but were slightly less potent, as evidenced by lower D-values

Modulation of efflux of daunorubicin and doxorubicin, which are additional established substrates of Pgp, MRP-1 and BCRPR482T, was also studied (Fig. 5). Ortataxel (Fig. 5a) effectively increased retention of daunorubicin and doxorubicin, in addition to mitoxantrone, in cell lines expressing Pgp (8226/Dox6, MCF7/R) and MRP-1 (HL60/Adr). As expected, it did increase retention of mitoxantrone but not of anthracyclines in 8226/MR20 cells and the parental 8226 and MCF7 cells, which express BCRPR482, for which the anthracyclines are not substrates. Moreover, ortataxel had no effect in MCF7/AdVp3000 cells, which overexpress BCRPR482T. tRA96023 had effects similar to those of ortataxel, except that it had no effect in HL60/Adr cells (MRP-1) (Fig. 5b).
Fig. 5a, b

Modulation of efflux of mitoxantrone, daunorubicin and doxorubicin by ortataxel (a) and by tRA96023 (b) in cell lines expressing different MDR-associated transport proteins. Parental cell lines: HL60 (none), 8226 (BCRPR482) and MCF7 (BCRPR482). MDR cell lines: 8226/Dox6 and MCF7/R (Pgp), HL60/Adr (MRP-1), 8226/MR20 (BCRPR482) and MCF7/AdVp3000 (BCRPR482T). Cells were incubated with 3 μM mitoxantrone, 3 μM daunorubicin or 3 μM doxorubicin for 30 min (uptake), washed and resuspended in drug-free medium, then incubated for 90 min with and without 10 μM ortataxel or 10 μM tRA96023 (efflux). Retention of each drug in the presence of each modulator was compared with retention in the absence of modulator by the Kolmogorov-Smirnov statistic, expressed as a D-value ranging from 0 to 1. D-values ≥0.20 (dashed lines) were considered to indicate modulation. The graphs show the mean values from triplicate experiments, with standard errors. Ortataxel (a) effectively increased mitoxantrone, daunorubicin and doxorubicin retention in cell lines overexpressing Pgp (8226/Dox6, MCF7/R) and MRP-1 (HL60/Adr). It also increased retention of mitoxantrone, but not anthracyclines, in 8226/MR20 cells, which overexpress BCRPR482, and to a lesser extent in the parental 8226 and MCF7 cells, which also express BCRPR482. This modulation profile is in agreement with the fact that anthracyclines are not substrates for BCRPR482. Moreover, ortataxel had no effect in MCF7/AdVp3000 cells, which overexpress BCRPR482T. The effects of tRA96023 were similar to those of ortataxel in all cells except HL60/Adr (MRP-1) (b)

Effects of ortataxel and tRA96023 on resistance mediated by Pgp, MRP-1 and BCRP

Because ortataxel is cytotoxic, activity was further characterized by comparing its cytotoxicity to those of paclitaxel and docetaxel in cell lines overexpressing Pgp, MRP-1 and BCRP (Fig. 6). 8226/Dox6 and MCF7/R cells, which overexpress Pgp, were highly resistant to paclitaxel and docetaxel, but only slightly resistant to ortataxel. HL60/Adr cells, which overexpress MRP-1, had slight resistance to paclitaxel and docetaxel, but none to ortataxel. 8226/MR20 and MCF7/AdVp3000 cells, which overexpress BCRP, were not resistant to any of the taxanes tested. Thus ortataxel is not a substrate for BCRP or MRP-1 and has a lower resistance profile in Pgp-overexpressing cell lines, as compared to paclitaxel and docetaxel, most likely due to its demonstrated modulation of Pgp and MRP-1.
Fig. 6

Relative resistance of cell lines overexpressing Pgp (8226/Dox6, MCF7/R), MRP-1 (HL60/Adr) and wild type (8226/MR20) and mutant (MCF7/AdVp3000) BCRP to paclitaxel, docetaxel and ortataxel, compared to parental cells. Relative resistance was calculated as the ratio (IC50 resistant cells/IC50 wild type cells). The graphs show the mean values from triplicate experiments, with standard errors. 8226/Dox6 and MCF7/R cells, which overexpress Pgp, were highly resistant to paclitaxel and docetaxel, but only slightly resistant to ortataxel. HL60/Adr cells, which overexpress MRP-1, had slight resistance to paclitaxel and docetaxel, but none to ortataxel. 8226/MR20 and MCF7/AdVp3000 cells, which overexpress BCRP, were not resistant to any of the taxanes tested

Because tRA96023 is noncytotoxic, its activity as a modulator was assessed by studying its potentiating effect on mitoxantrone and anthracycline cytotoxicity in cell lines overexpressing Pgp, MRP-1 and BCRP. The IC50 values of mitoxantrone, doxorubicin and daunorubicin were compared in the presence and absence of 10 μM tRA96023 in the resistant cell lines, in relation to parental cell lines (Table 1). Concordant with the drug retention data, the presence of tRA96023 decreased relative mitoxantrone resistance of cells expressing Pgp (8226/Dox6, MCF7/R) and BCRPR482 (8226/MR20) but not those expressing MRP-1 (HL60/Adr) or BCRPR482T (MCF7/AdVp3000), and decreased relative resistance to doxorubicin and daunorubicin only in cells expressing Pgp. Resistance modifying factors are shown graphically in Fig. 7.
Table 1

Modulation of cytotoxicity of mitoxantrone, daunorubicin and doxorubicin by tRA96023 (10 μM) in cell lines overexpressing Pgp, MRP-1, BCRPR482 or BCRPR482T and the corresponding parental cell lines. Resistance modifying factors (RMF) were calculated as the ratios of the IC50 values in the absence and presence of tRA. Each IC50 and RMF was calculated as the mean±SD of triplicate experiments. IC50 values are in micromoles

Cell line

Resistance mechanism

Mitoxantrone

Daunorubicin

Doxorubicin

IC50

RMF

IC50

RMF

IC50

RMF

8226 (parental)

BCRPR482

0.347±0.001

1.0±0.2

0.879±0.126

0.8±0.2

0.582±0.098

1.3±0.3

8226/Dox6

Pgp, BCRPR482

1.23±0.09

12.5±1.2

1.17±0.2

3.3±0.7

1.37±0.29

6.5±0.9

8226/MR20

BCRPR482

2.4±0.8

61.0±26.5

0.913±0.081

1.3±0.3

0.776±0.174

1.9±0.23

HL60 (parental)

None

0.018±0.007

1.8±0.2

0.007±0.002

2.7±0.8

0.014±0.006

1.5±0.3

HL60/Adr

MRP-1

4.7±0.66

1.6±0.3

0.647±0.094

1.1±0.4

5.2±0.62

1.35±0.05

MCF7 (parental)

BCRPR482

0.040±0.01

2.7±0.9

0.012±0.001

1.02±0.2

0.052±0.019

3.15±1.2

MCF7/R

Pgp

0.504±0.17

6.6±2.2

0.415±0.068

19.7±7.8

2.002±0.31

14.1±3.9

MCF7/AdVp3000

BCRPR482T

134.7±7.4

2.8±0.5

2.79±0.27

1.62±0.6

24.5±5.15

3.7±1.4

Fig. 7

tRA96023 modulation of mitoxantrone, doxorubicin and daunorubicin cytotoxicity in cell lines overexpressing Pgp (8226/Dox6, MCF7/R), MRP-1 (HL60/Adr) and wild type (8226/MR20) and mutant (MCF7/AdVp3000) BCRP. The resistance modifying factor was calculated as the ratio (IC50 drug/IC50 drug+modulator). The graphs show the mean values from triplicate experiments, with standard errors. Consistent with the drug retention data, tRA96023 modulated mitoxantrone cytotoxicity only in 8226/Dox6 and MCF7/R cells, which express Pgp, and in 8226/MR20 cells, which express BCRPR482, and modulated anthracycline cytotoxicity only in Pgp-expressing cells

Discussion

We demonstrated that synthetic taxanes modulate resistance mediated by the MDR-associated ABC transport proteins MRP-1 and BCRPR482, in addition to their known modulation of Pgp. Ortataxel, which is cytotoxic, increased mitoxantrone and anthracycline retention and was more cytotoxic than paclitaxel and docetaxel in cell lines overexpressing Pgp, MRP-1 and BCRPR482. tRA96023, which is noncytotoxic, enhanced mitoxantrone and anthracycline retention and cytotoxicity in cell lines overexpressing Pgp and BCRPR482, but not those overexpressing MRP-1. Thus synthetic taxane modulators may sensitize cancers with MDR mediated by multiple ABC proteins. The finding that neither taxane modulated drug transport mediated by BCRPR482T, in contrast to their effective modulation of BCRPR482, indicates that specific mutations in this transport protein may affect not only substrate specificity [15], but also modulator specificity.

The relevance of MDR proteins has been most fully studied in acute myeloid leukemia (AML), because of the ease of obtaining tumor cells for study and the ability to correlate MDR protein expression and function with heterogeneous treatment response. Pgp [7, 13, 18, 19, 20, 21, 23] and MRP-1 [18, 19, 21] have clinical relevance in AML, and the relevance of BCRP has also been suggested [37, 41, 43, 44]. BCRP has been wild type (BCRPR482) in all cases of AML studied to date [41]; whether mutant BCRP is present in other tumor types, or only in cell lines, is yet unknown. Broad-spectrum modulators have an obvious theoretical advantage in this and other malignancies with expression of multiple ABC transport proteins.

Most MDR modulation clinical trials to date have yielded disappointing results. Trials have only targeted Pgp, despite the fact that additional ABC proteins contribute to clinical MDR. The synthetic taxanes may be promising agents for clinical application because of their broad spectrum of modulation. Nevertheless, the safety and efficacy of clinical application of broad-spectrum modulators remains to be demonstrated.

The data presented here serve to demonstrate that taxane derivatives have the potential to modulate all three MDR pumps. The efficacy of ortataxel as a cytotoxic chemotherapeutic agent is currently being evaluated [6]. The present data suggest that this drug should be effective against tumors expressing MRP-1 and BCRP, as well as those expressing Pgp [45]. Nevertheless, due to the cytotoxic properties of ortataxel, it is not strictly an MDR modulator. tRA96023 is noncytotoxic, but its lack of modulation of MRP-1 makes it less than optimal for development as a broad-spectrum clinical modulator. However, tRA96023 is one of a large library of noncytotoxic taxane derivatives lacking the C13 tubulin-binding side chain [30, 32]. This library is currently being screened to identify noncytotoxic modulators of all three MDR-associated ABC proteins.

Copyright information

© Springer-Verlag 2004