Potentiates in Vivo the Therapeutic Efficacy of Doxorubicin against Multidrug Resistance Protein-expressing Tumors’

Intracellular glutathione (GSH) has been implicated as a regulatory determinant of multidrug resistance protein (MRP) function. The objective of the present study was to evaluate in vivo the ability of D,L-buthionine-(S,R)-sulfoximine (D,L-BSO), a potent inhibitor of GSH biosynthesis, to reverse MRP-mediated drug resistance to doxorubicin. Athymic nude mice (nu/nu) bearing advanced parental human fibrosarcoma HT1O8O and MRP-expressing HT1O8O/ DR4 tumors were treated with the maximum tolerated dose of doxorubicin (10 mg/kg, i.v. push). This therapy produced an overall response rate of 50% (20% complete response and 30% partial response) in mice bearing parental HT1O8O xenografts, whereas no significant antitumor activity against HT1O8O/DR4 tumors was observed. Treatment of mice bearing HT1O8O and HT1O8OIDR4 xenografts with a continuous i.v infusion of nontoxic doses of D,L-BSO (300 and 600 mg/kg/day) produced a 60% reduction of GSH plasma levels and greater than 95 % reduction in GSH tumor levels in both parental and multidrug-resistant tumors; however, this treatment possessed no in vivo antitumor activity by itself. Under these treatment conditions, a combination of D,L-BSO with the maximum tolerated dose of doxorubicin administered at 24 h during a 48-h i.v. infusion of D,L-BSO cornpletely restored the response of MRP-expressing HT1O8O/ DR4 tumors to doxorubicin (overall response rate, 63%; complete response rate, 38%) with no potentiation of host toxicity The D,L-BSO-induced in vivo reversal of MRPmediated drug resistance correlated in vitro with the restoration of intracellular doxorubicin retention in cultured HT1O8O/DR4 cells. Depletion of GSH by D,L-BSO in drugReceived 4/I 1/96; revised 8/28/96: accepted 9/18/96. I Supported in part by Program Grant CA 13038. CA 09581-07 T32 Training Grant. Grant ACS-DHP-l IS of the American Cancer Society. and by a special research fund of the West German Cancer Center. B. S. S. is a NIH T32 Laboratory Research Fellow.’ 2 To whom requests for reprints should be addressed, at Department of Experimental Therapeutics. Grace Cancer Drug Center. Roswell Park Cancer Institute, Elm and Carlton Streets, Buffalo, NY 14263. Phone: (716) 845-4532: Fax: (716) 845-8857. sensitive HT1O8O tumors that do not express MRP did not alter the in vivo response to doxorubicin. Using the same treatment schedule, dose, and administration of doxorubicin with and without D,L-BSO in nude mice bearing P-170 glycoprotein-expressing A2780IDx5 tumors, no potentiation of the therapeutic index of doxorubicin was found, demonstrating the in vivo selectivity of D,L-BSO-induced GSH depletion on MRP-function. The data reported herein indicate that in vivo function of MRP as a mediator of doxorubicin resistance requires the presence of sufficient GSH pools. D,L-BSO may provide an example of an effective in vivo modulator of MRP-mediated drug resistance. INTRODUCTION MDR3 has been associated with overexpression of a 190kDa et’flux pump ofthe ABC superfamily oftransporter proteins ( 1-5) termed MRP ( I ). Transfection of a MRP eDNA expression vector into HeLa, SW-IS73, and NIH/3T3 cells conferred a MDR phenotype (6-8) associated with alterations in drug accumulation and retention (7-9). Several agents capable of reversing Pgp-associated MDR. such as verapamil and cyclosporine analogues. have entered clinical trials; however, drug resistance-modifying agents effective in reversal of MRP-mediated MDR are limited ( 10-14). Recent studies indicate that MRP is not only an export pump for GSI-I conjugates ( I 5, 16) but also for glucuronidated xenobiotics (i.e.. glucuronosyl-etoposide, Ref 17). In addition, MRP has been related to the multispecific organic anion transporter ( 16). thus. conjugation or complex formation of drugs with GSH or negative-charged compounds has been proposed as well. However. the requirement of drug conjugation or complex formation for MRP-mediated drug transport remains controversial (18-20). In vitro, inhibition of intracellular GSH synthesis by D,LBSO resulted in increased drug accumulation and retention of anthracyclines (18-20), vincnistine (21, 22), etoposide (21 ). and rhodamine (18) in MRPbut not Pgp-mediated MDR (18-20. 23). Although the mechanisms of involvement of the GSH pathway in MRP-mediated MDR are still poorly understood. these in vitro studies suggest that sufficient intracellular GSH levels are required for MRP-mediated drug transport of at least several xenobiotics. We recently established MRP-overexpressing xenografts of the multidrug-resistant human fibrosarcorna cell line 3 The abbreviations used are: MDR. multidrug resistance: Pgp. PI 70 glycoprotein: MRP. multidrug resistance protein: I),L-BSO. i).i.-huthionine-(S,R)-sulfcaimine: mAb, monoclonal antibody: GSH. glutathione: Rh-I23, rhodamine-123: MTD, maximum tolerated dose: PR. partial response: CR, complete response: RTV. relative tumor volume. Research. on October 23, 2017. © 1996 American Association for Cancer clincancerres.aacrjournals.org Downloaded from 1962 In Vivo Effect of Glutathione Depletion on MRP Function HT1O8O/DR4 (3, 24) in athymic nude mice (nu/nu) highly resistant to doxorubicin and partially cross-resistant to paclitaxel treatment (14). Pynidine analogue PAK-lO4P (25), a potent modulator of MRPand Pgp-mediated MDR function (26, 27), restored in vivo drug sensitivity to doxorubicin (partially) and to paclitaxel (completely, Ref. 14). The present study investigated the in s’ii’o selectivity of D,L-BSO-induced GSH depletion by comparing its effect on the therapeutic efficacy of doxorubicin against parental HT1O8O and multidrug-resistant MRP-expressing HTIO8OIDR4 xenografts and Pgpbut not MRP-expressing A2780IDx5 xenografts. MATERIALS AND METHODS Mice. Eight to 12-week-old female athymic nude mice (nu/nu; body weight, 20-25 g) were obtained from Harlan Sprague-Dawley, Inc. (Indianapolis, IN). They were kept (five mice/cage) under specific pathogen-free conditions with water and food ad /ibitu,n. Tumor. The human fibrosarcoma cell line HTIO8O (parental) and the MDR subline HT1O8OIDR4 (MRPand LRPpositive, Pgp-negative) have been characterized earlier (3, 24, 28). Monolayer cultures of HT1O8O and HTIO8OIDR4 cell lines were maintained in Eagle’s MEM supplemented with Earle’s balanced salt solution, 10% heat-inactivated FCS, nonessential amino acids, and L-glutamine. HTIO8O and HTIO8OIDR4 Xenografts were established as described previously (14). Nonneerotic tumor tissue of approximately 50 mg was transplanted s.c. (via trocar), and drug treatments were initiated when tumor sizes reached 150-200 mg. The human ovarian cancer cell line A2780/wt (parental) and the MDR subline A278OIDxS [Pgp-positive (29), MRPnegative data not shown] have been characterized previously. A2780 and A2780IDx5 xenografts were established in athymic nude mice by injection of 106 cells s.c. Drug treatments were initiated when tumor sizes reached 150-200 mg. Immunohistochemical Characterization of MRP Expression. Tumor tissue of HT1O8O and HT1O8OIDR4 tumors was processed for paraffin sections (S rim) after 10% neutral buffered formalin fixation. Microwave antigen retrieval was applied in 6 M urea, and nonspecific staining was blocked with 0.03% casein in PBS as described previously (30). MRP-specific binding of mouse mAb MRPm6 (1 .0 ig/ml, Ref. 31 ; a gift from Prof. R. J. Scheper, Free University Hospital, Amsterdam, the Netherlands) was visualized using a polyvalent biotinylated secondary antibody and streptavidin-peroxidase reagent (Shandon-Lipshaw, Pittsburgh, PA) using 3,3’-diaminobenzidine (Sigma, St. Louis, MO) as chromogen. Isotype-matched mouse IgGl (Sigma) served as a negative control. Drug Treatment. Doxorubicin (Farmitalia Carla Erba SPA., Milan, Italy) was dissolved in sterile 0.9% saline and injected iv. push at the MTD of 10 mg/kg. D,L-BSO (Sigma Co.) was dissolved in sterile 0.9% saline, filtered through a 0.2-p.m polysulfone membrane filter (Whatman, Maidstone, United Kingdom), and administered by 48-h continuous iv. infusion at a dose of 300 mg/kg/day and 600 mg/kg/day starting at 24 h before doxorubicin administration. All solutions were prepared fresh and used immediately. MTD was defined as the maximum dose that could be administered to tumor-bearing mice without causing drug-related lethality and body-weight loss more than 20%. All studies were performed in accordance with Institutional Animal Care and Use Committee guidelines. Tumor Measurements and Body Weight. Tumor weight (milligrams) was determined by measurements of the longest axis (L) and short axis (W) of each tumor with a vernier caliper and calculated assuming unit density by the following formula: #{189} (L X W2). RTV was calculated using the formula RTV (%) = VJV, x V x 100, where V1 represents the tumor volume on day x and V represents the initial tumor volume. As a general policy, mice were sacrificed when the tumor sizes exceeded 2.0 g. To evaluate drug toxicity, body weights of mice were recorded at the time of tumor measurements. Antitumor Activity. Antitumor activity was assessed by maximum inhibitory rate of the RTV of the treated over the untreated controls. Tumor doubling time was defined as the mean time for the tumor to reach twice the initial size. PR was defined as a reduction in tumor size greater than 50%, and CR was defined as the inability to detect a tumor by palpation at the initial site of the tumor appearance for more than 60 days posttreatment (cures). GSH Determination. GSH levels in cultured cells. munine plasma, and xenografts were measured according to the GSH reductase recycling method of Griffith (32). To determine GSH content in tumor tissue, samples were washed immediately in ice-cold PBS after tumor extirpation, then homogenized in 10 volumes of ice-cold 125 mtvi phosphate buffer (pH 7.4) using S strokes of a polytrone (Kinematica Inc., Luzern, Switzerland). Samples were centrifuged at 120,000 X g for 20 mm at 2#{176}C, and the supernatant was deproteinized with a final concentration of 2% (w/v) sulfosalicylic acid (Sigma). Absorbance was measured at 412 nm for S mm at 30#{176}Cusing a computer-connected spectrophotometer (BIO-TEK Instruments, Inc., Winooski, VT). GSH concentrations were calculated by reference to a standard curve that was run with each batch of samples. Results are expressed as nanomoles/l06 cells, micromoles/liter plasma, and nanomoles/gram tumor weight. Quantitative Flow Cytometric Determination. Drug efflux studies in cultured HTIO8O and HTIO8O/DR4 cells were performed as described previously ( 14, 27). In brief, cells in exponential growth were pretreated for 24 h with 1-100 i.M D,L-BSO or drug-free medium, then exposed for 2 h at 37#{176} C to equicytotoxic concentrations of doxorubicin using the IC50 of HT1O8O cells (0.2 p.M doxorubicin) and HT1O8OIDR4 cells (55.0 I.LM doxorubicin), respectively, or for 20 mm to a noncytotoxic equimolar concentration of Rh1 23 ( I 3.0 pM; Molecular Probes, Eugene, OR). Samples were taken immediately after drug exposure (zero time, drug accumulation) and after 4 h in drug-free medium (drug retention) in the presence or absence of D,L-BSO. Previous studies from our laboratory have shown that the efflux of doxorubicin and Rh-l23 in HTIO8O/DR4 cells is time-dependent (27). Cellular drug concentrations were determined by analyzing cells with a FACScan (Beeton Dickinson, San Jose, CA). Excitation wavelength was 488 nm, and doxorubicin emission was captured using a 650 long-pass filter, whereas Rh123 emission was captured using a 530/30 nm band-pass filter. Results were calculated and analyzed with Win List software (Verity Software House, Topsham. ME). Cellular Research. on October 23, 2017. © 1996 American Association for Cancer clincancerres.aacrjournals.org Downloaded from Clinical Cancer Research 1963