ATP-dependent transport of lipophilic cytotoxic drugs by membrane vesicles prepared from MRP-overexpressing HL60/ADR cells.

MRP is an ATP-binding cassette family transporter that confers cellular resistance to a variety of natural product cytotoxic agents. However, the biochemical mechanism by which MRP confers resistance has not been established. To gain insight into its mechanism of action, the in vitro substrate specificity of MRP was examined by analyzing drug uptake into membrane vesicles prepared from MRP-overexpressing HL60/ADR cells. Compared to control HL60 membrane vesicles, HL60/ADR membrane vesicles exhibited markedly enhanced ATP-dependent transport of daunorubicin, etoposide, and vincristine. In contrast, little or no increased uptake was observed for vinblastine and Taxol. This pattern of in vitro substrate specificity was consistent with the analysis of the HL60/ADR drug resistance phenotype, which revealed substantial levels of resistance to anthracyclines, etoposide, and vincristine, but only slightly increased resistance to vinblastine and Taxol. Drug transport into HL60/ADR membrane vesicles was osmotically sensitive and dependent on ATP concentration, with a K(m) value of 45 microM for ATP. Lineweaver-Burk analysis indicated that substrate transport was concentration-dependent, with apparent K(m) values of 6.1, 5.7, and 5.5 microM for daunorubicin, etoposide, and vincristine, respectively. The P-glycoprotein-modulating agents cyclosporin A, PSC833, and verapamil, which have modest reversing effects on MRP-overexpressing cell lines, were weak competitive inhibitors of daunorubicin transport, with Ki values of 35, 84, and 95 microM, respectively. In addition, the glutathione analog azidophenacyl-glutathione, oxidized glutathione, and the LTD4 antagonist MK571 were competitive inhibitors of daunorubicin transport with Ki values of 69, 31, and 3.0 microM, respectively. Genistein, an MRP-specific modulating agent, and arsenate, a compound for which MRP has previously been reported to confer resistance, were also competitive inhibitors, with Ki values of 17 and 29 microM, respectively. These results are consistent with a previous report in which we demonstrated that HL60/ADR membrane vesicles transport azidophenacylglutathione and that transport of this agent is competitively inhibited by daunorubicin, vincristine, and etoposide [Shen et al., (1966) Biochemistry 35, 5719-5725]. Together, these uptake studies performed with HL60/ADR membrane vesicles constitute a consistent body of evidence that indicates that MRP transports both glutathione S conjugates and unaltered natural product drugs and support the idea that the direct transport of unaltered lipophilic cytotoxic drugs is the predominant biochemical mechanism whereby MRP confers multidrug resistance.

ATP-dependent uptake of natural product cytotoxic drugs by membrane vesicles establishes MRP as a broad specificity transporter.

MRP is a recently isolated ATP-binding cassette family transporter. We previously reported transfection studies that established that MRP confers multidrug resistance [Kruh, G. D., Chan, A., Myers, K., Gaughan, K., Miki, T. & Aaronson, S. A. (1994) Cancer Res. 54, 1649-1652] and that expression of MRP is associated with enhanced cellular efflux of lipophilic cytotoxic agents [Breuninger, L. M., Paul, S., Gaughan, K., Miki, T., Chan, A., Aaronson, S. A. & Kruh, G. D. (1995) Cancer Res. 55, 5342-5347]. To examine the biochemical mechanism by which MRP confers multidrug resistance, drug uptake experiments were performed using inside-out membrane vesicles prepared from NIH 3T3 cells transfected with an MRP expression vector. ATP-dependent transport was observed for several lipophilic cytotoxic agents including daunorubicin, etoposide, and vincristine, as well as for the glutathione conjugate leukotriene C4 (LTC4). However, only marginally increased uptake was observed for vinblastine and Taxol. Drug uptake was osmotically sensitive and saturable with regard to substrate concentration, with Km values of 6.3 microM, 4.4 microM, 4.2 microM, 35 nM, and 38 microM, for daunorubicin, etoposide, vincristine, LTC4, and ATP, respectively. The broad substrate specificity of MRP was confirmed by the observation that daunorubicin transport was competitively inhibited by reduced and oxidized glutathione, the glutathione conjugates S-(p-azidophenacyl)-glutathione (APA-SG) and S-(2,4-dinitrophenyl)glutathione (DNP-SG), arsenate, and the LTD4 antagonist MK571. This study establishes that MRP pumps unaltered lipophilic cytotoxic drugs, and suggests that this activity is an important mechanism by which the transporter confers multidrug resistance. The present study also indicates that the substrate specificity of MRP is overlapping but distinct from that of P-glycoprotein, and includes both the neutral or mildly cationic natural product cytotoxic drugs and the anionic products of glutathione conjugation. The widespread expression of MRP in tissues, combined with its ability to transport both lipophilic xenobiotics and the products of phase II detoxification, indicates that the transporter represents a widespread and remarkably versatile cellular defense mechanism.

Cellular and in vitro transport of glutathione conjugates by MRP.

MRP is a recently identified ATP-binding cassette transporter. We previously established that MRP confers resistance to a spectrum of natural product cytotoxic drugs [Kruh, G.D., (1994) Cancer Res. 54, 1649-1652], that expression of MRP is associated with enhanced drug efflux [Breuninger, L.M., (1995) Cancer Res. 55, 5342-5347], and that MRP transcript is widely expressed in human tissues and solid tumor cell lines [Kruh, G.D., (1995) J. Natl. Cancer Inst. 87, 1256-1258]. In the present study the relationship between MRP and drug glutathione S-conjugates was examined. We observed that MRP was labeled by azidophenacylglutathione (APA-SG), a photoaffinity analog of glutathione, and that inside-out membrane vesicles prepared from MRP-overexpressing HL60/ADR cells transported this compound. Transport into membrane vesicles was ATP-dependent, sensitive to osmolarity, and saturable with regard to APA-SG and ATP concentrations, with Km values of 15 and 61 microM, respectively. APA-SG transport was competitively inhibited by the natural product cytotoxic drugs daunorubicin, vincristine, and etoposide, with Ki values of 4.8, 3.8, and 5.5 microM, respectively. Oxidized glutathione, the drug-glutathione S-conjugate DNP-SG, the LTD4 antagonist MK571 and arsenate were also competitive inhibitors, with Ki values of 9.0, 23.4, 1.1, and 15.0 microM, respectively. Analysis of the fate of monochlorobimane in MRP transfectants revealed reduced intracellular concentrations of drug-glutathione S-conjugates associated with enhanced efflux and altered intracellular distribution. These results indicate that MRP can transport glutathione conjugates in vitro and in living cells and suggest the possibility that the transporter may represent a link between cellular resistance to some classes of cytotoxic drugs and glutathione-mediated mechanisms of resistance. In addition, the observation that both mildly cationic or neutral natural product cytotoxic drugs and anionic compounds such as DNP-SG, MK571, and arsenate are competitive inhibitors of MRP action suggests that the substrate specificity of the transporter is quite broad.

Expression of multidrug resistance-associated protein in NIH/3T3 cells confers multidrug resistance associated with increased drug efflux and altered intracellular drug distribution.

Multidrug resistance is a major obstacle to cancer treatment. Using an expression cDNA library transfer approach to elucidating the molecular basis of non-P-glycoprotein-mediated multidrug resistance, we previously established that expression of multidrug resistance protein (MRP), an ATP-binding cassette superfamily transporter, confers multidrug resistance (G. D. Kruh et al., Cancer Res., 54: 1649-1652, 1994). In the present study, we generated NIH/3T3 MRP transfectants without using chemotherapeutic drugs to facilitate the pharmacological analysis of the MRP phenotype. MRP transfectants displayed increased resistance to several lipophilic drugs, including doxorubicin, daunorubicin, etoposide, actinomycin D, vincristine, and vinblastine. However, increased resistance was not observed for Taxol, a drug for which transfection of MDR1 confers high levels of resistance. Verapamil increased the sensitivity of MRP transfectants relative to control transfectants, but reversal was incomplete for doxorubicin and etoposide, the drugs for which MRP conferred the highest resistance levels. For the latter two drugs, MRP transfectants, which were approximately 8- and approximately 10-fold more sensitive than control cells in the absence of verapamil, exhibited 3.8- and 3.3-fold relative sensitization with 10 microM verapamil, respectively, but remained approximately 2 and approximately 3-fold more resistant than control cells. Analysis of drug kinetics using radiolabeled daunorubicin revealed decreased accumulation and increased efflux in MRP transfectants. Confocal microscopic analysis of intracellular daunorubicin in MRP transfectants was consistent with reduced intracellular drug concentrations, and also revealed an altered pattern of intracellular drug distribution characterized by the initial accumulation of drug in a perinuclear location, followed by the development of a punctate pattern of drug scattered throughout the cytoplasm. This pattern was suggestive of a process of drug sequestration, possibly followed by vesicle transport. Both increased drug efflux and perinuclear drug accumulation are consistent with the reported localization of MRP in plasma and cytosolic membranes (N. Krishnamachary and M. S. Center, Cancer Res., 53: 3658-3663, 1993; M. J. Flens et al., Cancer Res., 54: 4557-4563, 1994). These results thus indicate that the drug specificity of MRP is quite similar to that of MDR1, but also suggest potential differences in Taxol specificity and the level of verapamil sensitivity. In addition, these results indicate that MRP functions to extrude drug from the cell, but additionally suggest the intriguing possibility that drug sequestration contributes to drug resistance by protecting cellular targets and/or contributing to drug efflux.

Hydrocortisone regulation of interleukin-6 protein production by a purified population of human peripheral blood monocytes.

The direct effect of the endogenous glucocorticoid (GC) hydrocortisone (HC) on interleukin-6 (IL-6) production was examined using purified populations of human peripheral blood monocytes (Mo). Lipopolysaccharide (LPS) induced IL-6 production by Mo in a dose-dependent fashion. IL-6 was detected in Mo supernatants as early as 2 hr after stimulation, with peak IL-6 production observed by 16 hr. Simultaneous addition of HC and LPS resulted in a significant decrease of IL-6 production 4 hr after LPS treatment, with maximum inhibition observed at 16-24 hr. An attenuation of the inhibitory effect of HC occurred with greater concentrations of LPS and with the delay of HC addition until after LPS. However, there was no correlation between the quantity of IL-6 produced by Mo and the level of HC inhibition. The inhibitory effect of HC was greater if LPS, rather than IL-1 beta, were used as a stimulus to induce IL-6 production. The EC50 of LPS-induced IL-6 production by HC was 2.0 x 10(-7) M. The inhibitory effect of HC on LPS-stimulated IL-6 production was GC specific and receptor mediated because: (i) equivalent inhibition was not observed with other endogenous steroids and (ii) equimolar amounts of the GC antagonist RU 486 blocked the GC-mediated effect.