(W1130-04-19) GSH-Dependent Inhibition of Multidrug Resistance-Associated Protein 1 (MRP1) by ZW-1226

Purpose: The multidrug resistance-associated protein 1 (MRP1) is a major ATP-binding cassette (ABC) efflux transporter that confers multidrug resistance (MDR) in cancers. Consequently, its selective inhibition is an attractive approach to enhancing the sensitivity of cancer cells to chemotherapy. Our previous research with membrane vesicles expressing ABC transporters demonstrated that the deazaflavin analog ZW-1226 is a potent and selective inhibitor of MRP1. It showed a potency 16-time greater than the widely used positive control MK-571 and displayed excellent selectivity indices, exceeding 100 over other major ABC transporters, in a GSH-dependent manner. This study aims to further confirm the GSH dependency of ZW-1226 through cell-based assays and computational methods.

Methods: To determine the inhibitory potency of ZW-1226 and known inhibitors (MK-571 and cyclosporine A) on MRP1, calcein retention assays were conducted on MRP1-overexpressing H69AR cells and the parental NCI-H69 cells. Data normalization was done against untreated parental NCI-H69 cells (100% inhibition) and untreated H69AR cells (0% inhibition). The effect of GSH was examined by pre-treating H69AR cells with buthionine sulfoximine (BSO), a GSH-depleting agent, followed by calcein retention assays with ZW-1226, MK-571, and cyclosporine A. To understand how GSH contributes to MRP1 inhibition by ZW-1226, molecular docking studies were performed with GSH and ZW-1226 using the cryo-EM structures of apo (PDB code: 5UJ9) and leukotriene C4 (LTC₄)-bound (PDB code: 5UJA) bovine MRP1 ¹. Molecular modeling was performed using the Schrödinger small molecule drug discovery suite 2021-3 (Schrödinger Inc., New York, USA).

Results: Calcein assay results indicate that calcein retention in MRP1-overexpressing H69AR cells is approximately 15% of that in the parent NCI-H69 cells after a 30-minute incubation with calcein-AM. Pre-treatment with ZW-1226 and established MRP1 inhibitors (MK-571 and cyclosporine A) led to a dose-dependent rise in calcein retention in H69AR cells, as depicted in Figure 1. The IC₅₀ for ZW-1226, the concentration needed to inhibit 50% of calcein efflux, resulting in higher calcein retention in H69AR cells, was found to be 0.5 µM (Figure 1A), consistent with the inhibitory potency previously seen in MRP1 membrane vesicles. ZW-1226 nearly restored calcein retention to the levels of parental NCI-H69 cells at doses above 8 µM. Conversely, the positive controls MK-571 (Figure 1B) and cyclosporine A (Figure 1C) only partially increased calcein retention, even at concentrations up to 200 µM. Specifically, cyclosporine A (CSA), a nonspecific inhibitor of P-gp, MRP1, and BCRP, reached a retention plateau of about 50% at a concentration of 50 µM, resulting in a high postulated IC₅₀ of 196 µM. The IC₅₀ of MK-571 was determined to be 89 µM, a potency 178-fold less than that of ZW-1226. In the parental NCI-H69 cells (Figure 1D), neither ZW-1226 nor MK-571 affected calcein retention at any concentration up to 200 µM. Moreover, pretreating H69AR cells with BSO significantly diminished the increase in calcein retention caused by ZW-1226 (Figure 2), demonstrating a GSH-dependent inhibition of MRP1. In contrast, the calcein retention induced by MK-571 and CSA was not affected by BSO pretreatments, indicating that their inhibitory effects on MRP1 are independent of GSH. Additionally, our molecular docking studies showed that ZW-1226 is poorly docked in the apo bovine MRP1 (5UJ9, Figure 3A), which represents the MRP1 conformation before accepting a substate, as indicated by the docking scores (Figure 3D). However, it aligns more closely with the cryo-EM structure of MRP1 bound to LTC₄ (5UJA, Figure 3B), a GSH-containing substrate. It forms a strong hydrogen bonding interaction with the F594 and a strong pi-pi interaction with N590 residues at LTC₄ binding pocket (Fig. 3B-B2). Both F594 and N590 have been reported to play an important role in the MRP1 transport activity. An apparent shift of ZW-1226 was also noted within the LTC₄-bound bovine MRP1 conformation, leading to an overlap between GSH and ZW-1226 in our docking poses (Figure 3C). Analysis of ETP, DOX, and E₂17βG in docking with LTC₄-bound MRP1 also suggests a potential overlap of the substrates with both GSH and ZW-1226 at the LTC₄ binding sites (data not shown). GSH and its analogs, including S-methyl GSH, γ-glu-leu-gly and ophthalmic acid, have been reported to trigger a conformational shift in MRP1, which increases the binding affinity of the modulators ². Our docking results, along with the literature report, appear to support a hypothesis that GSH or GSH-containing substrates may enhance the binding of ZW-1226 to MRP1, thereby occupying the substrate binding pockets to either block transport or render MRP1 catalytically inactive.

Conclusion: The present study demonstrates that ZW-1226 is a significantly more potent inhibitor of MRP1 compared to MK-571 and CSA. The use of the GSH-depleting agent BSO in calcein retention assays has conclusively established that the inhibitory action of ZW-1226 on MRP1 is dependent on GSH. Additionally, molecular docking studies reinforce this relationship, demonstrating a preferential binding of ZW-1226 to MRP1 in the GSH-containing LTC₄-bound state.

References: 1. Johnson, Z.L; Chen, J., Structural Basis of Substrate Recognition by the Multidrug Resistance Protein MRP1, Cell 168(6) (2017) 1075-1085 e9.
2. Hanssen, K.M., Wheatley, M.S., Yu, D.M.T., Conseil, G., Norris, M.D., Haber, M., Cole, S.P.C., Fletcher, J.I., 2022. GSH facilitates the binding and inhibitory activity of novel multidrug resistance protein 1 (MRP1) modulators. FEBS J 289, 3854-3875.

Acknowledgements: The authors would like to thank the Center for Drug Design in College of Pharmacy at University of Minnesota for supporting this research.

Figure 1. Inhibitory effect of ZW-1226 on calcein accumulation in MRP1-overexpressing H69AR cells. Representative dose-response curves of ZW-1226 (A), MK-571 (B), and CSA (C) in H69AR cells and NCI-H69 cells (D) using calcein retention assays. Calcein retention (%) indicates the proportion of calcein retained in H69AR cells compared to the parental NCI-H69 cells. For drug-treated NCI-H69 cells, this percentage represents the proportion of calcein retained in these cells compared to those treated with DMSO. Data are shown as mean ± SD from at least three independent studies (n ≥ 3).

Figure 2. Calcein retention in MRP1-overpressing H69AR cells with or without BSO pretreatment. Calcein retention (%) indicates the proportion of calcein retained in H69AR cells compared to the parental NCI-H69 cells. Data are shown as mean ± SD from three independent studies (n =3). One way or two-way ANOVA. *: p < 0.05. ns: not significant.

Figure 3. Molecular docking of ZW-1226 and GSH. (A) Docking of GSH and ZW-1226 with the apo bovine MRP1 (5UJ9). A1 and B1: the surface representations of apo bovine MRP1 docked with GSH and ZW-1226, respectively. A2 and B2: the cartoon representations of the domain structure of MRP1 with compound GSH and ZW-1226, respectively. (B) Docking of the GSH and ZW-1226 with the LCT4 bound bovine MRP1 (5UJA). A1 and B1: the surface representations of LCT4 bound bovine MRP1 docked with GSH and ZW-1226, respectively. A2 and B2: the cartoon representations of the domain structure of MRP1 with compound GSH and ZW-1226, respectively. (C) Overlay representation of GSH and ZW-1226. C1 and C2: docking overlay of GSH (teal blue) and ZW-1226 (orange) with the apo bovine MRP1 (5UJ9) and LTC4 bound bovine MRP1 (5UJA), respectively. (D) Docking scores of Etoposide, Doxorubicin, Estradiol-17β-d-glucuronide (E217βG), GSH, ZW-1226, and LTC4.