Mechanism Underlying the Reversal of Drug Resistance in P-Glycoprotein-Expressing Leukemia Cells by Pinoresinol and the Study of a Derivative

Abstract

P-glycoprotein (P-gp) is a membrane protein associated with multidrug resistance (MDR) due to its key role in mediating the traffic of chemotherapeutic drugs outside cancer cells, leading to a cellular response that hinders efforts toward successful therapy. With the aim of finding agents that circumvent the MDR phenotype mediated by P-gp, 15 compounds isolated from native and naturalized plants of Argentina were screened. Among these, the non-cytotoxic lignan (±) pinoresinol successfully restored sensitivity to doxorubicin from 7 μM in the P-gp overexpressed human myelogenous leukemia cells, Lucena 1. This resistance-reversing effect was confirmed by competitively increasing the intracellular doxorubicin accumulation and by significantly inhibiting the efflux of doxorubicin and, to a lesser extent, that of rhodamine 123. The activity obtained was similar to that observed with verapamil. No such results were observed in the sensitive parental K562 cell line. To gain deeper insight into the mode of action of pinoresinol, its effect on P-gp function and expression was examined. The docking simulations indicated that the lignan bound to P-gp at the apex of the V-shaped transmembrane cavity, involving transmembrane helices 4, 5, and 6, and partially overlapped the binding region of tariquidar, which was used as a positive control. These results would shed some light on the nature of its interaction with P-gp at molecular level and merit further mechanistic and kinetic studies. In addition, it showed a maximum 29% activation of ATP hydrolysis and antagonized verapamil-stimulated ATPase activity with an IC₅₀ of 20.9 μM. On the other hand, pinoresinol decreased the presence of P-gp in the cell surface. Derivatives of pinoresinol with improved activity were identified by docking studies. The most promising one, the non-cytotoxic 1-acetoxypinoresinol, caused a reversion of doxorubicin resistance from 0.11 μM and thus higher activity than the lead compound. It also caused a significant increase in doxorubicin accumulation. Results were similar to those observed with verapamil. The results obtained positioned these compounds as potential candidates for effective agents to overcome P-gp-mediated MDR, leading to better outcomes for leukemia chemotherapy.

Keywords: (±) pinoresinol; 1-acetoxy-(+)-pinoresinol; P-glycoprotein; multidrug resistance reversal; plant-derived compounds.

Figure 1

Chemical structures of compounds 1–16.

Figure 2

P-gp surface expression in (A) Lucena 1 and (B) K562 cells determined by flow cytometry. Unstained cells are shown with red histogram while FITC-conjugated mouse anti-human P-gp antibody stained cells are shown with blue histogram. Histograms are representative of three independent experiments.

Figure 3

Dose-response curves for cytotoxicity of doxorubicin (DOX) in Lucena 1 and K562 cells with and without pinoresinol (14) as determined by the resistance reversal assay. Values are expressed as mean ± SE of at least three independent experiments.

Figure 4

Flow cytometric analysis of the effect of pinoresinol (14) on the intracellular accumulation of doxorubicin (DOX). (A) Lucena 1 and (B) K562 cells were pre-incubated at different times with medium containing 14 at 112 μM before 1 h exposure to DOX. Each bar represents the mean ± SE. Significant differences from the ethanol control were determined at each time by using unpaired one-tailed Student's t-test (^***p < 0.001, ^**p < 0.01, ^*p < 0.05).

Figure 5

Time course of doxorubicin (DOX) accumulation in Lucena 1 cells treated with pinoresinol (14) at 112 μM or ethanol (control) and in K562 cells treated with ethanol at different periods of time. Data are expressed as mean ± SE. Significant differences at each time were determined by using two way analysis of variance (ANOVA) followed by the Bonferroni test (^{***, †††}p < 0.001, ^††p < 0.01, ^{*, †}p < 0.05); ^†: differences between Lucena 1 treated cells and Lucena 1 ethanol control and ^*: differences between K562 ethanol control and Lucena 1 ethanol control.

Figure 6

Lineweaver-Burk double reciprocal plot for the kinetic analysis of pinoresinol (14). Lucena 1 cells were cultured for 1 h with a series of concentrations of 14 before 1 h exposure to DOX. The lines were drawn using linear least squares fit. Values are expressed as mean ± SE.

Figure 7

Flow cytometric analysis of the effect of pinoresinol (14) on the efflux of doxorubicin (DOX) from (A) Lucena 1 and (B) K562 cells. After 1 h pre-incubation with 14 at 112 μM, verapamil or ethanol, cells were exposed to DOX for 1 h. Subsequently, cells were washed and then incubated in the presence of 14 at 112 μM, verapamil or ethanol at various time points in a DOX-free medium. Data points represent the mean ± SE. Significant differences from the control were determined by using unpaired one-tailed Student's t-test (^**p < 0.01, ^*p < 0.05).

Figure 8

Effect of a series of concentrations of pinoresinol (14) on the efflux of doxorubicin (DOX) in Lucena 1 cells. Data points represent the mean ± SE. Significant differences from the control were determined by using unpaired one-tailed Student's t-test (^***p < 0.001, ^**p < 0.01, ^*p < 0.05).

Figure 9

Flow cytometric analysis of the effect of pinoresinol (14) on the efflux of rhodamine 123 (Rho 123) in (A) Lucena 1 and (B) K562 cells. Both cell lines were pre-incubated with medium containing 14 at 112 μM, verapamil, trifluoperazine (TFP) or ethanol and then allowed to accumulate Rho 123 over 30 min. After washing, cells were incubated in probe-free medium in the presence of the assayed compounds for a further 30 min to allow Rho 123 extrusion. Each bar represents the mean ± SE. Significant differences from the control were determined by using unpaired one-tailed Student's t-test (^***p < 0.001, ^**p < 0.01, ^*p < 0.05).

Figure 10

Effect of 14 on ATPase activation and inhibition. Data points represent the mean ± SE.

Figure 11

Binding mode of the lowest energy conformation of (+)-pinoresinol (14) showing the main contacts of 14 with residues from TMH 4, 5, and 6.

Figure 12

Two different views of the lowest energy docked structure of 14. (A) Point of view looking to the inverted “V” within the bilayer plane, the same point of view as in Figure 11 and (B) looking from the intracellular side to outside (perpendicular to the bilayer plane). The binding pose of 14 (ball and sticks) is compared to the main and secondary sites of Rhodamine 123 (yellow tubes) and DOX (green tubes). The sites of lowest energy are opposite for Rho 123 and DOX, and the energy difference between the secondary site and the main (reported on Table 2) is below 0.5 kcal/mol in both cases. (C) Superimposition of the lowest energy poses of 14 (orange), 16 (red), and the known inhibitors verapamil (lime) and tariquidar (violet).

Figure 13

Effect of pinoresinol (14) on the surface expression of P-glycoprotein by flow cytometric analysis. (A) Lucena 1 and (B) K562 cells were stained with FITC-labeled mouse anti-human P-glycoprotein antibody at 24 and 48 h. Unstained cells are shown with red histogram, stained cells with blue histogram and cells stained and treated with 14 at 112 μM are shown with green histogram. Histograms are representative of three independent experiments.

Figure 14

Effect of a series of concentrations of 1-acetoxy-(+)-pinoresinol (16) on the accumulation of doxorubicin (DOX) in Lucena 1 cells. Significant differences from the control were determined by using unpaired one-tailed Student's t-test (^***p < 0.001, ^**p < 0.01, ^*p < 0.05).