A Role for iNOS in Erastin Mediated Reduction of P-Glycoprotein Transport Activity

Abstract

The blood-brain barrier is composed of both a physical barrier and an enzymatic barrier. Tight junction (TJ) proteins expressed between endothelial cells of brain capillaries provide the physical barrier to paracellular movement of ions and molecules to the brain, while luminal-facing efflux transporters enzymatically restrict the entry of blood-borne molecules from entering the brain. The expression and activity of ATP Binding Cassette transporters or "ABC" transporters in endothelial cells of the BBB and in human tumor cells are dynamically regulated by numerous signaling pathways. P-glycoprotein (P-gp), (ABCB1), is arguably the most studied transporter of the BBB, and in human cell lines. P-glycoprotein transport activity is rapidly inhibited by signaling pathways that call for the rapid production of nitric oxide (NO) from the inducible nitric oxide synthase enzyme, iNOS. This study investigated how nano-molar levels of the selective chemotherapeutic erastin affect the activity or expression of P-glycoprotein transporter in brain capillaries and in human tumor cell lines. We chose erastin because it signals to iNOS for NO production at low concentrations. Furthermore, erastin inhibits the cellular uptake of cystine through the X_C^- cystine/glutamate antiporter. Since previous reports indicate that NO production from iNOS can rapidly inhibit P-gp activity in tumor cells, we wondered if induction of iNOS by erastin could also rapidly reduce P-glycoprotein transport activity in brain endothelial cells and in human tumor cell lines. We show here that low concentrations of erastin (1 nM) can induce iNOS, inhibit the activity of P-glycoprotein, and reduce the intracellular uptake of cystine via the Xc- cystine/glutamate antiporter. Consistent with reduced P-glycoprotein activity in rat brain capillary endothelial cells, we show that human tumor cell lines exposed to erastin become more sensitive to cytotoxic substrates of P-glycoprotein.

Keywords: P-glycoprotein; blood–brain barrier; erastin; ferroptosis; iNOS.

Figure 1

Changes in P-glycoprotein (P-gp) transport activity in brain capillaries (N = 15–20 per data point) from six rats treated with erastin. (A) Representative confocal fluorescent and DIC images of brain capillaries after a 60-min incubation with 2.0 μM NBD-CSA; note the high luminal fluorescence in the control capillary and (B) decreased luminal fluorescence in capillaries exposed to 10.0 μM PSC833 (Scale bars, 10 μm). Representative confocal images of the luminal fluorescence (LF) from (C) male and female brain capillaries associated with P-glycoprotein (P-gp) transport activity at increasing doses of erastin in brain capillaries of SD rats. Bar graphs of P-glycoprotein transport activity determined by LF in (D) male and (E) female capillaries. Graphs of LF measuring P-glycoprotein (P-gp) transport activity in male (F) and female (G) capillaries (capillary N = 15–20 per experimental data point) exposed to 0.001 μM erastin over time. Dotted red line denotes multiple comparison. Significance is as compared to control unless otherwise specified: * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 2

Breast Cancer Resistance Protein (BCRP) and Multidrug Resistance Protein 2 (MRP2) transport activities in brain capillaries (N = 15–20) from six rats treated with erastin. (A) Males and (B) females are graphs of BCRP and MRP2 transport activity in brain capillaries of SD rats treated with 0.001 μM erastin for 3 h. SE and significance were determined by one-way ANOVA and Tukey multiple comparison. Significance is as compared to control unless otherwise specified: *** p < 0.001.

Figure 3

Transport reversibility following erastin removal. Capillaries treated with erastin (0.001 μM) for 3 h then removed. (A) Male and (B) female brain capillaries are graphs showing measured P-glycoprotein (P-gp) transport activities in brain capillaries (N = 15–20 per data point) from six rats before and after erastin removal. SE and significance were determined by one-way ANOVA and Tukey multiple comparison. Significance is as compared to control unless otherwise specified: * p < 0.05, *** p < 0.001.

Figure 4

Determining P-glycoprotein (P-gp) levels after erastin (0.001 μM) treatment. (A) Qualitative Western blots determined P-glycoprotein (P-gp) protein levels in erastin (Er), and vehicle control (VC) treated male and female SD rats (Animal N# = 6 pooled/well) brain capillary membrane lysates. (B) Quantitative P-glycoprotein Western blot of male rat brain capillary lysates performed in triplicate and measured by normalization to actin (6 rat brains/well). (C) Graph of P-glycoprotein band density units in males comparing erastin-treated (Er) to vehicle control (VC). Band density levels are found in Supplementary Figures S1 and S2.

Figure 5

Measuring erastin’s effect on cystine uptake and co-dosed with antioxidants, cysteine, or NAC. (A) (Left) Illustration of the assay used to measure X_C⁻ antiporter activity. (Right), Graph of measured fluorescence produced from selenocystine uptake through the X_C⁻ antiporter. Measured fluorescence of untreated rat brain capillaries with selenocystine (SC) or without selenocystine (-SC). SC + Er, selenocystine with erastin (0.001 μM). SC + SZ, selenocystine with 500.0 μM sulfasalazine an X_C⁻ inhibitor. (B) Male and (C) Female rat brain capillaries (N = 15–20) isolated from 6 rats per experiment and exposed to 0.001 μm erastin (Er), cystine alone (Cystine) or cystine with erastin (Cystine + Er). (D) Male and (E) female rat brain capillaries (N = 15–20) exposed to 0.001 μm erastin (Er), NAC alone or NAC with erastin (NAC + Er). SE and significance were determined by one-way ANOVA and Tukey multiple comparison. Significance is as compared to controls: ** p < 0.01, *** p < 0.001.

Figure 6

iNOS induction is required for erastin’s decrease in P-glycoprotein transport activity. Measurements of P-glycoprotein transport activity in (A) male and (B) female rat brain capillaries (N = 15–20) from 6 rats were treated with vehicle control (VC), erastin (Er), erastin with 100 μM of pan-NOS inhibitor L-NAME (L-NAME + Er), and erastin with 10 μM of the selective iNOS inhibitor 1400 W (1400 W + Er). (C) Western blot of vehicle and 0.001 μM erastin-treated male rat brain capillaries (pooled from 6 rats per well) and (D) graph of band density of iNOS normalized to loading control β-actin levels. SE and significance were determined by one-way ANOVA and Tukey multiple comparison. Significance is as compared to controls: *** p < 0.001.

Figure 7

Erastin increases the cytotoxicity of P-gp substrate in human cells in vitro that express P-glycoprotein. Graph representing (A) the percent of OVCAR-8 (non-expressing P-glycoprotein) and (B) NCI/ADR-RES (P-glycoprotein)-expressing cells surviving 72 h following adriamycin (0.01–10.0 μM) exposures with or without erastin (100 nM) or with erastin (100 nM) co-dosed with the iNOS inhibitor, 1400 W. Values represent three separate experiments carried out in triplicate. SE and significance were determined by one-way ANOVA and Tukey multiple comparison. Significance is as compared to control: * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 8

Proposed model of how erastin reduces P-gp transport. Erastin blocks transport of cystine into the capillary via system X_C⁻, decreasing intracellular cystine concentration and leading to a decrease in precursors for glutathione (GSH) production within the capillary. Decreases in GSH lead to increased oxidative stress, which indirectly leads to increases in iNOS activity that nitrosylate residues within P-glycoprotein. Although not a focus of this work, erastin also affects ion channels (VDAC) of the mitochondria and dysregulates TP53 expression and activity, which influences the expression of key genes important in both death and survival. Exogenous addition of the GSH precursors (NAC/cystine) mitigates the inhibitory effects of erastin on P-glycoprotein transport activity.