The antidote effect of quinone oxidoreductase 2 inhibitor against paraquat-induced toxicity in vitro and in vivo

Abstract

BACKGROUND AND PURPOSE The mechanisms of paraquat (PQ)-induced toxicity are poorly understood and PQ poisoning is often fatal due to a lack of effective antidotes. In this study we report the effects of N-[2-(2-methoxy-6H-dipyrido{2,3-a:3,2-e}pyrrolizin-11-yl)ethyl]-2-furamide (NMDPEF), a melatonin-related inhibitor of quinone oxidoreductase2 (QR2) on the toxicity of PQ in vitro & in vivo. EXPERIMENTAL APPROACH Prevention of PQ-induced toxicity was tested in different cells, including primary pneumocytes and astroglial U373 cells. Cell death and reactive oxygen species (ROS) were analysed by flow cytometry and fluorescent probes. QR2 silencing was achieved by lentiviral shRNAs. PQ (30 mg·kg(-1)) and NMDPEF were administered i.p. to Wistar rats and animals were monitored for 28 days. PQ toxicity in the substantia nigra (SN) was tested by a localized microinfusion and electrocorticography. QR2 activity was measured by fluorimetry of N-benzyldihydronicotinamide oxidation. KEY RESULTS NMDPEF potently antagonized non-apoptotic PQ-induced cell death, ROS generation and inhibited cellular QR2 activity. In contrast, the cytoprotective effect of melatonin and apocynin was limited and transient compared with NMDPEF. Silencing of QR2 attenuated PQ-induced cell death and reduced the efficacy of NMDPEF. Significantly, NMDPEF (4.5 mg·kg(-1)) potently antagonized PQ-induced systemic toxicity and animal mortality. Microinfusion of NMDPEF into SN prevented severe behavioural and electrocortical effects of PQ which correlated with inhibition of malondialdehyde accumulation in cells and tissues. CONCLUSIONS AND IMPLICATIONS NMDPEF protected against PQ-induced toxicity in vitro and in vivo, suggesting a key role for QR2 in the regulation of oxidative stress.

Figure 1

Protective effect of the QR2 inhibitor against PQ-induced cell death in vitro. (A) Structure of QR2 inhibitor NMDPEF. (B) Dose-dependent effect of NMDPEF against PQ-induced cell death in human astroglial cells U373. Cells were treated with 100 µM PQ and different concentrations of NMDPEF. Cell death was assessed by Trypan blue and FACS analysis 72 h post-treatment. (C) Long-term protective effect of NMDPEF (20 µM) against different concentrations of PQ was analysed in U373 cells for 7 days. (D) Phase-contrast micrographs of U373 cells treated with 100 µM PQ +/– 20 µM NMDPEF taken 72 h post-treatment. Bar 100 µm. (E,F) Weak and transient protective effect of melatonin (E) and a NADPH inhibitor apocynin (F) against PQ-induced cell death in U373 cells. The experiments were performed as in B but the cells were analysed at 66 and 72 h post-treatment. (G) Protective effect of NMDPEF against PQ-induced cell death in primary rat alveolar epithelial cells. Cell death was assessed by Trypan blue and FACS analysis. (H) Phase-contrast micrographs of primary pneumocytes treated with 20 µM PQ with or without 20 µM NMDPEF for 72 h. Bar 50 µM. The graphs in (B), (E) and (G) show the mean ± SEM of 3 independent experiments performed in triplicate (n= 9). The graphs C and G show the mean ± SD, n= 3, of a representative experiment out of at least three independent experiments performed in triplicate that yielded similar results. *P < 0.05, **P < 0.01; significant effect of NMDPEF, n= 9.

Figure 2

Effect of NMDPEF and PQ on QR2 activity is dependent on the assay. (A) QR2 reaction progress in the presence of substrate K3 (menadione, 100 µM), co-substrate BNAH (100 µM), U373 cell lysate (2.5 µg) and indicated conc. of NMDPEF or melatonin (100 µM) or apocynin (100 µM). The decrease in co-substrate (BNAH) fluorescence was measured by fluorimetry in triplicate. Best-fitted curves to all experimental points are shown. (B) QR2 activity in the presence of NMDPEF, melatonin (MLT) and apocynin (APO) at indicated concentrations. The reactions were carried out as described in (A) and QR2 activity was calculated. The graph shows the mean ± SEM of three independent experiments (n= 9). *P < 0.01; significantly different from control. (C) No effect of PQ on QR2 activity in vitro. U373 cell lysate (1 µg) was incubated with 100 µM PQ and 20, 1 and 0.1 µM NMDPEF in the presence or absence of K3. The graph shows the mean ± SEM of three independent experiments (n= 9). (D) Effects of PQ and NMDPEF on QR2 activity in whole cells. Cells were treated for 24 h with PQ (100 µM) with or without NMDPEF (1, 20 µM), lysed and 1 µg of the lysates (in quadruplicate) was assayed in the presence of different concentrations of substrate as indicated above. Michaelis-Menten curves were fitted by non-linear regression. The graph shows a representative experiment out of three independent experiments performed in quadruplicate, means ± SEM Note, QR2 activity is enhanced in cells exposed to PQ for 24 h and blocked by cytoprotective doses of NMDPEF.

Figure 3

QR2 inhibitor NMDPEF prevents ROS production induced by PQ, hydrogen peroxide and rotenone, but the protective effect is greater against PQ. The oxidative stress was measured by an MFI of MitoSOX using FACS. (A) U373 cells were treated with PQ (100 µM) or rotenone (10 µM) + /– NMDPEF (20 µM). Six, 10, 24 h later the cells were loaded with MitoSOX, trypsinized and analysed by FACS. (B) Representative fluorescence histograms of the experiment shown in A (two histograms for each 24 h treatment). (C) U373 cells were treated with PQ (100 µM) and or H₂O₂ (1 mM) and APF probe was used to quantify ROS levels. (D) Representative FITC channel fluorescence histograms of the experiment shown in C (two histograms for each 11 h treatment). Data represent the mean ± SEM, n= 6, of three independent experiments performed in duplicate. *P < 0.05, **P < 0.01, significant differences between untreated versus PQ-, rotenone- or H₂O₂-treated cells. ^#P < 0.05, significant effect of NMDPEF. (E-H) NMDPEF blocks lipid peroxidation in cells exposed to PQ. (E) Cells were treated for 20 h with 100 µM PQ plus 20 µM NMDPEF or with PQ plus vehicle (DMSO) and compared with control cells (DMSO treated). MDA was measured by HPLC. Mean ± SEM, n= 3, of three independent experiments. (F) Representative HPLC profiles of MDA-derivatization products extracted from control cells, (G) cells treated with PQ (100 µM) for 20 h and (H) cells co-treated with indicated concentrations of PQ and NMDPEF for 20 h. ABS, 280 nm absorbance; red circles, MDA derivatization product – 2,4-dinitrophenylhydrazone; ^#P < 0.05, significant difference between untreated and PQ-treated cells. *P < 0.05, significant effect of NMDPEF.

Figure 4

QR2 silencing attenuates cell death induced by PQ. (A) U373 cells were infected with four different lentiviruses encoding shRNA against human QR2 (shQR2) or control lentivirus [Lenti GFP and shRNA control (shCon)] and efficacy of QR2 silencing was assessed by Western blot 5 days after infection. Cell lysate (40 µg) was run on 8% SDS-PAGE gel. (B) In parallel, the cells analysed in A, were seeded in 24 well plates and incubated in the presence of PQ. Seventy-two hours post-treatment the PQ-induced toxicity was assessed by colorimetric assay for LDH content. This experiment was performed three times and yielded comparable results. Means ± SD, n= 4, of a representative experiment. The values start from the mean serum LDH background absorbance. (C) Human U373 cells infected as in A, with anti- hQR2 shRNA h10 and h11, lentiviruses effective in blocking the expression of QR2 (see Western blot in A), were incubated for 50 h with 1 mM PQ (left panel) or 72 h with 100 µM PQ (right panel). The PQ treatment was started 8 days post-infection (Day 0). Cell mortality was assessed by Trypan blue and FACS. Means ± SD, n= 3, of a representative experiment out of three experiments that yielded comparable results. (D) Western blot analysis of QR2 expression in U373 cells used in (C). Cells were lysed on day 0 of the experiment showed in C and analysed as in (A). (E) Western blot analysis of QR2 silencing in mouse epithelial cells EpH4. Cells were infected with lentiviruses producing shRNA against mouse QR2 (mQR2) and selected with puromycin for 8 days. Cell lysate (50 µg) was run on a gradient (4–12%) NuPage gel. (F) High resistance to a high and low dose of PQ in EpH4 shRNA m5 (shm5) expressing cells compared with pLKO-control shRNA cells. Eighty per cent confluent EpH4 cells, were treated with 100 µM PQ and analysed 30 h later (left panel) or with 10 µM PQ and analysed 72 h later (right panel) as described for C. Means ± SD, n= 3, of a representative experiment out of three independent experiments, performed in triplicate that yielded comparable results. *P < 0.05, statistically significant differences for shRNA QR2 cells versus the ‘no target’ shRNA pLKO control cells. ^#P < 0.05, significantly different from NMDPEF-untreated cells.

Figure 5

QR2 inhibitor prevents PQ-induced systemic toxicity and mortality in Wistar rats. (A) The animals (six per group) were injected i.p. with PQ (30 mg·kg⁻¹) followed by i.p. injections of NMDPEF (0.5 mg·kg⁻¹ or 0.75 mg·kg⁻¹ each) 2 and 4 h later (Day 0). The drug was also administered twice on the Day 1 and twice on the Day 2 post-PQ treatment, 0.5 mg·kg⁻¹ or 0.75 mg·kg⁻¹ each time, to the final dose 3 and 4.5 mg·kg⁻¹ respectively. The animals were monitored daily for signs of toxicity for 28 days post-treatment. (B) The animals surviving after the PQ treatment and controls were killed 3.5 months later and lungs were analysed for macroscopic signs of fibrosis. Note, enlarged lungs and extended areas of fibrosis in animals that survived from PQ treatment and normal size and normal appearance of lungs isolated form animals treated with 4.5 mg·kg⁻¹ of NMDPEF and PQ.

Figure 6

NMDPEF prevents ECoG discharges, lipid peroxidation and death induced by the infusion of PQ into the SN of Wistar rats. (A) Representative traces showing high voltage epileptogenic discharges induced by the infusion of PQ (25 µg) or vehicle with or without pre-treatment (20 min) with NMDPEF or a MnSOD mimetic M40401 (10 µg). Traces are representative of ECoG effect observed 30 min after administration of the herbicide. (B) Mortality following the infusion of PQ (25 µg, n= 10) and NMDPEF or M40401 as in A (n= 10, each treatment). (C) The effect of NMDPEF and M40401 (10 µg) on the latency (min) of seizures induced by PQ (25 µg). Mean ± SEM, n= 10. (D) MDA accumulation in the brain tissue in rats receiving PQ (25 µg) with or without pre-treatment with NMDPEF or M40401 as in A and B. SN homogenates were subjected to derivatization and analysed by HPLC Data shown are means ± SEM, n= 10. *P < 0.05, significantly different from PQ alone.