The Non-Specific Drp1 Inhibitor Mdivi-1 Has Modest Biochemical Antioxidant Activity

Abstract

Mitochondrial division inhibitor-1 (mdivi-1), a non-specific inhibitor of Drp1-dependent mitochondrial fission, is neuroprotective in numerous preclinical disease models. These include rodent models of Alzheimer's disease and ischemic or traumatic brain injury. Among its Drp1-independent actions, the compound was found to suppress mitochondrial Complex I-dependent respiration but with less resultant mitochondrial reactive oxygen species (ROS) emission compared with the classical Complex I inhibitor rotenone. We employed two different methods of quantifying Trolox-equivalent antioxidant capacity (TEAC) to test the prediction that mdivi-1 can directly scavenge free radicals. Mdivi-1 exhibited moderate antioxidant activity in the 2,2'-azinobis (3-ethylbenzothiazoline 6-sulfonate) (ABTS) assay. Half-maximal ABTS radical depletion was observed at ~25 μM mdivi-1, equivalent to that achieved by ~12.5 μM Trolox. Mdivi-1 also showed antioxidant activity in the α, α-diphenyl-β-picrylhydrazyl (DPPH) assay. However, mdivi-1 exhibited a reduced capacity to deplete the DPPH radical, which has a more sterically hindered radical site compared with ABTS, with 25 μM mdivi-1 displaying only 0.8 μM Trolox equivalency. Both assays indicate that mdivi-1 possesses biochemical antioxidant activity but with modest potency relative to the vitamin E analog Trolox. Future studies are needed to evaluate whether the ability of mdivi-1 to directly scavenge free radicals contributes to its mechanisms of neuroprotection.

Keywords: Complex I; ROS; free radical; mitochondria; mitochondrial fission; neurons; oxidative stress; oxygen; scavenger; superoxide.

Figure 1

Mitochondrial division inhibitor-1 (mdivi-1) structure, absorbance, and the free radicals used to measure its antioxidant activity. (A) 2,2′-Azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical. (B) 2,2-Diphenyl-1-picrylhydrazyl (DPPH) radical. (C) Mdivi-1 structure. (D) Absorbance spectra of mdivi-1 (100 μM) and dimethyl sulfoxide (DMSO, 0.2%) in solution. O.D., optical density.

Figure 2

Mdivi-1 dose-dependently depletes ABTS^•+. (A) Mdivi-1 (2.5–100 μM) or Trolox (0.78–25 μM) was added to ABTS^•+ and antioxidant activity was monitored by the ABTS^•+ decolorization assay as a decrease in optical density (O.D.), expressed as a percentage of the initial O.D. Each value is depicted as the mean ± SEM; n = 3 separate experiments. (B) Trolox-equivalent antioxidant capacity (TEAC) of mdivi-1 for the depletion of ABTS^•+. Bar graphs represent mean ± SEM from n = 3 separate experiments. One-way ANOVA, F-value (7, 16) = 95.29, p < 0.0001, was followed by Bonferroni’s multiple comparisons post hoc test comparing mdivi-1 treatment to the vehicle control (Ctrl). * p < 0.05, *** p < 0.001, **** p < 0.0001.

Figure 3

Mdivi-1 weakly but dose-dependently depletes DPPH radical. (A) Mdivi-1 (2.5–100 μM) or Trolox (0.78–25 μM) was added to DPPH radical and antioxidant activity was monitored by the DPPH decolorization assay as a decrease in optical density (O.D.), expressed as a percentage of the initial O.D. Each value is depicted as the mean ± SEM from n = 3 separate experiments. The inset in (A) shows the means ± SEM for the mdivi-1 group on a compressed scale (DPPH Radical (% O.D.), y-axis vs. Compound Conc. (μM), x-axis), fit by linear regression analysis (R² = 0.9139, p < 0.001). (B) Trolox-equivalent antioxidant capacity (TEAC) of mdivi-1 for the depletion of DPPH. Bar graphs represent means ± SEM from n = 3 separate experiments. One-way ANOVA, F-value (7, 16) = 48.44, p < 0.0001, was followed by Bonferroni’s multiple comparisons post hoc test comparing mdivi-1 treatment to Ctrl. * p < 0.05, *** p < 0.001, **** p < 0.0001.

Figure 4

Correlation of mdivi-1’s dose-dependent antioxidant capacity values measured by two different radical depletion assays. The Trolox-equivalent antioxidant capacity (TEAC) of mdivi-1 in the ABTS^•+ (y-axis) and DPPH radical (x-axis) depletion assays was found to be significantly correlated by linear regression analysis (R² = 0.8961, p = 0.0004). Dotted line depicts 95% confidence bands for the best-fit line. Note that the axes have different scales.

Figure 5

Effect of mdivi-1 on Amplex UltraRed oxidation rate. (A) Fluorescence in arbitrary units (A.U.) of Amplex UltraRed (Amplex) oxidation product monitored over time following xanthine oxidase (XO) addition in the absence (Ctrl, control) or presence of mdivi-1 (25–100 μM), DMSO (0.2%, vehicle for mdivi-1), Trolox (25 μM), glutathione (1 mM) or catalase (15,000 U/mL). Glutathione was prepared from powder in the aqueous Amplex assay medium and Trolox was diluted from a 50 mM stock prepared in water. Traces are means of quadruplicate wells and are representative of the experiment run on two different days. Trolox was only included on one of the days (depicted). (B) Amplex UltraRed oxidation rates expressed as a percentage of the Ctrl rate. Data are mean ± SEM, n = 4 for Trolox, and n = 8 for all others. An uncorrected representative glutathione trace is shown in (A), and both the uncorrected and corrected (corr, i.e., background-subtracted) rates are given in (B). Only the corrected rates were used for statistical analysis. A two-way mixed-model ANOVA followed by Bonferroni’s multiple comparisons post hoc test was used to test for differences among treatments, with treatment and experiment day as factors. Both treatment (F-value (8, 58) = 97.25, p < 0.0001) and the experiment day (F-value (1, 58) = 10.50, p = 0.002) had a significant effect, accounting for 92.18% and 1.244% of the total variation, respectively. **** p < 0.0001 compared to Ctrl, # p < 0.05 compared to Trolox, @ p < 0.05 compared to glutathione corr, ^^^^ p < 0.0001 compared to all other groups.