Targeting ferroptosis: A novel therapeutic strategy for the treatment of mitochondrial disease-related epilepsy

Abstract

Background: Mitochondrial disease is a family of genetic disorders characterized by defects in the generation and regulation of energy. Epilepsy is a common symptom of mitochondrial disease, and in the vast majority of cases, refractory to commonly used antiepileptic drugs. Ferroptosis is a recently-described form of iron- and lipid-dependent regulated cell death associated with glutathione depletion and production of lipid peroxides by lipoxygenase enzymes. Activation of the ferroptosis pathway has been implicated in a growing number of disorders, including epilepsy. Given that ferroptosis is regulated by balancing the activities of glutathione peroxidase-4 (GPX4) and 15-lipoxygenase (15-LO), targeting these enzymes may provide a rational therapeutic strategy to modulate seizure. The clinical-stage therapeutic vatiquinone (EPI-743, α-tocotrienol quinone) was reported to reduce seizure frequency and associated morbidity in children with the mitochondrial disorder pontocerebellar hypoplasia type 6. We sought to elucidate the molecular mechanism of EPI-743 and explore the potential of targeting 15-LO to treat additional mitochondrial disease-associated epilepsies.

Methods: Primary fibroblasts and B-lymphocytes derived from patients with mitochondrial disease-associated epilepsy were cultured under standardized conditions. Ferroptosis was induced by treatment with the irreversible GPX4 inhibitor RSL3 or a combination of pharmacological glutathione depletion and excess iron. EPI-743 was co-administered and endpoints, including cell viability and 15-LO-dependent lipid oxidation, were measured.

Results: EPI-743 potently prevented ferroptosis in patient cells representing five distinct pediatric disease syndromes with associated epilepsy. Cytoprotection was preceded by a dose-dependent decrease in general lipid oxidation and the specific 15-LO product 15-hydroxyeicosatetraenoic acid (15-HETE).

Conclusions: These findings support the continued clinical evaluation of EPI-743 as a therapeutic agent for PCH6 and other mitochondrial diseases with associated epilepsy.

Fig 1. EPI-743 rescues PCH6 patient fibroblasts from RSL3-induced lipid oxidation and ferroptosis.

(A) Dose-dependent reduction in viability of PCH6 patient cells by RSL3 treatment (N = 6–8 replicates for each patient cell summarized). Cell viability was assessed 18 h after RSL3 treatment using CellTiter-Glo 2.0 reagent to quantify cellular ATP. (B) The potency of RSL3-induced BODIPY 581/591 C11 lipid oxidation mirrors that of cytotoxicity (N = 6 replicates per patient culture). (C) Dose- and time-dependent BODIPY 581/591 C11 lipid oxidation in PCH6 patient fibroblasts, as measured by time-lapse microscopy (N = 2 independent experiments per patient culture, representative data from Subject 1 shown). (D) EPI-743 dose-dependently protected PCH6 patient fibroblasts from ferroptosis induced by RSL3 (2 μM, 18 h). Dose-response curves are representative of 5–9 independent experiments summarized in Table 2. (E) EPI-743 dose-dependently decreased the rate of cellular BODIPY 581/591 C11 oxidation following RSL3 (2 μM) treatment (N = 6 replicates per patient culture). (F) Time course of EPI-743 prevention of BODIPY 581/591 C11 lipid oxidation (N = 2 independent experiments per patient culture, representative data from Subject 1 shown). Mean ± SEM displayed along with best 4-parameter curve fits (A, B, D, E).

Fig 2. EPI-743 rescues PCH6 patient fibroblasts from iron / BSO-induced ferroptosis.

(A) PCH6 patient fibroblasts were sensitive to ferroptosis induced by BSO-mediated GSH depletion in the presence of varying concentrations of added iron(III) citrate (FeC). Viability relative to untreated control wells was measured by Calcein AM assay 48 h after BSO treatment. (B) EPI-743 treatment dose-dependently prevented FeC/BSO-induced ferroptosis in PCH6 patient fibroblasts. Mean ± SEM (N = 3) along with best-fit 4-parameter curve fits displayed; representative of two independent experiments. Refer to Table 2 for summary statistics of calculated rescue potencies.

Fig 3. EPI-743 hydroquinone is detected in PCH6 patient fibroblasts and inhibits human 15-LO.

(A) Dosed EPI-743 (1 μM, 3 h) was reduced to EPI-743 hydroquinone (EPI-743-HQ) in PCH6 patient fibroblasts from Subjects 1, 2, and 3. Mean ± SD displayed, N = 6 per subject. (B) The activity of purified human 15-LO was monitored via a ferrous oxidation-xylenol orange (FOX) method. EPI-743-HQ dose-dependently inhibited human 15-LO enzyme activity, whereas EPI-743 showed only 20% inhibition at 100 μM. Mean ± SD displayed, N = 2. The IC₅₀ for the displayed curve is 4.4 μM. (C, D) LC-MS/MS measurement of downstream products of 15-LO (15-HETE, C) and 12-LO (12-HETE, D) in PCH6 patient (Subject 1) fibroblasts following RSL3 (2 μM, 4 h) challenge, without or with EPI-743 (200 nM) co-treatment. Compound was dosed in the quinone form. Results shown are N = 17 replicates from three independent experiments. Mean ± SD displayed. ***, p<0.001; ns, not significant; by Kruskal-Wallis test followed by Dunn’s multiple comparisons test. We noted that cellular 15-hydroperoxyeicosatetraenoic acid (15-HpETE), the direct product of 15-LO acting upon arachidonic acid as substrate, was not consistently detected within the quantifiable range by LC-MS/MS in our studies.

Fig 4. ALOX15 siRNA-mediated gene knockdown partially abrogates ferroptosis in PCH6 patient fibroblasts.

(A-C) siRNA knockdown of ALOX15 (siALOX15) decreased the sensitivity of PCH6 patient cells to RSL3-induced cytotoxicity, as assessed by ATP levels at 18h. Summarized results in each PCH6 patient culture from three independent knockdown experiments. Means (circles) and best-fit curves (solid lines) from 4-parameter fits with associated 95% confidence bands (dotted lines) displayed. (D-I) siRNA-mediated knockdown of ALOX15 partially prevented BODIPY 581/591 C11 lipid oxidation induced by RSL3 (60 nM, panels D-F, or 167 nM, panels G-I) in fibroblasts from Subject 1 (D, G), Subject 2 (E, H), or Subject 3 (F, I). Cellular lipid oxidation was measured by changes in green fluorescence by time-lapse video microscopy. Mean ± SEM displayed, N = 6–8 per group. Representative results from two independent studies. Statistical comparison of Area Under the Curve values was performed by unpaired t-test relative to siControl; see S1 Table for summary of results. ALOX15-targeted siRNA transfection resulted in >90% reduction in the target mRNA, as assessed by quantitative RT-PCR.

Fig 5. EPI-743 prevents ferroptosis and lipid oxidation in Leigh syndrome patient-derived fibroblasts.

(A) Dose-dependent protection by EPI-743 from RSL3-induced cytotoxicity of Leigh syndrome patient-derived fibroblasts (Subject 4) as assessed by CellTiter-Glo 2.0. (B) Dose-dependent protection by EPI-743 from RSL3-induced lipid oxidization as assessed by BODIPY 581/591 C11 cellular fluorescence. (C) Dose-dependent protection by EPI-743 from FeC/BSO-induced cytotoxicity as assessed by Calcein AM fluorescence. Mean ± SD displayed, N = 6.