Unveiling the shield: Troglitazone's impact on epilepsy-induced nerve injury through ferroptosis inhibition

Abstract

Background: Epilepsy is a widespread central nervous system disorder with an estimated 50 million people affected globally. It is characterized by a bimodal incidence peak among infants and the elderly and is influenced by a variety of risk factors, including a significant genetic component. Despite the use of anti-epileptic drugs (AEDs), drug-refractory epilepsy develops in about one-third of patients, highlighting the need for alternative therapeutic approaches.

Aims: The primary aim of this study was to evaluate the neuroprotective effects of troglitazone (TGZ) in epilepsy and to explore the potential mechanisms underlying its action.

Methods: We employed both in vitro and in vivo models to assess TGZ's effects. The in vitro model involved glutamate-induced toxicity in HT22 mouse hippocampal neurons, while the in vivo model used kainic acid (KA) to induce epilepsy in mice. A range of methods, including Hoechst/PI staining, CCK-8 assay, flow cytometry, RT-PCR analysis, Nissl staining, scanning electron microscopy, and RNA sequencing, were utilized to assess various parameters such as cellular damage, viability, lipid-ROS levels, mitochondrial membrane potential, mRNA expression, seizure grade, and mitochondrial morphology.

Results: Our results indicate that TGZ, at doses of 5 or 20 mg/kg/day, significantly reduces KA-induced seizures and neuronal damage in mice by inhibiting the process of ferroptosis. Furthermore, TGZ was found to prevent changes in mitochondrial morphology. In the glutamate-induced HT22 cell damage model, 2.5 μM TGZ effectively suppressed neuronal ferroptosis, as shown by a reduction in lipid-ROS accumulation, a decrease in mitochondrial membrane potential, and an increase in PTGS2 expression. The anti-ferroptotic effect of TGZ was confirmed in an erastin-induced HT22 cell damage model as well. Additionally, TGZ reversed the upregulation of Plaur expression in HT22 cells treated with glutamate or erastin. The downregulation of Plaur expression was found to alleviate seizures and reduce neuronal damage in the mouse hippocampus.

Conclusion: This study demonstrates that troglitazone has significant therapeutic potential in the treatment of epilepsy by reducing epileptic seizures and the associated brain damage through the inhibition of neuronal ferroptosis. The downregulation of Plaur expression plays a crucial role in TGZ's anti-ferroptotic effect, offering a promising avenue for the development of new epilepsy treatments.

Keywords: Plaur; brain damage; epilepsy; ferroptosis; neuroprotection; troglitazone.

FIGURE 1

Troglitazone improves KA‐induced seizures and neuronal damage. (A) Schematic diagram of the experimental design. (B‐D) Effects of TGZ on the seizure score, seizure duration, and seizure numbers (in 30 m) 24 h after KA injection in different groups. (E) The illustration of Nissl staining in several groups' CA1 and CA3 hippocampus subregions 24 h after KA injection. Arrows indicate Nissl‐positive cells. Red scale bar indicates 100 μM; black scale bar indicates 50 μM. (F, G) Statistical analysis of viable neurons in CA1 and CA3 regions 24 h after KA injection in different groups. *p < 0.05; **,p < 0.01; ***p < 0.001; ***p < 0.0001. IH, surgical ipsilateral hippocampus; KA, kainic acid; NS, normal saline; TGZ, troglitazone. Shapiro–Wilk test is applied for normality checks. T‐test is used for two groups or ANOVA for multiple groups if data are normal; otherwise, we used Wilcoxon test for non‐normal data.

FIGURE 2

Ferroptosis is involved in the neuroprotection of troglitazone in KA‐induced epilepsy mouse. (A) Transmission electron micrograph of mitochondria in mouse right hippocampal cells 24 h after KA injection in different groups. (B, C) Statistical results for the ratio and percentage of mitochondrial major to miner axis. (D) MRNA expression of ferroptosis‐related genes Acsl4, Gpx4, Ptgs2 (encoding protein COX2), and Slc7a11. (E, F) Expression of ferroptosis‐related proteins. *p < 0.05; **p < 0.01; ***p < 0.001; ***p < 0.0001. KA, kainic acid; TGZ, troglitazone. Shapiro–Wilk test is applied for normality checks. T‐test is used for two groups or ANOVA for multiple groups if data are normal; otherwise, we used Wilcoxon test for non‐normal data.

FIGURE 3

Troglitazone significantly inhibits neuronal ferroptosis in glutamate‐induced cell death in mouse hippocampal HT22 cells. (A) Representative images of Hoechst 33342/PI staining 8 h after treated with 5 mM glutamate or 2.5 μM troglitazone in HT22 cells. (B) Statistical analysis of Hoechst 33342/PI staining results in different groups. (C) Measurement of cell viability upon 5 mM glutamate with or without 2.5 μM troglitazone with CCK8. (D) Lipid ROS production assessed 8 h after treated with 5 mM glutamate or 2.5 μM troglitazone in HT22 cells by flow cytometry using C11‐BODIPY. (E) Statistical analysis of flow cytometry results in different groups. (F) Change in mitochondrial membrane potential in different groups detected by JC‐1 fluorescent probe and examined by flow cytometry. (G) Statistical analysis mitochondrial membrane potential results in different groups. (H) MRNA expression of ferroptosis‐related genes Acsl4, Gpx4, Ptgs2, and Slc7a11. *p < 0.05; **p < 0.01; ***p < 0.001; ***p < 0.0001. Glu, glutamate; TGZ, troglitazone. Shapiro–Wilk test is applied for normality checks. T‐test is used for two groups or ANOVA for multiple groups if data are normal; otherwise, we used Wilcoxon test for non‐normal data.

FIGURE 4

Troglitazone was further shown to inhibit ferroptosis in erastin‐induced neuronal ferroptosis model. (A) Representative images of Hoechst 33342/PI staining 8 h after treated with 0.5 μM erastin or 2.5 μM troglitazone in HT22 cells. (B) Statistical analysis of Hoechst 33342/PI staining results in different groups. (C) Measurement of cell viability upon 0.5 μM erastin with or without 2.5 μM troglitazone with CCK8. (D) Lipid ROS production assessed 8 h after treated with 0.5 μM erastin or 2.5 μM troglitazone in HT22 cells by flow cytometry using C11‐BODIPY. (E) Statistical analysis of lipid ROS results in different groups. (F) Change in mitochondrial membrane potential in different groups detected by JC‐1 fluorescent probe and examined by flow cytometry. (G) Statistical analysis mitochondrial membrane potential results in different groups. (H) MRNA expression of ferroptosis‐related genes Acsl4, Gpx4, Ptgs2, and Slc7a11. (I) Laser confocal microscopy images is used to detect the content of ferrous ions. *p < 0.05; **p < 0.01; ***p < 0.001; ***p < 0.0001. Glu, glutamate; TGZ, troglitazone. Shapiro–Wilk test is applied for normality checks. T‐test is used for two groups or ANOVA for multiple groups if data are normal; otherwise, we used Wilcoxon test for non‐normal data.

FIGURE 5

Plaur as an important target of troglitazone to inhibit neuronal ferroptosis. (A) MRNA expression of previously reported ferroptosis‐related genes. (B‐C) Veen diagram shows the number of ferroptosis activation or inhibition DEGs between different groups. (D) Gene ontology (GO) annotations of DEGs. (E) KEGG pathway enrichment for DEGs analyzed by clusterProfiler package. (F) Hub gene regulatory network mapped by String website. Glu, glutamate; Era, erastin; TGZ, troglitazone.

FIGURE 6

Inhibition of Plaur suppresses seizures and neuronal damage in KA‐induced epilepsy mouse. (A) Schematic diagram of the experimental design. (B) Representative brain showing AVV injection effect and resulting GFP distribution. Blue, DAPI; green, GFP. (C) MRNA expression levels of Plaur in the surgical ipsilateral hippocampus of mice 28 days after AAV‐Plaur injection (relative to the left hippocampus). (D‐F) Effects of AAV‐Plaur on the seizure score, seizure duration, and seizure numbers (in 30 m) 24 h after KA injection in different groups. (G) The illustration of Nissl staining in several groups' CA1 and CA3 hippocampus subregions 24 h after KA injection. Arrows indicate Nissl‐positive cells. Red scale bar indicates 100 μM; black scale bar indicates 50 μM. (H) Statistical analysis of viable neurons in CA1 and CA3 regions 24 h after KA injection in different groups. (I) MRNA expression of ferroptosis‐related genes Acsl4, Gpx4, Ptgs2, and Slc7a11. *p < 0.05; **p < 0.01; ***p < 0.001; ***p < 0.0001. AAV‐P, AAV‐Plaur; CH, surgical contralateral hippocampus; IH, surgical ipsilateral hippocampus; KA, kainic acid; NS, normal saline. Shapiro–Wilk test is applied for normality checks. T‐test is used for two groups or ANOVA for multiple groups if data are normal; otherwise, we used Wilcoxon test for non‐normal data.

FIGURE 7

Working model summarizes the protective effect of troglitazone on KA‐induced seizures and neuronal damage. Era, erastin; Glu, glutamate; KA, kainic acid. Drawn by Figdraw (www.figdraw.com)

FIGURE 8

In the process of neuronal resistance to ferroptosis, Plaur may influence upstream regulators of ferroptosis such as HO‐1 and SLC7A11 through the RAS/MAPK pathway, thereby affecting ferroptosis. uPA, Plau; uPAR, Plaur.