Sodium valproate protects against neuronal ferroptosis in epilepsy via suppressing lysyl oxidase

Background and purpose: Epilepsy is a chronic neurological disease that is characterized by repetitive seizures. Seizures-related complications such as cognitive deficits, anxiety and sleep disorders seriously impact the life quality of patients. Antiepileptic drugs are widely used for the treatment of epilepsy. Sodium valproate is served as the first-line antiepileptic drugs and possesses various pharmacological effects on the brain. Sodium valproate exerts neuroprotective effects in acute nervous system diseases such as ischemic brain damage by inhibiting oxidative stress. However, the mechanism of neuroprotection of sodium valproate in epilepsy is unclear. Lysyl oxidase (Lox) is a monoamine oxidase that acts on extracellular matrix collagen and elastin and it can promote accumulation of oxidative stress. Our previous studies have confirmed that Lox is involved in ferroptosis, a novel iron-dependent and lipid peroxidation-mediated cell death pathway, during epilepsy. In this study, we would like to investigate whether sodium valproate can exert neuroprotective effects on kainic acid-induced epileptic seizures by inhibiting Lox-mediated ferroptosis. Methods: Epileptic mouse models were established by intracranial injection of 250 ng/μl kainic acid on right hippocampus. Sodium valproate and ferroptosis inhibitors were administrated by intraperitoneal injecting. The epileptic behavior of the mice within 4 hours was recorded after intracranial injection of kainic acid. Mouse hippocampus was acquired to analyze the mRNA expression of prostaglandin-endoperoxide synthase 2 (PTGS2) and the production of 4-hydroxynonenal (4-HNE). In vitro, the protective effects of sodium valproate on glutamate-induced HT22 cell damage model was assessed by PI/Hoechst staining; The levels of PTGS2, 4-HNE and lipid ROS were analyzed by RT-qPCR, western blot and flow cytometry, respectively. RT-qPCR and Western blot analysis the mRNA and protein expression of Lox in the glutamate-induced HT22 cell damage model. The Lox overexpression model was established by intracranial injection of AAV on right hippocampus. Results: Pretreatment with sodium valproate and ferroptosis inhibitors could significantly alleviate the epileptic seizures in the kainic acid induced epilepsy mouse model. Western blot and RT-qPCR results showed that sodium valproate and ferroptosis inhibitors significantly inhibited the levels of 4-HNE and PTGS2. PI/Hoechst staining showed that 1 mM sodium valproate exerted protective effect on glutamate-induced HT22 cell injury model. There was no significant difference observed between sodium valproate and ferroptosis inhibitors co-intervention group and sodium valproate intervention group on glutamate-induced cell injury model. And sodium valproate could significantly inhibit the production of lipid reactive oxygen species and 4-HNE. The expression of Lox was significantly increased in the glutamate-induced HT22 cell injury model, which could be reversed by pretreatment of sodium valproate. And β-aminopropionitrile (a specific inhibitor of Lox) could inhibit ferroptosis induced by glutamate, as well as ameliorate the epileptic seizures in the kainic acid induced epilepsy mouse model. Pretreatment with sodium valproate could not ameliorate the epileptic behavior in the Lox-overexpression mice. Western blot analysis showed that sodium valproate could not suppress the production of 4-HNE in kainic acid induced epileptic mice model. Conclusions: The neuroprotective effect of sodium valproate in epileptic seizures is closely related to the inhibition of ferroptosis. The inhibition of ferroptosis is involved in the neuroprotective effect of sodium valproate on glutamate-induced HT22 cell damage model. Sodium valproate may exert neuroprotective effects in kainic acid-induced epileptic seizures by abrogating Lox-mediated ferroptosis.