Abstracts

Rationale: Temporal lobe epilepsy (TLE), the most common form of epilepsy in adolescents and adults, is highly associated with drug resistance. In addition, TLE patients are at a higher risk of developing cognitive deficits and even dementia. Previous studies, including our own, have demonstrated the presence of amyloid-ß₄₂ (Aß₄₂) plaques and hyperphosphorylated tau (pTau) pathology in TLE, as seen in Alzheimer’s disease (AD), the most frequent cause of dementia. These findings suggest that AD-like pathology in TLE may contribute to cognitive deficits. However, the mechanisms responsible for these changes are unknown. We hypothesized that, similar to AD, endoplasmic reticulum (ER) stress and dysregulation of cell stress pathways may contribute to degeneration in TLE. The protein kinase R-like ER kinase (PERK) is one major kinase involved in the unfolded protein response due to ER stress and may lead to cell death and degeneration via phosphorylation of the translational factor eIF2a. The glycogen synthase kinase 3ß (GSK-3ß), the cyclin-dependant kinase 5 (CDK5) and the stress-activated kinase c-Jun N-terminal kinase (JNK) may be activated following ER stress and play a key role in the development of Aß₄₂ and tau pathology. Methods: Temporal lobe cortex and hippocampal tissue surgically resected from drug-resistant TLE patients (mean=35.8 years, n=13) and postmortem samples from age-matched control subjects (mean=37.6 years, n=10) were subjected to Western blot analysis. TLE samples were obtained from the Hospital of the University of Pennsylvania (Philadelphia, PA, USA). Control samples were obtained from the NIH NeurobioBank. The project was approved by the Institutional Review Board local ethic committee. The total expression and phosphorylation levels of ER stress marker PERK and its downstream target eIF2a, along with the stress kinases JNK, GSK-3ß, and CDK5 were quantified and compared between groups. PERK, JNK, and GSK-3ß are activated by phosphorylation, while CDK5 activity is regulated through cleavage of its activator p35 into the more active form p25. In contrast, the activity of eIF2a is downregulated when phosphorylated. Results: We observed an increase of phospho-PERK [Ser713]/PERK ratio, in TLE hippocampus (451% of controls, p<0.01) and cortex (1191% of controls, p<0.05). In parallel, we found a significant increase of phospho-eIF2a [Ser51]/eIF2a ratio in TLE hippocampus (177% of controls, p<0.05) and cortex (151% of controls, p<0.05). Phospho-JNK [Thr183/Tyr185]/JNK ratio was increased in TLE hippocampus (1866% of controls, p<0.0001), and cortex (866% of controls, p<0.0001). In contrast we observed a significant decrease of phospho-GSK-3ß [Tyr216]/GSK-3ß ratio in TLE hippocampus (22% of controls, p<0.001) and cortex (26% of controls, p<0.0001). The ratio p25/p35 was significantly decreased in TLE hippocampus (6% of controls, p<0.01) and cortex (24% of controls, p<0.0001). Conclusions: Our data demonstrate an upregulation of the stress kinase JNK following ER stress which may lead to cell death and degeneration in TLE via PERK/eIF2a pathway. As AD-like pathology may contribute to neuronal excitability and cognitive impairment in TLE, JNK and PERK pathways may represent relevant therapeutic targets. Funding: The National Institutes of Health NIH R01NS031718 (FEJ), 1R21NS105437-01A1 (FEJ), and R01NS101156 (DMT)