The Involvement of Apoptotic Neural Damages in a Patient with Intractable Temporal Lobe Epilepsy and in a Kindling Model

Abstract

Purpose: Human temporal lobe cpilepsy is frequently related to loss of hippocampal neurons, called hippocampal sclerosis. It is still controversial whether neuronal cell death is directly involved in the pathogenesis or epilepsy or whether it is a secondary phenomenon following recurrent seizures. The present study aims to clarify the contribution of neuronal apoptotic changes to the basic pathophysiology of epilepsy. We investigated apoptotic neural damage in the hippocampi of both a patient with intractable temporal lobe epilepsy and an amygdaloidkindling epilepsy model. Methods: The patient was a 30‐year old male who had a 15‐year history of drug‐resistant complex partial seizures. His seizure frequency increased to 5–10 seizures per week. Scalp EEG revealed right temporal‐dominant spikes, and spike and slow wave complexes. Frequent epileptic spike discharges from the right mid‐hippocampal region were detected by eicctrocorticography (ECoG) before temporal lobectomy. Samples of right hippocampal tissue were obtained during surgery. We cxamined apoptotic brain damage by i n situ terminal deoxynuclcotidyltraiisferase‐mediated dUTP nick‐end labeling of fragmented DNA (TUNEL) in the. hippocampal formation removed at surgery. We also examined the involvement of apoptotic brain damage in the kindling model of epilepsy. Sprague‐Dawley rats were kindled by a daily electrical stimulation in the left hasolateral amygdala with a SOHz biphasic square wave at an intensity just above the afterdischarge threshold for 2 seconds. The analysis was performed in the resected right hippocampus of the patient and in partially‐kindled (PK) and lully‐kindled (FK; 20 consecutive generalized convulsions) rats, 24 hours after the last seizure. After fixation i n 10% neutral‐buffered formalin, brain tissue was embedded i n paraffin. Coronal sections werc ording to ordinary TUNEL method. Immunoreactivity of Bax and the expression level of bax mRNA by in situ hybridination were alsn examined. Results: Apoptotic brain damage was observed in the spike‐detected and non spike‐detected areas of the patient's hippocampus. More DNA‐fragmented nuclei were seen i n non spike‐detected areas than in the spike‐detected areas. Only a few DNA‐fragmented nuclei were detected in the spike‐detected area. Compared with the spike‐detected area, an increased expression of Bax protein and bax mRNA levels was observed in non spike‐detected areas. In the kindling model, an increased number of TUNEL positive cells were secn in the CA3 and CA4 from the PK group rats. Only a few DNA‐fragmented nuclei were seen in the hippocampus from the FK group rats. The relative increase in expression levels of Bax protein and bax mRNA was observed in the CA3 and CA4 in both kindled groups. Conclusions: I1 we hypothesize that results from both human temporal lobe epilepsy and kindling models arise from a common basis in the pathophysiology of epilepsy, apoptotic cell death in hippocampus is induced not only by recurrent seizures, but also by factors rclated to the aquiaition process of epileptogenesis. The present results might be associated with the hippocainpal synaptic reorganization. However, clarification of this relationship requires further study. Our detection of changes in apoptosis‐related genes such as Bax and neuronal apoptotic cell death suggests their contribution to the pathogenesis of epilepsy.