Association of Immune Response and Neuronal Cell Death in the Rat Brain After Kainic Acid or Electrical Kindling Treatment

Abstract

Purpose: In this study, we tried to clarify the possible involvement of immune responses in epilepsy. This attempt was based on our previous reports on the occurrence of immune responses in the rat hippocampus undergoing kainic acid (KA)‐induced neuronal death and in the sclerotic hippocampus of patients with temporal lobe epilepsy. However, it remains unclear whether these immune responses elicited neuronal degeneration in the sclerotic hippocampus. We investigated the relation between neuronal death and microglial activation in rats either pretreated with KA or submitted to electrical kindling. Methods: KA treatment: Adult male Sprague‐Dawley (SD) rats received a single intraperitoneal injection of KA (12 mg/kg). Only rats showing generalized limbic seizures were used in this study. Rats injected with physiological saline were used as controls. At 6, 12, and 18 h; 1, 2, 3, and 7 days; and 2, 4, 6, 8, 10, 12, and 16 weeks after the treatment, the animals were killed. Electrical kindling: Adult male SD rats were stereotaxically implanted with a bipolar electrode into the left perforant pathway near the angular bundle. Two weeks after electrode implantation, rats were submitted to daily kindling. The unrestrained awake rats received kindling stimulation with a 2‐s train of 60‐Hz biphasic constant‐current 1.0‐ms square‐wave pulses twice a day (6 days per week). The evoked behavioral seizures were classified according to standard criteria. Twenty‐four hours after each behavioral seizure stage, the rats were hlled. For rapid kindling, the unrestrained awake animals were subjected to electrical kindling with hourly and repeated electrical stimulation until generalized seizures were elicited 5 times. Sham‐operated rats served as controls. At 1, 3, 6, 9, and 15 days after the final stimulation, the animals were killed. The activated microglial cells and T lymphocytes were detected immunohistochemically with the avidin‐biotinylated peroxidase complex method. The primary antibodies were directed to complement receptors CR3 (OX‐42). MHC 1 (OX‐18), MHC II (OX‐6). and T‐cell surface antigens. Neuronal death was evaluated by determining DNA fragmentation. For this purpose, coronal cryostat sections of the brain were processed according to the TUNEL technique (in situ end‐labeling method). Results: In control rats, numerous resting microglial cells were distributed throughout the brain. However, neither MHC I‐ nor MHC II‐positive staining was observed in the brain. In KA‐treated rats, reactive microglial cells with hypertrophic bodies and thick processes appeared in the CA1 and CA3 fields of the hippocampus, hilus of the dentate gyms, thalamus, amygdala, and piriform cortex as early as 6 h after injection. The number of these reactive microglial cells quickly increased and reached the maximum at 7 days after the KA injection. The expression of MHC I on reactive microglia, DNA fragmentation, and appearance of cytotoxic/suppressor T lymphocytes were noted at 18 h after injection. The helper/inducer T lymphocytes amd MHC II–positive microglial cells were detected in the same brain regions at 48 h after injection. The fragmentation of DNA and T lymphocytes was detected until 4 and 8 weeks, respectively. After 16 weeks, MHC I‐ or MHC II–positive reactive microglial cells or both were reduced in number. In the daily‐kindling model, only a small number of reactive microglia was observed. In contrast, DNA fragmentation, MHC I‐ or MHC II–positive cells, and T lymphocytes were not detected. In the rapid‐kindling model, there were no activated microglial cells, DNA fragmentation, MHC I‐ amd MHC II–positive cells, or T lymphocytes. Conclusions: In the kindling model of epilepsy, neither DNA fragmentation nor immune responses were detected. The result indicates that epileptic seizures do not depend on the immune responses. In the KA‐treated model of epilepsy, immune responses were closely related to DNA fragmentation, suggesting an association of immune responses with neuronal death. We therefore suggest that immune responses play an important role in the neuronal death process induces by KA.