Kainic acid (KA) was injected systemically, intracerebroventricularly (i.c.v.) and focally in the amygdala and other deep brain structures in the rat. EEG and behavioral changes were studied in relation to the neuropathology which developed subsequently.

Following intra-amygdaloid KA injection, diazepam blocked the epileptic events induced by the toxin, and abolished the neuronal loss usually seen in the lateral septum, claustrum, and contralateral cortex and hippocampus. The lesions in medial thalamic structures and ipsilateral hippocampus were also reduced by diazepam. Prior transection of the perforant path ipsilateral to the KA injection also decreased the severity of the electrographic and motor effects of the toxin and similarly reduced the extent of distant (‘remote’) pathological brain damage. Neither diazepam nor perforant path transection reduced the damage at the site of KA injection.

Kainic acid (0.4–2 μg) injected into the bed nucleus of the stria terminalis (BST) or the medial septum produced seizures with a longer latency and little brain damage outside the injection site. In contrast, intrastriatal KA injections were followed by ipsilateral hippocampal lesions.

i.c.v. Injection of KA (0.4–1.6 μg) produced a complex syndrome which included bilateral exophthalmos, mydriasis, foaming, tremor of the vibrissae, and paw and body tremor. The pattern of brain damage resembled that seen following intra-amygdaloid administration of the toxin. In addition, however, there was a bilateral necrosis of the pyriform and prepyriform cortices up to the rhinal fissure. Systemic administration of diazepam (i.p.) reduced the extent of the damage and in particular completely prevented the cortical damage.

Systemic administration of KA (9–15 mg/kg i.p.) readily produced motor and EEG seizures similar to those seen after intra-amygdaloid injection of the toxin. The pattern of brain damage was however more symmetrical than that which followed focal i.c.v. injection of the toxin and included necrosis of the pyriform cortex.

It is concluded that spread of seizure activity from the injection site plays a crucial role in the induction of ‘remote’ brain damage after focal intracerebral injections.