Abstracts

Rationale: Harmful algal blooms that generate the neurotoxin domoic acid (DA) are increasing in frequency and severity in coastal waters of the western United States. Previous studies discovered that DA-producing algae become concentrated in filter-feeding fish, such as anchovies, which are eaten by birds and mammals. California sea lions (Zalophus californianus) frequently are affected by DA toxicosis after feeding on these fish. Exposed sea lions display acute neurological signs including status epilepticus, and many die. A substantial fraction of those that survive develop behavioral abnormalities and epilepsy. In epileptic animals, MRI reveals hippocampal atrophy, and hematoxylin and eosin staining shows neuron loss in olfactory and limbic areas, especially hippocampus. The present study tested whether DA-exposed sea lions develop temporal lobe epilepsy. Methods: Hippocampi were isolated immediately after euthanasia from a DA group that consisted of 7 animals in which DA had been detected in feces or urine and/or they were in status epilepticus when admitted during an algal bloom, and 6 others that displayed signs of epilepsy including spontaneous seizures, abnormal EEG, and/or small hippocampal volume. Spontaneous, recurrent seizures were observed in 9/13 of these animals. In most cases, exact time of exposure to DA was unknown, but the minimum period between exposure and euthanasia was 6-210 d (51 ± 14 d, mean ± sem). Control tissue was obtained from 11 sea lions that were euthanized because of pneumonia, cancer, or shark bite wounds that were unresponsive to treatment. After immersion-fixation hippocampi were sectioned transversely from the septal pole to the temporal pole. Adjacent series of sections were processed for Nissl stain, somatostatin-immunocytochemistry, and Timm stain. Hilar neuron numbers were estimated by the optical fractionator method. Results: Nissl staining revealed obvious hilar neuron loss in 11/13 DA sea lions with variable additional loss of other hippocampal neurons. In addition, 2 animals originally categorized above as controls displayed hippocampal sclerosis. Control sea lions without hippocampal sclerosis had 333,000 ± 14,000 hilar neurons/hippocampus. Of the 13 animals with hippocampal sclerosis, 3 displayed severe bilateral hilar neuron loss, but in 10 hilar neuron numbers were reduced only unilaterally. When unilateral hilar neuron loss occurred, it was extreme, with numbers down to 4-19% of the contralateral hippocampus. In contrast, in epileptic kainate-treated rats neuron loss is more symmetrical and graded, with hilar neuron numbers ranging from 2-88% of controls. Somatostatin-immunoreactivity showed loss of many hilar interneurons but paradoxical exuberance of axons in the dentate gyrus of DA sea lions. Timm staining demonstrated aberrant mossy fiber sprouting into the inner molecular layer of the dentate gyrus in 13/13 in the DA group and in only 1/11 controls. Conclusions: DA-exposed sea lions display neuropathological findings, including unilateral hippocampal sclerosis and mossy fiber sprouting, that are similar to those reported for patients with temporal lobe epilepsy.