Abstract Seizures are induced when subjects are exposed to severe hypoxia. It is followed by ventilatory fall‐off and eventual respiratory arrest, which may underlie the pathophysiology of death in patients with epilepsy and severe respiratory disorders. However, the mechanisms of hypoxia‐induced seizures have not been fully understood. Because astrocytes are involved in various neurological disorders, we aimed to investigate whether astrocytes are operational in seizure generation and respiratory arrest in a severe hypoxic condition. We examined the effects of astrocytic activation blockade on responses of EEG and ventilation to severe hypoxia. Adult mice were divided into two groups; in one group ( n = 24) only vehicle was injected, and in the other group ( n = 24) arundic acid, an inhibitory modulator of astrocytic activation, was administered before initiation of recording. After recording EEG and ventilation by whole body plethysmography in room air, the gas in the recording chamber was switched to 5% oxygen (nitrogen balanced) until a seizure and ventilatory depression occurred, followed by prompt switch back to room air. Severe hypoxia initially increased ventilation, followed by a seizure and ventilatory suppression in all mice examined. Fourteen mice without arundic acid showed respiratory arrest during loading of hypoxia. However, 22 mice pretreated with arundic acid did not suffer from respiratory arrest. Time from the onset of hypoxia to the occurrence of seizures was significantly longer in the group with arundic acid than that in the group without arundic acid. We suggest that blockade of astrocytic activation delays the occurrence of seizures and prevents respiratory arrest.
Amyotrophic lateral sclerosis is a fatal neurodegenerative disease characterized by progressive motoneuron loss. Redistribution of transactive response deoxyribonucleic acid-binding protein 43 from the nucleus to the cytoplasm and the presence of cystatin C-positive Bunina bodies are considered pathological hallmarks of amyotrophic lateral sclerosis, but their significance has not been fully elucidated. Since all reported rodent transgenic models using wild-type transactive response deoxyribonucleic acid-binding protein 43 failed to recapitulate these features, we expected a species difference and aimed to make a non-human primate model of amyotrophic lateral sclerosis. We overexpressed wild-type human transactive response deoxyribonucleic acid-binding protein 43 in spinal cords of cynomolgus monkeys and rats by injecting adeno-associated virus vector into the cervical cord, and examined the phenotype using behavioural, electrophysiological, neuropathological and biochemical analyses. These monkeys developed progressive motor weakness and muscle atrophy with fasciculation in distal hand muscles first. They also showed regional cytoplasmic transactive response deoxyribonucleic acid-binding protein 43 mislocalization with loss of nuclear transactive response deoxyribonucleic acid-binding protein 43 staining in the lateral nuclear group of spinal cord innervating distal hand muscles and cystatin C-positive cytoplasmic aggregates, reminiscent of the spinal cord pathology of patients with amyotrophic lateral sclerosis. Transactive response deoxyribonucleic acid-binding protein 43 mislocalization was an early or presymptomatic event and was later associated with neuron loss. These findings suggest that the transactive response deoxyribonucleic acid-binding protein 43 mislocalization leads to α-motoneuron degeneration. Furthermore, truncation of transactive response deoxyribonucleic acid-binding protein 43 was not a prerequisite for motoneuronal degeneration, and phosphorylation of transactive response deoxyribonucleic acid-binding protein 43 occurred after degeneration had begun. In contrast, similarly prepared rat models expressed transactive response deoxyribonucleic acid-binding protein 43 only in the nucleus of motoneurons. There is thus a species difference in transactive response deoxyribonucleic acid-binding protein 43 pathology, and our monkey model recapitulates amyotrophic lateral sclerosis pathology to a greater extent than rodent models, providing a valuable tool for studying the pathogenesis of sporadic amyotrophic lateral sclerosis.
