Huperzine A (HUP) is a potent reversible inhibitor of acetylcholinesterase (AChE) that crosses the blood-brain barrier. Its ability to prevent seizures and subsequent hippocampal neuropathological changes induced by the organophosphate soman was studied in guinea pigs. Results were compared to guinea pigs treated with pyridostigmine (PYR, 0.2 mg/kg, subcutaneously). HUP pretreatment at 0.5 mg/kg, intraperitoneally, totally prevented seizures and ensured the survival of all animals for 24 h after intoxication. Hippocampal tissue was then free of any neuronal damage. Comparatively, all animals pretreated with PYR exhibited epileptic activity after soman poisoning and five of six animals died. Examination of the hippocampus of the only surviving guinea pig pretreated with PYR showed extensive neuropathological changes. Although HUP or PYR induced similar inhibitions of blood AChE activity, only HUP pretreatment led to a decrease in central AChE activity. In binding studies on guinea-pig brain homogenates, HUP had no affinity for muscarinic, alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) and gamma-aminobutyric acid (GABA)A receptors and only a very low one for N-methyl-D-aspartate (NMDA) receptors. In conclusion, HUP, unlike PYR, protects against soman-induced convulsions and neuropathological changes in the hippocampus. This efficacy seems to be related to a protection by HUP of both peripheral and central stores of AChE.

After intracerebral injection, some toxic secreted phospholipases A2 (sPLA2) can induce epileptic seizures which bases are currently ill known. We undertook the detailed study of the central neurotoxicity of paradoxin (PDX), an analog of taipoxin, in rodents. Since literature strongly suggests a high variability in the sPLA2 epileptogenic properties, we compared, in an acute model, PDX with crotoxin (CTX), known to induce seizures and that may bind to similar neuronal receptors. Related toxic enzymes (ammodytoxin A, ATX A, and CTX subunit CB) and the non neurotoxic sPLA2 from pancreas and PLA2 analog ammodytin L (AML) were also tested. Despite being highly neurotoxic, PDX did not induce either convulsions or long-lasting seizure fits. The results obtained with the other enzymes showed that toxic sPLA2s can effectively be differentiated based on two criteria: the presence of cortically recorded epileptic paroxysmal discharges (E) and convulsions (C). We thus propose to classify the toxic sPLA2s into different groups depending on their epileptogenic properties: E-C-(PDX), E+C+ (CTX, CB), and E-C+ (ATX A). The non toxic AML and pancreatic enzyme were E-C-. Moreover, the results obtained with AML, and preliminarily with chemically inhibited CB, suggested that phospholipid hydrolysis is important to trigger seizures and convulsions. However, PDX and CTX that possess highly different epileptogenic properties exerted comparable, although slightly different, catalytic activities. Similarly, histological evaluations of the brain of PDX and CTX-treated rats (H&E staining, GFAP immunodetection, hsp70 and c-fos mRNA detection) did not provide satisfactory clues to explain these large differences. Further studies are strongly required.