Mapping of cerebral metabolic activation in three models of cholinergic convulsions.

Glucose utilization of four cerebral cortex and 35 subcortical regions (CGU) was analyzed in three models of cholinergic seizures induced by the following compounds: 1) soman (pinacolylmethylphosphonofluoridate) an organophosphorus cholinesterase inhibitor, 100 microg/kg SC after pretreatment with pyridostigmine 26 microg/kg IM (n = 6); 2) physostigmine, a carbamate cholinesterase inhibitor, 1.31 mg/kg infused IV over 75 min (n = 6); and 3) pilocarpine, a direct cholinergic agonist, 30 mg/kg SC (n = 6). Physostigmine and pilocarpine were preceded by 3 mmol/kg LiCl IP 20 hrs earlier. Animals injected with saline SC (n = 6) were used as controls. Step-wise discriminant analysis successfully classified 100% of the cases into the four experimental groups with data from only six regions. Pyridostigmine-soman induced the most widespread and greatest increases in CGU. More restricted and lower levels of activation were observed with Li-pilocarpine while Li-physostigmine induced significant increases in CGU only in globus pallidus, entopeduncular nucleus, and substantia nigra. These three regions, which are functionally related, were also activated in the other two models of cholinergic convulsions and may represent the initial step in cholinergic activation of the CNS. Li-pilocarpine failed to activate most of the brainstem and the superior colliculus. All cortical regions were activated by Li-pilocarpine and pyridostigmine-soman, while they were inhibited by Li-physostigmine. This phenomenon may be due in part to the lack of activation with physostigmine of the basal forebrain nuclei (lateral septum, medial septum, vertical and horizontal limbs of the diagonal band, and substantia innominata) resulting in a decreased drive of cortical metabolism.