Abstracts

Rationale: The purpose of our study was (1) to characterize the morphology of methamphetamine (METH)-induced neuronal death; (2) to determine the regional distribution and severity of METH-induced neuronal death; and (3) to determine the role played, if any, by METH-induced electrographic seizure discharges. Methods: Anesthetized C57BL/6 male mice (25 g when given METH) underwent skull-screw implantation for EEG recording one week prior to receiving 40 mg/kg of free-base METH s.c. or i.p. in METH-treated mice or saline in 6 controls. They had EEGs, behavior and rectal temperatures monitored for at least 4 h, or until their monitored parameters returned to baseline. Twenty-four h after METH injection they were euthanized and underwent transcardiac brain perfusion-fixation with 4% phosphate-buffered paraformaldehyde, after which brains were removed, embedded in paraffin, sectioned at 6-µm thickness, stained with hematoxylin and eosin and examined. The data conformed to a Poisson distribution rather than a normal curve, so an analysis of deviance, with post-hoc comparisons under the Poisson model, was performed. Results: Fifty-eight per cent (7 of 12) had repetitive electrographic seizure discharges (RESDs). The time from METH injection to the appearance of RESDs was 47.8 ± 15.2 min, the duration of RESDs was 8.53 ± 4.21 min, and the total time that RESDs were present was 321 ± 124 min (mean ± SEM, n = 7). Only 43% of mice with RESDs (3 of 7) had the clonic forelimb movements characteristic of seizures induced in rodents (stage 3 behavioral seizures in the Racine classification). Sixty per cent of mice without RESDs (3 of 5) had clonic forelimb movements; none of the 3 had evidence of neuronal damage. One of the 5 mice without RESDs had a few acidophilic neurons in dorsal CA3a. Five of 7 mice with RESDs had acidophilic neurons (the light-microscopic appearance of necrotic neurons by electron microscopy) in the hippocampal CA1-CA3 regions and hilus, amygdala, piriform cortex and entorhinal cortex, and the overall mean damage scores were significantly greater than those for both the METH group without RESDs (0.69 ± 0.06 vs. 0.01 ± 0.01, p < 0.001) and controls (0.69 ± 0.06 vs 0.00 ± 0.00, p < 0.001). The damage scores for each brain region were also significantly greater than those for both the METH group without RESDs and controls (p < 0.001 for each comparison). Maximum rectal temperatures did not differ significantly in mice with and without RESDs following METH administration. Spearman correlation analysis showed that Tmax was not correlated with either the maximum damage score or the number of brain regions damaged (p = 0.324 and 0.381 respectively). Conclusions: Our findings indicate that in mice EEG recording to document RESDs is mandatory to document seizure activity and that clonic forelimb movements should not be used as an indicator of seizure activity. Moreover, METH-treated mice with RESDs had acidophilic neurons in many of the same brain regions as in generalized seizures, and METH-induced neuronal necrosis only occurred in mice with RESDs.