Our study was aimed at analyzing the basis for the apparent lack of perinatal sensitivity to the serotonergic neurotoxin 3,4-methylenedioxymethamphetamine (MDMA, "ecstasy"). MDMA (20 mg/kg s.c.) repeatedly administered to rat dams during gestation, did not affect [3H]paroxetine-labeled serotonin (5-HT) transporter density and 5-HT content in the offspring. A single dose of MDMA was then given to pups, not exposed prenatally to MDMA, at different postnatal ages (PND14, 21, 28 and 35). Long-term significant reductions in 5-HT levels in all the brain regions examined were only found at PND35. In a different set of experiments, MDMA administered at PND21 alone or in combination with (R)-1-(2,5-dimethoxy-4-iodophenyl)2-aminopropane (R-DOI, 0.5 mg/kg s.c.), or L-3,4-dihydroxyphenylalanine (L-DOPA, 80 mg/kg s.c.), caused a significant hyperthermia in the pups. However, only L-DOPA followed by MDMA caused a lasting reduction of 5-HT levels and 5-HT transporter density in the hippocampus and in the frontal cortex. In adult animals, no change in 5-HT levels and 5-HT transporter density in different brain regions was either found when MDMA was given to rats previously lesioned with 6-hydroxydopamine, but a significant reduction was again found in the lesioned animals receiving MDMA in combination with L-DOPA. These results appear to indicate that the hyperthermia induced by MDMA is not sufficient to produce lasting neurotoxic effects on the serotonergic system, at least at PND21, and support an important role for dopamine in the mechanism of neurotoxicity of MDMA, suggesting that an already developed dopaminergic system is necessary for the expression of the serotonergic deficits.