We studied in vivo expression of the serotonin transporter (SERT) protein after 3,4-methylenedioxymethamphetamine (MDMA), p-chloroamphetamine (PCA), or fenfluramine (FEN) treatments, and compared the effects of substituted amphetamines to those of 5,7-dihydroxytryptamine (5,7-DHT), an established serotonin (5-HT) neurotoxin. All drug treatments produced lasting reductions in 5-HT, 5-HIAA, and [(3)H]paroxetine binding, but no significant change in the density of a 70 kDa band initially thought to correspond to the SERT protein. Additional Western blot studies, however, showed that the 70 kDa band did not correspond to the SERT protein, and that a diffuse band at 63-68 kDa, one that had the anticipated regional brain distribution of SERT protein (midbrain>striatum>neocortex>cerebellum), was reduced after 5,7-DHT and was absent in SERT-null animals, was decreased after MDMA, PCA, or FEN treatments. In situ immunocytochemical (ICC) studies with the same two SERT antisera used in Western blot studies showed loss of SERT-immunoreactive (IR) axons after 5,7-DHT and MDMA treatments. In the same animals, tryptophan hydroxylase (TPH)-IR axon density was comparably reduced, indicating that serotonergic deficits after substituted amphetamines differ from those in SERT-null animals, which have normal TPH levels but, in the absence of SERT, develop apparent neuroadaptive changes in 5-HT metabolism. Together, these results suggest that lasting serotonergic deficits after MDMA and related drugs are unlikely to represent neuroadaptive metabolic responses to changes in SERT trafficking, and favor the view that substituted amphetamines have the potential to produce a distal axotomy of brain 5-HT neurons.