Pharmacological and neurochemical evidence suggest that brain dopamine systems, and the dopamine transporter in particular, contribute significantly to the behavioral effects and reinforcing properties of cocaine. The first objective of this study was to determine whether the brain distribution of cocaine supports these conclusions. A high resolution neuroanatomical map of cocaine disposition in brain after i.v. administration was developed. [3H]Cocaine ([3H](−)-cocaine) was administered to squirrel monkeys (Saimiri sciureus) at a- trace dose (0.001 mg/kg) and at doses at or above the threshold for producing behavioral effects (0.1 mg/kg, 0.3 mg/kg). After 15 min, ex vivo autoradiography revealed the highest accumulation of [3H]cocaine in dopamine-rich brain regions, including the caudate nucleus, putamen, and nucleus accumbens/ olfactory tubercle. The norepinephrine-rich locus coeruleus, the hippocampus, and amygdala also accumulated large quantities of [3H]cocaine. Moderately high levels were found in the stria terminalis, medial septum, substantia nigra, and other regions. Lowest levels were found in the cerebellum. A high and positive correlation was established for the brain distribution of [3H]cocaine administered at trace or at behaviorally relevant doses (r: 0.94; P < 0.001). To determine whether radioactivity represented [3H]cocaine or its metabolic products, tissue extracts from brain regions with high levels of cocaine were subjected to thin layer chromatography using two solvent systems. In caudate-putamen, nucleus accumbens, cortex, and hippocampus, radioactivity comigrated with standard [3H]cocane. In substantia nigra, less than 70% of the radioactivity comigrated with [3H]cocaine, suggesting that cocaine metabolites are generated more rapidly in the substantia nigra than in other brain regions. The second objective was to determine the brain distribution of mazindol, a potent norepinephrine and dopamine transport inhibitor with low abuse liability in humans. The disposition of intravenously administered [3H]mazindol in brain (0.001 mg/kg, 0.007 mg/kg) was surveyed by ex vivo autoradiography. In sharp contrast to [3H]cocaine distribution, the highest accumulation of [3H]mazindol was localized in the norepinephrine-rich pineal gland, discrete regions of the hypothalamus (paraventricular nucleus, supraoptic nucleus), and the locus coeruleus. Moderately high levels were detected in the caudate-putamen, nucleus accumbens, and other regions. The following conclusions were drawn: (1) Although dopamine-rich brain regions are principal targets of cocaine after i.v. administration to the nonhuman primate, other prominent targets of cocaine (locus coeruleus, hippocampus, and amygdala) may contribute to the acute and chronic effects of cocaine. (2) [3H]Cocaine levels in primate brain regions are positively correlated with increased glucose utilization in rodent brain regions [Porrino (1993), Psychopharmacoloogy, 112:343-351]. (3) Higher levels of [3H]cocaine metabolites detected in the substantia nigra may indicate that cocaine's effects in this region are © 1994 Wiley-Liss, Inc.