The mechanisms by which the atypical neuroleptic clozapine produces its therapeutic effects in the treatment of schizophrenia without causing the extrapyramidal side effects that are characteristic of most antipsychotic drugs remain unclear. Recently, a single injection of the typical antipsychotic haloperidol has been shown to increase c-fos expression in the striatum [Dragunow et al. (1990) Neuroscience 37, 287-294]. C-fos is a proto-oncogene that encodes a 55,000 mol. wt phosphoprotein, Fos, which is thought to assist in the regulation of "target genes" containing an AP-1 binding site. Because a wide variety of physiological and pharmacological stimuli increase c-fos expression, it has been proposed that Fos immunohistochemistry might be useful in mapping functional pathways in the central nervous system. The present experiments examined some potential neuroanatomical differences in the actions of clozapine and haloperidol by comparing their effects on c-fos expression in the medial prefrontal cortex, nucleus accumbens, striatum and lateral septum. The effects of the selective dopamine receptor antagonists SCH 23390 (D1) and raclopride (D2) were also examined. Haloperidol (0.5, 1 mg/kg) and raclopride (1, 2 mg/kg) produced large increases in the number of Fos-containing neurons in the striatum and nucleus accumbens. SCH 23390 (0.5, 1 mg/kg) reduced the number of Fos-positive neurons in the nucleus accumbens and striatum, and had no effect in the other regions. Neither haloperidol nor raclopride increased the number of Fos-positive neurons in the medial prefrontal cortex. Haloperidol, but not raclopride, produced a modest increase in c-fos expression in the lateral septal nucleus. Clozapine (10, 20 mg/kg) was without effect in the striatum; however, it significantly increased the number of Fos-positive neurons in the nucleus accumbens, medial prefrontal cortex and lateral septal nucleus. Destruction of mesotelencephalic dopaminergic neurons with 6-hydroxydopamine abolished the increase in Fos expression in the nucleus accumbens and striatum produced by haloperidol and raclopride, and also blocked the clozapine-induced increase in the nucleus accumbens. However, the inductive effects of clozapine and haloperidol on c-fos expression in the lateral septal nucleus and of clozapine in the medial prefrontal cortex were not affected by the 6-hydroxydopamine lesions. These results suggest that clozapine's unique therapeutic profile may be related to its failure to induce Fos in the striatum as well as its idiosyncratic actions in the lateral septum and medial prefrontal cortex. The effects of clozapine in these latter regions do not appear to be mediated by dopaminergic mechanisms.