There is growing interest in N-desmethylclozapine (NDMC), the major metabolite of clozapine, as a unique antipsychotic because it acts in vitro as a 5-HT(2) antagonist and as a partial agonist to dopamine D(2) and muscarinic receptors. To explore this, we compared NDMC to a typical (haloperidol), atypical (clozapine), and partial-agonist atypical (aripiprazole) antipsychotic in preclinical models. The comparison was carried out using: brain D(2) and 5-HT(2) receptor occupancy; animal models predictive of antipsychotic efficacy (amphetamine-induced hyperlocomotion (AIL) and conditioned avoidance response (CAR) models); measures predictive of side effects (catalepsy and prolactin elevation); and molecular markers predictive of antipsychotic action (striatal Fos induction). NDMC (10-60 mg/kg/s.c.) showed high 5-HT(2) (64-79%), but minimal D(2) occupancy (<15% at 60 mg/kg) 1 h after administration. In contrast to other antipsychotics, NDMC was not very effective in reducing AIL or CAR and showed minimal induction of Fos in the nucleus accumbens. However, like atypical antipsychotics, it showed no catalepsy, prolactin elevation, and minimal Fos in the dorsolateral striatum. It seems unlikely that NDMC would show efficacy as a stand-alone antipsychotic, however, its freedom from catalepsy and prolactin elevation, and its unique pharmacological profile (muscarinic agonism) may make it feasible to use this drug as an adjunctive treatment to existing antipsychotic regimens.