Modafinil is a stimulant compound with an unknown mechanism of action that has been used successfully to control excessive daytime sleepiness in narcoleptic patients in Europe (1,2). The compound also has a minimal effect on cataplexy and other accessory symptoms of the narcolepsy tetrad (1,2). It is well tolerated and is generally preferred to amphetamine and related compounds in the countries where it is available. The mode of action of modafinil is still not understood. The compound has been shown to decrease sleep and increase wakefulness in cats (3) and monkeys (4), and increase spontaneous locomotion in mice and to a lesser extent in rats (5). It also decreases barbitaland phenobarbital-induced sleep in mice and rats and does not produce stereotypies below very high doses (5). The compound is suggested to have no peripheral effects in rodents, dogs and monkeys (5-7). Surprisingly, the compound is also devoid of any sympathomimetic effects, including those at the cardiovascular level (2,6). The current hypothesis regarding the mode of action of modafinil involves the stimulation of alpha-l adrenergic mechanisms. The modafinil-induced increase in motor activity in mice is antagonized by central alpha-l antagonists such as prazosin or phenoxybenzamine, but not by dopamine antagonists (5). Similarly, the awakening properties of the compound (in monkeys and cats) were antagonized by prazosin, although antagonisms were only partial (3). Other experiments, however, have suggested a different mode of action. First, the compound does not bind alpha-l receptors in vitro (up to 10-3 Musing [3H]-prazosin in canine cortical membranes, data not shown). Secondly, our experiments in narcoleptic canines have shown that the compound does not modify canine cataplexy even at doses promoting alertness (7). Because canine cataplexy is very sensitive to compounds acting presynaptically and postsynaptically on adrenergic transmission (see reference 8 for review), the results suggest a nonadrenergic mode of action. Electrophysiological and volta metric data also do not support the hypothesis of catecholaminergic mechanism of action. High doses of modafinil have been shown not to modify the firing rate of noradrenergic neurons of the locus coeruleus or dopaminergic neurons of the ventral tegmental area or substantia nigra in rats (9). Similarly, modafinil does not modify catechol oxidation peaks as measured by voltametry in the mouse caudate nucleus and rat striatum or nucleus accumbens, including after pargyline treatment (10). To further study the profile of modafinil, the compound was evaluated using the NIMH/Novascreen® Psychotherapeutic Drug Discovery and Development Program. Binding assay experiments included receptors for adenosine, dopamine, ,),-aminobutyric acid (GABA), serotonin, N-methyl-D-aspartate, kainate, quisqualate, glycine (strychnine sensitive and insensitive), benzodiazepine, phencyclidine (PCP), MK-801, angiotensine, Arg-vasopressine, bombesin, cholecystokinin (central and peripheral), neurospeptide Y (NPY), substance K and P, neurotensin, somatostatin, vasoactive intestinal polypeptide (VIP), atrial natriuretic factor 1 (ANF1), epidermal growth factor, nerve growth factor, various ion channels (calcium channels N, T and L; chloride channels and low conduction potassium channels) and second messenger systems (adenylate cyclase, phorbol ester and inositol triphosphate). All results were negative when a concentration of 10-3 M was tested. In a second set of studies, we decided to explore the affinity of modafinil for various neurotransmitter uptake sites. The concentration tested was 10-4 M and uptake sites studied included adenosine CH-nitrobenzylthioinosine in rat cortex), choline CH-choline in rat brain), GABA CH-GABA in rat cortical membranes), dopamine CH-WIN in guinea pig stratum), norepinephrine CH-desmethylimipramine in rat cortex) and