Present pharmacotherapy of major depression is, in principle, based on enhancement of central monoaminergic neurotransmission. Clinical studies utilizing depletion experiments indicate that antidepressants which primarily enhance serotonergic or noradrenergic central activity, i.e. serotonin or nor-adrenaline reuptake inhibitors, largely work by two separate neuronal pathways. However, experimental studies have shown that noradrenaline may regulate serotonergic neurotransmission both at the serotonin cell body and nerve-terminal level. We therefore investigated the effects of the selective NRI reboxetine on serotonergic neuronal activity and extracellular levels of transmitter in the nerve-terminal area. In vivo electrophysiological experiments showed that low doses of reboxetine significantly enhance the firing rate of serotonergic neurons in the dorsal raphe nucleus of anaesthetized rats. Also, in the medial prefrontal cortex reboxetine (3 mg/kg s.c.) enhanced, whereas citalopram (3 mg/kg s.c.) reduced, extracellular concentrations of serotonin measured by means of microdialysis in awake rats, using a low dose of citalopram (0,5 micro M) in the perfusion solution. Local administration of reboxetine only induced an increase in cortical serotonin levels at very high concentrations (1000 micro M). Hence, NRIs may cause a secondary enhancement of central serotonergic activity by a mechanism separate from 5-HT reuptake inhibition; an effect that may contribute to their clinical antidepressant efficacy.