The release of 5-HT in terminal areas of the rodent brain is regulated by 5-HT1B receptors. Here we examined the role of 5-HT1B receptors in the control of 5-HT output and firing in the dorsal raphe nucleus (DR), median raphe nucleus (MnR) and forebrain of the rat in vivo. The local perfusion (30-300 microM) of the selective 5-HT1B receptor agonist CP-93,129 to freely moving rats decreased 5-HT release in the DR and more markedly in the MnR. Likewise, 300 microM CP-93,129 reduced 5-HT output in substantia nigra pars reticulata, ventral pallidum, lateral habenula and the suprachiasmatic nucleus. The effect of CP-93,129 was prevented by SB-224289, but not by WAY-100635, selective 5-HT1B and 5-HT1A receptor antagonists, respectively. SB-224289 did not alter dialysate 5-HT in any raphe nuclei. The intravenous administration of the brain-penetrant selective 5-HT1B receptor agonist CP-94,253 (0.5-2.0 mg/kg) to anesthetized rats decreased dialysate 5-HT in dorsal hippocampus and globus pallidus, increased it in MnR and left it unaltered in the DR and medial prefrontal cortex. SB-224289, at a dose known to block 5-HT1B autoreceptor-mediated effects (5 mg/kg), did not prevent the effect of CP-94,253 on MnR 5-HT. The intravenous administration of CP-94,253 (0.05-1.6 mg/kg) to anesthetized rats increased the firing rate of MnR, but not DR-5-HT neurons. The local perfusion of CP-94,253 in the MnR showed a biphasic effect, with 5-HT reductions at 0.3-3 microM and increase at 300 microM. These results suggest that 5-HT cell firing and release in midbrain raphe nuclei (particularly in the MnR) are under control of 5-HT1B receptors. The activation of 5-HT1B autoreceptors (possibly located on 5-HT nerve endings and/or varicosities within DR and MnR) reduces 5-HT release. The effects of higher concentrations of 5-HT1B receptor agonists seem more compatible with the activation of 5-HT1B heteroreceptors on inhibitory neurons.