Immunocytochemical studies of the turtle brain revealed the presence of serotonin (5-hydroxytryptamine, 5-HT) immunoreactive (5-HT-ir) processes in the granule and Purkinje cell layers, but not in the molecular layer (ML), of the cerebellar cortex. Immunoreactive axonal profiles were present throughout the granule cell layer (GCL) where they generally coursed in an anteroposterior direction and could frequently be seen to ascend toward the Purkinje cell layer (PCL). Occasional 5-HT-ir processes were observed adjacent to Purkinje cell somata. 2. The effects of exogenously applied serotonin on mossy fiber and parallel fiber evoked responses in turtle Purkinje cells were examined by use of intrasomatic and intradendritic recordings in an intact cerebellar preparation in vitro. 3. Bath application of serotonin (0.2-1.0 microM) produced a dose-dependent reduction in Purkinje cell membrane resistance, which was not correlated with changes in postsynaptic response amplitude. At 5-HT concentrations > 1.0 microM (0.01-5 mM), resistance values returned to control levels. No consistent changes in spike width or postspike afterhyperpolarization were seen in response to serotonin application, nor were endogenous pacemaker-like discharges affected. Firing rate, assessed as threshold response to depolarizing current injection (0.3-1.0 nA, 1 s duration), was increased in 51% and decreased in 40% of cells tested. 4. Single stimuli delivered to either the cerebellar peduncle or the GCL resulted in the activation of fast excitatory postsynaptic potentials (fEPSP). These responses were dose dependently reduced in amplitude by bath application of serotonin (0.2-1.0 microM). At concentrations ranging from 10 to 100 microM, the response amplitude following agonist application plateaued at approximately 70% of control value. With higher dose applications (0.5-5 mM) of serotonin, the response amplitude exhibited a steep reduction (from 65-10% of control value). 5. Brief trains of stimuli (5 stimuli, 50 Hz) delivered to either the cerebellar peduncle or the GCL resulted in the activation of slow excitatory postsynaptic potentials (sEPSP). The peak amplitude of this response was unaffected by bath application of serotonin at concentrations ranging from 0.2 to 100 microM. At higher concentrations (0.5-5 mM), the sEPSP peak amplitude was dose-dependently reduced, with the largest amplitude reduction seen after peduncular stimulation. 6. It is suggested that serotonin acts as a modulator of fast excitatory synaptic activity in the cerebellar cortex, while exerting little affect on slow excitatory events. The fact that serotonin preferentially affects fast excitatory transmission may have important implications for the integration of incoming sensory signals at both the granule and Purkinje cell level.
