In behaving animals the red nucleus produces sustained action potential discharge during movements of the limbs. These bursts are thought to encode parameters of movement and thereby represent motor commands. Similar bursts can be recorded in the in vitro brain stem-cerebellum from the turtle. In this preparation, sustained discharge of red nucleus neurons was postulated to be generated by N-methyl-D-aspartate-mediated cellular mechanisms acting in combination with positive feedback in a recurrent cerebellorubral network. The present study was designed to test this positive feedback hypothesis. During recording of sustained discharge in the deep cerebellar nuclei and cortex, the red nucleus was reversibly inactivated by microinjection. The positive feedback hypothesis would be supported if activity in the cerebellum was attenuated by inactivation of the red nucleus. A nonrecurrent source of excitation would have to be postulated if cerebellar activity was unaffected. 2. Extracellular single-unit recordings were made from neurons in the deep cerebellar nuclei, cerebellar cortex, and vestibular nuclei. Burst discharges were evoked by brief electrical stimuli applied to the spinal cord that activated sensory structures. During inactivation of the red nucleus, sensory projections to the cerebellum that may evoke burst discharge were unaffected. Pressure microinjections of cobalt, lidocaine, gamma-aminobutyric acid (GABA), or 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) were used to reversibly inactivate the red nucleus. Saline injections were also tested. 3. Sustained discharge of all neurons recorded in the lateral cerebellar nucleus was greatly attenuated or blocked completely by injection of the pharmacological agents into the red nucleus. These effects were reversible. Of the recordings in the cerebellar cortex, 63% of these were blocked. All four compounds tested were effective blockers of the bursts, although the effects of GABA were less potent than the others. Saline injections into the red nucleus showed no effect. Burst discharges of single units recorded in either the medial cerebellar nucleus or the vestibular complex, which do not receive input from the red nucleus, showed no effect of red nucleus inactivation. 4. The results showed that sustained discharge in the cerebellum was significantly attenuated by inactivation of the red nucleus even though sensory input that may trigger the bursts was intact. These data support the hypothesis that sustained discharge in the cerebellorubral circuit is generated by a distributed neuronal network that uses positive feedback. The results have implications for mechanisms underlying normal brain function and some motor disorders.
