164 Brain Research, 293 (1984) 1~ 1¢~ ~ [Zlsc~ic; BRE 20031 Strychnine eliminates alternating motor output during fictive locomotion in the lamprey AVIS H. COHEN and RONALD M. HARRIS-WARRICK Section of Neurobiology and Behavior, Seeley G. Mudd Hall, Cornell University, Ithaca, NY 14853 (U.S.A.) (Accepted October 18th, 1983) Key words: locomotion - - central pattern generator - - lamprey - - spinal cord - - strychnine - - oscillator The motor program for swimming in the lamprey includes rhythmic bursting of motoneurons with output from the two sides of a spinal segment alternating strictly out of phase. This motor program can be observed in vitro in the isolated spinal cord. Addition of strychnine (5-10/~M) can selectively eliminate the alternation between sides without blocking the temporal pattern of bursting on each side. This result suggests that temporal bursting on the right and left sides of the spinal cord is generated by independent neuronal os- cillators, while the alternation between the two sides results from crossed inhibitory coupling between these oscillators. In vertebrates and invertebrates, the motor com- mands for many rhythmic movements are generated within the central nervous system by neuronal en- sembles called central pattern generators (CPGs)6,12, 21. These rhythmic motor patterns are generally characterized by periodic bursts of moto- neuron activity with alternation in activity to func- tionally antagonistic muscles. During swimming in fish, this pattern is manifested as an alternation in the firing of motoneurons innervating the left and right muscles of a segmental myotome giving rise to a si- nusoidal movement of the body 7Ao. The source of this alternation has been proposed in many models of vertebrate CPGs for locomotion 9. In one class of models, alternation and rhythmic bursting are insep- arable, and result from intrinsic interactions within a single CPG network 1.17,2°. In the simplest model of this type, alternation and rhythmic bursting arise from reciprocal inhibition between two neuronal 'half-centers', each of which controls one half of the motor pattern 1. In another class of models 4.8,9 the CPG consists of several independent neuronal oscil- lator networks or 'unit CPGs', each of which contains all the elements needed to generate the rhythmic ac- tivity of a subset of the motor apparatus. Alternation would then arise from inhibitory coupling between the 'unit CPGs'. These two classes of models can be distinguished by blockade of reciprocal inhibition in the CPG networks (Fig. 1). If alternation arises from reciprocal inhibitory connections within a single CPG, blockade of inhibition should disrupt the pat- tern completely (Fig. 1A). If, however, alternation arises from reciprocal inhibition between indepen- dent 'unit CPGs', rhythmic motoneuron bursting will be preserved upon blockade of inhibition, but regular alternation will be disrupted (Fig. 1B). Using a phar- macological blockade of inhibitory pathways, we have studied the neuronal basis for alternation dur- ing 'fictive swimming' of the isolated lamprey spinal cord. Our results support the independent 'unit CPG" model. The preparation of the isolated spinal cord of the silver lamprey, Icthyomyzon unicuspis used to study the CPG for locomotion has been described else- whereL A dissected length of cord (25-50 segments long) was perfused at 4-6 ml/min with chilled lam- prey saline (mM: 115.0 NaC1, 2.l KC1, 2.6 CaC12, 2.0 MgCI2, 3 NaHCO 3, 3 glucose) to which drugs were added. Motoneuron activity was monitored extracel- lularly with bipolar suction electrodes placed on the two opposing ventral roots (VR) of a single segment. D-Glutamate (0.5-4 mM) was added to the saline to activate the motor pattern for swimming 5.~. In this 'fictive swimming' preparation, VR recordings dis- Correspondence." A. H. Cohen, Section of Neurobiology and Behavior, Seeley G. Mudd Hall, Corncll University. Ithaca. NY, U.S.A. 0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.