OSCILLATORY INTERACTION BETWEEN DORSAL ROOT EXCITABILITY AND DORSAL ROOT POTENTIALS IN THE SPINAL CORD OF THE TURTLE R. DELGADO-LEZAMA, J.-F. PERRIER and J. HOUNSGAARD* Department of Medical Physiology, Panum Institute, University of Copenhagen, Blegdamsvej 3, DK-2200 Copenhagen N., Denmark Abstract —The response to dorsal root stimulation, at one to two times threshold, was investigated in the isolated cervical enlargement of the turtle spinal cord. At frequencies near 10 Hz the synaptic response in motoneurons and the cord dorsum potential, after an initial lag time, oscillated in amplitude with a period of more than 1 s. The mono- and polysynaptyic postsynaptic response in motoneurons, the pre- and postsynaptic component of the cord dorsum potential and the dorsal root potential oscillated in synchrony. These oscillations were only observed with stimulus frequencies in the range 9–11 Hz. The oscillating response could only be evoked from stimulus sites to which dorsal root potentials were conducted from the spinal cord (2–3 mm). At more distant stimulus sites cyclic variations in amplitude of the cord dorsum potential and the synaptic response in motoneurons were not observed. During an oscillating spinal response to a stimulus train in one dorsal root ﬁlament, the response evoked by a stimulus in another short ﬁlament (2–3 mm) from the same root varied in amplitude with the induced oscillation. The spinal response to a stimulus in a longer ﬁlament (i.e. more than 3 mm) did not oscillate. It is argued that the oscillating responses described rely on interactions between distributed elements rather than on unit oscillators. We also show that primary afferent transmission is unaffected by the substantial variations in dorsal root potentials during oscillations.  1999 IBRO. Published by Elsevier Science Ltd. Key words: electrophysiology, spinal cord, presynaptic inhibition. Oscillatory and resonant activity play a role in several normal and abnormal functions of the CNS. These include locomo- tion, 13,21 sensation, 14 sleep 16 and epilepsy. 22 Often rhythmic activity can be attributed to the intrinsic properties of parti- cular neurons in a synaptically coupled network. 15,21 In other cases rhythmic activity has a distributed origin from synaptic interactions between neurons in a network. 9,11 In such cases unambiguous identiﬁcation of the mechanism underlying rhythmic activity is usually difﬁcult, experimentally and through modeling, because of overload of undetermined para- meters. The present study presents an example in which inter- action between distributed cellular elements in a spinal network is identiﬁed as the origin of the oscillating response to a train of stimuli imposed on dorsal root (DR) ﬁbres. It is shown that the oscillation involves an interaction between the excitability of DR ﬁbres, the amplitude of the dorsal root potential (DRP) and the number of afferents recruited by the stimulus. In the course of the study we also found that the postsynap- tic component of the cord dorsum potential (CDP) evoked by a stimulus in a peripheral nerve was unaffected by large varia- tions in amplitude of the DRP during DR stimulation at 10 Hz. The postsynaptic component of the CDP evoked by stimulation of a peripheral nerve was, however, depressed when preceded by a brief train of stimuli at high frequency applied to the DR. This shows that the amplitude of the DRP is not always a good correlate of presynaptic inhibition. EXPERIMENTAL PROCEDURES Preparation Adult turtles (Pseudemys scripta elegans) were anaesthetized with pentobarbitone (100 mg/kg, i.p.). The blood was removed by intraven- tricular perfusion with Ringer solution (6–10°C) of the following composition (mM): 120 NaCl, 5 KCl, 15 NaHCO 3 , 3 CaCl 2 ,2 MgCl 2 , 20 glucose. The cervical enlargement of the spinal cord with intact proximal nerve plexus was removed. The ventral roots were cut. Variable lengths of peripheral nerve, up to 3 cm, were left with the dorsal root in some experiments. Spinal cord sections of one to three segments in length were cut with a slicing machine. The preparation was glued vertically in a Plexiglas chamber and kept at room tempera- ture (20–22°C). For some experiments, ﬁlaments of a dorsal root were separated and mounted in separate suction electrodes. The preparation was continuously superfused with Ringer solution which was saturated with 0 2 (98%) and CO 2 (2%) in order to get a pH of 7.6. Recording and stimulation Intracellular recording in current-clamp mode was performed with an Axoclamp2A ampliﬁer. Sharp microelectrodes (50–60 MV) ﬁlled with potassium acetate (1 M) were used for intracellular recordings from motoneurons. Ipsilateral CDPs were recorded with a suction electrode (Fig. 1) using an AC ampliﬁer, gain 1000–10,000 (ISO Dam 8, World Precision Instrument). The same equipment was used to record DRPs from ipsilateral DR ﬁlaments. Extracellular stimulation was performed with suction electrodes by applying current pulses (Iso- Flex; AMPI) to an intact ipsilateral DR or to one or two DR ﬁlaments (Fig. 1). Signals processing All signals were digitized (sampling rate 16.6 to 32 kHz) with a 12- bit analogue-to-digital converter (DIGIDATA 2000 from Axon Instru- ments) and displayed by means of Axoscope software and stored on a hard disk for later analysis. Analysis were performed with Microcal Origin software (version 4.10 32-Bit). Drugs In some experiments the following drugs were applied to the normal medium: 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 20 mM; Oscillatory spinal response 731 731 Neuroscience Vol. 93, No. 2, pp. 731–739, 1999 Copyright  1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/99 $20.00+0.00 PII: S0306-4522(99)00187-6 Pergamon *To whom correspondence should be addressed. Abbreviations: AP-5, ( ^)-2-amino-5-phosphonopentanoic acid; CDP, cord dorsum potential; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; DR, dorsal root; DRP, dorsal root potential; EPSP, excitatory postsynaptic potential; IPSP, inhibitory postsynaptic potential; PAD, primary afferent depolarization.