RAPID COMMUNICATION Antidromic Modulation of a Proprioceptor Sensory Discharge in Crayfish

Bévengut, Michelle, François Clarac, and Daniel Cattaert. Antithe periphery in the sensory axons (El Manira et al 1991b). dromic modulation of a proprioceptor sensory discharge in crayfish. These spikes have no postsynaptic effect centrally (Cattaert J. Neurophysiol. 78: 1180–1183, 1997. In the proprioceptive neuet al. 1994; El Manira et al 1991b) but their peripheral rons of the coxo-basal chortotonal organ, orthodromic spikes conactions are yet unknown. Therefore, we wanted to test vey the sensory information from the cell somata (located peripherwhether the antidromic spikes were able to modify the orthoally) to the central output terminals. During fictive locomotion, dromic activities of the sensory neurons. In crayfish, the presynaptic depolarizations of these central terminals elicit bursts possibility of stimulating and recording intracellularly from of antidromic spikes that travel back to the periphery. To determine identified CBCO afferents enables us to investigate such a whether the antidromic spikes modified the orthodromic activity mechanism. of the sensory neurons, single identified primary afferents of the proprioceptor were recorded intracellularly and stimulated in in vitro preparations of crayfish nervous system. Depolarizing current M E T H O D S pulses were delivered in trains whose frequency and duration were controlled to reproduce bursts of antidromic spikes similar to those Experiments were performed in oxygenated physiological saline elicited during fictive locomotion. According to their frequencies, on 20 in vitro preparations of the nervous system of crayfish, these antidromic bursts reduce or suppress the orthodromic disProcambarus clarkii (El Manira et al. 1991a; Sillar and Skorupski charges in both positionand movement-sensitive neurons. They 1986). The preparation consisted of the last three thoracic and the induce both a long-lasting silence and a gradual recovery after their first abdominal ganglia of the ventral nerve cord dissected together occurrences. Neither the collision between the afferent and the with all the nerves of the two proximal segments of the left fifth efferent messages nor the release of serotonin by the sensory neupereiopod (Fig. 1A) . The strand of the coxo-basal chordontal organ rons can explain these results. We therefore conclude that anti(CBCO), containing the sensory cell bodies of this proprioceptor, dromic bursts produce a peripheral modulation of the orthodromic also was dissected intact, and its distal end was attached to an activity of the sensory neurons, modifying their sensitivity by electromagnetic puller (Ling Dynamic systems, VT101) controlled mechanisms yet unknown. by a home-made function generator. Extracellular activity in the CBCO nerve was recorded using ‘‘en passant’’ platinum wire electrodes (200 mm in diameter) insuI N T R O D U C T I O N lated with petroleum jelly (Vaseline) . The left fifth ganglion was desheathed to allow intracellular recordings from CBCO sensory In the central nervous system of both vertebrates and interminals (CBT) in the neuropile. Glass microelectrodes, filled with 3 M KCl (resistance Å 10–12 MV) , were connected to an vertebrates, presynaptic inhibition of primary afferents is Axoclamp 2A amplifier (Axon Instruments) . Data were recorded correlated with primary afferent depolarization (PAD) (Econ a digital tape recorder (Biologic-1801) and through appropriate cles et al 1962, 1963; Jiménez et al. 1988; Kennedy et al. interface and software (Cambridge Electronic Device) directly 1974; Kirk and Wine 1984; Sillar and Skorupski 1986). onto a personal computer. Centrally, PADs reduce the amplitudes of the orthodromic For each CBT, intracellular stimuli were delivered in trains sensory spikes and thereby of the transmitter release they whose frequency and duration were chosen. For each sequence of n induce, thus reducing the excitatory postsynaptic potentials trains at given parameters, a peristimulus histogram was calculated in the postsynaptic neurons (Cattaert et al. 1992, 1994; Hed(bins of 20 ms) and normalized by dividing the bin values by n . wig and Burrows 1996; Kirk 1985; Pearson and Goodman Thus in the results, normalized histograms express the averaged 1981). number of occurrences of the orthodromic spikes per bin against time. In vertebrates, PADs are associated with antidromic spikes recorded from leg primary afferents in dorsal root filaments during both fictive (Dubuc et al. 1985, 1988; Gossard et al. R E S U L T S 1989, 1991) and normal walking (Beloozerova and Rossignol 1994, 1995). In crayfish, during fictive locomotion, 90% During fictive locomotion, CBTs received spontaneous of the primary afferents of a leg proprioceptor ( the coxophasic burst of PADs (Fig. 1B1) . These PADs were able basal chordotonal organ, CBCO) receive phasic bursts of to elicit antidromic spikes in the sensory afferents (Fig. PADs (4–20 mV in amplitude) locked in phase with the 1B2) . The instantaneous firing frequency of the antidromic locomotor rhythm (El Manira et al 1991b). Moreover, PADs spikes within spontaneous bursts was between 2 and 100 Hz of large amplitudes (15–20 mV) seen in 40% of these afferfor doublets, with a mean instantaneous firing frequency from 20 to 50 Hz (unpublished data; the mean instantaneous ents are able to trigger antidromic spikes that travel toward