Mammalian peripheral C-fibre axons possess tetrodoxotin (TTX)-resistant sodium channels, a property that has been exploited to investigate responses to selective activation of C-fibres in the cat (Schomburg et al. 2000). The present study was designed to establish the viability of using TTX to give selective block of A-fibre conduction in the rabbit.

Experiments were performed in eight rabbits decerebrated under nitrous oxide/halothane (2-2.5 %) anaesthesia. The left sural nerve was cut peripherally and desheathed over a 20-30 mm length, part of which was laid in a plastic trough. Platinum stimulating electrodes were applied to the nerve just distal to and proximal to the trough. A third pair of electrodes was placed against the nerve at a central location to record afferent volleys. Reflex responses to sural nerve stimuli were recorded from the ipsilateral medial gastrocnemius (MG) muscle nerve. Stimuli were applied to the sural nerve in cycles that were repeated at 20 min intervals. Each cycle consisted of eight shocks given at 1 Hz, of 0 V, 11 times threshold (T, maximal for Aδ axons) and 111 T (maximal for C-fibres) applied to the distal stimulating electrodes, and 97 T applied to the proximal stimulating electrodes. After three control cycles, 15 µl TTX (citrate, dissolved in Ringer solution) was applied to the sural nerve in the trough. The concentrations used were 10, 20 and 30 nM. Two cycles were applied after the 10 and 20 nM applications, and four after the 30 nM dose. The animals were then spinalized at L1 and four more cycles were applied. MG responses were analysed in three time bands: phase 1 (5-12 ms post-stimulus); phase 2 (12-100 ms) and phase 3 (100-260 ms). Data are expressed as means ± S.E.M. Experiments were terminated by I.V. injection of saturated KCl solution.

The Aβ and Aδ components of the sural nerve compound action potentials elicited from the distal stimulating electrodes were abolished after application of 30 nM TTX, peak-to-peak amplitudes being 2 ± 1 and 1 ± 1 % of pre-TTX values, respectively. By contrast, the C-fibre wave was reduced to 70 ± 8 % of pre-TTX amplitude. Before TTX, stimulating at 11 or 111 T evoked MG reflexes that were significantly greater than 0 (single sample t tests, P < 0.01) in all three phases. After TTX, only the phase 3 response to 111 T stimulation was still significant, albeit at 38 ± 17 % of pre-TTX values. After spinalization the phase 3 response to 111 T increased significantly (paired t test, P < 0.05) but no reflexes appeared at earlier latencies. Volleys evoked from the proximal electrodes were unaffected by TTX and reflexes elicited from this location tended to increase throughout the experiment.

Thus TTX at 30 nM gives a selective block of A-fibre conduction in rabbit, and provides a useful method for studying the spinal actions of C-fibre afferents in isolation from other inputs.This work was supported by BBSRC.

Schomburg, E.D., Steffens, H. & Mense, S. (2000). Neurosci. Res. 37, 277-287.