Long-term effects of axotomy on neural activity during cat locomotion

1. Neural activity was recorded from cats during locomotion on a treadmill using electrodes in Silastic cuffs placed around the sciatic nerve and the lateral gastrocnemius-soleus, medial gastrocnemius, common peroneal and tibial nerve branches. Each branch gave characteristic patterns of activity which were studied before and after it was cut distal to the recording cuffs. Sensory and motor components were separated and measured using cross-correlation techniques. The amplitude of the cross-correlation peaks was compared with the amplitude of compound action potentials evoked by electrical stimulation and recorded from the same sites in the anaesthetized animal. 2. Sensory activity declined rapidly following axotomy and did not recover unless reinnervation occurred. Sensory activity even 5 months after nerve section and resuture had recovered to only a fraction of the control values. This reduction is attributed to a decline in the evoked compound potentials and to many fibres being unsuccessful in regenerating to appropriate sensory organs. 3. Motor activity declined more than could be accounted for by a decline in evoked potentials over the first month after axotomy. The extra reduction represents a decline in the number of impulses generated by alpha-motoneurones after axotomy. If regeneration was permitted, motor activity recovered to higher levels than did the evoked potentials for the whole nerve. Even if regeneration was prevented by nerve ligation, motoneurones continued to generate activity at a stable level over a period of months during which whole nerve compound potentials continued to decline. 4. The modulation of motor activity in ligated nerves during the step cycle was still appropriate to the required movement. Thus, activity recorded from severed nerves in human amputees may be useful in controlling powered artificial limbs. The persistence of motor activity may be responsible for the lesser degree of atrophy found in motor fibres than in sensory fibres following ligation (Hoffer, Stein & Gordon, 1979b).