Reversal of fusimotor reflex responses during locomotion in the decerebrate cat

The effect of brief trains of electrical stimulation, at 2, 3 and 20 x threshold (T), of cutaneous afferents in the medial plantar nerve on the discharges of single medial gastrocnemius static and dynamic gamma-efferents has been investigated at rest and during locomotion in a decerebrate car preparation. The units were classified as dynamic (10 units) or static (10 units) indirectly on the basis of their resting and locomotor discharge characteristics. Responses were assessed by calculating the change in mean gamma-rate during the 100 ms after stimulus onset compared with a control period. At rest, most dynamic neurones were inhibited by stimulation at 2T (9 of 10 units) and above. In contrast, the resting responses of most static neurones were excitatory al 2T (9 of 10 units) and 3T, while 20T produced static gamma-effects that varied in sign. During locomotion the responses of both types of gamma-efferent were phase related. Two patterns were observed with dynamic units. For seven dynamic neurones, at stimulus levels of 2T (7 units) and above, responses during electromyogram (EMG) bursts were inhibitory while those between bursts were not significantly different from zero. However, for three other dynamic units, a phase-related reversal of reflex responses was observed at some stimulus intensities (always 2T, 3 units) comprising inhibition during, and excitation between, EMG bursts. For static neurones, inhibitory (never excitatory) responses occurred during walking at stimulus intensities of 2T (10 units) and above. The locomotor responses of static units were maximum during (3 units) or between (7 units) EMG bursts and were minimum in the opposite phase of EMG activity. A task-related reversal of reflex responses was thus generally apparent (9 of 10 units) to low intensity stimulation (2T) for static gamma-efferents during locomotion (inhibition) compared with rest (excitation). During locomotion there was a significant linear relation between the magnitude of response and the background gamma-rate for static units and those dynamic units that did not exhibit phase-related reflex reversal (total, 17 units). For dynamic gamma-efferents, inhibition at rest and during locomotion occurred at short (spinal) latencies which were not significantly different and are consistent with the involvement of the same interneuronal pathway. We conclude that pathways of opposite sign may dominate the responses of fusimotor neurones to low threshold cutaneous afferents from the plantar surface of the foot depending on behavioural context. Furthermore, the cutaneous reflex responses of both types of gamma-motoneurones during locomotion appear to vary with the source of the afferent input and do not constitute a general excitatory drive. The results are discussed in relation to the role and reflex control of the fusimotor system.