Spinal recurrent inhibition influences the discharge patterns of motoneurons and spinal interneurons.(2) The precise pattern of this influence depends on the static and dynamic characteristics of this feedback system. It is thus of importance to quantify its characteristics as well as possible. We here compare nonlinear features (hysteresis) in Renshaw cells and recurrent inhibition in response to cyclic stimulation of motor axons. In pentobarbitone-anaesthetized or decerebrate cats, intracellular recordings were obtained from 26 hindlimb muscle nerves skeleto-motoneurons and extracellular recordings from nine Renshaw cells. Various hindlimb muscle nerves (dorsal roots cut) or ventral roots (dorsal roots intact) were prepared for electrical stimulation to elicit recurrent inhibition in motoneurons or discharges in Renshaw cells. Stimulus patterns consisted of repetitive pulse trains whose rates varied cyclically between around 10 pulses/s and several tens of pulses/s, at modulation frequencies between 0.1 and 1.0 Hz, in one of two waveforms: triangular or sinusoidal. Recurrent inhibitory potentials in motoneurons and discharge patterns of Renshaw cells were averaged with respect to triggers (cycle-triggers) marking a fixed phase in the stimulation cycle. In another two experiments, motor axons to hindlimb muscles (soleus and medial gastrocnemius) were stimulated with sinosoidal and distorted temporal patterns to show their effects on force production. Most often the cycle-averaged motoneuron membrane potential changed in a temporally asymmetrical way, i.e, it fairly rapidly hyperpolarized early in the stimulus cycle (during increasing rate) and then depolarized more slowly throughout the rest. Plotting hyperpolarization vs stimulus rate therefore yielded a hysteresis curve. Similarly, Renshaw cell discharge was more prominent during the increasing than the decreasing stroke of the cycle, again yielding a hysteresis curve compatible with that of recurrent motoneuron inhibition. Hence, Renshaw cells appear to impose their nonlinear dynamic behaviour on the entire system including the next synapse. This nonlinear behaviour of recurrent inhibition is bound to distort motoneuron discharge such that if the latter were sinusoidal without recurrent inhibition, it would be skewed, with recurrent inhibition, with slower rise and more rapid decline of firing rate during a cycle. This temporal motor axon activation pattern would distort muscle force time-course even further beyond what it would be with sinusoidal activation patterns (muscle hysteresis). Hence, in this form, recurrent inhibition cannot compensate for distortion of signal transmission through skeletal muscle. The hysteresis would have to be reversed, which might be effected by additional inputs modulating recurrent inhibition.
