Role of cutaneous and proprioceptive inputs in sensorimotor integration and plasticity occurring in the facial primary motor cortex

Abstract

Key points: Previous studies investigating the effects of somatosensory afferent inputs on cortical excitability and neural plasticity often used transcranial magnetic stimulation (TMS) of hand motor cortex (M1) as a model, but in this model it is difficult to separate out the relative contribution of cutaneous and muscle afferent input to each effect. In the face, cutaneous and muscle afferents are segregated in the trigeminal and facial nerves, respectively. We studied their relative contribution to corticobulbar excitability and neural plasticity in the depressor anguli oris M1. Stimulation of trigeminal afferents induced short-latency (SAI) but not long-latency (LAI) afferent inhibition of face M1, while facial nerve stimulation evoked LAI but not SAI. Plasticity induction was observed only after a paired associative stimulation protocol using the facial nerve. Physiological differences in effects of cutaneous and muscle afferent inputs on face M1 excitability suggest they play separate functional roles in behaviour. Abstract: The lack of conventional muscle spindles in face muscles raises the question of how sensory input from the face is used to control muscle activation. In 16 healthy volunteers, we probed sensorimotor interactions in face motor cortex (fM1) using short-afferent inhibition (SAI), long-afferent inhibition (LAI) and LTP-like plasticity following paired associative stimulation (PAS) in the depressor anguli oris muscle (DAO). Stimulation of low threshold afferents in the trigeminal nerve produced a clear SAI (P < 0.05) when the interval between trigeminal stimulation and transcranial magnetic stimulation (TMS) of fM1 was 15–30 ms. However, there was no evidence for LAI at longer intervals of 100–200 ms, nor was there any effect of PAS. In contrast, facial nerve stimulation produced significant LAI (P < 0.05) as well as significant facilitation 10–30 minutes after PAS (P < 0.05). Given that the facial nerve is a pure motor nerve, we presume that the afferent fibres responsible were those activated by the evoked muscle twitch. The F-wave in DAO was unaffected during both LAI and SAI, consistent with their presumed cortical origin. We hypothesize that, in fM1, SAI is evoked by activity in low threshold, presumably cutaneous afferents, whereas LAI and PAS require activity in (higher threshold) afferents activated by the muscle twitch evoked by electrical stimulation of the facial nerve. Cutaneous inputs may exert a paucisynaptic inhibitory effect on fM1, while proprioceptive information is likely to target inhibitory and excitatory polysynaptic circuits involved in LAI and PAS. Such information may be relevant to the physiopathology of several disorders involving the cranio-facial system.