Controlling the ‘simple’ – descending signals from the brainstem command the sign of a stretch reflex in a vertebrate spinal cord
A. Büschges
DOI: https://doi.org/10.1113/JP273352
2017-02-01
Journal of Physiology
Abstract:When humans and animals move around in their environment, external cues monitored by sense organs provide relevant signals to direct and generate these movements. In doing so, sensory signals not only serve higher order functions, like deciding which motor behaviour is appropriate and where to direct it, they also provide important information for its functional execution at the level of the neuromuscular system, such as a body to swim, legs for walking or wings for flying. For example, the visual signals perceived by the eyes direct locomotor movements, and sensory signals from proprioceptive sense organs along the body and its appendages that report movements of the body parts and forces generated by them provide the relevant information necessary to generate and promote movements (e.g. Büschges et al. 2011). In the latter case, sensory signals can induce reflexes that ensure functioning of the locomotor organs. For 40 years it has been known that even ‘simple’ reflexes in the motor system, like stretch reflexes of leg muscles, which serve to assure an animals’ posture, can be modulated and even reversed to promote movements. In the mid-70s, Forsberg (Forsberg et al. 1977) and Bässler (1976) reported for the first time for vertebrates and invertebrates alike that reflexes can be reversed for the execution of active movements. Today considerable detailed information is available about the local mechanisms governing this neural processing (e.g. summary in Clarac et al. 2000). Though evidence for descending influences has been recently reported (Martin et al. 2015), it is not known whether specific pathways from the brain exist to command such marked changes in reflex action. The study by Hsu et al. (2017) published in this issue of The Journal of Physiology has explored this question in the lamprey locomotor system. The spinal locomotor network of the lamprey is organized along segmental modules which are controlled by descending signals from the brainstem (e.g. summary in Grillner, 2003). In each segment of the spinal cord, central pattern generating (CPG) networks exist on both sides that, by mutual inhibition, generate left–right alternation of motor neuron activity. These motor neurons innervate the segmental myotomal muscles of the body wall, the lampreys’ locomotor organ. A movement-sensing sensory system on both sides of the spinal cord, the so-called spinal stretch receptors (SSRs), provides sensory feedback to the segmental CPGs that assists the generation of left–right alternation in motor activity. This means that when the myotomal muscles on the left side contract, and bend the body and spinal cord towards the same side, SSRs on the right side are activated. This in turn assists inactivation of the contralateral networks, and activation of the ipsilateral networks, i.e. the right side, and thereby promotes robustness of the cyclic motor activity. Hsu et al. provide clear-cut evidence that the influence of feedback from the SSRs is not fixed, but modified in a task-specific fashion depending on the locomotor behaviour to be generated. The authors took advantage of the accessibility of spinal stretch receptors that provide feedback about bending of the spinal cord to test their reflex effects in quiescence as well as during the generation of the neural output for four different motor behaviours, i.e. fast forward swimming, slow forward swimming, backward swimming and lateral turning. While fast forward swimming represents the ordinary form of locomotion in lamprey, slow forward swimming, backward swimming and lateral turning are generated in the case of escape, i.e. when the animal has to get away from an undesired condition. Body kinematics differs markedly between these locomotor behaviours, and body curvature and undulation are much larger during escape than during regular swimming. The authors tested the influence of feedback signals from SSRs on specific induction of the four locomotor behaviours. Interestingly, SSR signals have opposite effects in fast forward swimming compared to all escape behaviours tested. In fast forward swimming elicited by electrical stimulation of the mesencephalic locomotor region, feedback from SSRs supports activation of the silent ipsilateral side of the spinal network. In contrast, during escape behaviours elicited by stimulation of the trigeminal nerve, the same feedback signals assist contralateral motor activity. This change in reflex action represents a classical reflex reversal. Thus, the first significant observation of this study is a task-dependent reversal of a spinal stretch reflex in lamprey. The authors then identified the neural mechanism underlying the reversal of SSR-induced reflex action in the spinal cord. By means of intracellular recordings, they analysed the activity of reticulospinal neurons (RSNs) in the brainstem, which are known to serve in activating and inactivating downstream locomotor networks in the spinal cord through their caudal projections. Interestingly, individual RSNs appear to be activated with the generation of reflex reversal not only during the execution of escape motor activity, but also independently of an escape motor output. Even stronger electrical stimulation of the sites of origin of the RSNs in the brainstem alone was sufficient to evoke the reflex reversal without escape locomotor output being generated. Pharmacologically blocking excitatory NMDA-mediated glutamatergic synaptic inputs to RSNs resulted in a reduction of the strength of reflex reversal. These results clearly demonstrate that reflex reversal in the spinal cord is directly commanded by RSNs independently of the activity of segmental locomotor networks. These findings are of the utmost relevance to our current understanding of the generation and control of motor behaviours in general: spinal reflexes can apparently be under direct and specific descending control from the brainstem. This to-date unknown mechanism for reflex modulation and reversal in the spinal cord greatly extends our understanding of their origin. This is because previous studies in a variety of motor systems had pushed for the notion that reflex modulation and reflex reversal automatically result from local network activity, e.g. from CPGs for locomotor activity (Clarac et al. 2000). The next important step now will be to determine