The mammalian lumbar spinal cord has the capability to generate locomotor activity in the absence of input from the brain. Previously, we reported that transcutaneous electrical stimulation of the spinal cord at vertebral level T11 can activate the locomotor circuitry in noninjured subjects when their legs are placed in a gravity-neutral position (Gorodnichev RM, Pivovarova EA, Pukhov A, Moiseev SA, Savokhin AA, Moshonkina TR, Shcherbakova NA, Kilimnik VA, Selionov VA, Kozlovskaia IB, Edgerton VR, Gerasimenko IU. Fiziol Cheloveka 38: 46-56, 2012). In the present study we hypothesized that stimulating multiple spinal sites and therefore unique combinations of networks converging on postural and locomotor lumbosacral networks would be more effective in inducing more robust locomotor behavior and more selective control than stimulation of more restricted networks. We demonstrate that simultaneous stimulation at the cervical, thoracic, and lumbar levels induced coordinated stepping movements with a greater range of motion at multiple joints in five of six noninjured subjects. We show that the addition of stimulation at L1 and/or at C5 to stimulation at T11 immediately resulted in enhancing the kinematics and interlimb coordination as well as the EMG patterns in proximal and distal leg muscles. Sequential cessation of stimulation at C5 and then at L1 resulted in a progressive degradation of the stepping pattern. The synergistic and interactive effects of transcutaneous stimulation suggest a multisegmental convergence of descending and ascending, and most likely propriospinal, influences on the spinal neuronal circuitries associated with locomotor activity. The potential impact of using multisite spinal cord stimulation as a strategy to neuromodulate the spinal circuitry has significant implications in furthering our understanding of the mechanisms controlling posture and locomotion and for regaining significant sensorimotor function even after a severe spinal cord injury.

We attempted to elicit automatic stepping in healthy humans using appropriate afferent stimulation. It was found that continuous leg muscle vibration produced rhythmic locomotor-like stepping movements of the suspended leg, persisting up to the end of stimulation and sometimes outlasting it by a few cycles. Air-stepping elicited by vibration did not differ from the intentional stepping under the same conditions, and involved movements in hip and knee joints with reciprocal electromyogram (EMG) bursts in corresponding flexor and extensor muscles. The phase shift between evoked hip and knee movements could be positive or negative, corresponding to 'backward' or 'forward' locomotion. Such an essential feature of natural human locomotion as alternating movements of two legs, was also present in vibratory-evoked leg movements under appropriate conditions. It is suggested that vibration evokes locomotor-like movements because vibratory-induced afferent input sets into active state the central structures responsible for stepping generation.

Air-stepping can be used as a model for investigating rhythmogenesis and its interaction with sensory input. Here we show that it is possible to entrain involuntary rhythmic movement patterns in healthy humans by using different kinds of stimulation techniques. The subjects lay on their sides with one or both legs suspended, allowing low-friction horizontal rotation of the limb joints. To evoke involuntary stepping of the suspended leg, either we used continuous muscle vibration, electrical stimulation of the superficial peroneal or sural nerves, the Jendrassik maneuver, or we exploited the postcontraction state of neuronal networks (Kohnstamm phenomenon). The common feature across all stimulations was that they were tonic. Air-stepping could be elicited by most techniques in about 50% of subjects and involved prominent movements at the hip and the knee joint (approximately 40-70 degrees). Typically, however, the ankle joint was not involved. Minimal loading forces (4-25 N) applied constantly to the sole (using a long elastic cord) induced noticeable (approximately 5-20 degrees) ankle-joint-angle movements. The aftereffect of a voluntary long-lasting (30-s) contraction in the leg muscles featured alternating rhythmic leg movements that lasted for about 20-40 s, corresponding roughly to a typical duration of the postcontraction activity in static conditions. The Jendrassik maneuver per se did not evoke air-stepping. Nevertheless, it significantly prolonged rhythmic leg movements initiated manually by an experimenter or by a short (5-s) period of muscle vibration. Air-stepping of one leg could be evoked in both forward and backward directions with frequent spontaneous transitions, whereas involuntary alternating two-legged movements were more stable (no transitions). The hypothetical role of tonic influences, contact forces, and bilateral coordination in rhythmogenesis is discussed. The results overall demonstrated that nonspecific tonic drive may cause air-stepping and the characteristics and stability of the evoked pattern depended on the sensory input.

A new method for the activation of spinal locomotor networks (SLN) in humans by transcutane ous electrical spinal cord stimulation (tESCS) has been described. The tESCS applied in the region of the T11-T12 vertebrae with a frequency of 5-40 Hz elicited involuntary step like movements in healthy subjects with their legs suspended in a gravity neutral position. The amplitude of evoked step like movements increased with increasing tESCS frequency. The frequency of evoked step like movements did not depend on the frequency of tESCS. It was shown that the hip, knee, and ankle joints were involved in the evoked move ments. It has been suggested that tESCS activates the SPG (SLN) through in part, via the dorsal roots that enter the spinal cord. tESCS can be used as a noninvasive method in rehabilitation of spinal pathology.

No abstract