In the 1960s, it was discovered that electrical stimulation of a brain region located at the junction between the midbrain and hindbrain, the mesencephalic locomotor region (MLR), could elicit controlled locomotion in the cat. In the human, the MLR corresponds to the area containing the pedunculopontine (PPN), cuneiform and subcuneiform nuclei, and projections from these nuclei are crucial for controlling movement. Human subjects asked to imagine they are walking demonstrate increased activity in these nuclei during fMRI studies. Consequently, deep brain stimulation (DBS) of the PPN in particular has been under investigation for the alleviation of gait disturbances in Parkinson disease. Could stimulation of this target area aid in locomotor recovery from spinal cord injury?
In most cases of spinal cord injury, some nerve fibers remain intact, including some of those descending from the MLR, providing residual connections from the brain to the spinal cord below the level of injury. Bachmann et al (Deep Brain Stimulation of the Midbrain Locomotor Region Improves Paretic Hindlimb Function After Spinal Cord Injury in Rats. Sci Transl Med. 2013;23;5(208):208ra146.) recently showed that DBS of the MLR in rats with incomplete spinal cord sectioning improved their ability to walk and swim. First, however, the authors used retrograde axonal tracing from the lumbar spinal cord to demonstrate that brainstem reticular nuclei dominate the input to the spinal cord and are driven by the MLR. MLR stimulation was then shown to modulate the strength of locomotor output in intact, awake animals, using 0.5-ms cathodal pulses at 50 Hz. Remarkably, at 4 weeks after T10 spinal cord injury destroying 75% to 88% of the white matter, DBS of the MLR restored walking in spinal cord–injured rats close to prelesional performance, almost fully restored hindlimb function during swimming, and allowed functionally paralyzed animals to regain basic movements. Further investigations, using targeted injections of the GABAA receptor agonist muscimol, demonstrated that neurons of the MLR and not motor cortex were responsible for MLR DBS-induced stepping. Lastly, complete hemisection experiments showed that MLR BDS-induced stepping depended on the integrity of ipsilaterally descending reticulospinal fibers.
This new work suggests that MLR stimulation may be useful for improving gait in patients with spinal cord injury, particularly those who have been living with the injury for some time. That varying results on locomotion have been reported in patients undergoing PPN DBS for Parkinson disease, however, is a reminder that the functional anatomy of the human MLR is complex and not fully understood. Attempting to reproduce the results of Bachmann et al in a nonhuman primate spinal cord injury model would be a reasonable next step. Nonetheless, these results in the rat do suggest the potential for neuromodulation of midbrain locomotor control nuclei to treat patients with incomplete spinal cord injury.