By electrically stimulating the motor neurons of rats that have spinal-cord injury, in bursts that are attuned to the times at which the neurons receive voluntary motor commands, the animals' recovery can be improved.
People with spinal-cord injury face a host of challenges, including sensory, bowel, bladder and sexual dysfunction, paralysis and weakness. Although rehabilitation can help to improve motor and sensory function, recovery is limited. But writing in Proceedings of the National Academy of Sciences, McPherson et al.1 show that rats with spinal-cord injury can recover substantial motor ability when treated with a new type of electrical-stimulation therapy.
Rehabilitation from spinal-cord injury relies on the damaged nervous system adapting over time. For example, when a person with motor defects in their arm repeatedly practises hand grasps, their ability to do this improves as a result of changes in neural circuitry. This alteration in nervous communication patterns and connections to regain function is called neuronal plasticity. Therapeutic approaches that encourage neuronal plasticity therefore have the potential to improve recovery when combined with more-conventional rehabilitation strategies.
In one such approach, known as electrical stimulation, a barrage of electrical excitation is given to undamaged spinal-cord projections (called fibres) and the neurons that they target, boosting the body's own weak electrical inputs to these circuits. In addition to promoting neuronal plasticity, this technique can help to uncover latent motor functions that are presumably inactive owing to insufficient levels of excitation. For example, it has been used to demonstrate2 that even people with 'complete' spinal-cord injuries (who are paralysed below the level of the injury) retain some latent functional potential.
In the mid-twentieth century, the neuropsychologist Donald Hebb predicted that the synaptic connections between neurons can be altered, becoming stronger or weaker depending on the timing of their activation3. If the presynaptic neuron is activated before the postsynaptic neuron, then the synaptic connection is strengthened, increasing activation of the postsynaptic neuron in response to signals from the presynaptic neuron. This theory is summarized in a modern maxim4: “cells that fire together, wire together.” Thanks to improvements in microelectronics, it is now possible to create small devices that record electrical biological signals and then use those signals to trigger rapid electrical stimulation in the nervous system, allowing Hebb's theory to be tested directly in living organisms. A pivotal 2006 experiment5 in monkeys demonstrated that artificial coupling of areas in the brain's cerebral cortex can alter the function of the connected neurons. More-recent studies6,7,8,9 have begun to explore whether timed stimulation can strengthen synaptic connections in the brain and spinal cord.
McPherson and colleagues experimented with a variation on this technique in injured rats. They recorded an electrical signal from muscle that indicated that the rat was in the process of contracting that muscle, and used the signal to trigger an electrical-stimulation pulse in the spinal cord (Fig. 1). They hypothesized that, in this way, voluntary commands to induce muscle contraction (which are transmitted from the cerebral cortex and perhaps from other brain centres) would arrive at motor neurons in the spinal cord shortly before the neurons were activated by the stimulation pulse. This would strengthen the synaptic connections in the spinal cord. Once the connections had been strengthened, the voluntary commands from the brain might be sufficient to drive muscle contraction on their own.
First, the authors injured one side of the spinal cord of rats. This resulted in paralysis of the forelimb on the injured side, followed by limited natural recovery. The researchers then inserted microwires into the spinal cord, below the level of the injury. Such microwires can stimulate neurons, and even movement, with extremely low electrical currents (approximately ten times lower than that needed for neuronal stimulation at the surface of the spinal cord), allowing specific activation of motor neurons10. Wires were also implanted in forelimb muscles to allow recording of electrical signals and were used to trigger spinal-cord stimulation.
Six weeks after spinal-cord injury, the rats were unable to reach and grasp a small food pellet with the affected forelimb. All the animals were severely impaired, achieving only about 13% of their pre-injury scores in this reaching task. Over the following 13 weeks, they underwent rehabilitation training, practising reaching for 30 minutes each day, 5 days per week. One group of rats received only rehabilitative training; a second group received rehabilitation and electrical stimulation in the spinal cord that was triggered by muscle activity; a third group received rehabilitation and electrical stimulation in the spinal cord that was not linked to the timing of muscle activity.
Rats in the second group demonstrated substantial recovery. At the end of the regime, they had regained 63% of their pre-injury motor ability. By contrast, rats in the first and third control groups made much smaller improvements. Perhaps most importantly, the second group maintained motor function after a three-week follow-up period. By timing spinal-cord stimulation to coincide with the 'sweet spot' for motor-neuron activation, synaptic connections seemed to have been strengthened over time, allowing enhanced motor recovery.
Although spinal-cord circuitry is complex, and the exact mechanisms that underlie this dramatic effect remain to be fleshed out, McPherson and colleagues' results have substantial clinical relevance. This study is an example of therapeutic strategies for neural repair that go beyond simply exciting the injured nervous system with a blast of electrical stimulation. By mirroring the timing rules that the nervous system normally uses to enhance the strength of synaptic connections, it might be possible to guide neuronal plasticity in a more functional and adaptive way after injury.
Nevertheless, many theoretical and practical issues need to be addressed before such approaches are attempted in humans. For example, the safety of chronic electrical stimulation using electrodes in the spinal cord must be thoroughly tested. The timing of the stimulation may also need to be adjusted on the basis of differences in nerve-fibre size and speed of conduction in the human nervous system compared with that of rats. Nonetheless, the translation of techniques that exploit synaptic plasticity from bench to bedside is now a little closer.Footnote 1
McPherson, J. G., Miller, R. R. & Perlmutter, S. I. Proc. Natl Acad. Sci. USA 112, 12193–12198 (2015).
Angeli, C. A., Edgerton, V. R., Gerasimenko, Y. P. & Harkema, S. J. Brain 137, 1394–1409 (2014).
Hebb, D. O. The Organization of Behavior (Wiley, 1949).
Shatz, C. J. Sci. Am. 267, 60–67 (1992).
Jackson, A., Mavoori, J. & Fetz, E. E. Nature 444, 56–60 (2006).
Nishimura, Y., Perlmutter, S. I., Eaton, R. W. & Fetz, E. E. Neuron 80, 1301–1309 (2013).
Rebesco, J. M. & Miller, L. E. Prog. Brain Res. 192, 83–102 (2011).
Lucas, T. H. & Fetz, E. E. J. Neurosci. 33, 5261–5274 (2013).
Guggenmos, D. J. et al. Proc. Natl Acad. Sci. USA 110, 21177–21182 (2013).
Mushahwar, V. K. & Horch, K. W. Ann. NY Acad. Sci. 860, 531–535 (1998).
The author is co-founder and Interim CEO of NeuraLink Technologies, LLC, which is performing research and development on similar technologies.
Related links in Nature Research
About this article