MIT Technology Review

Thursday, May 31, 2012

Paralyzed Rats Walk Again

Spinal stimulation combined with assisted walking therapy generates new neural circuits and restores voluntary leg movement.

Rats paralyzed by spinal-cord injury can learn to control their hind limbs again if they are trained to walk in a rehabilitative device while their lower spine is electrically and chemically stimulated. A clinical trial using a similar system built for humans could begin in the next few years.

Researchers in Switzerland used electrical and chemical stimulation to excite neurons in the lower spinal cord of paralyzed rats while the rodents were suspended by a vest that forced them to walk using only their hind legs. The rehabilitative procedure led to the creation of new neuronal connections between the movement-directing motor cortex of the brain and the lower spine, the researchers report in Science.

Previous research has shown that it is possible to reverse some of the effects of spinal-cord injury by circumventing the normal connection between the brain and legs, which is broken by the injury. For example, walking can be triggered in spinal-cord-injured rats if their spine is stimulated. But until now, such movement has been involuntary. This new research shows that with a specialized training system, similar rats can regain voluntary control over their legs.

A report published last year showed the proof of principle "that this kind of approach can work in patients," says Grégoire Courtine, senior author of the rat study. In May 2011, 25-year-old Rob Summers, who had been paralyzed from the chest down in a car accident, was reported to stand on his own for a few minutes with electrical stimulation of his spinal cord. He could also take repeated steps on a treadmill with the stimulation, which activates regions in the lower spinal cord that control walking. The locomotion resulting from this kind of stimulation is automatic and involuntary and is thought to require no direct communication from the brain.

Courtine had previously shown that this type of automatic walking could initiate walking patterns in the hind limbs of spinal-cord-injured rats that were spinally stimulated while on a treadmill. Because the spinal column could control the walking pattern, Courtine suspected that only a weak signal from the brain would be necessary for the animals to start walking voluntarily.

To test whether the rats could recover brain-directed control of these movements, he and his team developed a robotic support system that suspends rats in a bipedal standing posture and helps with balance but does not provide any forward momentum. Ten paralyzed rats were trained daily to walk with stimulation both on a treadmill and in the robotic system. After two to three weeks, the rats took their first voluntary steps. "This is the first time we have seen voluntary control of locomotion in an animal with [an injury] that normally leaves it completely paralyzed," says Courtine. 

Key to this recovery was the active role of the rat's brain in wanting to move forward. The electrical and chemical stimulation puts the rat's nervous system in a state where walking is possible, says study co-author Janine Heutschi, and "then you need to make the rat to want to walk." The rats' desire to walk was motivated by chocolate rewards and vocal encouragement from the researchers (which you can hear in this video from the Swiss Federal Institute of Technology). The robotic suspension system forces the rodents to use their dormant hind limbs and not drag themselves forward with their still functional forelimbs.

The combination of electrochemical stimulation and active training, which included walking up stairs and around obstacles, resulted in new neuronal connections that bypassed the site of injury. "We promoted extensive remodeling of the neuronal connections not only at the site of the injury but throughout the central nervous system, including in the brain," says Courtine. What was most surprising, he says, was the fourfold increase in neuronal projections sent to the brain stem from the motor cortex, which provides conscious control of movements. "The motor cortex becomes the maestro of the reorganization process.

The conscious intent of the rats was necessary for the remodeling as well.  The nervous systems of rats that received the electrochemical stimulation but trained only on treadmills did not demonstrate the anatomical changes. "You need to incorporate an input from the brain," says Heutschi. "It doesn't work if the rat is on a treadmill; you have to force them to use the brain to control their hind limbs."

The clinical significance of the findings is unclear, according to Rutgers University neuroscientist Wise Young, because of the unusual surgical injury to the experimental rats (two cuts on each side of the cord at different heights). A more relevant injury would have been a spinal-cord contusion or bruise, he says.

However, other experts think the results are promising for those spinal-cord-injured patients who do not have a complete cut through the cord. Even though all connections between the brain and the lower spinal cord were disrupted in the experimental rats, "there are some remaining fibers, so the beauty of their technology is using the robotic training system to activate those remaining connections that can allow the cortex to control the limbs and to regain voluntary movement," says Zhigang He, a neuroscientist at Harvard Medical School. "This robotic training system makes that happen," he says.

Plans are under way to develop a human-sized version of the training system and to test its effects in clinical trials in Europe. Researchers at the Swiss Federal Institute of Technology and other European institutions are also working on an improved, implantable version of the electrical spinal stimulation system that may find its way into humans next year.