A rubbery ribbon of silicone, laced with cracked bits of gold that transmit nerve signals, has been spliced into the broken spinal cords of paralyzed rats, restoring their ability to move. The implant may be the first step towards helping paralyzed people in the same way.

Injuries that cause paralysis are like cuts in a telephone cable. Signals that start in the brain are supposed to travel down nerves in the spinal cord to muscles, but breaks in the nerves interrupt them. Patching the breaks with new wires, jumping over the cut in the phone line, should restore communication.

But it is an unfortunate paradox that, in people who cannot move, their spines still can. The nerves stretch and bend. Rigid wires implanted next to them rub them raw, creating scars and even more damage. And movement of the spine breaks the stiff electrodes. This rigidity has hobbled the whole spinal patch idea. "Other groups developed spinal implants, but they failed after a few weeks," says Gregoire Courtine, a neuroscientist at the Swiss Federal Institute of Technology in Lausanne. Successful implants that stimulate nerves, like pacemakers, do not come in direct contact with them.

Now Courtine and his colleagues have developed an implant made of stretchable silicone that can be placed right on nerve tissue, underneath the membrane that protects it, which is called the dura mater. In the January 9 issue of Science, the researchers reported that their new "e-dura" not only sends signals along the nerves, it has fluid channels that can deliver medication to nerve cells, stimulating them to heal. "This work represents a significant advance in the development of biocompatible devices," says Reggie Edgerton, director of the Neuromuscular Research Laboratory at the University of California, Los Angeles: In his own work, Edgerton has placed electrodes outside the dura mater and shown they can help paralyzed patients recover limited movement. But the therapy was limited by the external location of the device. Eventually, he thinks, the technique developed by Courtine could be used  "to control stimulating devices in a manner that will permit real-time fine tuning of the functional state of neural networks, making it easier for an individual perform movements rather automatically."

The Swiss group developed a ribbon of silicone that mimics the softness of the actual dura. They embedded it with gold wires to conduct nerve signals from one end to the other. Since the gold was too stiff, they fractured it with microcracks, enabling it to bend along with the silicone. The scientists then compared the performance of the e-dura to that of an implant made with stiffer wires. The different implants were placed in separate groups of rats with similar spinal cord damage. After six weeks, the rats with stiffer implants had more trouble walking and keeping their balance.

To test the therapeutic effects of the e-dura, the scientists implanted it in another group of paralyzed rats and used the implant to stimulate the animals with electrical signals as well as chemicals that improve nerve impulse transmission. Over six weeks, the rats showed they could walk. And because the e-dura delivered the chemicals directly to nerve tissue, scientists could use far less of it then when they tried an injectable form, thus reducing side effects. In fact, Courtine says, any side effects vanished.

Impressive, says Edgerton, although he cautions "the remaining challenges will be in establishing durability of such devices in real life." And of course, establishing that the implants work in real people, too.