“Through my experience with patients, I’ve recognized the huge potential of cochlear implants,” said Tobias Moser, director of the Institute for Auditory Neuroscience at the University Medical Center Göttingen and lead author of the new study. “At the same time I’ve also witnessed the shortcomings.”
Patients with cochlear implants often describe the sound quality through the devices as harsh and tinny. Listening to music or picking out one voice from several others is often impossible. The goal of the new research, said Moser, is to improve the technology and create “a more natural hearing so that patients can recognize the melody in music and speech.”
At the most basic level, the act of hearing is transforming sound into electrochemical signals, the language of neurons, that the brain can then interpret.
Much of this process occurs in the cochlea, a snail-shaped organ within the inner ear lined with specialized sensory cells called hair cells. When hair cells detect vibration through thin protrusions on their surface, they generate electrical current in neighboring nerve cells that travel to the brain.
In people who have dysfunctional or dead hair cells, cochlear implants work by electrically stimulating auditory nerve cells directly.
According to Dan Polley, an associate professor at Harvard Medical School and director of the Lauer Tinnitus Research Center who was not involved in the study, the successes and limitations of the cochlear implant are determined by the anatomy of the inner ear.
Within the cochlea, hair cells rest on an organic platform known as the basilar membrane that is floppy and wide on one end and narrow and taut on the other. This biomechanical organization causes hair cells to wiggle in response to specific sound frequencies or pitches that are mapped out smoothly from low to high, like keys on a piano.
When implanting cochlear implants, Polley explained, surgeons thread electrodes into specific locations along the basilar membrane to target nerves that are sensitive to particular frequencies.
With cochlear implants, however, electrical current spreads itself over a large area and activates not only the targeted nerves but also neighboring cells as well. This impercision distorts and muffles sound.
Moser and other scientists believe that using light, which can be more finely targeted, will improve the performance of implants.
“Using electricity to control neurons is like playing the piano with your elbow,” said Polley. “Whereas light is better. It is more like playing the piano while wearing mittens.”
Auditory neurons normally don’t respond to light. However, the rapidly growing field of optogenetics has made this possible. The key is to genetically engineer neurons so that they contain light-sensitive proteins that are found elsewhere in nature including bacteria, algae, and the human eye.
Moser and his team of scientists performed this technique in adult Mongolian gerbils instead of mice or rats, which were used in previous studies. Gerbils, in contrast to other rodents, are a better proxy for humans because they hear low frequencies used by the human ear and have relatively large cochleas that are only two-and-a-half times smaller than those of humans.
The researchers also used a new version of light-sensitive proteins, called CatCh, to increase how quickly nerve cells could respond to light stimulation. Previous versions of the protein worked sluggishly to move ions in and out of neurons — a critical step in generating electrical signals to the brain conveying the rapidly changing characteristics of sound like loudness and pitch.
This really hampers our ability to “capture the dynamics of speech,” said Polley. “If you have to deliver pulses [of light] more slowly, you can’t keep up.”
To test how quickly neurons equipped with CatCh could process information, researchers exposed neurons to rapid bursts of laser light, up to 300 flashes per second. They observed that individual neurons and groups of neurons were able to keep pace with the flashes by releasing their own corresponding surges of electrical energy.
“This is a big improvement over our previous work, “ said Moser. The response “is quick and approaches physiological performance of normal neurons.”
Aside from faster processing, the neurons also showed a graded response to different intensities of light. This suggests that the gerbils could experience an accurate representation of loudness that is proportional to how strongly neurons are activated by light.
But how could researchers be sure that all this promising electrical activity was actually creating a sense of hearing in gerbils?
Scientists first surgically implanted a loop of optical fibers — a primitive prototype of high-tech cochlear implants that could one day be used in humans — into the gerbils’ inner ear. They then trained the animals to jump over a fence-like obstacle dividing two halves of a box. Once the gerbils learned this behavior, scientists deafened the animals, causing them to become unresponsive to the loudspeaker. However, when researchers pulsed blue laser light through the optic fiber, the gerbils leapt into the air again.
“It’s not hearing until you measure behavior,” said Polley. “Because hearing is a psychological property. [This experiment] shows that the animals generalize light stimulation to sound, they treat it as if it were sound.”
Despite these dramatic results, Moser is quick to point out that light-based cochlear implants are not ready for human use.
The most obvious hurdle is that installing light-sensitive proteins into cells requires genetic engineering. In the case of gerbils, scientists accomplished the feat by injecting viruses carrying specially designed DNA into the animals’ ears. While gene therapy is being used to treat certain diseases in clinical trials — and has been approved for a rare form of blindness — it remains a long way from reality for most conditions.
“Right now,” said Polley, “ I doubt anyone would suggest that you get this gene therapy instead of cochlear implants.”
Even if gene therapy was a proven technology, light-based implants have other restrictions. In humans, bioengineers imagine using tiny light-emitting diodes (LEDs) to stimulate genetically engineered auditory neurons. However these machines consume a lot of power and dissipate heat, making it unclear how patients would wear or operate such a device.
“Optogenetics is the most sophisticated way we have to stimulate nerves in the cochlea,” said Adrien Eshraghi, a professor at the Miller School of Medicine and director of the Hearing Research Laboratory at the University of Miami who was not involved with the study. “It is still [in] the early basics of science research, but I think optogenetics has a good future.”
Another source of optimism is the power of the brain to adapt to new signaling inputs.
In the case of cochlear implants, patients initially experience the human voice as high-pitched and scratchy (think Mickey Mouse) but adapt over time. Although researchers don’t know what light will sound like, they expect a similar adjustment process.