ARTICLE IN BRIEF
MacArthur Fellow Sheila Nirenberg, PhD, discusses how she used optogenetics to develop a prosthetic device that helps blind mice see.
Sheila Nirenberg, PhD, sees the human brain as a complex mathematical problem, which is why as a post-doc at Harvard she pored through college textbooks teaching herself a lot of math that she wished she'd spent more time on in college. She knew that if she wanted to figure out how circuits work she had to become a code breaker — and that is what she did.
For the last 20 years, the neuroscientist has been using her passion for numbers and neurons to develop a way to help blind people see. Clearly, others outside the field are watching her work: In late September, she was one of 24 people chosen to receive what has become known as the MacArthur “genius” grants. As a fellow of the John D. and Catherine T. MacArthur Foundation, she will receive a “no-strings attached” stipend of $625,000 to be paid out in installments over the next five years.
Dr. Nirenberg, a professor of physiology and biophysics at Weill Cornell Medical College, loves a good problem and found herself early on in her career trying to unravel the mysteries of neural processing. She knew that if she could figure out neural codes — how neurons talk to one another — she could build new kinds of prosthetic devices.
WHY THE RETINA?
She puzzled over it. She chose to work on the retina as a model system because it was a self-contained organ that could be removed from an animal and studied in the lab. At the time, around six years ago, optogenetics, which uses genetics and optics to control the behavior or neurons in living tissue, was gaining attention. Using optogenetics, she explained, investigators could inject a virus that carries the gene for the protein channelrhodopsin into a neuron. The protein is sensitive to light, she said, and when it gets light, it triggers the neuron to fire an electrical pulse. The virus also carries a special signaling molecule that directs the channelrhodopsin into the retina's output cells.
Dr. Nirenberg knew that if she could figure out the neural code of the retina and combine it with optogenetics, she could produce a device that sends normal signals to the brain and produces a massively more powerful prosthetic device than had ever been tried before.
She found the code by showing normal animals images and recording from the ganglion cells. The data generated helped her develop a generalizable input-output relationship. The formula was a mathematical transformation that could translate essentially any image coming into the neural code the retina uses to communicate with the brain. In other words, it was a formula that translated images into a language the brain would understand.
Then, with the code in hand, Dr. Nirenberg set out to create a prosthetic device that could bypass the retina's input cells — the photoreceptors — and send the code for the visual information directly to the ganglion cells. (These ganglion cells are not damaged in macular degeneration or retinitis pigmentosa.) The ganglion cells do their job of getting the information to the brain.
Once she and colleagues got the device up and running they tested it on blind mice. They could record from electrodes placed in the retina and could “see” that the animal perceived the detailed visual information from the environment.
PROSTHETIC DEVICES FOR THE BLIND
With the success of the experiment, Dr. Nirenberg was again dreaming in numbers — the more than 10 million people in the world who are blind or facing vision loss due to diseases of the retina. The advances made in restoring sight remain a crude but hopeful attempt: Most rely on dozens of electrodes implanted into the eye to stimulate the retina. A stimulator chip encodes the visual information that it receives from a camera that is attached to glasses. The electrodes stimulate the ganglion cells to fire. There are about 10 different prototypes on the table and only one approved by the Food and Drug Administration (FDA). These devices provide the blind person with the perception of spots of bright light and high-contrast edges but not the actual image with all its complex detail.
Normally, an image enters the eyes and lands on photoreceptors. These cells extract the information and pass it through the retinal circuitry, which converts it into a pattern of electrical impulses that make their way to ganglion cells, which then send the code to the brain. Dr. Nirenberg knew that to build a device that could jump over the damaged front-end cells — the photoreceptors that are dying in macular degeneration and retinitis pigmentosa — and interact with the ganglion cells would require an encoder and a transducer.
EQUATIONS FOR RETINAL CODES
Knowing the retinal language was key to her success. She has a set of equations that act as a “code book” for images in its path. (She has already figured out the codes for the mouse retina and the monkey retina.) The technique that she developed starts with using optogenetics to modify proteins so that they are sensitive to light and then injecting them into the eye. A camera the size of a chip converts the output into the language of the retina. This information is beamed to the optogenetic proteins, which then take up the signal and transmit it to the brain via the ganglion cells and the optic nerve. The device does not require surgery to implant stimulating electrodes.
“A completely blind retina can now send out normal signals that the brain can understand,” the neurophysiologist explained. “No other device can do that.” (Other devices don't have the encoder.)
The goal, of course, is to help blind people see, said Dr. Nirenberg. The biggest hurdle is getting the FDA to green light the use of the optogenetic protein, and she needs money to support more research. Dr. Nirenberg is now in new territory. She is hoping to find investors willing to gamble that her device will transform the way blind people perceive their world. A few years ago she donned her usual outfit — a tailored silk shirt and faded jeans — and headed onto the TED stage to talk about her work. She is working on raising money so that she can eventually test the technology in patients. The MacArthur award will help, but she said that she needs about $3.5 million to get through the FDA process. She has already raised about $2 million.
Dr. Nirenberg's vision is sight for the blind, but she feels that her science of using neural codes to bypass damaged networks could be used to develop new prosthetic hearing devices and even a prosthetic that could help a stroke patient regain lost speech. She started a company called Bionic Sight. She also works intensely in designing robotic software.
The call from the MacArthur Foundation arrived three weeks before the rest of the world knew. “It was magical,” she said.
SHEILA NIRENBERG: ON THE ROAD TO THE MACARTHUR
Sheila Nirenberg grew up in Westchester and dreamed of becoming a writer. She quickly moved from English to psychology to neuroscience. During her undergraduate years, she signed on for a competition to solve a complex physical chemistry problem. She had to figure out how molecules moved through a medium. The answer was paradoxical, but she was right and won.
She went to Harvard for a PhD, working with Constance Cepko, PhD, a neurobiologist who was studying development and degeneration of the vertebrate retina. Her new student developed a novel technique to selectively kill different classes of cells in a network so that they could better understand how the classes of cells work together to solve problems. It was before scientists could knock out genes and there was no way to block cell-to-cell interactions.
After she finished her PhD, Dr. Nirenberg did a postdoc, also at Harvard; then a six-year stint as an assistant professor at the University of California, Los Angeles, where she got tenure. She worked on circuit breaking and neural coding. She was then recruited to Cornell and has been back in her home state ever since. The rest is MacArthur history.
TUNE IN: Find out how one MacArthur Fellow is making inroads in helping blind people see. Sheila Nirenberg, PhD, a professor of physiology and biophysics at Weill Cornell Medical College, was one of 24 people named in September as a 2013 MacArthur Fellow for her work in developing a prosthetic device to restore sight. Here in this TED video, she describes how the technique she developed to use neural codes to bypass damaged retinal networks in blind mice could be applied to develop new prosthetic hearing devices and even a prosthetic that could help a stroke patient regain lost speech. Watch the video here: http://bit.ly/aNQ4KB.
•. Nichols Z, Nirenberg S, Victor J. Interacting linear and nonlinear characteristics produce population coding asymmetries between ON and OFF cells in the retina. J Neurosci. 2013; :33:(37):14958–14973.
•. Boumash I, Roudi Y, Nirenberg S. A virtual retina for studying population coding. PLoS One. 2013; :8:(1):e53363.
•. Nirenberg S, Pandarinath C.. Retinal prosthetic strategy with the capacity to restore normal vision. Proc Natl Acad Sci USA. 2012; 109:(37):1501–1517.