ARTICLE IN BRIEF
A Stanford investigator discusses his work with optogenetics, an emerging field that combines optical and genetic techniques to probe neural circuits.
Imagine how much easier mapping circuits in the brain would be if every type of neuron came equipped with its own on-off switch you could control at will. That's the potential of a new technique called optogenetics, in which neurons in vivo are transfected with light-sensitive membrane channels, and controlled with an implanted optical fiber. The technique has already offered a new hypothesis for the mechanism of deep brain stimulation in Parkinson disease (PD), and has researchers thinking seriously about optogenetic therapy for a wide range of neurologic diseases in humans.
Karl Deisseroth, MD, PhD, associate professor of bioengineering and psychiatry at Stanford University in California, pioneered the technique, which employs light-activated membrane proteins — channelrhodopsins and halorhodopsins—found in microorganisms. Each protein is sensitive to a different frequency of light. “Every biologist is taught in school” about these proteins, Dr. Deisseroth said. “They've been studied for decades. But nobody had thought to put them into neurons.”
The channelrhodopsins are cation channels, while the halorhodopsins are chloride pumps. In neurons, activating a channelrhodopsin allows sodium or calcium ions in, lowering membrane potential and causing the neuron to fire. By pumping in chloride ions and raising the potential, the halorhodopsins do the opposite.
Viral transfection introduces the genes for the proteins into the mouse brain. “The problem with electrodes and drugs,” the alternatives for manipulating neurons, “is that they aren't specific for a particular kind of cell,” Dr. Deisseroth said. Genes, however, can be targeted to particular cell types, using a cell-specific promoter. All cells will take up the virus, but the promoter will restrict gene expression to the desired cell type.
The channels or pumps are stimulated by light of the appropriate frequency, delivered along a glass fiber, much thinner than current brain electrodes, placed within a millimeter of the target cells. The fiber can even be fused with a microelectrode to record neuronal activity in situ.
“There were a lot of reasons this shouldn't have worked,” Dr. Deisseroth said. But, he added, “it was a real ‘aha!’ moment,” when he and colleagues found that even their earliest, least refined attempts succeeded in controlling the targeted neurons.
Dr. Deisseroth has begun to use optogenetics to ask questions about circuits in the brain, including those involved in schizophrenia, depression, autism, narcolepsy, and PD. “I'm a psychiatrist, and also see patients in clinic with neurologic diseases,” he explained, adding that he has been frustrated with limitations of current treatments.
His investigation of subthalamic nucleus (STN) stimulation in PD, reported last April in Science, gives a flavor for how the technique is applied. Deep brain stimulation (DBS) with electrodes is a standard treatment for advanced disease, but it is unclear how it works. To address this question, Dr. Deisseroth expressed the two channel proteins in the STN of hemiparkinsonian rodents.
“The subthalamic nucleus is a predominantly excitatory structure embedded within an inhibitory network,” he explained, allowing investigators to target STN neurons with a promoter specific for glutamatergic neurons. With the optical fiber in place, they delivered yellow light to activate the chloride pump and suppress STN neuronal firing. A recording electrode indicated the neurons were successfully inhibited, but there was no effect on the behavior of the mice.
They tried again, this time delivering the channel proteins to astroglia, based on the idea that DBS may drive secretion of glial modulators that affect the STN circuitry. Again they could inhibit firing, but there was no behavioral effect. Neither did stimulating STN neurons (using blue light) have any effect.
“We were ready to give up because nothing was working,” he said. As a “last-ditch effort,” the investigators transfected axons arriving from the motor cortex. They found that light-induced high-frequency stimulation of these afferent fibers improved parkinsonian rotation to the point that the animals' behavior was indistinguishable from that of controls.
Dr. Deisseroth believes the effect he produced by stimulating these axons was to inhibit STN neurons. “This is not a refutation of previous models of DBS,” he said. But it does highlight the importance of the afferent system in STN control. However, he said, it is interesting that direct inhibition of STN neurons themselves from the point source of an optical fiber was not effective. “Probably what that means is that there really is a circuit property that has to change to get a therapeutic effect,” since a DBS electrode is also a point source, that modulation of afferents is critical to the effect of DBS.
“In this experiment, we broke apart the two possible effects of electrodes”— direct inhibition of STN neurons and inhibition of afferents—“and found that driving the axons was much more effective.”
More work remains to be done, but optogenetic therapy may become possible, delivering the channels to the brain using gene therapy, and delivering the light with an implanted fiber.
“It's all actually feasible — there is no fundamental barrier,” Dr. Deisseroth said, but he cautioned that such a direct translation is not essential. More importantly, these results “suggest a new site for therapeutic intervention,” namely the motor cortex, a possibility he and his collaborators are exploring. The results also suggest that axons might be a prime target within the STN, and even that drugs targeting these axons could be therapeutic.
The most important impact of optogenetics, according to Dr. Deisseroth, is that it provides insight and understanding of the inner workings of complex neural circuits.
“I think it's a fantastic technique,” said Andres Lozano, MD, PhD, professor in the department of surgery, and the Ron Tasker Chair in Stereotactic and Functional Neurosurgery at the University Health Network of the University of Toronto. “It is a wonderful new window into neural and circuit function, much more specific than electrical stimulation, because you can dissect out the relative attributes of having one element active or inhibited. This gives us the option to hone down on the mechanism.”
Dr. Lozano is less sure about the specific conclusions concerning STN circuitry. “They may be blocking only a small number of STN neurons” with the direct inhibition method, he suggested, leaving some doubt whether neuronal inhibition within the STN is as unimportant as the results imply. “The conclusions need to be validated with other methods. The [2009 Science] paper is more a proof of principle, that you can transfect elements of the circuitry, and selectively inhibit or activate neurons with light. That really is the main power.”
But Dr. Lozano is excited about the prospects for optogenetic therapy. “I think to scale up to the human will be a major undertaking, but it can be solved. It's tricky, but not impossible.”
“Gene therapy is being used for Parkinson disease now anyway,” he noted, with three clinical trials either completed or underway. “And there is no reason why these therapies cannot be combined,” Dr. Lozano said. “You can start envisaging a single catheter — what is essentially a portal — with the ability to deliver multiple therapeutics,” including light, electricity, and a gene therapy product.
One benefit might be to avoid some of the side effects of electrical stimulation due to current spread to nearby axons. “You could really tailor the stimulation to affect only those neural elements you choose, instead of anything in the vicinity, as occurs with electricity. Speech, dysarthria, eye deviation — these are all related to current spread to axons. We may be able to have a much smarter therapy.”
Neither is PD the only possible disease target. “Every indication for DBS that you can think of,” he said, including depression, Tourette syndrome, tremor, even Alzheimer disease — could potentially benefit.