Optogenetics Moves Towards the Clinic

by Jim Schnabel

February 8, 2016

Optogenetics is best known as a powerful toolkit for doing basic neuroscience. It involves placing opsin genes—which code for tiny, light-sensitive “switches” of neuronal activity—into a selected population of neurons. Researchers can then use a light source to flip those switches on or off, driving or suppressing the firing of the targeted neurons—for example in a live animal, to see how its behavior changes. In the decade since it emerged, optogenetics has been used in hundreds of studies to discover what various sets of neurons do in health and disease. Neuroscientists consider it revolutionary.

Now optogenetics is beginning its second act, as a basis for therapies. Optogenetic devices that are being tested in animals or are on the drawing boards would be able to directly manipulate neurons in live humans to treat, in principle, a variety of neuron-related disorders, from blindness and deafness to clinical depression.

“Using this technology you can control brain processes with the high speed and spatial resolution of a medical device, and the biochemical precision of a drug,” said Brian Chow, a researcher at the University of Pennsylvania who works in this area.

Treating blindness and deafness

Conceptually one of the simplest therapeutic applications of optogenetics is to insert opsin genes into neurons to make them responsive to natural light. Some forms of congenital retina degeneration, such as retinitis pigmentosa, seem particularly amenable to this type of therapy. The idea is to insert opsins into neurons in the degenerating retina, including any remaining rod or cone cells but also retinal neurons that aren’t normally light-sensing. This would boost the population of light-sensing neurons that send visual information back to the cortex. The cortex would learn to interpret the signals from the artificially light-sensitized cells as if they were rod or cone cells.

Chow noted that several companies already have done preclinical studies of this therapeutic strategy and are moving it to the clinic. “I wouldn’t be surprised if we saw clinical trials start in the very near future,” he said.

Most envisioned optogenetic therapies would be more complicated, and would involve not only the insertion of opsins into neurons to light-sensitize them, but also the use of light to activate or silence those neurons—for example, laser light beamed through a fiberoptic cable. Arguably the “optoprosthetic” devices that are nearest to the clinic are those that would replace electrical cochlear implants (CI) and other implants that aim to correct deafness.

“The electrical CI is really beneficial—more than 450,000 people use it and most of them can understand open speech, for example taking a telephone call,” said Tobias Moser, a researcher at the University of Goettingen who heads one opto-CI project. “The problem is that the electrical CI is really poor in its frequency resolution.”

An electrical CI works by using electrodes to stimulate auditory neurons in the cochlea. These neurons are arranged “tonotopically” so as to pick up the high frequency portions of the signal at the near end of the cochlea, low frequency portions at the far end, and middle-frequency portions in the middle. But as Moser points out, the electrical current flowing from a standard CIelectrode typically leaks from one frequency area to adjacent ones, thus muddying the signal. In principle, an opto-CI, using tightly focused spots of light, would be able to deliver frequency-specific signals much more precisely to the appropriate cochlear neurons, thus achieving much more lifelike frequency resolution.

Moser and his colleagues have been working towards developing an opto-CI with optogenetics experiments on rodents and monkeys, and have major, long-term funding from the European Research Council and the German government. “I feel quite optimistic that within five years we will be able to make a major leap forward and perhaps even finish preclinical studies in rodents—perhaps also in [monkeys],” Moser said.

Meanwhile another team of researchers from Harvard and Boston University has been investigating a different type of opto-prosthesis for treating deafness. Their device would stimulate auditory neurons that lie closer to the auditory cortex (compared to the cochlea) and thus in principle could treat patients whose cochlear neurons have degenerated.

In principle, optoprostheses could be used for an almost limitless number of brain-related medical needs. These would include connecting artificial limbs to their control networks in the motor cortex; suppressing seizure-triggering neurons to treat epilepsy; stimulating mood-related brain circuits to treat depression or anxiety; stimulating language centers to treat stroke-related speech deficits; replacing the motor-related neural signals that are lost in Parkinson’s disease; and repairing cognitive circuit problems in disorders such as schizophrenia, bipolar disorder, and ADHD. “Network neuroscience” researchers, who are trying to model brain networks to better understand and treat neurological disorders, consider optogenetics a broadly applicable technology for correcting errant brain-network activity.

The virus issue

Certainly there are hurdles that must be overcome if any of these hoped-for applications is to be realized.

Perhaps the most obvious hurdle is the perceived riskiness of the technique for inserting opsins into target neurons. Typically in animal models researchers use modified viruses to carry the opsin genes into the neurons—this is the “genetic” part of optogenetics. But the deliberate use of viruses in humans has traditionally been considered hazardous, since in principle it can cause harmful inflammation or even a cancer-triggering viral disruption of DNA.

The good news here is that the broader push for gene therapies has driven viral gene-delivery technology to the point that it is now being considered safe enough for some human applications. In particular, Moser pointed out, the adeno-associated virus (AAV)—which he is using to deliver opsin genes in his research—has been used, apparently safely, in dozens of human clinical trials including successful trials of gene replacement for the retinal disease Leber’s congenital amaurosis—considered a strong contender to be the first FDA-approved gene therapy.

Moser said that although he and his team plan to evaluate the safety specifically of his modified AAV opsin-gene-delivery virus in long-term studies in monkeys, it seems not to cause worrisome side effects in studies to date in mice. “I’m quite confident that this is not going to be a major hurdle in the end,” he said.

Speed and coding

The opsin protein, channelrhodopsin-2, that paved the way for the widespread use of optogenetics in neuroscience, was found in algae cells. Although it enables neuronal “firing” at rates high enough for many neuroscience applications, it doesn’t respond quickly enough to light pulses to be able to encode, for example, higher frequencies of speech and other ordinary sounds.

This hurdle may already have been cleared, however. In recent years scientists have isolated other opsins from algae and related organisms, and one of these, dubbed chronos, was recently reported to have faster properties that enable it to keep up with light pulses encoding high frequencies of sound. “Certainly since the advent of the chronos opsin we feel more comfortable,” said Moser.

Designing implantable, miniaturized, electro-optical prostheses represents another hurdle, as does the need to understand better how to translate, for example, sound waves into neuronal impulses to mimic normal auditory processing.

Possibly the greatest hurdle of all, for brain-related therapeutic optogenetics, is its invasiveness, since it requires at least the insertion of a fiber-optic cable into the brain.

“The threshold for implementing such technology ought to be somewhat high, since you have a combination of a gene therapy and an invasive device in the nervous system,” said Chow, though he noted that some planned opto-prosthetic applications would replace older therapies—such as deep brain stimulation—that are themselves invasive and risky.

Less invasive approaches

There are optogenetics-like approaches that would be inherently less invasive than optogenetics, because they would use other, skull-penetrating forms of energy to drive neurons, and thus wouldn’t require any through-the-skull cables or other hardware.

One such approach involves using radio waves to heat nanoparticles in neurons, which in turn open temperature-sensitive ion channels—a mechanism that can be used to drive neuronal firing. Another more recently described approach, sonogenetics, uses ion channels that are sensitive to the vibrations caused by ultrasound.

Sreekanth Chalasani at the Salk Institute, who has led the development of sonogenetics, is skeptical that medical ethics will allow, in most cases, optogenetics devices that are inserted through the skull and into the brain. “On the other hand,” he said, if human neurons can indeed be suitably and safely sensitized to ultrasound, “I’m pretty sure that institutions would be OK with [our using] a small cap with tiny ultrasound transducers. And I’m pretty sure we can convince Apple to design a little device that’ll be cute for people to wear.”

To the extent that such an approach is workable, it might well have a dramatic impact on the treatment of brain disorders. At the same time, though, its success could create a dramatic new ethical concern—that people’s brains could be “hacked” in this way.