“With Engineered Proteins, Scientists Use Optogenetics for the First Time to Help a Blind Patient See Again”, Megan Molteni2021-05-24 (, , )⁠:

Somewhere in Paris, in a white room, seated at a white table, a man wearing a headset reminiscent of those worn by VR gamers reached out with his right hand and placed his fingers on a black notebook. This simple motion, which he executed with confidence, was notable for one very important reason: The man had been blind for close to 4 decades.

What was different now was that as part of a clinical trial, genes had been injected into one of his eyes, causing neurons in the retina to produce a light-sensing protein normally found in the slimy bodies of green algae. When the black goggles he was wearing projected video images of his surroundings as a pulsed light beam onto those now-light-sensitive cells, the neurons fired, and the signal traveled up the optic nerve and into the visual processing center of the brain. The genetically modified neurons had become stand-ins for the photoreceptors he had lost many years before to a genetic disease called retinitis pigmentosa.

The man’s progress identifying objects inside the lab and out in the world were reported Monday in Nature Medicine. While he couldn’t see colors or fine details, the case study describes the first time optogenetic therapy successfully restored partial vision to a blind patient.

…For this study, researchers injected one eye of a 58-year old patient with an adenovirus-associated vector carrying the genetic instructions for a protein called ChrimsonR. When amber light strikes it, the protein shape-shifts, allowing ions to flow in and out of cells. The vector targeted retinal ganglion cells, which in a healthy eye, would gather signals from cones and rods and shuttle that information up to the brain’s visual cortex. Even in patients with advanced retinitis pigmentosa, these ganglion cells are still alive, but left idling without any information coming in. The addition of ChrimsonR allows them to sense light themselves.

Sahel and his collaborators had previously tried a different protein, one that is activated by blue-green light. And in mice it worked great. But that end of the visual spectrum is very energetic, and when they moved to testing in primate models, they encountered problems.

In a normal mammalian retina, one photo-sensing protein would activate another and another and another, resulting in an cascade that amplifies the signal. One protein can open up to 1 million ion channels. With optogenetics, one protein equals one channel, so scientists need to amplify the signal another way—by adding more light. That’s what the goggles are for. But too much blue-green light can be toxic to the remaining cells (the reason why you shouldn’t stare directly into the sun). By switching to ChrimsonR and amber light, the researchers were able to strike the right balance between effectiveness and safety.

…In addition to the notebook, the first patient was able to locate and count other objects, like cups and a small bottle of light green liquid. The bigger the objects and the higher the contrast, the more consistently he was able to spot them. The patient also reported being able to see crosswalks outside on the street and even count the number of white stripes. During the lab-testing portion of the study, the researchers used an EEG to record the neuronal activity across the man’s visual cortex, which suggested that the ChrimsonR activation was indeed propagating up to the brain.

None of these changes were immediate. It took 4 to 6 months post-injection for the proteins to be expressed in sufficient quantities, and a few months of training with the goggles for the patient to be able to orient the beam of light directly onto those protein-expressing cells in the retina. To locate the objects, the patient used his whole head to scan the area back and forth. And the vision that was returned to him was a grainy world of black and white contrast. To do things like read or recognize faces would require much higher resolution than what the optogenetic approach could provide.