A Nobel Prize for Controlling the Brain With Light

It's Tuesday, October sixth. The Nobel Prize in medicine was announced yesterday, and it went to something that solves a problem you probably didn't know existed. For most of neuroscience history, if you wanted to know what a particular group of brain cells does, you had two options. You could watch them. Stick an electrode in and listen to them fire, which tells you correlation but not causation. Or you could destroy them. Wipe out a region and see what the animal could no longer do, which is precise in space but not in time, and also you just destroyed brain tissue. Neither approach could answer the most basic question: if you turn this specific set of neurons on right now, what happens? Francis Crick, who co-discovered the structure of DNA, called this out in 1979. He wrote that neuroscience needed a method to control one type of brain cell while leaving the others untouched. He even suggested light might work. He was right. It just took another twenty-six years, a single-celled pond alga, and a Stanford psychiatrist to make it real.

The story starts with a green alga. It's a single-celled organism that lives in ponds and soil, and it does something you have probably seen: it swims toward light. To do that, it needs a way to detect light and convert that detection into motion. The protein that handles this is called channelrhodopsin. It sits in the cell membrane, and when blue light hits it, the protein changes shape and opens a pore. Ions rush through, creating the tiny electrical pulse that a neuron uses to send a message — that's what it means for a brain cell to fire.

Two researchers in Germany spent years trying to figure out exactly which protein was doing this. In 2002, they published the discovery of channelrhodopsin-1 in Science. A year later, they found channelrhodopsin-2, which was even better: it let through a wider range of ions and responded faster to light. They had identified the molecular light switch that algae use to navigate.

This was beautiful fundamental biology. But nobody had connected it to the human brain yet.

Karl Deisseroth (DICE-er-ahth) was the person who did. He's a psychiatrist by training, he still sees patients at Stanford, and he also has a PhD in neuroscience and a lab in the bioengineering department. In 2004, he was a young professor running a small lab, trying to figure out how to control neurons with precision. He had read about channelrhodopsin and thought: what if you could take the gene for this algal protein and put it into a mammalian neuron?

That is not as crazy as it sounds, but it's close. You're taking a light-sensitive protein from pond scum and expecting it to function inside a rat brain cell. The first experiment was in August 2004. Deisseroth and two graduate students took the channelrhodopsin-2 gene from the German researchers' lab, inserted it into cultured rat neurons using a virus, and then pulsed the dish with blue light. The neurons fired.

It worked on the first try. One of them later described it as serendipity: the protein straight out of the alga, completely unmodified, was good enough to control a mammalian neuron. They published in Nature Neuroscience in 2005.

Two years later, Deisseroth's lab made it work in the brains of living mice. They had created a light switch for any neuron they chose. The field got a name: optogenetics.

Before optogenetics, brain research operated on a model that was roughly watch what happens and guess at cause and effect. You'd see a brain region light up when someone felt fear, but you couldn't prove that region was causing the fear. Maybe it was just along for the ride. Maybe it was doing something else entirely and the fear happened somewhere else. Correlation is not causation, and the brain is the world capital of correlation-not-causation.

Optogenetics changed that. Now you can pick a specific type of neuron. Not a region, not a blob, but neurons that share a specific genetic signature — so you can target just the fear-related cells and leave the memory-related ones right next to them completely alone. And turn it on or off with a pulse of light. You can do it in a moving animal. You can do it at the exact speed the brain operates, which is milliseconds. And you can watch what behavior changes.

Here is what that has revealed. Turn on a specific set of neurons in the amygdala, a structure deep in the brain involved in emotion, and a mouse that was calmly exploring will freeze in terror. Turn them off, and a mouse that should be scared will walk around like nothing is wrong. You are not just correlating amygdala activity with fear. You are controlling fear with a light switch.

The same logic has been applied to memory. Activate the neurons that fired when a mouse learned a particular location, and the mouse will act as if it remembers being there, even if it has never been shocked there. You can create a false memory. Not in a vague, conceptual way. You can make a mouse behave as if something happened to it that did not happen to it.

In Parkinson's disease, researchers have used optogenetics to map which circuits in the basal ganglia, the deep brain region that deteriorates in Parkinson's, are actually responsible for the tremors and the freezing. Deep brain stimulation, the current surgical treatment, works by sending electrical pulses into that region. But it hits everything. It's like turning on every light in a building because you want to find one room. Optogenetics lets you test which specific neurons need to be stimulated and which ones you should leave alone. It's not a therapy yet. You'd need to genetically modify human brain cells. But it is building the map that future therapies will follow.

For depression, researchers have found circuits connecting the prefrontal cortex to the amygdala and the nucleus accumbens (uh-KUM-benz), a reward center, where turning certain neurons on relieves depressive behavior in mice and turning them off induces it.

The obvious question is whether this stays in mice or comes to humans. It's coming.

The first human clinical trial using optogenetics started in 2016. It targets retinitis pigmentosa (reh-tin-EYE-tiss pig-men-TOH-sah), a genetic disease that causes blindness by destroying the light-sensing cells in the retina. The optogenetic approach is different from anything you'd expect. Since the patient's own photoreceptors are gone, researchers inject a virus carrying the channelrhodopsin gene into the remaining cells of the retina. The virus works like a delivery truck. It carries the gene — which is just an instruction manual — into the cell, and the cell reads it and starts building channelrhodopsin proteins, the same light-switch protein the algae uses. Cells that were never meant to see now have light switches embedded in their membranes. The patient then wears goggles that translate the visual world into pulses of light at exactly the right wavelength. Partial vision restoration has been demonstrated in blind patients.

This is not a cure. It is a workaround: turning cells that were never meant to see into light sensors. But it works, and it suggests a template for everything else. If you can make any cell light-sensitive, you can control it. The Nobel committee called optogenetics the foundation of a new era in neuroscience.

Deisseroth is fifty-four. He and the two German researchers will split twelve million Swedish kronor, about one point two million dollars, three ways. Deisseroth won't be hard to find if you want to congratulate him. He still sees patients at Stanford Hospital. Psychiatrist, bioengineer, Nobel laureate.

The most powerful tool we have for mapping the human brain came from a single-celled organism that swims toward light. Twenty years ago, that sentence would have been science fiction. Today it's a Nobel Prize.

Stay informed, stay curious, and we'll see you tomorrow.