Advanced Brain Science: Optogenetics and Expansion Microscopy | Ed Boyden

Over the last decade, what we’ve been doing is trying to build tools that let us watch and control the operation of the brain. If we can understand the brain the way that we understand computers, for example, maybe we could understand the brain at such a level of detail that you could really comprehend how it generates things like thoughts and feelings, actions and sensations.

For example, one technology that we’ve developed is called optogenetics. In optogenetics, we install a gene that encodes for a light-sensitive protein in a cell or a set of cells in the brain. And then we can aim light at those cells down an optical fiber or with a scanning laser. So then you can play back activity to the brain. People have put artificial sensations into the brain. Can you figure out how a smell is represented, for example? People can trigger emotions, and some groups have done some pretty philosophically interesting experiments.

So, for example, a group at Cal Tech has activated certain clusters of cells deep, deep in the brains of mice. And if it’s the right cluster, you can actually trigger a mouse to become aggressive or violent. They’ll attack whatever’s next to them, even if it’s like a rubber glove, right? You can also study diseases. You can, for example, turn off overactive cells in a seizure and you can actually shut down seizures in animal models with epilepsy. These technologies are mostly being used in animals to reveal how brain circuits might be changed for therapeutic benefit.

So, for example, my group collaborated with another group to figure out that certain brain patterns actually might help clean up the debris in Alzheimer’s disease. From that knowledge, you can then go and develop other noninvasive strategies to try to help prevent, reduce the effects of, or reverse brain disorders. However, some people are exploring whether optogenetics might someday be used in humans directly.

And one area that’s of great interest is blindness. Millions of people cannot see because the photoreceptors in their eyes, the light-capturing cells, have died off. If you could convert the rest of the eye into a camera, though, by installing the optogenetic tools in spared cells of the eye, maybe you could help these people see again.

Another technology we’ve developed allows us to map the structure of the brain. The brain is really dense and complicated. In a cubic millimeter of your brain, you have around 100,000 cells called neurons, and they’re wired up. They’re connected at junctions called synapses. And there are about a billion synapses in that cubic millimeter. So mapping how the brain is wired up is a truly daunting task.

How can you image such a complex 3D structure with the nanoscale precision required to map the wiring? Well, we do it through a fairly unconventional way. In contrast to the last 300 years of imaging, where you use a lens to magnify light coming from a sample, we actually take pieces of brain and fuse them with a chemical that’s a lot like the stuff in baby diapers. And then we add water.

The baby diaper material swells and blows up the brain to make it 100 times or 1,000 times, or even more bigger by volume. So because we move all the molecules away from each other in a smooth, even fashion, we can map their relative organization. My hope is if we can map out the key geometry of the brain and how molecules are organized, maybe we could simulate a brain circuit while it’s doing something like constructing a decision, sensing something, or performing an action.

It’s not a very good metaphor, but imagine that the brain, you’re trying to solve the brain in the same way that you might solve a computer. You need to control the computer; that’s the keyboard. We use optogenetics. You need a map of the computer, the wiring; that’s what we’re using expansion microscopy for. And you need to watch the computer in action, the monitor. And we’re still working on those technologies.

I hope we’ll have that solved in the next couple of years. But if you can put those three things together – the wiring, the watching, and the control – you can do a lot of interrogation of how computational circuits work.