Nerves Signals and Light

16 September 2013

Interview with

Karl Deisseroth, Stanford University

Karl Deisseroth is based at Stanford University in California and in 2005 he Nerve Cellinvented a new technique called optogenetics; this is a way of controlling nerve cells using light.

Chris - Karl, how does it work?

Karl - Well, optogenetics works by using light to open ion channels. We've heard a lot about ion channels and how they can be controlled by the electrodes that Hodgkin and Huxley used. Of course, those could only be put into a squid axon and if you want to study the mammalian brain where things are a thousand times smaller, you've got problems with putting enough electrodes in. We've also heard about chemicals that can affect ion channels. We can open and then close them.  Those are great. Those can get all the way through the brain, but they don't have this millisecond precision that is the normal currency of information flow in the brain.  So, what if you could have both? What if, you could high speed and allow access to many different neurons at once, even in let's say, the mammalian brain? So, what we do is, we found that there are some microorganisms that make just the perfect tool for this. They do it for their own reasons.  Not to help us out, but it turns out to be the perfect thing. There are small algae, small kinds of bacteria as well that make light activated ion channels and they use this to sense what going on in their environment, to respond to it appropriately, to generate energy for a whole range of different functions.

Chris - So, these are marine microorganisms which just happen to make a chemical structure which is sensitive to light. They use it for one job, to regulate their behaviour that you're saying, you can steal that and use it for your purposes.

Karl - We did. We capitalised on the millions of years of evolution that went in to designing these beautiful proteins. We bring them into a mammalian neurons and we can turn mammalian neurons which normally don't care about light very much, turn them into very sensitive, precise light responding units.

Chris - So, you take the gene from the alga which would normally make that channel that the alga uses then you can put that gene into the nerve cell, making the nerve cell respond to light in the same way the algae would.

Karl - That's exactly right and it sounds a little farfetched and in fact, a lot of people thought that it would be very unlikely that genes from these algae or these ancient forms of bacteria would work well in mammalian cells, that they would get to the right place in the cell, that they would have the right supporting components that would help them function well. To some extent, they needed some help, we had to engineer the tools a bit to help them work well. In the end though, it works quite well now and we can flash on pulses of light of different colours. We can turn on and off different kinds of cells even while animals are freely behaving.

Chris - So, you put the gene from the alga into the nerve cell and it's in some way colour selected so you can use different colours to control different subsets of cells.

Karl - That's right. Some algae make proteins that respond to blue light, some archaebacteria make proteins that respond to amber light. In fact, there's a whole spectrum of these that are made by nature and that we can tune somewhat also with some protein engineering. We can put one kind of protein into one kind of cell, another kind of protein into another kind of cell, and even start to act like the conductor of the orchestra if you will, and have different elements, different instruments that play at different times or at the same time.

Chris - How do you put the genes into the brain cells that you want them to go into, in the animal that you want to study?

Karl - Well, this was the hardest thing after the initial proof of principle that these tools could work well in cells. The question became, how do you actually are going to make it useful? How are you going to target these into the cells of interest? This took a few years to sort out. It was helped by the fact that these are all-in-one proteins that it's just a single gene that's make a single protein that does all the jobs that gets the photons from light, and it delivers the ion flow. So, we only had to put one thing in and we use a range of genetic tricks. Actually, we can use little bits of DNA that are called promoters or enhancers and different cells that do different jobs will turn on or off these little bits of DNA that govern the expression or the production of proteins and cells. If we attach these little bits of DNA, the promoters or enhancers to the gene for the light responsive protein, and then we put that into all the cells in a region, it'll only get turned on in some, and that gives us a very powerful kind of specificity. There are other kinds too. For example, we can use the very remarkable shape and morphology of neurons. If we put the gene into one region of the brain, but we deliver light into another region of the brain that may be far away, we can selectively therefore stimulate the cells that live in one region, but send a connection to the other region where we're delivering the light. That turns out to be very powerful. We call that projection targeting and that, you don't actually need to know any genetics at all. You just need to know your anatomy.

Chris - How do you get the light that's going to do those jobs into the right bit of the nervous system in your animal?

Karl - Yes, that was another thing we had to sort out. In fact, a lot of the scepticism early on, people said, well, light doesn't penetrate very deeply into the brain and that's true. It will only go about a millimetre in before it's 99% scattered and absorbed. But we developed fibre optic methods for targeting very deep brain structures and so now, we can play in patterns of activity into very deep structures in the middle of the brain and the brain stem even while animals are behaving.

Chris - What can you do with this that we couldn't do before?

Karl - Well now we can play in precise patterns of activity into defined cells and I'll give you just one example. I'm interested in motivation and reward, what makes people want to do things, what makes them feel good about doing things. This is something that is impaired in depression and various kinds of psychiatric disease. And we're now playing in different patterns of activity into a kind of cell called the dopamine neuron and we're determining that some patterns of activity, but not others turn out to control feelings of reward or motivation. And that may help us understand these psychiatric diseases.

Add a comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.