In 2013, Garret Stuber, PhD, published a paper in the journal Science showing how an exquisitely precise research technique called optogenetics allowed his team to pinpoint the specific biological mechanisms in the brain that contribute to binge-eating. This groundbreaking work, according to UNC eating disorders expert Cynthia Bulik, PhD, “can lead us away from stigmatizing explanations that invoke blame and lack of will power” of people who suffer from eating disorders.
Since joining UNC in 2010 as an assistant professor of psychiatry and member of the UNC Neuroscience Center, Dr. Stuber has continued to perfect the research techniques that are now allowing him and other scientists to home in on the specific cells and cell types that contribute to various neuropsychiatric conditions, such as depression and autism. For his work, Dr. Stuber was awarded the 2013 Freedman Prize for Exceptional Basic Science Research by the Brain and Behavior Research Foundation, and this year he earned the Philip and Ruth Hettleman Prize for Artistic and Scholarly Achievement.
For our Five Questions feature, we sat down with Dr. Stuber to discuss how he became interested in neuroscience, why he came to UNC as a graduate student and later joined the faculty, and what his research entails.
In high school, I don’t think I even knew what neuroscience was. But as a kid I was always interested in taking stuff apart – electronics and computers – to figure out how stuff works.
I wasn’t interested in how the brain works until my first year in college, when I took a psychology course about neurons and how they were involved in behavior. That got me really interested. I took more classes with a behavioral neuroscience focus and volunteered in a lab when I was 19 with Dianne Lattemann at UW. My interest in neuroscience snowballed from there; I worked in that lab for three years.
That research was about the interaction between hormonal and metabolic states related to feeding and satiety signaling molecules and how this signaling could regulate the reward functions in the brain. This was back in the late 90s and early 2000s. We didn’t have all the cool techniques and approaches to do neuroscience that we have now. I really liked doing that kind of work, but for graduate school I really wanted to focus on reward-related behavior and things related to drug abuse, which was one of the reasons why I came to UNC; I wanted to work with Gina Carelli and Mark Wightman to look at dopamine signaling and that sort of thing.
Also, I really liked the diversity of the UNC faculty in terms of what they were working on. Other programs had real strengths in one focused area, but I felt that UNC had a lot of strengths, not only in things I was interested in but generally in neuroscience.
I was also looking for a change. I had spent most of my life in the Pacific Northwest, and I just wanted to experience a different part of the country.
For my postdoc, I wanted to take things in a more cellular and molecular direction and get more training in cell and slice physiology related to reward circuit function.
So I did my postdoc at UC-San Francisco with Antonello Bonci, who was a leader in [brain] slice physiology – [research in animals] – related to addiction and reward function and things like that.
When I finished graduate school in 2005, I thought, “Well, that chapter of my life is now closed; I probably won’t be coming back to Chapel Hill.”
To be honest, it was completely unexpected for me to come back to UNC. But it just so happened that when I went on the job market, UNC was one of the places hiring. Right after the job posted, I started talking to David Rubinow, chair of the psychiatry department, and we got along really well from the start. By that time I had other job offers but I decided that UNC was the best fit for me. The neuroscience community is very strong here and it has only gotten stronger since 2010. Also, the lifestyle and things like that in Chapel Hill were appealing.
Optogenetics is a technique that has evolved over the past 10 years. We use it to perform really precise manipulations on brain circuit function at the level of genetically defined cell populations. Within the brain, neurons are not universally the same. They’re all different and there are all sorts of different cell types that have different functions, connections, gene expression profiles, and things like that.
Given that there’s so much diversity in neuronal function, it’s been really difficult to understand the precise interconnected circuits and networks that orchestrate and encode aspects of behavior and function. This is where optogenteics and other approaches come in.
Using advances in molecular genetics, we can target very specific populations of neurons for manipulation. We might be working in a brain region where there are 50 different cell types all mixed together, but by using these tools we can selectively manipulate a subset of cells. We do that using light.
The idea is that we can genetically express parts of cells – light-gated channels and pumps – in only a subset of neurons. Then, using fiberoptics coupled to illumination sources, such as solid state lasers or LEDs, we can deliver light directly into the brain and modulate the activity of those cells we’ve targeted for manipulation.
So it’s a really nice synergistic approach that pulls on advances in molecular genetics, cell targeting, physics, optics, and neuroscience techniques.
Getting into optogenetics kind of just happened halfway through my postdoc. I was finishing my first project and the paper was accepted. I was mulling what to do next and reading the first papers about optogenetics, which was really preliminary at the time; only one or two labs in the world were using these approaches. Online I wound up chatting with someone from Karl Deisseroth’s lab at Stanford; he was one of the main inventors of this. I wound up working with one of his graduate students on some projects. I was at UCSF. I would drive down to Palo Alto every other week or so, and I’d help them with some of their projects and they’d help me learn these techniques. This evolved into a nice collaboration with the Deisseroth lab that we’ve continued up to this day.
I spent lot of time in the last years of my postdoc learning techniques but also developing them further to make them easier to use. It took me two years to really develop and know how to use these techniques. But now, in my lab, students can come in and learn it in a month or two and actually get data. So it’s come a long way.
The work has actually gone into a lot of different directions, and my lab has grown quite a bit. Originally my focus was on reward circuitry and implications for addiction. We’ve had a number of projects based on this. It turns out that a lot of aspects important for psychiatric disease states require the same sort of properly functioning reward circuits and neurons that also control motivated behavior.
So we became interested is studying neural circuits that regulate feeding behavior. In particular, we’ve been looking at different cell types and connections in the hypothalamus that can affect feeding behavior in animals.
Our other projects all fall under the umbrella of motivated behavior states but for different diseases.
My take on the implications of this work is that it’s a bit of a paradox – we’re studying neuropsychiatric related conditions from a basic science standpoint using animal models. The thing to always remember is that any of these conditions – whether it’s an eating disorder or depression or addiction – they’re all uniquely human conditions. They don’t exist in animals as they do in humans. So we don’t say we have an animal model of an eating disorder or depression because these conditions are so multifaceted and complex that you have to be studying a person to come close to fully grasping the condition.
What we try to do is focus on very precise aspects of these disorders. There are very well-defined phenotypes [characteristics and traits] that contribute to those more complex psychiatric states, such as an eating disorder. So that’s what we focus on – the more simple, well-defined phenotypes that we can model in animals.
One advantage of doing this is that we have exquisite cellular precision to uncover the actual cell types and neural connections that control these very specific aspects of behavior. It’s just not possible to do these experiments in humans.
Even using brain imaging, at best you’ll see a certain brain region of interest light up. Say, part of the cortex. But within that lit up area are billions of neurons of probably hundreds of cell types. Just understanding the connectivity and function at the cell level is not even close to feasible in humans right now. So we have to rely on basic science to uncover these things so we can have a better idea for how to create therapeutics and other interventions for people.
Also important to remember: most of the brain areas we’re interested in are evolutionary well-conserved brain structures – they exist in humans all the way down to the most primitive vertebrate animals. We think that, while humans have very developed and complicated cortical areas, a lot of subcortical areas that control innate behavior in humans are similar those same areas in animals. This is why we believe our basic science experiments can often tell us a lot about the human brain.
We’re still interested in optogenetics and circuit manipulations. But now we’re trying to understand how cells, networks of cells, and neural circuits are actually encoding aspects of behaviors. This requires not just manipulating those circuits but recording the activity of those cell populations.
Cell-based recording techniques have been around for years. We mostly rely on electrophysiology. We put wires in the brain and record action potentials – [electrical spikes that neurons fire] – things like that. But these techniques are fairly limited in terms of how many cells you can actually record from.
But using advances in optics and flourescent imaging, we’ve now moved into doing large scale imaging of neural activity over a vast number of cells at any given time. We can also do this in genetically defined cell populations as well.
We know what type of cell we can record from. We can image activity dynamics of hundreds of neurons right now. But this will easily escalate; soon we’ll be able to record thousands of cells simultaneously.
The data we’re getting are pretty hard to wrap your head around right now because we’re doing experiments that generate hundreds of gigabytes to terabytes of data in a single afternoon, and then we’re left saying, “well, what do we do with it; how do we analyze it all?”
One challenge will be pulling in people from outside of neuroscience who have expertise working with big data sets to develop strategies to analyze and make sense of the data and come up with new theories of how these different networks of cells interact with each other.
At some point you’d love to simplify it all and say “this” type of cell causes “this” type of behavior. Unfortunately, it’s not going to be that simple. What we’d like to do is be able to understand how genetically defined cell types are encoding aspects of behavior but also be able to experimentally repeat patterns of neural activity at the cellular level. Right now, using optogenetics, we can target a couple hundred or a thousand cells of a genetically defined type and we can uniformly stimulate or inhibit them. But that’s not really how the brain works. The brain is encoding different things at the cellular level. Some cells might be active and some might not be active. We really want to push toward real time recordings of cellular network dynamics but also have the ability to selectively perturb individual cells within that network.
How we will translate this work into humans, I don’t really know. But we do think that this kind of research will provide conceptual and theoretical frameworks for a better understanding of how neural circuits orchestrate behavior in humans. The idea, then, would be to design drugs to target different cell types that we learn are important for particular reasons.
That’s the end goal: to learn as much as we can to help as many people as we can.
Media contact: Mark Derewicz, 919-923-0959, email@example.com