Neural Circuits Research: How and Why

I wrote in my welcome message about my priorities. First, we need to fund excellent science. Second, we should support studies that will yield benefits on short, medium, and long-term timescales. I also have three particular areas of interest: neural circuits, computational and theoretical psychiatry, and suicide prevention. Here I will discuss an approach to translating neural circuit technology into novel treatment methods. These studies are an example of a research program with the potential to yield benefits in the medium-term.

“No way this will work.”

That is what I told my grad student, Nancy Padilla-Coreano, when she came to me with an idea. She had just spent the last three years on an experiment aimed at reducing activity in a specific component of a neural circuit we thought was critical for anxiety in mice. This circuit carries information from the hippocampus—a brain area involved in memory—to the prefrontal cortex—an area involved in interpreting information and making decisions. Using a carefully engineered virus, she was able to direct an inhibitory opsin—a protein that responds to light by decreasing neural activity—to the connections between these brain regions. She then used light to activate the opsin and inhibit circuit activity, which reduced anxiety in the mice. Nancy’s next idea was to try stimulating those inputs in a specific pattern, to see if she could increase anxiety instead of decreasing it. “Go ahead,” I said. “But that pattern won’t perfectly mimic brain activity. I’ll bet you it won’t work.” I promised a set of premium audio speakers for the lab if she proved me wrong.

Let’s just say the lab has a new set of speakers.

Such is the power of current technologies for manipulating activity in neural circuits, which are networks of interconnected nerve cells that work together to guide behavior. Neural circuit technologies allow us to dramatically modulate behavior in animals simply by turning up and down activity in a specific component of a circuit. The two mainstays of neural circuit technologies are optogenetics, which uses light to alter neural activity, and pharmacogenetics, which uses designer drugs to do the same. They have helped the emerging field of circuit neuroscience dramatically enhance our understanding of how behavior is produced by neural activity. We now know, for example, that the neural circuits responsible for learning about threats in the environment can be separated from those responsible for learning about rewards in the environment.^{1, 2, 3} Inhibit the first and you reduce fear; inhibit the second, you reduce motivation. Compare this to pharmacological treatments for anxiety, which can impair both, and you can imagine the immediate payoff if we could do for people what we could do in animals: turn down the fear circuit without turning down the reward circuit, and you would get the benefit without the side effect.

A lot has to happen first, though, if we are to use these technologies to improve treatment for mental illnesses. Two categories of advances are needed: advances in knowledge, and advances in technology.

First and foremost, we need to understand more about the neural circuits underlying the devastating symptoms of mental illnesses. This effort is the main purpose behind much of the neurobiological research supported by NIMH; we seek to understand what goes wrong in individuals with mental illnesses using a variety of experimental approaches, including using these circuit-based tools. We need to know which circuits are altered in disease and how; which circuit elements can be manipulated to reverse or compensate for these alterations; and at which points in time during the course of illness these manipulations are most effective.

In addition to seeking understanding, however, we need to further develop circuit-based technology if we are to apply these techniques to humans directly. For example, opto- and pharmacogenetics require targeting opsins or designer receptors within specific cell types and brain regions in order to work. We need to develop strategies that will enable such specificity in humans, without resorting to genetic tricks that are available in rodents but not in people. Current circuit-based techniques can be very invasive, requiring intracranial surgery and/or implanted devices. Developing methods that reduce or eliminate such invasive procedures would dramatically increase the potential reach of circuit-based therapies. Finally, we need sophisticated methods to test the efficacy of these approaches. These include verifying that the methods actually change neural activity in the expected ways, as well as quantitative behavioral tests that can accurately measure effects on behavior.

It might seem like science fiction, to imagine harnessing neural circuit technologies to reverse disease-related dysfunction deep within the living, thinking human brain. But more and more we can envision these methods as science fact rather than science fiction. Achieving that vision requires setting an ambitious agenda towards translation. And that is precisely what I am proposing here: that we in the psychiatric neuroscience community begin to think and plan with ambition, like Nancy did, taking on a long-shot project despite the doubts of her mentor. Nothing less will meet the tremendous challenges we face in trying to transform treatments for our patients who suffer from mental illnesses.

For the aficionados, I have articulated more of the details of this agenda in a perspective for scientific audiences.⁴

References

¹  Kim J et al. Antagonistic negative and positive neurons of the basolateral amygdala . Nat Neurosci. 2016 Dec;19(12):1636-1646.

²  Namburi P et al. A circuit mechanism for differentiating positive and negative associations . Nature. 2015 Apr 30;520(7549):675-8.

³  Gore F et al. Neural Representations of Unconditioned Stimuli in Basolateral Amygdala Mediate Innate and Learned Responses . Cell. 2015 Jul 2;162(1):134-45.

⁴  Gordon J. On being a circuit psychiatrist . Nat Neurosci. 2016 Oct 26;19(11):1385-1386.