How fear flows through the mind

Researchers have identified two neuron populations in the brain that work together to control fear impulses.

The findings, published this week in a pair of complementary papers in Nature, may someday facilitate the development of better therapeutic interventions for psychiatric illnesses such as post-traumatic stress disorder and phobias.

Screen capture from Carnival of Souls

"A real exciting aspect of this work is how we've now come to understand the regulation of complex emotion -- in this case fear -- at a single cell level," said Stephen Maren, director of the Neuroscience Graduate Program at the University of Michigan, who was not involved in the research. "That's a pretty impressive feat."

The amygdala, an almond-shaped structure in the brain involved in emotional memory and learning, has historically been considered key in processing fear impulses, but researchers at the California Institute of Technology wanted to understand the process at the level of cells. "Ultimately, we'd like a mechanistic understanding of how specific circuits, not only regions, generate brain functions like fear," said Wulf Haubensak, a postdoc and first author on the first paper.

After conducting systematic screens for genes marking neurons in the amygdala, the team focused on one -- a gene encoding protein kinase C delta, or PKC-δ, which was specifically expressed in a subpopulation of neurons that had not been studied in detail. "We decided to try and find out what these cells do," said David Anderson, a biologist and senior author on the paper.

They manipulated the activity of the neurons using transgenic mice in which only neurons that expressed PCK-δ in the amygdala were silenced. They were surprised with what they found. In classical studies, when a rat's amygdala is destroyed, it loses fear behaviors. Yet when the team silenced only the neurons that expressed PCK-δ, they saw an increase in fear impulses. "It implied there's something more going on than just a simple relay of fear signals [in the amygdala]," said Haubensak.

The team reasoned that the neurons normally suppress fear behavior, like one end of a seesaw resting on a garden hose and cutting off the flow of water. When the neurons are active, they cut off fear impulses from leaving the amygdala, but when a fear response pushes on the other end of the seesaw -- potentially other neurons in the area -- the seesaw lifts off the hose and fear impulses escape. Yet the team didn't have enough data to prove their hypothesis.

Serendipitously, three years ago, shortly after developing their seesaw hypothesis, Anderson met Andreas Luthi of the Friedrich Miescher Institute for Biomedical Research in Switzerland at a conference. Luthi was studying the same part of the amygdala but using a more conventional method, recording electrical activity while tracking behavioral responses to fearful stimuli. Luthi had identified two populations of neurons that responded to a frightening stimulus -- one that became activated and another which was inhibited. Two opposing populations of neurons acting, perhaps, like a seesaw.

The teams combined their methods in a sophisticated experiment and found that the neurons identified by Luthi's electrodes were the same labeled by Anderson's genetic marker. "It was really remarkable," said Anderson. In a deft action, the team linked electrophysiological evidence with molecular genetic approaches to identify a specific microcircuit in the brain, one of the first neuron-level pathways for emotion ever identified.

"This is clearly becoming the modern neuroscience strategy for understanding neural circuits -- to use a marriage of molecular, genetic, anatomical, electrophysiological, and behavioral approaches to get at these problems," said Maren. "It's a tour de force."

Anderson hopes to next try and understand how these circuits are altered in animal models of psychiatric disorders, which could lead to better therapies for illness. "It raises really interesting implications for clinical management of fear and anxiety," agreed Maren, because targeting genetically identified populations of cells with a drug could be more effective than general therapies. Many of today's drugs, for example, affect so many neurons in the brain, it's like pouring oil all over a car engine and hoping some of it happens to drip into the right spaces, said Anderson.

"The dream for the future would be to try to target drugs to specific neurons in specific circuits," he added. "If that works, you could give a therapy without any side effects."

Haubensak, W. et al., "Genetic dissection of an amygdala microcircuit that gates conditioned fear," Nature, 468:270-6, 2010.

Ciocchi, S. et al., "Encoding of conditioned fear in central amygdala inhibitory circuits," Nature, 468:277-82, 2010.


Related stories:

Brain proteins make fears last
[3rd September 2009]

Manipulating Memory
[1st March 2009]

The Brain on Stress
[1st August 2007]