Caltech-Led Team Pinpoints Aggression Neurons in the Brain

PASADENA, Calif.—Where does violence live in the brain? And where, precisely, does it lay down its biological roots? With the help of a new genetic tool that uses light to turn nerve cells on and off, a team led by researchers at the California Institute of Technology (Caltech) has tracked down the specific location of the neurons that elicit attack behaviors in mice, and defined the relationship of those cells to the brain circuits that play a key role in mating behaviors.

The researchers' hope is that these insights might lead to treatments that can specifically address impulsive violence, a category of behavior that has been historically difficult to grapple with from a medical or psychological perspective.

In a study published in this week's issue of the journal Nature, the researchers were not only able to localize the neuronal circuits mediating attack behavior in mice, but were able to determine that these circuits are "intimately associated, deeply intertwined," with another basic social-behavioral drive—mating—according to David J. Anderson, the Benzer Professor of Biology at Caltech and a Howard Hughes Medical Institute investigator.

Indeed, the neurons for violence and mating live so close together, in a brain region known as the ventrolateral subdivision of the ventromedial hypothalamus (VMH), that "they are like a salt-and-pepper mixture," says Anderson.

And if you think of the brain as the world and the hypothalamus as a country, he adds, then the ventromedial hypothalamus is like a state and the ventrolateral subdivision is like a city within that state. "We've found that these 'mating' and 'fighting' neurons are not only located in the same city, but potentially in the same neighborhood," he says.

To determine whether aggression and mating involve the same—or, rather, distinct but intermingled—neurons, Dayu Lin, a former postdoctoral fellow in Anderson's lab and now an assistant professor at New York University, carried out a series of challenging electrophysiological recording experiments in the VMH. Because the VMH is located deep within the brain, it is exceedingly difficult to target accurately. But by inserting a bundle of 16 wire electrodes into this region, Lin was able to get recordings from multiple neurons during repeated episodes of fighting or mating. "It's the first time in which it's been possible to record electrical activity from deep-brain hypothalamic structures in animals while they are engaging in aggressive and mating behaviors," says Anderson.

Using software developed by Allen E. Puckett Professor of Electrical Engineering Pietro Perona and senior postdoctoral scholar Piotr Dollar—and with the help of several Caltech undergraduates—the researchers annotated behavioral changes on a frame-by-frame basis from video taken at the same time as the electrical recordings were performed. This annotation allowed them to make correlations between neuronal activity and behavior with a temporal resolution of approximately 30 milliseconds.

These experiments indicated that while there is some overlap between "mating" neurons and "fighting" neurons, the majority of these cells are distinct, despite their close proximity. Perhaps most surprising, Anderson notes, is the way that the neurons responsible for aggression and mating communicate—or, rather, how they shut each other up. Sex and violence, it seems, are actually at odds: a neuron that is turned on during aggressive behavior will turn off during mating, and vice versa. "We found that they talk to each other in an inhibitory way," he says.

But a correlation between neuronal activity and fighting behavior doesn't indicate whether the activity causes the behavior or the behavior causes the activity. And so Lin, Anderson, and colleagues carried out experiments to activate or inhibit VMH neurons, to distinguish between those alternatives and to determine the effect of such manipulations on behavior.

In order to activate neurons in the VMH, they used a technique known as optogenetic stimulation. Using a disabled virus as a kind of "disposable molecular syringe," Lin injected VMH neurons with DNA that carries the code for channelrhodopsin-2, a protein from blue-green algae that increases neuronal activity in response to blue light. The sensitized neurons could then be turned on or off with the literal flip of a light switch, allowing the scientists to watch what happens to the behavior of an individual mouse.

Remarkably, says Anderson, for mice in which the injection was targeted to the correct location, blue light induced an attack—even toward an inanimate object such as an inflated latex glove. Conversely, using a technique developed by Caltech's Bren Professor of Biology Henry A. Lester that allowed the scientists to genetically inhibit neuronal activity, Lin and colleagues were able to show that neuronal activity in the VMH was necessary for normal aggressive behavior, as well.

"This answers an important, long-standing question in the field," says Anderson. "Are regions of the brain that can evoke aggression when artificially stimulated actually necessary for normal aggressive behavior? In this case, the answer is clearly 'yes.'"

The researchers also found that stimulating a male to be aggressive toward a female became more difficult as a mating encounter progressed to its consummatory phase. This result was consistent with the observation that neurons activated during fighting appear to become inhibited in the presence of a female. "The question," says Anderson, "is how that inhibition is achieved."

The answer may lead to new areas of research—and, perhaps, to new treatments for impulsive, violent behaviors. Specifically, notes Anderson, scientists can begin thinking about treatments that target violence-begetting neurons while sparing those involved in normal sexual behavior.

"For the last 500 years, we've really had no viable treatments for pathological violence other than execution or imprisonment," says Anderson. "And part of the reason is that we haven't understood enough about the basic neurobiology of aggression. The new studies are an important step in that direction."

In addition, he says, "mapping out the brain circuitry of aggression will provide a framework for understanding where and how in the brain genetic and environmental influences—nature vs. nurture—exert their influences on aggressive behavior."

The other authors on the Nature paper, "Functional identification of an aggression locus in the mouse hypothalamus," in addition to Anderson, Lin, Dollar, and Perona, are Caltech postdoctoral scholar Hyosang Lee and Maureen Boyle and Ed Lein from the Allen Institute for Brain Science in Seattle.

Their work was funded by the Weston-Havens Foundation, the Jane Coffin Childs Foundation, and the Howard Hughes Medical Institute.