A common brain pathway for anxiety and social behavior

Impaired social interaction is a common feature in autism, schizophrenia, depression, and anxiety, and it contributes to many of the problems that people with these conditions face. That is particularly true for adolescents with autism spectrum disorder, of whom about 40 percent are also diagnosed with anxiety.

A new study from Kay Tye’s laboratory at MIT found a circuit in the brain that might explain the link between impaired social interaction and anxiety in so many disorders. The circuit connects the amygdala, well known for its role in anxiety, with the hippocampus, important for learning, memory, and emotional responses.

Recently, the Tye Lab found that a discrete circuit connecting a subregion of the amygdala (the basolateral amygdala, or BLA) with the ventral hippocampus (vHPC) controlled anxiety. Activating it increased anxiety; inhibiting it decreased anxiety. In the latest study, the lab focused on this same circuit’s ability to modulate social behavior. Both studies were led by research associate Ada Felix-Ortiz.

“It’s exciting, because we found the actual locus in the brain of one population of synapses connecting these two regions that drive both social and anxiety behaviors,” says Tye, assistant professor in the Department of Brain and Cognitive Sciences and a principal investigator at the Picower Institute for Learning and Memory at MIT. “It provides insight on how a common brain circuit could control anxiety and social interaction.”

Research on social anxiety and brain disorders has often focused on changes in the amygdala, which underlies positive and negative emotions including fear, anxiety, motivation, reward processing, and sociability. The amygdala’s diverse role results from its “promiscuous” connections to many other brain regions, explains Tye, who has been working to disentangle these connections to understand each projection’s specific effect on behavior. That is important, because projections from the amygdala to two different brain regions can have opposing effects on anxiety in mice.

“This finding illustrates very elegantly how complex interactions between different parts of the brain, which are important for a range of complex behaviors, can be dissected and controlled,” says Amit Etkin, assistant professor of psychiatry and behavioral sciences at Stanford University, who was not involved with the study. “By understanding how the amygdala talks with a specific part of the hippocampus, the researchers were able to both increase and decrease social behavior in mice.”

To accomplish this, Tye applied optogenetics, the technique of activating or inhibiting specific neuron types with pulses of light, to manipulate neurons not where they live in the amygdala, as is typical, but where they reach out and touch other neurons in the hippocampus. Some of the test mice were expressing light-activated proteins in the BLA neurons. BLA neurons expressing channelrhodopsin-2 (ChR2) were activated when exposed to blue light via chronically implanted fiber optics in the vHPC. Neurons expressing halorhodopsin (NpHR) were inactivated when exposed to yellow light. (Control mice did not express a light-sensitive protein.)

To study social interactions in mice, Tye used two well-validated behavioral procedures. In the resident-intruder home-cage test, she measured the behavior of a test mouse (an adult male) accustomed to his home cage before and after a strange mouse (a juvenile male) was introduced to the cage. In the three-chamber sociability test, the intruder mouse is contained in a cup in one of the chambers and so cannot proactively initiate social interaction.

In both behavior tests, inhibiting the BLA-vHPC circuit (in NpHR mice) increased social behavior. Compared to control mice, home-cage mice became more interested in the strange intruder, spending more time sniffing, nuzzling, and closely following the novel stranger and less time exploring their own familiar cage. In contrast, activating the same circuit (in ChR2 mice) made the resident indifferent to the stranger, even avoiding the chamber with the stranger in the three-chamber cage. The effects did not impact other behaviors such as mobility or freezing (signs of locomotor activity and acute fear).

Significantly, the test mice with the activated BLA-vHPC circuit spent much more time self-grooming. “An increase in grooming is a primary characteristic in many mouse models of obsessive-compulsive behavior,” Tye says. “Maybe if you have social anxiety, an intruder might make you feel uncomfortable and lead to stereotypical compulsive behavior.”

A broad implication of this study concerns the growing understanding that autism, schizophrenia, and a range of psychiatric disorders are actually heterogeneous groupings of many subtypes, which helps explain why one treatment does not work for all people with the same diagnosis. If, for example, people with both autism and anxiety disorders have a distinct brain circuitry compared to those without overlapping diagnoses, they might benefit from a separate treatment that specifically targets that circuit.

“By combining insights from this study with the knowledge of there being amygdala changes in a range of psychiatric conditions, we gain a deeper understanding of how particular brain circuits underlie disruptions in social interactions common across mental illness,” Etkin says. “In this way, these findings also describe a potential target for interventions aimed at restoring normal social interactions, which would not be possible to the same degree without this beautiful study.”

The study was the featured article in the Jan. 8 Journal of Neuroscience.