The Anxious Circuit

Anxiety is a feeling we all face on a daily basis about our jobs, our relationships, and even the meaning of our lives. But when normal anxiety gets so severe that it interferes with daily functioning it becomes generalized anxiety disorder, a psychiatric diagnosis that 28% of people in the United States will suffer from during some period in their life, costing the economy billions of dollars annually.

But where is the seat of anxiety in the brain? In this article (Tae et al. 2011 Nature we will explore one of the first studies to ever identify a possible circuit for this important and pervasive feeling, which is not surprisingly located in the same prominent egg-shaped loci in the center of brain that control fear responses: the two amygdala.

The amygdala are made up of nuclei that have been discovered to be active in a variety of behavioral tasks related to emotional processing of experiences and sensory inputs. Most of our reaction to experience is categorized as “good” or “bad” by us, and this is an important survival mechanism (something that smells rotten disgusts us so we don’t eat it, and that’s generally great for us). We form memories, habits and reactions, and thoughts based in part on connections formed in the amygdala. This means that the amygdala are crucially important to the way we approach the world; understanding the amygdala can help us understand our own perception, in a way. More recently, one important study demonstrated a role for the amygdala beyond its classically understood role of assigning emotional meaning to stimuli we encounter at a given moment. This particular study added to the list of functions associated with the amygdala by demonstrating its role in general states of anxiety, not associated with any one set of stimuli (i.e. “unconditioned”).

Figure 1, taken from Tae et al. 2011  Nature

Figure 1, taken from Tae et al. 2011 Nature

The amygdala is made up of several nuclei, the most relevant of which are the positionally-named lateral, basolateral (BLA), and central nuclei (CeA). Roughly, the lateral nuclei receive sensory input which is forwarded to the basolateral nuclei, which form excitatory synapses onto the central nuclei, which go on to inhibit a diverse set of brain regions, some of which activate the fear response. Elegantly employing optogenetics for the first time in a moving animal, Tye et al. of the Deisseroth group discovered that manipulating a specific set of neuronal connections between the basolateral (BLA) and central (CeA, which is made up of the CeL and CeM cell-body clusters) nuclei resulted in anxiety-related behavioral changes.

To measure these changes, they took advantage of the fact that mice anxiously avoid open spaces. They examined mice by using two classic behavioral tests, that both take advantage of the phenomenon that mice are a bit fearful of open spaces, presumably because being in an open space makes them vulnerable targets for prey and other dangers in the wild. The first such test was an elevated-plus maze task: here they measured the time spent in the open arms of the maze. The second test was an open-field test, where they recorded the position of the mouse within a square room. In both tests, avoidance of the open space (measured as the amount of time spent outside  the open space) was used as a measure of anxiety.

The researchers first wanted to see if they could identify which neuronal circuits in the amygdala are responsible for anxious responses. They hypothesized that the BLA-CeA projections might be important so the researchers first compared the effects of stimulating one set of BLA synaptic projections into CeA dendrites to (as a negative control) generally stimulating BLA neurons whether or not they project onto the CeA (figure 1). They found that photostimulation of BLA terminals into the CeA resulted in behavioral metrics associated with less anxiety, including more open-arm and center-field times in the maze and field tests, respectively.

Next they attempted to determine the underlying circuit causing decreased anxious behavior in stimulated mice by observing which of the two nuclei in the CeA, the CeL (CeA Lateral part) or the CeM (CeA Medial part), were receiving excitatory input from the BLA. By tracking gene expression of the activity-dependent immediate early gene (c-fos), the researchers were able to identify which groups of the neurons in the CeA were activated during BLA terminal stimulation and found that CeL neurons showed increased activity, and thus CeL was directly downstream of BLA. To investigate the second step of their hypothesis, they authors took mouse brain-slices and directly recorded the activity of the cells in the CeL and CeM, finding that the CeL inhibited the CeM. Thus, there exists a BLA-CeL-CeM circuit, where information flows from the former to the latter.

These two pieces of evidence combined were not enough evidence to say for certain that the BLA-CeL-CeM circuit works as the authors hypothesized to modulate levels of general anxiety, because the optogenetics setup in this study only allows for the testing of one type of connection at at time. The authors did a series of control experiments where they showed they could abolish the effect by selectively blocking the excitatory neurotransmitter glutamate at the BLA-CeL junction or using the same photostimulation experiment with an optogenetic channel that inhibited the BLA-CeL link instead of exciting it. To sum it all up, these results suggest that the BLA-CeL excitatory synapse and the downstream CeL-CeM inhibitory synapses are together responsible for modulating general anxiety in the amygdala.

This study serves a good example of the use of simple, clear experiment coupled with rigorous controls to demonstrate a hypothesis. The implications of the study’s findings are also clinically relevant to our understanding of generalized anxiety disorder, which plagues an increasing millions of Americans every year.


Paper Link:

Amygdala circuitry mediating reversible and bidirectional control of anxiety

Kay M. Tye, Rohit Prakash, Sung-Yon Kim, Lief E. Fenno, Logan Grosenick, Hosniya Zarabi, Kimberly R. Thompson, Viviana Gradinaru, Charu Ramakrishnan & Karl Deisseroth

Useful review article: