Date published: 4/2/2018

We are studying critical survival responses to evade life-threatening danger, such as asphyxiation. Unfortunately, the circuits that control escape to these threats can also induce panic attacks. We investigate the mechanisms underlying both adaptive escape as well as maladaptive panic attacks.

It is estimated that 2.7% of adults in the USA have suffer from panic disorder every year. The core symptom of this disorder is the occurrence of repeated panic attacks, which can induce a catastrophic fear of death, pounding heart rate, among other symptoms. These extremely debilitating episodes can take a severe toll on patients. However, current therapies produce numerous side-effects and are often ineffective, pointing towards the need to better understand the circuits that produce panic.
Panic is typically commonly studied through escape behaviors in rodents, as panicolytic drugs decrease escape, stimulation of panic-inducing brain regions in humans induce escape in rodents, and stimuli that cause panic in humans (such as CO2 inhalation) induce escape. As the brain regions that control these key survival behaviors are evolutionarily conserved, studying the mechanisms underlying threat-induced escape in rodents reveal powerful insights into the neurobiology of panic attacks. Though escaping is not a common symptom in humans during panic attacks, it is noteworthy that an overwhelming urge to flee is often reported in these episodes.
The brain region most strongly implicated in the induction of panic is the periaqueductal gray (PAG). Indeed, electrical stimulation of the PAG in humans can cause a constellation of symptoms closely resembling panic attacks, and panic disorder patients have enlarged and hyperactive PAG relative to healthy controls. In rodents this region, among other functions, controls rapid escape reactions to intense threats, such as suffocation or imminent predatory attacks.  
Although the PAG is known to be a key node in the induction of escape and panic, it is unknown how different populations in the PAG modulate various panic symptoms. The PAG has many different types of cells, which produce distinct neurotransmitters and have unique anatomical features. We will characterize how distinct naturalistic panic-inducing threats activate these various populations of cells and which specific panic symptoms each population controls. We are also interested in studying if early-life stress hyper-sensitizes cells in the PAG, leading to disproportionately large responses to mild threats. To perform these studies we use a combination of modern system neuroscience techniques to observe and perturb neural activity in mice exposed to safe or panic-provoking situations, causally linking activation of discrete populations of brain cells with behavioral manifestations related to panic.