Stress can disrupt memory and lead to needless anxiety — here’s how

Stress makes mice form big bundles of neurons in the brain that disrupt memory formation, making them fearful of harmless situations1 — which might help to explain why stressed people often feel threatened in safe environments.
Researchers have long known that stress or trauma can lead people to fear harmless situations. For instance, after burning a finger on a hot pan, a stressed individual might subsequently avoid not only hot pans but the kitchen or cooking entirely. This kind of generalized fear is common in people with post-traumatic stress disorder (PTSD) and generalized anxiety disorder.
A study, published in Cell, describes how stress disrupts memory formation and, in particular, recollections of fearful events. The results could inform the development of therapies for people with PTSD and anxiety.
“This paper is really a tour-de-force,” says Ryuichi Shigemoto, a neuroscientist at the Institute of Science and Technology Austria, in Klosterneuburg. “They used so many different methods and techniques to prove this long pathway.”
Memories are packaged into groups of neurons, called engrams, which are active when a memory is being formed. Sheena Josselyn, a neuroscientist at the Hospital for Sick Children in Toronto, Canada, and her colleagues looked at whether stress disrupts engram formation and focused on a region of the brain called the amygdala, which is involved in stress and emotion response.
The study involved an elaborate three-step experiment in mice. First, they put some adult mice in a stressed state by injecting them with the stress hormone corticosterone or restraining them in a small tube for 30 minutes, which increased their corticosterone levels.
They then placed mice — some stressed, and others not — in a chamber and played a medium-pitched sound for 30 seconds, a neutral event. After a break, the mice went back into the chamber and experienced a high-pitched whistling sound for 30 seconds, which ended with a 2-second shock to the foot, to mimic a fearful event.
To test how the mice had stored the memories of these experiences, the researchers put the mice in a new environment and played the two tones — watching for how they responded.
The unstressed mice froze mostly when they heard the high-pitched whistling, whereas the stressed mice froze in response to both sounds, suggesting that they couldn’t distinguish between the neutral and fearful events.
The researchers used various techniques to visualize neural activity in the rodents. They found that, during memory formation, the unstressed mice formed small engrams in response to the whistle and foot shock, and these were only reactivated when exposed to the whistle. But the stressed mice formed larger engrams, which were reactivated when exposed to both sounds.
Further experiments uncovered the chain of events in the brain that created the larger engram in stressed mice. Under normal conditions, specific neurons in the amygdala block neuronal activity through the release of chemical messengers known as gamma-aminobutyric acid (GABA). This ensures that a small engram is created in response to a negative memory. “It’s sort of like the velvet rope at a nightclub: it only lets certain neurons into the nightclub,” says Josselyn. But under stress, excitatory neurons pump the brain with a neurotransmitter known as endocannabinoid, which binds to glucocorticoid receptors on those inhibitory neurons and prevents them from releasing GABA, resulting in larger engrams. In other words, the velvet rope drops, “and many neurons can get into this exclusive club”, says Josselyn.
The team were able to reverse the effects of stress on memory formation with two drugs, one of which is approved for terminating early pregnancy, mifepristone. The drugs either block the glucocorticoid receptors or the production of endocannabinoids, and the stressed mice recalled memories in the way that unstressed mice did. But researchers caution that the medications have side effects beyond the brain, and work only if they are given at the time the memory is formed, so are unlikely to be useful in people.
Josselyn and her colleagues are now trying to investigate whether engrams can be altered after a memory has formed, or whether there are other ways of mitigating the effects of stress on memory.