Stress makes mice’s memories less specific

Now, neuroscientists report November 15 in the Cell Press journal Cell that acute stress prevents mice from forming specific memories. Instead, the stressed mice formed generalized memories, which are encoded by larger numbers of neurons.

“We are now beginning to really understand how stress impacts aversive memories, and I think that’s good news for everybody,” says memory researcher and co-senior author Sheena Josselyn of The Hospital for Sick Children (SickKids) and the University of Toronto. “We were able to isolate the synaptic mechanisms that drove this and also show that this same phenomenon can be manipulated or blocked by using systemically available drugs.”

To test whether stress impacts memory specificity, the researchers trained mice to associate one sound with stress, and another sound with no stress. Then, they tested the mice’s ability to react appropriately to the different sounds.

Mice which had been placed in an acutely stressful, controlled experience exhibited defensive behavior regardless of which sound was played to them, suggesting that the stressful experience interfered with their ability to form specific memories. In contrast, control mice who had not been subjected to stress exhibited defensive freezing only in response to original sound.

Because the stressed mice had elevated levels of corticosterone in their blood, the researchers next tested whether corticosterone itself could impact memory formation. They showed that mice that received corticosterone prior to training were also unable to form specific memories to the two sounds, and that administering metyrapone, a chemical that inhibits glucocorticoid synthesis, restored the ability of stressed mice to form specific memories.

Specific memories are encoded by groups of neurons called engrams. Most engrams involve only a few neurons, but the researchers showed that the generalized engrams formed by stressed mice were larger, because inhibitory interneurons — gatekeeping cells that usually keep engrams exclusive — failed to do their job. This change, in turn, was driven by endocannabinoids that were released in the amygdala in response to corticosterone.

“When we manipulated endocannabinoid receptors in just one particular cell type in one brain region, it restored memory specificity and the size of the engram,” says stress researcher and co-senior author Matthew Hill of the University of Calgary. “This whole phenomenon is mediated by a very discrete microcircuit in the amygdala, but you can do a systemic pharmacological manipulation and still prevent it, which is very encouraging from the perspective of whether this could one day be translated for therapeutic use in humans.”

In future, the researchers want to investigate whether stress also impacts the specificity of non-aversive memories. They also plan to examine whether exogenous cannabinoids (e.g., cannabis) would have a similar effect on memory specificity, which could have implications for PTSD management.

“We only examined aversive threat memories, but it would be interesting to examine whether stress similarly increases the generalization of a rewarding memories,” says memory researcher and co-senior author Paul Frankland, also at SickKids and the University of Toronto.

“Given that this phenomenon involved the activation of endocannabinoid receptors, it would be very interesting to see if a stoned animal shows a similar generalization response,” says Hill. “That’s one of the things that I’d be curious to quickly run as a follow up, because if it did, that would have some interesting implications given that the whole conversation that exists right now around cannabis and PTSD is very confusing.”

This research was supported by the Dutch Research Council, Niels Stensen Fellowship, ZonMw Memorabel, Canadian Institutes of Health Research, Alzheimer Nederland, Toronto Cannabis and Cannabinoid Research Consortium, and the Brain Canada Foundation.