Read our COVID-19 research and news.

Using blue laser light, scientists manipulated neural circuits connecting the hippocampus and amygdala (regions in red) to transform negative memories into good ones.

Using blue laser light, scientists manipulated neural circuits connecting the hippocampus and amygdala (regions in red) to transform negative memories into good ones.

Adapted from Redondo et al., Nature (2014)

Researchers erase fearful memories in mice

By manipulating brain activity in mice with light, a research team has neutralized “fearful” memories in the rodents and even colored them with more positive emotions. Although the technique is too invasive to use on humans, researchers hope it will lead to future treatments for post-traumatic stress and anxiety disorders.

Our memory of the past is constantly under revision as we learn from new experiences, says Susumu Tonegawa, a neuroscientist at the Massachusetts Institute of Technology in Cambridge. In some cases, the natural malleability of memory can help people recover from trauma—therapists, for example, often ask patients to repeatedly confront their fears while experiencing a physical distraction such as rhythmic tapping, or summoning happier thoughts to counteract negative emotions. Such treatments do seem to alleviate suffering for some people, and researchers have long hypothesized that they might work by remodeling memory circuits in the brain to incorporate new, positive input. Until recently, however, there has been little scientific evidence to support that hypothesis, Tonegawa says.

To manipulate memory on a neuronal level, he and his colleagues use optogenetics, a tool that allows scientists to trigger neurons that have been genetically engineered to fire when light shines on them. The lab has also added a twist to the technology—"a technical tour de force" that allows the team to trace specific memories as they form, says Loren Frank, a neuroscientist at the University of California, San Francisco. In the new method, mice are genetically engineered to express a light-sensitive protein in neurons, or are injected with an inert virus that delivers the protein to specific brain regions. The protein is expressed only when a neuron fires, and only if the animals have been taken off a specific antibiotic. Using this technique, the team can briefly switch expression of the protein on or off so that it tags only those cells involved in encoding specific memories.

In 2013, Tonegawa and colleagues used the method to instill a fake memory in mice. In the experiment, published in Science, they traced two separate memories—one in the hippocampus, which encodes spatial aspects of memory, and the other in the amygdala, which assigns emotional value to events as we experience them. Then they used optogenetics to make both sets of cells fire together. Activating hippocampal neurons associated with a familiar cage and the amygdala neurons that fired when the mouse received a shock in the foot tricked the mice into later behaving as though they had been shocked in the cage, even though they had never been injured there.

In the new study, reported online this week in Nature, Tonegawa’s team decided to see if it could change bad memories into good ones. First, the researchers gave another group of male mice a mild electrical shock, then put them in a rectangular box where they could explore freely. When the rodents wandered into one side of the box, a laser implanted into their brains switched on and reactivated the cells that formed the memory of being shocked. Even though the danger was all in their heads, the mice still quickly learned to avoid that side of the box.

Next, the researchers took the mice out of the enclosure and provided what they consider to be a fail-safe source of rodent joy: contact with the opposite sex. For male mice never exposed to females before, "it's a big deal," says Roger Redondo, a postdoctoral student in Tonegawa's lab and lead author of the new work. "They go crazy." As the mice grew accustomed to their change in fortune, the team reactivated their memories of being shocked with the laser beam. They wanted to see if stimulating neurons involved in forming the original fearful memory would make them more susceptible to positive input from new experiences, or simply make the animals afraid of females. 

The mice did not develop a sudden phobia of the opposite sex—in fact, quite the opposite occurred, Redondo says. To see if the rodents’ old memories of fear had been supplanted by more agreeable associations, the group placed the rodents back in the enclosure and flipped the laser on again. Now, mice who had avoided the laser firing zone before sniffed around their enclosures looking for their former companions. They even seemed drawn to the laser zone, suggesting that the light stimulation might be triggering pleasant reminiscences.

Additional experiments showed that happy memories could also be soured if they were reactivated at the same time the mice were being shocked. Interestingly, the switch works only when applied to neurons in the hippocampus, and not the amygdala, Tonegawa says. Based on that finding and microscopic studies of the rodents' brain tissue, Tonegawa believes that there are two types of amygdala cell populations, one involved in fear memories and the other involved in positive memories. Hippocampal neurons appear to drive which of the two populations wins out, he says. Although it isn't possible to put optic fibers directly into the human brain, the connection between the two regions could be the target for drugs or other therapies, he suggests.

 It's unlikely that reactivating such a narrow subset of cells in mice truly triggers the kind of rich, sequential memory that allows us to remember which route we took to get to work and what we ate for breakfast, Frank notes. Such distinctions are important when attempting to apply the results to disorders such as post-traumatic stress, he says. Still, "this is first time that people have shown that you can activate a mental representation artificially and change what it’s connected to in the brain," he says. "It's a really impressive, massive step forward for the field."