Making memories

Researchers from The Scripps Research Institute have revealed the role of protein synthesis in encoding long-term memories — specifically, memories associated with fear of a specific environmental cue.

Associate Professor Sathyanarayanan V Puthanveettil, who led the research, explained that his team was looking into a subtype of contextual memory — that is, the memory of an experience in a specific context. “For example, if you have an exciting experience in a specific office room or place, you tend to recall that experience whenever you visit that room or place,” he said.

“In the case of contextual fear memory, one associates a fearful experience with a specific context. Because contextual fear memories are long-lasting, scientists intensely study them to understand the mechanisms of memory storage and retrieval, as well as memory loss.”

Writing in the journal Biological Psychiatry Cognitive Neuroscience and Neuroimaging, Associate Professor Puthanveettil and his colleagues said they were investigating protein synthesis — a requisite for the consolidation and storage of long-term memories — in the area of the medial prefrontal cortex (mPFC) known in rodents as the prelimbic (PL) cortex and in humans as the anterior cortex, which has been linked to processing emotional responses. While previous studies had suggested that the mPFC plays an important role in the consolidation of long-term memories, it was unknown whether it also played a role in encoding these memories.

The researchers sought to answer this question by employing contextual fear conditioning (CFC) in rodents, whereby an adult mouse would be put inside a fear-conditioning chamber and given three mild foot shocks. Associate Professor Puthanveettil explained, “The mice associate the memory of the foot shocks to the fear-conditioning chamber and exhibit a freezing behaviour in response to foot shocks.

“The next day, the mouse is put back into the fear-conditioning chamber and the animal will show freezing behaviour in the absence of foot shocks,” Associate Professor Puthanveettil continued. “This is because the mouse associates the foot shock experience to the fear-conditioning chamber.”

Following this contextual fear conditioning, the scientists imaged the mice’s brains in order to assess the changes in polyribosome-associated messenger RNAs (mRNAs) in the mPFC. “We identified several mRNAs that are differentially and temporally recruited to polyribosomes in the mPFC,” they said — a clear indication that this was where protein synthesis had taken place.

The researchers then went one step further, injecting the mice with an inhibitor called anisomycin (ANI) that would block new protein synthesis in the PL cortex. After injecting the mice immediately post-fear conditioning, then testing them 24 hours later, it was clear that the mice were no longer freezing once entering the fear-conditioning chamber — the memories had thus not taken hold.

But the researchers weren’t done yet, as they now wondered whether continuous protein synthesis in the PL cortex is required for encoding contextual fear memory. So rather than injecting the inhibitor straight away, the researchers waited until six hours after CFC training. The mice’s responses were now comparable to that of the placebo group, indicating that mPFC protein synthesis was only required for encoding in the immediate aftermath of the fear conditioning.

“There is [therefore] a specific time window during encoding that is sensitive to the disruption of protein synthesis in the PL cortex,” the scientists said.

Finally, the researchers analysed the expression of the protein Homer3, which is located in the PL cortex and whose mRNAs were found to be enriched in polyribosomes within the first hour following CFC training. “This suggested that Homer3 was produced in response to the training and therefore Homer3 plays a critical role in contextual fear memory storage,” Associate Professor Puthanveettil said.

Seeking to study the consequences of reducing Homer3 protein levels in neurons, the scientists injected ANI into the PL cortex, confirming that the inhibitor caused a significant decrease in Homer3 mRNA loaded onto polyribosomes. Next, they knocked down Homer3 mRNA in the PL cortex using small interfering RNAs (siRNAs). When tested 24 hours after training with CFC, mice injected with Homer3 siRNA exhibited reduced fear compared with the control group.

“We found that more of Homer3 proteins are produced during contextual fear memory storage and that reduction in Homer3 protein levels in neurons impedes formation of fear memory,” Associate Professor Puthanveettil said. “Thus, Homer3 has a critical role in contextual fear memory storage.”

The scientists had succeeded in identifying several molecular substrates of new protein synthesis in the mPFC and established that encoding of contextual fear memories requires new protein synthesis in the PL subregion of the mPFC. They therefore concluded that, in addition to the mPFC’s role in long-term memory storage, it also has an early role in memory consolidation and encoding contextual fear memories.

“This is very important and exciting because despite our understanding that production of new proteins is critical for memory storage, very few of newly produced proteins are known,” said Associate Professor Puthanveettil.

The researchers added that it remains to be determined if other subregions of the cortex are also involved in the synthesis of memory proteins. As noted by study co-author Bindu L Raveendra, “The specific roles of these subregions in encoding, expression and retrieval, as well as their underlying molecular mechanisms, remain to be unravelled.”

Image credit: ©stock.adobe.com/au/agsandrew