A process thought to destroy brain cells might actually help them store data

Recent research provides evidence that the nervous system actively promotes the formation of amyloid structures to stabilize long-term memories. While amyloids are often associated with neurodegenerative conditions, this study identifies a specific protein chaperone that drives the creation of beneficial amyloids in response to sensory experiences. These findings, which offer a new perspective on how the brain encodes information, were published in the Proceedings of the National Academy of Sciences.

Scientists have studied the biological basis of memory for decades. A prevailing model posits that long-term memory requires the physical alteration of synapses, the connections between neurons. This process involves changes in the proteins located at these synapses.

One specific protein, known as Orb2 in fruit flies, plays a central role in this process. Orb2 creates a stable memory trace by self-assembling into an amyloid, a tight stack of proteins that is durable and self-perpetuating.

Most research on amyloids focuses on their toxic role in diseases such as Alzheimer’s. In those contexts, proteins misfold and aggregate in ways that damage cells. However, the brain appears to use a similar aggregation mechanism for beneficial purposes. The question remained regarding how the brain ensures that Orb2 forms amyloids only when a memory needs to be stored and not at random times.

A research team led by Kyle Patton investigated the regulatory systems that might control this precise timing. They hypothesized that molecular chaperones, which are proteins that assist others in folding or assembling, might be responsible for this regulation.

Free daily newsletter

To identify the specific molecules involved, the researchers focused on the J-domain protein (JDP) family. This is a diverse group of chaperones known to regulate protein states. The team utilized Drosophila melanogaster, the common fruit fly, as their model organism. They examined 46 different JDPs found in the fly genome. The team narrowed their search to chaperones expressed in the mushroom body, a brain structure in insects that is essential for learning and memory.

The researchers conducted a genetic screen to determine which of these chaperones influenced memory retention. They used a classical conditioning experiment known as an associative appetitive memory paradigm. In this procedure, the researchers starved flies for a short period to motivate them. They then exposed the flies to two different odors. One odor was paired with a sugar reward, while the other was not. After training, the flies were given a choice between the two odors.

Most wild-type flies remember which odor predicts food for a certain period. The researchers genetically modified groups of flies to overexpress specific JDPs in their mushroom body neurons. They found that increasing the levels of one specific chaperone, named CG10375, significantly enhanced the flies’ ability to form long-term memories. The researchers named this protein “Funes,” inspired by a fictional character with the inability to forget.

The study showed that flies with elevated levels of Funes remembered the association between the odor and the sugar for much longer than control flies. This effect was specific to long-term memory. Short-term memory, which operates through different molecular mechanisms, appeared unaffected. This suggests that Funes plays a distinct role in the consolidation phase of memory storage.

To verify that Funes is necessary for memory—and not just a booster when artificially added—the team performed the reverse experiment. They used genetic tools to reduce the natural levels of Funes in the fly brain or to create mutations in the Funes gene.

Flies with reduced Funes activity were capable of learning the task initially. However, they failed to retain the memory 24 hours later. This indicates that Funes is an essential component of the natural machinery required for memory stabilization.

The researchers next investigated how Funes interacts with sensory information. Memory formation usually depends on the intensity of the experience. For example, a strong sugary reward creates a stronger memory than a weak one. The team tested Funes-overexpressing flies with lower concentrations of sugar and weaker odors.

Remarkably, flies with extra Funes formed robust memories even when the sensory cues were suboptimal. They learned effectively with much less sugar than typical flies required. This finding suggests that Funes helps signal the nutritional value or “salience” of the experience. It acts as a sensitizing agent, allowing the brain to encode memories of events that might otherwise be too faint to trigger long-term storage.

Following the behavioral tests, the researchers explored the molecular mechanism at play. They suspected that Funes acted by influencing Orb2, the memory protein known to form amyloids. They performed biochemical experiments to see if the two proteins interacted physically.

The results showed that Funes binds directly to Orb2. Specifically, it binds to Orb2 when it is in an oligomeric state, which is an intermediate stage between a single molecule and a full amyloid fiber.

The team then reconstituted the reaction in a test tube to observe it directly. They purified Funes and Orb2 proteins and mixed them in a controlled environment. When mixed, Funes accelerated the transition of Orb2 from these intermediate clusters into long, stable amyloid filaments. The researchers confirmed the presence of these structures using an amyloid-binding dye called Thioflavin T, which fluoresces when it attaches to amyloid fibers.

To ensure these laboratory-created fibers were the same as those found in living brains, the team utilized cryogenic electron microscopy (cryo-EM). This advanced imaging technique allows scientists to see the atomic structure of proteins.

The images revealed that the Orb2 amyloids created with the help of Funes were structurally identical to endogenous Orb2 amyloids extracted from fly heads. They possessed the same “cross-beta” architecture that characterizes functional amyloids.

The study further demonstrated that the “J-domain” of the Funes protein is essential for this activity. This domain is a specific section of the protein sequence that defines the JDP family.

The researchers generated a mutant version of Funes with a slight alteration in the J-domain. This mutant was able to bind to Orb2 but could not push it to form the final amyloid structure. When this mutant version was expressed in flies, it failed to enhance memory, confirming that the physical formation of the amyloid is the key to the memory-boosting effect.

Beyond structural formation, the researchers verified that these Funes-induced amyloids were functionally active. In the brain, Orb2 amyloids work by binding to specific messenger RNAs (mRNAs) and regulating their translation into new proteins.

The researchers used a reporter assay to measure this activity. They found that the amyloids facilitated by Funes successfully promoted the translation of target mRNAs, mimicking the natural biological process seen in memory consolidation.

One potential limitation of this study is its focus on Drosophila. While the fundamental molecular machinery of memory is highly conserved across species, it remains to be seen if a direct homolog of Funes performs the exact same function in mammals.

The human genome contains many J-domain proteins, and identifying which one corresponds functionally to Funes will be a necessary next step. The study suggests a link to human health, noting that some related chaperones have been genetically associated with schizophrenia, a condition that involves cognitive deficits.

Future research will likely investigate how Funes receives the signal to act. The current study shows that Funes responds to nutritional cues, but the precise signaling pathway that activates Funes remains to be mapped. Additionally, scientists will need to determine if Funes regulates other proteins beside Orb2. It is possible that this chaperone manages a suite of proteins required for synaptic plasticity.

This work challenges the traditional view that amyloid formation is merely a pathological accident. It provides evidence that the brain has evolved sophisticated machinery to harness these stable structures for information storage. By identifying Funes, the researchers have pinpointed a control switch for this process, offering a potential target for understanding how memories persist over a lifetime.

The study, “A J-domain protein enhances memory by promoting physiological amyloid formation in Drosophila,” was authored by Kyle Patton, Yangyang Yi, Raj Burt, Kevin Kan-Shing Ng, Mayur Mukhi, Peerzada Shariq Shaheen Khaki, Ruben Hervas, and Kausik Si.