Our brain's remarkable ability to form and store memories has long fascinated scientists, yet most of the microscopic mechanisms behind memory and learning processes remain a mystery. Recent research points to the importance of biochemical reactions occurring at postsynaptic densities—specialized areas where neurons connect and communicate. These tiny junctions between brain cells are now thought to be crucial sites where proteins need to organize in specific ways to facilitate learning and memory formation.
More specifically, a 2021 study revealed that memory-related proteins can bind together to form droplet-like structures at postsynaptic densities. What makes these structures particularly intriguing is their unique "droplet-inside-droplet" organization, which scientists believe may be fundamental to how our brains create lasting memories. However, understanding exactly how and why such complex protein arrangements form has remained a significant challenge in neuroscience.
Against this backdrop, a research team led by Researcher Vikas Pandey from the International Center for Brain Science (ICBS), Fujita Health University, Japan, has developed an innovative computational model that reproduces these intricate protein structures. Their paper, published online in Cell Reports on April 07, 2025, explores the mechanisms behind the formation of multilayered protein condensates. The study was co-authored by Dr. Tomohisa Hosokawa and Dr. Yasunori Hayashi from the Department of Pharmacology, Kyoto University Graduate School of Medicine, and Dr. Hidetoshi Urakubo from ICBS, Fujita Health University.
The researchers focused on four proteins found at synapses, with special attention to Ca²⁺/calmodulin-dependent protein kinase II (CaMKII)—a protein particularly abundant in postsynaptic densities. Using computational modeling techniques, they simulated how these proteins interact and organize themselves under various conditions. Their model successfully reproduced the formation of the above-mentioned "droplet-inside-droplet" structures observed in earlier experiments. Through simulations and detailed analyses of the physical forces and chemical interactions involved, the research team shed light on a process called liquid-liquid phase separation (LLPS); it involves proteins spontaneously organizing into condensates without membranes that sometimes resemble the organelles found inside cells.
Crucially, the researchers found that the distinctive "droplet-inside-droplet" structure appears as a result of competitive binding between the proteins and is significantly influenced by the shape of CaMKII, specifically its high valency (number of binding sites) and short linker length. These shape-related properties of CaMKII result in low surface tension and slow diffusion, allowing the protein condensates to remain stable for extended periods. This stability enables the sustained activation of downstream signaling pathways necessary for synaptic plasticity, which is the cellular basis for learning and memory. "Our results revealed new structure–function relationships for CaMKII as a synaptic memory unit. This is the first systematic and mechanistic study investigating the divergent structure of protein-regulated multiphase condensates," highlights Dr. Pandey.
These findings could pave the way toward a better understanding of the possible mechanisms of memory formation in humans. However, the long-term implications of this research extend well beyond basic neuroscience.
Defects in synapse formation have been associated with numerous neurological and mental health conditions, including schizophrenia, autism spectrum disorders, Down syndrome, and Rett syndrome. "Overall, the computational model developed in this study could serve as an important platform for investigating these conditions, potentially leading to new diagnostic tools and therapeutic approaches," explains Dr. Pandey.
Let us hope scientists continue to unravel the mysteries of how memories form at the molecular level, leading us to a more thorough comprehension of one of the brain's most fundamental and complex functions.
