When Renée Kosonen describes tau, she calls it an organizer. It's a simple word, but it explains something that has puzzled scientists for decades: why some memories last while others fade into nothing.
Now, a team of researchers from Flinders University in Adelaide, Australia, has shown that tau — a protein most famous for its link to Alzheimer's disease — is actually essential for creating memories that stick around. Their study, published in the journal Nature Communications, offers a surprising new picture of how healthy memory works.
"Our findings show that tau helps determine which cells are selected to store a memory, shaping how an experience forms a lasting memory trace," says Ms. Kosonen, a researcher at Flinders' Neuroscience and Dementia Research.
The researchers focused on tiny brain cells called engram cells — think of them as the brain's filing clerks, creating a physical record every time you have a new experience. Only a small number of these cells get chosen to store any given memory. The team discovered that tau is active during this critical selection process, helping decide exactly which cells will hold onto the experience for the long haul.
To test this, the scientists studied "remote memory" in mice — meaning memories that the animals recalled days or even weeks after something happened. They found something unexpected: tau isn't needed for learning something new or remembering it a few hours later. But without it, memories never become durable. They form in the moment, then quietly vanish.
"Why some memories last while others fade has long puzzled scientists and our study shows that tau plays a key role in how the brain forms long-lasting memories," says Associate Professor Arne Ittner, a neuroscientist from Flinders' College of Medicine and Public Health. "Without it, memories can still form in the moment, but they are weaker."
The team also discovered that tau acts like a noise-canceling device during memory formation. By reducing unnecessary background activity in the brain, it allows only a specific group of cells to become part of a memory — producing clearer, more stable traces.
During learning, tau undergoes a subtle chemical change called phosphorylation. This controlled change turns out to be a normal and essential part of healthy brain function, even though abnormal phosphorylation of tau is a well-known feature of Alzheimer's disease.
Perhaps most surprisingly, the researchers found that even when tau was completely absent, memory traces still existed inside the brain — they just couldn't be accessed naturally. When the scientists directly stimulated the engram cells, they could recover those memories. This suggests tau isn't needed to store memories themselves, but to connect natural cues — a smell, a song, a familiar face — with the ability to actually recall them.
The findings also hint at why dementia patients can sometimes learn new information but struggle to keep it. When disease-associated forms of tau were present during learning, they disrupted the creation of new memories. When they appeared after memories had already formed, they interfered with retrieval.
Because the research was conducted in mice, these findings can't be directly applied to human memory or Alzheimer's disease — yet. But they offer valuable clues that could shape future dementia research.
"Future research will hopefully be able to confirm concepts developed in our study in human memory and show their implication in dementia," says Associate Professor Ittner. For now, tau is no longer just the protein that goes wrong in Alzheimer's. It may also be one of the key reasons any of us can remember anything at all.