Alzheimer's may start when amyloid beta muscles tau off the cell's internal highways

For decades, Alzheimer's research has been organized around a deceptively simple idea: clumps of amyloid beta protein form plaques in the brain, and those plaques cause the disease. Remove the plaques, stop the disease.

Thousands of clinical trials have tested that logic. Nearly all have failed.

Now a study from UC Riverside offers an explanation for why - and it reframes the relationship between Alzheimer's two signature proteins in a way that could redirect therapeutic strategy entirely.

Same binding site, competing tenants

The study, led by chemistry professor Ryan Julian and published in Proceedings of the National Academy of Sciences, Nexus, starts with a structural observation that the field had overlooked. Tau protein normally stabilizes microtubules - the tiny tubes inside neurons that serve as highways for transporting essential molecules. Without functional microtubules, neurons cannot move the cargo they need to survive and communicate.

Julian's team noticed that the regions of tau that attach to microtubules bear a striking resemblance - in both size and structure - to amyloid beta (a-beta). That similarity raised an immediate question: can a-beta bind to microtubules too?

To find out, the researchers tagged a-beta with a fluorescent marker. When the protein's movements slowed and its fluorescent emission changed character, the team could see that a-beta had latched onto microtubules. More importantly, binding experiments showed that a-beta and tau attach with roughly the same strength.

That equal affinity is the critical finding. It means that if a-beta accumulates inside a neuron, it can physically displace tau from microtubule binding sites - not because it binds more strongly, but because there is simply more of it competing for the same spots.

When the highways collapse

The implications cascade from there. Displace tau from microtubules, and the transport system inside neurons starts breaking down. Molecules that need to travel from the cell body to distant synapses - or back again - cannot reach their destinations. The neuron's internal logistics fail.

But the damage does not stop at transport. Tau that is not bound to microtubules begins to misbehave. It aggregates into the neurofibrillary tangles that are Alzheimer's other pathological hallmark. It migrates into parts of the neuron where it does not belong. In this model, tau tangles are not a separate disease process - they are a downstream consequence of a-beta displacing tau from its normal job.

Why plaque removal has not worked

This framework explains a puzzle that has frustrated the field for years. If a-beta plaques cause Alzheimer's, then removing them should help. But drugs that clear extracellular plaques have shown, at best, modest clinical benefits. Julian's model suggests why: the plaques that form outside cells may not be the problem. The damage happens inside neurons, where a-beta competes with tau for microtubule binding. Clearing plaques from between cells does nothing to address the competition happening within them.

The model also fits with what we know about aging. The brain's internal recycling system - a process called autophagy - normally clears proteins like a-beta from inside cells. Autophagy slows with age. If a-beta clearance declines in older adults, the protein accumulates inside neurons and begins outcompeting tau at microtubule binding sites. This could explain why Alzheimer's risk climbs so sharply with age even in people without genetic risk factors.

Lithium, microtubules, and a consistent picture

Several otherwise disconnected observations click into place under this model. Recent epidemiological studies have found that lithium exposure correlates with lower Alzheimer's risk. Previous laboratory work showed that lithium stabilizes microtubules. If the disease begins with microtubule destabilization caused by tau displacement, then a drug that reinforces microtubules could counteract the damage - which is exactly what the lithium data suggest.

The model also explains why both a-beta and tau buildup are required for an Alzheimer's diagnosis. In Julian's framework, the two proteins are not independent pathologies but participants in a single process: a-beta displaces tau, tau aggregates, microtubules collapse, neurons degenerate.

What remains unproven

The binding competition was demonstrated in controlled laboratory conditions using purified proteins and isolated microtubules. Whether the same competition occurs at relevant concentrations inside living human neurons has not been shown. The intracellular environment is crowded with other proteins, and binding dynamics in a test tube do not always predict behavior in a cell.

The model is also, at this stage, a hypothesis supported by biophysical data rather than a clinically validated mechanism. Demonstrating that a-beta displaces tau from microtubules in animal models of Alzheimer's - and that preventing that displacement slows disease progression - would be the next critical test.

Julian acknowledged that the work is primarily about connecting existing findings into a coherent framework. "This idea helps make sense of many results that previously seemed unrelated," he said. "It gives us a clearer picture of what may be going wrong inside neurons and where new treatments might start."

If the hypothesis holds, the therapeutic target shifts from clearing protein aggregates to preventing a-beta from reaching microtubules in the first place - or strengthening the microtubule-tau interaction so that a-beta cannot displace it. Both are druggable strategies, and neither has been seriously pursued.