Modeling Tauopathy: How New Mouse Models Could Help Map Alzheimer’s Disease

Alzheimer’s disease is the most common form of dementia, affecting more than 55 million people worldwide. Characterized by progressive cognitive decline, memory loss and behavioral changes, it places immense emotional and economic stress on patients, families and healthcare systems. Despite decades of research, effective treatments remain elusive.

One reason for this challenge is the disease’s complex pathology. Alzheimer’s is defined by two major hallmarks in the brain: extracellular amyloid-beta plaques and intracellular neurofibrillary tangles composed of the tau protein.

While amyloid-beta has been a central focus in drug development, increasing evidence points to tau pathology as a more accurate predictor of disease progression.

This is where the work of Dr. Naoto Watamura could be vital.

Presenting at Technology Networks’ Innovations in Disease Modeling 2025 symposium, Dr. Watamura explained how his team at University College London’s Dementia Research Institute are using genetically engineered mouse models that more faithfully replicate human tauopathies.

These models, he says, hold promise for unlocking the molecular mechanisms behind Alzheimer’s and accelerating the search for new therapies.

Why tau matters

Dr. Watamura described tau as a “microtubule-associated protein,” central to neuronal structure and stability.

In Alzheimer’s disease, however, tau undergoes post-translational modifications like phosphorylation, causing it to dissociate from microtubules and form aggregates – key hallmarks of neurodegeneration.

“In abnormal conditions, such as disease, tau receives a bunch of a lot of the past translation modifications […] aggregating each other to form PHF-tau.”

But replicating human tauopathy in animals has been a persistent hurdle. Traditional transgenic models overexpress mutant forms of tau, such as 3R tau, leading to inconsistent phenotypes and results that are hard to interpret.

“This problem makes it difficult to distinguish disease associated phenotypes,” Watamura continued, “due to tau overexpressions.”

Modeling tauopathy

To overcome these limitations, Watamura first sourced mice with a genetic modification to simulate Alzheimer’s disease, a “knock-in” model known as App-NL-G-F.

To reduce the issue of tau overexpression, Watamura’s team further modified the mice by replacing their native mouse tau gene (sgRNA-MAPT-Exon10 P301) with its human counterpart mutations (sgRNA-MAPT-Intron10+3 and S305N) using CRISPR-based editing.

In the process, the team faced their challenges – “the most challenging part was designing the guide RNAs to target a specific region of the MAPT genes,” Watamura admitted – but ultimately they managed to achieve the innovation; the mice expressed all six human tau isoforms and displayed better balanced ratios of tau
3R and 4R.

“What I found here is the Intron10+3 and S305N mutations show the shift from the 3R to 4R tau,” Watamura said.

“The advantage of the tau knock-in is it’s not based on overexpression,” he emphasized. “So we can look at more physiological effects of the tauopathy.”

Crossing tauopathy models

Using immunostaining, silver staining and behavioral assays, Watamura and his team found that the mice exhibited decreased synaptic density, increased neurodegeneration and behavioral signs of “impaired short-term working memory,” “impaired long term spatial memory” and even “apathy, which often can be seen as a clinical feature of the frontotemporal dementia patients.”

In a novel step, Watamura also crossed his tauopathy models with amyloid-plaque forming lines to explore synergy.

“Once we crossed the App-NL-G-F knock-in mice with the tauopathy models, we found the increase of tau pathology […] and a significant increase of the dystrophic neurite pathology.”

This convergence mimics the dual pathology seen in human Alzheimer’s brains, providing new insights into how amyloid may exacerbate tau-related neurodegeneration.

Releasing tauopathy models to research

The potential applications of these models, Watamura hints, are vast.

“These mouse models potentially provide pathemechanistic insights into tauopathy/AD [Alzheimer’s disease] and serve as a useful platform for evaluating drug screening in vivo,” he told the Technology Networks audience.

Only time will tell what advances drug developers will make once equipped with these modified mice.