Summary: A new UCSF study shows how gene expression shapes where and how the Alzheimer’s-associated protein tau accumulates and spreads through the brain. Combining brain imaging, transcriptome maps and an advanced spread model, researchers identified four genetic pathways that either promote or protect against tau buildup, depending on whether their effects align with the brain’s connectivity. These findings challenge the idea that tau spreads only by passive diffusion and point to directed, network-based transport—highlighting new targets to slow or halt disease progression.
Alzheimer’s disease affects brain regions unevenly: areas such as the entorhinal cortex and hippocampus develop tau tangles early, while primary sensory cortices tend to be relatively spared. Tau normally stabilizes microtubules inside neurons, but in Alzheimer’s it misfolds, aggregates and forms toxic inclusions that impair neuronal function and eventually kill cells. Understanding why some regions are vulnerable and others resilient has been difficult because the spatial patterns of genetic risk have not matched simple maps of tau pathology.
Key Facts:
- Four gene pathway classes: Genes were grouped as promoting or resisting tau in ways that either align with, or act independently of, the brain’s network.
- Active, directed tau spread: Evidence supports tau transmission along neuronal connections with directional bias, not only passive diffusion through extracellular space.
- New intervention targets: Distinct biological roles for vulnerability and resilience genes highlight pathways—stress response, metabolism, immune function and amyloid processing—that may be targeted to alter tau spread.
Study overview
Researchers at the University of California, San Francisco developed an extended Network Diffusion Model (eNDM) to predict tau spread using healthy-brain connectome data. They applied the model to tau PET scans from 196 individuals at various stages of Alzheimer’s disease. By comparing observed tau patterns with model-predicted patterns, they isolated a residual signal—areas where tau accumulation could not be explained by network-driven spread alone. That residual highlighted locations where local factors, including gene expression, likely modulate vulnerability or resilience.
To examine genetic influences, the team compared the spatial expression of 100 Alzheimer’s risk genes—derived from the Allen Human Brain Atlas—with both the eNDM-predicted tau distribution and the residual tau signal. This allowed them to separate genes whose effects are aligned with connectivity from those that operate independently of network spread.
Four genetic signatures of vulnerability and resilience
The analysis revealed four distinct classes of genes:
- Network-Aligned Vulnerability (SV-NA): Genes whose expression promotes tau spread along connected pathways.
- Network-Independent Vulnerability (SV-NI): Genes that increase local susceptibility to tau accumulation in ways not explained by connectivity.
- Network-Aligned Resilience (SR-NA): Genes expressed in regions that resist tau despite being connected to vulnerable areas.
- Network-Independent Resilience (SR-NI): Genes that protect regions outside the typical network routes, acting as local defenses.
Functional analysis showed consistent biological differences among these classes. Network-aligned vulnerability genes were enriched for processes related to cell stress, metabolism and cell death—mechanisms that may facilitate tau propagation through connected neurons. Network-independent genes tended to implicate immune response and amyloid-β processing pathways—local mechanisms that influence resilience or susceptibility regardless of the wiring diagram.
Implications for how tau spreads
These findings reinforce and extend other UCSF work showing that tau propagation is not purely passive. Animal and modeling studies indicate tau moves trans-synaptically along axons and favors particular directions, consistent with an active transport process. By integrating network modeling with gene expression mapping, the current study shows that both network-based transmission and local genetic programs determine where tau accumulates and how pathology progresses.
“Think of the eNDM as a map predicting where tau is likely to travel next, based on healthy brain connections,” said senior author Ashish Raj, PhD. Comparing that map to observed data reveals the genetic ‘overrides’—local vulnerabilities or defenses—that modify disease spread.
Clinical and research relevance
By separating network-aligned from network-independent genetic effects, the study supplies a more nuanced framework for identifying therapeutic targets. Interventions could aim to block the mechanisms that facilitate trans-neuronal tau transport, bolster region-specific resilience pathways, or modulate immune and amyloid-processing functions that operate locally. The gene classes and associated biological processes reported here provide candidate pathways for such strategies.
Authors and funding
Additional authors include Farras Abdelnour, Benjamin Sipes, Daren Ma and Pedro D. Maia, PhD. The research was partially supported by NIH grants R01NS092802, RF1AG062196 and R01AG072753 awarded to Ashish Raj.
About this research news
Author: Melinda Krigel
Source: UCSF
Contact: Melinda Krigel – UCSF
Image: The image is credited to Neuroscience News
Original Research (Open access): “Selective vulnerability and resilience to Alzheimer’s disease tauopathy as a function of genes and the connectome” by Ashish Raj et al., published in Brain.
Abstract (condensed)
Alzheimer’s disease shows regionally distinct vulnerability to tau pathology: the entorhinal cortex and hippocampus show early tangles while primary sensory cortices remain resilient. To reconcile how local genetic programs and network-mediated spread together shape this selective vulnerability and resilience, the investigators used an extended network diffusion model fit to tau PET from 196 patients. Comparing observed and model-predicted tau allowed isolation of residual tau reflecting non-network factors. Correlating that residual with spatial expression of 100 Alzheimer’s risk genes from the Allen Human Brain Atlas revealed four gene classes—network-aligned and network-independent vulnerability and resilience—each with distinct functional enrichment. Network-aligned genes relate to stress, metabolism and cell death, while network-independent genes implicate amyloid processing and immune response. These results suggest multiple pathways by which genes modulate tauopathy and point to potential intervention targets.