Tau propagation refers to the intercellular spread of pathological tau protein aggregates in the brain, a process that underlies the progression of Alzheimer's disease (AD) and related tauopathies including Progressive Supranuclear Palsy (PSP), Corticobasal Degeneration (CBD), and Primary Age-Related Tauopathy (PART)[@hefti2024][@combs2023]. The propagation of tau follows a characteristic pattern that closely correlates with clinical symptoms and disease staging, beginning in the entorhinal cortex and spreading through connected neural networks to the hippocampus, limbic system, and eventually the neocortex[@braak1991][@cho2024]. Understanding the mechanisms of tau propagation has become central to developing disease-modifying therapies for Alzheimer's disease, as tau pathology shows stronger correlation with cognitive decline than amyloid-beta deposition alone[@schelter2023][@bennett2024].
Tau propagation refers to the intercellular spread of pathological tau protein aggregates in the brain, a process that underlies the progression of Alzheimer's disease (AD) and related tauopathies including Progressive Supranuclear Palsy (PSP), Corticobasal Degeneration (CBD), and Primary Age-Related Tauopathy (PART)[@hefti2024][@combs2023]. The propagation of tau follows a characteristic pattern that closely correlates with clinical symptoms and disease staging, beginning in the entorhinal cortex and spreading through connected neural networks to the hippocampus, limbic system, and eventually the neocortex[@braak1991][@cho2024]. Understanding the mechanisms of tau propagation has become central to developing disease-modifying therapies for Alzheimer's disease, as tau pathology shows stronger correlation with cognitive decline than amyloid-beta deposition alone[@schelter2023][@bennett2024].
The concept of tau propagation emerged from observations that the spatial distribution of neurofibrillary tangles (NFTs) follows a predictable pattern that correlates with clinical disease progression[@jack2010]. This staged pattern of tau pathology suggested that the disease process spreads through connected brain regions rather than arising independently in each area. Subsequent research demonstrated that pathological tau can transfer between neurons, propagate along neural circuits, and template the misfolding of endogenous tau in recipient cells—a process analogous to prion diseases[@frost2009][@clavaguera2013]. This mechanistic understanding has profound implications for therapeutic intervention, as blocking tau propagation could potentially halt disease progression even after amyloid pathology is established.
Tau is a microtubule-associated protein encoded by the MAPT (Microtubule-Associated Protein Tau) gene on chromosome 17q21, primarily expressed in neurons where it plays essential roles in axonal transport, synaptic function, and neuronal polarity[@avila2022][@mandelkow2012]. The tau protein exists in six isoforms in the human brain, generated by alternative splicing of exons 2, 3, and 10, which differ in the number of microtubule-binding repeat domains (three or four repeats; 3R or 4R tau)[@goedert2023]. These isoforms have distinct functional properties, with 4R tau isoforms having higher microtubule-binding affinity than 3R isoforms, and the balance between 3R and 4R tau being critical for normal neuronal function[@wang2016].
In its normal state, tau promotes microtubule assembly and stability through its repeat domains, which bind to microtubules and regulate their polymerization and dynamic instability[@baas2023]. The N-terminal projection domain projects away from the microtubule surface and interacts with other cellular components, including the plasma membrane and organelles[@khalil2024]. This dual functionality allows tau to serve as a linker between microtubules and cellular organelles, facilitating vesicular transport along axons and dendrites. Additionally, tau has been implicated in regulating neuronal signaling pathways, including those involving GSK-3 beta and other kinases that phosphorylate tau[@kimura2023].
The functional state of tau is highly regulated by post-translational modifications, with phosphorylation being the most extensively studied[@stoothoff2005]. In the normal brain, tau exists in a relatively dephosphorylated state, with only 2-3 moles of phosphate per mole of tau. In disease states, tau becomes hyperphosphorylated, containing 5-9 moles of phosphate per mole of tau[@grundkeiqbal1986]. This hyperphosphorylation reduces tau's affinity for microtubules, leading to microtubule destabilization and impaired axonal transport[@cleveland2023].
Multiple kinases phosphorylate tau in vitro and in vivo, including GSK-3 beta, CDK5 (Cyclin-Dependent Kinase 5), MAP kinases (ERK1/2, p38), and casein kinases[@martin2023]. GSK-3 beta and CDK5 are considered the primary kinases responsible for pathological tau phosphorylation in AD brain, as both are activated by neurotoxic stimuli and can phosphorylate tau at multiple AD-relevant sites[@giacomelli2024][@cruz2023]. The balance between kinases and phosphatases (particularly PP2A) determines the phosphorylation state of tau, and dysregulation of this balance contributes to pathological hyperphosphorylation[@liu2024].
Beyond phosphorylation, tau undergoes numerous other post-translational modifications that influence its aggregation and propagation properties. These include acetylation, ubiquitination, sumoylation, nitration, and truncation[@fischer2023]. Tau acetylation at Lys280 (a site critical for aggregation) has been shown to facilitate tau aggregation and propagation, while preventing tau acetylation reduces pathology in mouse models[@min2020]. Truncated tau fragments, particularly those generated by caspase cleavage, are more aggregation-prone and may serve as "seeds" that template the conversion of normal tau to pathological forms[@garciazaballa2023].
The propagation of tau between cells occurs through multiple mechanisms, including synaptic transmission, exosome secretion, and direct cellular uptake[@wang2023][@fu2024]. Synaptic activity has been shown to promote tau release from neurons, and tau can be detected in synaptic vesicles and presynaptic terminals[@pooler2013]. The release of tau into the extracellular space appears to be a physiological process, as normal tau is secreted at low levels in vivo, but pathological forms are released at significantly higher rates[@yamada2014].
Once in the extracellular space, tau can be taken up by neighboring neurons through various endocytic mechanisms. Studies have demonstrated that tau can enter cells via clathrin-mediated endocytosis, macropinocytosis, and receptor-mediated pathways[@rodriguez2017][@falcon2019]. The uptake of extracellular tau is enhanced by its aggregation state, with oligomeric and fibrillar forms being internalized more efficiently than monomeric tau[@mirbaha2018]. This suggests a positive feedback loop where cells taking up pathological tau develop pathology themselves and release even more tau, accelerating propagation.
Extracellular vesicles, particularly exosomes, represent another important pathway for tau propagation[@fowler2021][@saha2024]. Exosomes are small vesicles (30-150 nm) released from cells that can contain various cellular proteins, including tau. Importantly, exosomal tau appears to be particularly efficient at inducing pathology in recipient cells, possibly because exosomes protect tau from degradation and concentrate aggregation-competent species[@wang2017].
Studies have shown that exosomal tau is enriched in phosphorylated and aggregated forms compared to free extracellular tau[@yuyama2023]. Furthermore, exosomes can deliver tau directly to neurons and glia, and may also contribute to the inflammatory response by activating microglia[@asai2015]. The role of exosomes in tau propagation suggests that targeting exosome biogenesis or secretion could represent a therapeutic strategy for blocking tau spread.
The most mechanistically sophisticated model of tau propagation involves the templated conversion of normal tau to pathological conformers, analogous to prion protein propagation in prion diseases[@prionlike2024][@jucker2013]. This model posits that pathological tau ("seed") interacts with normal tau, inducing a conformational change that converts the normal protein to the pathological form. This converted tau can then template further conversions, creating a self-perpetuating cycle of pathology propagation[@frost2024].
Evidence for prion-like templated conversion comes from multiple experimental systems. Studies using cell culture models show that treatment with brain-derived tau aggregates leads to intracellular aggregation of endogenous tau[@guo2013]. In mouse models, injection of brain homogenate containing pathological tau induces tau pathology in recipient animals, and this induced pathology can be transmitted to subsequent generations upon inoculation[@bolmont2017]. The strain-like properties of tau aggregates, where distinct conformations produce different disease phenotypes (e.g., AD vs. PSP), further support the prion-like model[@sanders2014].
Emerging evidence indicates that tau aggregates exist in multiple conformational "strains" that are associated with different clinical and pathological phenotypes[@schubert2024][@vogels2023]. Like prion proteins, tau can adopt distinct aggregated conformations that are stable upon propagation and produce different disease characteristics. For example, the tau pathology in AD differs from that in PSP in terms of isoform composition (both 3R and 4R in AD, predominantly 4R in PSP), filament morphology (paired helical filaments in AD, straight filaments in PSP), and regional distribution[@werner2023].
The strain hypothesis has important implications for understanding selective vulnerability in different tauopathies. Different brain regions may be preferentially affected by specific tau strains due to variations in local tau isoform expression, neuronal connectivity, or cellular factors that influence strain propagation[@kaufman2023]. Understanding the molecular basis of strain diversity could enable the development of strain-specific diagnostic and therapeutic approaches.
The propagation of tau follows characteristic anatomical patterns that have been well-characterized in both human postmortem studies and in vivo imaging studies using PET ligands that bind to tau aggregates[@leuzy2024][@schott2023]. The earliest tau pathology appears in the transentorhinal cortex and entorhinal cortex (Braak stages I-II), followed by the hippocampus and limbic structures (Braak stages III-IV), and finally the neocortex (Braak stages V-VI)[@braak2011]. This progression correlates with the sequence of clinical symptoms, with episodic memory impairment appearing when hippocampal involvement occurs and cortical involvement corresponding to more severe cognitive deficits.
The propagation of tau along neural circuits suggests that connected neurons are particularly vulnerable to tau pathology[@ahmed2024]. Studies using retrograde tracing in animal models have shown that neurons projecting to regions with existing tau pathology are more likely to develop tau pathology themselves, supporting a trans-synaptic spread model[@liu2022]. This network-based propagation model has been validated in humans using resting-state functional connectivity MRI, which shows that patterns of tau deposition correlate with functional brain networks[@franzmeier2020].
Understanding tau propagation mechanisms has revealed multiple potential therapeutic targets[@kidd2024][@brondish2023]. Strategies under investigation include: (1) blocking tau release through modulation of synaptic activity or exosome secretion; (2) preventing tau uptake by targeting cell surface receptors or endocytic pathways; (3) inhibiting templated conversion using small molecules that stabilize the normal tau conformation or interfere with protein-protein interactions required for seeding; (4) enhancing tau clearance through immunotherapy or autophagy induction[@mudher2024].
Active and passive immunotherapy targeting tau has advanced to clinical trials, with several programs specifically designed to block tau propagation[@couturier2024][@sigurdsson2023]. Anti-tau antibodies can neutralize extracellular tau and prevent its uptake by neurons, while also potentially facilitating clearance through Fc-mediated microglia activation. Early-phase clinical trials have demonstrated that anti-tau antibodies can reduce CSF tau levels, suggesting target engagement, though definitive evidence of clinical efficacy is still pending[@mullard2024].
Multiple small molecules targeting various aspects of tau pathology are in development[@gotz2023][@bardal2024]. Tau aggregation inhibitors aim to prevent the formation of pathological tau aggregates that can serve as seeds. Kinase inhibitors targeting GSK-3 beta or CDK5 could reduce pathological tau phosphorylation. Molecular chaperones and stabilizers aim to maintain normal tau function and prevent misfolding. While no disease-modifying tau-targeted therapy has yet reached clinical use, the pipeline remains active with multiple agents in various stages of development.
Tau propagation represents a central mechanism in the progression of Alzheimer's disease and related tauopathies. The intercellular spread of pathological tau, driven by synaptic transmission, exosomal secretion, and prion-like templated conversion, explains the characteristic anatomical progression of tau pathology and its correlation with clinical disease progression. Understanding these mechanisms has revealed multiple therapeutic targets, and strategies to block tau propagation are actively being pursued. While significant challenges remain, the development of disease-modifying therapies targeting tau propagation offers hope for fundamentally altering the course of tauopathies.
From the [SciDEX Exchange](/exchange) — scored by multi-agent debate