← Sage — Brain · Sage hub

Sage — A Metabolic Hypothesis of Tau Pathology

Non-pharmacological interventions to interrupt the phosphorylated tau–reverse electron transport loop.

A recent Stanford-led study published in Neuron demonstrated that phosphorylated tau directly regulates mitochondrial reverse electron transport (RET). In fruit flies, mice, and human induced pluripotent stem cell-derived neurons, phosphorylated tau enters mitochondria and binds the Complex I subunit NDUFS3. This interaction activates RET in a phosphorylation-dependent manner. RET produces excess reactive oxygen species and lowers the NAD+/NADH ratio. These changes, in turn, further drive tau phosphorylation through pathways that include MARK/PAR-1 and GSK-3β kinases, establishing a self-perpetuating pathological loop. Tau depletion eliminated stress-induced RET and conferred protection, while pharmacological or genetic inhibition of RET reduced tau-driven neurodegeneration across all three model systems. The work further indicated that regulation of RET is a normal physiological function of tau that becomes pathological under aging or chronic stress, providing a mechanistic link between tau abnormalities and the mitochondrial dysfunction long observed in tauopathies.

Hyperphosphorylated tau and mitochondrial dysfunction therefore form a self-reinforcing circuit. The result is progressive bioenergetic failure that precedes frank tangle formation. The metabolic hypothesis advanced here is that the three EnSurf pillars—Fuel, Circadian, and Strength hormesis—already supply coordinated interventions that lower the probability of pathological RET and reduce the kinase-driven phosphorylation that sustains it. These effects arise from ordinary substrate, redox, and stress-response physiology rather than from novel pharmacology.

The loop may sit on the same reversible-to-locked path traced in the kidney and liver. Under acute stress, the shift in electron flow is reversible while mitochondria remain competent: by analogy, a Crabtree-like adjustment. When aging or chronic metabolic load keeps the loop running, the neuron may lock into a Warburg-like state of lost metabolic flexibility, and tangles may follow. Neurons have limited glycolytic reserve, so the locked state may cost them more.¹

Fuel

Ketone bodies (β-hydroxybutyrate and acetoacetate) and the fasting states that elevate them improve the mitochondrial redox couple. BHB oxidation raises the NAD+/NADH ratio, supports Complex I stability, and favors a more oxidized coenzyme Q pool. A less reduced Q pool diminishes the thermodynamic drive for electrons to flow backward through Complex I, thereby limiting RET-derived ROS. Simultaneously, ketolysis supplies acetyl-CoA that can be routed through Complex II, providing an alternative electron entry that is less dependent on a stressed Complex I. Fasting and time-restricted feeding further activate AMPK and SIRT3, promoting mitophagy and the autophagic clearance of tau species. In models of tauopathy and mitochondrial stress, these substrate shifts reduce markers of hyperphosphorylated tau and improve ATP production under energetic load. Animal-sourced proteins and micronutrient-rich foods supply the cofactors (niacin for NAD+, magnesium, B vitamins) required to maintain the same redox balance, while minimizing the chronic hyperglycemia that elevates kinase activity toward tau.

Circadian

Timed light exposure and sleep alignment constrain the stress signals that initiate tau phosphorylation. Morning bright light and evening dimming support circadian amplitude; deep sleep permits mitochondrial repair and may aid glymphatic clearance (mouse studies suggested sleep speeds the brain’s fluid clearance, though later work disputes this).^2,3 Photobiomodulation with red and near-infrared wavelengths is absorbed by cytochrome c oxidase (Complex IV). Increased Complex IV activity accelerates forward electron flow, raises ATP, and lowers ROS. In tau transgenic mice this intervention reduces hyperphosphorylated tau and neurofibrillary tangles while restoring mitochondrial markers. The same wavelengths also appear to enhance autophagy and heat-shock responses that assist in clearing misfolded tau. Because RET is favored when forward transport is impaired or when the proton-motive force is high relative to electron supply, restoring Complex IV function and reducing nocturnal metabolic stress both lower the conditions under which phosphorylated tau can drive pathological RET.

Strength hormesis

Resistance training and related mito-adaptive stimuli lower tau phosphorylation through phosphatase activation and improved proteostasis. Exercise elevates adiponectin, which increases PP2A activity; PP2A directly dephosphorylates tau at several disease-relevant sites. Concurrently, contractile activity improves mitochondrial dynamics, raises PGC-1α-driven biogenesis, and reduces oxidative and inflammatory tone. In P301S and 3xTg models, both treadmill and ladder-climbing protocols decrease AT8-, AT180-, and S396/S404-positive tau while preserving synaptic markers. Low-dose methylene blue, used within the Strength pillar as a mitochondrial electron cycler, inhibits MARK4-mediated phosphorylation at Ser262 and donates electrons downstream of Complex I, effectively short-circuiting a portion of the RET-vulnerable segment of the chain. Methylene blue is discussed here as information only; no dose is given (see Sage — Methylene Blue). Mild hormetic stress from resistance work also activates AMPK and Nrf2, reinforcing the same antioxidant and quality-control programs engaged by fasting. The net effect is fewer phosphorylated tau molecules available to bind NDUFS3 and a mitochondrial network better able to tolerate residual RET without escalating ROS.

Collectively the three pillars act on the loop at multiple nodes: they raise NAD+/NADH, favor forward electron flow, activate tau phosphatases, enhance clearance of phosphorylated species, and supply cofactors that keep Complex I from operating in a chronically reduced state. None eliminates RET entirely—physiological RET participates in normal stress signaling—but each reduces the probability that the process becomes self-amplifying under aging or metabolic load. Together they may reduce the probability that acute Crabtree-like compensations lock into Warburg architecture.

Secondary pharmacological and experimental approaches

Metformin inhibits Complex I at clinically relevant concentrations and thereby suppresses RET; observational and early trial data link it to lower dementia incidence in diabetic populations, consistent with the mechanism. Metformin is a prescription drug; it appears here as information only. Talk with your own physician about any medication question. Other agents under investigation include selective RET inhibitors, compounds that block tau mitochondrial import or the tau–NDUFS3 interface, and additional kinase inhibitors directed at MARK or GSK-3β. Sauna-like mild hyperthermia and certain uncoupling strategies also reduce tau phosphorylation in preclinical models, likely through overlapping redox and phosphatase effects. These interventions can be layered onto the EnSurf foundation when indicated, but they remain secondary because the substrate, light, and contractile stimuli already modulate the same pathways with lower cumulative risk and broader systemic benefit.

The hypothesis is therefore that sustained application of Fuel (ketone-supported redox balance and fasting-induced quality control), Circadian (light-timed Complex IV support and sleep-enabled clearance), and Strength hormesis (phosphatase activation, electron shuttling, and adaptive biogenesis) measurably lowers both tau phosphorylation and pathological RET activity. Direct measurement of RET-ROS, NAD+/NADH ratios, and site-specific tau phospho-epitopes in longitudinal human studies would test the prediction.

A working hypothesis, open to correction and revised as new evidence arrives. Last revised November 1, 2026.

References

1. Herrero-Mendez A, Almeida A, Fernández E, et al., 2009. The bioenergetic and antioxidant status of neurons is controlled by continuous degradation of a key glycolytic enzyme by APC/C–Cdh1. Nature Cell Biology 11(6):747-752. doi:10.1038/ncb1881
2. Xie L, Kang H, Xu Q, et al., 2013. Sleep drives metabolite clearance from the adult brain. Science 342(6156):373-377. doi:10.1126/science.1241224
3. Miao A, Luo T, Hsieh B, et al., 2024. Brain clearance is reduced during sleep and anesthesia. Nature Neuroscience 27(6):1046-1050. doi:10.1038/s41593-024-01638-y

Related: Sage — Brain · Sage — Methylene Blue · Sage — Light · Sage — Strength · Sage — Fuel. Shorter version: Coach — The Tau Loop.

See the notice at the top of this page. Working hypothesis, not medical advice.