The Convergence of Bredesen’s Multimodal ReCODE Approach & Cell-Type-Directed Network Therapies in Alzheimer’s Disease: By Dr. Jeff Bland

We’re honored to share this important piece from Dr. Jeffrey Bland, President of the Personalized Lifestyle Medicine Institute (PLMI).

In this article, Dr. Bland bridges two worlds—cutting-edge molecular neuroscience and clinical precision medicine—showing how the latest Cell research aligns with Dr. Dale Bredesen’s ReCODE Protocol. It’s a powerful look at how network-based thinking is reshaping our understanding of Alzheimer’s disease and paving the way for truly personalized care.

---

The Convergence of Bredesen’s Multimodal ReCODE Approach and Cell-Type-Directed Network Therapies in Alzheimer’s Disease

Jeffrey S. Bland, Ph.D.
President, Personalized Lifestyle Medicine Institute

Alzheimer’s disease (AD) has humbled the one-target/one-drug paradigm for decades. Even when a target is biologically relevant—amyloid or tau, for example—clinical benefits have often been modest because AD is not a single-node failure but a network disorder that unfolds differently across cell types and over time. Dale Bredesen’s ReCODE program anticipated this reality from the clinical side: rather than pursuing a silver bullet, it orchestrates many small levers—nutrition, metabolic repair, micronutrients, detoxification, sleep and circadian hygiene, physical and cognitive training—to push multiple pathways in a healthier direction at once. In 2025, Li and colleagues provided a rigorous molecular counterpoint to that philosophy. By building cell-type-specific transcriptional maps of AD and then matching them to drug perturbation profiles, they identified a combination that “network-corrects” disease signatures across neurons and glia—letrozole (neuronal) plus irinotecan (glial)—and showed that this pair improved memory and broadly alleviated pathology in a stringent mouse model while being supported by real-world evidence in human medical records. The conceptual bridge between ReCODE and Li et al. is clear: multi-lever, cell-aware interventions can outperform single-target strategies in a heterogeneous disease.

From “disease of plaques” to a “disease of cell networks”

Li et al. pooled single-nucleus RNA-seq from multiple cohorts to extract cell-type-specific AD signatures across six major brain populations: excitatory neurons, inhibitory neurons, microglia, astrocytes, oligodendrocytes, and oligodendrocyte precursor cells (OPCs). They found distinct, and sometimes oppositely regulated, gene programs across these types, with only modest overlap—an expression landscape that explains why one drug aimed at a single hallmark is unlikely to help every cell state that is “off” in AD. Their computational repurposing pipeline then queried these signatures against the Connectivity Map, nominating drugs predicted to reverse (i.e., normalize) the dysregulated patterns within each cell type. Crucially, they did not stop at lists: they integrated real-world evidence from millions of electronic medical records and prioritized a minimal pair—letrozole for neuronal signatures, irinotecan for glial signatures—that was associated with lower AD risk in exposed populations and then validated the duo in vivo.

Experimentally, the combination outperformed either single agent: only the pair rescued both short- and long-term spatial memory and produced broad improvements across pathology measures, including hippocampal atrophy, amyloid plaques, and (importantly) phosphorylated tau—an index tightly coupled to cognitive decline.

Single-nucleus profiling after treatment showed reversal of human AD signatures in a cell-type-specific manner and enrichment of neuroprotective pathways such as long-term potentiation, cAMP and calcium signaling, circadian entrainment, cholesterol metabolism in astrocytes, and oxidative-stress programs in microglia. In short: fix the network, not merely the node.

Why a multimodal lifestyle-nutrient program is a natural “network-corrector”

Bredesen’s ReCODE arrays interventions across six practical domains—Inflammation, Atrophic/Mitochondrial, Glycotoxic, Toxic, Vascular, and Hormonal—precisely because each domain modulates different subnetworks of the AD phenotype. Read through the lens of Li et al., ReCODE’s levers map onto cell-type vulnerabilities:

1) Excitatory and inhibitory neurons: plasticity, energy, and synaptic signaling.
Li et al. found neuronal signatures linked to synaptic function, long-term potentiation, and signaling axes such as estrogen, cAMP, and calcium. Nutritional strategies that raise neuronal “signal-to-noise” and energy availability speak directly to these modules. Ketoflexible nutrition (time-restricted feeding, low-glycemic Mediterranean-style diet) increases β-hydroxybutyrate, a signaling metabolite that supports BDNF, mitochondrial biogenesis, and more stable neuronal firing. Omega-3 DHA and EPA remodel synaptic membranes and promote pro-resolving mediators that damp synapse-toxic inflammation, while polyphenols (curcumin, catechins, anthocyanins, quercetin) enhance CREB-BDNF pathways and modulate kinases that influence tau phosphorylation. Methylation cofactors (folate, B12, B6, riboflavin) and niacinamide (NAD⁺ repletion) help tune neuronal epigenetics and energy metabolism. Sleep optimization, bright-day/dark-night circadian practices, and aerobic plus resistance exercise further potentiate LTP and cAMP/PKA signaling—behavioral “drugs” that aim at precisely the neuronal pathways Li et al. show are off-axis.

2) Microglia: the inflammatory rheostat.
Microglia are the brain’s innate immune sentinels; in AD they often get stuck in a high-alert phenotype, amplifying MAPK/NF-κB signaling, impairing phagocytosis, and producing reactive oxygen species. Li et al. demonstrate that glial signatures are as decisive as neuronal ones, and their combination therapy tamped down microgliosis and restored glial-related pathways.

ReCODE connects here through: omega-3-derived SPMs (resolvins, protectins, maresins) that flip microglia from M1-like to pro-resolving states; vitamin D sufficiency for immunoregulation; and polyphenols that activate NRF2 and inhibit NLRP3 inflammasomes. The gut–brain axis is a potent microglial dial: prebiotic fibers and targeted probiotics increase butyrate, an HDAC-inhibiting short-chain fatty acid that re-programs microglial gene expression toward resolution. Sleep (deep NREM, glymphatic clearance), stress-reduction and vagal-tone practices (box breathing, paced exhalation), and regular physical activity add non-pharmacologic pressure to the same inflammatory network nodes that Li et al. targeted pharmacologically.

3) Astrocytes: cholesterol, glutamate, and circadian glue.
Astrocytes choreograph cholesterol shuttling (APOE, ABCA1), buffer glutamate, and gate calcium and circadian rhythms—exactly the pathways Li et al. show are perturbed and then normalized by combination therapy. Nutritionally, choline and phosphatidylcholine support phospholipid and lipoprotein remodeling; niacinamide and ribose help normalize astrocyte NAD⁺ redox; MCT oil and mild ketosis augment astrocyte-neuron lactate shuttling; and carotenoids (lutein, astaxanthin) and sulforaphane support antioxidant and detox transcriptional programs. Behaviorally, circadian-consistent feeding (most calories in daylight, ≥3 hours before sleep), morning light, and regular sleep–wake timing stabilize astrocytic clock genes that influence synaptic homeostasis—an inherently cell-type-aware lifestyle prescription aligned with Li et al.’s pathway map.

4) Oligodendrocytes and OPCs: myelin maintenance as cognition insurance.
White-matter integrity and remyelination capacity depend on lipid supply, mitochondrial ATP, and low oxidative stress. Li et al. observed AD-linked disruptions and treatment-driven improvements in oligodendrocyte/OPC pathways including differentiation, cAMP/calcium signaling, and ROS control. ReCODE’s diet—rich in omega-3s, monounsaturated fats, and fat-soluble antioxidants—rebuilds the lipid backbone of myelin while creatine, CoQ10, and ergothioneine support glial mitochondria. B-vitamins and methyl donors influence expression of myelin genes (e.g., MBP), while resistance exercise and glycemic control reduce the glyco-oxidative drag that impairs oligodendroglial function. In this frame, “eat for myelin” is not a slogan; it is a cell-type-directed tactic.

Convergence of methods: drugs, foods, and lifestyle habits as edges on the same graph

The Li et al. framework begins with human data—post-mortem single-cell brain maps—then chooses drugs predicted to reverse those gene programs, validates them against EMR-derived risk signals, and finally demonstrates superiority of a combination in a mouse model with both amyloid and tau pathologies.

ReCODE starts with clinical phenotype and risk drivers—insulin resistance, inflammation, toxin burden, nutrient depletion, circadian and sleep disruption—and systemically removes friction while providing substrates for repair. Yet both approaches can be plotted on the same graph: nodes are pathways (LTP, cAMP, calcium, MAPK/NF-κB, cholesterol handling, oxidative phosphorylation), edges are interventions (letrozole/irinotecan on the drug side; DHA, polyphenols, D, methyl donors, ketosis, sleep, exercise, stress recovery on the lifestyle side). Where the graph densifies—where multiple edges converge on the same dysregulated subnetwork—we expect clinical benefit.

That expectation is borne out in Li et al.: the combination normalized dozens of human AD signature genes per cell type and enriched neuroprotective pathways such as long-term potentiation, circadian entrainment, and calcium/cAMP signaling, while simultaneously reducing microgliosis and rescuing CA1 neuronal loss and hippocampal volume; only the combination showed consistent memory rescue in both short- and longer-term probes.

The implication for clinical practice is not that everyone should take oncology drugs, but that multi-lever strategies—including diet and lifestyle—deserve to be engineered cell-type-first. In practical terms:

• Design for neurons: ketoflexible Mediterranean eating, time-restricted feeding, aerobic + resistance training, sleep consolidation, DHA/EPA, polyphenols, magnesium, niacinamide, B-vitamins.

• Design for microglia: omega-3-SPM axis, vitamin D sufficiency, polyphenol-rich foods, prebiotic fibers and targeted probiotics, stress-to-calm transitions (vagal tone), heat/cold hormesis as tolerated.

• Design for astrocytes: choline/PC, MCTs for lactate shuttling, circadian-consistent feeding and light, carotenoids, sulforaphane, glutathione precursors.

• Design for oligodendrocytes/OPCs: lipid quality (DHA/olive-oil-centric fats), creatine/CoQ10/ergothioneine, methylation sufficiency, glycemic control, progressive strength training.

Beyond AD: a template for neurodegeneration

Because Li et al.’s methodology is agnostic to disease label—map cell-type programs, choose edges that reverse them, validate in the clinic and the lab—it generalizes to Parkinson’s disease, ALS, and mixed vascular-degenerative cognitive impairment. ReCODE’s scaffolding, likewise, is portable: restore metabolic flexibility, extinguish chronic inflammation, remove toxic friction, feed synapses and myelin, stabilize circadian timing, and protect sleep. In both cases, success depends on respecting cellular context. The neuron cannot thrive if the astrocyte starves for NAD⁺; synapses will not stabilize if microglia remain in a danger-signaling loop; myelin cannot be maintained without lipid building blocks and ATP. The right “cocktail” varies by person because the relative contribution of each cell-type disturbance varies by person.

A practical synthesis

Clinicians and researchers can now meet in the middle. From the lab bench, cell-type-directed combination therapies offer a blueprint for pharmacologic synergy in heterogeneous disease, with the Li et al. study providing compelling preclinical efficacy and human-data triangulation.

From the clinic, ReCODE shows how a well-designed multimodal program can operationalize the same logic with foods, nutrients, and habits—safe, scalable tools that modulate the very pathways Li et al. highlighted: synaptic plasticity, cAMP/calcium signaling, circadian entrainment, lipid transport and cholesterol homeostasis, oxidative phosphorylation, and the inflammatory rheostat across microglia and astrocytes.

The old question—Which single target matters most?—is giving way to a better one: Which set of cell-type networks do we need to normalize together for this individual? Li et al. and Bredesen arrive at the same answer from opposite directions. The future of AD care is not a bigger hammer; it is a smarter orchestra.

References

• Li Y., Pereda Serras C., Blumenfeld J. et al. (2025). Cell-type-directed network-correcting combination therapy for Alzheimer’s disease. Cell 188, 5516–5534. https://doi.org/10.1016/j.cell.2025.06.035

• Coleman K., Tatonetti N.P. (2025). Decoding Alzheimer’s disease at the cellular level reveals promising combination therapy. Cell 188, 5433–5435.

• Parolo S. et al. (2023). Single-cell-led drug repurposing for Alzheimer’s disease. Sci Rep 13, 222.

• Bredesen D.E. et al. (2016). Reversal of cognitive decline: a novel therapeutic program. Aging 8, 1250–1258.

• Bredesen D.E. et al. (2020). Precision medicine approach to Alzheimer’s disease: the ReCODE trial. J Alzheimers Dis 73, 1003–1024.

• Calder P.C. et al. (2020). Specialized pro-resolving mediators in inflammation. Prog Lipid Res 78, 101047.

• Tan J.Z. et al. (2022). Docosahexaenoic acid and synaptic plasticity. Nutrients 14, 2138.

• Wang Y. et al. (2021). Polyphenols in neuroprotection and mitochondrial function. Mol Neurobiol 58, 543.

• Musiek E.S., Holtzman D.M. (2016). Mechanisms linking circadian clocks, sleep, and neurodegeneration. Science 354, 1004–1008.

Find the original piece here.