Natural Compounds Target Multiple Aging Pathways as AI Accelerates Discovery
A 2025 review maps how polyphenols, terpenoids, and marine metabolites modulate Nrf2, mTOR, SIRT1, and AMPK to slow aging—and how AI is speeding up the search.
Summary
This 2025 perspective from Biomolecules surveys how natural products—polyphenols, terpenoids, alkaloids, polysaccharides, peptides, and marine metabolites—target key aging pathways including Nrf2/ARE, NF-κB, AMPK/PGC-1α, mTOR, SIRT1, and FOXO. Compounds like EGCG, resveratrol, quercetin, curcumin, berberine, spermidine, and ginsenosides each address distinct hallmarks of aging such as oxidative stress, inflammaging, mitochondrial decline, and impaired autophagy. The review highlights synergistic combination strategies, AI-assisted mechanism discovery via molecular docking with neural network scoring, green extraction using deep eutectic solvents, and nanostructured delivery systems. Together, these advances provide a roadmap for translating natural product science into validated nutraceuticals and therapies targeting healthy aging.
Detailed Summary
Aging is driven by interconnected biological processes—oxidative stress, chronic low-grade inflammation, mitochondrial dysfunction, and cellular senescence—that collectively accelerate age-related diseases including cardiovascular disorders, neurodegeneration, metabolic syndromes, and skin photoaging. This 2025 perspective article published in Biomolecules synthesizes current research on natural products as multi-pathway anti-aging agents, with emphasis on mechanistic clarity, combination synergy, AI-assisted discovery, and improved formulation.
The mechanistic core of the review maps specific natural compounds to defined molecular targets. Polyphenols such as EGCG and resveratrol activate the Nrf2/ARE axis, upregulating antioxidant enzymes like HO-1 and glutathione peroxidase. Flavonoids, terpenoids, and alkaloids suppress NF-κB, MAPK, and JAK/STAT signaling to reduce TNF-α, IL-1β, and IL-6—collectively countering 'inflammaging.' Ginsenosides and quercetin support mitochondrial biogenesis via AMPK/PGC-1α. Curcumin, spermidine, and berberine regulate autophagy and proteostasis through mTOR, SIRT1, and FOXO pathways. Notably, nordihydroguaiaretic acid extended median lifespan in male UM-HET3 mice in rigorous multi-site testing, providing direct in vivo longevity evidence for antioxidant natural products.
The review devotes substantial attention to synergy strategies. Examples span enzyme-level pairs (glyceollin plus luteolin at a 3:7 ratio achieving a combination index of 0.642 for α-glucosidase inhibition), tissue-level two-extract blends (Rhynchosia nulubilis and Polygonum multiflorum at 4:1 for dermal papilla cell proliferation under androgen stress), and equal-ratio tri-herbal formulations (TADIOS: Taraxacum officinale, Dioscorea batatas, Schizonepeta tenuifolia) that outperform individual extracts on inflammatory endpoints. These examples illustrate that ratio optimization and mechanistic complementarity are practical levers for achieving enhanced bioactivity.
Artificial intelligence integration is presented as a transformative acceleration layer. Molecular docking enumerates ligand-target interactions for aging-relevant proteins, while artificial neural network (ANN) scoring refines pose selection and affinity ranking. A concrete example involves RSM-ANN-genetic algorithm workflows applied to ultrasound-assisted extraction of Allium sativum, simultaneously optimizing yield and antioxidant potency. CNN-based scorers in GNINA 1.3 are noted to outperform classical scoring functions in structure-based virtual screening. On the formulation side, deep eutectic solvents improve compound yield and stability over conventional solvents, Box-Behnken design has optimized nanostructured lipid carriers for quercetin skin delivery, and RSM-ANN pipelines connect extraction parameters directly to bioactivity outcomes.
The authors acknowledge significant caveats: AI approaches face challenges in data availability, model interpretability, and algorithmic bias. They call for transparent algorithms and tight coupling of computational predictions with experimental validation. Future progress is conditioned on rigorous compositional verification (HPLC, LC-MS/MS, NMR), robust dose-response characterization, clear synergy metrics, appropriate controls, and safety data relevant to older adults. The perspective frames these requirements not as obstacles but as the foundation for clinically meaningful, reproducible natural product science.
Key Findings
- EGCG and resveratrol activate Nrf2/ARE, boosting HO-1 and glutathione peroxidase antioxidant defenses.
- Ginsenosides and quercetin stimulate AMPK/PGC-1α, supporting mitochondrial biogenesis and energy metabolism.
- Nordihydroguaiaretic acid extended median lifespan in male UM-HET3 mice in rigorous multi-site testing.
- Glyceollin–luteolin 3:7 combination achieved a synergy index of 0.642 for α-glucosidase inhibition.
- CNN-based scoring in GNINA 1.3 outperforms classical docking functions, accelerating natural product target identification.
Methodology
This is a perspective/review article, not an original research study; it synthesizes published experimental findings, computational methods, and formulation strategies without generating new primary data. The authors draw on in vitro enzyme assays, cell culture models, animal lifespan studies, and computational docking/AI workflows reported in cited literature. No IRB approval or patient data were involved.
Study Limitations
As a perspective article, the paper does not present new experimental data and relies on heterogeneous cited studies with varying model systems, doses, and endpoints. AI-driven discovery tools discussed face unresolved challenges in data quality, model interpretability, and generalizability. The authors explicitly note that compositional standardization, safety evaluation in elderly populations, and clinical validation remain outstanding requirements for the field.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.