Fisetin is a senotherapeutic that extends health and lifespan

TL;DR

Screened a panel of 10 flavonoids (plus pirfenidone control) for senolytic activity in vitro; fisetin was the most potent. Showed in two distinct in vivo regimens that fisetin reduces senescent-cell burden in multiple tissues of aged mice, and — when fed as a chronic diet starting at 85 weeks of age — extends both median and maximum lifespan in WT mice. Confirmed senolytic activity in human adipose tissue explants ex vivo. Foundational paper underpinning current human senolytic trials.

Design

The paper contains two distinct in vivo experiments, often conflated:

Experiment A — Acute senescence reduction in aged WT mice

• Subjects: 22–24 month old WT C57BL/6 mice (pure background)
• Dose: 100 mg/kg fisetin in PEG400/ethanol vehicle by oral gavage, 5 consecutive days
• Sacrifice: 72 h after final dose
• Endpoint: SA-β-gal+ cells in inguinal fat (n = 6–7 per group)
• Companion: also tested in p16Ink4a-INK-ATTAC reporter mice with CyTOF-based identification of which p16+ cell types were cleared

Experiment B — Late-life chronic intervention for lifespan

• Subjects: f1 C57BL/6 × FVB/n hybrid WT mice (NOT pure C57BL/6)
• Treatment start: 85 weeks of age (~20 months; paper notes this is “roughly equivalent to age 75 years in humans”)
• Diet: 500 ppm fisetin in chow (~60 mg/kg/day calculated from food intake) ad libitum, vs control diet with no drug
• n: 8–9 mice per group
• Statistic: Log rank (Mantel-Cox) survival test
• Endpoints: lifespan (median + maximum); composite lesion scores in brain/kidney/liver/lung/forepaw; serum amylase/ALT/MCP-1; SASP factors in tissues

Companion experiments

• In vitro flavonoid screen: Ercc1-/- MEFs (a progeroid model); 11 compounds at 5 μM (resveratrol, fisetin, luteolin, rutin, EGCG, curcumin, pirfenidone, apigenin, myricetin, catechin, quercetin); SA-β-gal+ readout via IN Cell Analyzer 6000.
• Cell-type validation: HUVECs (apoptosis via caspase 3/7 in fisetin-treated senescent HUVECs); IMR90 cells (senescence reduction).
• Progeroid mouse studies: Ercc1-/Δ mice (a hypomorphic progeroid model) treated with intermittent 500 ppm fisetin chow.
• Human ex vivo: human omental adipose tissue explants from 3 female subjects (1 lean BMI ~26, 1 obese BMI 45.6, ages 55–66) treated with 20 μM fisetin × 48 h; SA-β-gal and multiplex SASP cytokines.

Key results

Acute fisetin reduces senescent cells in multiple tissues

In 22–24 month WT C57BL/6 mice (Experiment A): single 5-day course of 100 mg/kg fisetin gavage produced significant reduction in SA-β-gal+ cells in inguinal fat (Fig 4A, p<0.01).

CyTOF in p16Ink4a-INK-ATTAC mice (Fig 4C) showed cell-type-specific senolysis — significant reductions in p16+ cells among:

• c-Kit+ mesenchymal stem/progenitor cells
• CD4+ T lymphocytes
• CD8+ T lymphocytes
• NK1.1+ NK cells
• CD146+CD31+ endothelial cells

But no significant effect on p16+ macrophages or dendritic cells. Fisetin is not a pan-senolytic — it acts on specific cell populations.

Independent confirmation via CENP-B+ cell quantification (Fig 4D) in the same populations.

Late-life fisetin extends lifespan in WT mice

In 85-week-old f1 C57BL/6;FVB/n WT mice (Experiment B; n=8–9/group): chronic 500 ppm fisetin chow extended both median and maximum lifespan (Fig 5A,B; Log rank test, *p<0.05). This is the headline lifespan finding.

Healthspan markers improved in old WT mice

• Composite lesion scores reduced in brain, kidney, lung (Fig 5D; n=3–8/group)
• Serum amylase (pancreatic marker) and ALT (liver enzyme) significantly reduced
• Serum MCP-1 (a SASP factor) reduced
• Liver HNE adducts (oxidative stress marker) reduced
• Liver GSH:GSSG ratio improved (antioxidant capacity)
• Tissue SASP marker expression reduced in fat, spleen, liver, kidney, and peripheral CD3+ T cells (Fig 5F-J)

Human ex vivo confirmation

Human adipose tissue explants treated with 20 μM fisetin × 48 h showed:

• Significant reduction in SA-β-gal+ cells (Fig 4E, p<0.01)
• Significant reduction in conditioned-media IL-6, IL-8, and MCP-1 (Fig 4F)
• IL-10 and GM-CSF not significantly affected

n=3 subjects only — preliminary but mechanistically supportive of human translatability.

Proposed mechanism

The paper does not identify a single direct molecular target. Discussion notes:

• Fisetin “blocks the PI3K/AKT/mTOR pathway” (citing prior work, ref 80) — disrupting this pathway is one of the SCAP (Senescent-Cell Anti-Apoptotic Pathway) targeting strategies.
• Fisetin is a topoisomerase inhibitor and increases hSIRT1 catalytic activity in vitro.
• Acts as a direct ROS scavenger and inhibits TNFα, IL-6, NF-κB activity.
• Cell-type-dependent mode of action: in HUVECs fisetin induces apoptosis (caspase 3/7); in MEFs it reduces senescence markers without cell killing — apoptotic senolysis vs senomorphic depending on cell type.

The paper does not position fisetin as a Bcl-xL/Bcl-2 inhibitor (that mechanism applies to Navitoclax / A1331852, mentioned as a separate senolytic class).

Pharmacokinetics

The Discussion cites half-lives of 0.09 h (rapid) and 3.1 h (terminal) for fisetin (compared to 3.1 h for quercetin). This is extremely short. Authors argue the persistent effect after intermittent dosing favors a “hit-and-run” senolytic mechanism (one-time elimination of senescent cells) rather than continuous-occupancy receptor pharmacology.

Average dietary fisetin intake in Japan is reported as ~0.4 mg/day — orders of magnitude below the experimental doses.

Extrapolation to humans

Dimension	Status	Notes
Pathway conserved in humans?	yes	Senescence biology and SASP conserved; PI3K/AKT/mTOR conserved
Phenotype conserved in humans?	partial	Human ex vivo explants show senescent-cell clearance + SASP reduction (n=3); whole-organism effect untested in vivo
Replicated in humans?	in-progress	Multiple human trials initiated post-publication; see fisetin § Human evidence for current NCT registry status

Limitations

• Lifespan study used a hybrid background (F1 C57BL/6;FVB/n), not the more common pure C57BL/6. Strain-effect generalizability is uncertain.
• Lifespan study n is small (8–9/group). Replication in larger cohorts would strengthen confidence.
• Human evidence is ex vivo only (n=3). No in vivo human pharmacodynamic data in this paper.
• Mechanism-of-action remains incompletely resolved despite the in vivo phenotype.
• Authors have a financial interest disclosed (patents on senolytic drugs, owned by Mayo Clinic; J.L.K., T.T., Y.Z., M.X., and T.P. listed as conflict).
• “No adverse effects of fisetin have been reported, even when given at high doses” — author claim, but the paper does not include a dedicated long-term safety arm.

Cited by

[Auto-populated by Obsidian backlinks panel]