A pomegranate-derived metabolite and a green tea polyphenol walk into an Alzheimer's mouse model, and the mitochondria seem to respond. That is the short version of a new paper in Mitochondrion (2026). The longer version has a catch, and it matters if you are tempted to read this as a supplement recommendation.
Bottom line up front: this is mouse research. The animals received injections, not oral supplements. The authors themselves say key questions remain open.
What the researchers tested
A team from Texas Tech University Health Sciences Center and AbSynaptex LLC studied humanized amyloid-β knock-in (hAbKI) mice, a model used in Alzheimer's disease research. The two compounds were:
- Urolithin A (UA): described in the paper as a mitophagy-promoting metabolite. Mitophagy is the cell's process for clearing damaged mitochondria.
- Epigallocatechin-3-gallate (EGCG): the green tea polyphenol the authors credit with antioxidant, anti-inflammatory and anti-amyloidogenic properties.
Mice started treatment at 3 months of age. They received either UA (2.5 mg/kg) or UA plus EGCG (25 mg/kg) by intraperitoneal injection, three times a week, for four months. The team then assessed a wide panel of measures at 7 months.
What changed in the treated mice
The authors report shifts across several layers of mitochondrial biology.
Fission and fusion. Expression of the fission genes Drp1 and Fis1 went down. Expression of the fusion genes Mfn1, Mfn2 and Opa1 went up. Immunoblotting supported these transcriptional changes.
Biogenesis. Genes tied to making new mitochondria, including PGC1α, Nrf1 and TFAM, were increased.
Mitophagy and autophagy markers. Levels of PINK1, Parkin, ATG5, LC3B, Beclin1 and BCL2 rose.
Structure and energy. Transmission electron microscopy showed longer mitochondria and less fragmentation. Mitochondrial ATP levels rose, while hydrogen peroxide and 4-hydroxynonenal (markers of oxidative stress) fell.
Neuronal and brain-chemistry readouts. Both regimens were associated with:
- higher synaptic and dendritic measures
- lower Iba1 and GFAP (markers linked to glial activation)
- higher NeuN (a neuronal marker)
- lower soluble Aβ40 and Aβ42
- better performance on several behavioral tasks
Where the combination stood out
Across multiple mitochondrial and neuronal endpoints, the response was greater with UA plus EGCG than with the regimens described as either treatment alone. The authors conclude that UA modulates several components of mitochondrial quality control in these mice. They also conclude that adding EGCG is associated with broader improvements in mitochondrial and neuronal homeostasis.
The wording is deliberately cautious. "Associated with" is not the same as "caused by," and the abstract does not claim the combination is proven superior in every measure.
The catch: what the study did not show
The authors state plainly that further work is needed to:
- Define EGCG's specific contribution. The design suggests the combination did more, but it does not fully separate what EGCG itself is doing.
- Directly assess mitophagic flux. The study measured mitophagy- and autophagy-associated markers, which is not the same as showing the cleanup process actually ran faster.
- Directly assess mitochondrial respiration. ATP levels were measured, but respiration was not.
Other limits follow from the design:
- Animal model only. Mice engineered for amyloid-β biology do not capture the full complexity of human Alzheimer's disease.
- Injection route. Intraperitoneal dosing bypasses digestion and gut metabolism. It says little about what happens after swallowing a capsule.
- Short window, early start. Treatment began at 3 months and ended at 7 months. The paper does not show whether the effects would hold in older animals or after disease is established.
- Mostly molecular readouts. Many findings are gene and protein markers, which are useful signals but not clinical outcomes.
Practical takeaways
Dose. There is no established human dose for this combination. The mouse doses (2.5 mg/kg UA and 25 mg/kg EGCG, injected) should not be converted into a supplement plan. For context only, human studies of oral urolithin A in other settings have used roughly 500-1,000 mg per day. Those studies did not test this combination and do not show it works for Alzheimer's-related mitochondrial changes.
Timing. In the mice, dosing was three times per week for four months, starting early. Whether any similar schedule or window makes sense in people is unknown.
Who it's for. Right now, it is for researchers. Nothing in this paper supports using UA or EGCG to prevent or treat Alzheimer's disease. Anyone with a diagnosis, or concerned about memory, should talk with a clinician before adding supplements. As general background not drawn from this paper, concentrated green tea extracts have been linked to liver problems in some people, which is another reason to seek medical advice.
Why it's still worth watching
Mitochondrial dysfunction and weakened quality control are increasingly recognized as contributors to Alzheimer's disease. A pairing that touches fusion, biogenesis, mitophagy-linked pathways and oxidative stress at once is an appealing hypothesis. This paper adds a coherent preclinical signal. The next steps are the ones the authors name: measure mitophagic flux and respiration directly, and isolate EGCG's role. Until then, treat it as a promising mouse finding, not a protocol.
Source: Kshirsagar S, et al. Urolithin A plus EGCG enhances mitophagy associated pathways, mitochondrial dynamics, and bioenergetic function in humanized amyloid-β knock-in mice. Mitochondrion. 2026;92:102237. doi:10.1016/j.mito.2026.102237. PMID: 43009032.