Drug blocking EP2 restored senescent-cell clearance in old mice, easing memory and frailty
A prostaglandin pathway breakdown let EP2 sabotage macrophage cleanup; targeting it improved ageing outcomes in mice.

A study led by Katrin Andreasson at Stanford University found that age-related buildup of senescent (zombie-like) cells in mice is tied to prostaglandin E2 driving EP2 signaling on tissue macrophages. Blocking EP2, either genetically or with an experimental oral drug for two months, improved inflammation, muscle loss, visceral fat, mobility, and memory tests.
If your immune system’s cleanup crew stops showing up, “ageing” stops being a vibe and becomes biochemistry. In a new Science study (DOI: 10.1126/science.aei9816), researchers report that blocking a single signaling pathway helped older mice clear senescent, zombie-like cells and improved multiple signs of ageing, including memory performance and frailty-like outcomes.
The key mechanism centers on prostaglandin E2, a signaling molecule that increases with age. Andreasson and her colleagues compared mice aged 6 to 8 months (roughly equivalent to people in their 20s) with mice aged 23 to 25 months (roughly equivalent to people in their late 60s or 70s). The older mice had more senescent cells in organs including the liver and spleen, and also in bone marrow. The senescent cells turned out to be neutrophils, a type of immune cell normally responsible for first-line defense against infection. Under healthy conditions, macrophages that live in tissues remove damaged cells and debris, but this ability declines with age.
So what breaks the cleanup? The team traced the decline to prostaglandin E2 activity, which overstimulates a receptor called EP2 on tissue-resident macrophages. When EP2 is pushed too hard, the macrophages become less able to remove senescent cells. This is the kind of pathway detail executives should care about because it changes the intervention logic. The research is not trying to kill senescent cells directly. As Derek Gilroy of University College London (not involved in the research) put it, the attractive part is “repairing the body’s own waste-disposal system that should have removed them in the first place.”
To test whether fixing EP2 signaling translates into better ageing outcomes, the researchers ran two complementary experiments. First, they genetically modified the EP2 gene from tissue-resident macrophages in older mice. Those animals cleared more senescent neutrophils. They also showed signs of healthier ageing: lower levels of inflammation, reduced muscle loss, less visceral fat, and better mobility, compared with unmodified older mice. In memory tests, the modified mice performed almost as well as young mice. Andreasson said the magnitude of the effect surprised the team.
Second, they used an experimental drug that blocks EP2. When given orally to older mice for two months, it produced similar age-related improvements to the genetically modified approach. The fact that the effect was seen with an orally administered drug is a practical step forward for any translational conversation. In drug development, “works in mice” is one thing. “Works when administered orally and for a defined period” is often what separates an interesting mechanism from a pathway that could actually become an intervention.
Still, Gilroy’s caution matters for anyone tracking this space. EP2 is part of a normal signaling system, so blocking it throughout the body could have unwanted effects. The mice did not show known side effects in the study, but Gilroy suggested it might be safer to target EP2 in ageing macrophages specifically. That’s a governance and R&D planning issue as much as it is biology: if the receptor has roles beyond macrophage cleanup, the drug design and delivery strategy become central to risk management.
For boardrooms and investment committees, there is another important layer: human evidence so far is supportive but correlative. The researchers reported similar patterns in human tissue, with liver samples from older people showing higher EP2 activity and more senescent neutrophils. But Gilroy noted that the human data are supportive, not yet causal. The team has not shown that blocking EP2 can restore neutrophil clearance in aged human tissue.
Looking ahead, the researchers are planning to study whether this process affects conditions such as Alzheimer’s disease. That matters because it reframes the stakes from “general ageing improvements” to possible links with specific neurodegenerative outcomes. If subsequent work confirms causality in human tissues and establishes safety and durability, this pathway could become a more credible contender in the broader anti-ageing and immune modulation landscape.
Bottom line: the study points to a concrete target, EP2, and a concrete failure mode, age-driven prostaglandin E2 overstimulation that weakens macrophage cleanup of senescent neutrophils. For executives watching longevity, immunology, and regenerative medicine converge, this is the kind of mechanistic clarity that can unlock new programs, reshape risk assessments, and force competitors to answer the same question: can you restore the body’s waste-disposal system without breaking other normal signaling systems? The paper’s next steps will determine whether the promise stays confined to mice or starts earning real-world attention.
This story's Key Insights and Take-aways are locked.
Create a free account to unlock Executive Actions for one credit.
Register to UnlockAlways free for Executives Club members. Join the Club
More in Science
University of Arizona shows graphene nanoribbons survive gamma radiation for fusion sensing
Gamma radiation tests point to a sturdier sensor path for fusion reactors, helping shrink a major grid-readiness hurdle.

New Horizons finds 6 Pluto landslides, with debris aprons that could bury a small city
A NASA imaging analysis of Pluto’s Sputnik Planitia shows six landslides, proving the dwarf planet is still geologically alive.

Parisi and Zamponi prove a+b=1 after 40 Claude prompts for jamming math
The Nobel physicist team used Claude to turn a decade-old jamming relation into a publishable proof in Journal of Statistical Mechanics.

