"We are not simply placing a material into the brain," Segura said. "We are engineering a local environment that can coordinate several parts of the repair response."

What signals worked best?

The researchers tested EVs loaded with different molecular signals. One combination stood out: IL-4 and C1q. These two molecules drew macrophages and a surprisingly persistent population of neutrophils into the stroke cavity.

Neutrophils usually cause inflammation and tissue damage during the first hours of a stroke. The new results suggest a second, helpful role at later times.

"This result changes how we think about neutrophils after stroke," said Shangjing Xin, lead scientist of the study and a postdoctoral fellow in the Segura Laboratory. "Their role appears to depend on when they arrive, where they are located, and the signals they receive from their surroundings."

When the researchers reduced neutrophil numbers in the scaffold, blood vessel formation dropped sharply. The scaffold itself also remodeled less. That finding pointed to a constructive role for the cells in healing.

What changed inside the treated brain?

New blood vessels grew throughout the stroke cavity as immune cells moved into the area. Researchers also counted more axons — the long fibers brain cells use to send signals — both within and around the damaged region.

The structural changes translated into measurable motor gains. Treated mice made fewer forelimb placement errors on a grid-walking test. Their performance became statistically indistinguishable from that of healthy controls by week eight.

Was the scaffold actually necessary?

Yes. When the team injected EVs alone, without the MAP scaffold, blood vessel repair did not improve. The porous architecture, and the ability to anchor EV signals inside the damaged region, appeared essential to the repair response.

What are the next steps?

The treatment remains preclinical. All experiments so far used mouse models with direct injection into the brain. Further work must clarify safety, identify which immune-cell populations drive the benefit, and test the approach in larger animals that more closely resemble human stroke.

Segura's lab now plans to derive EVs from human induced pluripotent stem cells. These cells could yield a more scalable, clinically relevant source and give researchers tighter control over the signals inside each vesicle.

"You do not restore an ecosystem simply by containing the initial damage," Segura said. "You have to create the conditions that allow life to return."