Image: UZH
Researchers at the University of Zurich transplanted human neural progenitor cells into mice after an induced ischemic stroke. The cells survived the five-week analysis window, mostly became neurons, and formed connections with the host brain. Treated mice climbed and walked better than untreated controls. Christian Tackenberg, scientific head of division at UZH’s Institute for Regenerative Medicine, said the cells “transformed into neurons, which actually even communicated with the already existing brain cells.” Two papers — Nature Communications in September 2025 and Advanced Science online in May 2025 — report the results. No human dose, safety profile, or trial timeline exists.
What the Grafts Did in the Lesion
The team reprogrammed human somatic cells into induced pluripotent stem cells, then directed them toward a neural progenitor fate with collaborators at USC and Kyoto University’s Center for iPS Cell Research and Application. Those progenitors went into cortex damaged by a permanent experimental stroke.
Rebecca Weber and colleagues found the xenografts did more than occupy dead space. Host tissue around the graft showed more blood-vessel growth, less inflammation, and a tighter blood-brain barrier. Single-nucleus RNA work indicated many of the new neurons were GABA-producing cells, a population the stroke had cut by about half. Movement recovery was scored in part with AI-assisted gait and fine-motor analysis.
“Our findings show that neural stem cells not only form new neurons, but also induce other regeneration processes,” Tackenberg said.
Timing Was Not Immediate
A companion Advanced Science study found grafting one week after stroke produced better early graft survival than grafting immediately after the insult. That interval, if it holds in further animal work, would sit after emergency clot care. It is a scheduling clue in mice, not a protocol for an ICU.
The manufacturing protocol avoided animal-derived reagents, which may ease some regulatory steps. Direct injection into injured brain remains invasive. The group is looking at endovascular delivery so cells could travel through vessels instead of a craniotomy. Uncontrolled growth is a known risk for stem-cell-derived products. Researchers are building a biological kill switch if cells proliferate abnormally.
What Mice Cannot Answer
The animals had modified immune systems so they would not reject human cells. That design lets scientists watch the graft. It does not reproduce a patient’s immune system, stroke size, age, or rehab. Animal models do not fully match human recovery of speech or skilled hand use.
Stroke still leaves a large share of survivors with lasting motor or language loss because dead circuits do not grow back on their own. That clinical gap is why the work exists. It does not license a clinic to sell injections. Priorities before any human test are longer follow-up, a delivery method that does not require open-brain surgery, and safety in animals with intact immunity.
The papers show a graft can survive, become neurons, talk to host cells, and coincide with better climbing in a mouse. That is the claim. Everything else is a research queue.
