Neuroscientists at Stanford University have genetically modified mice so that human brain cells can grow and work within their brains, a development that could offer a new way to test therapies for neurological and psychiatric conditions.
The team, led by Prof Sergiu Pașca, engineered mice to develop almost none of their own cerebral cortex. They then implanted brain organoids, small pieces of brain-like tissue grown from reprogrammed human skin cells, into the animals' brains.
The transplanted cells survived, multiplied and organised themselves within the mice's existing brain circuitry. Over a period ranging from a few weeks to a few months, the human cells grew to occupy most of the available space. Nearly half the brain volume of the modified mice was replaced with human cells.
The work, published in the journal Nature on 16 September 2026, showed that the implanted tissue could form cell types found in human and other primate brains that do not occur naturally in mice. The finding suggests that human-derived neural tissue can both survive and functionally integrate in a living mammalian brain across species boundaries.
About six months after surgery, the researchers assessed the animals in basic behavioural tests, including observations of how they moved around a small table-top arena. Pașca said the mice performed 'largely as [the normal] mice did' and stated: 'They don't have any enhancement'.
The experiments did show a difference between the engineered animals that received human cells and those that did not. Mice with missing brain tissue and no transplant had trouble recalling which parts of a maze they had explored. Animals given human cells performed better in the memory-based maze task.
The results may help address a persistent problem in brain research. Scientists often rely on animal models to study disease and assess possible drugs, yet treatments that appear promising in those models can fail in human trials. Pașca said psychiatry has 'one of the lowest success rates for clinical trials'.
He added: 'Even drugs that actually make it to clinical trial - that seem to be working really well in animal models - fail dramatically in clinic'.
The extensive integration of human cells could create a platform for testing treatments in vivo, meaning within a living organism. The Stanford researchers said the approach could be 'transformative' for some complex conditions, including epilepsy, autism and cerebral palsy.
Important questions remain. Researchers do not yet know whether organoid transplants would integrate in the same way in species other than mice, or which molecular processes lead the human cells to organise within mouse brain circuitry. The long-term effects on the mice and their lifespan are also unclear, as is the extent to which the animals can model specific neurological or psychiatric disorders beyond basic memory tasks.
The research also raises questions about the ethical limits of introducing human neural tissue into animals. Pașca said he was not concerned that the mice could develop human cognitive capacities, citing their relatively small brains and the large evolutionary distance between humans and mice.
He drew a firm boundary around primates. Pașca cautioned against placing human brain organoids into a monkey engineered to lack a cortex, describing primates as 'a very clear red line'. Last year, he convened a group of ethics experts to examine the implications of neural organoid technology.
Comments (0)