Hijacking skull immune cells to bypass the blood-brain barrier for brain drug delivery

Development of effective medications for central nervous system (CNS) disorders has long been plagued by a high failure rate in clinical trials. One major obstacle is the limited permeability of the blood brain barrier (BBB), which often leads to insufficient drug accumulation and consequently unsatisfactory therapeutic outcomes. Despite recent advancements in multiple drug delivery platforms, efficacious drug delivery to the CNS remains a daunting challenge.

Recent discoveries in neuroimmunology, however, have revealed a previously underappreciated pathway connecting the skull bone marrow to the brain. Immune cells residing in the calvarial bone marrow can migrate through skull-meninges microchannels, allowing them to access brain tissue rapidly while bypassing the classical BBB. This finding has opened new possibilities for delivering therapeutic materials directly to the brain.

A research team led by Mingjun Zhang from Tsinghua University and Yilong Wang from Beijing Tiantan Hospital, has now demonstrated how this pathway can be harnessed for targeted CNS therapy. In a study published in Cell (Janurary 16, 2026）, the researchers reported a strategy in which nanoparticles “hijack” calvarial immune cells, using them as carriers to transport therapeutic cargo directly into the brain for stroke treatment.

By injecting albumin-based nanoparticles into the skull bone marrow, the team achieved efficient uptake through local myeloid immune cells. Compared with conventional intravenous administration, this approach resulted in minimal systemic exposure and markedly improved delivery efficiency to the brain. In models of neuroinflammation or acute ischemic stroke, the nanoparticle-loaded immune cells migrated through skull-meninges channels and preferentially accumulated at brain lesion sites, enabling targeted drug delivery while bypassing the BBB.

Using the neuroprotective drug nerinetide (NA1) as a model drug, the researchers showed that this delivery strategy significantly reduced infarct size, brain edema and neurological deficits in mice, even at doses as low as one fifteenth of those required for intravenous delivery. Notably, therapeutic benefits were observed when treatment was initiated up to 4.5 hours after stroke onset, a clinically relevant time window. Long-term follow-up further revealed sustained preservation of brain structure, accompanied by improved functional recovery and survival.

To explore clinical feasibility, the team conducted the first in human exploratory study involving patients with malignant middle cerebral artery infarction. The skull bone marrow injection procedure was well tolerated, with no serious adverse events observed during follow-ups, and early signs of neurological improvement were observed in treated patients.

“This work demonstrates that the calvarial bone marrow is not just a passive reservoir of immune cells, but a functional gateway to the brain,” said Dr. Xize Gao, the lead author of the study. “By leveraging immune cell trafficking, we can establish an effective immune-guided route for delivering therapeutics to the CNS.”

Beyond drug delivery, the team noted, the skull bone marrow-brain pathway may present alternative approaches for a variety of different carriers, such as stem cells or CAR-T cells, as well as functional nanomaterials designed for wireless, non-invasive brain interaction. “It points to a new class of brain interfaces that operate not only through information transport, but also via controlled matter delivery,” they said, “integrating therapy, immunomodulation, and brain-machine interaction.”

In summary, this promising approach offers a versatile platform for next-generation CNS treatments. By bypassing the blood-brain barrier, it can enable targeted delivery for a wide range of CNS disorders, while also laying the foundation for advanced, minimally invasive brain-machine interfaces.

Other contributors to the study include Dr. Xize Gao, Dr. Xiangrong Liu and Dr. Nanxing Wang as co-first authors, along with collaborators from Tsinghua University, Beijing Tiantan Hospital, Northwestern Polytechnical University and Peking University.

About the Authors

Dr. Mingjun Zhang is a professor in the School of Biomedical Engineering at Tsinghua University, China, and Dr. Xize Gao is a post-doctoral research associate at Zhang’s Lab. Their research focuses on the interdisciplinary frontier of neural engineering and micro-/nano-medicine, with an emphasis on developing micro/nanorobotic systems and intelligent closed-loop brain-machine interfaces (BMIs) to overcome biological barriers for integrated matter, energy, and information exchange with the nervous system. Their research group explores embodied-intelligence BMI paradigms characterized by enhanced perception, predictive decision-making, and rapid execution, advancing brain-machine integrated intelligent robotic systems and providing key technologies for brain science research, neurological disease diagnosis and therapy, and brain-machine intelligence fusion with functional and behavioral enhancement. For more information, please see their research homepage at http://www.bmi-robotics.com/index.php/publications/