Cerebral organoids reveal how Ebola virus hides in the brain for months

Ebola does not always leave when the acute infection ends. Months or years after a person survives the initial illness, the virus can remain hidden in the body, particularly in the brain—a place where the immune system operates with deliberate restraint to protect delicate neural tissue. This persistence creates a dangerous paradox: the virus stays alive and infectious, capable of triggering disease relapse or, in rare cases, spreading to others. Until recently, scientists understood very little about how the virus manages this feat.

Researchers at the Icahn School of Medicine at Mount Sinai and the Bernhard Nocht Institute for Tropical Medicine set out to solve the puzzle using an unconventional tool: lab-grown brain tissue. They created cerebral organoids—three-dimensional structures built from human stem cells that mimic the architecture and cell types of the central nervous system. These miniature brains allowed them to observe Ebola virus behavior over extended periods in a human context, something that would be far more difficult and ethically fraught to study in living patients or even in animal models. The work, published in Nature Microbiology, reveals how the virus persists and why survivors sometimes develop severe inflammation of the brain, eyes, and meninges months after they should have recovered.

The organoids proved remarkably faithful to what happens in actual human infection. When researchers introduced Ebola virus into the tissue, it replicated for up to 120 days—four months of continuous viral activity. The virus infected multiple cell types: neurons, astrocytes, and microglia, the brain's resident immune cells. It spread in two ways simultaneously—directly from one infected cell to its neighbor, and by budding from host cells in the classical manner. This was not a dormant virus hiding in a cellular vault. It was actively replicating, remaining fully infectious. The researchers call this "productive persistence," a state in which the virus maintains its ability to cause harm even as it evades complete elimination.

The organoids mounted an immune response. They produced pro-inflammatory cytokines, chemical signals that should have alerted the body's defenses to the threat. Yet the immune system could not finish the job. The virus persisted despite the inflammatory activity, and that mismatch—inflammation without successful viral clearance—mirrors what happens in surviving patients. Some develop meningoencephalitis, severe inflammation of the brain and its protective membranes, months after the acute phase has passed. The organoid model suggests this delayed inflammation is not a separate complication but a direct consequence of the virus's long-term presence in immune-privileged tissue.

How does Ebola accomplish this feat? The research team identified two key mechanisms. First, the virus generates defective genomes—incomplete or mutated versions of its genetic material that can suppress its own replication, allowing it to persist in a controlled, attenuated state rather than burning through the tissue and triggering immediate death. Second, as the virus replicates over weeks and months, its genetic machinery—unlike human machinery—cannot proofread its own copying. Mutations accumulate. Some of these mutations have been observed in naturally infected patients and are thought to reduce viral replication. Others were entirely new, never before documented in human survivors. Whether these novel mutations actively drive persistence or merely result from it remains an open question.

The significance extends beyond understanding Ebola alone. The organoid model works for other filoviruses too—Sudan virus, Reston virus, and Marburg virus all replicated in the tissue. This suggests the persistence mechanism may be fundamental to the entire family of filoviruses, not unique to Ebola. The model also offers a practical advantage: it allows researchers to study human neural tissue without relying on animal experiments, potentially reducing the use of primates and other animals in infectious disease research. And it creates a platform for testing antivirals and other treatments in a human context before they reach clinical trials.

Lina Widerspick, the first author of the study, notes that the organoids enable detailed investigation of mechanisms that might otherwise remain hidden. Gustavo Palacios, a co-author and expert in Ebola genomics, emphasizes that the virus behaves in these organoids much as it does in actual human infections, validating the model's relevance. The next steps are clear: deeper investigation of the long-term virus-host interactions, expansion to less-studied filoviruses like Reston, Taï Forest, Bombali, and Bundibugyo virus, and a push to understand whether the newly discovered mutations are causally linked to persistence or merely passengers along for the ride. The work opens a window into why some diseases never truly leave.