Back

Microbiota-derived extracellular vesicles link intestinal dysbiosis to neuroimmune activation in long COVID

Aranguren, M.; Doyon-Laliberte, K.; Boncheva, I.; Villard, A.; Desjardins, A.; Darbinian, E.; Patel, S.; DuSablon, C.; Rivera Conde, E.; Cabrera Munoz, D.; Purchase, L.; Piscopo, V. E. C.; Alluli, A.; Benaliouad, F.; Sirois, J.; Durcan, T.; Masse, C.; Mlaga, K. D.; Chandrasekaran, P.; Poudrier, J.; Falcone, E. L.

2026-03-02 microbiology
10.64898/2026.02.28.708602 bioRxiv
Show abstract

Post COVID-19 condition (Long COVID, LC) is frequently accompanied by persistent neurological symptoms, but the mechanisms linking intestinal dysbiosis to neuroinflammation remain unclear. Here we identify gut microbiota-derived extracellular vesicles (GMEVs) as functional mediators linking LC-associated dysbiosis to systemic and neuroimmune inflammation. In a longitudinally characterized cohort, individuals with LC and neurological symptoms exhibit a persistent intestinal microbiome signature. Transplantation of LC-associated microbiota into germ-free mice induces intestinal barrier disruption and neuroinflammatory phenotypes. GMEVs from individuals with LC activate inflammasome-associated programs and impair epithelial barrier function, promote inflammatory responses in macrophages, and induce coordinated pro-inflammatory transcriptional programs in human induced pluripotent stem cell (iPSC)-derived microglia. Chronic oral administration of LC-derived GMEVs remodels the microbiota and induces intestinal and systemic inflammation with glial activation in vivo. Together, these findings support a vesicle-centered framework in which microbiota-derived extracellular vesicles translate dysbiosis into sustained immune and neuroimmune activation in a post-viral inflammatory state.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.