Back

A scalable human-zebrafish xenotransplantation model reveals gastrosome-mediated processing of dying neurons by human microglia

Villani, A.; Wittmann, J.; Wyss, T.; Mallona Gonzalez, I.; Santisteban Ortiz, I.; Tichy, N.; Pena, M.; Pal, A. A.; Gilmour, D.; Schafer, S.; Peri, F.

2025-08-28 cell biology
10.1101/2025.08.27.672578 bioRxiv
Show abstract

Microglia engulf dying neurons through efferocytosis, a critical function in both development and disease. How microglia process the engulfed neuronal material--especially lipids--remains poorly understood, despite its central role in neurodegeneration. Thus, we developed HuZIBRA, a scalable in vivo xenotransplantation model in which human iPSC-derived microglia-like cells (iMGLs) are introduced into the developing zebrafish brain (zf-hiMG), a system characterized by high levels of neuronal cell death and amenable to precise genetic and pharmacological manipulation. We show that human microglia-like cells recognize and engulf apoptotic zebrafish neurons, indicating conserved efferocytic mechanisms. In these cells, engulfed neuronal material accumulates into a distinct, lipid-rich intracellular compartment, the gastrosome, which we also observed in iMGLs placed in a human brain-like environment. The size of the human gastrosome dynamically reflects neuronal cell death levels and is regulated by key genes, including TREM2 and SLC37A2. Pharmacological inhibition of the cholesterol transporter NPC1 induces gastrosome expansion and lipid accumulation, recapitulating pathological features of Niemann-Pick disease type C. Thus, HuZIBRA provides a powerful in vivo platform to uncover cell-autonomous adaptive responses of human microglia to apoptotic and metabolic stress, with the gastrosome emerging as a key integrator of neuronal debris processing and disease-relevant lipid metabolism.

Published in Communications Biology (predicted rank #15) · training set

Matching journals

The top 5 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.