Back

Disrupted astrocyte-endothelial crosstalk drives hemangioblastoma lesions in VHL disease

De Andres-Laguillo, M.; Garcia-Gonzalez, I.; Rocha, S. F.; Garcia-Cabero, A.; Ruiz-Garcia, S.; Diago-Domingo, L.; Dadana, A.; Cusso, L.; De Bock, K.; Enriquez, J. A.; Benedito, R.

2025-09-18 cancer biology
10.1101/2025.09.15.675862 bioRxiv
Show abstract

Hemangioblastomas (HBs) are highly vascularized central nervous system (CNS) tumours that can become life-threatening, especially in the context of Von Hippel-Lindau (VHL) disease, caused by the loss of VHL function. The limited pharmacological options targeting VHL-HBs stem from an incomplete understanding of their cellular origin, development, and molecular pathogenesis. Here we use advanced mouse genetics to show that mosaic deletion of Vhl in Apln+ cells leads to the formation of precursor tumour-like lesions, composed by clusters of Vhl-knockout (VhlKO) astrocytes and surrounding Vhl-wild-type (VhlWT) vessels that become malformed, resembling early-stage HBs linked to VHL disease. VhlKO astrocytes morphologically and transcriptomically resembled the reactive astrocytes characteristic of ischemic CNS injury. They exhibited metabolic rewirement towards glycolysis and upregulation of cell growth pathways. They also expressed several secreted proangiogenic molecules that activate and prevent the normal maturation of neighbouring vessels, leading to VHL-HBs. Temporal conditional genetic analysis revealed that Vhl loss need to happen during postnatal development for HBs to form, and that lesions become quiescent in early adulthood. HIF-2 deletion, or MTORC1 inhibition with rapamycin, efficiently inhibited VHL-HBs growth and the associated vascular malformations. Our work shows that the loss of Vhl in single astrocytes induces their growth and pathogenic crosstalk with neighbouring endothelial cells, driving hemangioblastoma development in VHL disease. Our new somatic mosaic mouse models will also enable testing of novel drugs against this disease.

Matching journals

The top 4 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.