Back

Endothelial gain of PIEZO/TRPV mechanosensors drives cavernous malformations

Faurobert, E.; de Jong, J.; Pasquier, C.; Manicom Ramsamy, T.; Ahmed, A.; Onisiforou, A.; Roedel, C. J.; Senger, F.; Bouin, A.-P.; Leverrier-Penna, S.; Rey, E.; Vorsselmans, S.; Poveda Vicedo, C.; Bryant, E.; Shapeti, A.; Kroenke, N.; Feuerhake, F.; Hartmann, C.; Petrakakis, I.; Oddou, C.; Linhares, D.; Cardinaels, R.; Planus, E.; Rocha, S.; Samii, A.; Bertalanffy, H.; destaing, o.; Abdelilah-Seyfried, S.; Penna, a.; Kar, S.; Van Oosterwyck, H.

2026-10-05 cell biology
10.64898/2026.10.01.756057 bioRxiv
Show abstract

Vascular disease often begins as a failure to read mechanical force. Cerebral cavernous malformations (CCMs) are leaky, brain-invading capillary clusters that cause stroke and seizures and arise from endothelial loss of the CCM complex. They form exclusively in low-flow vessels but existing models cannot explain this selectivity. We previously established that CCM-mutant endothelium inverts its adhesive balance, weakening cell-cell junctions while building matrix-degrading adhesions. The mechanosensors imposing this switch are unknown. Here we show that the mechanosensitive Ca2+ channels PIEZO1, TRPV2 and TRPV4 are overexpressed in the endothelium of CCM lesions in a 42-patient cohort, and that lesion volume scales with their expression level. Mechanistically, PIEZO1 partners with TRPV4 to destabilise cell-cell junctions and with TRPV2 to reinforce cell-matrix adhesions; silencing TRPV4 restores junctional integrity and abolishes KLF4 induction, whereas silencing TRPV2 or inhibition with tranilast abolishes matrix degradation and invasion. In vivo, trpv4 knockdown or inhibition with GSK2798745 restores atrioventricular canal morphogenesis in ccm2-mutant zebrafish. Gain of mechanosensory channel function is thus a causal step in CCM pathogenesis, upstream of the KLF program. Combined therapeutic targeting of TRPV2 and TRPV4 with inhibitors validated in humans offers a route to therapy.

Matching journals

The top 3 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.