Allosteric Antagonist Modulation of TRPV2 by Piperlongumine Impairs Glioblastoma Progression

Using machine learning to identify biological targets for natural products with anticancer properties and unknown modes of action is gaining momentum. Herein, we employ machine intelligence to deconvolute the phenotypic effects of the natural product Piperlongumine (PL) and establish an unprecedented link to allosteric modulation of the human transient receptor potential vanilloid 2 (hTRPV2) channel. Using cryo-electron microscopy (cryo-EM), we determined the structure of the PL-bound full-length rat TRPV2 channel at an overall resolution of 3.4 Å. We disclose binding of PL to a novel allosteric pocket which is responsible for a new mode of anticancer activity by PL, in particular against the aggressive glioblastoma (GBM), where TRPV2 is overexpressed. By creating CRISPRi TRPV2 knockdown GBM cells we found that the down-regulation of TRPV2 reduces sensitivity to PL and decreases ROS production. By analyzing GBM patients’ samples, we associated TRPV2 overexpression with tumor grade, disease progression and poor prognosis. Finally, formulating PL for sustained local therapy, we obtained tumor remission in a murine model of orthotopic GBM. Altogether, our strategy for target identification counter cycles established methods is broadly applicable and leverages data-motivated research hypotheses for the swift discovery of new biology and therapeutics. 12 Samples, wild type and TRPV2 CRISPRi KD and 2 treatments