Discovery reveals potent anti-ovarian cancer agents targeting tubulin via colchicine site, indicating deep learning's role in drug design.
To address the toxicity of current microtubule inhibitors, we employed the GeminiMol deep learning model to screen the Zinc20 database, identifying a novel 4,5-dihydropyrrolo[3,4-c]pyrazol-6(2H)-one scaffold (Y1) targeting the colchicine binding site. Subsequent optimization culminated in Y60s, a potent antiproliferative agent (SKOV3 IC50 = 0.025 μM) that inhibits clonogenic formation, migration, and invasion of ovarian cancer cells. Y60s inhibited tubulin polymerization which in turn induced G2/M arrest and apoptosis in SKOV3 cells. Y60s demonstrated potent antitumor activity without observable toxicity in an SKOV3 xenograft model. The cocrystal structure of Y8 in complex with tubulin was resolved, confirming the key binding mode of 4,5-dihydropyrrolo[3,4-c]pyrazol-6(2H)-one compounds. This work showcases Y60s as a promising novel tubulin inhibitor and highlights the utility of deep learning model for rapid identification of bioactive compounds from large chemical databases.
No takes yet. Share an insight, caveat, or question.
Yu et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: