Back

Photocaged chloroquine derivatives for the light-dependent inhibition of autophagy in cancer stem cells

Alonso-Manresa, S.; Serra, C.; Munoz, L.; Bataller, M.; Garcia-Mayea, Y.; Pajarin, M. E. L.; Prats, B. G.; Zamora, S. M.; Riascos, Z. V. D.; Llebaria, A.; Josa-Cullere, L.

2025-11-27 cancer biology
10.1101/2025.11.24.690081 bioRxiv
Show abstract

Chloroquine (CQ) and hydroxychloroquine (HCQ) inhibit autophagy and have shown promise as adjuvant anticancer agents, particularly for targeting therapy-resistant cancer stem cells (CSCs). However, their clinical utility is limited by systemic toxicity and poor tumour selectivity. Here we report the design, synthesis, and photochemical evaluation of [7-(diethylamino)coumarin-4-yl]methyl (DEACM)-caged CQ and HCQ derivatives as visible-light-activated autophagy inhibitors. Selective caging of the aliphatic amine fully suppressed biological activity in the dark and enabled rapid, efficient release of the parent drugs upon illumination. The lead compound 1C displayed robust light-dependent cytotoxicity across multiple cancer cell lines and, upon photoactivation, recapitulated CQs effects on LC3-II accumulation. In CSC-enriched tumourspheres, illumination of 1C completely abolished spheroid formation, demonstrating precise spatiotemporal control of stemness suppression. Importantly, ex vivo and in vivo studies confirmed that visible light penetrates tumour tissue sufficiently to activate 1C and release CQ within the tumour. These findings establish the first proof of concept for light-controlled autophagy inhibition and provide a blueprint for spatiotemporally confined anticancer therapies based on photopharmacological modulation of CSCs.

Published in ACS Chemical Biology (predicted rank #5) · training set

Matching journals

The top 9 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.