Back

Combined celecoxib and NDGA induces a non-canonical, autophagy-independent redox cell death via p62/KEAP1 in colorectal cancer

Yadala, R.; palakurthi, S.; Pawar, S.

2026-10-06 cancer biology
10.64898/2026.09.29.753356 bioRxiv
Show abstract

Colorectal cancer (CRC) exhibits substantial intratumoral heterogeneity, with differences in oncogenic driver mutations and apoptotic competence contributing to therapeutic resistance. Here, we investigated whether combined celecoxib and NDGA treatment (C+N) induces redox-mediated cell death via the p62/KEAP1 axis, offering a therapeutic approach that can overcome genetic diversity in CRC. We used two genetically distinct CRC cell lines, HCT116 and HT29, harbouring KRAS and BRAF mutations respectively, with differing p53 status, as models of tumor heterogeneity. C+N exhibited profound cytotoxicity in both cell lines, independent of COX-2/5-LOX expression and apoptotic competence. Additionally, C+N induced marked oxidative stress with mitochondrial ROS, lipid peroxidation, and labile ferrous iron accumulation, visualized as distinct JC-1 green mitochondrial puncta and FerroOrange red iron puncta in both cell lines. The antioxidant NAC reversed C+N-induced death, but inhibitors of apoptosis, autophagy, necroptosis, or ferroptosis did not, indicating a mechanism distinct from these pathways. Mechanistically, the combination induced p62 Ser349 phosphorylation, promoting KEAP1 degradation and NRF2 stabilization independently of ULK1-mediated autophagy; however, NRF2 stabilization was uncoupled from HO-1 induction. Our work reveals that C+N induces a ULK1/autophagy-independent, redox-mediated cell death mechanism that circumvents apoptotic competence, highlighting redox disruption as a genotype-independent therapeutic strategy to overcome CRC heterogeneity.

Matching journals

The top 7 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.