Back

Fasting disrupts the InsP6 HDAC3 axis to drive ER stress-mediated clearance of DNA-damaged cells and enforce tissue quality control.

Chetterjee, S.; Sin, Z.; Van, R. A.; Tran, N.; Tran, T.; Shukla, A.; Jeusset, L. M.-P.; Koshkaryan, G.; Saldana, K.; Muneer, R.; Viera Preval, L.; Gu, L.; Bhat, F.; Ritter, K.; Huang, K.; Huff-Hardy, K.; Rood, R.; Gremida, A.; Gregory, M.; Chen, C. H.; Weng, M.; Dekker, F. J.; Deepak, P.; Jessen, H. J.; McManus, K. J.; Han, M. V.; Guha, P.

2026-05-27 cell biology
10.64898/2026.05.24.727426 bioRxiv
Show abstract

Fasting drives metabolic adaptation but also elicits acute cellular stress. How this stress shapes tissue integrity is unknown. Here, we show that in the intestine, fasting depletes growth factor signaling, which triggers cellular stress. This response functions as a tissue quality-control checkpoint that selectively eliminates pre-existing DNA-damaged cells while sparing healthy counterparts. A short-term fast diminishes TGF-{beta} signaling and elicits endoplasmic reticulum (ER) stress, driving DNA-damaged intestinal cells beyond an apoptotic threshold, thereby reducing the inflammatory burden. Mechanistically, loss of TGF-{beta} signaling triggers FBXO22-Cullin1-mediated degradation of the inositol kinase IPMK, leading to depletion of inositol hexaphosphate (InsP). InsP loss attenuates HDAC3 activity and initiates coordinated epigenetic and post-translational reprogramming, thereby increasing CDK5RAP3 abundance. Elevated CDK5RAP3 inhibits ribosomal RPL26 UFMylation, thereby amplifying ER stress and selectively licensing apoptosis in DNA-damaged cells. Collectively, fasting disrupts a TGF-{beta}-InsP6-HDAC3 axis to drive ER stress-dependent clearance of DNA-damaged cells, enforcing tissue quality control.

Matching journals

The top 6 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.