DNA methylation protects cancer cells against non-canonical senescence

Aberrant DNA methylation patterns are a hallmark of human cancers, prompting targeting of DNA methylation machinery to be explored as an anti-tumor strategy. DNA methyltransferase 1 (DNMT1) and ubiquitin like With PHD and Ring finger domains 1 (UHRF1) are crucial enzymes for DNA methylation maintenance. Unlike DNMT1, UHRF1 is overexpressed in most cancers and is a proven oncogene in vivo. Our previous work revealed that the degree of DNA methylation loss induced by UHRF1 depletion is more severe than DNMT1 depletion in colorectal cancer cells, indicating that targeting UHRF1 may be a more effective approach. Using an advanced chemical/genetic system, the auxin-inducible degron (AID) technology, we monitored the long-term effects of removing UHRF1 and/or DNMT1 in cancer cells, deciphering the mechanistic basis through bioinformatic tools. We found that chronic DNA demethylation triggers cancer cells to undergo cellular senescence. The loss of UHRF1 results in a more rapid and profound senescence phenotype than DNMT1 loss. Intriguingly, this senescence is not accompanied by DNA damage and functions independently of canonical senescence pathways such as p53 and p16/pRB. Non-canonical pathway investigations revealed that cytosolic p21 accumulation contributes to antiapoptosis during senescence. The suppression of cyclic GMP-AMP synthase (cGAS) also alleviates the senescence-associated secretory phenotype (SASP) and lysosomal activity, and this process is independent of stimulator of interferon genes (STING). Finally, xenograft experiments verified our findings in vivo. Our results highlight how DNA methylation protects cancer cells from non-canonical senescence and demonstrate the potential of targeting UHRF1 as a novel avenue for anticancer treatments.