Alexei G. from (KO) mice showed decreased LC3B accumulation, increased SQSTM1 expression, and reduced the number of autophagic vesicles (indicated by the electron microscopic images) under starvation treatment (Fig.?2aCd). In homolog significantly decreased PRKCA the level of cleaved GFP, which serves as a degradation product in autolysosomes18, under both normal and starvation conditions (Fig.?2f). Moreover, loss of suppressed the number of GFP::LGG-1 (LC3 homolog) dots, which was observed in the seam cells and pharyngeal muscle upon starvation (Fig.?2gCh and Supplementary Fig.?4a). These results suggest that ENDOG-promoted autophagy is conserved across species, including mouse, or mouse livers after starvation for 24?h (or mouse livers after starvation for 24?h (LD: lipid drop; N: nuclear; red arrow: autophagic vesicle; fat body cells after starvation for 4?h (GFP-NLS-labeled cells circled by dotted line express RNAi targeting ENDOG. Cells outside the circled dotted line are wild-type and used as controls; scale bar?=?10?m). f Representative western blots of GFP-LGG-1 and ACTB in control (GFP::LGG-1) and ENDOG loss function mutant (after starvation for 4?h. Graph, quantification of cleaved-GFP (correspond to a product of degradation in autolysosomes) ((white arrows: GFP-LGG-1 puncta; and are generated preferentially when ENDOG fragments double-stranded DNA33. We suspected that ENDOG nuclease activity might induce the DNA damage response pathway. We demonstrated that ENDOG promoted DNA damage both under normal and stress conditions (Supplementary Fig.?10C13). DNA damage has been reported as an Lck inhibitor 2 early event during starvation-induced autophagy. Under starvation, PARP-1/AMPK and ATM/CHK2 pathways were activated and eventually promoted autophagy21,22. In the present study, ENDOG promoted starvation-induced DNA damage through PARP-1/AMPK pathway, which could repress the mTOR activity and initiate autophagy (Fig.?5). In contrast to the wild-type ENDOG, the endonuclease activity deficient form of ENDOG could not activate the PARP-1/AMPK pathway or induce mTOR repression and autophagy promotion (Supplementary Fig.?13d, e). In addition, recent study showed that DNA damage induced CHK2-mediated FOXK phosphorylation, which traped FOXK in the cytoplasm through interacting with 14-3-3, resulting in promoting the transcriptional of autophagy-related genes (ATGs) and facilitating Lck inhibitor 2 autophagy21. However, the transcription of autophagy-related genes was?not affected in ENDOG overexpression or knockout cells (Supplementary Fig.?1c, d). This suggested that ENDOG-mediated DNA damage response promotes autophagy in a CHK2-FOXK-axis independent manner. Previous studies showed that DNA damage agents (such as camptothecin, etoposide, and Lck inhibitor 2 temozolomide)34,35 and ionizing radiation36 promoted autophagy initiation. Consistently, we found etoposide treatment promoted autophagosome formation (GFP-LC3 dots) both in wild-type and ENDOG-overexpressing cells (Supplementary Fig.?10). Moreover, by using the ATM Lck inhibitor 2 inhibitor KU-60019, which only inhibits ATM but not ATR, to block the?DNA damage response, ENDOG-induced DNA damage and autophagosome formation could be partially repressed (Fig.?6hCk and Supplementary Figs.?10, 11). Additionally, ENDOG without endonuclease activity did not induce the?DNA damage response or autophagy (Supplementary Figs.?12, 13). Chemical inhibition of ENDOG activity also repressed DNA damage and autophagy (Supplementary Fig.?14). These data suggest that ENDOG-promoted autophagy is endonuclease activity-dependent. ENDOGs N-terminus (1C48 aa) is a mitochondrial targeting sequence (MTS) that is removed after ENDOG is imported into the mitochondria13,37. Interestingly, we found that ENDOG without MTS failed to induce DNA damage response, but it could still promote autophagy (Supplementary Fig.?12). Therefore, we propose that ENDOG without MTS lost its ability to cause DNA damage in nuclei, because it did not mature properly without being imported into mitochondria13 or because it was affected in another unknown way. However, ENDOG without MTS maintained its binding affinity with 14-3-3, so it can still repress mTOR pathway and promote autophagy initiation. This may be the reason why Del 1C48-ENDOG did not induce DNA damage but still promoted autophagy under the etoposide treatment (Supplementary Fig.?12). In summary, our findings indicate that ENDOG conservatively promotes autophagy in multiple species. In mammalian cells, ENDOG promotes autophagy through the suppression of mTOR by its phosphorylation-mediated interaction with 14-3-3 and its endonuclease activity-mediated DNA damage response. Methods Animals The ENDOG-knockout mice (C57BL/6) were.