Fig 1.
ULK1 undergoes electrophoretic mobility shift and is highly phosphorylated in mitosis.
(A) ULK1-ATG13 shows a mobility shift in mitosis. HeLa cells synchronized by double-thymidine release in the presence or absence of nocodazole were subjected to SDS-PAGE and western blot analysis. (B) ULK1-ATG13 undergoes band shift during mitotic progression. HeLa cells synchronized by double-thymidine and RO-3306 were released into mitosis for western blot analysis. (C) ULK1 is phosphorylated in nocodazole-arrested mitosis. The 293T cells overexpressing FLAG-tagged mULK1 or GFP were synchronized by single-thymidine and nocodazole. The immunoprecipitates using the FLAG antibody were subjected to Coomassie Brilliant Blue R-250 staining and western blot analysis. Statistical analysis for relative serine/threonine phosphorylated ULK1 was shown in Fig 1C, lower panel. n = 4, **p < 0.01. (D) The immunoprecipitates from Fig 1C were treated with or without λPP in the presence or absence of PPIs and then subjected to western blot analysis. (E) ULK1 undergoes mobility shift in both nocodazole- and STLC-arrested mitosis. HeLa cells synchronized with single-thymidine and nocodazole or STLC were analyzed by western blot analysis. The upper panel shows the immunoblotting and the lower panel shows the ratio of upshifted and nonshifted ULK1. One-way ANOVA followed by Tukey’s multiple comparison test was used for the analysis. n = 5, ***p < 0.001. (F) ULK1 undergoes phosphorylation-induced electrophoretic mobility shift in single-thymidine and nocodazole synchronized mitotic HCT 116 and RPE1 cells analyzed by western blot analysis. Numerical data underlying the figure panels are available in S1 Data. Asyn, asynchronous; ATG, autophagy-related; DT-R2h, double-thymidine block and release for 2 hours; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFP, green fluorescent protein; IP, immunoprecipitation; mULK1, mouse ULK1; n.s., not significant; HCT 116, human colorectal cancer cells; PPI, phosphatase inhibitor; RPE1, human retinal pigmented epithelial cells; STLC, S-trityl-L-cysteine; ULK1, unc-51-like autophagy activating kinase 1; λPP, lambda phosphatase.
Fig 2.
ULK1-ATG13 can be recognized by CDK substrate phosphorylation antibodies.
(A-B) ULK1-ATG13 upshifted band in mitosis could be recognized by CDK substrate-specific antibodies. The 293T cells stably expressing FLAG-tagged mULK1 or ATG13 were synchronized by single-thymidine and released in the presence or absence of nocodazole. The coimmunoprecipitates and input were immunoblotted with specific antibodies. The phospho-signal was quantified relative to the FLAG antibody–detected bands (mULK1/ATG13-3FLAG) in IP; the IP signal of mULK1/ATG13-3FLAG was quantified relative to its input. Various cell cycle markers were detected to show the respective phases of cell cycle. ATG, autophagy-related; CDK, cyclin-dependent kinase; FIP200, FAK family-interacting protein of 200 kDa; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFP, green fluorescent protein; IP, immunoprecipitation; mULK1, mouse ULK1; p-MAPK, phospho mitogen-activated protein kinase; ULK1, unc-51-like autophagy activating kinase 1.
Fig 3.
ULK1 is a substrate of CDK1/cyclin B in mitosis.
(A) Illustration of cell synchronization and kinase inhibitors treatment. HeLa cells were synchronized with single-thymidine and nocodazole. Kinase inhibitors were added for different timepoints, ranging from 3 minutes to 1.5 hours. (B) CDK1 inhibitors, but not Aurora kinase, mTORC1, and other CDK inhibitors, abolished the ULK1 band shift in mitosis. HeLa cells synchronized and treated as (A) were subjected to western blot analysis. (C-D) ULK1 coimmunoprecipitates with CDK1 and vice versa. 293T cells stably overexpressing FLAG-tagged mULK1 or CDK1 were synchronized with single-thymidine and nocodazole and the coimmunoprecipitates were subjected to western blot analysis. (E) ULK1 is upshifted and phosphorylated by purified CDK1/cyclin B complex in vitro. Purified CDK1/cyclin B complex as kinase and the ULK1 immunoprecipitates from asynchronous 293T cells overexpressing FLAG-tagged mULK1 as substrate with or without RO-3306 were subjected to in vitro kinase assay and western blot analysis. CDK, cyclin-dependent kinases; CST, Cell Signaling Technology; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFP, green fluorescent protein; IP, immunoprecipitation; mTORC1, mammalian target-of-rapamycin complex 1; mULK1, mouse ULK1; PVDF, polyvinylidene fluoride; ULK1, unc-51-like autophagy activating kinase 1; λ-PP, lambda phosphatase.
Fig 4.
ATG13 is a substrate of CDK1/cyclin B in mitosis.
(A) Mitotic ATG13 undergoes mobility upshift in mitosis in electrophoresis. The 293T cells overexpressing FLAG-tagged mULK1 or GFP were synchronized by single-thymidine in the presence or absence of nocodazole and coimmunoprecipitated by the FLAG antibody followed by western blot analysis. Size of endogenous ATG and expressed FLAG-tagged ATG13 were not distinguishable in electrophoresis here. (B) The effect of AMPK inhibition on ULK1 and ATG13 mobility shift in mitosis. The AMPK inhibitor Compound C was used to evaluate the role of AMPK on ULK1 or ATG13 mobility shift as Fig 3B. Phospho-AMPKα-T172 and phospho-Acetyl-CoA Carboxylase (Ser79) were used to indicate AMPK inhibition. (C) CDK1 inhibitor RO-3306 decreases the ATG13 band shift in mitosis. HeLa cells synchronized and treated as Fig 3A were subjected to western blot analysis. (D) ATG13 is phosphorylated and interacts with CDK1/cyclin B1 in mitosis. The 293T cells overexpressing FLAG-tagged ATG13 were synchronized by single-thymidine in the presence or absence of nocodazole. The coimmunoprecipitates by FLAG antibody were subjected to western blot analysis. (E-F) ATG13 is directly phosphorylated by purified CDK1/cyclin B complex in vitro. The ATG13 immunoprecipitates from asynchronous 293T cells overexpressing FLAG-tagged ATG13 or ULK1 and purified CDK1/cyclin B complex were subjected to in vitro kinase assay with or without RO-3306 followed by western blot analysis. (E) shows representative western blot analysis of the ATG13 immunoprecipitates as substrate, and (F) shows the representative western blot analysis of the ULK1-WT coimmunoprecipitates as substrate. AMPK, AMP-activated protein kinase; ATG, autophagy-related; CDK, cyclin-dependent kinase; FIP200, FAK family-interacting protein of 200; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GFP, green fluorescent protein; IP, immunoprecipitation; mULK1, mouse ULK1; ULK1, unc-51-like autophagy activating kinase 1; WT, wild type.
Fig 5.
ULK1 and ATG13 phosphorylation sites in mitosis by CDK1/cyclin B.
(A) The mitotic-specific phosphorylated sites identified by MS in mitotic mULK1 compared with asynchronous mULK1. (B-C) S622/T635/T653 phosphorylation contributes to ULK1 mobility shift in mitosis. The 293T cells overexpressing FLAG-tagged mULK1-S622/T635/T653 mutants were synchronized with single-thymidine and nocodazole. Then immunoprecipitation for single mutant with FLAG antibody (B) or western blot analysis for double and triple mutant (C) was performed. (D-E) More sites contribute to ULK1 band shift. Based on triple 3A mutant, the other 8 Ser/Thr sites were mutated into Ala. The 293T cells expressing various mutants were synchronized into mitosis with thymidine and nocodazole for immunoprecipitation by FLAG antibody and western blot analysis. (F) ULK1-11A mutant was not upshifted or phosphorylated by CDK1/cyclin B kinase complex in vitro. The ULK1-11A mutant immunoprecipitates from asynchronous 293T cells overexpressing FLAG-tagged mULK1 and purified CDK1/cyclin B complex were tested in an in vitro kinase assay and western blot analysis. (G) The phosphorylated sites identified by MS in mitotic ATG13 compared with asynchronous ATG13. (H-I) ATG13-T342/T332/S44/S224 phosphorylation contributes to ATG13 mobility shift in mitosis. HeLa-ATG13 KO cells (H) or 293T cells (I) overexpressing FLAG-tagged ATG13-T342/T332/S44/S224 mutants were synchronized with single-thymidine and nocodazole. Then western blot analysis for 4-site mutant (H) or immunoprecipitation and input for mutants with FLAG antibody (I) was performed. 3A, S622/T635/T653A; 5A, 3A-S479&S543A; 7A, 5A-S411&S413A; 9A, 5A-S413&T401&S403&S405A; 10A, 9A-T282A; 11A, 10A-T502A; ATG, autophagy-related; CDK, cyclin-dependent kinase; GFP, green fluorescent protein; IP, immunoprecipitation; MS, mass spectrometry; mULK1, mouse ULK1; ULK1, unc-51-like autophagy activating kinase 1; WT, wild type.
Fig 6.
ULK1-ATG13 phosphorylation in mitosis promotes mitotic autophagy.
(A) The localization of mutant ULK1-11A and ATG13-4A in mitosis. Cells released from thymidine block for 10 hours were fixed with 3.7% formaldehyde for 20 minutes at room temperature and then for blocking and FLAG antibody, Alexa-488 conjugated secondary antibody, DAPI staining. Scale bar, 10 μm. (B) The ULK1 kinase activity is not affected by the 11A mutation. ATG13-Ser355 is the substrate of ULK1, which corresponds to Ser318 of ATG13 isoform 2. ULK1-KO 293T cells reconstituted with ULK1-WT, ULK1-11A, or ULK1-K46I were plated and lysed with M-PER. The western blot for ATG13-Ser355 indicated that ULK1 activity was not affected by 11A mutation. n = 3. (C) Autophagy activity was examined in ULK1&ATG13 double WT or mutant (“Mut”) cells expressing GFP-LC3-RFP by flow cytometry. ULK1&ATG13 double WT or mutant cells, HeLa, or HeLa- ULK1&ATG13 DKO cells stably expressing GFP-LC3-RFP were established by infection of retrovirus packaged by pMRX-IP-GFP-LC3-RFP and helper plasmids Vsvg and pMLV. Mitotic cells were collected by shake-off from thymidine and nocodazole synchronized cells for flow cytometry detection. The GFP/RFP ratio inversely correlated with autophagy activity. n = 4, ***p < 0.001, ****p < 0.0001. (D) The autophagic flux of HeLa-DKO cells stably overexpressing double WT or mutant FLAG-tagged mULK1 and ATG13. Cells synchronized to mitosis with thymidine and nocodazole were shaken off and treated with or without autophagy inhibitor 25 μM CQ for 1 hour. The left panel is a representative western blot, and the right panel is the statistical result for LC3B-II and p62. n = 3, *p < 0.05, **p < 0.01. (E) LC3B puncta number was counted, and the micrographs were captured by Zeiss LSM710 confocal microscope. Scale bar, 10 μm. n = 43, **p < 0.01. Numerical data underlying the figure panels are available in S1 Data. 4A, T342/T332/S44/S224A; 11A, S622&T635&T653&S479&S543&S413&T401&S403&S405&T282&T502A; ATG, autophagy-related; CQ, chloroquine; Ctrl, control; DAPI, 4 DAPdiamidino-2-phenylindole; DKO, double knockout; GFP, green fluorescent protein; KO, knockout; M-PER, Mammalian Protein Extraction Reagent; mULK1, mouse ULK1; n.s., not significant; RFP, red fluorescent protein; ULK1, unc-51-like autophagy activating kinase 1; WT, wild type.
Fig 7.
ULK1-ATG13 and their mitotic phospho-regulation by CDK1 are required for cell cycle progression.
(A) ULK1-ATG13 DKO inhibits S/G2 and G2/M transitions. HeLa WT or ULK1&ATG13-DKO cells synchronized into mitosis were subjected to PI and pH3(S10) co-staining for cell cycle and mitotic index analysis by flow cytometry. n = 3, *p < 0.05, **p < 0.01. (B) Representative western blot analysis suggests that ULK1&ATG13 DKO inhibits mitotic entry, which is shown by mitotic markers and CDK1 substrate phosphorylation. (C) Mitotic exit of ULK1, ATG13, or ULK1&ATG13 KO cells. Cells were synchronized into mitosis with thymidine and NOC and released into NOC-free complete DMEM medium for different timepoints and then subjected to either PI and pH3(S10) co-staining or cyclin B1 staining for cell cycle, mitotic index, and cyclin B1 level analysis by flow cytometry. n = 3, *p < 0.05, **p < 0.01. (D) ULK1&ATG13 DKO inhibits cell proliferation. HeLa WT or ULK1&ATG13 DKO cells were plated at 1 × 105 cells/mL and cultured for 1, 2, or 3 days. The cell number was counted by flow cytometry and the doubling time was calculated. TD indicates the average cell doubling time and is calculated as: TD = t*[lg2/(lgNt − lgN0)], where t is the culture time, Nt is the cell number after culturing, and N0 is the original cell number plated. n = 3, *p < 0.05, ***p < 0.001. (E-G) The role of ULK1-ATG13 phosphorylation in cell cycle progression and cell proliferation. ULK1&ATG13-DKO cells reconstituted with WT mULK1-3FLAG and ATG13-3FLAG or mutant (“Mut”) mULK1-11A and ATG13-4A, HeLa, or ULK1&ATG13-DKO cells were treated as (A)/(B) and (D). The mitotic exit (G1%) is the percentage of cells in G1 phase when released for indicated time. n = 3, **p < 0.01, ***p < 0.001. (H) The relative tumor volume growth curve of nude mice bearing different tumors with or without SBI-0206965. The nude mice were injected with cells (1 × 107) in 100 μL PBS/Matrigel Matrix (1:1). Seven days postimplantation, 5 mice in each group were injected with 0.5% (M/V) methyl cellulose or SBI-0206965 in 0.5% methyl cellulose (20 mg/kg/d) every day for 33 days. Tumor growth was evaluated every day and tumor volume was calculated as: volume = 1/2 (length × width2). (I) Tumor growth in nude mice bearing WT or KO cells with or without ULK1 kinase inhibitor SBI-0206965. The protocols were indicated as Fig 7H and the mice were sacrificed, and tumors were harvested and weighed up at the end of the experiment. n = 5, *p < 0.05, ***p < 0.001. Numerical data underlying the figure panels are available in S1 Data. 4A, T342/T332/S44/S224A; 11A, S622&T635&T653&S479&S543&S413&T401&S403&S405&T282&T502A; ATG, autophagy-related; CDK, cyclin-dependent kinase; DKO, double knockout; DMEM, Dulbecco’s Modified Eagle Medium; KO, knockout; mULK1, mouse ULK1; Myt1, myelin transcription factor 1; NOC, nocodazole; n.s., not significant; pH3(S10), phospho histone H3 serine 10; PI, propidium iodide; ULK1, unc-51-like autophagy activating kinase 1; WT, wild type.