Fine-tuning of ULK1 mRNA and protein levels is required for autophagy oscillation

Abstract

Autophagy is an intracellular degradation pathway whose levels are tightly controlled to secure cell homeostasis. Unc-51-like kinase 1 (ULK1) is a conserved serine-threonine kinase that plays a central role in the initiation of autophagy. Here, we report that upon autophagy progression, ULK1 protein levels are specifically down-regulated by the E3 ligase NEDD4L, which ubiquitylates ULK1 for degradation by the proteasome. However, whereas ULK1 protein is degraded, ULK1 mRNA is actively transcribed. Upon reactivation of mTOR-dependent protein synthesis, basal levels of ULK1 are promptly restored, but the activity of newly synthesized ULK1 is inhibited by mTOR. This prepares the cell for a new possible round of autophagy stimulation. Our results thus place NEDD4L and ULK1 in a key position to control oscillatory activation of autophagy during prolonged stress to keep the levels of this process under a safe and physiological threshold.

Figure 1.

ULK1 mRNA and protein levels are regulated in a controlled time course during autophagy. (A) HeLa cells were treated with EBSS for the indicated time periods in the presence or not of Clq, and the levels of ULK1, ACTIN, and LC3 were detected by WB. Densitometric analysis of ULK1 over ACTIN is also shown (right graph). Data are expressed as the mean ± SEM (n = 4). (B) qPCR analysis of ULK1 mRNA levels in HeLa cells incubated in EBSS medium for the indicated time periods. Data are expressed as mean ± SEM (n = 3). (C) qPCR analysis of ULK1 mRNA levels in HeLa cells incubated in EBSS medium for 4 h in the presence or not of Act D. Data are expressed as mean ± SEM (n = 3). (D) HeLa cells were treated with EBSS for the indicated time periods in the presence or not of MG132. Protein levels of ULK1 and TUBULIN were detected by WB. (E) HeLa cells were transfected with ubiquitin-HA and treated with EBSS for the indicated time periods in the presence of MG132. Protein extracts were immunoprecipitated in denaturing conditions using an anti-ULK1 antibody; ubiquitin, ULK1, and ACTIN were analyzed by WB. In A and B, data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey post hoc test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001. In C, data were analyzed by two-way ANOVA followed by Bonferroni’s multiple comparison post test. ***, P < 0.001.

Figure 2.

ULK1 interacts with the E3 ligase NEDD4L. (A) Protein extracts from HeLa cells were immunoprecipitated using anti-NEDD4L, anti-ULK1 antibodies and with IgG as a negative control. NEDD4L and ULK1 were analyzed by WB. The two NEDD4L bands correspond to two different spliced isoforms. (B) HeLa cells were transiently transfected with Myc-ULK1 and then grown in normal or EBSS medium for 1 and 3 h, respectively. Protein extracts were immunoprecipitated using an anti-Myc antibody or with IgG as a negative control; Myc-ULK1 and NEDD4L were analyzed by WB. (C) HeLa cells were grown either in normal or in EBSS medium for 90 min, respectively. Protein extracts were immunoprecipitated using an anti-NEDD4L antibody or with IgG as a negative control. ULK1 and NEDD4L analyzed by WB. The band density ratio of immunoprecipitated ULK1, relative to immunoprecipitated NEDD4L normalized on the input amount of ULK1, is analyzed (n = 3). Data are expressed as the mean ± SD, and statistical analysis was performed using an unpaired Student's t test. **, P < 0.01. (D) Scheme of NEDD4L full length and fragments showing the N-terminal C2 domain, four WW domains (black boxes), and the HECT domain. HeLa cells were cotransfected with vectors encoding Myc-ULK1 together with NEDD4L-HA fragments encoding the WW1, WW2, and WW3-4 domains, respectively. Protein extracts were immunoprecipitated using an anti-HA antibody; Myc-ULK1 and NEDD4L-HA fragments are analyzed by WB. Asterisk represents unspecific bands.

Figure 3.

NEDD4L ubiquitylates ULK1 and induces its degradation via proteasome. (A) HeLa cells were transfected with empty or NEDD4L-HA vectors; levels of ULK1, NEDD4L, and TUBULIN were detected by WB. Densitometric analysis of ULK1 over TUBULIN is also shown (right graph). Data are expressed as the mean ± SD (n = 3) and statistical analysis was performed using an unpaired Student's t test. **, P < 0.01. (B) HeLa cells were transfected with cDNAs coding for NEDD4L^WT and NEDD4L^CA HA-tagged proteins; ULK1, NEDD4L, and TUBULIN protein levels were detected by WB. Densitometric analysis of ULK1 over TUBULIN is also shown (right graph). Data are expressed as the mean ± SEM (n = 4). (C) HeLa cells were transfected with NEDD4L-HA in the presence or not of MG132 for 4 h. Levels of ULK1, NEDD4L, and TUBULIN were detected by WB. Densitometric analysis of ULK1 over TUBULIN is also shown (right graph). Data are expressed as the mean ± SEM (n = 3). (D) HeLa cells were cotransfected with a vector encoding a 6xHIS-tag ubiquitin and Myc-ULK1 in combination or not with NEDD4L^WT and NEDD4L^CA in the presence of MG132. Protein extracts were prepared in a denaturing urea buffer and subjected to Ni-NTA purification. The amount of ubiquitylated Myc-ULK1 copurified with 6xHIS-ubiquitin was evaluated by WB. (E) An in vitro ubiquitylation assay was performed by mixing immunopurified Myc-ULK1 and NEDD4L-HA, together with recombinant HIS-tag ubiquitin. ULK1 ubiquitylation is evaluated using an anti-Ubiquitin antibody to detect the incorporation of recombinant HIS-ubiquitin. The levels of ULK1 and NEDD4L were analyzed by WB. (F) HeLa cells were cotransfected independently with vectors encoding for 6xHIS-tag specific ubiquitin constructs and Myc-ULK1 in combination or not with HA-tagged NEDD4L^WT or NEDD4L^CA as a negative control. Protein extracts were prepared as in (D) and the amount of ubiquitylated ULK1 copurified with 6xHIS-Ubiquitin was evaluated by WB. The higher expression levels of NEDD4L^CA construct are caused by the lack of activity on itself (Bruce et al., 2008). In B and C, data were analyzed by one-way ANOVA followed by Tukey post hoc test. *, P < 0.05.

Figure 4.

NEDD4L degrades ULK1 during prolonged starvation. (A) NEDD4L expression was down-regulated in HeLa cells using specific RNAi oligonucleotides (oligos; siRNA NEDD4L#1) or unrelated oligos as negative control (siRNA CTRL). Densitometric analysis of ULK1 over ACTIN is also shown (right graph). Data are expressed as the mean value ± SD (n = 3), and statistical analysis was performed using an unpaired Student's t test. **, P < 0.01. (B) NEDD4L expression was down-regulated in HeLa cells using specific RNAi oligos targeting NEDD4L 3′UTR (siRNA NEDD4L#2) or unrelated oligos as negative control (siRNA CTRL). Some of them were then transfected with empty vector or NEDD4L-HA. Levels of ULK1, NEDD4L, and ACTIN were detected by WB. Densitometric analyses of both ULK1 and NEDD4L over ACTIN are also shown (right graphs). Data are expressed as the mean value ± SEM (n = 3). (C) NEDD4L expression was down-regulated in HeLa cells as in (A) and autophagy was induced by starving cells for the indicated time periods. Levels of ULK1 and ACTIN were detected by WB. Densitometric analysis of ULK1 over ACTIN is also shown (below graph). Data are expressed as the mean ± SEM (n = 3). Two different expositions (low and high) for ULK1 bands are shown. (D) HeLa cells were cotransfected with cDNAs coding for Myc-tagged ULK1^WT or ULK1^S1047A or ULK1^K46I together with NEDD4L^WT-HA in the presence or absence of MG132 or lactacystin for two different time periods (1 h and 2 h). Levels of ULK1, NEDD4L, and ACTIN were detected by WB. Densitometric analysis of ULK1 over ACTIN is also shown (below graph). Data are expressed as the mean ± SEM (n = 3). (E) HeLa cells were cotransfected with a vector encoding 6xHIS-tag Ubiquitin and in some of them with NEDD4L-HA and autophagy was induced for 3 h in the presence of MG132. Protein extracts were prepared in a denaturing urea buffer and subjected to Ni-NTA purification. The amount of ubiquitylated NEDD4L copurified with 6xHIS-ubiquitin was evaluated by WB. (F) HeLa cells were treated with EBSS for the indicated time periods, and the levels of p-NEDD4L, NEDD4L, and ACTIN were detected by WB. (G) qPCR analysis of NEDD4L mRNA levels in HeLa cells incubated in EBSS medium for the indicated time periods. Data are expressed as mean ± SEM (n = 3). In B–D and G, data were analyzed by one-way ANOVA followed by Tukey post hoc test. *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

Figure 5.

NEDD4L down-regulation increases autophagy. (A) NEDD4L expression was down-regulated in HeLa by RNAi. Cells were nutrient-starved for the indicated time periods in the presence or not of Clq. Protein extracts were analyzed by WB for the expression of LC3, ACTIN, and NEDD4L. Densitometric analysis of LC3II+Clq-LC3II control over ACTIN band is shown. Data are expressed as the mean ± SEM of three independent experiments (n = 3). (B) HeLa cells were treated as in A. Protein extracts were analyzed by WB for the expression of p62 and ACTIN. (C and D) NEDD4L expression was down-regulated by RNAi as in A; cells were nutrient-starved for the indicated time periods and fixed and labeled with anti-ATG16L or anti-LC3 antibodies (red puncta) and visualized by confocal microscopy. Analysis of the number of both ATG16L and LC3 puncta occurrence per cell is shown in the graph. Data are expressed as the mean ± SEM (n = 3); representative images are shown. Bar, 20 µm (>50 cells analyzed per sample). In A, data were analyzed by one-way ANOVA followed by Tukey post hoc test. *, P < 0.05. In C and D, data were analyzed by two-way ANOVA followed by Bonferroni’s multiple comparison post test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 6.

NEDD4L ubiquitylates ULK1 at lysine 925 and lysine 933. (A) HeLa cells were transfected with ULK1^WT, ULK1^K925R, and ULK1^K933R HA-tagged proteins and treated with CHX (50 µM) for different time periods. Levels of ULK1 and TUBULIN were detected by WB. Densitometric analysis of ULK1 over TUBULIN is also shown (right graph). Data are expressed as the mean ± SEM (n = 3). (B and C) HeLa cells were transfected ULK1^WT, ULK1^K925R, ULK1^K933R, and ULK1^K925R+K9333R HA-tagged proteins together with NEDD4L-HA in the presence or not of MG132 (4 h) and all were treated with CHX for 3 h. Levels of ULK1, NEDD4L, and TUBULIN were detected by WB. (B) Densitometric analysis of ULK1+MG132/ULK1 control over TUBULIN bands is also shown. Data are expressed as the mean ± SEM (n = 3). (D and E) HeLa cells were transfected with ULK1^WT, ULK1^K925R, ULK1^K933R, and ULK1^K925R+K933R HA-tagged proteins in the presence or not of MG132 and autophagy was induced with EBSS for 4 h. The levels of ULK1 and TUBULIN were detected by WB. (D) Densitometric analysis of ULK1+MG132/ULK1 control over TUBULIN bands is also shown. Data are expressed as the mean ± SEM (n = 3). (F) HeLa cells were transfected with a vector encoding a 6xHIS-tag ubiquitin together with ULK1^WT, ULK1^K925R and ULK1^K933R HA-tagged proteins in the presence of NEDD4L. Protein extracts were prepared in a denaturing urea buffer and subjected to Ni-NTA purification. The amount of ubiquitylated ULK1 copurified with 6xHIS-ubiquitin was valuated by WB. The intensity of ULK1 bands is shown. Data are the mean ± SEM (n = 3). (G) HeLa cells were transfected with a vector encoding a 6xHIS-tag ubiquitin together with ULK1^WT and ULK1^K925R+K933R mCherry-tagged proteins in the presence of NEDD4L. Protein extracts were prepared in a denaturing urea buffer and subjected to Ni-NTA purification. The amount of ubiquitylated ULK1 copurified with 6xHIS-ubiquitin was evaluated by WB. (H) Endogenous ULK1 was silenced by RNAi; cells were then transfected with ULK1^WT or ULK1^K925R+K933R mCherry-tagged plasmids. Cells were nutrient-starved for the indicated time periods, fixed and labeled with anti-ATG16L antibody (green puncta), and visualized by confocal microscopy. Analysis of the number of ATG16L occurrence per cell is shown in the graph. Data are expressed as the mean value± SEM (n = 3). Representative images are shown. Bars, 20 µm (>50 cells analyzed per sample). In A, B, D, and F, data were analyzed by one-way ANOVA followed by Tukey post hoc test. *, P < 0.05; ****, P < 0.0001. In H, data were analyzed by two-way ANOVA followed by Bonferroni’s multiple comparison post test. **, P < 0.01; ****, P < 0.0001. For all ULK1 mutant constructs the same amount of DNA is used for the transfection; the differences are caused by the increased stability of these mutants.

Figure 7.

ULK1 restoration requires de novo mTOR-dependent protein synthesis. (A) HeLa cells were treated with EBSS for the indicated time periods. The levels of p-mTOR, mTOR, ULK1, p-ULK1 S757, ULK1, and ACTIN were detected by WB. Densitometric analysis of p-mTOR over ACTIN is also shown (right graph). Data are expressed as the mean ± SEM (n = 4). (B) HeLa cells were treated with EBSS for the indicated time periods; at 4 h of starvation, rapamycin (100 nM) was added. Levels of p-mTOR, mTOR, ULK1, p-ULK1 S757, and ACTIN were detected by WB. Densitometric analysis of ULK1 over ACTIN is also shown (right graph). Data are expressed as the mean ± SD (n = 3), and statistical analysis was performed using unpaired Student's t test. *, P < 0.05. (C) HeLa cells were treated with EBSS for the indicated time periods and at 4 h of starvation, 250 nM Torin1 or 50 µM CHX was added. Levels of p-mTOR, mTOR, p-p70S6K, ULK1, and ACTIN were detected by WB. (D) HeLa cells were treated with EBSS for 6 h in the presence or not of Act D. Levels of ULK1 and ACTIN were detected by WB. Densitometric analysis of ULK1 over ACTIN is also shown (right graph). Data are expressed as the mean ± SD (n = 3) and statistical analysis was performed using unpaired Student's t test. *, P < 0.05. (E) qPCR analysis of polysomal recruitment of ULK1 and ACTIN mRNAs in control (CTRL) and EBSS-starved (2 h EBSS and 6 h EBSS) HeLa cells. Densitometric analysis of the signal in each fraction was performed, and the results were represented as the percentage of total signal in all fractions. Data represent mean ± SEM (n = 3). (F) NEDD4L expression was down-regulated by RNAi. Cells were nutrient-starved for 12 h in the presence or not of Baf A1 and analyzed by MTS assay. NEDD4L down-regulation was detected by WB. In A and C, data were analyzed by one-way ANOVA followed by Tukey post hoc test. *, P < 0.05; **, P < 0.01; ***, P < 0.001. In E and F, data were analyzed by two-way ANOVA followed by Bonferroni’s multiple comparison post test. ***, P < 0.001; ****, P < 0.0001.

Figure 8.

Autophagy restimulation requires mRNA preaccumulation and is amplified by NEDD4L down-regulation. (A) HeLa cells were treated with EBSS for the indicated time periods. After 6 h, in some samples, full medium was added for 15 min, and then cells were starved again for 1 and 2 h, respectively. Protein levels of ULK1 and ACTIN were detected by WB. Densitometric analysis of ULK1 over ACTIN is also shown (below graph). Data are expressed as the mean ± SEM (n = 3). (B) HeLa cells were treated with EBSS as in A. In some samples, Act D was added for the first 4 h of starvation. Then, cells were analyzed both by WB and by immunofluorescence. Protein levels of ULK1 and ACTIN were detected by WB. ATG16L puncta occurrence (red puncta) was visualized by confocal microscopy. Analysis of the number of ATG16L puncta per cell is shown in the graph. Data are expressed as the mean ± SEM (n = 3); representative images are shown. Bar, 20 µm (>50 cells analyzed per sample). (C) NEDD4L expression was down-regulated in HeLa cells by using specific RNAi oligos (siRNA NEDD4L#1) or unrelated oligos as a negative control (siRNA CTRL). Cells were nutrient-starved as in A. Protein extracts were next analyzed by WB for the expression of ULK1, NEDD4L, and ACTIN. ATG16L puncta occurrence (red puncta) was visualized by confocal microscopy. Analysis of the number of ATG16L puncta per cell is shown in the graph. Data are expressed as the mean ± SEM (n = 3); representative images are shown. Bar, 20 µm (>50 cells analyzed per sample). Data represent mean ± SEM (n = 3). In A, data were analyzed by one-way ANOVA followed by Tukey post hoc test. **, P < 0.01. In B and C, data were analyzed by two-way ANOVA followed by Bonferroni’s multiple comparison post hoc test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 9.

A cyclic scheme representing autophagy oscillatory regulation. (A) In enduring starvation, NEDD4L promotes ULK1 degradation while ULK1 mRNA is actively transcribed. When the autophagy recycling activity makes more nutrients available for protein synthesis, mTOR is reactivated (in conditions of prolonged starvation), and ULK1 mRNA is actively translated upon p70S6K phosphorylation. Then the mTOR kinase function inhibits again the ULK1 protein by phosphorylation, thus making the system ready to respond a putative new stimulus (early starvation), capable to reactivate autophagy. This regulation allows oscillatory autophagy behavior, both preventing excessive autophagy and preparing the cell for a prompt autophagy response upon its reinduction.