GCN2 upregulates autophagy in response to short-term deprivation of a single essential amino acid

Figures & data

Figure 1. Leucine deprivation rapidly triggers an autophagy-dependent upregulation of protein breakdown that requires GCN2. (A) The upregulation of proteolysis during leucine deprivation results from mobilization of the autophagic process. Protein degradation was measured using L-[35S]Met radiolabeling of cellular proteins and pulse-chase approach. Radio-labeled L-[³⁵S]Met wild-type (WT) MEFs were incubated in the control (Ctr) or leucine-devoid (-Leu) medium for 2 h without inhibitor (vehicle) or in the presence of either an inhibitor of autophagy (3-methyladenine; 3-MA, 10 mM), or an inhibitor of proteasome-dependent proteolysis (lactacystin; Lacta, 2 µM). Three independent experiments were performed. Results are expressed as variation in proteolysis rate in response to leucine deprivation compared to control proteolysis rate (control medium). Bar values are mean ± SEM (*, p < 0.05 relative to the control, Student’s t-test). (B) Electron microscopy images of autophagic structures (arrows) following incubation of WT MEFs in the Ctr or -Leu medium for 2 h. (C) Impact of Gcn2 knock-out (KO) on variations in proteolysis resulting from leucine deprivation. WT and gcn2 KO MEFs were incubated in the Ctr or -Leu medium for 2 h, and protein degradation was measured as described for (A). Three independent experiments were performed and results were analyzed and expressed as in (A).

Figure 2. GCN2 is required for upregulating autophagy in response to short-term leucine deprivation in MEFs. (A) Immunoblot analyses of total protein extracts of WT and gcn2 KO MEFs either incubated in the Ctr or -Leu medium for 1 h. Representative immunoblots and quantification of relative intensities of P-[S278]-ATG16L1 to GAPDH, LC3-II to LC3-I and p62 to GAPDH are shown (four independent experiments). Bar values are mean ± SEM (*, p < 0.05 relative to Ctr of the same cell type, Student’s t-test). As previously observed by others [52], the anti-ATG16L1 antibody detected two isoforms (B) Representative images and quantification of LC3-labeled punctae. WT and gcn2 KO MEFs were cultured in the Ctr or -Leu medium in the presence of 20 µM chloroquine for 2 h. Endogenous LC3 was detected by immunofluorescence and average number of punctae per cell was determined (40-50 cells were analyzed per condition in three independent experiments). Bar values are mean ± SEM (*, p < 0.05 relative to Ctr of the same cell type, Student’s t-test). (C) Immunoblot analyses of total protein extracts of gcn2 KO MEFs overexpressing either GFP as a control or GCN2. Two pMX-Gcn2-transduced populations are shown, one total population (type #1) and one clone (type #2). Cells were either incubated in the Ctr or -Leu medium for 2 h. Representative immunoblots and relative quantifications of LC3-II to LC3-I and P-[S278]-ATG16L1 and p62 to GAPDH are given (three independent experiments). Bar values are mean ± SEM (*, p < 0.05 relative to Ctr of the same cell type, Student’s t-test).

Figure 3. GCN2 activity is required to upregulate autophagy in response to leucine deficiency in mouse liver. After overnight fasting, WT and gcn2 KO mice were fed the Ctr or -Leu experimental diet for 1 h. (A) Plasma leucine concentrations. Bar values are mean ± SEM (*, p < 0.05 relative to the Ctr for the same genotype, Student’s t-test; n=3). (B) Representative immunoblots of liver total protein extracts and relative quantifications of LC3-II to LC3-I and P-[S278]-ATG16L1 to GAPDH are given (6 mice per group from two independent experiments). Bar values are mean ± SEM (*, p < 0.05 relative to the Ctr for the same genotype, Student’s t-test). It is noticeable that the batch of anti-LC3B antibody used here detected a non-specific protein (n.s.) above LC3-I signal.

Figure 4. In response to short-term leucine deficiency, eIF2α phosphorylation is an essential event, whereas ATF4 is not required to upregulate autophagy. Immunoblot analyses of total protein extracts of WT and atf4 KO MEFs (A) and eIF2α ^S51A/S51A (eIF2α ^AA) MEFs (B) either incubated in the Ctr or -Leu medium for 1 h. In each case, representative immunoblots and relative quantifications of LC3-II to LC3-I and P-[S278]-ATG16L1 and p62 to GAPDH are shown (four independent experiments). Bar values are mean ± SEM (*, p < 0.05 relative to Ctr of the same cell type, Student’s t-test).

Figure 5. The upregulation of autophagy associated to GCN2 activation can occur without decreasing the phosphorylation level of either [S757]-ULK1 or [T389]-S6K1. Immunoblot analyses of total protein extracts from WT MEFs exposed for 1 h to the Ctr, -Leu or lysine-devoid (-Lys) medium (A) or L-histidinol-containing- (Hol, 4 mM) medium (B). In each case, representative immunoblots and relative quantifications of P-[S278]-ATG16L1 to GAPDH, LC3-II to LC3-I and P-[S757]-ULK1 to ULK1 are given (three independent experiments). Bar values are mean ± SEM (*, p < 0.05 relative to Ctr, Student’s t-test).