An Atg4B mutant hampers the lipidation of LC3 paralogues and causes defects in autophagosome closure

Abstract

In the process of autophagy, a ubiquitin-like molecule, LC3/Atg8, is conjugated to phosphatidylethanolamine (PE) and associates with forming autophagosomes. In mammalian cells, the existence of multiple Atg8 homologues (referred to as LC3 paralogues) has hampered genetic analysis of the lipidation of LC3 paralogues. Here, we show that overexpression of an inactive mutant of Atg4B, a protease that processes pro-LC3 paralogues, inhibits autophagic degradation and lipidation of LC3 paralogues. Inhibition was caused by sequestration of free LC3 paralogues in stable complexes with the Atg4B mutant. In mutant overexpressing cells, Atg5- and ULK1-positive intermediate autophagic structures accumulated. The length of these membrane structures was comparable to that in control cells; however, a significant number were not closed. These results show that the lipidation of LC3 paralogues is involved in the completion of autophagosome formation in mammalian cells. This study also provides a powerful tool for a wide variety of studies of autophagy in the future.

Figure 1.

Cells expressing Atg4B^C74A exhibit a block in autophagic degradation. (A) MCF7 cells stably expressing GFP-LC3 were infected with adenovirus bearing mStrawberry (Mock), mStrawberry-Atg4B^WT (WT), or mStrawberry-Atg4B^C74A (C74A) and incubated for 40 h. The cells were then cultured in HBSS for 2 h, fixed, and observed using fluorescence microscopy. Bar, 10 μm. (B) PC12 cells were infected with adenovirus bearing GFP (−), 3xFlag-tagged wild-type Atg4B (WT), Atg4B^C74A (CA), or Atg4B^C74S (CS). Cells were cultured either in growth or starvation medium, and lysates were examined by Western blotting using each antibody. Top panel, anti-Flag; middle panel, anti-LC3; bottom panel, anti-α-tubulin. (C) 293A cells stably expressing empty vector (Mock) or mStrawberry-Atg4B^C74A (Atg4B^C74A) were grown in growth medium (F), HBSS (S), or HBSS with 100 nM wortmannin (W) for 2 h. Cell lysates were examined by Western blotting using each antibody. From top panel, anti-RFP, anti-p62, anti-LC3, and anti-α-tubulin. (D) 293A cells stably expressing empty vector (Mock) or shRNA against LC3 (LC3 KD) were cultured in HBSS for 2 h and collected. Cell lysates were examined by Western blotting using each antibody. From top panel, anti-LC3, anti-GABARAP, and anti-α-tubulin. (E) Mock, mStrawberry-Atg4B^C74A-expressing (Atg4B^C74A), or LC3-knockdown (LC3-KD) 293A cells were grown in growth medium (F), HBSS (S), or HBSS with 100 nM wortmannin (W) for 2 h. Long-lived protein degradation was scored as described in Materials and Methods.

Figure 2.

Effect of Atg4B^C74A overexpression on the Atg16L complex. (A) 293A cells stably expressing empty vector (Mock) or mStrawberry-Atg4B^C74A (Atg4B^C74A) were cultured in growth medium or HBSS for 2 h, and Western blotting was performed using each antibody. From top panel, anti-RFP, anti-Atg5, anti-LC3, anti-α-tubulin. Anti-RFP antibody reacts with mStrawberry. (B) Cytosolic fractions of 293A cells stably expressing empty vector or mStrawberry-Atg4B^C74A were separated by size exclusion chromatography. Fractions were subjected to Western blotting using the indicated antibodies. The positions of the molecular-mass standards are shown. Vo, void fraction.

Figure 3.

Stable complex formation between excess Atg4B^C74A and LC3 prevents access to Atg7. (A) 293A cells stably expressing empty vector (Mock) or mStrawberry-Atg4B^C74A (Atg4B^C74A) were transfected with Myc-LC3-HA (WT) or Myc-LC3^G120A-HA (GA). Thirty-six hours after transfection, cells were cultured in growth medium or HBSS for 2 h, and Western blotting were performed. From top panel, anti-RFP, anti-Myc, and anti-α-tubulin. (B) PC12 cells were infected with adenovirus bearing 3xFlag-tagged wild-type Atg4B (WT), Atg4B^C74A (CA), or Atg4B^C74S (CS). After 40-h incubation, cell lysates were subjected to immunoprecipitation with anti-Flag M2-conjugated agarose beads. Coimmunoprecipitated molecules were examined by Western blotting using each antibody. From top panel, anti-Flag, anti-LC3, and anti-GATE16. Total cell lysate (Input) and immunoprecipitated proteins (IP) are shown. (C) 293A cells stably expressing empty vector (Mock) or mStrawberry-Atg4B^C74A (Atg4B^C74A) were transfected with GFP-LC3 and Myc-Atg7 as indicated. Thirty-six hours after transfection, cell lysates were examined by Western blotting using each antibody. From top panel, anti-RFP, anti-myc, anti-GFP, and anti-α-tubulin.

Figure 4.

The inhibitory effect of Atg4B mutant on LC3 lipidation is suppressed by exogenous LC3 in a dose-dependent manner. (A) NIH3T3 cells were infected with different amounts of retroviruses bearing mStrawberry-Atg4B^C74A, and stable transformants were selected. The stable cells were cultured in HBSS (Starved) for 1 h, and cell lysates were examined by Western blotting using each antibody. From top panel, anti-Atg4B, anti-RFP, anti-LC3, and anti-α-tubulin. (B) NIH3T3 cells stably expressing mStrawberry-Atg4B^C74A were infected with different amounts of retroviruses bearing GFP-LC3 and then double stable transformants were selected. Parent NIH3T3 cells and the stable transformants were cultured in HBSS (Starved) for 1 h, and cell lysates were examined by Western blotting using each antibody. From top panel, anti-RFP, anti-GFP, anti-LC3, and anti-α-tubulin.

Figure 5.

Effect of Atg4B^C74A overexpression on GFP-Atg5–positive membrane structures. (A and B) NIH3T3 cells stably expressing GFP-Atg5 or both GFP-Atg5 and mStrawberry-Atg4B^C74A were grown in growth medium (F), HBSS (S), or HBSS with 100 nM wortmannin (W) for 1 h and then fixed. Three-dimensional image stacks were obtained from sequential optical sections acquired 0.3 μm apart by confocal laser scanning microscopy (FV1000, Olympus) (A). Bar, 10 μm. The number of GFP-Atg5 puncta was counted in more than 100 cells. The value indicated is the mean ± SD (B). (C) NIH3T3 cells stably expressing GFP-Atg5 or both GFP-Atg5 and mStrawberry-Atg4B^C74A were grown in HBSS for 1 h and directly observed by time-lapse video microscopy. The duration of each GFP-Atg5 puncta was measured for more than 50 cases. The value indicated is the mean ± SD.

Figure 6.

Ultrastructual analysis of autophagic membranes in Atg4B^C74A-overexpressing cells. NIH3T3 cells stably expressing empty vector (Mock) (A–E) or mStrawberry-Atg4B^C74A (F–J) were cultured in HBSS for 1 h, fixed, and subjected to conventional electron microscopic analysis. (B–E) Typical autophagic structures in mock cells; isolation membrane (B), autophagosomes (C–E). (G–J) Typical autophagic structures in mStrawberry-Atg4B^C74A-expressing cells; isolation membranes (G and H), closed double-membrane structures (I and J). Examples of electron-dense structures (black arrows), closed autophagic membranes (white arrows), and open autophagic membranes (white arrowheads) are indicated. Bar, 500 nm. (K) The number of autophagic structures in mock and mStrawberry-Atg4B^C74A-expressing cells. ▩, open autophagic structures; ■, closed autophagic structures. Data are the means ± SD of triplicates from representative experiments. (L) The ratio of open structures to total autophagic structures. Data are the means ± SD of triplicates from representative experiments. *p < 0.05. (M) The length of autophagic membranes in mock and mStrawberry-Atg4B^C74A-expressing cells. ▩, open autophagic structures; ■, closed autophagic structures. For the length of autophagic membranes determination, ImageJ version 1.40 was used (http://rsb.info.nih.gov/ij/). The value indicated is the mean ± SD. At least 20 samples were examined for each structure. *p < 0.05; NS, not significant. (N) The ratio of the length of open to closed autophagic structures in mock and mStrawberry-Atg4B^C74A-expressing cells. *p < 0.05. (O) NIH3T3 cells expressing both GFP-Atg5 and mStrawberry-Atg4B^C74A were grown in HBSS for 1 h and fixed. The localization of GFP-Atg5 was examined by gold-enhanced immunogold electron microscopy using an anti-GFP antibody. Bar, 500 nm.