S-acylation of p62 promotes p62 droplet recruitment into autophagosomes in mammalian autophagy

Abstract

p62 is a well-characterized autophagy receptor that recognizes and sequesters specific cargoes into autophagosomes for degradation. p62 promotes the assembly and removal of ubiquitinated proteins by forming p62-liquid droplets. However, it remains unclear how autophagosomes efficiently sequester p62 droplets. Herein, we report that p62 undergoes reversible S-acylation in multiple human-, rat-, and mouse-derived cell lines, catalyzed by zinc-finger Asp-His-His-Cys S-acyltransferase 19 (ZDHHC19) and deacylated by acyl protein thioesterase 1 (APT1). S-acylation of p62 enhances the affinity of p62 for microtubule-associated protein 1 light chain 3 (LC3)-positive membranes and promotes autophagic membrane localization of p62 droplets, thereby leading to the production of small LC3-positive p62 droplets and efficient autophagic degradation of p62-cargo complexes. Specifically, increasing p62 acylation by upregulating ZDHHC19 or by genetic knockout of APT1 accelerates p62 degradation and p62-mediated autophagic clearance of ubiquitinated proteins. Thus, the protein S-acylation-deacylation cycle regulates p62 droplet recruitment to the autophagic membrane and selective autophagic flux, thereby contributing to the control of selective autophagic clearance of ubiquitinated proteins.

Keywords: APT1; S-acylation; ZDHHC19; autophagy; autophagy receptor; liquid-liquid phase separation; p62 droplet; p62 protein; protein posttranslational modification; selective autophagy.

Graphical abstract

Figure 1

p62 undergoes S-acylation at Cys289 and Cys290 (A) S-acylation of endogenous p62 detected by ABE assay in NRK and HeLa cells. Endogenous p62 was immunoprecipitated with anti-p62 antibody. HAM, hydroxylamine. (B) S-acylation of FLAG-p62 detected by chemical reporters in HEK293T cells. The cells transiently expressing FLAG-p62 were metabolically labeled with Alk-C16 or Alk-C18 (100 μM) for 12 h. FLAG-p62 was immunoprecipitated with anti-FLAG antibody. Alk-C16, palmitic acid alkyne; Alk-C18, stearic acid alkyne. (C) S-acylation of p62 detected by ABE assay in HEK293T cells incubated in starvation medium (EBSS) and treated with puromycin, MG132, or Torin 1. (D) S-acylation of FLAG-p62 mutants detected by ABE assay. Because of p62 oligomerization, anti-FLAG antibody precipitated exogenous FLAG-p62 (the upper band) as well as p62 (the lower band). (E) Scheme of the fusion motifs consisting of FLAG-GFP and peptides derived from the p62 sequence. Three peptides, including p62 (285–294), p62 (283–296), and p62 (279–300), were selected to define the region of p62 required for S-acylation. The peptides containing the mutation C289,290S were designed as controls. (F) S-acylation of FLAG-GFP fusions detected by ABE assay in (E). (G) Conservation analysis of p62 S-acylation sites Cys289 and Cys290 among various species. PB1, Phox1, and Bem1p domain; ZZ, ZZ-type zinc-finger domain; TB, TRAF6-binding domain; LIR, LC3-interacting region; KIR, Keap1-interacting region; UBA, ubiquitin-associated domain. See also Figure S1.

Figure 2

ZDHHC19 mediates p62 S-acylation (A) Representative images of HA-ZDHHC19 localization in p62-KO NRK cells stably expressing GFP-p62 (green). HA-ZDHHC19 was transfected and immunostained with anti-HA antibody (red). The nucleus was visualized by DAPI (blue). Scale bars, 10 μm. (B) Immunoblot analysis of FLAG-p62-immunoprecipitated (IP) proteins in HEK293T cells co-expressing FLAG-p62 with HA-ZDHHC19. (C) Immunoblot analysis of HA-ZDHHC19-IP proteins in NRK cells co-expressing FLAG-p62 with HA-ZDHHC19. (D) Scheme of TurboID-mediated proximity labeling of ZDHHC19. Cells were transfected with the plasmid encoding ZDHHC19-TurboID and labeled with D-biotin. (E) Immunoblot analysis of ZDHHC19-TurboID-labeled proteins. Total proteins were precipitated from cell lysates and enriched with streptavidin-coated magnetic beads. Biotinylated proteins were eluted and further analyzed by western blotting with HRP-conjugated streptavidin. Biotinylated p62 was analyzed by western blotting with anti-p62 antibody. (F) S-acylation level of p62 detected by ABE assay in ZDHHC19-KO HEK293T cells. (G) Quantification of p62 S-acylation levels as in (F). Data represent the mean ± SEM of three independent experiments. ^∗∗∗p < 0.001, Student’s t test. See also Figure S2.

Figure 3

S-acylation facilitates the degradation of p62 and ubiquitinated proteins (A) Immunoblot analysis of FLAG-p62 WT and FLAG-p62^C289,290S proteins in p62-KO NRK cells. Cells were transiently transfected with FLAG-p62 WT or FLAG-p62^C289,290S and treated with Torin 1 (1 μM) for 1 h, with or without Baf-A1 (1 μM). (B) Quantification of FLAG-p62 WT and FLAG-p62^C289,290S protein levels as in (A). (C) Immunoblot analysis of p62 protein upon transiently expressing ZDHHC19 WT or ZDHHC19^C142S in NRK cells. (D) Quantification of p62 protein levels as in (C). (E) Immunoblot analysis of p62 protein and ubiquitinated proteins in WT or ZDHHC19-KO HEK293T cells. Cells were transfected with empty vector or HA-ZDHHC19 and then incubated with Torin 1 (1 μM) for 4 h, with or without Baf-A1 (1 μM). (F) Quantification of p62 protein levels as in (E). (G) Immunoblot analysis of p62 protein in ATG7-KO cells upon HA-ZDHHC19 expression. HEK293T WT or ATG7-KO cells were transfected with empty vector or HA-ZDHHC19 and then treated with Torin 1 (1 μM) for 1 h. (H) Quantification of p62 protein levels as in (G). (I) Immunoblot analysis of total ubiquitinated proteins in WT or p62-KO NRK cells transiently expressing FLAG-p62 WT or FLAG-p62^C289,290S. (J) Quantification of total ubiquitinated proteins as in (I). (K) Immunoblot analysis of p62 and total ubiquitinated proteins in NRK cells transiently expressing ZDHHC19. (L) Quantification of total ubiquitinated protein levels as in (K). (M) Immunoblot analysis of p62 protein and total ubiquitinated proteins upon ZDHHC19 expression in NRK cells. (N) Quantification of total ubiquitinated protein levels as in (M). Data in (B), (D), (F), (H), (J), (L), and (N) represent the mean ± SEM of three independent experiments. ns, no significant difference; ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; Student’s t test. See also Figure S3.

Figure 4

S-acylation promotes the formation of small LC3-positive p62 puncta (A) Representative images of p62 puncta in GFP-p62 WT or GFP-p62^C289,290S cells. GFP-p62 WT or GFP-p62^C289,290S was stably expressed in p62-KO NRK cells. Cells were fixed with paraformaldehyde and observed under confocal microscopy. The arrows indicate p62 puncta with a diameter of more than 1 μm. Scale bars, 10 μm. (B) Quantification of the number of p62 puncta with a diameter of more than 1 μm in each cell as in (A). ^∗∗∗p < 0.001, Student’s t test. (C) Immunoblot analysis of p62 proteins in GFP-p62 WT or GFP-p62^C289,290S cells. (D) Representative images of p62 puncta in ZDHHC19-KO HEK293T cells. Scale bars, 10 μm. (E) Quantification of the number of p62 puncta with a diameter of more than 1 μm in each cell as in (D). ^∗∗∗p < 0.001, Student’s t test. (F) Representative images of p62 puncta in GFP-p62 NRK cells transiently expressing HA-ZDHHC19 WT or HA-ZDHHC19^C142S. Scale bars, 10 μm. (G) Quantification of the number of p62 puncta with a diameter of more than 1 μm in each cell as in (F). ^∗∗∗p < 0.001, Student’s t test. (H) Representative images of p62 puncta in GFP-p62 WT or GFP-p62^C289,290S NRK cells transiently expressing HA-ZDHHC19. Scale bars, 10 μm. (I) Quantification of the number of p62 puncta with a diameter of more than 1 μm in each cell as in (H). ns, no significant difference. ^∗∗∗p < 0.001, Student’s t test. See also Figure S4.

Figure 5

S-acylation regulates autophagic membrane localization of p62 droplets (A) Representative images of autophagosome formation in GFP-p62 WT or GFP-p62^C289,290S cells upon autophagy induction. p62-KO NRK cells stably expressing GFP-p62 WT or GFP-p62^C289,290S (green) were incubated with Torin 1 (1 μM) for 3 h, followed by immunostaining with anti-LC3A/B antibody (red). Scale bars, 10 μm. (B) Quantification of LC3-positive p62 puncta in GFP-p62 WT or GFP-p62^C289,290S cells upon autophagy induction as in (A). ^∗∗∗p < 0.001, Student’s t test. (C) Representative images of autophagosome formation in mCherry-LC3 HEK293 cells upon the expression of ZDHHC19 WT or ZDHHC19^C142S. HEK293 cells stably expressing mCherry-LC3 (red) were transiently expressed with HA-ZDHHC19 WT or HA-ZDHHC19^C142S, followed by immunostaining with anti-p62 antibody (green) and anti-HA antibody (purple). Scale bars, 10 μm. (D) Quantification of LC3-positive p62 puncta in each cell as in (C). ^∗∗∗p < 0.001, Student’s t test. (E) Representative images of p62 puncta and autophagosome in WT or ZDHHC19-KO HEK293T cells. HEK293T cells transiently expressing mCherry-LC3 (red) were treated with or without Torin 1 (1 μM) for 3 h and immunostained with anti-p62 antibody (green). Scale bars, 10 μm. (F) Quantification of LC3-positive p62 puncta in each cell as in (E). ^∗∗∗p < 0.001, Student’s t test. (G) Representative images of p62 puncta and autophagosome in GFP-p62 WT or GFP-p62^C289,290S cells upon ZDHHC19 expression. p62-KO NRK cells stably expressing GFP-p62 WT or GFP-p62^C289,290S (green) were transfected with HA-ZDHHC19 and immunostained with anti-LC3A/B antibody (red) and anti-HA antibody (purple). Scale bars, 10 μm. (H) Fluorescence recovery after photobleaching (FRAP) assays of GFP-p62 WT and GFP-p62^C289,290S puncta. Data from four independent experiments were analyzed. Scale bars, 1 μm. See also Figure S5.

Figure 6

S-acylation enhances the affinity of p62 for LC3-positive membranes in vitro (A) Left, representative images of GFP-p62-positive GUVs. GFP-p62 (155–440) or GFP-p62 (155–440)-palm (green) was incubated with rhodamine labeled GUVs (red) for 0–60 min. The fluorescence was monitored by laser scanning confocal microscope. Right, schematic diagram of p62 protein recruitment to GUVs. GFP-p62 (155–440)-palm, palmitate-maleimide conjugated MBP-GFP-p62 (155–440)^C331S. Scale bars, 4 μm. (B) Schematic diagram of p62 co-precipitated with total cellular membrane in vitro. (C) Immunoblot analysis of mCherry-p62 and mCherry-p62-palm proteins co-precipitated with total cellular membrane in vitro as in (B). The levels of Na⁺/K⁺ ATPase and GADPH were detected as internal references for total cellular membrane and cytoplasm, respectively. (D) Immunoblot analysis of GFP-p62 and GFP-p62-palm proteins co-precipitated with total cellular membrane in vitro as in (B). (E) Schematic diagram of GFP-p62-palm co-precipitated with LC3-positive or LC3-negative membranes. (F) Immunoblot analysis of GFP-p62-palm co-precipitated with LC3-rich and RavZ-treated membrane in vitro as in (E). (G) Immunoblot analysis of GFP-p62-palm co-precipitated with LC3-positive and LC3-negative membrane from in vitro ATG7-KO HEK293T cells as in (E). (H) Schematic diagram of reconstitution of S-acylated p62 in total cellular membrane in vitro. S-acylated p62 and non-acylated p62 proteins were purified from p62-KO NRK cells transiently expressing FLAG-p62 WT and FLAG-p62^C289,290S, respectively. The total cellular membranes were prepared from p62-KO NRK cells and further incubated with the purified p62 proteins. (I) Immunoblot analysis of mammalian cell-expressed FLAG-p62 WT and FLAG-p62^C289,290S proteins co-precipitated with total cellular membrane in vitro as in (H). See also Figure S6.

Figure 7

APT1 mediates deacylation of p62 (A) S-acylation of endogenous p62 detected by ABE assay in the presence of various thioesterases. (B) Immunoblot analysis of p62 and total polyubiquitinated proteins in HEK293T cells transiently expressing HA-APT1 WT or HA-APT1^S119A. Cells were transfected with HA-APT1 WT or HA-APT1^S119A and then incubated with Torin 1 (1 μM) for 4 h, with or without Baf-A1 (1 μM). (C) Quantification of total ubiquitinated proteins as in (B). Data represent the mean ± SEM of three independent experiments. ns, no significant difference; ^∗∗p < 0.01; ^∗∗∗p < 0.001; Student’s t test. (D) Quantification of p62 protein levels as in (B). Data represent the mean ± SEM of three independent experiments. ns, no significant difference; ^∗∗p < 0.01; ^∗∗∗p < 0.001; Student’s t test. (E) Immunoblot analysis of p62 protein in APT1-KO HEK293T cells expressing empty vector, HA-APT1 WT, or HA-APT1^S119A. (F) Quantification of the p62 protein level as in (E). Data represent the mean ± SEM of three independent experiments. ns, no significant difference; ^∗∗p < 0.01; Student’s t test. (G) Representative images of p62 puncta in GFP-p62-expressing p62-KO NRK cells upon HA-APT1 expression. GFP-p62 NRK cells (green) were transiently transfected with HA-APT1 and immunostained with anti-HA antibody (red). Scale bars, 10 μm. (H) Quantification of the number of p62 puncta in each cell as in (G). ^∗∗∗p < 0.001, Student’s t test. (I) Representative images of p62 puncta in APT1-KO HEK293T cells. Scale bars, 10 μm. (J) Quantification of the number of p62 puncta in each cell as in (I). ^∗∗∗p < 0.001, Student’s t test. (K) Representative images of p62- and LC3-positive puncta in mCherry-LC3-expressing HEK293 cells transiently expressing HA-APT1. Cells were incubated with Torin 1 (1 μM) for 1 h, followed by immunostaining with anti-p62 antibody (green) and anti-HA antibody (purple). Scale bars, 10 μm. (L) Quantification of the number of LC3-positive p62 puncta in each cell as in (K). ns, no significant difference; ^∗∗∗p < 0.001; Student’s t test. (M) Analysis of Triton X-100 soluble and insoluble fractions of p62 in HEK293T cells transiently expressing HA-APT1. (N) Analysis of Triton X-100 soluble and insoluble fractions of p62 in APT1-KO cells. See also Figure S7.

Scheme 1

Synthesis route of palmitate-maleimide