Regulation of Golgi turnover by CALCOCO1-mediated selective autophagy

Abstract

The Golgi complex is essential for the processing, sorting, and trafficking of newly synthesized proteins and lipids. Golgi turnover is regulated to meet different cellular physiological demands. The role of autophagy in the turnover of Golgi, however, has not been clarified. Here we show that CALCOCO1 binds the Golgi-resident palmitoyltransferase ZDHHC17 to facilitate Golgi degradation by autophagy during starvation. Depletion of CALCOCO1 in cells causes expansion of the Golgi and accumulation of its structural and membrane proteins. ZDHHC17 itself is degraded by autophagy together with other Golgi membrane proteins such as TMEM165. Taken together, our data suggest a model in which CALCOCO1 mediates selective Golgiphagy to control Golgi size and morphology in eukaryotic cells via its interaction with ZDHHC17.

Figure 1.

CALCOCO1 interacts with ZDHHC17 via zDABM motif. (A) Volcano plot showing some of the Golgi- and ER-localized proteins that were identified in the CALCOCO1-interactome screen. Three independent immunoprecipitations and MS analyses were conducted with EGFP-CALCOCO1 as bait. Only some of the identified proteins are shown. (B) HeLa CALCOCO1 KO cells were transiently cotransfected with Myc-ZDHHC17 and EGFP-CALCOCO1 cultured in full medium and immunostained with anti-GM130 antibody. We used CALCCO1 KO cells to avoid interference by endogenous CALCOCO1. Scale bars represent 5 µm. (C) Fluorescence intensity analysis of EGFP-CALCOCO1 in the cells analyzed in B. The error bars represent mean ± SD of three independent experiments with 100 cells per experiment. (D) GST, GST-ZDHHC17, or GST-ZDHHC17(N100A/W130A) was used in pull-down against in vitro translated Myc-CALCOCO1 or Myc-CALCOCO1-mutzDABM (mutant zDABM). (E) Domain architecture of human CALCOCO1 showing the location of zDABM motif relative to other domains. (F) GST or GST-ZDHHC17 was used in pull-down against in vitro translated Myc-CALCOCO1 or Myc-CALCOCO1 Δ145–513. (G) GST, GST-ZDHHC17, or GST-ZDHHC17(N100A/W130A) was used in pull-down against in vitro translated Myc-TAX1BP1 or Myc-TAX1BP1-ΔzDABM (ΔzDABM, zDABM deleted). MW, molecular weight.

Figure S1.

Mutation of zDABM motif of CALCOCO1 reduces its Golgi localization. (A) HeLa cells were cotransfected with EGFP-CALCOCO1 and either Myc-ZDHHC17 or Myc-ZDHHC13, grown in full medium and imaged by confocal microscopy. Bars represent 5 µm (main) and 1 µm (insets). (B) HeLa cells were cotransfected with EGFP-CALCOCO-mutZDABM and Myc-ZDHHC17 and then treated as in A. Bars represent 5 µm (main) and 1 µm (insets). (C) Colocalization analysis of Myc-ZDHHC17 and EGFP-CALCOCO1 in the cells analyzed in A and B. The error bars represent mean ± SEM of three independent experiments per condition with 100 cells per experiment. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as ***, P < 0.001. corr. coeff., correlation coefficient. (D) HeLa CALCOCO1 KO cells were transfected with EGFP-ZDHHC17; 24 h after transfection, they were either left untreated or starved (HBSS) for 4 h with Baf A1 treatment, and then immunostained for endogenous LC3B and LAMP2. The arrows point to colocalization. Bars represent 5 µm (main) and 1 µm (insets).

Figure 2.

CALCOCO1 recruits starvation-induced Golgi fragments to autophagosomes. (A) HeLa WT cells were transfected with EGFP-ZDHHC17; 24 h after transfection, they were either left untreated or starved (HBSS) for 4 h with Baf A1 treatment, then stained for endogenous TMEM165. Bars represent 5 µm (main) and 1 µm (insets). (B) HeLa WT cells were cotransfected with Myc-ZDHHC17 and EGFP-CALCOCO1; 24 h after transfection, they were either left untreated or starved for 4 h with Baf A1 treatment and then immunostained for endogenous LC3B. The arrows in A and B point to colocalization. Bars represent 5 µm (main) and 1 µm (insets). (C) HeLa WT and HeLa-CALCOCO1 KO cells were transfected with EGFP-ZDHHC17; 24 h after transfection, they were either left untreated or starved for 6 h. The number of cells with fragmented Golgi phenotype was manually counted and weighted against the total number of cells. (D) Colocalization analysis of ZDHHC17 and LC3B in WT cells analyzed in B and CALCOCO1 KO cells analyzed in Fig. S1 D. The error bars in C and D represent mean ± SEM of three independent experiments per condition with 100 cells per experiment. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as ***, P < 0.001; * P < 0.01. corr. coeff., correlation coefficient. (E) EGFP-CALCOCO1 was transiently expressed in HeLa WT cells; 24 h after transfection, the cells were either left untreated or starved and treated with Baf A1. EGFP-CALCOCO1 was then immunoprecipitated (IP) from cell extracts using GFP-TRAP. EGFP-CALCOCO1 and coprecipitated endogenous LC3B were visualized by Western blotting using anti-GFP and anti-LC3B antibodies. MW, molecular weight.

Figure 3.

ZDHHC17 promotes degradation of Golgi by autophagy during starvation. (A and B) MEF parental and Atg5 KO cells were left untreated, treated with Baf A1 for 6 h or MG132 for 6 h, or starved for 6 h with or without Baf A1. Cell lysates were analyzed by immunoblotting with the indicated antibodies. In A, the panels are collected from more than one Western blot experiment, but for clarity, only a single GAPDH loading control is shown. Numbers below the blots represent relative intensity of the bands in the blots shown, normalized against loading control. The asterisks (*) indicate unspecific bands. The bars in B represent the mean ± SD of band intensities relative to the loading control from three independent experiments quantified using ImageJ. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as ***, P < 0.001; **, P < 0.005; *, P < 0.01. (C–E) HeLa cells stably expressing tetracycline-inducible EGFP-ZDHHC17 were left uninduced or induced for 48 h and then treated as in A. Cell lysates were analyzed by immunoblotting with the indicated antibodies. Numbers below the EGFP-ZDHHC17 blot represent relative intensity of the bands normalized against the loading control. Data in D and E are presented as mean ± SD of three independent experiments. Statistical comparison was analyzed as in B; significance is displayed as ***, P < 0.001; **, P < 0.005; *, P < 0.01. (F) HeLa WT cells were transfected with EGFP-ZDHHC17; 24 h after transfection, were either left untreated or starved (HBSS) for 4 h with Baf A1 treatment, then immunostained for endogenous LC3B and LAMP2. The arrows point to colocalization. Bars represent 5 µm (main) and 1 µm (insets). (G) Colocalization analysis of EGFP-ZDHHC17 and LAMP2 in WT HeLa cells analyzed in F, CALCOCO1 KO cells analyzed in Fig. S1 D, and ATG7 KO cells analyzed in Fig. S2 D. The error bars represent mean ± SEM of three independent experiments per condition with 100 cells per experiment. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as ***, P < 0.001. MW, molecular weight.

Figure S2.

Golgiphagy during starvation requires macroautophagy and depends on the ZDHHC17-CALCOCO1 interaction. (A) HeLa parental or ATG7 KO cells were left untreated, treated with either Baf A1 for 6 h or MG132 for 6 h, or starved for 6 h with or without Baf A1. Cell lysates were analyzed by immunoblotting with the indicated antibodies. The panels are collected from more than one Western blot experiment, but for clarity, only a single GAPDH loading control is shown. Numbers below the blots represent relative intensity of the bands in the blots shown, normalized against loading control. (B and C) HeLa CALCOCO1 KO cells stably expressing inducible EGFP-ZDHHC17 (B) or EGFP-ZDHHC17 N100A/W130A (C) were left uninduced or induced for 24 h and then left untreated or treated as indicated. Cell lysates were analyzed by immunoblotting with the indicated antibodies. Numbers below the EGFP-ZDHHC17 blot represent relative intensity of the bands normalized against the loading control. (D) ATG7 KO HeLa cells were transfected with EGFP-ZDHHC17; 24 h after transfection, they were either left untreated or starved (HBSS) for 4 h with Baf A1 treatment, then immunostained for endogenous LC3B and LAMP2. Bars represent 5 µm (main) and 1 µm (insets). MW, molecular weight.

Figure 4.

Depletion of CALCOCO1 decreases Golgi degradation by autophagy. (A and B) WT (A) and CALCOCO1 KO (B) HeLa cells were transiently transfected with mCherry-EYFP-ZDHHC17 or mCherry-EYFP-TMEM165. 24 h after transfection, the cells were treated or not with HBSS as indicated. The arrows point to red-only puncta. Bars represent 5 µm (main) and 1 µm (insets). (C and D) The fraction of red-only puncta in the cells shown in A and B and in ATG7 KO HeLa cells (see Fig. S3 A) were counted and are shown as a percentage of total puncta. The error bars represent mean ± SD of red-only puncta percentages of three independent experiments. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test; significance is displayed as ***, P < 0.001.

Figure S3.

CALCOCO1-dependent Golgiphagy is mediated by macroautophagy and depends on the ATG8 interaction motifs, but not the FFAT-like motif. (A) ATG7 KO cells were transiently transfected with mCherry-EYFP-ZDHHC17 or mCherry-EYFP-TMEM165. 24 h after transfection, the cells were treated or not with HBSS as indicated. The arrow points to red-only puncta. Bars represent 5 µm (main) and 1 µm (insets). (B) Western blot analysis of HeLa WT or HeLa CALCOCO1 KO cells treated as indicated and analyzed by immunoblotting with the indicated antibodies. Numbers below the blots represent relative intensity of the bands in the blots shown, normalized against loading control. The panels are collected from more than one Western blot experiment, but only a single GAPDH loading control is shown. (C and D) HeLa-CALCOCO1 KO cells reconstituted with EGFP-CALCOCO1(mutLIR+Δ623–691) were left uninduced or were induced for 24 h and then either left untreated or treated as indicated. Cell lysates were analyzed by immunoblotting using the indicated antibodies. The error bars in D represent the mean ± SD of band intensities relative to the loading control from three independent experiments quantified using ImageJ. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. (E and F) HeLa-CALCOCO1 KO cells reconstituted with EGFP-CALCOCO1 Δ145–513 (E) or EGFP-CALCOCO1 Δ623–691 (F) were left uninduced or were induced for 24 h and then treated as indicated. Cell lysates were analyzed by immunoblotting using the indicated antibodies. Numbers below the blots represent relative intensity of the bands in the blots shown, normalized against loading control. MW, molecular weight.

Figure 5.

CALCOCO1 mediates degradation of Golgi by autophagy. (A and B) HeLa-CALCOCO1 KO cells reconstituted with inducible EGFP-CALCOCO1 were left uninduced or induced for 48 h and then left untreated, treated with Baf A1 for 6 h or MG132 for 6 h, or starved (HBSS) for 6 h with or without Baf A1. Cell lysates were analyzed by immunoblotting with the indicated antibodies. In A, the panels are collected from more than one Western blot experiment, but for clarity, only a single GAPDH loading control is shown. Numbers below the GABARAP blot represent relative intensity of the bands normalized against loading control. The bars in B represent the mean ± SD of band intensities relative to the loading control from three independent experiments quantified using ImageJ. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as **, P < 0.005; *, P < 0.01. (C and D) HeLa-CALCOCO1 KO cells reconstituted with EGFP-CALCOCO1-mutZDABM were left uninduced or were induced for 24 h and then either left untreated or treated as indicated. Cell lysates were analyzed by immunoblotting using the indicated antibodies. Numbers below the EGFP-CALCOCO1-mutZDABM blot represent relative intensity of the bands normalized against loading control. The error bars in D represent the mean ± SD of band intensities relative to the loading control from three independent experiments quantified using ImageJ. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as **, P < 0.005. (E) HeLa-CALCOCO1 KO cells reconstituted with inducible expression of EGFP-CALCOCO1 were induced for 24 h and then left untreated or treated with Baf A1 or starved for the indicated times. The abundance of the Golgi apparatus was then quantified from wide-field fluorescence images with immunostaining of endogenous GM130, acquired in random fashion. For each condition analyzed, 25 regions of interest (typically containing 1,200–1,800 cells in total) were randomly chosen across each well, and the average fluorescence intensity of GM130-positive structures contained inside the total cell area population of chosen regions was measured. Images were analyzed in Volocity software using a custom-made measurement protocol to segment images into populations of objects representing nuclei, total cell area, and Golgi. The error bars represent mean ± SD of the intensities of GM130-positive structures in the analyzed cells. Statistical comparison was analyzed by one-way ANOVA followed by Tukey multiple comparison test. Significance is displayed as ***, P < 0.001; *, P < 0.01. (F) A model of the CALCOCO1 bridging the gap between the Golgi and autophagic membrane. MW, molecular weight.