GMAP is an Atg8a-interacting protein that regulates Golgi turnover in Drosophila

Abstract

Selective autophagy receptors and adapters contain short linear motifs called LIR motifs (LC3-interacting region), which are required for the interaction with the Atg8-family proteins. LIR motifs bind to the hydrophobic pockets of the LIR motif docking site (LDS) of the respective Atg8-family proteins. The physiological significance of LDS docking sites has not been clarified in vivo. Here, we show that Atg8a-LDS mutant Drosophila flies accumulate autophagy substrates and have reduced lifespan. Using quantitative proteomics to identify the proteins that accumulate in Atg8a-LDS mutants, we identify the cis-Golgi protein GMAP (Golgi microtubule-associated protein) as a LIR motif-containing protein that interacts with Atg8a. GMAP LIR mutant flies exhibit accumulation of Golgi markers and elongated Golgi morphology. Our data suggest that GMAP mediates the turnover of Golgi by selective autophagy to regulate its morphology and size via its LIR motif-mediated interaction with Atg8a.

Keywords: CP: Cell biology; CP: Immunology; Golgi; Golgiphagy Drosophila; LIR motif; LIR motif docking site; autophagy.

Graphical abstract

Figure 1

Characterization of Atg8a^K48A/Y49A (LDS) mutant flies (A) Genomic DNA from Atg8a ^K48A/Y49A mutant flies was extracted, and the sequenced results confirmed the successful incorporation of the K48A/Y49A mutation. (B) Wild-type, Atg8a^KG07569, and Atg8a^K48A/Y49A mutant flies were aged for 2 weeks. Western blot analysis of lysates from whole flies showed that Ref(2)P, Kenny, and ubiquitin were accumulated in both Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies. (C) Quantification of the western blottings in (B) shows significant accumulations of the aforementioned proteins in both Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies. (D and E) Confocal images from 2-week-old adult brains. Ref(2)P (green) and ubiquitin (red) aggregates (arrows) can be seen in Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies and not in wild-type flies. DNA was dyed with Hoechst (blue). Scale bars, 60 μm. (F) Survival test of wild-type, Atg8a ^KG07569, and Atg8a^K48A/Y49A mutant flies. The results show that Atg8a and Atg8a^K48A/Y49A mutant flies have a short lifespan. Bar charts show means ± SD. Statistical significance was determined using two-tailed Student’s t test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. Number of biological repeats (N = 3 for all figures). Genotypes for all figures: control: w¹¹¹⁸/Y, Atg8a LDS: Atg8a ^K48A/Y49A/Y, Atg8a: Atg8a ^KG07569/Y.

Figure 2

Quantitative proteomic analysis of Atg8a^K48A/Y49A mutant flies (A) Principal-component analysis (PCA) of wild-type, Atg8a^KG07569, and Atg8a^K48A/Y49A mutant adult Drosophila heads. Two-week-old male flies were selected and their heads were collected to perform the proteomic analysis. Four biological replicates were performed for each sample. PCA divided the 12 protein samples into three obvious groups. (B) Venn diagram representing upregulated proteins in Drosophila mutant flies. The cut-off p value was set as <0.05 together with a difference of more than 2-fold between mutant and wild-type Drosophila heads. Twenty-nine proteins passed these two criteria and showed upregulated expression in both of Atg8a^KG07569 and Atg8a^K48A/Y49A mutants. (C) The iBAQ intensity is used to show upregulation of Ref(2)P and GMAP. Bar charts show means ± SD. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. Genotypes for all figures: wild-type: w¹¹¹⁸/Y, Atg8a LDS: Atg8a ^K48A/Y49A/Y, Atg8a: Atg8a ^KG07569/Y.

Figure 3

Accumulation of GMAP in Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies (A) Western blot analysis shows that GMAP is accumulated in both Atg8a ^KG07569 and Atg8a^K48A/Y49A mutant flies. (B) Quantification of GMAP in Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies. (C) Confocal images from 2-week-old adult brains, GMAP (green) (arrows) and ubiquitin (red) aggregates can be seen in Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies. DNA was dyed with Hoechst (blue). Scale bars, 10 μm. (D) Average GMAP puncta size is larger in Atg8a^KG07569 and Atg8a^K48A/Y49A mutant flies compared with wild-type flies. Bar charts show means ± SD. Statistical significance was determined using two-tailed Student’s t test. ^∗p < 0.05. Number of biological repeats (N = 3 for all figures). Genotypes for all figures: wild-type: w¹¹¹⁸/Y, Atg8a LDS: Atg8a ^K48A/Y49A/Y, Atg8a: Atg8a ^KG07569/Y.

Figure 4

GMAP interacts with Atg8a via a LIR motif (A) Structure of GMAP. GMAP is a coiled-coil protein which has 12 coiled-coil domains (gray) and a GRAB domain (red). Yellow represents the predicted LIR motif. (B) GMAP has a predicted LIR motif at position 320–325. (C and D) GST-pulldown assay between GST-tagged Atg8a-WT or GST-tagged Atg8a-LDS mutant and His-tagged GMAP. GMAP interacts with Atg8a-WT but significantly less with Atg8a-LDS. GST was used as negative control. (E and F) GST-pulldown assay between GST-tagged Atg8a-WT and His-GMAP or His-GMAP LIR mutant. GMAP interacts with Atg8a. Point mutations of the GMAP LIR motif in positions 322 and 325 by alanine substitutions of the aromatic and hydrophobic residues (F322A and V325A) significantly reduced its binding to Atg8a. GST was used as negative control. A truncated form of GMAP (1–490 aa) was used. Bar charts show means ± SD. Statistical significance was determined using two-tailed Student’s t test. ^∗∗p < 0.01. Number of biological repeats (N = 3 for all figures).

Figure 5

GMAP regulates Golgi turnover via autophagy (A) Confocal images showing co-localization of endogenous Atg8a and the GMAP under starvation conditions in adult fat body in control and GMAP ^F322A/V325A mutant flies. (B) Western blots showing accumulation of the Golgi marker GM130 in Atg8a ^KG07569 and Atg8a^K48A/Y49A mutant flies compared with wild-type flies. (C) Western blots showing accumulation of GM130 in GMAP-RNAi lines compared with control RNAi and its quantification shown below. (D) Western blots showing accumulation of GM130 in GMAP ^F322A/V325A mutant flies compared with wild-type flies. (E) Immunofluorescence confocal microscopy of Drosophila brain showing increased accumulation of the cis-Golgi marker (GM130) and the altered morphology of Golgi in GMAP ^F322A/V325A and Atg8a ^KG07569 mutant flies. Scale bars, 10 μm. (F) Average GM130 puncta size is larger in GMAP ^F322A/V325A and Atg8a ^KG07569 mutant flies compared with wild-type flies. Bar charts show means ± SD. Statistical significance was determined using two-tailed Student’s t test. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001. Number of biological repeats (N = 3 for all figures). Genotypes: (A) Control: w¹¹¹⁸/Y, GMAP: GMAP ^F322A/V325A/Y. (B) Control: w¹¹¹⁸/Y, Atg8a LDS: Atg8a ^K48A/Y49A/Y, Atg8a: Atg8a ^KG07569/Y. (C) Control: yw¹¹¹⁸;P{attP,y[+],w[3`]}/+;da-GAL4/+, GMAP-RNAi: GMAP-RNAi/+; da-GAL4/+. (D and E) Control:w¹¹¹⁸/Y, GMAP: GMAP ^F322A/V325A/Y, Atg8a: Atg8a ^KG07569/Y.