Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 45 (1), 114-131.e8

The RAB11A-Positive Compartment Is a Primary Platform for Autophagosome Assembly Mediated by WIPI2 Recognition of PI3P-RAB11A

Affiliations

The RAB11A-Positive Compartment Is a Primary Platform for Autophagosome Assembly Mediated by WIPI2 Recognition of PI3P-RAB11A

Claudia Puri et al. Dev Cell.

Abstract

Autophagy is a critical pathway that degrades intracytoplasmic contents by engulfing them in double-membraned autophagosomes that are conjugated with LC3 family members. These membranes are specified by phosphatidylinositol 3-phosphate (PI3P), which recruits WIPI2, which, in turn, recruits ATG16L1 to specify the sites of LC3-conjugation. Conventionally, phosphatidylinositides act in concert with other proteins in targeting effectors to specific membranes. Here we describe that WIPI2 localizes to autophagic precursor membranes by binding RAB11A, a protein that specifies recycling endosomes, and that PI3P is formed on RAB11A-positive membranes upon starvation. Loss of RAB11A impairs the recruitment and assembly of the autophagic machinery. RAB11A-positive membranes are a primary direct platform for canonical autophagosome formation that enables autophagy of the transferrin receptor and damaged mitochondria. While this compartment may receive membrane inputs from other sources to enable autophagosome biogenesis, RAB11A-positive membranes appear to be a compartment from which autophagosomes evolve.

Keywords: RAB11A; WIPI2; autophagy; autophagy platform; macroautophagy; mitophagy; recycling endosome; transferrin receptor.

Figures

None
Figure 1
Figure 1
WIPI2 Interacts with RAB11A (A) GFP-WIPI2 and GFP were immuno-precipitated using GFP-TRAP on lysates from HeLa cells stably expressing GFP-WIPI2 or GFP-empty vector under starvation conditions (EBSS) for 2 hr; blots were probed as indicated. (B) Binding of GST-WIPI2 to RAB11A-liposomes was analyzed by probing for liposome-bound GST (GST antibody lanes 1–2) or GST-WIPI2 (WIPI2 antibody, lanes 3–4) (see STAR Methods). Data are means ± SEM, n = 3; One-way ANOVA with post hoc Tukey's test, ∗∗p < 0.01; NS, not significant. (C) HeLa cells treated with control or RAB11A siRNA, starved for 2 hr, and labeled for WIPI2 and ATG16L1. Quantification of WIPI2-, ATG16L1-single-positive structures (number/cell) is shown. WIPI2-ATG16L1 double-positive structures are expressed as percentage of total WIPI2 or ATG16L1 structures. Data are means ± SEM (n = 6 for WIPI2, n = 3 for ATG16L1, 50 cells per condition); two-tailed paired t test, ∗∗p < 0.01, ∗∗∗∗p < 0.0001; NS, not significant. (D) HeLa cells treated as in (A) transfected with GFP-WIPI2 in combination with mCherry-empty or mCherry-RAB11A WT or mutants. Quantification of WIPI2 structures/cell is shown. Data are means ± SEM (n = 3, 40 cells per condition); One-way ANOVA with post hoc Tukey's test, ∗∗p < 0.01, ∗∗∗∗p < 0.0001; NS, not significant. (E) HeLa cells transfected with GFP-RAB11A WT or mutants were starved for 1 hr and processed for GFP-TRAP. The amount of WIPI2 pull-down by RAB11A WT and mutants is shown. Data are means ± SEM (n = 3 independent experiments); one-way ANOVA with post hoc Tukey's test, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001; NS, not significant. (F) Alignment of amino acid sequence of the RAB11-binding domain (RBD) of RAB11FIPs with WIPI2 (isoform B, residues 133–161). Colored box, white character strict identity. Consensus sequence for RAB11 binding is shown; hydrophobic residues forming a RAB11-binding patch are marked with asterisks. (G) HeLa cells transfected with GFP-WIPI2 WT, GFP-WIPI2 LE115AT and GFP-WIPI2 YI120FE and starved for 2 hr were processed for GFP-TRAP as in (A). Data are means ± SEM, n = 5; one-way ANOVA, p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001. (H) Binding of WT, LE115AT, YI120FE, and FRRG223FTTG WIPI2-FLAG recombinant proteins to RAB11A-containing liposomes was measured in vitro by liposome sedimentation assay (see STAR Methods). WIPI2 association with liposomes is shown. Data are means ± SEM, n = 4; one-way ANOVA with post hoc Tukey's test, ∗∗∗∗p < 0.0001; NS, not significant.
Figure 2
Figure 2
WIPI2 Mutants Defective in RAB11A Binding Fail to Localize on Autophagosomes and to Sustain Autophagosome Formation (A) Representation of the human WIPI2 protein harboring 7 WD repeats. (B and C) HeLa cells were transiently transfected with GFP-WIPI2 WT and mutants in combination with FLAG or FLAG-ATG4BC74A, starved for 2 hr, and labeled for FLAG and ATG16L1. The number of WT or mutant WIPI2 dots is shown. Data are means ± SEM, n = 3, 50 cells per condition; one-way ANOVA with post hoc Tukey's test, p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (D–F) HeLa cells transfected with control or WIPI2 siRNA were transfected with GFP-empty, or GFP-WIPI2 WT and mutants, incubated for 2 hr (D) or 4 hr (E) in EBSS, and labeled for ATG16L1 (D) or LC3 (E). (D-E) ATG16L1 or LC3 vesicles number/cell are shown. In (F), WIPI2-silenced cells were transfected with GFP-empty or GFP-WIPI2 WT and mutants in combination with HA-tagged Htt-Q74 and processed for immunofluorescence. Number of cells with Htt-Q74 aggregates is shown. Data are means ± SEM, n = 3, 40 cells (D and E) or 800 cells (F) per condition; one-way ANOVA with post hoc Tukey's test, p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001; NS, not significant. (G) CRISPR/Cas9 ATG16L1 knockout and control cells were transfected with GFP-WIPI2 WT, starved for 2 hr, and labeled for RAB11A. WIPI2 puncta/cell are shown. Data are means ± SEM, n = 3, 25 cells per condition; one-way ANOVA with post hoc Tukey's test, p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001; NS, not significant. Also see Video S1. (H) Binding of purified WIPI2-FLAG proteins (WT and mutants) was measured by liposome sedimentation assay (see STAR Methods). WIPI2 binding to PI3P-liposomes is shown. Data are means ± SEM, n = 3; one-way ANOVA with post hoc Tukey's test, p < 0.05; NS, not significant. (I) Liposomes carrying RAB11A, PI3P or combination of both were tested for the ability to recruit WIPI2 by liposome sedimentation assay (see STAR Methods). WIPI2 membrane binding is shown. Data are means ± SEM, n = 5; one-way ANOVA with post hoc Tukey's test, p < 0.05, ∗∗∗p < 0.001. (J) HeLa cells transiently transfected with GFP-FRB-RAB5A and RFP-FKBP-RAB11A were treated with 250 nM AP21967 for 30 min in full-media and stained for WIPI2 and anti-RFP. The graphs on the left show the relative fluorescence intensity of the area under the arrows; the histogram on the right shows quantification of RAB11A/WIPI2 translocation to early endosomes by addition of AP21967. Data are means ± SEM, n = 3; one-way ANOVA with post hoc Tukey's test, ∗∗∗∗p < 0.0001.
Figure 3
Figure 3
Autophagic Proteins Are in RAB11A Compartments (A) Colocalization of different autophagic proteins with RAB11A or calnexin (MAM) in basal or under hypotonic medium treatment is shown. Data are means ± SEM, n = 3, 20 cells per condition; two-tailed paired t test; p < 0.05; ∗∗∗p < 0.001; NS, not significant. Relevant images in Figure S3. (B) HeLa transfected with GFP-DFCP1 were labeled for RAB11A and calnexin and visualized on Elyra (Zeiss) superresolution microscope. (C) GFP-RAB11A HeLa labeled for ATG14 and calnexin and processed for 3D z-stack reconstruction using Volocity software. (D) HeLa cells treated with VPS34-IN1 were stained for RAB11A and PI3P and the overlap between PI3P/RAB11A was measured. Data are means ± SEM, n = 3, 35 cells per condition; one-way ANOVA, ∗∗p < 0.01. (E) HeLa cells transfected with mCherry-RAB11A and GFP-FYVE4X were recorded under starvation in presence or absence of VPS34-IN1. Corresponding regions of the GFP-FYVE4X and Cherry-RAB11A channels are shown with dashed circles. (F) HeLa cells transfected with GFP-WIPI2, starved for 2 hr, and labeled for RAB11A and ATG14. Corresponding regions of the different channels are shown with dashed circles.
Figure 4
Figure 4
LC3 Is in RAB11A Compartments (A) HeLa cells labeled for RAB11A and LC3. The graphs show the relative fluorescence intensity of the area under the arrow. See also Video S2. (B) HeLa cells transfected with RFP-LC3 and labeled for RAB10. (C) Colocalization of RFP-LC3 with RAB11A or calnexin in cells treated or not with hypotonic medium or nocodazole. Data are means ± SEM, n = 3, 20 cells per condition; two-tailed paired t test, ∗∗p < 0.01; NS, not significant. Relevant images in Figure S4I. (D) HeLa cells transfected with PAM-mCherry-LC3 and processed for whole-mount immunoelectron microscopy (see STAR Methods). The asterisk indicates the LC3-positive structure. The edges of the recycling endosomes (green) and the autophagic structure (red) are highlighted below. (E and F) HeLa cells transfected with RFP-GFP tandem fluorescent-tagged LC3 were labeled for the RAB11A and processed for superresolution microscopy. The arrow indicates an acidified red-only autophagosome separate from the RAB11A compartment. In (F), the quantification of the distance (nm) of LC3 dots (red only or yellow) from the closest RAB11A tubule was assessed using Volocity 6.3 Software. Data are means ± SEM (30 structures per condition); two-tailed paired t test, ∗∗∗∗p < 0.0001. (G and H) HeLa cells treated with ATG2A/B siRNA, transfected with RFP-LC3 and labeled for RAB11A were processed for superresolution microscopy and the distance of LC3 dots from the closest RAB11A tubule quantified (G) as in (F). Data are means ± SEM, 130 structures per condition; two-tailed paired t test, ∗∗∗∗p < 0.0001. (I) HeLa cells transfected with GFP-WIPI2, mCherry-RAB11A and CFP-LC3 were imaged for 30 min (see also Video S3). The area delimited by a circle shows a new autophagosome forming on recycling endosome. Dashed line circles were used in time frames where the fluorescence signal disappears.
Figure 5
Figure 5
TfR Is Recruited to LC3 Vesicles in Starvation Conditions (A) HeLa cells transfected with GFP-LC3, were loaded with transferrin Alexa 555 for 1 hr in Hank’s balanced salt solution or full medium and chased for 15 min. The Tf/LC3 and LC3/Tf signal overlaps are shown. Data are means ± SEM, n = 3; two-tailed paired t test, ∗∗∗p < 0.001. (B) HeLa cells treated as in (A) were processed for immunogold labeling on cryosections. Arrows specify autophagosome double membranes, black outer, white inner. (C) HeLa cells starved for 1 hr and loaded with anti-TfR antibody were processed for pre-embedding electron microscopy. (D) HeLa cells treated with ATG2A/B siRNA (left) were processed as in (C). HeLa cells transfected for Flag-ATG4BC74A were loaded with Tf and processed as in (B). LD, lipid droplet. Arrows specify autophagosome double membranes as in (B). (E) HeLa cells treated with control or ATG7 siRNA and/or transfected with ATG7 were processed for immunoblotting and the TfR levels measured. Data are means ± SD, n = 3; two-tailed paired t test, ∗∗∗p < 0.001. (F) HeLa cells treated with control or RAB11A or WIPI2 siRNAs were processed as in (E). Data are means ± SD, n = 3; two-tailed paired t test, ∗∗p < 0.01, ∗∗∗p < 0.001. (G) HeLa cells starved for 1 hr, loaded with Ferrofluid-Tf Alexa 488 for 1 hr and chased for 15 min in starvation medium. The cells were then fragmented and the membranes containing Ferrofluid-Tf488 (bound) or not containing Ferrofluid-Tf488 (unbound) were separated and processed for immunoblotting (see Figure S5G). (H) HeLa cells transfected with control or ATG2A/B siRNA were processed as in (G). The amount of LC3II in bound fraction in ATG2 knockdown condition is expressed as percentage of control. Data are means ± SEM, n = 3. (I) HeLa cells treated with control or ULK1 siRNA and processed as in (G). The LDH activity was measured in the bound fraction. Data are means ± SEM, n = 3; two-tailed paired t test, ∗∗∗p < 0.001.
Figure 6
Figure 6
Photo-Damaged Mitochondria Are Engulfed by RAB11A-Positive Membranes (A and B) GFP-RAB11A HeLa cells transfected with RFP-LC3, incubated with MitoTracker Deep Red, and processed for live imaging. Mitochondria were photo-damaged using 488-nm laser light. See also Video S4. (C) GFP-RAB11A HeLa cells transfected with BFP-Sec61, labeled with MitoTracker Red, and processed as in (A). See also Videos S5 and S6. The relative fluorescence intensity of the area under the arrow is shown. Arrowheads indicate RAB11A engulfment of mitochondria with no ER. (D) The percentage of photo-damaged mitochondria surrounded by the different markers was measured in HeLa cells transfected as indicated and processed as in (A). Relevant images are shown in Figure S6. Data are means ± SEM, n = 3, 10 cells per condition; two-tailed paired t test; p < 0.05; ∗∗∗p < 0.001; NS, not significant.
Figure 7
Figure 7
Mitochondria Are Engulfed by RAB11A-Positive Membranes during Parkin-Dependent and -Independent Mitophagy (A) HeLa cells treated with ATG2A/B siRNA were processed as in Figure 5D. (B) GFP-RAB11A HeLa cells treated with DFP, labeled with MitoTracker Red, and imaged. Line scan analysis of the area under the arrow is shown in the graph on the right. (C) HeLa cells treated as in (B) were stained for LC3 and RAB11A. Z-stacks of the area in the inset were processed by ZEN software and presented as max z-projected confocal images. (D) HA-PARKIN HEK cells transfected with GFP-RAB11A, labeled with MitoTracker Red, and treated with CCCP were imaged. Line scan analysis of the area under the arrow is shown in the graph on the right. (E) Lysates from HA-PARKIN HEK cells treated with control, RAB11A, or WIPI2 siRNA and incubated with DMSO or 5 μM CCCP for 6 hr were probed as indicated. Data are means ± SD, n = 3; two-tailed unpaired t test, p < 0.05. (F) The cartoon summarizes events that lead to formation of the autophagosome: coincident detection of PI3P and RAB11A by WIPI2 (1) and recruitment of ATG16L1 complex (2–3). Here we show that all the relevant events preceding the completion of autophagosome formation (4) occur on the same membrane district (the platform), the RAB11A-positive membranes. Light green, inactive RAB11A, dark green, active RAB11A.

Comment in

  • Phagophores Evolve From Recycling Endosomes
    C Puri et al. Autophagy 14 (8), 1475-1477. PMID 29940791.
    The membrane origins of autophagosomes have been a key unresolved question in the field. The earliest morphologically recognizable structure in the macroautophagy/autopha …

Similar articles

See all similar articles

Cited by 19 PubMed Central articles

See all "Cited by" articles

References

    1. Allen G.F., Toth R., James J., Ganley I.G. Loss of iron triggers PINK1/Parkin-independent mitophagy. EMBO Rep. 2013;14:1127–1135. - PMC - PubMed
    1. Axe E.L., Walker S.A., Manifava M., Chandra P., Roderick H.L., Habermann A., Griffiths G., Ktistakis N.T. Autophagosome formation from membrane compartments enriched in phosphatidylinositol 3-phosphate and dynamically connected to the endoplasmic reticulum. J. Cell Biol. 2008;182:685–701. - PMC - PubMed
    1. Bago R., Malik N., Munson M.J., Prescott A.R., Davies P., Sommer E., Shpiro N., Ward R., Cross D., Ganley I.G. Characterization of VPS34-IN1, a selective inhibitor of Vps34, reveals that the phosphatidylinositol 3-phosphate-binding SGK3 protein kinase is a downstream target of class III phosphoinositide 3-kinase. Biochem. J. 2014;463:413–427. - PMC - PubMed
    1. Bento C.F., Ashkenazi A., Jimenez-Sanchez M., Rubinsztein D.C. The Parkinson's disease-associated genes ATP13A2 and SYT11 regulate autophagy via a common pathway. Nat. Commun. 2016;7:11803. - PMC - PubMed
    1. Biazik J., Yla-Anttila P., Vihinen H., Jokitalo E., Eskelinen E.L. Ultrastructural relationship of the phagophore with surrounding organelles. Autophagy. 2015;11:439–451. - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources

Feedback