Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 37 (5)

p62 Filaments Capture and Present Ubiquitinated Cargos for Autophagy

Affiliations

p62 Filaments Capture and Present Ubiquitinated Cargos for Autophagy

Gabriele Zaffagnini et al. EMBO J.

Abstract

The removal of misfolded, ubiquitinated proteins is an essential part of the protein quality control. The ubiquitin-proteasome system (UPS) and autophagy are two interconnected pathways that mediate the degradation of such proteins. During autophagy, ubiquitinated proteins are clustered in a p62-dependent manner and are subsequently engulfed by autophagosomes. However, the nature of the protein substrates targeted for autophagy is unclear. Here, we developed a reconstituted system using purified components and show that p62 and ubiquitinated proteins spontaneously coalesce into larger clusters. Efficient cluster formation requires substrates modified with at least two ubiquitin chains longer than three moieties and is based on p62 filaments cross-linked by the substrates. The reaction is inhibited by free ubiquitin, K48-, and K63-linked ubiquitin chains, as well as by the autophagosomal marker LC3B, suggesting a tight cross talk with general proteostasis and autophagosome formation. Our study provides mechanistic insights on how substrates are channeled into autophagy.

Keywords: aggrephagy; cargo receptor; phase transition; quality control; selective autophagy.

Figures

Figure 1
Figure 1. p62 and ubiquitin‐positive proteins spontaneously cluster in solution

Schematic representation of the p62 architecture. Mutations tested in this study are indicated. PB1: Phox and Bem1p domain, ZZ: Zinc finger, LIR: LC3‐interacting region, UBA: ubiquitin‐associated domain.

Schematic representation of the constructs employed in the clustering assay.

Representative micrographs of a clustering assay with the proteins indicated on the left. GFP and mCherry fluorescence are consistently displayed in cyan and magenta, respectively.

Quantification of particle number, size, and estimated volume determined from the data shown in (C).

Quantification of cluster formation by the indicated p62 mutants.

Data information: For all the graphs, averages and SDs from at least three independent replicates are shown.
Figure EV1
Figure EV1. p62 and ubiquitin‐positive proteins spontaneously cluster in solution

Representative micrographs of a cluster formation assay conducted with 2 μM mCherry‐p62, 2 μM GFP‐p62, and 5 μM GST‐4xUb taken after 60 min from the reaction start.

Schematic representation of the particle analysis workflow.

Panel identical to Fig 1C, except for the fact that background was not subtracted from the pictures.

Representative structured illumination microscopy (SIM) micrographs of p62 clusters formed with the indicated proteins. GFP‐p62‐containing clusters were formed for 30 min, and then, the respective mCherry‐p62 mutants were added and incubated for another 30 min. Representative images of at least 10 particles per sample.

Figure 2
Figure 2. Endogenously tagged p62 is oligomeric and forms clusters with ubiquitin

Top: schematic representation of the STG‐p62 construct. Bottom: characterization of several STG‐p62 clones in comparison with parental Hap1 cells.

Wild‐type or STG‐p62 Hap1 cells were mock‐ or puromycin‐treated for 2 h and stained for ubiquitin. The endogenous p62 was stained with an anti‐p62 antibody in the parental Hap1 cells, while the fluorescence of the GFP tag was recorded in STG‐p62 cells. Arrowheads indicate colocalizing puncta.

Purified STG‐p62 was tested for cluster formation in the presence of 20 μM GST‐4xUb. Averages and SDs of three independent replicates are shown for the +GST‐4xUb sample and of two for STG‐p62 alone.

FCS analysis of diffusion (D) and brightness (E) of the GFP1‐5 ruler and of STG‐p62 in the cytoplasm of Hap1 cells. Dashed lines represent the linear regression of the GFP1‐5 ruler data. Regression coefficients are indicated. For all data points, averages and SDs (on both axes for STG‐p62) are shown (see Materials and Methods for a list of the number of measurements performed for each construct). MW: molecular weight. D: diffusion coefficient (μm2/s).

Source data are available online for this figure.
Figure EV2
Figure EV2. Endogenously tagged p62 is oligomeric and forms clusters with ubiquitin

Characterization of the p62 degradation pattern upon starvation in two STG‐p62 clones. Wild‐type and ATG5 −/− Hap1 cells were included as positive and negative controls, respectively. Baf.: bafilomycin A1. STG‐p62 clone 3D was employed for further characterization and in the experiments shown in the main Figure. Anti‐GAPDH blot is shown as loading control.

Characterization of the LC3B lipidation pattern in wild‐type and STG‐p62 Hap1 cells upon starvation and/or bafilomycin A1 treatment. Anti‐GAPDH blot is shown as loading control.

Wild‐type Hap1 cells transfected with the GFP1‐5 ruler constructs were lysed and subjected to Western blotting against GFP. A mock transfection was used as a control.

Representative autocorrelation curves for the GFP1‐5 ruler constructs and STG‐p62. Curves were normalized setting the first value to 1 and the last to 0. Three random curves per sample were averaged and plotted. See Materials and Methods for the detailed number of curves analyzed per sample.

Top: a representative autocorrelation curve of STG‐p62 (cyan) was fitted assuming either a single diffusing species (red) or two (blue). Bottom: plot of the residuals (data minus fit) for both models.

Figure 3
Figure 3. p62‐dependent cluster formation is triggered by poly‐ubiquitinated proteins

Representative micrographs of cluster formation assays conducted with GST‐tagged linear tetra‐, tri‐ or di‐ubiquitin. The total amount of ubiquitin moieties was kept identical in each condition. Scale bars: 10 μm.

Quantification of the data shown in (A).

Quantification of cluster formation assays conducted with the indicated proteins. mCherry‐p62 was pre‐incubated with the indicated 4xUb variants, followed by the addition of streptavidin immediately before imaging.

Representative micrographs of the indicated samples taken 60 min after addition of streptavidin.

Quantification of cluster formation assays conducted with the indicated proteins.

Quantification of cluster formation assays conducted with 2 μM mCherry‐p62 incubated with the indicated concentration of GST‐GFP‐4xUb. Box: the highest particle number per field detected in each sample was plotted against the relative GST‐GFP‐4xUb concentration. Data points were fitted to a sigmoidal curve. The regression coefficient is indicated.

Quantification of cluster formation assays conducted with GFP‐p62 and the indicated GST‐tagged ubiquitin chains.

Quantification of cluster formation assays performed with mCherry‐p62 and GST‐4xUb in the presence of the indicated free ubiquitin chains. Ubiquitin chains were either pre‐mixed with p62 before the addition of GST‐4xUb (H), or added 3 min after the addition of GST‐4xUb to p62 (I).

Quantification of cluster formation assays conducted with mCherry‐2xFKBP‐p62 in the presence of AP20187, GST‐4xUB, and the indicated Ub chains.

Representative micrographs of glutathione beads coated with the indicated GST‐tagged proteins and incubated with the indicated mCherry‐p62 variants.

Data information: In all pictures, the brightness was adjusted to highlight the profile of the beads. For all graphs, averages and SDs from at least three independent replicates are shown.
Figure EV3
Figure EV3. p62‐dependent cluster formation is triggered by poly‐ubiquitinated proteins

Representative images of cluster formation assays conducted with 5 μM GST‐4xUb and the indicated mCherry‐p62 concentrations. Images were taken before the addition of GST‐4xUb or 60 min after. Representative images of three independent replicates.

Quantification of p62 cluster formation in the presence of the indicated free ubiquitin chains. GST‐4xUb was used as a positive control. Averages and SDs from three independent replicates are shown. Samples were imaged for 15 min and then again after 45 min. Dashed lines represent the extrapolated curves for a full time lapse. For GST‐4xUb data points from the peak of the curve to the latest time point (t = 45 min) were fitted to a single exponential decay (R 2 = 0.9973). Box: representative micrographs of the indicated samples 45 min after addition of Ub chains. Pictures are displayed in false colors (ImageJ: fire).

Left: Coomassie Brilliant Blue‐stained gel showing the assembly of GST‐tagged K48‐ and K63‐linked ubiquitin chains. Sup.: supernatant. Right: purified GST‐tagged ubiquitin chains were subjected to Western blotting with the indicated antibodies.

Left: quantification of aggregation assays conducted with 2 μM mCherry‐p62 and 20 μM GST‐4xUb in the presence or absence of 200 μM GFP‐Ub, GFP, or Ub. Averages and SDs from four independent experiments are shown.

mCherry‐2xFKBP‐p62 was subjected to size‐exclusion chromatography in the presence or absence of the homodimerizer AP20187. Fractions were collected and run on a SDS–PAGE gel. The corresponding elution volume is indicated for each fraction.

Cluster formation assays conducted with mCherry‐p62 WT or ‐2xFKBP in the presence or absence of GST‐4xUb and AP20187. Average and SDs of three independent experiments are shown.

Figure 4
Figure 4. p62 and ubiquitin show different motilities within the clusters

mCherry‐p62/GST‐GFP‐4xUb clusters were photobleached, and fluorescence recovery in the bleached area was monitored. The mCherry‐tagged low‐complexity domain (LCD) of hnRNPA1 (hnRNPA1LCD) was included as a reference for a liquid droplet‐forming protein. Left: representative micrographs of bleached particles. White arrows indicate the position and orientation of the kymographs shown in Fig EV4A. Right: quantification of fluorescence recovery. Individual traces of all particles analyzed from three independent replicates are displayed.

p62 puncta in STG‐p62 cells were photobleached, and the fluorescence recovery was monitored. Left: representative micrographs of bleached cells. White squares indicate the bleached puncta. Insert: magnification of the bleached puncta. Right: quantification of fluorescence recovery. Averages and SDs of 28 (unbleached) and 29 (bleached) puncta from three independent experiments are shown.

Representative electron micrographs of negatively stained p62 filaments incubated in the presence or absence of GST‐4xUb for the indicated times.

Figure EV4
Figure EV4. p62 and ubiquitin show different motilities within the clusters

Representative kymographs of the samples shown in Fig 4A.

Structured illumination imaging (SIM) was performed with GFP‐ and mCherry‐p62‐containing particles formed as indicated on the left. Representative micrographs of at least 20 particles per sample from two independent experiments.

Figure 5
Figure 5. NBR1 cooperates with p62

Top: schematic representation of the domain architecture of the GFP‐NBR1 construct. PB1: Phox and Bem1p domain, ZZ: Zinc finger, CC: coiled coil, LIR: LC3‐interacting region, UBA: ubiquitin‐associated domain. Bottom: GFP‐NBR1 was recruited to glutathione beads coated with the indicated GST fusion proteins and imaged by spinning disk microscopy. Representative micrographs of at least 70 beads imaged per sample from two independent experiments.

mCherry‐p62 variants were recruited to GFP‐ or GFP‐NBR1‐coated beads and imaged as in (A). (B) Representative micrographs of at least 70 beads per sample from three independent replicates. (C) Quantification of the data shown in (B). Data were normalized to p62 wild‐type binding to GFP‐NBR1 within each replicate. Averages and SDs of three independent experiments are shown.

2 μM mCherry‐p62 and/or GFP‐NBR1 were incubated in the presence of 5 μM GST‐4xUb. Representative micrographs (D) and quantification of particle number for each channel (E) are shown. Averages and SDs from at least three independent experiments are shown.

Figure EV5
Figure EV5. NBR1 cooperates with p62

Left: quantification of cluster formation assays conducted with 2 μM GFP‐p62 or GFP‐NBR1 in the presence of 20 μM GST‐4xUb. Averages and SDs from three independent replicates are shown. Right: Representative Coomassie‐stained gel of the samples collected after imaging and run on an SDS–PAGE gel.

mCherry‐OPTN was recruited to GST‐ or GST‐4xUb‐coated beads and imaged by spinning disk microscopy. Representative images of at least 80 beads imaged per sample from three independent experiments.

Left. Representative micrographs of cluster formation assays conducted with GFP‐p62 and GST‐4xUb in the presence or absence of mCherry‐OPTN. The displayed pictures were taken 60 min after the addition of GST‐4xUb. Middle and right: quantification of the number of particles formed by mCherry‐OPTN (middle) and GFP‐p62 (right) in the indicated samples. All graphs show averages and SDs from three independent experiments.

Figure EV6
Figure EV6. LIR‐mediated cross talk between cluster formation and the autophagy machinery

Representative micrographs of clusters formed with 1 μM mCherry‐p62 WT or LIR mutant and 20 μM GST‐4xUb for 10 min, followed by the addition of 1 μM GFP or GFP‐LC3B immediately prior to imaging. Clusters deposited on the bottom of the plate were imaged by spinning disk microscopy at 63× magnification.

GFP‐p62‐ and GST‐4xUb‐containing clusters were formed in the presence of pre‐mixed 20 μM mCherry or mCherry‐LC3B employing p62 wild type or the LIR mutant. The maximal number of particles per field formed per sample was taken as readout. Data were normalized to the samples containing p62 and GST‐4xUb only for both p62 wild type and the LIR mutant. Averages and SDs from three independent experiments are shown. P‐values were calculated with a two‐tailed unpaired Student's t‐test.

Quantification of cluster formation reactions conducted with 2 μM mCherry‐2xFKBP‐p62 in the presence of 5 μM GST‐4xUb and 20 μM of the indicated proteins. Averages and SDs from three independent replicates are shown.

Representative SIM micrographs of p62 clusters formed with the indicated proteins. GFP‐p62‐containing clusters were formed for 30 min, and then, the respective mCherry‐p62 mutants were added and incubated for another 30 min. Representative images of at least 10 clusters per sample are shown.

The indicated p62 variants were recruited to glutathione beads coated with the indicated GST‐tagged proteins and imaged under a spinning disk microscope (E) or a conventional confocal microscope (F). Representative pictures of two independent experiments per condition are shown.

Figure 6
Figure 6. LIR‐mediated cross talk between cluster formation and the autophagy machinery

Cluster formation assays were conducted in the presence or absence of 20 μM mCherry‐LC3B or mCherry, either pre‐mixed with p62 or added 1 min after the addition of GST‐4xUb. Quantification of particle number (left) and size (right) are shown.

Cluster formation assays were conducted with mCherry‐p62 wt or LIR mutant in the presence of GST‐4xUb. Left: representative images of the samples two minutes after the addition of GST‐4xUb. Middle: quantification of particle count. The dashed line indicates the time point of images displayed on the left. Right: input gel of the samples after reaction. The relative position of marker bands is indicated.

Top left: Western blots of lysates of CRISPR/Cas9 genome‐edited STG‐p62 wild‐type and LIR mutant cells. Clone Lir1B was used for further experiments shown here. Bottom left: Representative pictures of STG‐p62 WT or LIR mutant Hap1 cells treated with puromycin for 90 min and stained with anti‐Ub conjugates antibody (FK2). Right: quantification of cytoplasmic p62 puncta per cell (top), and of their co‐localization with ubiquitin upon puromycin treatment (bottom). In total, at least 1,440 cells were counted per condition.

Data information: For all graphs, averages and SDs from at least three independent replicates are shown. Source data are available online for this figure.
Figure 7
Figure 7. p62‐dependent aggrephagy compensates for proteasomal deficiency or overload
Model for p62‐dependent cargo nucleation and autophagic degradation of poly‐ubiquitinated proteins, and its cross talk with the UPS.

Similar articles

See all similar articles

Cited by 27 PubMed Central articles

See all "Cited by" articles

References

    1. Babu JR, Geetha T, Wooten MW (2005) Sequestosome 1/p62 shuttles polyubiquitinated tau for proteasomal degradation. J Neurochem 94: 192–203 - PubMed
    1. Bertipaglia C, Schneider S, Jakobi AJ, Tarafder AK, Bykov YS, Picco A, Kukulski W, Kosinski J, Hagen WJ, Ravichandran AC, Wilmanns M, Kaksonen M, Briggs JA, Sachse C (2016) Higher‐order assemblies of oligomeric cargo receptor complexes form the membrane scaffold of the Cvt vesicle. EMBO Rep 17: 1044–1060 - PMC - PubMed
    1. Bjorkoy G, Lamark T, Brech A, Outzen H, Perander M, Overvatn A, Stenmark H, Johansen T (2005) p62/SQSTM1 forms protein aggregates degraded by autophagy and has a protective effect on huntingtin‐induced cell death. J Cell Biol 171: 603–614 - PMC - PubMed
    1. Cha‐Molstad H, Sung KS, Hwang J, Kim KA, Yu JE, Yoo YD, Jang JM, Han DH, Molstad M, Kim JG, Lee YJ, Zakrzewska A, Kim SH, Kim ST, Kim SY, Lee HG, Soung NK, Ahn JS, Ciechanover A, Kim BY et al (2015) Amino‐terminal arginylation targets endoplasmic reticulum chaperone BiP for autophagy through p62 binding. Nat Cell Biol 17: 917–929 - PMC - PubMed
    1. Cha‐Molstad H, Yu JE, Feng Z, Lee SH, Kim JG, Yang P, Han B, Sung KW, Yoo YD, Hwang J, McGuire T, Shim SM, Song HD, Ganipisetti S, Wang N, Jang JM, Lee MJ, Kim SJ, Lee KH, Hong JT et al (2017) p62/SQSTM1/Sequestosome‐1 is an N‐recognin of the N‐end rule pathway which modulates autophagosome biogenesis. Nat Commun 8: 102 - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources

Feedback