Direct visualization of degradation microcompartments at the ER membrane

Abstract

To promote the biochemical reactions of life, cells can compartmentalize molecular interaction partners together within separated non-membrane-bound regions. It is unknown whether this strategy is used to facilitate protein degradation at specific locations within the cell. Leveraging in situ cryo-electron tomography to image the native molecular landscape of the unicellular alga Chlamydomonas reinhardtii, we discovered that the cytosolic protein degradation machinery is concentrated within ∼200-nm foci that contact specialized patches of endoplasmic reticulum (ER) membrane away from the ER-Golgi interface. These non-membrane-bound microcompartments exclude ribosomes and consist of a core of densely clustered 26S proteasomes surrounded by a loose cloud of Cdc48. Active proteasomes in the microcompartments directly engage with putative substrate at the ER membrane, a function canonically assigned to Cdc48. Live-cell fluorescence microscopy revealed that the proteasome clusters are dynamic, with frequent assembly and fusion events. We propose that the microcompartments perform ER-associated degradation, colocalizing the degradation machinery at specific ER hot spots to enable efficient protein quality control.

Keywords: ERAD; cdc48; cryo-electron tomography; phase separation; proteasome.

Fig. 1.

In Chlamydomonas cells, proteasomes are concentrated at the nuclear envelope and in cytosolic puncta. (A) Live Chlamydomonas mat3-4 cells expressing the tagged proteasome subunit Rpn11-mVenus, imaged in 3D by wide-field deconvolution fluorescence microscopy. Left column: maximum-intensity projection of Rpn11-mVenus, showing localization to the nuclear envelope and cytosolic puncta. Middle column: Rpn11-mVenus (green) overlaid with chlorophyll autofluorescence (magenta). Right column: both fluorescence signals overlaid on a bright-field image, with protruding flagella distinguishing the apical (api) and basal (bas) sides of the cell. (B) Transmission electron microscopy overview image of a Chlamydomonas cell thinned to ∼200 nm with a cryo-FIB. The nucleus, ER, Golgi, and chloroplast are pseudocolored as indicated. (C) Diagram of Chlamydomonas organelle architecture (Left; colored as in B) and fluorescence localization from A (Right). The Rpn11-mVenus puncta are predominantly localized to the cytoplasm between the nucleus and the chloroplast, a region occupied by ER and Golgi. (D) Histogram of the number of cytosolic puncta per cell. N = 565 cells. (E) The intensity of Rpn11-mVenus fluorescence at the nuclear envelope and within the cytosolic puncta, normalized by fold change over each cell’s cytosolic background. Puncta have nearly twice the intensity of the nuclear envelope, indicating higher proteasome concentration. Error bars show SD. (Scale bars: 2 μm in A and B.)

Fig. 2.

Imaged with in situ cryo-ET, cytosolic proteasomes cluster at the ER membrane. (A) Slice through a tomogram that targets the cytoplasm of a Chlamydomonas mat3-4 cell, showing the ER and the Golgi. (B and C) Two different Z slices through the proteasome cluster boxed in A. Red arrows, proteasomes; yellow arrows, Cdc48 (Movie S4). (D) Corresponding segmentation (Golgi, dark gray; ER, light gray; other organelles, white) with in situ subtomogram averages of proteasomes (red, 19.5-Å resolution), Cdc48 (yellow, 32.8-Å resolution), free cytosolic ribosomes (light blue, 18.8-Å resolution), and membrane-bound ribosomes (dark blue, 21.9-Å resolution) mapped back into the cellular volume. (E) Enlarged views of the in situ subtomogram averages. (Scale bars: 200 nm in A, 50 nm in B and C, and 10 nm in E.)

Fig. 3.

Whole-cell mass spectrometry shows that Cdc48 is the most abundant type II AAA-ATPase and confirms the relative abundance of complexes identified by in situ cryo-ET. (A) Scatter plot of the proteome from Chlamydomonas mat3-4 cells (the same strain used for cryo-ET). Protein abundance is plotted as intensity versus the iBAQ value (intensity-based absolute quantification; raw protein intensity divided by the number of peptides) (99). Measured proteasome, ribosome, Cdc48, NSF, and Pex1/6 subunits are marked with red, blue, yellow, light green, and dark green circles, respectively. Note that because Cdc48 and NSF form homo-hexamers, the number of macromolecular complexes is 6-fold lower than the protein abundance. Pex1 and Pex6 form a hetero-hexamer, so the number of macromolecular complexes is 3-fold lower than the protein abundance. (B) The relative levels of 3 common cytosolic type II AAA-ATPases from the whole-cell proteomics, normalized to Cdc48. (C) The relative levels of ribosome, proteasome, and Cdc48 complexes from the proteomics (blue) and cryo-ET (orange). Error bars show SD (between subunits for the mass spectrometry and between tomograms for the cryo-ET). Concentrations are normalized by the ribosome levels. The plot on the Right shows a zoom-in on the proteasomes and Cdc48 to more clearly display their relative abundance. The slightly lower concentration of proteasomes determined by cryo-ET is likely due to the high abundance of proteasomes in the nucleus (5); in this analysis, only cytosolic proteasomes were quantified by cryo-ET, whereas whole cells were measured by mass spectrometry.

Fig. 4.

Proteasomes and Cdc48 form globular ribosome-excluding microcompartments. (A and B) Two close-up views of proteasome clusters within the cell, looking toward the ER membrane (gray) from the cytosol. Top image: view displaying proteasomes (red), Cdc48 (yellow), and ribosomes (membrane-bound, dark blue; free, light blue). Bottom image: the same view with only the ribosomes displayed, revealing that ribosomes are excluded from the proteasome cluster and adjacent patch on the ER membrane (red dashed line). (C) PCA of proteasome cluster shape. Left side: eigenvectors (3 shades of purple) are drawn for a model circle and sphere (Top 2 rows), as well as an example cluster (Bottom row, red dots are proteasome center positions). Right side: the corresponding ratio of eigenvalues for each shape, with SDs displayed as bidirectional arrows. N = 6 proteasome clusters (SI Appendix, Figs. S6 and S7). (D) Radial concentrations of proteasomes, Cdc48, and ribosomes outward from the centroid positions of the proteasome clusters reveal a cellular microcompartment of distinct composition. Proteasomes are strongly accumulated in the microcompartment. Cdc48 are found throughout the cytosol but peak at the microcompartment border. Ribosomes are only found beyond the microcompartment border, reaching a constant concentration throughout the cytosol. The fairly broad transition from proteasomes to ribosomes at the border is primarily due to combining microcompartments of different sizes (range of diameters on long axis: 157 to 235 nm) for this analysis. Error bars show SD. (E) Distances from every proteasome to every cytosolic Cdc48, performed separately for cluster proteasomes (Left plot, yellow) and noncluster proteasomes (Right plot, yellow). The analysis was repeated on simulated data where the same number of Cdc48 complexes was randomly placed into the same cellular volumes (gray). The experimental data for the cluster proteasomes shows a nonrandom peak at <200 nm, indicating that Cdc48 clusters together with proteasomes in the microcompartments.

Fig. 5.

Analysis of proteasome states within degradation microcompartments. (A) Comparison of assembly state and functional state frequencies between the cluster and noncluster cytosolic proteasomes. Cluster proteasomes have a higher percentage of double-capped (80%) compared to the cytosolic population (61%). For both populations, the majority of proteasome caps are in the ground state (69 to 74%). Double-capped, purple; single-capped, lavender; ground state, green; substrate-processing state, pink; unclassified, gray. (B) Functional states of cluster proteasome caps as a function of distance to the ER membrane. The percentages of processing caps are written in pink over each bar in the graph. Proteasomes <20 nm from ER have a significantly higher fraction of processing caps compared to the rest of the cluster proteasomes (for statistical test, see Methods). Subtomogram averaging of these ER-proximal proteasome caps reveals an extra density that likely corresponds to engaged substrate (dashed Inset, red: fitted proteasome molecular structure; see SI Appendix, Fig. S13 for evidence that the extra density is bound at the proteasome’s substrate engagement site). (C) Mapping each of the ER-proximal processing state proteasomes (red) back into the cellular volumes reveals that the extra density (orange) always connects to the ER membrane (gray), consistent with a putative substrate.

Fig. 6.

Molecular architecture of non–membrane-bound degradation microcompartments at the ER. Proteasomes (red) and Cdc48 (yellow) cluster together to form concentrated ∼200-nm microcompartments that exclude cytosolic ribosomes (light blue) (Fig. 4 and SI Appendix, Figs. S6 and S7). These microcompartments directly contact small patches on the rough ER membrane (gray) that are devoid of membrane-bound ribosomes (dark blue) and may be enriched in ERAD substrates. The proteasomes closest to the membrane have regulatory caps that are in the substrate-processing conformation and are engaged at their substrate-binding sites with densities emanating from the ER (Fig. 5 and SI Appendix, Fig. S13). Thus, these densities are likely substrates that are undergoing removal from the ER directly by proteasomes (a noncanonical pathway we term “direct ERAD”). Cdc48 complexes are enriched at the microcompartment periphery, where they may be handing substrates to the proteasomes in the final step of the canonical ERAD pathway. Unlike proteasomes, Cdc48 is commonly found all along the ER membrane (82% of ER-proximal proteasomes, but only 20% of ER-proximal Cdc48, are localized to the microcompartments). Therefore, Cdc48 might extract substrates from the ER, then diffuse through the cytosol until it encounters a proteasome, the majority of which are clustered in microcompartments. Cytosolic diffusion of Cdc48, followed by substrate hand-off to proteasomes, could explain the peripheral localization of Cdc48 around the proteasome clusters. The microcompartments may additionally serve as degradation centers for cytosolic proteins. The microcompartments are positioned away from the larger ribosome-free region of the ER where COPII-coated vesicles bud en route to the Golgi. This spatially segregates the ERAD and secretory pathway machinery, and may aid in sorting ER proteins between trafficking and degradation.