Temporal control of acute protein aggregate turnover by UBE3C and NRF1-dependent proteasomal pathways

A hallmark of neurodegenerative diseases is the progressive loss of proteostasis, leading to the accumulation of misfolded proteins or protein aggregates, with subsequent cytotoxicity. To combat this toxicity, cells have evolved degradation pathways (ubiquitin-proteasome system and autophagy) that detect and degrade misfolded proteins. However, studying the underlying cellular pathways and mechanisms has remained a challenge, as formation of many types of protein aggregates is asynchronous, with individual cells displaying distinct kinetics, thereby hindering rigorous time-course studies. Here, we merge a kinetically tractable and synchronous agDD-GFP system for aggregate formation with targeted gene knockdowns, to uncover degradation mechanisms used in response to acute aggregate formation. We find that agDD-GFP forms amorphous aggregates by cryo-electron tomography at both early and late stages of aggregate formation. Aggregate turnover occurs in a proteasome-dependent mechanism in a manner that is dictated by cellular aggregate burden, with no evidence of the involvement of autophagy. Lower levels of misfolded agDD-GFP, enriched in oligomers, utilizes UBE3C-dependent proteasomal degradation in a pathway that is independent of RPN13 ubiquitylation by UBE3C. Higher aggregate burden activates the NRF1 transcription factor to increase proteasome subunit transcription, and subsequent degradation capacity of cells. Loss or gain of NRF1 function alters the turnover of agDD-GFP under conditions of high aggregate burden. Together, these results define the role of UBE3C in degradation of this class of misfolded aggregation-prone proteins and reveals a role for NRF1 in proteostasis control in response to widespread protein aggregation.


Cell lysis and immunoblotting assay
Protocols for these procedures are provided at: dx.doi.org/10.17504/protocols.io.4r3l226e4l1y/v1.Briefly, cells were cultured in the presence of the corresponding stress to 60-80% confluency in 6well plates, 10 cm or 15 cm dishes.After removing the media, the cells were washed with DPBS three times.To lyse cell urea buffer (8M urea, 50 mM TRIS 7.5, 150 mM NaCl, containing mammalian protease inhibitor cocktail (Sigma), Phos-STOP, and 20 unit/ml Benzonase (Millipore)) was added directly onto the cells.Cell lysates were collected by cell scrapers and sonicated on ice for 10 seconds at level 5, and lysates were cleared by centrifugation (15000 rpm, 10 min at 4 °C).The concentration of the supernatant was measured by BCA assay.For immunoblotting, the whole cell lysate was denatured by the addition of LDS sample buffer supplemented with 100 mM DTT, followed by boiling at 95 °C for 5 minutes.10-20 µg of each lysate was loaded onto the 4-20% Tris-Glycine gel (Thermo Fisher Scientific), Bolt 8% Bis-Tris gel, or 4-12% NuPAGE Bis-Tris gel (Thermo Fisher Scientific), followed by SDS-PAGE with Tris-Glycine SDS running buffer (Thermo Fisher Scientific) or MOPS SDS running buffer (Thermo Fisher Scientific), respectively.The proteins were electro-transferred to nitrocellulose membranes and then the total protein was stained using Ponceau (Thermo Fisher Scientific).The membrane was then blocked with LI-COR blocking buffer at room temperature for 1h.Then membranes were incubated with the indicated primary antibodies (4°C, overnight), washed three times with TBST (total 30 min), and further incubated either with fluorescent IRDye 680RD Goat anti-Mouse IgG H+L, or IRDye 800CW Goat anti-Rabbit IgG H+L secondary antibody at (1:10,000) at room temperature for 1h.After thorough wash with TBST for 30 min, near infrared signal was detected using OdysseyCLx imager (LI-COR Bioscience) and quantified using ImageStudioLite (LI-COR).

Confocal Microscopy
Protocols for imaging can be found at: DOI: dx.doi.org/10.17504/protocols.io.q26g717m8gwz/v1.Fixed cells: Cells were plated onto 18 or 22 mm-glass coverslips (No. 1.5, 22x22 mm glass diameter, VWR 48366-227) the day before imaging.After aggregation, cells were fixed using 4% PFA followed by permeabilization with 0.5% Triton-X100.Cells were blocked in 3% BSA for 30 minutes, followed by incubation in primary antibodies for 1h at room temperature.Cells were washed 3 times with DPBS + 0.02% tween-20, followed by incubation in secondary (alexafluor conjugated secondary antibodies) for 1h at room temperature.Coverslips were then washed 3 times with DPBS + 0.02% tween-20 and mounted onto glass slides using mounting media (Vectashield H-1000) and sealed with nail polish.The cells were imaged using a Yokogawa CSU-W1 spinning disk confocal on a Nikon Ti motorized microscope equipped with a Nikon Plan Apo 100x/1.40N.A objective lens, and Hamamatsu ORCA-Fusion BT CMOS camera.For the analysis, the equal gamma, brightness, and contrast were applied for each image using FiJi software (FiJi ImageJ V.2.0.0 https://imagej.net/Fiji)(8).For quantification, at least 3 separate images.Live cell: Cells were plated onto glass bottom dishes the day before imaging.Before imaging, washout was started to begin aggregation in PhenolRed free dmem.The cells were imaged using a Yokogawa CSU-W1 spinning disk confocal on a Nikon Ti motorized microscope equipped with a Nikon Plan Apo 100x/1.40N.A objective lens, and Hamamatsu ORCA-Fusion BT CMOS camera, and a live cell chamber with temperature and carbon dioxide control.For the analysis, the equal gamma, brightness, and contrast were applied for each image using FiJi software (FiJi ImageJ V.2.0.0 https://imagej.net/Fiji)(8).Plots and statistics were performed using python packages matplotlib and python stats.Image analysis was performed using Fiji (8) and CellProfiler (9,10).Maximum intensity projections of z stacks were converted to tiff images in Fiji, then analyzed in CellProfiler by segmenting nuclei and aggregates and calculating parameters such as number, area and intensity of segmented objects.The output csv file was then loaded into a custom Python script for further analysis and data visualization using pandas, seaborn, numpy, and matplotlib packages.Statistics were performed using stats package.Corresponding p-values were included in figures and figure legends.

Mass spectrometry of purified proteasomes
Mass spectrometry protocol can be found at: DOI: dx.doi.org/10.17504/protocols.io.rm7vzjej4lx1/v1.Purified and eluted proteasomes were treated with 10 mM TCEP, 15 mM iodoacetamide, and 10 mM DTT to reduce and alkylate proteins.Methanol-chloroform precipitation was performed on proteins at ratios of methanol to chloroform to water of 4:1:3.Precipitated proteins were washed once with methanol, then spun at 16,000xg, and methanol was removed.Precipitated proteins were resuspended in 100 μL of 200 mM EPPS (pH = 8.0) and digested overnight with endoproteinase Lys-C at 23°C, followed by addition of Trypsin for 6 hours at 37°C, both with a mass ratio of 1:100 protease to total protein.Samples were labeled with 6 μL of TMTpro reagents at a stock concentration of 10 ng/μL and addition of 30 μL of acetonitrile for 1 hour at RT and then quenched with 0.5% vol/vol final concentration of hydroxylamine.Samples were then combined, acidified with 1% formic acid to achieve a pH ~1.5, then subjected to StageTip fractionation, then dried again and resuspended in 1% formic acid 5% acetonitrile for mass spectrometry analysis.Mass spectra were collected on an Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Fisher Scientific) using a gradient of acetonitrile from 3 to 13% for the first 83 minutes, then 13 to 28% for the final 5 minutes.An SPS-MS3 method was employed using Real Time Search as previously described (12,13).Mass spectra were searched using the Comet ( 14) using a reference human proteome from UniProt (The UniProt Consortium, 2019).A 2% FDR was applied to peptide-spectrum matches (PSM).Protein abundance was quantified from a globally protein-normalized sum of the PSM reporter ion intensity with a signal:noise ratio > 100.Normalization ratios for these proteasome-specific samples were calculated as the average of a and b proteasome subunits in each TMT channel and multiplied to each channel.
Cells were vitrified using a home-made manual, gravity-driven plunge-freezer.Before plunge-freezing, cells were treated with 10% glycerol in culture medium as a cryoprotectant for 1 -5 min (15).EM grids were blotted for 8 s with filter paper from the backside, immediately frozen in liquid ethane/propane and kept at liquid nitrogen temperature.
Grids were mounted on a transfer shuttle designed for a cryo-loading system ( 16) and loaded into a Quanta 3D FEG dual-beam cryo-FIB/SEM (FEI).Grids were sputtered with platinum (10 mA, 30 s) in a PP3000T loading system (Quorum) to reduce charging effects during electron imaging.Grids were then coated with organometallic platinum as a protective layer for ion beam milling.Grids were imaged with the scanning electron beam operated at 5 kV / 12 pA.Regions of interest (ROI) were thinned down at tilt angles of 18° -20° with the focused ion beam operated at 30 kV.The beam currents where set to 1 nA at approximately 1 µm distance from the ROI, 500 pA at 750 nm, 300 pA at 400 nm, 100 pA at 250 nm, 50 pA at 100 nm and 30 pA at 75 nm for fine polishing.If grids were designated for Volta phase plate (VPP) (17) imaging, grids were sputtered once more with platinum (10 mA, 5 s) after milling to increase conductivity of the lamellae.Samples vitrified at 10 minutes post S1 removal were imaged with a Titan Krios cryo-TEM (Thermo Fisher Scientific) operated at 300 kV, equipped with a FEG, post-column energy filter (Gatan) and VPP.Samples vitrified at 6 hours post S1 removal were imaged using a Polara F30 cryo-TEM (FEI) operated at 300 kV, equipped with a FEG and post-column energy filter (Gatan).In both cases, tiltseries were acquired on K2 Summit direct electron detectors (Gatan) in dose fractionation mode (0.08 frames per second) using SerialEM (18).Tilt-series were taken at -0.5 µm defocus for VPP imaging at 33,000x magnification (pixel size: 4.21 Å) on the Titan Krios and at -9 μm defocus at 22,500x magnification (pixel size: 5.22 Å) on the Polara.The tilt-series were acquired at an angular increment of 2° and typically ranged from -50° to 60°.The total dose was restricted to approximately 120 electrons / Å² per tilt-series.
Frames were aligned and combined using an in-house software based on previous work (19).Tilt-series were aligned using patch tracking from the IMOD software package (20) and reconstructed with weighted back projection.Tomograms were binned four times to a pixel size of 16.84 Å (Titan Krios) or 20.88 Å (Polara) to increase contrast.

Supplemental Figures
Fig. S2.UBE3C knockdown impacts agDD-GFP degradation.(A) mRNA levels for UBE3A, B, and C were measured by qPCR relative to control cells in biological triplicate.Error bars represent SEM.(B) Plot of GFP fluorescence at the indicated day post-infection with CRISPRi sgRNAs.(C) Plot of GFP fluorescence intensity versus cumulative cell fraction for the indicated cell lines without S1 washout or 4 h post-S1 washout.

Fig. S3 .
Fig. S3.UBE3C is required for acute agDD-GFP degradation.(A) Sequence analysis of two independent clones of UBE3C -/-HEK293T cells, showing in-frame deletions in the indicated alleles.The position of the gRNA is indicated by the yellow shading.(B) Immunoblot of wild-type and UBE3C -/-(clone 1) cell extracts using a-UBE3C antibodies, and a-PCNA as a loading control.(C) Histograms of agDD-GFP florescence levels in wild-type or UBE3C -/-HEK293T agDD-GFP low, intermediate, or high cells at the indicated times post-S1 washout measured by flow cytometry.(D) Histograms of agDD-GFP florescence levels in HeLa dCas9 agDD-GFP or agDD-GFP HIGH cells with or without cycloheximide (CHX) with or without S1 washout for 4 h measured by flow cytometry.

Fig. S4 .
Fig. S4.RPN13 ubiquitylation on K21 and K34 is not required for agDD-GFP degradation.(A)Sequence analysis of three independent clones of RPN13 K21R;K34R HEK293T cells, showing the desired edits in the indicated alleles.The position of the gRNA is indicated by the yellow shading.(B) Tandem mass tagging (TMT)-based proteomic analysis of proteasomes purified from wild-type and RPN13 K21R;K34R (Clone 3) cells using GST-UBL affinity resin.(C) Immunoblotting of input, flowthrough and GST-UBL elution of proteasomes from wild-type and RPN13 K21R;K34R HEK293T cells (Clones 1 and 2).Cells were either subjected to S1 washout for 1 h or treated with BTZ for 1 h prior to proteasome isolation.Blots were probed with a-RPN13, a-UBE3C, and a-PSMA3.(D) Histograms of agDD-GFP florescence levels in HEK293T agDD-GFP or agDD-GFP HIGH cells with or without RPN13 K21R;K34R editing to demonstrate similar levels of agDD-GFP fluorescence measured by flow cytometry.

Fig. S5 .
Fig. S5.NRF1 is activated in response to widespread protein misfolding.(A) HEK293T agDD-GFP cells were either subjected to S1 washout (WO) or treated BTZ (10 nM) for the indicated time periods.Cell extracts were either subjected to immunoblotting with the indicated antibodies or subjected to EndoH treatment to remove sugar modifications prior to immunoblotting.(B) HeLa dCas9 agDD-GFP LOW cells were treated as in panel A for the indicated times followed by immunoblotting of extracts with the indicated antibodies.a-Tubulin was used as a loading control.Markers are fluorescent and are therefore detected by LiCOR.

Fig. S6 .
Fig. S6.Genetic modulation of the NRF1 pathway alters aggregated protein degradation dynamics.(A) mRNA levels for NRF1 and DDI2 were measured by qPCR relative to control cells.Each point represents the value for individual replicates (n=3 biological replicates).(B) Immunoblots of HeLa dCas9 agDD-GFP HIGH cells with or without ectopic expression of NRF1 or a constitutively active NRF1 18ND mutant probed with a-NRF1 and a-HSP90.(C) Histograms of agDD-GFP florescence levels in HeLa dCas9 agDD-GFP HIGH cells with or without NRF1 or the NRF1 18ND mutant measured by flow cytometry.