Cell Culture
HEK293T and U2OS cells were obtained from ATCC to create all cell lines. Cells were maintained in Dulbecco’s DMEM with 10% fetal bovine serum, 1% penicillin-streptomycin, and 1% GlutaMax. v2L tau biosensor cells were used for all seeding assays. Details on these biosensors have been recently published (8). Cell lines were checked approximately biweekly for mycoplasma contamination (Venor-GEM Mycoplasma Detection kit).
Proteomics Screen
T225 flasks were coated with 10mL of 0.01mg/mL poly-D-lysine (PDL) for 3h in the incubator and rinsed with PBS before plating cells. 22 million cells were plated in 25ml/T225 flask and allowed to settle overnight. The following day, cells were treated with 50nM tau + Lipofectamine-2000 complexes (or 50nM α-synuclein + Lipofectamine as a negative control) which were incubated for 20min at RT prior to addition to cells. Cells were incubated with the fibrils for 5h. Thirty minutes before the 5h time point, cells were treated with BP (biotin phenol) at a final concentration of 500µM at 37ºC. At 5h, cells were treated with H2O2 at a final concentration of 1mM and the flasks were agitated at RT for 1min. The biotinylation reaction was quenched with the quenching buffer followed by three additional rinses. Quenching buffer was also used to scrape the cells to collect the cell pellets. This buffer was prepared as previously described in the APEX2 labeling protocol (67).
On-Bead Trypsin Digestion
Protein concentrations were normalized across all the samples (~1mg of starting lysate) based on the Pierce 660 assay readings and protein abundances from shotgun proteomics analysis of trypsin digests of these samples by the UT Southwestern Proteomics Core Facility. Lysates (1 mg) were incubated with 250µL of magnetic streptavidin beads at 4ºC for overnight incubation ~16h. The next day, the beads were concentrated by magnet, and washed 2x with 200µL of 50mM Tris-HCl pH 7.5 followed by 2x with 2M urea + 50mM Tris-HCl pH 7.5. The beads were then incubated with 80µL 2M urea + 100μL 0.5ug/µl trypsin + 20µL 10mM DTT to achieve a final urea concentration of 1mM and a ratio of 1: 20 for trypsin: lysate, for 1h at 25ºC with shaking at 1000rpm in a thermomixer. The beads were washed 2x with 60µl of 2M urea + 50mM Tris-HCl pH 7.5 and the two washes were combined with the supernatant. The eluate was reduced with DTT at a net concentration of 4mM by incubating for 30min with shaking at 1000rpm, 25ºC. The samples were alkylated with 10mM iodoacetamide for 45min at 25ºC with shaking at 1000rpm.
50mM Tris-HCl pH 7.5 was then added to achieve a final urea concentration of 0.73M. Samples were incubated overnight (~15h) at 37ºC with shaking at 1000rpm to allow complete trypsin digestion. The samples were removed from the thermomixer and spun down. Trypsin was quenched by acidifying the samples to pH <3 with the addition of formic acid at a final concentration of 1%.
TMT Mass Spectrometry
5μL of 10% trifluoroacetic acid (TFA) was added to each sample, and solid-phase extraction was performed on each sample using an Oasis HLB 96-well uElution plate (Waters). Eluates were dried and reconstituted in 50μL of 100 mM triethylammonium bicarbonate (TEAB). 10μL of each sample was labeled with 4μL of a different TMT10plex reagent (Thermo Scientific, label TMT10-131 not used). Samples were quenched with 1μL of hydroxylamine, mixed, and dried in a SpeedVac. Samples were reconstituted in 2% acetonitrile, 0.1% formic acid to a concentration of 0.5 ug/μL,
2μL of each TMT sample were injected onto an Orbitrap Fusion Lumos mass spectrometer coupled to an Ultimate 3000 RSLC-Nano liquid chromatography system. Samples were injected onto a 75μm i.d., 75-cm long EasySpray column (Thermo) and eluted with a gradient from 0-28% buffer B over 180 min. Buffer A contained 2% (v/v) acetonitrile (ACN) and 0.1% formic acid in water, and buffer B contained 80% (v/v) ACN, 10% (v/v) trifluoroethanol, and 0.1% formic acid in water. The mass spectrometer was operated in positive ion mode with a source voltage of 1.8 kV and an ion transfer tube temperature of 275°C. MS scans were acquired at 120,000 resolution in the Orbitrap and top speed mode was used for SPS-MS3 analysis with a cycle time of 2.5 s. MS2 was performed with CID with a collision energy of 35%. The top 10 fragments were selected for MS3 fragmentation using HCD, with a collision energy of 55%. Dynamic exclusion was set for 25 s after an ion was selected for fragmentation.
Proteomics Data Analysis
Raw MS data files were analyzed using Proteome Discoverer v2.4 (Thermo), with peptide identification performed using Sequest HT searching against the human protein database from UniProt. Fragment and precursor tolerances of 10 ppm and 0.6 Da were specified, and three missed cleavages were allowed. Carbamidomethylation of Cys and TMT labeling of N-termini and Lys sidechains were set as a fixed modification, with oxidation of Met set as a variable modification. The false-discovery rate (FDR) cutoff was 1% for all peptides.
For every biological replicate, absolute abundance of each protein was first normalized to the total protein abundance of a particular lysate sample to account for any differences in total protein concentrations across samples before comparison. These values were used to calculate the relative enrichment of proteins specific to tau seeding: Protein Abundance Ratio= (sAPEX2 P301S + seeds)/(sAPEX2 P301S – seeds).
Normalized protein abundance ratios for sAPEX2 P301S and sAPEX2 alone (negative control) treated with and without tau fibrils were compared using unpaired t- test on three independent biological replicates; two-stage step-up (Benjamini, Krieger, and Yekutieli), FDR 1.00%). Positive relative abundance values on the graph indicate enrichment in the aggregation proteome. Statistical significance was determined based on q value.
Generation of TDP-43 Biosensor Cell Line
FM5 CMV promoter plasmids containing the mClover3 and mRuby3 fluorophores were digested with Esp3I restriction enzyme. WT TDP-43 (a.a. 274-414) was cloned into the cut plasmids using Gibson assembly to generate constructs expressing TDP-43 tagged to the respective fluorophores. Constructs were sequence verified using UTSW’s Sanger sequencing core and used for making lentivirus.
HEK293T cells were transduced with the TDP-43 lentivirus and were sorted into monoclonal cells for highest expression of the two fluorophores. Monoclonal cell line that responded with maximum seeding to different FTLD TDP-43 lysates was chosen for the seeding assay reported here. Cells expressing the individual fluorophores were also sorted as fluorophore compensation controls for detecting FRET on the flow cytometer.
Biosensor Seeding Assay
All biosensor assays were performed with “naked seeding” (no cation-based transfection reagent). Cell lines were plated at a density of 15,000 cells/well of a 96 well plate and allowed to settle overnight. Cells were treated with an appropriate concentration of recombinant tau fibrils for 48h, and then harvested for flow cytometry. Fibril preps were sonicated in a water bath sonicator for 1min at 65Amp prior to cell treatment. Recombinant tau fibrils were prepared as previously described (30).
For seeding with brain lysates and for the a-synuclein biosensor seeding assay, 8,000 cells/well (10,000 cells/well for TDP-43 biosensors) were plated in 96 well plates and seeding was monitored for 72h. In the case of brain homogenates, biosensors were treated with 25µg of tauopathy lysates and 10µg of FTLD lysate. These were sonicated for 1min at 65 Amp in a water bath sonicator. For seeding with a-synuclein, fibrils were sonicated for 5min total, 1min on /1min off at 65 Amp, and used at a final concentration of 400nM.
All seeding results are reported as % FRET+ cells. The FRET data were plotted by subtracting the background signal (no exogenous tau added) which was negligible for all conditions (no FRET recorded in the absence of tau seeds), unless otherwise specified.
Brain Homogenization
Brain tissue from clinically and neuropathologically characterized cases of AD, CBD, and FTLD TDP-43 were obtained from UTSW and Washington University in St. Louis. All human tissues used in these experiments were derived from autopsy subjects. Since deceased subjects are not considered human subjects for research purposes, these studies were exempt from human subjects research regulations and did not require IRB approval. Brain samples were weighed and added to 1X TBS supplemented with cOmplete Ultra (Roche) protease inhibitor to prepare a 10% w/v solution. The brains were homogenized using a probe homogenizer to obtain a slurry that was sonicated for 15min total, 1min on/ 30sec off. The sonicated samples were centrifuged at 4°C for 15min at 21,300 x g. Protein concentration of the supernatant was measured using Pierce 660 assay and was subsequently used for naked seeding.
Uptake Assay
The uptake assay was performed as previously described (30). v2L cells were plated overnight at a density of 8,000 cells/well of a 96 well plate. Cells were treated with 25nM of AF-647 labeled tau fibrils or AF-647 dye alone as a negative control. After 4h of incubation with the fibrils, cells were harvested with 0.25% trypsin for flow cytometry.
The labeled fibrils used in this assay were obtained by incubating recombinant tau fibrils (8μM, 200μL) with lyophilized AF-647 dye (25μg) for 1h at room temperature (RT) followed by quenching the reaction with 100mM glycine and subsequent dialysis in a 3500kDa dialysis cassette.
The median fluorescence intensity (MFI) values representing the amount of tau internalized were plotted after subtracting the background MFI of the dye alone signal for all conditions. These MFI values were then normalized relative to the appropriate control condition of the respective experiment (DMSO, NTG, or Scr controls).
Compound Treatments
96 well plates were coated with 0.01mg/mL PDL and incubated at 37°C for 3h followed by washout with PBS. v2L cells were plated at a density of 15,000 cells/well and allowed to settle overnight. Cells were treated with different compounds (ML-240, NMS-873, MG132, and LLOMe) for about an hour upon which 25nM of recombinant tau fibrils were introduced to the media. After four hours of incubation with the fibrils (five hours with compounds), the media was replaced with fresh media and uptake or seeding was monitored for 4h and 48h, respectively.
Flow Cytometry
To harvest cells for flow cytometry, media was removed, and cells were treated with 0.05% trypsin (0.25% trypsin for uptake assay) for 5min at 37°C (0.25% trypsin, 15min at 37°C in case of PDL coated plates). Trypsin was quenched with cold media and cells were resuspended a few times before transferring the suspension to 96-well round-bottom plates which were centrifuged at 1000 rpm for 5min. Supernatant was removed and the cell pellets were resuspended in 2% paraformaldehyde (PFA) and allowed to fix for 10min at RT. Cells were spun down again, PFA was removed, and cells were resuspended in PBS and stored at 4°C until ready to be run on a flow cytometer (BD LSRFortessa) for quantifying the FRET signal.
Cloning
FM5 vector with UBC promoter was used to clone all the APEX constructs. sAPEX fragments (AP and EX) were PCR amplified from the constructs provided by the laboratory of Alice Ting (Stanford). Amplified sAPEX fragments were appended on to the c-terminus of RD tau fragments via a linker using overlap PCR. Using Gibson assembly, the final gene fragments were cloned into FM5 UBC plasmid, which was cut with Esp3I. All Gibson reaction products were transformed into Stbl3 bacterial cells. Bacterial colonies were inoculated, DNA was purified using Qiagen miniprep kit, and the sequences were verified using Sanger sequencing at UTSW’s sequencing facility.
Lentivirus Production
Low passage HEK293T cells were plated at ~ 70% confluency in 6 well plates and allowed to settle overnight. A master mix was prepared using 400ng of plasmid of interest, 400ng of VSVG, and 1200ng of PSP plasmids required for virus packaging, along with 7.5mL of TransIT 293T and 120ul of OMEM per well of a 6 well plate. The master mix was allowed to incubate at RT for 30min upon which it was added to the cells drop-wise. The virus was harvested 48h later by collecting the media, spinning it for 5min at 1000rpm, and then freezing the aliquoted supernatant.
Differentiation and Culturing of Human iPSC-derived Cortical Neurons
We utilized the integrated, inducible, and isogenic Ngn2 iPSC line (i3N). It was previously shown that expression of the transcription factor neurogenin-2 (Ngn2) induces rapid differentiation of iPSCs into cortical glutamatergic neurons (68). This iPSC line harbors a doxycycline-inducible mouse Ngn2 transgene at an adeno-associated virus integration site 1 (AAVS1) safe-harbor locus, allowing for a simplified differentiation protocol (69,70). iPSCs were dissociated with Accutase (Sigma, A6964) and plated onto basement membrane extract-coated plates (R&D, 3434-001-02). Ngn2 expression was induced with 2 μM doxycycline hyclate (Sigma, D9891) in KSR media alone with 10 μM SB431442 (R&D, 1614), 2 μM XAV939 (Stemgent, 04–0046) and 100 nM LDN-193189 (Stemgent, 04-0074) (doxycycline hyclate is maintained in all medias going forward). On day 2, cells were fed with a 1:1 ratio of KSR media + SB/XAV/LDN and N2-supplmented neural induction media with 2 μg/ml puromycin (Life Technologies, A1113803). On day 3, cells were fed with N2-supplemented neural induction media. On day 4, cells were dissociated with Accutase and plated onto PDL-(Sigma, P1149) and laminin-(Life Technologies, 23 017-015) coated tissue culture plates. Cells were subsequently maintained with neurobasal media (Life Technologies, 21103049) supplemented with NeuroCult SM1 (StemCell Technologies, 05711) and 10 ng/ml brain-derived neurotrophic factor (R&D, 248-BD-005/CF) until collected.
Compound Treatment and Seeding Assay in Neurons
Differentiated human neurons were treated with tau RD P301S lentivirus (tau-clover at an MOI of 3; tau-ruby at an MOI of 2). 48h later, the transduced cells were simultaneously incubated with 15nM FL, WT recombinant tau fibrils and the VCP inhibitors, ML-240 and NMS-873 at 100nM for 48h. For acute exposure with tau fibrils and the compounds, the neuronal biosensors were treated with 1μM ML-240 and 100nM NMS-873 for 4h followed by media replacement. FRET signal, used as a measure of tau aggregation was monitored over 48h using ImageXpress Confocal HT.ai High-Content Imaging System (Molecular Devices). 4 images per well, with 5 wells per condition of a 96 well plate, were collected for the clover, ruby, and FRET channels. Single color controls were used to correct for fluorophore bleed through using an automated algorithm on MATLAB. Single color controls were used to measure the bleedthrough coefficient of each channel by fitting the correlation of pixel intensities in the single color channel and the “empty” channel with a linear fit. Images from the acceptor/donor channels of real data were then multiplied by the resulting slopes (i.e. bleedthrough coefficients) to approximate the amount of bleedthrough. Finally, FRET channel images were corrected by subtracting these bleedthrough images from the FRET images. The corrected images were analyzed to calculate the FRET area using an ImageJ macro and the quantified FRET area was plotted for different conditions.
CRISPR/Cas9 Screen for VCP Cofactors
CRISPR constructs for the cofactors were outsourced to Twist Biosciences for synthesis. Constructs not synthesized by the company were cloned in the lab using standard ligation reactions. Four guides per gene were chosen from the Brunello library deposited online and ordered as duplex DNA from IDT. LentiCRISPRv2 (Addgene #52961) was cut using Esp3I, and T4 ligase was used for all ligation reactions of the guides into the plasmid. Stbl3 bacteria were transformed with the ligated products, selected colonies were inoculated, mini-prepped using Qiagen kit, and the purified DNA was sequence-verified. Pooled lentivirus was prepared with four constructs per gene and v2L cells were transduced with the virus at the desired MOI. After 24h, cells were expanded in puromycin media (2mg/mL) for selection of the KO population. Selected populations were eventually used for seeding and uptake assays. The KO was verified using western blot.
siRNA Knockdown
siRNAs were ordered from Origene. 300,000 v2L cells were plated in 6 well plates and allowed to settle overnight. The next day, cells were treated with 100nM of each siRNA, with a total of three siRNAs per gene using RNAiMax Lipofectamine (Thermo) as a transfection vehicle at 7.5µl/well. After 48h of transfection, the cells were plated in 96 well plates for seeding and uptake assays. Cells were also used for western blot to verify the knockdown.
Western Blot
Cell pellets were lysed in RIPA buffer and allowed to sit on ice for 5min followed by a 15,000g spin for 10min at 4°C. The supernatants were used to determine the protein concentrations using Pierce 660 assay. 15mg of total protein was treated with SDS buffer and heated at 95°C for 10min. Samples were loaded onto 4-12% bis-tris gels and the proteins were transferred onto nitrocellulose membranes using the Biorad turbo transfer machine. All incubations for subsequent steps were done in TBS + 0.05% Tween-20 (TBST). The membranes were first incubated in blocking buffer (5% milk powder + TBST) for 1h at RT, followed by primary antibodies in the blocking buffer at 4°C with overnight shaking. After the primary antibody incubation, the membranes were washed 3x with TBST, 10min each. Then, appropriate HRP-conjugated secondary antibodies in blocking buffer were added to the membranes for a 1.5h incubation at RT. Membranes were again washed 3x in TBST followed by a single wash in TBS alone before reading the HRP signal using the Thermo ECL kit.
Statistical Analysis
Proteomics data was analyzed using unpaired t- test, two-stage step-up (Benjamini, Krieger, and Yekutieli), FDR 1.00%. One-Way ANOVA (Šídák method) with a 95% confidence interval was used for all the other statistical analyses unless otherwise stated. Paired t-test was used to analyze the neuron data.
The P values are described as follows: ns = not significant/ P>0.05, * = P ≤ 0.05, ** = P ≤ 0.01, *** = P ≤ 0.001, **** = P ≤0.0001.
Graphics
Biorender.com was used to create the graphics presented here and has granted publication and licensing rights (agreement number: RX26PLQVWQ).