Post-mortem human subjects and clinical data

For this data set, human post-mortem cortex specimens from DLB patients (n = 30), and age-matched controls (n = 29) were obtained from the Harvard Brain Tissue Resource Center (HBTRC) acquired through the NIH NeuroBioBank (U.S Department of Health and Human Services, National Institutes of Health). Control subjects had no clinical diagnosis of synucleinopathy, and DLB subjects fulfilled the pathological criteria for synucleinopathy disease. Neuropathological details for subjects are summarized in Table S1 and S2. Pathological and clinical information, if available, were de-identified.

Sarkosyl fractionation of post-mortem tissue

A sarkosyl fractionation protocol was carried out to enrich the insoluble pathological α-synuclein and tau in all patients. Briefly, approx. 150–160 mg of post-mortem frozen cortical brain tissue per patient was homogenized in 640 µl TBS lysis buffer (50 mM Tris–HCL, 150 mM NaCl, 0.5 mM MgSO4, 10 mM ethylene diamine tetraacetic acid (EDTA), 10 mM ethylene glycol tetraacetic acid (EGTA), 1 mM Dithiothreitol (DTT), 10 mM Nicotinamide, 2 µM Trichostatin A, phosphatase inhibitor cocktail (Sigma) and protease inhibitor cocktail (Roche)), using Precellys® 24 tissue homogenizer (5500 speed, 3 cycles of 20 s with pause of 30 s). Homogenized tissue was clarified by centrifugation at 14,000 rpm for 20 min at 4 °C. The resulting supernatant (termed S1) was transferred to a new tube. The pellets were solubilized by adding 640 µl 1X Salt Sucrose Buffer (0.8 M NaCl, 10% sucrose, 10 mM Tris–HCL, 1 mM EDTA, 1 mM EGTA, 10 mM Nicotinamide, 2 µM Trichostatin A, phosphatase inhibitor cocktail (Sigma) and protease inhibitor cocktail (Roche)), and homogenized as previously, sonicated (1 min, 10 s on 5 s off, 20% amplitude) using a probe, and centrifuged. The resulting supernatant is termed S2, and the pellets are discarded. To S1, 640 µl of 2X Salt Sucrose Buffer was added, and too both S1 and S2 1% final concentration of sarkosyl was added. Samples were incubated for 1.5 h on the thermomixer (300 rpm at room temperature), and ultracentrifuged for 1.5 h (50,000 rpm at 4 °C). The resulting supernatants (sarkosyl soluble) were transferred to new tubes and the pellets (sarkosyl insoluble) were stored at −80 °C for further processing. Sarkosyl insoluble pellets are reconstituted in pellet buffer (50 mM Tris–HCL, 5% Sodium dodecyl sulfate (SDS), 10 mM Nicotinamide, 2 µM Trichostatin A, 8 M Urea, phosphatase inhibitor cocktail (Sigma) and protease inhibitor cocktail (Roche), 8 M Urea) and sonicated for analysis.

Preparation of samples for mass spectrometry analysis

Following total protein concentration quantification with the bicinchoninic acid assay (Pierce™ BCA Protein Assay Kit, Thermo Scientific), 50 µg of total protein sample was digested using Single-pot, Solid-phase-enhanced sample preparation (SP3). Briefly, samples were diluted in 50 mM ammonium bicarbonate (ABC) and 8 M Urea buffer, 10 mM DTT was added to the samples and incubated for 30 min at 55 °C at 600 rpm. Alkylation was induced with the addition of 1% acrylamide and incubation at room temperature for 30 min. The alkylation reaction was quenched with the addition of 20 mM DTT and incubated at room temperature. Samples were loaded onto a combination of hydrophobic and hydrophilic magnetic beads (Cytiva LifeSciences) and binding was induced using 70% ethanol. Following incubation, the samples were washed with 2 washes of 80% ethanol, and 100% acetonitrile (ACN). Protein mixtures were digested with 12.5 ng/µl trypsin (sequencing grade modified trypsin, Promega, Madison, WI) for 16h at 37 °C. Digested peptides were removed from the magnetic beads and acidified using formic acid (FA). Peptides were desalted using C18 microspin columns (SEMSS18V, Nest Group, MA). Vacuum-dried peptides were reconstituted in MS sample buffer (5% FA, 5% ACN).

LC–MS/MS measurements

Soluble and insoluble fractions of all samples were analyzed using the timsTOF Pro2 mass spectrometer coupled with ultra-high-pressure nano-flow liquid chromatography nanoElute system (Bruker, Germany). Peptides were loaded on to a reverse phase 25 cm aurora series C18 analytical column (25 cm x 75 µm ID, 1.6 µm C18) fitted with captive spray insert (Ionopticks, Australia). Column temperature was maintained at 50 °C and mobile phase A (2% acetonitrile, and 0.1% formic acid in water) and mobile phase B (0.1% formic acid in acetonitrile) was used for the separation of peptides with 400nL/min constant flow using a linear gradient starting from 0 to 30% in 90 min, followed by an increase to 80% B within 10 min, followed by washing and re-equilibration for 20 min. The mass spectrometer was operated in Data-Dependent Acquisition (DDA) mode using Parallel Accumulation Serial Fragmentation (PASEF). Full mass spectra were acquired within a mass range of 100–1700 m/z and an ion mobility (1/K0) range from 0.60–1.60. A top 10 PASEF method was employed, where 10 PASEF MS/MS scans were triggered per acquisition cycle for the most abundant precursor ions, followed by dynamic exclusion to avoid repeated fragmentation of the same precursor.

Immunohistochemistry and analysis

Chromogenic immunohistochemistry was used to detect phosphorylated tau and alpha synuclein. Five-micron thick sections were cut from formalin-fixed paraffin-embedded sections from the same Dorsolateral Prefrontal cortex Brodmann area 9 (BA9) from 16 donors from the DLB cohort. The sections were mounted on slides to produce four slides per donor. Slides were deparaffinized and rehydrated with xylenes and graded alcohol. After antigen unmasking with a citrate buffer and incubating in Peroxidized 1 (Biocare Medical, cat#PX968), slides were blocked with Background Punisher (Biocare Medical, cat#BP974). Two sections per donor were incubated in anti-human PHF-Tau AT8 (Thermo Scientific, 1:400, mouse monoclonal, cat#MN1020), and two sections per donor were incubated in purified anti-a-Synuclein LB509 (Sigma Millipore, 1:4000, mouse monoclonal, cat#MABN824). Slides were then incubated in Mach 4 Mouse Probe HRP Secondary antibody, followed by incubation in Mach 4 HRP polymer (Biocare Medical, micro-polymer detection, cat# M4U534L). All slides were then visualized with DAB peroxidase substrate, counterstained with hematoxylin, cleared, and mounted.

Brightfield whole slide scans were performed using an Olympus VS120 Slide Scanner at 20 × magnification (Neurobiology Imaging Facility, Boston, MA). A semi-automated quantitative analysis of phosphorylated tau expression was performed on each image using HALO™ Image Analysis software (Indica Labs, Albuquerque, NM). Quantification of tau positive cells across all images of tau-stained slides was carried out using the HALO™ IHC Multiplex module. The IHC Multiplex module was tailored with customized parameters, a custom-trained nucleus segmentation classifier, and a custom-trained quality control tissue classifier. A total cell count was acquired by counting all nuclei positive for hematoxylin. Quantification of tau positive cells was shown as the number of tau positive cells over the total cell count (hematoxylin positive). Additionally, the images were analyzed using the HALO™ Area Quantification Module tailored with the same custom-trained nucleus segmentation classifier and custom-trained quality control tissue classifier. Quantification of tau expression was shown as the percentage tau positive tissue area over the total section area. Quantification of α-synuclein positive cells across all images of α-synuclein stained slides was also performed using the HALO™ Area Quantification Module tailored with a custom-trained nucleus segmentation classifier and custom-trained quality control tissue classifier. Quantification of α-synuclein expression was shown as the percentage α-synuclein positive tissue area over the total section area.

Multiplex immunofluorescence and Imaging

Formalin-fixed paraffin-embedded (FFPE) cortical brain tissue sections from one DLBTau⁻ and one DLBTau⁺ patient were used for immunofluorescence analysis. Five-micron thick sections were cut and mounted on slides. Tissue sections were deparaffinized using xylene and rehydrated through graded ethanol into distilled water. Antigen retrieval was performed by heating the slides in citrate buffer (pH 6.0) at 95 °C for 20 min, followed by cooling to room temperature. To prevent non-specific binding, slides were blocked with 5% normal goat serum in PBS for 1 h at room temperature.

The sections were incubated with the following primary antibodies, each diluted at 1:100: recombinant Alexa Fluor® 488 anti-α-synuclein aggregate (Abcam, cat# ab216124, clone MJFR-14–6-4–2), Alexa Fluor® 647 beta-amyloid (Cell Signaling Technology, cat# D3D2N), and Alexa Fluor® 750 phospho-tau (T217) (Cell Signaling Technology, cat# E9Y4S). DAPI (Cell Signaling Technology) was used as a nuclear stain. All sections were incubated with the primary antibodies overnight at 4 °C in a humidified chamber.

Following incubation, slides were washed with PBS and mounted with an anti-fade mounting medium. Fluorescence imaging was performed using the Advanced Solutions BioAssemblyBot 200 CELL DIVE Automation workstation and the CELL DIVE multiplex imaging system (Leica Microsystems). Images were acquired at 20 × magnification using the CELL DIVE Mx-workflow software, which applied automatic image corrections including autofluorescence removal, distortion correction, blank glass subtraction, and flat-field correction. Fluorescent signals were detected at AF-488 nm for α-synuclein, AF-647 nm for beta-amyloid, and AF-750 nm for phospho-tau.

Image analysis and visualization were conducted using AIVIA software (version 14.1, Leica Microsystems). Brightness and contrast were adjusted, and images were captured at full slide size, as well as at 100 μm and 50 μm resolutions, for figure preparation. Co-localization analysis of the different fluorescent markers was performed to evaluate the spatial distribution of α-synuclein aggregates and phospho-tau.

Software, statistical analysis and data visualization for proteomic data

DDA timsTOF data files were analyzed using Fragpipe (version 17.1) with MSFragger (version 3.4) for proteome analyses, and MASCOT software (version 2.6.1) for PTM identification. Briefly, raw data was analyzed in MSfragger, and peptide list searched against the Homo sapiens Uniprot protein sequence database (January 2021). The following parameters were applied: a minimum of two unique peptides per protein for identification, stricttrypsin cut at lysine and arginine with up to two missed cleavages, mass tolerance set to 20 ppm. Oxidation of M, acetylation of N-terminal and K, phosphorylation of S,T,Y, and ubiquitination of K were chosen as variable modifications and propionylation of cysteine as fixed modification. False discovery rate (FDR) was set to 1% on peptide and protein levels with a minimum length of six amino acids. For all other search parameters, the default settings were used.

All statistical analyses were performed using Perseus (version 2.0.5.0), GraphPad Prism (version 10.1.12), Excel (version 2408) and Cytoscape (version 3.9.1) with the STRING and STRING Enrichment applications. For differential expression analysis in Perseus, protein groups were filtered to remove contaminants, reverse hits, and those identified only by site. Using the MSFragger output files, label-free protein quantification intensity values were Log2-transformed to normalize the wide range of protein abundance values and facilitate statistical analysis. Proteins were filtered for a minimum of 70% valid values in at least one treatment group, and missing values were not imputed. Differentially expressed proteins were identified using a t-test with a permutation-based false discovery rate (FDR) threshold of < 0.05. Volcano plots for figures were generated in GraphPad Prism.

Box-and-whisker plots, created in GraphPad Prism, were used to display the Log2 intensities of proteins of interest, showing maximum and minimum values per group, with statistical significance between cohorts assessed by Student’s t-test (p-value < 0.05). Log2-transformed tau protein intensities were used for clustering analysis, where a classical k-means clustering with two clusters was performed using Orange (version 3.36.2), initialized with the k-means + + method and followed by 50 re-runs and 1000 iterations. A 2D scatter plot with jittering was generated to reduce overplotting, based on cluster number and Log2 intensities. Excel was used to create histograms for peptide intensities in proteins of interest.

Pathway enrichment analysis was conducted using Cytoscape with the STRING Enrichment plugin, focusing on Gene Ontology (GO) Biological Pathways. Final figures were prepared using BioRender (app.biorender.com).

Western blot analysis

Pooled protein samples were prepared from all groups: controls (CT, n = 30), DLBTau⁻ (n = 21), and DLBTau⁺ (n = 9). For each group, 10 µg of pooled protein was mixed with Invitrogen™ NuPAGE™ LDS Sample Buffer (4X) and Invitrogen™ NuPAGE™ Sample Reducing Agent (10X). Samples were denatured at 70 °C for 10 min using a thermomixer. Each sample (20 µL) and Invitrogen™ SeeBlue™ Plus2 Pre-stained Protein Standard were loaded onto 10-well 4–20% Mini-PROTEAN TGX Precast gels (Bio-Rad). The gels were run for approximately 1 h at a constant voltage of 150 V.

Proteins were transferred onto PVDF membranes using the Bio-Rad Trans-Blot Turbo Transfer System for 7 min. The membranes were blocked and subsequently incubated overnight at 4 °C with the following primary antibodies: APP/β-Amyloid (NAB228, 1:500, mouse monoclonal, cat#2450, Cell Signalling Technology), Clusterin (D4B6P, 1:500, rabbit monoclonal, cat#42,143, Cell Signalling Technology), EGF Receptor (D1D4J XP®, 1:500, rabbit monoclonal, cat#54,359, Cell Signalling Technology), and Ubiquitin (P4D1, 1:500, mouse monoclonal, cat#3936, Cell Signalling Technology). Membranes were washed and incubated with Invitrogen™ IRDye 680 goat anti-rabbit and IRDye 800 donkey anti-mouse secondary antibodies.

Total protein was assessed using the LI-COR Revert™ 700 Total Protein Stain, and imaging was performed on a Bio-Rad ChemiDoc MP Imaging System. Total protein was visualized at IRDye 680, while the primary antibodies were detected at IRDye 680 (for rabbit) and IRDye 800 (for mouse).

Images were analyzed using ImageJ software (version 1.53c). The area under the total protein lanes and specific bands of interest were calculated. The intensity of the bands of interest was normalized to the total protein stain for each lane. Ratios of protein intensities were calculated by comparing the normalized intensities to that of the control group. For presentation, the images were adjusted for brightness and contrast and cropped for Supplementary Fig. 5. Bar graphs were created using GraphPad Prism (version 10.1.12) and depict normalized intensities and ratio intensities for all antibodies of interest, representing one replicate per group, with each group being a pooled sample.