Human post-mortem tissues

Fresh-frozen human post-mortem brain tissues from Parkinson disease patients (n = 19) and age-matched controls (n = 23) were obtained from the MRC London Neurodegenerative Diseases Brain Bank (6 Parkinson disease, 10 control; King’s College, London, UK), the Parkinson UK Brain Bank (4 Parkinson disease; Imperial College, London, UK), and the Sydney Brain Bank (9 Parkinson disease, 13 control; Sydney, Australia). Diagnoses of Parkinson disease were determined clinically by the donors' physicians. Pathological identification of Lewy pathology and dopamine neuron loss in the SNc by brain bank neuropathologists confirmed clinical findings. All Parkinson disease cases were free of other neurological or neuropathological conditions. Genotyping confirmed the absence of SOD1 mutations, as described below. Age-matched control cases were free of any clinically diagnosed neurological disorders and neuropathological abnormalities. Ethics approval was obtained from the University of Sydney Human Research Ethics Committee (approval number 2019/309). Demographic and clinical information for all cases are detailed in Supplementary Table 1. Diagnostic groups were matched for sex, age, and post-mortem interval (Supplementary Table 2). Fresh-frozen tissue samples were randomly numbered by a secondary investigator (V.C. or S.G.) prior to experimentation to blind primary investigators (B.G.T., A.H.A) to case diagnoses.

SOD1 genotyping

SOD1 genotyping was performed in ten Parkinson disease and ten age-matched control cases as previously described [58]. DNA was extracted from fresh-frozen human brain tissue from the anterior cingulate or occipital cortex (OCx) using the DNeasy DNA extraction kit (Qiagen, Hilden, Germany, #69506), according to the manufacturer’s instructions. All five exons of SOD1, and at least 10 bp of flanking sequence were sequenced using PCR amplification and Sanger sequencing in duplicate. All results were independently analyzed by two team members. Remaining Parkinson disease and control cases were genotyped previously [86].

Animals

All methods conformed to the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes [64], with protocols approved by the Animal Ethics Committee at the University of Melbourne (Ethics ID: 1814531.3) and ratified by the University of Sydney Animal Ethics Committees. Hemizygous male mice expressing seven-transgene copies of the human SOD1WT mouse strain (B6.Cg-Tg(SOD1)2Gur/J) were crossbred with female Ctr1+/- (Slc31a1tm2.1Djt/J) mice to produce the novel hSOD1WT/Ctr1+/- (SOCK) mouse strain. Both hSOD1WT and Ctr1+/- mouse lines were sourced from The Jackson Laboratory (Bar Harbour, Maine, USA), and housed in filter top enclosures (12/12 h light/dark cycle, 22 °C, 45% humidity) with cardboard boxes and tubes for environmental enrichment. Enclosures contained Breeder’s Choice Cat Litter with paper tissues provided for bedding, and ad libitum access to standard chow pellets and water.

Tail snips were obtained from all mice prior to the age of weaning (3 weeks) for commercial genotyping of both Slc31a1 and hSOD1WT genes. Mice were bred and aged to 1.5, 3, 6, and 12 months-of-age (group sizes displayed in Supplementary Table 3), before being anesthetized with a lethal dose of xylazine (16 mg/kg body weight) and ketamine (120 mg/kg body weight). Mice were then perfused through the left ventricle with ice-cold 0.1 M phosphate buffer saline (PBS; pH 7.4, 4 °C) supplemented with phosphatase inhibitors (Phosphatase Inhibitor Cocktail 2; Sigma), protease inhibitors (Complete EDTA-free Tablets; Roche), and heparin (20 U/mL) and brain, lumbar spinal cord, and liver tissues harvested. Brains were bisected sagittally into two hemispheres, before regions of interest (midbrain, striatum, and cortex) were micro-dissected from the left hemisphere and stored with liver tissues at −80 °C for downstream biochemical analyses. The right brain hemisphere and entire lumbar spinal cord were postfixed in 4% paraformaldehyde overnight in preparation for immunohistochemical analyses.

Fresh human and mouse tissue preparation for biochemical measurements

Fresh tissues were homogenized in 10 × homogenization buffer volume (µL) per mg tissue weight (20 mM Tris-base pH 7.4 containing EDTA-free protease inhibitor (Sigma-Aldrich) and phosphatase inhibitor (Roche) using a Kontes pestle pellet mechanical tissue grinder (Sigma-Aldrich). Following homogenization, extracts were incubated at 4 °C for 30 min before protein concentration was determined using a bicinchoninic acid assay according to the manufacturer’s instructions (Thermo-Fisher Scientific).

Measurement of enzymatically active SOD1 metallation

The metal content of dimeric SOD1 was quantified using size-exclusion chromatography coupled with native isoelectric focusing and synchrotron radiation X-ray fluorescence analyses, according to our published method [74, 87]. Briefly, brain tissue homogenates were subjected to size-exclusion chromatography [Superdex 75 Increase 10/300 GL column connected to an Äkta Pure chromatography system, cooled to 4 °C, 100 mM ammonium acetate solution (pH 7.4) as the eluant] to purify enzymatically active SOD1 dimers from fresh-frozen post-mortem tissues. SOD1-containing fractions were then applied along immobilized pH gradient strips (ReadyStrip IPG Strips, pH 4–7, Bio-Rad) in duplicate and native isoelectric focusing was performed to further purify SOD1 according to its isoelectric point (pI). SOD1 was identified within the first duplicate IPG strips using nitroblue tetrazolium activity staining and metal quantification was performed on the second duplicate IPG strip at the pI of active SOD1. Copper and zinc quantification by synchrotron X-ray fluorescence (SXRF) was performed at the microprobe of the Hard X-ray Micro/Nano-Probe beamline P06 at the synchrotron PETRA III (DESY) in Hamburg (Germany) [75] using a Vortex silicon-drift X-ray detector with Cube preamplifier. SXRF analyses were carried out with an X-ray beam of 14 keV photon energy, 0.5 × 0.5 mm2 beam size, and 1.1 × 1011 ph/s and in-house code enabled online data analysis. Measurements were performed in triplicate or quintuplicate for each sample, which were averaged and used to generate mean Cu/Zn ratios and sd. Copper measurements represent total cuprous and cupric ions, while zinc measurements only represent Zn2+ ions as this is the only oxidation state of this metal. Limit of detection was calculated from blank measurements and only results above 3 LOD were retained for final analysis.

SOD1 immunoprecipitation and preparation for proteomic mass spectrometry

SOD1 protein was immunoprecipitated from human and mouse brain tissue homogenates as previously described [87]. Briefly, 10 mg of Dynabeads M-280 Tosylactivated (Invitrogen, Carlsbad, CA, USA) were conjugated to 100 µg polyclonal SOD1 antibody (Enzo Life Sciences, Farmingdale, NY, USA; Supplementary Table 4) at 40 mg beads/mL in coupling buffer (0.1 M boric acid, pH 9.5; 1.2 M ammonium sulfate) overnight at 37 °C. Dynabeads were blocked with 0.5% bovine serum albumin (BSA) in PBS (pH 7.4), washed with 0.1% BSA in PBS (pH 7.4), and incubated with tissue homogenates (200 µg total protein) diluted in PBS (pH 7.4) to 40 mg beads/mL overnight at 4 °C. Following PBS washes, immunoprecipitated proteins were eluted from Dynabeads using successive 10 min incubations with 0.1 M glycine (pH 3), eluants neutralized using an equivalent volume of ammonium bicarbonate (pH 8), and extracts dried under pressure using a vacuum concentrator. Dried SOD1 immunoprecipitates were resuspended in 50 mM ammonium bicarbonate (pH 8) containing 6 M urea, reduced with dithiothreitol (DTT; 10 mM final) for 30 min at 56 °C, alkylated with iodoacetamide (IAA; 20 mM final) for 30 min at room temperature in the dark, and finally quenched with a further 10 mM DTT for 30 min at room temperature. Samples were diluted fivefold using 50 mM ammonium bicarbonate (pH 8) to decrease the concentration of urea to 1.2 M and acetonitrile added (10% final), before in-solution digestion performed overnight at room temperature using 0.2 µg sequencing-grade modified trypsin (Promega, Madison, WI, USA). Samples were then acidified using trifluoroacetic acid, desalted using Pierce C18 Tips (ThermoFisher Scientific, USA) according to the manufacturer’s instructions, and dried under pressure using a vacuum concentrator. Samples were resuspended in loading buffer (0.1% formic acid, 3% ACN) and transferred to reverse-phase high-performance liquid chromatography system (HPLC) vials immediately prior to mass spectrometry analyses.

Mass spectrometry data acquisition and analysis

Label-free Fourier Transform Mass Spectrometry was employed to analyze immunoprecipitated protein extracts at Sydney Mass Spectrometry (Sydney, New South Wales, Australia). Analyses were performed using an UltiMate 3000 RSLCnano system (ThermoFisher Scientific, USA) coupled online via a Nanospray Ion Source (ThermoFisher Scientific, USA) to a Q Exactive HF-X Hybrid Quadrupole-Orbitrap Mass Spectrometer (ThermoFisher Scientific, USA). Peptide digests were loaded onto an in-house packed ReproSil-Pur 120 C18-AQ analytical column (75 µm id × 40 cm, 1.9 µm particle size; Dr Maisch, Ammerbuch, Germany) regulated to 60 °C using a PRSO-V2 Sonation column oven (Sonation, Baden-Wuerttemberg, Germany). A binary gradient of solvent A (0.1% formic acid in MilliQ water) to solvent B (0.1% formic acid in 80% ACN diluted with MilliQ water) was used for peptide elution at a separation flow rate of 300–450 nL/min over 90 min. The mass spectrometer operated in positive ion mode at a 2.4 kV needle voltage. Data were acquired using Xcalibur software (Version 4.4.16.14, ThermoFisher Scientific, USA) in a data-independent (DIA) mode. The MS was operated in a data-independent fashion with 20 dynamic DIA segments covering the mass range from 350 to 1650 m/z. The resolution for the MS1 scan was set to 60 k with a max injection time of 50 ms and an AGC target of 3e6. The DIA scans were acquired in the Orbitrap with a resolution of 30 k after fragmentation in the HCD cell (max injection time: auto; AGC target: 3e6; fixed first mass: 300 m/z; loop count: 1; MSX count: 1; isolation window: 26–589 m/z).

Raw DIA data were processed using Spectronaut software’s directDIA workflow (Version 19, Biognosys, Zurich, Switzerland), whereby raw data files were first used to generate project-specific spectral libraries in silico using Spectronaut Pulsar. Separate libraries were generated for each experimental question; the first used raw data files from Parkinson disease and control SNc extracts to address SOD1 PTM changes in Parkinson disease, while the second utilized raw data from SOCK and hSOD1WT mice, as well as human control SNc. Library generation was performed using BGS factory settings, with the Human Uniprot fasta file employed as a protein database for searches. Two missed cleavages were allowed and the false discovery rate (FDR) controlled at 1% for both PSM and protein group levels. Peptide identification and label-free quantification in each individual sample were then performed using default BGS factory settings, with spectra screened against Uniprot entry P00441 (SODC-HUMAN), corresponding to human SOD1. Cross-run normalization was disabled, PTM localization was enabled (summative PTM consolidation strategy and a probability cutoff of 0.75), and analyses were performed with a log2 ratio candidate filter of 0.58, a confidence (Q) candidate filter of 0.05 and multiple comparisons testing correction enabled. Carbamidomethyl (C) was included as a fixed modification, while modifications of interest were included individually in variable modifications alongside acetylation (N-term) and oxidation (M). PTMs of interest for comparison between Parkinson disease and control extracts were analyzed in separate analysis batches, and included; acetylation (K; + 42.01), acetylglucosamine (NST; 203.08) carboxymethyllysine (K; 58.01), deamidation (N/Q; + 0.98), kynurenine (W; + 3.99), glycation (K/R; + 108.02 with neutral loss of three water molecules [41]), glycosylation (NST; + 162.05), nitration (W; + 44.99), oxidation (H/W; + 15.99), phosphorylation (S/T; + 79.97), succinylation (K; + 100.02), and ubiquitination (GlyGly, K; 114.04). A maximum of five modifications per peptide was allowed, with two missed trypsin cleavages. SOD1 protein was not identified in negative control immunoprecipitates prepared using Dynabeads that were not conjugated to our capture antibody, suggesting negligible false discovery of SOD1 protein in tissue extracts [87]. No differences in the relative levels of PTMs of interest were identified between immunoprecipitated and non-immunoprecipitated commercial SOD1 protein in a previous study [87], implying our immunoprecipitation protocol did not significantly alter PTMs of interest.

SOD1 and copper chaperone for SOD1 protein quantification

Immunoblotting for SOD1 and copper chaperone for SOD1 (CCS) proteins was performed by probing membranes first for CCS, and then simultaneously probing membranes for SOD1 and GAPDH following stripping of CCS antibodies. Protein samples (0.5 μg for hSOD1WT SOD1 overexpressing and SOCK and 2.5 μg for Ctr1+/- and WT mouse) were incubated in loading buffer [17.5% sodium dodecyl sulfate, 50% glycerol, 400 mM dithiothreitol (DTT), 0.3 M Tris-base (pH 6.8), and 0.25% bromophenol blue] for 45 min at 56 °C to reduce and denature sample protein content, before being loaded onto 4–12% Bis–Tris Criterion pre-cast gels (Bio-Rad, Hercules, CA) and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis in a Mini-PROTEAN Tetra Cell system at 180 V for 40 min at 4 °C (Bio-Rad). A region-specific loading control, prepared by combining equal amounts of protein from 18 samples of each region (3-month-old mice, n = 9 hSOD1WT-expressing, n = 9 WT for hSOD1WT), was loaded onto each gel. Separated proteins were transferred to Immobilon-PSQ PVDF (Millipore, Billerica, MA) membranes overnight at 9 V at 4 °C, before being air-dried overnight and then blocked in 5% skim milk (Bio-Rad, Hercules, CA) in phosphate buffer saline containing 0.1%Tween®20 (PBST) (Sigma-Aldrich, St. Louis, MO) for 1 h at room temperature. Membranes were then incubated with primary antibody against CCS (1:2000, Rabbit anti-CCS, raised against peptides corresponding to amino acid residues 252–270 of the human CCS sequence, gifted by Dr Isil Keshin, Umeå University, Sweden) diluted in 1% skim milk in PBST overnight at 4 °C, followed by incubation with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (1:5000, Bio-Rad, Hercules, CA) for 2 h at room temperature. Protein signals were obtained using an ECL western blotting detection system (Bio-Rad, Hercules, CA) as per the manufacturer’s instructions, and developed using the iBright imaging system (Invitrogen). Membranes were then incubated in a stripping buffer (25 mM Glycine, 1.5% SDS, pH 2.0) to remove primary and secondary antibodies, re-blocked with PBST and 5% skim milk and incubated with primary antibody against rabbit anti-SOD1 (1:2000, Enzo, NY, USA) and rabbit anti-GAPDH (1:10,000 Merck, G9545) overnight at 4 °C. Membranes were incubated with HRP goat anti-rabbit IgG (1:5000 Bio-Rad, Hercules, CA) and protein signals were obtained using an ECL western blotting detection system (Bio-Rad, Hercules, CA). Antibody details are presented in Supplementary Table 4. CCS, SOD1, and GAPDH signal intensities were quantified by densitometry using iBright analysis software v5.2.2 (Invitrogen). CCS and SOD1 values were first normalized to the corresponding GAPDH values and then normalized to the loading control value in each gel. GAPDH values were unchanged between genotypes at each age in all regions, validating the choice of this protein as a housekeeping gene (Supplementary Fig. 1).

Quantification of tissue metal levels

Metal levels in soluble and insoluble tissue extracts were quantified using inductively coupled plasma-mass spectrometry (ICP-MS), according to previously published methods by our group [26]. Twenty-to-thirty microliters of tissue homogenate were dried down and digested overnight using concentrated nitric acid (50 µL, 70%, Suprapur grade, Merk Millipore) at room temperature. Samples were then digested for a further 30 min at 70 °C, incubated with concentrated hydrogen peroxide (30%, VWH International, PA, USA) for 60 min at 70 °C, and then diluted to 2 mL with 1% nitric acid (1:10 v/v; Suprapur grade, Merk Millipore) prior to analysis. Total metal levels in each sample were measured in triplicate using a Perkin Elmer Nexion 300X Inductively Coupled Plasma Mass Spectrometer. Buffer controls containing 1% nitric acid were incorporated every 20 samples. Helium (4 mL/min) was used as a collision gas for the removal of polyatomic interferences. Measured mass-to-charge (m/z) ratios were 63 (Cu) and 66 (Zn). External calibration was performed using S24 multi-element standards (High Purity Standards, USA) diluted in 1% HNO3, while rhodium (Rh; m/z = 45) was used as reference element via online introduction with a Teflon T-piece. Measurements were background corrected to metal levels in buffer controls, adjusted for dilution factors and standardized against original wet tissue weights. Samples below the instrument’s limits of detection were excluded from analyses.

SOD activity measurement

SOD1 antioxidant activity was quantified in tissue extracts using a commercial SOD Assay Kit (Cat. #19160, Sigma-Aldrich, USA) according to the manufacturer’s instructions [85]. Briefly, samples containing 2 µg protein were diluted serially between 10- and 1000-fold and the assay signal measured in triplicate. A bovine SOD standard was used to generate a standard curve relating SOD activity to assay signal, which was applied to sample dilution curves to obtain SOD activity measurements. Total SOD activity in each sample was normalized to SOD1 protein levels measured using immunoblotting, which yielded a measure of SOD activity per unit of SOD1 protein in each sample.

Fixed mouse tissue preparation for immunostaining

Following overnight fixation in 4% paraformaldehyde, mouse brain and lumbar spinal cord tissues were incubated in 30% (w/v) sucrose solution in a 50 mL sample collection tube for 24–48 h until they sunk to the bottom of their container. They were then embedded in Tissue-Tek® optimal cutting temperature medium (Sakura Finetek, Nagano, Japan; #4583) and stored at −80 °C. Fifty micrometer thick serial brain tissue sections were then cut from Bregma 2.53 mm to 3.04 mm using an Epredia™ CrytoStar™ NX50 cryostat (ThermoFisher Scientific) to ensure the entire rostro-caudal SNc was collected, while the lumbar spinal cord was cut into serial 30 µm sections. These were divided into three free-floating section series and stored in 0.1 M PBS containing 0.02% sodium azide at 4 °C until staining.

Nigral dopamine neuron stereology

Free-floating brain tissue sections for dopamine neuron stereology were incubated for 30 min in citrate buffer (pH 6.0; Fronine) at 95 °C and were then cooled to room temperature before proceeding with 3,3-diaminobenzidine (DAB) staining. Endogenous peroxidase activity was quenched by pre-treating sections in a 3% H2O2 solution in 50% ethanol solution for 30 min at room temperature. Sections were washed in PBS-T, then blocked for 1 h at room temperature using 0.5% BSA and 1% casein in PBS-T. Sections were incubated in an anti-tyrosine hydroxylase (TH) primary antibody (Merck Millipore, USA, #AB152, 1:5000) at 4 °C. Primary antibodies were detected using a biotinylated goat anti-rabbit IgG secondary antibody (Vector Laboratories, USA; #BA-1000, 1:200), followed by a tertiary antibody (Vector Laboratories, USA; #PK-7200, 1:100) and visualized using DAB (Sigma-Aldrich, USA). Sections were washed and mounted onto Superfrost® Plus silica glass microscope slides (ThermoFisher Scientific), and counterstained with cresyl violet prior to coverslipping. All images were captured using an Olympus VS120 Virtual Slide Microscope (Olympus, Japan) at 20 × magnification with 4 µm-depth intervals. Quantitative stereological analysis of TH-positive neurons in the SNc was completed using VS-DESKTOP (Olympus, version 2.91). Coronal plates of the anatomical distribution of the SNc and its lateral, dorsomedial, and ventral subregions were established according to published literature to ensure that regions of interest were consistently defined in all imaged animals [23]. TH-positive neuron staining containing Nissl bodies in SNc subregions were counted. Neuronal density was calculated as the number of SNc dopamine neurons divided by the total volume of the SNc. Ten percent of images from each cohort were independently counted by two researchers to measure the interrater reliability of counts, demonstrating excellent interrater reliability of these measurements (Cronbach’s α = 0.953, n = 17).

Immunofluorescent staining

Free-floating brain and spinal cord tissue sections for immunofluorescent staining were brought to room temperature and pre-treated with 0.3% Triton X-100 made in PBS (PBS-Tx) for 45 min to increase tissue permeability, before antigen retrieval performed using citrate buffer at 95 °C (Vector Laboratories, USA). After blocking with 10% normal horse serum made in 0.1% PBS-Tx, sections were incubated with appropriate primary antibodies (Supplementary Table 4) at 37 °C for 1.5 h and then at 4 °C for two nights. Following PBS-Tx washes, appropriate non-spectrally overlapping fluorescent secondary antibodies (Supplementary Table 4) were then incubated under the same conditions as the primary antibodies, before cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) for 20 min at room temperature. Sections were mounted and coverslipped using SlowFade™ Glass Antifade Mountant (ThermoFisher Scientific, USA). Image acquisition was performed using a Nikon C2 + Confocal Microscope System and Nikon NIS-elements software (Nikon, v5.20.02). Large-scale format images were first acquired at 10 × magnification to select three regions of interest from three sections spanning the rostro-caudal SNc. A 60 × magnification oil immersion lens was used to image those regions of interest at 1024 × 1024 pixel resolution with 0.3 µm Z-step through the entire tissue thickness. A negative control was run by substituting the primary antibody with normal serum where no positive immunolabelling was detected, thus validating the absence of non-specific fluorescent signals (Supplementary Fig. 2).

Quantification of SOD1 in nigral dopamine neuron soma

Confocal microscope images of brain tissue sections stained using SOD1 UßB, TH, and DAPI were converted from ND2 format to 3DTIFF using Fiji (NIH, v1.53J). Images were loaded on Avizo™ 3D (2021.2, ThermoFisher Scientific) and a subvolume of the image was extracted for downstream analysis. Segmentation of TH+ neuron cell bodies was performed by manually tracing a region of interest on every second slice and interpolating the remaining slices based on the traced regions of interest. All slices were visually checked to ensure that the region of interest was captured correctly. Slices interpolated inaccurately were corrected using the brush tool or repeated with an increased number of manually segmented regions of interest. A threshold intensity for SOD1 was assigned to analyzed images to identify and quantify SOD1 aggregates. Variable illumination and image intensity between images required an individual threshold for each image. Threshold intensities for SOD1 were individually assigned for each captured region of interest. Separate images for the TH+ neuron cell bodies and SOD1 protein threshold were exported in a 3DTIFF format. A Python script was subsequently applied to all analyzed images to extract 3D measurements of SOD1 aggregate volume and its proportion inside and outside TH+ neuron cell bodies. This script is publicly available on GitHub (owner: Richard Harwood, repository: image_analysis_SOD1).

Lumbar spinal cord motor neuron stereology

Mouse lumbar spinal cord sections (50 sections/mouse on average) for motor neuron stereology were immunostained for the motor neuron markers choline acetyltransferase (ChAT) and Islet-1 (ISL-1) and imaged using an Olympus VS200 virtual slide scanner (Olympus, Japan) at 40 × magnification using the maximum intensity projection function. Spinal cord sections were confirmed as being from the lumbar region using anatomical features identified in an atlas of the mouse spinal cord [91]. Once images were obtained, left and right gray matter horns that were ventral to the spinal canal were then converted to.tiff files separately using QuPath (v0.5.1) software. Ventral motor neurons that were considered positive for both ISl-1 and ChAT were then segmented for counting by manual and automated methods using FIJI imaging software. Briefly, automated segmentation was performed via an FIJI software script that imported the multiplex images into FIJI and split them into single channel images. Nuclei that were immunopositive for ISL-1 (488 nm channel) were then pre-processed using enhanced contrast (saturation: 0.2%, normalize histogram) and background subtraction (rolling ball radius: 100 pixels), before image thresholding applied using the yen method, which was chosen as the most appropriate of 17 methods trialed. Fill holes and erode were applied to binary masks followed by segmentation using analyze particle analysis (size: 0.5–infinity, circularity: 0.1–1). To confirm double staining of motor neurons, segmented ISL-1 nuclei were then overlayed on ChAT immunopositive staining (647 nm channel, min and max intensity: 50–700) and each ISL-1 nuclei particle expanded proportional to its respective Feret diameter. Average 488 nm and 647 nm intensities and area measures were recorded in overlap zones, while background 647 nm intensity was measured as a 500 × 500 pixel box overlayed over areas of background on a subset of the data (30% images). Positive motor neurons were counted when they had both ISL-1 nuclei segmentation and proximal ChAT staining above background. Counts were then normalized to a 1 mm length of spinal cord by dividing the counts by the combined length of spinal cord comprised by the sections counted (e.g., 50 sections of 30 µm thickness = 1.5 mm length of spinal cord). Manual segmentation of ventral spinal cord motor neurons was performed on 10% of the dataset to confirm reliability of the automated data produced, whereby motor neurons that were positive for both ISL-1 and ChAT staining were counted in the extracted images. Manual counting demonstrated excellent reliability for the automated segmentation (Cronbach’s α = 0.985, n = 75).

Quantification of striatal dopamine levels and turnover

Fresh-frozen striatal mouse tissue was homogenized by pulse sonication (30% duty cycle, output 2, 2 × 30 s intervals) in a solution of 150 mM phosphoric acid and 500 µM diethylenetriaminepentaacetic acid. Total protein concentration of the tissue homogenate was assessed using the bicinchoninic acid assay according to the manufacturer’s protocol (Thermo-Fisher Scientific). Homogenized tissue was centrifuged at 16,000 g for 45 min at 4 °C and the remaining supernatant was spun in 3 kDa cut-off Amicon® centrifugal filter tubes (Merck Millipore) at 14,000 g for 90 min at 4 °C. Filtrates were collected and stored at −80 °C for downstream processing.

Quantification of dopamine, 3,4-dihydroxyphenylacetic acid (DOPAC), and homovanillic acid (HVA) levels was conducted using HPLC, consisted of a pump module (Shimadzu Prominence LC-20AD, Shimadzu Corporation) coupled to a reversed-phase Gemini C18 110 Å column (5 μm pore size, 150 × 4.6 mm; Phenomenex) and an electrochemical detector (Antec Leyden) with a glassy carbon-working electrode maintained at + 0.82 V against a Ag/AgCl reference electrode. Both detector and column were maintained at 40 °C. Twenty microlitres of prepared samples were injected via a Prominence autosampler (Shimadzu Prominence SIL-20A, Shimadzu Corporation, Kyoto, Japan). The mobile phase consisted of 0.01 M sodium phosphate monobasic [84% (v/v)) and methanol (16% (v/v)] containing 0.1 mM ethylenediaminetetraacetic acid, 0.65 mM 1-octanesulfonic acid and 0.5 mM triethylamine. Mobile phase was adjusted to pH 3.8 using 1 M hydrochloric acid and filtered through 0.22 μm Whatman® filter circles and then pumped at a flow rate of 1 mL/min. Calibration curves prepared from starting concentrations of analytical standards of dopamine (1 μM), DOPAC (0.5 μM), and HVA (3 μM) were run at the beginning of each day to quantify any variability in the HPLC system. The Shimadzu integrated workstation LabSolutions software (Version 5.57; Shimadzu) was used to calculate the area under the curve for each peak of interest, and data were subsequently normalized to the total protein concentration of each sample. Dopamine turnover was calculated as the ratio of HVA to dopamine. Neurotransmitter concentrations are expressed as ng/mg of protein (mean ± SEM).

Quantification of α-synuclein phosphorylation

Total and phosphorylated serine 129 (pS129) α-synuclein were quantified in mouse midbrain tissue extracts using AlphaLISA SureFire Ultra assays for these targets (Revvity, Total α-synuclein; ALSU-TASYN-B, Phospho-α-synuclein (Ser129); ALSU-PASYN-B) according to the manufacturer’s instructions. We first identified optimal dilution factors required to bring total and pS129 α-synuclein levels in extracts to within the linear dynamic range of the total and pS129 assay (25-fold for pS129 assay, 125-fold for total assay). Extracts were then diluted in assay lysis buffer to their optimal dilution factors and applied to total and pS129 α-synuclein assays. Total α-synuclein concentration in samples was calculated using a standard curve generated by measuring serial dilutions of purified wild-type α-synuclein (RP-009; Proteos, Kalamazoo, MI, USA) using the total α-synuclein kit. The concentration of pS129 α-synuclein in these samples was calculated using a standard curve generated by measuring serial dilutions of purified pS129 α-synuclein (RP-004; Proteos) using the pS129 α-synuclein kit. Concentrations in diluted samples were then multiplied by their dilution factors and expressed as ng/mg total protein to account for differences in protein concentration in the original extracts. The proportion of α-synuclein phosphorylated at S129 was calculated by dividing ng pS129 α-synuclein/mg total protein by the ng total α-synuclein/mg total protein, expressed as a percentage.

Animal weight and locomotor function

At each age of interest, animal weight was recorded prior to the assessment of motor function using the rotarod test according to published methods [69]. Mice were initially habituated on the stationary rod, and then trained to walk on the rotating rod for 5 days prior to blinded data collection. Mice were tested on the rotarod twice a week during the data collection phase of the study. The rotation speed of the dowel was accelerated from 4 to 40 rpm over 180 s and the length of time spent on the dowel (latency to fall) was recorded up to a maximum of 180 s.

Statistical analyses

Statistical analyses were performed using RStudio (Build 524) and SPSS 28.0 (IBM Corp, NY, USA). Outliers were defined by SPSS as ‘extreme values’ ≥ 3 × the interquartile range (or ± 2 standard deviations) and excluded from the analysis. Parametric tests or descriptive statistics with parametric assumptions (standard one- and two-way ANOVA, Pearson’s r and t test) were used for variables meeting the associated assumptions, with data normality assessed using the Shapiro–Wilk test, Levene’s test, and Brown–Forsythe tests. One- and two-analysis of variance (ANOVA) were paired with Dunnett’s multiple comparisons post hoc test to assess pairwise comparisons between experimental groups for a given variable. Non-parametric tests or statistics (Kruskal–Wallis test and Mann–Whitney U test) were used for variables where the observed data did not fit the assumptions of parametric tests, with Dunn’s multiple comparisons post hoc tests to assess pairwise comparisons between select diagnostic groups for a given variable. Where large differences between groups existed (e.g., SOD1 protein expression and SOD1 proteinopathy), data were log transformed prior to application of statistical tests. Cronbach’s alpha was used to measure interrater reliability between researchers performing quantitative stereology. Significance level was defined as p < 0.05 for all statistical tests. Graphs were generated using GraphPad Prism 9.4.0 (Graph-Pad, CA, USA).