TDP-43-stratified single-cell proteomic profiling of postmortem human spinal motor neurons reveals protein dynamics in amyotrophic lateral sclerosis

Abstract

Unbiased proteomics has been employed to interrogate central nervous system (CNS) tissues (brain, spinal cord) and fluid matrices (CSF, plasma) from amyotrophic lateral sclerosis (ALS) patients; yet, a limitation of conventional bulk tissue studies is that motor neuron (MN) proteome signals may be confounded by admixed non-MN proteins. Recent advances in trace sample proteomics have enabled quantitative protein abundance datasets from single human MNs (Cong et al., 2020b). In this study, we leveraged laser capture microdissection (LCM) and nanoPOTS (Zhu et al., 2018c) single-cell mass spectrometry (MS)-based proteomics to query changes in protein expression in single MNs from postmortem ALS and control donor spinal cord tissues, leading to the identification of 2515 proteins across MNs samples (>900 per single MN) and quantitative comparison of 1870 proteins between disease groups. Furthermore, we studied the impact of enriching/stratifying MN proteome samples based on the presence and extent of immunoreactive, cytoplasmic TDP-43 inclusions, allowing identification of 3368 proteins across MNs samples and profiling of 2238 proteins across TDP-43 strata. We found extensive overlap in differential protein abundance profiles between MNs with or without obvious TDP-43 cytoplasmic inclusions that together point to early and sustained dysregulation of oxidative phosphorylation, mRNA splicing and translation, and retromer-mediated vesicular transport in ALS. Our data are the first unbiased quantification of single MN protein abundance changes associated with TDP-43 proteinopathy and begin to demonstrate the utility of pathology-stratified trace sample proteomics for understanding single-cell protein abundance changes in human neurologic diseases.

Keywords: TDP-43 proteinopathy; amyotrophic lateral sclerosis (ALS); human tissue pathology; laser capture microdissection; motor neuron; nanoPOTS; retromer complex; single cell proteomics; stathmin 2 (STMN2).

Figure 1.. Ultrasensitive single cell proteomic mapping of ALS motor neurons

(A) Age and postmortem interval (PMI) of ALS and control (CTL) donor samples (n=3 per diagnosis) selected for single motor neuron (MN). Individual MNs (n=6 per donor) were identified in H&E-stained frozen tissue sections and excised by laser capture microdissection. Individual captured MNs were processed using the nanoPOTS workflow in parallel to “boost” samples (10 MN-equivalent) collected from each donor to facilitate match-between-runs. Samples were block randomized and analyzed by LC-MS/MS for protein identification and quantitation (B) High confidence (1% global FDR) Master Proteins (HCMPs) detected (ID) and quantified (Qt) across single MN samples isolated from ALS or CTL donor tissues (C) Pair-wise Pearson correlations between individual MN samples based on HCMP abundances and mean (µ) Pearson correlation score per comparison class (ALS:ALS, ALS:CTL, CTL:CTL) (***, p<0.001; ****, p<0.0001, 1-way ANOVA + Tukey’s MHC) (D) Volcano plot of differential protein abundance (log₂(ALS/CTL)) in ALS vs. CTL MNs based on normalized protein abundance values. Significantly dysregulated proteins (5% FDR (Benjamini-Hochberg-corrected p-value) ∩ |log₂(ALS/CTL)|≥1) are indicated in blue; green marker overlays indicate ribosomal proteins; red marker overlays indicate neurofilaments. (E) Protein-protein interaction network of significantly differentially abundant proteins (5% FDR ∩ |log₂(ALS/CTL)|≥1). Node color indicates fold change in normalized protein abundance (log₂(ALS/CTL); edges denote high confidence STRING-db (v11.5) protein interactions (interaction score cutoff = 0.7) (F) KEGG pathway annotations significantly enriched among differentially abundant proteins. FDR reflects Benjamini-Hochberg-corrected p-value.

Figure 2.. Absence of STMN2 protein detection in human ALS MNs parallels decreased abundance of Stathmin-2 (STMN2) RNA and absence of functional TDP-43 interaction partners..

(A) Rank-ordered median MS1 intensities for proteins detected in either CTL (blue) or ALS (orange) single MNs span five orders of magnitude (B) Uniquely identified and overlapping proteins detected in ALS or CTL single MN samples (C) Functional interaction network of proteins uniquely detected in CTL single MNs and their relationship to TDP-43 (node manually added). Magenta node borders indicate reported interaction (direct or indirect) with TDP-43; purple node border indicates a downstream splice target of TDP-43, STMN2; edges indicate physical protein-protein interactions annotated by the STRING database (v11) (D) Dual expression of STMN2 RNA (red, ISH) and TDP-43 protein (teal, IHC) in human ventral motor neurons from ALS and CTL donors (E) Quantitation of STMN2 RNA signal in ventral MNs from ALS (n=5) and CTL (n=10) and donors as determined by relative percentage of STMN2-immunoreactive (STMN2ir) cell area; violin plots show total percent STMN2ir area in MNs and non-MN ventral horn tissue for each donor; *, p<0.05 ; **, p<0.01 unpaired t-test with Welch’s correction (F) Detection of cryptic exon-containing and canonical STMN2 transcript expression in adjacent ALS MN cross-sections.

Figure 3.. Significant disruption of proteostasis, mitochondrial dysfunction, and induction of pro-apoptotic signaling are apparent prior to overt TDP-43 aggregation

(A) Schematic of single MN selection for laser capture microdissection based on dual detection of TDP-43 and ChAT by immunohistochemistry in immediately adjacent human tissue sections. MNs selected for LC-MS/MS analysis by nanoPOTS were stratified based on the presence of TDP-43 inclusions (0, CTL – normal appearing; 1, ALS – normal appearing; 2, ALS – mild inclusions; 3, ALS moderate inclusions; 4, ALS – severe inclusions) for subsequent pseudotemporal comparisons (B) Frequency of STMN2 protein detection across single MNs corresponding to different TDP-43 inclusion strata (C) Volcano plot comparisons of differential protein abundance between sequential TDP-43 inclusion strata based on normalized protein abundance values (log₂([Stage n+1]/[Stage n])). Significantly dysregulated proteins (5% FDR (Benjamini-Hochberg-corrected p-value) ∩ |log₂([Stage n+1]/[Stage n]) ≥1.5) are indicated in purple (i) NON (ALS) vs. CTL (CTL); (ii) MLD (ALS) vs. NON (ALS); (iii) MOD (ALS) vs. MLD (ALS); (iv) SEV (ALS) vs. MOD (ALS) (D) Overlap in proteins designated as significantly differentially expressed in (top) “Early ALS” (NON vs. CTL, |log₂([NON]/[CTL])| ≥1.5) following TDP-43-focused MN selection and in (bottom) ALS vs. CTL |log₂([ALS]/[CTL])| ≥1.0) following TDP-43-agnostic cell selection (E) Overlap in protein complexes and functional pathways represented among the subset of proteins designated as significantly differentially expressed in (i) “Early ALS” (NON vs. CTL, |log₂([NON]/[CTL])| ≥1.5) following TDP-43-focused MN selection or (i) in ALS vs. CTL |log₂([ALS]/[CTL])| ≥1.0) following TDP-43-agnostic cell selection. Node color indicates fold change in abundance between groups; edges indicate high confidence protein-protein interactions curated by the STRING database (v11.5).

Figure 4.. Neurofilament protein and retromer complex component abundances track inversely with increasing TDP-43 aggregation in postmortem human MNs

(A) Volcano plot of Spearman rank correlations between individual protein abundances per single MN and corresponding TDP-43 stratum; purple highlight, significantly (Bonferroni q < 0.05) correlated proteins; green dashed lines indicate neurofilament protein correlations (heavy (NEFH), medium (NEFM), light (NEFL)) (B) Distribution of protein abundances in individual MNs across TDP-43 strata for neurofilament proteins (NEFH, NEFM, NEFL) (C) (i) KEGG pathways over-represented and (ii) physical protein complexes represented among proteins comprising the top 5% of negative correlators with respect to TDP-43 strata; (D) TDP-43-inclusion-strata-associated protein abundance trajectories for retromer complex components SNX6, SNX12, VPS26A, VPS26B, VPS29, VPS35; GARP/EARP complex components VPS51 and VPS52; HOPS/CORVET complex proteins VPS18; and ESCRT-III complex proteins VPS4B and CHMP4B (E) Glial:Neuronal enrichment scores for proteins significantly positively correlated with increasing TDP-43 inclusion strata determined based on log₂-transformed LFQ ratios between individual glial and neuronal populations measured in (Sharma et al., 2015). (F) TDP-43 pathology-associated protein abundance trajectories for the intermediate filament protein SYNM and the presynaptic neurexin protein CNTNAP2. Box plots indicate median ± IQR; whiskers indicate full data range per stage.

Figure 5.. Modeling motor neuron degeneration in TDP-43 pseudotime: early and sustained retromer complex dysfunction promotes accumulation of TDP-43 accompanied by modulation of the neuronal-glial axis

Maintenance of proteostasis in (A) healthy MNs through clearance of cytoplasmic TDP-43 via (1) proteosomal degradation or (2) recycling through the endo-lysosomal pathway. Protein sorting is carried out in the early endosome via activity of cargo selective vacuolar sorting proteins (VPS) and nexins (SNX) for retrograde trafficking or lysosome-mediated degradation (B) Diminished abundance of VPS and SNX proteins in early ALS (NON) leads to inefficient protein sorting and trafficking, promoting cytoplasmic accumulation and aggregation of TDP-43 proteins\