ALS-associated TDP-43 aggregates drive innate and adaptive immune cell activation

Abstract

Amyotrophic lateral sclerosis (ALS) is the most common and fatal motor neuron disease. Approximately 90% of ALS patients exhibit pathology of the master RNA regulator, transactive response DNA binding protein (TDP-43). Despite the prevalence TDP-43 pathology in ALS motor neurons, recent findings suggest immune dysfunction is a determinant of disease progression in patients. Whether TDP-43 aggregates elicit immune responses remains underexplored. In this study, we demonstrate that TDP-43 aggregates are internalized by antigen-presenting cell populations, cause vesicle rupture, and drive innate and adaptive immune cell activation by way of antigen presentation. Using a multiplex imaging platform, we observed enrichment of activated microglia/macrophages in ALS white matter that correlated with phosphorylated TDP-43 accumulation, CD8 T cell infiltration, and major histocompatibility complex expression. Taken together, this study sheds light on a novel cellular response to TDP-43 aggregates through an immunological lens.

Keywords: Immunity; Neuroscience; Omics.

Graphical abstract

Figure 1

TDP-43 aggregates are internalized by human monocyte-derived macrophages, which promotes their activation (A) Immunoblot depicts the dose-dependent uptake of insoluble, hyper-phosphorylated (pTDP-43) and total GFP-tagged TDP-43a (tTDP-43) in primary hMDM. Data are represented as mean ± standard deviation from n = 3 independent experiments. (B) Laser-scanning confocal micrograph of early internalization of TDP-43a in primary hMDM counter-stained with phalloidin. Pseudo-colored phalloidin stain depicts relative actin intensity (scale bars, 10 μm). (C) Representative 3-dimensional renderings of complete and incomplete TDP-43a internalization with respect to hMDM plasma membrane. Green denotes TDP-43a, magenta denotes wheat-germ agglutinin (scale bars, 10 μm). (D) Quantification of complete TDP-43a internalization in the presence of phagocytosis inhibitor cytochalasin D, endocytosis inhibitor dynasore, and micropinocytosis inhibitor EIPA. Data points depict the frequency of cells with at least one completely internalized TDP-43a particle relative to the total number of cells in a randomized field of view. Datapoints are colored (magenta, green, and blue) to denote fields of view from n = 3 independent experiments from 3 unique human donors. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Ordinary one-way ANOVA with Tukey’s multiple comparison test. Data are represented as mean ± standard deviation. (E) Volcano plots depict differentially expressed intracellular proteins (proteome, left) and extracellular secreted proteins (secretome, right) from primary hMDM cultures stimulated with TDP-43a compared to vehicle control. (F) Network analyses depict integrated secretome and intracellular proteomes that converge on actin remodeling (left), autophagy (middle), and phagocytosis pathways (right). Data represent n = 3 independent experiments; FDR <0.05; Student’s t test.

Figure 2

TDP-43 aggregates compromise intracellular vesicle integrity Airyscan and confocal micrographs (630X) illustrate TDP-43a co-localization with early endosome marker Rab5a (scale bars, 10 μm and 1 μm for insets) (A), acidified vesicle marker LysoTracker (scale-bar 15-μm and 5-μm for insets) (B), and ruptured autophagosome markers LC3B and galectin-3 (scale bar, 1 μm) (C). Representative images were acquired from n = 3 independent experiments. (D) Transmission electron micrographs depict whole-cell hMDM (top left; scale bars, 10 μm), hMDM vesicles following vehicle treatment (upper right; scale bars, 1 μm), and vesicles following TDP-43a treatment for 1-h (lower left; scale bars, 0.5 μm) and 16-h (lower right; scale bars, 1 μm). Representative vesicles are shaded yellow. Representative images were acquired from n = 2 independent experiments.

Figure 3

TDP-43 aggregates induce gene expression changes in primary human monocyte-derived macrophages (A) An MA plot depicts differential analysis of RNA-seq data of cells treated with either TDP-43a or vehicle for 12 h. Differential genes are depicted in blue or red. (B) A bar plot depicts the number of differential genes detected when comparing TDP-43a, _oAβ_1-42, or LPS to cells treated with PBS for the equivalent amount of time. Red represents upregulated genes and blue represents down-regulated genes. (C) Gene Ontology enrichment analysis depicts functional gene clusters that were differentially upregulated in response to TDP-43a treatment for 12 h. (D) Venn diagram depicting unique and/or overlapping genes between TDP-43a, _oAβ_1-42, or LPS treatment. (E) Heat maps depict the magnitude of expression change of each of the 34 conserved genes between TDP-43a, _oAβ_1-42, or LPS treatments. Data represent n = 2 independent experiments.

Figure 4

Global immunophenotypic changes occur in response to TDP-43 aggregates (A) Schematic of mass cytometry-based TDP-43a internalization assay, Aggre-Gate. Purified TDP-43a was partially reduced with DTT and coupled to tellurium maleimide (¹³⁰TeMal) via thiol-reactive chemistry. Labeled TDP-43a were added to cell cultures and internalization was determined by Boolean gating and/or spanning-tree progression of density-normalized events (SPADE) following mass cytometry analysis. (B) Concatenated dot-plots depict kinetics of TDP-43a internalization in classical monocytes from n = 3 independent experiments from a single genotype. (C) Linear regression of the percentage of TDP-43a positive classical monocytes with increasing incubation time. Gating scheme for identification of total monocytes is illustrated in Figure S4B. (D) SPADE analysis depicting TDP-43a accumulation amongst major immune cell-types. (E) Concatenated dot-plots of monocytes, dendritic cells, B cells, natural killer (NK) cells, and CD8 T cells positive for TDP-43a. (F) Quantification of the percent of cells positive for TDP-43a following a 24-h stimulation of bulk PBMC from n = 3 independent experiments from 3 unique donors (magenta, blue, and green datapoints). Analyzed by two-way ANOVA with Sidak’s multiple comparison’s test. Gating scheme for identification of immune cell subsets is illustrated in Figures S4B and S4C. (G–J) Quantification of relative CD127 expression levels as a readout of T cell effector activity following TDP-43a stimulation from n = 3 independent experiments from 3 unique donors. Arcsin(h) transformed ratios of CD127 intensity across CD4 effector (G), CD4 central (H), CD8 effector (I), and CD8 central (J) memory T cells. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Parametric, paired Student’s t test. See also Figures S1–S4.

Figure 5

TDP-43 aggregates stimulate antigen presentation and activation of naive T cells (A) Representative confocal micrographs (200X) of Raji-Jurkat synapses in the presence of isotype control (vehicle), TDP-43a purification eluates, or PHA. Pseudo-colored actin intensity images illustrate intercellular contacts characterized by actin polarization (scale bars, 100 μm and 20 μm insets). (B) Quantification of (A) by one-way ANOVA with Tukey’s multiple comparisons test. Graph depicts the frequency of cells forming contacts relative to the total number of cells in a randomized field of view. Data bars represent percentages of single and multiple cell contacts per condition, relative to the number of contacts in an image. Data points represent n = 3 independent experiments. Data are represented as mean ± standard deviation. (C) Representative confocal micrographs (left; 200X; scale bars, 50 μm) of primary hMDM immune synapses in the presence of isotype control (vehicle) or TDP-43a purification eluates. Arrowheads indicate hMDM (ActiStain-555; magenta) and arrows indicate regions of intercellular contact between hMDM and syngeneic PBMCs (CellTracker green). High magnification Airyscan micrograph (right; 630X; scale bars ,10 μm) illustrates actin-polarized contact between hMDM and PBMC (asterisk). Images are representative of n = 2 independent experiments. (D) Schematic of live-cell calcium imaging co-culture assay to monitor naive T cell activation following TDP-43a stimulation of hMDM. (E) Time-series widefield fluorescence micrographs of intercellular contact between hMDM and naive CD8 T cell with intracellular calcium release. Dashed line denotes hMDM plasma membrane, and arrowheads denote site of intercellular contact (scale bars, 10 μm). (F) Quantification of T cell calcium signaling in both naive CD4 and CD8 T cell co-cultures with hMDM. Each data point represents single-cell calcium signaling events from a 40-min imaging period within a single field of view. One field of view per coverslip. Data points originated from n = 3 independent experiments from 3 unique donors for vehicle and TDP-43a stimulation, and a single donor for HIV stimulation. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Ordinary one-way ANOVA with Dunnett’s multiple comparisons test. (G and H) Quantification of CD4 (G) and CD8 (H) T cell activation through analysis of calcium signaling in the presence of neutralizing antibodies to MHC-II and MHC-I, respectively. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. One-way ANOVA with Tukey’s multiple comparisons test. n = 2 independent experiments from a single whole-blood donor. See also Figure S6 and Video S1.

Figure 6

Imaging mass cytometry of ALS white matter reveals distinct CD8 T cell infiltrates that correlate with TDP-43 pathology burden (A) Representative overlays of the white and gray matter of motor cortex from n = 3 independent ALS and control donors (scale bars, 50 μm and 20 μm inset). (B) Quantification of CD68⁺ microglia (left) and perivascular/parenchymal CD3⁺CD8⁺ T cell (right) frequencies in donor brain tissue analyzed by imaging mass cytometry. Single points represent averaged frequencies from 1 to 3, 1.0 mm² ablations per group. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. Two-way ANOVA with Tukey’s multiple comparisons test. (C) Overlays and corresponding single-channel images of activated myeloid cells/microglia expressing antigen presentation machinery (CD68 and HLA-DR) and interacting with CD3⁺CD8⁺ T cells in the perivascular space of ALS white matter. Nuclei counter-labeled blue. Vasculature marked with dashed line. Arrowheads denote phosphorylated TDP-43 (pTDP-43) pathology, arrows denote interactions between antigen-presenting cells, CD8 T cells, and pTDP-43. Images were despeckled and adjusted similarly between control and ALS for brightness/contrast using FIJI (scale bars, 50 μm). (D) Pearson correlation analyses of immune features in ALS white matter. Individual dots represent respective cell frequencies from eight individual 1.0 mm² control (blue) and ALS white matter (red) regions of interest from n = 3 donors per group. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. See also Figures S7–S9. Images were acquired from n = 3 independent imaging sessions.