2024.01.07.574541v2.full-html.html

bioRxiv preprint

Abstract

Amyotrophic lateral sclerosis is the most common 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 pathology elicits disease-modifying

immune responses in ALS remains underexplored. In this study, we demonstrate that TDP-43 pathology is

internalized by antigen presenting cells, causes vesicle rupture, and leads to innate and adaptive immune cell

activation. Using a multiplex imaging platform, we observed interactions between innate and adaptive immune

cells near TDP-43 pathological lesions in ALS brain. We used a mass cytometry-based whole-blood

stimulation assay to provide evidence that ALS patient peripheral immune cells exhibit responses to TDP-43

aggregates. Taken together, this study provides a novel link between TDP-43 pathology and ALS immune

dysfunction, and further highlights the translational and diagnostic implications of monitoring and manipulating

the ALS immune response.

Introduction

ALS is the most common, fatal motor neuron disease worldwide. The disease is primarily characterized by the

progressive degeneration of motor neurons in the motor cortex and spinal cord (1). The average post-

diagnosis survival interval is approximately 2-5 years (2), however a subset of patients can survive >10 years

post-diagnosis (3). Recent evidence suggests systemic immune perturbations modify ALS presentation and

progression (4–6). Indeed, disparities in the levels of nearly every immune population including innate and

adaptive immune cells have been documented in ALS (7–11). Additionally, loss-of-function mutations in

several ALS-associated risk genes, such as C9ORF72 (chromosome 9 open reading frame 72) and TBK1

(TANK-binding kinase1) were shown to promote autoimmune-like inflammatory responses (12, 13). However,

C9ORF72 and TBK1 mutations only constitute 5-7% and 1%, respectively, of all ALS cases (14, 15). In 2006,

Neumann and colleagues identified pathological inclusions of TDP-43 as an underlying feature to motor neuron

degeneration in the vast majority of nearly all ALS cases, including > 90% of those with sporadic ALS (16, 17).

TDP-43 regulates the expression or stability of approximately 6,000 transcripts across diverse

biological processes in diverse cell types such as neuroglia and peripheral immune cells (18–22). In disease,

TDP-43 mislocalizes to the cytoplasm and forms post-translationally modified aggregates (23–25). We recently

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

demonstrated that TDP-43 aggregates generated in human cells and then isolated by biochemical methods

contained nearly 2,000 sequestered proteins. These include classically defined damage associated molecular

patterns (DAMPs) such as heat shock proteins, heterogeneous nuclear ribonucleoproteins, and nuclear pore

proteins. We also demonstrated that monocyte-derived macrophages and microglia internalize TDP-43

aggregates (26). Other studies have shown that TDP-43 dysfunction leads to dysregulation of STING antiviral

immune responses in neurons (27). Here, we aimed to address the following question: do pathological TDP-43

species serve as antigens that stimulate disease-modifying immune cells that are relevant to ALS?

Results

TDP-43 aggregates are phagocytosed, trafficked to autophagolysosomes, and promote acute

activation of primary monocyte-derived macrophages

Our prior study showed that immunopurified, human TDP-43 aggregates (referred to as TDP-43a) are readily

internalized by both primary murine microglia and primary human monocyte-derived macrophages

(hMDM)(26). Additionally, we noted transcriptional similarity between human microglia and our hMDM cultures

(28). For this reason, we used hMDM for studying interactions with TDP-43a. We first assessed the dynamics

of TDP-43a internalization using biochemical methods. By immunoblot, we observed significant dose-

dependent aggregate internalization by hMDM following stimulation with TDP-43a at 0.025% (**p = 0.0051)

and 0.1% (****p < 0.0001) (Figure 1A). Laser-scanning confocal microscopy confirmed internalization via

active, actin-dependent transport (Figure 1B). These results suggest that actin-dependent phagocytosis is the

dominant mode of TDP-43a internalization. However, several internalization pathways could employ actin-

polymerization. Thus, to conclusively assess phagocytosis as the dominant route by which TDP-43a is

internalized, we used a panel of pharmacological inhibitors to target different internalization routes. These

included Dynasore to inhibit dynamin-mediated endocytosis and 5-(N-ethyl-N-isopropyl)-Amiloride (EIPA) to

inhibit micropinocytosis. Following inhibition and stimulation with TDP-43a, we counterstained the hMDM cell

membrane with wheat-germ agglutinin (WGA) and quantified complete internalization using laser-scanning

confocal microscopy (Figure 1C-D). With respect to vehicle control (27.13% ± 3.528%), a partial inhibitory

effect was observed with Dynasore (15.75% ± 2.84%; *p = 0.018) and EIPA (15.13% ± 2.88%; *p = 0.011);

however, cytochalasin D treatment consistently inhibited internalization with 100% efficacy (****p < 0.0001).

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

This led us to assess the functional response of hMDM to TDP-43a internalization using a systems-based

proteomics approach.

We performed combinatorial secretome and proteomics analysis on hMDM cultures stimulated with

TDP-43a for 16 hours and identified several features of macrophage activation in the global intracellular

proteome and secretome. These included increased secretion of macrophage migration inhibitory factor (MIF)

(Log

fold = 3.84; FDR < 0.05) and IL-1 converting enzyme or caspase 1 (Log

fold = 3.67; FDR < 0.05), and

reduced secretion of immunosuppressive immune checkpoint molecule CD276 (Log

fold = -2.14; FDR < 0.05),

among others (Supplementary Tables S1A-B). By merging these data into a network analysis, we identified

upregulated signaling pathways including actin remodeling, phagocytosis, and autophagy (Figure 1E). These

findings confirm that extracellular TDP-43a augments innate immune signaling and scavenging pathways,

supporting the ‘DAMP-like’ properties of TDP-43a.

TDP-43a compromises autophagy and promotes vesicle rupture

Our proteomics analysis highlighted autophagy as a significant response following TDP-43a

stimulation. Given the intimate link between TDP-43 dysfunction and autophagic dysregulation (29), and the

fact that autophagic dysregulation can drive maladaptive immune responses (30), we next interrogated TDP-

43a trafficking and autophagic integrity in hMDM. We first performed immunofluorescence and Airyscan

confocal imaging where we observed co-localization between TDP-43a and markers of early endosomes

(Rab5) (Figure 2A). We then validated lysosomal targeting by performing live-cell imaging using the vesicle

acidification probe, LysoTracker-Red. Indeed, we observed co-localization between TDP-43a and

LysoTracker-positive vesicles (Figure 2B). Furthermore, Airyscan confocal imaging of hMDM stimulated with

TDP-43a for 16 hours revealed the presence of LC3-positive autophagosomes co-localizing with galectin-3 and

TDP-43a, suggesting TDP-43a burden promotes vesicle rupture and downstream vesicle turnover (Figure

2C)(31, 32). We further characterized the ultrastructural integrity of vesicles by transmission electron

microscopy (TEM) of hMDM stimulated with TDP-43a for 1 hour and 16 hours to simulate acute and chronic

exposures, respectively (Figure 2D). By 16 hours we observed greater incidences of irregular, enlarged

vesicles. We also identified features of vesicle degradation and turnover, indicated by increased presence of

multivesicular bodies (MVBs). These ultrastructural features were not present in unstimulated or 1-hour

stimulated hMDM. To validate whether autophagic processing is abnormal in the presence of pathological

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

TDP-43a, hMDM were stimulated with TDP-43a and lysates were collected at 0, 0.5, 4, 8, 16, and 24 hours.

Lysates were then probed for markers of autophagic flux, LC3-I and LC3-II, to assess autophagy induction

(Figure 2E). We noted a distinctive loss of both LC3B-I and LC3B-II, rather than ratiometric loss of LC3B-I and

gain of LC3B-II, which is indicative of dysregulated autophagy (33). Additional readouts of autophagic

dysfunction include increased expression of the autophagy adaptor p62 (34). Indeed, increased p62 levels

correlated with LC3B-I and LC3B-II between 8 and 24 hours (Spearman’s r > 0.4) (Figure 2E). Together, these

results suggest that TDP-43 pathology induces autophagic dysfunction and vesicular damage characterized by

the mobilization of the inflammatory lectin galectin-3.

TDP-43a drives a reactive transcriptome in primary monocyte-derived macrophages

To determine if TDP-43a elicited an innate immune response at the transcriptional level, we conducted RNA-

sequencing on hMDM treated with either PBS or TDP-43a for 12 hours. We identified 262 differentially

expressed genes (Figure 3A; DESeq2, p < 0.01, absolute fold change > 1.15), including those with known

roles in innate immune response including galectin-3, TLR4, CCL22, SIGLEC1, and complement receptor 1

(CR1). Additional differentially expressed genes included signaling regulators JAK1 and STAT5B, suggesting

broad immune modulation following TDP-43a stimulation. We note that these changes were more targeted and

specific (both in number and magnitude) compared to a 6-day, ‘chronic’ amyloid beta oligomer (

1-42

)

stimulation (712 differentially expressed genes) and a more potent 12-hour lipopolysaccharide (LPS)

stimulation (5940 differentially expressed genes) (Figure 3B; Supplementary Table S3A-C). TDP-43a gene

expression differences were distinct compared to

1-42

, as only 48 genes showed a common stimulation-

dependent regulation, and among these, the magnitude of differential expression was unique between TDP-

43a and

1-42

(Figure 3C). Gene ontology (GO) clustering indicated that TDP-43a preferentially regulated

innate immune responses (cytokine production involved in immune response), blood cell differentiation

(myeloid cell differentiation and regulation of hematopoiesis) and interleukin-2 signaling (Figure 3D).

Importantly, interleukin-2 signaling is a major orchestrator of innate-to-adaptive immune activation,

predominantly through T-cell activation and maintenance. These data suggest that internalized TDP-43a can

activate innate immune responses and thereby influence the crosstalk between innate and adaptive immune

cells.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

TDP-43a promotes global immunophenotypic changes

Given that there are numerous immune cell types with the potential to phagocytose and react to TDP-43a, we

asked how broadly TDP-43a impacts the human immune landscape using ex vivo human immune cell cultures.

Based on our mechanistic studies of TDP-43a internalization in hMDM, we reasoned that any professional

phagocyte may internalize TDP-43a. To address this question, we optimized a novel mass cytometry-based

aggregate internalization assay, that we termed Aggre-Gate, in bulk PBMCs that allowed for simultaneous

global immunophenotyping and cell population-wide localization of TDP-43a (Figure 4A schematic). To monitor

localization of TDP-43a, aggregates were first coupled to a Tellurium-Maleimide adduct (130Te) and used to

stimulate bulk primary human PBMC cultures. We confirmed that our assay follows linear kinetics of

internalization in phagocytes such as classical monocytes by traditional Boolean gating (Figure 4B-C and

Supplementary Figure S4A-B). We next used Spanning-Tree Progression Analysis of Density-Normalized

Events (SPADE) to stratify PBMCs and overlay relative 130Te intensity to identify TDP-43a internalization on a

cell cluster-specific basis. We observed TDP-43a internalization by all available phagocytic and antigen

presenting cells in PBMC preparations, including classical monocytes (77.13% ± 3.19%; ****p < 0.0001) and

dendritic cells (65.28% ± 8.91%; ****p < 0.0001) by SPADE and manual analyses. Strikingly, this analysis

platform identified TDP-43a internalization in B-cells, an unconventional phagocyte implicated in both antigen-

presentation and antibody production (8.683% ± 3.63%; *p = 0.0263). Finally, as expected, classically

described non-phagocytic cells such as CD8 T-cells (

p > 0.999) and natural killer (NK) cells (

p = 0.9978) did

not significantly internalize TDP-43a (Figure 4E-F).

In parallel, our TDP-43a mass cytometry assay also allowed us to profile adaptive immune cell

activation following acute, 24-hour stimulation with TDP-43a. Our phenotyping strategy included markers of

proliferation, early activation, tissue residency/homing, and immune checkpoints (Table 4). Unless otherwise

specified, CD4 and CD8 T-cell subsets refer to the TCR

αβ

subtype, as this is the predominant TCR

responsible for antigen recognition. Both CD4 and CD8 T-cell subsets exhibited significant reduction in surface

expression of CD127 levels following TDP-43a treatment, which is associated with immune activation,

increased effector activity, and senescence (35–37). CD127 downregulation was evident in CD4 effector T-

cells (-0.232 ± 0.06 transformed intensity units; *p = 0.0186) (Figure 4G), CD4 central memory T-cells (-0.276 ±

0.09 transformed intensity units; *p = 0.0332) (Figure 4H), CD8 effector T-cells (-0.225 ± 0.04 transformed

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

intensity units; **p = 0.0082) (Figure 4I), and CD8 central memory T-cells (-0.261 ± 0.08 transformed intensity

units; *p = 0.0393) (Figure 4J).

TDP-43a stimulation promotes antigen presentation and activation of naïve T-cells

Classically, protein-driven T-cell immunity requires an essential intercellular signaling event between

antigen presenting cells and naïve CD4 and/or CD8 T-cells before memory responses are established. This

activation can be monitored via transient intracellular calcium release within T-cells, which is the downstream

result of T-cell receptor-mediated PKC

1 signal transduction (38). To address antigen presentation capabilities

of TDP-43a-primed innate immune cells, we first adopted an established co-culture assay using Raji Burkitt

lymphoma and Jurkat T-cell lines (39). This model was chosen as Raji and Jurkat cell lines have compatible

MHC and TCRs, a necessity when assessing classical antigen presentation (40). Raji B-cells were labeled with

CellTracker DeepRed and stimulated with immunopurified TDP-43a or isotype control product for 16-hours.

Jurkat T-cells were labeled with CellTracker green and added to Raji cultures for 45 minutes and labeled with

phalloidin to detect actin-dense immune synapses that form at junctions between APC and T-cells. The

frequency of actin-dense intercellular contacts was quantified using laser-scanning confocal microscopy

(Figure 5A-B). Phytohemagglutinin (PHA) was used as a positive control to non-specifically activate Raji cells

and encourage immune synapse formation. TDP-43a (3.507% ± 0.17%; ****p < 0.0001) and PHA (2.533% ±

0.27%; ****p < 0.0001) stimulated Raji cells exhibited significantly higher frequencies of actin-polarized cell

clusters with Jurkat T-cells relative to isotype control (0.513% ± 0.25%). We note that TDP-43a-stimulated Raji

cells showed an even higher level of synapse formation compared to PHA (**p = 0.0049). However, since PHA

is not a peptide antigen, we cannot directly compare the antigenic potency between TDP-43a and PHA.

To further validate this finding in a primary human system, we used syngeneic macrophage-

mononuclear cell co-cultures to ensure HLA haplotype compatibility to appropriately assess TCR signaling. In

this paradigm, hMDM were stimulated with TDP-43a or isotype control material for 16-hours. In parallel, frozen,

syngeneic peripheral blood mononuclear cells (PBMCs) were thawed and allowed to recover overnight. At the

time of assay, PBMCs were labeled with CellTracker Green (CTG), co-cultured with macrophages for 45

minutes, and subsequently fixed and labeled with phalloidin prior to analysis by Airyscan confocal microscopy.

Strikingly, we noticed the presence of CTG-positive macrophage synapses with TDP-43a treatment (Figure

5C), but a complete absence of CTG labeled cells in the isotype control treatment. The lack of synapses in the

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

isotype control condition is likely due to the inability to elicit hMDM activation leading to reduced chemotactic

cues that recruit CTG-positive PBMCs. Supporting this possibility, our secretome analysis revealed significant

increase in a notable T-cell chemotactic factor, CXCL16 (Log2fold change 2.836, FDR < 0.05, *p = 0.034)

following TDP-43a stimulation. We then performed a separate live-cell imaging analysis where mixed hMDM-

PBMC cultures were imaged immediately following the addition of TDP-43a or isotype control product.

Compared to control, TDP-43a stimulated cultures exhibited robust migration of PBMCS to hMDM that were

actively interacting with aggregates visible by phase-contrast microscopy (Supplementary Video SV1). Taken

together, these data suggest hMDM stimulated with TDP-43a exhibit heightened chemo-attractive properties

that induce recruitment and engagement of adaptive immune cells.

Our data so far suggests that TDP-43 aggregates can drive immune synapses between antigen

presenting cells and adaptive immune cells in several co-culture models of TDP-43a stimulation. However, it is

still unclear whether immune synapse formation in response to TDP-43a leads to cellular activation. Since

sustained memory T-cell responses indicate prior activation of naïve T-cells, we asked whether TDP-43a

drives antigen presentation and activation of naïve T-cell populations.

To address this question, we turned to a sensitive live-cell calcium imaging and co-culture assay

(Figure 5D schematic). A functional readout of T-cell activation during antigen presentation is the immediate

release of intracellular calcium following TCR engagement (38). Thus, we stimulated hMDM for 16 hours with

TDP-43a or isotype control product. In parallel, syngeneic naïve CD4 and CD8 T-cells were enriched by

negative magnetic activated cell sorting (MACS). Here, naïve T-cells are defined as CD45RO-, CD45RA+,

CCR7+, and CD27+ (Supplementary Figure S5A). As a positive control, we included co-cultures where hMDM

were stimulated with HIV virions, a previously established method to induce hMDM-T-cell antigen presentation

and calcium signaling (41). Next, hMDM and either naïve CD4 or CD8 T-cells were co-cultured and

immediately imaged for 45 minutes to monitor migration, cellular contact, and calcium activity. Strikingly, we

observed recruitment of T-cells to the hMDM plasma membrane (Figure 5E). Following recruitment, T-cells

formed transient immune synapses, and elicited rapid intracellular calcium responses. Quantification of calcium

spiking events revealed significant increases in both naïve CD4 (0.634 spikes/minute; **p = 0.0062) and CD8

T-cells (0.698 spikes/minute; ***p = 0.0002) following co-culture with TDP-43a-stimulated hMDM relative to

isotype control stimulations (0.308 spikes/minute) (Figure 5F). These findings suggest TDP-43a can drive

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

naïve T-cell activation via contact-mediated antigen presentation, and thus satisfy early requirements for

formation of conventional memory T-cell responses.

Imaging mass cytometry reveals antigen presenting microenvironments in ALS brain

To support our in vitro and ex vivo assays, we determined whether the molecular and cellular features of

innate-to-adaptive immune activation are present in the ALS brain. Using formalin-fixed paraffin embedded

ALS and control brain sections, we performed a high parameter imaging mass cytometry assay to enable the

simultaneous labeling of TDP-43 pathology (Figure 6A) as well as a panel of heterogenous brain cell types

including brain-resident immune cells (Figure 6B) using a 17-plex antibody panel that included functional

markers of antigen presentation and memory T-cells (Table 5). We observed in vivo features of innate and

adaptive immune activation in ALS tissue that recapitulated features identified in vitro. These included

microenvironments containing activated microglia (CD68+) associated with pathological TDP-43 (phosphor-

serine 409/410) (Figure 6B, arrowheads). Additionally, we identified a subset of myeloid cells that were positive

for CD163, a marker associated with infiltrating macrophages (42), galectin-3, and MHC-II (HLA-DR). We also

observed intercellular contacts between galectin-3+, HLA-DR+, CD163+ myeloid cells and CD8 T-cells (Figure

6B, white arrows). Interestingly, these CD8 T-cells were also positive for the memory T-cell marker CD45RO

but were granzyme B-negative, although granzyme B-positive lymphocytes negative for T-cell markers were

identified in the gray matter (Supplementary Figure S6).

CD8 T-cell subpopulations show divergent responses to TDP-43a in ALS patients

Our data thus far suggest TDP-43 pathology alters immunophenotypes both in vitro and ex vivo. However, it

remains unknown whether immune responses to TDP-43 pathology are clinically relevant. Therefore, we

interrogated the relationship between immune populations, TDP-43a reactivity, and clinical progression of ALS.

First, we employed a whole-blood immune profiling assay (Figure 7A), which allows us to analyze the entirety

of a patient’s immune profile without fractionation and cryopreservation, which minimizes frequency artifacts in

sensitive cell types such as monocytes and NK cells (43). Furthermore, we were able to characterize

granulocytes that are typically absent during PBMC profiling protocols (44).

ALS patients were categorized as fast-progressing (

ALSFRS/t > 1.0; n=6), standard-progressing (0.5

ALSFRS/t < 1.0; n=8), and-slow progressing (

ALSFRS/t < 0.5; n=9) based on the rate of change to their

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

revised ALS functional rating scale (ALSFRS-R) calculated between clinical visits (Table 6). Per-patient

immune profiles assayed through mass cytometry were then analyzed with VoPo, a machine learning pipeline

for identifying immunological correlates of particular clinical outcomes (Stanley et al., 2020). Briefly, Vopo first

integrates cells across all profiled samples to identify a common set of cell-populations in an agnostic manner.

Frequencies and functional marker readouts within each of these uncovered populations can then be

systematically compared to identify distinct cell subtypes that may be protective against fast progression (45).

Our analyses of the patient cohorts included several patient group comparisons to highlight the most relevant

immune populations in each of the fast, standard, or slow progression categories (Figure 7B and

Supplementary Figure S7).

Using VoPo, we first compared whole-blood profiles between fast- and standard-progressing ALS

patients. Candidates were validated by manual gating and statistical analysis. We identified a statistically

significant increased frequency of the mucosal associated invariant T-cell/natural killer T-cell (MAIT/NKT) niche

in fast-progressing patients compared to standard-progressing patients (0.10% vs. 0.23% of CD45+

leukocytes, respectively; *p = 0.0221, Mann-Whitney U test). Additionally, we observed a significant reduction

in senescent NK cells in fast-progressing patients (3.48% vs 2.16% of CD45+ leukocytes, respectively; **p =

0.0027, Mann-Whitney U test) (Figure 7C). We next compared whole-blood profiles between standard- and

slow-progressing patients. We identified a significant increase in plasmacytoid dendritic cells (0.13% vs 0.24%

of CD45+ leukocytes, respectively; *p = 0.0115, Mann-Whitney U test) and terminally differentiated CD8 T-cells

re-expressing CD45RA (21.07% vs 47.21% of CD8+ T-cells, respectively; *p = 0.0418, Mann-Whitney U test)

in slow-progressing patients relative to standard-progressing (Figure 7D). We observed an increasing trend

(though not significant) in myeloid dendritic cells (

p = 0.0549) and eosinophils (

p = 0.175, Mann-Whitney U

test) in slow-progressing patients relative to standard-progressing patients.

We next wanted to assess whether immune population differences can be used to segregate slow-,

standard-, and fast-progressing patients as a proof-of-principle for predictive modeling of ALS progression. To

do so, we performed principal component analysis on whole blood profiles. At this sample size, we identified

relative segregation of standard-progressing patients, while slow- and fast-progressing patients appeared to

cluster together (Figure 7E). In further support of this, VoPo analysis revealed a lack of differentiating signal

when comparing slow- against fast-progressing patients in several clusters, including the CD8 TEMRA cluster

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

(CD8 T-cell 1). However, signal in this cluster returned when we compared standard to non-standard profiles

(aggregated slow- and fast-progressing profiles), with enrichment in the non-standard progressing group

(Figure 7F). Although slow- and fast-progressing patients had similar CD8 TEMRA profiles based on VoPo, we

suspected that this particular sub-population may be functionally distinct and/or uniquely regulated in slow-

versus fast-progressing subtypes. Indeed, CD8 TEMRA have been characterized as functionally dichotomous,

with cytotoxic and regulatory roles governed by the IL-7/CD127 (IL-7R) axis (45). We reasoned that bona fide

functional readouts of CD8 TEMRA activity would provide additional information for distinguishing between

fast- and slow-progressing immune profiles.

To test this, we performed a proof-of-concept clinical case assessment. We implemented a blinded

whole-blood stimulation and mass cytometry assay with the goal of capturing functional differences, rather than

population differences, of ALS CD8 TEMRA following stimulation with TDP-43a. We performed the assay on 3

different ALS donors, one each of a fast-, standard-, and slow-progressing patient, and 3 same-visit control

samples taken from either a caretaker or cohabitating donor that is considered environmentally matched to the

respective patient. This design allowed us to reference ALS responses against natural immunological

variability that might arise from different genotypes and environments. Patient and control whole blood were

stimulated for 16-hours with TDP-43a or dimethyl sulfoxide (DMSO) to establish baseline. We first opted to

assess markers of CD8 TEMRA senescence, an inherent property of cytotoxic T-cells with regulatory and

pathological implications (45). Specifically, we calculated the ratio of proliferative (CD127+CD57-) to

senescent-like (CD127-CD57+) CD8 TEMRA. Each ratio was normalized to the respective DMSO control.

Notably, the slow-progressing patient TEMRA exhibited a 2.74-fold increase (non-inferential) in conversion of

proliferative CD127+CD57- cells to CD127-CD57+ senescent-like TEMRA following stimulation with TDP-43a.

This conversion was absent in the matched control, as well as fast- and standard-progressing patients and

respective controls (Figure 8A). Next, we assessed expression changes of CD103 as a readout of tissue-

residency, a property that can minimize or exacerbate chronic tissue damage (46). Remarkably, the slow-

progressing patient TEMRA population exhibited the largest loss of CD103+ TEMRA (-2.75%; non-inferential),

compared to controls, standard-progressing (-0.76%; non-inferential) and fast-progressing (+0.14%; non-

inferential) patients (Figure 8B). While this data indicates proof-of-concept functional differences across all

three disease progression states, we emphasize that larger sample sizes for whole-blood stimulation and

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

profiling will be required for future quantitative ALS biomarker studies. These data suggest that ALS patients

exhibit dichotomous immune responses to human TDP-43 pathology that may delineate slow-, standard-, and

fast-progressing ALS patients. We note that, since TDP-43a contains numerous co-aggregating factors that

are targets of innate and adaptive immunity, the identity of the specific antigen(s) triggering this response is not

yet known. Regardless of the exact self-antigens, these data warrant further investigation into TDP-43a

reactivity and the potential for immune monitoring as a readout for ALS phenotypes

Discussion

We discovered a novel link between innate and adaptive immune activation in response to pathological TDP-

43a. It was recently demonstrated that ALS adaptive immune responses are restricted to self-proteins (8) and

not pure TDP-43 protein (47). Using physiologically relevant TDP-43 aggregates containing accessory co-

aggregating DAMPs that we isolated from human cells, we provide evidence that this pathological TDP-43

species acts like a classically defined antigen with potential to induce innate and adaptive immune activation.

We assessed the ability of TDP-43 aggregates to 1) serve as a substrate for innate immune cell scavenging, 2)

drive innate immune cell activation, 3) augment antigen presentation to naïve CD4 and CD8 T-cells, and 4)

elicit memory T-cell responses in ALS immune cells.

When TDP-43a was used to stimulate hMDM, we observed phagocytosis-dependent internalization that

correlated with general immune reactivity (i.e., release of pro-inflammatory factors such as caspase 1, MIF,

and CXCL16), vesicle rupture, and autophagic dysfunction. It remains to be explored whether this autophagic

dysfunction is a key event in the activation of both CD4 and CD8 T-cells described herein. Classically, cargo

destined for degradation via autophagy is loaded onto MHC-II for antigen presentation to CD4 T-cells.

Conversely, intracellular antigenic cargo is degraded via proteasome and loaded onto MHC-I via TAP for

antigen presentation to CD8 T-cells. In certain host-defense states, non-conventional presentation routes of

internalized cargo can lead to both class I and class II presentation. Such events are termed cross-antigen

presentation and are characterized by downstream activation of both CD4 and CD8 T-cells (48). Our

observations that TDP-43a promotes autophagic dysfunction and vesicle rupture as evidenced by galectin-3

imaging, TEM, proteomics analyses, and biochemical assays, leads us to speculate that a vesicle escape

mechanism may drive cross-presentation naïve CD4 and CD8 T-cell activation and T-cell memory responses.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

While it has been shown that TDP-43 pathology can compromise both autophagy and proteasomal processing

in ALS-affected neurons, further experiments in immune cells are warranted to assess this hypothesis.

Our RNA-sequencing data further validated the galectin-3 feed-forward pathway of innate immune

activation following vesicle rupture, including increased expression of LGALS3 and TLR4 transcript in hMDM

following TDP-43a stimulation. We also gleaned information regarding ALS immunophenotypes including

reduced TGF

and STAT5 transcript levels, factors necessary for maintenance of regulatory T-cells (49) that

are reduced in ALS patients and inversely correlate with autoimmune-type Th17 T-cells (50–52). Moreover, we

observed increased expression of CCL22, a homing cytokine for CCR4-expressing NKT cells (53–55), which

our VoPo analysis highlighted in fast-progressing ALS patients. Finally, GO Term analysis highlighted the IL-2

signaling, a major T-cell regulatory pathway, following TDP-43a stimulation. These data further support

intercellular, innate-to-adaptive immune signaling as a link between early innate responses to TDP-43a and

clinically relevant adaptive immune signatures in ALS patients.

Using multiple primary and cell culture models of antigen presentation, we demonstrated that TDP-43a

stimulation of antigen presenting cells formed immune synapses with T-cells (Figure 5). With respect to naïve

T-cells, we observed intracellular activation via calcium signaling, a hallmark event for TCR-mediated T-cell

activation and adaptive immunity. Our high parameter neuroimmune imaging platform highlighted intercellular

contacts between innate and adaptive immune cells surrounding TDP-43 pathology in the ALS brain. We

emphasize that the exact peptide(s) presented on MHC in these experiments are unidentified. At the moment,

we cannot 1) conclude what protein within the TDP-43 aggregate is responsible for eliciting this reaction, nor 2)

assume it is necessarily the same peptide sequence or protein source across human donors. Additional

variables could include the highly polymorphic nature of MHC and its impact on peptide-selection and

presentation, or natural variation of thymic selection against anti-self T-cells that may exist on an individualistic,

disease-specific, or disease subtype-specific basis. Nevertheless, these are certainly topics for future global

immunomic studies.

Finally, we implemented machine learning to identify peripheral immune signatures among ALS

patients that were associated with ALS progression rates. Surprisingly, ALS subtype-specific immune

populations exhibited functional responses when stimulated with TDP-43 pathology. We were surprised to

observe that, relative to standard-progressing patients, both slow and fast-progressors exhibited increased

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

representation of various innate and adaptive immune populations, one of these being CD8 TEMRA cells. We

note an overall higher representation of CD8 TEMRA cells in ALS, which is in line with previous reports (8).

However, we found that CD8 TEMRA cells—typically regarded as terminally differentiated with high cytolytic

potential—were overrepresented in non-standard-progressing ALS patients. One consideration is that TEMRA

cells could have dichotomous, regulatory properties in slow progressors but cytotoxic properties in fast-

progressors. For example, if CD8 TEMRA cells can adopt senescent phenotypes (CD127-CD57+) and regress

from sites of tissue inflammation (CD103-), a patient may be protected from advanced ALS phenotypes.

However, if CD8 TEMRA maintain proliferative capacity (CD127+CD57-) and reside in the CNS (CD103+),

then off-target tissue damage may expedite disease progression. We therefore suggest implementation of

functional assays, rather than solely population-based assays, that may provide greater clinical utility when

combined with standard profiling approaches. By analyzing whole-blood, but not PBMCs, within 4-hours of a

patient visit, we were able to identify changes to NKT cells in fast-progressing ALS. Though larger sample

sizes in the future are necessary to extend these findings, we suspect that critical cellular and functional

information may be lacking in cryopreserved PBMC preparations (8, 56, 57). Very few studies have been able

to evaluate NKT cells, which are significantly impacted by cryopreservation damage following PBMC isolation

and storage (43) (58). In summary, this study sheds light on how TDP-43 proteinopathy may be intimately

linked to immune dysfunction in ALS and further highlights whole-blood immune assays as a potential

precision-medicine approach for diagnosing and predicting clinical features of ALS progression.

Materials and Methods

Reagents, cell lines, and cell culture

Raji Burkitt lymphoma cell line (ATCC, CCL-86) and Jurkat T-cell clone E6-1 (ATCC, TIB-152) were procured

from UNC Tissue Culture Facility (UNC-TCF). Cells were cultured in vented T-75 flasks in Roswell Park

Memorial Institute (RPMI) 1640 media (Thermo Scientific, A1049101) supplemented with 10% fetal bovine

serum, 2.0 mM L-glutamine, and 1X penicillin/streptomycin. Cells were maintained between 1.0-2.0x

cells/mL.

To passage, cells were collected by centrifugation at 300X RCF for 5 minutes. The cell pellet was resuspended

in 1.0 mL of calcium/magnesium-free PBS and incubated at 37°C for 5 minutes. Cell suspension was strained

through a 40-micron nylon strainer, counted as above, and re-seeded to a fresh T-75 flask. For a complete list

of key resources, see Table 1.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Peripheral Blood Mononuclear Cell (PBMC) Isolation

PBMCs were isolated from de-identified buffy-coated leukocytes obtained from New York Blood Center (Long

Island City, NY) following screening against bacteremia and viremia. Briefly, anticoagulated blood (heparin or

ACD citrate as indicated) was diluted 1:1 with 1X sterile phosphate-buffered saline (PBS). Diluted blood was

overlaid atop Ficoll-Hypaque density gradient (Sigma, GE17-1440-02) and centrifuged at 120X RCF for 20

minutes with zero brake. Buffy coat was isolated, washed in 1X PBS, centrifuged as above twice, and cleared

with 1X erythrocyte (RBC) lysis buffer for 10 minutes with gentle agitation. Cleaned PBMCs were then washed

with 1X PBS, counted using Trypan Blue exclusion hemocytometry, and plated according to assay or

differentiation protocol. For storage of PBMCs, cells were cryopreserved in 10% DMSO in antibiotic-free media

containing 10% FBS (Sigma, F2442), and 2 mM L-Glutamine (Gibco, 25030-081) in Dulbecco’s Modified Eagle

Media (DMEM; Gibco, 31-053-028).

Primary human monocyte-derived macrophage (hMDM) culture

Freshly isolated PBMCs were plated at a density of 1.0x

cells/mL into ultralow adhesion plastic tissue culture

dishes (Corning CLS3473), glass bottom slides (Cellvis, C8-1.5H-N), or glass-bottom culture dishes (Nunc,

150680) in complete DMEM consisting of 10% FBS, 2.0 mM L-glutamine, 1X Penicillin/Streptamycin (Gibco,

15140122). Monocytes were allowed to adhere to substrate for 5 days at 37°C, 5% CO

. At day 5 in culture,

media was exchanged with complete media plus 15 ng/mL carrier-free recombinant human granulocyte-

macrophage colony stimulating factor (rhGM-CSF; R&D,

215-GM-010/CF). Thereafter, differentiated

macrophages were maintained with half-volume media exchanges performed every other day.

TDP-43 aggregate purification

The following is an abridged protocol based on Evangelista et al., 2023. All buffers and plastics were sterile

and procedures performed in a laminar flow hood. Briefly, HEK293A (ATCC, CRL1573) cells (ATCC,

CRL1573) were seeded to 15-cm tissue culture plates such that they reached 80% confluence at day of

transfection. Cells were transfected with 8 µg of pcDNA3.1-GFP-mNLS-K145Q using Fugene 6 Transfection

reagent (Promega, E2692) per manufacturer protocol. Cells were harvested 48-hours post-transfection in

sterile 1X RIPA buffer supplemented with protease, phosphatase, and RNAse inhibitors (Proega, PRN2111).

Cells were passed through 18-, 21-, and 25-gage syringes, vortexed at top speed for 30 seconds, and then

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

treated with RQ1 RNAse-free DNAse (Promega, M6101) for 30 minutes at 37°C. Sarkosyl was added to a final

concentration of 0.5%. Lysates were sheared as above, then centrifuged at 100,000X RCF for one hour at

4°C. The supernatants were removed. Crude pellets were stored at -80 indefinitely until immunoprecipitation.

To prepare immunoprecipitation matrix, 30 µL of Protein A/G magnetic agarose (Thermo Scientific, 78609) was

incubated with 15 µg of TDP-43 antibody (Proteintech, 10782-2-AP) for one hour at 4°C. Residual antibody

was removed, beads were washed twice with 1X RIPA, and twice with 1X PBS. BS3 cross-linker (Thermo

Scientific, 21580) was diluted to 5 mM with 1X sterile PBS. Antibody-bead complexes were crosslinked for 30

minutes at room temperature on an orbital shaker. Crosslinking was quenched with a 1:10 ratio of 1.0 M Tris

pH 7.5. Non-crosslinked antibody was removed with 2 washes of Pierce Gentle Antibody elution buffer

(Thermo Scientific, 21013), and the beads washed twice more with 1X RIPA. Beads were then blocked

overnight in non-protein block buffer (Licor Bioscience, 927-60001) with 1mM DTT and 1.0% Tween-20 at 4°C

with orbital rotation. For control samples, Rabbit isotype control antibody was used in place of 10782-2-AP. To

immunoprecipitated TDP-43 aggregates, pellets were retrieved from -80°C. and sonicated in 500 µL of

extraction buffer (1X RIPA supplemented with 0.5% Sarkosyl, 1.0% Tween-20, 1.0 mM DTT, protease,

phosphatase, and RNAse inhbitors). Bead slurry and lysate were combined and allowed to bind overnight at

4°C with constant rotation. The next day, supernatants were removed, beads were washed 3X in extraction

buffer, and aggregates were eluted with a 3-series wash with 5.0 M NaCl warmed to 55°C. Aggregates were

pelleted at 100,000x RCF for 1 hour, and sonicated into a 1.0 mL solution (15 second pulses, 30 second

intermission on ice, for 4 cycles) using sterile 1X PBS unless otherwise specified.

Immunofluorescence analysis

Coverslips were gently rinsed in room temperature PBS and then immediately fixed in fresh, 4.0%

paraformaldehyde (PFA; Electron Microscopy Services, 15713S) for 10 minutes at room temperature. Cells

were rinsed 3X in PBS, then permeabilized using 0.2% Triton X-100 in PBS for 8 minutes at room temperature.

Cells were blocked in 2.0% normal goat serum (Sigma, NS02L-1ML) + 0.2% Triton X-100 (PBSTN) for one

hour at room temperature. Cells were incubated with primary antibodies in 2.0% PBSTN overnight at 4°C.

Cells were rinsed in PBST and subsequently stained with fluorophore conjugated secondary antibodies in

PBSTN for one hour at room temperature protected from light. Cells were preserved in ProLong Diamond anti-

fade mounting medium (Thermo Scientific, P36965).

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

hMDM aggregate uptake assay for confocal microscopy

hMDM were cultured in glass bottom dishes. Cells were pre-charged with 10 µM Cytochalasin D (Sigma,

C8273), 80 µM Dynasore (Sigma, D7693), 10 µM EIPA (MedChem Express, HY-101840), or 0.05% DMSO

control for 30 minutes at 37°C. Cells were then stimulated with 0.025% (v/v) TDP-43a and returned to

incubator for 4 hours. Cells were rinsed 1X in PBS, then immediately fixed for 10 minutes at room temperature

with 4.0% PFA. Cells were rinsed 2X in PBS and stained with 10 µg/mL Wheat-Germ Agglutinin (WGA)

AlexaFluor 647 conjugate (Invitrogen, W32466) for 10 minutes at room temperature. Cells were rinsed 2X in

PBS and mounted with ProLong Diamond antifade mounting media.

Confocal microscopy and live-cell imaging

Images were obtained on an inverted Zeiss 800/Airyscan laser-scanning confocal microscope fitted with 405,

488, 561, and 640 nm diode lasers and gallium arsenide phosphide (GaAsP) detectors. Live cells were

incubated in a 37°C heated stage in a humidified 5% CO

chamber. Image analyses were performed in FIJI

(version 1.53t) (59). Brightness and contrast were similarly adjusted between treatments.

Solubility fractionation and immunoblotting

6-well tissue culture plates were removed of media and rinsed 1X in 1.0 mL PBS. Cells were collected in 250

µL of 1X radioimmunoprecipitation assay buffer (RIPA) by scraping over ice. In all cases, RIPA buffer was

supplemented with protease and phosphatase inhibitors. Lysates were sonicated 20-times at 25% amplitude

with a QSonica hand sonicator probe over ice. Samples were centrifuged at 21,000X RCF for 30 minutes at

4°C. Supernatants were transferred to chilled microcentrifuge tubes and stored at -80°C as the soluble fraction.

Pellets containing insoluble fraction were resuspended in Urea plus protease and phosphatase inhibitors, and

sonicated for 10 pulses as above. Samples were centrifuged as above, and the supernatants stored as the

insoluble fraction. To prepare samples for sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-

PAGE) analysis, protein lysate concentrations were normalized by bicinchoninic assay. 5.0-10.0 µg of protein

lysate were denatured in 1X reducing Laemmli buffer, boiled at 98°C for 5 minutes, and electrophoresed on

12% SDS-PAGE gel. Proteins were transferred to nitrocellulose membrane at 4°C, 100V, for 75 minutes.

Membranes were blocked in Tris buffered saline-0.2% Tween (TBS-T) with 2.0% skim-milk for 30 minutes at

room temperature with rocking. Membranes were incubated with primary antibodies (Table 2) at 4°C overnight

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

in 0.2% Milk-TBS. Membranes were washed 3X in TBS-T, incubated with respective secondary antibodies

(Table 3) for 60 minutes at room temperature with rocking and washed as above. Membranes were imaged by

enhanced chemiluminescence and quantified using ImageStudo Lite.

Secretome and proteome analysis

Differentiated macrophages in 10-cm dishes underwent complete media change to serum-free DMEM. Cells

were then stimulated with 0.025% (v/v) aggregate-containing media or equal volume of isotype control material

and allowed to incubate for 16-hours. Cell culture media was removed and centrifuged at 300 RCF for 5

minutes at 4°C to remove cell debris, followed by filtration through a 0.2 µm syringe filter. Cells were harvested

by gently scraping in ice-cold PBS, centrifuged as above and supernatant was removed. All components were

then stored at -80°C until a full set of replicates was obtained. For media analysis, proteins were precipitated

with deoxycholate and trichloroacetic acid. Next, protein precipitates and cell pellets were solubilized in 8M

urea. Proteins were then reduced, alkylated, and digested with Trypsin at room temperature overnight.

Digested peptides were then acidified with trifluoroacetic acid, desalted, and reconstituted in 0.1% formic acid.

Samples were analyzed by reverse phase liquid chromatography (LC)-MS/MS using Velos Orbitrap mass

spectrometer (Thermo). Experiments were calibrated to a mass accuracy of < 0.05.

Electron Microscopy

Adherent cells were fixed in 2% paraformaldehyde/2.5% glutaraldehyde in 0.15 M sodium phosphate buffer,

pH 7.4 for 1 hour at room temperature, then stored in fixative at 4

C until processed for TEM. Fixative was

aspirated and cells were washed three times in 0.15 M sodium phosphate buffer. After washing, reduced

osmium tetroxide (1% osmium tetroxide with 1.25% potassium ferrocyanide in 0.15 M sodium phosphate

buffer) was added to cover the cellular monolayer and incubated at room temperature for 1 hour. Cells were

washed three times with deionized water and dehydrated through an ascending series of ethanol (30%, 50%,

75%, 90%, 100%, 100% 100%). They were then infiltrated with three exchanges of Polybed 812 epoxy resin

(Polysciences, Inc., Warrington, PA) before being embedded in fresh 100% Polybed 812 epoxy resin and

allowed to cure at 60

C until hardened. Resin blocks were sectioned at 80 nm using a diamond ultra knife on a

Leica UCT7 ultramicrotome and mounted on 200 mesh copper grids. Grids were stained with 4% aqueous

uranyl acetate for 12 minutes followed by Reynold’s lead citrate for 8 minutes, based on Reynolds et al., 1963

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

(60). Samples were viewed using a JEOL JEM-1230 transmission electron microscope operating at 80 kV

(JEOL USA, Inc., Peabody, MA) and images were obtained using a Gatan Orius SC1000 CCD Digital Camera

and Gatan Microscopy Suite 3.0 software (Gatan, Inc., Pleasanton, CA).

RNA Sequencing

RNA-seq library preparation. Total RNA was extracted using RNeasy Mini Kit (QIAGEN) guided by the

manufacturer’s instructions. Each sample's RNA concentration and RNA integrity number (RIN) were

measured using Qubit and Agilent Tapestation 4150 system. The depletion of ribosomal RNAs (rRNAs) and

the generation of stranded RNA-seq libraries were performed using KAPA RNA HyperPrep with RiboErase kit

with 500 ng of isolated RNA as input and following the guidelines. Final libraries were quantified and

normalized by Qubit and DNA Tapestation and then combined to a total 4 nM library. The combined library was

subsequently subjected to paired-end 75bp read sequencing in the Illumina Nextseq550 platform.

RNA-seq data analysis. The quality of the reads was assessed by FastQC (version 0.11.9). Low-quality reads

and adapters were trimmed by using Trim Galore (version 0.6.7). The remaining reads were mapped to the

hg19 transcriptome (GENCODE release19) and quantified by Salmon (v. 1.10.0). The estimated counts for

each sample were summarized by using tximport (v. 1.28.0). Differential gene analysis was applied to the

counts matrix with DESeq2 (v. 1.40.2) using a design adjusting for technical bias when calculating differences

between treatment groups (~Replicates+Condition). The method “apeglm” was used for effect size shrinkage

(LFC estimates). Genes with an FDR-adjusted p-value below 0.05 (Wald test) and an absolute fold change

greater than 1.15 were selected as the differential genes when comparing the treated samples to their

corresponding controls. The gene symbols and Entrez IDs were annotated to the data by using the package

AnnotationHub (v. 3.8.0) and annotables (v. 0.2.0). The universal gene list and differential gene lists were then

prepared for generating functional enrichment results using over-representation analysis from the GO

database by clusterProfiler (v. 4.8.3), along with its complementary packages. The GO terms were identified

with both FDR-adjusted p-value and multiple testing corrected p-value lower than 0.05. The number of overlap

genes between different conditions was conducted by the package eulerr (v. 7.0.0) displaying the proportion of

area related to the numbers.

Aggre-Gate: TDP-43a internalization assay by mass cytometry

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

130

Tellurium maleimide (130TeMal) coupling. TDP-43a were purified as above. After the final spin, the

supernatant was removed and replaced with 500 µL of sterile Maxpar PBS, vortexed for 5 seconds, and

collected by centrifugation for 30 minutes at 100,000x RCF. The supernatant was removed and 100 mL of

fresh sterile Maxpar PBS was added. The pellet was sonicated for two cycles of 20, 1-second pulses at 25%

amplitude with a 30 second incubation on ice between cycles. Next, 1M DTT was added to a final

concentration of 0.5 mM and the sample partially reduced for 10 minutes at room temperature. The reaction

was quenched with 900 µL of sterile Maxpar Cell Staining Buffer (CSB), followed by centrifugation at 100,000

RCF for 30 minutes. The supernatant was removed, replaced with 1.0 mL of sterile Maxpar PBS, vortexed, and

centrifuged as above. The partially reduced TDP-43a pellet was sonicated in 100 µL Maxpar PBS as above,

and 1.1 µL of 10 mM 130TeMal was added for a final concentration of 100 µM. Coupling occurred for 15

minutes at room temperature with constant mixing by vortex at low speed. The reaction was quenched and

centrifuged as above. 1.0 mL of sterile PBS was added to the TDP-43a pellet and sonicated for two, 20-pulse

cycles on ice as above. Aggregate preparations were then tested for microbial and endotoxin contamination by

culturing a sample in nutrient agar followed by OD600 measurement (<0.001 arbitrary absorbance units) and

LAL assay (<0.05 EU/mL), respectively.

Cell stimulations. On Day 0, PBMCs were thawed in pre-warmed complete RPMI1640 supplemented with 50

units/mL of benzonase endonuclease. Cells were pelleted at 300x RCF for 3 minutes and the supernatant

discarded. Cells were resuspended in 5.0 mL of complete RPMI1640 plus benzonase and transferred to a

vented T-25 flask, allowed to recover overnight. The next morning, cells were collected by centrifugation. The

cells were resuspended in complete media to count viable cells by trypan blue exclusion. 2.0e6 cells were

added to a 24-well round-bottom tissue culture plate in complete RPMI. As indicated, TDP-43a was added to a

final concentration of 0.025% (v/v) and returned to 5% CO

incubation. Stimulations were performed in reverse

chronological order (i.e. starting with longest time-points) then collected at once for cytometry staining to

normalize phenotypic drift in bulk immune cultures along incubation.

Mass cytometry staining and acquisition. Following stimulation, cells were harvested in Maxpar PBS and

centrifuged at 300x RCF for 5 minutes. As needed, cells were barcoded using 1.0 µM TeMal and Selenium

maleimide (SeMal) barcodes in Maxpar PBS for 15 minutes at room temperature. TeMal isotopes were

synthesized by Dr.Youngran Seo at the University of North Carolina Department of Chemistry according to the

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

published protocol of Willis et al (61). Reactions were quenched with 3 volumes of CSB and centrifuged at

300x RCF for 5 minutes. Cells were pooled in 100 µL volumes of CSB. Fc receptors were blocked with 10 µL

TruStain FCX receptor block (BioLegend, 422301) for 10 minutes at room temperature. Blocked cells were

added to 110 µL of CSB and transferred to a Maxpar Direct Immunophenotyping Assay (MDIPA; Standard

BioTools, 201334) tube along with 50 µL of custom checkpoint antibody cocktail (For complete suspension

mass cytometry antibody panels see Table 4). Cells were stained for 30 minutes at room temperature. Cells

were washed twice in 3.0 with mL of CSB followed by centrifugation. Intracellular Ki67 staining was performed

using the FoxP3 transcription factor staining kit per manufacturer protocol (ThermoFisher, 00-5523-00). Cells

were fixed in 2% paraformaldehyde for 1 hour at 4°C, pelleted at 800X RCF for 7 minutes, and resuspended in

1.0 mL of Maxpar Fix/Perm buffer containing 1:3000 Cell ID-Iridium DNA intercalator for overnight incubation at

4°C. The next day, cells were pelleted at 800 RCF for 7 minutes, resuspended in 200 µL of Fix/Perm buffer,

and stored at -80°C until batch analysis. On day of analysis, cells were washed once with CSB, once with Cell

Acquisition Solution (CAS; Standard BioTools), then filtered through a 40-micron filter and diluted in CAS

containing 10% EQ Calibration Beads (Standard BioTools) at 0.5 million cells per mL and acquired on a mass

cytometer (Helios). Mass cytometry data were normalized using Fluigidm CyTOF software (v7.0). Sample files

were pre-processed to remove beads, debris, dead cells, and doublets for further downstream analysis. For

SPADE analysis, down-sampling was set to 10%, with a node setting of 75.

Raji-Jurkat immune synapse assay

Raji cells were stained with 1.0 µM Celltracker Deep Red (Invitrogen, C34565) for 30 minutes in supplement-

free RPMI1640 at 37°C. 2.0e6 cells were added to vented flow cytometry tubes in complete RPMI 1640. As

indicated 0.025% (v/v) TDP-43a, isovolumetric equivalent of isotype control material, or (0.5) µg/mL

phytohemagglutinin (PHA) were added to tubes for a 16-hour simulation period at 37°C. The next day, Jurkat

T-cells were incubated with 1.0 µM CellTracker Green (Invitrogen, C2925) for 30 minutes at 37°C. 2.0e6 Jurkat

cells were added to flow cytometry tubes containing Raji cells and incubated at 37°C for 45 minutes. Cells

were gently pipetted and strained through a 70 µm strainer directly into 1 volume equivalent of 8% PFA (for a

final concentration of 4%). Cells were fixed for 10 minutes at room temperature in the dark. Cells were pelleted

at 800 RCF for 5 minutes, then resuspended in PHEM buffer (5 mM HEPES, 60 mM PIPES, 10 mM EGTA, 2

mM MgCl

pH 7.0 with KOH) plus 0.1% (w/v) digitonin. To label immunological synapses, ActiStain-555

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

(phalloidin) was added to a final concentration of 14 µM and incubated for 30 minutes at room temperature in

the dark. Cells were then rinsed in 1X PBS and incubated with 1ug/mL DAPI in PBS prior to mounting on

microscope slides with Prolong Diamond Antifade mounting media. Synapse frequencies were quantified as at

least 1 Raji and 1 Jurkat T-cell containing a polarized actin contact site from randomized fields of view obtained

on a 20X air objective. Descriptively, synapses were then categorized as pair (one Raji per one Jurkat) or

clusters.

Macrophage-PBMC co-culture assay

hMDM were sub-cultured to glass coverslips. hMDM were stimulated with 0.025% (v/v) immunopurified TDP-

43a and volumetric equivalent of isotype control product for 16 hours. In parallel, cryopreserved syngeneic

PBMCs were thawed and allowed to recover overnight in complete RPMI 1640 plus benzonase as above. The

next day, 2e6 PBMCs per condition were labeled with 1 µM CellTracker green for 30 minutes at 37°C in

supplement-free RPMI 1640. The reaction was quenched and washed with complete media. 1e6 labeled

PBMCs were added drop wise to stimulated hMDM and were returned to 37°C, 5% CO

for 45 minutes.

Without disturbing the cells, a 1-well volume equivalent of 8% PFA was added per well (final concentration 4%)

and allowed to incubate for 10 minutes at room temperature. Media was removed and fresh fixative was added

for an additional 10 minutes. Coverslips were rinsed and permeabilized in PHEM buffer. Actin was stained

using Acti-Stain 555 as above, nuclei were counter-labeled with DAPI, and immune synapses were imaged on

an LSM 800 confocal microscope at 20X and 63X magnification with Airyscan.

hMDM-T-cell calcium imaging assay

Macrophages were stimulated for 16-hours with 0.025% (v/v) TDiP aggregates or isovolumetric isotype control

material in complete media in Nunc glass-bottom microscope dishes. In parallel, naïve CD4 and CD8 T-cells

were isolated from syngeneic cryopreserved PBMC aliquots using the MojoSort magnetic negative selection

kits (BioLegend, 480041 and 480045). Naïve T-cells were incubated in complete medium supplemented with

10 IU/mL recombinant human IL-2 (R&D Systems, 202-IL-050/CF). As a positive control, macrophages were

treated with 10 pg/mL neutralized, CCR5 tropic, HIV virions. Following macrophage stimulations, macrophages

and T-cells were separately loaded with calcium indicator dye using a modified protocol. Here, cells were

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

washed in HEPES-buffered artificial cerebrospinal fluid (137 mM NaCl, 5 mM KCl, 2.3 mM CaCl

, 1.3 mM

MgCl

, 20 mM glucose, and 10 mM HEPES, pH 7.4) and pre-charged with 25% (v/v) Fluo-4 acetoxymethyl

(Invitrogen, # F14201) for 30 minutes. T-cells were added to macrophages at a ratio of 1 (macrophage) : 2 (T-

cell) and imaged on an Olympus IX71 inverted microscope. Time-lapsed images were measured every 6

seconds for 6 minutes for 20 minutes using Metamorph (Molecular Devices). Change to fluorescence intensity

was calculated using baseline calcium levels per cell.

Qualitative Imaging Mass Cytometry

Antibody Conjugation. Antibody panel configuration was performed by Standard BioTools to ensure optimal

isotope assignment based on antigen abundance. Approximately 100 µg of each antibody (carrier- free) were

coupled to lanthanide metals using previously described protocols with the Maxpar X8 labeling kit (Standard

BioTools, 201169B). Briefly, antibodies were partially reduced with TCEP buffer (SBT, 77720) at 37°C.

Following 112 reduction, antibodies were then incubated with an excess of metal-loaded MaxPar X8 polymer

for 90 min at 37°C. Labeled antibodies were purified using a 50 kDa size exclusion centrifugal filter unit. Final

antibody concentration was determined by A

280

. For complete list of IMC probes, refer to Table 5.

Tissue staining. 10 µM formalin-fixed paraffin embedded (FFPE) spinal cord sections from ALS and control

spinal patients were obtained from the Veterans Affairs ALS Brain Biorepository. FFPE sections were de-

paraffinized in xylene followed by progressive rehydration in a series of graded alcohols. Heat-induced epitope

retrieval with pH 9.0 buffer was performed prior to blocking in PBS + 3% BSA for 45 minutes. Tissues were

then incubated with antibodies overnight at 4°C in PBS + 0.5% BSA. Tissues were washed in PBS + 0.2%

Triton X-100 prior to nuclear staining with Iridium intercalator (1:800) for 30 minutes. Finally, tissues were

washed with water and dehydrated at room temperature. Regions of interest (ROI) were previously determined

on an immediate serial section using chromogenic immunohistochemistry, focusing on areas positive for

pS409/410 (1D3). 1 mm x 1 mm ROIs were placed around 2 pS409/410-positive lesions. Synchronized laser

ablation was performed at 200 Hz over a 2-hour period, generating regions with 1 µm

resolution. Data were

exported to MCD Viewer (Standard BioTools) where they were converted to 32-bit TIFF files. Each channel

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

was despeckled and composite images were created using FIJI (version 1.53t). Brightness and contrast for

each marker was adjusted similarly across each ROI.

Whole blood immune profiling and stimulation

General immunophenotyping. ALS patient blood was collected at Duke University and UNC Chapel Hill

Neurology clinics under approved Institutional Review. Blood was collected in 10 mL ACD citrate collection

tubes and processed within 4 hours of collection. 300 µL of anticoagulated blood were then added to one tube

of the Maxpar Direct Immunophenotyping Assay and incubated for 30 minutes at room temperature. Blood was

briefly agitated by tapping every 10 minutes to ensure even staining. Stained blood was then transferred to

cryovials containing 420 µL of SmartTube Proteomic Stabilizing agent (SmartTube Inc,

501351692), mixed,

and allowed to stabilize for 10 minutes at room temperature. Samples were then stored at -80°C until batch

processing and mass cytometry analysis. One day prior to analysis, stabilized blood was thawed at room

temperature and transferred to 10 mL of 1X Thaw/Lyse buffer (SmartTube Inc.) and allowed to incubate at

room temperature for 10 minutes with orbital rotation. Lysed material was pelleted by centrifugation at 500

RCF for 5 minutes. Lysis was repeated one time. Cells were then washed in 5 mL of CSB and resuspended in

1mL of 1X Maxpar PBS. Cells were strained through 70 µM strainer caps into flow cytometry tubes containing

1 mL of 4% fresh PFA (final concentration 2%). Cells were fixed for 1 hour at 4°C. Cells were pelleted at 800

RCF for 7 minutes, then resuspended in 1 mL of Maxpar Fix/Perm buffer containing 1:3000 Cell-ID Iridium

Intercalator. All samples contained 500,000 universal donor PBMCs barcoded with 130TeMal to aid in

normalization. A minimum of 300,000 events were collected in all conditions.

Whole-blood TDP-43a stimulations. For whole-blood stimulation assays, ALS and control patient blood was

collected in 5 mL Heparin tubes. We opted for heparin tubes as these appeared to be preferred for whole-

blood stimulation assays, although a comparative analysis suggests no significant difference in lymphocyte-

antigen response between ACD citrate and heparin (62). 300µL of blood was added to sterile vented flow

cytometry tubes and received either 3.0% (v/v) TDP-43a, 0.5 µg/mL PHA, or isovolumetric DMSO control that

never exceeded 0.05% in culture. Samples were incubated for 16-hours in 5% CO

at 37°C. Blood was

transferred to MDIPA reaction tubes with 50 µL of immune checkpoint antibody cocktail. A metal-minus-one

control for CD103 gating was established by staining whole blood with MDIPA, excluding the CD103 probe.

Following staining, blood was stabilized, cleaned, and prepared for batch analysis as above.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Statistical analysis and data reporting

Where appropriate, data were assessed for normality by Kruskall-Wallis test. Non-parametric data were

analyzed by non-parametric two-tailed t-test, or multivariable Mann-Whitney U-test. Parametric data were

analyzed by two-tailed t-test, one-way ANOVA with Bonferroni correction or Tukey’s multiple comparison test,

or Two-way ANOVA with Sidak’s multiple comparison test. For correlation analysis, Spearman’s r was used.

For primary PBMC stimulation data, paired t-tests were used to match genotypes. Arcsin(h) transformed data

of surface marker expression changes were analyzed via paired t-test. Statistical analysis was performed using

GraphPad Prism Version 9 for Windows (San Diego, CA). Cut-off for statistical significance were as follows: *p

< 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Contributions

B.A.E. conceptualized study, designed, and performed experimentation. J.V.R. and S.C. assisted in confocal

microscopy experimentation and analysis. K.J.P and S.R.C. assisted in mass cytometry analysis and

biochemical assays. O.K.A assisted in biochemical assays. J.M. and K.W. performed electron microscopy

sample preparation and image acquisition. S.S., J.C., imaging mass cytometry panel design and image

acquisition. L.X. and X.C. mass spectrometry analysis. J.A.E., histology. M.A.I, suspension mass cytometry

panel design and sample acquisition. R.T., X.L., and R.B., patient care for study participants, managed clinical

data, and sample acquisition. N.S., VoPo analysis of mass cytometry datasets. R.M. assisted with live-cell

calcium imaging and analysis, and primary cell culture. T.J.C. supervised generation, acquisition, and analyses

of all research data. B.A.E. wrote the manuscript, which was reviewed and edited by all co-authors.

Conflict of Interest

The authors declare no competing conflicts of interest.

Study Approval

The study was approved by the Institutional Review Pro00109640 of Duke University and 22-2638 of

University of North Carolina at Chapel Hill. All participants were studied after providing written informed

consent.

Acknowledgments

The Microscopy Services Laboratory, Department of Pathology and Laboratory Medicine, is supported in part

by P30 CA016086 Cancer Center Core Support Grant to the UNC Lineberger Comprehensive Cancer Center.

UNC Mass Cytometry Core University Cancer Research Fund (UCRF) UNC Cancer Center Core Support

Grant P30CA016086.

J.V.R. and S.C. were supported

by the National Institute of General Medical Sciences of

the National Institutes of Health under award number R35 GM133460. T.J.C. and was supported in part by

National Institute of Neurological Disorders and Stroke R01 NS105981. D.H.P, I.Y.Q., Y.W., and T.J.C were

supported in part by the National Institute of Aging R01 AG066871. Clinical work was supported in part by the

North Carolina Translational and Clinical Sciences (TraCS) pilot award 550KR282107 awarded to R.T. and

X.L. B.A.E was supported in part by the National Institute of Neurological Disorders and Stroke F31 NS122242.

S.R.C was supported by National Institutes of Aging T35 AG038047. ALS tissue specimens were provided by

the Department of Veterans Affairs Biorepository, VA Merit review BX002466.

This work was supported by the

Standard BioTools Discovery Lab at Standard BioTools Inc. Assay development such as staining, titrations,

and data acquisition were performed as paid services. We’d like to thank N. Lane, K. Mottershead, R.J. Perna,

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

A. Sidders, E. Swanson, for strategic discussion. We’d like to thank C. Simmons, M. Chopra

, M. Ward, and H.

Zampa for

patient recruitment, enrollment, and sample acquisition. Finally, we sincerely thank the ALS patients

and family members for their selfless contribution to this body of work.

Table 1. Key Resource Table

Reagent

Vendor

Catalogue

20% Paraformaldehyde

EMS

15713

ActiStain 555

Cytoskeleton Inc.

PHDH1

Bis-sulfosuccinimidyl suberate

(BS3)

Thermo Scientific

21580

CellTracker Deep Red

Invitrogen

C34565

CellTracker Green CMFDA Dye

Invitrogen

C2925

Cytochalasin D

Sigma

C8273-1MG

DAPI Invitrogen D1306

Dimethyl Sulfoxide

Sigma

D2650

Dulbeccom's Modified Eagle

Media

Gibco 31-053-028

Dynasore Sigma

D7693-5MG

EIPA MedChem

Express

HY-101840

Fetal bovine serum

Sigma

F2442

Ficoll Paque Plus

Sigma

GE17-1440-02

Fluo-4-AM Invitrogen F14201

FoxP3 Fixation/Permeabilization

kit

Thermofisher Cat#

00-5523-00

Fugene 6 transfection reagent

Promega

E2692

HEK293A ATCC CRL1573

Human TruStain FcX

Biolegend

422301

Intercept (TBS) blocking buffer

Licor Bioscience

927-60001

Jurkat T-cell lymphoma clone E6-

ATCC TIB-152

L-glutamine Gibco 25030-081

LysoTracker Deep Red

Invitrogen

L12492

Maxpar Direct

Immunophenotyping Assay kit

Standard Biotools

201334

MojoSort Human CD4 Naïve T-

Cell Isolation kit

BioLegend 480041

MojoSort Human CD8 Naïve T-

Cell Isolation kit

BioLegend 480045

Phytohemagglutinin-L Sigma

11249738001

Pierce Gentle Ag/Ab Elution

buffer, pH6.6

Thermo Scientific

21013

Protein A/G magnetic Agarose

Thermo Scientific

78609

Raji B-cell lymphoma line

ATCC

CCL-86

rhGM-CSF R&D

Systems

215-GM-010/CF

rhIL-2 R&D

Systems

202-IL-010

Roswell Park Memorial Institute

1640 media

Thermo Scientific

A1049101

RQ1 RNAse-Free DNAse I

Promega

M6101

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Smart Tube Proteomic Stabilizer SmartTube

Inc.

501351692

SuperAsein RNAse inhibitor

Promega

PRN2111

Ultrafree cenrifugal filter, 0.1-

micron

Sigma UFC30VV25

Wheat Germ Agglutinin

AlexaFluor 647 conjugate

Invitrogen W32466

Table 2. Primary antibodies

Antibody

Vendor

Dilution

Catalogue

RRID

TDP-43 pS409/410

Proteintech

IB: 1:1000

80007-1-RR

AB_2882937

Total TDP-43

Proteintech

IB: 1:2000

TDiP: 40 µg

10782-2-AP

AB_615042

GFP Aves

IB:

1:1000

GFP1202

AB_2734732

GAPDH Millipore

IB:

1:1000 ABS16

AB_10806772

LC3B CST

IB:

1:1000

IF: 1:100

2775S

AB_915950

p62 CST

IB:

1:1000

5114S

AB_10624872

Galectin-3 BioLegend IF:

1:100

126701

AB_1134255

Rab5a

Santa Cruz

IF: 1:100

SC46692

AB_628191

Table 3. Secondary antibodies

Antibody

Vendor

Dilution

Catalogue

HRP Goat anti-Rabbit

Invitrogen

IB: 1: 3000

A31458

HRP Goat anti-Chicken

Invitrogen

IB:1: 3000

A16054

HRP Goat anti-Mouse

Invitrogen

IB:1:1000

A31430

AlexaFluor 594

conjugated goat anti-

Rabbit

Invitrogen IF:

1:500 A11012

AlexaFluor 594

conjugated goat anti-

Mouse

Invitrogen IF:

1:500 A32742

AlexaFluor 647

conjugated goat anti-

Rabbit

Invitrogen IF:

1:500 A32733

Table 4. Suspension mass cytometry probes

Assembly

Isotope

Antibody

Clone

Dilution

MDIPA Kit

89Y CD45 HI30

141Pr CD196/CCR6

G034E3

143Nd CD123 6H6

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

144Nd CD196/CCR6 HIB19

145Nd CD4 RPA-T4

146Nd CD8a RPA-T8

147Sm CD11c Bu15

148Nd CD16 3G8

149Sm CD45RO UCHL1

150Nd CD45RA HI100

151Eu CD161

HP-3G10

152Sm CD194/CCR4 L291H4

153Eu CD25 BC96

154Sm CD27 O323

155Gd CD57 HNK-1

156Gd CD183/CXCR3 G025H7

158Gd CD185/CXCR5 J252D4

160Gd CD28 CD28.2

161Dy CD38 HB-7

163Dy CD56/NCACM

NCAM16.2

164Dy TCRgd B1

166Er CD294 BM16

167Er CD197/CCR7

G043H7

168Er CD14 63D3

170Er CD3 UCHT1

171Yb CD20 2H7

172Yb CD66b G10F5

173Yb HLA-DR LN3

174Yb IgD IA6-2

176Yb CD127 A019D5

103Rh Cell-ID

Intercalator

Expansion Panel

169Tm CD69 FN50

0.125

µL/test

165Ho CD103

Ber-ACT8

0.125

µL/test

159Tb

CD366

F38-2E2

2.0 µL/test

162Dy CD95 DX2

0.125

µL/test

142Nd

CD134

ACT35

1.0 µL/test

209Bi CD137 4B4-1

1.0

µL/test

175Lu CD279

EH12.2H7

0.25

µL/test

198Pt Ki67 Ki-67

0.25

µL/test

159Tb HLA-ABC W6/32

0.25

µL/test

Barcodes

76Se

76SeMal

In-house

Cell: 98 µM

77Se

77SeMal

In-house

Cell: 98 µM

78Se

78SeMal

In-house

Cell: 98 µM

124Te

124TeMal

In-house

Cell: 2 µM

126Te

126TeMal

In-house

Cell: 2 µM

128Te

128TeMal

In-house

Cell: 2 µM

130Te

130TeMal

In-house

Cell: 2 µM

TDiP: 100 µM

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Table 5. Imaging mass cytometry probes

Assembly

Isotope

Antibody

Clone

Dilution

Conjugation

Neural Landscape

150Nd Total

TDP-43 6H6E12 1:250 In-house

159Tb CD68

KP1

1:200

SBT

161Dy pS409/410

(pTDP-43)

1D3

1:100

In-house

164Dy GFAP

GA-5

1:700

In-house

169Tm

Myelin Basic Protein

EPR21188

1:800

In-house

194Pt MAP2 MT-08

1:150

In-house

Immune Landscape

143Nd CD45RA

HI100

1:200

In-house

144Nd CD31 EPR3094

1:50

In-house

147Sm CD163 EDHu-1

1:200 SBT

153Eu Mac2

(Galectin-3) M3/38

1:50

SBT

156Gd CD4 EPR6855

1:200

SBT

162Dy CD8a C8/144B

1:200 SBT

167Er Granzyme

B EPR20129-

217

1:750 SBT

170Er CD3 Polyclonal

1:200

SBT

173Yb CD45RO UCHL1

1:200 SBT

174Yb HLA-DR

LN3 1:50 SBT

191/193Ir DNA

N/A 1:800 SBT

Table 6. Patient demographics for whole-blood analysis

Identifier

Age of

onset

Sex

ALSFRS-

R/time*

Classifier

HIV

Riluzole*

Edaravone*

Sodium

phenylbutyrate

and taurursodiol*

Duke-812

0.60

Standard

no risks

Duke-813

0.15

Slow

no risks

yes

Duke-814

1.00

Standard

negative

Duke-815

0.70

Standard

no risks

yes

Duke-816

0.07

Slow

negative

Duke-817

0.42

Slow

no risks

Duke-818

0.37

Slow

no risks

yes

Duke-819

4.50

Fast

no risks

yes

Duke-824

0.38

Slow

negative

Duke-825

0.20

Slow

negative

Duke-826

1.40

Fast

negative

yes

Duke-827

1.40

Fast

no risks

yes

Duke-828

1.00

Standard

no risks

yes

Duke-829

0.40

Slow

no risks

Duke-830

0.74

Standard

negative

Duke-831

1.20

Fast

negative

Duke-832

2.20

Fast

negative

yes

Duke-835

0.50

Standard

negative

yes

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Figure Legends

Figure 1. TDP-43 aggregates are phagocytosed, trafficked to autophagolysosomes, and promote acute

activation of primary human monocyte-derived macrophages. (A) Immunoblot depicts the dose-

dependent uptake of insoluble, hyper-phosphorylated, GFP-tagged TDP-43a in primary hMDM. (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. (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. (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. Color depicts

data point with respect to donor genotype. N = 3 independent experiments from 3 different genotypes. *P <

0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Ordinary one-Way ANOVA with Tukey’s multiple comparison’s

test. Scale-bar 10-µm. (E) Volcano plots depict differentially expressed intracellular (left) and secreted (middle)

proteomes from primary hMDM cultures stimulated with TDP-43a. Network analyses depict differentially

expressed proteins related to phagocytosis (upper right) and actin remodeling (lower right) pathways.

Performed in triplicate (n = 3); FDR < 0.05; student’s t-test.

Figure 2. TDP-43 aggregates compromise autophagy and promote vesicle rupture. Airyscan and

confocal micrographs (630X) illustrate the extent of TDP-43a co-localization with early endosome marker

Rab5a (A), acidified vesicle marker LysoTracker (B), and ruptured autophagosome markers LC3B and

Galectin-3 (C). (D) Transmission electron micrographs depict whole-cell hMDM (top left; scale-bar 10-µm,

hMDM vesicles following vehicle treatment (upper right; scale-bar 1-µm), and vesicles following TDP-43a

treatment for 1-hour (lower left; scale-bar 0.5-µm) and 16-hours (lower right; scale-bar 1-µm). Representative

vesicles are shaded yellow. (E) Representative immunoblots of LC3-I, LC3-II, and p62 kinetics in hMDM

stimulated with TDP-43a. N = 2 independent experiments. Spearman’s r correlation analysis of LC3B-I, LC3B-

II, and p62 changes after incubation with TDP-43a.

Figure 3. Distinct and specific gene expression changes in primary human monocyte-derived

macrophages challenged with TDP43 aggregates. (A) An MA plot depicts differential analysis of RNA-seq

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

data of cells treated with either TDP-43a or PBS for 12 hours. Differential genes are depicted in blue or red. (B)

A bar plot depicts the number of differential genes detected when comparing TDP-43a, AB, 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

enriched in those genes upregulated in response to 12 hours of TDP43a treatment. (D) Venn diagram

depicting unique and conserved genes that differentially expressed following TDP-43a or

1-42

treatment.

Heat maps depict the magnitude of expression change of each of the 48 conserved genes between TDP-43a

and

1-42

treatments. Performed in duplicate, n = 2.

Figure 4. Global immunophenotypic changes in response to TDP43 aggregates. (A) Schematic of mass

cytometry-based TDP-43a internalization assay, Aggre-Gate. Purified TDP-43a is partially reduced with DTT

and coupled to Tellurium maleimide (

130

TeMal) via thiol-reactive chemistry. Labeled TDP-43a are added to cell

cultures and internalization is 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 cultures of a single genotype. (C)

Linear regression of the percent classical monocytes positive for TDP-43a with increasing incubation time with

TDP-43a. Gating scheme for identification of total monocytes is illustrated in Supplementary Figure 4B. (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-hour stimulation of

bulk PBMC from N = 3 genotypically 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 Supplementary Figure 4B. (G-J) Quantification of relative CD127 expression levels as a readout

of T-cell effector activity following TDP-43a stimulation from N = 3 genetically 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, ****P < 0.0001. Parametric, paired student’s t-

test.

Figure 5. TDP-43 aggregates stimulate antigen presentation and activation of naïve T-cells. (A)

Representative confocal micrographs (200X) of Raji-Jurkat synapses in the presence of isotype control

(vehicle) or TDP-43a purification products, or PHA. Pseudo-colored actin intensity images illustrate intercellular

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

contacts characterized by actin polarization. (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 relative percentages of single and multiple cell contacts per

condition, relative to the number of contacts in an image. Data points represent N = 3 independent passages.

(C) Representative confocal micrographs (left; 200X) of primary hMDM immune synapses in the presence of

isotype control (vehicle) or TDP-43a purification product. 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) illustrates actin-polarized contact between

hMDM and PBMC (asterisk). (D) Schematic of live-cell calcium imaging co-culture assay to monitor naïve T-

cell activation following TDP-43a stimulation of hMDM. (E) Time-series widefield fluorescence micrographs of

intercellular contact between hMDM and naïve CD8 T-cell with intracellular calcium release. Dashed line

denotes hMDM plasma membrane, and arrowheads denote site of intercellular contact. (F) Quantification of T-

cell calcium signaling in both naïve CD4 and CD8 T-cell co-cultures with hMDM. Each data point represents

single-cell calcium signaling events from a 40-minute imaging period within a single field of view. One field of

view per coverslip. Inscribed values denote the total number of coverslips (n) analyzed across all genotypes

(N). All data points originated from N = 3-5 unique genotypes. *P < 0.05, **P < 0.01, ***P < 0.001, ****P <

0.0001. Ordinary one-Way ANOVA with Dunnett’s multiple comparisons test.

Figure 6. Imaging mass cytometry (IMC) profile of immune microenvironments near sites of TDP-43

pathology in ALS motor cortex. (A) Qualitative, representative overlays of the gray (left) and white (right)

matter of motor cortex from a single control and ALS motor cortex tissue section. Insets represent magnified

views of single feature markers for neurons (MAP2), TDP-43 pathology (pTDP-43), astrocytes (GFAP), and

microglia (CD68). Nuclei counter-labeled blue. (B) Overlays and corresponding single-channel images of

activated myeloid cells/microglia expressing antigen presentation machinery (Gal3, CD68, HLA-DR, CD163)

and interacting with CD3+CD8+ T-cells in the parenchyma of ALS white matter. Nuclei counter-labeled blue.

Vasculature marked with dashed line. Arrowheads denote 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.

Figure 7. Whole-blood immune profiling reveals niche immune populations that correlate with ALS

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

progression-rate. (A) Schematic depicts whole-blood profiling strategy using computational and manual

analyses (top). Table summarizes sample sizes and progression rate classifiers determined using the rate of

change of the revised ALS functional rating scale (ALSFRS-R). (B) Annotated tSNE plot denotes immune sub-

populations and associated phenotyping markers of whole-blood leukocytes. (C) VoPo analyses and resulting

per-cluster enrichment scores between standard- and fast-progressing ALS patients. Quantification of manual

gating for NK cell 1 cluster and NKT cell cluster depicts reduced late natural killer cell frequencies (top) and

increased natural killer T-cells (bottom) in fast-progressing patients relative to standard-progressing patients.

Analyzed by Mann-Whitney U test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (D) VoPo analysis and

cluster-specific enrichment scores between standard- and slow-progressing ALS patients. Quantification of

manual gating for dendritic cell cluster and CD8 T-cell cluster 1 depicts increased plasmacytoid dendritic cell

(top) and CD8 TEMRA cell (bottom) frequencies in slow-progressing patients relative to standard-progressing

patients. Analyzed by Mann-Whitney U test. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (E) Principal

component analysis of ALS whole-blood profiles depicts discrete clustering of standard- and non-standard (fast

+ slow) patient whole-blood profiles. (F) VoPo analyses and resulting per-cluster enrichment scores between

slow- and fast-progressing patients (left) and standard- and non-standard-progressing patients (right). Circle

illustrates penetrance of CD8 TEMRA enrichment score in non-standard ALS patients.

Figure 8. Clinical case assessment: TDP-43a whole-blood stimulation induces functional responses

that are dependent on disease progression rate. (A) Normalized ratios of senescent (CD127-CD57+) to

proliferative (CD127+CD57-) CD8 TEMRA T-cells following TDP-43a stimulation of slow-, standard-, and fast-

progressing patient whole-blood cultures (top). Control samples are environmentally matched to each ALS

patient. Mass cytometry contour plots depict frequency changes of proliferative (magenta) and senescent

(green) CD8 TEMRA cells. Non-inferential, N = 1 patient and control pair, per progression-rate category. (B)

Percent difference of CD103+ CD8 TEMRA cells as a readout of tissue-homing potential following TDP-43a

stimulation of slow-, standard-, and fast-progressing patient whole-blood cultures (top). Control samples are

environmentally matched to each ALS patient. Mass cytometry contour plots depict CD103+ CD8 TEMRA cells

(cyan). Non-inferential, N = 1 patient and control pair, per progression-rate category.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

References

1. Brown RH, Al-Chalabi A. Amyotrophic Lateral Sclerosis. N Engl J Med. 2017;377(2):162–172.

2. Huisman MHB, et al. Population based epidemiology of amyotrophic lateral sclerosis using capture-

recapture methodology. J Neurol Neurosurg Psychiatry. 2011;82(10):1165–1170.

3. Wijesekera LC, et al. Natural history and clinical features of the flail arm and flail leg ALS variants.

Neurology. 2009;72(12):1087.

4. Abu-Rumeileh S, et al. Diagnostic-prognostic value and electrophysiological correlates of CSF biomarkers of

neurodegeneration and neuroinflammation in amyotrophic lateral sclerosis. J Neurol. 2020;267(6):1699–1708.

5. Thonhoff JR, Simpson EP, Appel SH. Neuroinflammatory mechanisms in amyotrophic lateral sclerosis

pathogenesis. Curr Opin Neurol. 2018;31(5):635–639.

6. Masrori P, et al. The role of inflammation in neurodegeneration: novel insights into the role of the immune

system in C9orf72 HRE-mediated ALS/FTD. Mol Neurodegener 2022 171. 2022;17(1):1–23.

7. Murdock BJ, et al. NK cells associate with ALS in a sex- and age-dependent manner. JCI Insight.

2021;6(11). https://doi.org/10.1172/JCI.INSIGHT.147129.

8. Campisi L, et al. Clonally expanded CD8 T cells characterize amyotrophic lateral sclerosis-4. Nature.

2022;606(7916):945–952.

9. Choi SJ, et al. High neutrophil-to-lymphocyte ratio predicts short survival duration in amyotrophic lateral

sclerosis. Sci Reports 2020 101. 2020;10(1):1–6.

10. Rolfes L, et al. Amyotrophic lateral sclerosis patients show increased peripheral and intrathecal T-cell

activation. Brain Commun. 2021;3(3). https://doi.org/10.1093/BRAINCOMMS/FCAB157.

11. Cui C, et al. Correlation between leukocyte phenotypes and prognosis of amyotrophic lateral sclerosis.

Elife. 2022;11. https://doi.org/10.7554/ELIFE.74065.

12. McCauley ME, et al. C9orf72 in myeloid cells suppresses STING-induced inflammation. Nature.

2020;585(7823):96–101.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

13. Oakes JA, Davies MC, Collins MO. TBK1: a new player in ALS linking autophagy and neuroinflammation.

Mol Brain. 2017;10(1):1–10.

14. DeJesus-Hernandez M, et al. Expanded GGGGCC hexanucleotide repeat in noncoding region of

C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011;72:245–256.

15. Renton AE, et al. A hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21-

linked ALS-FTD. Neuron. 2011;72:257–268.

16. Deerlin VM Van, et al. TARDBP mutations in amyotrophic lateral sclerosis with TDP-43 neuropathology: a

genetic and histopathological analysis. Lancet Neurol. 2008;7:409–416.

17. Neumann M, et al. Ubiquitinated TDP-43 in fronto-temporal lobar degeneration and amyotrophic lateral

sclerosis. Science (80- ). 2006;314:130–133.

18. Buratti E, Baralle FE. Characterization and Functional Implications of the RNA Binding Properties of

Nuclear Factor TDP-43, a Novel Splicing Regulator of CFTR Exon 9. J Biol Chem. 2001;276(39):36337–

36343.

19. Buratti E, et al. TDP-43 binds heterogeneous nuclear ribonucleoprotein A/B through its C-terminal tail: An

important region for the inhibition of cystic fibrosis transmembrane conductance regulator exon 9 splicing. J

Biol Chem. 2005;280(45):37572–37584.

20. Buratti E, et al. Nuclear factor TDP-43 can affect selected microRNA levels. FEBS J. 2010;277(10):2268–

2281.

21. Colombrita C, et al. TDP-43 and FUS RNA-binding proteins bind distinct sets of cytoplasmic messenger

RNAs and differently regulate their post-transcriptional fate in motoneuron-like cells. J Biol Chem.

2012;287(19):15635–15647.

22. Polymenidou M, et al. Long pre-mRNA depletion and RNA missplicing contribute to neuronal vulnerability

from loss of TDP-43. Nat Neurosci. 2011;14(4):459–468.

23. Neumann M, et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral

sclerosis. Science (80- ). 2006;314(5796):130–133.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

24. Arai T, et al. TDP-43 is a component of ubiquitin-positive tau-negative inclusions in frontotemporal lobar

degeneration and amyotrophic lateral sclerosis. Biochem Biophys Res Commun. 2006;351(3):602–611.

25. Baloh RH. TDP-43: The relationship between protein aggregation and neurodegeneration in amyotrophic

lateral sclerosis and frontotemporal lobar degeneration. FEBS J. 2011;278(19):3539–3549.

26. Evangelista BA, et al. Tandem detergent-extraction and immunoprecipitation of proteinopathy: Scalable

enrichment of ALS-associated TDP-43 aggregates. iScience. 2023;26(5):106645.

27. Yu CH, et al. TDP-43 Triggers Mitochondrial DNA Release via mPTP to Activate cGAS/STING in ALS.

Cell. 2020;183(3):636.

28. Quiroga IY, et al. Synthetic amyloid beta does not induce a robust transcriptional response in innate

immune cell culture systems. J Neuroinflammation. 2022;19(1). https://doi.org/10.1186/S12974-022-02459-1.

29. Budini M, et al. Autophagy and Its Impact on Neurodegenerative Diseases: New Roles for TDP-43 and

C9orf72. Front Mol Neurosci. 2017;10:170.

30. Qian M, Fang X, Wang X. Autophagy and inflammation. Clin Transl Med. 2017;6(1):24.

31. Jiang P, et al. Impaired endo-lysosomal membrane integrity accelerates the seeding progression of

synuclein aggregates. Sci Rep. 2017;7(1):7690.

32. Paz I, et al. Galectin-3, a marker for vacuole lysis by invasive pathogens. Cell Microbiol. 2010;12(4):530–

544.

33. Mizushima N, Yoshimori T. How to interpret LC3 immunoblotting. Autophagy. 2007;3(6):542–545.

34. Wu Y, et al. P62/SQSTM1 accumulation due to degradation inhibition and transcriptional activation plays a

critical role in silica nanoparticle-induced airway inflammation via NF-

B activation. J Nanobiotechnology.

2020;18(1):1–15.

35. Wu H, et al. Expression of KLRG1 and CD127 defines distinct CD8+ subsets that differentially impact

patient outcome in follicular lymphoma. J Immunother cancer. 2021;9(7). https://doi.org/10.1136/JITC-2021-

002662.

36. Lim HW, Kim CH. Loss of IL-7 receptor alpha on CD4+ T cells defines terminally differentiated B cell-

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

helping effector T cells in a B cell-rich lymphoid tissue. J Immunol. 2007;179(11):7448–7456.

37. Benito JM, et al. Down-regulation of interleukin-7 receptor (CD127) in HIV infection is associated with T cell

activation and is a main factor influencing restoration of CD4(+) cells after antiretroviral therapy. J Infect Dis.

2008;198(10):1466–1473.

38. Trebak M, Kinet JP. Calcium signalling in T cells. Nat Rev Immunol 2018 193. 2019;19(3):154–169.

39. Bello-Gamboa A, et al. Imaging the Human Immunological Synapse. J Vis Exp. 2019;2019(154).

https://doi.org/10.3791/60312.

40. Fukumoto R, et al. Inhibition of T-cell receptor signal transduction and viral expression by the linker for

activation of T cells-interacting p12(I) protein of human T-cell leukemia/lymphoma virus type 1. J Virol.

2007;81(17):9088–9099.

41. Killebrew DA, et al. Suppression of HIV-associated Macrophage Activation by a p75 Neurotrophin Receptor

Ligand. J Neuroimmune Pharmacol. 2022;17(1–2):242–260.

42. Fabriek BO, Dijkstra CD, van den Berg TK. The macrophage scavenger receptor CD163. Immunobiology.

2005;210(2–4):153–160.

43. Li B, et al. Comprehensive evaluation of the effects of long-term cryopreservation on peripheral blood

mononuclear cells using flow cytometry. BMC Immunol. 2022;23(1). https://doi.org/10.1186/S12865-022-

00505-4.

44. Jaatinen T, Laine J. Isolation of mononuclear cells from human cord blood by Ficoll-Paque density

gradient. Curr Protoc Stem Cell Biol. 2007;Chapter 2. https://doi.org/10.1002/9780470151808.SC02A01S1.

45. Stanley N, et al. VoPo leverages cellular heterogeneity for predictive modeling of single-cell data. Nat

Commun 2020 111. 2020;11(1):1–9.

46. Smolders J, et al. Tissue-resident memory T cells populate the human brain. Nat Commun 2018 91.

2018;9(1):1–14.

47. Ramachandran S, et al. Low T-cell reactivity to TDP-43 peptides in ALS. Front Immunol. 2023;14.

https://doi.org/10.3389/FIMMU.2023.1193507/FULL.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

48. Pishesha N, Harmand TJ, Ploegh HL. A guide to antigen processing and presentation. Nat Rev Immunol.

2022;22(12):751–764.

49. Passerini L, et al. STAT5-signaling cytokines regulate the expression of FOXP3 in CD4+CD25+ regulatory

T cells and CD4+CD25- effector T cells. Int Immunol. 2008;20(3):421–431.

50. Beers DR, et al. ALS patients’ regulatory T lymphocytes are dysfunctional, and correlate with disease

progression rate and severity. JCI Insight. 2017;2(5). https://doi.org/10.1172/JCI.INSIGHT.89530.

51. Rentzos M, et al. Interleukin-17 and interleukin-23 are elevated in serum and cerebrospinal fluid of patients

with ALS: a reflection of Th17 cells activation? Acta Neurol Scand. 2010;122(6):425–429.

52. Saresella M, et al. T helper-17 activation dominates the immunologic milieu of both amyotrophic lateral

sclerosis and progressive multiple sclerosis. Clin Immunol. 2013;148(1):79–88.

53. Kim CH, Johnston B, Butcher EC. Trafficking machinery of NKT cells: shared and differential chemokine

receptor expression among V

24+V

11+ NKT cell subsets with distinct cytokine-producing capacity. Blood.

2002;100(1):11–16.

54. Semmling V, et al. Alternative cross-priming through CCL17-CCR4-mediated attraction of CTLs toward

NKT cell-licensed DCs. Nat Immunol. 2010;11(4):313–320.

55. Yoshie O, Matsushima K. CCR4 and its ligands: from bench to bedside. Int Immunol. 2015;27(1):11–20.

56. Yazdani S, et al. T cell responses at diagnosis of amyotrophic lateral sclerosis predict disease progression.

Nat Commun. 2022;13(1). https://doi.org/10.1038/S41467-022-34526-9.

57. Yildiz O, et al. Senescent-like Blood Lymphocytes and Disease Progression in Amyotrophic Lateral

Sclerosis. Neurol Neuroimmunol neuroinflammation. 2022;10(1).

https://doi.org/10.1212/NXI.0000000000200042.

58. Rentzos M, et al. Alterations of T cell subsets in ALS: a systemic immune activation? Acta Neurol Scand.

2012;125(4):260–264.

59. Schindelin J, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 2012 97.

2012;9(7):676–682.

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Figure 1

10 µm

Maximum intensity projection

High

intensity

Low

intensity

16 _colors LUT

Optical section

2 µm

TDP-43a

Actin

Actin Intensity

Evangelista et al., 2024

Insoluble

Soluble

TDP-43a (

% v/v

)

pTDP-43

tTDP-43

GAPDH

TDP-43

-75

-50

-37

0.00 0.025 0.05 0.10

Transmitted

Surface

Complete

Internalization

Incomplete Internalization

Global

ARL3

MIF

HDHD1

MCOLN1

CHCHD2

LAIR1

TRAPPC9

TMP1

CTTN

ERAP1

STRAP

CASP1

MIF

CD276

TNFAIP8

CXCL16

PDCD5

Secretome

BRK1

GNG2

SCARF1

Phagocytosis

Actin Remodeling

Log2(TDP/Ctrl)

-Log(P-value)

-75

GFP

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

15 µm

Figure 2

TDP-43a

Evangelista et al., 2024

Airyscan; Optical section

1 µm

Inset

Autophagosome

TDP-43a

LC3

Galectin-3

Airyscan; Optical section

2 µm

p62

Early Endosome

Inset

1 µm

10 µm

Rab5

Inset

Acidified Vesicle

TDP-43a

LysoT
racker

Live-cell; Optical section

5 µm

10 µm

1 µm

0.5 µm

1 µm

hMDM + vehicle

TDP-43a (1.0-hr)

TDP-43a (16-hr)

TDP-GFP

TDP-43

p62

GAPDH

-75

-50

-37

TDP-43a (

)

0.0

4.0

0.5

8.0 16 24

-15

LC3B -I-

LC3B-II-

Insoluble

Soluble

Time (h)

ize

)

p62
LC3B-I
LC3B-II

1.00

0.71

0.49

1.00

0.83

LC3B-I

LC3B-II

p62

LC3B-I

LC3B-II

Spearman r

0 0.5 1.0

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Evangelista et al., 2024

262

(145/117)

712

(417/295)

5940

(3042/2898)

2000

4000

expressed genes

TDP-43a 0.5d

PBS 0.5d

Aβ 6d

PBS 6d

LPS 0.5d

PBS 0.5d

TLR4

TIMP3

TGFBI

JAK1

CCL22

LGALS3

SIGLEC1

STAT5B

TRIM69

CR1

CSF1

BAG6

TIMP1

CD101

−4

−2

100

1000

10000

100000

mean of normalized counts

log2 RNA (TDP-43a/PBS)

Down

ry response to wounding

regulation of metallopeptidase activity

smooth muscle cell migration

regulation of smooth muscle cell migration

cytokine production involved in immune response

regulation of myeloid leukocyte di erentiation

regulation of production of molecular

mediator of immune response

positive regulation of hemopoiesis

positive regulation of leukocyte di erentiation

myeloid leukocyte di erentiation

regulation of myeloid cell di erentiation

immune response–activating cell

surface receptor signaling pathway

regulation of leukocyte di erentiation

regulation of leukocyte cell−cell adhesion

leukocyte cell−cell adhesion

regulation of hemopoiesis

response to salt

regulation of anatomical structure size

positive regulation of cell development

myeloid cell di erentiation

0.02

0.04

0.06

GeneRatio

TDP-43a 0.5d vs PBS 0.5d

p.adjust

0.027

0.026

0.025

0.024

Count

664

214

Aβ

1-42

TDP-43a

Log2(RNA)

Aβ

1-42

(6d)

STAB1

MRC1

SLC7A5

ZNF260

RAB1A

FCGR2C

TNFSF15

RUNX1

SDHD

ENO2

C10orf32

CHD3

ADO

SPTBN1

FAM49A

HSPH1

ABCA1

LRP1

DYNC1H1

ATP6V0C

FKBP1A

SLC15A3

HNRNPD

G6PD

CTNS

DCSTAMP

SPATA18

SLC44A2

LY6E

IFI6

CCL22

SIGLEC1

ECE1

APOC1

DHRS9

CYB561A3

APOBEC3A

CRABP2

DSP

NPR 1

COX5B

MDC 1

SLC25A23

FAIM2

GPR85

NPTX2

DBN1

CYP4F22

-2

-1

TDP-43a (0.5d)

NPTX2

CYP4F22

GPR85

NPR 1

SLC25A23

DSP

SPATA18

DHRS9

DBN1

ECE1

APOC1

SIGLEC1

APOBEC3A

LY6E

COX5B

CRABP2

FAIM2

CCL22

IFI6

ZNF260

G6PD

CYB561A3

C10orf32

CTNS

RAB1A

ADO

HSPH1

DCSTAMP

SLC44A2

FKBP1A

SDHD

SLC15A3

ATP6V0C

ABCA1

DYNC1H1

HNRNPD

LRP1

FAM49A

SPTBN1

ENO2

RUNX1

TNFSF15

CHD3

MRC1

SLC7A5

FCGR2C

STAB1

MDC 1

-2

-1

Figure 3

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

T = 0h

T = 0.5h

T = 4.0h

T = 8.0h

T = 24.0h

CD45 (89Yb)

TDP-43a (130Te)

Monocytes

B-cells

Dendritic cells

NK-cells

CD8 T-cells

Monocytes

Vehicle

TDP-43a

CD45 (89Yb)

TDP-43a (130Te)

Counts

CD127 (176Yb)

CD8 Effector Memory

CD127 (176Yb)

Counts

CD4 Central Memory

CD127 (176Yb)

Counts

CD4 Effector Memory

CD127 (176Yb)

Counts

CD8 Central Memory

0.61%

0.66%

0.46%

0.15%

0.13%

77.13%

62.48%

8.68%

1.15%

0.27%

0.00%

22.26%

11.61%

42.41%

82.24%

TDP-43a (130T

Figure 4

Evangelista et al., 2024

<0.0001

0.0263

0.9978

>0.9999

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Immune synapse

TDP-43a

Actin foci

TCR

Antigen

MHC

hMDM Co-culture

hMDM Monoculture

hMDM

T-cell

CD4

CD8

Vehicle

TDP43a

HIV

Figure 5

Evangelista et al., 2024

100 µm

20 µm

Raji

Jurkat

Inset

Actin Intensity

TDP-43a

PHA

Vehicle

n from N >3 5

Single Contacts
Multiple Contacts

57%

43%

61%

39%

High

intensity

Low

intensity

16 _colors LUT

T= 00:06

T= 01:00

T= 02:30

T= 02:42

T= 03:42

T= 07:22

T= 05:42

T= 08:54

T= 10:54

T= 12:18

T= 14:12

T= 18:42

Widefield

Optical section

TDP-43a

50 µm

Vehicle

10 µm

Actin Intensity

Airyscan; Optical section

No synapse

Synapse

Optical section

Magnification

Actin

PBMC

DAPI

High

intensity

Low

intensity

16 _colors LUT

hMDM

T-cell

High

intensity

Low

intensity

16 _colors LUT

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

B-cell 1

(CD27+IgD+CXCR5+)

B-cell 2

(CD27+IgD+CXCR5-)

Basophils

NK cell 1

(CD57+)

NK cell 2

(CD57-)

Monocyte 1

(CD38-CD14-CD16+)

Monocyte 2

(CD38+CD14+CD16-)

CD4 T-cell 3

(CD25+)

CD8 T-cell 1

(CCR7-CD27-CD45RA+)

TCRγδ T-cell

MAIT/NKT

(CD3+CD56+CD28+CD161+)

CD8 T-cell 2

(CCR7+CD27+CD45RA+)

CD8 T-cell 3

(CCR7-CD27-CD45RO+)

CD4 T-cell 4

(CD45RO+CXCR3+)

CD4 T-cell 5

(CD45RO+CXCR3-)

CD4 T-cell 2

(CCR6-CCR7+CD45RA+CD27+)

CD4 T-cell 1

(CCR6+CCR7+CD45RA+CD27+)

Eosinophils

Neutrophils

Dendritic cell

ALS Whole-blood Immune Profiling

Slow

Standard

Fast

Classifier

ΔALSFRS-R

0 < x < 0.5 0.5 < x < 1.0 1.0 < x

Standard

Slow

Standard vs. Slow

(frequency)

CD8 T-cell 1

Dendritic Cell

Plasmacytoid Dendritic Cell

Percent of CD45+

Standard Slow

Percent of CD8+

CD8 TEMRA

Standard

Non-Standard

Standard vs. Non-standard

(frequency)

Evangelista et al., 2024

Standard vs. Fast

(frequency)

Standard

Fast

NK Cell 1

MAIT/NKT

Late Natural Killer Cell

Percent of CD45+

Standard Fast

MAIT/NKT

Percent of CD45+

Standard Fast

Slow vs. Fast

(frequency)

Slow

Fast

tSNE 1

tSNE 2

Probe 1

Probe 2

tSNE 1

tSNE 2

Figure 7

−25

PC1

PC2

Class

fast

slow

typical

CD8 T-cell 1

(TEMRA)

CD8 T-cell 1

(TEMRA)

The copyright holder for this preprint

this version posted January 10, 2024.

bioRxiv preprint

Control

ALS

Control

ALS

Control

ALS

CD127 (176Yb)

CD57 (155Gd)

CD127 (176Yb)

CD57 (155Gd)

CD127 (176Yb)

CD57 (155Gd)

Slow

Standard

Fast

Control

ALS

Control

ALS

Control

ALS

DMSO

TDP-43a

DMSO

TDP-43a

Senescent TEMRA (Normalized Ratio)

Evangelista et al., 2024

Slow

Standard

Fast

Control

ALS

Control

ALS

Control

ALS

CD103+ TEMRA

(Percent difference)

CD103+ TEMRA

(Percent difference)

CD103+ TEMRA

(Percent difference)

DMSO

TDP-43a

CD127 (176Yb)

CD103 (165Ho)

CD127 (176Yb)

CD103 (165Ho)

CD127 (176Yb)

CD103 (165Ho)

Control

ALS

Control

ALS

Control

ALS

Figure 8

The copyright holder for this preprint

this version posted January 10, 2024.