The mixed Treg EVs were derived from media byproduct of ALS patient Treg expansions or control patient Treg expansions. Leukapheresis products were collected from ALS patients at the Blood Donor Center of Houston Methodist Hospital according to internal procedures or from Gulf Coast Regional Blood Center. For the ALS expanded process prior to their use in their clinical trial, Treg cells were isolated and cultured in the University of Texas Health Science Center at Houston (UTHealth) in the Judith R. Hoffberger Cellular Therapeutics Laboratory that is a FDA-registered, CAP- and FACT-accredited current Good Manufacturing Practice (cGMP) facility. For the control patient expanded Tregs, isolation and expansion protocols were done in the research setting while following the clinical trial expansion protocols and operations to produce a consistent, reproducible, and sterile process. For the cell isolations, patients’ CD4+CD25+ Tregs were isolated from their leukapheresis products through an initial CD8+ and CD19+ cell depletion step (CD8+ and CD19+ depletion reagent, Miltenyi Biotec) followed by a CD25+ cell enrichment step (CD25+ enrichment reagent, Miltenyi Biotec) using the automated CliniMACS Plus System (Miltenyi Biotec). Treg cells from the isolation process were then cultured in ex vivo Treg expansions. Tregs were expanded in TexMACS GMP medium containing 1% human AB serum, MACS GMP ExpAct Treg (CD3/CD28 beads, Bead:cell=4:1), IL-2 (500IU/ml) and Rapamycin (100 nM). Culture of the Tregs was performed in a TerumoBCT Quantum Bioreactor. IL-2 and Rapamycin were replenished every 2-3 days. Restimulation of Tregs was performed by adding MACS GMP ExpAct Treg (CD3/CD28 beads, Bead:cell=1:1) at day 15 only if needed to reach the dose of Tregs necessary for the clinical trial. After 15-25 days in culture, supernatants of Treg cultures were collected and stored at -20 C for EV isolation at a later date. Note that this Treg expansion process utilized a 1% serum supplement during the expansion process which results in a mixture of EVs derived from the Treg cell expansions and the media serum supplement. Control population of media EVs was generated by isolating EVs from the unused TexMACS media with the 1% serum supplement added. Media EVs are used as a control for the differential effects of the Treg EV mixed population being analyzed.

For the experiments using Treg EVs without the serum EVs (enriched Treg EVs), we replicated the smaller-scale expansion protocol used by the lab in previous studies but added the inclusion of exosome-depleted serum supplement instead of the normal serum supplement (36, 38). CD4+CD25+ Tregs were isolated from healthy control blood using the Human CD4+CD25+ Regulatory T Cell Isolation Kit (Miltenyi Biotec) according to the manufacturer’s instructions. Tregs were expanded in TexMACS GMP medium containing 5% exosome-depleted FBS (Gibco), MACS GMP ExpAct Treg (CD3/CD28 beads, Bead:cell=4:1), IL-2 (500IU/ml) and Rapamycin (100 nM). IL-2 and Rapamycin were replenished every 2-3 days. Restimulation of Tregs was performed by adding MACS GMP ExpAct Treg (CD3/CD28 beads, Bead:cell=1:1) at day 15. After 25 days in culture, supernatant from the Treg culture was collected and stored at -20 C for EV isolation at a later date.

Treg EVs were isolated using either the polyethylene glycol precipitation method (PEG) that was initially used for small-scale studies or via tangential flow filtration (TFF) that was later optimized for larger scale EV isolations (50). ExoQuick-TC reagent (System Biosciences, SBI) was used for the PEG isolations and protocol conducted according to manufacturer’s instructions. Briefly, media from Treg expansion cultures were centrifuged at 3000 x g for 15 minutes to remove cells and debris. PEG reagent was added to spun supernatant at 1:5 ratio of PEG: tissue culture media, mixed thoroughly, and refrigerated overnight at 4C. Mixture was then centrifuged at 1500 x g for 30 minutes, supernatant aspirated, spun again at 1500 x g for 10 minutes, and supernatant aspirated again. EV pellet was resuspended in sterile PBS and diluted for nanoparticle analysis using Nanosight NS300 nanoparticle analyzer for Treg EV size/concentration analysis. EVs were stored at -20C while limiting freeze/thaw cycles. For TFF isolations, Treg expansion media was collected and processed through a KrosFlo K2Ri TFF system (Repligen/Spectrum) via a two-step process for the isolation, concentration and diafiltration of Treg EVs. Step 1 utilized a Midi 20cm 0.65um mPES 0.75mm hollow fiber filter (Repligen) for the elimination of any cells, beads, and/or debris from the tissue culture media. Step 1 permeate was used as the input for further processing in Step 2. Step 2 utilized a Midi 20cm 500kD mPES 0.5mm hollow fiber filter (Repligen) for the concentration of Treg EVs in the retentate while soluble material smaller than the filter passed through to the permeate. Following concentration, a 10X diafiltration step took place with the same filter/setup in order to wash and buffer exchange the Treg EVs into sterile PBS solution. Final concentrated Treg EV product is found in the Step 2 retentate, and aliquots of final and intermediate products were obtained for nanoparticle analysis using the Nanosight NS300 nanoparticle analyzer for size distribution, particle concentration, and TFF EV yield.

EV readings were obtained using Nanosight NS300 (Malvern Panalytical) particle analyzer. EV samples were optimally diluted to appropriate concentrations for readings and analyzed using continual measurement at a constant flow of 50 (arbitrary units) with 3 recordings/analyses of 60 seconds each with the following parameters: camera level (12–15), temperature (22C), and detection threshold (5). Concentration was recorded as particles/ml and size statistics were recorded as mean and mode.

Surface protein expression on the EV populations was assessed using the MACSPlex Exosome Kit (Miltenyi Biotec) according to manufacturer’s instructions and analyzed on MACSQuant Analyzer flow cytometer (Miltenyi Biotec). Briefly, EV populations were incubated overnight with a cocktail of various fluorescently labeled bead populations coated with specific antibodies targeting different surface epitopes. Exosome detection reagents are used to form sandwich complexes on the beads that are then analyzed based on their unique fluorescent characteristics. Distinct positive populations can then be measured with the MACSQuant flow cytometer.

Induced pluripotent stem cell (iPSC) derived myeloid cells were used to obtain a consistent inflammatory response independent of any patient-specific confounding factors. Pro-inflammatory iPSC-derived myeloid cells were generated using protocols previously developed/described and recapitulated over time by our lab for multiple studies (36, 38, 39, 51). Briefly, a 4-step culturing process allows for the generation of CD14+ from control patient iPSC lines. CD14 myeloid cells are isolated using positive magnetic selection with Miltenyi Biotec CD14 beads, isolation columns, and MACS magnetic cell separator. For differentiation into M0 macrophages, CD14 cells are cultured in complete RPMI media (10% fetal bovine serum, 25mM HEPES, 1mM sodium pyruvate, 1xnonessential amino acids, 55uM 2-mercaptoethanol, 100 units/ml penicillin, and 100ug/ml streptomycin) supplemented with 50 ng/ml GMCSF (R&D systems) for 7 days to create M0 cells for M1 use. M0 cells are then primed with 0.1ng/ml Lipopolysaccharide (LPS) (Sigma) and 0.2 ng/ml IFN-γ (Invitrogen) in order to polarize myeloid cells to become pro-inflammatory (M1).

M0 (GMCSF) cells are detached using enzyme-free dissociation buffer, pelleted, and plated at a density of 50,000 cells/well in 96 well, flat bottom plates. M1 cells are primed with 0.1ng/ml LPS (Sigma) and 0.2 ng/ml IFN-γ (Invitrogen) for 1 hour to polarize to M1 cells. Treg EVs (1x10⁸ particles) are added into cultures following M1 polarization for overnight time point (~18 hrs) followed by collection of the confitioned media for M1-derived cytokine/protein analysis. Supernatants were collected from co-culture groups and IL-6 protein amounts were assessed using ELISA-based immunoassays (Invitrogen). Pro-inflammatory output is measured in IL-6 production by M1 cells and normalization of suppression in the figures is compared to max output of IL-6 from pro-inflammatory M1 cells without treatment. For Tresp proliferation assays, Tresp cells from an allogeneic control are isolated using Miltenyi Biotec magnetic bead/column reagents and protocols to negatively isolate CD4+CD25- T cells (Tresps) from control patient peripheral blood samples. Tresps are plated at 50,000 cells per well in 96 well, round-bottom plates and stimulated with CD3/28 beads (Miltenyi Biotec). Treg EVs are added to the cultures in escalating doses and remain in Tresp culture through co-culture experiment. Following 4 days in culture, Tresps are pulsed with tritium and proliferation is determined by examining tritium incorporation 18 hours after tritium pulsing. Results of the assay are reported as the percent inhibition of Tresp proliferation compared to Tresps that are activated but not treated with EVs.

All animal experimental procedures were approved by the Houston Methodist Research Institute’s Institutional Animal Care and Use Committee in compliance with the National Institutes of Health guidelines. All mice were monitored for pain, distress, and adverse effects from EV treatments. For the acute LPS-induced neuroinflammation mouse model, C57Bl6 WT mice were injected intraperitoneally (IP) with 2mg/kg LPS (Sigma; O111:B4) followed by peripheral (IV) or intranasal administration of various Treg EV doses after 2 hours post-LPS injection. IV administration of Treg EVs were dosed with either 1x10⁹, 1x10¹⁰, or 1x10¹¹ particles via tail vein injection and intranasal administration of enriched Treg EVs were given 1x10⁹ particles. Dosing paradigms were within effective dose ranges previously reported (52). Mice were then sacrificed at 14-16 hours post model initiation followed by removal of organs for RNA analysis, specifically spleens for further immune cell isolation/analysis and different regions of the brain. Immune cells were isolated from freshly isolated spleens by extracting single cell suspension using a 40um cell strainer followed by magnetic, bead-based isolation techniques for myeloid cells (CD11b+ isolation kit; Miltenyi Biotec) and T cell populations (CD4+CD25+ isolation kit; Miltenyi Biotec). Isolated immune cells were flash frozen for subsequent RNA isolation and transcript analysis using RT-PCR. For CNS tissue, mouse brains were freshly isolated and different regions of the brain were dissected on ice and flash frozen for RNA isolation and transcript analysis using RT-PCR.

For the ALS mouse model, we utilized transgenic mice harboring the SOD1-G93A mutation that previously was described as a motor neuron degeneration model for ALS (53). SOD1 mice began phenotype assessments starting at day 70 followed by intranasal injections of enriched Treg EVs (1x10⁹ particles) beginning at day 90 at continuous intervals of every two weeks until they reached their ethically defined endpoint. Mouse phenotype was assessed using a modified “BASH scoring system” whereby SOD1 mice gain a degenerative point from 0 (no symptoms) to 6 (ethical endpoint) as phenotype worsens with disease progression (54–56). The phenotypes assessed and points added are as follows but not necessarily in this order: +1 Tremulousness, +1 Gait abnormalities, +1 Hindlimb weakness/paresis, +1 Weight loss of more than 10% adult weight, +1 Spasticity to one or both hindlimbs, +1 Paralysis. When animals reach ethical endpoint, they are sacrificed and organs are harvested, RNA isolated, and RT-PCR done for transcript analysis. Specifically, diseased lumbar spinal cords were isolated from mice to examine pro-inflammatory transcripts using RT-PCR. Dosing information prior to sacrifice were as follows: Number of doses (PBS 4, 5, 5, 5; Treg EV 6, 4, 5, 5, 10, 10); Average doses (PBS 4.75; Treg EV 6.67); Median doses (PBS 5; Treg EV 5.5). Follow up days from first treatment to sacrifice: Average days (PBS 61.75; Treg EV 72.83); Median days (PBS 61.5; Treg EV 74.5); Minimum days (PBS 57; Treg EV 51); Maximum days (PBS 67; Treg EV 84).

RNA was isolated from cells and tissues using Trizol reagent followed by Direct-zol RNA MiniPrep Plus Kit (Zymo Research) according to manufacturer’s recommendations. Quantitative RT-PCR was performed using one-step RT-PCR kit with SYBR Green (Bio-Rad) and an iQ5 Multicolor Real-Time PCR detection system (Bio-Rad). Primers for RT-PCR (IL-6, IL-1β, TNF, IL-10, Arg-1, IFN-γ, FOXP3, and CD206) were acquired from Bio-Rad and run according to the manufacturer’s protocols. The relative expression level of each mRNA was assessed using the ΔΔCt method and normalized to β-actin/controls.

Levels of CD73 (Abcam) and CTLA4 (Abcam) protein were measured by ELISA-based immunoassays according to manufacturer’s protocols. From the Treg EV and media EV preparations, a volume equivalent of 10ug of protein was added as input to each well of the ELISAs in triplicate for consistent protein normalization across samples. Treg cell protein from expanded Treg cells was also added in the same amount to directly compare to cell amounts. Following incubation protocols and washes, the immunoassay absorbances were read using a microplate plate reader at the assay’s specified wavelength.

Western blot analysis was used to identify Treg-related proteins in protein preparations from Treg EVs, medias EVs, and Treg cells. A volume equivalent of 30ug of protein was loaded as input for each well from the respective EV or cell sample onto a polyacrylamide gel (Biorad) and run at a voltage of 70 V for 30min and then changing to 120V for about an hour. Blots were transferred to nitrocellulose membrane, rinsed with Tris-buffered saline with 0.1% Tween (3X), and blocked with 6% powdered milk for 2 hours at room temperature. Subsequent steps required blots be incubated with 1:1000 dilution of either anti-CTLA4 (Abcam) or ICOS (Abcam) primary antibodies overnight at 4C followed by 1:2000 dilution of goat anti-rabbit secondary antibody at room temperature for 2 hours. Blots were then developed and exposed at various times to reveal optimal protein banding. Afterward, membranes were exposed to Ponceau S stain (ThermoFisher) for visual total protein normalization of the induvial sample lanes. Same protocols were used for negative control markers for EV isolation but with 1:500 dilution anti-Calnexin (ThermoFisher) or 1:1000 dilution of anti-GM130 (Abcam) followed by 1:2000 dilution goat anti-mouse secondary or goat anti-rabbit secondary, respectively.

Flow cytometry was used to assess Treg cells following isolation from leukapheresis product and following expansion protocols. Antibodies (anti-human) against the following targets were used in the staining and analysis: CD3 Alexa Fluor 700 (Invitrogen), CD4 V500 (BD Biosciences), CD8 eFluor 450 (Invitrogen), CD25 PerCP-Cy 5.5 (BD Biosciences), FOXP3 Alexa Fluor 488 (Invitrogen). Appropriate isotype controls were set for gating schemes and to establish background parameters. Live/Dead fixable blue dead cell UV stain kit was used to assess viability of the cells. For intracellular FOXP3 staining, cells were fixed and permeabilized using the FOXP3/Transcription Factor Staining Buffer Set (BD Biosciences) prior to staining. Cells were subsequently analyzed using a BD LSRII flow cytometer with BD FACSDIVA software.

The isolation and ex vivo expansion of patient Treg cells generate an enriched CD4+CD25+ population of highly suppressive Treg cells ( Table 1 ). Specifically, isolation protocols using the CliniMACS system initially depleted CD8+ and CD19+ cells followed by a CD25+ enrichment step that produced a CD4+CD25+ population that is greater than or equal to 70% (% of total CD4 population). Ex vivo expansion of these CD4+CD25+ cells results in a highly pure population of Treg cells typically reflected by a CD4+CD25+ population greater than or equal to 95% (% of total CD4+ population). Additionally, potential contaminating cell populations such as CD8+ and CD19+ cell populations remained below 0.07% when analyzed following a depletion step and below 1.5% following an enrichment step of the isolation ( Supplementary Table 1 ). In addition to the high purity, we find that ex vivo expanded Treg cells are highly functional in their ability to suppress T cell proliferation with greater than 88% suppression of proliferation across patients at a 1:1 ratio (Treg: Tresp) ( Table 1 ). These cells also express increased amounts of both CD25 and FOXP3 protein noted via flow cytometry analysis of MFI levels from CD4+CD25+ cells. The isolation and expansion protocols provide an enriched CD4+CD25+ suppressive Treg cell population from which EVs can be isolated.

During the expansion process of the CD4+CD25+ cells following isolation, EV-enriched media is generated and collected as a byproduct. EVs from the byproduct were isolated, initially via polyethylene glycol (PEG) precipitation and later by tangential flow filtration (TFF) for scale, and analyzed using Nanosight nanoparticle analysis. The nanoparticle analysis of the EVs isolated during the expansion of diverse Treg populations demonstrated a consistent size distribution between 50nm and 150nm (Mean= 94.5nm, Mode= 76.8nm, D10 = 56.6nm, D50(Median)= 86.4nm, D90 = 146.9nm) ( Figure 1A ). Scanning electron microscopy confirmed Treg vesicle size and shape ( Figure 1B ). These Treg EVs were found to express the combination of confirmed vesicle surface markers of CD9, CD63, and CD81; media EVs only expressed the CD9 vesicle marker ( Figure 1C ). Additionally, these Treg EVs were positive for surface markers of CD2, CD4, CD25, CD44, CD29, CD45, and HLA-DRDPDP; media EVs did not express any of these markers ( Figure 1D ). Treg EVs derived from expansion of both ALS patients and controls show the same vesicle and Treg-conserved markers when run using the MACSPlex exosome assay ( Supplementary Figures 1C, D ). Characterization of the Treg EVs demonstrate a size distribution consistent with an EV definition along with unique markers that are differentially expressed compared to the media EVs. Along with characterization of the EVs, purity of the EV isolations were assessed using suggested markers of contamination that are not of plasma membrane or endosomal origin, specifically GM130 and calnexin ( Supplementary Figure 3 ) (57, 58). Treg EVs from ALS patients and controls along with media EVs did not express either GM130 or calnexin while the Treg cells did suggesting purity of the EV populations.

In vitro co-culture of Treg EVs with pro-inflammatory myeloid cells (M1) and T cells were performed to test suppressive capacity. The Treg EVs suppressed pro-inflammatory iPSC-derived M1 cell IL-6 protein production at 1x10⁸ particles per 50,000 activated M1 cells; media EVs used as the differential control demonstrated very little suppressive function ( Figure 1E ). Treg EVs in increased doses were added to responder T cells (Tresp) resulting in a dose-dependent inhibition of Tresp proliferation. At a dose of 5x10⁶ particles, suppression was 65% and suppression increased to 82% at a dose of 1x10⁷ ( Figure 1F ). At higher doses than 1x10⁷ EVs, we found increased media EV suppression that looks to be a culturing artifact; potentially derived from large quantities of EVs disrupting Tresp and CD3/CD28 bead interactions in the smaller wells of the 96-well plate that the assay is optimized for. Differential isolation techniques using either PEG or TFF isolations have no consequence on Treg EV suppressive function ( Supplementary Figures 1A, B ). Additionally, a direct suppressive functional comparison between Treg EVs mesenchymal stem cell EVs (MSC EVs) demonstrated that Treg EVs are more suppressive in co-culture with both pro-inflammatory myeloid cells and Tresp proliferation ( Supplementary Figures 1E, F ). Overall, Treg EVs derived from ex vivo expanded Treg cells demonstrated a unique and Treg-conserved signature along with suppressive function in vitro that was comparable to expanded Treg cells.

Treg EVs were administered via tail vein injection (IV) at different doses in mice following 2 mg/kg IP LPS injection to induce inflammation ( Figure 2A ). Following overnight time point after 14-16 hours, mice were sacrificed, and spleens dissected to isolate immune cells including CD11b+ myeloid cells, CD4+ CD25+ Treg cells, and CD4+ CD25- effector T cells. Transcript analysis of spleen-derived CD11b+ myeloid cells showed an induction of pro-inflammatory transcripts such as IL-6, iNOS, IL-1β, and IFN-γ following LPS-induced inflammation induction in Treg EV untreated mice ( Figure 2B ). Increasing peripheral doses of Treg EVs demonstrated a corresponding reduction in spleen-derived myeloid pro-inflammatory IL-6 transcripts at doses of 1x10¹⁰ (61%) and 1x10¹¹ (75%) while also reducing iNOS transcripts in these same cells at doses of 1x10¹⁰ (64%) and 1x10¹¹ (85%). A decreasing trend in IL-1β and IFN-y transcripts was observed with the best results at the higher doses administered. Anti-inflammatory transcripts were assayed from these myeloid cells following LPS activation and subsequent Treg EV peripheral treatment. Additionally, Treg EVs at a dose of 1x10¹¹ particles induced the production of anti-inflammatory transcripts such as MRC1 (mannose receptor/CD206) and CD163 in these pro-inflammatory myeloid cells compared with LPS injected only animals ( Figure 2C ).

T cell populations were also isolated from the spleens of these animals to assess the immune-modulating effects of the Treg EVs on CD4+CD25+ Treg populations and CD4+CD25- effector T cell (Teff) populations. The Treg population demonstrated a decrease in the FOXP3 expression following LPS-induced inflammation. Treatment with Treg EVs increased FOXP3 transcript levels in a dose-dependent fashion with an increase in expression at the 1x10¹¹ dose ( Figure 2D ). Additionally, an increase in IL2RA transcripts (CD25) was observed at the highest Treg EV doses. The CD4+CD25- Teff cells did not exhibit activation in multiple inflammatory transcripts investigated at this time point including TNF-α and IFN-y. With respect to the Treg EV treatments, there were no observable inflammatory or anti-inflammatory immune transcript alterations or cell proliferation in the Teff cell population following Treg EV treatment.

Since IV administration of Treg EVs modulates peripheral immune cell signatures in the LPS-induced model of inflammation, the extent of neuroinflammatory immune modulation in inflamed mice following IV administration of Treg EVs was evaluated. In the same treated animals, brain regions were dissected, and transcript analysis was performed to assess the potential immunomodulation of peripheral Treg EVs in reducing neuroinflammation ( Figure 3A ). In the hippocampus of the affected mice, IV delivered Treg EVs produced only a modest reduction in pro-inflammatory transcripts of IL-6 while IL-1β had a mixed result; both cytokines were reduced at the highest dose of 1x10¹¹ particles ( Figure 3B ). In the cortex, there was a trend for a dose-dependent suppression of IL-6 and IL-1β ( Figure 3C ). TNF-α transcripts in both regions were not elevated in LPS-induced animals.

Tregs were expanded utilizing an exosome-depleted serum to obtain a population of enriched Treg EVs. Mice were injected with LPS peripherally to induce peripheral and central nervous system inflammation followed by intranasal administration of 1x10⁹ enriched Treg EVs ( Figure 4A ). Enriched Treg EVs suppressed hippocampal IL-6 and IL-1β transcripts generated by the LPS injections; there was no change in TNF-α transcripts following treatment ( Figure 4B ). In examining the cortex, there was a treatment-specific reduction in IL-6 transcripts while IL-1β and TNF-α transcripts remained elevated ( Figure 4C ). Modulation of peripheral inflammation was assayed by analyzing inflammatory transcript changes in splenic-derived CD11b+ myeloid cells. Interestingly, there was a robust decrease in myeloid IL-6 and TNF-α transcripts following intranasal enriched Treg EV treatment ( Figure 4D ). These results suggest an enhanced immune-modulating response of neuroinflammation following treatment with enriched Treg EVs given intranasally.

An ALS mouse model for motor neuron degeneration and the accompanying inflammatory responses was used to examine the effects of chronic intranasal administration of enriched Treg EVs on reducing neuroinflammation and associated disease. Intranasal treatments of 1x10⁹ particles of enriched Treg EVs were initiated at day 90 when the animals were already showing signs of motor neuron disease ( Figure 5A ). Enriched Treg EV treatment significantly increased the survival in the treated group compared to PBS intranasally administered controls ( Figure 5B ). Additionally, we observed that the enriched Treg EV treatments slowed disease progression in the later, more rapid progressing stages of disease (54–56) ( Figure 5C ). When examining the disease duration from first symptom to end stage disease, enriched Treg EVs increased disease duration (symptom onset to sacrifice) to 85 days compared with disease duration of 69 days in the PBS treated mice ( Figure 5D ). Average lifespan of the treated animals was increased in the enriched Treg EV treated animals compared to PBS controls at 162.8 days vs 151.7 days, respectively (p=0.09) ( Figure 5E ). Additionally, follow-up time from first treatment to sacrifice increased in the Treg EV treated animals compared to controls (average days: PBS 61.75, Treg EV 72.83; median days: PBS 61.5, Treg EV 74.5; minimum days: PBS 57, Treg EV 51; maximum days: PBS 67, Treg EV 84). With respect to overall intranasal doses administered, mice treated with PBS ended with 4 to 5 intranasal doses with an average of 4.75 doses while the Treg EV treated group ultimately received 4 to 10 intranasal doses with an average dose of 6.67 in their group. Following animal end stage disease and subsequent sacrifice, we extracted RNA from the lumbar portions of their diseased spinal cord to evaluate treatment-associated inflammatory changes via transcript analysis ( Figure 5F ). Examination of pro-inflammatory transcripts found decreases in TNF-α and IFN-y transcripts in lumbar spinal cords of enriched Treg EV treated animals while IL-6 and IL-1β transcripts trended in the same decreasing direction ( Figure 5F ). Additionally, levels of FOXP3 transcripts were increased with Treg EV treatment; myeloid-specific CD206 transcripts were also increased in Treg EV treated ALS mice compared with PBS treated mice. Together, these data show that intranasal Treg EVs can increase survival, ameliorate motor neuron disease progression, decrease pro-inflammatory mechanisms in the spinal cord, and increase beneficial anti-inflammatory signatures indicative of increased Treg and anti-inflammatory myeloid cell contributions.