Regulatory T Cell Activation, Proliferation, and Reprogramming Induced by Extracellular Vesicles

Abstract

Background:

Extracellular vesicles (EVs) from heart stromal/progenitor cells modulate innate immunity, with salutary effects in a variety of cardiac disease models. Little is known, however, about the effects of these EVs on adaptive immunity.

Methods:

Ex vivo differentiation of naïve CD4⁺ T cells was conducted to assess the effect of EVs on cytokine production and proliferation of Th1, Th2, Th17, and regulatory T (T_reg) cells. These effects were further tested in vivo using the experimental autoimmune myocarditis (EAM) model.

Results:

Using differentiated CD4⁺ T cells, we show that EVs secreted by human-derived heart stromal/progenitor cells selectively influence the phenotype, activity, and proliferation of regulatory T (T_reg) cells. Exposure of T_reg cells to EVs results in faster proliferation, augmented production of IL-10, and polarization toward an intermediate FOXP3⁺RORγt⁺ phenotype. In experimental autoimmune myocarditis, EVs attenuate cardiac inflammation and functional decline, in association with increased numbers of splenic IL10⁺-T_reg cells.

Conclusions:

T cell modulation by EVs represents a novel therapeutic approach to inflammation, harnessing endogenous immunosuppressive mechanisms that may be applied in solid organ transplantation, graft-versus-host disease, and autoimmune disorders.

Graphical Abstract

Introduction

CD4⁺ T cells are essential mediators of adaptive immunity, capable of mounting inflammatory responses against microbial pathogens, maintaining surveillance against tumors, and establishing long-term immunologic memory. During initial activation, naïve CD4⁺ T cells can differentiate into various effector T cells (e.g., Th1, Th2, Th17), depending on the cytokine milieu and expression of lineage-specific transcription factors.¹ In addition to these conventional effector subtypes, CD4⁺ T cells can differentiate into regulatory T (T_reg) cells, which suppress various immune responses and represent potential therapeutic agents in solid organ transplantation, graft-versus-host disease, and autoimmune disorders.² Conventional effector T cells and T_reg cells are defined by their cytokine profiles and function, but there is substantial plasticity, such that a particular cell type can switch its cytokine profile and phenotype.³ Equally important is the ability of T_reg cells to adopt features that overlap with effector T cells (e.g., Th17-like T_reg cell), which enables tailored, cell-specific suppression.⁴

Environmental cues play an important role in differentiation, function, and plasticity of CD4⁺ T cells. Extracellular vesicles (EVs) can influence cells of adaptive immunity,^{5, 6} but mechanistic details remain incomplete. Given their disease-modifying bioactivity in multiple inflammatory disease models,^{7, 8} EVs secreted by heart stromal/progenitor cells (hEVs) merit particular attention in terms of innate and adaptive immunity. Within innate immunity, hEVs have already demonstrated immunomodulatory properties by polarizing macrophages toward an anti-inflammatory, pro-reparative phenotype.^9–12 These phenotypic changes led to cardioprotective effects in animal models of myocardial ischemia. Little is known, however, about the effects of hEVs in adaptive immunity. Here, we demonstrate that hEVs increase proliferation of T_reg cells, augment their production of IL-10, and reprogram T_reg cells into an intermediate FOXP3⁺RORγt⁺ phenotype with suppressive capacity.

Materials and Methods

Detailed methods are provided in the Supplementary Material. All datasets used for this study are available upon reasonable request.

Ex vivo T cell experiments

Ex vivo differentiation of naïve CD4⁺ T cells was conducted to assess the effect of hEVs on phenotype and behavior of conventional effector CD4⁺ T cells (Th1, Th2, Th17) and T_reg cells. Naïve CD4⁺ T cells were isolated from FOXP3-IRES-mRFP mice (The Jackson Laboratory), polarized, and co-cultured with either hEVs or vehicle. Rates of proliferation, cytokine production, and the global transcriptome were analyzed.

Experimental autoimmune myocarditis

Experimental autoimmune myocarditis (EAM) was induced in female Lewis rats (Charles River) and female BALB/c mice (Taconic) using purified cardiac myosin and αMHC peptide, respectively, emulsified in complete Freund’s adjuvant. Animals were injected with either hEVs or vehicle on day 10 and were euthanized on days 21–28 for functional and immunologic analyses. All animal protocols were approved by the local Institutional Animal Care and Use Committee (IACUC: 008416, 005460, 007407).

Extracellular vesicles

hEVs used in the ex vivo T cell experiments and mouse EAM experiments were isolated from heart stromal/progenitor cells grown from human endomyocardial biopsies, following approval by the local Institutional Review Board (IRB: 00014713). hEVs used in rat EAM experiments were isolated from rat cardiac cells, to facilitate direct comparison with published work using these cells in rat EAM.¹³ Regardless, hEVs from heart stromal/progenitor cells of different species appear to be roughly bioequivalent^14–17 and work well in xenogeneic applications.^{9, 10, 18}

Results

Heart-derived EVs (hEVs) influence T cell proliferation in a cell-specific and dose-dependent manner

Naïve CD4⁺ T cells were isolated from mice (Figure 1A,B) and differentiated into conventional effector T cells (Th1, Th2, Th17) or induced regulatory T (iT_reg) cells (Figure S1A,B). After 5 days of culture, we determined the effects of varying doses of human hEVs (or vehicle) on proliferation rates of each T cell population. hEVs were isolated from media conditioned with heart-derived stromal/progenitor cells for 15 days and had a mean and mode size of ~130 nm and ~120 nm, respectively (Figure 1C,D). Characterization of hEVs demonstrated expression of canonical EV markers: CD9, CD63, and CD81 (Figure 1E). Compared to controls, Th1 and Th2 cells exposed to hEVs had similar rates of proliferation, while Th17 cells had a 7–11% reduction in proliferation (Figure 1F,G). In contrast, iT_reg cells exposed to hEVs (iT_reg-hEV) proliferated at higher rates compared to iT_reg cells exposed only to vehicle (iT_reg-Ctrl) (Figure 1F,G). At a cell-to-EV ratio of 1:1000, hEVs elicited a strong proliferative response (iT_reg-hEV 75.3 ± 0.6% versus iT_reg-CTL 62.5 ± 1.5%, p < 0.0001) (Figure 1H, Figure S1C-F). Notably, exposure to hEVs resulted in distinct, dose-dependent responses among the different types of CD4⁺ T cells: no change in proliferation of Th1 and Th2 cells, decrease in proliferation of Th17 cells, and increase in proliferation of iT_reg cells. Thus, the effects of hEVs on iT_reg cells appear to be specific.

Figure 1. Dose-dependent and cell-specific response in proliferation after hEV exposure.

A. Experimental outline for in vitro T cell analysis: isolation of the total CD4⁺ T cell population from FOXP3 reporter mice, enrichment of naïve CD4⁺ T cells, and subsequent differentiation into conventional effector T cells (Th1, Th2, Th17) and induced regulatory T (iT_reg) cells. Following differentiation, cells were cultured with either vehicle or human hEVs. B. Sorting strategy for naïve CD4⁺ T cells (red gate). C. Particle concentration and mean and mode sizes of hEVs by nanoparticle tracking analysis. D. Transmission electron microscopy images of hEVs. Bars, 200 nm. E. Analysis of EV markers, including CD9, CD63, and CD81, by western blot (n=1/group). F. Proliferation of effector T cells and iT_reg cells after exposure to vehicle or varying doses of hEVs (ratio expressed as cell-to-EV). Top two rows are light microscopy images of the cells. Bars, 100 μm. Bottom rows represent quantification by flow cytometry using dilution of the proliferation dye, CellTrace Violet (CTV). G. Quantification of the change in the rate of proliferation with increasing doses of hEVs relative to vehicle control. %Δ in proliferation = [(proliferation with hEV – proliferation with vehicle)/proliferation with vehicle] x 100. H. Proliferation of iT_reg cells following exposure to hEVs (cell-to-EV ratio of 1:1000) and vehicle control (****P < 0.0001, n=6–7/group, two-tailed Student’s t-test). Data are mean ± SEM.

iT_reg cells primed with hEVs, but not fibroblast EVs, produce higher levels of IL-10

To see if phenotypic changes accompany the observed changes in proliferation, we stimulated effector T cells and iT_reg cells after 5 days of culture and analyzed their signature cytokines. Compared to vehicle, human hEVs had little effect on the expression of characteristic cytokines by conventional effector T cells (Figure 2A, Figure S2A). In contrast, exposure of iT_reg cells to hEVs resulted in ~1000-fold increase in expression of Il10, relative to vehicle (Figure 2B, Figure S2B). Expression of other iT_reg cell markers was either unchanged (Ctla4, Icos, and Tgfb1) or decreased (Il12a and Ebi3), suggesting that hEVs target specific pathways, rather than indiscriminately activating iT_reg cells. At the protein level, analysis of conditioned media revealed much greater secretion of IL-10 by iT_reg cells that had been exposed to hEVs (Figure 2C). To determine whether the effects on Il10 expression were specific to hEVs, we compared iT_reg cell responses to hEVs with those to human dermal fibroblast EVs, which are similar in size distribution (Figure S2C) but functionally inert in other assay systems.⁸ Indeed, iT_reg cells primed with hEVs outperformed iT_reg cells primed with fibroblast EVs in terms of Il10 expression (Figure 2D). Enhanced IL-10 production was evident 5 days after hEV exposure, but not two days earlier (Figure 2E). Thus, hEVs (but not fibroblast EVs) augment production of IL-10 by iT_reg cells and increase their proliferation, while little effect is seen across effector T cells.

Figure 2. hEVs augment production of IL-10 by iT_reg cells.

A. Expression of signature cytokines by human hEV-primed effector T cells, including Ifng and Tnfa by Th1 cells, Il4 and Il13 by Th2 cells, and Il17a and Il17f by Th17 cells, relative to vehicle-primed cells (n=3–5/group). Data are mean ± SD. B. Quantification of T_reg markers and cytokines expressed by iT_reg-hEV relative to iT_reg-Ctrl (n=5–8/marker). Data are mean ± SD. C. Protein concentration of IL-10 quantified by ELISA in the culture media of iT_reg-hEV and iT_reg-Ctrl (****P < 0.0001, n = 5/group, two-tailed Student’s t-test). Data are mean ± SEM. D. Expression of Il10 by iT_reg cells on day 5 following exposure to hEVs or human dermal fibroblast EVs (fbEVs), relative to iT_reg-Ctrl (****P < 0.0001, n = 6–10/group, two-tailed Student’s t-test). Data are mean ± SEM. Gene expression data analyzed using the 2^−ΔΔCt method after normalization against Gapdh expression and vehicle controls. E. Representative flow cytometry plots and quantification of intracellular IL-10 staining, as percent of total iT_reg cells on days 3 and 5 (ns = 0.1539 between iT_reg-Ctrl and iT_reg-hEV on day 3; ****P < 0.0001 between iT_reg-Ctrl on day 3 and iT_reg-Ctrl on day 5; ****P < 0.0001 between iT_reg-hEV on day 3 and iT_reg-hEV on day 5; ****P < 0.0001 between iT_reg-Ctrl on day 5 and iT_reg-hEV on day 5; n=5/group, two-way ANOVA with Sidak’s multiple comparisons test). Data are mean ± SEM.

Infusion of hEVs attenuates disease severity and expands the IL10⁺-T_reg cell population in a model of myocarditis

To test whether hEVs affect the T_reg cell population and IL-10 production in vivo, we investigated experimental autoimmune myocarditis (EAM), a myosin-induced and T-cell driven model of myocarditis.¹⁹ Heart-derived stromal/progenitor cells have previously been shown to attenuate inflammation and preserve cardiac function in a rat model of EAM;¹³ we found similar effects with hEVs derived from rat hearts. Following induction of disease (Figure 3A,B), rats receiving hEVs exhibited improved left ventricular ejection fraction (Figure 3C) and less cardiac inflammation (Figure 3D). Although we did not characterize the inflammatory infiltrate in our study, rats with EAM receiving the parent stromal/progenitor cells have been shown to exhibit decreased CD3⁺ lymphocyte infiltrates.¹³

Figure 3. hEVs from rats attenuate left ventricular functional decline and inflammation in experimental autoimmune myocarditis.

A. Timeline of the experimental autoimmune myocarditis (EAM) model in rats. On day 10, rats received either a vehicle infusion (EAM+vehicle) or hEVs (EAM+hEV). B. Representative immunofluorescence images of cardiac tissue in healthy controls (Ctrl) and diseased animals (EAM). Bar, 25 μm for Ctrl; 75 μm for EAM. C. Representative echocardiographic M-mode images in short axis on day 28 with quantification of left ventricular function (LVEF), left ventricular end systolic diameter (LVESD), and left ventricular end diastolic diameter (LVEDD). (LVEF: ****P < 0.0001 between Ctrl and EAM+vehicle on day 28; **P = 0.0032 between EAM+vehicle and EAM+hEV on day 28. LVESD: ****P<0.0001 between Ctrl and EAM+vehicle; ns = 0.0735 between EAM+vehicle and EAM+hEV on day 28. LVEDD: *P = 0.0237 between Ctrl and EAM+vehicle; ns = 0.9994 between EAM+vehicle and EAM+hEV on day 28; n = 10–12/group, two-way ANOVA with Tukey’s multiple comparisons test). Data are mean ± SEM. D. Representative cardiac cross sections using hematoxylin and eosin staining and a semiquantitative assessment of inflammation. Bars, 200 μm. Data are individual scores with the median denoted by a bar.

In EAM mice, where T cell populations can be better-sorted than in rats,^{20, 21} we tested whether human hEVs can expand the overall T_reg cell population and IL-10 production, as our ex vivo experiments have shown. Human-derived, rather than murine-derived, cells were used as the EV source to explore the translational value of hEVs. Mice received hEVs on day 10 after induction of disease and were euthanized on day 21 (Figure 4A), a time point that corresponds with peak inflammation.²² Animals receiving hEVs weighed slightly more overall, but splenic sizes and overall splenocyte counts were similar in both groups (Figure 4B,C). Among splenocytes, there were no significant differences in CD4⁺ cells or CD4⁺FOXP3⁺ cells (Figure 4D-F), but the number of CD4⁺FOXP3⁺IL10⁺ cells was significantly higher in animals that received hEVs, relative to vehicle controls (Figure 4G). Within the hearts of EAM mice, transcription levels of Foxp3 were increased by ~2 to 3-fold in animals that received hEVs, while transcription levels of Il10 were similar (Figure S3). Thus, hEV augment IL-10-producing T_reg cells in vitro and within mouse spleens in vivo.

Figure 4. Increased number of IL-10⁺ T_reg cells within spleens of mice with experimental autoimmune myocarditis following hEV infusion.

A. Experimental design for autoimmune myocarditis, which involves induction of disease on days 0 and 7, and intravenous injection of human hEVs or vehicle on day 10. B. Total mass on day 0 (baseline) and day 21 (end point) and spleen mass on day 21 of animals in the vehicle control and hEV groups (ns = 0.2019 between control and hEV groups for animal mass on day 0; **P = 0.0097 between control and hEV groups for animal mass on day 21; ns = 0.2019 between control and hEV groups for spleen mass on day 21; n=8/group, two-tailed Student’s t-test). Data are mean ± SEM. C. Total number of splenocytes per mg of spleen tissue (ns = 0.9201 between control and hEV groups; n=8/group, two-tailed Student’s t-test). Data are mean ± SEM. D. Representative flow cytometry plots from spleens of vehicle- and hEV-treated animals on day 21. E. Quantification of the total number of CD4⁺ cells per mg of spleen tissue after gating on live cells (ns = 0.4175, n=8/group, two-tailed Student’s t-test). Data are mean ± SEM. F. Quantification of the number of CD4⁺FOXP3⁺ cells per mg of spleen tissue (ns = 0.6178, n=8/group, two-tailed Student’s t-test). Data are mean ± SEM. G. Quantification of the number of CD4⁺FOXP3⁺IL10⁺ cells per mg of spleen tissue (*P = 0.0487, n=8/group, two-tailed Student’s t-test). Data are mean ± SEM.

Transcriptome analysis of iT_reg-hEV reveals an intermediate phenotype with features of T_reg and Th17 cells

Given the changes observed among iT_reg cells exposed to hEVs, we sought to further characterize this cell population. In a global transcriptome analysis (Figure 5A), iT_reg cells upregulated expression of Rorc (RORγt), Il17a, and Il17f in response to human hEVs (Figure 5B), a profile characteristic of Th17 cells.^{23, 24} Meanwhile, iT_reg-hEV preserved their expression of Il2ra (CD25), Foxp3, Ctla4, and Tgfb1 (Figure 5B,C), which are defining markers for T_reg cells. Consistent with Figure 2, iT_reg-hEV expressed much higher levels of Il10 (Figure 5B), which acts not only to suppress inflammation but also to maintain the T_reg phenotype.²⁵ To exclude the possibility that hEVs are polarizing T_reg cells toward T regulatory 1 (Tr1) cells, which also secrete high levels of IL-10, we assessed the expression of LAG-3 and CD49b (characteristic markers of Tr1 cells).²⁶ Expression levels of LAG-3 and CD49b were similar between iTreg-Ctrl and iTreg-hEV (Figure S4), suggesting that iT_reg-hEV are distinct from Tr1 cells. Taken together, exposure to hEVs induces an intermediate phenotype, sharing characteristics of FOXP3⁺ T_reg cells and RORγt⁺ Th17 cells (Figure 5D). T_reg cells are known to express both FOXP3 and RORγt,^27–29 transcription factors that together allow for selective suppression of Th17 cells but, in order to do so, additionally require transcription factor c-Maf (encoded by Maf).³⁰ Indeed, expression of Maf was much higher among iT_reg-hEV, compared to iT_reg-Ctrl (Figure 5B). Finally, we noted upregulation of Il21, a cytokine that stimulates proliferation of activated T cells.³¹ Thus, the transcriptome profile of iT_reg-hEV points to an intermediate FOXP3⁺RORγt⁺ cell type, capable of enhanced proliferation upon activation.

Figure 5. iT_reg cells exposed to human hEVs exhibit an intermediate FOXP3⁺RORγt⁺ phenotype with T_reg cell and Th17 features.

A. Transcriptome analysis of iT_reg-Ctrl and iT_reg-hEV (n=3/group). B. RNA-sequence volcano plot with –log₁₀[adjusted P value] (y axis) and log₂[fold change] (x axis), highlighting significantly upregulated (red) and downregulated (blue) genes in iT_reg-hEV, relative to iT_reg-Ctrl. C. Log₁₀ expression of Foxp3 in naïve T cells (Th0), iT_reg-Ctrl, and iT_reg-hEV, relative to average Foxp3 expression among Th0 cells (n=4–6/group). Data are mean ± SEM. D. Merged contour plots for FOXP3 and RORγt expression in iT_reg-Ctrl and iT_reg-hEV. Plots are representative of two independent experiments.

iT_reg-hEV cells exhibit suppressive capacity in vitro

Given the changes in iT_reg cells toward an intermediate FOXP3⁺RORγt⁺ phenotype, we wanted to assess whether human hEVs alter iT_reg cells’ suppressive ability using an in vitro assay with responder CD8⁺ T cells, antigen presenting cells, and iT_reg cells.²⁹ Both iT_reg-Ctrl (Figure 6A,B) and iT_reg-hEV (Figure 6C,D) suppressed proliferation of CD8⁺ T cells, verifying that the central immunosuppressive phenotype is not impeded by hEVs. Therefore, iT_reg-hEV represent an intermediate FOXP3⁺RORγt⁺ cell type with enhanced proliferation and IL-10 production and preserved suppressive capacity.

Figure 6. iT_reg-hEV maintain suppressive capacity in vitro.

A. Representative histograms show proliferation of CD8⁺ T cells after co-culture with antigen presenting cells and iT_reg-Ctrl at various ratios. B. Quantification of proliferation and suppression of CD8⁺ T cells co-cultured with iT_reg-Ctrl; n=4/group. Data are mean ± SEM. % Suppression = [(% proliferating with no iT_reg - % proliferating with iT_reg)/% proliferating with no iT_reg) x 100]. C. Representative histograms show proliferation of CD8⁺ T cells after co-culture with antigen presenting cells and iT_reg-hEV at various ratios. D. Quantification of proliferation and suppression of CD8⁺ T cells co-cultured with iT_reg-hEV; n=3–4/group. Data are mean ± SEM. % Suppression = [(% proliferating with no iT_reg - % proliferating with iT_reg)/% proliferating with no iT_reg) x 100].

Discussion

EVs are nanometer-sized, lipid-bilayer particles laden with noncoding RNAs and a diverse signaling cargo.³² In the context of immunity, EVs derived from both immune and non-immune parent cells have demonstrated bioactivity in dendritic cells, monocytes, macrophages, natural killer cells and T lymphocytes.^{6, 33} Exposure to EVs can induce either activating or inhibitory effects, depending on the origin and state of the parent cell that produced the EVs. Thus, the underlying mechanisms of EV immunomodulation are diverse and warrant further investigation to better define their therapeutic and diagnostic potential.

Here, we have demonstrated that hEVs elicit cell-specific and dose-dependent responses among effector and regulatory CD4⁺ T cells. The biological effect of hEVs was most pronounced in T_reg cells, with increased rates of proliferation, augmented production of IL-10, and modified expression of lineage-specific transcription factors. Effects of hEVs on cells of innate immunity are well-described; for example, hEVs polarize macrophages toward a distinctive phenotype, with decreased expression of proinflammatory genes (Nos2 and Tnf), increased expression of anti-inflammatory genes (Arg1, Il4ra, Tgfb1), and enhanced efferocytosis.^{10, 11} These changes are mediated through transfer of EV cargo, including noncoding RNAs and micro-RNAs (e.g., Y RNA fragments and miR-181b). Fragments of Y RNA are of particular interest, as they constitute the largest fraction of small RNAs within hEVs. The most abundantly expressed fragment, EV-YF1, induces production and secretion of IL-10 by macrophages, with downstream cardioprotective effects.⁹ Therefore, this RNA fragment could potentially account for the increase in IL-10 we have observed, but this possibility remains untested. In terms of proliferation, miR-155 is known to promote T_reg cell expansion and therefore could be the responsible factor within hEV cargo, but this too is conjectural.³⁴

The increase in IL-10 production by T_reg cells, coupled with their enhanced proliferation, can theoretically dampen systemic and local inflammation. Although T_reg cells employ other mechanisms of suppression (e.g., TGFβ, IL-35, CTLA4, CD25), only expression of IL-10 was increased following hEV administration in our study (Figure 2B, Figure 5B), suggesting that the suppressive effects we observed in vitro were related to changes in IL-10. In EAM, we showed that hEVs have anti-inflammatory and cardioprotective effects, which correlated with increased numbers of IL10⁺-T_reg cells. Further characterization of iT_reg-hEV revealed a transcriptional profile with intermediate features of T_reg and Th17 cells. Dual expression of FOXP3 and RORγt by T_reg cells has been described in rodents and humans.^{27–29, 35} In humans, FOXP3⁺RORγt⁺ T_reg cells are capable of producing IL-17, yet retain their suppressive function.^{35, 36} Other studies have demonstrated enhanced suppressive function by FOXP3⁺RORγt⁺ T_reg cells.²⁹ Because T_reg cells can acquire features specific to the type of cell they aim to suppress,⁴ FOXP3⁺RORγt⁺ T_reg cells are well-suited to suppress Th17 cells. This is particularly important in EAM, where higher levels of Th17 cells account for increased EAM susceptibility.³⁷ Furthermore, Th17-mediated inflammation favors adverse cardiac remodeling and progression to dilated cardiomyopathy in mice and humans with myocarditis.^{38, 39} Thus, the intermediate FOXP3⁺RORγt⁺ phenotype induced by hEV represents a potential mechanism for the functional benefits observed in EAM, but further investigation (e.g., depletion and adoptive transfer studies) is needed to test this possibility.

The ability to modulate T_reg cells is an attractive therapeutic approach to inflammation that relies on harnessing an endogenous mechanism of immunosuppression. Multiple (>50) clinical trials are exploring this general approach in the setting of solid organ transplantation, graft-versus-host disease, and autoimmune disorders. Most of these trials, however, rely on a challenging adoptive-transfer protocol that involves isolation of autologous T_reg cells, expansion of these cells ex vivo, and reinfusion of the expanded population back into the patient.² This paradigm is limited by the scarcity of T_reg cells in the peripheral blood (the most common source for T_reg cells), as well as uncertainties about the suppressive and proliferative potential of T_reg cells after ex vivo expansion. Treatment with hEVs can potentially harness T_reg cells in vivo, ^{40, 41} circumventing the limitations of adoptive transfer. We already have circumstantial evidence of hEV bioactivity in patients with COVID-19, a syndrome characterized by a maladaptive, hyperinflammatory host response and lymphocytopenia.⁴² Therapy using the parent cardiac stromal/progenital cells in 4 of 6 such critically-ill patients was associated with clinical improvement, including higher lymphocyte counts and lower levels of inflammatory biomarkers.⁴³ Beyond COVID-19, these parent cells have demonstrated signs of therapeutic efficacy in clinical trials of ischemic^{44, 45} and dilated⁴⁶ cardiomyopathy, and in Duchenne muscular dystrophy.⁴⁷ Our findings that hEVs are able to augment T_reg cell proliferation and IL-10 production, along with prior evidence of immunomodulatory effects by the parent cells,^{12, 13, 48–50} may shed light on the mechanisms underlying disease-modifying bioactivity. Not only in COVID-19 or myocarditis, but more generally for inflammatory disorders involving innate and adaptive immunities, hEVs represent logical next-generation therapeutic candidates.