In-vitro immunomodulatory efficacy of extracellular vesicles derived from TGF-β1/IFN-γ dual licensed human bone marrow mesenchymal stromal cells

Abstract

Background

Mesenchymal stromal cells (MSCs) possess strong immunomodulatory properties, making them attractive candidates for regenerative medicine and immune-related therapies. Pre-activation, or licensing, of MSCs with cytokines such as interferon-gamma (IFN-γ) and transforming growth factor-beta 1 (TGF-β1) has been shown to enhance their immunosuppressive efficacy. Recent attention has turned to extracellular vesicles (EVs) released by licensed MSCs as a cell-free therapeutic alternative.

Methods

Small EVs were isolated from MSCs licensed with a combination of IFN-γ and TGF-β1. These EVs were characterized according to standardized criteria. Their immunomodulatory effects were assessed in vitro using two human immune models: a THP-1-derived macrophage polarization system and a peripheral blood mononuclear cell (PBMC) co-culture assay. Pro/anti-inflammatory molecules secretion, T cell proliferation, and regulatory T cell induction were quantified. Dimensionality reduction using t-distributed stochastic neighbor embedding (t-SNE) was applied to multiparametric flow cytometry data for immune profiling. In addition, publicly available transcriptomic datasets (GSE122091 and GSE46019) were analyzed to identify differentially expressed genes (DEGs) in IFN-γ– and TGF-β1–licensed MSCs, providing insight into potential molecular drivers of EV-mediated immunoregulation.

Results

Licensed EVs significantly inhibited pro-inflammatory THP-1 macrophage activation and promoted an anti-inflammatory phenotype, with reduced secretion of tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β), increased IL-10 production, and decreased nitric oxide (NO) levels.. Compared to EVs from non-licensed MSCs, licensed EVs induced a greater proportion of regulatory T cells and exhibited enhanced suppression of allogeneic T cell proliferation. t-SNE analysis revealed a distinct immunoregulatory signature induced by licensed EVs, characterized by the emergence of a non-proliferative lymphocyte subset with elevated co-expression of CD4, CD25, and FOXP3. Transcriptomic analysis further revealed seven overlapping DEGs between IFN-γ– and TGF-β1–licensed MSCs, including both upregulated (GPR68, LIMK2, LIPG) and downregulated (EFNA5, PRKG1, DCLK1, TRIM2) genes, several of which are functionally implicated in EV-mediated immune regulation.

Conclusions

Small EVs derived from IFN-γ and TGF-β1-licensed MSCs exhibit demonstrate dose-dependent immunomodulatory trends in vitro, with enhanced effects observed at higher concentrations.. These findings suggest their potential utility in modulating both innate and adaptive immune responses, warranting further investigation for their application as a cell-free therapeutic strategy in immune-mediated conditions.

Graphic Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-025-04476-2.

Background

Mesenchymal stromal cells (MSCs) have garnered significant attention for their robust immunomodulatory capabilities (1–3), establishing them as a promising therapeutic approach in regenerative medicine (4) and immune regulation (5). Through the secretion of bioactive molecules (6) and direct interactions (7) with immune cells, MSCs effectively modulate both innate (8) and adaptive (9) immune responses. These distinctive properties enable MSCs to suppress inflammation, promote tissue repair, and maintain immune homeostasis, making them invaluable candidates for treating a wide spectrum of inflammatory (2, 10) and autoimmune disorders (11, 12).

Recent studies have demonstrated that the immunomodulatory potential of MSCs can be significantly enhanced when the cells are licensed/pre-activated/primed/preconditioned with pro-inflammatory cytokines such as Interferon-γ (IFN-γ) (13, 14). This licensing process activates intracellular signaling cascades in MSCs, resulting in the upregulation of key immunomodulatory molecules, including indoleamine 2,3-dioxygenase (IDO) (15, 16), prostaglandin E2 (PGE2) (17), and programmed death-ligand 1 (PD-L1) (15, 17). These changes further amplify MSCs' ability to suppress inflammatory responses, enhancing their therapeutic potential in immune-mediated diseases. Recently, we showed for the first time that murine MSCs licensed with the anti-inflammatory cytokine Transforming Growth Factor-β1 (TGF-β1) demonstrated greater immunomodulatory potency than naïve MSCs in a mouse corneal transplant model (18). Furthermore, another study from our group revealed that TGF-β1-licensed MSCs increased the frequency of regulatory T cells in an in vitro co-incubation model with peripheral blood mononuclear cells (PBMCs) (19). Both TGF-β1 and IFN-γ licensed MSCs also showed greater ability to reduce pro-inflammatory macrophage secreting TNF-α and IL-1β (19). Collectively, these findings highlight the enhanced immunomodulatory potency of MSCs when licensed with IFN-γ or TGF-β1.

Similarly, extracellular vesicles (EVs) derived from cytokine-licensed MSCs have emerged as a novel focus of research due to their ability to deliver bioactive molecules that mediate immunomodulation (20–23). These EVs, including exosomes and microvesicles, carry proteins, lipids, and nucleic acids reflective of their parent cells’ activated state (24). Small EVs, typically ranging from 30 to 150 nm in diameter, are particularly noted for their role in intercellular communication and their potential for therapeutic immunomodulatory/immunoregulatory applications (25–27). Emerging evidence suggests that cytokine-activated MSC-EVs may exhibit enhanced immunosuppressive and anti-inflammatory effects compared to EVs derived from non-licensed MSCs (21, 28–30). Our previous study specifically focused on cytokine-licensed MSC-derived apoptotic bodies (ApoBDs), a distinct subpopulation of EVs, and demonstrated that they possess superior immunomodulatory potency, particularly in inhibiting allogeneic T cell proliferation, compared to naïve MSC-EVs or those licensed with single cytokines such as IFN-γ or TGF-β1 (31). However, despite these promising preclinical findings, the effects of cytokine-licensed MSC-EVs remain poorly understood, warranting further investigation into their therapeutic potential.

Extracellular vesicles derived from MSCs also offer unique advantages over traditional cell-based therapies. As a cell-free product, EVs minimize the risks associated with immune rejection (24, 32). Additionally, EVs are easier to produce on a large scale (33), exhibit superior storage stability (34), and allow for simplified quality control processes. These features make EVs particularly appealing for therapeutic applications, as they can deliver bioactive molecules reflective of their parent cells' functionality while avoiding the challenges of maintaining cell viability and engraftment in vivo. Consequently, MSC-EVs are increasingly recognized as a next-generation therapeutic platform for regenerative medicine and immune modulation, potentially overcoming limitations faced by MSC-based therapies.

To evaluate the immunomodulatory properties of MSC-EVs in vitro, various experimental models have been developed. These models often involve co-culture systems where MSC-EVs are incubated with immune cells such as T cells (35, 36), macrophages (37, 38), or dendritic cells (39) under inflammatory conditions. Functional assays, including T cell proliferation suppression, macrophage polarization, and cytokine secretion profiling, are used to quantify the immunosuppressive effects of MSC-EVs. Such in vitro models provide crucial insights into the mechanisms by which MSC-EVs exert their immunomodulatory functions, forming the foundation for subsequent in vivo and clinical investigations.

Herein, we focused on investigating the immunomodulatory activity of small EVs derived from TGF-β1/IFN-γ-licensed MSCs due to the enhanced licensing effects of dual TGF-β1/IFN-γ treatment. EVs were isolated and purified using chromatography and subsequently characterized according to MISEV guidelines. Our findings revealed that EVs from both naïve and TGF-β1/IFN-γ-licensed MSCs (referred to as naïve EVs and licensed EVs, respectively) were capable of polarizing THP-1-differentiated pro-inflammatory macrophages toward the anti-inflammatory phenotype, with reduced secretion of tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β), increased IL-10 production, and decreased nitric oxide (NO) levels.. Notably, licensed EVs demonstrated a stronger capacity to inhibit allogeneic T-cell proliferation and to induce regulatory T cells (Tregs). Using a PBMC model analyzed through dimensionality reduction by t-distributed stochastic neighbor embedding (t-SNE), we further identified the unique fingerprint induced by licensed EVs based on the Treg cell staining panel. Transcriptomic analysis of licensed MSCs revealed seven shared differentially expressed genes under IFN-γ and TGF-β1 stimulation, several of which (e.g., GPR68, LIMK2, LIPG) are potentially involved in EV-mediated immunoregulation. In summary, cytokine licensed EVs exhibited superior immunomodulatory properties in vitro and hold promise for greater therapeutic efficacy in future in-vivo studies.

Methods

MSC isolation, culture with cytokines, and characterization

Human MSCs were isolated from the bone marrow of three healthy volunteers at Galway University Hospital under an ethically approved protocol using a standardized procedure (ref. 08/May/14). Written consent was obtained from all donors. Briefly, bone marrow cell suspensions were processed using a Ficoll density gradient to isolate the nucleated cell fraction, which was then washed and resuspended in MSC culture medium. After 24 h of cultivation, nonadherent cells were removed, fresh medium was added, and fibroblast-like colonies were expanded to confluence before passaging.

The MSCs were cultured in T175 flasks with 1,000,000 cells/cm² seeding density in α-MEM (BioSciences) supplemented with 1% penicillin/streptomycin (Sigma-Aldrich), 10% fetal bovine serum (FBS) (Sigma-Aldrich), and 1 ng/mL FGF-2 (Sigma-Aldrich) and incubated at 37 °C with 5% CO₂. The media was changed every 2–3 days and the cells were passaged at 90% confluency. At Passage four, after 24 h, the cells were washed twice with PBS and then 20 mL of licensing medium, a xeno-free medium (40) supplemented with 50 ng/mL IFN-γ (PeproTech, ThermoFisher Scientific) and 50 ng/mL TGF-β1 (Bio-Techne/R&D Systems), was added to the flask following by 72 h incubation at 37 °C with 5% CO₂.

Their characterization involved assessing the expression of specific surface markers (CD73, CD44, CD45, CD90, CD11b, and HLA-DR) by flow cytometry and evaluating differentiation potential. For flow cytometry, MSCs were harvested and incubated with anti-human antibodies (see Supplementary Material for details) diluted in FACS buffer. Data were acquired using a Cytek Northern Lights™ 3000 flow cytometer and analyzed with FlowJo version 10 (Tree Star Inc.). Differentiation assays were performed following established methods (19, 31), with MSCs from three donors (donor 1, 2, and 3) characterized individually.

PBMC isolation

Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood samples obtained from four healthy volunteers following written informed consent approved by Galway University Hospital Clinical Research Ethics Committee (C.A 2534). Blood was collected in 5 mL EDTA-coated Vacutainer® tubes (BD Medical Supplies, Crawley, UK). To isolate PBMCs, 3 mL of anti-coagulated blood was layered over 3 mL of endotoxin-free Ficoll-Premium (Sigma-Aldrich, Wicklow, Ireland) in 15 mL tubes and centrifuged at 400 g for 22 min at 18 °C. The mononuclear cell layer ("buffy coat") was carefully extracted using a plastic Pasteur pipette and transferred to fresh tubes. PBMCs were washed twice with 10 mL DPBS (Thermo Fisher Scientific) and centrifuged at 400 g for 5 min at 25 °C. Cell count and viability were assessed using Trypan Blue exclusion.

THP-1 and PBMC culture

THP-1 cells (provided by the research group of Dr Aideen Ryan, Discipline of Pharmacology and Therapeutics, School of Medicine, University of Galway, Galway, Ireland) and human PBMCs were cultured in RPMI-1640 medium (Sigma-Aldrich) supplemented with 10% heat-inactivated FBS (Thermo Fisher Scientific), 1% L-glutamine, and 1% penicillin/streptomycin (both from Sigma-Aldrich).

Macrophage induction

To induce differentiation, THP-1 cells were treated with 100 ng/mL PMA (Sigma-Aldrich) for 24 h, resulting in macrophage-like THP-1 cells adhering to the flask. Following the removal of non-adherent, undifferentiated cells, the differentiated cells were further stimulated with 100 ng/mL LPS (Sigma-Aldrich) and 20 ng/mL IFN-γ (PeproTech) for an additional 24 h to generate pro-inflammatory macrophage-like THP-1 cells.

MSC-EVs isolation and characterization

MSCs were cultured in xeno-free naïve or IFN-γ and TGF-β1 cytokines dual-licensed media for 72 h at 37 °C with 5% CO₂. Afterwards, the conditioned media was collected and centrifuged for 10 min at 400 × g in order to remove the cell debris. The supernatant was further centrifuged for 30 min at 2,000 × g to eradicate any remaining debris and the large apoptotic bodies. Utilizing Amicon Ultra-15 centrifugal filter units with 100 kDa MWCO filters, the condition media was concentrated. MSC-EVs were isolated by running the concentrated conditioned media through size exclusion chromatography (SEC) q-EV 35 nm Gen 2 columns (iZON), following manufacturer’s protocol. The fractions 1–3 post-void volume were pooled as MSC-EVs containing fractions, aliquoted in 0.1 μm filtered-PBS, and stored at -80 °C until further experiments.

The isolated naïve and cytokine-licensed MSC-EVs were characterized following MISEV 2023 guideline (41). The MSC-EVs were analyzed for their size distribution and concentration by diluting the samples in 0.1 μm filtered ultra-pure water (1:1000) and using Nanoparticle Tracking Analysis (NTA, Nanosight NS500, Malvern Zetasizer, Worcestershire, UK). Transmission electron microscope (TEM) was employed to observe the morphology of the MSC-EVs by adsorbing them into Formvar/carbon coated grids (Agar Scientific, UK) and analyzing at 100.0 kV voltage and 80,000 × direct magnification using Hitachi 7500 microscope (Hitachi, Japan). The surface biomarkers of the MSC-EVs were investigated by staining the MSC-EVs by CD9 (Catalog number: 555372, BD Pharmingen), CD63 (Catalog number: 557305, BD Pharmaceutics), and CD81 (Catalog number: 551108, BD Pharmaceutics) antibodies at 1:20 concentration in 0.1 μm filtered PBS. The stained MSC-EVs were analyzed by Cytek Northern Lights™ 3000 high resolution flow cytometer. The Apogee size calibration beads (Apogee flow systems, UK), 0.1 μm filtered PBS, Unstained MSC-EVs, and free antibodies were used as control groups for setting the forward scatter (FSC) and side scatter (SSC) voltages, gating the background noise, and confirming the MSC-EVs Ab-staining, respectively.

MSC-EV/macrophage co-incubation

1 × 10^5 THP-1-differentiated pro-inflammatory macrophages were seeded and incubated with 1 × 10^7, 5 × 10^7, 1 × 10^8, and 5 × 10^8 naïve or licensed MSC-EVs, respectively, for 48 h. Brefeldin A was added at the 42nd hour and maintained until the end of incubation. Cells were harvested and stained using cytokine and polarization panels (Supplementary Material). Data acquisition was performed on a Cytek Northern Lights™ 3000 flow cytometer and analyzed with FlowJo version 10 (Tree Star Inc.).

The supernatant from the co-culture was collected, and IL-10 levels were quantified using the LEGEND MAX™ Human IL-10 ELISA Kit with Pre-coated Plates (BioLegend, Cat. No. 430607), according to the manufacturer’s instructions.

Nitrite levels in culture supernatants were measured using the Griess assay. Equal volumes of Reagent A (1% sulfanilamide in 5% phosphoric acid; Sigma-Aldrich, S9251 and 695,017) and Reagent B (0.1% N-(1-naphthyl)ethylenediamine dihydrochloride in 5% phosphoric acid; Sigma-Aldrich, N9125 and 695,017) were freshly mixed. Then, 100 μL of sample (supernatant and standard reference) was combined with 100 μL of Griess reagent (A + B) in a 96-well plate. After color development at room temperature, absorbance was measured at 543 nm.

MSC-EV/PBMC Co-incubation

PBMCs were stained with CellTrace dye (Thermo Fisher Scientific, Dublin, Ireland) following the manufacturer’s protocol and seeded into 96-well round-bottom plates (Sarstedt) at a concentration of 1 × 10^5 cells/100 μL of complete medium, with or without 2 × 10^4 Human T-Activator CD3/CD28 Dynabeads® (Thermo Fisher Scientific, Dublin, Ireland). Naïve and licensed MSC-EVs were added at concentrations of 1 × 10^7, 5 × 10^7, 1 × 10^8, and 5 × 10^8, respectively, and incubated with PBMCs for 72 h. After incubation, cells were harvested and stained using the proliferation and Treg panels (described in Supplementary Material). Data acquisition was performed on a Cytek Northern Lights™ 3000 flow cytometer and analyzed using FlowJo version 10 (Tree Star Inc.).

Clustering approach: DBSCAN

We employed the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) to identify clusters in the t-SNE-reduced data. Unlike K-means, DBSCAN does not require predefining the number of clusters; instead, it identifies clusters based on the density of data points in a given region.

DBSCAN has two primary parameters: 1. Epsilon (eps): The maximum distance within which points are considered neighbors; 2. MinPts (min_samples): The minimum number of points required to form a dense region.

Data points are categorized as follows: 1. Core points: Points with at least min_samples neighbors within a radius of eps. 2. Border points: Points that fall within the eps radius of a core point but do not themselves meet the min_samples requirement. 3. Noise points: Points that do not belong to any cluster and are labeled as noise.

The clustering process was as follows: 1. The K-distance graph was used to determine the optimal value for eps. The "elbow point" in the graph represents a significant transition, indicating the distance threshold for dense regions. 2. We empirically set min_samples to align with the dimensionality of the t-SNE-reduced data (commonly 5–20 for 2D data)0.3. DBSCAN was executed with the identified parameters to detect clusters and outliers.

The R script was in Supplementary Material.

Bubble plot visualization and analysis

To visualize the distribution of variables across clusters and groups, a custom bubble plot was generated using the R programming language. Data preprocessing and visualization were performed with the dplyr, tidyr, and ggplot2 packages. The key steps involved in the visualization are as follows: Normalization: 1) Bubble sizes were scaled relative to the maximum proportion within each cluster, ensuring comparability across clusters. 2) Colors were scaled for each variable based on the normalized range of median expression values, with white representing the minimum and purple the maximum.

The plot was created using geom_point in ggplot2, where the bubble size reflected the proportion of each group, and the color gradient represented the normalized expression values.

Public transcriptomic data acquisition and analysis

Transcriptomic datasets GSE122091 (comparing IFN-γ–licensed versus naïve MSCs; embryonic stem cell–derived MSCs treated with 100 U/mL IFN-γ for 24 h) and GSE46019 (comparing TGF-β1–licensed versus naïve MSCs; bone marrow–derived MSCs treated with 1 ng/mL TGF-β1 for 12 h) were retrieved from the NCBI Gene Expression Omnibus (GEO) database. Differential gene expression analysis was performed using logFC thresholds of > 0.5 (upregulated) and < –0.5 (downregulated), with adjusted P < 0.1. Overlapping differentially expressed genes (DEGs) between the two datasets were identified and visualized using volcano plots (R studio 4.4.2) and bar graphs (GraphPad Software 8.0.2).

Statistical analysis and cell heterogeneity statement

Data are presented as mean ± SD. Statistical significance for Fig. 5C was determined using one-way analysis of variance (ANOVA) with Tukey's multiple comparisons test. For all other analyses, two-way ANOVA with Sidak’s multiple comparisons test was employed. Differences were considered statistically significant at p < 0.05. Statistical analyses were performed using GraphPad Software (version 8.0.2). Details regarding donor heterogeneity are provided in the Supplementary Material.

Fig. 5.

Cluster analysis based on t-distributed stochastic neighbor embedding (t-SNE) dimensionality reduction and regulatory T cell (Treg) staining panel. A Marker expression profile for each cluster. B Cluster fingerprints for each group. C Frequency distribution of each cluster across the experimental groups. Statistical significance was assessed using one-way ANOVA with Tukey's multiple comparisons test (*P < 0.05; **P < 0.01; ***P < 0.001). (D) Bubble plot summarizing cluster and experimental group data. Bubble size represents the frequency of each cluster, normalized to the largest bubble within the three comparison units. The intensity of purple reflects marker expression levels, normalized by each row. Data were collected via flow cytometry, with dimensionality reduction performed using t-SNE and clustering achieved using the DBSCAN package. All results represent three independent experiments. (Naïve MSC-EVs: EVs derived from naïve MSCs; Licensed MSC-EVs: EVs derived from TGF-β1/IFN-γ-licensed MSCs)

Results

MSC characterization

This section is in Supplementary Materials.

MSC-EV characterization of naïve and cytokine licensed MSCs

MSC-EVs were isolated from naïve and cytokine-licensed MSC-conditioned medium using iZON q-EV Gen 2 SEC columns. NTA analysis was used to analyze the size and concentration of the MSC-EVs. Naïve MSC-EVs showed a mode size of 78.8 nm and 1.42e + 10 particles/mL concentration, while for the licensed-MSC-EVs the mode size was 86.3 nm and concentration of 3.79e + 10 particles/mL (Fig. 1A). MSC-EVs were adsorbed into carbon coated grids and visualized using TEM microscope. As shown in Fig. 1B, naïve and licensed MSC-EVs showed lipid bilayer spherical shape with size around 70 nm and 80 nm, respectively. Flow cytometry analysis was used to analyze the surface biomarkers of the naïve and licensed MSC-EVs. The SSC and FSC were set by analyzing Apogee beads, and filtered PBS, unstained MSC-EVs and antibody only samples were used for gating strategy (Supplementary Material Fig. 1). Both naïve and licensed MSC-EVs were positive for the CD9, CD63, and CD81 surface biomarkers (Fig. 1C). Naïve MSC-EVs were 65.7% CD9⁺, 65.6% CD63⁺, and 54.0% CD81⁺. Expression of CD9, CD63, and CD81 were 80.8%, 81.1%, and 66.7% for the licensed-MSC-EVs, respectively (Fig. 1D).

Fig. 1.

Naïve and cytokine-licensed mesenchymal stromal cell-derived extracellular vesicles (MSC-EVs) characterization. A Mode size and concentration of the naïve and licensed MSC-EVs B Representative TEM images of the naïve and licensed MSC-EVs C Representative pseudo-color plots showing the expression of CD9, CD63, and CD81 surface biomarkers in naïve and licensed MSC-EVs. D Percentage of the surface biomarkers expression of the naïve and licensed MSC-EVs. All results represent three independent experiments

MSC-EV/macrophage co-incubation

THP-1 cells and THP-1-differentiated macrophages were extensively utilized as substitutes for monocytes and macrophages (42, 43), respectively. This model has also been applied in our and other’s research (19, 31, 44–46). Briefly, THP-1 cells were induced into an pro-inflammatory phenotype and co-incubated with varying concentrations of MSC-EVs.

Regarding the anti-inflammatory macrophage phenotype surface markers, as shown in Fig. 2A, both naïve and licensed MSC-EVs upregulated CD163 and CD206 even at the lowest MSC-EV concentration (Fig. 2A). This indicates that both naïve and licensed MSC-EVs can polarize pro-inflammatory macrophages into the anti-inflammatory phenotype. Interestingly, there was no significant difference between naïve and licensed MSC-EVs in their ability to upregulate CD163 and CD206 expression in macrophages.

Fig. 2.

Polarization assay of THP-1-differentiated macrophages by mesenchymal stromal cell-derived extracellular vesicles (MSC-EVs). A Surface protein expression of CD163 and CD206 and B intracellular cytokine levels of IL-1β and TNF-α in macrophages after incubation with gradient concentrations of naïve and licensed MSC-EVs. All results represent three independent experiments. C Relative concentration of IL-10 in cell culture supernatants measured by enzyme-linked immunosorbent assay. D Relative production of nitric oxide (NO) assessed by Griess reagent assay. Data were presented as mean ± SD and were analyzed by flow cytometry. Statistical significance was assessed using two-way ANOVA with Šídák's multiple comparisons test. ns: not significant; *P < 0.05; **P < 0.01; ***P < 0.001. (Naïve MSC-EVs: EVs derived from naïve MSCs; Licensed MSC-EVs: EVs derived from TGF-β1/IFN-γ-licensed MSCs)

Pro-inflammatory macrophages initiate signaling events that lead to the release of pro-inflammatory cytokines IL-1β and TNF-α, triggering a cascade of inflammatory reactions. To investigate whether MSC-EVs can inhibit the secretion of these cytokines, naïve and licensed MSC-EVs were co-incubated with pro-inflammatory macrophages. The cytokine secretion was blocked intracellularly using Brefeldin A for 6 h, and cytokine levels were assessed through intracellular flow cytometry staining (Fig. 2B). Although the differences were not statistically significant, the data suggest a trend indicating that both naïve and licensed MSC-EVs inhibit the secretion of IL-1β and TNF-α.

The anti-inflammatory cytokine IL-10 was quantified by ELISA. As shown in Fig. 2C, IL-10 secretion by macrophages increased in a dose-dependent manner following treatment with MSC-EVs. At the highest dosage, both naïve and licensed EVs significantly enhanced IL-10 production, indicating their capacity to promote an anti-inflammatory macrophage phenotype. Notably, licensed EVs induced significantly higher IL-10 levels compared to naïve EVs at the same concentration (the highest dosage), suggesting a stronger immunomodulatory potential.

As shown in Fig. 2D, the inflammatory mediator nitric oxide (NO) was quantified using the Griess assay. A significant reduction in NO production was observed only at the highest dose of licensed EVs, indicating their superior capacity to induce an anti-inflammatory macrophage phenotype.

In summary, both naïve and licensed MSC-EVs promote the polarization of pro-inflammatory macrophages into the anti-inflammatory phenotype by upregulating CD206 and CD163 expression. Additionally, they reduce the secretion of TNF-α and IL-1β, as indicated by numerical trends. Moreover, ELISA results demonstrated a dose-dependent increase in IL-10 production following EV treatment, with licensed EVs inducing significantly higher levels than naïve EVs. Consistently, NO production—commonly associated with inflammatory activation—was significantly reduced by licensed EVs at the highest dose. These results collectively support the enhanced immunomodulatory capacity of licensed MSC-EVs in promoting an anti-inflammatory macrophage phenotype.

MSC-EV/PBMC co-incubation: T cell proliferation assay

The T cell proliferation assay is considered the gold standard for evaluating immunomodulatory potency, as a lower frequency of proliferated T cells indicates stronger immunomodulatory effects. In this study, PBMCs were stimulated with anti-CD3/CD28 beads and incubated with various concentrations of MSC-EVs. The gating strategy for identifying CD3⁺, CD4⁺, and CD8⁺ T cells is summarized in Fig. 3A.

Fig. 3.

T cell proliferation assay using mesenchymal stromal cell-derived extracellular vesicles (MSC-EVs). A Gating strategy for CD3⁺, CD4⁺, and CD8⁺ T cells from peripheral blood mononuclear cells (PBMCs). B Percentage of proliferated CD3⁺, CD4⁺, and CD8⁺ T cells following incubation with gradient concentrations of naïve and licensed MSC-EVs. C Representative CellTrace Violet intensity histograms for CD3⁺, CD4⁺, and CD8⁺ T cells after 72 h of incubation with 5E8 naïve and licensed MSC-EVs. The median fluorescence intensity of D PD-L1 and F CD73 of CD3⁺ cells. The histogram on the right represents the results of incubation with 5E8 EVs. All results represent three independent experiments. Data were presented as mean ± SD and were analyzed by flow cytometry. Statistical significance was assessed using two-way ANOVA with Šídák's multiple comparisons test. ns: not significant; *P < 0.05; **P < 0.01; ***P < 0.001. (Naïve MSC-EVs: EVs derived from naïve MSCs; Licensed MSC-EVs: EVs derived from TGF-β1/IFN-γ-licensed MSCs)

Licensed MSC-EVs significantly inhibited the proliferation of CD3⁺, CD4⁺, and CD8⁺ T cells, but only at the highest dose tested (Fig. 3B and 3C). At lower doses, no statistically significant differences were observed although a trend towards inhibition of proliferation was evident. In contrast, naïve MSC-EVs did not exhibit any significant inhibitory effect on T cell proliferation at any of the tested doses. The CTV intensities of T cells treated with the highest dose of MSC-EVs are visualized in Fig. 3C, where an additional generation of T cells was clearly observed in the group treated with naïve MSC-EVs.

In summary, licensed MSC-EVs significantly inhibited the proliferation of CD3⁺ effector CD4 and CD8 T cells, whereas naïve MSC-EVs showed no such effect.

MSC-EV/PBMC co-incubation: PD-L1 and CD73 expression in CD3⁺ T Cells

Next, the expression levels of PD-L1 and CD73 in CD3⁺ T cells were analyzed, as both PD-L1 (47, 48) and CD73 (35, 49) are critical immunomodulatory proteins. As shown in Fig. 3D, naïve MSC-EVs had a minimal effect on upregulating PD-L1 expression in CD3⁺ T cells. In contrast, licensed MSC-EVs significantly and potently upregulated PD-L1 expression, which is consistent with their inhibitory effect on T cell proliferation described earlier. This difference is further supported by the high intensity of PD-L1 observed in the right histogram.

Interestingly, in our study, we observed a notable upregulation of PD-L1 expression on CD3⁺ T cells following stimulation with anti-CD3/CD28 beads. While PD-L1 is traditionally associated with tumor cells and professional antigen-presenting cells such as macrophages and dendritic cells (50), recent findings have shown that PD-L1 can also be inducibly expressed on activated T cells—particularly CD4⁺ memory subsets—under specific inflammatory or regulatory conditions (51). In this context, PD-L1⁺ T cells may contribute to local immune modulation or even promote regulatory mechanisms, raising the possibility that they play a functional role in immune suppression or exhaustion. Given this context-dependent expression pattern, the immunological significance of PD-L1⁺ T cells warrants further investigation, particularly in the setting of immunoregulatory EV therapies.Regarding CD73 expression, both naïve and licensed MSC-EVs induced CD73 in CD3⁺ T cells. However, no statistically significant differences were observed between the effects of naïve and licensed MSC-EVs on CD73 upregulation.

In summary, both naïve and licensed MSC-EVs upregulated CD73 expression in CD3⁺ T cells. However, only licensed MSC-EVs significantly upregulated PD-L1 expression in CD3⁺ T cells.

MSC-EV/PBMC co-incubation: Treg induction

Regulatory T cells (Tregs) are a critical immunomodulatory subpopulation within PBMCs, and a high frequency of Tregs indicates strong immunosuppressive potency (52). The experimental model used here was consistent with previous sections, with a staining panel specifically designed for Treg detection. The gating strategy for identifying Tregs is illustrated in Fig. 4A, where CD4, CD25, and FOXP3 triple-positive cells were classified as Tregs. CellTrace Far Red (CTFR) dye was employed to distinguish between proliferated and non-proliferated Tregs.

Fig. 4.

Regulatory T cell (Treg) induction assay using mesenchymal stromal cell-derived extracellular vesicles (MSC-EVs). A Gating strategy for proliferated Tregs from peripheral blood mononuclear cells (PBMCs). B Frequency of Tregs following incubation with gradient concentrations of naïve and licensed MSC-EVs (left), and representative dot plots showing Tregs induced by 5E8 naïve and licensed MSC-EVs (right). C Frequency of non-proliferated (right) and proliferated Tregs (middle) after incubation with gradient concentrations of naïve and licensed MSC-EVs, and representative CellTrace Far Red intensity histograms of Tregs after 72 h of incubation with 5E8 naïve and licensed MSC-EVs. All results represent three independent experiments. Data were presented as mean ± SD and were analyzed by flow cytometry. Statistical significance was assessed using two-way ANOVA with Šídák's multiple comparisons test. ns: not significant; *P < 0.05; **P < 0.01; ***P < 0.001. (Naïve MSC-EVs: EVs derived from naïve MSCs; Licensed MSC-EVs: EVs derived from TGF-β1/IFN-γ-licensed MSCs)

As shown in Fig. 4B, similar to the T cell proliferation assay, licensed MSC-EVs demonstrated a significantly stronger capacity to induce a higher frequency of Tregs, particularly at the two highest doses tested, with statistical significance. In contrast, naïve MSC-EVs showed limited efficacy in inducing Tregs. The FOXP3 intensity histogram further supports these findings, as more Tregs were induced by licensed MSC-EVs compared to naïve MSC-EVs.

Next, the origin of the induced Tregs was analyzed. Considering the previous observation that licensed MSC-EVs inhibited T cell proliferation, we investigated whether Treg proliferation, as a subset of T cells, was similarly affected. As shown in Fig. 4C, the left histogram illustrates the gating of proliferated and non-proliferated Tregs. The majority of Tregs induced by both licensed and naïve MSC-EVs originated from non-proliferated T cells, although the overall number was much lower for naïve MSC-EVs. The right diagram in Fig. 4C indicates that licensed MSC-EVs promoted Treg proliferation more effectively, resulting in a higher frequency of proliferated Tregs and a lower frequency of non-proliferated Tregs, despite the majority of Tregs being derived from the non-proliferated T cells.

In conclusion, licensed MSC-EVs induced a higher frequency of Tregs compared to naïve MSC-EVs, which largely maintained the baseline Treg frequency. The majority of Tregs originated from non-proliferated cells. Licensed MSC-EVs appear to induce differentiation of non-proliferated T cells into Tregs while also promoting the expansion of existing Tregs.

MSC-EV/PBMC co-incubation: Treg fingerprinting using t-SNE-based dimensionality reduction

Given the multi-marker staining employed in this study, t-SNE-based dimensionality reduction was conducted to analyze the multi-dimensional characteristics of Tregs (53, 54). The markers included CD4, CD25, FOXP3, and CTFR, and the t-SNE method reduced these dimensions into two comprehensive parameters: t-SNE 1 and t-SNE 2. Cell clustering was performed using the DBSCAN method (55, 56), which automatically identified clusters based on the t-SNE results.

Five distinct clusters were identified, as shown in Figs. 5A and 5B. Figure 5A illustrates the marker expression profiles for each cluster. Notably, cluster 3 (light blue) exhibited the highest expression of CD4, CD25, FOXP3, and CTFR, representing non-proliferated typical Tregs. In contrast, cluster 0 showed the lowest expression of these markers, representing highly proliferated non-Treg cells. Clusters 1, 2, and 4 exhibited intermediate marker expression levels for all four markers (CD4, CD25, FOXP3, and CTFR), which were consistently highest in Cluster 3 and lowest in Cluster 0. The expression hierarchy across these clusters followed the order: Cluster 3 > Cluster 1 > Cluster 4 > Cluster 2 > Cluster 0. Notably, higher proliferation (indicated by lower CTFR staining) was observed in Clusters 1, 2, and 4, correlating with a progressively less Treg-like phenotype. This pattern suggests that as cells exhibit increased proliferation, their regulatory characteristics diminish, moving them away from the classical Treg profile seen in Cluster 3 toward a phenotype more closely resembling effector or non-regulatory T cells, as typified by Cluster 0.

This pattern suggests that increased proliferation is associated with a progressive loss of regulatory characteristics, moving cells away from the classical Treg phenotype observed in Cluster 3 toward a more effector-like or non-regulatory profile, as seen in Cluster 0. Given that FOXP3, CD4, and CD25 expression decreases with proliferation, this implies that the majority of Tregs are derived from a predefined differentiation pathway rather than arising through extensive proliferation of existing Tregs.

Figure 5B depicts the t-SNE fingerprints for each experimental group, with cells colored according to their cluster assignment in Fig. 5A. The PBMC group treated with 5 × 10^8 MSC-EVs was used as the representative example due to its potent effects observed earlier. Biological replicates were merged to generate the t-SNE plot, and the frequency of each cluster was quantified in Fig. 5C. Licensed MSC-EVs induced a higher number of non-proliferated Tregs (cluster 3) compared to both the control and naïve MSC-EVs. Conversely, the frequency of cluster 2, representing less typical Tregs, was significantly lower in the licensed MSC-EV group compared to the control but higher than the naïve MSC-EV group. For naïve MSC-EVs, the frequencies of all Treg-related clusters (1, 2, 3, and 4) were lower than both the control and licensed MSC-EVs, while the frequency of non-Tregs (cluster 0) was higher.

In summary, licensed MSC-EVs effectively promoted non-proliferated typical Tregs (cluster 3), whereas naïve MSC-EVs showed limited capacity to induce Tregs.

Finally, all findings were summarized in the bubble plot in Fig. 5D. Here, bubble size represents cluster frequency normalized to the largest bubble in each three-comparison unit, and color intensity indicates marker expression normalized to the strongest expression within each row. Consistent with the previous observations, cluster 3 represented non-proliferated typical Tregs with high expression of CD4, CD25, FOXP3, and CTFR. The marker expression levels followed the sequence: cluster 3 > cluster 1 > cluster 4 > cluster 2 > cluster 0. Cluster 0, characterized by highly proliferated non-Treg-like cells, showed minimal marker expression. Licensed MSC-EVs generated a larger cluster 3 bubble, demonstrating their ability to promote Treg induction and differentiation, whereas naïve MSC-EVs had limited effect.

Parental MSC Transcriptome Suggests Molecular Drivers of Licensed EV Immunomodulation

To investigate the potential molecular differences between naïve and licensed MSC-derived EVs, we analyzed publicly available transcriptomic datasets. Specifically, we retrieved two open-access GEO datasets: GSE122091, which compares naïve MSCs with IFN-γ–licensed MSCs, and GSE46019, which compares naïve MSCs with TGF-β1–licensed MSCs. In both datasets, genes with logFC < –0.5 and adjusted P < 0.1 were considered downregulated, while genes with logFC > 0.5 were considered upregulated. As shown in Fig. 6A and 6B, three genes (GPR68, LIMK2, and LIPG) were consistently upregulated under both IFN-γ and TGF-β1 stimulation, whereas four genes (EFNA5, PRKG1, DCLK1, and TRIM2) were consistently downregulated. Given that the molecular profiles of EVs closely reflect those of their parental cells (57), these seven genes are strong candidates for being similarly differentially expressed in licensed EVs compared to naïve EVs.

Fig. 6.

Transcriptomic comparison of mesenchymal stromal cells (MSCs) treated with IFN-γ and TGF-β1 reveals shared differentially expressed genes, addressing the possible related molecules. A Volcano plot of differentially expressed genes (DEGs) between untreated MSCs and MSCs treated with IFN-γ (GSE122091). Genes upregulated in the IFN-γ group are shown in red (logFC > 0.5, adjusted P < 0.1), and downregulated genes are shown in blue (logFC < –0.5, adjusted P < 0.1). B Volcano plot of DEGs between untreated MSCs and MSCs treated with TGF-β1 (GSE46019), with the same thresholds as in (A). C Expression levels of three genes (GPR68, LIMK2, and LIPG) commonly upregulated in both IFN-γ-treated and TGF-β1-treated MSCs, shown as normalized signal intensity. D Expression levels of four genes (DCLK1, EFNA5, PRKG1, and TRIM2) commonly downregulated under both stimulation conditions. Error bars represent mean ± SD from three biological replicates as illustrated in the GSE46019 and GSE122091

The normalized expression values of these genes in both datasets are illustrated in Fig. 6C and 6D. Seven genes (GPR68, LIMK2, LIPG, EFNA5, PRKG1, DCLK1, and TRIM2) that were consistently up- or downregulated in both IFN-γ– and TGF-β1–licensed MSCs. Although their precise roles in EVs remain to be fully elucidated, accumulating evidence suggests they may contribute to the immunomodulatory functions of EVs. For instance, GPR68, a proton-sensing G protein-coupled receptor, has been implicated in modulating immune responses in acidic inflammatory microenvironments (58) and may influence EV-mediated signaling. LIMK2 phosphorylates and inactivates cofilin to regulate actin cytoskeleton dynamics (59), and thus may modulate the localized F-actin remodeling (60) essential for multivesicular body–plasma membrane fusion (61, 62), thereby influencing EV biogenesis and cargo sorting. LIPG, involved in lipid metabolism (63), could modify the lipid composition of EVs (64) and alter their immunological properties (65). Conversely, EFNA5 and PRKG1 both function as key mediators of cell-to-cell signaling—EFNA5 via incorporation into exosomal Eph/ephrin complexes (66) and PRKG1 through modulation of the NO/cGMP pathway—and their decreased expression may alter the cargo and signaling capacity of extracellular vesicles (67), thereby reshaping vesicle-driven communication with immune cells (68). DCLK1, a stemness-related kinase, and TRIM2, an E3 ubiquitin ligase, may modulate EV content related to inflammation and immune suppression (69–71). Collectively, these genes represent promising candidates for mediating the differential immune effects of licensed versus naïve MSC-EVs and warrant further functional validation.

Discussion

This study demonstrates the potent immunomodulatory effects of licensed MSC-EVs in regulating T cell and macrophage activity, highlighting their therapeutic potential in modulating immune responses. Our findings show that licensed MSC-EVs significantly inhibited the proliferation of CD3⁺ effector CD4 and CD8 T cells, while also promoting the polarization of pro-inflammatory macrophages into the anti-inflammatory phenotype. Additionally, MSC-EVs effectively upregulated the expression of key immunomodulatory markers, including PD-L1 and CD73, and induced higher frequencies of Tregs. In contrast, naïve MSC-EVs exhibited limited immunomodulatory capacity in our selected doses, emphasizing the enhanced functionality of the licensed effect by dual TGF-β1 and IFN-γ.

These results align well with previous reports demonstrating the immunomodulatory properties of other extracellular vesicles derived from MSCs such as MSC-ApoBDs (19), particularly in their ability to modulate T cell and macrophage activity. However, our study is among the first to directly compare licensed and naïve MSC-EVs, revealing a superior capacity of licensed MSC-EVs to upregulate PD-L1 and promote Treg differentiation. Unlike prior studies that broadly characterized MSC-EVs’ effects, we employed advanced t-SNE-based dimensionality reduction and clustering analysis, providing a detailed “fingerprint” of Treg subpopulations. This approach revealed that licensed MSC-EVs predominantly induce non-proliferated typical Tregs, a finding that has not been previously reported.

The superior immunomodulatory effects of licensed MSC-EVs may be attributed to their enhanced ability to upregulate PD-L1 and CD73 expression. PD-L1 is a well-known immune checkpoint molecule critical for maintaining immune tolerance (47, 48), while CD73 plays a pivotal role in generating extracellular adenosine (72, 73), which suppresses effector T cell activity (74) and promotes Treg differentiation (75, 76). Licensed MSC-EVs’ dual ability to induce Tregs and inhibit effector T cells suggests a coordinated mechanism that modulates both pro- and anti-inflammatory pathways. Additionally, the observation that the majority of Tregs induced by licensed MSC-EVs originated from non-proliferated cells implies that licensed MSC-EVs may facilitate the differentiation of resting T cells into Tregs.

These findings underscore the potential of licensed MSC-EVs as a therapeutic tool for treating immune-related disorders, such as autoimmune diseases, chronic inflammation, and transplant rejection. By selectively enhancing Treg induction and macrophage polarization, licensed MSC-EVs may offer a more targeted and effective approach to restoring immune homeostasis compared to conventional MSC-EVs or other cell-based therapies. Moreover, the ability to modulate key immune checkpoints, such as PD-L1, positions licensed MSC-EVs as a promising candidate for combination therapies in oncology and chronic inflammatory diseases.

Compared with Zhang et al. (77), who demonstrated enhanced Treg induction by exosomes from MSCs stimulated with 1,000 IU/mL IFN-γ and 10 ng/mL TGF-β, our study employed a balanced cytokine dose (50 ng/mL each) and quantified EVs by particle number using NTA. We also adopted size-exclusion chromatography to ensure higher EV purity. While both studies observed increased Treg differentiation following dual cytokine licensing, our results additionally revealed dose-dependent effects on macrophage polarization, IL-10 secretion, NO reduction, and immune checkpoint expression. Notably, functional differences between naïve and licensed MSC-EVs in our study were more pronounced at higher EV doses, highlighting the importance of precise particle-based dosing in functional immunomodulation.

Despite the promising findings, this study has several limitations. First, all experiments were conducted using in vitro models, including the THP-1 monocytic cell line and PBMCs from a limited number of healthy donors, which may not fully recapitulate the complex cellular interactions and microenvironment of immune responses in vivo. Second, the relatively small sample size limits the statistical power and generalizability of the findings, especially when inter-donor variability is considered. Third, while we identified immunomodulatory markers such as LIPG and GPR68 based on publicly available transcriptomic datasets, the underlying molecular mechanisms driving the differences between licensed and naïve MSC-EVs remain incompletely understood. These findings should be further validated using targeted transcriptomic or proteomic profiling of EVs derived from our specific MSC populations, coupled with qPCR-based confirmation to ensure biological relevance and experimental specificity. Forth, the study primarily focused on T cells and macrophages, leaving the effects of MSC-EVs on other immune cell types, such as dendritic cells, B cells, or NK cells, unexamined. Finally, the absence of long-term in vivo validation restricts the current translational relevance of these findings, and future studies should incorporate disease-relevant animal models to assess efficacy, biodistribution, and safety profiles.

Future studies should validate the immunomodulatory efficacy of licensed MSC-EVs in vivo using disease-relevant animal models, such as murine models of uveitis or graft-versus-host disease. To elucidate the underlying mechanisms, integrated proteomic and small RNA sequencing of licensed versus naïve MSC-EVs should be conducted to identify differentially enriched immunoregulatory molecules. In parallel, functional assays targeting additional immune subsets, including dendritic cells, B cells, and NK cells, will help define the broader immunological impact of licensed EVs. Long-term studies assessing biodistribution, pharmacokinetics, and repeated-dose safety in vivo are also necessary to establish a translational foundation. Finally, comparative analyses of EVs generated under different cytokine combinations or licensing durations may inform standardized manufacturing protocols for future clinical applications.

Conclusion

In conclusion, this study demonstrates that MSC-EVs licensed with IFN-γ and TGF-β1 exhibit enhanced immunomodulatory trends compared to naïve MSC-EVs in vitro. These effects include reduced T cell proliferation, increased Treg differentiation, modulation of macrophage polarization, and upregulation of immune checkpoint markers, with several findings being more pronounced at higher EV doses. While these results are promising, further research is needed to clarify the underlying molecular mechanisms, including the identification of key RNA and protein cargo. In vivo studies in disease-relevant models will also be essential to validate the therapeutic potential of licensed MSC-EVs and assess their safety and translational feasibility in the treatment of immune-mediated disorders.

Abbreviations

MSC

Mesenchymal stromal cell

ApoBD

Apoptotic body

EV

Extracellular vesicle

IFN-γ

Interferon-gamma

TGF-β1

Transforming growth factor-beta 1

PBMC

Peripheral blood mononuclear cell

TNF-α

Tumor necrosis factor-alpha

IL-1β

Interleukin-1 beta

Treg

Regulatory T cell

t-SNE

T-distributed stochastic neighbor embedding

FBS

Fetal bovine serum

PBS

Phosphate-buffered saline

MFI

Mean fluorescence intensity

ANOVA

Analysis of variance

PD-L1

Programmed death-ligand 1

IDO

Indoleamine 2,3-dioxygenase

PGE2

Prostaglandin E2

NO

Nitric oxide

DEGs

Differentially expressed genes

NTA

Nanoparticle tracking analysis

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Jiemin Wang, Seyedmohammad Moosavizadeh, Manon Jammes, Abbas Tabasi and Trung Bach. The first draft of the manuscript was written by Jiemin Wang. Aideen E Ryan and Thomas Ritter supervised the study and were responsible to direct and comment the manuscript. All authors read and approved the final manuscript.

Funding

Jiemin Wang acknowledges the financial support from the China Scholarship Council (202006370067). This work was partly funded by the European Union under grant agreement number 101080611 (Restore Vision). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Health And Digital Executive Agency (HADEA). Neither the European Union nor HADEA can be held responsible for them. This work was also supported by Science Foundation Ireland 18/EPSRC-CDT/3583 and the Engineering and Physical Sciences Research Council EP/S02347X/1. This work has also emanated from research supported in part by a grant from Science Foundation Ireland (SFI) and the European Regional Development Fund (ERDF) under grant number 13/RC/2073_P2. Aideen E Ryan is supported through a Research Ireland FFP Award (19/FFP/6446).

Availability of data and materials

The R code used in this study is provided in the supplementary files. The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request. GEO accession numbers GSE122091 and GSE46019 can be accessed at https://www.ncbi.nlm.nih.gov/geo.

Declarations

Ethics approval and consent to participate

Peripheral blood mononuclear cells (PBMCs) were isolated from healthy donors under a research protocol titled “Isolation of PBMCs from healthy donors”, which was approved by the Clinical Research Ethics Committee at University Hospital Galway (Approval number: Ref. C.A. 2534, Date of approval: November 2020). Additionally, human bone marrow-derived mesenchymal stromal cells (MSCs) were obtained under a separate project titled “Isolation of human bone marrow-derived mesenchymal stromal cells”, approved by the Clinical Research Ethics Committee of Galway University Hospital (Approval number: Ref. 08/May/14, Date of approval: May, 2014).

The THP-1 cell line was kindly provided by Dr. Aideen Ryan’s laboratory and was originally purchased from the American Type Culture Collection (ATCC, Catalog No. TIB-202). THP-1 cells were established by ATCC from a human donor with appropriate Institutional Review Board (IRB) approval and informed consent procedures in place.

Consent for publication

Not applicable.

Competing interests

The authors declared no conflicts of interest.