Tonsil-derived mesenchymal stem cells alleviate skin inflammation by modulating neutrophil extracellular trap formation and T cell migration
Hyun Ju Kim, Kyung-Ah Cho, So-Youn Woo
TL;DR
Tonsil-derived stem cells reduce skin inflammation by changing how neutrophils and T cells behave, offering a new treatment approach for inflammatory skin diseases.
Contribution
The study reveals a novel mechanism by which tonsil-derived MSCs modulate neutrophil extracellular traps and T cell migration to alleviate skin inflammation.
Findings
T-MSCs reduced DNCB-induced skin inflammation by decreasing epidermal thickness and neutrophil infiltration.
T-MSCs suppressed neutrophil accumulation in vivo despite enhancing NET formation in vitro.
T-MSCs modulated T cell subset distribution and activation in skin and lymphoid organs.
Abstract
Skin inflammation arises from complex interactions among immune cells, particularly T cells and neutrophils. Mesenchymal stem cells (MSCs) exhibit potent immunomodulatory properties, but the specific roles of tonsil-derived MSCs (T-MSCs) in regulating neutrophil extracellular trap (NET) formation and cell death, as well as T cell migration in inflammatory skin conditions, remain poorly defined. In this study, the therapeutic effects and mechanisms of T-MSCs were investigated in a 2,4-dinitrochlorobenzene (DNCB)-induced skin inflammation model, with a focus on NET formation and T cell migration. T-MSCs were intravenously administered to mice with DNCB-induced skin inflammation; inflammation severity and immune cell dynamics were evaluated using histological analysis, flow cytometry, immunostaining, microarray profiling, NET assays, and T cell migration assays. T-MSC treatment reduced…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Fig 1
Fig 2
Fig 3
Fig 4
Fig 5
Fig 6
Fig 7
Fig 8
Fig 9
Fig 10- —http://dx.doi.org/10.13039/501100003725National Research Foundation of Korea (NRF)
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsNeutrophil, Myeloperoxidase and Oxidative Mechanisms · Mesenchymal stem cell research · Immune cells in cancer
Introduction
Skin inflammation results from complex interactions among immune cells, particularly T cells and neutrophils. Chronic inflammatory skin diseases, such as atopic dermatitis and psoriasis, involve sustained immune activation that contributes to substantial morbidity. To elucidate the mechanisms of immune regulation in skin inflammation, 2,4-dinitrochlorobenzene (DNCB)-induced dermatitis models have been widely utilized because they effectively reproduce allergic contact dermatitis and atopic-like skin inflammation [1,2]. DNCB acts as a hapten that binds to skin proteins, initiating a T cell–mediated immune response and recruiting neutrophils and other immune cells to the affected tissue. Previous studies have demonstrated the immunomodulatory effects of exosomes derived from tonsil mesenchymal stem cells (T-MSCs) in C57BL/6 mice under inflammatory conditions [3]. In the previous study, skin inflammation was induced with DNCB on mice and C57BL/6 mice were used to ensure consistency with established models and enable investigation of T-MSC–mediated immunoregulatory mechanisms in vivo.
Mesenchymal stem cells (MSCs) have emerged as potent immunomodulatory agents. They can differentiate into multiple cell types and regulate immune responses by suppressing T cell proliferation, modulating macrophage polarization, and interacting with neutrophils [4–6]. Among MSCs, T-MSCs provide distinct advantages, i.e., they can be obtained noninvasively from discarded tonsillectomy tissue, display superior proliferation rates, and exhibit broader differentiation potential, making them particularly attractive for clinical applications [7–9]. Furthermore, their low immunogenicity enhances both safety and therapeutic feasibility. T-MSCs have shown robust therapeutic effects in inflammatory skin diseases [10].
Although the general mechanisms of MSC-mediated immunoregulation are well characterized [7,11,12], recent findings indicate that MSCs can suppress excessive neutrophil extracellular trap–associated cell death (NETosis) through multiple pathways, thereby reducing tissue damage [13]. However, these effects vary across MSC sources and inflammatory contexts because the ability to regulate neutrophil functions—including NETosis—depends on the tissue of origin, activation state, and surrounding microenvironment [4].
NETosis is a specialized form of neutrophil cell death in which decondensed chromatin and antimicrobial proteins are released into the extracellular space to form neutrophil extracellular traps (NETs) [14–16]. This process is essential for pathogen defense but, when excessive, promotes tissue injury and chronic inflammation. The direct effects of T-MSCs on NETosis in the context of skin inflammation remain incompletely defined.
In inflammatory skin conditions, NETosis has a dual role, i.e., excessive NET formation exacerbates tissue damage and sustains inflammation, whereas controlled NETosis facilitates pathogen clearance and may aid in resolution [10,17]. Beyond antimicrobial and tissue-damaging functions, NETs influence adaptive immunity by modulating dendritic cell activity and directly priming T cells, thus shaping the magnitude and quality of T cell–mediated inflammation [18,19]. Inflammatory skin diseases often involve NET-associated cytokines such as interleukin-17A (IL-17A) and tumor necrosis factor-alpha (TNF-α), which drive persistent leukocyte infiltration and amplify local inflammation [20]. NETs can also activate plasmacytoid dendritic cells and promote interferon-alpha (IFN-α) production, further intensifying autoimmune responses [21].
Although the immunomodulatory effects of T-MSCs have been demonstrated in models of inflammatory skin disease, where they suppress T and B cell responses and enhance anti-inflammatory cytokine production [10]. However, their direct impact on NETosis and subsequent regulation of T cell migration remains unclear. Despite its potential influence on both acute and chronic phases of inflammation, even the interplay between NETosis and T cell dynamics has not been fully investigated in DNCB-induced dermatitis model [18–21]. Neutrophil NETosis can serve as a central driver of inflammation, but T cell migration to inflamed skin and lymphoid organs also critically determines the persistence of immune responses [22–24]. The mechanisms by which T-MSCs interplay and regulate these processes remain unknown. Elucidation of such mechanisms may provide novel insights into the immunomodulatory functions of T-MSCs and their therapeutic potential for inflammatory skin diseases.
This study investigates the effects of T-MSCs on NETosis and T cell migration in DNCB-induced skin inflammation. By exploring the immunomodulatory mechanisms of T-MSCs, this work aims to establish their therapeutic potential in inflammatory skin diseases and contribute to the development of novel cell-based strategies to correct immune dysregulation in dermatological conditions.
Materials and methods
T-MSC culture
T-MSCs (CLZ-2002), provided by Cellatoz (Sungnam, Korea), were cultured in Dulbecco’s modified Eagle medium (DMEM) low-glucose formulation (Welgene, Gyeongsan, Korea) supplemented with 10% (v/v) fetal bovine serum (FBS; Welgene), 100 U/mL penicillin, and 100 µg/mL streptomycin (Capricorn Scientific, Ebsdorfegrund, Germany). Cells were maintained at 37°C in a humidified atmosphere containing 5% CO_2_, and the culture medium was replaced every 3 days. For subculture, cells were washed once with phosphate-buffered saline (PBS; 137 mM sodium chloride, 2.7 mM potassium chloride, 10 mM sodium phosphate, 1.76 mM potassium phosphate, pH 7.4; Welgene). One milliliter of trypsin–ethylenediaminetetraacetic acid (EDTA; Welgene) was then added to each flask; this was followed by incubation for 10 minutes at 37°C. Detached cells were collected and centrifuged at 1,200 rpm for 5 minutes. T-MSCs between passages 7 and 9 were used for all experiments.
T-MSC differentiation
For adipogenic differentiation, T-MSCs were seeded at a density of 1 × 10^4^ cells/mL in 12-well plates and cultured for 21 days in StemPro™ Adipocyte Differentiation Basal Medium supplemented with StemPro™ Adipogenesis Supplement (A1007001, Gibco). After differentiation, cells were fixed with 4% formalin for 5 minutes at room temperature and stained with Oil Red O working solution (00625, Sigma-Aldrich, St. Louis, MO, USA) for 10 minutes. For osteogenic differentiation, T-MSCs were seeded at the same density and cultured for 21 days in StemPro™ Osteocyte/Chondrocyte Differentiation Basal Medium supplemented with StemPro™ Osteogenesis Supplement (A1007201, Gibco). Cells were fixed with 60% isopropyl alcohol for 5 minutes and stained with 2% Alizarin Red S solution (pH 4.2) (A5533; Sigma-Aldrich) for 3–5 minutes. Cells were visualized under a microscope (CKX41; Olympus Corporation, Tokyo, Japan). For chondrogenic differentiation, T-MSCs were cultured in StemPro™ Osteocyte/Chondrocyte Differentiation Basal Medium supplemented with StemPro™ Chondrogenesis Supplement (A1007101, Gibco). To establish three-dimensional pellet culture, T-MSCs were seeded at a density of 1.5 × 10^6^ cells/mL in 15-mL tubes and cultured for 21 days. Sections were stained with Alcian Blue solution (A5268, Sigma-Aldrich) for 30 minutes, and counterstained with Nuclear Fast Red solution (N8002; Sigma-Aldrich) for 5 minutes. Slides were then cleared in xylene and mounted with resinous medium (Shandon Synthetic Mountant™, 6769007; Thermo Fisher Scientific). Images were acquired using a DP28 microscope (Olympus Corporation).
Cell culture
HL-60 cells (human promyelocytic leukemia cell line; American Type Culture Collection [ATCC] CCL-240) were maintained at 37°C in a humidified 5% CO_2_ atmosphere in Roswell Park Memorial Institute (RPMI) 1640 medium (Welgene) supplemented with 10% (v/v) FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin. Neutrophil differentiation was induced by culturing HL-60 cells for 4 days in the presence of 0.8% dimethylformamide (DMF) [25]. To confirm differentiation, cytospin preparations were made using a Cytospin 3 Centrifuge (Thermo Shandon, Thermo Fisher Scientific) at 800 rpm for 5 minutes. Slides were fixed in methanol (Merck, Darmstadt, Germany) for 1 minute, stained with Giemsa solution (48900; Sigma-Aldrich) for 20 minutes, rinsed with distilled water, mounted, and examined under a DP28 microscope.
Jurkat cells (human T cell leukemia line; ATCC TIB-152) were maintained at 37°C in a 5% CO_2_ atmosphere in RPMI 1640 medium supplemented with 10% (v/v) FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin.
Flow cytometry analysis
To confirm that the cultured T-MSCs displayed MSC characteristics, flow cytometry was performed using the following surface markers: anti-human CD11b, anti-human CD34, anti-human CD45, anti-human CD73, anti-human CD105 and anti-human CD90 antibodies. Isotype-matched controls were included for comparison. To verify neutrophil differentiation of HL-60 cells, flow cytometry was conducted using neutrophil-specific markers: anti-human CD11b, anti-human CD35, and anti-human CD71 antibodies. T cell subsets were analyzed with the following antibodies from isolated cells from skin, LNs, and SP: anti-mouse CD4, anti-mouse IFN-γ, anti-mouse IL-4, anti-mouse IL-17, and anti-mouse Foxp3 antibodies. Activation and adhesion markers of the T cells were analyzed with anti-mouse CD62L, anti-mouse CD69, and anti-mouse CD49d antibodies. For neutrophil identification in skin, anti-mouse Ly-6G and anti-mouse CD11b antibodies were used. Isotype-matched controls were included for each antibody.
All antibodies were purchased from BioLegend (San Diego, CA, USA) and antibodies were listed in Table 1. For surface staining prior to analysis, cells were pelleted and resuspended at a density of 0.5–1 × 10^6^ cells in 100 µL of flow cytometry buffer (PBS, pH 7.5, containing 0.1% sodium azide and 0.5% FBS). Antibodies (0.2–0.5 µg) were added directly to the suspension and incubated on ice for 30 minutes. After incubation, cells were washed with 1 mL of flow cytometry buffer and centrifuged at 1,200 rpm for 5 minutes. The supernatant was discarded, and the cell pellet was resuspended in 400–500 µL of fresh flow cytometry buffer before transfer to polypropylene tubes for analysis. All samples were analyzed using an ACEA NovoCyte 3000 flow cytometer (ACEA Biosciences, San Diego, CA, USA) equipped with the NovoSampler Pro Combo. Data were processed with NovoExpress software (version 1.6.2, ACEA Biosciences).
Table 1: List antibodies used in this study.
DNCB-induced skin inflammation model
Female C57BL/6J mice, 8 weeks old, were purchased from RaonBio (Seoul, South Korea). Mice were housed under specific pathogen-free (SPF) conditions at 21–23°C with 51–54% humidity, maintained on a 12-hour light/dark cycle, and provided food and water ad libitum. All procedures were approved by the Ewha Womans University College of Medicine Animal Care and Use Committee (EWHA MED IACUC 21-031-6). Mice were randomly assigned to four groups: vehicle control, vehicle + T-MSCs, DNCB (138630; Sigma-Aldrich), and DNCB + T-MSCs. A 1% DNCB solution was prepared in a mixture of acetone (179124; Sigma-Aldrich, 75%) and olive oil (25%). The vehicle solution consisted of the same acetone/olive oil mixture without DNCB. Each group received 200 µL of the respective solution applied to shaved dorsal skin on days 1, 2, 3, 8, and 9. On day 2, T-MSCs (5 × 10^5^ cells in 200 µL PBS) were administered by intravenous injection via the tail vein to the designated groups. Vehicle groups received 200 µL PBS by intravenous injection as controls. For the drug-treated group, the mice were housed individually to prevent attacks. Humane endpoints were established such that any mouse exhibiting a body weight loss exceeding 20% of its original weight during the experiment would be euthanized using carbon dioxide (30–70% volume/min). Mouse body weight was recorded daily to monitor changes associated with treatment or inflammation. After 10 days, the mice were euthanized using carbon dioxide, and skin, inguinal and axillary lymph node (LN), and spleen (SP) tissues were collected.
Hematoxylin and eosin (H&E) staining
To evaluate histological changes in skin, LNs, and SP among the four groups, tissues were fixed in 4% paraformaldehyde and submitted to DK Korea (Seoul, Republic of Korea) for paraffin embedding, sectioning, and H&E staining. Histological features, including epidermal hyperplasia, medullary sinus expansion, and follicular structure, were assessed as indicators of inflammation. Slides were examined using a DP28 microscope; skin sections were imaged at 100 × magnification, whereas LN and SP sections were viewed at 40 × magnification. Tissue parameters were quantified using ImageJ software (version 1.54, https://imagej.nih.gov/ij/download.html). Epidermal thickness was measured in skin sections, medullary sinus area was documented in LN sections, and the proportion of B cell follicle area relative to total area was recorded in SP sections.
Cell isolation
Dorsal skin was excised and placed in 3–4 mL of enzyme solution consisting of 6 mg/mL collagenase type I (9001-12-1; StemCell Technologies, Vancouver, BC, Canada), 10 µg/mL DNase (New England Biolabs, Ipswich, MA, USA), 10% FBS, and 1% penicillin/streptomycin in RPMI 1640, in a six-well plate. After 10 minutes of incubation at 37°C, tissue was mechanically dissociated using scissors and a syringe plunger. The suspension was transferred to a flask, supplemented with enzyme solution to a final volume of 10 mL, and stirred for 1 hour at 37°C. The suspension was then filtered through a 70-µm cell strainer (93070; SPL Life Sciences, Pocheon-si, Gyeonggi-do, South Korea), centrifuged at 1,200 rpm for 5 minutes at room temperature, and treated with ammonium–chloride–potassium (ACK) lysis buffer (8.26 g NH_4_Cl, 1 g KHCO_3_, and 37.2 mg Na_2_EDTA dissolved in 850 mL H_2_O, pH adjusted to 7.2–7.4, final volume 1,000 mL) to lyse red blood cells. Cells were washed twice with tissue buffer (10% FBS, 10 mM EDTA, 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid [HEPES], 100 U/mL penicillin, 100 µg/mL streptomycin, and 1 mM sodium pyruvate in PBS), centrifuged, and resuspended for subsequent analysis. LNs and SP were isolated and placed in tissue buffer. Tissues were mechanically dissociated through a 70-µm cell strainer, centrifuged at 1,200 rpm for 5 minutes at 4°C, treated with ACK lysis buffer for 2–3 minutes at room temperature, and washed twice with tissue buffer prior to cell counting. For intracellular staining in flow cytometry, cells were permeabilized in PBS containing 0.5% saponin and 1% bovine serum albumin (BSA; BSAS01, BOVOGEN, Vienna, Austria).
Immunohistochemistry
For immunohistochemical detection of myeloperoxidase (MPO) in skin sections, paraffin-embedded tissues were deparaffinized in xylene (JUNSEI Chemical, Tokyo, Japan) twice for 5 minutes each, followed by rehydration through a graded ethanol series (100%, 95%, 90%, 80%, 70%) for 3 minutes each. Slides were immersed in distilled water for 10 minutes and washed in PBS using a programmable digital rocker (RK-1D; DAIHAN Scientific, Seoul, South Korea) at 30 rpm. Endogenous peroxidase activity was blocked with Peroxidase Blocking Solution (S2023; Dako, Santa Clara, CA, USA) for 30 minutes at room temperature in a humidified chamber. After washing in PBS for 5 minutes, protein blocking was performed with Protein Block (X0909; Dako) for 15 minutes at room temperature. Sections were incubated overnight at 4°C with anti-MPO primary antibody (1:8000). The following day, slides were equilibrated to room temperature for 1 hour and washed with PBS for 5 minutes. Secondary antibody incubation was performed for 20 minutes using the rabbit alkaline phosphatase-conjugated secondary antibody included in the ENZO kit (POLYVIEW® IHC reagent, ADI-950-113-0100; ENZO Life Sciences, Farmingdale, NY, USA). Slides were developed with the stable alkaline phosphatase substrate provided in the same kit. After substrate development, sections were counterstained with hematoxylin for 60 seconds, washed under running tap water for 5 minutes, dipped three times in 1% acid alcohol (100% ethanol + 1% HCl), and washed again under running water for 5 minutes. Sections were dehydrated through a graded ethanol series (70%, 80%, 90%, 95%, and 100%) for 3 minutes each, cleared in xylene for 5 minutes, and mounted with synthetic mountant. Images were acquired at 40× and 200 × magnification using a DP28 microscope.
Immunofluorescence staining
To investigate immune cell migration within the skin, staining for cutaneous lymphocyte antigen (CLA) was performed. Paraffin was removed from mouse tissue sections by immersing slides twice in xylene for 5 minutes each, followed by rehydration through a graded ethanol series (100%, 95%, 90%, 80%, and 70%) for 3 minutes each. Slides were then immersed in distilled water for 10 minutes and washed in PBS for 5 minutes. To block nonspecific binding, slides were incubated in 1% BSA in PBS containing 0.1% Tween-20 (P1379; Sigma-Aldrich) for 30 minutes. The primary anti-CLA antibody (1:50, Biolegend) was applied and incubated overnight at 4°C. The following day, slides were equilibrated to room temperature for 1 hour, rinsed three times with PBS for 5 minutes each, and incubated with goat anti-rat IgM µ-chain secondary antibody (1:100, Abcam) in the dark for 1 hour at room temperature. After three additional washes with PBS in the dark at 30 rpm, slides were mounted with DAPI, sealed with cyanoacrylate adhesive, and imaged at 40× and 100 × magnification using a DP28 microscope.
Microarray data analysis
Gene expression profiling was performed using Affymetrix Mouse Clariom-S arrays by Macrogen Inc. (Seoul, Korea). Total RNA extracted from SP samples of vehicle, vehicle + T-MSCs (V + T), DNCB, and DNCB + T-MSCs (D + T) groups was analyzed. Raw data (CEL files) were processed with Affymetrix Power Tools using the Signal Space Transformation–Robust Multichip Analysis (SST-RMA) algorithm for background correction, normalization, and probe summarization. Differentially expressed genes (DEGs) were identified using thresholds of absolute fold change ≥ 1.5 and raw p-value < 0.05. Hierarchical clustering was performed with Euclidean distance and complete linkage. Functional annotation and enrichment analysis were carried out using g:Profiler (https://biit.cs.ut.ee/gprofiler/) to identify significantly enriched Gene Ontology (GO) terms in biological process, cellular component, and molecular function categories. The Database for Annotation, Visualization and Integrated Discovery (DAVID) was used to analyze enriched biological processes and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. For DAVID analysis, the top 100 upregulated and top 100 downregulated genes, ranked by fold change, were selected. Visualization of enriched GO terms and pathways was performed in R (version 3.3.3). Top enriched terms were displayed as dot plots, with gene ratio, adjusted p-value, and intersection size indicated. All gene analysis data were deposited in National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO accession ID: GSE315896).
NET formation assay
To evaluate NETosis in differentiated HL-60 (dHL-60) cells, a fluorescence-based assay was performed using NucGreen™ Dead 488 ReadyProbes™ Reagent (SYTOX™ Green Nucleic acid stain (5 mM in DMSO); Thermo Fisher Scientific) [26]. Cells were incubated with two drops of NucGreen Dead 488 Reagent per milliliter of medium for 15 minutes in the dark (0.5 mM of final concentration). After incubation, the staining solution was removed, and cells were washed 2 or 3 times with PBS (1,200 rpm, 5 minutes). Stained cells were then prepared as cytospin slides, fixed in methanol for 1 minute, and mounted with coverslips. Fluorescence images were acquired using a DP28 microscope. NETosis was quantified by analyzing SYTOX™ Green–stained extracellular DNA structures with ImageJ software (version 1.54p) and fluorescence-positive areas were regarded as indicators of NET formation.
Transwell migration assay
A Transwell migration assay was performed using Corning® 24 mm Transwell® inserts with 8.0-µm pore polycarbonate membranes or Corning® 6.5 mm Transwell® inserts with 5.0-µm pore membranes. T-MSCs were seeded in the lower chamber at a concentration of 1 × 10^6^ cells/mL in low-glucose DMEM supplemented with 10% (v/v) FBS, 100 U/mL penicillin, and 100 µg/mL streptomycin, and incubated overnight. Before the addition of dHL-60 cells, medium in the lower chamber was replaced with RPMI 1640. The upper chamber was then loaded with HL-60 or dHL-60 cells, with or without DNCB treatment, at 1 × 10^5^ cells/mL in RPMI 1640. After 3 hours of incubation, Jurkat cells were added to the upper chamber at 1 × 10^5^ cells/mL in RPMI 1640. The assay proceeded for an additional 3 hours. At the end of the incubation period, Jurkat cells that had migrated to the lower chamber were collected and analyzed by flow cytometry. Anti-CD3 antibody (BioLegend) staining was conducted to identify migrated Jurkat cells.
Preparation of conditioned media (CM)
T-MSCs were cultured to approximately 80% confluence. Cells were harvested and centrifuged at 1,200 rpm for 5 minutes. The pellet was resuspended in PBS, and cell counts were determined. T-MSCs were then seeded at a density of 1 × 10^7^ cells/mL in serum-free, antibiotic-free low-glucose DMEM and cultured for 48 hours. After incubation, the culture suspension was collected and centrifuged at 1,200 rpm for 5 minutes. The supernatant was carefully recovered, avoiding the cell pellet.
For downstream assays, HL-60 cells were seeded at 1 × 10^5^ cells/mL in serum-free, antibiotic-free RPMI 1640 medium and incubated for 3 hours as untreated controls. dHL-60 cells were also seeded at 1 × 10^5^ cells/mL in the same medium and cultured for 3 hours under the following conditions: untreated control, treatment with 1 mL of T-MSC CM, or treatment with phorbol 12-myristate 13-acetate (PMA, P8139; Sigma-Aldrich, 20 nM), GSK 484 hydrochloride (1652591-81-5; Axon Medchem, Groningen, The Netherlands, 10 nM), or diphenyleneiodonium chloride (DPI, BML-CN240; ENZO Life Sciences, 10 nM). All treatments were applied at the seeding step. After 3 hours of incubation, cell suspensions were collected and centrifuged at 1,200 rpm for 5 minutes. Supernatants were carefully recovered, avoiding the pellets. The collected supernatants were concentrated using Omega membrane filters (Pall, Port Washington, NY, USA) via centrifugation at 5,300 × g for 1 hour with a Sorvall LYNX4000 centrifuge (Thermo Fisher Scientific). Concentrated CM was collected from the membrane and stored in 5-mL tubes at –80°C until further use.
Western blot analysis
Samples of HL-60 CM, dHL-60 + PMA CM, dHL-60 + T-MSC CM, dHL-60 + GSK 484 CM, dHL-60 + T-MSC + GSK 484 CM, dHL-60 + DPI CM, and dHL-60 + T-MSC + DPI CM were prepared with 6 × loading dye. Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis using 10 µL of sample per lane. The stacking gel was run at 85 V, and the separating gel was run at 100 V. Proteins were transferred to polyvinylidene fluoride membranes at 100 V for 1 hour. Membranes were blocked with 3% skim milk (30620074−1; Bioworld, Downers Grove, IL, USA) in TBST (40 mL 1 M Tris pH 8.0, 120 mL 5 M NaCl, 4 mL Tween 20) for 30 minutes with gentle agitation at 10 rpm, followed by a brief rinse in TBST. Primary antibodies were diluted in 3% BSA with 0.02% sodium azide in TBST. Membranes were incubated overnight at 4°C with gentle agitation (10 rpm) with one of the following primary antibodies: anti-histone H3 (citrulline R26, Abcam), or anti-β-actin (Santa Cruz Biotechnology, CA, USA). After primary antibody incubation, membranes were washed three times in TBST for 10 minutes each at 43 rpm. Secondary antibody incubation was performed with goat anti-rabbit IgG (H + L)-HRP (Bio-Rad, Hercules, CA, USA, 1:3000) or goat anti-mouse IgG (H + L)-HRP (Bio-Rad, 1:3000) in TBST for 1 hour at room temperature with gentle agitation (10 rpm). Membranes were then washed three times in TBST for 10 minutes each at 43 rpm. Protein bands were visualized using the EZ-Western Lumi Femto Kit (DG-WF100; DoGenBio). Equal volumes of solutions A and B were mixed and applied evenly to the membranes. Images were captured with a LAS-3000 image analyzer (Fujifilm, Tokyo, Japan).
Statistical analysis
All statistical analyses were performed using GraphPad Prism (version 10; GraphPad Software, San Diego, CA, USA). Data are expressed as mean ± standard deviation (SD) unless otherwise noted. Two-way analysis of variance (ANOVA) was used to evaluate differences among groups, followed by Tukey’s post hoc test for multiple comparisons. A p-value < 0.05 was considered statistically significant.
Results
Characterization of T-MSCs
To confirm the MSC identity and multi-lineage differentiation potential of T-MSCs, a series of characterization assays was performed. T-MSCs exhibited typical MSC features, including fibroblast-like, spindle-shaped morphology, adherence to plastic surfaces, and formation of a homogeneous monolayer (Fig 1A). Flow cytometry analysis demonstrated that T-MSCs were positive for CD73, CD90, and CD105; they were negative for CD11b, CD34, and CD45 (Fig 1B). Under differentiation conditions, T-MSCs successfully differentiated into adipogenic, osteogenic, and chondrogenic lineages, as confirmed by Oil Red O, Alizarin Red S, and Alcian Blue staining, respectively (Fig 1C).
Characteristics and multilineage differentiation potential of T-MSCs.(A) Morphology of cultured T-MSCs observed under light microscopy (original magnification, 200×). (B) Flow cytometric analysis of surface marker expression in T-MSCs. Histograms show absence of CD11b, CD34, and CD45 (negative markers) and expression of CD73, CD90, and CD105 (positive markers). Red areas represent isotype controls; green areas represent specific antibody staining. (C) Multilineage differentiation potential of T-MSCs. Left: adipogenic differentiation visualized by Oil Red O staining of lipid droplets. Center: osteogenic differentiation shown by Alizarin Red S staining of calcium deposits. Right: chondrogenic differentiation demonstrated by Alcian Blue staining of proteoglycans (original magnification, 200×).
Characterization of dHL-60 neutrophils
HL-60 cells efficiently differentiated into neutrophil-like cells after treatment with 0.8% DMF for 4 days (Fig 2A). Giemsa staining revealed the characteristic multinucleated morphology of neutrophils in dHL-60 cells (Fig 2B). Flow cytometry analysis confirmed neutrophil differentiation, with increased expression of CD11b (51.4%) and CD35 (73.2%) and decreased expression of CD71 (98.4%) relative to undifferentiated HL-60 cells (Fig 2C).
Neutrophil-like differentiation of HL-60 cells.(A) Morphological changes in HL-60 cells during differentiation. Cells were treated with 0.8% dimethylformamide (DMF) and observed at Day 0 (undifferentiated) and Day 4 (differentiated). Images were captured using an Olympus DP28 microscope (original magnification, 200×). (B) Giemsa staining of differentiated HL-60 cells. Cells were collected by cytospin centrifugation, fixed in methanol, and stained with Giemsa solution. Images were captured using an Olympus DP28 microscope (original magnification, 200×). Differentiated HL-60 cells are indicated by arrows. Scale bar = 10 µm. (C) Flow cytometric analysis of neutrophil surface markers. Expression of CD11b, CD35, and CD71 was compared between Day 0 (undifferentiated) and Day 4 (differentiated) HL-60 cells.
DNCB-induced skin inflammation model and effects of T-MSC treatment
To investigate the therapeutic effects of T-MSCs on DNCB-induced skin inflammation, an in vivo mouse model was established. The experimental timeline, including DNCB application and T-MSC administration, is shown in Fig 3A. Body weight was monitored throughout the study. Compared to day 1, DNCB-treated mice exhibited an 8.1% body weight reduction by day 3, while DNCB + T-MSC-treated mice showed a 7.9% reduction on day 3 and 10% on day 4, respectively. By day 5, DNCB-treated mice recovered only 3% of their lost weight, whereas DNCB + T-MSC-treated mice showed a 5% weight gain, indicating faster recovery with T-MSC treatment (Fig 3B).
Effects of T-MSCs in the DNCB-induced skin inflammation model.(A) Experimental scheme showing the timeline of DNCB application and T-MSC injection in the mouse model of skin inflammation. DNCB or vehicle was applied on days 1, 2, 3, 8, and 9. T-MSCs or PBS were administered intravenously on day 2. Mice were sacrificed on day 10 for analysis. (B) Body weight changes over 10 days following DNCB treatment and T-MSC injection. The graph shows body weight as a percentage of baseline (day 0) for each group (n = 4 per group). (C) Representative photographs of dorsal skin from each group taken at sacrifice on day 10. (D) Macroscopic images of lymph nodes (LN) and spleens (SP) harvested from each group. (E) Quantification of LN length across groups. DNCB treatment significantly increased LN length compared with vehicle (n = 2 per group). (F) Quantification of SP length across groups. Data were analyzed using two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison ( P < 0.05 and ** P < 0.01). Vehicle (V), vehicle + T-MSCs (V + T), DNCB (D), and DNCB + T-MSCs (D + T).*
Gross examination of dorsal skin revealed distinct differences among groups (Fig 3C). DNCB-treated mice developed visible signs of inflammation, including erythema and scaling, which were attenuated in the DNCB + T-MSC group. These findings suggest that T-MSC administration partially ameliorated the skin inflammation induced by DNCB.
The LNs and SP were also examined to assess the systemic effects of treatment (Fig 3D). Organ size was quantified by measuring LN and SP lengths (Fig 3E and 3F). Mice in the DNCB group exhibited enlargement of both LN and SP compared with vehicle-treated controls, whereas T-MSC treatment attenuated this enlargement. Specifically, LN length in the DNCB + T-MSC group (3.45 ± 0.64 mm) was slightly greater than that in the vehicle group (2.55 ± 0.35 mm, not statistically significant) but was significantly reduced compared with DNCB group (5.65 ± 0.07 mm).
Similarly, DNCB treatment significantly increased SP length (20. 1 ± 1.7 mm) relative to vehicle group (14.3 ± 0.57 mm). Although SP length in the DNCB + T-MSC group (16.9 ± 1.13 mm) remained greater than that in the vehicle group (not statistically significant), it was significantly reduced compared with the DNCB group.
These findings indicate that DNCB induces pronounced inflammatory changes not only in the skin and but also in secondary lymphoid organs, and T-MSC treatment mitigates inflammation in the skin, SP, and LNs, supporting its potential therapeutic role in DNCB-induced skin inflammation and associated systemic immune responses. In contrast, mice treated with T-MSC alone did not exhibit significant increase in LN (2.85 ± 0.6 mm) or SP (14.3 ± 0.6 mm) size compared with vehicle-treated controls.
To further assess tissue-specific alterations, histological analysis was performed using H&E staining (Fig 4). Skin sections (Fig 4A) from DNCB-treated mice exhibited typical inflammatory features, including epidermal thickening and immune cell infiltrations. Quantification of epidermal thickness (Fig 4D) confirmed a significant increase in the DNCB group (2.07 ± 0.69 um) compared with the vehicle group (0.43 ± 0.11 um). Treatment with T-MSCs (1.01 ± 0.26 um) reduced epidermal thickness relative to the DNCB group, indicating an anti-inflammatory effect, with a trend toward normalization.
Histological analysis of inflammatory features by H&E staining.(A) Representative H&E-stained dorsal skin sections from each group (Vehicle, Vehicle + T-MSCs, DNCB, and DNCB + T-MSCs) at 100 × magnification. Scale bar = 100 µm. (B) H&E-stained LN sections from each group at 4 × magnification. Scale bar = 50 µm. (C) H&E-stained spleen (SP) sections from each group at 4 × magnification. Scale bar = 50 µm. (D) Quantification of epidermal thickness across groups. Twenty measurements per group (5 per slide, 4 slides; n = 20) were included. (E) Quantification of LN medullary sinus area as a percentage of total LN area. Three slides per group (n = 3) were analyzed. (F) Quantification of splenic follicle area as a percentage of total SP area. Eight slides per group (n = 8) were analyzed. Images were analyzed with ImageJ, and results were analyzed by two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison ( P < 0.05, ** P < 0.01, ***P < 0.001 and ***P < 0.0001). ns: not significant. Vehicle (V), vehicle + T-MSCs (V + T), DNCB (D), and DNCB + T-MSCs (D + T).
LN sections (Fig 4B) revealed structural alterations, particularly within the medullary sinus region. DNCB treatment resulted in LN enlargement with expansion of the medullary areas. Quantitative analysis (Fig 4E) confirmed a significant increase in medullary sinus area in the DNCB group (7.0 ± 1.4%) compared with the vehicle group (0.6 ± 0.4%). In the DNCB + T-MSC group, the medullary sinus area showed a further increase (7.73 ± 2.91%) relative to the DNCB group; however, this difference did not reach statistical significance.
Similarly, SP sections (Fig 4C) demonstrated alterations in the organization of B cell follicular regions. DNCB treatment disrupted splenic architecture, particularly within follicular areas. Quantification of follicular area relative to total spleen area (Fig 4F) revealed a reduction in DNCB-treated mice (11.6 ± 3.1%) compared with the vehicle group (22.2 ± 9.9%). The DNCB + T-MSC group (12.2 ± 5.6%) exhibited a modest increase in follicular area compared with the DNCB group, suggesting partial restoration of splenic architecture and a trend toward normalization, although the effect did not reach statistical significance.
Collectively, these results demonstrate that DNCB induces pronounced histological alterations in the skin and secondary lymphoid tissues. T-MSC treatment significantly alleviated skin inflammation, whereas effects observed in the SP and LNs were modest and did not reach statistically significance, indicating variable or limited impacts on these organs.T cell subsets and activation markers in the DNCB-induced skin inflammation model
Flow cytometry was performed to analyze CD4 + T cell subsets (Th1, Th2, Th17, and Treg) and activation/adhesion markers (VLA4, CD69, and CD62L) in the skin, LNs, and SP of mice subjected to DNCB treatment with or without T-MSC administration (Fig 5).
Countined.
In the skin, analysis of CD4 + T cell subset revealed that Th1 and Th2 populations were significantly reduced in the DNCB + T-MSC group (2.22 ± 0.14% and 2.62 ± 0.19%, respectively) (Fig 5A). Compared with the vehicle group, DNCB group did not show significant increases in Th1 (4.14 ± 0.19%), Th2 (4.42 ± 0.20%), Th17 (1.51 ± 0.26%) nor Treg (1.50 ± 0.13%) populations. Notably, Th2 cells were markedly increased in the skin of the in T-MSC-only group (7.15 ± 0.04%). Th17 and Treg subsets exhibited decreasing trends in the DNCB + T-MSCs group; however, these changes did not reach statistical significance.
In the LNs, CD4 + T cell subset analysis demonstrated significant increases in Th1 (3.42 ± 0.44%), Th17 (3.25 ± 0.15%), and Treg (4.22 ± 0.28%) populations following DNCB exposure (Fig 5B). In DNCB-treated mice receiving T-MSCs, Th1 (3.80 ± 0.65%) and Th2 (6.04 ± 1.17%) populations showed slight increases, whereas Th17 (2.95 ± 0.35%) and Treg (3.71 ± 0.31%) populations were modestly reduced compared with DNCB-only group; however, these differences were not statistically significant.
In the SP, CD4 + T cell subsets did not exhibit statistically significant changes across groups (Fig 5C). Nevertheless, Th1 and Th2 populations tended to increase following DNCB treatment (1.86 ± 0.32% and 2.49 ± 0.88%, respectively) and decrease after DNCB + T-MSC treatment (1.54 ± 0.40% and 1.80 ± 0.19%), whereas Th17 and Treg populations showed the opposite pattern.
Assessment of activation and adhesion markers on CD4 + T cells revealed marked reduction in VLA4 + cells in the skin across the T-MSC-only (3.56 ± 1.65%), DNCB (2.70 ± 0.82%), and DNCB+T-MSCs (1.42 ± 0.23%) groups compared with the vehicle group (13.67 ± 2.75%) (Fig 5D). In the LN, the DNCB group exhibited a decrease in VLA4 + cells (4.13 ± 3.67%) relative to vehicle controls (14.61 ± 1.28%). Although the DNCB + T-MSCs group showed a partial increase in VLA4 + cells (7.96 ± 1.57%) compared with the DNCB-only group, this difference was not statistically significant.
Similarly, the proportions of CD69 + CD4 + T cells were reduced in both the DNCB-only (8.97 ± 0.71%) and DNCB+T-MSC (10.17 ± 1.02%) groups compared with the vehicle group (17.14 ± 1.28%). CD62L + CD4 + T cells were decreased in the T-MSC-only (16.61 ± 2.72%), DNCB (10.07 ± 0.98%), and DNCB+T-MSC (10.57 ± 1.33%) groups relative to vehicle controls (28.87 ± 0.91%). Overall, expression of activation and adhesion markers on LN CD4 + T cells showed a trend toward recovery following T-MSC administration in DNCB-treated mice, although these changes did not reach statistical significance (Fig 5E).
In spleen, among CD4 + T cells, VLA4+ (2.83 ± 0.41%) and CD62L+ (4.85 ± 1.10%) populations were reduced follwing DNCB treatment compared with the T-MSC-only group (4.73 ± 0.88% for VLA4 and 7.85 ± 0.95% for CD62L), but did not differ significantly from vehicle controls (3.74 ± 0.63% for VLA4 and 6.24 ± 0.77% for CD62L) (Fig 5F). Collectively, these results suggest that T-MSCs exert immunomodulatory effects in DNCB-induced inflammation by suppressing T cell activation in the skin, altering T cell subset distribution in lymphoid organs, and partially restoring activation marker expression in LNs and SP.
To specifically identify skin-homing T cells infiltrating the skin, immunofluorescence staining for CLA was performed on skin sections (Fig 6A). Quantitative image analysis using ImageJ revealed a marked increase in CLA+ cells following DNCB-induced inflammation (31.0 ± 3.0) compared with vehicle controls (4.3 ± 0.6), which was further enhanced by T-MSC administration (41.3 ± 2.1). In contrast, the T-MSCs-only group (9.3 ± 1.5 cells) did not exhibit a significant increase in CLA+ cells (Fig 6B).
Immunofluorescence analysis of cutaneous lymphocyte antigen (CLA).(A) Representative immunofluorescence images of skin sections from each group (Vehicle, Vehicle + T-MSCs, DNCB, and DNCB + T-MSCs) stained for CLA (red) and DAPI (blue). (B) Quantification of CLA+ cells per field of view across groups. Data were analyzed using two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison (* P < 0.01 and ***P < 0.0001).
Microarray analysis revealed that T-MSCs modulate inflammatory and neutrophil-related pathways in DNCB-induced skin inflammation
To investigate molecular mechanisms underlying the effects of T-MSCs on DNCB-induced skin inflammation, microarray analysis was performed on SP samples from vehicle, vehicle + T-MSC, DNCB, and DNCB + T-MSC groups. Hierarchical clustering of gene expression profiles revealed distinct patterns among the four groups, indicating that DNCB treatment markedly altered gene expression compared with the vehicle group, and that T-MSC administration further modulated this expression profile (Fig 7A). These results demonstrate that T-MSCs exert a substantial influence on gene expression in DNCB-induced skin inflammation, supporting their potential therapeutic role.
Continued.
Gene ontology analysis showed that both DNCB and DNCB + T-MSC groups exhibited increased expression of genes associated with inflammatory responses, innate immunity, and chemotaxis compared with the vehicle group (Fig 7B). Several inflammatory response-related genes were upregulated in both the DNCB and DNCB+T-MSC groups, including Camp (cathelicidin, NM_009921), Elane (neutrophil expressed elastase, NM_015779), S100a8 (calgranulin A, NM_013650), Chil3 (chitinase-like 3, NM_009892), Anxa1 (Annexin A1, NM_010730), and S100a9 (calgranulin B, NM_001281852). In contrast, Hp (haptoglobin, NM_017370), Lbp (lipopolysaccharide binding protein, NM_008489), and Alox5 (arachidonate 5-lipoxygenase, NM_009662) were selectively upregulated in the DNCB group. Inflammatory response-related genes that were downregulated in both the DNCB and DNCB+T-MSC group included Cxcl9 (NM_008599), and Dusp10 (dual-specificity phosphatase 10, NM_022019). In addition, Il1b (interleukin 1 beta, NM_008361) was specifically downregulated in the DNCB + T-MSC group (data not shown).
Notably, the DNCB + T-MSC group uniquely exhibited upregulation of apoptosis- and neutrophil-associated genes, including S100a8, S100a9, Lcn2 (lipocalin 2, NM_008491), and Pglyrp1 (peptidoglycan recognition protein 1, NM_009402), with Lcn2 showing the most pronounced increase (Fig 7C). These genes are closely associated with neutrophil function and the regulation of inflammatory responses.
KEGG pathway analysis revealed that both DNCB and DNCB + T-MSC groups displayed enhanced activity in the IL-17 signaling and neutrophil extracellular trap (NET) formation pathways (Fig 7D). In contrast, the DNCB + T-MSC group exhibited reduced activity in several inflammation-associated pathways, including T cell receptor signaling, leukocyte transendothelial migration, and NF-κB signaling. Direct comparison between the DNCB and DNCB + T-MSC groups confirmed that T-MSC treatment modulated pathways related to immune responses and intracellular signaling. Moreover, the DNCB + T-MSC group showed increased activity in nucleocytoplasmic transport and autophagy pathways, which may contribute to regulation of inflammatory processes (Fig 7E and 7F).
T-MSCs modulated neutrophil infiltration and activation in DNCB-induced skin inflammation
To evaluate the effects of T-MSCs on neutrophil infiltration and activation in DNCB-treated mice, skin samples were analyzed by immunohistochemistry and flow cytometry. Immunohistochemical staining for MPO revealed a marked increase in MPO expression in DNCB-treated skin (4.33 ± 0.20) compared with the vehicle group (0.34 ± 0.01) (Fig 8A and 8B). Increased MPO expression in skin tissue indicates enhanced neutrophil infiltration and activation in DNCB-induced skin inflammation. Given the established role of MPO in NET formation, these findings can support increased NETosis in DNCB-treated mice.
Myeloperoxidase (MPO) expression and neutrophil infiltration.(A) Representative immunohistochemistry images of skin sections from each experimental group stained for MPO. Scale bar = 100 µm. (B) Quantification of MPO expression in skin sections across groups. Data were analyzed using two-way ANOVA (*** indicates P < 0.0001). (C) Representative flow cytometry plots showing neutrophil populations (CD11b⁺Ly-6G⁺) in skin samples from each group. (D) Quantification of neutrophil infiltration in skin samples across groups. Data were analyzed using two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison (*** P < 0.001 and **** P < 0.0001).*
Notably, treatment with T-MSCs significantly reduced MPO expression in DNCB-treated skin (1.95 ± 0.22), suggesting that T-MSCs modulate neutrophil infiltration and/or activation. Consistent with these findings, flow cytometry analysis demonstrated a significant increase in neutrophil (Ly6G+CD11b+) infiltration in the skin following DNCB treatment (35.13 ± 2.76) compared with the vehicle group (3.60 ± 0.22) (Fig 8C and 8D). This increase was significantly attenuated by T-MSC administration in DNCB-treated mice (27.95 ± 1.48), indicating an anti-inflammatory effect of T-MSCs, at least in part through the regulation of neutrophil recruitment or activation.
Evaluation of T-MSC effects on NET formation in dHL-60 cells
To further assess the impact of T-MSCs on neutrophil function, in vitro experiments were performed using differentiated HL-60 (dHL-60) cells as neutrophil-like cells (Fig 9A). NETosis, defined as the formation of extracellular traps and detected by SYTOX™ Green–stained extracellular DNA structures, was slightly increased in dHL-60 cells treated with DNCB (1.48 ± 0.57 arbitrary units) compared with untreated dHL-60 cells (1.10 ± 0.04), although this difference was not statistically significant. Moreover, co-culture with T-MSCs, either in the absence or presence of DNCB, did not attenuate the SYTOX™ Green–positive area, and these differences also did not reach statistical significance (Fig 9B).
*NET formation analysis.(A) Representative images of differentiated HL-60 (dHL-60) cells used as neutrophil-like cells for in vitro experiments. (B) Quantification of NETosis in dHL-60 cells under various treatment conditions: untreated, DNCB-treated, co-cultured with T-MSCs, or co-cultured with T-MSCs and DNCB. Data were analyzed using two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison (*P < 0.05 and P < 0.01). (C) Immunoblot analysis of citrullinated histone H3 (CitH3) and β-actin in dHL-60 cells treated with PMA, T-MSCs, PAD4 inhibitor (GSK484) and NADPH oxidase inhibitor (DPI) for NETosis quantification. (D) Densitometric quantification of CitH3 expression normalized to β-actin. Bars represent mean ± SD of three independent experiments. Data were analyzed using two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison ( P < 0.05). ns: not significant.
Western blot analysis was performed to further evaluate NETosis by assessing the expression of citrullinated histone H3 (CitH3) in dHL-60 cells (Fig 9C and 9D). As expected, stimulation with phorbol 12-myristate 13-acetate (PMA), a protein kinase C activator, increased CitH3 expression. Co-culture with T-MSCs also resulted in elevated CitH3 levels. GSK484 hydrochloride, a selective and reversible peptidylarginine deiminase 4 (PAD4) inhibitor, unexpectedly increased CitH3 expression compared with untreated dHL-60 cells. Furthermore, the addition of T-MSCs to GSK484-treated dHL-60 cells further enhanced CitH3 expression.
In addition, treatment with diphenyleneiodonium chloride (DPI), a NADPH oxidase (NOX) inhibitor, increased CitH3 expression, and this effect was modestly reduced by co-culture with T-MSCs; however, the difference was not statistically significant. Collectively, these results suggest that T-MSCs do not inhibit NET formation in this in vitro model.
T-MSCs influence T cell migration in the presence of neutrophil-like cells
Jurkat T-cell migration was assessed using Transwell assays with two different pore sizes (5 µm and 8 µm), as illustrated in Fig 10A. Experimental conditions included Jurkat cells cultured alone, Jurkat cells co-cultured with differentiated HL-60 (dHL-60) cells, or Jurkat cells co-cultured with DNCB-treated dHL-60 cells, with or without T-MSCs placed in the lower chamber. Migrated Jurkat cells were quantified by flow cytometry based on CD3 expression (Fig 10B).
Transwell migration assay of Jurkat cells.(A) Experimental design of the Transwell migration assay. Jurkat cells (human T cell leukemia line) were seeded in the upper chamber either alone, co-cultured with dHL-60 cells (neutrophil-like cells), or co-cultured with DNCB-treated dHL-60 cells. T-MSCs were placed in the lower chamber. Two pore sizes (5 µm and 8 µm) were tested. (B) Flow cytometry-based quantification of Jurkat migration under different conditions. Groups included Jurkat alone, Jurkat + dHL-60, Jurkat + DNCB-treated dHL-60, and each of these with or without T-MSCs. Migrated cells were quantified by flow cytometry for both pore sizes. (C) Quantification of Jurkat migration using 5 µm pore Transwells. Bar graphs show the percentage of migrated cells. (D) Quantification of Jurkat migration using 8 µm pore Transwells. Data were analyzed using two-way ANOVA, followed by Tukey’s post hoc test for multiple comparison (a: P < 0.001 for Jurkat vs. Jurkat + T-MSCs, b: P < 0.001 for dHL-60 + Jurkat vs. dHL-60 + Jurkat + T-MSCs, c: P < 0.001 for dHL-60 + DNCB + Jurkat vs. dHL-60 + DNCB + Jurkat + T-MSCs, d: P < 0.001 for Jurkat + T-MSCs vs. dHL-60 + Jurkat + T-MSCs, and e: P < 0.001 for Jurkat + T-MSCs vs. dHL-60 + DNCB + Jurkat + T-MSCs).
Using both 5-µm and 8-µm pore sizes, Jurkat cell migration was influenced by the presence of T-MSCs as well as dHL-60 cells (Fig 10C and 10D). The presence of T-MSCs in the lower chamber significantly increased Jurkat cell migration regardless of whether dHL-60 cells or DNCB-treated dHL-60 cells were present. In contrast, co-culture with dHL-60 cells in the upper chamber reduced Jurkat cell migration compared with Jurkat cells cultured alone.
Collectively, these findings suggest that soluble factors derived from T-MSCs enhance Jurkat T-cell migration, whereas factors associated with dHL-60 cells inhibit T-cell migration toward T-MSCs, indicating that distinct chemotactic cues from these cell types differentially regulate Jurkat T cell migratory behavior.
Discussion
In this study, T-MSCs demonstrated significant therapeutic effects in a DNCB-induced skin inflammation model by modulating both neutrophil and T-cell responses. Administration of T-MSCs substantially reduced skin inflammation, which was associated with decreased neutrophil infiltration and altered T-cell subset distribution. These findings indicate that T-MSCs exert coordinated immunomodulatory effects on both innate and adaptive immune compartments in inflamed skin.
A particularly notable observation following T-MSC administration was the reduction in neutrophil accumulation at sites of inflammation, as evidenced by decreased MPO expression and reduced numbers of Ly6G⁺CD11b⁺ cells (Fig 8), in contrast to the robust neutrophil infiltration typically observed during acute skin inflammation. Gene expression analyses further revealed that T-MSC treatment modulated neutrophil-associated genes and inflammatory signaling pathways, suggesting a role for T-MSCs in regulating neutrophil function during inflammatory resolution. Notably, neutrophil elastase was highly expressed in both the DNCB and DNCB + T-MSC groups, whereas interleukin-1β was specifically downregulated in the DNCB + T-MSC group.
Western blot analyses of NETosis-related markers supported a context-dependent regulatory effect of T-MSCs on NET formation. PMA stimulation increased citrullinated histone H3 (CitH3), consistent with enhanced NETosis, and co-culture with T-MSCs further increased CitH3 levels. Pharmacological modulation of NETosis using the PAD4 inhibitor GSK484 or the NOX inhibitor DPI also enhanced CitH3 expression, effects that were partially modified by T-MSC co-culture. Collectively, these findings indicate that T-MSCs do not inhibit NET formation in this in vitro model.
Although earlier studies have largely emphasized the NETosis-suppressive effects of MSCs [13], emerging evidence indicates that MSC immunomodulatory properties vary depending on tissue origin, activation status, and the surrounding inflammatory milieu [4]. In this context, the ability of T-MSCs to fine-tune NETosis may represent a novel mechanism by which they regulate immune cell dynamics and contribute to the resolution of skin inflammation [18]. Consistent with recent reports, controlled NETosis may aid inflammatory resolution by promoting the removal of activated neutrophils rather than perpetuating tissue injury [15,16].
In parallel with their effects on neutrophils, T-MSCs significantly reduced infiltration and activation of Th1 and Th2 cells in the skin, indicating a strong local immunosuppressive effect [27,28]. In secondary lymphoid organs, including lymph nodes and spleen, T-MSCs altered T-cell subset distribution and activation marker expression, suggesting broader immunoregulatory activity affecting both local and systemic immune responses [5,6,10]. In this study, RNA expression data showed that Cxcl9 was downregulated in both the DNCB and DNCB+T-MSC group. By attenuating T-cell activation and migration into inflamed skin while modulating immune responses in lymphoid tissues, T-MSCs may promote both local resolution and systemic immune homeostasis.
Accumulating evidence indicates that NETs can influence the migration and activation of other immune cells, including T cells, within inflamed tissues [26,29]. In our model, enhanced or altered NETosis following T-MSC treatment may have contributed to reduced T-cell infiltration by modifying the inflammatory microenvironment, including chemokine gradients and adhesion-dependent migration. Supporting this possibility, Transwell migration assays demonstrated that T-MSCs enhanced Jurkat T-cell migration in a chemotactic context, whereas neutrophil-like cells impeded migration, indicating that distinct soluble cues from these cell types differentially regulate T-cell trafficking. The use of two pore sizes (5 µm and 8 µm) enabled refined assessment of T-cell migration under varying physical constraints, further supporting the role of microenvironmental factors in regulating immune cell movement.
Together, these findings underscore the complex interplay among T-MSCs, neutrophils, and T cells within the inflamed skin microenvironment and support the emerging concept that MSCs regulate immune responses through both direct and indirect mechanisms. Several limitations should be addressed in future studies. First, the causal relationship between T-MSC–modulated NETosis and inhibition of T-cell migration requires further mechanistic validation. In this study, Jurkat cell migration was assessed only under co-culture conditions with T-MSCs and dHL-60 cells; therefore, direct comparisons of the migratory effects of T-MSCs and dHL-60 cells alone are warranted. In addition, based on the microarray data, future studies should aim to identify chemotactic factors that regulate immune cell migration in inflammatory skin disease. Second, direct detection and characterization of NETs in vivo are needed. Finally, it will be important to directly compare the therapeutic effects of T-MSCs, conditioned medium, and purified extracellular vesicles in the DNCB-induced dermatitis model.
In conclusion, this study identifies a novel aspect of T-MSC–mediated immunomodulation in skin inflammation, centered on coordinated regulation of neutrophil function and T-cell migration. By selectively modulating critical immune cell interactions, T-MSCs may provide a targeted strategy for restoring immune balance in dermatological and potentially other inflammatory diseases.
Supporting information
S1 FileRaw Data.(ZIP)
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Riedl R, Kühn A, Rietz D, Hebecker B, Glowalla K-G, Peltner LK, et al. Establishment and characterization of mild atopic dermatitis in the DNCB-Induced Mouse Model. Int J Mol Sci. 2023;24(15):12325. doi: 10.3390/ijms 241512325 37569701 PMC 10418750 · doi ↗ · pubmed ↗
- 2Ye S, Zhu L, Ruan T, Jia J, Mo X, Yan F, et al. Comparative study of mouse models of atopic dermatitis. Heliyon. 2025;11(2):e 41989. doi: 10.1016/j.heliyon.2025.e 41989 40013261 PMC 11862492 · doi ↗ · pubmed ↗
- 3Cho K-A, Cha J-E, Kim J, Kim Y-H, Ryu K-H, Woo S-Y. Mesenchymal Stem Cell-Derived Exosomes Attenuate TLR 7-Mediated Mast Cell Activation. Tissue Eng Regen Med. 2022;19(1):117–29. doi: 10.1007/s 13770-021-00395-4 34792754 PMC 8782981 · doi ↗ · pubmed ↗
- 4Wang L-T, Lee W, Liu K-J, Sytwu H-K, Yen M-L, Yen BL. Mesenchymal Stromal/Stem Cells Know Best: The Remarkable Complexities of Its Interactions With Polymorphonuclear Neutrophils. Stem Cells. 2024;42(5):403–15. doi: 10.1093/stmcls/sxae 011 38310524 · doi ↗ · pubmed ↗
- 5Shi Y, Su J, Roberts AI, Shou P, Rabson AB, Ren G. How mesenchymal stem cells interact with tissue immune responses. Trends Immunol. 2012;33(3):136–43. doi: 10.1016/j.it.2011.11.004 22227317 PMC 3412175 · doi ↗ · pubmed ↗
- 6Uccelli A, Moretta L, Pistoia V. Mesenchymal stem cells in health and disease. Nat Rev Immunol. 2008;8(9):726–36. doi: 10.1038/nri 2395 19172693 · doi ↗ · pubmed ↗
- 7Oh S-Y, Choi YM, Kim HY, Park YS, Jung S-C, Park J-W, et al. Application of Tonsil-Derived Mesenchymal Stem Cells in Tissue Regeneration: Concise Review. Stem Cells. 2019;37(10):1252–60. doi: 10.1002/stem.3058 31287931 PMC 6852396 · doi ↗ · pubmed ↗
- 8Ryu K-H, Cho K-A, Park HS, Kim J-Y, Woo S-Y, Jo I, et al. Tonsil-derived mesenchymal stromal cells: evaluation of biologic, immunologic and genetic factors for successful banking. Cytotherapy. 2012;14(10):1193–202. doi: 10.3109/14653249.2012.706708 22900958 · doi ↗ · pubmed ↗