The Standard Reference Material (SRM) 1650b diesel particulate matter was obtained from the National Institute of Standards and Technology (NIST, Gaithersburg, MD, USA). The DEP were suspended in dimethyl sulfoxide as a 10 mg/ml stock solution.

Wharton’s jelly-derived mesenchymal stem cells (WJ-MSCs) were isolated from umbilical cords of healthy full-term babies as previously described (10). The procedures for tissue harvesting and obtaining informed consent were approved by the Asan Medical Center Institutional Review Board (Protocol no. 2015–3030). WJ-MSCs were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Tissue Culture Biologicals, Tulare, CA, USA), 1% Glutamax (Gibco), 1% antibiotic/antimycotic solution (Gibco), 25 ng/ml epidermal growth factor (EGF; Peprotech, Rocky Hill, NJ, USA), and 50 ng/ml basic fibroblast growth factor (bFGF; Peprotech) in a humidified atmosphere containing 5% CO₂ at 37℃. Cells were passaged every 3∼4 days using 0.05% trypsin/EDTA. The cells were pretreated with N-acetyl-l-cysteine (NAC; 5 mM; Sigma-Aldrich, St. Louis, MO, USA), U0126 (10 μM; Cell Signaling Technology, Danvers, MA, USA), or parthenolide (PTL; 5 μM; Sigma-Aldrich) before stimulation with DEP (10 μg/ml) as indicated.

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich) assay was performed to assess the effect of DEP on the viability of WJ-MSCs. WJ-MSCs were seeded into a 24-well plate at 5×10⁴ cells/well with and without various concentrations of DEP for 48 hr. MTT reagent (0.5 mg/ml; Sigma-Aldrich) was added, and the cells were cultured for 3 hr at 37℃. The resulting formazan precipitate was solubilized in 0.04 M HCl, and the absorbance was measured at 570 nm using a Synergy^TM H1 microplate reader (Biotek, Winooski, WT, USA).

For apoptosis assessment, WJ-MSCs were seeded into a 96-well plate at 1×10⁴ cells/well with DEP (0∼40 μg/ml) for 48 hr. The cells were stained with Annexin-V-FITC (BD Biosciences, Franklin Lakes, NJ, USA) and 7-AAD (BD Biosciences) following the manufacturer’s instructions and analyzed by flow cytometry.

WJ-MSCs were seeded into a 24-well plate at 1.5×10⁵ cells/well and cultured overnight to 100% confluency. The cells were incubated with 10 μg/ml mitomycin C (MMC; Sigma-Aldrich) for 1 hr to render them mitotically inactive. The cells were scratched with a 200 ul pipette tip and treated with 0∼10 μg/ml DEP. The cells migrating into the scratched area were photographed at 0, 18, and 36 hr, and migration distances were calculated using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

WJ-MSCs were seeded into a 12-well plate at 5×10⁴ cells/well and incubated with various concentrations of DEP for 48 hr. When the cells reached 70∼80% confluence, StemMACS^TM Adipodiff Media and OsteoDiff Media (Miltenyi Biotec, Bergisch Gladbach, Germany) were added to assess adipogenic and osteogenic differentiation, respectively. The induction media were changed every three days. After three weeks, calcium deposits and lipid droplets were stained with Alizarin Red S (Sigma-Aldrich) and Oil Red O (Abcam, Cambridge, UK), respectively. For quantification, Alizarin-Red-S-stained cells were dissolved in 100 mM cetylpyridinium chloride (Sigma-Aldrich), and the absorbance was measured at 590 nm. Oil-Red-O-stained cells were dissolved in 100% isopropanol, and the absorbance was measured at 490 nm using a spectrophotometer.

RNA sequencing (RNA-seq) was performed by Bioneer Co. (Daejeon, Korea) using Illumina technology. Total RNA was extracted from the WJ-MSCs incubated with or without 5 μg/ml of DEP using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The mRNA sequencing library was generated using the Illumina TruSeq strand mRNA sample preparation kit (Illumina, San Diego, CA, USA) and sequenced using NovaSeq 6000 (2×150 paired end sequencing, Illumina) according to the manufacturer’s protocol. After removing the adapter sequence and filtering the low-quality readings using an in-house script, the filtered readings were mapped to the reference human genome using TopHat (15). For differential expression analysis, the gene expression of each group was quantified using cufflink v2.1.1 (16) by mapping the readings to the human gene annotation database (hg19). Next, the differentially expressed genes (DEGs) in the control and DEP-treated WJ-MSCs were identified based on absolute log2-fold change ≥1 and p<0.05 using Cuffdiff (17, 18). Then, heatmaps were generated using an in-house script, and clustering analysis was performed using a hierarchical clustering method. Volcano plots of DEGs were prepared using GraphPad Prism software (version 8.0.1; GraphPad Software, San Diego, CA, USA).

We identified gene sets based on gene ontology (GO) into three categories: biological processes (BP), cellular components (CC), and molecular function (MF). The significance of the gene sets was calculated using gene set enrichment analysis (GSEA v3.0, https://www.gsea-msigdb.org/gsea/index.jsp). GO-based trend testing was performed using Fisher’s exact test.

Total RNA was extracted using TRIzol reagent (Invitrogen), and 2 μg of total RNA was converted to cDNA using Superscript^TM III reverse transcriptase (Invitrogen). Real-time PCR was performed using the SYBR Green Master Mix (Applied Biosystems, Foster City, CA, USA) and measured using a MX3000P thermal cycler (Agilent Technologies, Santa Clara, CA, USA). β-actin was used as the reference gene for normalization. The primer sequences for qRT-PCR are listed in Supplementary Table S1.

Intracellular reactive oxygen species (ROS) were detec-ted using the peroxide-sensitive fluorophore 2’,7’-dichlorofluorescin diacetate (DCFDA, Sigma-Aldrich). WJ-MSCs were incubated with various concentrations of DEP for 48 hr. After incubation, cells were washed with PBS and incubated with 10 μM DCFDA in serum-free culture medium for 30 min at 37℃. The mean DCFDA fluorescence was analyzed using an Attune NxT flow cytometer (Thermo Fisher Scientific, Waltham, MA, USA).

Umbilical cord blood (UCB) samples were obtained from the Catholic Hematopoietic Stem Cell Bank (CHSCB) under approval of Institutional Review Board (IRB) of the Seoul National University (IRB No. E2011/001-010). Human mononuclear cells (MNCs) were isolated by Ficoll gradient centrifugation as previously described (10), and were cultured in RPMI 1640 medium (Gibco) supplemented with 10% FBS.

WJ-MSCs were pre-cultured with or without NAC for 1 hr, and then various concentrations of DEP were added for 48 hr. Subsequently, the WJ-MSCs were treated with 10 μg/ml MMC for 1 hr to arrest cell proliferation, and the cells were seeded at 1×10⁴ cells per well in a 96-well plate. After 24 hr, MNCs were labeled with 2 μM 5,6-carboxyfluorescein succinimidyl ester (CFSE; Thermo Fisher Scientific), and 1×10⁵ MNCs were added to each well containing WJ-MSCs, in the presence of anti-CD3/CD28 microbeads (Gibco) and recombinant human IL-2 (30 U/ml; PeproTech). After six days of incubation, the cells were stained with fluorescence-labeled human monoclonal antibodies against CD45-APC-H7, CD3-BV510, CD4-APC, and CD8-BV421 (BD Biosciences). Proliferation of total T cells and T cell subpopulations was measured by dilution of CFSE using a FACScanto flow cytometer (BD Biosciences). Viable lymphocytes were gated on 7AAD-negative cells.

To silence the expression of cFos, WJ-MSCs were transiently transfected with specific siRNAs (Bioneer) using the Lipofectamine^TM RNAiMAX reagent (Invitrogen) according to the manufacturer’s protocol. After 4 hr of transfection, 10 μg/ml of DEP was added, and the cells were incubated at 37℃ for 24 hr.

The cells were lysed using RIPA buffer (Thermo Fisher Scientific), and the protein concentrations were determined using a BCA assay kit (Thermo Fisher Scientific). Equal amounts of proteins were obtained after SDS-PAGE separation and analyzed using primary antibodies against cFos, ERK, p-ERK, and p-IκBα (Cell Signaling Technology), and GAPDH (Santa Cruz Biotechnology, Dallas, TX, USA). The bands were visualized using an enhanced chemiluminescence assay kit (Thermo Fisher Scientific) and luminescent image analyzer (LAS-3000 system; Fujifilm, Tokyo, Japan). For quantification, ImageJ software (National Institutes of Health) were used to analyzed the bands intensity.

All animal experiments were carried out in accordance with the approved by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University (protocol no. SNU-201021-2-1). Colitis was induced by administration of 3% dextran sulfate sodium (DSS; MP Biomedicals, Santa Ana, CA, USA) in drinking water for 7 days. 8-week-old C57BL/6 mice were randomly assigned to 5 groups (n=10 per group: (1) Negative control, (2) DSS only, (3) DSS with WJ-MSC, (4) DSS with DEP-treated WJ-MSC (DEP group), (5) DSS with NAC pretreated DEP-treated WJ-MSC (NAC group). Mice were injected intraperitoneally with PBS or cells (2×10⁶ cells in 200 μl) on day 1 after DSS drinking. The severity of colitis was assessed daily using the disease activity index (DAI). All the mice were sacrificed on day 12 and their colon lengths and weights were measured. Myeloperoxidase (MPO) activity assay and histopathological evaluation were performed as previously described (11).

For statistical analysis, all experiments were performed at least in triplicate. Where data were normally distributed, the significance was determined using Student’s t-test or a one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test. The data are presented as the mean±standard deviation. All statistical analyses were performed using GraphPad Prism software (version 8.0.1).

Many studies have shown that DEP activate pro-infla-mmatory responses and affect various cellular functions in diverse cell types (19-22). We therefore investigated the effects of DEP on stemness and immunomodulatory functions of WJ-MSCs. First, we assessed DEP-induced toxicity using the MTT assay; incubation with 10 μg/ml or lower concentrations of DEP did not affect the cell viability, but the viability was significantly reduced at 20 μg/ml and 40 μg/ml DEP (Fig. 1a). Moreover, higher concentrations of DEP significantly induced early and late apoptosis (Fig. 1b). Based on these results, non-apoptotic concentrations of DEP (0∼10 μg/ml) were used in subsequent experiments.

To test the migration and differentiation potential of DEP-treated WJ-MSCs, a scratch wound assay was performed, and in vitro differentiation (adipogenesis and osteogenesis) was assessed. The cell-based wound healing assay showed that migration of WJ-MSCs incubated with 10 μg/ml DEP was significantly reduced compared with that of control WJ-MSCs (Fig. 1c). When cultured under osteogenesis differentiation conditions, calcium deposition was reduced by incubation with DEP in a concentration-de-pendent manner (Fig. 1d), as was the expression of the osteogenic differentiation marker RUNX2 (Fig. 1e). However, there were no significant effects on adipogenic differentiation at any DEP concentration tested (Supplementary Fig. 1a and 1b). These results indicate that DEP inhibits migration and osteogenic differentiation of WJ-MSCs.

RNA sequencing (RNA-seq) was performed to examine changes in the gene expression profile of DEP-treated WJ-MSCs. Global differences in gene expression profiles were visualized by the hierarchical clustering heat map and the volcano plot. Clustering analyses showed 504 differentially expressed genes (DEG) in control MSCs and DEP-treated WJ-MSCs (Fig. 2a). These genes were used in gene ontology (GO)-term analyses to identify overrepresented biological functions. The enriched GO-categories identified in DEP-treated WJ-MSCs included cellular processes, metabolic processes, and biological regulation (Supplementary Fig. 2a and 2b). Among the most upregulated genes related to metabolic processes in DEP-treated WJ-MSCs, we focused on cFos, as it provides a possible link to DEP-induced pathology (Fig. 2b). We confirmed that cFos mRNA expression was significantly increased after incubation of WJ-MSCs with DEP (Fig. 2c).

As DEP-induced pathology is largely due to ROS-mediated cellular toxicity, we assessed induction of ROS in DEP-treated WJ-MSCs using DCFDA staining. At 10 μg/ml DEP, intracellular ROS increased significantly (Fig. 2d), and expression of ROS-related genes SOD2, Nrf2, and NOS2 was increased (Fig. 2e). We next investigated the regulation of inflammatory genes by DEP in WJ-MSCs. Expression of pro-inflammatory genes IFNγ, IL-1β, IL-6, and NLRP3 was significantly upregulated by DEP treatment, and the anti-inflammatory cytokine IL-10 was downregulated (Fig. 2f). These results indicate that DEP upregulates ROS-related genes and pro-inflammatory genes in WJ-MSCs.

To determine whether DEP-induced ROS are implicated in the immunosuppressive properties of WJ-MSCs, we performed a T cell proliferation assay in the presence of DEP and an ROS inhibitor (N-acetyl-l-cysteine, NAC). DEP-treated WJ-MSCs with or without added NAC were co-cultured with CFSE-labeled human umbilical cord blood-derived mononuclear cells that were activated with CD3/CD28 Dynabeads and IL-2. DEP inhibited WJ-MSC-mediated reduction of the proliferation of total lymphocytes and of CD4＋ and CD8＋ T lymphocytes in a concentration-dependent manner. This inhibitory effect was significantly attenuated by the ROS inhibitor (Fig. 3a and 3b, Supplementary Fig. 3a). The percentage of MNCs per cell division shifted to later cycles (cycle 4, 5) depending on the concentration of DEP and returned to the previous cycle (cycle 2, 3) when the WJ-MSCs were incubated with NAC (Fig. 3c, Supplementary Fig. 3b). Taken together, these data indicate that DEP-induced ROS are a major inhibitor of the immunosuppressive effects of WJ-MSCs.

A previous study showed that cFos is a direct downstream target of ROS and is considered an ROS-induced pro-inflammatory gene (23). Therefore, we investigated whether ROS are implicated in DEP-induced cFos in WJ-MSCs. ROS inhibitor treatment significantly inhibited DEP-induced cFos mRNA and protein expression in WJ-MSCs (Fig. 3d and 3e). To investigate involvement of cFos in the immunomodulatory effects of WJ-MSCs, WJ-MSCs were transfected with specific siRNAs against cFos and cultured in the presence of 10 μg/ml of DEP. We confirmed siRNA-mediated knockdown of cFos at both the mRNA and protein levels in DEP-treated WJ-MSCs (Fig. 4a and 4b). Knockdown of cFos reversed the increase in T cell proliferation in DEP-treated WJ-MSCs (Fig. 4c and 4d, Supplementary Fig. 4a). As indicated in Fig. 4e, the percentages of MNCs per cell division in cFos-knockdown WJ-MSCs increased in the 2nd and 3rd division cycles compared with DEP-treated WJ-MSCs (Fig. 4e, Supplementary Fig. 4b). In DEP-treated cFos-depleted WJ-MSCs, the osteogenic potential was markedly greater than in DEP-treated control WJ-MSCs (Fig. 4f). In addition, up-regulated pro-inflammatory cytokines, such as IL-1β, IL-6, and NLRP3, in DEP-treated WJ-MSCs were decreased after cFos depletion whereas the anti-inflammatory cytokine IL-10 was increased (Fig. 4g). These results indicate that DEP inhibits the immunomodulatory function of WJ-MSCs through cFos signaling.

Increased concentrations of proinflammatory cytokines in human airway epithelial cells after DEP exposure have been reported to be associated with activation of NF-κB, cFos, and MAPK (24). Similarly, DEP increased phosphorylation of ERK and IκBα in a concentration-de-pendent manner in WJ-MSCs (Fig. 5a). To examine whether ERK and NF-κB activation is involved in DEP-induced cFos signaling, WJ-MSCs were pre-treated with the ERK kinase inhibitor U0126 and the NF-κB inhibitor parthenolide (PTL). The U0126 and PTL almost completely blocked the activation of ERK and NF-κB, respectively (Fig. 5b), and U0126 but not PTL significantly inhibited DEP-induced cFos expression (Fig. 5c). Taken together, these results indicate that inhibition of the immunomodulatory function of WJ-MSCs by DEP is mediated through ROS/ERK/cFos signaling.

To evaluate whether DEP-induced ROS affected the therapeutic effect of WJ-MSCs in DSS-induced colitis, WJ-MSCs cultured with DEP and NAC were administered to mice 1 day after colitis induction (Fig 6a). As shown in Fig. 6a and 6b, mice in the DEP group displayed continuous body weight loss and remarkably elevated disease activity index (DAI). However, the NAC group was significantly ameliorated body weight loss and DAI as well as increased the survival rate (Fig. 6a∼c). Simultaneously, the colons of the DEP group were obviously shortened when compared to the WJ-MSC group whereas NAC group significantly increased the colon length and decreased the neutrophil infiltration (Fig. 6d and 6e). Histological examination showed that NAC group significantly rescued the submucosal thickening, the destruction of the epithelium, infiltration of inflammatory cells (Fig. 6f). The NAC group also had reduced the histopathological scores compared with the DEP group (Fig. 6g). Expression of pro‐inflammatory cytokines, including IFN‐γ and TNFα was decreased in colon tissue of NAC group compared to DEP group (Fig. 6h). Th1, Th17 and Treg, known to play an important role in the pathogenesis of inflammatory bowel disease, were analyzed in the colon of mice with experimental colitis. The DEP group significantly increased the expression levels of Th1 and Th17 lineage transcription factors compared to the WJ-MSC group, whereas the NAC group significantly decreased the expression levels of Th1 and Th17 transcription factors (Fig. 6i). However, Treg did not show any significant difference among groups. Taken together, these data indicate that DEP-induced ROS is involved in inhibiting the therapeutic effect of WJ-MSCs in DSS-induced colitis.