Low dose IFNγ remodels enhancer landscape to potentiate mesenchymal stromal cell activation Authors: Junxin Lin, Dengfeng Ruan, Qiongying Hu, Xuan Cao, Wenyan Zhou, Siwang Hu Published: 2025 ABSTRACT AbstractBackgroundInterferon gamma (IFNγ) priming is a prominent approach to enhance the immune-regulatory capacity of mesenchymal stromal cells (MSCs). Despite its potential, the concentration of IFNγ vary between studies, and how MSCs response to different concentrations of IFNγ is unclear.MethodsMSCs were treated with various concentrations of IFNγ (0, 1, 5, 10, 50, and 100 ng/mL) for 48 h. Followed by RNA sequencing (RNA-seq) and Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) to profile gene expression and chromatin accessibility, respectively. Integrative analysis was conducted to elucidate the epigenetic regulation of MSCs response to different doses of IFNγ.ResultsWe found that low dose IFNγ (1 ng/mL) treatment was sufficient to induced significant changes in gene expression and chromatin accessibility. RNA-seq revealed that IFNγ activated genes involved in immune response and inactivated genes related to cell-cycle regulation. ATAC-seq showed that IFNγ reduced global enhancer accessibility of MSCs, possibly by downregulation of the SWI/SNF complex. Moreover, IFNγ selectively opened enhancer regions associated with immune response genes while closing regions linked to proliferation-related genes. Motif enrichment analysis identified binding sites for transcription factors such as IRF and AP-1 family members in open and closed regions, respectively, highlighting selective regulatory mechanisms.ConclusionsThis study demonstrates that low dose IFNγ is sufficient to modulate the transcriptional and epigenetic landscape of MSCs by selectively altering enhancer landscape. Our results provide insights into the regulatory mechanisms underlying IFNγ priming and offer a resource for the rational design of strategies to optimize MSC licensing for therapeutic applications.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13287-025-04801-9.Keywords: Mesenchymal stromal cells, Interferon gamma, Gene expression, Chromatin accessibility, Enhancer Background Mesenchymal stromal cells (MSCs) have emerged as a pivotal component in the field of regenerative medicine since their initial discovery in the 1970s [1]. Initially isolated from bone marrow, MSCs have since been identified in various tissues, including adipose tissue, umbilical cord blood, and dental pulp [2]. These cells are characterized by their ability to self-renew and differentiate into multiple cell lineages. Moreover, MSCs possess potent immunomodulatory properties, enabling them to modulate the immune response and promote tissue repair. These unique properties have led to extensive research into their clinical applications, including the treatment of orthopedic injuries, skin disorders, cardiovascular diseases, autoimmune disorders, and neurodegenerative diseases [3, 4]. Extensive research has demonstrated that the immunosuppressive properties of MSCs are not inherently present but are instead induced through inflammatory factors [5]. Thus, preconditioning or priming of MSCs with inflammatory factors has become a common strategy to enhance the immunomodulatory functions of MSCs. Among the factors, interferon-gamma (IFNγ) is one of the the most widely used due to its powerful ability to induce immune response [5, 6]. IFNγ primes MSCs by upregulating the expression of various soluble and membrane-bound molecules, such as indoleamine 2,3-dioxygenase (IDO) and programmed death-ligand 1 (PD-L1), which play crucial roles in suppressing cytotoxic T-cell responses [7–9]. This priming process can improve the therapeutic efficacy of MSCs in various clinical applications, including retinal damage repair and graft vs. host disease [10, 11]. Nevertheless, a challenge in the field is the lack of consensus on the optimal concentration of IFNγ for MSC priming. On the one hand, different studies have employed varying concentrations of IFNγ, ranging from a few to hundreds of nanograms per milliliter, which has led to differential effects of in vitro MSC activation and in vivo treatment [12]. On the other hand, recent study has shown that a high dose of IFNγ can significantly up-regulate the expression of major histocompatibility complex (MHC) class I and II molecules on MSCs, resulting in increased cellular immunogenicity [13]. This variability in IFNγ concentration not only affects the reproducibility of MSC priming effects but also poses challenges in translating these findings into consistent and effective clinical therapies. Thus, there is an urgent need for an understanding of the responses of MSCs to treatment of different concentrations of IFNγ, as well as the underlying mechanisms. To address this issue, this study employs an integrative approach combining ATAC-seq and RNA-seq analyses to elucidate the changes in gene expression profiles and chromatin accessibility of MSCs upon stimulation with different doses of IFNγ (Fig. 1). We confirmed that low dose IFNγ treatment is sufficient to selectively open enhancers to potentiate mesenchymal stromal cells for the activation of immune-response related genes. These results not only enhance our understanding of IFNγ priming but also pave the way for more effective and reproducible MSC-based therapies. Schematic diagram of the experimental design of the study. Mesenchymal stromal cells were treated with 0, 1, 5, 10, 50, 100 ng/mL IFNγ, and then subjected to RNA-seq and ATAC-seq for gene expression and chromatin accessibility profiling. An integrative analysis was performed to identify IFNγ dose-dependent transcriptomic alterations of MSCs and underlying regulatory mechanisms Materials and methods Isolation and culture of mesenchymal stromal cells MSCs were isolated from adipose tissue as previously described [4]. Briefly, the adipose tissues were digested for 4 h at 37 °C using collagenase type I (Gibco, 17100017). The reaction was neutralized with an equal volume of culture medium. Once the adherent cells achieved confluence, they were passaged using 0.25% trypsin (Gibco, 25200056) at a density of 5000 cell/cm2. MSCs were cultured at 37 °C in an atmosphere of 5% CO2 using serum-free culture medium as previously described [14]. All MSCs used in this study were derived from passages 2 to 4. Flow cytometry For flow cytometry analysis, cells were dissociated and suspended in 100 µL of 1% BSA in PBS and incubated with antibody (CD73-PE (BioLegend, 344003), CD90-APC (BioLegend, 328113), CD105-PE (BioLegend, 800503), CD45-PE/C7 (BioLegend, 368531), CD34-AF700 (BioLegend, 343621)) for 30 min on ice in darkness. All staining was analyzed using CytoFLEX S flow cytometer (Beckman Coulter) and data were analyzed using the FlowJo software. Multilineage differentiation For adipogenic differentiation, cells were plated in 24-well-plate (5 × 104 cells per well) and cultured in adipogenic medium (MSC medium supplemented with 0.5 mM hydrocortisone (Macklin, H811182), 100 µM indomethacin (Macklin, I811784), 0.5 mM 3-iso-butyl-1-methylxanthine (Macklin, I811775), and 0.1 µM dexamethasone (Macklin, D829854)). The medium was changed every 2 days for 2 weeks. Positive induction was detected by Oil Red O staining. For osteogenic differentiation, cells were plated in 24-well-plate (1 × 104 cells per well) and cultured in osteogenic medium (MSC medium supplemented with 0.1 µM dexamethasone, 0.2 mM L-ascorbic acid (Macklin, A800295), and 10 mM glycerol 2-phosphate disodium salt hydrate (Macklin, G806967)). The medium was changed every 2 days for 2 weeks. Positive induction was detected by alizarin red staining. For chondrogenic differentiation, 2 × 105 cells were resuspended in 10 µL medium and plated at the center of each well in a 24-well plate. Following 6 h of cell attachment, 1 mL chondrogenic induction medium, which consisted of high-glucose DMEM, 1% ITS (Sigma-Aldrich, I3146), 100 nM dexamethasone, 1% P/S, 10 ng/mL TGF-β3 (MedChemExpress, HY-P7120), 50 µg/mL L-ascorbic acid, and 100 µg/mL sodium pyruvate (Gibco, 11360070), was added. The medium was changed every 2 days for 3 weeks. Positive induction was detected by alcian blue staining. Stimulation of MSCs MSCs were pretreated with or without recombinant human IFNγ (Novoprotein, C014) as follows. When MSCs reached 70% confluence, the culture medium was supplemented with IFNγ to achieve final concentrations of 1, 5, 10, 50, and 100 ng/mL. Based on recent RNA-seq and H3K27ac ChIP-seq data showing robust transcriptional changes and enhancer remodeling of MSCs after 48 h of IFNγ stimulation, we adopted the same incubation time [13]. After 48 h of treatment, cells were harvested for subsequent analyses. For RNA-seq (n = 3 per group) and ATAC-seq (n = 1 per group), cells from the same batch and passage were used to minimize variability introduced by cell culture process. RNA-seq library Preparation and sequencing RNA was extracted from MSCs using a cell RNA isolation kit (Vazyme, RC102). Only high-quality RNA samples (OD260/280 = 1.8 ~ 2.2 and RIN > 9) were used to construct sequencing library. Library preparation and sequencing were performed at Shanghai Majorbio Bio-pharm Biotechnology Co., Ltd., China. The RNA-seq libraries were prepared using Illumina® Stranded mRNA Prep, Ligation (San Diego, CA) using 1 µg of total RNA. Briefly, mRNA was isolated according to polyA selection method by oligo(dT) beads and then fragmented by fragmentation buffer firstly. Secondly double-stranded cDNA was synthesized with random hexamer primers. Then the synthesized cDNA was subjected to end-repair, phosphorylation and adapter addition according to library construction protocol. Libraries were size selected for cDNA target fragments of 300–400 bp use magnetic beads followed by PCR amplification. All libraries were sequenced on Illumina NovaSeq X Plus platform and 150 bp paired-end reads were generated. RNA-seq data analysis All RNA-seq data were quality controlled and trimmed by fastp (v0.23.2) to remove low quality reads or adaptor sequences [15]. The clean reads were mapped to human reference genome hg38 by STAR (v2.7.10a) [16]. The gene expression level was calculated by featureCounts (v2.0.6) [17]. Differentially expressed genes (DEGs) were calculated using the R package DESeq2 [18]. DEGs were defined as log2(fold change) >1 or < -1 and FDR < 0.05. Gene ontology (GO) analysis on DEGs was performed using clusterProfiler [19] with default parameters. ATAC library Preparation and sequencing ATAC-seq libraries were prepared using an open chromatin profile kit (Novoprotein, N248) following the manufacturer’s instruction. Shortly, 50,000 cells were lysed in 45 µL of cold lysis buffer for 5 min. After lysis, the samples were centrifuged at 500 x g for 5 min at 4 °C to collect nuclei and processed for tagmentation in a transposition reaction mix at 37 °C for 30 min. After tagmentation, DNA fragments were purified and amplified by PCR. The amplified products were then purified using DNA clean beads for sequencing. All libraries were sequenced on Illumina NovaSeq 6000 platform and 150 bp paired-end reads were generated (Shanghai Majorbio Bio-Pharm Technology Co., Ltd., China). ATAC-seq data analysis Raw ATAC-seq data generated in this study were quality controlled, trimmed, and aligned to human gemone hg38 using STAR (v2.7.10a) with parameters --outFilterMultimapNmax 1000 --outFilterMismatchNmax 3 --outSAMmultNmax 1 [16]. Duplicated reads were removed using picard (http://broadinstitute.github.io/picard). ATAC-seq peaks were defined using MACS2 (v2.2.9.1) callpeak command with default parameters [20]. Bigwig tracks were generated using deepTools (v3.5.5) bamCoverage command by normalizing read counts to reads per kilobase of bin per million of reads sequenced (RPKM) using bin size of 10 bp [21]. ATAC-seq signal over region of interest were visualized using deepTools (v3.5.5) [21]. Genome and motif annotation over ATAC peak summits was performed using Homer (v4.11) findMotifsGenome.pl command [22]. Differentially accessible regions were identified as described before [13]. To determine the regions that are open or closed by IFNγ, we calculated the read density for each ATAC peak by dividing the number of mapped reads within each peak by the total number of mapped reads and the peak length. Regions were considered differentially accessible if the fold change in read density between IFNγ-treated and naive MSCs was greater than 2 or less than 0.5. RAD analysis To explore the relationship between chromatin accessibility and transcriptional regulation, we performed region associated differentially expressed genes (RAD) analysis using the web application developed by Guo et al. [23]. We input the common up- and down-regulated genes in MSCs induced by IFNγ as the DEGs lists and the common open and closed ATAC peak regions as Genomic Regions of Interest (gROI) file. For submit options, we selected “GRCh38” as the reference genome and entered “1000, 500, 200, 100, 50, 10, 5, and 0 kb” as the customized peak extend distance. The “hypergeometric test” was chosen for statistical test. ChIP-seq data analysis SMARCC1 and SMARCA4 ChIP-seq data analyzed in this study were generated in a previous publication [24]. Raw data was downloaded from Gene Expression Omnibus (GEO) database under accession number GSE215025 (Run ID: SRR21838595 and SRR21838596). Quality control was performed by fastp (v0.23.2). Then reads were aligned by bowtie2 ( v2.5.4) with parameters: --end-to-end --very-sensitive --no-mixed --no-discordant --phred33 -I 10 -X 700 [25]. PCR duplicates were removed using using picard (http://broadinstitute.github.io/picard). Peaks were defined using MACS2 (v2.2.9.1) callpeak command with parameters: -f BAM -g hs --keep-dup all --q 0.01 --call-summits [20]. Bigwig tracks were generated using deepTools (v3.5.5) bamCoverage command by normalizing to RPKM using bin size of 10 bp [21]. Reverse transcription-quantitative PCR (RT-qPCR) RT-qPCR was performed as previously described. In each experiment, RNA was isolated using TRIzol reagent (Invitrogen, 15596018CN). cDNA was synthesized from 2 mg of RNA with NovoScript®Plus All-in-one 1st Strand cDNA Synthesis SuperMix (Novoprotein, E047). Quantitative PCR] was performed using NovoStart® Universal Fast SYBR qPCR SuperMix (Novoprotein, E401) on a Light Cycler apparatus (Roche 480II). Values were calculated with the 2-DD Ct method and normalized to Gapdh expression. Primers used in this study were summarized in Supplementary Table 1. Western blotting Cells were washed with ice-cold PBS and lysed with RIPA lysis buffer (Beyotime, P0013B) supplemented with protease and phosphates inhibitor cocktail (NCM Biotech, P002). Protein quantification was performed using BCA kit (Beyotime, P0010S). Protein lysates were then subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 8% gels and electro-transferred onto a polyvinylidene fluoride (PVDF) membrane. The membranes were blocked with 10% skimmed milk at room temperature for 1 h and incubated with primary antibodies (anti-WARS, Abways, CY8407, 1:2000; anti-COL1A1, Abways, CY5120, 1:2000 anti-GAPDH, Proteintech, 10494-1-AP, 1:20000) overnight at 4 °C. Then, the membranes were washed three times with Tris buffer saline with Tween 20 (TBST) and incubated with the secondary antibodies (HUABIO, HA1001) at room temperature for 2 h. After washing, the membranes were developed by a digital imaging system (ImageQuant 800, Cytiva). T cell suppression assay Human CD3⁺ T cells were purchased from Shanghai Jinyuan Biotechnology Co., Ltd (JY-J089). Upon receipt, cells were labeled with 5 µM CFSE (Ambeed, A1148024) according to the manufacturer’s protocol. CFSE-labeled T cells were stimulated with 5 µg/mL phytohemagglutinin (Ambeed, A1145366) and co-cultured with MSCs stimulated with different concentrations of IFNγ at a ratio of 1:5 (MSCs: T cells) in RPMI 1640 medium supplemented with 10% fetal bovine serum. After 3 days of co-culture, T cell proliferation was analyzed by flow cytometry based on CFSE dilution. Cell proliferation assay Cell proliferation assay were performed by directly counting cell number and using a counting kit-8 (CCK-8, Beyotime, C0042). For direct cell counting, cells were plated at equal density and cultured for 48 h. Then the cells were fixed using 4% paraformaldehyde and stained with DAPI for 15 min. The number of cells were quantified using ImageJ based on the DAPI fluorescence signal [26]. For CCK-8 assay, cells were treated with different doses of IFNγ for 1, 3, 5 days. They were then incubated with 10 µL of CCK-8 solution diluted in 100 µL of culture medium for 2 h. The optical density (OD) values at 450 nm were measured using a microplate reader (Tecan’s Spark, Australia). Statistical analysis Data are represented as means ± SD. Statistical analyses were performed using R programming language. Significance was assayed using the Wilcox rank-sum test, and Studen’s t-test as indicated in the figure legend. Statistical significance was indicated with asterisks: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Results Characterization of MSCs gene expression profile in response to IFNγ To investigate the dose-dependent effects of IFNγ on gene expression of mesenchymal stromal cells (MSCs), we isolated MSCs from adipose tissue and characterized their identity by assessing surface marker expression and their potential for adipogenic, osteogenic and chondrogenic differentiation (Fig. S1). The results showed that the isolated cells meet the criteria for MSC as defined by the International Society for Cell Therapy [27]. Subsequently, we treated MSCs with 0, 1, 5, 10, 50, and 100 ng/mL IFNγ for 48 h and performed RNA-sequencing (RNA-seq). Principal component analysis (PCA) and sample distance analysis results demonstrated a clear separation between naïve and IFNγ-treated MSCs, indicating that gene expression profiles are significantly altered upon IFNγ treatment (Fig. 2A and S2). The first principal component (PC1) explains 93% of the variance, whereas the second principal component (PC2) accounted for only 6% of the variance, suggesting that PC1 accounted for the majority of variance between samples and the transcriptomic differences among MSCs treated with varying IFNγ concentrations were not dominant (Fig. 2A). To validate PCA analysis results, we examined PC1 loadings and found WARS and COL1A1 as the genes with the highest positive and negative contributions, respectively (Fig. S3A). This indicates that WARS was highly expressed in IFNγ-treated MSCs, while COL1A1 was lowly expressed in these cells compared to naïve MSCs. These differential expression patterns were confirmed by RT-qPCR and western blot analysis, supporting the robustness of PCA analysis (Fig. S3B and C). Gene expression profiling of MSCs upon IFNγ treatment. A Principal component analysis (PCA) of RNA-seq data. B Venn diagram describing unique and overlap genes activated (left) and inactivated (right) by different concentration of IFNγ. C Line plot showing the expression of common activated or inactivate genes in each group. D Enriched gene ontology terms of common activated genes. E Enriched gene ontology terms of common inactivated genes To further characterize IFNγ-induced gene expression changes, we identified differentially expressed genes (DEGs) across various concentrations relative to the control group. Herein, we refer genes that are up-regulated following IFNγ treatment as IFNγ-activated genes, and those that are down-regulated as IFNγ-inactivated genes. Comparison of DEGs from each contrast showed that 2282 genes were consistently activated, and 1753 genes were consistently inactivated across all IFNγ concentrations (Fig. 2B and C). Gene functional enrichment analysis demonstrated that common activated genes were involved in pathways related to response to type II interferon, positive regulation of cytokine production, T-cell mediated immunity, regulation of innate immune response, as well as antigen processing and presentation (Fig. 2D). Common inactivated genes were mainly involved in cell cycle regulation, DNA replication and ribosome biogenesis, indicating that cell proliferation was inhibited upon IFNγ treatment (Fig. 2E). Additionally, 78, 29, 223, 59, and 35 genes were uniquely activated, and 83, 61, 128, 66 and 46 genes were uniquely inactivated by different concentrations of IFNγ (Fig. 2B). However, these specifically activated or inactivated genes were not enriched in specific functional terms, indicating these changes may be random. We also noticed that there were more genes specifically activated or inactivated with the treatment of 10 ng/mL of IFNγ, while these genes were not enriched for pathways that may not closely related to immunoregulation (Fig. S4), indicating that expression of these genes might be randomly shifted with the treatment of IFNγ. These results suggest that IFNγ predominantly exerts its effects on gene expression through shared mechanisms rather than dose-specific responses. Chromatin accessibility profiles of MSCs in response to IFNγ To explore the regulatory mechanisms underlying IFNγ-mediated gene expression changes we performed ATAC-seq on MSCs treated with varying IFNγ concentrations. ATAC-seq data quality was validated by expected fragment length distributions, which exhibited both nucleosome-free and single-nucleosome fragments (Fig. S5). The distribution of chromatin accessibility in protein-coding regions also followed expected patterns, with high accessibility near transcription start sites and lower accessibility at gene body and near transcription end sites (Fig. S6). To investigate the effect of IFNγ on the overall chromatin accessibility of MSCs, PCA and Pearson correlation analysis were performed based on the signals of merged peaks from all samples. The PCA plot sorted the principal components according to the amount of data variability and showed that the samples were separated by IFNγ treatment, with PC1 account for 95.8% variance and PC2 account for only 1.5% (Fig. 3A). Meanwhile, pair-wise Spearman correlation analysis further confirmed that chromatin accessibility profiles were highly similar among IFNγ-treated samples but distinct from naive MSCs (Fig. S7). These results suggest that, like transcriptomic changes, IFNγ treatment induces profound chromatin remodeling even at low concentrations. IFNγ treatment reduces global enhancer accessibility through down-regulation of SWI/SNF expression. A Principal component analysis (PCA) of ATAC-seq data. B Metaplots showing ATAC-seq signals over all accessible regions, enhancer, and promoter regions. C Boxplots showing ATAC-seq signals over all accessible regions, enhancer and promoter regions. Two-sided Wilcoxon rank-sum tests between IFNγ treatment group and naive group were performed. *: p < 0.05; n.s.: no significant. D Bar plots showing expression level of SWI/SNF complex genes in naive and IFNγ-treated MSCs. *: p < 0.05, Student’s t-test. E RT-qPCR analysis of SMARCA4 and SMARCC1 expression in MSCs treated with 0, 1, 5, 10, 50, 100 ng/mL of IFNγ. Data are presented as mean ± SD. *p < 0.05, **p < 0.01. F Metaplots showing SMARCA4 and SMARCC1 signals over common open and closed regions upon IFNγ treatment. G Screenshot of SMARCA4, SMARCC1 and ATAC signals on chromosome 17 Next, we compared global ATAC-seq signals between different groups and found that the overall chromatin accessibility was decreased following IFNγ treatment (Fig. 3B and C). While promoter regions showed minimal changes in accessibility across conditions, enhancer regions exhibited a significant reduction in signal intensity in IFNγ-treated MSCs (Fig. 3B and C). Since SWI/SNF is an important complex that regulates chromatin accessibility, we sought to explore whether the SWI/SNF complex is involved in the regulation of IFNγ-induced chromatin accessibility changes. We examined the expression of SWI/SNF complex genes and found that several key components, including SMARCA4 and SMARCC1 were significantly down-regulated (Fig. 3D and E). We next analyzed the ChIP-seq data of SMARCA4 and SMARCC1 and found that these two proteins preferentially bind regions opened after IFNγ stimulation but not the closed regions (Fig. 3F and G). These findings suggest that IFNγ treatment caused chromatin condensation at enhancer regions, and this process may be mediated by downregulation of SWI/SNF complex components. Differentially accessible regions in MSCs following IFNγ stimulation To investigate IFNγ-induced chromatin accessibility changes, we identified the differentially accessible regions (DARs) in MSCs treated with different IFNγ concentrations when compared to naïve MSCs. The total number of open or closed regions remained relatively stable across conditions, but within each concentration, significantly more regions were closed than opened following IFNγ treatment (Fig. S8), in line with previous finding that IFNγ treatment reduced overall chromatin accessibility of MSCs. Venn diagram showed that there were 14,575 commonly opened regions (COR) and 20,390 commonly closed regions (CCR) with the treatment of different concentrations of IFNγ, accounting for a significant proportion of DARs (Fig. 4A). Heatmap plots and metaplots demonstrated that COR exhibited stronger ATAC signals in IFNγ-treated MSCs than in naive MSCs, while CCR exhibited stronger ATAC signals in naive MSCs than in IFNγ-treated MSCs (Fig. 4B and C). Meanwhile, we can also observe that the signal intensity of CCR was stronger than COR, which is consistent with the finding that IFNγ induced overall decrease in chromatin accessibility as mentioned earlier (Fig. 4B and C). Identification and characterization of common open and closed regions upon IFNγ treatment. A Venn diagram describing unique and overlap accessible regions open or closed with the treatment of IFNγ. B Heatmap showing ATAC-seq signals of common open (up) and closed (below) regions. C Metaplots showing ATAC-seq signals over common open (up) and closed (below) regions. D Genomic annotation of all accessible regions in each group, as well as common open and closed regions. 3’UTR: 3′ untranslated regions; 5’UTR: 5′ untranslated regions; TTS: transcription termination site To reveal the distribution characteristics of these differential peaks in the genome, we performed genomic annotation on all peaks and common open or closed regions of different samples. The results showed that compared to the overall peak distribution, COR and CCR were significantly enriched in intergenic and intronic regions rather than promoters (Fig. 4D), which further confirm the notion that IFNγ mainly regulate the enhancer landscape of MSCs. Selective accessible enhancer potentiate activation for immune response genes To investigate the functional consequences of IFNγ-induced chromatin remodeling, we employed region associated DEG (RAD) analysis to assess the relationship between IFNγ-COR and DEGs. With this analysis, we discovered that genes up-regulated upon IFNγ stimulation were significantly enriched within 200 kb next to IFNγ-COR (Fig. 5A). In contrast, no significant association was found between the IFNγ-COR and genes down-regulated by IFNγ (Fig. 5A). More specifically, 75.1% of IFNγ-activated genes were located within 50 kb of IFNγ-COR (Fig. 5B and C). For example, we found that key immune response genes, including STAT1 and IRF1, exhibited both significant transcriptional upregulation and increased chromatin accessibility near their loci upon IFNγ stimulation, reinforcing the role of chromatin remodeling in IFNγ-driven gene activation (Fig. 5D). Given that STAT1 can enhance the T cell inhibition mediated by MSCs [28], we used flow cytometry analysis to examine the inhibitory effect of MSCs treated with different concentrations of IFNγ on T proliferation. The results showed that IFNγ treatment could significantly amplify MSCs’ suppression effect on T cell proliferation even at low doses (Fig. 5E), indicating that changes in chromatin accessibility and gene expression are associated with the functional alterations of MSCs. Selectively accessible enhancers upon IFNγ stimulation are critical for immune response of MSCs. A RAD analysis for the association between common open ATAC regions with common activated and inactivated genes induced by IFNγ. ****p < 0.0001, hypergeometric test. B Pie chart showing the fraction of common activated genes located < 50 Kb or > 50 Kb of common open ATAC regions. C Scatter plot showing the expression of common activated genes located < 50 Kb of common open ATAC regions. D Screenshot of genomic regions around STAT1 and IRF1. E Percentage of proliferative T cells when stimulated alone or stimulated when co-cultured with MSCs treated with different concentrations of IFNγ. *p < 0.05, **p < 0.01, ****p < 0.0001. F Scatter plot showing unbiased motif enrichment analysis of common open ATAC regions To identify the potential regulators of IFNγ-COR, we conducted unbiased motif enrichment analysis to determine the potential transcription factors (TF) that bind to these elements. The analysis revealed preferential enrichment for several transcription factor families, such as basic leucine-zipper (bZIP), interferon regulatory factors (IRF), zinc finger, as well as ETS-IRF complex (Fig. 5F). Among these, well-known IFNγ-regulated genes such as ISRE, IRF1,and IRF2 were identified. Notably, we also found that the IFNγ-COR was enriched for the binding of the PU.1-IRF8 complex. This finding supports the concept that latent enhancers are established through the cooperative binding of IRFs with the lineage-determining factor PU.1 to open chromatin. In conclusion, IFNγ-induced chromatin remodeling significantly promotes the upregulation of key immune response genes near IFNγ-COR, and this process is regulated by cooperative binding of transcription factors such as PU.1 and IRF transcription factor family. IFNγ-induced enhancer closure is associated with proliferation-related gene downregulation We also examined the relationship between IFNγ-CCR and DEGs. As expected, RAD analysis revealed significant enrichment of IFNγ-inactivated genes within 200 kb of IFNγ-CCR, while no significant association observed between IFNγ-up-regulated genes and IFNγ-CCR (Fig. 6A). Notably, 69.8% of IFNγ-inactivated genes were located within 50 kb of IFNγ-closed regions, including cell proliferation regulators, such as FOS and EGR1, linking IFNγ-mediated chromatin condensation to cell cycle inhibition (Fig. 6B-D). Cell number counting and proliferation assay confirmed that IFNγ treatment suppressed the proliferation of MSCs (Fig. 6E-G). Motif analysis further revealed enrichment of bHLH, bZIP, homeobox and Runttranscription factor families, which were proven to be critical for MSCs proliferation (Fig. 6H). These findings suggest that IFNγ-induced enhancer closure is closely associated with the downregulation of proliferation-related genes, demonstrating that IFNγ-mediated chromatin remodeling plays a central role in orchestrating MSC gene expression and cell behavior. IFNγ treatment closes enhancers related with MSCs proliferation. A RAD analysis for the association between common closed ATAC regions with common activated and inactivated genes induced by IFNγ. ****p < 0.0001, hypergeometric test. B Pie chart showing the fraction of common activated genes located < 50 Kb or > 50 Kb of common closed ATAC regions. C Scatter plot showing the expression of common activated genes located < 50 Kb of common closed ATAC regions. D Screenshot of genomic regions around EGR1 and FOS. E DAPI staining of MSCs treated with 0, 1, 5, 10, 50, 100 ng/mL of IFNγ. Scale bar: 100 μm. F Quantification of cell number based on DAPI signal. G Cell proliferation assay using CCK8. Data are presented as mean ± SD. **p < 0.01, ***p < 0.001. H Scatter plot showing unbiased motif enrichment analysis of common closed ATAC regions. Cell viability of MSCs following IFNγ treatment Discussion IFNγ is a critical regulator of mesenchymal stromal cells (MSCs) function, with profound effects on both gene expression and chromatin accessibility. In this study, integrated analysis of transcriptomic and chromatin accessibility data revealed that 1 ng/mL IFNγ priming significantly induces gene expression reprogramming and chromatin remodeling in MSCs, with minimal dose-dependent differences observed across concentration groups. This finding holds critical implications for MSC clinical applications. Previous studies indicate that high-dose IFNγ (e.g., 10–100 ng/mL) may enhance MSC immunosuppressive capacity but concurrently elevate immunogenicity [13]. Our study further corroborates from an epigenetic perspective that low-dose IFNγ sufficiently activates MSC immunomodulatory potential, suggesting 1 ng/mL as a sufficient pretreatment concentration in preclinical studies to balance efficacy, cost, and safety. However, functional validation is required to confirm whether in vitro and in vivo immunosuppressive efficacy at this dose matches high-dose regimens, thereby refining the biological validation of dose-response relationships. This study systematically delineated IFNγ-induced chromatin remodeling and transcriptional changes in MSCs, identifying transcription factor binding motifs within chromatin regions that underwent specific opening or closure post-pretreatment. Results demonstrated that immune response-related PU.1-IRF8 composite binding sites were opened, while motifs for bHLH and Runt families, which related to proliferation and differentiation, were closed following IFNγ exposure. Previous mechanistic studies showed that combining IFNγ with the mTOR inhibitor rapamycin synergistically enhances immunosuppressive gene expression in MSCs [29]. Thus, our findings provide a theoretical foundation for optimizing IFNγ pretreatment efficacy. Co-activation of PU.1-IRF8 during IFNγ pretreatment may represent a strategy to further amplify MSC immunosuppressive potency, while suppression of bHLH and Runt family transcription factors may enhance its beneficial immunomodulatory effects while mitigating potential drawbacks related to proliferation and multi-differentiation. This study reveals that IFNγ-induced chromatin accessibility changes exhibit distinct functional characteristics: open regions are enriched with enhancers of immune response-related genes, while closed regions predominantly harbor enhancers of proliferation and differentiation-associated genes. This regulatory mechanism drives the transition of MSCs from a multipotent state to an immunomodulatory phenotype by restricting transcription factor access (e.g., bHLH, Runt family) to proliferation or differentiation-related enhancers while facilitating the binding of immune-responsive transcription factors (e.g., IRF, ISRE) to enhancers of immunoregulatory genes. Notably, chromatin remodeling coincides with a global reduction in chromatin accessibility, reflecting a bidirectional “global silencing–selective activation” regulatory strategy. This pattern may represent an evolutionary adaptive mechanism in MSCs under inflammatory microenvironments—suppressing non-essential gene expression to reduce metabolic burden while prioritizing immunosuppression functions to maintain tissue homeostasis. We identified significant downregulation of SMARCA4 and SMARCC1, encoding the SWI/SNF chromatin remodeling complex, suggesting SWI/SNF as a potential key mediator of MSC adaptive strategies in inflammatory settings. Additionally, IFNγ pretreatment alters MSC metabolic profiles, which may reciprocally amplify its regulatory effects. However, metabolic responses to IFNγ vary markedly across MSC sources [30, 31]. Future studies integrating transcriptomic, epigenomic, and metabolomic analyses could comprehensively delineate source-specific functional and metabolic adaptations to IFNγ, thereby informing the development of personalized pretreatment protocols tailored to MSC origin. While our data show a general correlation between open chromatin regions and increased gene expression in IFNγ-treated cells, we also observed instances where genes exhibited elevated expression without significant changes in chromatin accessibility near them. This observation raises important questions about the underlying mechanisms driving gene expression changes in the absence of detectable alterations in chromatin accessibility. Indeed, IFNγ activates gene expression through multiple mechanisms. For instance, IFNγ can enhance histone acetylation at the promoter regions of colitogenic mediators to promote the development of the colitogenic monocyte and macrophage lineage [32]. Additionally, IFNγ-induced transcriptional responses require Mediator kinase-dependent RNA polymerase II pause release, highlighting the critical role of RNA polymerase processivity in activating IFNγ-responsive genes [33]. Furthermore, a recent study revealed that WTAP/YTHDF1-mediated m6A modification increases the stability of immunosuppressive molecules, thereby enhancing their mRNA expression and amplifying IFNγ-induced immunosuppressive properties of human MSCs [34]. Given these findings, future investigations should aim to integrate additional epigenomic and epitranscriptomic data, such as histone modification profiles, RNA polymerase II occupancy and RNA modification landscapes, to provide a more comprehensive understanding of the mechanisms driving gene expression changes in response to IFNγ treatment. Another limitation of this study is the single replicate for ATAC-seq analysis, which restricts statistical power and the reliable distinction of biological signal from technical noise. Therefore, the chromatin accessibility findings should be viewed as exploratory and hypothesis-generating. They provide a valuable set of candidate regulatory regions for future validation in studies with robust experimental replicates, which will be essential to definitively establish the epigenetic mechanisms involved in IFNγ regulation of MSC behavior. Conclusion In summary, our study reveals that low-dose of IFNγ profoundly modulates the transcriptional and epigenetic landscape of MSCs. These changes balance immune activation and proliferation control, highlighting the dual role of IFNγ in MSC function. Our findings provide insights into the regulatory mechanisms of IFNγ and suggest new avenues for optimizing MSC-based therapies to enhance their therapeutic potential in regenerative medicine and immunomodulation. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Acknowledgements The authors declare that they have not use AI-generated work in this manuscript. Author contributions Junxin Lin: Conceptualization, Investigation, Formal analysis, Writing - Original Draft, Funding acquisition; Dengfeng Ruan: Resources, Investigation; Qiongying Hu: Resources; Xuan Cao: Resources; Wenyan Zhou: Conceptualization, Investigation, Writing – Review and Editing, Funding acquisition; Siwang Hu: Conceptualization, Project administration, Writing – Review and Editing, Funding acquisition. Funding This work was supported by National Natural Science Foundation of China (82302784), Zhejiang Provincial Natural Science Foundation of China (LQN25C100007), Zhejiang Provincial Program for Medicine and Health (2022KY446 and 2023KY1347), and The Technology Plan Project of Taizhou (24ywa52). Data availability The dataset supporting the conclusions of this article is included within the article and its additional file. Declarations Ethics approval and consent to participate Human adipose tissues were obtained from patients undergoing specific surgical procedure with the approval of the ethics committee of Second Affiliated Hospital, School of Medicine, Zhejiang University, and written informed consent was obtained from the donors in accordance with the Declaration of Helsinki. Title: Heterogeneity analysis of mesenchymal stromal cells (Ref no: IR2021641, dated: 20.12.2021). The original source (Shanghai Jinyuan Biotechnology Co., Ltd) of human CD3⁺ T cells has confirmed that there was initial ethical approval for collection of human cells, and that the donors had signed informed consent. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Wenyan Zhou, Email: zhouwenyan@tzc.edu.cn. Siwang Hu, Email: siwang_h@wmu.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The dataset supporting the conclusions of this article is included within the article and its additional file.