DNMT1 recruits RUNX1 and represses FOXO1 transcription to inhibit anti-inflammatory activity of regulatory T cells and augments sepsis-induced lung injury

Abstract

Sepsis-induced lung injury (ALI) is a critical condition characterized by excessive immune responses and tissue damage. Previous evidence has underscored an upregulation pattern of DNA methyltransferase 1 (DNMT1) in sepsis. This study reveals the key role of DNMT1 in modulating regulatory T cell (Treg) activity in septic ALI. A septic mouse model was generated through cecal ligation and puncture. Treatment with either DNMT1 antagonist Thioguanine (ThG) or AAV-sh-DNMT1 significantly reduced immune cell infiltration, reduced production of pro-inflammatory cytokines, and increasing production of anti-inflammatory cytokines in the bronchoalveolar lavage fluid (BALF) of mice, alongside improved lung pathology and integrity. Furthermore, the DNMT1 inhibition or silencing significantly enhanced population of FOXP3⁺ Tregs in the BALF and lung tissue. Similar trends were observed in mice with specific DNMT1 deletion in CD4⁺ T cells (DNMT1-CD4-ko). Regarding the mechanism, we observed that DNMT1 represses transcription of forkhead box O1 (FOXO1) by recruiting RUNX family transcription factor 1 (RUNX1) to the FOXO1 promoter. FOXO1-specific knockout in CD4⁺ T cells reduced anti-inflammatory activity of Tregs. Additionally, administration of the CD25 antibody exacerbated sepsis-induced ALI in DNMT1-CD4-ko mice. Collectively, these findings illustrate that targeting DNMT1 interacts with RUNX1 to repress transcription of FOXO1, which reduces immunomodulatory activity of Tregs and augments inflammatory cascades in septic lung injury.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10565-025-10069-9.

Introduction

Sepsis is a life-threatening condition marked by dysregulated immune responses to infection, often resulting in organ dysfunction, including severe lung injury (Gotts and Matthay 2016). Acute lung injury (ALI) or acute respiratory distress syndrome (ARDS) is a common complication of sepsis and represents a major cause of morbidity and mortality in affected patients (Zou et al. 2023). The pathogenesis of sepsis-induced ALI involves a complex interplay of immune cells, pro-inflammatory cytokines, and immune suppressive mechanisms, all of which exacerbate tissue damage and compromise respiratory function (Niederman et al. 2021). Given the lack of effective therapies, understanding the molecular regulators of immune responses in sepsis-induced ALI is critical for developing new therapeutic strategies (Wu et al. 2023).

DNA Methyltransferase 1 (DNMT1) is a key epigenetic regulator known to mediate DNA methylation, a process essential for gene silencing and chromatin remodeling (Wong 2020). DNMT1 is upregulated in various immune cells and is significant in immune responses (Morales-Nebreda et al. 2019). Recent studies have revealed that aberrant DNMT1 activity contributes to inflammatory diseases and cancers by silencing tumor suppressor genes and modulating immune cell differentiation (Klein 1993). In the context of sepsis, DNMT1 overexpression has been associated with increased inflammation, tissue damage, and impaired immune regulation (Luo et al. 2019; Mateska et al. 2023). Our previous research has confirmed the involvement of DNMT1 in lipopolysaccharide-induced sepsis in mice (Ding et al. 2022). Additionally, DNMT1’s role in sepsis is believed to involve the suppression of critical anti-inflammatory pathways, particularly those related to regulatory T cells (Tregs), which are vital for controlling excessive immune activation during inflammation (Wang et al. 2024; Wang et al. 2017). Notably, DNMT inhibition through 5-aza-2'-deoxycytidine (Aza) has been found to mitigate inflammation induced pyroptosis and apoptosis during endotoxemia-induced ALI (Samanta et al. 2018). Targeting DNMT1 may also potentially alleviate inflammatory cascades in septic injury.

Tregs are a specialized subset of CD4⁺ T cells that keep immune homeostasis and repress autoimmunity by suppressing excessive immune responses (Sakaguchi et al. 2020). Treg differentiation is primarily triggered by Forkhead box protein 3 (FOXP3), and activated Tregs mainly secret interleukin (IL)−10 and transforming growth factor beta (TGF-β) to maintain self-tolerance and suppress inflammation (Noack and Miossec 2014; Pawlak et al. 2020). Tregs are crucial for tissue repair and protection during acute inflammatory conditions, including sepsis (Allos et al. 2019; Kawai et al. 2018; Liu et al. 2024). Compelling evidence states that the failure to maintain sufficient Treg function can lead to unchecked immune responses, resulting in severe tissue damage (Lin et al. 2018; Wang et al. 2022). Compared to survivors, deceased septic patients have a higher Th17 (pro-inflammatory CD4⁺ cells)/Treg ratio in the circulation (Gupta et al. 2016). Therefore, maintaining or restoring immunomodulatory activity of Tregs may represent a promising strategy to alleviate inflammatory cascades in sepsis and the associated tissue injury.

Forkhead box O1 (FOXO1) is part of the larger FOXO family of proteins, which are involved in controlling diverse processes such as cell proliferation, apoptosis, autophagy, oxidative stress and metabolic dysregulation (Xing et al. 2018). This factor is one of the major downstream of the PI3K signaling pathway and plays a crucial role in regulating the activity and function of various immune cells (Carrette et al. 2009; Li et al. 2024). Recent evidence has highlighted the important role of FOXO1 in modulating the stemness, efficacy, and memory programming in chimeric antigen receptor T cells (Chan et al. 2024), implicating its potential in the management of human malignancies. Specifically, FOXO proteins also exert critical functions in the early differentiation of Treg lineage, and nuclear localization of FOXO1 has been suggested to improve the expression of FOXP3 and enhance Treg stability (Ouyang et al. 2010; Yang et al. 2024). FOXO1 plays an essential role in promoting Treg-mediated immune suppression and tissue protection by activating genes involved in anti-inflammatory pathways (Ren et al. 2023; Ren et al. 2022). Notably, a recent study by Ge and colleagues demonstrated that FOXO1 mitigated inflammatory responses and pathological damages in septic ALI (Ge et al. 2024), suggesting FOXO1 as a promising factor for immune regulation in sepsis management.

RUNX family transcription factor 1 (RUNX1), a transcription factor widely known for its role in hematopoiesis, has also been implicated in immune regulation and inflammation (Illango et al. 2022; Lin et al. 2017). RUNX1 can either activate or repress transcription of target genes depending on its interaction partners (Barutcu et al. 2016; Riddell et al. 2020). In this study, the authors identified a DNMT1/RUNX1/FOXO1 cascade that is potentially linked to Treg regulation and immune responses in the context of sepsis-induced ALI.

Material and methods

Generation of the mouse model of sepsis-induced ALI

All animal experiments followed institutional ethical guidelines and were approved by the Institutional Animal Care and Use Committee of Shanghai Pulmonary Hospital (Approval number: K21-354). Male C57BL/6 mice (6-8 weeks old) were purchased from Charles River Laboratories and housed under specific pathogen-free conditions with a 12-h light-dark cycle and ad libitum access to food and water.

Sepsis was induced via cecal ligation and puncture (CLP). Mice were anesthetized with 3% isoflurane in oxygen, and a midline abdominal incision was made to expose the cecum. The cecum was ligated just below the ileocecal valve and punctured twice using a 21-gauge needle, followed by gentle extrusion of fecal material. The cecum was repositioned, and the peritoneum and skin were sutured. Mice were resuscitated with 1 mL subcutaneous saline. Sham-operated mice underwent the same procedure without cecal ligation or puncture.

Issaeva and colleagues applied Thioguanine (ThG) at 1.5 mg/kg in a mouse model on tumors (Issaeva et al. 2010). Since this study focused on septic lung injury, ThG was administered via intratracheal injection to achieve local delivery to the lungs. After several attempts, we identified ThG (Sigma-Aldrich) administration at 3 mg/kg/day for 5 consecutive days as an optimal regimen to alleviate inflammatory response and improve lung pathology in mice without causing significant advert effects. Therefore, this regimen was used in subsequent experiments.

Unless otherwise specified, ThG (3 mg/kg/day) was dissolved in dimethyl sulphoxide (DMSO) and administered via intratracheal instillation for 5 consecutive days, starting on the day of CLP surgery. Control mice received an equivalent volume of vehicle solution via the same route.

For gene knockdown studies, adeno-associated virus (AAV9)-encapsulated short hairpin RNA (shRNA) targeting DNMT1 was delivered via intratracheal injection (5 × 10¹¹ vg per mouse) 3 d before CLP induction. Mice were monitored every 6 h for signs of morbidity, and humane endpoints were applied in accordance with approved protocols.

At 7 d post-CLP, mice were euthanized using intraperitoneal overdose of nembutal, and lung tissues were collected for histopathological, molecular, and immunological analyses. Bronchoalveolar lavage fluid (BALF) was harvested for assessment of immune cell infiltration and cytokine secretion.

BALF collection and analysis

BALF was obtained by inserting a sterile 22-gauge catheter into the trachea, and 1 mL of sterile PBS was instilled and retrieved gently to avoid cellular damage. This procedure was repeated twice, and the total fluid recovered was centrifuged at 300 g for 10 min at 4°C. Afterward, the supernatant was collected and stored at –80°C for further cytokine analysis, and cell pellets were resuspended in 100 μL PBS for cell counting and differential cell analysis.

Total cells in the BALF were counted by a hemocytometer. For differential cell counting, cytospin was prepared by centrifuging 100 μL of BALF cell suspension at 500 g for 5 min onto microscope slides using a Shandon Cytospin III centrifuge (Thermo Fisher Scientific). Slides were air-dried, fixed in methanol, and stained with Giemsa stain (Sigma-Aldrich) to differentiate macrophages, neutrophils, eosinophils, and lymphocytes. A total of 300 cells per slide were counted by two operators blinded to the grouping details, and the relative percentages of each cell type were calculated.

Cytokine assessment using Enzyme-linked immunosorbent assay

Cytokine concentrations in the BALF supernatant, including tumor necrosis factor-α (TNF-α), IL-1β, monocyte chemotactant protein-1 (MCP-1), IL-10, and TGF-β1 were assessed using the corresponding ELISA kits (Thermo Fisher Scientific). All procedures were conducted strictly adhering to the manufacturer's protocols. Briefly, BALF samples were diluted 1:2 with assay buffer and loaded to 96-well plates precoated with capture antibodies for 2 h. After washing, detection antibodies were applied, followed by incubation with streptavidin-HRP conjugate. Tetramethylbenzidine substrate solution was then added, and the reaction was terminated with sulfuric acid. Optical density at 450 nm was read using a microplate reader (BioTek Instruments). A standard curve for each cytokine was generated using recombinant cytokines provided in the kits, and concentrations were examined by interpolation.

Evans blue dye extravasation assay

To determine pulmonary vascular permeability, mice were intravenously injected with 1% Evans Blue dye (20 mg/kg, Sigma-Aldrich) 1 h before euthanasia. Following euthanasia, lungs were perfused through the right ventricle with 10 mL of PBS to remove intravascular dye. Lungs were excised, weighed, and homogenized in 1 mL of formamide (Sigma-Aldrich). Samples were incubated at 60 °C for 18 h to extract the dye. After centrifugation at 12,000 g for 15 min, OD value was read at 620 nm using a spectrophotometer. Evans Blue concentration was quantified using a standard curve and normalized to lung tissue weight.

Histological analysis

Mouse lung tissues were fixed overnight in 4% paraformaldehyde at 4 °C, dehydrated, and embedded in paraffin. Sections were cut at 5 μm thickness using a rotary microtome (Leica RM2235) and mounted on glass slides. The prepared sections were subjected to hematoxylin and eosin (H&E) staining to evaluate to assess histological changes including alveolar wall thickening, edema, and inflammatory cell infiltration. Additionally, mucus production was evaluated using and periodic acid-Schiff (PAS) staining. The results were evaluated by two pathologists blinded to the groups using a light microscope (Leica DM5000B). A semiquantitative scoring system was used to evaluate lung injury based on the extent of alveolar collapse, infiltration of inflammatory cells, and hemorrhage.

Immunofluorescence staining for FOXP3⁺ Tregs

Lung sections were deparaffinized, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 6.0) at 95 °C for 20 min. Afterward, sections were sealed with 5% bovine serum albumin for 1 h and incubated overnight with FOXP3 antibody (1:100, Abcam) at 4 °C and probed with fluorophore-conjugated secondary antibodies (Thermo Fisher Scientific) for 1 h. Nuclei were counterstained with 4', 6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich), and slides were mounted with Fluoromount-G (SouthernBiotech). Immunofluorescence images were captured using a Zeiss LSM 880 confocal microscope. of the FOXP3⁺ Tregs were quantified by counting positive cells in 10 randomly selected fields (200 x) using ImageJ software.

Generation of DNMT1-CD4-knockout (ko) mice

DNMT1fl/fl mice (Jackson Laboratory) were allowed to mate with CD4-Cre mice to generate DNMT1-CD4-ko offspring. Genotyping was conducted by polymerase chain reaction (PCR) using genomic DNA isolated from tail biopsies. DNMT1fl/fl littermates were utilized as controls. Mice at 6-8 weeks of age were used for sepsis-induced ALI studies as described above.

Treg differentiation from CD4⁺ T cells

CD4⁺ T cells were isolated from the spleens of DNMT1fl/fl or DNMT1-CD4-ko mice using magnetic-activated cell sorting with a CD4⁺ T Cell Isolation Kit (Miltenyi Biotec). The cells were purified using flow cytometry, making sure a purity exceeding 95% prior to subsequent experiments. Naive CD4⁺ T cells were stimulated with plate-bound anti-CD3 (2 μg/mL) and anti-CD28 (2 μg/mL) antibodies in the presence of TGF-β1 (5 ng/mL) and IL-2 (10 ng/mL) for 72 h to induce differentiation into Tregs. Cells were harvested for RNA sequencing (RNA-seq) and protein analysis.

Isolation of Tregs from mice and co-culture experiments

To evaluate the role of DNMT1 in Treg-mediated immunosuppressive activity, CD4⁺CD25⁺ Tregs were isolated from the spleens and lymph nodes of either DNMT1fl/fl or DNMT1-CD4-KO mice using magnetic-activated cell sorting (MACS). Single-cell suspensions were prepared by mechanical dissociation and filtration through a 70 μm cell strainer. CD4⁺CD25⁺ Tregs were enriched using a CD4⁺CD25⁺ Regulatory T Cell Isolation Kit (Miltenyi Biotec, Cat# 130-091-041) following the manufacturer's protocol. Cell purity (>90%) was confirmed by flow cytometry. Responder CD4⁺CD25^- conventional T cells (Tcon) were isolated from wild-type C57BL/6 mice and labeled with carboxyfluorescein diacetate succinimidyl ester (CFSE, Thermo Fisher, Cat# C34554) at a final concentration of 5 μM in PBS at 37 °C for 10 min. The labeling reaction was quenched with RPMI-1640 medium containing 10% FBS and washed twice. For suppression assays, CFSE-labeled Tcon cells (1 × 10⁵ per well) were co-cultured with Tregs from DNMT1fl/fl or DNMT1-CD4-ko mice in round-bottom 96-well plates. Cells were stimulated with anti-CD3 (2 μg/mL, BioLegend, Cat# 100314) and anti-CD28 (1 μg/mL, BioLegend, Cat# 102112) in the presence of irradiated splenocytes as antigen-presenting cells. After 72 h, cells were harvested and analyzed by flow cytometry. Tcon proliferation was quantified based on CFSE dilution.

RNA-seq analysis

Total RNA was extracted from differentiated Tregs using the RNeasy Mini Kit (Qiagen). RNA quality was assessed using an Agilent Bioanalyzer 2100, and samples with RNA integrity numbers (RIN) greater than 7 were used for library preparation. RNA-seq libraries were generated with the NEBNext Ultra RNA Library Prep Kit for Illumina, and sequencing was carried out on an Illumina NovaSeq 6000 platform. Differential expression analysis was performed using DESeq2, with genes showing adjusted p-values < 0.05 considered significantly differentially expressed.

Quantitative PCR (qPCR) analysis

qPCR was conducted to validate RNA-seq results. Total RNA was reverse-transcribed using the PrimeScript RT Reagent Kit (Takara), and qPCR was performed using SYBR Green PCR Master Mix on a CFX96 Real-Time PCR System (Bio-Rad). Primers for FOXO1 and GAPDH (used as an internal control) were designed and synthesized by Sangon Biotech. Relative gene expression levels were calculated using the 2^−ΔΔCt method.

Western blot (WB) analysis

Protein extracts were prepared from Tregs using RIPA buffer. Protein concentration was determined using the BCA Kit (Thermo Fisher Scientific). Equal amounts of protein (30 μg) were separated by SDS-PAGE and transferred onto PVDF membranes (Millipore). Membranes were blocked with 5% non-fat milk in TBST for 1 hour and then probed with primary antibodies against FOXO1 (1:1000, Cell Signaling Technology) and β-actin (1:2000, Abcam) overnight at 4°C. After washing, membranes were incubated with HRP-conjugated secondary antibodies, and signals were visualized using ECL detection reagents (Thermo Fisher Scientific)

Chromatin immunoprecipitation (ChIP) assays

ChIP assays were performed using the SimpleChIP Kit (Cell Signaling Technology). Tregs, differentiated from CD4⁺ T cells, were cross-linked with 1% formaldehyde for 10 min to preserve protein-DNA interactions. Cross-linking was quenched by adding glycine to a final concentration of 0.125 M for 5 min. Cells were lysed in the provided lysis buffer, and chromatin was sheared into 200-600 bp fragments using micrococcal nuclease digestion and sonication (Bioruptor Plus, Diagenode). For immunoprecipitation, 5 μg of sheared chromatin was incubated overnight at 4 °C with 5 μg of either anti-DNMT1 (Cell Signaling Technology) or anti-RUNX1 (Abcam) antibodies, or normal rabbit IgG (negative control). Protein A/G magnetic beads (Cell Signaling Technology) were added, and the complexes were washed with low- and high-salt wash buffers. Cross-links were reversed by incubating with ChIP elution buffer and proteinase K at 65 °C for 4 h. The immunoprecipitated DNA was purified using the kit’s spin columns. ChIP-qPCR was performed using SYBR Green Master Mix on a CFX96 Real-Time PCR Detection System (Bio-Rad). Primers targeting the FOXO1 promoter region, including the CpG island, were designed using Primer3 software. Fold enrichment was calculated relative to input DNA and normalized to the IgG control.

Luciferase reporter assays

The FOXO1 promoter region (including the CpG island) was amplified from mouse genomic DNA and cloned into the pGL3-Basic vector (Promega). The promoter-reporter construct was verified by Sanger sequencing. For reporter assays, 293 T cells were seeded in 24-well plates at 2 × 10⁵ cells per well and transfected using Lipofectamine 3000 with 500 ng of the FOXO1 promoter-pGL3 construct and 50 ng of the Renilla luciferase plasmid (pRL-TK) as a control. Co-transfection experiments were performed with DNMT1 shRNA or RUNX1 overexpression plasmids, along with the FOXO1 promoter reporter construct. DNMT1 shRNA and RUNX1 overexpression plasmids were obtained from Addgene. After 48 h, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega). Firefly luciferase activity was normalized to Renilla luciferase activity, and relative luciferase units were calculated as a percentage of the control group.

Bisulfite sequencing PCR (BSP-qPCR)

Genomic DNA was extracted from Tregs using the DNeasy Blood & Tissue Kit (Qiagen). Bisulfite conversion of unmethylated cytosines to uracil was performed using the EZ DNA Methylation-Gold Kit (Zymo Research). Converted DNA was amplified by PCR with primers targeting the CpG-rich region in the FOXO1 promoter. Primers were designed using MethPrimer software to ensure optimal specificity for bisulfite-treated DNA. The amplified PCR products were purified using the QIAquick PCR Purification Kit (Qiagen) and cloned into the pCR4-TOPO vector (Thermo Fisher Scientific) for sequencing. A minimum of 10 individual clones from each sample were sequenced to assess the methylation status of each CpG site. Methylation levels were determined by comparing cytosine to thymine conversions at CpG sites using ClustalW alignment. For quantitative methylation analysis, BSP-qPCR was performed on bisulfite-treated DNA. Primers specific to the FOXO1 CpG island were used in qPCR reactions with SYBR Green PCR Master Mix. Methylation levels were calculated using the 2^−ΔΔCt method and normalized to input DNA.

Biological replicates and statistical analysis

Data are presented as the mean ± standard error of the mean (SEM). Unless otherwise specified, for cellular experiments, each experiment was performed using cells isolated from six independent mice, with each measurement conducted in technical triplicates. For animal experiments, each group consisted of six mice, and three independent replications were conducted for each sample. The data from each independent experiment are represented as individual dots in the graphs. Comparisons between two groups were analyzed using unpaired two-tailed Student's t-tests. For multiple group comparisons, one-way or two-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was applied. Normality of data distribution was assessed using the Shapiro-Wilk test. For non-normally distributed data, non-parametric tests (e.g., Kruskal-Wallis with Dunn’s post hoc test) were used. Statistical analyses were performed using GraphPad Prism 9.0 and R version 4.3.1. A p-value of < 0.05 was considered statistically significant. No data points were excluded, unless due to technical errors.

Results

DNMT1 inhibition attenuates sepsis-induced ALI in mice

Our previous research has confirmed the involvement of DNMT1 in lipopolysaccharide-induced sepsis in mice (Ding et al. 2022). To further investigate the specific role of DNMT1 in sepsis-induced ALI, a mouse model of sepsis was generated through CLP challenge, followed by administration of either the DNMT1 antagonists ThG or AAV9-mediated DNMT1 knockdown (5×10¹¹vg per mouse) via intratracheal injection (Fig. 1A, Fig. S1A-B). Notably, it was observed that the populations of macrophages, neutrophils, eosinophils, and lymphocytes in the BALF were substantially increased in septic mice, which were then significantly by either ThG treatment or DNMT1 knockdown (Fig. 1B). Additionally, the concentrations of pro-inflammatory cytokines, including TNF-α, IL-1β, and MCP-1, as well as the levels of anti-inflammatory cytokines, including IL-10 and TGF-β1, were upregulated in the BALF of septic mice compared to sham-operated ones. Notably, the ThG treatment or DNMT1 knockdown substantially reduced the levels of pro-inflammatory cytokines while further elevating the levels of anti-inflammatory cytokines Fig. 1C). Furthermore, Evans blue staining showed a significant increase in vascular permeability in lung tissues of septic mice, which was substantially alleviated after pharmacological inhibition or knockdown of DNMT1 (Fig. 1D). Pathological changes in lung tissues developed in septic mice were also alleviated by ThG treatment or DNMT1 knockdown, including the reduction of inflammatory cell infiltration, partial recovery of alveolar structure, decreased alveolar wall thickening, and reduced interstitial edema and hemorrhage. Additionally, the mucus secretion in lung tissues was decreased, and the amount of glycoprotein deposition in the airways and alveoli was noticeably reduced following DNMT1 inhibition or knockdown (Fig. 1E-F). Furthermore, flow cytometry and immunofluorescence assays confirmed that the population of FOXP3⁺ Tregs in BALF and lung tissue was increased significantly following ThG treatment or DNMT1 knockdown, indicating a local immunoregulatory effect within the pulmonary microenvironment (Fig. 1G-H). Additionally, it was observed that the number of FOXP3⁺ in the peripheral blood of mice was not significantly altered following either ThG or AAV9-shRNA treatment (Fig. 1I), suggesting that the intratracheal administration of ThG or AAV9-shRNA primarily promotes local expansion or recruitment of Tregs in the pulmonary microenvironment.

Fig. 1.

DNMT1 inhibition attenuates sepsis-induced ALI in mice. A Schematic presentation of the septic mouse model induced via CLP challenge, followed by intratracheal injection of either DNMT1 antagonist ThG (3 mg/kg/day) or AAV9-encapsulated sh-DNMT1. B Number of macrophages, lymphocytes, neutrophils, and eosinophils in mouse BALF. C ELISA analysis of TNF-α, IL-6, IL-1β, MCP-1, IL-10 and TGF-β1 concentrations in mouse BALF. D Evans Blue staining of vascular permeability in mouse lung tissue. E HE staining to analyze pathological injury in mouse lung tissues. F PAS staining to assess glycoprotein deposition in mouse lung tissues. G Flow cytometric analysis of CD4⁺CD25⁺FOXP3⁺ T cells in mouse BALF. H Immunofluorescence detection of FOXP3⁺ cells in mouse lung tissue (Red: FOXP3⁺; Blue: DAPI). Each group contains six mice, and three independent replications were conducted for each sample Differences between groups were compared by one-way ANOVA followed by Tukey’s post-hoc tests. p < 0.05 was considered statistically significant

DNMT1 deficiency in CD4⁺ T cells alleviates sepsis-induced AKI in mice by enhancing the immunosuppressive function of Tregs

Tregs are known to provide tissue protection and coordinate lung tissue repair, promoting the return of homeostatic pulmonary function (Jovisic et al. 2023). This led us to hypothesize a correlation between DNMT1 and Treg activation. To explore this, we generated DNMT1-CD4-Cre mice with specific DNMT1 deletion in CD4⁺ T cells (DNMT1-CD4-ko), followed by CLP challenge to induce sepsis. Compared to DNMT1fl/fl mice, the DNMT1-CD4-ko mice showed significantly alleviated septic injury after CLP, as manifested by reduced inflammatory cells and cytokines in BALF whereas increased concentrations of anti-inflammatory cytokines (IL-10 and TGF-β1) (Fig. 2A-B), accompanied by alleviated vascular permeability in lung tissues (Fig. 2C). Furthermore, the inflammatory cell infiltration, alveolar structure, and decreased edema and hemorrhage in the lung interstitium were alleviated in the DNMT1-CD4-ko mice as well (Fig. 2D-E). Additionally, Masson's trichrome staining showed that the fibrosis in mouse lung tissues was reduced upon DNMT1 knockout in CD4⁺ T cells (Fig. 2F). Consistent trends were identified in terms of the expression of fibrosis markers FN1, SPP1, and COL1A1 (Fig. 2G). As anticipated, the population of Tregs was notably increased in lung tissues after CD4⁺ T cell-specific DNMT1 knockout (Fig. 2H-I). Consistently, the number of CD4⁺CTLA4⁺ T cells was significantly increased in BALF from DNMT1-CD4-ko mice (Fig. 2J).

Fig. 2.

DNMT1 deficiency in CD4⁺ T cells alleviates sepsis-induced AKI in mice by enhancing the immunosuppressive function of Tregs. DNMT1-CD4-Cre mice with specific DNMT1 deletion in CD4+ T cells (DNMT1-CD4-ko) were generated, followed by CLP challenge to induce sepsis. CLP-challenged DNMT1fl/fl mice were set to controls. A Number of macrophages, neutrophils, eosinophils, and lymphocytes in mouse BALF. B ELISA analysis of TNF-α, IL-6, IL-1β, MCP-1, IL-10 and TGF-β1 concentrations in mouse BALF. C Evans Blue staining of vascular permeability in mouse lung tissue. D HE staining to analyze pathological injury in mouse lung tissues. E PAS staining to assess glycoprotein deposition in mouse lung tissues. F Masson's trichrome staining to detect collagen deposition in lung tissues. G qPCR analysis of mRNA levels of fibrosis-related genes (FN1, SPP1, and COL1A1) in mouse lung tissues. H Flow cytometry analysis of CD4⁺CD25⁺FOXP3⁺ T cells in mouse BALF. I Immunofluorescence detection of FOXP3⁺ cells in mouse lung tissue (Red: FOXP3⁺; Blue: DAPI). J Flow cytometric analysis of CD4⁺CD25⁺CTLA4⁺ T cells in mouse BALF. Each group contains six mice, and three independent replications were conducted for each sample. Differences between groups were compared by unpaired t tests, or by the two-way ANOVA followed by Tukey’s post-hoc tests. p < 0.05 was considered statistically significant

To further confirm the role of DNMT1 in Treg activity, Tregs were isolated from either DNMT1fl/fl or DNMT1-CD4-ko mice and purified. These cells were then co-cultured with CFSE-labeled CD4⁺CD25^- Tcon cells. Notably, upon co-culturing with Tregs extracted from DNMT1-CD4-ko mice, the number of CFSE⁺CD4⁺CD25⁺ T cells was significantly increased (Fig. S2A). Additionally, the concentrations of IL-10 of TGF-β1 in the co-culture supernatant were increased in this setting (Fig. S2B). This evidence suggests that DNMT1 deficiency in CD4⁺ T cells increases Treg differentiation and enhances their anti-inflammatory activity.

DNMT1 represses FOXO1 expression

To further investigate genes regulated by DNMT1, we isolated naïve CD4⁺ T cells from both DNMT1fl/fl and DNMT1-CD4-ko mice. These cells were stimulated with anti-CD3 and anti-CD28 antibodies, followed by induction with TGF-β1 and IL-2 to promote their differentiation into Tregs (Fig. 3A). RNA seq analysis was then performed to identify differentially expressed genes (DEGs) in Tregs derived from the two groups of mice. A total of 224 DEGs were identified, with FOXO1 showing significant upregulation in DNMT1-deficient Tregs (Fig. 3B). This increase in FOXO1 expression was further confirmed by qPCR and WB analyses (Fig. 3C-D).

Fig. 3.

DNMT1 represses FOXO1 expression. A Naive CD4⁺ T cells were isolated from DNMT1fl/fl or DNMT1-CD4-ko mice. These cells were the stimulated with These cells were stimulated with anti-CD3 and anti-CD28 antibodies, followed by induction with TGF-β1 and IL-2 to promote their differentiation into Tregs. B Volcano plots for DEGs in Tregs from DNMT1fl/fl and DNMT1-CD4-ko mice identified through RNA-seq analysis. C~D qPCR and WB analysis of FOXO1 levels in Tregs. E CpG island upstream of the FOXO1 promoter predicted using the TUBIC software. F BSP-qPCR detection of the CpG island in FOXO1 promoter in Tregs. G Binding of DNMT1 to the FOXO1 promoter determined using ChIP-qPCR assays. H Regulation of DNMT1 on the transcription activity of FOXO1 determined using luciferase reporter gene assays. Six independent experiments were performed. Differences between groups were compared by unpaired t tests, or by the two-way ANOVA followed by Tukey’s post-hoc tests. p < 0.05 was considered statistically significant

As an epigenetic regulator, DNMT1 represses gene expression through DNA methylation of promoter regions. Using the UCSC Genome Browser, we identified a GC-rich region (~237 bp) upstream of the FOXO1 promoter in the mouse mm39 genome (chr3:52175032–52177481). Further analysis with TUBIC software predicted the presence of a CpG island in this region (Fig. 3E). To validate this, BSP-qPCR was further performed, identifying that DNMT1 deletion significantly reduced CpG island methylation in this region of Tregs (Fig. 3F).

To determine whether DNMT1 directly regulates FOXO1 expression via promoter binding, we performed ChIP-qPCR assays, which showed that DNMT1 knockdown led to reduced enrichment of FOXO1 promoter fragments in precipitates reacted with anti-DNMT1 (Fig. 3G). Additionally, we constructed a FOXO1 promoter-luciferase reporter (pGL3 vector) and co-transfected it with DNMT1 shRNA into 293 T cells. Notably, the DNMT1 silencing led to a significant increase in luciferase activity (Fig. 3G), suggesting that DNMT1 negatively regulates FOXO1 transcription by binding to its promoter.

DNMT1 recruits RUNX1 to repress FOXO1 transcription

The regulation of gene transcription typically involves transcription factors. To identify transcription factors forming a repressive complex with DNMT1, we performed immunoprecipitation-mass spectrometry (IP-MS). RUNX1 as the transcription factor most strongly associated with DNMT1 (Fig. 4A). The double label fluorescence assays confirmed the co-localization and interaction between DNMT1 and RUNX1 in both extracted Tregs and lung tissue sections (Fig. 4B). Immunofluorescence staining also suggested that the nuclear localization of RUNX1 was reduced in Tregs extracted from DNMT1-CD4-ko mice (Fig. 4C). Furthermore, Co-IP assays confirmed the direct binding between DNMT1 and RUNX1 (Fig. 4D). Additionally, ChIP assays identified that RUNX1 overexpression increased the enrichment of FOXO1 promoter fragments in the immunoprecipitated complexes (Fig. 4E), and the luciferase assays suggested that the RUNX1 overexpression reduced luciferase activity of the reporter containing FOXO1 promoter (Fig. 4F), which consequently led to reduced mRNA and protein levels of FOXO1 (Fig. 4G-H). These results indicate that DNMT1 interacts with RUNX1 to repress FOXO1 transcription.

Fig. 4.

DNMT1 recruits RUNX1 to repress FOXO1 transcription. A Volcano plots of DNMT1-interacting proteins identified by IP-MS. B Co-localization of DNMT1 and RUNX1 in extracted Tregs or mouse lung tissue sections determined using double-label fluorescence staining assays (Red: RUNX1⁺; Green: DNMT1⁺; Blue: DAPI). C Immunofluorescence detection of RUNX1 nuclear localization in Tregs (Red: RUNX1⁺; Green: DNMT1⁺; Blue: DAPI). D The interaction between RUNX1 and DNMT1 determined using Co-IP and WB assays. E Binding of RUNX1 to the FOXO1 promoter determined using ChIP-qPCR assays. F Regulation of RUNX1 on the transcription activity of FOXO1 determined using luciferase reporter gene assays. G-H qPCR and WB analysis of RUNX1 and FOXO1 levels in 293 T cells after RUNX1 overexpression. Six independent experiments were performed. Differences between groups were compared by unpaired t tests, or by the two-way ANOVA followed by Tukey’s post-hoc tests. p < 0.05 was considered statistically significant

FOXO1 deficiency in CD4⁺ T cells reduces anti-inflammatory activity of Tregs

To further analyze the role of FOXO1 in Treg activity and the progression of septic lung injury, FOXO1-CD4-ko mice were generated. Following CLP challenge, FOXO1-CD4-ko mice exhibited a notable increase in inflammatory cells and cytokines in BALF, accompanied by decreased concentrations of IL-10 and TGF-β1 (Fig. 5A-B). The vascular permeability (Fig. 5C). inflammatory cell infiltration and tissue damage (Fig. 5D), mucus production (Fig. 5E), and fibrosis (Fig. 5F) in the lung tissues were also augmented upon FOXO1 deletion in CD4⁺ T cells. Flow cytometry and immunofluorescence confirmed that the number of Tregs was markedly reduced in FOXO1-CD4-ko mice (Fig. 5G-H). Furthermore, the number of CD4⁺CTLA4⁺ T cells in the mouse BALF was substantially decreased upon FOXO1 knockout (Fig. 5I).

Fig. 5.

FOXO1 deficiency in CD4⁺ T cells reduces anti-inflammatory activity of Tregs. FOXO1-CD4-ko mice were generated, followed by CLP challenge to induce sepsis. CLP-challenged DNMT1fl/fl mice were set to controls. A Number of macrophages, neutrophils, eosinophils, and lymphocytes in BALF. B ELISA analysis of inflammatory cytokines NF-α, IL-6, IL-1β, MCP-1, IL-10 and TGF-β1 concentrations in mouse BALF. C Evans Blue staining of vascular permeability in mouse lung tissue. D HE staining to analyze pathological injury in mouse lung tissues. E PAS staining to assess glycoprotein deposition in mouse lung tissues. F Masson's trichrome staining to detect collagen deposition in lung tissues. G Flow cytometry analysis of CD4⁺CD25⁺FOXP3⁺ T cells in mouse BALF. H Immunofluorescence detection of FOXP3⁺ cells in mouse lung tissues (Red: FOXP3⁺; Blue: DAPI). J Flow cytometric analysis of CD4⁺CD25⁺CTLA4⁺ T cells in mouse BALF. Each group contains six mice, and three independent replications were conducted for each sample. Differences between groups were compared by unpaired t tests. p < 0.05 was considered statistically significant

ThG also alleviates septic lung injury in FOXO1-CD4-ko mice

Interestingly, in additional experiments, we observed that treatment with ThG also reduced the number of inflammatory cells and decreased the concentrations of inflammatory cytokines in the BALF of FOXO1-CD4-ko mice with CLP-induced sepsis (Fig. 6A-B), along with decreased vascular permeability in lung tissues (Fig. 6C). Histopathologic analyses identified reduced inflammatory infiltration, reduced alveolar wall thickening, and decreased mucus and glycoprotein deposition in treated mice (Fig. 6D-E). Additionally, the number of FOXP3⁺ Tregs in mouse lung tissues was also increased upon ThG treatment (Fig. 6F-G). These observations indicate that DNMT inhibition also increases Treg population and activity through FOXO3-independent mechanisms.

Fig. 6.

ThG also alleviates septic lung injury in FOXO1-CD4-ko mice. A Number of macrophages, neutrophils, eosinophils, and lymphocytes in BALF. B ELISA analysis of inflammatory cytokines in BALF. C Evans Blue staining of vascular permeability in mouse lung tissue. D HE staining to analyze pathological injury in mouse lung tissues. E PAS staining to assess glycoprotein deposition in mouse lung tissues. F Flow cytometry analysis of CD4⁺CD25⁺FOXP3⁺ T cells in mouse BALF. G Immunofluorescence detection of FOXP3⁺ cells in mouse lung tissues (Red: FOXP3⁺; Blue: DAPI). Each group contains six mice, and three independent replications were conducted for each sample. Differences between groups were compared by unpaired t tests. p < 0.05 was considered statistically significant

To substantiate the DNMT1-FOXO1-Treg cascade, CD4⁺ T cells extracted from DNMT1fl/fl and DNMT1-CD4-ko mice were additionally treated with the FOXO1-specific antagonist AS1842856 (30 nM). Notably, the AS1842856 treatment substantially reduced the mRNA and protein levels of FOXP3 (Fig. S3A-B) and, consequently, reduced the populations of CD4⁺CD25⁺FOXP3⁺ cells (Fig. S3C). Accordingly, the concentrations of IL-10 and TGF-β1 in the cell culture system were reduced by the AS1842856 treatment as well (Fig. S3D).

CD25 antibody exacerbates sepsis-induced ALI in DNMT1-CD4-ko mice

To further investigate whether the DNMT1–FOXO1 axis mediates the progression of sepsis-induced AKI through modulation of Tregs, we administered anti-CD25 to DNMT1-CD4-ko mice to deplete Tregs. Flow cytometry and immunofluorescence staining confirmed the loss of Tregs within mouse lung tissue following anti-CD25 treatment (Fig. 7A-B). Consequently, the inflammatory cell infiltration and production of proinflammatory cytokines in the BALF were increased (Fig. 7C-D). These were accompanied by increased vascular permeability (Fig. 7E), exacerbated lung injury (Fig. 7F), increased mucus production (Fig. 7G), and fibrosis (Fig. 7H). Collectively, these findings suggest that DNMT1, by recruiting the transcription factor RUNX1, suppresses FOXO1 expression and thereby attenuates Treg polarization, playing a critical role in the pathogenesis of sepsis-induced lung injury.

Fig. 7.

CD25 antibody exacerbates sepsis-induced ALI in DNMT1-CD4-ko mice. DNMT1-CD4-ko mice were treated with anti-CD25 to deplete Tregs, followed by sepsis modeling. A Flow cytometry analysis of CD4⁺CD25⁺FOXP3⁺ T cells in mouse BALF. B Immunofluorescence detection of FOXP3⁺ cells in mouse lung tissues (Red: FOXP3⁺; Blue: DAPI). C Number of macrophages, neutrophils, eosinophils, and lymphocytes in BALF. D ELISA analysis of inflammatory cytokines in BALF. E Evans Blue staining of vascular permeability in mouse lung tissue. F HE staining to analyze pathological injury in mouse lung tissues. G PAS staining to assess glycoprotein deposition in mouse lung tissues. H Masson's trichrome staining to detect collagen deposition in lung tissues. Each group contains six mice, and three independent replications were conducted for each sample. Differences between groups were compared by unpaired t tests. p < 0.05 was considered statistically significant

Discussion

Our study has revealed that DNMT1 is significantly overexpressed in sepsis-induced ALI, and its inhibition by ThG led to repressed lung inflammation, decreased cytokine production, and improved vascular permeability. Mechanistically, DNMT1 suppresses FOXO1 expression by recruiting RUNX1, a known transcriptional factor, to the FOXO1 promoter. This repression of FOXO1 impairs Treg function, thereby exacerbating inflammation. In DNMT1-CD4 ko mice, FOXO1 expression was restored, Tregs increased, and lung injury was significantly attenuated, underscoring the critical role of the DNMT1-FOXO1 axis in regulating Treg-mediated immune responses during sepsis.

Our previous research has identified the upregulation pattern of DNMT1 in septic mice (Ding et al. 2022). Additionally, Zhang and colleagues demonstrated that DNMT1 contributed to cell apoptosis and inflammation in lipopolysaccharide-induced septic conditions in animal and cell models (Zhang et al. 2022). Samamta and teammates reported that the application of DNMT inhibitor Aza reduced inflammation and cell death in endotoxemia-induced ALI (Samanta et al. 2018). Importantly, in this study, we further identified that either ThG-mediated pharmacological inhibition or AAV-sh-RNA-mediated gene silencing of DNMT1 significantly reduced immune cell infiltration, reduced production of pro-inflammatory cytokines, and increasing production of anti-inflammatory cytokines in the BALF of mice, alongside improved lung pathology and integrity. This evidence supports the potential pathogenic role of DNMT1 in sepsis and the associated tissue damage. Hu Y and Jiang T demonstrated that DNMT1 promotes hypermethylation of Treg-associated genes in inflammatory diseases, such as multiple sclerosis, impairing Treg function and exacerbating inflammation (Hu et al. 2023; Jiang et al. 2021). In patients with abdominal arterial aneurysm, the extracted Tregs showed higher DNA methylation rate, with higher expression of DNMT1 and DNMT3A detected (Xia et al. 2019). Furthermore, Fang and colleagues demonstrated that the Aza-mediated DNMT inhibition enhance the differentiation of Tregs from naïve CD4⁺ T cells extracted from the peripheral blood of patients with chronic HBV infection (Fang et al. 2021). Earlier studies have also demonstrated that DNMT1 inhibition could lead to Treg by reducing DNA methylation of FOXP3 (Lal et al. 2009; Wang et al. 2013). DNMT1’s regulation of Treg function has been identified as a common mechanism that spans both acute and chronic inflammatory conditions (Gilbert et al. 2012; Jiménez et al. 2024). Consistent with these insights, we identified that intratracheal pharmacological inhibition or gene silencing of DNMT1 increased the population of CD4⁺CD25⁺FOXP3⁺ Tregs in both BALF and lung tissue of septic mice. Furthermore, similar trends were observed in mice with specific DNMT1 deletion in CD4⁺ T cells (DNMT1-CD4-ko), further substantiating the role of DNMT1 in reducing Treg abundance and activity in septic ALI.

Regarding the functional mechanisms in the associated events, we performed RNA-seq analysis and identified FOXO1 as a significant DEG between Tregs from DNMT1fl/fl and DNMT1-CD4-ko mice. Additionally, RUNX1 was identified as an important interacting protein of DNMT1 through IP-MS analysis. RUNX1, a transcription factor traditionally associated with hematopoiesis, has also been implicated in immune regulation (Padella et al. 2022). As aforementioned, RUNX1 can either activate or repress gene transcription depending on its interacting partners. Here, we performed integrated ChIP-qPCR and luciferase assays, confirming that DNMT1 recruits RUNX1 to the promoter of FOXO1, leading to transcriptional repression of the FOXO1 gene. Interestingly, a previous study by Chao et al. demonstrated that RUNX1, which can be interacted by FOXP3, reduced apoptosis in CD4⁺ T cells and ameliorate inflammatory cascades in septic mice (Chao et al. 2023). Here, we observed that FOXO1-specific knockout in CD4⁺ T cells reduced anti-inflammatory activity of Tregs, leading to exacerbated lung pathology in septic mice. These observations partially align with the findings by Ge et al., where FOXO1 alleviated inflammatory responses and improved lung pathology in septic ALI by promoting M2 polarization of macrophages, a large subgroup of immunosuppressive cells (Ge et al. 2024). FOXO1 has also been identified to improve FOXP3 expression and enhance Treg stability (Shen et al. 2025; Wang et al. 2013; Yang et al. 2024). Consistently, we identified that the loss of FOXO1 in CD4⁺ T cells led to a reduction in Treg abundancy and activity in septic mice.

Notably, we also observed that ThG-mediated DNMT1 inhibition still increased Treg population in FOXO1-CD4-ko mice, indicating that DNMT1 deficiency can also enhance Treg differentiation and function through FOXO1-independent mechanisms. DNMT1 is a global epigenetic regulator that likely affects multiple transcriptional programs beyond FOXO1. As discussed above, previous evidence has demonstrated the direct role of DNMT1 in modulating FOXP3 methylation (Lal et al. 2009; Wang et al. 2013). Additionally, our RNA-seq data also revealed moderate upregulation of other transcription factors such as CTLA4, in DNMT1-CD4-ko mice. The number of CTLA4⁺CD4⁺ T cells in mouse lung tissues was reduced in both DNMT1-CD4-ko mice and FOXO1-CD4-ko mice. Nevertheless, to substantiate the involvement of FOXO1 in Treg differentiation and function following DNMT1 loss, we extracted CD4+ T cells extracted from DNMT1fl/fl and DNMT1-CD4-ko mice and treated them with the FOXO1-specific antagonist AS1842856. The AS1842856 treatment substantially reduced the populations of CD4+CD25+FOXP3+ cells, providing further evidence that FOXO1 is, at least in part, implicated in Treg differentiation and function in the DNMT1-deficiency scenario.

Although this study offers valuable insights into the role of the DNMT1/RUNX1/FOXO1 cascade in sepsis-induced ALI, there are limitations. First, while we observed that blocking FOXO1 reduces Treg activity in DNMT-loss conditions, we did not perform rescue experiments like artificially restoring FOXO1 in DNMT1-overexpressing conditions, primarily due to financial constraints. In addition, the functions of the DNMT1/RUNX1/FOXO1 axis in other types of immune cells crucial in sepsis pathogenesis, such as macrophages, neutrophils, and dendritic cells, were not concerned in the current data. Furthermore, this is an animal-based study, further validation in human clinical trials is necessary to confirm the therapeutic potential of DNMT1 inhibitors like ThG in treating sepsis-induced ALI. Finally, the long-term safety and efficacy of DNMT1 inhibitors require further investigation, as ThG has potential side effects that may limit its clinical application.

Conclusion

In conclusion, our research establishes DNMT1 as an essential regulator in sepsis-induced ALI by repressing FOXO1 and impairing Treg function. Inhibiting DNMT1 restores FOXO1 expression and enhances Treg-mediated immune suppression, leading to reduced inflammation and improved lung integrity. Targeting DNMT1 could be a promising approach for managing lung injury and other inflammatory diseases (Fig. 8). Future research should substantiate the therapeutic potential of DNMT1 inhibition in human patients and investigate its effects on a broader range of immune cells to fully elucidate its role in inflammation.

Fig. 8.

Graphic Abstract. DNMT1 as an essential regulator in sepsis-induced ALI by repressing FOXO1 and impairing Treg function. Inhibiting DNMT1 restores FOXO1 expression and enhances Treg-mediated immune suppression, leading to reduced inflammation and improved lung integrity

Author contribution

Jurong Ding: Conceptualization, methodology, formal analysis, data curation, writing—original draft. Benyong Xu: Investigation, methodology, validation, writing—review and editing. Mingyan Wu: Data analysis, visualization. Mengling Zhan: Resources, project administration. Shanmei Wang: Supervision, funding acquisition, writing—review and editing. Haiwen Lu: Project administration, funding acquisition, supervision, writing—final approval.

Funding

This research is supported by Bethune Foundation project (bj-rw2020001j). 

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Shanghai Pulmonary Hospital Ethics Committee. All animal experiments complied with the guidelines of the Shanghai Pulmonary Hospital Animal Care and Use Committee (Approval number: K21-354).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.