Introduction

Inflammatory diseases are recognized as a leading cause of death worldwide1. Excessive inflammation disrupts immune equilibrium, resulting in pain and tissue damage. Consequently, numerous therapeutic strategies have been developed to mitigate inflammation, including nonsteroidal anti-inflammatory drugs (NSAIDs), glucocorticoids, and disease-modifying antirheumatic drugs (DMARDs)2,3. In addition, cellular therapy has gained attention as a novel therapeutic approach for immune regulation4,5,6. Regulatory T (Treg) cells play an important role in modulating adaptive immune responses and have been employed in clinical trials for refractory Crohn’s disease and type 1 diabetes7,8.

Macrophages are highly plastic innate immune cells that rapidly respond to changes in the microenvironment9. They can polarize along a spectrum, and the two extremes of this spectrum are classically activated (M1) macrophages and alternativly activated (M2) macrophages10. M1 polarization is typically induced by LPS and interferon (IFN)-γ, while M2 polarization is driven by interleukin (IL)−4. M2 macrophages display anti-inflammatory and tissue reparative functions, characterized by the expression of genes such as arginase 1 (Arg1, Arg1), resistin-like molecule alpha (Retnla, Fizz1), mannose receptor C type 1 (Mrc1, CD206), early growth response 2 (Egr2, Egr2), and chitinase-like 3 (Chil3, Ym1). In contrast, M1 macrophages are proinflammatory and express high levels of tumor necrosis factor (Tnfa, TNF)-α, IL-6, inducible nitric oxide synthase (Nos2, iNOS), CXCL9 (Cxcl9), and CD86 (Cd86), contributing to host defense. However, excessive activation of M1-like proinflammatory macrophages could result in maladaptive immune responses, contributing to diseases such as acetaminophen-induced liver injury, inflammatory bowel disease (IBD), and diet-induced obesity11,12,13,14. Therefore, identifying strategies to alleviate excessive M1 macrophage activation in inflammatory diseases is of pivotal importance.

Conventional Treg cells display a crucial role in regulating macrophage polarization through multiple mechanisms. For instance, Treg cells facilitate macrophage emigration and M2-like polarization during the regression of atherosclerosis15. Soluble factors derived from Treg cells have been shown to inhibit NF-κB signaling and support M2 polarization in inflammatory environment16. CD4+CD25+ Treg cells have been reported to promote M2-like macrophage differentiation and inhibit M1 polarization, partly through IL-10, TGF-β, and arginase in SCID mice17. Proto et al. further revealed that Treg cell-derived IL-13 induces macrophages to produce IL-10 and enhances apoptotic cell engulfment via an autocrine-paracrine mechanism18. These findings highlight the key role of Treg cells in regulating the development and function of macrophages.

Our previous studies and those of others have demonstrated that B cells possess the ability to convert CD4+CD25− naïve T cells into CD4+Foxp3− Treg cells, termed Treg-of-B cells, in the presence of anti-CD3 and anti-CD28 stimulation, without additional cytokines19,20,21. Although Treg-of-B cells lack Foxp3 expression, they upregulate a range of regulation-associated markers, including cytotoxic T-lymphocyte-associated protein 4 (CTLA4, CD152), lymphocyte-activation gene 3 (LAG3, CD223), inducible T cell costimulator (ICOS, CD278), glucocorticoid-induced TNFR-related protein (GITR, CD357), TNFRSF4 (OX40, CD134) and programmed cell death protein 1 (PD1, CD279). Different from other inducible Tregs (iTregs), Treg-of-B cells represent a novel subset of Treg cells22. Treg-of-B cells exhibit immunomodulatory capacity and inhibit lymphocyte proliferation through both IL-10- and tumor growth factor (TGF)-β-dependent and cell-cell contact mechanisms23,24,25,26. Moreover, Treg-of-B cells have been shown to regulate macrophage-mediated inflammation in murine models of collagen-induced arthritis24 and monosodium urate (MSU)-induced air-pouch gout models27.

In the present study, the modulatory capacity of Treg-of-B cells on macrophage polarization was investigated using bone marrow-derived macrophages (BMDMs). Treg-of-B cells induced M2-like phenotype and part of M1-like markers in unpolarized (M0-like) macrophages, as evidenced by the expression of Arg1, Retnla, Mrc1, Egr2, and Nos2. In M1-like macrophages, Treg-of-B cells inhibited the expression of Nos2, Tnfa, Cd86, and Cxcl9, as well as the production of TNF-α, IL-6, and nitrite. Treg-of-B cells also secreted Th2 cytokines, including IL-10, IL-4, and IL-13, and their production was enhanced when cocultured with macrophages. Mechanistically, the regulatory effects of Treg-of-B cells were mediated through cell-cell contact and the contact-dependent induction of soluble factors, particularly Th2 cytokines. Furthermore, Treg-of-B cells were found to decrease NF-κB pathway activation, possibly through accumulating IκBα and inhibiting RNA expression of Kruppel-like factor 4 (Klf4). In summary, these findings suggest that Treg-of-B cells modulate macrophage polarization in inflammatory environments.

Methods

Animals

All BALB/c mice (5- to 7-week-old for BMDM differentiation, 6- to 8-week-old for Treg-of-B cells) were purchased from the National Laboratory Animal Center under specific pathogen-free conditions. All mice were housed in the Laboratory Animal Center, College of Medicine, National Taiwan University, and handled in accordance with the approved protocols of the Institutional Animal Care and Use Committee at the College of Medicine, National Taiwan University (license 20210137). This study is reported in accordance with ARRIVE guidelines.

Preparation and polarization of BMDMs

Preparation of BMDMs was performed as described previously28,29. Briefly, bone marrow cells were harvested from 5- to 7-week-old female BALB/c mouse femurs and tibias. After removing red blood cells with 1× RBC lysis buffer (420301, BioLegend, USA), bone marrow cells were washed and filtered through a 70 μm cell strainer. The cells were suspended in RPMI 1640 medium (SH30027.02, Gibco, Thermo Fisher Scientific, USA) supplemented with 10% fetal bovine serum (10437028, Gibco), 292 µg/ml L-glutamine (BII03-020-1B, Biological Industries, Israel), 1% nonessential amino acids (11140050, Gibco), 1% HEPES buffer solution (03-025-1B, Gibco), 2 mM sodium pyruvate (SI-S8636, Gibco), 0.05 mM 2-mercaptoethanol (M7522, Gibco), and 1% penicillin–streptomycin-amphotericin (BII03-033-1B, Biological Industries). A total of 6 × 106 cells were seeded in 10 mL medium in a petri dish and supplemented with 20 ng/ml M-CSF (576406, BioLegend). On Day 3, 5 mL of medium containing 20 ng/ml M-CSF was carefully added. The cells were then incubated for 5–7 days at 37°C and 5% carbon dioxide. For collection, the attached cells were washed twice with Ca-and Mg-free cold PBS, and incubated with cold Accumax™ cell detachment solution (STEMCELL, Canada) for 15 min at room temperature. The suspended BMDMs were collected and washed with MACS buffer (PBS with 0.5% bovine serum albumin and 2 mM EDTA) for subsequent analysis. For polarization, cells were plated at a density of 1.5 × 106 cells/1.5 mL each well in a 6-well plate and rested overnight. For M1 polarization, 100 ng/ml LPS (L2880, MilliporeSigma, USA) and 20 ng/ml IFN-γ (315-05, PeproTech Inc., USA) were used; for M2 polarization, 20 ng/ml IL-4 (214 − 14, PeproTech) and 20 ng/ml IL-13 (575902, BioLegend) were used, both for 24 h.

Preparation of Treg-of-B cells

Culture of Treg-of-B cells was performed as described previously30. In brief, B cells were isolated by positive selection using magnetic bead-conjugated antibodies (anti-mouse CD45R/B220 magnetic particles, 551513, BD, USA). CD4+ T cells were separated using the EasySep™ CD4+ T cell isolation kit (19852, STEMCELL) and then negatively selected using a magnetic field column. The suspended cells were washed three times with MACS buffer. CD4+ cells were incubated with PE-conjugated anti-CD25 (558642, BD) and then anti-PE magnetic particles (557899, BD) in the dark. CD4+CD25- T cells were isolated in suspension using a magnetic field column. The purity of the cells was confirmed by flow cytometry and was at least 90%.

Preparation of Treg-of-B cells was performed as described previously23. Briefly, B220+ B cells and CD4+CD25- T cells were cocultured at a ratio of 1:1 in RPMI 1640 medium supplemented with 5% FBS, 292 µg/ml L-glutamine, 1% nonessential amino acids, 1% HEPES, 2 mM sodium pyruvate, 0.05 mM 2-mercaptoethanol, and 1% penicillin–streptomycin-amphotericin. Cells were seeded in a 48-well plate, with each well containing 106 CD4+CD25- T cells in the presence of 1 µg/mL anti-CD3 (100340, BioLegend) and 1 µg/mL anti-CD28 (102116, BioLegend) antibodies. After 3 days, Treg-of-B cells were collected using Ficoll-Paque PLUS (GE17-1440-02, MERCK, Germany) to remove dead cells and B220 magnetic beads to deplete B cells.

Coculture of BMDMs and Treg-of-B cells

BMDMs were prepared as described and plated at 1.5 × 106 cells/1.5 mL each well in a 6-well plate. The next day, 1.5 × 106 Treg-of-B cells were added to BMDMs in the presence of 1 µg/ml anti-CD3 and 1 µg/ml anti-CD28 antibodies for 24 h. Coculture conditions were added LPS/IFN-γ and IL-4/IL-13 for M1 and M2 polarization, respectively. Transwell experiments were performed in 6-well plates. In the neutralization experiment, anti-IL-4 antibody (504122, BioLegend), anti-IL-13 antibody (16-7135-85, eBioscience, USA), anti-IL-10 antibody, or corresponding control isotype antibodies were added to the coculture at a final concentration of 10 µg/ml.

Flow cytometry

Cells were suspended in FACS buffer (PBS with 2% FBS and 1% NaN3) and stained with fluorescence-conjugated antibodies against molecules and the corresponding control isotype at 4℃ for 30 min in the dark. Monoclonal antibodies against CD11b, F4/80, CD80, CD206, and Annexin V for BMDMs; B220 for B cells; and CD4, CD25, Foxp3, ICOS, OX40, PD1, GITR, CTLA4 and LAG3 for T cells were purchased from eBioscience. Intranuclear staining was performed following intracellular antigen staining protocols. Cells were analyzed using FACSCalibur™ (BD) with CellQuest™ v4.0.2 (BD) and analyzed by FlowJo v10.7.2 (BD).

Real-time polymerase chain reaction (quantitative PCR, qPCR)

Total RNA was extracted using TRIzol reagent (PT-kp200, Invitrogen™ Thermo Fisher Scientific) and then reverse transcribed to cDNA using RevertAid reverse transcriptase (EP0442, Thermo Fisher Scientific), random hexamer primers (5070029000, Genestar Biotech, Taiwan), dNTPs (BM0110-0001, Genestar Biotech), and DEPC H2O according to the manufacturer’s instructions. Quantitative PCR was performed using the IQ2 SYBR green ROX mix (BB-DBU-008, Bio-Genesis, Taiwan) in the StepOnePlus™ Real-Time PCR machine (Applied Biosystems, Life Technologies, USA). A primer list is provided in Table S1. The data were normalized to the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (Gapdh), calculated using the 2-△△Ct method, and expressed in relative quantification (R.Q.).

Western blot

Western blotting was performed as described previously27. Cells were homogenized with RIPA lysis buffer (Invitrogen™ Thermo Fisher Scientific) containing phosphatase/protease inhibitor cocktail. The protein concentration was determined using a Pierce™ BCA protein assay kit (Invitrogen™ Thermo Fisher Scientific) according to the manufacturer’s instructions. Equal amounts of protein were separated by SDS-PAGE and transferred onto a PVDF membrane. The membrane was incubated overnight at 4℃ with different primary antibodies: iNOS (anti-rabbit, 1:1000, 13120, Cell Signaling Technology, USA), Arg-1 (anti-rabbit, 1:1000, 93668, Cell Signaling Technology), IκB-α (anti-rabbit, 1:500, sc-371, Santa Cruz Biotechnology, USA), and phospho-IκB-α (anti-mouse, 1:1000, sc-8404, Santa Cruz Biotechnology). The reactive bands were detected using chemiluminescent HRP substrates (34080, Thermo Fisher Scientific) and ECL. The photographs were obtained by Chemidoc XRS (Bio-Rad, USA) and processed by Image Lab V6.1.0 (Bio-Rad). Full-length blots are provided in Supplementary Fig. 4.

Enzyme-linked immunosorbent assay (ELISA)

The supernatant was collected, and the levels of IL-2, IL-4, IL-10, IL-13, IL-17, IFN-γ, and TNF-α were determined using DuoSet ELISA Development Systems (R&D, USA) according to the manufacturer’s instructions. The absorbance at optical density (OD) 450 nm and 550 nm was measured on a VERSA max tunable microplate reader (Molecular Devices, USA) with SoftMax Pro Software v5.4.1 (Molecular Devices).

Nitrite quantification

Levels of nitrite in the supernatant were measured immediately by the Griess Reagent Kit (Invitrogen™ Thermo Fisher Scientific) according to the manufacturer’s instructions.

[3H]-incorporation assay

Purified 105 CD4+ T cells were stimulated with mitomycin C (MMC, M4287, MilliporeSigma)-treated 105 splenocytes in the presence of 1 µg/ml anti-CD3 and 1 µg/ml anti-CD28 antibodies in each well of a 96-well round-bottom plate. Treg-of-B cells and CD4+ T cells were seeded at a ratio of 1:1 and cultured at 37℃ and 5% CO2. After 3 days, 1 µCi [3H]-thymidine (PerkinElmer, USA) was added to each well for an additional 16–18 h. Cells were collected onto glass fiber paper using a harvester. The incorporation of [3H]-thymidine was measured by a β-counter (Packard Instrument Co., USA). The results are presented in counts per minute (c.p.m.).

Statistical analysis

Two-way repeated measures analysis of variance (ANOVA) or one-way ANOVA with Tukey’s multiple comparisons test was used to analyze the data by GraphPad Prism 8 Software. Data are presented as the mean ± SD. A P value < 0.05 was considered significant.

Results

Treg-of-B cells induce an intermediate phenotype in macrophages

We first confirmed that Treg-of-B cells exhibited characteristics consistent with previous reports22,30. Naïve B cells induced naïve splenic CD4+CD25− T cells (Fig. 1a) to differentiate into CD25+Foxp3− Treg-of-B cells (Fig. 1b and c), which expressed conventional Treg-associated markers, including CTLA4, LAG3, ICOS, GITR, OX40, and PD1. Treg-of-B cells secreted higher levels of IL-10, IL-4, and IL-13, but lower levels of IL-2 and IFN-γ (Fig. 1d), and suppressed CD4+ T-cell proliferation in a dose-dependent manner (Fig. 1e and f).

Fig. 1
figure 1

Characterization of B-cell-induced regulatory T (Treg-of-B) cells. (a) Purity of splenic B cells, CD4+CD25− T cells (CD25− T), and Treg-of-B cells. (b) Representative histogram of markers on Treg-of-B cells. Red lines indicate specific antibody staining; clear histograms indicate corresponding isotype controls. (c) Representative histogram of Foxp3 on CD4+CD25− T (CD25− T), CD4+CD25+ (nTreg), and Treg-of-B cells. (d) Cytokine production from activated CD4+CD25− T (CD25− T) cells, Treg-of-B cells, and nTreg cells. (e) Suppressive effects of Treg-of-B cells and nTreg cells on CD4+ responder T-cell proliferation (Resp T) upon stimulation with MMC-treated splenocytes and anti-CD3/CD28 monoclonal antibodies (APC + mAb). (f) Dose-dependent suppression by Treg-of-B cells at 0.5×, 1×, and 2× ratios. Data are presented as mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; ****, P < 0.0001 compared to Resp T or indicated groups.

Full size image

BMDMs were induced with high viability and over 90% purity using macrophage colony-stimulating factor (M-CSF) (Supplementary Fig. 1). To investigate the modulatory effects of Treg-of-B cells on macrophage polarization, unpolarized M0-like macrophages were cocultured with either Treg-of-B cells or CD4+CD25− T (CD25− T) cells (Fig. 2a). Surprisingly, Treg-of-B cells induced RNA expression of certain M1-like markers, including inducible Nos2 and Cxcl9, without affecting Tnfa and Cd86 (Fig. 2b). Simultaneously, Treg-of-B cells increased RNA expression of M2-related genes, including Arg1, Retnla, Mrc1, and Egr2 (Fig. 2c), although the increases were less pronounced than those observed in fully polarized M2 macrophages (Supplementary Fig. 2). In contrast, CD25− T cells enhanced M1-like gene expression but did not upregulate M2-related genes. Together, Treg-of-B cells modulated macrophages to exhibit an intermediate phenotype, characterized by both M2-related gene expression and some M1-associated markers.

Fig. 2
figure 2

Treg-of-B cells induce an intermediate macrophage phenotype. (a) Experimental design for investigating the modulatory effects of Treg-of-B cells or CD4+CD25− T (CD25− T) cells on unpolarized M0-like macrophages (BMDM). (b) RNA expression of M1-associated genes in M0-like macrophages. (c) RNA expression of M2-associated genes in M0-like macrophages. Data are presented as the mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Full size image

Treg-of-B cells downregulate the M1-like macrophage phenotype

TLR ligands and/or IFN-γ are known to drive macrophage polarization toward the M1 phenotype, characterized by the production of proinflammatory cytokines, reactive oxygen species (ROS) and reactive nitrogen species (RNS), and enhanced phagocytic capacity31,32,33. LPS/IFN-γ stimulation induced M1-related phenotypes, including Nos2, Tnfa, Cd86, Cxcl9, and CD80 (Supplementary Fig. 2). In contrast, IL-4/IL-13 stimulation promoted M2-related phenotypes, such as Arg1, Retnla, Chil3, Egr2, and CD206. Consistent with previous studies, M2-like macrophages exhibited significantly higher CD206 expression, whereas in vitro-induced M1-like macrophages expressed lower amounts of CD20634. To assess the regulatory impact of Treg-of-B cells, macrophages were polarized toward the M1 phenotype in the presence or absence of Treg-of-B cells (Fig. 3a). M1-like proinflammatory macrophages express iNOS, which competes with arginase 1 for L-arginine and catalyzes its conversion into L-citrulline and nitric oxide (NO)35. NO is spontaneously oxidized into nitrite, which serves as a measurable indicator of reactive nitrogen intermediates36. In M1-like macrophages, Treg-of-B cells reduced the expression of M1-associated genes, such as Nos2, Tnfa, Cd86, and Cxcl9 (Fig. 3b), and suppressed TNF-α, IL-6 (Fig. 3f), and nitrite production (Fig. 3e). This inhibitory effect was not observed in cocultures with CD25− T cells. Moreover, the modulation of M1 polarization by Treg-of-B cells was dose-dependent (Fig. 3g). Interestingly, coculture with Treg-of-B cells led to a modest increase in M2-related gene expression (Fig. 3c), although these levels remained lower than those seen in fully polarized M2 macrophages (Supplementary Fig. 2). Overall, Treg-of-B cells attenuated M1-like phenotype but did not fully abolish it.

Fig. 3
figure 3

Treg-of-B cells downregulate the M1-like macrophage phenotype. (a) Experimental design assessing the effect of Treg-of-B cells or CD4+CD25− T (CD25− T) cells on macrophages (BMDM) under M0 (RPMI only) or M1 (LPS + IFN-γ) polarization. (b) RNA expression of M1-associated genes in macrophages. (c) RNA expression of M2-associated genes in macrophages. (d) Protein expression of iNOS in macrophages. (e) Nitrite concentration in culture supernatants. (f) TNF-α and IL-6 levels in culture supernatants. (g) Dose-dependent suppression of M1-related gene expression by Treg-of-B cells at 1× and 3× ratios. Data are presented as the mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Full size image

Treg-of-B cells enhance anti-inflammatory cytokine production in cocultured macrophages

Treg-of-B cells have been reported to secrete anti-inflammatory and Th2 cytokines, such as IL-4, IL-10, and TGF-β23, which are known to suppress NF-κB activation37. To investigate the role of these cytokines in macrophage regulation, cytokine production in cocultures was assessed (Fig. 4a). The levels of IL-10, IL-4, and IL-13 were significantly increased when Treg-of-B cells were cocultured with macrophages under both M0 and M1 polarization, but not in cocultures with CD25− T cells (Fig. 4b). Freshly isolated (rest) Treg-of-B cells expressed higher levels of Il10, Il4, and Il13 than CD25− T cells, and this phenomenon was augmented following coculture with macrophages (Fig. 4c). Notably, these cytokine RNA expressions were also upregulated in macrophages cocultured with Treg-of-B cells (Fig. 4d), suggesting bidirectional induction of Th2 cytokines in both Treg-of-B cells and macrophages.

Fig. 4
figure 4

Treg-of-B cells enhance anti-inflammatory cytokine production in cocultured macrophages. (a) Experimental design to explore the effects of Treg-of-B cells on cytokine production in macrophages cocultured with either Treg-of-B cells or CD4+CD25− T (CD25− T) cells under M0 and M1 polarization. (b) IL-10, IL-4, and IL-13 concentrations in coculture supernatants. (c) RNA expression in Treg-of-B cells or CD25− T cells freshly isolated (rest) and after coculture with macrophages (cocultured). (d) RNA expression in macrophages. Data are presented as the mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Full size image

Treg-of-B cells modulate M1 polarization via cell-cell contact and soluble factors

Previous studies have demonstrated that Treg cells suppress macrophage foam-cell constitution through both cell-cell contact and soluble factors38. To understand the mechanisms by which Treg-of-B cells regulate M1 polarization, conditioned medium (CM) from Treg-of-B cell and macrophage cocultures were used to treat M1-like macrophages (Fig. 5a). The CM group significantly downregulated M1-related gene expression (Fig. 5b), reduced nitrite production (Fig. 5c), and decreased TNF-α and IL-6 levels (Fig. 5d), suggesting that soluble factors contributed to the suppression of M1 polarization. To further validate this, a transwell system was used to separate macrophages from Treg-of-B cells (TW group) or from Treg-of-B cells and macrophages (TW + M group) (Fig. 5e). The TW group abrogated the inhibition observed in direct coculture (CC group), as evidenced by the restoration of Nos2, Tnfa, Cd86, and Cxcl9 RNA expression (Fig. 5f). Notably, the TW + M group partially retained the suppression of Nos2, Tnfa, and Cd86 RNA expression. Collectively, these findings indicate that Treg-of-B cells inhibit the M1-like phenotype primarily through cell-cell contact and, at least in part, via contact-dependent induction of soluble factors.

Fig. 5
figure 5

Treg-of-B cells modulate M1 polarization via cell-cell contact and soluble factors. (a) Experimental design to assess the effect of conditioned medium (CM) from Treg-of-B and macrophages cocultures. (b) RNA expression of M1-associated genes in macrophages. (c) Nitrite concentration in culture supernatants. (d) TNF-α and IL-6 levels in culture supernatants. (e) Transwell system design to separate Treg-of-B cells and macrophages under M0 or M1 polarization: direct coculture (CC), transwell-separated Treg-of-B cells (TW), and transwell-separated Treg-of-B + macrophages (TW + M). (f) RNA expression of M1-related genes in macrophages. Data are presented as the mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 compared to the corresponding M1-like macrophages or indicated groups.

Full size image

IL-10, IL-4, and IL-13 contribute to Treg-of-B-mediated regulation of macrophages

Given the upregulation of IL-4, IL-10, and IL-13 in Treg-of-B cell cocultures (Fig. 4c), the role of these cytokines was analyzed using neutralizing antibodies (Fig. 6a). Neutralization of IL-4 significantly restored nitrite production in M1-like macrophages cocultured with Treg-of-B cells (Fig. 6c), although Nos2 RNA expression showed only a slight and non-significant increase (Fig. 6b). IL-10 neutralization slightly elevated Tnfa and Cd86 RNA expression (Fig. 6b), consistent with a trend toward increased TNF-α levels (Fig. 6d). Neutralization of IL-13 and IL-10 significantly reversed the TNF-α suppression. Additionally, IL-6 levels were modestly increased upon IL-10 blockade and significantly increased when both IL-10 and IL-4 were neutralized (Fig. 6d). Together, these data implicate IL-4, IL-10, and IL-13 as key contributors to the regulatory effects of Treg-of-B cells.

Fig. 6
figure 6

IL-10, IL-4, and IL-13 contribute to Treg-of-B-mediated regulation of macrophages. (a) Experimental scheme showing Treg-of-B cells cocultured with macrophages in the presence of neutralizing antibodies against IL-4 (α-IL-4), IL-13 (α-IL-13), and IL-10 (α-IL-10) or corresponding isotype controls (IC). (b) RNA expression of M1-related genes in macrophages. (c) Nitrite levels in culture supernatants. (d) TNF-α and IL-6 levels in culture supernatants. Data are presented as the mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 compared to the corresponding M1 or indicated cells.

Full size image

Treg-of-B cells inhibit NF-κB activation and Klf4 expression in macrophages

NF-κB activation plays an important role in M1 polarization and TNF-α production. Upon activation, IκBα is phosphorylated and degraded, leading to NF-κB complex nuclear translocation39. Next, the effect of Treg-of-B cells on IκBα expression in macrophages was analyzed (Fig. 7a). M1-like macrophages exhibited IκBα degradation compared with M0-like macrophages (Fig. 7b). However, IκBα expression gradually accumulated over time in M1-like macrophages cocultured with Treg-of-B cells.

Fig. 7
figure 7

Treg-of-B cells inhibit NF-κB activation and Klf4 expression in macrophages. (a) Experimental design to analyze IκBα expression in macrophages cocultured with Treg-of-B cells under M0 or M1 polarization. (b) Time-course analysis of IκBα expression in M1-like macrophages cocultured with Treg-of-B cells for 0.5, 1, and 2 h. (c) Experimental design for dose-dependent analysis. (d) RNA expression of Klf4 in macrophages cocultured with Treg-of-B cells at 1× and 3× ratios. Data are presented as the mean ± SD from three independent experiments. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparisons test. *, P < 0.05; ***, P < 0.001; ****, P < 0.0001 compared to the corresponding M0 or indicated groups.

Full size image

KLF4 is a transcription factor known to cooperate with NF-κB members such as p65 (RelA) and p300/CBP to activate Nos2 expression40. Accordingly, the expression of Klf4 in M1-like macrophages was examined under different coculture ratios with Treg-of-B cells (Fig. 7c). M1-like macrophages exhibited elevated Klf4 expression (Fig. 7d), but this was reduced significantly upon coculture with increasing numbers of Treg-of-B cells. Taken together, these results indicate that Treg-of-B cells downregulate M1-like macrophages by inhibiting NF-κB signaling and Klf4 expression.

Treg-of-B cells attenuate M1-like macrophage activation via contact and Th2 cytokines

Upon LPS/IFN-γ stimulation, macrophages displayed M1-like features, including increased expression of Nos2, Cd86, and Cxcl9 as well as proinflammatory cytokines such as TNF-α and IL-6 (Supplementary Fig. 3). CD4+Foxp3− Treg-of-B cells secreted IL-10, IL-4, and IL-13 at steady state and further elevated cytokine production upon interaction with macrophages. These cells facilitated the expression of M2-related genes, such as Arg1, Retnla, Mrc1, and Egr2 and also Nos2 in macrophages. Importantly, Treg-of-B cells suppressed the production of nitrite, TNF-α, and IL-6 in M1-like macrophages, with IL-10, IL-4, and IL-13 contributing, at least in part, to the regulation. Moreover, Treg-of-B cells increased IκBα accumulation and decreased Klf4 expression, collectively reducing NF-κB activation. These findings highlight the regulatory potential of Treg-of-B cells in modulating macrophage responses in inflammation.

Discussion

In this study, Treg-of-B cells induced the expression of M2-associated genes, including Arg1, Retnla, Mrc1, and Egr2, as well as the M1-related Nos2 in M0-like macrophages, leading to the emergence of an intermediate phenotype (Fig. 2). When cocultured with Treg-of-B cells (Fig. 3) or treated with conditioned medium from Treg-of-B cell and macrophage cocultures (Fig. 5), M1-like macrophages showed attenuated expression of Nos2, Tnfa, Cd86, and Cxcl9, as well as decreased production of TNF-α, IL-6, and nitrite. Coculture with Treg-of-B cells triggered bidirectional regulation of Th2 cytokines, including IL-10, IL-4, and IL-13 (Fig. 4), between macrophages and Treg-of-B cells. Mechanistically, the regulatory effect was mediated through both cell-cell contact and contact-dependent induction of soluble factors. Among these, IL-4, IL-10, and IL-13 played roles in regulating inflammatory mediators such as nitrite, TNF-α, and IL-6 (Fig. 6). These effects were associated with preserved IκBα expression, attenuated activation of the NF-κB pathway, and reduced Klf4 expression in macrophages (Fig. 7). Taken together, these findings reveal that Treg-of-B cells regulate macrophage polarization under inflammatory conditions.

While conventional CD4+Foxp3+ Treg cells primarily promote M2 polarization through IL-1015,18 and TGF-β17, Foxp3− Treg-of-B cells produce IL-10, IL-4, and IL-13 (Fig. 4), suggesting a distinct regulatory mechanism. Treg-of-B cells inhibited M1-associated markers but did not completely polarize macrophages toward an M2 phenotype. Specifically, Treg-of-B cells attenuated Nos2 expression and nitrite production in M1-like macrophages but did not completely abolish them (Fig. 3), and even elevated Nos2 expression in M0-like macrophages primarily via contact-dependent mechanisms (Fig. 2). Consistently, TNF-α and IL-6 production remained detectable in M1-like macrophages cocultured with Treg-of-B cells, indicating the presence of an intermediate macrophage phenotype. These observations support the concept that macrophage polarization exists along a continuum spectrum rather than as a strict binary state41. Treg-of-B cells appear to modulate macrophage plasticity, shifting them toward an intermediate phenotype that limits excessive inflammation.

Macrophages are typically polarized toward the M1 phenotype by stimulation with microbial products and/or proinflammatory cytokines, particularly LPS and IFN-γ10. The current study employed LPS/IFN-γ stimulation to induce M1-like markers, such as TNF-α, IL-6, Nos2, Cd86, and Cxcl9 (Fig. 3). Treg-of-B cells decreased the expression of M1-associated phenotypes, including TNF-α, Nos2, Cd86, and Cxcl9 at the RNA level, as well as TNF-α and IL-6 at the protein level in the coculture system (Fig. 3). In addition, nitrite levels were decreased in the coculture medium. Collectively, these findings indicate that Treg-of-B cells downregulate M1-like macrophage phenotypes.

Treg-of-B cells have previously been reported to produce higher levels of IL-10 and IL-4 than splenic CD4+CD25− T cells23. IL-10 and surface LAG3 on Treg-of-B cells synergistically suppressed CD4+ T cell proliferation, and LAG3+ Treg-of-B cells were shown to inhibit proinflammatory cytokines in collagen-induced arthritis24. Treg-of-B cells have also been shown to suppress NLRP3 inflammasome activation in BMDMs, primarily via a contact-dependent pathway27. In this study, IL-4 significantly inhibited IL-6 production in M1-like macrophages, especially when combined with IL-10 (Fig. 6). IL-4 also regulated nitrite levels, although this effect was not reflected at the transcriptional level. IL-10 played a role in the modulation of TNF-α, particularly in combination with IL-13. As reported previously, IL-13 derived from Treg cells can induce IL-10 production in macrophages18, which might explain the synergistic effects observed between IL-10 and IL-13. It is worth noting that IL-10 deficiency does not impair the suppressive function of Treg-of-B cells in experimental colitis23, suggesting that Treg-of-B cells possess both contact-dependent and soluble mediators in their modulatory mechanism.

Upon coculture with Treg-of-B cells, macrophages prevented IκBα levels under M1 polarization and decreased Klf4 RNA expression in a dose-dependent manner (Fig. 7). These findings are consistent with previous work showing that Treg-of-B cells decrease IκBα degradation and suppress NLRP3 inflammasome activation27. IL-10 and IL-13 are known to suppress the NF-κB pathway through stabilizing IκBα37 while IL-4 and IL-13 activate STAT6 to upregulate M2-associated genes42. KLF4 has been identified as a context-dependent transcription factor that can interact with p300/CBP and cooperate with p65 at the Nos2 promoter in response to proinflammatory cytokines40. However, KLF4 is also markedly induced in M2 macrophages, where it cooperates with STAT6 to activate M2-associated Arg143,44. These findings suggest that Treg-of-B cells enhance an M2-like phenotype through the expression of IL-4, IL-13, and Klf4, while Klf4 also contributes to the induction of M1-associated Nos2 in macrophages. This dual role supports the notion that Treg-of-B cells modulate macrophages toward a mixed phenotype.

Conclusions

In summary, this study demonstrates that Treg-of-B cells regulate macrophages by inducing an intermediate phenotype, characterized by elevated expression of M2-like markers and partial retention of M1-like characteristics. Both cell-cell contact-dependent mechanisms and contact-dependent induction of soluble factors, particularly IL-10, IL-4, and IL-13, are involved in the regulatory function of Treg-of-B cells. This study also suggests that Treg-of-B cells reduce NF-κB pathway activation, possibly through inhibition of Klf4 expression. Given their high recovery rate30 and demonstrated regulatory functions on both T cells and myeloid cells, Treg-of-B cells represent a potential cellular therapeutic strategy. They have been applied in various models to alleviate dysregulated immune responses, including macrophage-mediated collagen-induced arthritis24, air-pouch gout27, Th2-dominant allergic asthma19, Th1/Th17-initiated IBD23, and contact-dependent hypersensitivity45. However, both the current and previous reports have relied exclusively on murine in vitro systems. Whether Treg-of-B cells can be generated under physiological conditions and whether human Treg-of-B cell equivalents exhibit similar characteristics remains to be elucidated. Further research is required to identify the specific soluble immunomodulatory factors involved in decreasing M1 polarization and to explore the therapeutic potential of Treg-of-B cells in clinical settings.