IL-35 enhances IL-10⁺ breg-mediated immunoregulation and attenuates inflammation and fibrosis in systemic sclerosis

Abstract

Objective

This study aimed to explore the role of interleukin-35 (IL-35)-mediated regulatory B cells (Bregs) in modulating inflammation and fibrosis in systemic sclerosis (SSc).

Methods

Peripheral blood and skin samples from 132 treatment-naïve SSc patients and 58 healthy controls (HCs) were analyzed for IL-35 levels, B cell subsets, and cytokine profiles. CD19⁺ B cells were stimulated with IL-35 or anti-IL-35 monoclonal antibody (mAbIL-35) and co-cultured with autologous CD3⁺ T cells to evaluate immunoregulatory function. A bleomycin-induced SSc mouse model was used to assess in vivo effects.

Results

SSc patients exhibited elevated plasma IL-35, IL-10, IL-6, and BAFF levels, along with increased CD19⁺ B cells and IL-6⁺ effector B cells. However, IL-35⁺ and IL-10⁺ Bregs were significantly decreased in both blood and skin. IL-10, IL-6, and TGF-β1 mRNA levels were increased in skin lesions. IL-35 and IL-10 levels correlated with lung function, skin scores, and inflammatory markers. In vitro, IL-35 promoted IL-10⁺ Breg expansion and cytokine secretion, and suppressed Th1, Th17, and CD8⁺IFN-γ⁺ T cell responses via an IL-10-dependent mechanism. In vivo, IL-35 alleviated fibrosis and inflammation in bleomycin-treated mice, whereas mAbIL-35 exacerbated it.

Conclusion

While IL-35 is compensatorily elevated in SSc, its endogenous levels are insufficient to curb the disease. However, the therapeutic administration of exogenous IL-35 demonstrates significant potential.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13075-026-03735-8.

Introduction

Systemic sclerosis (SSc) is an autoimmune disorder characterized by a triad of pathological features: immune dysregulation, widespread vascular abnormalities, and progressive fibrosis of the skin and internal organs [1]. Patients show marked expansion of lymphocyte subsets, particularly hyperactivated B cells with distinct transcriptional profiles linked to disease severity [2]. Within the SSc microenvironment, fibrosis and B cells engage in a self-perpetuating cycle. Activated fibroblasts release mediators such as interleukin-6 (IL-6) and transforming growth factor-β (TGF-β), which recruit and further activate B cells [3, 4], In turn, these B cells produce profibrotic cytokines that promote collagen production and sustain fibroblast activation [5].

A critical counterbalance to this process is provided by regulatory B cells (Bregs). Bregs function to suppress the pro-inflammatory effects of effector B cells (Beffs) through the secretion of immunomodulatory cytokines like IL-10 and IL-35, and via cell-surface molecules such as PD-L1 and FasL [6–11]. In SSc, reduced Bregs and IL-10 correlate with interstitial lung disease progression [12, 13], suggesting Breg dysfunction contributes to pathogenesis.

IL-35 is a heterodimeric cytokine composed of p35 and Epstein-Barr virus-induced gene 3 (EBI3) subunits [14], As a new member of the IL-12 cytokine family, it exhibits anti-inflammatory and immunosuppressive properties [15, 16]. Our previous work has established that IL-35 is elevated in the plasma of SSc patients. These levels correlated with regulatory T cell (Treg) frequency and pulmonary HRCT scores, while showing an inverse relationship with Th1 cell counts and forced vital capacity (FVC) [17]. Furthermore, we demonstrated that recombinant human IL-35 (rhIL-35) could suppress proliferation in a co-culture system of patient-derived CD4⁺ T cells and human fibroblasts. This was accompanied by an upregulation of IL-10 and a concurrent downregulation of IL-17 A, α-SMA, and COL-1 [18]. In experimental autoimmune encephalomyelitis (EAE) and uveitis (EAU)mouse models, IL-35 exerts immunoregulatory functions in B cells, and expansion of IL-35⁺/IL-10⁺ Bregs ameliorates disease severity [19, 20], indicating that increased Breg frequency may improve autoimmune diseases.

This study is to investigate the immunomodulatory effects of IL-35 on human primary B-cell subsets secretory functions in SSc. We aim to characterize IL-35–B cell correlations and define the resulting impact on the dysregulated inflammatory and fibrotic pathways characteristic of SSc in vitro and in vivo approaches.

Materials and methods

Patients and samples

Peripheral blood samples were collected from 132 treatment-naïve or treatment-discontinued (≥ 3 months) SSc patients attending the Rheumatology and Immunology Department at The First Affiliated Hospital of Guangxi Medical University (January 2022-December 2024). All patients fulfilled either the 1980 American College of Rheumatology (ACR) [21] or 2013 ACR/European League Against Rheumatism (EULAR) classification criteria for SSc [22]. Exclusion criteria included: coexisting autoimmune/autoinflammatory diseases, active infections, malignancies, transmissible diseases, or severe chronic multiorgan disorders. Fifty-eight age- (± 3 years) and sex-matched healthy controls (HCs) were recruited.

Clinical samples

Peripheral blood (4–10 mL) was drawn from SSc patients and HCs into heparinized or acid citrate dextrose (ACD) tubes. HCs were enrolled from our hospital’s Health Examination Center. Additionally, matched skin samples (n = 9/group) were collected, including lesional skin from SSc patients and normal skin from HCs (obtained as discarded surgical specimens from elective orthopedic/plastic procedures at our institution).

Modified Rodnan Skin Score (mRSS): Skin thickness was assessed at 17 body sites by gentle palpation and scored from 0 (normal) to 3 (severe thickening). Total scores ranged from 0 to 51 [23].

High-Resolution Computed Tomography (HRCT) Score: Two blinded radiologists independently scored each lung lobe for ground-glass opacity, fibrosis, honeycombing, and bronchiectasis based on percentage involvement (0: 0%; 1: 1–25%; 2: 26–50%; 3: 51–75%; 4: 76–100%). Total scores ranged from 0 to 80 [24].

CD19⁺ B cell isolation, culture, and treatment

CD19⁺ B cells were isolated from peripheral blood mononuclear cells (PBMCs) using a human CD19⁺ Isolation Kit (Stemcell, Canada), with purity exceeding 95% as verified by flow cytometry (FCM), and were subsequently cultured at 1 × 10⁶ cells/ml in lymphocyte medium (Stemcell, Canada) supplemented with 100 U/ml penicillin/streptomycin and 15 ng/ml lipopolysaccharide (LPS) in a 5% CO₂ incubator at 37 °C. The cells were divided into three treatment groups receiving either 100 ng/ml rhIL-35 (Peprotech, Cranbury, NJ, USA; Cat# 200-37−100UG), 10 ng/ml anti-IL-35 monoclonal neutralizing antibody (mAbIL-35)(Thermo Fisher Scientific, Waltham, MA, USA; Cat# PA5-46993), or vehicle control (human serum albumin, HSA), with dosages determined based on previous studies and preliminary experiments [25, 26]. At 120 h post-treatment, a portion of cells was analyzed by FCM while the remaining cells were washed with phosphate-buffered saline (PBS), resuspended in fresh medium at 1 × 10⁶ cells/ml, and cultured for an additional 72 h before supernatant collection for enzyme-linked immunosorbent assay (ELISA) analysis.

CD3⁺ T lymphocyte isolation, coculture, and treatment

CD3⁺ T cells were isolated from PBMCs using a human CD3⁺ Selection Kit (Stemcell, Canada), with purity exceeding 95% as confirmed by FCM, and cultured at 1 × 10⁶ cells/ml in lymphocyte medium (Stemcell, Canada) supplemented with 100 U/mL penicillin/streptomycin and 25 µl/ml anti-CD3/CD28 T cell agonist (Stemcell, Canada) under 5% CO₂ at 37 °C. The cells were then divided into five groups for transwell co-culture, with the lower chamber containing T cells and the upper chamber respectively receiving: (1) vehicle (HSA) control, (2) autologous B cells, (3) IL-35-treated B cells, (4) mAbIL-35-treated B cells, (5) IL-35 + mAbIL-10-treated B cells. After 96 h of co-culture, a portion of cells was immediately analyzed by FCM, while the remaining cells were washed, resuspended in fresh medium (1 × 10⁶ cells/ml), and cultured for an additional 72 h before supernatant collection for ELISA analysis.

ELISA

Cytokine levels in plasma or culture supernatants were measured using specific ELISA kits according to the manufacturers’ protocols. Kits from R&D Systems (Minneapolis, MN, USA) were used for IL-10 (sensitivity: 3.9 pg/mL) and IL-6 (sensitivity: 0.626 pg/mL). Kits from Youpin Biotech (China) were used for IL-35, TGF-β1, BAFF, IL-17 A, IFN-γ, and IL-4, with sensitivities of 0.626, 31.25, 31.25, 31.25, 31.25, and 31.25pg/mL, respectively.

Histopathological assessment

All skin sections were obtained from the paravertebral region of the lower back in mice. Sections were stained with hematoxylin and eosin (H&E) and Masson’s trichrome.

Western blot analysis

Protein samples extracted from mouse skin tissues were separated by 10% SDS-PAGE and electrotransferred onto PVDF membranes (Millipore, USA). After blocking with 5% skim milk for 1 h, the membranes were incubated overnight at 4 °C with primary antibodies against either α-SMA (Cell Signaling Technology, USA; 1:1000) or GAPDH (Abcam, UK; 1:20,000). Following incubation with HRP-conjugated secondary antibodies (Cell Signaling Technology, USA) at room temperature for 1 h, protein bands were visualized using an ECL detection system (Tanon, China) and quantified using ImageJ software (National Institutes of Health, USA).

Quantitative Real-Time PCR (qPCR)

Relative mRNA expression levels of human skin were measured by RT-qPCR using the 2⁻^ΔΔCt method, normalized to β-actin levels as the endogenous control. Each sample was assayed in technical duplicate (two replicates per reaction), with three independent biological replicates performed for each condition. Primer sequences, included those for β-actin (forward: AGATCAAGATCATTGCTCCTCCTG, reverse: AGTCATAGTCCGCCTAGAAGCAT), IL-6 (forward: TAACCACCCCTGACCCAACC, reverse: AACAATCTGAGGTGCCCATGC), IL-10(forward: CCTGCCTAACATGCTTCGAG, reverse: GGTTCTCAGCTTGGGGCATC), EBI3 (forward: CATCATCAAGCCCGACCCTC, reverse: CCCTGACGCTTGTAACGGAT), P35 (forward: TCCTCCCTTGAAGAACCGGA, reverse: GACAACGGTTTGGAGGGACC), and TGF-β1 (forward: TTATTGAGCACCTTGGGCACT, reverse: TGTCTTCTTCACTATCCCCCACT), with β-actin serving as the internal control for normalization.

Measurement of hydroxyproline content in mouse skin

The hydroxyproline content in mouse skin was determined using a hydroxyproline assay kit (Solarbio, China). Briefly, frozen skin samples were weighed, minced. The mixture was boiled in a metal bath until complete dissolution was achieved. Based on preliminary experiments, the resulting supernatant was appropriately diluted prior to analysis. All subsequent procedures were performed strictly according to the manufacturer’s instructions. The hydroxyproline content in the tissue was calculated as specified in the kit protocol.

Immunofluorescence

For immunofluorescence co-localization analysis, random 200× magnification images were acquired by fluorescence microscopy. Using ImageJ software, we quantified the percentage of co-localized cells (co-localized cells/total cells ×100%) to compare SSc and HC groups. The following reagents were used: Rat polyclonal CD19 antibody (Columbia, MD, USA; Cat# 32727), a Mouse IL-6 antibody (Santa Cruz Biotechnology, USA; Cat# SC-28343), Mouse monoclonal CD19 antibody (Novus Biologicals, USA; Cat# NBP2-25196SS), Rat Anti-IL-10 antibody (Abcam, Cambridge, UK; Cat# ab133575), and Rat IL-35 Antibody (Abbexa, Cambridge, UK; Cat# Abx274658),

Flow cytometry

Human PBMCs or murine splenic single-cell suspensions were cultured in complete medium containing 50 ng/mL phorbol 12-myristate 13-acetate (PMA), 750 µg/mL ionomycin, 5 µg/mL brefeldin A (BFA), 15 µg/mL LPS, and 10% fetal bovine serum (FBS) at 37 °C with 5% CO₂ for 5 h. Following incubation, cells were treated with Fc receptor blocking reagent (BioLegend, USA) for 5 min at room temperature, then stained with Zombie Red™ Fixable Viability Dye (BioLegend, USA) for 15 min at 4 °C in the dark. Subsequent surface marker staining (30 min, 4 °C) was followed by fixation/permeabilization using BD Cytofix/Cytoperm™ Kit (BD Biosciences, USA) for intracellular staining with fluorochrome-conjugated antibodies.Surface markers were detected using FITC Mouse Anti-Human CD19, APC-Cy7 Mouse Anti-Human CD19, and PE-Cy7 Rat Anti-Mouse CD19 (all from BD, USA), along with FITC Mouse Anti-Human CD138 and BV510 Mouse Anti-Human CD3 (BD, USA). Cytokine profiling included PE Monoclonal Antibody IL-12 p35 (cross-reactive for human/mouse; eBioscience), Human/Mouse IL-12/IL-35 p35 Alexa Fluor^® 647-conjugated Antibody (R&D Systems), PE-Cy7 Rat anti-human IL-10 (BD, USA), and BV421 Rat anti-mouse IL-10 (BD, USA). Transcription factor analysis employed BB700 Mouse Anti-Human FoxP3 (BD, USA). For Th cell characterization, we used PE-Cy7-Mouse Anti-Human IFN-γ, Alexa Fluor 647-Mouse Anti-Human IL-17 A, APC-Cy7-Mouse Anti-Human CD8 (BD, USA), and PE-IL-4 Monoclonal Antibody (eBioscience, USA), PE-IL-6 Monoclonal Antibody(eBioscience, USA).Appropriate isotype controls were implemented: PE-/PE-Cy7-/APC-/BV421-/BB700-Mouse Anti-Human IgG1 (BD, USA), PE- Rat IgG1κ (eBioscience, USA), Mouse IgG1 AF 647-/700-conjugated Antibody (R&D Systems), and PE-Cy7- Rat IgG1κ (BioLegend). All staining procedures followed manufacturer-recommended concentrations with spectral compensation adjustments for multicolor flow cytometry.

BLM-induced SSc mouse model and administration of IL-35 and mAbIL-35

Thirty-two 8-week-old female BALB/c mice (Vital River, China) were equally divided into four groups to establish a bleomycin-induced systemic sclerosis model. All groups received intravenous tail vein injections every other day according to the following schedule: days 1, 3, 5, 7, and 9 (total of five injections).The control group received daily subcutaneous dorsal injections of phosphate-buffered saline plus every other day intravenous injections of phosphate-buffered saline. The model group [27] received daily subcutaneous dorsal injections of bleomycin(BLM) (1 mg/ml; Hanhui Pharmaceutical, China) plus every other day intravenous injections of phosphate-buffered saline. The IL-35 group received daily subcutaneous dorsal injections of bleomycin plus every other day intravenous injections of recombinant murine IL-35(rmIL-35) (10 µg/ml; Chimerigen, USA, # CHIMF-11135-C025). The mAbIL-35 group received daily subcutaneous dorsal injections of bleomycin plus every other day intravenous injections of anti-interleukin-35 monoclonal antibody (5 µg/ml; Sigma-Aldrich, Merck KGaA, Germany, #MABF848). All injections were administered in a volume of 100 µl. After 21 days, tissues were collected following euthanasia (approved by Guangxi Medical University’s Animal Ethics Committee) (Fig. 4A). Additionally, both the mAbIL-35 group and its corresponding isotype control antibody (InVivoPlus mouse IgG2b isotype control, BioXcell, USA, Cat# BP0086) group were included in the BLM mouse model, with results provided in Supplementary Material 1.

Fig. 4.

Effects of interleukin-35 on inflammation and fibrosis in a mouse model of subcutaneous bleomycin injection. A The experimental timeline. B Flow cytometry analysis of splenic IL-10+Breg and IL-35+Breg cells. C Plasma concentrations of IL-17 A, interferon-γ (IFN-γ), and IL-10. D Skin sections stained with hematoxylin & eosin (H&E) and Masson’s trichrome, along with dermal thickness measurements. E Western blot analysis of α-SMA (alpha-smooth muscle actin) in skin tissue and semi-quantitative results. F Quantification of hydroxyproline content in skin tissues. G.Correlation heatmap. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; ns, not significant. Scale bar: 50 μm. Magnification: 200×

Statistical analysis

The normality of all data distributions was formally assessed using the Shapiro-Wilk test. Data conforming to a normal distribution are presented as mean ± standard deviation and were analyzed using independent samples t-tests or one-way analysis of variance (ANOVA). Non-normally distributed data are expressed as median (interquartile range, IQR) and were analyzed via Mann-Whitney U tests or Kruskal-Wallis H tests. For the analysis of categorical variables, the chi-squared test was applied. Correlation analyses were performed using Pearson’s product-moment correlation for normally distributed data and Spearman’s rank correlation for non-normally distributed data. A two-tailed P value < 0.05 was considered statistically significant. All statistical calculations were performed using GraphPad Prism 8.0 and SPSS 26.0 software.

Result

Baseline characteristics of patients with SSc

A total of 132 patients with SSc were enrolled, including 85 females and 47 males, with a mean age of 53.8 ± 10.1 years. No significant differences were observed in sex and age between the groups. The median disease duration was 2 years. Baseline clinical data comparisons between groups are shown in Table 1.

Table 1.

Demographic, clinical, and laboratory characteristics of systemic sclerosis patients and healthy donors

Characteristic	Healthy controls (n = 58)	Systemic sclerosis(n = 132)
Gender, (male/female)	20/38	47/85
Age, mean (SD), years	53.53 ± 5.789	53.77 ± 10.10
Disease duration (years)	-	2.79 ± 2.05
lcSSc/dcSSc, n(%)	-	13(9.8)/119(90.2)
Anti-Scl-70(+), n(%)	-	103(78.0)
Anti-ANA(+), n(%)	-	107(81.1)
Anti-ACA(+), n(%)		4(3.0)
ESR, mean (SD), mm/h	-	27.0(16.0, 43.8)
CRP, mean (SD), mg/dl	-	7.0(1.7, 20.4)
Skin, n(%)	-	132(100)
Interstitial Lung Disease, n(%)	-	113(85.6)
Pulmonary hypertension, n (%)	-	22(16.6)
Bone and joint, n(%)	-	71(53.7)
Gastrointestinal tract, n(%)	-	17(12.8)
mRSS	-	16.7 ± 9.6
HRCT score	-	31.8 ± 16.0
FVC%pred(n = 80)	-	84.2 ± 16.2
FEV1%pred(n = 80)	-	85.8 ± 18.0
DLCO%pred(n = 80)	-	59.3 ± 18.0

Abbreviation: lcSSc limited cutaneous systemic sclerosis, dcSSc diffuse cutaneous systemic sclerosis, Anti-Scl-70 Anti-topoisomerase I antibody, Anti-ANA Anti-nuclear antibody, Anti-ACA Anti-centromere antibody, ESR Erythrocyte sedimentation rate, CRP C-reactive protein, mRSS Modified Rodnan skin score, HRCT High-resolution computed tomography, FVC%pred Forced vital capacity (% predicted), FEV1%pred Forced expiratory volume in 1 s (% predicted), DLCO%pred Diffusing capacity of the lungs for carbon monoxide (% predicted)

Expression of peripheral blood B - cell subsets and related cytokines

Peripheral blood CD19⁺ B cell and CD19⁺IL-6⁺ Beff frequencies were significantly higher in SSc patients than in HC: 9.267 ± 3.587% vs. 7.608 ± 3.480% (P = 0.0031) and 4.816 ± 4.294% vs. 2.159 ± 1.052% (P < 0.001, respectively) (Fig. 1A). Conversely, frequencies of CD19⁺IL-10⁺ Breg and CD19⁺IL-35⁺ Breg cells were reduced in SSc: 2.715 ± 1.316% vs. 3.358 ± 1.227% (P = 0.006) and 0.4342 ± 0.3027% vs. 1.435 ± 1.003% (P < 0.001, respectively) (Fig. 1A). Plasma levels of IL-35 [17.67 (12.89, 21.99) pg/ml vs. 13.91 (8.202, 16.47) pg/ml, P = 0.011], IL-10 [23.81 (13.73, 40.22) pg/ml vs. 15.01 (7.16, 25.65) pg/ml, P = 0.011], IL-6 (15.87 ± 13.23 pg/ml vs. 5.18 ± 5.79 pg/ml, P = 0.027), and B-cell activating factor (BAFF) (139.20 ± 18.91 pg/ml vs. 127.01 ± 9.15 pg/ml, P = 0.027) were all significantly increased in SSc patients compared with HC (Fig. 1B).

Fig. 1.

Expression levels of B cell subsets and related cytokines in peripheral blood of systemic sclerosis patients and healthy controls. A Flow cytometry gating strategy and staining patterns showing frequencies of B cells, interleukin-10+ regulatory B cells (IL-10+Breg), IL-35+Breg cells, and IL-6+ effector B cells (IL-6+Beff). B Plasma concentrations of IL-35, IL-10, B cell activating factor (BAFF), and IL-6. C Relative mRNA expression of IL-10, IL-6, and transforming growth factor-β1 (TGF-β1) in skin tissue. D Immunofluorescence images of IL-10+Breg cells in skin tissue and their proportion among CD19+B cells (scale bar: 50 μm, 400× magnification). E Immunofluorescence images of IL-35+Breg cells in skin tissue and their proportion among CD19+B cells (scale bar: 50 μm, 400× magnification). F Heatmap showing correlations between circulating B cell subsets/cytokines and clinical parameters. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns, not significant

Expression of skin B - cell subsets and related cytokines

Compared with HC, SSc patients exhibited no significant difference in the frequency of skin CD19⁺ cells. However, skin IL-10⁺ Breg cell and IL-35⁺ Breg cell frequencies were significantly reduced in SSc: 1.279 ± 0.6382% vs. 4.572 ± 2.418% (P = 0.0017) and 1.372 ± 0.6364% vs. 5.665 ± 2.163% (P = 0.0002, Fig. 1D, E), respectively. No notable change was observed in IL-6⁺ Beffs frequencies between groups.

Quantitative analysis of skin tissue revealed upregulated mRNA expression of IL-10 (1.491 ± 0.6141 vs. 0.5216 ± 0.3078, P = 0.0017), IL-6 (0.7568 ± 0.4783 vs. 0.2925 ± 0.1861, P = 0.0045), and TGF-β1 (1.608 ± 0.7755 vs. 0.7730 ± 0.6570, P = 0.0182) in SSc patients compared with HC (Fig. 1C).

Correlation analysis

Correlation analysis in patients with SSc revealed the following associations: Circulating IL-35 levels were positively correlated with erythrocyte sedimentation rate (ESR), IL-10⁺ Breg cell frequencies, forced expiratory volume in 1 s (FEV1% pred), and FVC% pred. Conversely, IL-35 exhibited inverse correlations with BAFF, CD19⁺ B cell proportions, HRCT scores, and modified Rodnan skin score (mRSS). Circulating IL-10 levels showed positive associations with IL-35, IL-10⁺ Breg frequencies, FEV1% pred, and FVC% pred, but a negative correlation with HRCT scores. Circulating IL-6 levels were positively associated with mRSS, whereas BAFF levels negatively correlated with both FEV1% pred and FVC% pred. CD19⁺ B cell proportions correlated positively with HRCT scores and mRSS but negatively with ESR. IL-10⁺ Breg cell frequencies were positively associated with IL-35 levels and FVC% pred but negatively correlated with HRCT scores and mRSS. IL-35⁺ Breg cell frequencies demonstrated a significant negative correlation with mRSS(Fig. 1F).

Proportion of Breg subsets and cytokine secretion in activated primary B lymphocytes

After activated by LPS, the proportion of IL-10⁺Breg subsets and IL-10 secretion in activated primary B cells from SSc patients were significantly lower compared to HCs: IL-10⁺Breg proportion: 1.95 (1.00, 2.36) vs. 2.53 (2.30, 3.73), P = 0.0379;IL-10 secretion (pg/mL): 6.467 (5.59, 8.62) vs. 17.97 (7.75, 53.77), P < 0.05 (Fig. 2A、C).

Fig. 2.

Effect of Interleukin-35 on Breg Subset Variation and Cytokine Secretion. A IL-10-secreting regulatory B cells (IL-10+Breg). B Supernatant IL-35 concentration. C Supernatant IL-10 concentration. P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001; ns, not significant

Effects of IL-35 on Breg subset differentiation and cytokine secretion in primary B lymphocytes in vitro

After the addition of rhIL-35, the proportions of IL-10⁺Bregs in both SSc patients and HC increased significantly in the rhIL-35 group [6.14 (4.37, 14.5) vs. 1.95 (1.00, 2.36), P < 0.001; 10.60 (5.06, 20.7) vs. 2.53 (2.30, 3.73), P < 0.001, respectively]. Additionally, the concentrations of IL-10 and IL-35 were significantly elevated (P < 0.05) (Fig. 2). Following rhIL-35 intervention, numerical increases were observed in the proportions of IL-35⁺ Bregs and CD19⁺CD138⁺ cells, although these changes did not reach statistical significance (P > 0.05). In the mAbIL-35 group, IL-35 concentration exhibited a downward trend, while no significant differences were observed in the remaining indicators (P > 0.05).

RhIL-35 mediates the suppression of T-cell subset proliferation and secretion through an IL-10-dependent pathway in B cells

Compared to HCs, the concentration of IL-17 A in the culture supernatant of activated primary CD3⁺ T cells from SSc patients was significantly increased (P < 0.05). After Transwell co-culture with autologous CD19⁺ B cells, Compared to the control groups (including T cells alone and B + T cell co-cultures) of both SSc and HC, the T cells + autologous rhIL-35-treated B cells group showed significantly decreased proportions of CD3⁺CFSE^low T cells, CD4⁺CFSE ^low T cells, and CD8⁺CFSE^low T cells. Additionally, the proportions of Th17 CFSE ^low cells, Th17 cells, Th1 CFSE ^low cells, Th1 cells, CD8⁺IFN-γ⁺ cells, and CD8⁺IFN-γ⁺ CFSE ^low cells were significantly reduced, while the proportions of Treg CFSE ^low cells and Treg cells were significantly increased. The Th17/Treg ratio was significantly decreased, and the secretion of IL-17 A and IFN-γ was significantly reduced, with all differences being statistically significant (P < 0.05). Although the Th1/Th2 ratio showed a decreasing trend, the difference was not statistically significant (P > 0.05). In contrast, the T cells + autologous mAbIL-35-B cells group did not exhibit any significant differences in these parameters. Neutralizing IL-10 reversed the IL-35-mediated effects of B cells on T cell subset proliferation (Fig. 3).

Fig. 3.

Interleukin-35 (IL-35) mediates B cell suppression of effector T cell proliferation and cytokine secretion in an IL-10-dependent pathway. A Heatmap of various T cell subsets proliferation. B Secretion of IL-17 A. C Secretion of IFN-γ. *P < 0.05 compared with the T cells + autologous rhIL-35-treated B cells group; #P < 0.05 compared with the HC group receiving the same treatment

IL-35 increases the frequency of Breg cells in the spleen of BLM-induced SSc mice and reduces inflammatory cytokines

Compared with the control group, the proportion of IL-10⁺Breg cells in the spleen was significantly increased in the model group, and further elevated in the IL-35 group (P < 0.05), while no significant difference was observed in the mAbIL-35 group. Compared with the model group, the proportion of IL-35⁺Breg cells in the spleen was significantly higher in the IL-35 group (P < 0.05), whereas no significant differences were detected in the control or mAbIL-35 groups (P > 0.05). Compared with the control group, plasma concentrations of IFN-γ and IL-17 A were significantly elevated in the model group (P < 0.05). These levels were significantly lower in the IL-35 group than in the model group (P < 0.05), while the mAbIL-35 group showed no significant differences (P > 0.05). Compared with the model group, plasma IL-10 levels were significantly increased in the IL-35 group (P < 0.05), with no significant changes observed in the control or mAbIL-35 groups (P > 0.05) (Figs. 4B, C).

IL-35 alleviates skin fibrosis in BLM-injected mice

To evaluate the effects of IL-35 and mAbIL-35 on skin fibrosis, histopathological examination of skin tissue sections revealed that the dermal thickness in control mice was 57.29 ± 18.82 μm, with no significant collagen proliferation. In contrast, BLM-injected model mice exhibited marked collagen proliferation and a significant increase in dermal thickness (118.20 ± 13.64 μm, P < 0.05). The IL-35 group showed mild collagen proliferation and a reduced dermal thickness (89.97 ± 11.96 μm, P < 0.05 vs. model group), whereas the mAbIL-35 group displayed aggravated collagen proliferation and a further increase in dermal thickness (133.30 ± 14.00 μm, P < 0.05 vs. model group) (Fig. 4D).Compared with the control group, model mice demonstrated significantly elevated expression of α-SMA and hydroxyproline, while the IL-35 group showed lower levels than the model group, and the mAbIL-35 group exhibited higher expression (P < 0.05) (Fig. 4E) (Full uncropped Western blot membranes for all protein analyses are available in Supplementary File 2). Due to unstable/intermittent lung fibrosis development in our BLM model, we could not reliably evaluate fibrotic changes in lung tissues. This limitation underscores the challenge of standardizing pulmonary fibrosis in murine SSc models.

Correlation analysis indicated that IL-10 levels were negatively correlated with Hydroxyproline content, dermal thickness at the injection site, and IFN-γ levels. Additionally, IL-10⁺ Breg cells showed a negative correlation with Hydroxyproline content, dermal thickness, IFN-γ, and IL-17 A, whereas IL-35⁺ Breg cells were negatively correlated with plasma IFN-γ and IL-17 A levels (Fig. 4F).

Discussion

Cytokines and immune cells play pivotal roles in SSc’s immunoinflammatory and fibrotic processes [5, 28–30]. IL-35 has been shown to inhibit Th17 cell function and suppress the proliferation and secretory activity of human fibroblasts [18]. It also promotes the expansion of Tregs and Bregs [31]. Exogenous IL-35 can effectively increase IL-10 production in B cells in a concentration-dependent manner. Furthermore, IL-35 promotes the phosphorylation of STAT1 and STAT3 in B cells, enhancing their ability to suppress IFN-γ and TNF-α secretion by T cells [25]. While most studies report elevated serum IL-35 levels in early-stage SSc patients versus late-stage or healthy controls [30, 32], some show decreased plasma levels negatively correlating with mRSS [33], suggesting disease stage- and medication-dependent variations. Our findings align with majority reports, showing significantly increased plasma IL-35 and IL-10 in SSc patients versus HCs. IL-35 positively correlated with IL-10, IL-10⁺Bregs, IL-35⁺Bregs, ESR, and pulmonary function (FEV1%, FVC%), while negatively correlating with CD19⁺ B cells, HRCT scores, and mRSS, indicating its potential inhibitory roles in both inflammatory and fibrotic processes of SSc and SSc-ILD.

Our study observed elevated circulating IL-35 and IL-10 levels but reduced proportions of IL-35⁺ Breg and IL-10⁺ Breg cells in SSc patients, suggesting these cytokines may primarily originate from other immune sources such as Tregs [34–38]. Previous studies have shown an increased proportion of IL-35-secreting Tregs in SSc peripheral blood [38], and activated macrophages in SSc patients exhibit enhanced IL-10 secretion [39]. The elevated IL-35 levels could reflect an early stress response or fibroinflammatory activation (potentially via TGF-β/Smad pathways) and might function as a compensatory feedback mechanism to moderate disease progression [40–42]. However, this interpretation remains speculative, as our study lacks longitudinal or direct functional evidence to establish causality.

B cells play pivotal roles in SSc pathogenesis by regulating T cell and fibroblast activity through antigen presentation, cytokine secretion, and direct cellular interactions [2, 12, 43, 44]. Importantly, Bregs maintain immune tolerance through multiple mechanisms including IL-10, IL-35 and TGF-β secretion [6–11]. In SSc patients, decreased Breg proportions and reduced IL-10 secretion are associated with the development and progression of ILD [12, 13]. These patients demonstrate significant reductions in both CD19⁺CD24^hiCD38^hi and CD19⁺CD24^hiCD27^hi Breg subsets, along with diminished IL-10 secretory capacity upon Breg activation [13]. O. Aravena et al. reported that TIM-1⁺ B cells from healthy controls effectively suppress CD4⁺ T-cell activity in vitro, whereas this immunosuppressive function is markedly impaired in SSc patients [45], indicating a Breg functional defect in SSc. Our results are consistent with these observations, showing that LPS-stimulated B cells from SSc patients exhibit both quantitative and functional Breg deficiencies, including significantly reduced IL-10⁺ Breg frequencies (p < 0.01) and lower IL-10 secretion (p < 0.05) compared to healthy controls.

The immunoregulatory capacity of IL-35 in B cells has been well-established in autoimmune models, where IL-35⁺Bregs may be essential for EAE recovery [19] and IL-35-induced Bregs suppress EAU by modulating Th17/Th1/Treg responses [20]. Our study extends these findings to human systems, demonstrating that IL-35 treatment enhances B cell regulatory function through: (1) increased IL-10/IL-35 secretion and IL-10⁺Breg proportions; (2) suppression of CD3⁺/CD4⁺/CD8⁺ T cell proliferation; and (3) modulation of inflammatory T cell subsets (reduced Th1/Th17, increased Tregs). These effects were IL-10-dependent, as neutralization abolished immunosuppression. Notably, mAbIL-35 (EBI3-targeting) showed no significant impact, possibly due to incomplete signaling blockade or limited basal IL-35 production.

SSc involves complex clinical manifestations that cannot be fully reproduced by any single animal model. In this study, we employed the BLM-induced dermal fibrosis model, one of the most widely used in SSc research [27], which offers high reproducibility and rapidly induces a robust fibrotic response—features essential for initial therapeutic assessment. This model also recapitulates several key disease processes relevant to our investigation, including dermal fibrosis, B-cell activation, and dysregulation of cytokines such as IL-17 A and IFN-γ, thereby providing a pertinent platform to evaluate the immunomodulatory role of IL-35. Nevertheless, we acknowledge that the BLM model does not replicate certain autoimmune components of human SSc, such as autoantibody-dependent pathology or systemic autoimmunity. Therefore, while our results clearly demonstrate the efficacy of IL-35 in this inflammatory-fibrotic setting, it remains to be determined whether its protective effects extend to antibody-driven fibrotic mechanisms. We propose that future studies involving adoptive transfer of pathogenic antibodies (e.g., anti-topoisomerase I), graft-versus-host disease and antibody-induced nephritis models would help clarify the broader applicability of IL-35–based immunomodulation in fibrosis diseases.

High-dose IL-35 has been demonstrated to possess dual anti-inflammatory and anti-fibrotic properties across multiple chronic inflammatory conditions. Supporting evidence comes from studies on systemic lupus erythematosus, coronary artery diseases, and pulmonary fibrosis, where its therapeutic potential has been studied [25, 46, 47]. Our study extends these findings by demonstrating that despite elevated IL-35 levels, fibrosis continues to progress in patients. This suggests that the increased IL-35 likely represents a compensatory feedback response that ultimately proves insufficient. In contrast, our experimental models show that administration of IL-35 at supraphysiological concentrations significantly attenuates both inflammation and fibrosis. These results indicate that achieving therapeutic efficacy may require supraphysiological enhancement of IL-35 signaling.

IL-35 distinguishes itself from other Breg-regulating cytokines (such as IL-21 and IL-10) by uniquely expanding IL-10⁺ Bregs while simultaneously suppressing effector T cells and fibrosis - a dual mechanism that is crucial in SSc. Unlike IL-10, which primarily exerts anti-inflammatory effects, IL-35 concurrently enhances STAT1/STAT3-dependent Breg suppression of IFN-γ/TNF-α and inhibits fibrosis [25], a hallmark of SSC progression. While IL-21 promotes Breg plasticity, its pro-fibrotic role limits therapeutic utility [48, 49]. Thus, IL-35’s dual modulation of immune and fibrotic pathways positions it as a promising SSC treatment target, warranting further clinical exploration. It is also important to note that the p35 detection methods used in our flow cytometry, immunofluorescence, and PCR assays may recognize the p35 subunit present in IL-12. Therefore, to specifically identify IL-35, we adopted a double-positive strategy targeting both p35 and EBI3 in our assays. Similarly, since the IL-35 used in both in vitro and in vivo experiments may also cross-react with the p35 subunit of IL-12, we performed neutralization targeting the EBI3 subunit to help clarify the distinct functional contributions of the IL-35 heterodimer.

This study has several limitations: (i) The underlying molecular mechanisms for both the impaired IL-10 production in SSc B cells and its enhancement by IL-35 remain unclear and require further elucidation. (ii) the lack of longitudinal patient analysis to track IL-35 and Breg dynamics across disease stages; (iii) no direct evaluation of cellular sources of IL-35 in human SSc samples; and (iv) potential differences between murine models and human SSc in IL-35 response. Future studies employing single-cell RNA sequencing, B-cell receptor profiling, and detailed signaling analyses may help clarify Breg dysfunction in SSc and identify key regulators of IL-35-mediated IL-10 production. Future work should also explore IL-35’s function in diverse SSc models, for example, through adoptive transfer models employing pathogenic antibodies (e.g., anti-topoisomerase I), to better evaluate its therapeutic scope across different fibrotic and immune mechanisms.

In summary, peripheral blood levels of IL-35, IL-10, IL-6, and BAFF are significantly elevated in SSc patients compared to healthy controls, accompanied by an imbalance in total B-cell and Breg proportions. Our data indicate that IL-35 can increase the frequency of IL-10⁺ Bregs and enhance secretion of IL-10 and IL-35 in primary SSc B cells. We propose that IL-35 may promote SSc B cells to secrete IL-10, thereby inhibiting the proliferation and cytokine secretion of Th17, Th1, and CD8⁺ T cells, and subsequently suppressing immune inflammation and fibrosis (Fig. 5). While IL-35 is compensatorily elevated in SSc, its endogenous levels appear inadequate to control disease progression. Nevertheless, the therapeutic administration of exogenous IL-35 demonstrates considerable potential warranting further investigation.

Fig. 5.

IL-35 ameliorates systemic sclerosis by enhancing IL-10⁺ regulatory B cell function and suppressing effector T cell proliferation/cytokine production, while neutralizing IL-35 exacerbates skin fibrosis

Authors’ contributions

Wen Zeng designed the study, conducted all experiments, and wrote the manuscript. Dongmei Pan performed flow cytometry and immunofluorescence assays using human peripheral blood and skin tissues, and contributed to manuscript writing. Chunxiu Lu carried out flow cytometry, PCR, and ELISA experiments with human peripheral blood and skin tissue samples. Jie Pan participated in experimental design. Xu Wang and Mu Huang were responsible for mouse model establishment. Hanyou Mo and Letting Zheng contributed to study design. Ling Lei participated in experimental design and manuscript review. All authors critically reviewed and approved the final manuscript.

Funding

This study was supported by the Guangxi Key Research and Development Program Special Fund for Science and Technology Development (2024AB17049), the Guangxi Natural Science Foundation Key Project Joint Special Project for Research on Regional High-Incidence Diseases (2023GXNSFDA026061), and the Self-funded Scientific Project of the Health Commission of Guangxi Zhuang Autonomous Region (Z-A20240534).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The study was approved by the Institutional Review Board (GXMUUH-2021-IRB-045), with written informed consent obtained from all participants. Clinical trial number: not applicable.

Consent for publication

All authors have read and approved the final manuscript and consent to its publication.

Competing interests

The authors declare no competing interests.