There were 14 subjects with mild-to-moderate allergic asthma enrolled in this study. Inclusion criteria included a physician diagnosis of allergic asthma and serum allergen-specific IgE>0.35 kUA/L. Exclusion criteria included upper respiratory infection within 4 weeks, prednisone use within 3 months, pregnancy, and other lung diseases or autoimmune diseases. All patients were taking low- or medium-dose ICS-LABA as their daily treatment and were well controlled. Detailed patient information is shown in Table 1. The study was approved by the Ethic Committee of Ruijin Hospital (No. 2019-YK061). All participants signed an informed consent form prior to the study.

Eosinophils and naïve CD4⁺ T cells from allergic asthmatics were separated from peripheral blood by magnetic cell separation system (MACS) using Eosinophil Isolation Kit (#130-092-010, Miltenyi) and Naïve CD4⁺ T Cell Isolation Kit II (#130-094-131, Miltenyi) as previously described (17). Briefly, peripheral blood was layered on Ficoll (GE Healthcare) and centrifuged at 800 g for 20 min. Eosinophils were negatively isolated by depletion of non-eosinophils (magnetically labeled with a cocktail of antibodies and anti-biotin monoclonal antibodies conjugated to MicroBeads) from the granule cell layer suspension. The magnetically labeled non-eosinophils are depleted by retaining them on an MACS column in the magnetic field of a separator, while the unlabeled eosinophils pass through the column. Naïve CD4⁺ T cells were negatively sorted from mononuclear cell layer suspension. Purity of isolated cells was checked by flow cytometry and was always >95%. Purified eosinophils were resuspended at 1.0x10⁶/mL in RPMI 1640 (Hyclone) supplemented with 10% FBS, 1% penicillin/streptomycin (Servicebio), and 10 ng/ml IL-5 (Peprotech). Eosinophils were stimulated with 100 μg/ml house dust mite extracts (HDM, Greer Lab) and/or 1 ng/ml IL-25 (Peprotech) at 37°C in a humidified 5% CO₂ incubator for 18 h, and then cells were harvested for co-culture or direct flowcytometry analysis.

After 18 h stimulation with HDM and/or IL-25, eosinophils (1.0x10⁶/ml) were resuspended with autologous naïve CD4⁺ T cells (10⁶/ml) in 1:1(v:v) RPMI 1640 (Hyclone) with ImunoCult-XF T Cell Expansion Medium (Stemcell) supplemented with 10 ng/ml IL-2 (Peprotech), 10 ng/ml IL-5 (Peprotech) and 25 μl/ml ImmunoCult Human CD3/CD28 T Cell Activator (stem cell). Cells were cultured at 37°C in a humidified 5% CO₂ incubator for 48 h. Individual naïve CD4⁺ T cells (1.0x10⁶/ml) groups served as control, cultured in the same condition medium as above. Four hours before the end of culture, 1:1000 Monensin Solution (Biolegend) was added for intracellular cytokines analysis.

The flow cytometry analyses were performed with a Cytoflex LX instrument (Beckman). Dead cells were excluded by Zombie UV (Biolegend). For eosinophil surface marker analysis, cells were incubated with Fc Receptor Blocking Solution (Biolegend) to eliminate Fc receptor-mediated antibody binding, and then followed with staining of APC/Cy7 anti-human CD16 antibody (Clone:3G8, BD sciences), APC anti-human Siglec-8 antibody (Clone:7C9), PE anti-human CD40 antiobody (Clone : HB14), PE/Cyanine7 anti-human HLA-DR antibody (Clone:L243), Brilliant Violet 421 anti-human CD80 antibody (Clone:L307.4, BD sciences), BB515 anti-human CD86 antibody (Clone: FUN-1, BD sciences), PE/Cyanine7 anti-human PD-L1 antibody (Clone:29E.2A3), PE anti-human OX-40L antibody (Clone:11C3.1) for 20 min at 20°C in the dark, then cells were washed, fixed with Fixation Buffer (Biolgend) and resuspended in Cyto-last Buffer (Biolegend) until analysis.

To measure cytokines production of T cells after co-culture, following surface staining of FITC anti-human CD3 antibody (Clone: UCHT1), PE/Cyanine7 anti-human CD4 antibody (Clone: RPA-T4), cells were fixed by Fixation Buffer (Biolgend) for 20 min at 20°C avoided from light, and then washed twice with Intracellular Permeabilization Buffer. Cells were stained with APC/Fire 750 anti-human IFN-γ antibody (Clone:4S. B3), Brilliant Violet 421 anti-human IL-4 antibody (Clone: G077F6), PE anti-human IL-9 antibody (Clone : MH9A4), APC anti-human IL-17A antibody (Clone: BL168) in remaining permeabilization buffer for 30 min at 20°C in the dark, and then washed and resuspended in FACS buffer for analysis. All antibodies were purchased from Biolegend unless otherwise specified.

IL-25-deficient mice(Il25^-/-)were obtained from Tsinghua University, which have been described in previous studies (18, 19). The genotyping primers were as follows: forward, 5′-CTGCTCCAGTCAGCCTCTCT-3′; reverse1, 5′AGCAGCTGGGCAAGTGAC-3′; reverse2,5′ AGGTGGAGAAAGTGCCTGT-3′. All mice were bred and maintained under specific pathogen-free conditions in Shanghai Biomodel Organism Science and Technology Development Co., Ltd. Sex-matched littermate wild-type (WT) and Il25^-/- mice 6 to 8 weeks of age were mainly used for experiments. All animal studies were approved by the Ruijin Hospital Animal Ethics Committee.

The HDM-induced mouse asthma model was established as described previously (20). Briefly, Il25^-/- and littermate WT mice were sensitized intranasally with 100 μg HDM in 40 μL PBS on Day 0 and challenged intranasally with 10 μg HDM on Days 7 to 11, harvested on Day 14. Histopathologic analysis was done to identify airway inflammation. Bronchoalveolar lavage fluid (BALF) was centrifuged and supernatants were frozen at -80°C for subsequent cytokine analysis. Cytokines IL-4, IL-5, IFN-γ, and IL-13 levels in BALF and serum IgE levels were measured by ELISA according to the manufacturer’s instructions (BioLegend).

In some experiments for investigating the role of eosinophils in the sensitization period, C57bL/6J mice were sensitized with 100 μg HDM before the first challenge and lungs were harvested for flow cytometry analysis at indicated times. To study the effect of IL-25, mice were treated intranasally with 1 μg IL-25(R&D) in 40 μL PBS and lungs were digested to obtain the single cell suspension for flow cytometry analysis 24 h later. To analyze antigen processing by eosinophil, 100 μg HDM and 50 μg DQ-OVA (Thermofisher, US) in 40 μL PBS were intranasally administrated to Il25^-/- and littermate WT mice. Lungs were collected and analyzed by flow cytometry 3 days later.

The ex vivo culture of BmEOS was adapted from the previously published protocols with minor modifications (21, 22). Briefly, bone marrow (BM) was flushed out from the femur and tibiofibula of C57bl/6J mice. After erythrocyte lysis, BM cells were seeded at 1×10^6/ml in IMDM media supplemented with 20% fetal bovine serum (Gibco), 1% penicillin/streptomycin, 2 mM L-glutamine (Gibco), 1 mM sodium pyruvate (Gibco), and 50 mM 2-mercaptoethanol (Sigma-Aldrich) in the presence of 100 ng/mL FLT-3L and 100 ng/mL recombinant murine stem cell factor (SCF, Peprotech), cultured from Day 0 to Day 4. On Day 4 and 8, the media containing SCF and FLT3-L was replaced with media containing 10 ng/mL recombinant mouse interleukin-5 (rmIL-5, R&D Systems) only. Every other day, from this point forward, media was replaced with fresh media containing rmIL-5. After Day 10, cells were seeded and used for subsequent experiments.

BmEOS were seeded in 48 wells plates and stimulated for 2 days with or without HDM and IL-25 in complete medium. To assess DQ-OVA uptake by BmEOS, BmEOS were treated with DQ-OVA for 12 h before being collected, washed, and further analyzed using flow cytometry. Cells cultured in the absence of DQ-OVA were set as a negative control.

Spleens were removed from C57bl/6J 6~8-week-old mice, mechanically disrupted, treated with erythrocyte lysis buffer, and run through 70-μm cell strainer. Suspensions were resuspended in MACS buffer, labelled with an antibody cocktail, and naïve CD4⁺ T cells were negatively separated using the mouse Naïve CD4⁺ T cell Isolation kit (#130-104-453, Miltenyi), in accordance with manufacturer’s instructions. The ex vivo co-culture of BmEOS and CD4⁺T was adapted from the previously published literature with modifications (7). High purity isolated naïve CD4⁺ T cells (1*10^6/ml) were co-cultured with the BmEOS (5*10^5/ml) in supplemented IMDM medium with or without 100 ug/ml HDM in the presence or absence of 50 ng/ml IL-25, with 10 ng/ml IL-2 and 10 ng/ml GM-CSF (R&D) treatment to sustain T cells survival. As controls, CD4⁺ T cells alone were also cultured. Cells were incubated for 96 h at 37°C in a 5% CO₂ humidified atmosphere. During the last 5 h, cells were re-stimulated with cell stimulation cocktail (combined with phorbol 12-myristate 13-acetate (PMA), ionomycin, brefeldin A, and monensin) (eBioscience #00-4975-03) and then were harvested and stained with fluorescent antibodies for flow cytometric analysis.

BALF was collected from lungs via two intratracheal infusions with 0.8 ml PBS using a 20G flushing needle. To obtain single-cell suspensions from lung tissues, lungs were intubated and digested for 1 h at 37°C with intermittent pipetting every 15 min in Hank’s Balanced Salt Solution(HBSS, Invitrogen) media containing 1 mg/mL type 1A collagenase (Sigma) and 30 μg/mL DNase I (Sigma). The digested lung fragments were passed through 70 μm filters, treated with erythrocyte lysis buffer, washed, and resuspended in PBS with 1% FBS.

For surface staining, single-cell suspensions were incubated with fluorochrome-conjugated antibodies cocktails for 30 min at 4°C. For intracellular cytokine staining, cells were fixed and permeabilized using Fixation/Permeabilization Buffer Set (eBioscience) according to the manufacturer’s protocol. Fluorochrome-conjugated anti-mouse mAbs used in these experiments included the following: CD45-BV510, Siglec-F-AF647, CD11c-FITC, CD11b-PECY7, CD4-PECY7, IL-4-BV421, IL-17A-AF647, IFN-γ-FITC, IL-9-PE, CD44-FITC (Miltenyi), CD62L-APC, and CD3-PERCPCY5.5. A fixable viability dye eF780 was used to distinguish viable cells from nonviable cells. All antibodies were purchased from BD Biosciences unless otherwise specified. Flow cytometry data were collected using BD Fortessa or Beckman Coulter CytoFlex S and analyzed by FlowJo X software.

Flow cytometry data were analyzed by CytExpert (Beckman, US) or FlowJo X software. Statistical significance was performed by analysis of one-way ANOVA with post-hoc Holm-Sidak’s test or a two-sided unpaired Student’s t-test. Statistical analysis was performed using Graphpad Prism 8.0 (Graphpad software, US). All data were presented as mean ± SD. P values less than 0.05 were considered statistically significant.

Effective antigen presentation requires expression of both MHC-II and co-stimulatory molecules on APCs, so we first examined whether IL-25 had an impact on the expression of those co-stimulatory molecules of eosinophils from allergic asthmatics in vitro. We sorted eosinophils from peripheral blood and confirmed the purity (Figure 1A). When stimulated with both HDM and IL-25 for 18 h, the expression of HLA-DR, PD-L1, and OX-40L of cultured eosinophils was found to be significantly increased compared with the control group. HLA-DR expression was also elevated in the HDM+IL-25 group compared to HDM or IL-25 alone group. The expression of PD-L1 was significantly higher when cultured with HDM+IL-25 than HDM alone (Figures 1B, C).

Eosinophils were pulsed with HDM and/or IL-25 for 18 h, and then went through a co-culture procedure with autologous naïve CD4⁺ T cells for up to 48 h. Intracellular cytokines expression was tested by flow cytometry. Signature effector cytokines IFN-γ, IL-4, IL-9, and IL-17A were selected to stand for Th1, Th2, Th9, and Th17 cells, respectively. When treated with both IL-25 and HDM, eosinophils further promoted differentiation of autologous naïve CD4⁺ T cells toward Th2 cells, which was unseen in IL-25 or HDM mono-stimulation group. Comparatively, in groups where eosinophils were absent, IL-25 and HDM separately were unable to promote Th cell differentiation effectively (Figure 2).

To determine whether IL-25 is essential for induction of allergic airway inflammation, we subjected IL-25-deficient (Il25^-/-) and littermate wild-type mice to allergic airway inflammation induced by HDM, the most common allergen of allergic asthma in Chinese patients (23) (Figure 3A). After a sensitization and challenge phase with HDM, WT mice developed a severe lung pathology, increased serum IgE, and high Th2-cell-mediated inflammatory responses. We found that Il25^-/- mice, compared with WT mice, had an alleviated airway pathology via H&E staining and periodic acid-Schiff (PAS) assay (Figure 3C), decreased IgE production (Figure 3B) and cellular infiltration in the airways (Figure 3D), less eosinophils (Figure 3E), and reduced type 2-cytokines level, including IL-4, IL-5, and IL-13 (Figures 3F–H). These data suggest that IL-25 is required to maintain disease severity during allergen-mediated Th2 immune responses.

IL-25 was shown in previous research to directly activate conventional DCs in vivo (24). Thus, we sought to determine the effect of IL-25 in vivo on eosinophils. We administered recombinant murine IL-25 (1 μg/mouse) or PBS to WT mice. After 24 h, lungs were harvested and analyzed for infiltrated eosinophils. There were significantly more eosinophils in lungs of IL-25-treated mice compared with PBS-treated control mice (Figure 4A). Thus, IL-25 promotes eosinophils to recruit to lungs in vivo.

Previous studies have identified eosinophils process antigen to promote Th2 cells expansion (6–8). Since the antigen-presenting role occurs during the sensitization phase, we wondered whether IL-25 affects eosinophils during the sensitization phase. A sensitized dose of HDM was administrated intranasally into WT mice and we explored changes in lung eosinophils from day 0 to day 5.We found that eosinophils were significantly increased on the third day after HDM sensitization, suggesting it may exert an antigen-presenting role at this time (Figure 4B). To determine whether IL-25 contributes to the role of eosinophils in sensitization response, we analyzed eosinophils in IL-25^-/- mice 3 days after HDM sensitization (Figure 4C). IL-25^-/- mice exhibited reduced eosinophils compared with WT mice, both in lung and BALF (Figures 4D, E), which demonstrated IL-25 was required in promoting eosinophils recruitment to airway to exert its role in the sensitization phase, before the onset of the effector stage.

Considering the predominant role for antigen uptake in antigen-presenting cells, DQ-OVA, which becomes fluorescent upon processing by antigen-presenting cells, was co-administered intranasally with HDM into IL-25KO and littermate WT mice. Antigen processing by eosinophil was analyzed 72 h later (Figure 5A). IL-25^-/- mice displayed reduced lung eosinophils (Figure 5B) and DQ-labelled eosinophils (Figure 5C) compared to their littermate WT mice. Also, IL-25KO mice had lower mean fluorescence intensity (MFI) for eosinophils than WT mice, indicating that IL-25 deficiency affected DQ-OVA uptake in HDM-treated mice (Figure 5D). The data suggested that IL-25 promotes antigen uptake by eosinophils in vivo.

Next, we generated eosinophils from murine bone marrow (Figure 6A) and identified that BmEOS became mature and reached at high purity after Day 10 (Figures 6B–D). To assess whether IL-25 modulated the ability of eosinophils to uptake antigen in vitro, we treated BmEOS with DQ-OVA when pulsed with or without HDM. When pulsed with HDM, the uptake ability of BmEOS slightly raised but without significant statistical difference compared with the control group. Pulsed with HDM and IL-25 significantly increased uptake of DQ-OVA by BmEOS, as compared to pulsed with HDM alone (Figure 6E). These data suggested that IL-25 promoted the ability of eosinophils to take up the antigen in vitro.

To further investigate the effect of IL-25 on eosinophils in initiating Th cell responses, BmEOS from WT mice and naïve CD4⁺ T cells from spleens were co-cultured in the absence or presence of HDM and IL-25 (Figures 7A, B). The results showed that the percentage of CD4⁺ IL-4⁺ Th2 cells was significantly increased in BmEOS-T co-culture system with HDM plus IL-25 treatment, as compared with HDM treatment alone. When culturing naïve CD4⁺T cells alone, there were no detectable differences in the percentage of CD4⁺ IL-4⁺ Th2 cells among vehicle-treated, HDM-treated, and HDM plus IL-25-treated T cells, suggesting that HDM plus IL-25 treatment did not lead to CD4⁺ Th2 cells-intrinsic differentiation (Figure 7C). In addition, HDM-treated co-cultures developed an increase in the percentages of CD4⁺IFN-γ⁺Th1 cells and CD4⁺IL-9⁺Th9 cells than the vehicle-treatment group. However, there were no significant differences in the proportions of Th1, Th9, and Th17 cells between HDM treatment and HDM plus IL-25 treatment group, no matter in BmEOS-T co-culture or cultures of T cells alone (Figure 7D). These led to the conclusion that IL-25 could promote eosinophil-induced Th2 cell polarization.