T cell activator‐carrying extracellular vesicles induce antigen‐specific regulatory T cells

Abstract

The mechanism of antigen‐specific regulatory T cell (T_reg) induction is not yet fully understood. Curcumin has an immune regulatory function. This study aims to induce antigen‐specific T_regs by employing extracellular vesicles (EVs) that carry two types of T cell activators. Two types of T cell activators, ovalbumin (OVA)/major histocompatibility complex‐II (MHC‐II) and tetramethylcurcumin (FLLL31) (a curcumin analog) were carried by dendritic cell‐derived extracellular vesicles, designated OFexo. A murine model of allergic rhinitis (AR) was developed with OVA as the specific antigen. AR mice were treated with a nasal instillation containing OFexo. We observed that OFexo recognized antigen‐specific T cell receptors (TCR) on CD4⁺ T cells and enhanced Il10 gene transcription in CD4⁺ T cells. Administration of the OFexo‐containing nasal instillation induced antigen‐specific type 1 T_regs (Tr1 cells) in the mouse airway tissues. OFexo‐induced Tr1 cells showed immune suppressive functions on CD4⁺ T cell proliferation. Administration of OFexo efficiently alleviated experimental AR in mice. In conclusion, OFexo can induce antigen‐specific Tr1 cells that can efficiently alleviate experimental AR. The results suggest that OFexo has the translational potential to be employed for the treatment of AR or other allergic disorders.

Administration of OFexo can induce Ag‐specific Tr1 cells in the airway tissues, which efficiently alleviate experimental AR. The results suggest that OFexo has the translational potential to be employed for the treatment of AR or other allergic disorders.

INTRODUCTION

Allergic rhinitis (AR) is an adverse immune response to airborne antigens [1]. The underlying mechanism of AR is not yet fully understood. The clinical symptoms of AR include sneezing, nasal itch, profound nasal discharges and nasal congestion. AR may complicate nasal polyps, rhinosinusitis or asthma [2]. Although administration of corticosteroid nasal spray or antigen histamines can effectively alleviate AR clinical symptoms [3], the therapeutic effects only last for a short time. In fact, AR has been a general health problem globally. It is necessary to further elucidate the pathogenesis of AR and devise novel, safe and more effective remedies for the treatment of AR.

Immune pathologically, AR features as a T helper type 2 (Th2)‐biased immune response in the nasal mucosa [4]. The AR nasal tissues are overpopulated by Th2 cells and saturated with Th2 cytokines [4]. Th2 cytokines, especially interleukin (IL)‐4, induce immunoglobulin (Ig)E production by plasma cells [5]. IgE makes mast cells sensitized. Re‐exposure to specific antigens activates sensitized mast cells; the cells release allergic mediators to induce AR attacks [6]. As immune responses in the body are generally regulated by the immune regulatory system, the skewed Th2 response in the AR nasal tissues suggests a dysfunctional status of the immune regulatory system in the nasal tissues [7]. The underlying mechanism needs to be further investigated.

The immune regulatory system in the body consists of immune regulatory cells, such as regulatory T cells (T_regs) and regulatory B cells (B_regs) and immune regulatory mediators, such as transforming growth factor (TGF)‐β and IL‐10 [8, 9]. Previous studies indicate that the low frequency of immune regulatory cells or/and low production of immune regulatory mediators are associated with the pathogenesis of allergic diseases [10, 11]. Currently, however, the regulatory remedies to recover dysfunctional immune regulatory system are limited.

Published data indicate that extracellular vesicles (EVs) have immune regulatory functions [12]. Many studies have employed EVs to suppress immune inflammation [13]. Dendritic cell (DC)‐derived EVs carry major histocompatibility antigens (MHC‐II) that allow EVs to contact T cells by interacting with the T cell receptors [14]. Previous reports indicate that curcumin, the extracts of the rhizomes of Curcuma longa L. [15], has active immune regulatory functions [16]. However, although curcumin can interact with various immunomodulators, such as DCs, macrophages, T cells and B cells, as well as cytokines and gene transcription factors involved in various immune responses [17], the regulatory effects of curcumin are non‐specific. Therefore, we hypothesize that to equip curcumin with a targeting device may promote its immune regulatory efficacy and make it more specific. To test this, we generated DC‐derived EVs, which carried FLLL31, an analog of curcumin, and ovalbumin (OVA)/MHC‐II complex; the latter served as a targeting device aiming at the antigen‐specific T cell receptor (TCR) [18, 19]. This EV was designated OFexo. The immune regulatory effects of OFexo on inducing antigen‐specific T_regs and alleviating experimental AR were assessed in the present study.

MATERIALS AND METHODS

Reagents

The antibodies of OVA (clone no. 2D11), MHC‐II (7‐1H), c‐Maf (E‐7), Pol II (mara‐3), IL‐10 (A‐2, AF647), CD9 (C‐4), CD63 (MX49.129.5), CD81 (B‐11) and pre‐TCR‐α clustered regularly interspaced short palindromic repeats (CRISPR). Plasmids (m), antibodies for flow cytometry, including CD3 (PS1, AF488), CD4 (5B4, AF546), CD25 (AF594) EpCAM (AUA1, AF488), CD19 (B‐1, AF546), CD11c (N418, AF594), CD117 (E‐3, AF647), FcεRI (H‐5, AF790) and CD154 [F‐1, antigen‐presenting cells (APC], were purchased from Santa Cruz Biotech (Santa Cruz, California, USA). Reagents and materials for real‐time–polymerase chain reaction (RT–qPCR), Western blotting and Siglec F antibody (1RNM44N, AF700) were purchased from Invitrogen (Carlsbad, California, USA). Antigen‐p‐Tyr (P‐Tyr‐1000) antibody was purchased from Cell Signaling Technology (Danvers, Massachusetts, USA). The fluorochrome‐labeled FLLL31, AS101, OVA and fluorescein isothiocyanate (FITC) labeling kits were purchased from Sigma Aldrich (St Louis, Missouri, USA). Enzyme‐linked immunosorbent assay (ELISA) kits of OVA‐specific IgE, mouse mast cell protease‐1, IL‐4, IL‐5 and IL‐13 were purchased from Dakewe BioMart (Beijing, China). Magnetic beads coated with antigen‐CD9 antibody were purchased from AMS Biotechnology (Abingdon, UK). CD4⁺CD62L⁺ T cell isolation and plasmacytoid dendritic cell isolation kits were purchased from Miltenyi Biotech (San Diego, California, USA). Mite allergens were purchased from WolwoBiotech (Hangzhou, China). The level of endotoxin in all reagents was less than 0.001 units/ml, as assessed by the limulus amebocyte lysate gel‐clot assay (E‐Toxate kit; Sigma‐Aldrich).

Generation of bone marrow‐derived DCs (BMDC)

The femurs were cut from naive mice upon euthanasia. The bone marrows (BM) were flushed out from the femurs with saline. Red blood cells in the BMs were lysed with a red blood cell lysis buffer. BM cells were cultured in modified Dulbecco’s Eagle’s medium in the presence of granulocyte macrophage–colony‐stimulating factor (GM‐CSF) (200 U/ml). Nine days later, non‐adherent cells were collected and assessed by flow cytometry. Approximately 90% cells were CD11c⁺. The cells were used as BMDC in the present study.

Generation of EVs

BMDCs were cultured in the presence of OVA (5 mg/ml) or/and FLLL31 (5 mg/ml) overnight. The culture superantigen was collected and processed with established procedures [20] to purify EVs, which was then captured by magnetic beads coated with antigen‐CD9 antibody to be further purified in a magnetic field device [21]. The purified EVs were also CD9, CD63, CD81 and MHC‐II‐positive, as shown by Western blotting (Supporting information, Figure S1a–f). Four types of EVs were prepared: OFexo: containing OVA and FLLL31; Oexo: containing OVA; Fexo: containing FLLL31; and Eexo: did not contain either OVA or FLLL31. These EVs also carried co‐stimulatory molecules CD80, CD83 and CD86, the levels of which were not significantly different between different EVs (Supporting information, Figure S2g–i). For comparison use, MFexo [containing mite extracts (5 mg/ml in the culture) and FLLL31] was prepared. Conversely, T cells (purified from the mouse spleen), instead of BMDCs, were employed to generate EVs to prepare OFexo to be used as control EVs. Additionally, OFexo was also generated with EVs isolated from plasmacytoid dendritic cells to be used as a comparison. We also tested the effects of prepared EVs on regulating the CD4⁺ T cell frequency in naive mice. We found that administration of neither of the EVs altered the CD4⁺ T cell frequency in the airways or spleen of naive mice (Supporting information, Figure S1j,k).

Assessment of generated EVs binding to antigen‐specific CD4⁺ T cells

Generated EVs were labeled with FITC with a FITC‐labeling reagent kit following the manufacturer’s instructions. OVA‐specific CD4⁺ T cells were isolated from the DO11.10 mouse spleen by fluorescence activated cell sorter (FACS) and cultured in the presence of FITC‐EVs for 3 h. The cells were then analyzed by FACS to assess the FITC‐EVs binding to the surface of CD4⁺ T cells.

Induction of Tr1 cells in the mouse airway tissues

BALB/c mice were treated with OFexo in nasal drops (5 mg/ml saline; 50 µl per nostril) daily for 7 days. The mice were killed on day 8. The nasal mucosa and lungs were excised. Tr1 cells in the tissue were assessed by FACS.

Induction and purification of OVA‐specific Tr1 cells in vitro

CD4⁺CD62L⁺ T cells were isolated from the spleen of naive mice by FACS and cultured in the presence of OFexo (10 µg/ml) for 72 h. CD4⁺CD25⁺ IL‐10⁺ Tr1 cells were assessed by FACS to verify that Tr1 cells were generated. To isolate OVA‐specific Tr1 cells, the OFexo‐treated cells were cultured in the presence of OVA (5 µg/ml) overnight. The CD154⁺ T cells (a marker of T cell activation) were purified by FACS and used as OVA‐specific Tr1 cells. As assessed by flow cytometry, these purified CD154⁺ T cells were also CD4⁺ IL‐10⁺ (Supporting information, Figure S2).

Mice

Male BALB/c 6–8‐week‐old mice were purchased from the Beijing Experimental Animal Center (Beijing, China). DO11.10 mice were purchased from Jackson Laboratory (Bar Harbor, Maine, USA). The mice were maintained in a specific pathogen‐free facility at Shanxi Medical University. The mice could freely access food and water. The animal experimental procedures were reviewed and approved by the Animal Ethical Committee at Shenzhen University (no. SZUAE20190302155); all methods were performed in accordance with the relevant guidelines and regulations of animal experiments.

AR mouse model development

Following the published procedures [22], BALB/c mice were sensitized by back skin injection with OVA (0.1 mg/mouse mixed in 0.1 ml alum) on days 0, 3 and 7, respectively. From days 9 to 15, mice were challenged with nasal instillation (50 µl per nostril per day containing OVA at 5 mg/ml) daily. Mice were euthanized on day 16. Nasal itch (nasal scratches) and sneezing were recorded during 2 h after the last challenge with OVA. Truncated blood samples were collected at euthanasia. The serum was isolated from blood samples. The nasal mucosa was excised at euthanasia by scraping. Protein extracts were prepared from the nasal mucosal samples.

Assessment of the effects of OFexo on suppressing experimental AR

Immediately after sensitization (from day 16), AR mice were treated with nasal instillation (containing OFexo or control EVs, 5 mg/ml; saline was used for control group) at 50 µl/nostril daily for 7 days. The AR response was assessed following published procedures [22].

Blocking IL‐10 in mice

Mice were injected intraperitoneally (i.p.) (with AS101 [27 μg in 200 μl phosphate‐buffered saline (PBS)] daily 30 min before each treatment with OFexo.

Cell culture

Cells were cultured in RPMI‐1640 medium. The medium was supplemented with 2 mM glutamine, antibiotics (100 U/ml penicillin and 0.1 mg/ml streptomycin) and 10% fetal calf serum. Cell viability was assessed by the Trypan blue exclusion assay that was greater than 99% before using in further experiments.

Preparation of mononuclear cells from the airway tissues

The airway tissues (including nasal mucosa and lungs) were cut into small pieces (2 × 2 × 2 mm) and incubated with intravenous (i.v.) collagenase (1 mg/ml) at 37°C for 0.5 h with mild agitation. Single cells were collected by filtering the cells through a cell strainer (70 µm) and centrifugation.

Flow cytometry

Cells were harvested from relevant experiments. For the surface staining, cells were incubated with fluorescence‐labeled antibodies (see figure legends for details) or isotype IgG for 30 min at 4°C. For the intracellular staining, cells were treated with fixative/permeable reagents first, and were then incubated with fluorescence‐labeled antibodies or isotype IgG for 30 min at 4°C. After washing with PBS, cells were analyzed with a flow cytometer (FACSCanto II; BD Biosciences, San Jose, California, USA). The data were analyzed with the FlowJo software package with the data obtained from isotype IgG staining as a gating reference.

Isolation of immune cells by FACS

CD4⁺CD62L⁺ T cells and plasmacytoid dendritic cells were isolated from the prepared airway mucosa mononuclear cells or spleen by magnetic cell sorting with commercial reagent kits following the manufacturer’s instructions. To purify cells from airway tissue‐isolated single cells, EpCAM was used as the epithelial cell marker; CD19 was used as the B cell marker; CD11c and Siglec F were used as eosinophil markers; CD117 and FcεRI were used as mast cell markers; single cells were labeled with these antibodies (labeled with fluorescence) and sorted by FACS. Cell purity was greater than 95% as assessed by FACS.

RT–qPCR

The total RNAs were extracted from cells harvested from relevant experiments and converted to cDNA with a reverse transcription kit following the manufacturer’s instructions. The samples were amplified in a qPCR device with the SYBR Green master mix and the presence of primers of IL‐10 (ATAACTGCACCCACTTCCCA and GGGCATCACTTCTACCAGGT) or CD226 (CCACTTAAAGAGCCCAGGGA and GTGTTCAGGCCAAAAGAGCA). The results are calculated with 2^−∆∆Ct method and presented as relative changes.

Protein extraction

Cells were collected from relevant experiments, and lysed by incubating in a lysis buffer for 30 min. Lysates were centrifuged at 13 000 g for 10 min. Superantigen was harvested and used as the cytosolic protein extracts. The pellets were resuspended in a nuclear lysis buffer and incubated for 30 min. Lysates were centrifuged for 10 min at 13 000 g . Superantigen was harvested and used as the nuclear protein extract. All the procedures were performed at 4°C. Protein extracts were stored at −80°C until use.

Western blotting

The total proteins were extracted from cells collected from relevant experiments, fractioned by sodium dodecyl sulphate–polymerase gel electrophoresis (SDS‐PAGE) and transferred onto a polyvinylidene fluoride (PVDF) membrane. After blocking with 5% skimmed milk for 30 min, the membrane was incubated with the primary antibodies of interest (antibody types are detailed in the figures) overnight at 4°C, washed with Tris‐buffered saline containing 0.1% triton X‐100 (TBST) three times, incubated with peroxidase‐labeled secondary antibodies for 2 h at room temperature and washed with TBST three times. The immunoblots on membranes were developed by enhanced chemiluminescence and photographed in an imaging device.

Immunoprecipitation (IP)

Existing immunocomplexes in protein extracts were precleared by incubating with protein G sepharose beads for 2 h. Beads were removed by centrifugation at 5000 g for 10 min. Superantigen was collected and incubated with antibodies of interest (or isotype IgG used as a negative control) overnight to form immune complexes. Samples were incubated with protein G sepharose beads for 2 h to adsorb the immune complexes. Beads were collected by centrifugation for 10 min at 5000 g . Proteins on the beads were eluted with an eluting buffer and analyzed by Western blotting. All the procedures were performed at 4°C.

Chromatin immunoprecipitation (ChIP)

Cells were fixed with 1% formalin for 15 min to cross‐link the chromatin and surrounding proteins, lysed with RIPA lysis buffer and sonicated to shear the DNA into small pieces. Samples were then processed with the IP procedures (see above). After eluting, DNA in the samples were recovered using a DNA extracting kit following the manufacturer’s instruction, and analyzed by qPCR in the presence of Il10 promoter primers (CTGTGCCAACGAAGATCCTC and AACATTCGCCTAGAGTCCCC). Results were normalized to relative changes against the input. All the procedures were performed at 4°C.

Enzyme‐linked immunosorbent assay (ELISA)

The specific IgE, mouse mast cell protease‐1 and Th2 cytokines in the serum or nasal mucosa extracts were measured by ELISA with commercial reagent kits following the manufacturer’s instructions.

Depletion of TCR in CD4⁺ T cells

CD3⁺CD4⁺ T cells were isolated from the DO11.10 mouse spleen by FACS. The cells were treated with TCR‐α CRISPR reagent kits following the manufacturer’s instruction. The cells were analyzed for the effects of TCR depletion by Western blotting and flow cytometry.

Statistics

The difference between two groups was determined using Student’s t‐test. Analysis of variance (ANOVA) followed by Dunnett’s test was carried out for multiple comparisons; p < 0.05 was considered statistically significant.

RESULTS

OFexo induces antigen‐specific Tr1 cells in airway tissues

Naive DO11.10 mice were treated with nasal instillation (containing OFexo or control EVs) daily for 7 consecutive days. Mice were killed on day 8. The airway tissues, including nasal mucosa and lungs, were excised upon death. AMCs (airway mononuclear cells) were isolated from the airway tissues and analyzed by FACS. The results showed that administration of OFexo markedly increased IL‐10⁺ CD25⁺ T cells (Tr1 cells) in the airway tissues, while administration of Oexo (containing OVA alone) or Eexo (neither FLLL31 nor OVA) or Fexo (containing FLLL31 alone) did not show such an effect (Figure 1a–d). We found that the IL‐10⁺ CD25⁺ Tr1 cells [23] were also CD49b⁺ lymphocyte‐activation gene 3 (LAG‐3⁺) (Figure 1e,f), consistent with published data [24]. To test if the OFexo‐induced Tr1 cells were antigen‐specific, CD49b⁺ LAG‐3⁺ Tr1 cells were purified from AMCs and cultured in the presence of OVA (BSA was used as an irrelevant control antigen) overnight. As expected, a significant increase in CD154 (a T cell activation marker [25]) expression was detected in OFexo‐induced Tr1 cells (Figure 1g,h). The OVA‐specific IgE in the serum was under detectable levels (data not shown). The results demonstrate that OFexo can induce OVA‐specific Tr1 cells in the airway tissues. The results were reproduced in in‐vitro experiments, in which exposure to OFexo, but not Oexo or Fexo, in the culture induced antigen‐specific Tr1 cells (Supporting information, Figure S2). After exposure to specific antigens in the culture, the expression of CD154 and CD69 was also increased in Tr1 cells (Supporting information, Figure S3). Conversely, as shown by FACS data, administration of either OFexo or Oexo and Eexo or Fexo did not increase IL‐10 expression in airway epithelial cells, B cells, eosinophils and mast cells (Supporting information, Figure S4).

FIGURE 1.

Administration of OFexo induces antigen‐specific type 1 regulatory T cells (T_reg) (Tr1) cells in the mouse airway tissues. DO11.10 mice were treated with nasal instillation containing OFexo [or control extracellular vesicles (EVs)] daily for 7 consecutive days. The airway tissues (including nasal mucosa and lungs) were collected from mice. Airway mononuclear cells (AMCs) were isolated from the tissues, cultured in the presence of phorbol myristate acetate (PMA) (50 ng/ml) overnight with addition of brefeldin A (10 µg/ml) in the last 4 h, and analyzed by fluorescence activated cell sorter (FACS). (a,b) CD3⁺ CD4⁺ T cells were gated first. (c) Gated dot‐plots show representative CD25⁺ interleukin (IL)‐10⁺ Tr1 cell populations in AMCs. (d) Box‐plots show summarized Tr1 cell counts. (e) Gated plots show the frequency of CD49b⁺ antigen‐3⁺ cells in the CD25⁺ IL‐10⁺ Tr1 cells, as shown in (c). (f) Box‐plots show summarized CD49b⁺ antigen‐3⁺ cell counts. (g) CD49b⁺ antigen‐3⁺ cells were isolated from OFexo‐treated mice and cultured in the conditions denoted above each subpanel overnight. Cells were analyzed by flow cytometry. Gated histograms show CD154⁺ Tr1 cell frequency. (h) Box‐plots show summarized CD154⁺ Tr1 cell counts. OFexo: EVEVs containing ovalbumin (OVA) and tetramethylcurcumin (FLLL31). Fexo: EVEVs containing FLLL31. Oexo: EVs containing OVA. Eexo: empty EVs. Each group consists of six mice. Samples from individual mice were processed separately. The box‐plots data are presented as median [interquartile range (IQR)]; *p < 0.01 [analysis of variance (ANOVA) followed by Dunnett’s test], compared with (a). Each experiment was repeated three times

OFexo recognizes antigen‐specific TCR on CD4⁺ T cells

The data reported above suggest that OFexo is capable of recognizing antigen‐specific TCR on CD4⁺ T cells. To test this, proteins were extracted from OFexo and analyzed by IP. The MHC‐II/OVA complexes were detected in the protein extracts (Figure 2a). As OFexo was generated from BMDCs cultured in the presence of OVA and FLLL31 overnight, the data imply that BMDCs capture OVA and presumably process it; the latter formed complexes with MHC‐II, and was carried by BMDC‐derived EVs. Thus, we inferred that OFexo could bind to the antigen‐specific TCR of CD4⁺ T cells. To this end, CD4⁺ T cells were isolated from the DO11.10 mouse spleen by FACS; the cells were cultured in the presence of FITC‐labeled OFexo for 3 h and analyzed by FACS. The results showed that OFexo bound to the surface of DO11.10 CD4⁺ T cells, but not in BALB/c CD4⁺ T cells. Depletion of TCR (Supporting information, Figure S5) abolished the binding (Figure 2b,c).

FIGURE 2.

OFexo binds antigen‐specific T cell receptor (TCR) on antigen‐specific CD4⁺ T cells. (a) Protein extracts were prepared with OFexo and analyzed by immunoprecipitation assay. The immunoblots show an ovalbumin/major histocompatibility complex‐II (OVA/MHC‐II). (b,c) CD4⁺ T cells were isolated from the DO11.10 mouse spleen and cultured in the presence of fluorescein isothiocyanate (FITC)‐labeled extracellular vesicles (EVs) for 30 min. The cells were analyzed by flow cytometry. (b) Box‐plots show that OFexo binds to the surface of OVA‐specific CD4⁺ T cells. (c) Representative flow cytometry plots show FITC‐labeled OFexo bound on the surface of OVA‐specific CD4⁺ T cells (isolated from DO11.10 mice). OFexo: EVs containing OVA and tetramethylcurcumin (FLLL31). Fexo: EVs contain FLLL31, no OVA. TCRd: DO11.10 CD4⁺ T cells with TCR depletion by clustered regularly interspaced short palindromic repeats (CRISPR). Box‐plots data are presented as median [interquartile range (IQR)]. Each dot in bars presents data obtained from one sample; *p < 0.01 [analysis of variance (ANOVA) followed by Dunnett’s test], compared with (a). The data represent six independent experiments

OFexo enhances Il10 gene transcription in OVA‐specific CD4⁺ T cells

Next, we assessed the effects of OFexo on regulating the Il10 gene transcription in CD4⁺ T cells. Naive CD4⁺ T cells were isolated from the DO11.10 mouse spleen and exposed to OFexo in the culture for 48 h. We observed that exposure to OFexo, but not Oexo or Fexo, increased the expression of c‐Maf, the Il10 gene transcription factor, in CD4⁺ T cells (Figure 3a). We further found that OFexo increased the levels of c‐Maf in the Il10 promoter (Figure 3b) and promoted the Il10 gene transcription activity (Figure 3c). We also found that OFexo did not alter the Il10 gene transcription activities in CD4⁺ T cells isolated from BALB/c mice. To gain further insight into the mechanism by which OFexo increases IL‐10 expression in CD4⁺ T cells, cells treated with the procedures of Figure 3a–c were also analyzed by Western blotting. We found that exposure to OFexo, but not Oexo or Fexo, markedly promoted c‐Maf phosphorylation. The results demonstrate that OFexo can up regulate the Il10 gene transcription in OVA‐specific CD4⁺ T cells in an antigen‐specific manner. Additionally, we observed that the expression of CD226, a molecule that can be up‐regulated upon T cell activation [26], was markedly up‐regulated by exposure to OFexo (Figure 3e).

FIGURE 3.

OFexo increases Il10 gene transcription in OVA‐specific CD4⁺ T cells. CD4⁺ T cells were isolated from the DO11.10 or BALB/c ($) mouse spleen by fluorescence activated cell sorter (FACS). The cells were cultured in the presence of OFexo, Oexo or Fexo, respectively, at 400 ng/ml [without the presence of phorbol myristate acetate (PMA)]. Two days later, the cells were harvested and analyzed by real‐time–quantitative polymerase chain reaction (RT–qPCR) and chromatin immunoprecipitation (IP). (a) Interleukin (IL)‐10 mRNA expression in CD4⁺ T cells. (b) c‐Maf levels in the Il10 promoter of CD4⁺ T cells. (c) Pol II levels in the Il10 promoter of CD4⁺ T cells. (d) Immunoblots show c‐Maf phosphorylation (p‐c‐Maf). (e) CD226 mRNA levels. Box‐plots data are presented as median [interquartile range (IQR)]. Each dot in bars presents data obtained from one sample; *p < 0.01 [analysis of variance (ANOVA) followed by Dunnett’s test], compared with group saline. The data represent six independent experiments

OFexo‐induced Tr1 cells have immune‐suppressive functions

To test the immune‐suppressive functions of OFexo‐induced Tr1 cells, we generated Tr1 cells by OFexo in vitro. Effector CD4⁺CD25⁻ T_effs were isolated from the naive DO11.10 mouse spleen. Tr1 cells were co‐cultured with T_effs at a ratio of 1:5 in the presence of OVA (the specific antigen) and DCs for 3 days. The cells were analyzed by FACS. The results showed that the Tr1 cells efficiently suppressed T_eff proliferation (Figure 4a,b), which was mimicked by using culture superantigen of Tr1 cell culture superantigen that could be blocked by the presence of an antigen‐IL‐10 antibody (Figure 4c,d). The results demonstrate that OFexo‐induced Tr1 cells have immune‐suppressive functions.

FIGURE 4.

OFexo‐induced antigen‐specific type 1 regulatory T cells (T_reg) (Tr1) cells have immune suppressive functions. OFexo‐induced Tr1 cells were prepared and cultured with effector CD4⁺CD25⁻ T cells (T_effs), labeled with carboxyfluorescein succinimidyl ester (CFSE, isolated from DO11.10 mice) were co‐cultured at a ratio of 1 × 10⁵:5 × 10⁵ cells/ml in the presence of ovalbumin (OVA) (5 µg/ml) and dendritic cells (DC) for 3 days. (a) Gated histograms show proliferating T_effs. (b) The bars show summarized proliferating T_effs. nTc: naive CD4⁺ T cells. (c,d) T_effs were prepared from BALB/c mice and treated with the procedures denoted above each subpanel (the conditioned medium is the supernatant of OVA‐specific Tr1 cells after exposure to OVA overnight). Anti‐CD3/CD28 antibody: 5 µg/ml of each. Anti‐interleukin (IL)‐10 antibody: 20 ng/ml. The gated histograms indicate proliferating T_effs (c). Box‐plots show median [interquartile range (IQR)] of proliferating T_effs; *p < 0.01 [analysis of variance (ANOVA) followed by Dunnett’s test], compared with group OVA alone (b) or group (a). The data represent six independent experiments

OFexo inhibits AR in mice

An AR mouse model was developed with OVA as the specific antigen. The mice were treated with nasal instillation containing OFexo or control EVs for 7 consecutive days. Upon challenge with the specific antigen (OVA), AR mice showed the AR response in the local tissues (sneezing; Figure 5a), an allergic response in the serum (increase in specific IgE and mMCP1; Figure 5b,c) and an increase in Th2 cytokines in the nasal mucosa (Figure 5d–f). The AR response was efficiently suppressed by the administration with OFexo, which was abolished by the presence of the IL‐10 inhibitor. We then isolated AMCs for analysis by flow cytometry. We found a significantly smaller number of Tr1 cells in AR mice than in naive control mice. Administration of OFexo, but not Oexo nor Fexo, markedly increased the Tr1 cell frequency in AR AMCs. As expected, inhibition of IL‐10 abolished the OFexo‐increased Tr1 cells in the AR airway tissues (Figure 6). Additionally, treating AR mice sensitized to mite with OFexo did not show inhibiting effects on AR response. Treating AR mice sensitized to OVA with MFexo (exos containing mite extracts and FLLL31) also did not alleviate AR symptoms. OFexo generated from T cells did not show any effects on Tr1 generation. The Tr1‐induction by OFexo could be blocked by concurrent administration of antigen MHC‐II antibodies (Figure S6). As plasmacytoid DCs have an immune regulatory function [27]; OFexo was also generated from plasmacytoid DC‐derived extracellular vesicles and used to treat AR mice. The results showed that these OFexo also had similar suppressant effects on experimental AR (Supporting information, Figure S7).

FIGURE 5.

Inhibitory effects of OFexo on experimental allergic rhinitis (AR). AR mice were treated with nasal drops containing OFexo, Oexo or Fexo at 50 µl/nostril (0.2 mg/ml) daily for 7 consecutive days. AR response was assessed. (a) Sneezing counts after antigen challenge. (b,c) Serum levels of specific immunoglobulin (Ig)E (sIgE) and mouse mast cell protease‐1 (mMCP1). (d–f) T helper type 2 (Th2) cytokines in nasal tissue protein extracts. #Mice were treated with AS101 (an IL‐10 inhibitor). Data of box‐plots are presented as median [interquartile range (IQR)]. Each dot inside bars presents data obtained from one sample; *p < 0.01 [analysis of variance (ANOVA) followed by Dunnett’s test], compared with the ‘AR mice’ group. Each group consists of six mice

FIGURE 6.

OFexo induces antigen‐specific type 1 regulatory T cells (T_reg) (Tr1) cells in the airway tissues of allergic rhinitis (AR) mice. The airway tissue samples were taken from mice described in Fig. 5. Airway mononuclear cells (AMCs) were isolated and analyzed by flow cytometry. (a) Representative flow cytometry plots show CD4⁺ T cells (the gated plots) in AMCs. (b) CD4⁺ T cell frequency in AMCs. (c) Representative flow cytometry plots show Tr1 cells (the gated plots) in AMCs. (d) Tr1 cell frequency in AMCs. Box‐plots data are presented as median [interquartile range (IQR)]. Each dot inside bars presents data obtained from one sample; *p < 0.01 [analysis of variance (ANOVA) followed by Dunnett’s test], compared with group (b). Each group consists of six mice. Samples from individual mice were processed separately

DISCUSSION

In this study, we developed extracellular vesicles carrying two types of T cell activators to be used in the treatment of experimental AR. The EVs were derived from BMDCs that carried FLLL31, an analog of curcumin, a complex of MHC‐II and OVA; the latter was a specific antigen used in the experimental AR development. We found that administration of OFexo efficiently inhibited experimental AR by increasing Tr1 development in the mouse airway tissues.

Curcumin has been employed in the regulation of immune disorders for years. Studies such as those by Kinney et al. reported that curcumin could inhibit mastocytosis in food allergy [28]. Curcumin could also inhibit experimental asthma by regulating metalloproteinase‐9 [29]. Because curcumin also suppresses skewed Th1 responses [30], it may have an immune regulatory capacity. Our data indicate that the FLLL31, an analog of curcumin, indeed has an immune regulatory function. Importantly, the constructed OFexo can specifically focus upon antigen‐specific CD4⁺ T cells to induce antigen‐specific Tr1 cells in the airway tissues. These Tr1 cells have the immune suppressive function in experimental AR.

The data show that after capture by BMDCs, FLLL31 could be released by BMDCs that was carried by extracellular vesicles. The extracellular vesicles also carried other components: the complexes of MHC‐II/antigen. The MHC‐II/antigen complexes can present antigens to specific CD4⁺ T cells that was also reported by other investigators [31]. By exploiting this feature, we found that OFexo could recognize OVA‐specific CD4⁺ T cells. It is known that, to activate CD4⁺ T cells, two activating signals are required. The MHC‐II/OVA complexes serve as one of them; the FLLL31 serves as another, as we observed that FLLL31 significantly increase the phosphorylation of c‐Maf in CD4⁺ T cells, indicating that FLLL31 can activate CD4⁺ T cells, as shown by the present data, while details of the biochemical processes by which FLLL31 modulates TCR activities on CD4⁺ T cells are to be further investigated. Others also found that FLLL31 had effects on suppressing immune disorders via suppressing the activation of macrophages and lymphocytes [19]. Its analog, curcumin, can also regulate T cell activities, as shown in an experimental autoimmune encephalomyelitis study [32]. The present data demonstrate that OFexo carries two types of T cell activators, the antigen/MHC‐II complexes and the FLLL31, which efficiently activate CD4⁺ T cells and induce antigen‐specific Tr1 cells.

The c‐Maf is a transcription factor of the Il10 gene [33]. Liu et al. found that c‐Maf was an essential transcription factor of the Il10 gene in B cells and was required in the induction of regulatory B cells [34]. Others found that c‐Maf was also required in the induction and maintenance of Tr1 cells [35]. Our results have expanded the previous findings [33, 35] by revealing that FLLL31‐OVA‐containing EVs (OFexo) could up‐regulate the phosphorylation of c‐Maf in CD4⁺ T cells, which was in synergy with the OVA/MHC‐II complex to increase the frequency of Tr1 cells in the airway mucosa.

EVs have been employed in the research of immune regulation. DC‐derived EVs facilitate cytotoxic CD8⁺ T cell differentiation to fulfill antigen tumor activities while T_reg‐derived EVs suppress cytotoxic CD8⁺ T cell function to benefit tumor growth [36]. IL‐10‐ and IL‐4‐primed DC‐derived EVs can suppress experimental delayed hypersensitivity [37]. Experimental rheumatoid arthritis also can be suppressed by DC‐derived EVs [38]. Our study showed that the FLLL31‐carrying EVs had the desired effect on inducing immune regulatory cells and suppressing experimental AR.

In summary, the present data show that OFexo can suppress experimental airway allergy mainly via inducing Tr1 cells, and also suggest that OFexo has the translational potential to be employed in the treatment of allergic disorders.

CONFLICTS OF INTEREST

We do not have any conflicts of interest concerning this work.

AUTHOR CONTRIBUTIONS

L.H.M., H.Y.H., Q.R.J., Y.N.S., G.H.W., Y.Z., L.T.Y. and T.L. performed experiments, analyzed data and reviewed the manuscript. P.C.Y., Y.F. and Z.G.L. organized the study and supervised experiments. P.C.Y. designed the project and prepared the manuscript.

Supporting information

Fig S1‐7

Mo L‐H, Han H‐Y, Jin Q‐R, Song Y‐N, Wu G‐H, Zhang Y, et al. T cell activator‐carrying extracellular vesicles induce antigen‐specific regulatory T cells. Clin Exp Immunol. 2021;206:129–140. 10.1111/cei.13655

DATA AVAILABILITY STATEMENT

All the data are included in this paper and the online supplemental materials.