Critical roles of regulatory B and T cells in helminth parasite‐induced protection against allergic airway inflammation

Summary

The prevalence of allergic asthma and incidences of helminth infections in humans are inversely correlated. Although experimental studies have established the causal relation between parasite infection and allergic asthma, the mechanism of the parasite‐associated immunomodulation is not fully elucidated. Using a murine model of asthma and nematode parasite Heligmosomoides polygyrus, we investigated the roles of regulatory B cells (B_reg) and T cells (T_reg) in mediation of the protection against allergic asthma by parasite. H. polygyrus infection significantly suppressed ovalbumin (OVA)‐induced allergic airway inflammation (AAI) evidenced by alleviated lung histopathology and reduced numbers of bronchoalveolar inflammatory cell infiltration, and induced significant responses of interleukin (IL)‐10⁺ B_reg, IL‐10⁺ T_reg and forkhead box protein 3 (FoxP3)⁺ T_reg in mesenteric lymph node and spleen of the mice. Adoptive transfer of IL‐10⁺ B_reg and IL‐10⁺ T_reg cell prevented the lung immunopathology in AAI mice. Depletion of FoxP3⁺ T_reg cells in FoxP3‐diphtheria toxin (DT) receptor transgenic mice by diphtheria toxin (DT) treatment exacerbated airway inflammation in parasite‐free AAI mice and partially abrogated the parasite‐induced protection against AAI. IL‐10⁺ B_reg cells were able to promote IL‐10⁺ T_reg expansion and maintain FoxP3⁺ T_reg cell population. These two types of T_regs failed to induce CD19⁺ B cells to transform into IL‐10⁺ B_reg cells. These results demonstrate that B_reg, IL‐10⁺ T_reg and FoxP3⁺ T_reg cells contribute in A discrepant manner to the protection against allergic airway immunopathology by parasiteS. B_reg cell might be a key upstream regulatory cell that induces IL‐10⁺ T_reg response and supports FoxP3⁺ T_reg cell population which, in turn, mediate the parasite‐imposed immunosuppression of allergic airway inflammation. These results provide insight into the immunological relationship between parasite infection and allergic asthma.

Concurrent infection of Heligmosomoides polygyrus (Hp) parasite protected mice from OVA‐induced allergic airway inflammation (AAI). Depletion of FoxP3⁺ T_reg cells in FoxP3‐diphtheria toxin (DT) receptor transgenic mice by DT treatment exacerbated airway inflammation in parasite‐free AAI mice and partially abrogated the parasite‐induced protection against AAI.

Introduction

Asthma and allergic diseases have become increasingly prevalent during the past few decades in children in developed countries and in populations undergoing urbanization 1. Analysis of the epidemiological data reveals that the increased prevalence of asthma in developed countries is associated with the decreased frequencies of microbial infections during the same period of time 2, 3, 4, suggesting that microbial infections modulate the host immune response to protect against the immunopathologies. Helminth parasites are known to modulate the host’s immune response and induce immunosuppression 5. Human studies and meta‐analyses on epidemiological data provide evidence to support the inverse association between helminth infection and asthma and other atopic symptoms 6, 7, 8. Laboratory animal models offer useful tools for study of the immunological relationship between parasite infection and asthma. Acute and chronic phases of Trichinella spiralis infection suppress ovalbumin (OVA)‐induced allergic airway inflammation (AAI) 9. Infection of mice with intestinal nematode parasite Heligmosomoides polygyrus suppressed allergen‐induced airway eosinophilia and bronchial hyper‐reactivity 10, 11. Similar protection against allergen‐induced airway inflammation and hyper‐responsiveness is also observed in mice infected with the male Schistosoma mansoni parasite 12. These studies with animal models provide direct evidence for the causal relationship of helminth infection and allergic asthma.

Allergic asthma is a T helper type 2 (Th2)‐driving immunopathology in the respiratory system 13. The observations that infection with Th2 response‐inducing helminth prevents or alleviates the asthma manifestation suggest that immunoregulatory mechanisms, other than Th1/Th2 reciprocal suppression, might be involved in protection against asthma by parasites. Nematode parasite infections in mice induced interleukin (IL)‐10 production, and protection against OVA‐induced AAI by parasite was abrogated in IL‐10‐deficient mice 10, indicating that the protective effect is IL‐10‐dependent. Adoptive transfer of CD4⁺CD25⁺forkhead box P3 (FoxP3)⁺ regulatory T cells from parasite‐infected mice conferred protection against the OVA‐induced inflammation in the lung 9, 10, 11, 14. A study by Wilson et al. showed that CD19⁺ B cells from H. polygyrus‐infected IL‐10‐deficient C57BL/6 mice were able to suppress allergen‐induced airway eosinophilia, and the suppressive effect was not observed in the B cells from uninfected donor mice 15. These findings demonstrate that multiple factors of the host’s immune regulatory arm are involved in the modulation of allergic immunopathology by helminth parasites.

In the present study, we used the murine gastrointestinal parasite H. polygyrus and OVA‐induced AAI asthma model to investigate the immunological relationship between helminth parasite infection and allergic asthma. In agreement with the findings by others, H. polygyris infection strongly ameliorates the inflammatory pathology in AAI animals. More importantly, IL‐10⁺CD19⁺ regulatory B cells (B_reg), IL‐10⁺CD4⁺ regulatory T cells (IL‐10⁺ T_reg) and FoxP3⁺CD4⁺ regulatory T cells (FoxP3⁺T_reg) induced in the gut‐associated lymphoid tissue during parasite infection contribute, in a sequential manner, to the regulation of allergic inflammatory response of lung by intestinal parasite infection.

Materials and methods

Animals and parasites

Female C57BL/6 mice, 6–8 weeks old, were purchased from Vital River Laboratories (Beijing, China). IL‐10 green fluorescent protein (GFP) reporter mice 16 and Depletion of Regulatory T cells (DEREG) mice, that co‐express diphtheria toxin (DT) receptor and GFP protein under the control of the foxp3 gene locus 17, were obtained from Jackson Laboratory (Bar Harbor, ME, USA). H. polygyrus parasites were maintained in the laboratory via passage in BALB/c mice. For H. polygyrus infection, mice were given 300 third‐stage larvae (L3) by oral gavage. H. polygyrus antigen was prepared following the procedure described previously 18. Briefly, 3 weeks after infection with 500 H. polygyrus L3, adult worms were collected from the small intestine of mice and washed extensively with sterile phosphate‐buffered saline (PBS) containing antibiotics. The adult worms were homogenized with a glass tissue grinder, and the homogenate was centrifuged (3000 g, 10 min). The soluble fraction of the homogenate was collected, the protein concentration was determined with a protein assay kit (Bio‐Rad Laboratories, Hercules, CA, USA), and the material was stored at –20°C. Mice were housed in the specific pathogen‐free animal facility of the Guangzhou Institutes of Biomedicine and Health. All experiments were carried out in accordance with the National Animal Protection Guidelines of China and approved by the Institutional Animal Care and Use Committee.

Induction of allergic airway inflammation (AAI)

AAI was induced in mice using the OVA sensitizing/challenging protocol 19, 20, with some modifications. Briefly, groups of mice were sensitized on days 0 and 7 by intraperitoneal (i.p.) administration of 25 μg OVA (Sigma‐Aldrich, St Louis, MO, USA) with 2 mg aluminum hydroxide (Thermo Scientific, Rockford, IL, USA) in 100 μl PBS. Mice were challenged with aerosolized OVA (1% solution, 30 min) three times on days 14, 15 and 16, and euthanized on day 18. To determine the effect of parasite infection on AAI, mice were infected with 300 H. polygyrus L3s 14 days before the first OVA‐sensitization.

Bronchoalveolar lavage (BAL) cell analysis

Forty‐eight hours after final challenge, mice were anesthetized, the tracheas were cannulated and airspaces of lung were lavaged with 600 μl PBS, followed by two 600 μl washes. BAL cell suspensions were centrifuged at 300 g and pellets were collected and suspended in medium. BAL fluids from the first lavage were collected analysis of cytokines by enzyme‐linked immunosorbent assay (ELISA). Cell counts were performed on individual samples. In some experiments, lung‐draining lymph nodes 21 were collected, a single‐cell suspension was made and cell numbers were determined. To differentiate the cell types, BAL cells were stained with anti‐CD11c‐allophycocyanin (APC) (clone N418; eBioscience, San Diego, CA, USA) and anti‐Siglec‐F‐phycoerythrin (PE) (clone E50‐2440BD; BD Biosciences, San Jose, CA, USA) fluorescent monoclonal antibodies (mAbs). Alveolar macrophage was defined as CD11c⁺SiglecF⁺ and eosinophils as SiglecF⁺CD11c⁻ 22, 23. Flow cytometry analysis was performed on a fluorescence activator cell sorting (FACS)Calibur™ (BD Biosciences).

Histopathology

The mice were euthanized and lungs were removed, fixed in a 4% formaldehyde solution for 48 h and embedded in paraffin. Tissue sections (3 µm thick) were prepared and stained with hematoxylin and eosin (H&E) (Sigma‐Aldrich) following the standard procedures. Alcian blue periodic acid‐Schiff (AB‐PAS) staining was also performed on sections to identify mucus‐containing goblet cells.

Cell culture

Spleens and mesenteric lymph nodes (MLN) single‐cell suspensions were prepared in RPMI‐1640 supplemented with 25 mM HEPES, 2 mM L‐glutamine (Invitrogen, Carlsbad, CA, USA), 10% heat‐inactivated fetal bovine serum (HyClone, Logan, UT, USA), 200 µg/ml penicillin and 200 U/ml streptomycin (HyClone) (complete medium). Red blood cells (RBCs) were lysed with 0·175 M NH₄Cl. Cell viability was determined by trypan blue (Invitrogen) exclusion and was always > 95%. For cell culture, 0·5 ml cell suspensions in complete medium (1 × 10⁷ cells/ml) were cultured in 48‐well plates (Greiner Bio‐One, Frickenhausen, Germany) with OVA (20 µg/ml) or H. polygyrus antigen (20 µg/ml) and incubated for 48 h at 37°C in a humidified CO₂ incubator. Supernatants were collected and the cytokine levels were determined by ELISA. In some experiments, spleen and MLN cells were stained with fluorescent mAb and sorted for adaptive transfer.

For cell co‐culture, spleen cell suspensions were made from IL‐10‐GFP and DEREG mice. Cells were stained with anti‐CD4‐PE (clone GK1.5; eBioscience) or anti‐CD19‐APC (clone 1D3; eBioscience) mAbs and sorted on FACS Aria II U (BD Biosciences) to collect IL‐10⁺ T_reg, FoxP3⁺ T_reg or IL‐10⁺ B_reg cells. Purified target (2 × 10⁵) and effector cells (2 ×10⁵) in 0·2 ml complete medium were co‐cultured in U‐bottomed 96‐well plates for 48 h. The cultured cells were harvested, stained for surface markers and analyzed by flow cytometry.

Cytokine and antibody ELISA

Levels of cytokines in sera, BAL fluids and cell culture supernatants and total IgE in sera were determined using an ELISA kit (R&D Systems, Minneapolis, MN, USA), according to the manufacturer’s instructions. Cytokine and antibody levels in samples were calculated by linear regression against the standards.

Transfer of IL‐10⁺ B_reg and IL‐10⁺ T_reg cells

Spleens were collected from H. polygyrus‐infected or uninfected IL‐10‐GFP reporter mice and single‐cell suspension was made as described in the previous section. Total B or T cells were first enriched by negative selection using magnetic bead columns (Miltenyi Biotec, Bergisch Gladbach, Germany). The enriched B and T cells were stained with anti‐CD19‐APC or anti‐CD4‐PE antibodies, respectively, sorted and CD19⁺GFP⁺ B cells (B_reg) and CD4⁺GFP⁺ T cells (IL‐10⁺ T_reg) were collected. The cells were washed with PBS twice and aliquots of 0·1 ml cell suspension (5×10⁵) were injected to mice intravenously. To examine the in‐vivo induction of IL‐10⁺ T_regs and FoxP3⁺T_regs by B_reg, the sorted IL‐10⁺ B_reg cells and IL‐10^– B cells (3 × 10⁶ in 0·1 ml) were injected to mice i.v. The recipient mice were euthanized 48 h later, the spleen cells were stained and the frequencies of IL‐10⁺T_reg and FoxP3⁺ T_reg cells in CD4⁺ T cell population were determined by flow cytometry.

In vivo depletion of FoxP3⁺ T_reg cells in DEREG mice

To deplete FoxP3⁺T_reg cells in DEREG mice, 1 µg of DT (Merck Millipore, Darmstadt, Germany) was injected to mice i.p. four times during the experiment. Depletion efficiency was evaluated by examination of CD4⁺ FoxP3⁺ cells in spleen cells from the DT‐treated mice by flow cytometry 24 h after the final DT injection.

Statistical analysis

All experiments were performed two to three times and each experimental group consists of three to five mice, as indicated in the figure legends. Data were shown as means [± standard error of the mean (s.e.m.)]. Statistical analysis was performed using GraphPad Prism (GraphPad software Inc., San Diego, CA, USA). Statistical significance of the differences between experimental groups was determined by two‐tailed Student’s t‐test. P‐values less than 0·05 were considered statistically significant.

Results

Suppression of airway inflammation and Th2 immune response by H. polygyrus infection

To test whether the intestinal parasite infection modulates the airway inflammation induced by OVA, a group of naive mice and a group of H. polygyrus‐infected mice (day 14 of infection) were treated with OVA to induce AAI. Naive untreated and H. polygyrus‐infected mice were included as control groups. Two days after the final OVA challenge, lung histology and cellular infiltration were examined (Supporting information, Fig. S1). Examination of H&E‐stained lung sections showed that parasite infection did not result in apparent histological changes in lung compared to that of naive animals. Lung sections from AAI mice showed considerable cellular infiltrates. As observed in the previously reported studies 10, 11, 12, AAI mice pre‐infected with H. polygyrus showed significantly reduced lung histopathology compared to the AAI animals (Fig. 1a). AB‐PAS staining of lung sections revealed a severe hyperplasia of goblet cells and mucus production in bronchi of AAI mice. However, these histopathological changes were suppressed in H. polygyrus‐infected AAI mice (Fig. 1b).

Figure 1.

Suppression of allergic airway inflammation (AAI) and T helper type 2 (Th2) cytokine responses by Heligmosomoides polygyrus. Four groups of C57BL/6 mice were treated following the experimental protocol as described in Supporting information, Fig. S1 and euthanized at day 18. (a) Photomicrographs of lung sections stained with hematoxylin and eosin (H&E). (b). Photomicrographs of lung sections stained with Alcian blue periodic acid‐Schiff (AB‐PAS) showing mucus‐containing goblet cells. Scale bar = 100 µm. (c). Total number of bronchoalveolar lavage (BAL) cells. (d) Total number of eosinophils in BAL determined by percentage of CD11c^–Siglec‐F⁺ cells in total BAL cells. (e) Total number of lung‐draining lymph nodes cells. (f–h) Interleukin (IL)‐4, IL‐5 and IL‐13 levels in BAL fluid determined by enzyme‐linked immunosorbent assay (ELISA). Results presented are from one of three experiments. Data are means ± standard error of the mean (s.e.m.) of three to four mice per group. Statistical significances are indicated by *P < 0·05, **P < 0·01, ***P < 0·001.

Analysis of BAL cells also revealed substantial cellular infiltrates into lung tissue of AAI mice. In parasite‐infected AAI mice, however, the cellular influx was significantly reduced compared with that of AAI mice (P < 0·05) (Fig. 1c). Compared with the naive mice, the frequency and number of eosinophils were significantly increased in AAI mice, and H. polygyrus infection suppressed eosinophil response (Fig. 1d, Supporting information, Fig. S2). Lung‐draining lymph nodes showed increased cellularity in AAI mice, and this response was suppressed in H. polygyrus‐infected AAI mice (Fig. 1e). We also observed that H. polygyrus infection given 3 or 4 weeks before OVA sensitization was able to suppress the AAI immunopathologies (data not shown).

BAL fluids were collected and analyzed by ELISA to determine the cytokine levels. BAL fluid from H. polygyrus‐infected mice showed increased levels of IL‐4 (P < 0·01), IL‐5 and IL‐13 compared to the samples from naive mice (both P < 0·05). BAL fluids from AAI mice contained high levels of these Th2 cytokines. BAL samples from H. polygyrus‐infected AAI mice showed significantly reduced IL‐4, IL‐5 and IL‐13 levels (P < 0·01 and P < 0·001) (Fig. 1f‐h). Cytokine production in vitro by spleen cells showed different patterns from that of BAL fluids. Spleen cells from parasite‐infected mice and parasite‐infected AAI mice produced significantly higher levels of IL‐4 and IL‐5 than the cells from AAI mice (Supporting information, Fig. S3). IL‐13 was undetectable in the supernatants of spleen cell culture.

Responses of IL‐10 cytokine and IL‐10⁺ B_reg and IL‐10⁺ T_reg cells to H. polygyrus infection

The observation that Th2‐inducing H. polygyrus infection suppressed Th2‐driven AAI suggests that other immunoregulatory factors may be involved in modulation of the AAI by the parasite. We then examined IL‐10 production in sera of H. polygyrus‐infected and found that IL‐10 levels in serum increased significantly 1 and 2 weeks following infection, and declined to lower levels 3 weeks after infection (Fig. 2a). In addition, BAL fluids and cell culture supernatants of spleen cells from parasite‐infected AAI mice stimulated in vitro with OVA and parasite antigen showed higher levels of IL‐10 than that from AAI mice (P < 0·01 and 0·05) (Fig. 2b,c).

Figure 2.

Responses of interleukin (IL)‐10 cytokine in serum and IL‐10⁺ regulatory B cells (B_reg) and IL‐10⁺ regulatory T cells (T_reg) in mesenteric lymph node (MLN) and spleen following Heligmosomoides polygyrus infection. (a). Serum levels of IL‐10. (b). IL‐10 levels in bronchoalveolar lavage (BAL) fluids of the four groups of mice. (c). Levels of IL‐10 produced by spleen cells in vitro. Spleen cells from four groups of mice were cultured in medium (naive group), in the presence of H. polygyrus antigen (HpAAI) (20 μg/ml, Hp group), ovalbumin (OVA) [20 μg/ml, allergic airway inflammation (AAI) group] or Hp antigen and OVA (Hp+AAI group). Cell culture supernants were harvested and IL‐10 level was determined by enzyme‐linked immunosorbent assay (ELISA). (d). Percentage of IL‐10⁺ B_reg cells in CD19⁺ B cells in mesenteric lymph nodes (MLN). (e) Number of IL‐10⁺ B_reg cells in MLN. (f) Percentage of IL‐10⁺ B_reg cells in CD19⁺ B cells in spleen. (g). Percentage of IL‐10⁺ T_reg cells in CD4⁺ T cells in MLN. (h). Number of IL‐10⁺ T_reg cells in MLN. (i). Percentage of IL‐10⁺ T_reg cells in CD4⁺ T cells in spleen. Results from one of two experiments are shown. Data are means ± standard error of the mean (s.e.m.) of three to five mice per group. Statistical significances are indicated by *P < 0·05, **P < 0·01, ***P < 0·001. In (a,d–g), statistical significance of the differences versus day 0 are indicated.

To determine the cellular sources of IL‐10, we analyzed the frequency of CD19⁺IL‐10⁺ B_reg cells in MLN and spleen of the mice following H. polygyrus infection. The frequency of IL‐10⁺CD19⁺ B_reg cells was increased significantly in MLN of the infected mice and peak level occurred 1 week following infection (P < 0·001 versus day 0) (Fig. 2d). Parasite infection also induced increased frequency of IL‐10⁺CD4⁺T_reg cells that peaked 1 week after H. polygyrus infection (P < 0·001 versus day 0) (Fig. 2g). We also examined the absolute numbers of IL‐10⁺ B_reg and IL‐10⁺T_reg cells in MLN during the course of H. polygyrus infection and observed that peak numbers of IL‐10⁺ B_reg cells occurred 2 weeks after infection, but the number of IL‐10⁺ T_reg cells peaked 3 weeks post‐infection (P < 0·001 versus day 0) (Fig. 2e,h). The frequencies of IL‐10⁺ B_reg and IL‐10⁺ T_reg cells in spleen also increased and peak levels were detected 2 weeks after parasite infection (P < 0·001 versus day 0) (Fig. 2f,i).

Suppression of AAI by transferred IL‐10⁺ B_reg and IL‐10⁺ T_reg cells

The increase in frequencies and numbers of IL‐10⁺ B_reg and IL‐10⁺ T_reg cells in H. polygyrus‐infected mice suggests that these immunoregulatory cells may play roles in modulation of OVA‐induced AAI. To confirm the functional association between the B_reg and IL‐10⁺ T_reg cells and the protection against allergic inflammation in the lung, IL‐10⁺ B_reg and IL‐10⁺ T_reg cells, as well as their IL‐10^– counterparts, were purified from spleens of H. polygyrus‐infected IL‐10‐GFP mice and transferred to mice 1 day before OVA sensitization. In comparison with the AAI control group, mice receiving IL‐10⁺ B_reg cells showed reduced cell infiltration in the lung, significantly less numbers of total BAL cells (P < 0.05) and eosinophils (P < 0·05) and reduced serum levels of IgE (P < 0·01). The mice receiving IL‐10^– B cells showed lung cellular infiltrates, the numbers of BAL cells and eosinophil and IgE production that were comparable to those of AAI animals (Fig. 3a–c, Supporting information, Fig. S4a). Adoptive transfer of IL‐10⁺ T_reg cells also suppressed lung immunopathology, as shown by lung histology sections, BAL total cell and eosinophil numbers and serum IgE levels (all P < 0·01), but this protective effect was not seen in transfer of IL‐10^– T cells (Fig. 3d–f, Supporting information, Fig. S4b). Transfer of IL‐10⁺ B_reg and IL‐10⁺ T_reg cells from uninfected mice to AAI mice were also able to suppress the AAI immunopathology in recipient mice (Supporting information, Fig. S5a–d).

Figure 3.

Suppression of lung immunopathology by adoptively transferred interleukin (IL)‐10⁺ regulatory B cells (B_reg) and interleukin (IL)‐10⁺ regulatory T cells (T_reg) cells. (a). Photomicrographs of hematoxylin and eosin (H&E)‐stained lung sections from the three groups of allergic airway inflammation (AAI) mice receiving phosphate‐buffered saline (PBS), IL‐10^– B cells or IL‐10⁺ B_reg cells. (b,c) Total number of bronchoalveolar lavage (BAL) cells (b) and eosinophils (c). (d) Photomicrographs of H&E‐stained lung sections from the three groups of AAI mice receiving PBS, IL‐10^–CD4⁺ T cells or IL‐10⁺ T_reg cells. (e,f) Total number of BAL cells (e) and eosinophils (f). Results from one of two experiments are presented. Scale bar = 100 µm. (b,c,e,f) Means ± standard error of the mean (s.e.m.) of three to four mice per group. Statistical significance is indicated by *P < 0·05, **P < 0·01.

Interaction between IL‐10⁺ B_reg and IL‐10⁺ T_reg cells

Both IL‐10⁺ B_reg and IL‐10⁺ T_reg cells showed protective roles against OVA‐induced AAI in the adaptive transfer experiments, but the relationship between the two types of immunoregulatory cell is not known. To address this question, in‐vitro cell co‐culture experiments were performed to analyze their interaction. Effector IL‐10⁺ B_reg and IL‐10^– B cells were isolated from IL‐10‐GFP mice and co‐cultured with target CD4⁺ T cells. CD4⁺ T cells co‐cultured with IL‐10⁺ B_regs showed significantly increased frequency of IL‐10⁺ T_reg cells and the increase of IL‐10⁺ T_reg cell frequency showed a dose–response to the increasing effector : target ratios. CD4⁺ T cells co‐cultured with IL‐10^– B cells also showed an increase in frequency of IL‐10⁺ T_reg cells compared with CD4⁺ T cells cultured alone (P < 0·05), but the response was lower than the CD4⁺T cells co‐cultured with IL‐10⁺ B_regs (Fig. 4a). When this co‐culture assay was performed in Transwell culture plates, promotion of IL‐10⁺ T_reg cell proliferation by IL‐10⁺ B_reg cells was abrogated (Fig. 4b). We then determined whether IL‐10⁺ T_reg cells were able to promote IL‐10⁺ B_reg cell differentiation, and observed that B cells co‐cultured with IL‐10⁺ T_reg cells or IL‐10^‐CD4⁺ T cells showed no significant change in proportion of IL‐10⁺ B_reg cells (Fig. 4c). Further IL‐10⁺ B_reg cell transfer experiment was performed to determine whether the B_reg cells were able to promote IL‐10⁺ T_reg cell expansion in vivo. CD4⁺ T cells from spleen of mice receiving IL‐10⁺ B_reg cells showed significantly increased frequency of IL‐10⁺ T_reg cells in comparison with the mice that received no cells or IL‐10^– B cells (Fig. 4d).

Figure 4.

Analysis of interaction between interleukin (IL)‐10⁺ regulatory B cells (B_reg) cells and IL‐10⁺ regulatory T cells (T_reg) cells. (a). CD4⁺ T cells were co‐cultured with IL‐10^– B cells (E : T ratio 1 : 1) or IL‐10⁺ B_reg cells (E : T ratio 0.5 : 1, 1 : 1, 2 : 1) and percentage of IL‐10⁺CD4⁺ T_reg cells in CD4⁺ T cells were examined. (b) CD4⁺ T cells were co‐cultured with IL‐10^–B cells or IL‐10⁺ B_reg cells in Transwell cell culture plates and percentage of IL‐10⁺ T_reg cells in CD4⁺ T cells were examined. (c). CD19⁺ B cells were co‐cultured with IL‐10^–CD4⁺ T cells or IL‐10⁺CD4⁺ T_reg cells and percentage of IL‐10⁺ B_reg cells in total CD19⁺ B cells were examined. (d). IL‐10^–CD19⁺B cells and IL‐10⁺CD19⁺ B_reg cells were isolated and transferred to naive IL‐10 green fluorescent protein (GFP) mice. Frequency of IL‐10⁺ T_reg cells in total CD4⁺ T cells in spleen was determined 48 h after cell transfer. Results presented are from one of two experiments. Data shown are means ± standard error of the mean (s.e.m.) of four cell culture samples. Statistical significance is indicated by *P < 0·05, **P < 0·01.

Role of FoxP3⁺ T_reg cells in mediation of parasite‐induced suppression of AAI

FoxP3⁺ T_reg cells are known to play an important role in counteracting inflammation and maintenance of immune homeostasis, but its role in prevention from lung immunopathology is not fully understood. Analysis of the change in frequency of CD4⁺FoxP3⁺ T_reg cells revealed that the frequency of FoxP3⁺ T_reg cells in total CD4⁺ T cells were increased in MLN and spleen and peaked 2 weeks following H. polygyrus infection (P < 0·05–0·001 versus day 0) (Fig. 5a,b). In the in‐vitro cell co‐culture assay, we found that CD4⁺ T cells co‐cultured with IL‐10⁺ B_reg cells showed a higher frequency of FoxP3⁺ T_reg cells than the CD4⁺ T cells cultured alone or co‐cultured with IL‐10^– B cells (Fig. 5c). However, the FoxP3⁺ T_reg cells did not show the ability to induce IL‐10⁺ B_reg cell proliferation (Fig. 5d). Passive transfer of IL‐10⁺ B_reg cells did not increase the frequency of FoxP3⁺ T_reg cells in spleen CD4⁺ T cells in vivo (data not shown).

Figure 5.

Roles of forkhead box protein 3 (FoxP3)⁺ regulatory T cells (T_reg) cells in mediation of parasite‐associated immunosuppression of ovalbumin (OVA)‐induced lung immunopathology. (a,b) Response of FoxP3⁺ T_reg cells in mesenteric lymph nodes (MLN) (a) and spleen (b) during the course of Heligmosomoides polygyrus infection. (c) Stimulation of FoxP3⁺ T_reg cell proliferation by interleukin (IL)‐10^– B cells and IL‐10⁺ regulatory B cells (B_reg) cells in vitro. (d). Stimulation of IL‐10⁺ B_reg cell proliferation by FoxP3^–CD4⁺ T cells and FoxP3⁺CD4⁺ T_reg cells in vitro. (e) Photomicrographs of hematoxylin and eosin (H&E)‐stained lung sections of naive, allergic airway inflammation (AAI) or H. polygyrus antigen (HpAAI) DEREG mice that were either untreated or treated with diphtheria toxin (DT). Scale bar = 100 µm. (f,g) Total number of bronchoalveolar lavage (BAL) cells (f) and eosinophils (g) of naive, AAI or Hp+AAI DEREG mice that were either untreated or treated with DT. Results presented are from one of two experiments. Data are means ± standard error of the mean (s.e.m.) of four cell culture samples. Statistical significances are indicated by *P < 0·05, **P < 0·01, ***P < 0·001. (a,b) Statistical significance of the differences versus day 0 are indicated.

The in‐vivo depletion approach was then used in DEREG mice to determine the role of FoxP3⁺ T_reg cells in mediation of suppression of AAI by parasite. In the preliminary experiments, we found that DT treatment of DEREG mice depleted 60–75% FoxP3⁺ T_reg cells in spleen (Supporting information, Fig. S6a). Treatment of wild‐type naive and AAI C57BL/6 mice with DT had no effect on the BAL cell numbers (Supporting information, Fig. S6b).

Groups of DEREG mice that were either untreated (naive), sensitized and challenged with OVA for induction of AAI (AAI) or infected with H. polygyrus prior to induction of AAI (Hp+AAI) were given four treatments of DT before and during OVA sensitization (Supporting information, Fig. S6c). Three corresponding control groups were not treated with DT. As observed in wild‐type C57BL/6 mice (Fig. 1), AAI DEREG mice that were not treated with DT developed severe lung cellular infiltrates and showed increased numbers of BAL total cells and eosinophils in comparison with the naive DEREG mice (P < 0·01), and Hp+AAI DEREG mice without DT treatment showed alleviated lung histopathology and reduced numbers of total cells and eosinophils in BAL compared with the DEREG AAI mice (Fig. 5e–g). The AAI DEREG mice that were depleted of the FoxP3⁺ T_reg cells by DT treatment showed exacerbated lung histopathology and significantly increased numbers of BAL total cells and eosinophils (P < 0·05–0·01) compared with the AAI DEREG mice that were not treated with DT. The Hp+AAI DEREG mice treated with DT showed some lung cellular infiltrates and higher numbers of BAL cells compared with the Hp+AAI DEREG mice that were not treated with DT (P < 0·05) (Fig. 5f,g). These results demonstrated that absence of FoxP3⁺ T_reg cells resulted in more severe allergic inflammation in both parasite‐free and parasite‐infected AAI mice.

Discussion

Epidemiological studies have revealed an inverse association between the increased frequency of allergic asthma and reduced prevalence of parasitic infections during the past few decades in the populations of industrialized countries. Laboratory studies with animal models provide firm evidence for the causal relationship between the parasite infection and protection against allergic asthma 24, 25, 26. In this study, we used an OVA‐AAI asthma model and H. polygyrus, a nematode parasite that dwells in the host’s intestine during its entire life cycle, to investigate the immune mechanisms underlying the protection against allergic asthma by concurrent parasite infection. In agreement with findings by others 10, 11, 15, we observed that H. polygyrus infection established before induction of AAI significantly suppressed the lung immunopathology evaluated by pulmonary histopathological changes and infiltration of inflammatory cells (Fig. 1a–d). AAI mice that harbor Th2‐inducing H. polygyrus infection showed increased levels of immunosuppressive cytokine IL‐10 and reduced levels of IL‐4, IL‐5 and IL‐13 in local lung airspace fluid than the parasite‐free AAI mice (Figs. 1f–h, Fig. 2b), indicating that immunomodulatory pathways other than Th1/Th2 antagonism may operate to modulate the immunopathology in the local inflammatory site by parasite. Further experiments revealed that H. polygyrus infection induced increased level of IL‐10 in the periphery, increased frequencies of IL‐10‐producing B_reg cells, IL‐10‐producing T_reg and FoxP3⁺ T_reg cells in MLN and spleen (Figs. 2, 5a,b). Passive transfer of the IL‐10⁺ B_reg and IL‐10⁺ T_reg cells protected the AAI mice from developing severe immunopathology, mimicking the effects of live parasite infection (Fig. 3). Furthermore, depletion of FoxP3⁺Treg cells in vivo in H. polygyrus‐infected AAI mice partially abrogated the immune protection against AAI by parasite (Fig. 5). These results demonstrate that these three types of immunoregulatory cells are involved in the mediation of suppression of allergic inflammation in lung by concurrent parasite infection.

FoxP3⁺ T_reg and IL‐10⁺ T_reg cells are known to play critical roles in control of excessive immune response to maintain the immunohomeostasis and immune tolerance 27, 28. IL‐10⁺ B_reg cells are also recognized in recent years as important immunosuppressive cells and critical in suppression of immunopathology 29, 30. The roles of these immunoregulatory cells in modulation of lung allergic response by parasite have been studied, but inconsistent results were obtained. It was reported that CD4⁺ T cells or CD19⁺ B cells from H. polygyrus‐infected IL‐10‐deficient mice were able to transfer protection from the Der p 1 allergen‐induced allergic response, and depletion of CD25⁺ cells with mAb in vivo partially blocked the suppression of lung allergic response by parasite 11, 15. However, study by Kitagaki et al. showed that suppression of airway allergic response by H. polygyrus infection was absent in IL‐10‐deficient mice 10. With passive transfer of purified IL‐10⁺ B_reg or IL‐10⁺ T_reg cells and in‐vivo depletion of FoxP3⁺ T_reg approaches, we demonstrated that IL‐10⁺ B_reg cells and FoxP3⁺ T_reg cells play important roles in modulation of allergic response in lung by parasite. Our study further demonstrates for the first time, to our knowledge, that IL‐10⁺ T_reg cells are an important modulator involved in the suppression of allergic response by parasite. The observation that IL‐10⁺ B_reg and IL‐10⁺ T_reg cells from uninfected and parasite‐infected mice showed similar potency to suppress the AAI response in the recipient mice (Fig. 3, Supporting information, Fig. S5) indicated that these immunoregulatory cells exert their suppressive effect in an antigen‐unspecific manner.

Although IL‐10⁺ B_regs, IL‐10⁺ T_regs and FoxP3⁺ T_regs have been shown to mediate the immunosuppression of allergic diseases by parasites, the relationship and interaction between these immune regulatory cells are not fully known. IL‐10⁺ T_reg and FoxP⁺ T_reg are two distinct types of immune regulatory T cells. Analysis of the functional marker expression revealed that a minor portion of these two types of T_regs co‐express IL‐10 and FoxP3 (Supporting information, Fig. S7). In arthritis and tumor animal model studies, it was observed that B cells or IL‐10⁺ B_reg cells are required for maintenance of normal levels of IL‐10⁺ T_regs and FoxP3⁺ T_regs and sufficient immunosuppression 31, 32. We observed that H. polygyrus infection induced increased frequencies and numbers of IL‐10⁺ B_regs and IL‐10⁺ T_regs (Fig. 2). In‐vitro cell co‐culture assay revealed that IL‐10⁺ B_reg cells were able to promote CD4⁺ cells to differentiate into IL‐10⁺ T_reg cells, but IL‐10⁺ T_reg cells were unable to induce IL‐10⁺ B_reg cell expansion. Furthermore, transfer of IL‐10⁺ B_reg cells induced proliferation of IL‐10⁺ T_reg cells (Fig. 4). These results, taken together, suggest a possible series of events, i.e. parasite infection activates and induces IL‐10⁺ B_reg cell response that promotes IL‐10⁺ T_reg cell expansion, and the expanded IL‐10⁺ T_reg cell population exerts the immunosuppression function to prevent the excessive inflammation.

Carter et al. reported that induction of IL‐10⁺ T_reg by B_reg requires cell–cell contact 33. A study by Mann et al. demonstrated that B_reg cells regulate T_reg cells by cell interaction through a CD80/CD86‐CD28‐dependent mechanism 34. Mizoguchi et al. also observed that B cell transfer to suppress colitis requires the presence of CD40 and CD86 35. It is clear that IL‐10 is not sufficient to induce T_regs directly, but requires a cognate interaction mediated by co‐stimulatory molecules on the surface of B cells and CD4⁺ T cells. Our observation that IL‐10⁺ B_reg cells, when separated from CD4⁺ T cells in the Transwell cell culture system, were not able to induce IL‐10⁺ T_reg cell proliferation (Fig. 4b), supports the notion that co‐stimulatory molecules are involved in the cell–cell contact for induction of T_regs by IL‐10⁺ B_reg cells 33, 34, 35.

The frequency of FoxP3⁺ T_reg cells was increased in both MLN and spleen following H. polygyrus infection. To determine whether this FoxP3⁺ T_reg response requires IL‐10⁺ B_reg cells, a cell co‐culture assay was performed. The CD4⁺ T cells cultured alone for 48 h showed reduced the frequency of FoxP3⁺ T_regs (Fig. 5c) compared with the frequency of FoxP3⁺ T_regs in ex‐vivo cell samples (Fig. 5b, day  0), and the CD4⁺ T cells co‐cultured with IL‐10⁺ B_regs showed elevated levels of FoxP3⁺ T_regs. Passive transfer of IL‐10⁺ B_regs failed to induce FoxP3⁺ T_reg expansion in vivo (data not shown). These results suggest that IL‐10⁺ B_reg cells are required for maintenance, but not for induction, of the FoxP3⁺ T_reg cell population 31. Our previous study showed that H. polygyrus infection induced expansion of a subset of dendritic cells with CD11c^lowCD45RB^mid phenotypes that were able to promote proliferation of FoxP3⁺ T_reg cells 36.

Depletion of FoxP3⁺ T_reg cells in vivo in AAI DEREG mice enhanced the allergic airway inflammation, indicating that FoxP3⁺ T_reg cells are essential for control of severity of AAI in parasite‐free conditions (Fig. 5e–g). Removal of FoxP3⁺ T_regs in Hp+AAI mice also resulted in more severe allergic inflammation compared with the Hp+AAI mice that were not depleted of FoxP3⁺ T_reg cells. This increment in severity of allergic inflammation in Hp+AAI mice may result from the loss of FoxP3⁺ T_reg cells that intrinsically occur in mice to control the basic severity of AAI. However, infection of H. polygyrus induced substantial increase of FoxP3⁺ T_reg cells in MLN and spleen. The increased severity of allergic inflammation in DT‐treated Hp+AAI mice may also be attributable to the loss of parasite‐induced FoxP3⁺ T_reg cells.

Helminth parasites may modulate the response of host immune system directly or via alteration of gut microbiota. A recent study by Zaiss et al. showed that H. polygyrus infection altered the intestinal microbiota community and transfer of the modified microbiota conferred protection against allergic asthma 37. We found that mouse spleen and MLN B cells stimulated with bacterial lipopolysaccharide (LPS) and cytosine–phosphate–guanosine (CpG) in vitro showed increased frequency of IL‐10⁺ B_reg cells 38 (and unpublished data). It is possible that the nematode parasite‐modified intestinal microbiota preferentially promotes IL‐10⁺ B_reg response, a pathway that warrants further study.

In conclusion, this study demonstrates the protective effects of intestinal parasite infection on the OVA‐induced allergic airway inflammation. This immune protection is mediated by B_reg, IL‐10⁺ T_reg and FoxP3⁺ T_reg cells. Our results also demonstrate that IL‐10⁺ B_reg cells is one of the key upstream immunoregulatory cells that promote expansion of IL‐10⁺ T_reg and maintenance of FoxP3⁺ T_reg cells. These findings not only provide new insight into the complex immunological relationship between helminth parasite infection and allergic immunopathology, but also reveals the potential targets for development of immune therapy against allergic asthma.

Disclosures

The authors declare no conflicts of interest.

Author contributions

Z. S., J. L. and X. G. conceived and designed the study. X. G. and X. R. performed the experiments. Q. W., Y. L. and Z. Y. contributed to experimental design and data analysis. X. G. and Z. S. prepared the manuscript.

Supporting information

Fig. S1. Schematic of experimental design. Two groups of C57BL/6 mice were infected with 300 H. polygyrus third‐stage larvae, and one of them were left untreated (Hp group) and another group were sensitized by i.p. injection of OVA in PBS 14 days (as day 0 of the experiment) after parasite infection. After second OVA sensitization at day 7, mice were challenged with three aerosolized OVA treatment at day 14‐16 (Hp+AAI group). Mice that were untreated (Naïve group) or OVA‐sensitized and challenged as described above (AAI group) were included. All mice were sacrificed at day 18.

Fig. S2. Representative flow cytometric profiles of bronchoalveolar cells from the four groups of mice stained with fluorescent anti‐CD11c and anti‐Siglec‐F mAbs. CD11c^highSiglec‐F⁺ population is defined as alveolar macrophages and CD11c^‐Siglec‐F⁺ population is defined as eosinophils.

Fig. S3. Cytokines production in vitro by splenic cells from the four groups of mice. Spleens were collected from the four groups mice, single cell suspensions were prepared and cultured in the presence of parasite antigen (Hp group cells), OVA (AAI group cells) or parasite antigen and OVA (Hp+AAI group cells). Supernatants were collected and levels of IL‐4 (a) and IL‐5 (b) were determined by ELISA. Results shown are from one of two experiments. ***, P < 0.001.

Fig. S4. Serum levels of total IgE in mice. Groups of mice were sensitized and challenged with OVA to induce AAI and transferred of IL‐10⁺Breg cells (a) or IL‐10⁺Treg cells (b). AAI mice receiving PBS, IL‐10^‐CD19⁺ B cells (a) or IL‐10^‐CD4⁺ T cells (b) were included as control groups. Mice were sacrificed and serum levels of total IgE were quantified by ELISA. The data shown are mean ± SEM of 3‐4 mice per group. ***, P < 0.01.

Fig. S5. Adoptive transfer of IL‐10⁺Breg and IL‐10⁺Treg cells collected from normal uninfected mice. IL‐10⁺CD19⁺Breg and IL‐10^‐CD19⁺ B cells (a, b) and IL‐10⁺CD4⁺Treg and IL‐10^‐CD4⁺ cells (c, d) were isolated from spleens of normal uninfected mice by magnetic beads negative selection and cell sorting. Cells were adoptively transferred into AAI mice. Total number of BAL cells (a, c) and eosinophils (b, d) were determined in recipient mice at sacrifice. Data shown are mean ± SEM of 4‐6 mice per group. **, P < 0.01.

Fig. S6. Diphtheria toxin (DT) treatment to delete Foxp3⁺ Treg cells in vivo. (a) Efficiency of Foxp3⁺ Treg cell depletion in Dereg mice by DT treatment. (b) Effect of DT treatment on bronchoaleveolar cells of wild‐type C57BL/6 naive and AAI mice. (c) Schematic of DT treatment experiment. DT treatment were given four times at day ‐1, 3, 8 and 12 as highlighted in red.

Fig. S7. Co‐expression of IL‐10 and Foxp3 in IL‐10⁺Treg and Foxp3⁺Treg cell populations. Spleen cells were collected from IL‐10‐GFP mice and stained with anti‐CD4‐PE and intracellularly stained with anti‐Foxp3‐APC. GFP⁺ (IL‐10⁺) and Foxp3⁺ populations in gated CD4⁺ cells were analyzed by flow cytometry.