Helminth Products Protect against Autoimmunity via Innate Type 2 Cytokines IL-5 and IL-33, Which Promote Eosinophilia

This article is featured in In This Issue, p.519

The incidence of autoimmune and allergic diseases is increasing significantly in developed countries. Conversely, the prevalence of these diseases remains relatively low in rural regions of the developing world and this has been associated with a higher corresponding incidence of infections with helminth parasites (1). Protective immunity to helminths is mediated by Th2 responses, but because helminths also induce regulatory immune responses, the immune system is generally poor at clearing the parasites, usually resulting in chronic infections that remain asymptomatic in the majority of patients (1). The regulatory immune responses induced during infection with helminths not only compromise host immunity and thus clearance of the parasites; they can also suppress inflammation associated with allergic and autoimmune disorders (1).

Patients with multiple sclerosis (MS) who are infected with helminth parasites have less severe disease than uninfected patients do (2). Moreover, helminth-infected patients with MS have more regulatory T cells (Tregs) and myelin-specific T cells that secrete IL-10 and TGF-β (2). Remarkably, MS-associated disease activity increases after treatment with anti-helminthic drugs (3). Studies in mouse models have demonstrated that infection with helminth parasites protects against autoimmune diseases mediated by IL-17– or IFN-γ–secreting T cells (4–7). We have reported that live infection with the helminth parasite Fasciola hepatica attenuated the clinical severity of experimental autoimmune encephalomyelitis (EAE) in a TGF-β–dependent manner and was associated with the expansion of Tregs (7).

Administration of live helminths is under clinical evaluation as a therapy for MS and other autoimmune diseases (8, 9). Although the initial findings look promising, safety and standardization considerations may hamper advancement of this approach for the treatment of inflammatory disorders in humans. Alternatively, immunosuppressive products secreted by helminths have considerable potential as immunotherapeutics. For example, ES-62 derived from Acanthocheilonema viteae attenuated the symptoms of collagen-induced arthritis (10) and Schistosoma mansoni soluble egg Ag (SEA) protected against EAE, colitis, and diabetes (11–13). Much of the evidence to date suggests that helminths suppress autoimmunity through the induction of Tregs (1). The role of helminth-induced innate type 2 responses has received less attention.

This study was designed to evaluate the immunomodulatory properties of F. hepatica excretory-secretory products (FHES), in particular their ability to suppress immune responses that mediate autoimmune diseases. Our findings demonstrate that administration of FHES suppresses autoantigen-specific Th1 and Th17 responses and significantly delays onset and attenuates the clinical signs of EAE. The protective effect was independent of Tregs and IL-4, but was mediated by IL-5 and IL-33, which promoted expansion and activation of eosinophils. Although a member of the IL-1 family, IL-33 is a type 2 biased cytokine that induces the production of IL-5 and has previously been associated with immunity to helminth infection (14–16). Our findings demonstrate that bystander suppression of T cell–mediated autoimmune disease by helminth parasites can be mediated by innate type 2 rather than adaptive regulatory immune responses.

C57BL/6 and BALB/c were bred in-house or obtained from Harlan UK. IL-10^−/− and IL-4^−/− mice on a C57BL/6 background were obtained as breeding pairs from Jackson Laboratories and bred in-house. IL-33^−/− mice (17) (accession number CDB0631K; http://www.cdb.riken.jp/arg/mutant%20mice%20list.html) were provided by Susumu Nakae (University of Tokyo, Japan). Mice were housed under specific pathogen-free conditions and were 6–12 wk old at the initiation of each experiment. All animal experiments were conducted in accordance with the recommendations and guidelines of The Health Products Regulatory Authority, the competent authority in Ireland responsible for the implementation of Directive 2010/63/EU on the protection of animals used for scientific purposes in accordance with the requirements of the Statutory Instrument No. 543 of 2012. Animal experiments were carried out under license (B100/2412) approved by The Health Products Regulatory Authority and in accordance with protocols approved by Trinity College Dublin Animal Research Ethics Committee.

Live adult flukes were collected from infected bovine livers at a local abattoir (Kildare Chilling), washed in several changes of PBS containing 100 μg/ml penicillin and streptomycin and transported to the laboratory. Live flukes were incubated overnight (18 h) in PBS containing 100 μg/ml penicillin and streptomycin in cell culture incubator at 37°C and 5% CO₂. FHES in the supernatant fluid was harvested, cleared by centrifugation at 13,000 × g for 30 min, filtrated through a 0.22-μm filter, and stored at −80°C and used within 6 mo of preparation. Protein concentration was adjusted to 2 mg/ml. C57BL/6 or IL-33^−/− mice were injected i.p. with PBS or FHES (50 μg), daily for 5 or 6 consecutive days. On days 6 or 7, peritoneal exudate cells (PECs) were isolated from mice by peritoneal lavage. Peritoneal exudate fluid, for the quantification of IL-33 and IL-5, was collected using lavage with 250 μl of PBS. The relative expression of il33 (Mm00505403_m1) mRNA was determined using predesigned TaqMan gene expression assays (Applied Biosystems). In some experiments, mice were injected with anti–IL-5 (TRFK5 100 μg) or control Ig on days 0, 3, and 5.

C57BL/6 mice or IL-4^−/−, IL-10^−/−, or IL-33^−/− mice were immunized s.c. with 100 μg myelin oligodendrocyte glycoprotein (MOG) peptide 35-55 (Cambridge BioSciences) emulsified in CFA containing 4 mg/ml heat-killed Mycobacterium tuberculosis (Chondrex). Mice were injected (i.p.) with 250 ng pertussis toxin (Kaketsuken) on days 0 and 2. Disease severity was recorded as follows: grade 0, normal; grade 1, limp tail; grade 2, wobbly gait; grade 3, hind limb weakness; and grade 4, hind limb paralysis. Mice were injected s.c. with PBS or FHES (50 μg) every second day beginning on day −1 or at onset of first symptoms (days 5−7). For treatment with anti–TGF-β or anti–IL-5, mice were injected i.p. every second day from day −1 to day 17 with 100 μg anti–TGF-β (clone 1D11, reactive to mouse TGF-β1, TGF-β2, and TGF-β3), anti–IL-5 (TRFK5, 100 μg) or with an isotype-matched control Ab (anti–β-galactosidase, GL113). For depletion of Tregs, mice were injected i.p. with 500 μg anti-CD25 (PC61) on days −5 and −2 so as not to eliminate effector T cells during the induction of EAE. A similar protocol has been shown facilitate a dramatic reduction in the numbers of Foxp3⁺ T cells in mice with EAE for up to 20 d (18). For exogenous administration of IL-33, mice were injected i.p. with every second day starting on day −1 with 500 ng mouse recombinant mature IL-33 (IL-33_109−266; Immunotools). Brains were isolated after intracardial perfusion of mice with PBS. The relative expression of il17 (Mm00439619_m1), ifng (Mm00801778_m1), il4 (Mm00445259_m1), and il13 (Mm00134204_m1) mRNA was determined using predesigned TaqMan gene expression assays. 18s rRNA was used as an endogenous control. Samples were assayed on an Applied Biosystems 7500 Fast Real-Time PCR machine. In some experiments, mononuclear cells were isolated from the brains by Percoll density (70%/40%) centrifugation. To assess MOG-specific and FHES-specific T cell responses ex vivo, brain mononuclear cells or spleen cells (2 × 10⁶ cells/ml) were restimulated with medium alone, MOG (increasing to 100 μg/ml) or heat-inactivated FHES (HI-ES, 50 μg/ml). After 72 h, supernatants were collected, and the concentrations of IL-17, IFN-γ, and IL-4 were quantified with ELISA.

CD11b⁺SSC^highF4/80^loGR-1^lo were purified from the PEC of mice treated for 6 d with FHES or mature IL-33 (0.3 μg) i.p. every 2 d for 6 d. Anti–Siglec-F could not be used to isolate eosinophils because of its proapoptotic effect on eosinophils (19); therefore, we devised an alternative FACS-based negative selection method that eliminated lymphocytes, monocytes, macrophages, and neutrophils. First, eosinophils were gated based on forward scatter and side scatter; this excluded most other cell types including lymphocytes and monocytes and most of the macrophage populations, but not neutrophils. Selection of CD11b^intermediate cells, excluded nonmyeloid CD11b⁻ cells and CD11b^high large peritoneal macrophages and selection of F4/80⁻ cells eliminated small peritoneal macrophages. Gating on Ly6G^low cells excluded Ly6G^high neutrophils. FACS analysis with anti–Siglec-F revealed that the CD11b⁺SSC^highF4/80^loGR1^lo–sorted population was greater than 95% eosinophils.

Bone marrow–derived eosinophils were prepared by culturing, bone marrow cells with stem cell factor (100 ng/ml) and FLT3-L (100 ng/ml). On day 4, nonadherent cells were recultured in a new flask containing mouse recombinant IL-5 (10 ng/ml). On day 10, cells were counted, and the percentage of Siglec-F⁺ CD11b⁺ eosinophils was determined with flow cytometric analysis to be >93% pure. Eosinophils (0.33–0.66 × 10⁶) were transferred into mice on days 0, 6, and 11 of EAE.

Bone marrow–derived immature dendritic cells (DCs) were prepared by culturing bone marrow cells from C57BL/6 mice in J558-conditioned medium containing GM-CSF. Bone marrow–derived DCs (BMDCs) were prepared by culturing bone marrow cells from C57BL/6 mice in medium supplemented with 20 ng/ml GM-CSF derived from the supernatant of the J558 cell line. On day 3, fresh medium containing 20 ng/ml GM-CSF was added. On day 6, semiadherent cells were removed using 0.02% EDTA (Sigma-Aldrich). Cells were recultured in medium supplemented with 20 ng/ml GM-CSF. On day 8, fresh medium containing 20 ng/ml GM-CSF was added. On day 10, loosely adherent cells were collected, washed, and used for assays. To determine the effect of FHES on cytokine production, DCS were cultured with LPS in the presence or absence of increasing concentrations of FHES.

Spleen cells, BMDCs, PECs, purified CD4⁺ T cells, or brain mononuclear cells were recovered and blocked by incubating in Fcγ blocker (BD Pharmingen; 1 μg/ml) for 10 min. To discriminate live from dead cells, cells were stained with LIVE/DEAD Aqua (Life technologies) for an additional 30 min. To identify innate cells, staining was performed with fluorescent Abs directed against CD11b (clone, M1/70), Ly6G (1A8), Siglec-F (E50-2480), CD11c (N418), F4/80 (BM-8), Ly6C (AL-21), MHC class II (I/A-I/E, M5/114.15.2), CD3 (145-2C11), and CD19 (1D3). To identify T cells, cells were then stained with anti-CD3, CD4 (GK1.5), CD8 (53-6.7), and CD25 (PC61.5). For assessment of T cell responses by intracellular cytokine staining (ICS), cells were stimulated for 5 h with PMA (25 ng/ml) and ionomycin (500 ng/ml) in the presence of brefeldin A (5 μg/ml) for the final 4 h. Cells were blocked and stained with LIVE/DEAD Aqua as before. Cells were then fixed and permeabilized (Fix and Perm Cell Permeabilization Kit; Caltag Laboratories) and stained for intracellular IL-17 (TC11-18H10), IFN-γ (XMG1.2), IL-4 (BVD6-24G2), IL-5 (TRFK5), and IL-13 (eBio13A). To identify intranuclear transcription factors, T cells were surface stained and permeabilized using Foxp3 staining buffers (eBioscience) according to the manufacturer’s instructions before staining with anti-Foxp3 (FJK-16A). Cells were acquired using aCyan ADP (Dako Cytomotion), FACS Canto II (BD), or LSRFortessa (BD). Analysis was completed using FlowJo version 9.2 (Tree Star).

After fixation in 10% neutral buffered formalin, coronal sections (3 mm) of brain and spinal cord were embedded in paraffin wax, sectioned at 4 μm and stained with Gill-2 H&E to assess inflammation and Luxol fast blue to assess demyelination. Results were interpreted by a trained pathologist.

Statistical analyses were performed using GraphPad Prism statistical analysis software. Group differences were analyzed with unpaired Student t test or one-way ANOVA with multiple comparisons, followed by Bonferroni posttest comparisons, for three or more groups. Differences between groups for clinical scores in EAE were analyzed by two-way ANOVA with repeated measures and Bonferroni posttest comparisons; p values ≤0.05 were considered significant.

We examined the effect of FHES on the development of autoimmunity using the EAE model and found that treatment of mice with FHES delayed the onset and significantly attenuated the clinical severity of disease when administered before (Fig. 1A) or 5 d after induction of EAE (Fig. 1B). The effect of FHES treatment was similar when given from days −1 or 5, suggesting that it attenuates disease not by preventing induction of Th1/Th17 cells, but on their migration to the CNS or their effector function. Clear histopathologic changes were observed in untreated mice with EAE, with infiltrating cells, in the meninges, and in the perivascular spaces with multifocal extension into the white matter of the spinal cord (Fig. 1C). These inflammatory changes were associated with myelin degeneration (Fig. 1D). By contrast, no histopathologic changes were observed in the mice treated with FHES (Fig. 1C, 1D).

Brain mononuclear cells, isolated from mice treated with FHES produced significantly less IFN-γ and IL-17 than cells from PBS-treated mice did after restimulation with MOG ex vivo (Fig. 1E). There was also a significant reduction in il17 and ifng mRNA expression in the brains of mice treated with FHES (Fig. 1F). FACS analysis of cells in the brain revealed that the infiltration of T cells and neutrophils during EAE was significantly reduced in mice treated with FHES (Fig. 1G, 1H), whereas macrophages infiltration were unchanged (data not shown). Finally, FACS analysis of mononuclear cells isolated from the perfused brains 14 d after induction of EAE showed a significant reduction in infiltration of CD4⁺ T cells that produced IFN-γ, IL-17 or both cytokines in mice treated with FHES (Fig. 1I, 1J). There was a reciprocal enhancement of IL-5–secreting CD4⁺ in mice treated with FHES (Fig 1J). These data demonstrate that FHES significantly reduced the severity of EAE and that protection is associated with a reduction in infiltrating Th1 and Th17 cells and enhancement of Th2 cells into the CNS. These data suggest that treatment with FHES suppresses CNS inflammation by inhibiting the differentiation or expansion of encephalitogenic T cells.

Experimental models of autoimmunity have revealed that helminth-induced protection against disease is primarily mediated by anti-inflammatory cytokines and Tregs (1). We have shown that infection with live F. hepatica induces IL-10 and TGF-β production by DCs and Tregs and expands the number of Tregs in vivo (7). Indeed, protection induced by infection with the live parasite was found to be TGF-β dependent. It has also been reported that somatic extracts from F. hepatica enhance Treg differentiation (20). In this study, we found that FHES induced dose-dependent production of TGF-β by GM-CSF–expanded BMDCs (Fig. 2A). At high concentration, FHES enhanced LPS-induced IL-10 by GM-CSF–expanded BMDCs (Fig. 2A). The GM-CSF–expanded DCs were 85–90% CD11C⁺, and 45% were CD11c⁺MHC class II^high. This finding is consistent with the recent demonstration that GM-CSF–expanded BMDCs are a heterogeneous population of DCs and monocyte-derived macrophages (21). Indeed, we found that FHES also enhanced TGF-β production by FLT3L-expanded BMDCs and M-CSF–expanded macrophages (data not shown), suggesting that FHES induces anti-inflammatory cytokine production by macrophages and DCs. Surprisingly, although treatment of mice with FHES significantly enhanced TGF-β, it did not alter Foxp3 expression by CD4⁺ T cells in spleens or lymph nodes (Fig. 2B). Likewise, in the brains of mice with EAE, FHES did not increase in the expression Foxp3 by CD4⁺ T cells (Fig. 2C). In fact, because of the lower numbers of CD4⁺ T cells in the brains of FHES-treated mice, there was a marked decrease in the absolute number of Foxp3⁺ T cells in the brains of these animals (Fig. 2D).

The protective effect of FHES on EAE was maintained in IL-10^−/− mice (Fig. 2E). Surprisingly, EAE was not exacerbated in untreated IL-10^−/− compared with untreated wild type (WT) mice. This is at variance with a report by Bettelli et al. (22), who found more severe disease in IL-10^−/− mice. However, our findings and those of Dai et al. (23), who reported similar disease in WT and IL-10^−/− mice, may reflect the fact that the EAE scores were relatively higher in the WT mice, making it more difficult for us to see exacerbation of disease in the IL-10^−/− mice. Administration of depleting Ab to CD25, which has been shown to reduce the numbers and function of Tregs significantly in mice with EAE (18), did not reverse the protective effect of FHES (Fig. 2F). Blocking TGF-β in vivo reduced the protective effect of FHES, although not significantly, and anti–TGF-β also increased the severity of EAE in mice that were not treated with FHES (Fig. 2G). These findings demonstrate that FHES enhances induction of innate anti-inflammatory cytokines, and although TGF-β can contribute, the protective effect of FHES does not appear to be mediated by IL-10 or Tregs, suggesting that other suppressive mechanisms must be involved.

Th2 cells and IL-4 signaling are required for protective immunity against helminth parasites (24) and F. hepatica–infected hosts mount anti-parasite Th2 responses (25). In this study, we addressed the hypothesis that FHES can suppress autoimmunity through IL-4, or by switching the MOG-specific Th1/Th17 response toward a nonpathogenic Th2 response (26). We found that coinjection of OVA and FHES enhanced OVA-specific IL-4, IL-5, and IL-13 production, demonstrating that FHES promotes Th2 responses to unrelated Ags in vivo (data not shown). Therefore, we examined the possibility that treatment with FHES promotes MOG-specific Th2 responses in mice with EAE. An examination of the brains of mice with EAE revealed that treatment with FHES increased the expression of IL-4 and IL-13 by CD4⁺ T cells restimulated with PMA and ionomycin ex vivo (Fig. 3A) and increased the absolute number of Th2 cells in the brains (Fig. 3B). However, an examination of MOG-specific cells in the brain revealed that brain mononuclear cells from FHES-treated mice produced less IL-4, IL-5, and IL-13 than untreated mice did after stimulation with MOG ex vivo (Fig. 3C). In the periphery, spleen cells from mice with EAE treated with FHES produced less MOG-specific IL-4, IFN-γ, and IL-17 (Fig. 3D). Interestingly, spleen cells from FHES-treated mice cells produced more IL-4, IL-5, and IL-13 when restimulated with FHES Ags, but not with MOG (Fig. 3E). These results suggest that although FHES promotes helminth-specific Th2 responses, it does not switch the dominant MOG-specific Th1/Th17 profile toward a Th2 response in FHES-treated mice with EAE. Furthermore, we show that the protective effect of FHES against EAE was maintained in IL-4^−/− mice (Fig. 3F). These results demonstrate that although FHES promotes Th2 responses, it does not protect mice against EAE via IL-4.

Our data suggest that anti-inflammatory cytokines, Tregs or Th2 cells are not central to the attenuating effect of FHES against EAE. Therefore, we investigated the possibility that protection is mediated by innate immune cells. Infection with helminths has been associated with immunosuppressive activities on innate cells, including M2 macrophages and eosinophils (1, 27, 28).

Eosinophilia is a feature of F. hepatica infection in ruminants (29) and mice (7). In this study, we found that daily administration of FHES for 5 d significantly increased the proportion (Fig. 4A) and total number of eosinophils in the peritoneal cavity (Fig. 4B) and bone marrow (Fig. 4C). IL-5 is the major differentiating factor for eosinophils in vivo (30), and eotaxin (CCL11) is an eosinophil-recruiting chemokine. Injection of FHES significantly enhanced the concentration of IL-5 in the peritoneal cavity (Fig. 4D) and the serum concentration of eotaxin (Fig. 4E). In mice with EAE, treatment with FHES significantly enhanced the frequency of eosinophils in the blood, inguinal lymph node, spleen, and brain (Fig. 4F–H) Furthermore, FHES treatment of mice with EAE enhanced the concentration of IL-5 in the serum (Fig. 4I). Collectively, these findings demonstrate that the protective effect of FHES against EAE is associated with eosinophilia and infiltration of eosinophils into the CNS.

We have shown that treatment of mice with FHES induced production of IL-5, a cytokine known to be a major differentiating factor for eosinophils in vivo (30). Therefore, we examined the role of IL-5 in FHES-induced eosinophilia. Treatment of mice with anti–IL-5 neutralizing Ab completely abolished FHES-induced eosinophilia in the blood and reduced the proportion and absolute number (Fig. 5A–C) of eosinophils in the peritoneal cavity. Furthermore, protection against EAE induced by FHES was significantly reversed after in vivo neutralization of IL-5, whether the anti–IL-5 and FHES was administered the day before induction of EAE (Fig. 6A) or from day 5 (Fig. 6B). Treatment with anti–IL-5 reversed eosinophilia induced by FHES in mice with EAE (Fig. 6C). Assessment of CNS infiltrating cells revealed that the reduction in IFN-γ and IL-17–secreting cells after treatment with FHES was reversed by neutralization of IL-5 (Fig. 6D). These findings demonstrate that FHES promotes eosinophilia through IL-5 production and that helminth-induced IL-5 is a critical pathway in protection against EAE.

In helminth infection, IL-33 induces the production of innate IL-5 that promotes eosinophilia. IL-33 also plays a role in protective immunity to the helminth parasites (14). Helminths induce production of IL-33, IL-25, and thymic stromal lymphopoietin in vivo mainly from epithelial cells (1). In this study, we found that FHES significantly enhanced IL-33 production in vivo (Fig. 7A), but failed to induce IL-25 or thymic stromal lymphopoietin production (data not shown). Although IL-33 is primarily found preformed in epithelial cells and endothelial cells, its expression is enhanced during inflammation (31, 32). Administration of FHES to mice enhanced the expression of il33 mRNA by PEC (Fig. 7B). Furthermore, daily administration of FHES for 7 d enhanced expression of the IL-33R (ST2) by peritoneal CD4⁺ T cells (Fig. 7C). We demonstrated that FHES-induced eosinophilia was reversed in IL-33^−/− mice (Fig. 7D, 7E). These data suggest that IL-33 mediates the protective effect of FHES on EAE. Indeed, the protective effect of FHES was lost in IL-33^−/− mice (Fig. 7F).

Finally transfer of purified eosinophils from FHES-treated mice (days 0, 6, and 11) significantly reduced the severity of EAE (Fig. 7G). Furthermore, adoptive transfer of eosinophils derived from mice that were injected with IL-33 reduced the severity of EAE (Fig 7H). In contrast, transfer of eosinophils that were differentiated from bone marrow did not alter the course of disease (Fig 7H). Despite the short half-life of activated eosinophils in vivo, it is remarkable that three injections of FHES-elicited or IL-33–induced eosinophils were capable of modulating the course of EAE. To our knowledge, these findings demonstrate for the first time that helminth-induced IL-33 can protect against autoimmunity in a mouse model through the activation of eosinophils.

To confirm the protective role of helminth-induced IL-33, we examined the ability of exogenously administered IL-33 to recruit eosinophils and protect against EAE. Administration of mature IL-33_109–266 to naive mice greatly enhanced the frequency (Fig. 8A) and absolute number (Fig. 8B) of eosinophils in the peritoneal cavity. IL-33_109–266 also increased the concentration of IL-5 in serum (Fig. 8C). Injection of mice with IL-33_109–266 from day −1, and throughout the course of disease, delayed the onset and reduced the severity of EAE (Fig. 8D). Finally, the protective effect of IL-33_109–266 on EAE was reversed by neutralization of IL-5 (Fig. 8E). Taken together, these data demonstrate that administration of recombinant IL-33_109–266 can replicate the effects of FHES in protecting animals against EAE. These findings demonstrate that mature IL-33 and its downstream targets, IL-5 and eosinophils, can mediate helminth-induced modulation of autoimmunity.

This study demonstrates that helminth products can protect against an experimental autoimmune disease caused by pathogenic Th1 and Th17 cells through activation of IL-5 and IL-33. The hygiene hypothesis has been based on the premise that suppression of autoimmunity by helminth parasites is mediated by Th2 cells or Tregs and the immunosuppressive cytokines that they secrete. Unexpectedly, we found that suppression of autoreactive T cells was not mediated by Tregs or the anti-inflammatory cytokines IL-10 or IL-4, but through the induction of the type 2 cytokines IL-33 and IL-5, which promote recruitment and activation of eosinophils.

In general, helminths induce mixed type 2/regulatory immune responses and use a variety of mechanisms to suppress the anti-parasite immune response of the host, which can have a bystander effect on immune responses to self-antigens (1). The immunosuppressive effect of helminth-induced Tregs and anti-inflammatory cytokines in mediating the protective effect of parasite infections against allergy and autoimmunity has emerged as a plausible explanation for the hygiene hypothesis (33). It has been reported that SEA enhanced the expression of Foxp3 in naive NOD-specific T cells in a TGF-β–dependent manner (34). Similarly, infection with Litomosoides sigmodontis–protected mice from diabetes in a Treg- and TGF-β–dependent manner (5, 35). Furthermore, we have previously shown that infection with live F. hepatica attenuated EAE in a TGF-β–dependent manner and this was associated with an expansion of Tregs (7). The live fluke probably exerts distinct immunomodulatory effects to FHES, because of the additional effect of tegumental and somatic Ags. Total extract of F. hepatica, a soluble preparation of homogenized adult flukes that contains tegumental and somatic Ags, was shown to protect mice from collagen-induced arthritis in a TGF-β and Treg-dependent manner (20). In contrast, the current study demonstrates that ES fraction suppresses not by Tregs but by modulation of innate immune responses, specifically the induction of IL-5 and IL-33. The physical interaction of the parasite with the host during live infection, including tissue translocation, mating, and egg production, may induce immune modulation that is distinct from that induced by its secreted molecules and explain why administration of ES products differs from live infection in the ability to promote Tregs.

Helminths are a broad group of evolutionarily distinct parasites that can infect different tissues. Intestinal nematodes appear to induce characteristic type 2 and regulatory responses when compared with tissue resident trematodes (that include F. hepatica), which initially induces mixed Th1/Th2 responses and then progresses to more polarized Th2 responses (36, 37). Thus, different helminths may use distinct and overlapping mechanisms of suppression that are not limited to the induction of Tregs but also include suppressive type 2 responses (1). Th2 cells have also been implicated in helminth-induced protection against autoimmunity; Trichinella spiralis and Heligmosomoides polygyrus prevented the onset of Th1-mediated diabetes in NOD mice by inducing Th2 responses (38). Likewise, infection of mice with EAE with S. mansoni reduced MOG-specific IFN-γ while enhancing systemic IL-4 and IL-5 (39). Zheng et al. (40) showed that treatment with SEA enhanced MOG-specific IL-4 and suppressed IFN-γ in mice with EAE. Furthermore, we have previously demonstrated that F. hepatica infection induces Th2 responses which can suppress a protective Th1 response against a bacterial pathogen Bordetella pertussis (41). Although we found that FHES promoted induction of Th2 cells, the attenuation of EAE by FHES was independent of IL-4. Moreover, FHES did not modulate the MOG-specific response toward a Th2-dominated response, rather the Th2 response was specific to Ags found within the ES products themselves. Although IL-4 and the IL-4Rα have been studied extensively as mediators of helminth-induced suppression of autoimmunity, the effect of IL-5 has been largely overlooked. To our knowledge, this study provides the first evidence that IL-5 may have a key role in helminth-induced attenuation of autoimmunity. Consistent with these findings, it has been reported that administration of recombinant IL-5 ameliorated experimental autoimmune neuritis in rats (42). However, the protective effect of exogenously administered IL-5 in this model appeared to be dependent on Tregs (42), suggesting species-specific differences in the response to IL-5. In contrast to that study, endogenously induced IL-5 in our model appears to function through the action of eosinophils.

IL-5 promotes the differentiation, maturation, and survival of eosinophils (30). Mice deficient in the IL-5Rα have reduced basal levels of eosinophils and display delayed clearance of the nematode Angiostrongylus cantonensis (43). We found that FHES-mediated protection against EAE was associated with eosinophilia and eosinophil infiltration into the CNS. Furthermore, FHES-induced eosinophilia and protection against EAE was abrogated following neutralization of IL-5, suggesting that helminth-induced IL-5 mediates protection against EAE through the action of eosinophils. These findings contrast the effects of certain nematode species that suppress, rather than promote, eosinophils. Necator americanus secretes a metalloprotease that cleaves eotaxin, whereas H. polygyrus ES products inactivates IL-33, reducing IL-5–mediated mobilization of eosinophils (44, 45).

Once viewed as end-stage effector cells, eosinophils are known to have a range of immunomodulatory functions, including the production of regulatory cytokines, stored preformed in vesicles for immediate release upon activation (46). Eosinophils also act as APCs (47–50) and may help to re-educate pathogenic T cells toward a suppressive or anergic phenotype. Although the numbers of eosinophils were enhanced in the PECs and blood after treatment of mice with FHES, the number of basophils was unchanged, whereas the number of mast cells was increased in blood after injection of FHES (data not shown). Furthermore, although FHES induced IgE-independent mast cell degranulation, it did not induced cytokine secretion by mast cells and inhibiting mast cell degranulation in vivo using cromolyn sodium did not inhibit the recruitment of eosinophils in response to FHES (data not shown).

To our knowledge, this study is the first to define a role for IL-33 in helminth-induced regulation of autoimmunity. Unlike other IL-1 family members, IL-33 preferentially induces type 2 response. IL-33 is a required for the protective innate immunity to a number of helminth parasites (15, 16). There are conflicting reports regarding the role of IL-33 in the pathogenesis of autoimmune diseases. Expression of ST2, the receptor for IL-33 is enhanced in the spinal cord of mice with EAE and neutralization of IL-33 at the induction phase inhibited IFN-γ and IL-17 production and attenuated disease, whereas administration of exogenous IL-33 at the induction phase of EAE worsened disease (51). Furthermore, administration of IL-33 increases the severity of experimental arthritis in mice (52), whereas treatment with anti-ST2 attenuated the severity of collagen-induced arthritis (53). However, another report showed that the course of disease in the EAE model was not altered in ST2-deficient mice and suggested that IL-33 is redundant in the pathogenesis of EAE (17). In contrast, Jiang et al. (54) reported that exogenous IL-33, when given only after peak of disease, had a partially protective effect against EAE and was associated with type 2 cytokines and M2 macrophages. Furthermore, a recent study on autoimmune uveitis, a disease with an etiology similar to EAE, demonstrated that ST2-deficient mice develop more severe disease and that administration of exogenous IL-33 was protective and associated with enhanced type 2 responses (55). The conflicting data regarding IL-33 may be explained by the fact that full-length IL-33 is posttranslationally processed by endogenous enzymes from mast cells and neutrophils into mature products with both increased (56) and diminished (57) activity. Interestingly, another study demonstrated that full-length IL-33 exerts distinct activity to mature IL-33 that is independent of ST2 and associated with recruitment of neutrophils rather than eosinophils (58). We used the processed or mature form of murine IL-33, IL-33_109–266, which is more potent in promoting type 2 responses than full-length IL-33 (56). Our data, together with published reports, suggest that full-length and mature IL-33 may have overlapping and distinct activities and depending on the model or context may also exert reciprocal proinflammatory or anti-inflammatory effect.

Our data provide strong evidence of a protective role for IL-33 in autoimmune disease. IL-33^−/− mice had delayed onset of EAE, mature IL-33 reduced the severity of EAE, and the protective effect of FHES was lost in mice lacking IL-33. We also found that treatment of mice with IL-33 increased the concentrations of IL-5 in serum and neutralization of IL-5 reversed the protective effect of IL-33 in the EAE model. Collectively, these findings demonstrate that IL-33 induced by helminth products can protect against autoimmunity through induction of IL-5 and consequent mobilization and activation of eosinophils with regulatory activity. This study has uncovered a novel mechanism of immune suppression by helminth products that could be exploited for the treatment of autoimmune diseases in humans.

We thank Barry Moran for assistance with flow cytometry, Prof. Joseph Cassidy for interpretation of lung histology, and the veterinary staff members at Kildare Chilling Ltd for help isolating F. hepatica flukes.

The authors have no financial conflicts of interest.