Interleukin 25 regulates type 2 cytokine-dependent immunity and limits chronic inflammation in the gastrointestinal tract

To test whether IL-25 could induce a protective type 2 response and confer immunity to Trichuris in a normally susceptible strain, infected AKR mice were treated with recombinant IL-25 (rIL-25). Although control-treated mice exhibited a polarized Trichuris-specific IFN-γ response in the absence of detectable IL-4 (Fig. 1 A), administration of rIL-25 resulted in the induction of an antigen-specific IL-4 response and extinguished IFN-γ production (Fig. 1 A). In addition, rIL-25–treated AKR mice exhibited the hallmarks of a protective T_H2 cell response, including reduced levels of Trichuris-specific IgG2a (Fig. S1), an increase in the numbers of goblet cells in the cecum (Fig. 1 B) and increased expression of RELMβ (Fig. 1 C). Consistent with the development of a Trichuris-specific type 2 response, rIL-25 treatment conferred resistance to infection, as untreated AKR mice had a high parasite burden, whereas treated AKR mice achieved sterile cure (Fig. 1 D). IL-25–induced immunity was lymphocyte dependent, as infected SCID mice treated with rIL-25 did not exhibit goblet cell responses (Fig. 1 B) and were unable to eradicate infection (Fig. 1 D). Thus, exogenous IL-25 can promote protective lymphocyte-dependent type 2 responses and act as an effective therapeutic treatment to enhance type 2 immunity in the GI tract.

We next tested whether early differences in expression of IL-25 was a predictor of infection outcome in genetically resistant versus susceptible strains. Although no infection-induced increases in IL-25 mRNA were observed after infection with Trichuris (unpublished data), we found that resistant BALB/c mice expressed constitutively higher levels of IL-25 mRNA in the cecum but not the mesenteric lymph node (mLN) compared with susceptible AKR mice, whereas both strains expressed equivalent but low levels of IL-25 mRNA in the mLN (Fig. 2 A). Furthermore, IL-25 receptor (IL-25R) mRNA expression by BALB/c mice was elevated in both compartments (Fig. 2 A). To determine the cellular source of IL-25, we examined the expression of a lacZ reporter construct inserted into the IL-25 locus. This reporter system allows for the amplification of the endogenous IL-25 response by using the fluorescent β-galactosidase substrate, FDG. Upon examination of the mLN and cecal patch (CP) (the murine equivalent of the appendix) for expression of the lacZ/IL-25, we identified that populations of CD4⁺ and CD8⁺ T cells residing specifically in the CP of naive mice expressed β-galactosidase activity (Fig. 2 B). We observed no FDG staining in CD19⁺, FcεRI⁺, CCR3⁺, NK1.1⁺, CD11c⁺, or CD11b⁺ cells from the mLN or CP (unpublished data). Thus, IL-25 is constitutively expressed by a subset of T cells that are found in the CP of resistant mice.

To directly investigate the functional significance of endogenous IL-25 in the development of protective type 2 cytokine responses in the GI tract, IL-25^−/− mice were generated by homologous recombination and infected with Trichuris. IL-25^−/− mice on a C57BL/6 genetic background were born in expected Mendelian ratios, developed normally, and exhibited no defects in lymphocyte development in the thymus and periphery (unpublished data). After infection, WT mice exhibited enhanced expression of IL-4 and IL-13 mRNA in the draining mLN and elevated IL-4 mRNA levels in the cecum at day 20 after infection with a negligible increase in IFN-γ mRNA expression at both sites (Fig. 3 A). In contrast, IL-4 and IL-13 mRNA levels were significantly reduced in infected IL-25^−/− mice with a coincident elevation of IFN-γ mRNA levels (Fig. 3 A). Levels of secreted cytokine protein after antigen restimulation of mLN cells were consistent with ex vivo mRNA levels (unpublished data). The mRNA of other cytokines implicated in resistance to Trichuris, including TNF-α (21) and IL-1β (22), were not expressed at significantly different levels in the mLN between WT and IL-25^−/− mice (unpublished data). IL-25^−/− mice also exhibited lower serum levels of Trichuris-specific IgG1 (Fig. S2 A) and total IgE (Fig. S2 B), and elevated levels of Trichuris-specific IgG2a (Fig. S2 A), further reflecting a shift from a protective type 2 response to a nonprotective type 1 response in the absence of IL-25. Infection of WT mice with Trichuris resulted in increased numbers of goblet cells in the cecum (Fig. 3 B) and elevated secretion of RELMβ into the GI lumen (Fig. 3 C). In contrast, infected IL-25^−/− mice failed to exhibit any significant goblet cell hyperplasia (Fig. 3 B) or detectable secretion of RELMβ (Fig. 3 C). Interestingly, mucosal mast cell responses (numbers of mast cells in the gut and serum levels of mouse mast cell protease-1) that are also associated with type 2 responses at mucosal sites (14, 23) were unimpaired in the absence of IL-25 (Fig. S3, A and B), demonstrating that IL-25 differentially regulates specific components of type 2 cytokine-dependent inflammation in the GI tract. Nevertheless, consistent with a defective type 2 immune response, IL-25^−/− mice were unable to eradicate infection and were as susceptible as RAG-deficient mice, whereas WT mice were resistant to infection, clearing worms by day 20 after infection (Fig. 3 D). Thus, endogenous IL-25 is a critical component of type 2 cytokine-dependent immunity in the GI microenvironment.

Although the cytokines that drive T_H1 cell differentiation and type 1 inflammation are well characterized, the factors that promote the development of type 2 responses in vivo remain elusive (24). Based on higher levels of IL-25/IL-25R mRNA in genetically resistant mouse strains, coupled with the results demonstrating that exogenous IL-25 can drive expression of T_H2 cytokines, whereas type 2 immunity is severely impaired in IL-25^−/− mice, we hypothesized that this cytokine may be the critical factor required for the development of T_H2 cells and type 2 responses in vivo. To test whether IL-25 expression was essential for immunity to Trichuris, infected IL-25^−/− mice were treated with a combination of anti–IL-12 and anti–IFN-γ monoclonal antibodies. Previous studies have shown that in vivo depletion of type 1 cytokines can promote type 2 cytokine-dependent resistance to Trichuris in normally susceptible hosts (16, 19). Blockade of type 1 cytokines in infected IL-25^−/− mice was associated with elevated production of Trichuris-specific IL-4, IL-5, and IL-13 in the draining mLN (Fig. 4 A), enhanced serum IgE responses (Fig. S4 A), and a recovery of intestinal goblet cell responses (Fig. 4 B). Furthermore, anti–IL-12/anti–IFN-γ treated IL-25^−/− mice exhibited reduced Trichuris-specific IFN-γ production (Fig. 4 A) and lower serum IgG2a levels (Fig. S4 B). Although control-treated IL-25^−/− mice displayed high worm burdens, inhibition of type 1 responses resulted in successful expulsion of Trichuris in IL-25^−/− mice (Fig. 4 C). Contrary to the hypothesis that IL-25 is essential for type 2 responses, these results show that expression of IL-4, IL-5, and IL-13 was recovered in the absence of IL-25 if endogenous type 1 cytokines were blocked. Therefore, these data show that the requirement for IL-25 in the development of type 2 immunity is conditional and suggest that, in addition to promoting type 2 responses, IL-25 may function to actively inhibit expression of type 1 cytokines after helminth infection.

In support of an antiinflammatory function of IL-25, we observed a critical role for this cytokine in limiting chronic infection-induced intestinal inflammation. Genetically susceptible strains of mice, such as AKR, as well as mice deficient in IL-4Rα signaling, develop persistent chronic infections but display moderate infection-induced inflammation characterized by minimal crypt hyperplasia and no loss in gross tissue morphology (references 17, 21, 25 and unpublished data). Although we observed no differences in steady-state intestinal structure between AKR, WT, or IL-25^−/− mice (Fig. 5 A, top), chronically infected IL-25^−/− mice developed severe intestinal inflammation characterized by a transmural infiltrate of lymphocytes, crypt elongation, disrupted intestinal architecture, and a complete absence of goblet cells (Fig. 5 A, bottom). Consistent with the hypothesis that IL-25 may act to limit inflammation, the infection-induced pathology in IL-25^−/− mice was associated with heightened levels of the proinflammatory cytokines IFN-γ and IL-17 in the mLN and cecum (Fig. 5, B and C). IL-10^−/− mice also develop severe Trichuris-induced intestinal inflammation that can be fatal (20). However, the destructive inflammation observed in IL-25^−/− mice was not associated with defective expression of IL-10 in the mLN (Fig. 5 D), suggesting a novel IL-10–independent, IL-25–dependent pathway in the regulation of GI inflammation. Interestingly, mice deficient in NF-κB1 (p50) develop severe intestinal inflammation after infection with Trichuris (19). IL-25 has been shown to activate NF-κB (12), suggesting that IL-25– induced NF-κB1 activation may be a critical component in ameliorating inflammation and maintaining the integrity of the mucosal barrier.

These results demonstrate that IL-25 plays a dual role in both the development of immunity and in limiting inflammation in the GI tract. As IL-25 is produced constitutively by T cells present in the CP of naive mice, this cytokine is most likely operating early after infection by driving expression of T_H2 cytokines and simultaneously limiting the production of proinflammatory cytokines. Consistent with this, IL-25^−/− mice fail to develop a polarized type 2 immune response and are unable to clear infection unless endogenous type 1 cytokine responses are blocked. This ability to limit proinflammatory responses is critical, as chronically infected IL-25^−/− mice develop severe intestinal pathology characterized by high levels of IFN-γ and IL-17. IFN-γ and IL-17 have been implicated in the pathogenesis of inflammatory bowel disease (26, 27). However, whether the intestinal pathology observed in Trichuris-infected IL-25^−/− mice is mediated by dysregulated IL-12–dependent, IFN-γ–producing T_H1 cells or by an IL-23–dependent, IL-17–producing T cell population is still not known. It is possible that IL-25 directly inhibits the production of IFN-γ or IL-17 from T cells, or alternatively limits their expression through influencing expression of IL-12/IL-23. We observed no significant difference in the expression levels of the mRNA for the IL-12 and IL-23 subunits p19, p35, and p40 between WT and IL-25^−/− mice in the mLN after infection with Trichuris (unpublished data). Therefore, it is unclear at present how IL-25 is limiting inflammation.

Recent studies have shown that IL-17–producing CD4⁺ T cells mediate chronic inflammation in the central nervous system of animals that develop experimental autoimmune encephalitis (EAE), an animal model of multiple sclerosis that is characterized by chronic inflammatory responses and demyelination of neurons (28, 29). Supporting a role for IL-25 in limiting production of proinflammatory cytokines and destructive tissue inflammation, IL-25^−/− mice are highly susceptible to EAE, exhibiting reduced expression of T_H2 cytokines, increased numbers of IL-17^pos CD4⁺ T cells in the central nervous system and rapid onset and severity of disease (unpublished data). Together with results presented here, these studies implicate IL-25 as a global inhibitor of chronic inflammatory responses. Therefore, manipulation of IL-25/IL-25R signaling offers the potential to promote or block type 2 responses associated with infection, asthma, or allergic inflammation and may provide a novel therapeutic approach for the treatment of multiple diseases associated with chronic inflammation.

IL-25^−/− mice were generated by homologous recombination (HR). After a screen of a C57BL/6J genomic BAC library (Genome Systems) with a full-length mIL-25 cDNA probe, a clone was identified and used to generate a shotgun plasmid library (SeqWright). Using a genomic subclone (F6), a BglII site was introduced in exon1 at the ATG via site-directed mutagenesis. This clone was digested with BglII and a 5,442-bp 5′ region of homology was cloned into the unique BglII site of pbgal-Basic (CLONTECH Laboratories, Inc.). A 1,049-bp 3′ region of homology was generated by PCR using a genomic subclone (C6), incorporating unique SpeI and NotI restriction sites, and cloned into pBS-lox-Neo-lox. This plasmid was digested with SalI and the lox-Neo-lox-3′ region of homology was ligated into the unique SalI site of the 5′ arm-pbgal-Basic, yielding the final targeting vector. The NheI linearized targeting vector was electroporated into C57BL/6 ES cells and Neo-resistant clones were analyzed for HR using a PCR based screening strategy followed by Southern blot confirmation with 5′ and 3′ probes. A confirmed HR clone was electroporated with a Cre recombinase expression containing plasmid and several Neo-sensitive clones were screened using Southern blot to identify the Neo-flipped HR IL-25-LacZ ES line. This line was injected into C57BL/6 blastocysts and IL-25-LacZ-knockin (KI)^−/− progeny were obtained by intercrossing IL-25–LacZ-KI^+/− animals. A PCR-based genotyping strategy was developed to track the IL-25 WT and KI alleles as follows: IL-25 targeted allele, sense 5′-GCTGACTCTCAACATTCTACTCC TCC-3′, antisense 5′-CCTGCTGCTTCAGGTAGGGCTTTG-3′; wild-type IL-25 allele, sense 5′-CTACAGACAGGCTCCCACATGGACC-3′, antisense 5′-CCTGCTGCTTCAGGTAGGGCTTTG-3′.

C57BL/6J, AKR, SCID (CBySmn.CB17-Prkdc^scid/J), and RAG^−/− (B6.129S7-Rag1^tm1Mom/J) mice were obtained from The Jackson Laboratory. Animals were maintained in a specific-pathogen free environment and tested negative for pathogens in routine screening. All experiments were performed following the guidelines of the University of Pennsylvania Institutional Animal Care and Use Committee. Trichuris muris was maintained in genetically susceptible animals. Isolation of Trichuris excretory-secretory antigen and eggs was performed as described previously (14). Mice were infected on day 0 with 150–200 embryonated eggs, and parasite burdens were assessed on various days after infection.

Monoclonal antibodies against IL-12 (C17.8) and IFN-γ (XMG-6) were purified from ascites by ammonium sulfate precipitation and extensively dialyzed against PBS. Mice were treated i.p. with 500 μg of each antibody every 4 d between days 4 and 20 after infection. rIL-25 was synthesized and purified as described previously (11). Mice were treated i.p. with 10 μg of rIL-25 every other day between days 8 and 18 after infection.

mRNA expression levels were analyzed using real-time PCR. Total RNA from tissue was isolated, cDNA was prepared, and 10–25 ng of cDNA was used to analyze gene expression in either the SYBR green real-time PCR assay or using predesigned TaqMan probe sets for IL-25, IL-25R, IL-4, IL-13, IL-10, IL-17, IFN-γ, ubiquitin, and β-actin (Applied Biosystems). Reactions were run on the GeneAmp 5700 Sequence Detection System (Applied Biosystems), and expression was normalized to the housekeeping genes ubiquitin or β-actin.

At necropsy, the mesenteric LN was harvested, and single cell suspensions were prepared in DMEM supplemented with 10% heat-inactivated FBS, 2 mM glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 25 mM Hepes, and 5 × 10⁻⁵ M 2-ME. Cells were plated in medium alone or in the presence of Trichuris muris ES Ag (50 μg/ml). Cell-free supernatants were harvested after 48 h, and analyzed for cytokine secretion by sandwich ELISA or cytometric bead array (mouse Th1/Th2 cytokine CBA; BD Biosciences). For determination of lacZ expression, the fluorescent β-galactosidase substrate fluorescein di-β-d-galactopyranoside (FDG; Invitrogen) was added to cells isolated from the mLN or CP of naive or infected IL-25^−/− mice that express lacZ in place of IL-25. After 20 min on ice, cells were surface stained with fluorescently labeled antibodies against CD4, CD8, CCR3, CD11b, CD11c, CD19, FcεRI, and NK1.1 (eBioscience) and analyzed by flow cytometry on a FACSCalibur using CELLQuest Pro software (BD Biosciences).

Serum was analyzed by ELISA for Trichuris-specific IgG1 and IgG2a as described previously (19). Total serum IgE was analyzed with an OptEIA IgE ELISA kit following the manufacturer's recommendations (BD Biosciences).

Segments of cecum were removed and fixed in 4% paraformaldehyde. For detection of intestinal goblet cells, 5-μm paraffin-embedded sections were cut and stained with Alcian blue-periodic acid Schiff's reagent or with hematoxylin and eosin. Isolation of proteins from stool samples was performed as described previously (15). Equal amounts of protein were analyzed by SDS-PAGE and immunoblotted for RELMβ with a polyclonal rabbit α-murine RELMβ antibody. Intestinal mast cells were identified by immunohistochemistry using rat anti–mouse mast cell protease-1 (mMCP-1) antibody as described previously (30). Concentrations of mMCP-1 in serum were quantitated as described previously (30).

Results represent the mean ± SEM unless otherwise stated. Statistical significance was determined by Student's t test.

Trichuris-specific serum IgG2a response in AKR mice after infection with Trichuris with or without exogenous rIL-25 is shown in Fig. S1. Fig. S2 depicts the serum immunoglobulin responses (IgG1, IgG2a, and IgE) of wild-type and IL-25^−/− mice after infection with Trichuris. Fig. S3 shows the local and systemic mucosal mast cell responses in wild-type and IL-25^−/− mice after infection with Trichuris. Fig. S4 shows that blocking IL-12 and IFN-γ in IL-25^−/− mice after infection with Trichuris results in decreased serum IgG2a and increased IgG1 and IgE responses.

The authors thank S. Longworth and J. Northrup for assistance with lacZ/FDG staining and E.J. Pearce and P. Scott for critical reading of the manuscript.

This work was supported by National Institutes of Health grant nos. AI61570 (to D. Artis) and AI42334 (to C.A. Hunter); Molecular Biology Core, Morphology Core and Pilot Feasibility Program of the National Institute of Diabetes and Digestive and Kidney Diseases Center grant no. DK50306 (to D. Artis); the Crohn's and Colitis Foundation of America's William and Shelby Modell Family Foundation Research Award (to D. Artis); and the Irvington Institute for Immunological Research Postdoctoral Fellowship (to C. Zaph). Schering-Plough Biopharma (formerly DNAX Research Inc.) is funded by the Schering-Plough Corporation.

The authors have no conflicting financial interests.