Thymic Stromal Lymphopoietin-Dependent Basophils Promote Th2 Cytokine Responses following Intestinal Helminth Infection

Promotion of Th2 cytokine responses by CD4⁺ T cells, including expression of IL-4, IL-5, IL-9, and IL-13, is required for pathogen clearance and tissue repair following exposure to helminth parasites (1–5). However, Th2 cytokine responses can also promote the pathological changes associated with asthma and allergic diseases at multiple barrier surfaces (6, 7). Recent studies have identified that in addition to their well-established role as late-phase effector cells, basophils can express MHC class II, secrete IL-4, migrate into lymph nodes (LNs), and promote optimal Th2 cytokine-mediated immune responses after exposure to some, but not all, allergens or helminth parasites (8–17).

The predominately T cell-derived cytokine IL-3 is a primary factor involved in basophil maturation, activation, and trafficking (9, 18–20). However, because IL-3 is not required for basophil development (18), it was hypothesized that other factors could regulate basophil development and/or activation. Consistent with this, recent studies indicate that basophils are a heterogeneous population of cells, whose differentiation can be promoted by the epithelial cell-derived cytokine thymic stromal lymphopoietin (TSLP) cooperatively or independently of IL-3–IL-3R interactions (21). Further, TSLP-elicited basophils exhibit a distinct pattern of gene expression compared with classical IL-3–elicited basophils, respond more robustly to stimulation by the IL-1 family cytokines IL-18 and IL-33, and produce higher levels of IL-4 and IL-6 (21). The identification of functional heterogeneity within the basophil lineage may, in part, explain the differential requirements for basophils in promoting optimal Th2 cytokine responses depending on the pathogen or allergen examined. Although TSLP-dependent basophils promote Th2 cytokine-associated inflammation in a mouse model of atopic dermatitis (21), the functional potential of IL-3–dependent versus TSLP-dependent basophils in helminth-induced Th2 cytokine-mediated inflammation is unknown.

In the current study, we demonstrate that after infection with the intestinal helminth parasite Trichinella spiralis, IL-3–IL-3R–independent, TSLP-dependent basophils are rapidly recruited into multiple tissue compartments, including intestinal-draining LNs. Critically, depletion of basophils diminished the magnitude of infection-induced CD4⁺ Th2 cytokine responses, suggesting that the rapid generation of “early responder” TSLP-dependent basophil populations contributes to an environment permissive for optimal Th2 cell differentiation that is required for immunity to invading helminths.

Wild-type (WT) C57BL/6 mice and TSLPR^−/− mice were obtained from Charles River Laboratories through arrangement with Amgen. IL-3R^−/− mice (Csf2rb2^tm1Cgb Csf2rb^tm1Clsc) and BaS-TRECK mice were bred at the University of Pennsylvania and maintained in a specific pathogen-free environment. All experiments were performed with age-, gender-, and genetic background-matched mice to minimize variations in infection-induced immune responses. All experiments were performed according to guidelines from University of Pennsylvania Institutional Animal Care and Use Committee-approved protocols. Methods for maintenance, recovery, infection, and isolation of Trichinella Ag were performed as previously described (22, 23). Mice were infected with 300 Trichinella muscle larvae by oral gavage and were sacrificed at days 2, 4, 7, or 12 postinfection (p.i.) for assessment of basophil responses or were sacrificed at day 12 p.i. for analysis of peak infection-induced Th2 cytokine responses, worm burdens, or humoral responses. At necropsy, single-cell suspensions of mesenteric lymph node (mLN) were prepared by passage through a 70-μm nylon mesh filter. Splenocytes were isolated by homogenization followed by RBC lysis. Blood was collected by cardiac puncture, serum was isolated, and peritoneal exudate cells were recovered by lavage with 10 ml cold PBS. Bone marrow was isolated from femurs and single-cell suspensions made by filtration through 70-μm nylon mesh filters and RBC lysis.

Mice were treated with neutralizing mAb against mouse TSLP (obtained from Amgen) or mouse IL-3 (34D.11) by i.p. injection with 0.25 mg Ab 4 h prior to infection and every 3 d p.i. Control mice received equivalent amounts of rat Ig (control Ig). Mice were depleted of basophils by i.p. injection of 10 μg anti-FcεRI Ab (MAR-1; eBioscience) or were given control hamster Ig (eBioscience) on days 0, 1, 2, 7, 8, and 9 p.i. BaS-TRECK mice were given i.p. injections of diphtheria toxin (500 ng) on days −1, 4, and 9 after Trichinella infection.

Cell preparations were surface stained with anti-mouse fluorochrome-conjugated mAbs against CD3ε (145-2C11), CD4 (GK1.5), CD11c (N418), CD19 (1D3), CD49b (DX5), CD123 (5B11), MHC class II (AF6-120.1), FcεRIα (MAR-1), c-kit (2B8), IgE (R35-72), and TSLPR (obtained from Amgen). For staining with intracellular cytokines, cells were stimulated for 4 h with 50 ng/ml PMA, 500 ng/ml ionomycin, and 10 μg/ml brefeldin A (Sigma Aldrich), stained with cell surface Abs, fixed with paraformaldehyde, permeabilized in saponin, and then stained with fluorochrome-labeled anti–IL-4 (11B11) and anti–IL-13 (eBio13A) Abs. All Abs were from eBioscience unless specified otherwise. Cells were analyzed by flow cytometry using a FACSCanto or LSRII (BD Biosciences), and further analysis was performed using FlowJo software (Tree Star).

Single-cell suspensions of mLN from naive or infected mice were plated at 6 million cells/ml in complete medium (DMEM; Life Technologies) supplemented with 10% heat-inactivated FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM l-glutamine, and 50 μM 2-mercaptoethanol and stimulated for 48 h with 50 μg/ml Trichinella Ag. Supernatants or serum samples were assayed for IL-3, IL-4, IL-5, IL-13, and IgE using standard sandwich ELISA protocols (eBioscience).

RNA from 1-cm sections of small intestine was isolated by homogenization in TRIzol using a TissueLyzer (Qiagen) followed by phenol–chloroform extraction and isopropanol precipitation. cDNA was generated per standard protocol with Superscript reverse transcriptase (Invitrogen) and used as input for real-time PCR. Real-time data were analyzed using the ΔΔCT method whereby actin served as the endogenous gene. All reactions were run on an ABI 7500 Fast Real-Time PCR System (Applied Biosystems). Samples are normalized to naive controls.

Groups of animals were compared using Mann–Whitney U tests or Student t tests where applicable, and p values ≤0.05 were considered significant.

Basophils have been implicated as important regulators and effectors of Th2 cytokine responses, and the recent identification of heterogeneity in the basophil lineage has provoked renewed interest in the factors that regulate basophil development, activation, and function. In the current study, we used infection with the helminth parasite Trichinella spiralis to investigate regulation of basophil responses. Trichinella is a self-limiting natural intestinal nematode parasite of mice that undergoes a transient intestinal phase where the parasite occupies a partially intracellular niche within intestinal epithelial cells (24). Trichinella infection elicits peripheral basophilia (25) and CD4⁺ Th2 cytokine responses coincident with expulsion of the intestinal stage of the parasite (26); however, the temporal and spatial kinetics of Trichinella-induced basophilia have not been reported. To address this, basophil responses in the blood, spleen, and mLN of WT mice were examined in the first 12 d p.i. Trichinella infection resulted in significant increases in frequencies of basophils (identified as CD3⁻, CD4⁻, CD19⁻, c-kit⁻, CD49b⁺, IgE⁺ cells) in the blood and spleen between day 2 and day 12 p.i. compared with naive mice (Fig. 1A). Most strikingly, whereas basophils were found in very low frequencies within mLN of naive mice, Trichinella infection resulted in a 10-fold increase in frequencies of basophils in the mLN by day 4 p.i., correlating with significant increases in total numbers of basophils in the mLN, which persisted until day 12 p.i. (Fig. 1A, 1B). Phenotypic analysis of basophils in the mLN demonstrated expression of subunits of the receptors for both TSLP and IL-3 (TSLPR and CD123) (Fig. 1C). Together, these data demonstrate that Trichinella infection elicits the rapid population expansion of IL-3 and/or TSLP-responsive basophil populations.

Given that both IL-3 and TSLP can regulate basophil responses (18, 21), we sought to test the relative contributions of each of these cytokines to Trichinella-induced basophilia and CD4⁺ Th2 cytokine responses. First, WT mice were infected with Trichinella, and the expression of TSLP mRNA and IL-3 were assessed in the small intestine or serum at days 2, 4, 7, and 12 p.i. Trichinella infection elicited a 4-fold increase in TSLP mRNA expression by day 2 p.i., and TSLP expression returned to basal levels by day 7 p.i. (Fig. 2A). In contrast, serum IL-3 levels were not elevated above those seen in naive animals until day 12 p.i. (Fig. 2B). Collectively, these data demonstrate that Trichinella infection induces rapid and transient TSLP expression but more delayed increases in IL-3 levels. Next, WT mice were infected with Trichinella and received either isotype control or neutralizing anti–IL-3 or anti-TSLP mAbs, and basophil responses were assessed at day 4 p.i., a time point at which significant splenic and mLN basophilia were observed (Fig. 1A). Trichinella-infected mice treated with control Ig did not exhibit increases in the frequencies of basophils in the blood, consistent with data presented in Fig. 1A, and these responses were not affected by ablation of TSLP or IL-3 (Fig. 2C, 2D). However, frequencies and absolute numbers of splenic basophils were increased in Trichinella-infected mice treated with control Ig or anti–IL-3 mAb (Fig. 2E, 2F). In contrast, treatment of mice with anti-TSLP mAb completely abolished infection-induced splenic basophil population expansion (Fig. 2E, 2F). While anti–IL-3 mAb treatment did partially reduce frequencies and total numbers of basophils in the mLN (Fig. 2G, 2H), consistent with a role for IL-3 in mediating basophil homing into LNs during infection with other helminth species (9), anti-TSLP mAb treatment had a greater effect and reduced mLN basophil numbers to those seen in naive animals (Fig. 2G, 2H). Collectively, these data suggest that Trichinella-induced basophil responses are highly dependent on TSLP–TSLPR signaling.

Critically, when infection-induced CD4⁺ Th2 cytokine responses were assessed at day 12 p.i., mice treated with either control Ig or anti–IL-3 mAb displayed increases in frequencies of mLN CD4⁺ T cells that coexpress the Th2 cytokines IL-4 and IL-13, whereas mice treated with anti-TSLP mAb exhibited a significantly diminished response (Fig. 2I). Together, these data suggest that although both TSLP and IL-3 may be required for optimal basophil responses, TSLP appears to play a dominant role in regulating basophil responses and the magnitude of CD4⁺ Th2 cytokine responses after Trichinella infection.

In a mouse model of atopic dermatitis, TSLP-elicited basophil responses were independent of IL-3–IL-3R interactions (21). We sought to test whether TSLP-dependent basophil responses after Trichinella infection were dependent or independent of the IL-3–IL-3R pathway. To test this, WT or mice deficient in both β-chains of the IL-3R (Csf2rb2^−/− Csf2rb^−/−) were infected with Trichinella and basophil responses examined. Critically, both WT and Csf2rb2^−/− Csf2rb^−/− mice exhibited similar Trichinella-induced increases in frequencies of basophils in the blood (Fig. 3A, 3D), spleen (Fig. 3B), and mLN (Fig. 3C). In addition, both WT and Csf2rb2^−/− Csf2rb^−/− mice exhibited increases in total numbers of spleen and mLN basophils at day 4 p.i. (Fig. 3E, 3F), indicating that IL-3–IL-3R interactions are not required for Trichinella-induced basophil responses.

To test directly whether the IL-3R–independent basophil responses that develop after Trichinella infection were dependent on TSLP–TSLPR interactions, Trichinella-infected Csf2rb2^−/− Csf2rb^−/− mice were treated with either control Ig or neutralizing anti-TSLP mAbs. Whereas infection of Csf2rb2^−/− Csf2rb^−/− mice treated with control Ab resulted in pronounced basophil population expansion in the blood (Fig. 3G), spleen (Fig. 3H), and mLN (Fig. 3I), treatment with anti-TSLP mAb significantly diminished these responses, suggesting that TSLP directly promotes basophilia independently of IL-3–IL-3R interactions. Taken together, these data indicate that Trichinella infection is a potent stimulus for the rapid development of TSLP-dependent, IL-3–independent basophil responses.

To examine the influence of TSLP–TSLPR interactions on the induction of Trichinella-induced basophilia and Th2 cytokine responses, we used mice genetically deficient in TSLPR. WT or TSLPR^−/− mice were infected with Trichinella, and basophil responses and Th2 cytokine responses were examined at day 4 or day 12 p.i., respectively. Whereas blood basophil responses in WT and TSLPR^−/− mice were comparable (Fig. 4A), infected TSLPR^−/− mice exhibited reduced frequencies of basophils in the bone marrow (Fig. 4B), spleen (Fig. 4C), and mLN (Fig. 4E) and of total numbers of basophils in the spleen (Fig. 4D) and mLN (Fig. 4F) compared with WT mice, consistent with results observed after Ab-mediated TSLP ablation (see Fig. 2). Further, TSLPR^−/− mice exhibited significantly diminished CD4⁺ Th2 cytokine responses in the mLN compared with WT mice as measured by ex vivo intracellular cytokine staining (Fig. 4G, 4H) and production of IL-4, IL-5, and IL-13 by mLN cells after restimulation with Trichinella Ag (Fig. 4I–K). These data indicate that TSLP–TSLPR interactions are critically important for both Trichinella-induced basophil responses and Th2 cytokine responses, provoking the hypothesis that TSLP may also regulate Th2 cytokine responses by eliciting basophil populations.

Basophils can promote Th2 cytokine-mediated inflammation and act as late-stage effector cells in some models of helminth infection or allergy (10, 11, 13, 27), but not in others (8, 12, 14). Paradoxical reports on the requirement of basophils for promoting optimal Th2 cell responses may be explained by previously unrecognized functional heterogeneity between IL-3 and TSLP-elicited basophils (21). Because early Trichinella-induced basophil responses are TSLP dependent, we tested the contribution of basophils to Trichinella-induced Th2 cytokine responses. We used two common methods for depleting basophils: anti-FcεRI mAb treatment or diphtheria toxin (DT)-mediated depletion in BaS-TRECK mice. Treatment of mice with anti-FcεRI mAb depleted splenic and mLN basophils in infected mice (Fig. 5A). Analysis of Trichinella-induced CD4⁺ Th2 cytokine responses at day 12 p.i. revealed that anti-FcεRI mAb-treated mice displayed significantly reduced frequencies and total numbers of CD4⁺ T cells that coexpress IL-4 and IL-13 in the mLN compared with control Ig-treated mice (Fig. 5B, 5C). It has been reported that anti-FcεRI mAb treatment can target mast cell or dendritic cell (DC) populations in some settings (8, 27, 28). Although no alteration of peritoneal cavity mast cell or mLN FcεRI⁺ DC responses was observed after anti-FcεRI mAb treatment (Supplemental Fig. 1A, 1C, 1D), anti-FcεRI mAb-treated mice did exhibit reduced frequencies of tissue-resident mast cells in the small intestine after Trichinella infection (Supplemental Fig. 1B). Therefore, as an alternative approach to selectively deplete basophils, BaS-TRECK (BaS-DTR⁺) mice were used, which allow lineage-specific basophil deletion via expression of the human DTR under the control of the proximal 3′UTR element in the mouse il4 locus (21, 29). These studies were critical to avoid the potential off-target effects of anti-FcεRI Ab treatment on mast cell populations (Supplemental Fig. 1B). Treatment of Trichinella-infected BaS-DTR⁺ and littermate BaS-DTR⁻ mice with DT resulted in complete ablation of basophils in the spleen and mLN (Fig. 5D). Critically, basophil-depleted mice exhibited significantly reduced frequencies and total numbers of CD4⁺ T cells that coexpress IL-4 and IL-13 in the mLN (Fig. 5E, 5F), indicating impaired induction of Th2 cytokine responses. Basophil-depleted mice also exhibited significantly reduced IL-4, IL-5, and IL-13 production by mLN cells after restimulation with Trichinella Ag compared with control mice (Fig. 5G–I) and reduced serum IgE titers (Fig. 5J). Consistent with previous studies demonstrating that the intestinal phase of Trichinella infection is self-limiting and that mice naturally expel the parasite even in the absence of lymphocytes (30), basophil depletion by either anti-FcεRI Ab treatment or DT-mediated ablation did not significantly affect the rate of intestinal worm expulsion (day 12 p.i.: control Ig 20 ± 5 worms versus anti-FcεRI mAb 39 ± 13 worms) or DT-mediated basophil depletion (BaS-DTR⁻ 17 ± 5 worms versus BaS-DTR⁺ 23 ± 6 worms). Together, these data indicate that TSLP-dependent basophil responses are a key contributor to the promotion of optimal CD4⁺ Th2 cytokine responses after Trichinella infection.

The generation of CD4⁺ Th2 cytokine responses is critical for immunity to parasitic helminths and is also responsible for the chronic inflammation associated with many allergic diseases (1, 6, 7, 31); however, the early events that drive Th2 cytokine production remain incompletely understood. Recent evidence that basophils can contribute to optimal Th2 cytokine-mediated immune responses after infection with some, but not all, parasitic helminth species and allergens (8–17) and that functional heterogeneity exists in the basophil lineage (21) provoked the question as to whether differences in the phenotype of responding basophil populations may explain the paradoxical roles for these cells in regulating Th2 cytokine responses. Data from the current study demonstrate that after gastrointestinal helminth infection, TSLP rapidly elicits the population expansion of TSLP-dependent, IL-3–independent basophils. TSLP-dependent basophils are rapidly recruited into LNs, and depletion of basophils results in impaired CD4⁺ Th2 cytokine responses after infection. These data indicate that TSLP-dependent basophil populations are critical for promoting optimal CD4⁺ Th2 cytokine responses after infection with a gastrointestinal helminth.

TSLP is a primarily epithelial-derived cytokine that, along with other epithelial-derived cytokines IL-25 and IL-33 (32–35), has been implicated in regulating Th2 cytokine responses after exposure to infectious or allergic stimuli (36–38). For example, TSLP–TSLPR signaling is important for Th2 cytokine responses after infections with some helminth species (37, 39). However, the mechanisms by which TSLP promotes Th2 cytokine responses are not fully understood, and TSLP can have effects on diverse cell types, including DCs via limitation of IL-12 p40 expression and upregulation of OX40L, which creates a more Th2-permissive environment (38, 40). Data from the current study implicate TSLP–TSLPR interactions as critically important for Trichinella-induced basophil responses in the bone marrow, spleen, and intestinal-draining LNs and also the magnitude of the infection-induced Th2 cytokine response. This provokes the hypothesis that TSLP regulates Th2 cytokine responses in part by eliciting basophil populations that can be rapidly recruited into intestinal-draining LNs to influence developing Th2 cell responses. Because Trichinella infection elicits rapid TSLP expression at the site of infection, which precedes any increase in circulating IL-3 levels, this may explain how TSLP is more important in regulating the acute basophil response elicited after infection. Given that TSLP-elicited basophils are more potent IL-4 producers than classical IL-3–elicited basophils and more responsive to stimulation with the epithelial-derived cytokine IL-33 (21), these studies suggest that tissue-resident epithelial cells may be central for rapidly regulating the differentiation, mobilization, and activation of TSLP-dependent basophils.

T cell-derived IL-3 is an important factor involved in basophil maturation, activation, and trafficking (9, 18, 19, 41–43). Although basophil responses after Trichinella infection and in a model of atopic dermatitis are largely IL-3–IL-3R–independent (21), IL-3 ablation did result in reduced recruitment of basophils into intestinal-draining LNs after Trichinella infection. Collectively, these data provoke the hypothesis that TSLP and IL-3 may cooperate to regulate optimal basophil responses. However, additional studies are needed to interrogate further the contributions of TSLP and/or IL-3 to basophil responses in the context of health and disease.

In addition to TSLP and IL-3, IL-18 (44), IL-33 (45, 46), GM-CSF (45), IgE (47, 48), IgD (49), C5a (50), and immune complexes (17, 51) have also been demonstrated to regulate basophil activation, development, or homing. Consistent with this, TSLPR^−/−IL-3R^−/− mice still exhibit circulating basophil populations (21), and in the current study, TSLP ablation in IL-3R^−/− mice failed to completely inhibit basophil homing to the mLN. The mechanisms by which basophil responses can be regulated are complex and are most likely regulated by multiple factors. Further research is required to classify the molecular and cellular factors involved in regulation of distinct IL-3 or TSLP-elicited basophil populations during different modes of inflammation. The development of genetically modified mice with cell-specific deletions in either IL-3R or TSLPR will provide new tools to interrogate the relative contribution of these distinct granulocyte populations in regulation of inflammation and immunity.

In conclusion, data from the current study provoke a model of the initial cellular immune response to a helminth infection, whereby epithelial cells at the site of infection produce TSLP, which elicits an “early responder” population of basophils that are immediately mobilized prior to T cell activation. Later during infection, when the effector CD4⁺ T cell response has been established, levels of IL-3 are elevated, and a “late responder” population of classical basophils can be elicited and maintained by IL-3 (9, 18–20). The identification and functional classification of TSLP-dependent basophils in a natural helminth infection model, coupled with their role in promoting atopic dermatitis (21), suggests these cells may be a critical population in promoting Th2 cytokine-associated inflammation in a variety of inflammatory or infectious settings. As such, this cell population could represent a new target in the design of therapeutic intervention strategies to promote or limit Th2 cytokine-dependent immunity and inflammation.

We thank members of the Artis laboratory for discussions and critical reading of the manuscript. We also thank the Matthew J. Ryan Veterinary Hospital Pathology Laboratory, the Penn Microarray Facility, and the Mucosal Immunology Studies Team of the National Institute of Allergy and Infectious Diseases for expertise and resources. We would also like to thank the Abramson Cancer Center Flow Cytometry and Cell Sorting Resource Laboratory for technical advice and support.

M.R.C. is employed by Amgen and supported by Amgen Inc. The other authors have no financial conflicts of interest.