Helminths exploit intrinsic regulatory pathways of the mammalian immune system to dampen the immune response directed against them. In this article, we show that infection with the parasitic nematode Strongyloides ratti induced upregulation of the coinhibitory receptor B and T lymphocyte attenuator (BTLA) predominantly on CD4+ T cells but also on a small fraction of innate leukocytes. Deficiency of either BTLA or its ligand herpes virus entry mediator (HVEM) resulted in reduced numbers of parasitic adults in the small intestine and reduced larval output throughout infection. Reduced parasite burden in BTLA- and HVEM-deficient mice was accompanied by accelerated degranulation of mucosal mast cells and increased Ag-specific production of the mast cell–activating cytokine IL-9. Our combined results support a model whereby BTLA on CD4+ T cells and additional innate leukocytes is triggered by HVEM and delivers negative signals into BTLA+ cells, thereby interfering with the protective immune response to this intestinal parasite.
The mammalian immune system is a complex network of activating and inhibiting regulatory circuits that allow the rapid induction of immune responses to pathogens while facilitating homeostasis in the absence of infection. B and T lymphocyte attenuator (BTLA; CD272) is an Ig domain superfamily protein that delivers inhibitory signals via an intracellular ITIM upon engagement by its ligand, herpes virus entry mediator (HVEM) (1–3). Although BTLA is constitutively expressed on B cells, it is predominantly Foxp3− effector T cells that upregulate BTLA upon activation (4). Maintenance of homeostasis by BTLA-mediated negative coregulation was shown in murine models of autoimmune hepatitis, autoimmune colitis, and graft-versus-host disease (1). The BTLA ligand HVEM is a member of the TNFR superfamily and is expressed on a broad range of hematopoietic and nonhematopoietic cells. While delivering negative signals into BTLA-expressing cells, HVEM engagement leads to costimulatory signaling into the HVEM-expressing cells via NF-κB induction (5). Moreover, HVEM can bind to three ligands in addition to BTLA: one coinhibitory receptor (CD160) and two costimulatory receptors (lymphotoxin-like, exhibits inducible expression, competes with HSV gpD for HVEM, a receptor expressed by T lymphocytes [LIGHT] and lymphotoxin α) (6). Thus, the BTLA–HVEM-mediated network of costimulation and coinhibition is a classic example of the regulatory circuits that enable the immune system to rapidly respond to infections without losing homeostatic control (1, 5, 7).
The role of BTLA–HVEM-mediated immune regulation during helminth infection has not been investigated. Helminths are large multicellular parasites that are exposed to the immune system of their host. To survive and to avoid the induction of immune pathology, helminths use existing regulatory elements of the immune system, such as regulatory surface receptors, to suppress the immune response that is directed toward them (8).
We used experimental infection of mice with the parasitic nematode Strongyloides ratti to investigate helminth-induced immune modulation. Thereby, S. ratti serves as a model for transient gastrointestinal nematode infections. Infective third-stage larvae (iL3) dwell in the environment and actively penetrate the skin of their host (9). Larvae migrate within 2 d to the head, are swallowed, and reach the small intestine. They molt via a fourth larval stage to become parasitic adults that live embedded in the intestinal mucosa. Five days postinfection (p.i.), adults start to reproduce, and eggs and hatched first-stage larvae leave the host with the feces. Immune-competent mice terminate infection within 30 d and remain semiresistant to subsequent infections (10).
To our knowledge, we report for the first time that BTLA–HVEM-mediated regulation dampens the protective immune response to S. ratti infection and promotes survival of parasitic adults in the small intestine. Mice lacking either BTLA or its ligand HVEM displayed decreased intestinal parasite burden in the context of increased IL-9 production and accelerated mast cell degranulation.
Materials and Methods
S. ratti life cycle and infections
Animal experimentation was conducted at the animal facility of the Bernhard Nocht Institute for Tropical Medicine in agreement with the German animal protection law under the supervision of a veterinarian. BTLA−/− mice (2), a kind gift of Dr. Kenneth Murphy (Howard Hughes Medical Institute, Washington University School of Medicine, St. Louis, MO), BTLA−/−RAG1−/− mice, and HVEM−/− mice (11) were cohoused with the C57BL/6 and RAG1−/− control groups for ≥4 wk prior to infection. For selected experiments (Fig. 1C), heterozygous BTLA−/− and BTLA+/− or BTLA+/+ littermates were generated by backcrossing BTLA−/− mice to C57BL/6 mice. Mice were infected by s.c. injection of 1000 S. ratti iL3 in the footpad, and parasite burden in tissue and intestine, as well as larval output, was quantified as described (10, 12). For IL-9 blockade, mice received 100 μg anti–IL-9 mAb (MM9C1; Bio X Cell, West Lebanon, NH) or isotype control i.p. at days 0 and 3 of S. ratti infection.
Cytokine production and mast cell activation
Mice were sacrificed at day 6 p.i., mesenteric lymph node (mLN) cells (5 × 105/well) were stimulated with S. ratti Ag or anti-CD3 (145-2C11), and cytokines in the culture supernatants were measured as described (13
Statistical analysis was performed with GraphPad Prism software (La Jolla, CA) using two-way ANOVA for repeated experiments. Group sizes in experiments were always ≥3. The data are presented as mean ± SEM.
Results and Discussion
Monitoring the expression of regulatory receptors during murine S. ratti infection, we observed increased BTLA expression specifically in the mLNs, but not on leukocytes, in blood, or in the peritoneum, of S. ratti–infected mice (Supplemental Fig. 1). Within the mLNs, BTLA expression was induced predominantly on CD4+ T lymphocytes and to a lesser extent on CD8+ T lymphocytes. Constitutive BTLA expression on B cells was not modulated during infection. BTLA also was induced on a small fraction of CD19− TCR− cells in the mLNs of S. ratti–infected mice that displayed variable CD11b, CD11c, and Gr-1 expression and were CD49b (DX5) and F4/80 negative.
To evaluate a possible biological function for BTLA-mediated signaling during S. ratti infection, we compared the parasite burden in BTLA−/− and BTLA-expressing mice. Although the numbers of tissue-migrating larvae in the lung and head were similar (Fig. 1A, 1B), BTLA−/− mice displayed significantly reduced numbers of parasitic adults in the small intestine at day 6 p.i. compared with BTLA-expressing wild-type mice (Fig. 1C). Reduced parasite burden in BTLA−/− mice was reflected by the reduced release of S. ratti–derived DNA in the feces as an indicator of larval output throughout infection until clearance by day 29 p.i. (Fig. 1D).
HVEM is the only ligand described for BTLA (3). Compared with wild-type mice, HVEM−/− mice displayed increased numbers of tissue-migrating larvae in lung and head (Fig. 2A, 2B). Interestingly, and despite the initially increased larval burden in the tissue, numbers of parasitic adults in the small intestine were reduced in HVEM−/− mice (Fig. 2C). Subsequent larval output in the feces also was drastically reduced in the absence of HVEM at all time points examined p.i. (Fig. 2D).
Taken together, our results show no impact of BTLA-mediated regulation during the control of tissue-migrating larvae. HVEM apparently promoted eradication of migrating larvae during the first 2 d of infection independently of BTLA in the wild-type state. HVEM, as a signaling receptor itself, may have responded to engagement by alternative HVEM ligands, such as LIGHT, CD160, or lymphotoxin α (5, 7).
With regard to intestinal immunity to S. ratti, our results clearly show that BTLA and its ligand HVEM interfere with efficient eradication of parasites from the small intestine of infected mice. Control of Strongyloides infection in the intestine was shown to depend on a functional Th2 response (14, 15) and activated mast cells (16). To analyze the immunological mechanism that reduced parasite burden in BTLA−/− and HVEM−/− mice, we quantified the cytokine response during S. ratti infection (Fig. 3A). Although IL-13 and IL-10 production by mLN cells was comparable in all mouse strains, the absence of either BTLA or its ligand HVEM increased IL-9 production in response to both CD3 engagement and S. ratti–specific stimulation (Fig. 3A and data not shown).
IL-9 is a cytokine with pleiotropic functions (17) that promotes expulsion of intestinal parasites (18, 19). We demonstrated recently that neutralization of IL-9 increased parasite burden and reduced mast cell degranulation in S. ratti–infected BALB/c and C57BL/6 mice (13). To quantify mast cell degranulation during infection, we measured the serum concentration of mcpt-1 that is specifically released by mucosal mast cells (20). Compared with wild-type mice, both BTLA and HVEM deficiency were accompanied by increased early degranulation of mucosal mast cells at 2 and 3 d post–S. ratti infection (Fig. 3B). Furthermore, in vivo neutralization of IL-9 partially abrogated the improved host defense observed in BTLA−/− and HVEM−/− mice (Fig. 3C). Collectively, our data suggest that accelerated IL-9–driven mast cell activation contributed to improved expulsion of S. ratti from the intestine of BTLA−/− and HVEM−/− mice.
Our results agree with previous studies showing that negative coregulation via BTLA dampens the adaptive immune response during murine Plasmodium infection. Decreased parasitemia and increased cytokine production by BTLA−/− CD4+ T cells were observed in a model for protective immunity in the absence of BTLA-mediated regulation (21). Conversely, an increase in negative signaling into BTLA-expressing cells via BTLA ligation prevented the genesis of cerebral malaria in a model of immune pathology (22). The BTLA–HVEM pathway also plays a central role in the regulation of T cell–mediated intestinal inflammation in the absence of infection. HVEM expressed by intestinal epithelial cells ameliorated T cell transfer colitis by engagement of BTLA on the adoptively transferred CD4+ T cells (23). HVEM expression on T cells apparently promoted colitis, because HVEM deficiency in transferred T cells resulted in less severe colitis (24), thus emphasizing the dual nature of HVEM-BTLA–mediated regulation (7).
Although negative regulation of T and B cells and NKT cells by BTLA is well established (1), accumulating evidence suggests that BTLA may regulate innate immune cells in Plasmodium (21) or Listeria infection (25) or during experimental sepsis (26, 27).
We observed BTLA upregulation on CD11b+ and CD11c+ but not on F4/80+ non-B non-T cells (Supplemental Fig. 1). To elucidate the contribution of these BTLA+ innate leukocytes to the regulation of acute nematode infection, we analyzed the impact of BTLA deficiency in the absence of adaptive immunity using RAG1−/− and BTLA−/−RAG1−/− mice. Although RAG1−/− mice are unable to terminate infection (data not shown), initial control of parasite burden can be maintained independently of B and T cells. Thus, RAG1−/− and wild-type mice displayed comparable numbers of parasitic adults in the small intestine at day 6 p.i. (Fig. 4). BTLA deficiency in the presence of B and T cells strongly reduced parasite burden. In the absence of adaptive immunity (i.e., comparing RAG1−/− mice and BTLA−/−RAG1−/− mice), the advantage of BTLA deficiency was still significant but less pronounced.
Our combined results could be explained by a model whereby BTLA, predominantly on CD4+ T cells, is triggered by HVEM and delivers negative coregulation into BTLA+ cells, thus interfering with the protective immune response to the intestinal parasite. BTLA expression on innate leukocytes may contribute to this negative coregulation.
Although immune-competent wild-type mice clear S. ratti infection within 30 d, even this short survival of parasites in their host critically depends on immune evasion. We showed previously that Foxp3+ regulatory T cells (Tregs) promote survival of S. ratti in the small intestine of BALB/c mice by dampening IL-9–driven mast cell activation (13). Because Foxp3+ Tregs did not play a central role in the survival of S. ratti, within the even more susceptible C57BL/6 mice, we hypothesized that redundant alternative regulatory pathways contributed to immune evasion on this genetic background. In the current study we provide evidence that, in addition to Tregs, the BTLA–HVEM network represents one additional layer of regulation that can be exploited by intestinal parasites to delay immune-mediated expulsion.
The authors have no financial conflicts of interest.
We thank Dr. Kenneth Murphy for providing the BTLA−/− mice, Erin C. Boyle for critical reading of the manuscript, and Andreas Hahn for assistance with statistics.
M.B. is supported by Deutsche Forschungsgemeinschaft Grant 3754/2-1. T.J. is supported by Sonderforschungsbereich 841, a special funding tool for large collaborative research by the Deutsche Forschungsgemeinschaft.
The online version of this article contains supplemental material.
Abbreviations used in this article:
- B and T lymphocyte attenuator
- herpes virus entry mediator
- infective third-stage larvae
- lymphotoxin-like, exhibits inducible expression, competes with HSV gpD for HVEM, a receptor expressed by T lymphocytes
- mouse mast cell protease-1
- mesenteric lymph node
- regulatory T cell.
- Received October 6, 2014.
- Accepted December 12, 2014.
- Copyright © 2015 by The American Association of Immunologists, Inc.