H:S Blodbank KI2033, Copenhagen University Hospital, Blegdamsvej 9, DK-2100 Copenhagen, Denmark
http://www.100md.com
感染与免疫杂志 2006年第6期
Department of Microbiology and Immunology, Thomas Jefferson University, Philadelphia, Pennsylvania 19107
Department of Biochemistry and Molecular Biology, Mayo Clinic Scottsdale, Scottsdale, Arizona 85259
Department of Pathobiology, University of Pennsylvania, Philadelphia, Pennsylvania 19104
ABSTRACT
Several studies have demonstrated roles for eosinophils during innate and adaptive immune responses to helminth infections. However, evidence that eosinophils are capable of initiating an immune response to parasite antigens is lacking. The goal of the present in vitro study was to investigate the potential of eosinophils to serve as antigen-presenting cells (APC) and initiate an immune response to parasite antigens. Purified eosinophils were exposed to soluble Strongyloides stercoralis antigens, and the expression of various surface markers involved in cell activation was examined. Antigen-exposed eosinophils showed a sixfold increase in expression levels of CD69 and major histocompatibility complex (MHC) class II, a fourfold increase in levels of T-cell costimulatory molecule CD86, and a twofold decrease in levels of CD62L compared to eosinophils cultured in medium containing granulocyte-macrophage colony-stimulating factor. The ability of eosinophils to present antigen to T cells was determined by culturing them with T cells in vitro. Eosinophils pulsed with antigen stimulated antigen-specific primed T cells and CD4+ T cells to increase interleukin-5 (IL-5) production. The blocking of MHC class II expression on eosinophils inhibited their ability to induce IL-5 production by CD4+ T cells in culture. Antigen-pulsed eosinophils were able to prime nave T cells and CD4+ T cells in culture and polarized them into Th2 cells producing IL-5 similar to that induced by antigen-loaded dendritic cells. These results demonstrate that eosinophils are capable of activating antigen-specific Th2 cells inducing the release of cytokines and assist in the priming of nave T cells to initiate Th2 responses against infection. This study highlights the potential of eosinophils to actively induce immune responses against infection by amplifying antigen-specific Th2-cell responses.
INTRODUCTION
Eosinophils have been shown to function as antigen-presenting cells (APC) in several experimental allergy model systems (24). Antigen-loaded eosinophils present antigen to primed T cells and increase Th2 cytokine production (12, 31). The eosinophils migrate into local lymph nodes and localize in the T-cell-rich paracortical zones, where they stimulate the expansion of CD4+ T cells. Antigen-exposed eosinophils were apparently activated, based on their increased expression levels of CD80, CD86, and major histocompatibility complex (MHC) class II molecules (5, 12, 25). The antigen-loaded eosinophils also promoted the production of interleukin-5 (IL-5) when placed in culture with antigen-specific CD4+ T cells isolated from allergic mice (12). There are conflicting reports, however, regarding the ability of eosinophils to act as APC to stimulate nave T cells. Eosinophils that were derived from the peritoneal cavities of IL-5-transgenic (TG) mice and pulsed with ovalbumin (OVA) in vitro were capable of sensitizing nave mice to OVA (12). In contrast to this, eosinophils purified from the bronchoalveolar lavage fluid of OVA-sensitized mice were unable to induce the proliferation of naive T cells in the draining lymph nodes of the lungs (28). Evidence that eosinophils are capable of initiating an immune response to parasite antigens is lacking, although it was demonstrated that granulocyte-macrophage colony-stimulating factor (GM-CSF)-activated eosinophils are capable of acting as specific antigen-presenting cells to a T-cell clone derived from mice infected with the cestode Mesocestoides corti (3). In addition, eosinophils recovered from mice infected with the nematode Brugia malayi express high levels of MHC class II molecules, suggesting the potential of these cells to act as APC (14).
Protective immunity to Strongyloides stercoralis infection in mice has been shown to be dependent on eosinophils during both the innate and the adaptive immunity responses (8, 21). IL-5–/– mice, which are incapable of augmenting blood and tissue eosinophil levels, have diminished larval killing during the innate immune response. Complementary studies using IL-5-TG mice, which overexpress IL-5 and develop a profound systemic eosinophilia, demonstrated elevated levels of larval killing in primary infections (8). Adaptive protective immunity to S. stercoralis infection in mice has been shown to be dependent on CD4+ T cells and in particular on the Th2 cytokines IL-4 and IL-5 (20). IL-5–/– mice failed to develop adaptive protective immunity to infection with S. stercoralis. However, the adoptive transfer of eosinophils into IL-5–/– mice at the time of immunization with live S. stercoralis larvae (L3) reconstituted the ability of IL-5–/– mice to develop adaptive protective immunity against the infection. The eosinophils did not have a direct role in killing the larvae but rather induced the production of protective antibody (Ab) that in conjunction with complement and other cells, eliminated the challenge infection (8). Therefore, eosinophils may have a dual function during the development of adaptive protective immunity to S. stercoralis infection in mice. The first function is to kill the larvae through an innate mechanism, as has been previously reported (8). This results in converting the comparatively large larvae into a form that can be phagocytized by APC. It is hypothesized in the present study that eosinophils represent the APC that initiate the adaptive protective immune response.
The present in vitro study utilized the S. stercoralis infection of mice as a model to investigate whether eosinophils have the potential to act as APC in the initiation and development of a protective Th2 response during helminth infections. Purified eosinophils were exposed to S. stercoralis antigens, and activated, antigen-pulsed eosinophils were analyzed for their potential to present antigens to T cells. Results show that eosinophils become activated after exposure to antigens, are capable of priming nave T cells, and can induce an antigen-specific Th2 response through MHC class II peptides.
MATERIALS AND METHODS
Experimental animals and parasites. BALB/cJ and C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME). The IL-5-transgenic mice (mouse line NJ.1638) were bred in the Laboratory of Animal Sciences facility at Thomas Jefferson University (Philadelphia, PA). All mice were housed in filter top microisolator boxes under light- and temperature-controlled conditions.
S. stercoralis L3 organisms were obtained from the fresh stools of a laboratory dog infected with the parasite according to methods previously described (1). Larvae were collected from charcoal cultures and washed by centrifugation and resuspension in a 1:1 mixture of Iscove modified Dulbecco medium (Sigma) and NCTC-135 (Sigma) with a mixture of 100 U of penicillin, 100 μg of streptomycin per ml (Gibco, Grand Island, NY), and 25 μg of levofloxacin per ml (Ortho-McNeil Pharmaceutical, Raritan, NJ).
Antigen preparation. Soluble larval antigens from S. stercoralis larvae were prepared as described previously (9). Briefly, L3 were washed in phosphate-buffered saline (PBS) supplemented with 100 U of penicillin and 100 μg of streptomycin per ml and stored at –80°C. L3 were thawed and homogenized in the presence of a protease inhibitor cocktail (Sigma, St. Louis, MO) and then sonicated. The homogenized and sonicated L3 were incubated in PBS at 4°C for 18 h with continuous mixing. PBS-soluble antigens were removed, filter sterilized, and stored at –80°C. PBS-insoluble proteins were resuspended in 20 mM Tris-HCl-0.5% deoxycholic acid (DOC; Sigma) with continuous mixing for 12 h at 4°C. The DOC-soluble antigens were then dialyzed (6- to 8-kDa cutoff) against PBS for 12 h, concentrated, filter sterilized, and stored at –80°C. Protein concentration was quantitated by a Micro BCA protein assay kit (Pierce, Rockford, IL). Lipopolysaccharide (LPS) content in the soluble antigen preparations was measured using a Limulus amebocyte lysate assay; LPS was detected at about 1 ng of LPS/mg of S. stercoralis antigens. As no difference was observed in the ability of these antigens to induce T cells, either DOC-soluble or PBS-soluble antigens were used in the present study.
Immunization and isolation of T cells. C57BL/6J mice were immunized with live S. stercoralis L3 as described previously (1). The immunization protocol consisted of two subcutaneous injections with 5,000 live L3 administered 2 weeks apart. Four weeks after the second injection, mice were sacrificed and primed T cells and CD4+ T cells were isolated from the spleens using magnetic-activated cell-sorting columns (Miltenyi Biotech, Auburn, CA). T cells and CD4+ T cells were purified by positive selection with Ab-conjugated magnetic beads specific for CD90 (Thy 1.2) and CD4 (L3T4), respectively. Nave T cells and CD4+ T cells were obtained from the spleens of C57BL/6 mice.
Isolation of eosinophils. Eosinophils were isolated from the spleens of nave IL-5-TG mice, following the method previously described for isolating them from blood (2). In brief, single-cell suspensions of spleens were layered onto a Percoll gradient (60% Percoll [ = 1.084] 1x Hanks balanced salt solution, 15 mM HEPES [pH 7.4], and 0.003 N HCl) and centrifuged (45 min, 1,200 x g, 4°C). The buffy coat was recovered and washed twice in PBS containing 2% fetal bovine serum (FBS). Eosinophils were isolated using magnetic-activated cell-sorting columns. Specifically, B cells and T cells were removed by positive selection with Ab-conjugated magnetic beads specific for CD45-R (B220) B cells and CD90 (Thy 1.2) T cells. Cells were placed in a Cytospin 3 apparatus (Shandon, Pittsburgh, PA) and stained for differential counts with a Hema 3 stain set (Fisher Diagnostics, Middletown, VA) to check for purity. Approximately 95% of the cells were eosinophils, along with neutrophils (1 to 2%), macrophages (2 to 4%), and lymphocytes (1 to 2%) as contaminating cells.
Activation of eosinophils. For activation of eosinophils, cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated and filtered FBS (HyClone, Logan, UT), 2 mM L-glutamine (GIBCO), 100 U/ml penicillin plus 100 μg/ml streptomycin (Gibco), and 50 μM 2-mercaptoethanol (2-ME; Sigma) either in the presence or in the absence of 5 ng/ml GM-CSF (Peprotech, Rocky Hill, NJ) and pulsed with 100 μg/ml S. stercoralis antigens or 1 μg/ml LPS (Sigma) for 4 h. To determine the activation state of the eosinophils, cells were harvested, washed, and resuspended in fluorescence-activated cell sorter buffer (1x phosphate-buffered saline, 0.2% bovine serum albumin fraction V, 4 mM sodium azide) and the Fc receptors on cells were blocked using Fc-Block CD16/CD32 Ab (BD Pharmingen, San Jose, CA) for 20 min on ice. The expression of various surface molecules was quantified by immunofluorescent staining using monoclonal antibodies (MAbs) specific for I-Ab, CD69, CD62L, CD29 (integrin-1), CD86, CD80, CD11c (all purchased from BD Pharmingen), and CCR3 (R & D system, Minneapolis, MN), conjugated to either fluorescein isothiocyanate, phycoerythrin, or allophycocyanin. Samples were analyzed using a FACSCalibur flow cytometer and CellQuest software (Becton Dickinson, Franklin Lakes, NJ).
Generation of dendritic cells and macrophages. Dendritic cells (DC) and macrophages were generated from bone marrow by collecting bone marrow from femurs and tibias from mice. Dendritic cells were generated as previously described (11). In brief, cells were seeded into petri dishes at 2 x 105/ml in 10 ml RPMI 1640 supplemented with 10% heat-inactivated and filtered FBS, 2 mM L-glutamine, 100 U/ml penicillin plus 100 μg/ml streptomycin, and 50 μM 2-ME, with the addition of 20 ng/ml GM-CSF. On day 3, a further 10 ml of medium containing 20 ng/ml GM-CSF was added. On day 6, 10 ml culture supernatant was removed and replaced with 10 ml fresh culture medium containing 20 ng/ml GM-CSF. For generation of immature DC, on day 8 plates were fed as on day 6, but only 5 ng/ml GM-CSF was added in fresh media, and cells were harvested 18 h later (day 9). For activation of DC, cells were treated in the same way, with the addition of the S. stercoralis antigens (100 μg/ml) for the final 18-h incubation. To generate macrophages from bone marrow cells, cells were cultured in Dulbecco modified Eagle medium (Sigma) supplemented with 20% heat-inactivated and filtered fetal calf serum (HyClone), 2 mM L-glutamine, 100 U/ml penicillin plus 100 μg/ml streptomycin, and 50 μM 2-ME, with the addition of 30% L929 cell-conditioned medium as a source of macrophage colony-stimulating factor (26), at 37°C in a 5% CO2 atmosphere as previously described (4), with minor modifications. At day 3, culture supernatants were removed and replaced with fresh medium. Cells were harvested on day 6 using ice-cold PBS (pH 7.2). For activation of macrophages, cells were treated with the addition of the S. stercoralis antigens (100 μg/ml) and cultured for the final 18 h. Cells were fixed with 4% paraformaldehyde and used for in vitro assays.
Determination of the ability of eosinophils to stimulate T cells. Eosinophils and DC were pulsed with S. stercoralis antigens (100 μg/ml) for 18 h in RPMI medium containing GM-CSF (5 ng/ml) and fixed in paraformaldehyde (4%) prior to coculture with T cells. Purified, primed T cells or primed CD4+ T cells (2.5 x 106) were cultured with either unpulsed or antigen-pulsed S. stercoralis eosinophils (5 x 105) in Dulbecco modified Eagle medium (Sigma) supplemented with 10% heat-inactivated and filtered fetal calf serum (HyClone), 2 mM L-glutamine (Life Technologies), 100 U/ml penicillin plus 100 μg/ml streptomycin (Life Technologies), and 50 μM 2-ME (Sigma). Cultures were also set up with antigen-pulsed or unpulsed S. stercoralis DC (5 x 105) to compare the potentials of eosinophils to stimulate T cells. After 96 h, supernatants were collected from the culture and analyzed for IL-5, IL-4, and gamma interferon (IFN-) production. To determine whether eosinophils can prime nave cells, purified nave T cells or CD4+ T cells were cultured with antigen-pulsed eosinophils or DC. After 48 h of culture, recombinant interleukin-2 was added at 20 ng/ml and cells were then rested for 72 to 96 h, before being restimulated with anti-CD3 Ab (BD Pharmingen) (0.5 μg/well) for 72 h. The supernatants harvested from these cultures were analyzed for IL-5 and IL-4 production by sandwich enzyme-linked immunosorbent assays (ELISAs) using paired MAbs (TRFK.5 and TRFK.4 for detection of IL-5 and BVD6-24G2 and BVD4-1D11 for detection of IL-4; BD Pharmingen). IFN- production was measured using an ELISA kit (BD Pharmingen) as directed.
MHC class II restriction of antigen presentation by eosinophils. To determine whether antigen presentation by eosinophils was through MHC class II peptides, antigen-pulsed eosinophils or unpulsed eosinophils were fixed and incubated with monoclonal antibodies specific to MHC class II molecules (I-Ab; BD Pharmingen) for 30 min at 37°C to block the presentation of MHC class II peptides, as described previously (3). Eosinophils were washed twice with PBS (pH 7.2) to remove unbound antibodies and then cultured with primed CD4+ T cells. In some studies, paraformaldehyde-fixed control eosinophils (5 x 105) and eosinophils pulsed with antigen, obtained from IL-5-TG mice of the C57BL/6J background (H-2b haplotype), were cultured with 2.5 x 106 primed CD4+ T cells isolated from either immunized BALB/cJ (H-2d haplotype) or C57BL/6J (H-2b haplotype) mice. Alternatively, antigen-pulsed and fixed eosinophils were mixed with H-2d haplotype dendritic cells (2.5 x 104; 5% of the eosinophil population) or macrophages (2.5 x 104; 5% of the eosinophil population) derived from the bone marrow of nave BALB/cJ mice, and the mixture was cultured with 2.5 x 106 primed CD4+ T cells from immunized BALB/cJ mice. After 96 h, supernatants from these in-vitro cultures were analyzed for IL-5 production by ELISA.
Statistical analysis. Statistical analysis of the data was performed using MGLH multifactorial analysis of variance with Systat version 11 software (Systat, Evanston, IL). Fisher's least-significant-difference test was performed for post hoc analyses. Probability values of less than 0.05 were considered significant.
RESULTS
Activation of Ag-pulsed eosinophils. Purified eosinophils were exposed to S. stercoralis antigens in the presence of GM-CSF to determine whether the antigens would activate the eosinophils. Antigen-pulsed eosinophils in the presence of GM-CSF showed (i) a sixfold increase in expression levels of CD69 and MHC class II (Fig. 1A, panel III), (ii) a fourfold increase in expression levels of CD86 (Fig. 1B, panel III), and (iii) a twofold decrease in expression levels of CD62L (Fig. 1C, panel III) compared to cells incubated only in medium or in medium containing GM-CSF. No differences were seen in levels of CD29 (Fig. 1C, panel III) and CD80 expression (data not shown). To determine whether the S. stercoralis antigens require GM-CSF for activation of eosinophils, purified eosinophils were exposed to antigen in the absence of GM-CSF and analyzed for the expression patterns of the surface proteins. Antigen-pulsed eosinophils in the absence of GM-CSF showed a fourfold increase in expression levels of CD69 and MHC class II (Fig. 1A, panel IV) and a fivefold increase in expression levels of CD86 (Fig. 1B, panel IV). There was also a twofold increase in expression levels of CD29 and a threefold reduction in those of CD62L (Fig. 1C, panel IV). Eosinophils incubated with LPS (1 μg/ml) did not show any significant change in surface levels of CD69, MHC class II, CD80, and CD86 (data not shown). Overall, the increase in expression levels of CD69, MHC class II, CD86, and CD29 and the decrease in CD62L expression levels on the surfaces of eosinophils were observed on eosinophils exposed to S. stercoralis antigens regardless of the presence of GM-CSF, thereby demonstrating that S. stercoralis antigens could directly activate eosinophils.
Antigen-pulsed eosinophils stimulate primed T cells to produce IL-5. Eosinophils pulsed with antigen in the presence of GM-CSF were fixed in paraformaldehyde and then cultured with T cells isolated from mice immunized with S. stercoralis. There was a significant increase in IL-5 production by primed T cells cultured with antigen-pulsed eosinophils compared to production levels in T cells cultured with untreated eosinophils (Fig. 2A). Primed CD4+ T cells isolated from immunized mice and placed in culture with fixed, antigen-pulsed eosinophils also showed a significantly increased production of IL-5 (Fig. 2B). Experiments were performed to determine whether the induction of IL-5 production by primed T cells was due to the presentation of antigen by eosinophils in conjunction with MHC class II molecules. Eosinophils pulsed with antigen were treated with anti-MHC class II MAb prior to incubation with primed CD4+ T cells. The blocking of MHC class II molecules on the surfaces of eosinophils, through treatment with the anti-MHC class II MAb, eliminated the ability of antigen-pulsed eosinophils to stimulate IL-5 production by primed CD4+ T cells (Fig. 2C).
Possible contaminating dendritic cells and macrophages within eosinophil preparation are not responsible for antigen presentation to T cells. The purified eosinophils, used in the experiments described above, contained approximately 5% contaminating cells consisting of a mixture of lymphocytes, neutrophils, dendritic cells, and macrophages. Experiments were performed to determine whether the functional APC in the purified eosinophil population were eosinophils or whether they were the contaminating dendritic cells and macrophages. In these studies, eosinophils were purified from C57BL/6J (H-2b haplotype) mice, whereas dendritic cells and macrophages were recovered from BALB/cJ (H-2d haplotype) mice. Antigen-pulsed and fixed eosinophils from C57BL/6 mice induced IL-5 production from primed C57BL/6J mouse-derived CD4+ T cells but not from cells isolated from immunized BALB/c mice, because of the differences in MHC class II molecules (Fig. 3). Cultures were then set up with BALB/c mouse-derived CD4+ T cells and antigen-pulsed eosinophils from C57BL/6J mice. To these cultures were added antigen-pulsed dendritic cells or macrophages from BALB/c mice in numbers equal to the 5% contamination of the eosinophils population. IL-5 production by BALB/c mouse-derived CD4+ T cells did not increase by the addition of the pulsed H-2d haplotype dendritic cells or macrophages (Fig. 3). These experiments therefore demonstrate that the induction of IL-5 production by primed CD4+ T cells was due to the presentation of antigen by eosinophils and not by other cells contaminating the eosinophils population.
Antigen-pulsed eosinophils prime nave T cells to produce IL-4 and IL-5. T cells and CD4+ T cells, isolated from the spleens of nave C57BL/6 mice, were cultured with fixed eosinophils pulsed with S. stercoralis antigen in the presence of GM-CSF. The levels of IL-5 produced by nave T cells (Fig. 4A) and CD4+ T cells (Fig. 4B) were significantly increased after culture with antigen-pulsed eosinophils. Experiments were performed to compare the antigen-presenting potential of eosinophils and dendritic cells to induce primary and secondary Th1 and Th2 responses to S. stercoralis antigen in T cells. Both eosinophils and dendritic cells did not induce the production of IFN- from nave and primed T cells. However, dendritic cells were superior to eosinophils in their ability to induce IL-5 (Fig. 5A) and IL-4 (Fig. 5C) production by primed T cells. Eosinophils and dendritic cells showed similar potentials to prime nave T cells to produce IL-5 (Fig. 5B) and IL-4 (Fig. 5D).
DISCUSSION
The objectives of these studies were to determine whether eosinophils become activated after exposure to helminth antigens and whether they were potentially APC for initiation and restimulation of immune responses to helminth infections. The exposure of nave eosinophils to S. stercoralis antigens and GM-CSF increased expression levels of CD69, which serves as an activation marker for both human and mouse eosinophils (18, 22). Moreover, expression levels of MHC class II peptides and the T-cell costimulatory molecule CD86 were also elevated on eosinophils exposed to both antigens and GM-CSF but not on eosinophils exposed to only GM-CSF. Studies of eosinophil activation in mice infected with Nippostrongylus brasiliensis have shown increased CD29 and decreased CD62L levels on eosinophils found within inflammatory tissues compared to those on eosinophils from the blood (29). CD29 is a member of the very-late-activation integrin family and known to function as an adhesion and signaling molecule. CD62L (L-selectin) plays a critical role in initiating the interactions of cells with the activated endothelium at an inflammatory site and is characteristically down-modulated in response to cell activation (6, 16, 19). Increased CD29 and decreased CD62L levels were also observed in the present study after eosinophils were exposed to S. stercoralis antigens. Soluble S. stercoralis antigens in the absence of GM-CSF were also capable of activating mouse eosinophils. Helminths have been shown to produce homologues of several mammalian cytokines (13), and it is possible that the soluble extract of S. stercoralis antigens contains homologues of mammalian cytokines that may facilitate the direct activation of eosinophils.
In vitro experiments were performed to investigate whether activated eosinophils could act as APC and stimulate T cells to produce IL-5. Before culture with T cells or CD4+ T cells, the eosinophils were fixed to avoid degranulation, block alloreactivity, and preserve eosinophils during subsequent culture (3, 30). Fixed, antigen-pulsed APC have been shown to have unchanged expression levels of MHC class II molecules and to be competent for stimulating the proliferation of antigen-reactive T cells (17). GM-CSF was used during in vitro pulsing of eosinophils with antigen, as it stimulates the functional activity of eosinophils and also maintains maximum viability of cells (27). Purified T cells and CD4+ T cells isolated from immunized mice and placed in culture with antigen-pulsed eosinophils were stimulated to produce IL-5. A similar observation has been made in an experimental mouse model for allergic diseases (12). Furthermore, the blocking of MHC class II peptides completely inhibited the stimulation of primed CD4+ T cells, thereby demonstrating that stimulation of CD4+ T cells by antigen-pulsed eosinophils is MHC class II dependent, as has previously been reported (3, 27).
It was essential, however, to prove that T-cell stimulation and IL-5 production in culture were induced by the antigen-pulsed eosinophils and not by the contaminating cells (5%) in the eosinophil preparation. Experiments were performed to capitalize on the genetic difference between BALB/cJ and C57BL/6J mice in their MHC class II molecules. The actual number of cells that comprised the 5% contamination of the eosinophil population was determined (2.5 x 104), and that quantity of pure dendritic cells or macrophages derived from the bone marrow of BALB/c mice was mixed with eosinophils isolated from IL-5-TG mice of the C57BL/6J background. The data from these experiments demonstrated that the few contaminating dendritic cells and macrophages in the eosinophil preparation could not have accounted for the observed T-cell stimulation and IL-5 production. This conclusion is in agreement with previous studies that have demonstrated that using fewer than 2 x 104 APC in culture failed to stimulate the proliferation of T cells (7, 27). It was therefore concluded that the antigen-pulsed eosinophils presented antigen to the primed T cells, which were then induced to produce IL-5.
There have been conflicting reports on the capability of eosinophils to present antigen to nave T cells. Eosinophils isolated from the bronchoalveolar lavage fluid of OVA-sensitized and -challenged mice fail to induce the division of OVA-T-cell-receptor-transgenic T cells, whereas dendritic cells induced T-cell divisions (28). This conclusion differs from the results of another study that demonstrated that OVA-pulsed eosinophils are capable of sensitizing mice when injected repeatedly into the peritoneal cavity (12). As eosinophils are associated with helminth parasites at the initial stage of infection (8, 10, 15, 23), it is possible that these cells capture antigens from the helminths, migrate to T-cell-rich regions, and present antigens to T cells to initiate antigen-specific T-cell responses. Therefore, the present study also examined the ability of antigen-exposed eosinophils to prime the nave T cells. IL-5 production from nave T cells or CD4+ T cells cultured with antigen-loaded eosinophils was significantly higher than that from cells cultured with untreated eosinophils. These data demonstrated that activated eosinophils are capable of functioning as a professional APC with the potential to prime nave T cells and initiate the development of an antigen-specific immune response against infection. More importantly, the data demonstrate that the antigen-presenting capacity of an eosinophil to induce a Th2 response was equivalent to that of a traditional APC, such as a dendritic cell in a primary response. Antigen-pulsed dendritic cells were, however, superior to eosinophils in their ability to induce IL-4 and IL-5 production from primed T cells.
The results of the present report therefore support the study hypothesis that eosinophils, which come in direct contact with parasites during the innate immune response, may capture parasitic antigens and present them to T cells in the induction of the adaptive immune response to S. stercoralis infection in mice. Furthermore, the observation that eosinophils and dendritic cells function equally well at inducing the primary response to the infection suggests that eosinophils may be essential as the APC for the induction of adaptive immunity. This conclusion is supported by previous data which showed that the immunization of IL-5–/– mice resulted in diminished antibody responses, even though there were abundant dendritic cells in these mice. IL-5–/– mice reconstituted with eosinophils at the time of immunization developed sustained antibody responses equal to those in wild-type mice, suggesting that eosinophils were required only at the initiation of the response, which would be consistent with a role for the eosinophils as APC (8).
In summary, the present study demonstrated that S. stercoralis antigens activate eosinophils and induce the expression of MHC class II and T-cell costimulatory molecules and that these activated eosinophils can stimulate T cells for antigen-specific immune responses. In addition, this study demonstrated that eosinophils have the ability to prime nave T cells and assist the immune system in the initiation and development of antigen-specific T-cell responses against S. stercoralis. These results suggest that in addition to their role as terminal effector cells in helminth infections, eosinophils may also serve as specific antigen-presenting cells and prime the immune response during infection. Significantly, these data also demonstrate that eosinophils can communicate with T cells and suggest that eosinophils are integral in the interface between the innate and adaptive immune responses.
ACKNOWLEDGMENTS
This work was supported by NIH grants RO1 AI47189 and RO1 A1 22662 and the Mayo Foundation.
We thank Jessica Hess and Amy O'Connell for critical reading of the manuscript and Juergen Landmann for expert technical assistance.
REFERENCES
1. Abraham, D., H. L. Rotman, H. F. Haberstroh, W. Yutanawiboonchai, R. A. Brigandi, O. Leon, T. J. Nolan, and G. A. Schad. 1995. Strongyloides stercoralis: protective immunity to third-stage larvae in BALB/cByJ mice. Exp. Parasitol. 80:297-307.
2. Borchers, M. T., P. J. Justice, T. Ansay, V. Mancino, M. P. McGarry, J. Crosby, M. I. Simon, N. A. Lee, and J. J. Lee. 2002. Gq signaling is required for allergen-induced pulmonary eosinophilia. J. Immunol. 168:3543-3549.
3. Del Pozo, V., B. De Andres, E. Martin, B. Cardaba, J. C. Fernandez, S. Gallardo, P. Tramon, F. Leyva-Cobian, P. Palomino, and C. Lahoz. 1992. Eosinophil as antigen-presenting cell: activation of T cell clones and T cell hybridoma by eosinophils after antigen processing. Eur. J. Immunol. 22:1919-1925.
4. Descoteaux, A., G. Matlashewski, and S. J. Turco. 1992. Inhibition of macrophage protein kinase C-mediated protein phosphorylation by Leishmania donovani lipophosphoglycan. J. Immunol. 149:3008-3015.
5. Duez, C., A. Dakhama, A. Tomkinson, P. Marquillies, A. Balhorn, A. B. Tonnel, D. L. Bratton, and E. W. Gelfand. 2004. Migration and accumulation of eosinophils toward regional lymph nodes after airway allergen challenge. J. Allergy Clin. Immunol. 114:820-825.
6. Elangbam, C. S., C. W. Qualls, Jr., and R. R. Dahlgren. 1997. Cell adhesion molecules—update. Vet. Pathol. 34:61-73.
7. Handzel, Z. T., W. W. Busse, J. B. Sedgwick, R. Vrtis, W. M. Lee, E. A. Kelly, and J. E. Gern. 1998. Eosinophils bind rhinovirus and activate virus-specific T cells. J. Immunol. 160:1279-1284.
8. Herbert, D. R., J. J. Lee, N. A. Lee, T. J. Nolan, G. A. Schad, and D. Abraham. 2000. Role of IL-5 in innate and adaptive immunity to larval Strongyloides stercoralis in mice. J. Immunol. 165:4544-4551.
9. Herbert, D. R., T. J. Nolan, G. A. Schad, S. Lustigman, and D. Abraham. 2002. Immunoaffinity-isolated antigens induce protective immunity against larval Strongyloides stercoralis in mice. Exp. Parasitol. 100:112-120.
10. Korenaga, M., Y. Hitoshi, K. Takatsu, and I. Tada. 1994. Regulatory effect of anti-interleukin-5 monoclonal antibody on intestinal worm burden in a primary infection with Strongyloides venezuelensis in mice. Int. J. Parasitol. 24:951-957.
11. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and G. Schuler. 1999. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223:77-92.
12. MacKenzie, J. R., J. Mattes, L. A. Dent, and P. S. Foster. 2001. Eosinophils promote allergic disease of the lung by regulating CD4(+) Th2 lymphocyte function. J. Immunol. 167:3146-3155.
13. Maizels, R. M., A. Balic, N. Gomez-Escobar, M. Nair, M. D. Taylor, and J. E. Allen. 2004. Helminth parasites—masters of regulation. Immunol. Rev. 201:89-116.
14. Mawhorter, S. D., E. Pearlman, J. W. Kazura, and W. H. Boom. 1993. Class II major histocompatibility complex molecule expression on murine eosinophils activated in vivo by Brugia malayi. Infect. Immun. 61:5410-5412.
15. Meeusen, E. N., and A. Balic. 2000. Do eosinophils have a role in the killing of helminth parasites Parasitol. Today 16:95-101.
16. Mengelers, H. J., T. Maikoe, B. Hooibrink, T. W. Kuypers, J. Kreukniet, J. W. Lammers, and L. Koenderman. 1993. Down modulation of L-Selectin expression on eosinophils recovered from bronchoalveolar lavage fluid after allergen provocation. Clin. Exp. Allergy 23:196-204.
17. Moreno, J., and P. E. Lipsky. 1986. Functional heterogeneity of human antigen-presenting cells: presentation of soluble antigen but not self-Ia by monocytes. J. Clin. Immunol. 6:9-20.
18. Nishikawa, K., T. Morii, H. Ako, K. Hamada, S. Saito, and N. Narita. 1992. In vivo expression of CD69 on lung eosinophils in eosinophilic pneumonia: CD69 as a possible activation marker for eosinophils. J. Allergy Clin. Immunol. 90:169-174.
19. Reyes-Reyes, M., N. Mora, G. Gonzalez, and C. Rosales. 2002. beta1 and beta2 integrins activate different signalling pathways in monocytes. Biochem. J. 363:273-280.
20. Rotman, H. L., S. Schnyder-Candrian, P. Scott, T. J. Nolan, G. A. Schad, and D. Abraham. 1997. IL-12 eliminates the Th-2 dependent protective immune response of mice to larval Strongyloides stercoralis. Parasite Immunol. 19:29-39.
21. Rotman, H. L., W. Yutanawiboonchai, R. A. Brigandi, O. Leon, G. J. Gleich, T. J. Nolan, G. A. Schad, and D. Abraham. 1996. Strongyloides stercoralis: eosinophil-dependent immune-mediated killing of third stage larvae in BALB/cByJ mice. Exp. Parasitol. 82:267-278.
22. Rumbley, C. A., and S. M. Phillips. 1999. The schistosome granuloma: an immunoregulatory organelle. Microbes Infect. 1:499-504.
23. Sasaki, O., H. Sugaya, K. Ishida, and K. Yoshimura. 1993. Ablation of eosinophils with anti-IL-5 antibody enhances the survival of intracranial worms of Angiostrongylus cantonensis in the mouse. Parasite Immunol. 15:349-354.
24. Shi, H. Z. 2004. Eosinophils function as antigen-presenting cells. J. Leukoc. Biol. 76:520-527.
25. Shi, H. Z., A. Humbles, C. Gerard, Z. Jin, and P. F. Weller. 2000. Lymph node trafficking and antigen presentation by endobronchial eosinophils. J. Clin. Investig. 105:945-953.
26. Stanley, E. R., and P. M. Heard. 1977. Factors regulating macrophage production and growth. Purification and some properties of the colony stimulating factor from medium conditioned by mouse L cells. J. Biol. Chem. 252:4305-4312.
27. Tamura, N., N. Ishii, M. Nakazawa, M. Nagoya, M. Yoshinari, T. Amano, H. Nakazima, and M. Minami. 1996. Requirement of CD80 and CD86 molecules for antigen presentation by eosinophils. Scand. J. Immunol. 44:229-238.
28. van Rijt, L. S., N. Vos, D. Hijdra, V. C. de Vries, H. C. Hoogsteden, and B. N. Lambrecht. 2003. Airway eosinophils accumulate in the mediastinal lymph nodes but lack antigen-presenting potential for naive T cells. J. Immunol. 171:3372-3378.
29. Voehringer, D., K. Shinkai, and R. M. Locksley. 2004. Type 2 immunity reflects orchestrated recruitment of cells committed to IL-4 production. Immunity 20:267-277.
30. Weller, P. F., T. H. Rand, T. Barrett, A. Elovic, D. T. Wong, and R. W. Finberg. 1993. Accessory cell function of human eosinophils. HLA-DR-dependent, MHC-restricted antigen-presentation and IL-1 alpha expression. J. Immunol. 150:2554-2562.
31. Xie, Z. F., H. Z. Shi, X. J. Qin, L. F. Kang, C. P. Huang, and Y. Q. Chen. 2005. Effects of antigen presentation of eosinophils on lung Th1/Th2 imbalance. Chin. Med. J. (Engl. Ed.) 118:6-11.(Udaikumar M. Padigel, Jam)
Department of Biochemistry and Molecular Biology, Mayo Clinic Scottsdale, Scottsdale, Arizona 85259
Department of Pathobiology, University of Pennsylvania, Philadelphia, Pennsylvania 19104
ABSTRACT
Several studies have demonstrated roles for eosinophils during innate and adaptive immune responses to helminth infections. However, evidence that eosinophils are capable of initiating an immune response to parasite antigens is lacking. The goal of the present in vitro study was to investigate the potential of eosinophils to serve as antigen-presenting cells (APC) and initiate an immune response to parasite antigens. Purified eosinophils were exposed to soluble Strongyloides stercoralis antigens, and the expression of various surface markers involved in cell activation was examined. Antigen-exposed eosinophils showed a sixfold increase in expression levels of CD69 and major histocompatibility complex (MHC) class II, a fourfold increase in levels of T-cell costimulatory molecule CD86, and a twofold decrease in levels of CD62L compared to eosinophils cultured in medium containing granulocyte-macrophage colony-stimulating factor. The ability of eosinophils to present antigen to T cells was determined by culturing them with T cells in vitro. Eosinophils pulsed with antigen stimulated antigen-specific primed T cells and CD4+ T cells to increase interleukin-5 (IL-5) production. The blocking of MHC class II expression on eosinophils inhibited their ability to induce IL-5 production by CD4+ T cells in culture. Antigen-pulsed eosinophils were able to prime nave T cells and CD4+ T cells in culture and polarized them into Th2 cells producing IL-5 similar to that induced by antigen-loaded dendritic cells. These results demonstrate that eosinophils are capable of activating antigen-specific Th2 cells inducing the release of cytokines and assist in the priming of nave T cells to initiate Th2 responses against infection. This study highlights the potential of eosinophils to actively induce immune responses against infection by amplifying antigen-specific Th2-cell responses.
INTRODUCTION
Eosinophils have been shown to function as antigen-presenting cells (APC) in several experimental allergy model systems (24). Antigen-loaded eosinophils present antigen to primed T cells and increase Th2 cytokine production (12, 31). The eosinophils migrate into local lymph nodes and localize in the T-cell-rich paracortical zones, where they stimulate the expansion of CD4+ T cells. Antigen-exposed eosinophils were apparently activated, based on their increased expression levels of CD80, CD86, and major histocompatibility complex (MHC) class II molecules (5, 12, 25). The antigen-loaded eosinophils also promoted the production of interleukin-5 (IL-5) when placed in culture with antigen-specific CD4+ T cells isolated from allergic mice (12). There are conflicting reports, however, regarding the ability of eosinophils to act as APC to stimulate nave T cells. Eosinophils that were derived from the peritoneal cavities of IL-5-transgenic (TG) mice and pulsed with ovalbumin (OVA) in vitro were capable of sensitizing nave mice to OVA (12). In contrast to this, eosinophils purified from the bronchoalveolar lavage fluid of OVA-sensitized mice were unable to induce the proliferation of naive T cells in the draining lymph nodes of the lungs (28). Evidence that eosinophils are capable of initiating an immune response to parasite antigens is lacking, although it was demonstrated that granulocyte-macrophage colony-stimulating factor (GM-CSF)-activated eosinophils are capable of acting as specific antigen-presenting cells to a T-cell clone derived from mice infected with the cestode Mesocestoides corti (3). In addition, eosinophils recovered from mice infected with the nematode Brugia malayi express high levels of MHC class II molecules, suggesting the potential of these cells to act as APC (14).
Protective immunity to Strongyloides stercoralis infection in mice has been shown to be dependent on eosinophils during both the innate and the adaptive immunity responses (8, 21). IL-5–/– mice, which are incapable of augmenting blood and tissue eosinophil levels, have diminished larval killing during the innate immune response. Complementary studies using IL-5-TG mice, which overexpress IL-5 and develop a profound systemic eosinophilia, demonstrated elevated levels of larval killing in primary infections (8). Adaptive protective immunity to S. stercoralis infection in mice has been shown to be dependent on CD4+ T cells and in particular on the Th2 cytokines IL-4 and IL-5 (20). IL-5–/– mice failed to develop adaptive protective immunity to infection with S. stercoralis. However, the adoptive transfer of eosinophils into IL-5–/– mice at the time of immunization with live S. stercoralis larvae (L3) reconstituted the ability of IL-5–/– mice to develop adaptive protective immunity against the infection. The eosinophils did not have a direct role in killing the larvae but rather induced the production of protective antibody (Ab) that in conjunction with complement and other cells, eliminated the challenge infection (8). Therefore, eosinophils may have a dual function during the development of adaptive protective immunity to S. stercoralis infection in mice. The first function is to kill the larvae through an innate mechanism, as has been previously reported (8). This results in converting the comparatively large larvae into a form that can be phagocytized by APC. It is hypothesized in the present study that eosinophils represent the APC that initiate the adaptive protective immune response.
The present in vitro study utilized the S. stercoralis infection of mice as a model to investigate whether eosinophils have the potential to act as APC in the initiation and development of a protective Th2 response during helminth infections. Purified eosinophils were exposed to S. stercoralis antigens, and activated, antigen-pulsed eosinophils were analyzed for their potential to present antigens to T cells. Results show that eosinophils become activated after exposure to antigens, are capable of priming nave T cells, and can induce an antigen-specific Th2 response through MHC class II peptides.
MATERIALS AND METHODS
Experimental animals and parasites. BALB/cJ and C57BL/6J mice were purchased from The Jackson Laboratory (Bar Harbor, ME). The IL-5-transgenic mice (mouse line NJ.1638) were bred in the Laboratory of Animal Sciences facility at Thomas Jefferson University (Philadelphia, PA). All mice were housed in filter top microisolator boxes under light- and temperature-controlled conditions.
S. stercoralis L3 organisms were obtained from the fresh stools of a laboratory dog infected with the parasite according to methods previously described (1). Larvae were collected from charcoal cultures and washed by centrifugation and resuspension in a 1:1 mixture of Iscove modified Dulbecco medium (Sigma) and NCTC-135 (Sigma) with a mixture of 100 U of penicillin, 100 μg of streptomycin per ml (Gibco, Grand Island, NY), and 25 μg of levofloxacin per ml (Ortho-McNeil Pharmaceutical, Raritan, NJ).
Antigen preparation. Soluble larval antigens from S. stercoralis larvae were prepared as described previously (9). Briefly, L3 were washed in phosphate-buffered saline (PBS) supplemented with 100 U of penicillin and 100 μg of streptomycin per ml and stored at –80°C. L3 were thawed and homogenized in the presence of a protease inhibitor cocktail (Sigma, St. Louis, MO) and then sonicated. The homogenized and sonicated L3 were incubated in PBS at 4°C for 18 h with continuous mixing. PBS-soluble antigens were removed, filter sterilized, and stored at –80°C. PBS-insoluble proteins were resuspended in 20 mM Tris-HCl-0.5% deoxycholic acid (DOC; Sigma) with continuous mixing for 12 h at 4°C. The DOC-soluble antigens were then dialyzed (6- to 8-kDa cutoff) against PBS for 12 h, concentrated, filter sterilized, and stored at –80°C. Protein concentration was quantitated by a Micro BCA protein assay kit (Pierce, Rockford, IL). Lipopolysaccharide (LPS) content in the soluble antigen preparations was measured using a Limulus amebocyte lysate assay; LPS was detected at about 1 ng of LPS/mg of S. stercoralis antigens. As no difference was observed in the ability of these antigens to induce T cells, either DOC-soluble or PBS-soluble antigens were used in the present study.
Immunization and isolation of T cells. C57BL/6J mice were immunized with live S. stercoralis L3 as described previously (1). The immunization protocol consisted of two subcutaneous injections with 5,000 live L3 administered 2 weeks apart. Four weeks after the second injection, mice were sacrificed and primed T cells and CD4+ T cells were isolated from the spleens using magnetic-activated cell-sorting columns (Miltenyi Biotech, Auburn, CA). T cells and CD4+ T cells were purified by positive selection with Ab-conjugated magnetic beads specific for CD90 (Thy 1.2) and CD4 (L3T4), respectively. Nave T cells and CD4+ T cells were obtained from the spleens of C57BL/6 mice.
Isolation of eosinophils. Eosinophils were isolated from the spleens of nave IL-5-TG mice, following the method previously described for isolating them from blood (2). In brief, single-cell suspensions of spleens were layered onto a Percoll gradient (60% Percoll [ = 1.084] 1x Hanks balanced salt solution, 15 mM HEPES [pH 7.4], and 0.003 N HCl) and centrifuged (45 min, 1,200 x g, 4°C). The buffy coat was recovered and washed twice in PBS containing 2% fetal bovine serum (FBS). Eosinophils were isolated using magnetic-activated cell-sorting columns. Specifically, B cells and T cells were removed by positive selection with Ab-conjugated magnetic beads specific for CD45-R (B220) B cells and CD90 (Thy 1.2) T cells. Cells were placed in a Cytospin 3 apparatus (Shandon, Pittsburgh, PA) and stained for differential counts with a Hema 3 stain set (Fisher Diagnostics, Middletown, VA) to check for purity. Approximately 95% of the cells were eosinophils, along with neutrophils (1 to 2%), macrophages (2 to 4%), and lymphocytes (1 to 2%) as contaminating cells.
Activation of eosinophils. For activation of eosinophils, cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated and filtered FBS (HyClone, Logan, UT), 2 mM L-glutamine (GIBCO), 100 U/ml penicillin plus 100 μg/ml streptomycin (Gibco), and 50 μM 2-mercaptoethanol (2-ME; Sigma) either in the presence or in the absence of 5 ng/ml GM-CSF (Peprotech, Rocky Hill, NJ) and pulsed with 100 μg/ml S. stercoralis antigens or 1 μg/ml LPS (Sigma) for 4 h. To determine the activation state of the eosinophils, cells were harvested, washed, and resuspended in fluorescence-activated cell sorter buffer (1x phosphate-buffered saline, 0.2% bovine serum albumin fraction V, 4 mM sodium azide) and the Fc receptors on cells were blocked using Fc-Block CD16/CD32 Ab (BD Pharmingen, San Jose, CA) for 20 min on ice. The expression of various surface molecules was quantified by immunofluorescent staining using monoclonal antibodies (MAbs) specific for I-Ab, CD69, CD62L, CD29 (integrin-1), CD86, CD80, CD11c (all purchased from BD Pharmingen), and CCR3 (R & D system, Minneapolis, MN), conjugated to either fluorescein isothiocyanate, phycoerythrin, or allophycocyanin. Samples were analyzed using a FACSCalibur flow cytometer and CellQuest software (Becton Dickinson, Franklin Lakes, NJ).
Generation of dendritic cells and macrophages. Dendritic cells (DC) and macrophages were generated from bone marrow by collecting bone marrow from femurs and tibias from mice. Dendritic cells were generated as previously described (11). In brief, cells were seeded into petri dishes at 2 x 105/ml in 10 ml RPMI 1640 supplemented with 10% heat-inactivated and filtered FBS, 2 mM L-glutamine, 100 U/ml penicillin plus 100 μg/ml streptomycin, and 50 μM 2-ME, with the addition of 20 ng/ml GM-CSF. On day 3, a further 10 ml of medium containing 20 ng/ml GM-CSF was added. On day 6, 10 ml culture supernatant was removed and replaced with 10 ml fresh culture medium containing 20 ng/ml GM-CSF. For generation of immature DC, on day 8 plates were fed as on day 6, but only 5 ng/ml GM-CSF was added in fresh media, and cells were harvested 18 h later (day 9). For activation of DC, cells were treated in the same way, with the addition of the S. stercoralis antigens (100 μg/ml) for the final 18-h incubation. To generate macrophages from bone marrow cells, cells were cultured in Dulbecco modified Eagle medium (Sigma) supplemented with 20% heat-inactivated and filtered fetal calf serum (HyClone), 2 mM L-glutamine, 100 U/ml penicillin plus 100 μg/ml streptomycin, and 50 μM 2-ME, with the addition of 30% L929 cell-conditioned medium as a source of macrophage colony-stimulating factor (26), at 37°C in a 5% CO2 atmosphere as previously described (4), with minor modifications. At day 3, culture supernatants were removed and replaced with fresh medium. Cells were harvested on day 6 using ice-cold PBS (pH 7.2). For activation of macrophages, cells were treated with the addition of the S. stercoralis antigens (100 μg/ml) and cultured for the final 18 h. Cells were fixed with 4% paraformaldehyde and used for in vitro assays.
Determination of the ability of eosinophils to stimulate T cells. Eosinophils and DC were pulsed with S. stercoralis antigens (100 μg/ml) for 18 h in RPMI medium containing GM-CSF (5 ng/ml) and fixed in paraformaldehyde (4%) prior to coculture with T cells. Purified, primed T cells or primed CD4+ T cells (2.5 x 106) were cultured with either unpulsed or antigen-pulsed S. stercoralis eosinophils (5 x 105) in Dulbecco modified Eagle medium (Sigma) supplemented with 10% heat-inactivated and filtered fetal calf serum (HyClone), 2 mM L-glutamine (Life Technologies), 100 U/ml penicillin plus 100 μg/ml streptomycin (Life Technologies), and 50 μM 2-ME (Sigma). Cultures were also set up with antigen-pulsed or unpulsed S. stercoralis DC (5 x 105) to compare the potentials of eosinophils to stimulate T cells. After 96 h, supernatants were collected from the culture and analyzed for IL-5, IL-4, and gamma interferon (IFN-) production. To determine whether eosinophils can prime nave cells, purified nave T cells or CD4+ T cells were cultured with antigen-pulsed eosinophils or DC. After 48 h of culture, recombinant interleukin-2 was added at 20 ng/ml and cells were then rested for 72 to 96 h, before being restimulated with anti-CD3 Ab (BD Pharmingen) (0.5 μg/well) for 72 h. The supernatants harvested from these cultures were analyzed for IL-5 and IL-4 production by sandwich enzyme-linked immunosorbent assays (ELISAs) using paired MAbs (TRFK.5 and TRFK.4 for detection of IL-5 and BVD6-24G2 and BVD4-1D11 for detection of IL-4; BD Pharmingen). IFN- production was measured using an ELISA kit (BD Pharmingen) as directed.
MHC class II restriction of antigen presentation by eosinophils. To determine whether antigen presentation by eosinophils was through MHC class II peptides, antigen-pulsed eosinophils or unpulsed eosinophils were fixed and incubated with monoclonal antibodies specific to MHC class II molecules (I-Ab; BD Pharmingen) for 30 min at 37°C to block the presentation of MHC class II peptides, as described previously (3). Eosinophils were washed twice with PBS (pH 7.2) to remove unbound antibodies and then cultured with primed CD4+ T cells. In some studies, paraformaldehyde-fixed control eosinophils (5 x 105) and eosinophils pulsed with antigen, obtained from IL-5-TG mice of the C57BL/6J background (H-2b haplotype), were cultured with 2.5 x 106 primed CD4+ T cells isolated from either immunized BALB/cJ (H-2d haplotype) or C57BL/6J (H-2b haplotype) mice. Alternatively, antigen-pulsed and fixed eosinophils were mixed with H-2d haplotype dendritic cells (2.5 x 104; 5% of the eosinophil population) or macrophages (2.5 x 104; 5% of the eosinophil population) derived from the bone marrow of nave BALB/cJ mice, and the mixture was cultured with 2.5 x 106 primed CD4+ T cells from immunized BALB/cJ mice. After 96 h, supernatants from these in-vitro cultures were analyzed for IL-5 production by ELISA.
Statistical analysis. Statistical analysis of the data was performed using MGLH multifactorial analysis of variance with Systat version 11 software (Systat, Evanston, IL). Fisher's least-significant-difference test was performed for post hoc analyses. Probability values of less than 0.05 were considered significant.
RESULTS
Activation of Ag-pulsed eosinophils. Purified eosinophils were exposed to S. stercoralis antigens in the presence of GM-CSF to determine whether the antigens would activate the eosinophils. Antigen-pulsed eosinophils in the presence of GM-CSF showed (i) a sixfold increase in expression levels of CD69 and MHC class II (Fig. 1A, panel III), (ii) a fourfold increase in expression levels of CD86 (Fig. 1B, panel III), and (iii) a twofold decrease in expression levels of CD62L (Fig. 1C, panel III) compared to cells incubated only in medium or in medium containing GM-CSF. No differences were seen in levels of CD29 (Fig. 1C, panel III) and CD80 expression (data not shown). To determine whether the S. stercoralis antigens require GM-CSF for activation of eosinophils, purified eosinophils were exposed to antigen in the absence of GM-CSF and analyzed for the expression patterns of the surface proteins. Antigen-pulsed eosinophils in the absence of GM-CSF showed a fourfold increase in expression levels of CD69 and MHC class II (Fig. 1A, panel IV) and a fivefold increase in expression levels of CD86 (Fig. 1B, panel IV). There was also a twofold increase in expression levels of CD29 and a threefold reduction in those of CD62L (Fig. 1C, panel IV). Eosinophils incubated with LPS (1 μg/ml) did not show any significant change in surface levels of CD69, MHC class II, CD80, and CD86 (data not shown). Overall, the increase in expression levels of CD69, MHC class II, CD86, and CD29 and the decrease in CD62L expression levels on the surfaces of eosinophils were observed on eosinophils exposed to S. stercoralis antigens regardless of the presence of GM-CSF, thereby demonstrating that S. stercoralis antigens could directly activate eosinophils.
Antigen-pulsed eosinophils stimulate primed T cells to produce IL-5. Eosinophils pulsed with antigen in the presence of GM-CSF were fixed in paraformaldehyde and then cultured with T cells isolated from mice immunized with S. stercoralis. There was a significant increase in IL-5 production by primed T cells cultured with antigen-pulsed eosinophils compared to production levels in T cells cultured with untreated eosinophils (Fig. 2A). Primed CD4+ T cells isolated from immunized mice and placed in culture with fixed, antigen-pulsed eosinophils also showed a significantly increased production of IL-5 (Fig. 2B). Experiments were performed to determine whether the induction of IL-5 production by primed T cells was due to the presentation of antigen by eosinophils in conjunction with MHC class II molecules. Eosinophils pulsed with antigen were treated with anti-MHC class II MAb prior to incubation with primed CD4+ T cells. The blocking of MHC class II molecules on the surfaces of eosinophils, through treatment with the anti-MHC class II MAb, eliminated the ability of antigen-pulsed eosinophils to stimulate IL-5 production by primed CD4+ T cells (Fig. 2C).
Possible contaminating dendritic cells and macrophages within eosinophil preparation are not responsible for antigen presentation to T cells. The purified eosinophils, used in the experiments described above, contained approximately 5% contaminating cells consisting of a mixture of lymphocytes, neutrophils, dendritic cells, and macrophages. Experiments were performed to determine whether the functional APC in the purified eosinophil population were eosinophils or whether they were the contaminating dendritic cells and macrophages. In these studies, eosinophils were purified from C57BL/6J (H-2b haplotype) mice, whereas dendritic cells and macrophages were recovered from BALB/cJ (H-2d haplotype) mice. Antigen-pulsed and fixed eosinophils from C57BL/6 mice induced IL-5 production from primed C57BL/6J mouse-derived CD4+ T cells but not from cells isolated from immunized BALB/c mice, because of the differences in MHC class II molecules (Fig. 3). Cultures were then set up with BALB/c mouse-derived CD4+ T cells and antigen-pulsed eosinophils from C57BL/6J mice. To these cultures were added antigen-pulsed dendritic cells or macrophages from BALB/c mice in numbers equal to the 5% contamination of the eosinophils population. IL-5 production by BALB/c mouse-derived CD4+ T cells did not increase by the addition of the pulsed H-2d haplotype dendritic cells or macrophages (Fig. 3). These experiments therefore demonstrate that the induction of IL-5 production by primed CD4+ T cells was due to the presentation of antigen by eosinophils and not by other cells contaminating the eosinophils population.
Antigen-pulsed eosinophils prime nave T cells to produce IL-4 and IL-5. T cells and CD4+ T cells, isolated from the spleens of nave C57BL/6 mice, were cultured with fixed eosinophils pulsed with S. stercoralis antigen in the presence of GM-CSF. The levels of IL-5 produced by nave T cells (Fig. 4A) and CD4+ T cells (Fig. 4B) were significantly increased after culture with antigen-pulsed eosinophils. Experiments were performed to compare the antigen-presenting potential of eosinophils and dendritic cells to induce primary and secondary Th1 and Th2 responses to S. stercoralis antigen in T cells. Both eosinophils and dendritic cells did not induce the production of IFN- from nave and primed T cells. However, dendritic cells were superior to eosinophils in their ability to induce IL-5 (Fig. 5A) and IL-4 (Fig. 5C) production by primed T cells. Eosinophils and dendritic cells showed similar potentials to prime nave T cells to produce IL-5 (Fig. 5B) and IL-4 (Fig. 5D).
DISCUSSION
The objectives of these studies were to determine whether eosinophils become activated after exposure to helminth antigens and whether they were potentially APC for initiation and restimulation of immune responses to helminth infections. The exposure of nave eosinophils to S. stercoralis antigens and GM-CSF increased expression levels of CD69, which serves as an activation marker for both human and mouse eosinophils (18, 22). Moreover, expression levels of MHC class II peptides and the T-cell costimulatory molecule CD86 were also elevated on eosinophils exposed to both antigens and GM-CSF but not on eosinophils exposed to only GM-CSF. Studies of eosinophil activation in mice infected with Nippostrongylus brasiliensis have shown increased CD29 and decreased CD62L levels on eosinophils found within inflammatory tissues compared to those on eosinophils from the blood (29). CD29 is a member of the very-late-activation integrin family and known to function as an adhesion and signaling molecule. CD62L (L-selectin) plays a critical role in initiating the interactions of cells with the activated endothelium at an inflammatory site and is characteristically down-modulated in response to cell activation (6, 16, 19). Increased CD29 and decreased CD62L levels were also observed in the present study after eosinophils were exposed to S. stercoralis antigens. Soluble S. stercoralis antigens in the absence of GM-CSF were also capable of activating mouse eosinophils. Helminths have been shown to produce homologues of several mammalian cytokines (13), and it is possible that the soluble extract of S. stercoralis antigens contains homologues of mammalian cytokines that may facilitate the direct activation of eosinophils.
In vitro experiments were performed to investigate whether activated eosinophils could act as APC and stimulate T cells to produce IL-5. Before culture with T cells or CD4+ T cells, the eosinophils were fixed to avoid degranulation, block alloreactivity, and preserve eosinophils during subsequent culture (3, 30). Fixed, antigen-pulsed APC have been shown to have unchanged expression levels of MHC class II molecules and to be competent for stimulating the proliferation of antigen-reactive T cells (17). GM-CSF was used during in vitro pulsing of eosinophils with antigen, as it stimulates the functional activity of eosinophils and also maintains maximum viability of cells (27). Purified T cells and CD4+ T cells isolated from immunized mice and placed in culture with antigen-pulsed eosinophils were stimulated to produce IL-5. A similar observation has been made in an experimental mouse model for allergic diseases (12). Furthermore, the blocking of MHC class II peptides completely inhibited the stimulation of primed CD4+ T cells, thereby demonstrating that stimulation of CD4+ T cells by antigen-pulsed eosinophils is MHC class II dependent, as has previously been reported (3, 27).
It was essential, however, to prove that T-cell stimulation and IL-5 production in culture were induced by the antigen-pulsed eosinophils and not by the contaminating cells (5%) in the eosinophil preparation. Experiments were performed to capitalize on the genetic difference between BALB/cJ and C57BL/6J mice in their MHC class II molecules. The actual number of cells that comprised the 5% contamination of the eosinophil population was determined (2.5 x 104), and that quantity of pure dendritic cells or macrophages derived from the bone marrow of BALB/c mice was mixed with eosinophils isolated from IL-5-TG mice of the C57BL/6J background. The data from these experiments demonstrated that the few contaminating dendritic cells and macrophages in the eosinophil preparation could not have accounted for the observed T-cell stimulation and IL-5 production. This conclusion is in agreement with previous studies that have demonstrated that using fewer than 2 x 104 APC in culture failed to stimulate the proliferation of T cells (7, 27). It was therefore concluded that the antigen-pulsed eosinophils presented antigen to the primed T cells, which were then induced to produce IL-5.
There have been conflicting reports on the capability of eosinophils to present antigen to nave T cells. Eosinophils isolated from the bronchoalveolar lavage fluid of OVA-sensitized and -challenged mice fail to induce the division of OVA-T-cell-receptor-transgenic T cells, whereas dendritic cells induced T-cell divisions (28). This conclusion differs from the results of another study that demonstrated that OVA-pulsed eosinophils are capable of sensitizing mice when injected repeatedly into the peritoneal cavity (12). As eosinophils are associated with helminth parasites at the initial stage of infection (8, 10, 15, 23), it is possible that these cells capture antigens from the helminths, migrate to T-cell-rich regions, and present antigens to T cells to initiate antigen-specific T-cell responses. Therefore, the present study also examined the ability of antigen-exposed eosinophils to prime the nave T cells. IL-5 production from nave T cells or CD4+ T cells cultured with antigen-loaded eosinophils was significantly higher than that from cells cultured with untreated eosinophils. These data demonstrated that activated eosinophils are capable of functioning as a professional APC with the potential to prime nave T cells and initiate the development of an antigen-specific immune response against infection. More importantly, the data demonstrate that the antigen-presenting capacity of an eosinophil to induce a Th2 response was equivalent to that of a traditional APC, such as a dendritic cell in a primary response. Antigen-pulsed dendritic cells were, however, superior to eosinophils in their ability to induce IL-4 and IL-5 production from primed T cells.
The results of the present report therefore support the study hypothesis that eosinophils, which come in direct contact with parasites during the innate immune response, may capture parasitic antigens and present them to T cells in the induction of the adaptive immune response to S. stercoralis infection in mice. Furthermore, the observation that eosinophils and dendritic cells function equally well at inducing the primary response to the infection suggests that eosinophils may be essential as the APC for the induction of adaptive immunity. This conclusion is supported by previous data which showed that the immunization of IL-5–/– mice resulted in diminished antibody responses, even though there were abundant dendritic cells in these mice. IL-5–/– mice reconstituted with eosinophils at the time of immunization developed sustained antibody responses equal to those in wild-type mice, suggesting that eosinophils were required only at the initiation of the response, which would be consistent with a role for the eosinophils as APC (8).
In summary, the present study demonstrated that S. stercoralis antigens activate eosinophils and induce the expression of MHC class II and T-cell costimulatory molecules and that these activated eosinophils can stimulate T cells for antigen-specific immune responses. In addition, this study demonstrated that eosinophils have the ability to prime nave T cells and assist the immune system in the initiation and development of antigen-specific T-cell responses against S. stercoralis. These results suggest that in addition to their role as terminal effector cells in helminth infections, eosinophils may also serve as specific antigen-presenting cells and prime the immune response during infection. Significantly, these data also demonstrate that eosinophils can communicate with T cells and suggest that eosinophils are integral in the interface between the innate and adaptive immune responses.
ACKNOWLEDGMENTS
This work was supported by NIH grants RO1 AI47189 and RO1 A1 22662 and the Mayo Foundation.
We thank Jessica Hess and Amy O'Connell for critical reading of the manuscript and Juergen Landmann for expert technical assistance.
REFERENCES
1. Abraham, D., H. L. Rotman, H. F. Haberstroh, W. Yutanawiboonchai, R. A. Brigandi, O. Leon, T. J. Nolan, and G. A. Schad. 1995. Strongyloides stercoralis: protective immunity to third-stage larvae in BALB/cByJ mice. Exp. Parasitol. 80:297-307.
2. Borchers, M. T., P. J. Justice, T. Ansay, V. Mancino, M. P. McGarry, J. Crosby, M. I. Simon, N. A. Lee, and J. J. Lee. 2002. Gq signaling is required for allergen-induced pulmonary eosinophilia. J. Immunol. 168:3543-3549.
3. Del Pozo, V., B. De Andres, E. Martin, B. Cardaba, J. C. Fernandez, S. Gallardo, P. Tramon, F. Leyva-Cobian, P. Palomino, and C. Lahoz. 1992. Eosinophil as antigen-presenting cell: activation of T cell clones and T cell hybridoma by eosinophils after antigen processing. Eur. J. Immunol. 22:1919-1925.
4. Descoteaux, A., G. Matlashewski, and S. J. Turco. 1992. Inhibition of macrophage protein kinase C-mediated protein phosphorylation by Leishmania donovani lipophosphoglycan. J. Immunol. 149:3008-3015.
5. Duez, C., A. Dakhama, A. Tomkinson, P. Marquillies, A. Balhorn, A. B. Tonnel, D. L. Bratton, and E. W. Gelfand. 2004. Migration and accumulation of eosinophils toward regional lymph nodes after airway allergen challenge. J. Allergy Clin. Immunol. 114:820-825.
6. Elangbam, C. S., C. W. Qualls, Jr., and R. R. Dahlgren. 1997. Cell adhesion molecules—update. Vet. Pathol. 34:61-73.
7. Handzel, Z. T., W. W. Busse, J. B. Sedgwick, R. Vrtis, W. M. Lee, E. A. Kelly, and J. E. Gern. 1998. Eosinophils bind rhinovirus and activate virus-specific T cells. J. Immunol. 160:1279-1284.
8. Herbert, D. R., J. J. Lee, N. A. Lee, T. J. Nolan, G. A. Schad, and D. Abraham. 2000. Role of IL-5 in innate and adaptive immunity to larval Strongyloides stercoralis in mice. J. Immunol. 165:4544-4551.
9. Herbert, D. R., T. J. Nolan, G. A. Schad, S. Lustigman, and D. Abraham. 2002. Immunoaffinity-isolated antigens induce protective immunity against larval Strongyloides stercoralis in mice. Exp. Parasitol. 100:112-120.
10. Korenaga, M., Y. Hitoshi, K. Takatsu, and I. Tada. 1994. Regulatory effect of anti-interleukin-5 monoclonal antibody on intestinal worm burden in a primary infection with Strongyloides venezuelensis in mice. Int. J. Parasitol. 24:951-957.
11. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and G. Schuler. 1999. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223:77-92.
12. MacKenzie, J. R., J. Mattes, L. A. Dent, and P. S. Foster. 2001. Eosinophils promote allergic disease of the lung by regulating CD4(+) Th2 lymphocyte function. J. Immunol. 167:3146-3155.
13. Maizels, R. M., A. Balic, N. Gomez-Escobar, M. Nair, M. D. Taylor, and J. E. Allen. 2004. Helminth parasites—masters of regulation. Immunol. Rev. 201:89-116.
14. Mawhorter, S. D., E. Pearlman, J. W. Kazura, and W. H. Boom. 1993. Class II major histocompatibility complex molecule expression on murine eosinophils activated in vivo by Brugia malayi. Infect. Immun. 61:5410-5412.
15. Meeusen, E. N., and A. Balic. 2000. Do eosinophils have a role in the killing of helminth parasites Parasitol. Today 16:95-101.
16. Mengelers, H. J., T. Maikoe, B. Hooibrink, T. W. Kuypers, J. Kreukniet, J. W. Lammers, and L. Koenderman. 1993. Down modulation of L-Selectin expression on eosinophils recovered from bronchoalveolar lavage fluid after allergen provocation. Clin. Exp. Allergy 23:196-204.
17. Moreno, J., and P. E. Lipsky. 1986. Functional heterogeneity of human antigen-presenting cells: presentation of soluble antigen but not self-Ia by monocytes. J. Clin. Immunol. 6:9-20.
18. Nishikawa, K., T. Morii, H. Ako, K. Hamada, S. Saito, and N. Narita. 1992. In vivo expression of CD69 on lung eosinophils in eosinophilic pneumonia: CD69 as a possible activation marker for eosinophils. J. Allergy Clin. Immunol. 90:169-174.
19. Reyes-Reyes, M., N. Mora, G. Gonzalez, and C. Rosales. 2002. beta1 and beta2 integrins activate different signalling pathways in monocytes. Biochem. J. 363:273-280.
20. Rotman, H. L., S. Schnyder-Candrian, P. Scott, T. J. Nolan, G. A. Schad, and D. Abraham. 1997. IL-12 eliminates the Th-2 dependent protective immune response of mice to larval Strongyloides stercoralis. Parasite Immunol. 19:29-39.
21. Rotman, H. L., W. Yutanawiboonchai, R. A. Brigandi, O. Leon, G. J. Gleich, T. J. Nolan, G. A. Schad, and D. Abraham. 1996. Strongyloides stercoralis: eosinophil-dependent immune-mediated killing of third stage larvae in BALB/cByJ mice. Exp. Parasitol. 82:267-278.
22. Rumbley, C. A., and S. M. Phillips. 1999. The schistosome granuloma: an immunoregulatory organelle. Microbes Infect. 1:499-504.
23. Sasaki, O., H. Sugaya, K. Ishida, and K. Yoshimura. 1993. Ablation of eosinophils with anti-IL-5 antibody enhances the survival of intracranial worms of Angiostrongylus cantonensis in the mouse. Parasite Immunol. 15:349-354.
24. Shi, H. Z. 2004. Eosinophils function as antigen-presenting cells. J. Leukoc. Biol. 76:520-527.
25. Shi, H. Z., A. Humbles, C. Gerard, Z. Jin, and P. F. Weller. 2000. Lymph node trafficking and antigen presentation by endobronchial eosinophils. J. Clin. Investig. 105:945-953.
26. Stanley, E. R., and P. M. Heard. 1977. Factors regulating macrophage production and growth. Purification and some properties of the colony stimulating factor from medium conditioned by mouse L cells. J. Biol. Chem. 252:4305-4312.
27. Tamura, N., N. Ishii, M. Nakazawa, M. Nagoya, M. Yoshinari, T. Amano, H. Nakazima, and M. Minami. 1996. Requirement of CD80 and CD86 molecules for antigen presentation by eosinophils. Scand. J. Immunol. 44:229-238.
28. van Rijt, L. S., N. Vos, D. Hijdra, V. C. de Vries, H. C. Hoogsteden, and B. N. Lambrecht. 2003. Airway eosinophils accumulate in the mediastinal lymph nodes but lack antigen-presenting potential for naive T cells. J. Immunol. 171:3372-3378.
29. Voehringer, D., K. Shinkai, and R. M. Locksley. 2004. Type 2 immunity reflects orchestrated recruitment of cells committed to IL-4 production. Immunity 20:267-277.
30. Weller, P. F., T. H. Rand, T. Barrett, A. Elovic, D. T. Wong, and R. W. Finberg. 1993. Accessory cell function of human eosinophils. HLA-DR-dependent, MHC-restricted antigen-presentation and IL-1 alpha expression. J. Immunol. 150:2554-2562.
31. Xie, Z. F., H. Z. Shi, X. J. Qin, L. F. Kang, C. P. Huang, and Y. Q. Chen. 2005. Effects of antigen presentation of eosinophils on lung Th1/Th2 imbalance. Chin. Med. J. (Engl. Ed.) 118:6-11.(Udaikumar M. Padigel, Jam)