Helminth 2-Cys peroxiredoxin drives Th2 responses through a mechanism involving alternatively activated macrophages

Abstract

During helminth infections, alternatively activated macrophages (AAMacs) are key to promoting Th2 responses and suppressing Th1-driven inflammatory pathology. Th2 cytokines IL-4 and/or IL-13 are believed to be important in the induction and activation of AAMacs. Using murine models for the helminth infections caused by Fasciola hepatica (Fh) and Schistosoma mansoni (Sm), we show that a secreted antioxidant, peroxiredoxin (Prx), induces alternative activation of macrophages. These activated, Ym1-expressing macrophages enhanced the secretion of IL-4, IL-5, and IL-13 from naive CD4⁺ T cells. Administration of recombinant FhPrx and SmPrx to wild-type and IL-4^−/− and IL-13^−/− mice induced the production of AAMacs. In addition, Prx stimulated the expression of markers of AAMacs (particularly, Ym1) in vitro, and therefore can act independently of IL-4/IL-13 signaling. The immunomodulatory property of Prx is not due to its antioxidant activity, as an inactive recombinant variant with active site Cys residues replaced by Gly could also induce AAMacs and Th2 responses. Immunization of mice with recombinant Prx or passive transfer of anti-Prx antibodies prior to infection with Fh not only blocked the induction of AAMacs but also the development of parasite-specific Th2 responses. We propose that Prx activates macrophages as an initial step in the induction of Th2 responses by helminth parasites and is thereby a novel pathogen-associated molecular pattern.—Donnelly, S., Stack, C. M., O'Neill, S. M., Sayed, A. A., Williams, D. L., Dalton, J. P. Helminth 2-Cys peroxiredoxin drives Th2 responses through a mechanism involving alternatively activated macrophages.

Alternatively activated macrophages (AAMacs) are found in all Th2 cytokine environments regardless of whether activation is induced by parasitic, asthmatic, or tumor-associated inflammation (1,2,3,4,5). AAMacs have been classified as a phenotype of macrophage induced by IL-4 and/or IL-13 (6). Several markers have been identified as characteristic of this IL-4/IL-13 activated phenotype, such as Arg-1, Mannose receptor, Fizz1, and Ym1 (7,8,9).

During helminth infections, AAMacs not only play key roles in the enhancement of Th2 cell differentiation (10, 11) but are also involved in the suppression of Th1 inflammatory responses (12, 13). Ym1 is the most abundantly expressed gene in helminth parasite-activated macrophages, encoded by more than 10% of gene transcripts of nematode-activated macrophages (9) and expressed over 350-fold more than any other marker during infection with Taenia crassiceps(14). Ym1 is a member of a family of mammalian proteins that are highly homologous to chitinases of lower organisms (15); although lacking chitinase activity, it possesses heparin binding characteristics (16) and can promote the differentiation of Th2 cells (4, 17). Recent studies have shown that the expression of Ym1 induced by helminth infection can occur independently of IL-4 and IL-13 signaling, suggesting that its induction is critical to the development of Th2 polarization (14, 18, 19).

The liver (Fasciola hepatica) and blood (Schistosoma spp.) flukes are important global helminth parasites of humans and their livestock (20). Successful vaccination against infection with the liver fluke F. hepatica is dependent on preventing the development of Th2 responses and the induction of Th1 responses (21). In the case of the blood fluke Schistosoma mansoni, Th2 immune responses are instrumental in the survival of the parasite (22), whereas worm rejection, for example, induced by exposure to radiation-attenuated cercariae is associated with Th1 responses (23). The polarization of the immune response toward a favorable Th2 phenotype is, therefore, critical to both parasites.

F. hepatica induces the expression of Ym1-expressing AAMacs as early as day 4 after infection (24, and unpublished data) and polarized antigen-specific Th2 responses are detected in splenocytes within 7 days of infection (24, 25). During infection with S. mansoni, a weak Th1 immune response develops in the first 5 to 6 wk of infection and shifts to an intense Th2 cytokine environment when eggs produced by fecund females become trapped in the liver and intestinal tissues (26). S. mansoni Th2-driven inflammatory responses develop toward egg antigens and their secretions (27) concurrent with the appearance of AAMacs (12).

We previously demonstrated that molecules secreted by the helminth parasite F. hepatica can induce the recruitment and alternative activation of macrophages when administered to mice (24). Fractionation studies indicated that the molecule peroxiredoxin (Prx), an antioxidant enzyme, was one of the key players in this function (24). Here we show that soluble extracts of F. hepatica and S. mansoni containing Prx induced the expression of Ym1 in macrophages. In addition, purified recombinant F. hepatica and S. mansoni Prx stimulated the expression of Ym1 in macrophages in vivo independently of IL-4 and IL-13 signaling, and these macrophages enhanced the production of IL-4, IL-5, and IL-13 by naive T cells in culture. Antibody-mediated neutralization of parasite-secreted Prx during infection of F. hepatica blocked the expression of Ym1 in peritoneal macrophages and significantly reduced the development of Th2 responses. We therefore propose that helminth Prx induces Ym1 production in macrophages as a mechanism of polarizing T-cell responses and is therefore a novel pathogen-associated molecular pattern.

MATERIALS AND METHODS

Parasites and recombinant proteins

Soluble extracts of whole adult liver fluke (SFh) and excretory/secretory (ES) products were prepared as described elsewhere (31). Soluble egg antigen (SEA) and extracts of whole adult Schistosoma parasites (SWAP) were purchased from the Schistosome Biological Supply Program (Theodor Bilharz Research Institute, Giza, Egypt).

The gene encoding peroxiredoxin derived from F. hepatica (rFhPrx) was synthesized with a SnaBI restriction site incorporated at the 5′ end of the gene and an AvrII restriction site and His₆-tag sequence at the 3′ end (Geneart, Regensburg, Germany). Similarly, the gene encoding mouse peroxiredoxin 2 (rMoPrx2) was synthesized with KpnI and HindIII restriction sites and with a His₆-tag sequence at the 3′ end. rFhPrx, SmPrx1, SmPrx2, and rMoPrx2 were expressed in Escherichia coli and purified by nickel-chelate affinity chromatography as described previously (28,29,30). Correct folding of recombinants was assessed by measuring their antioxidant activity and by assessing their migration in SDS-PAGE under reducing and nonreducing conditions, as described below. A functionally inactive variant recombinant FhPrx (rvFhPrx) was prepared by synthesizing a gene with the 2 reactive Cys⁴⁷ and Cys¹⁷⁰ residues replaced by Gly residues (28, 29). M1 aminopeptidase of Plasmodium falciparum used in this study as a recombinant control was expressed using a similar method. Recombinant F. hepatica cathepsin L1 (FhCL1) expressed in the yeast Pichia pastoris was also used as a control (31). Residual bacterial endotoxin was removed from purified recombinant proteins by phase separation using Triton X-114 (32) and determined to be lipopolysaccharide (LPS) free by a commercial assay (PTS; Charles River Laboratories, Wilmington, MA, USA).

Assessment of antioxidant activity of recombinant Prx

The specific activity of rFhPrx and rvFhPrx was determined by the reduction of H₂O₂ in a reaction containing E. coli thioredoxin and thioredoxin reductase (Sigma–Aldrich, St. Louis, MO, USA), as described elsewhere (28). Consumption of NADPH at 22°C was monitored at A₃₄₀ for 1 min in a spectrophotometer after adding hydrogen peroxide. The specific activity of the FhPrx was determined by the following equation: units of Prx/mg of protein = {[(DA/DT·Vt)/(6.22·Vs)]·1000}/[protein] (mg/ml), where 6.22 is the extinction coefficient of NADPH (mM⁻¹ cm⁻¹), Vt is the total reaction volume, Vs is the sample volume, and 1000 is used to convert to nanomoles.

SDS-PAGE and Western blot analysis

Protein samples (25 μg) were mixed 3:1 with sample buffer [125 mM Tris-HCl, 20% glycerol (v/v), 4% SDS, and 0.01% bromphenol blue] with or without the reducing agent dithiothreitol (DTT, 40 mM) and analyzed by 12.5% SDS-PAGE gel. The proteins were transferred to a nitrocellulose membrane (Schleicher & Schuell BioScience Inc., Keene, NH, USA), which was then blocked with 5% nonfat dry milk in PBS Tween (0.05% v/v). The membrane was incubated with a 1:1000 dilution of mouse antiserum raised against recombinant FhPrx and, subsequently, with alkaline phosphatase-conjugated goat anti-mouse immunoglobulin G (IgG; Sigma-Aldrich) for 1 h. Bound antibody was visualized by the addition of 3,3′-diaminobenzidine peroxidase substrate (28).

Animal infection and treatment with recombinant Prx

Six- to 8-wk-old female BALB/c mice (Australian Resource Centre, Perth, WA, Australia), IL-4^−/− and IL-13^−/− (BALB/c background; Australian National University, Canberra, ACT, Australia) were maintained according to the guidelines of the UTS/Royal North Shore Hospital Animal Care and Ethics Committee. Mice were orally infected with 10 metacercariae of F. hepatica (Elizabeth Macarthur Agricultural Institute, Camden, NSW, Australia), which produced infection in 100% of animals. After the experimental period of 7 days, mice were necropsied, and peritoneal exudate cells (PECs) were harvested by thorough washing of the peritoneal cavity with 10 ml of sterile PBS.

To demonstrate the role of helminth molecules in immunomodulation, 5 μg of parasite extracts or purified recombinant Prx was injected intraperitoneally 6 or 9 times, delivered on alternate days, and PECs were harvested as described above 2 days after the final injection. As a negative control for recombinant Prx activation, LPS E. coli serotype 111:B4 (Sigma-Aldrich) was delivered in the same manner at a concentration of 26 EU/ml, the same quantity of LPS present in a preparation of recombinant protein prior to phase separation.

Antibody-mediated neutralization of FhPrx

In the first experiments, mice were given 3 subcutaneous injections of 10 μg of rFhPrx mixed with 10 μg of QuilA adjuvant, 3 wk apart. Two weeks later they were orally infected with 10 metacercaria of F. hepatica. In the second experiments, polyclonal anti-Prx IgG was isolated by protein G binding (Pierce Biotechnology, Rockford, IL, USA) from a pool of serum from 7 sheep that were each vaccinated with 100 μg FhPrx 3 times using QuilA as an adjuvant (these animals exhibit a mean 52% protection against a subsequent infection with F. hepatica, unpublished). Mice were given an intraperitoneal injection of 500 μg IgG 10 h prior to oral infection with 10 metacercaria of F. hepatica.

Purification of macrophages from PECs

Following determination of viability by trypan blue exclusion, PECs were adjusted to 5 × 10⁶ cells/ml in supplemented (Eagle’s) modified essential medium (MEM) and cultured in 6-well plates (Corning Costar, Cambridge, MA, USA). After 2–3 h incubation at 37°C, nonadherent cells were removed by washing with warm MEM. The remaining adherent cells were removed with cold PBS and a scraper and readjusted to 1 × 10⁶ cells/ml for experimentation (24).

Coculture of PECs and CD4⁺ cells

PECs were obtained as described previously, and the macrophage population was removed by adherence. Splenocytes were isolated from naive mice, enriched for CD4⁺ T cells by negative selection (Miltenyi Biotec, Bergisch Gladbach, Germany), and stimulated with 1 μg/ml anti-CD3 (BD Biosciences Pharmingen, San Diego, CA, USA) for 3 h at 37°C. Either total PECs or PECs depleted of macrophages were then added to the CD4⁺ T cells at a ratio of 1:5, and cultures were subsequently maintained at 37°C and 5% CO₂ for 72 h. Supernatants were then harvested and analyzed for IL-4, IL-5, IL-13, and IFN-γ production by ELISA.

RNA extraction and reverse transcriptase-polymerase chain reaction (RT-PCR)

RNA was recovered from macrophages by direct addition of Trireagent (Sigma-Aldrich) to the cell pellet of 1 × 10⁶ cells, isolated as described above. Total cellular RNA was subsequently extracted according to the manufacturer’s specifications and 50 ng was subjected to first-strand cDNA synthesis with oligo-dT primers and AMV RT (Promega, Madison, WI, USA). A 5 μl aliquot of the resultant cDNA was amplified with primers specific for β-actin (Stratagene, La Jolla, CA, USA) and Ym1 (27) using cycle conditions as described elsewhere (24). All PCR products were electrophoretically analyzed on 1% agarose gel and visualized by ethidium bromide staining.

Activation of in vitro macrophages

Peritoneal macrophages (1×10⁶ cells/ml) were induced to express Ym1 by incubating them in the presence of 10 μg of parasite extract, rFhPrx, rvFhPrx, rSmPrx1, or LPS as a negative control. After 16 h incubation at 37°C, cells were washed with PBS, and then RNA was recovered from cultured cells by direct addition of Trireagent and extraction as described.

Cytokine assays

T-cell cytokine production was assessed by culturing spleen cells (2×10⁶ cells/ml) with parasite extracts and/or recombinant Prx. Control stimuli included medium alone or anti-CD3 (2 μg/ml) and phorbol myristate acetate (PMA; 25 ng/ml) (24). Supernatants were removed after 72 h and the concentrations of IL-4, IL-5, IL-13, and IFN-γ measured by immunoassay using pairs of commercially available monoclonal antibodies (BD Biosciences Pharmingen).

Statistical analysis

The statistical significance of difference was determined by the 2-tailed Student’s t test. Values of P < 0.05 were considered significant.

RESULTS

Parasite extracts induce the expression of Ym1 in macrophages and the development of antigen-specific Th2 responses

FhPrx is expressed by the infective juvenile and adult F. hepatica parasites (28) and is abundant in both SFh and in medium in which parasites were cultured for 6 h, termed ES products (refs. 24, 28; Supplemental Fig. 1A, B). Separation of ES by HPLC yields 2 major protein peaks, termed PI (>200 kDa) and PII (60–20 kDa) (Supplemental Fig. 1C–E). Total ES products and PI, but not PII, induced the expression of Ym1 in peritoneal macrophages when injected into the peritoneum of BALB/c mice every other day for 12 days (Fig. 1A). This treatment coincided with the development of antigen-specific Th2 responses, as measured by the secretion of IL-4 and IL-5 in the absence of any IFNγ (Fig. 1B). Mass spectroscopic analysis revealed that Prx was the major protein in PI (unpublished results), which was confirmed by immunoblotting using anti-FhPrx antiserum (Supplemental Fig. 1D, E). PII consisted of a cathepsin l-cysteine protease only (31).

Figure 1.

Development of Th2 immunity in response to parasite antigens is associated with increased Ym1 expression in macrophages. A, C) RT-PCR was used to assess the expression of Ym1 in peritoneal macrophages isolated from BALB/c mice given 3 intraperitoneal injections of 5 μg of F. hepatica parasite ES, high molecular mass fraction PI, or the low molecular mass fraction PII (A), and SEA or SWAP (C). Images are representative of a total of 6 mice examined in each case, with the result for all 6 the same. B, D) Spleen cell cultures from the same mice were stimulated with 20 μg of the relevant injected antigen or PMA and αCD3 for 72 h, and the cytokines secreted into the supernatants were measured by ELISA. Data are means ± se, 3 mice/group, analyzed in triplicate; results are representative of 2 separate experiments.

Two schistosome Prxs, SmPrx1 and SmPrx2, are expressed by male and female adult worms, although a much higher level of Prx protein is found in SEA compared to SWAP (33, 34). SmPrx1 is also secreted by schistosome eggs in culture, suggesting that it may participate in the induction of T-cell-mediated granulomatous response to eggs (33, 36). Like F. hepatica ES, delivery of SEA also activated macrophages expressing Ym1 (Fig. 1C) and induced antigen-specific Th2 splenic responses (Fig. 1D). In contrast, administration of the same quantity of SWAP did not induce significant levels of Ym1 expression (Fig. 1C) and produced a mixed Th1/Th2 antigen-specific response (Fig. 1D). Extended delivery of SWAP (9 injections of 5 μg) did induce the expression of Ym1 in macrophages, although the antigen-specific splenic T-cell responses remained of mixed phenotype (data not shown).

Thus, the abundance of Prx in parasite extracts correlates with their ability to induce Ym1 expression and promote polarized Th2 immune responses.

Recombinant helminth Prxs induce the expression of Ym1 in macrophages and the development of Th2 immune responses

A comparison of the amino acid sequence of FhPrx with other known 2-Cys Prxs revealed similarities extending over their entire sequence lengths. FhPrx is 68 and 69% identical to 2 Prxs of S. mansoni, SmPrx1 and SmPrx2, respectively, and exhibited 61% identity to both mouse and human Prx2 (Fig. 2A). The other members of the human 2-Cys Prx family, Prx1, 3, and 4, are 59, 58, and 55% identical to FhPrx. In all 2-Cys Prxs, the regions surrounding the active cysteine residues Cys⁴⁷ and Cys¹⁷⁰ are conserved and create the active site scaffold essential to the antioxidant activity of this group of enzymes (28, 36, 37) (Fig. 2A).

Figure 2.

Helminth Prxs induce the expression of Ym1 in peritoneal macrophages and the production of antigen-specific Th2 immune responses. A) Primary sequence alignment of F. hepatica Prx, S. mansoni Prx2 and Prx1, human Prx2, and mouse Prx2. The arrow indicate the redox-active Cys⁴⁷ and Cys¹⁷⁰ residues that were replaced by Gly to generate an enzymatically inactive recombinant F. hepatica Prx. B) Recombinant Prx derived from F. hepatica (rFhPrx) and S. mansoni (rSmPrx) electrophoresed under reducing (addition of 40 mM DTT prior to electrophoresis) and nonreducing conditions. Monomeric and dimeric forms of the enzymes are indicated by the arrow and arrowhead, respectively. C) RT-PCR was used to assess the expression of Ym1 in peritoneal macrophages isolated from BALB/c mice given 3 intraperitoneal injections of 5 μg of rFhPrx, rSmPrx1, and a recombinant cathepsin L derived from F. hepatica. Images are representative of a total of 6 mice examined in each case, with the result for all 6 the same. D) Spleen cell cultures from the same mice were stimulated with 20 μg of rFhPrx, rSmPrx1, or rFhCL1 and PMA/αCD3 for 72 h, and the cytokines secreted into the supernatants were measured by ELISA. Data are means ± se, 3 mice/group, analyzed in triplicate; results are representative of 2 separate experiments.

rFhPrx, rSmPrx1, and rSmPrx2 were produced in E. coli. Under nonreducing conditions, rFhPrx, rSmPrx1, and rSmPrx2 (data not shown) migrated at ∼50 kDa in 12.5% SDS-PAGE (Fig. 2B), consistent with the enzymes’ dimeric structure, which is brought about by intersubunit disulfide bonds between the active site Cys residues to form a reactive center (28, 35). Reduction of the intersubunit disulfide bridges by the addition of DTT to sample buffers prior to electrophoresis results in both recombinant Prxs migrating as monomers of ∼26 kDa (Fig. 2B), which is in agreement with the molecular mass of the helminth Prxs predicted from the amino acid sequence and the additional 6× His-Tag. Native Prx from both F. hepatica and S. mansoni also demonstrates this dimerization property (refs. 28, 30, 32; Supplemental Fig. 1).

Delivery of the rFhPrx, rSmPrx1, and rSmPrx2 (data not shown) to mice as 6 intraperitoneal injections of 5 μg protein over 12 days induced the expression of Ym1 in peritoneal macrophages (Fig. 2C). Stimulation of spleen cells from the same mice with recombinant Prx showed a polarized Th2 response characterized by the secretion of IL-4 and no IFNγ (Fig. 2D). Similar experiments performed with another helminth secreted recombinant enzyme FhCL1 (Fig. 2C, D) or bacterial LPS (data not shown) did not induce Ym1-expressing macrophages nor promote the development of Th2 immune responses.

The ability of helminth Prx to induce Ym1 is independent of its antioxidant activity

To examine whether the biological activity of Prx was involved in the induction of Ym1 expression, we prepared a variant recombinant form of rFhPrx in which the redox-active Cys⁴⁷ and Cys¹⁷⁰ were replaced with Gly (rvFhPrx) (see Fig. 2A). As a consequence of these replacements, rvFhPrx does not form disulfide linkages, remains in monomeric form in nonreducing conditions, and therefore lacks a reactive center (Fig. 3A).

Figure 3.

Helminth Prx induces the expression of Ym1 in macrophages independently of antioxidant activity. A) Active recombinant wild-type rFhPrx and a recombinant variant, rvFhPrx, were electrophoresed under reducing (with addition of 40 mM DTT prior to electrophoresis) and nonreducing conditions (without DTT). Monomeric and dimeric forms of the enzymes are indicated by the arrow and arrowhead, respectively. Note that rFhPrx but not rvFhPrx dimerizes in the absence of DTT. B) Specific activity of rFhPrx and rvFhPrx was determined by NADPH (0.25 mM) consumption. Hydrogen peroxide (0.25 mM) was provided as a substrate. TR, E. coli thioredoxin; Trx, E. coli thioredoxin reductase. C) RT-PCR detection of Ym1 expression in peritoneal macrophages isolated from BALB/c mice given 3 intraperitoneal injections of 5 μg of rFhPrx and rvFhPrx. Images are representative of a total of 6 mice examined in each case, with the result for all 6 the same. D) Spleen-cell cultures from the same mice were stimulated with 20 μg of either rFhPrx or rvFhPrx and PMA/αCD3 for 72 h, and the cytokines secreted into the supernatants were measured by ELISA. Data are means ± se, 3 mice/group, analyzed in triplicate; results are representative of 2 separate experiments.

To confirm that rvFhPrx lacked antioxidant activity, the specific activity of this and wild-type rFhPrx was measured in an NADPH oxidation assay. Using this system, NADPH consumption occurred when the active rFhPrx was coupled with E. coli thioredoxin and thioredoxin reductase, with the enzyme exhibiting a specific activity of 2500 nmol min⁻¹ mg⁻¹ (Fig. 2B). Addition of rvFhPrx resulted in no consumption of NADPH, confirming the absence of enzyme activity in the recombinant enzyme. rSmPrx1 exhibited similar activity to rFhPrx in these assays (data not shown).

Removal of enzyme activity had no effect on the ability of Prx to induce either Ym1 expression in macrophages (Fig. 3C) or the development of antigen-specific Th2 immune responses (Fig. 3D) following intraperitoneal delivery to mice. Therefore, the Prx-mediated activation of macrophages is independent of the enzyme’s peroxidase activity.

Mammalian Prx induces the expression of Ym1 in macrophages and Prx-specific Th2 immune responses

Considering that Prxs derived from different helminth parasites induced Ym1 expression and that this occurred independently of peroxidase activity, we examined the possibility that the effect was not unique to helminth Prxs, but rather conserved among the 2-Cys Prx family. Accordingly, we prepared rMoPrx2 in E. coli. As observed for rFhPrx, rMoPrx2 migrated at ∼50 kDa under nonreducing conditions in 12.5% SDS-PAGE, and as a monomer of ∼26 kDa under reducing conditions (Fig. 4A).

Figure 4.

Mouse Prx induces the expression of Ym1 in macrophages and the production of antigen-specific Th2 immune responses. A) recombinant rMoPrx and rFhPrx were electrophoresed under reducing (with addition of 40mM DTT prior to electrophoresis) and nonreducing conditions (without DTT). Monomeric and dimeric forms of the enzymes are indicated by the arrow and arrowhead, respectively. B) RT-PCR detection of Ym1 expression in peritoneal macrophages isolated from BALB/c mice given 3 intraperitoneal injections of 5 μg of ES, rFhPrx, and rMoPrx. Images are representative of a total of 4 mice examined in each case, with the result for all 4 the same. C) Spleen-cell cultures from the same mice were stimulated with rMoPrx or PMA/αCD3 for 72 h, and the cytokines secreted into the supernatants were measured by ELISA. Data are means ± se, 4 mice/group, analyzed in triplicate. D) Spleen cells isolated from mice injected with 5 μg of ES, rFhPrx, or rMoPrx were stimulated with 20 μmg/ml of each antigen or PMA/αCD3 for 72 h, and the cytokines secreted into the supernatants were measured by ELISA. Data are means ± se, 4 mice/group, analyzed in triplicate.

Administration of the rMoPrx2 to mice as 6 intraperitoneal injections of 5 μg protein induced the expression of Ym1 in peritoneal macrophages, similar to both the Prx-containing ES products of F. hepatica and recombinant rFhPrx, (Fig. 4B). Stimulation of spleen cells from these mice with rMoPrx2 induced the secretion of the Th2 cytokines IL-4 and IL-5, but not the Th1 cytokine IFNγ (Fig. 4C). Moreover, spleen cells isolated from rMoPrx2-treated mice also produced a Th2 cytokine profile (IL-4 and IL-5, but not IFNγ) in response to stimulation with either F. hepatica ES or rFhPrx (Fig. 4D, and data not shown). Conversely, splenocytes taken from mice given 6 injections of either ES or rFhPrx secreted both IL-4 and IL-5 in response to stimulation with rMoPrx2 (Fig. 4D). Thus, the ability of Prx to induce expression of Ym1 in macrophages and promote Th2 responses is common to 2-Cys Prxs independent of the originating species.

Helminth Prx activates macrophages independently of IL-4 and IL-13

Recent studies showed that Ym1 is the only marker of alternative activation induced by helminth infection in the absence of IL-4 and/or IL-13 signaling (14, 18, 19). We observed that macrophages isolated from wild type BALB/c mice and IL-4 deficient or IL-13 deficient mice (on a BALB/c background) infected with F. hepatica for 7 days exhibited increased expression of Ym1. Furthermore, injection of cytokine-deficient mice with F. hepatica ES products, rFhPrx or rSmPrx1 also induced the expression of Ym1 in macrophages (Fig. 5A).

Figure 5.

Parasite Prx induces the expression of Ym1 in macrophages independently of IL-4 and IL-13. A) Peritoneal macrophages isolated from BALB/c mice 7 days after an oral infection of F. hepatica (BALB/c) were analyzed for the induction of Ym1 by RT-PCR and compared to macrophages derived from nontreated (BALB/c PBS) mice and infected IL-4^−/− and IL-13^−/− mice. Similarly, peritoneal macrophages from BALB/c, IL-4^−/− and IL-13^−/− mice given 3 intraperitoneal injections of 5 μg of F. hepatica ES products, or rFhPrx and rSmPrx1 were analyzed by RT-PCR. Data are representative of the findings from 4 mice/group. B) Peritoneal macrophages were isolated from naive BALB/c mice, cultured ex vivo for 16 h at 37°C in the presence of 10 μg of F. hepatica ES, purified native Prx (nFhPrx), rFhPrx, SEA, rSmPrx1, SWAP, and E. coli LPS. Extracted mRNA was then analyzed for Ym1 expression by RT-PCR. Results are representative of 3 independent experiments.

Since both IL-4 and IL-13 signal through the IL-4Rα subunit, it is possible that the lack of Th2 dependence observed in this study is due to the singular deficit of either cytokine rather than the direct activation by parasite Prx. Therefore, we investigated whether helminth Prx could stimulate the up-regulation of Ym1 in the absence of Th2 cytokines by carrying out in vitro studies. In the complete absence of exogenous cytokine stimulation, parasite secretions which contain Prx (ES, PI, and SEA) and recombinant Prx (rFhPrx and SmPrx1) stimulated the expression of Ym1 in peritoneal macrophages in vitro (Fig. 5B). Incubation of macrophages with SWAP induced some expression of Ym1, possibly reflecting the presence of SmPrx in this preparation. As a control for specificity, we demonstrated that stimulation of the peritoneal macrophages with LPS or recombinant FhCL1 did not induce the expression of Ym1 (Fig. 5B and data not shown).

Prx-activated macrophages enhance the production of Th2 cytokines from naive CD4⁺ T cells

To examine whether Prx-activated, Ym1-expressing macrophages can directly influence the differentiation of Th2 immune responses, we stimulated naive CD4 T cells with αCD3 in the presence of PECs taken from mice that were treated with either rFhPrx or F. hepatica ES products (Fig. 6). When cultured with PECs isolated from control PBS-treated mice, the CD4⁺ T cells secreted a mixture of both Th1 (IFNγ) and Th2 (IL-4, IL-5, and IL-13) cytokines. However, the production of Th2 cytokines was significantly increased, coincident with a reduction in IFNγ production, when naive T cells were cultured with PECs isolated from mice given rFhPrx or ES products (Fig. 6B). When macrophages were removed from the PECs taken from rFhPrx- and ES-mice, the remaining cells (which showed no Ym1 expression; Fig. 6A) did not induce an alteration in cytokine production by T cells. Significantly, addition of the isolated Ym1 expressing macrophages to naive CD4⁺ T cells resulted in a polarized Th2 response to stimulation with αCD3 (data not shown), supporting our conclusion that the effect on T cells is mediated by Ym1-expressing macrophages.

Figure 6.

Parasite-activated macrophages polarize naive CD4⁺ T cells toward a Th2 phenotype in vitro. A) PECs derived from BALB/c mice given 6 intraperitoneal injections (5 μg) of rFhPrx or PBS were harvested or depleted of macrophages (PEC-Mθ) and analyzed by RT-PCR for the expression of Ym1. B) PECs or PEC-Mθ were cultured with naive splenic CD4⁺ T cells and αCD3 (1 μg/ml) for 72 h. C) PECs or PEC-Mθ isolated from mice given 3 i.p. injections of F. hepatica ES were cultured with naive splenic CD4⁺ T cells and αCD3 (1 μg/ml) for 72 h. Levels of IL-4, IL-5, IL-13, and IFNγ in the culture supernatants were determined by ELISA. Data are means ± se of triplicate cultures and are representative of 3 independent experiments. Statistical significance is based on comparisons relative to cocultures containing PECs isolated from mice receiving injections of PBS.

Neutralization of helminth Prx reduces Ym1 in macrophages and the development of parasite-specific Th2 responses

To support our hypothesis that the induction of Ym1 expression in macrophages promoted the differentiation of Th2 cells, and that Prx was essential to this induction in helminth infection, we performed antibody blocking experiments in mice (Fig. 7). First, BALB/c mice were immunized with rFhPrx (in QuilA adjuvant) 3 times before an infection with F. hepatica. Seven days later, peritoneal macrophages were analyzed for Ym1 expression and spleen cells for parasite antigen-specific cytokine secretion. The data revealed that expression of Ym1 in macrophages was inhibited in 12 of 15 rFhPrx-immunized mice but not in any of the adjuvant control mice (Fig. 7A). Correlating with the absence of Ym1 was a significant reduction in the production of parasite antigen-specific Th2 cytokines and the secretion of IL-4 and IL-5 in response to stimulation with PMA/αCD3. No change was observed in the levels of IFNγ produced (Fig. 7A). In a second experiment, we passively administered anti-Prx IgG (500 μg of polyclonal total IgG), isolated from sheep that were protectively vaccinated with rFhPrx to mice 10 h before an oral infection of F. hepatica. These antibodies caused a reduction in macrophage Ym1 expression and a reduction in Th2 cytokine production at 7 days postinfection compared to mice administered control sheep IgG (Fig. 7B).

Figure 7.

Neutralization of Prx inhibits macrophage Ym1 expression and reduces Th2 polarization. BALB/c mice were immunized with rFhPrx (10 μg) in adjuvant (QuilA) and subsequently challenged with an oral infection of F. hepatica. A) After 7 days, peritoneal macrophages were removed and analyzed for Ym1 expression by RT-PCR. Images are representative of a total of 15 mice examined. Spleen cell cultures from the same mice were stimulated with 20 μg of parasite antigens (ES and rFhPrx) or PMA (25 ng/ml) and αCD3 (2 μg/ml), for 72 h, and the levels of IL-4, IL-5, and IFNγ in the culture supernatants were measured by ELISA. Data are means ± se, 10 mice/group, analyzed in triplicate; results are representative of 2 separate experiments. Statistical significance is based on comparisons relative to nonimmunized (PBS) group using the Student’s t test. B) RT-PCR analysis of Ym1 mRNA in macrophages isolated from BALB/c mice treated with anti-Prx IgG (+) or preimmune IgG (–) and challenged with a 7 day infection of F. hepatica. Spleen cells from the same mice were stimulated with parasite ES or PMA/αCD3 for 72 h, and levels of IL-4 in the supernatant were measured by ELISA. Data are means ± se, 5 mice, analyzed in triplicate.

DISCUSSION

Despite the biological diversity among helminth parasites, they all induce similar immune responses in their hosts, characterized by a potent Th2 response and reduced Th1 response (38, 39). AAMacs are also a hallmark of these infections and, indeed, much of the characterization of this phenotype of macrophage has been based on cells isolated from mice that were chronically infected with helminths (7, 8, 9). While there are several markers of alternative activation (e.g., Fizz and Arg1), the chitinase-like Ym1 protein is most abundantly expressed in helminth-activated macrophages (9, 14). Chitinases are components of an evolutionary conserved innate immune response against chitin-containing pathogens such as parasites, but no antiparasitic properties have been identified for Ym1. However, it has been suggested that Ym1 expressed by antigen-presenting cells plays a role in Th2 differentiation (17), which is supported by data showing that Ym1-expressing macrophages adoptively transferred into naive recipients significantly altered the antigen-specific T-cell response toward a Th2 profile in vivo(40).

AAMacs are recruited to the site of helminth infection as early as 24 h after initial infection (24, 39), and their activation is generally considered to occur only in the context of a Th2 cytokine environment (41). However, while the expression of most markers of alternative activation are dependent on the presence of the Th2 cytokines IL-4 and/or IL-13, studies have shown that macrophages isolated from parasite-infected mice deficient in IL-4Rα signaling pathways still express Ym1 (14, 18, 19, 40). This observation would suggest that helminth parasites, or molecules secreted by them, can directly activate macrophages to express Ym1. Our present studies would support this proposal as we have shown that soluble extracts and secretory antigens derived from the helminths F. hepatica and S. mansoni can induce the recruitment and activation of Ym1-expressing macrophages in mice. We have also identified a molecule, the antioxidant Prx, that is secreted by both parasites and induces Ym1 expression in macrophages in the absence of cytokine stimulation. Neutralization of Prx by specific antibodies not only prevented its capacity to induce the expression of macrophage Ym1 in F. hepatica-infected mice, but also prevented T cells from differentiating toward a Th2 phenotype. These data show that Prx-induced Ym1 is a critical step in the process that leads to the induction of Th2 responses by helminth parasites.

The helminth parasites F. hepatica and S. mansoni are both digenetic trematodes and are thus closely related organisms. However, Prxs are produced by more diverse helminth parasites such as nematodes, and their structure is very well conserved (the trematode and nematode Prxs share ∼66–68% sequence identity at the amino acid level) (36, 37, 42). Therefore, we propose that the strategy of Th2 induction via Ym1-expressing macrophages could be a generic mechanism utilized by helminth parasites. In support of this idea, we have shown that a recombinant form of the Prx expressed by the gastrointestinal nematode of sheep, Hemonchus contortus, also induces the expression of Ym1 in macrophages following intraperitoneal delivery to BALB/c mice (unpublished results). Accordingly, given the highly conserved nature of Prxs among helminth parasites, these molecules may represent a novel pathogen-associated molecular pattern. Considering that Th2 responses are essential to the parasite’s survival in the host, we have exploited FhPrx as a potential vaccine; vaccination of sheep with recombinant FhPrx provided high levels of protection (52%) against a heterologous challenge with F. hepatica parasites (43).

Because of the ability to detoxify hydrogen peroxide, helminth Prx was initially believed to be a housekeeping molecule involved in the protection against reactive oxygen species (ROS) generated during regular cellular metabolism (35, 36). Moreover, since Prxs are secreted extracorporeally by parasites, a role in the inactivation of ROS released by the host’s immune effector cells such as eosinophils, neutrophils, and macrophages was also proposed (42, 44). While these may be functions of helminth Prx as an antioxidant enzyme, we have used an inactive variant of F. hepatica Prx to demonstrate that the activation of macrophages by Prx is not dependent on the enzyme’s antioxidant activity. Helminth Prx-mediated activation of macrophages likely involves direct interaction of a conserved Prx motif with a presently unknown receptor molecule. The involvement of a conserved motif is further supported by our observation that rMoPrx2, which shares 61% sequence identity to helminth Prx, also induced the expression of Ym1 in peritoneal macrophages. Furthermore, T cells primed in vivo by exposure to F. hepatica ES or rFhPrx responded in vitro to stimulation with rMoPrx2 by secreting Th2 cytokines (and, conversely, spleen cells from rMoPrx2-treated mice responded similarly to ES and rFhPrx), indicating the recognition of a shared motif within the Prx molecules.

Studies of human 2-Cys Prxs have shown that they can interact with various proteins and modulate their activities. For example, the interaction of human Prx-1 with both c-Myc and c-Abl alters the biological function of these oncoproteins (45, 46). Human Prx-4 binds to, and regulates the function of, the thromboxane A₂ G-protein coupled receptor (47). Most significantly, a recent report showed that a mutant recombinant of human Prx-1 that lacked antioxidant activity interacts with the androgen receptor, enhancing its transactivation (48). All of these examples support our proposal for a new function for Prx in macrophage activation mediated through protein-receptor binding.

The discovery of a novel function for Prx is a significant finding, with implications beyond parasite immune-modulation. These enzymes are located intracellularly in most cells, where they are known to have multiple antioxidant functions such as the regulation of transcription activation and the protection of cells from oxidative stress (49). However, Prx is also actively secreted from lung cancer cells in vitro and in vivo(50), and its overexpression in breast cancer cells, esophageal cell carcinoma, and pancreatic adenocarcinoma acts as a biomarker for early detection of disease (50, 51). The clinical significance of Prx detection remains undefined, although it has been suggested that the abnormal redox status of tumor cells induces Prx as a cellular resistance to oxidative damage. As AAMacs are a major component of the cellular infiltrate that aids tumor progression (8, 52), and given that we found that rMoPrx2 is capable of driving Th2 responses, it is possible that tumor cells secrete Prx to modulate the surrounding immune microenvironment to one that is beneficial for their proliferation and migration.

The discovery of Prx as a molecule that instructs Th2 responses via Ym1-expressing macrophages provides a model that can be used to dissect the mechanisms behind Th2-driven inflammatory disorders. In addition, preliminary data demonstrates the ability of FhPrx to induce the expression of Ym1 in dendritic cells in the peritoneal cavity of mice following intraperitoneal delivery (data not shown). Characterization of the structural motif on these Prx molecules that binds to and activates antigen presenting cells and elucidation of the cognate receptor has potential not only for the development of antihelminth treatments but also as prospective immuno- or chemotherapeutics for human inflammatory disorders.