Role for CTLA-4 but not CD25+ T cells during Schistosoma mansoni infection of mice

Authors


: Dr Padraic Fallon, Institute of Molecular Medicine, St James's Hospital, Trinity College Dublin, Dublin 8, Ireland (e-mail: pfallon@tcd.ie).

SUMMARY

Schistosoma mansoni infection of mice increases the frequency of cells that are CD4+CD25+ in the acute (4 and 8 weeks) and chronic (16 week) stages of infection. Depletion of > 85% of CD25+ cells in the acute or chronic stages of schistosome infection caused no overt changes in morbidity or immunological responses. The absence of effect in mice with CD25+ cells depleted may be due to the preferential expression of IL-4 and IL-10, two cytokines that are protective in schistosome infection, on CD25 CD4+ cells. We also assessed infection-induced changes of other regulatory markers, GITR, CD103 and CTLA-4 on CD4+ cells. We identified a marked expansion of CTLA-4+ population on CD25 CD4+ cells in acute and chronic infection. Blocking of CTLA-4 during acute, but not chronic infection, caused significant weight loss and altered the type 2 cytokine response of mice, with increased IL-4 and IL-5 production associated with significantly more Th2 cells and eosinophils in the liver granuloma. This study illustrates the complexity of regulation of T cells in schistosome infection and highlights a specific role for CTLA-4+, but not CD25+ cells, in the regulation of Th2 responses in helminth infection.

Abbreviations:
ALT

alanine aminotransaminase

AST

aspartate aminotransaminase

AW

adult worm antigen

GITR

glucocorticoid-induced tumour necrosis factor receptor family related protein

SEA

soluble egg antigen

Treg

T regulatory

INTRODUCTION

A central cause of the immunopathology during infection with the helminth parasite Schistosoma mansoni is the granulomatous inflammation evoked by the parasite's eggs that become trapped in various tissues (1). These eggs are laid in the vasculature 4–5 weeks after infection and they are responsible for the stimulation of the characteristic type 2 cytokine-dominated responses in infected mice (2,3). An obligate role for T cells in schistosome granuloma formation was initially shown using various mice with T cell deficiencies (4–6). More recent studies have demonstrated that granuloma formation around the schistosome egg is dependent on CD4+, and not on CD8+, T cells (7–10).

Although CD4+ cells are known to be essential to schistosome egg granuloma formation, the role of distinct CD4+ cell subsets in schistosome infection has not been fully addressed. There are various subsets of CD4+ T cells with regulatory activity, with these cells limiting the immune response and having a maintenance role in preventing autoimmune disease. T regulatory (Treg) CD4+ cells include those that can occur naturally and cells that are induced by certain stimuli (11–13). Natural Treg cells were initially characterized by surface expression of the activation marker IL-2α receptor, CD25, on CD4+ cells. The levels of such natural CD4+CD25+ Treg cells have been shown to be expanded by various pathogens and are implicated in limited immunopathology during pathogen infection (13). With respect to murine S. mansoni infections, it has been shown that CD4+CD25+ Treg cells have an important role during infection as they are a source for IL-10 (14), with IL-10 known to prevent pathology during schistosome infection of mice (15–18). In another study it was also demonstrated that CD4+CD25+ Treg cells from S. mansoni-infected mice may contribute to Th2 cell polarization during infection by suppression of the Th1 cell response (19).

In addition to CD25, other putative surface markers for Treg cells include glucocorticoid-induced tumour necrosis factor receptor family related protein (GITR) and the integrin CD103 (20). A hallmark of Treg cells is the expression of forkhead family transcription factor Foxp3 (21). Although expressed on activated T cells, the co-stimulatory molecule CTLA-4 is also associated with regulatory activity (22). CTLA-4 deficient mice develop massive lymphoproliferation and early death (23,24), and CTLA-4 also plays a vital role in Th2 cell differentiation (25,26).

In this study we show that although the levels of CD4+CD25+ cells are up-regulated by S. mansoni infection, depletion of CD25 cells during infection showed no obligatory role for CD25+ cells in immune regulation in infection. We therefore evaluated if schistosome infection induced alterations in known regulatory markers on CD4+ but CD25 spleen cells. We identified that schistosome infection expanded a CD4+CD25CTLA-4+ spleen cell population. A functional role for CTLA-4 in suppression of type 2 cytokine responses in schistosome infection was shown when CTLA-4 was blocked with mAb during acute but not chronic infection.

MATERIALS AND METHODS

Mice and parasite infections

BALB/c and CD1 strain mice were bred in-house. Bicistronic 4get IL-4 reporter mice (27), on a BALB/c background, were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and bred in-house. Mice were housed in a specific pathogen-free facility in individually ventilated and filtered cages under positive pressure (Techniplast, UK). A Puerto Rican strain of S. mansoni was maintained by passage in CD1 strain mice and albino Biomphalaria glabrata snails. Six- to eight-week-old BALB/c or 4get mice were infected percutaneously with 40 cercariae. All animal experiments were performed in compliance with Irish Department of Health and Children regulations.

Parasitology and pathology

Schistosoma mansoni adult worm (AW) and soluble egg antigens (SEA) were prepared as described (28). Portal perfusion for worm recovery, liver, intestinal and faecal egg counts were performed as described (29). Parasite fecundity was determined by dividing total tissue (liver and intestine) egg counts by the number of worm pairs present. Livers from mice were taken and fixed in 10% formaldehyde saline for histology. Sections of liver were stained with haematoxylin and eosin for egg granuloma diameter measurements, Giemsa-stained for eosinophil quantification and Martius Scarlet Blue stained for visualization of collagen deposition. Hepatic fibrosis was measured by quantification of hydroxyproline (30). Hepatic hydroxyproline was measured in weighed samples of livers from acutely or chronically infected mice and also from uninfected age- and sex-matched mice. Hepatic collagen is expressed as the increase in infected mice above levels in age–sex matched uninfected mice adjusted to microgram collagen per 10 000 liver eggs, as described (31). To quantify the extent of liver damage in mice, plasma was recovered for analysis of aspartate aminotransaminase (AST) and alanine aminotransaminase (ALT) levels, as described (29,32).

mAbs and cell depletions

The following mAbs were used for flow cytometry: APC-Alexafluor and Tri-conjugated anti-CD4 (CT-CD4), Tri-conjugated anti-CD8 (5H10), Tri-conjugated anti-CD19 (6D5), Tri-conjugated F4/80 (F4/80), FITC-conjugated anti-IL-10 (JES5-2A5), PE-conjugated IL-4 (BVD6-24G2) and PE- or APC-conjugated anti-CD25 (PC61 5·3) were obtained from Caltag Medisystems (UK). Purified anti-CD25 (PC61) was also biotinylated in-house using a commercial kit (Pierce, USA). PE-conjugated anti-CCR3 (83101), anti-GITR (108619) and rat IgG2a isotype were from R&D Systems (UK). Purified or PE-conjugated anti-CTLA-4 (UC10-4F10-11), biotinylated anti-hamster IgG (G70-204, G94-56), FITC-conjugated anti-CD103 (M290), FITC-conjugated anti-CD25 (7D4), FITC-conjugated anti-rat IgG and Streptavidin-FITC were all obtained from BDPharmingen (UK). FITC- or PE-conjugated anti-Foxp3 mAb (FJK-16s) were obtained from eBiosciences (USA). FITC-conjugated anti-T1/ST2 (DJ8) was obtained from MD Biosciences (Morwell Diagnostic GmbH, Germany).

Depleting/blocking mAbs against CD25 (PC61 5·3) and CTLA-4 (UC10-4F10-11) were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). A hamster monoclonal mAb (IgG) against dinitrophenol (UC8-1B9), also purchased from ATCC, was used as a control mAb for the anti-CTLA-4 mAb. Rat IgG (Sigma, Dorset, UK) or a rat IgG1 mAb against trinitrophenyl (1B7·11; from ATCC) was used as control for the anti-CD25 mAb. The anti-CD4 (YTS191) and anti-CD8 (YTS169) hybridomas, originally from Prof. Herman Waldmann (University of Oxford, UK), was provided by Prof. Anne Cooke (University of Cambridge, UK). The above hybridoma cell lines were cultured in standard conditions using roller bottles, and supernatants were precipitated in 50% ammonium sulphate followed by extensive dialysis against endotoxin-free Dulbecco's PBS (pH 7·2; Sigma). Antibody was purified on Protein G separation columns (Sigma) and protein was quantified before use. All antibodies were tested for endotoxin contamination and confirmed to have < 0·5 EU/mg (Chromogenic LAL, Biowhittaker, MD).

In all in vivo depletion experiments the efficacy of the dose of mAb used and frequencies of injections was initially optimized to deplete target cells over 4 weeks, with doses of 200 µg to 1 mg per week tested. To deplete cells during infection mice were treated i.p. with 200 µg of mAb administered three times a week. The levels of CD4+CD25+ cells in the spleen, mesenteric lymph nodes and liver granulomas, isolated as described below, were checked by flow cytometry weekly in 2–3 representative mice treated with anti-CD25 (PC61) mAb. For cytometry detection of CD4+CD25+ cell levels the depleting mAb (PC61) was used or, as this data may be inaccurate due to the epitope on IL-2 receptor been prebound by PC61 in vivo, a different CD25 mAb clone (7D4), that reacts against a different epitope than PC61 on the IL-2 receptor was used (33). A > 85% depletion of CD25+ cells was achieved throughout the 4-week depletion regime.

Cell culture and immunological analysis

Cells were isolated from granulomas in the livers of infected mice as described (29). Spleen or mesenteric lymph nodes were passed through 70 µm sieves (BD Biosciences, San Jose, CA, USA) to prepare single cell suspensions. Splenocytes were depleted of erythrocytes by lysis with ammonium chloride solution. Cells were cultured in RPMI 1640 (Biowest, UK) supplemented with 10% (v/v) heat-inactivated FCS (Labtech, UK), 100 mm l-glutamine (Gibco, Carlsbad, CA, USA), 100 U/mL penicillin and 100 µg/mL streptomycin (Gibco). Depletion of CD4+, CD8+ or B220+ cells in vitro was performed using anti-CD4 (YTS191), anti-CD8 (YTS169) and anti-B220 (RA3-3A1) mAb and two rounds of complement-mediated lysis (Lo-Tox rabbit complement; Cedarlane Laboratories, Hornby, Canada). Cells were unstimulated (media) or stimulated with 0·5 µg/mL soluble anti-CD3ɛ mAb (clone 145-2C11), SEA (20 µg/mL) or AW (20 µg/mL) at 37°C. Supernatants were harvested after 24 h (IL-2) or 72 h (IL-4, IL-5, IL-10, IL-13, IFN-γ and TGF-β). Sandwich ELISAs were performed to quantify levels of specific cytokines in the supernatants from cell cultures. Reagents for detection of IL-2, IL-4 and IL-5 were from BDPharmingen (UK). IL-13, IFN-γ and IL-10 reagents were purchased as a DuoSet ELISA development system from R&D Systems (UK). TGF-β reagents were purchased as a kit from Promega (Madison, WI, USA), and total TGF-β was quantified on acid-treated samples.

For cell proliferation assays, single cell suspensions were cultured in 96-well U-bottomed plates for 72 h at 37°C. All samples were set up in triplicate wells. Cultures were pulsed with 1 µCi/well [3H]thymidine (Amersham Pharmacia Biotech, Piscataway, NJ) for the last 16 h of culture. Cells were harvested, and [3H]thymidine incorporation was determined.

To label cells with CFSE, spleen cells were washed in PBS before staining with 0·5 µm CFSE at 37°C, then CFSE was quenched with washing with ice-cold RPMI containing 10% FCS. Labelled cells were cultured for 24–72 h. Cells were harvested and incubated with TRI-conjugated antibodies against CD4, CD8, CD19, CD11b or PE-conjugated anti-CCR3, and division of individual cell populations analysed using Cellquest software as described below.

Flow cytometry

Surface marker expression and intracellular cytokine staining was performed as described (10,28,29,34,35). Eosinophils within liver granuloma cells were stained by flow cytometry; eosinophils were CD4CD8CD19 and CCR3+ (36,37). Detection of Foxp3 by intracellular staining was performed according to manufacturers’ instructions (eBioscience, San Diego, CA, USA). IL-4-eGFP+ cells in the spleen or isolated granulomas of 4get mice were analysed by flow cytometry, and the absolute number of IL-4-eGFP+ CD4+ cells determined. Data were collected on a FACScan flow cytometer (Becton-Dickinson, San Jose, CA) and analysed using cellquest software. T1/ST2 staining and 4-colour immunofluorescence staining was performed in the Conway Institute (University College Dublin, Ireland) using a DAKO Cyan cytometre and summit™ software. Quadrants and gates were drawn based on isotype-control Ig staining and were plotted on logarithmic scales.

RESULTS

Levels of CD4+CD25+ T cells are increased during Schistosoma mansoni infection

We first tested if the frequencies of CD4+CD25+ T cells increased during the peak of inflammation during schistosome infection of mice, week 8 post-infection, and found infected mice had a significant (P < 0·01) increase in spleen CD4+CD25+ cells compared to levels seen in uninfected mice (Figure 1a). As CD4+ cell activity is modulated during acute to chronic stages of infection we analysed the frequencies of CD4+CD25+ cells in the spleens of mice infected with S. mansoni for 4, 8 and 16 weeks (Figure 1b). Interestingly, the frequencies of CD4+CD25+ cells had already increased in the spleens of 4 week-infected mice (P < 0·05), which is prior to eggs being laid, indicating that the initial larval/worm stages of infection can induce expansion of CD4+CD25+ cells (Figure 1b). The level of CD4+CD25+ cells increased further by week 8, and remained significantly elevated (P < 0·05) in the chronic stage of infection (week 16) (Figure 1b).

Figure 1.

CD4+CD25+ cells are up-regulated throughout Schistosoma mansoni infection. (a) Expression of CD25 on CD4+ spleen cells from naïve and 8-week S. mansoni-infected mice. (b) Percentage of CD4+ spleen cells expressing CD25+ in mice infected with S. mansoni for 4, 8 or 16 weeks, and expression on age- and sex-matched uninfected mice. Data shown are Mean ± SD from four individual mice with similar levels seen in at least three separate experiments per time-point. Student's t-test was used to test for statistical difference at each time point. (c) Foxp3 expression on CD4+CD25+ cells from uninfected and mice infected for 4, 8 or 16 weeks. Data shown is from individual mice and are representative from at least two separate experiments. Isotype gates for CD25 and Foxp3 are shown. Flow cytometry was performed on the spleen cells from 16 week-infected mice on separate days from others.

The transcription factor Foxp3 is another marker of regulatory cells that is predominately expressed on CD4+CD25+ cells (21). We investigated expression of Foxp3 on CD4+CD25+ cells during the acute and chronic stages of infection. Despite the increased frequency of spleen CD4+CD25+ cells during infection, there was little or no increase in the percentage of CD4+CD25+ cells that co-expressed Foxp3 at week 4 or 8, with a slight increase at week 16 post-infection (Figure 1c). The increase in CD4+CD25+ cells at week 4, and the greater increase observed at weeks 8 and 16 of infection was largely confined to a Foxp3 population (Figure 1c), with these CD4+CD25+Foxp3 cells possibly representing activated T cells induced by the S. mansoni infection.

Depletion of CD25+ cells during acute or chronic Schistosoma mansoni infection has no effect

The expansion of CD4+CD25+ T cells during the first 4–8 weeks of S. mansoni infection (Figure 1) is temporally coincident with the essential role for CD4+ T cells in the granulomatous response during the acute stages of infection of mice (10,38). Therefore we used mAb to deplete either CD4+ or CD25+ cells during the acute 4–8 weeks of infection. The protective role of CD4+ cells in acute schistosome infection was illustrated by a range of parameters, including the progressive weight loss and deaths of anti-CD4 mAb-treated mice (Figure 2a); which was due to hepato-intestinal pathology (data not shown (10)). In contrast, depletion of CD25+ cells caused no mortalities, with anti-CD25 mAb-treated animals having comparable weight changes as seen in control mAb-treated mice (Figure 2a). Additionally, anti-CD25 mAb treatment did not alter parasite fecundity, egg production or excretion as well as egg granuloma size, eosinophil content, liver fibrosis or hepatocyte damage (Table 1; data not shown). Furthermore, spleen and mesenteric lymph nodes cells from anti-CD25 mAb-treated mice had comparable production of Th2 (IL-4, IL-5), Th1 (IFN-γ) or regulatory (IL-10 and TGF-β) cytokines as control mAb-treated mice when stimulated with anti-CD3 mAb or parasite worm or egg antigens (Figure 2b,c). IL-2 production was also equivalent between groups (Figure 2b,c). It has been suggested that CD25+ cells may preferentially function in the chronic stages of S. mansoni infection (39). We therefore depleted CD25+ cells in mice from week 12 to 16 during the chronic infection. Depletion of CD25+ cells in chronic infection had no effect on any parameter investigated (Table 1, Figure 3a,b).

Figure 2.

CD4+ cells are protective but no role for CD25+ cells in acute schistosome infection. BALB/c mice were infected with S. mansoni and control mAb or mAb against CD4+ (YTS191) or CD25+ (PC61) were administered three times per week from week 4 post-infection. (a) Changes in body weight. Weights were measured routinely up to 8 weeks post-infection and are shown as percentage weight change from week 4. Mortalities are represented by a black cross. The number of deaths/total number in each group is shown in brackets. Data shown are from five to seven mice per group, and are representative of five separate experiments. Production of cytokines by cells from the spleen (b) and (c) mesenteric lymph node from infected mice. Cells were unstimulated (media), or stimulated with 0·5 µg/mL anti-CD3ɛ mAb, AW (20 µg/mL) or SEA (20 µg/mL) for 24 h (IL-2) or 72 h (IL-4, IL-5, IFN-γ, IL-10 and TGF-β). Cytokines in cell supernatants were detected by ELISA and expressed as pg/mL.

Table 1.  Parasitology and liver granuloma size in S. mansoni-infected mice treated with anti-CD25 mAb or control mAb in acute or chronic stages of infectiona
 EggsbGranuloma
Worm pairsLiverIntestinesExcretedcGranuloma diameter (µm)Eosinophils %
  • a

    Data are Mean + SEM from four to six mice per group;

  • b

    Eggs per worm pair or in tissue are × 103;

  • c

    c Eggs in faeces (eggs per gram).

Acute infection
 Control mAb3·8 ± 0·4 8·9 ± 1·112·3 ± 1·8379 ± 37398 ± 4454 ± 11
 Anti-CD25 mAb3·9 ± 0·6 8·2 ± 1·413·2 ± 1·2356 ± 43411 ± 3252 ± 9
Chronic infection
 Control mAb4·1 ± 0·917·9 ± 2·815·8 ± 1·8203 ± 24227 ± 3141 ± 10
 Anti-CD25 mAb3·7 ± 1·119·2 ± 1·615·3 ± 1·9210 ± 19218 ± 3739 ± 7
Figure 3.

Depletion of CD25+ cells does not alter chronic schistosome infection. Schistosoma mansoni-infected mice were treated with and control mAb or mAb against CD25 (PC61) three times per week from week 12 post-infection. (a) Weight changes from week 12 to 16 post-infection. Data shown are from four to six mice per group, and are representative of three separate experiments. (b) Cytokine production from spleen or mesenteric lymph node cells from infected mice. Cells were unstimulated (media), treated with anti-CD3 mAb, AW or SEA. IL-4, IL-10, IFN-γ and TGF-β levels were measured by ELISA. (c) Efficacy of CD25+ cell depletion in spleen, mesenteric lymph node, and liver granuloma in both acute and chronic infection. Cells were stained for CD4 and CD25 and percentage of CD4+ cells that were CD25+ was measured. Values are mean + SD from four individual anti-CD25 mAb treated infected mice compared control mAb-treated infected mice. (d) Levels of Foxp3 expression on CD25+ and CD25 CD4 cells from control mAb or anti-CD25 mAb-treated infected spleens. Data shown represents mean + SD from two to four individual mice from at least two separate experiments.

In both acute and chronic infection experiments the efficacy of CD25 cell depletion was checked by flow cytometry. After 4 weeks of anti-CD25 mAb treatment, at the end of acute or chronic infection, flow cytometry confirmed > 85% depletion of CD25+ cells in the spleens, mesenteric lymph nodes or liver granuloma cells (Figure 3c). A representative image of spleen cells from control mAb-treated or anti-CD25 mAb-treated mice is shown in Figure 3(d). So despite the marked increase in CD4+CD25+ cells in mice during schistosome infection, depletion of the CD25+ cells did not alter cytokine responses, granuloma formation and immunopathology during acute or chronic stages of an S. mansoni infection. We have not addressed functional activation vs. depletion of CD25 cells after PC61 treatment (40). However, when we checked for Foxp3 expressing cells in infected mice treated with anti-CD25 mAb, there was an increase in Foxp3+ population in the CD25 cell population (Figure 3d). These data show that while CD4+ cells are involved in schistosome infection, depletion of the CD25+ cells has no effect in acute or chronic infection, it is indicative that a CD4+ but CD25 cell may have a regulatory role in schistosome infection.

Schistosoma mansoni infection induces production of IL-4 and IL-10 from CD4+ cells that are CD25

Mice deficient in IL-4 succumb to pathology in acute S. mansoni infection (17,41,42), highlighting that IL-4 is essential in infection to prevent mortalities. As CD4+ cells are a major source of IL-4 in infection (28), we checked for IL-4 production from the CD25+ or CD25 populations. The ~5-fold expansion of Th2 cells in the spleen of mice following schistosome infection was preferentially due to an increase in the frequencies of IL-4 producing CD4+CD25 cells, with significantly (P < 0·005) more IL-4 producing cells in the CD25 subpopulation (Figure 4a). Using the Th2 cell specific marker T1/ST2, which is up-regulated on Th2 cells in schistosome infection (43), it was also confirmed that T1/ST2 expression is predominately on the CD25 fraction of CD4+ T cells in infected mice (data not shown). In support of a CD25 cell source for IL-4, infected mice treated with anti-CD25 mAb there was intact, or possible elevated, IL-4 production by cells from the spleen or mesenteric lymph node (Figures 2b and 3b).

Figure 4.

CD4+CD25 spleen cells produce IL-4 and IL-10 during S. mansoni-infection. Spleen cells were isolated from uninfected mice or 7–8 week infected mice. Representative images of (a), intracellular IL-4 and (b), intracellular IL-10 producing CD4+CD25+ and CD4+CD25 cell populations. Bar graph shows Mean + SEM of frequency of cell with intracellular cytokine in CD25+ or CD25 CD4+ cells in infected mice. Student's t-test was used to test for statistical differences between CD25 and CD25+ cells, from three to six mice per group. Data shown are representative from two to three separate experiments. All gates or quadrants were set on appropriate mAb isotype controls.

We also looked at production by CD4+CD25+ or CD4+CD25 cells of the regulatory cytokine IL-10, a cytokine with an essential role in schistosome infection (15,17,18). In infected mice there is a marked expansion of spleen CD4+ cells that produce IL-10, with significantly (P < 0·05) more CD4+ IL-10 producing cells that were CD25 than CD25+ (Figure 4b). Thus based on intracellular IL-10 detection, the depletion of CD25+ cells would not lower IL-10 levels, which is what was detected in spleen or mesenteric lymph node cells from infected mice treated with anti-CD25 mAb (Figures 2b and 3b). Therefore as both IL-4 and IL-10 are produced by CD25 Th2 cells it may explain why depletion of CD25+ cells had a negligible effect on S. mansoni infection, as the CD4+ cells producing IL-4 or IL-10 would not be removed.

Schistosoma mansoni infection induces elevation in a CD4+CD25CTLA-4+ population

We tested for differential expression of putative regulatory markers on CD4+ cells that were CD25+ or CD25 during acute infection. Expression of CD103 was marginally elevated on CD4+ spleen cells from infected mice, with CD103 expression significantly up-regulated (P < 0·01) in the CD25+ relative to CD25 cell population (Figure 5a). The expression of the regulatory marker GITR was markedly increased on cells from infected mice but, similar to CD103, expression was predominately on CD4+CD25+ cells, with significantly (P < 0·01) more GITR expression on CD25+ vs. CD25 cells (Figure 5b).

Figure 5.

CTLA-4 is up-regulated on CD4+CD25+ and CD4+CD25 spleen cells during S. mansoni-infection. Spleen cells were isolated from uninfected mice or 7–8 week infected mice. Representative images of the expression (a) CD103, (b) GITR and (c) CTLA-4, on CD4+CD25+ and CD4+CD25 cell populations was analysed. Bar graph shows Mean + SEM of expression of each surface marker on CD25+ or CD25 cells in infected mice. Student's t-test was used to test for statistical differences in marker expression between CD25 and CD25+ cells, from three to six mice per group. Data shown are representative from two to four separate experiments. All gates or quadrants were set on appropriate mAb isotype controls.

Levels of the regulatory marker CTLA-4 were also assessed on cells from uninfected and S. mansoni-infected mice. Similar to CD103 and GITR there was an increase in CTLA-4 expression on the CD4+CD25+ population in infected mice compared to uninfected animals (Figure 5c). Interestingly, S. mansoni infection also induced a marked three- to fourfold increase in CTLA-4 expression on the CD25 CD4+ cell population, with significantly (P < 0·05) greater CD25 cells expressing CTLA-4 than CD25+ cells (Figure 5c).

It has previously been observed that the phenotype observed in mice deficient in CTLA-4 is similar to that which develops in Foxp3 (scurfin) deficient mice, with both groups developing massive lymphoproliferation and early death (23,44). As Foxp3 is potentially one of the more robust markers for Treg cell activity (45), we examined Foxp3 and CTLA-4 co-expression on CD4+CD25+ and CD4+CD25 spleen cells from uninfected mice and animals infected for 4 and 8 weeks. Consistent with Foxp3 being a marker for CD4+CD25+ cells, > 95% of such cells expressed Foxp3 in uninfected mice, whereas there was a slight reduction in Foxp3+ cells by week 8 of infection (Figure 6). Within the CD4+CD25+ cell population there was increased expression of CTLA-4 in cells from mice with an 8-week schistosome infection, with almost 40% of CD4+CD25+ cells from infected mice co-expressing Foxp3 and CTLA-4. In addition, by week 8 of infection ~5% of CD4+CD25+ cells also expressed CTLA-4 but were negative for Foxp3, compared to < 1% of such cells from uninfected or 4-week infected mice (Figure 6). Strikingly, in the CD4+CD25 cell population from 8-week-infected mice there was 10-fold expansion of CTLA-4+Foxp3 cells relative to these cells in uninfected mice (Figure 6). This specific expansion during acute infection of a CD4+CTLA-4+ cell population, that was CD25 and Foxp3, prompted us to examine the role of CTLA-4 during infection.

Figure 6.

Alterations in CTLA-4 and Foxp3 expression on CD25+ or CD25 CD4+ spleen cells from mice with acute infections. Cells were gated as CD4+CD25+ or CD4+CD25 and expression of CTLA-4 and Foxp3 on each population was assessed. CTLA-4 and Foxp3 mAb iostypes are shown. Data are representative of two separate experiments, and was analysed using Summit™ software.

Blocking CTLA-4 alters Th2 responses in acute, but not chronic, schistosome infection in mice

Schistosome-infected mice were administered a blocking mAb against CTLA-4 or a control hamster mAb from week 4 to week 8 of infection. Mice with CTLA-4 blocked had significant (P < 0·05–0·0001) progressive loss of weight compared to control hamster mAb-treated mice, with 15%–20% loss in body weight by week 8 (Figure 7). However, despite the marked weight-loss in anti-CTLA-4 mAb-treated mice these animals did not appear moribund and in seven experiments there were no deaths. Anti-CTLA-4 mAb-treated mice had comparable schistosome worm burdens and fecundity as control infected mice (data not shown). At autopsy there was no overt exacerbation of intestinal pathology in anti-CTLA-4 mAb-treated mice. Consistent with normal infection-associated changes in the intestines, the ability of CTLA-4 mAb-treated mice to excrete eggs was also normal (321 ± 79 eggs per gram faeces in treated mice compared to 328 ± 66 eggs per gram faeces in control mice; mean ± SEM, n = 5–6 mice).

Figure 7.

Weight changes in S. mansoni-infected mice treated with control or anti-CTLA-4 mAb. Mice were treated with mAb from week 4 post-infection, and mice were weighed periodically until week 8 post-infection. Data are mean values from seven to eight infected mice per group, and are representative of at least six separate experiments. Student's t-test was used to test for differences in weight loss between groups (*P < 0·05; **P < 0·01, ***P < 0·0001).

There were no gross changes in the appearance of livers of CTLA-4 mAb-treated mice with these animals having comparable plasma transaminase levels, which are markers for hepatocyte damage, as control mAb-treated mice (data not shown). The size of the granuloma around schistosome eggs in the liver was larger in anti-CTLA-4-mAb treated mice, but the increase in size was not statistically significant from untreated mice (Figure 8a). Additionally, there was also a nonsignificant increase in liver fibrosis (Figure 8b), with collagen deposition on Martius Scarlet Blue-stained sections indicative that the increase was possibly due to the larger-sized granulomas rather than a generalized increase in hepatic fibrosis in anti-CTLA-4 mAb-treated mice (data not shown). Most notably, on Giemsa-stained liver sections there was a significantly greater influx of eosinophils into the granuloma of anti-CTLA-4 mAb-treated mice compared to untreated mice (Figure 8c; P < 0·05). Rather than relying solely on histological counts for eosinophils we isolated liver granuloma cells and used flow cytometry to quantify eosinophils (CD4CD8CD19CCR3+). By flow cytometry there was also a significant (P < 0·01) increase in eosinophils within the liver of anti-CTLA-4 mAb-treated mice compared to levels in granuloma cells from untreated infected mice (Figure 8d). We also infected 4get mice, which have IL-4 linked to eGFP, and treated the mice with anti-CTLA-4 mAb to address the direct effects of treatment on Th2 (CD4+IL4-GFP+) cells within the liver granuloma. There were significantly more (P < 0·05) Th2 cells within the granuloma of anti-CTLA-4 mAb-treated mice compared to untreated mice (Figure 8e). These data indicate that blocking anti-CTLA-4 during acute schistosome infections causes marked weight loss, but not deaths or overt increased pathology in mice, and causes a significant increase in the eosinophil and Th2 cell content within the liver granuloma.

Figure 8.

Blocking of CTLA-4 during acute infection increases eosinophils and Th2 cells in the liver granuloma. (a) Size (diameter) of liver granuloma and (b), levels of hepatic fibrosis were nonsignificantly (NS) elevated in anti-CTLA-4 mAb-treated groups. (c) Numbers of eosinophils within the liver, determined by counting histology sections and expressed as a percentage of total granuloma cells. (d) Detection of eosinophils in the liver granuloma by flow cytometry. Images are representative of percentage of non B, non T (CD4, CD8, CD19) cells that were CCR3+ from pools of livers from three infected mice. Absolute numbers of eosinophils were determined and expressed as a bar graph, with mean ± SEM from three to four replicates. (e) Frequencies of Th2 cells (CD4+ and IL-4-GFP+) in granulomas from pools of livers from untreated or anti-CTLA-4 mAb-treated S. mansoni-infected 4get mice. An infected wild-type mouse is shown for gating as a GFP control. Absolute numbers of Th2 cells were determined in the livers from four to five replicates and data are expressed as a bar graph. Student's t-test was used to compare differences between groups.

Mice with chronic 16-week infections also have greater up-regulation in the expression of CTLA-4 on CD4+ CD25 cells than seen on the CD25+ cells (Figure 9a). However, when mice were treated with anti-CTLA-4 mAb during the chronic stages (weeks 12–16) of infection there was no difference between groups in body weight or mortalities when compared to infected mice treated with a control hamster mAb (Figure 9b). The size of liver granulomas (259 µm ± 27 vs. 244 µm ± 33) and eosinophil content (42% ± 9 vs. 48% ± 13) were unaltered between control and anti-CTLA-4 mAb-treated groups, respectively. Spleen cells and mesenteric lymph node cells (data not shown), from chronically infected mice treated with anti-CTLA-4 mAb had comparable cell prolifer1ation and production of a range of cytokines as mice treated with control mAb (Figure 9c,d). However, there was some spontaneous IL-5 production from unstimulated cells in anti-CTLA-4 mAb-treated chronically infected mice (Figure 9d). Despite CTLA-4 expression being up-regulated in both acute and chronic stage of schistosome infection, blocking CTLA-4 altered the infection only in acute stages.

Figure 9.

Blocking of CTLA-4 during chronic S. mansoni infection does not induce any weight loss in mice. (a) Representative image of expression of CTLA-4 on CD25 and CD25+ CD4+ spleen cells from uninfected and chronically infected (week 16) mice. (b) Mice were infected with S. mansoni and control mAb or mAb against CTLA-4+ were administered three times per week from week 12 post-infection. Weights are shown as percentage weight change from week 12. Data shown are from six mice per group, and are representative of three separate experiments. (c) Proliferation of spleen cells from infected or anti-CTLA-4 mAb-treated infected mice. Cells were stimulated with media, anti-CD3 mAb, AW or SEA. (3H)thymidine incorporation is expressed as counts per minute (CPM). (d) Cytokine production from stimulated spleen cells from infected or anti-CTLA-4 mAb-treated infected mice. Data in (a–c) were from mice culled at week 16 post-infection.

To address effects of anti-CTLA-4 treatment on cellular responses in mice with acute infections we removed the spleens and mesenteric lymph nodes from schistosome-infected untreated or anti-CTLA-4 mAb-treated mice. Spleen cells from anti-CTLA-4 mAb-treated mice had greater relative cell proliferation to various in vitro stimulations when compared to control infected mice, with marked spontaneous proliferation of cells from these mice when cultured in media (Figure 10a). Using CFSE staining of spleen cells from anti-CTLA-4 mAb-treated mice we have been unable to determine distinct cell types spontaneously proliferating (data not shown).

Figure 10.

Spontaneous production of IL-5 by spleen cells from infected mice treated with anti-CTLA-4 mAb. Infected mice were treated as described in Figure 6, and culled on day 56 post-infection. (a) Proliferation of spleen cells expressed as CPM. (b) Cytokine production from stimulated spleen cells. (c) Proliferation of mesenteric lymph node cells from infected mice. (d) Cytokine production from stimulated mesenteric lymph node cells. Data shown is from pools of spleen or MLN cells from three individual mice, and are representative of at least four separate experiments.

The production of a range of cytokines by spleen cells from the control and anti-CTLA-4 mAb-treated groups was also examined (Figure 10b). Although there was no difference between groups in production of the Th2 cytokine IL-13, the levels of IL-4 were elevated in schistosome antigen-stimulated, but not anti-CD3 treated, cells from the anti-CTLA-4 mAb-treated mice (Figure 10b). Strikingly, there was spontaneous production of the Th2 cytokine IL-5 in the media of the CTLA-4 mAb-treated group, similar to what was observed in chronically infected mice (Figure 9d), but at much higher levels. There was no difference in levels of IL-5 in response to anti-CD3 or antigen-specific stimuli. Levels of the Th1 cytokine IFN-γ were comparable between groups, with slightly less IFN-γ production in the anti-CTLA-4 mAb-treated group upon anti-CD3 stimulation. IL-10 and TGF-β production in response to anti-CD3, and antigen-specific AW and SEA, was equivalent between groups. In contrast, mesenteric lymph node cells from anti-CTLA-4 mAb-treated mice had comparable cell proliferation, with some spontaneous proliferation, and comparable production of all cytokines tested as control mice (Figure 10c,d). Consistent with an elevated Th2 response in anti-CTLA-4 mAb-treated mice, in these animals there was a significant increase (P < 0·05) in total serum IgE (48·7 ± 15·5 µg/mL) relative to IgE detected in serum of untreated mice (31·3 ± 15·5 µg/mL; mean ± SD; Student's t-test of 10 mice per group).

Spleen cells from S. mansoni-infected 4get mice that were treated with anti-CTLA-4 mAb were examined for levels of IL-4-eGFP+ cells. Schistosoma mansoni infection caused a significant ~10-fold increase (P < 0·005) in spleen IL-4-eGFP+ CD4+ T compared to levels detected in spleen cells from uninfected mice (Figure 11a). The anti-CTLA-4 mAb-treated 4get mice had a significant (P < 0·05) elevation in the frequency of splenic IL-4-eGFP+ CD4+ T cells compared to control mAb-treated mice (Figure 11a), which was similar to the increase in Th2 cells that was observed in the liver granuloma of these mice (Figure 8e).

Figure 11.

IL-5 from anti-CTLA-4 mAb-treated mice is predominately derived from non CD4+ cells. (a) Representative images of increase in frequencies of Th2 cells (CD4+ IL-4-GFP+) in spleen cells from uninfected 4get mice, and anti-CTLA-4 mAb-treated infected 4get mice relative to levels in infected 4get mice not treated with mAb. Data from three individual mice per group were used to determine absolute numbers in the spleen. Student's t-test was used to compare differences between groups. (b) IL-4 and IL-5 production by whole spleen cells (Total) or CD4+ cell-depleted spleen cells from untreated or anti-CTLA-4 mAb-treated S. mansoni infected mice. CD4+ cells were depleted by incubating cells with anti-CD4+ (YTS191) followed by complement depletion and cells were stimulated for 72 h with media, anti-CD3, AW or SEA. Levels of the cytokine IL-4 and IL-5 were detected by ELISA. Spleens were removed 56 days post-infection. Data are pooled cells from three spleens and are representative of two to three separate experiments.

In view of the elevated IL-4 and IL-5 in spleen, but not mesenteric, cell cultures (Figure 10b,d) and increased IL-4-eGFP+ CD4+ cells in the spleen of anti-CTLA-4 mAb-treated mice (Figure 11a) we addressed if CD4+ cells are the primary source of the IL-4 and IL-5 in these mice. CD4+ cells were depleted in vitro and cytokine production in response to anti-CD3 or schistosome AW or SEA stimulation was tested. The production of the schistosome-antigen-induced IL-4 was almost completely abrogated when CD4+ cells were depleted from spleen cells from both groups, with a marked reduction in IL-4 induced by anti-CD3 treatment (Figure 11b), indicating that in both groups of mice IL-4 is predominately from CD4+ cells. IL-5 release from in vitro stimulated cells from infected and anti-CTLA-4 mAb-treated infected mice was also produced primarily from CD4+ cells (Figure 11b). In contrast, the spontaneous production of IL-5 in unstimulated (media) cells from spleens of anti-CTLA-4 mAb-treated mice was not diminished when CD4+ T cells were depleted (Figure 11b). The spontaneous production of IL-5 by spleen cells from anti-CTLA-4 mAb-treated mice was also not reduced when CD8+ cells or B (B220+) cells were depleted in vitro (data not shown). This suggests an alternative, non T/non B, cellular source for IL-5 production when CTLA-4 is blocked in acute schistosome infection.

DISCUSSION

In this study, we first demonstrate that in S. mansoni infection of mice there is an expansion in CD4+ cells that co-express the activation marker CD25. However, despite the expansion of these CD4+CD25+ cells during infection, the in vivo depletion of CD25+ cells from week 4 to 8 or 12 to 16 of infection had no effect on schistosome infection and did not alter parasite-induced immune responses. These data based on cell depletion during infection indicate no function for CD25-expressing cells in S. mansoni infection of mice. As we have confirmed an essential protective role for CD4+ cells in infection, our data indicates that a CD4+, but CD25 cell population may regulate immunopathology in schistosome infection. CTLA-4 expression on CD25 CD4+ cells was up-regulated during infection, with anti-CTLA4 mAb treatment altering Th2 responses in acute infection.

Potentially, the essential role for CD4+ cells in acute infection could be due to their production of the Th2 cytokine IL-4, which has an important host protective role (17,41,42), or IL-10, which also potently prevents morbidity and mortalities in acute infection (15–18). However, spleen and mesenteric cells from CD25+ cell-depleted mice had normal or elevated production of IL-4 and IL-10 ex vivo (Figures 2b,c and 3b). Furthermore, CD25+ cell-depletion may have not dramatically altered IL-4 and IL-10 production as there was a substantial population of IL-4- or IL-10-producing CD25 CD4+ cells in infected mice. In support of these data, a recent paper has also shown that both IL-4 and IL-10 is preferentially produced by CD25 CD4+ cells in schistosome infection (46).

Previous studies have already highlighted a potential role for CD4+CD25+ T cells and IL-10 producing T cells in schistosome infection (14,19,39). Hesse and coworkers demonstrated a central role for both IL-10-producing innate effectors and Treg cells in reduction of morbidity and mortality in chronic schistosomiasis (14). Within liver CD4+ granuloma cells of chronically infected mice (> 12 weeks), the main producer of IL-10 was the CD4+CD25+ subset, with these cells having a significant role in transferring protection from morbidity during infection (14). Despite showing elevation of CD4+CD25+ cells at 4, 8 and 16 weeks post-infection, depletion of CD25+ cells had no influence on any parameters studied in acute or chronic stages of schistosome infection. However, as the frequency of CD4+CD25+ cells was up-regulated at week 4 of infection, it is possible that CD4+CD25+ cells may have a role in the initial stages of infection.

The CD25 cell depletion regime used in this study caused a > 85% reduction in CD25+ cells, measured weekly over 4 weeks, in the spleen, mesenteric lymph nodes and liver granuloma of infected mice. This level of CD25 cell depletion is comparable to other in vivo studies where > 85% depletion was sufficient to show biological effects for CD4+CD25+ Treg cell activity (47–50). However, a recent study has issued questions on interpretation of in vivo depletion of CD25+ cells using the anti-CD25 mAb (PC61) used here (40). We have not addressed if there was functional inactivation of mice treated with PC61 (40). In this study, following anti-CD25 mAb treatment there was a ~30% reduction in Foxp3 expression on CD4+ cells, but we do not know if the intact Foxp3+ cells are functional in vivo and compensated for the absence of CD25+ cells. Others have highlighted that using CD25 as a target for selective depletion of Treg cells is confounded by the removal of activated effector CD25+ T cells as well as CD4+CD25+ naturally occurring Treg cells (51). Our data shows that in schistosome infection the increase in expression of CD25 on CD4+ cells is due to increased cell activation during infection rather than an expansion of CD4+CD25+ naturally occurring Treg cells. That we have seen no role for CD25+ cells in this study may indicate that during schistosome infection there are other additional regulatory mechanisms that function when CD25+ cells are depleted. The CD4+CD25Foxp3+ cell population that was increased in frequency when CD25 cells were depleted may have regulatory function, although we have not addressed this.

Foxp3 is considered to be a better marker than CD25, with some groups preferring to use it as an exclusive marker for Treg cells (45). However, despite the increase in spleen CD4+CD25+ cells following schistosome infection, we observed no increase in intracellular Foxp3+-expressing cells at week 4 or 8 post-infection, with only a slight increase by week 16 (Figure 1c). Recently, Singh and colleagues also have found that mRNA expression for Foxp3 in the spleens of 8 week-infected mice was comparable to mRNA levels in uninfected spleens, with an increase in Foxp3 mRNA in spleens of chronic (week 16) infected mice (39). As we used intracellular detection, we could discriminate the CD4+ subpopulation that expressed Foxp3. There was no increase in Foxp3 expression on CD4+CD25+ cells from acute 4 and 8 week-infected mice vs. levels in cells from uninfected mice, similar to what was seen by detection of Foxp3 mRNA (19).

During infection the expression of the Treg markers CD103 and GITR were predominately associated with the CD25+ subpopulation of CD4+ cells. In contrasts, infection induced a significant increase in a population of CTLA-4+ CD4+ cells that were CD25. When CTLA-4 was blocked in acute infection, mice suffered significant weight loss with no mortalities. Additionally, when we blocked CTLA-4 in mice with heavier infections (100 cercariae) we also observed increased IL-4 and IL-5 production and weight loss, but no increase in mortalities (data not shown). Weight loss in acute infection in CTLA-4 mAb treated mice was associated with selective effects on type 2 responses, with elevated IL-4 and IL-5 production, and significant increases in eosinophils and Th2 cells in liver granulomas. In other studies it has been shown that in schistosome infection of transgenic mice with an extreme type 2 polarized response there is exacerbated pathology, weight loss and mortalities (52,53). However, in this study the magnitude of increase IL-4 and IL-5 production was not as high as in studies in transgenic mice (17,53), which might explain the weight loss without overt severe pathology or mortalities. In contrast to acute infection, and despite having up-regulation in CTLA-4 expression on CD4+ cells, blocking mAb treatment during chronic infection had no effects on any parameter tested, with the exception of some elevated IL-5 production in unstimulated spleen cells.

Our observation of increased IL-4 and IL-5 when CTLA-4 is blocked in mice with acute schistosome infections is comparable to earlier studies in mice infected with the gastro-intestinal nematode Nippostrongylus brasiliensis (54). Anti-CTLA-4 mAb treatment of N. brasiliensis-infected mice caused a dramatic reduction in worm burden, with enhanced production of Th2 cytokines IL-4 and IL-5 in the mesenteric lymph nodes that drain the site of infection (54). In this study there was also an increase in IL-4 production in anti-CTLA-4 mAb-treated mice, involving an expansion of spleen and granuloma Th2 cells, and a consequential elevation in serum IgE. Surprisingly, the marked spontaneous production of IL-5 in the media of cultured spleen cells from anti-CTLA-4 mAb treated mice was not from CD4+, or from CD8+ or B cells. IL-5 may also be produced by non T cell sources, such as eosinophils and mast cells (55). Previously, eosinophils from S. mansoni-infected mice were shown to produce a major portion of Th2 cytokines, including IL-5, in the granuloma milieu (56). In view of the increase in granuloma eosinophils in anti-CTLA-4 treated mice it is possible that the spontaneous release of IL-5 is from eosinophils, but using intracellular cytokine staining we have been unable to as yet confirm this.

Patients infected with the filarial nematode Brugia malayi have elevated CTLA-4+ expression on CD4+ cells compared to uninfected individuals, with CTLA-4 expression associated with suppression of T cell reactivity (57). Steel and coworkers found that in vitro blocking of CTLA-4 in PBMC of infected patients resulted in a substantial increase in IL-5 expression and a decrease in IFN-γ production in response to microfilarial antigen (57), in contrast when PBMCs were exposed to live parasites blocking CTLA-4 restored both Th1 and Th2 cytokine genes (58). In the murine studies with schistosome infection described here, there was also an increase in CTLA-4 expression on CD4+ cells and we also observed an increase in IL-5 production when CTLA-4 was blocked in vivo. This is suggestive that a common mechanism in the regulation of type 2 cytokine biased responses evoked by helminth parasites may involve expression of CTLA-4 on CD4+ cells to suppress the magnitude of Th2 cell activity, as shown in N. braziliensis (54), B. malayi (57) and S. mansoni (this study).

In a pulmonary type 2 allergen-sensitization model blocking CTLA-4, using the same mAb and mouse strain as this study, induced elevated IL-4 and IL-5 production, increased IgE as well as elevated lung eosinophilia (59), which we observed in mice with CTLA-4 blocked during acute schistosome infection. Similarly, blocking CTLA-4 in a peanut allergen oral-sensitization model also up-regulated polyclonal Th2 cytokine responses and total serum IgE (60). However, whereas in the allergen pulmonary sensitization model (59) TGF-β was reduced, in our study we did not see a reduction in TGF-β production in cells from mice with CTLA-4 blocked, which may indicate that TGF-β is not involved in the CTLA-4-mediated mechanism of suppression in schistosome infection. Our data in schistosome infection using CTLA-4 mAb blocking experiments indicates that the increased expression of CTLA-4 on CD4+ cells may function to block excessive production of IL-4, and in particular IL-5.

Although expression of CTLA-4 on CD4+ cells is up-regulated in the chronic stages of infection, blocking mAb treatment had no major effect. Similar to acute infection, cells from chronically infected mice treated with anti-CTLA-4 mAb had some spontaneous in vitro IL-5 release, but had no weight loss or other changes. This could be due to CTLA-4 functioning to dampen the magnitude of the potent initial Th2 response evoked in the acute stages of infection. In contrast, in the down-modulated stages of chronic infection the Th2 response is reduced by CTLA-4-independent mechanisms. In support of this it has been shown that blocking CTLA-4 mAb treatment altered the initial induction of tolerance to allergen in mice, but once inhalation tolerance was established anti-CTLA-4 mAb had no effect (61).

In summary, we have shown that although there is induction of CD4+CD25+ cells in schistosome infection of mice, based on depletion with mAb we do not observe a critical role for these cells in acute or chronic infection. The increase in CD25+CD4+ but Foxp3 cells in schistosome infection is more indicative of cell activation rather than a regulatory activity. We have shown that S. mansoni infection also induces a marked CTLA-4+ cell population on CD4+ cells that were CD25. Blocking mAb treatment demonstrate that CTLA-4 plays a role in regulating the Th2 response, and thereby limiting type 2-associated responses such as eosinophilia, in acute infection but not chronic stages of infection. This study illustrates the complexity of regulation of T cells in schistosome infection and highlights a specific role for CTLA-4+, but not CD25+ cells, in the regulation of Th2 responses in helminth infection.

ACKNOWLEDGEMENTS

We are grateful to Niamh Mangan for assistance and comments on the manuscript. This work was supported by the Wellcome Trust, the Irish Research Council for Science, Engineering and Technology, and the Higher Education Authority Programme for Research in Third Level Institutions.

Ancillary