Functional Analysis of Effector and Regulatory T Cells in a Parasitic Nematode Infection

Abstract

Parasitic nematodes typically modulate T-cell reactivity, primarily during the chronic phase of infection. We analyzed the role of CD4-positive (CD4⁺) T effector (T_eff) cells and regulatory T (T_reg) cells derived from mice chronically infected with the intestinal nematode Heligmosomoides polygyrus. Different CD4⁺ T-cell subsets were transferred into naïve recipients that were subsequently infected with H. polygyrus. Adoptive transfer of conventional T_eff cells conferred protection and led to a significant decrease in the worm burdens of H. polygyrus-infected recipients. Roughly 0.2% of the CD4⁺ T cells were H. polygyrus specific based on expression of CD154, and cells producing interleukin 4 (IL-4) and IL-13 were highly enriched within the CD154⁺ population. In contrast, adoptive transfer of T_reg cells, characterized by the markers CD25 and CD103 and the transcription factor Foxp3, had no effect on the worm burdens of recipients. Further analysis showed that soon after infection, the number of Foxp3⁺ T_reg cells temporarily increased in the inflamed tissue while effector/memory-like CD103⁺ Foxp⁺ T_reg cells systemically increased in the draining lymph nodes and spleen. In addition, T_reg cells represented a potential source of IL-10 and reduced the expression of IL-4. Finally, under in vitro conditions, T_reg cells from infected mice were more potent suppressors than cells derived from naïve mice. In conclusion, our data indicate that small numbers of T_eff cells have the ability to promote host protective immune responses, even in the presence of T_reg cells.

Infections with parasitic nematodes have been shown to markedly modulate the host's immune system as a means to escape immune effector mechanisms and to ensure parasite survival within an immunocompetent host (35, 55). Highly effective immune evasion mechanisms are key elements of the long-lasting persistence of parasitic worms. Both parasite-specific and nonspecific immune suppression are well documented (19, 47, 60). Usually, acute infection stimulates antigen-specific T-cell proliferation, but with increasing exposure to parasite antigens, the immune system becomes hyporesponsive (34).

Immune responses against gastrointestinal nematodes include strong immunoglobulin G1 (IgG1) and IgE responses, eosinophilia, intestinal mastocytosis, goblet cell hyperplasia, and smooth-muscle hypercontractility (1, 12, 70). Th2-driven effector mechanisms were shown to be protective against gastrointestinal helminths, whereby interleukin 13 (IL-13) and IL-4 play an important role in primary, as well as secondary, infections (7, 13, 15, 53). Recently, alternatively activated macrophages have been introduced as an IL-4-dependent effector population essential for protective immunity to challenge infections with Heligmosomoides polygyrus (4). However, there is a lack of information on the activity of CD4-positive (CD4⁺) T-cell subsets during the chronic phase of primary nematode infections (when T-cell reactivity is suppressed) and on the T-cell subsets that contribute to host-protective or parasite-beneficial immune responses. In contrast, the interplay of T-cell subsets and regulation of effector responses in bacterial (28) and protozoan infections, like those with Leishmania (9, 38, 66) and Plasmodium (21, 44) species, are well defined.

It has been suggested that nematode infections reduce both Th1- and Th2-mediated responses by profoundly influencing regulatory pathways (5, 35). Regulatory T (T_reg) cells represent a subset of CD4⁺ T cells that are critically involved in balancing the reactivity of the immune system and preventing autoimmunity (40, 50). Alongside their role in preventing autoimmune reactions, T_reg cells have been shown to control excessive inflammatory responses against pathogens (20, 48). On the other hand, a strict control of T effector (T_eff) cell responses by T_reg cells can promote pathogen persistence (9, 28, 44, 67). Some cell markers used to identify T_reg cells are CD25, glucocorticoid-induced tumor necrosis factor receptor family-related gene (GITR), and the transcription factor forkhead box transcription factor P3 (Foxp3). In addition, the integrin α_E(CD103)β₇ is a marker for a subset of highly potent effector/memory-like T_reg cells, and CD4⁺ CD103⁺ cells were found to be the most potent suppressors of inflammatory processes in disease models, such as colitis and arthritis (22, 23, 30, 54). Here, we used CD25, CD103, and Foxp3 to identify T_reg cells. For the characterization of parasite-specific T_eff cells, we used CD154, a marker recently shown to exhibit exquisite specificity for antigen-activated T cells (15, 26).

To investigate the roles of different CD4⁺ T-cell subsets during the chronic phase of a primary nematode infection, we used the gastrointestinal nematode H. polygyrus, which resides in the proximal third of the mouse duodenum for up to several months during primary infection. Mice become infected by ingestion of infective larvae (L3) that invade the duodenal wall, where development to the L4 stage takes place. L4 reenter the gut lumen and mature to adults, which are chronically maintained during primary infection (42). We used H. polygyrus infection to address the role of CD4⁺ T_eff cells and T_reg cells with regard to worm expulsion and the role of T_reg cells in modulating T_eff cell function. Our data demonstrate that adoptive transfer of T_eff cells leads to protective immune responses in which antigen-specific CD4⁺ T cells produce predominantly IL-4 and IL-13. In contrast, CD4⁺ T_reg cells from chronic infection show no effect on the adult worm burden after adoptive transfer, although they are highly potent suppressors in vitro.

MATERIALS AND METHODS

Mice, parasites, and infection.

H. polygyrus was maintained by serial passage in BALB/c mice. The infective larvae (L3) were obtained from feces culture, extensively washed, and stored in distilled water at 4°C. BALB/c and C57BL/6 mice were purchased from the Bundesinstitut für Risikobewertung, Berlin, Germany. C57BL/6-Tg(CAG-EGFP)C15-001-FJ001Osb mice were a gift of M. Okawa, Osaka University. Male mice at 7 to 10 weeks of age were infected with 200 L3 using a gavage tube. Infection was surveyed by fecal egg count. The worm burden was determined by collecting adult worms from the small intestine on the day of dissection. Animals were housed and handled following national guidelines and as approved by our animal ethics committee.

Preparation of adult worm antigen.

Soluble worm antigen was prepared from adult worms kept in culture in RPMI medium containing 100 U/ml penicillin and 100 μg/ml streptomycin for 24 h. The worm material was homogenized and sonicated (1 min; 60 W) on ice in phosphate-buffered saline (PBS) (pH 7.4). The homogenate was centrifuged (20 min; 20,000 × g; 4°C), and the supernatant was passed through a 0.4-μm filter (Schleicher & Schuell, Germany) for sterilization. The protein content was determined by the bicinchoninic acid test (Pierce). Antigen extracts were stored at −80°C until they were applied.

Antibodies, staining, and sorting reagents.

The following antibodies and secondary reagents were purchased from BD Biosciences (Heidelberg, Germany): αCD4 fluorescein isothiocyanate (FITC)/PerCP (RM4-5), αCD103 phycoerythrin (PE) (M290), αCD25 allophycocyanin (APC)/PerCP-Cy5.5 (PC61), αCD3e (145-2C11), αCD28 (37.51), αCD40 (HM40-3), αIL-10 PE (JES5-16E3), αIL-4 FITC (11B11), α-gamma interferon (αIFN-γ) FITC (XMG1.2), αCD8 PE (53-6.7), SA-PE-Cy7, and SA-APC. αCD103 biotin (M290), αCD19 FITC (ID3), αFcR II/III (2.4G2), and α-digoxigenin PE were kindly provided by the German Arthritis Research Center (Berlin, Germany). αCD154 (CD40L) APC (MR1) was obtained from Miltenyi Biotec (Bergisch-Gladbach, Germany). αIL-13 (38213.11) was obtained from R&D systems (Wiesbaden, Germany) and coupled with digoxigenin at the German Arthritis Research Center. Foxp3 staining was performed using the PE α-mouse Foxp3 staining set (clone FJK-16S) purchased from eBioscience (San Diego, CA). αCD4, αCD90, αPE, and αAPC microbeads, as well as FITC MultiSort kits, were obtained from Miltenyi Biotec.

Phenotypic analysis of lymphocytes by flow cytometry.

Mesenteric lymph node cells (MLNC) and splenocytes from naïve and H. polygyrus-infected mice were dissociated by passing organs through a steel mesh in PBS, pH 7.4, containing 0.2% bovine serum albumin. In some experiments, intraepithelial lymphocytes (IEL) and lamina propria lymphocytes (LPL) were isolated from the small intestine. After visible Peyer's patches were removed, the small intestine was opened longitudinally; washed in PBS, pH 7.4; and incubated in RPMI at 37°C, 150 rpm for 40 min. The small intestine was washed twice in PBS, and the supernatants were collected for isolation of IEL. The organ was cut into pieces and incubated (40 to 50 min; 37°C; 150 rpm) with collagenases VIII and D (40 μg/ml each; both from Sigma), and then tissue was removed with a mesh. The supernatants were spun, and the cells were layered on a column of Percoll (GE Healthcare, Uppsala, Sweden) with a 40%-70% gradient. The cells were spun at room temperature and 2,200 × g for 20 min, and then collected from the interphase, washed, and kept in PBS-0.2% bovine serum albumin. For detection of changes in lymphocyte composition, cell suspensions (1 × 10⁶ total cells) were stained with αCD4, αCD8, and αCD19 monoclonal antibodies (MAbs). T_reg cells were detected by staining them for CD4, CD25, and CD103. Nonspecific binding of the MAbs was blocked by the addition of αFcgRII/III (20 μg/ml). Intracellular detection of Foxp3 was performed according to the manufacturer's instructions. For intracellular detection of CD154 and cytokines, cells were fixed in PBS containing 2% formaldehyde for 15 min at room temperature. After permeabilization with 0.5% saponin (Sigma), the cells were blocked with whole rat IgG (0.1 mg/ml) for 15 min at 4°C to reduce nonspecific binding of MAbs and stained with αCD154 and two of the α-mouse cytokine MAbs for 30 min at 4°C. For combined detection of CD154 and Foxp3, CD154 was stained on the cell surface directly during in vitro stimulation in complete culture medium as described below (see “Culture conditions”). Cytometric analysis was performed using FACSCalibur or LSRII (BD Biosciences) and FlowJo software (Tree Star, Inc.).

Isolation of T-cell subsets.

The separation of T-cell subsets for transfers and in vitro stimulation was performed as follows. Cells were stained for CD25 (APC) and CD103 (PE). CD25⁺ and CD103⁺ cells were enriched by the AutoMACS magnetic separation system using αAPC and αPE magnetic beads. For isolation of the different regulatory subsets, the bead-positive fraction was stained with FITC-labeled αCD4, and the CD4⁺ CD25⁺ CD103⁺ cells and CD4⁺ CD25⁺ CD103⁻ cell subsets were separated using a FACS Diva cell sorter (BD, Heidelberg, Germany). After complete removal of cells expressing CD25 and/or CD103, the negative fraction was used to isolate conventional CD4⁺ T cells using αCD4 beads. Naïve splenocytes depleted of T cells using αCD90 beads were irradiated (30 Gy) and used as antigen-presenting cells for in vitro assays. For some adoptive transfers, the whole CD4⁺ CD103⁺ subset (irrespective of CD25 expression) was isolated. Therefore, cells were stained for CD4 (FITC), CD25 (APC), and CD103 (PE). CD4⁺ cells were isolated using the FITC-MultiSort kit by AutoMACS. After removal of the beads, CD103⁺ cells were isolated from CD4⁺ cells using αPE beads.

Culture conditions.

Cell cultures were performed in cRPMI (BioChrom, Berlin, Germany) containing 10% fetal calf serum, 20 mM l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin as quadruplicates on 96-well plates. Culture of complete and T_reg cell-depleted MLNC was performed with 3.5 × 10⁵ cells for 72 h and a concentration of 12 μg/ml of adult worm antigen, and then the supernatants were harvested for cytokine detection. Culture of regulatory subsets and CD4⁺ CD25⁻ CD103⁻ responder cells for detection of polyclonal response to αCD3 stimulation (1 μg/ml) was performed with 2.5 × 10⁴ CD4⁺ cells and 5 × 10⁴ antigen-presenting cells per well. The cells were incubated for 48 h, followed by the addition of 1 μCi of [methyl-³H]thymidine (Amersham Pharmacia Biotech, Gent, Belgium) per well for 20 h to measure proliferation. In coincubation assays, the indicated numbers of T_reg cells were added to naïve responder cells and antigen-presenting cells and treated as described above. In some assays, bone marrow-derived dendritic cells (DC) were used as antigen-presenting cells, together with sorted CD4⁺ T-cell populations. For the generation of DC, bone marrow was isolated from tibias and femurs of 6- to 8-week-old naïve BALB/c mice, and the cells were kept in 24-well plates at a concentration of 1.5 × 10⁶/ml in cRPMI supplemented with 20 ng/ml of granulocyte-macrophage colony-stimulating factor (PeproTech, Hamburg, Germany) for 6 days, followed by incubation with 10 μg/ml of adult worm antigen for 12 h. Control cells were left untreated. DC (1 × 10⁴) were cultured for 72 h with 5 × 10⁴ T cells, and then the supernatants were harvested for cytokine enzyme-linked immunosorbent assay (ELISA) and the cells were snap-frozen in liquid nitrogen and stored at −80°C for real-time PCR. Recombinant mouse IL-2 (10 ng/ml; PeproTech, Hamburg, Germany) and 1 μg/ml αCD28 were added to some cultures for optimal stimulation. For measurement of CD154 (CD40L) expression of T cells and detection of cytokines in antigen-specifically activated cells, splenocytes and MLNC from infected animals were incubated in 24-well plates at a concentration of 4 × 10⁷ cells per ml with 20 μg/ml of adult worm antigen and 1 μg/ml αCD28 for 12 h. To survey cytokine production, brefeldin A (5 μg/ml; Sigma) was added after the first 2 h of stimulation. After 12 h, the cells were washed and prepared for flow-cytometric analysis as described above. For surface staining of CD154, cells were incubated as indicated above but without the addition of brefeldin A. αCD154-APC, αFcR II/III MAbs (20 μg/ml), whole rat IgG (10 μg/ml), and αCD40 (to avoid rapid removal of CD154 from the cell surface after binding of CD40 expressed on antigen-presenting cells) were added to the culture. After 12 h, the cells were washed and stained for CD4, CD103, and Foxp3. We omitted staining of CD25 in these assays due to the unreliability of the marker with respect to T_reg characterization after restimulation.

Histology.

Tissue samples from the proximal third of the small intestines of naïve and H. polygyrus-infected mice were fixed in 4% phosphate-buffered formalin, embedded in paraffin, and used for cross sections. Immunohistology for Foxp3-expressing cells was performed as described elsewhere (32, 33). Foxp3⁺ cells were counted in 10 high-power fields (40-fold magnification) randomly distributed in sections from each animal (Peyer's patch areas were excluded).

Cytokine analysis and quantitative PCR.

IL-4, IL-10, and IFN-γ in cell culture supernatants were quantified using OptEIA ELISA kits (BD Biosciences) according to the manufacturer's instructions. IL-13 and active transforming growth factor β1 (TGF-β1) were detected using DuoSets from R&D Systems. Active TGF-β1 was analyzed in culture supernatants without acidification. Transcript quantification by real-time PCR of IL-4 and IL-10 in distinct CD4⁺ T_eff and T_reg populations was performed after coincubation of T cells with naïve or H. polygyrus antigen-pretreated bone marrow-derived DC (see “Culture conditions” above). RNA extractions were performed using the RNeasy Mini Kit (Qiagen, Hilden, Germany), followed by digestion of DNA using the RNase-free DNase set (Qiagen) according to the manufacturer's instructions. RNA was reverse transcribed using the TaqMan reverse transcription reagent (Applied Biosystems, Warrington, United Kingdom) and oligo(dT)s. Quantitative real-time PCR was performed with the 7300 Real-Time PCR System (Applied Biosystems) using TaqMan reagents (Applied Biosystems). PCR amplifications were done in triplicates containing 3 μl of cDNA, 2 μl of 20× TaqMan-labeled primer mixture, and 10 μl of 2× TaqMan PCR buffer. The 20× TaqMan primer mixture consisted of two unlabeled PCR primers (900 nM [each] final concentration) and one 6-carboxyfluorescein dye-labeled TaqManMGB probe (250 nM final concentration). All primers were obtained from Applied Biosystems (IL-4 assay identifier, Mm00445259_m1; IL-10 assay identifier, Mm00439616_m1; GAPDH [glyceraldehyde-3-phosphate dehydrogenase] assay identifier, Mm99999915_g1). Real-time PCR was performed using the following conditions: 10 min of denaturation at 95°C, followed by 40 amplification cycles of 15 s at 95°C and 60 s at 60°C. The relative amounts of IL-10 and IL-4 mRNA were normalized to the endogenous reference GAPDH. Quantification of transcripts in cells cultured in the presence of DC pretreated with H. polygyrus antigen was done relative to cells cultured with naïve DC using the 2^−ΔΔCT method as described elsewhere (31).

Adoptive-transfer experiments.

Sorted CD4⁺ T-cell subpopulations (5 × 10⁵ cells per animal) were injected intraperitoneally into naïve mice in 0.2 ml of sterile PBS. Control animals received PBS only. One day after transfer, the mice were infected with approximately 200 L3 larvae. Four weeks after infection, animals were sacrificed, and the number of adult worms in each animal was determined and calculated as a percentage of the exact dose of applied L3 (set as 100%). The success of the infection was determined by surveying the fecal egg output starting on day 10 postinfection (p.i.). To survey cell survival and to trace transferred cells, the mice received 1 × 10⁷ carboxyfluoroscein succinimidyl ester (CFSE)-labeled or enhanced green fluorescent protein (EGFP)-expressing CD4⁺ cells. Reanalysis by flow cytometry was performed 6 days after transfer to the spleen, MLN, and small intestine.

Statistical analysis.

Statistical analysis was performed using GraphPad Prism software (San Diego, CA). Statistical significance as indicated in the figure legends was analyzed by either the Mann-Whitney test or analysis of variance (ANOVA), in combination with Bonferroni posttests.

RESULTS

Analysis of lymphocytes in MLNs of mice chronically infected with H. polygyrus.

To gather information about changes occurring in the MLN after infection with H. polygyrus, we analyzed the lymphocyte composition of the MLN in the chronic phase of infection (28 days p.i.). The absolute cell numbers obtained from the MLN showed a 5.2-fold increase (P < 0.05) in infected compared to naïve animals (Table 1). CD4⁺ and CD8⁺ T cells, as well as B cells, were analyzed by flow cytometry staining. We found a 4.7-fold (P < 0.05) increase in absolute CD4⁺ and a 4.3-fold (P < 0.05) increase in absolute CD8⁺ T-cell numbers. The strongest increase in total numbers, however, was found for B cells (8.3-fold; P < 0.05) (Table 1). Interestingly, the number of CD25⁺ CD103⁺ T cells increased 7.8-fold (P < 0.05) within the CD4⁺ T-cell compartment, surpassing the outgrowth of CD25⁺ CD103⁻ T cells (5.1-fold; P < 0.05) and CD25⁻ CD103⁻ T_eff cells (4.5-fold; P < 0.05) (Table 1).

TABLE 1.

Total MLNC numbers and lymphocyte composition of MLNs from naïve and H. polygyrus-infected animals

Cell type	No.		Increase (fold)
	Naïve	28 days p.i.
CD4+	9.32 × 106 ± 1.40 × 106b	43.54 × 106 ± 5.17 × 106b,c	4.7
CD25− CD103−a	8.28 × 106 ± 1.21 × 106	36.93 × 106 ± 4.35 × 106c	4.5
CD25+ CD103−a	0.79 × 106 ± 0.12 × 106	4.06 × 106 ± 0.29 × 106c	5.1
CD25+ CD103+a	0.21 × 106 ± 0.06 × 106	1.64 × 106 ± 0.28 × 106c	7.8
CD8+	2.89 × 106 ± 1.01 × 106	12.71 × 106 ± 3.19 × 106c	4.3
B cells (CD19+)	4.67 × 106 ± 2.12 × 106	38.99 × 106 ± 15.54 × 106c	8.3
Total	1.72 × 107 ± 0.29 × 107	8.96 × 107 ± 1.1 × 107c	5.2

^aCD4⁺ T-cell subset.

^bMean ± standard error of the mean of four (naïve) and five (28-day p.i.) animals. Similar data were obtained in four experiments.

^cStatistical significance for comparison of naïve to infected animals as determined by the Mann-Whitney test (P < 0.05).

Analysis of T_reg cells during the course of infection.

To further characterize the regulatory CD4⁺ T-cell compartment, we analyzed the frequency of T_reg cells in the MLNs and spleens of H. polygyrus-infected mice at different time points after infection. Flow cytometry analysis was performed after staining of CD4, CD25, CD103, and the transcription factor Foxp3. By detecting CD25 and CD103, we were able to distinguish between CD4⁺ CD25⁺ CD103⁻ naturally occurring and CD4⁺ CD25⁺ CD103⁺ effector/memory-like T_reg cells (Fig. 1A). Foxp3 expression levels were >90% in CD4⁺ CD25⁺ CD103⁻ cells at most time points analyzed (naïve mice and 3, 21, and 28 days p.i.), with the exception of 6 days p.i., showing a significant decrease in Foxp3⁺ cells within this compartment (naïve, 95.75% ± 0.05%; 6 days p.i., 86.72% ± 1.11%; P < 0.03), arguing for an increased proportion of recently activated effector cells present in MLNC at this time point (Fig. 1B). The CD4⁺ CD25⁺ CD103⁺ subset displayed Foxp3 expression levels of >96% at all time points analyzed (Fig. 1B and not shown), while few (<1.8%) CD4⁺ CD25⁻ CD103⁻ cells expressed Foxp3 (not shown). Comparing percentages of CD4⁺ Foxp3⁺ cells between naïve and infected animals (3, 6, 21, and 28 days p.i.) revealed no significant differences (Fig. 1E and not shown). Thus, the increase in total T_reg cells during infection reflects the increase in CD4⁺ cell numbers. However, we detected significant changes concerning the frequency of effector/memory-like T_reg cells after infection with H. polygyrus. Within the MLN draining the site of infection, we determined a significant increase (P < 0.03) in CD103⁺ T_reg percentages as early as day 6 days p.i. in comparison to naïve animals (Fig. 1C). Within the spleen, a significantly increased percentage was detected from day 12 p.i. onward (P < 0.03) (Fig. 1D). The higher frequency of T_reg cells expressing CD103 was stable in both lymphoid compartments until the chronic phase of infection (Fig. 1C and D) and returned to the level of naïve mice when adult worms were expelled after 8 to 12 weeks (data not shown). The level of CD25⁺ CD103⁻ T_reg cells showed no such increase (data not shown). When analyzing the ratio of CD103⁺ and CD103⁻ T_reg cells after gating on all Foxp⁺ cells, we rather determined a decrease in CD103⁻ cells in infected animals at the chronic stage, counterbalancing the increase in CD103⁺ cells (Fig. 1F). This finding argues for a conversion of naturally occurring T_reg cells into effector/memory-like T_reg cells, probably directly related to the infection.

FIG. 1.

Comparison of T_reg numbers in naive and H. polygyrus (H.p.)-infected mice. (A) T_reg cells were detected by flow cytometry based on surface expression of CD4, CD25, and CD103. Shown are plots derived from CD4⁺ MLNC of naïve and chronically infected animals. The data are representative of four animals per group and four independent experiments. (B) The predominantly regulatory phenotype of CD4⁺ CD25⁺ CD103⁻ (top) and CD4⁺ CD25⁺ CD103⁺ (bottom) cells derived from MLNs could be confirmed by intracellular staining of Foxp3. Four animals per group were analyzed in two independent experiments. (C and D) The frequencies of CD25⁺ CD103⁺ cells within CD4⁺ lymphocytes as detected in MLNs (C) and spleens (D) of animals at different time points after infection (black bars) compared to naïve controls (open bars). The data were derived from four animals per group, and analysis of T_reg cell numbers was performed at least twice for each time point. Means plus standard errors of the mean (SEM) are shown. (E) Frequencies of total Foxp3⁺ cells within CD4⁺ lymphocytes from MLN. Means plus SEM of four animals per group are shown. The data are representative of two independent experiments. (F) Proportion of CD25⁺ CD103⁻ and CD25⁺ CD103⁺ cells in CD4⁺ Foxp3⁺ cells. Means plus SEM of four animals per group are shown. The data are representative of two independent experiments. (G) Cross sections of the proximal third of the small intestine were stained for Foxp3-expressing cells. Foxp3⁺ T_reg cells were found in the epithelium and in the lamina propria (depicted by black arrowheads). The white arrowheads indicate tissue-dwelling larvae of H. polygyrus at day 6 p.i. (H) Significant increases in T_reg cells within the intestine were detected on days 3, 6, and 12 p.i. Stained cells in 10 high-power fields (HPF) (40-fold magnification) per animal were counted. Means of single animals (closed circles) and means of groups (horizontal lines) are shown. *, significant difference between naïve and infected animals as determined by the Mann-Whitney test (P < 0.05).

To test whether the changes in T_reg numbers found in the lymphatic organs were also reflected at the site of inflammation, we detected Foxp3-expressing cells in sections of the proximal third of the small intestine (Fig. 1G). In clear contrast to the findings in lymphatic organs, we detected a transient maximal accumulation of Foxp3⁺ cells at day 6 p.i. (P < 0.016), which gradually returned to the basal level seen in naïve animals at later time points of infection (Fig. 1H). The kinetics of Foxp3⁺ cells was mirrored by a maximal production of IL-10 and TGF-β1 in the small-intestinal tissue at day 6 p.i. (data not shown). Taken together, these data show that T_reg cells with an effector/memory-like phenotype increase strongly during infection with H. polygyrus in the MLN and spleen, while Foxp3-expressing cells accumulate only transiently at the site of infection.

Cytokine production of complete and CD25/CD103-depleted MLNC.

MLNC of H. polygyrus-infected mice were isolated at different time points after infection and cultured in the presence of H. polygyrus antigen to analyze their cytokine profile. Our data show that the key Th2 cytokines IL-4 and IL-13 in particular, as well as IL-10, were readily produced in high concentrations during the early phase of infection (6 days p.i.), accompanied by a weak IFN-γ response (Fig. 2A to D). Similar amounts of cytokines were found during the acute phase (12 days p.i.), except for IFN-γ, which was only marginally produced at this time point (Fig. 2A). At the chronic phase of infection (28 days p.i.), all cytokines analyzed were produced in smaller amounts, especially IL-13 and IL-10. Although it did not reach statistical significance, the trend toward lower cytokine production by MLNC from the chronic phase argues for a generally down-regulated parasite-specific response. Active TGF-β1 was detected only in very small amounts (<15 pg/ml) irrespective of the time point (data not shown). Interestingly, depletion of cells with a mainly regulatory phenotype (carrying the surface markers CD25 and CD103) (Fig. 2E) resulted in drastic changes in cytokine production of the remaining MLNC (Fig. 2A to D). First, in the early phase of infection (6 days p.i.), all cytokines analyzed were produced in smaller amounts after depletion of CD25⁺ and CD103⁺ cells. The finding of lower IL-4, IL-13, IL-10, TGF-β1 (not shown), and IFN-γ levels in cultures depleted of cells carrying CD25 and/or CD103 argues for depletion of not only T_reg cells, but also recently activated T_eff cells at this early time point. We could confirm the presence of effector cells within the CD4⁺ CD25⁺ CD103⁻ subset at 6 days p.i. by detection of Foxp3 expression (showing a decline at 6 days p.i.) (Fig. 1B) and by analyzing IL-4 and IL-10 mRNA levels in different CD4⁺ T-cell subsets kept in cocultures with DC presenting H. polygyrus antigens (Fig. 2F). We determined that CD4⁺ CD25⁺ CD103⁻ T cells expressed large amounts of IL-4 and IL-10 transcripts at this early time point, indicating the presence of T_eff and T_reg cells within this population. In contrast, the CD4⁺ CD25⁺ CD103⁺ T_reg population from the early infection exclusively expressed high levels of IL-10 (Fig. 2F). As we did not analyze the segregation of the IL-10 production to Foxp⁺ and Foxp3⁻ cells within the CD4⁺ CD25⁺ CD103⁻ compartment, we cannot exclude a contribution of recently activated effector cells to the IL-10 production at the early time point. Hence, the depletion of cells expressing CD25 and/or CD103 at the early time point removed a significant proportion of recently activated T_eff cells, as well as T_reg cells, as seen by diminished production of the Th2 cytokines IL-4, IL-13, and IL-10 by the remaining cells. The depletion had no effect with respect to IL-10 production at later time points (Fig. 2D), arguing that cells other than T_reg cells may also represent important IL-10 sources during the acute and chronic phases. Interestingly, the IL-4 response in the chronic phase of infection was more vigorous after depletion, indicating a suppressive effect of T_reg cells on the Th2 response in the chronic phase of infection (Fig. 2B).

FIG. 2.

Cytokine production of MLNC during H. polygyrus infection. (A to D) The cytokine response of MLNC after restimulation with H. polygyrus adult worm antigen was analyzed with cells from animals at days 6, 12, and 28 p.i. The cytokine response of complete MLNC (open bars) was compared to that of MLNC depleted of cells expressing CD25 and CD103 (hatched bars). IFN-γ (A), IL-4 (B), IL-13 (C), and IL-10 (D) production was analyzed. The means plus standard deviations (SD) of three cell pools of two animals each are shown. *, statistical significance comparing cytokine responses before and after depletion as determined by two-way ANOVA, followed by Bonferroni posttests (P < 0.05). The data are representative of two independent experiments. (E) The efficiency of T_reg depletion was determined by flow cytometry. The expression of CD25 and CD103 (upper row) and Foxp3 (lower row) in CD4⁺ cells before (left) and after (right) depletion is shown. (F) The relative expression of IL-4 and IL-10 mRNAs by CD4⁺ T-cell subsets after coincubation with H. polygyrus-primed DC was assessed at day 6 p.i. Means plus SD of three measurements are shown.

Adoptive transfer of CD4⁺ T cells from H. polygyrus-infected mice.

To analyze the functions of different CD4⁺ T-cell subsets, we performed adoptive transfers of CD4⁺ T_eff cells in comparison to CD4⁺ T_reg subsets derived from chronically H. polygyrus-infected mice (28 days p.i.). T_eff cells (CD4⁺ CD25⁻ CD103⁻) and CD4⁺ T cells with a regulatory phenotype, namely, CD4⁺ CD25⁺ CD103⁻ and CD4⁺ CD25⁺ CD103⁺ T cells, were transferred to naïve recipients. Control animals received PBS only. One day after transfer, the recipients were infected with a defined dose of larvae. The purity of T_eff cells was >98% for expression of CD4⁺, with <3% remaining CD25⁺ and/or CD103⁺ cells, whereas the T_reg compartments were >90% CD4⁺ CD25⁺ CD103⁻ cells and >89% CD4⁺ CD25⁺ CD103⁺ cells, respectively (Fig. 3A). Four weeks after infection, the adult worm burdens of recipients and control animals were assessed. We detected a significant reduction of 43.7% (P < 0.008) in the worm burden in animals receiving T_eff cells (CD4⁺ CD25⁻ CD103⁻) compared to the PBS control group (Fig. 3B). In contrast, transfer of CD4⁺ CD25⁻ CD103⁻ T cells from naive mice had no influence (data not shown), indicating that the reduction was due to parasite-specific T_eff cells. The protective role could be solely ascribed to CD4⁺ T cells, as the transfer of CD4⁻ cells had no effect (data not shown). Animals receiving CD4⁺ CD25⁺ CD103⁻ T_reg cells (comprising >90% Foxp3⁺ cells) had worm burdens comparable to that of the PBS control. Similarly, the transfer of CD4⁺ CD25⁺ CD103⁺ T_reg cells (comprising >95% Foxp3⁺ cells) did not result in significant changes. In contrast, transfer of a heterogeneous T-cell population containing T_reg and T_eff cells, namely, CD4⁺ CD103⁺ T cells (with a purity of >95% CD4⁺ cells and >85% CD103⁺ cells but comprising only about 70% Foxp3⁺ cells) led to a significant decrease in worm numbers (P < 0.03) comparable to pure CD4⁺ T_eff cells (Fig. 3B). We verified cell survival in separate experiments using CFSE-labeled CD4⁺ cells (data not shown). In addition, we performed transfers with EGFP-expressing CD4⁺ cells from chronically infected donors. The recipients were infected the following day. Six days after transfer, spleen and MLN, as well as IEL and LPL from the small intestine, were analyzed for transferred cells (Fig. 3C and D). We were able to trace EGFP⁺ cells in all analyzed organs with no significant accumulation in any of the compartments. The slightly elevated numbers of EGFP⁺ cells within MLNC compared to splenocytes merely reflects the higher CD4⁺ T-cell numbers found in lymph nodes (Fig. 3D). We also analyzed percentages of T_reg cells according to expression of CD25, CD103, and Foxp3 within the EGFP⁺ population and could show that T_reg cells survived the transfer (data not shown).

FIG. 3.

Influence of adoptive CD4⁺ T-cell transfer on adult worm burden. T cells of the indicated subtypes (5 × 10⁵) obtained from MLNs and spleens of chronically H. polygyrus-infected mice were transferred to recipients that were subsequently infected with H. polygyrus larvae. Control animals received PBS only. (A) Purity of transferred cells as determined by flow cytometry. Representative data from one of three independent experiments are shown. (B) Adult worm burden in recipients 28 days p.i. Worm counts are shown as percentages of the number of applied larvae. Group sizes varied between 5 and 20 animals. The data originated from three individual experiments. Individual worm counts and medians are shown. The asterisks show statistical significance as determined by a Kruskal-Wallis test followed by a Mann-Whitney test: *, P < 0.05; **, P < 0.01. (C) Tracing of transferred cells in C57BL/6 mice receiving 1 × 10⁷ CD4⁺ cells from chronically infected EGFP-expressing donors. The recipients were infected with H. polygyrus the following day. Examples of flow cytometry plots derived from splenocytes, MLNC, and small-intestinal IEL and LPL 6 days after transfer are shown. (D) Percentages of EGFP⁺ cells within lymphocytes of recipients. Means plus standard deviations for four animals are shown. The data are representative of two experiments.

Hence, CD4⁺ T_eff cells from the chronic phase of an H. polygyrus infection transferred protection to naïve recipients in contrast to naturally occurring or effector/memory-like T_reg cells, which had no intrinsic effect on worm development. Furthermore, transfer of a mixed T-cell population comprising effector and regulatory T cells showed that T_reg cells were not efficient in suppressing the protective effect mediated by T_eff cells.

Cytokine production of CD4⁺ T-cell subsets in the chronic phase of infection.

We then analyzed the cytokine production of the CD4⁺ T-cell subsets used for the transfer experiments to gain information about mediators involved in worm expulsion or establishment. H. polygyrus-treated DC were used as antigen-presenting cells to stimulate the T-cell subsets. We determined that T_eff cells (CD4⁺ CD25⁻ CD103⁻) released IL-4 and IL-13 in the presence of H. polygyrus-primed DC in contrast to T_reg cells (CD4⁺ CD25⁺ CD103⁻ or CD4⁺ CD25⁺ CD103⁺) (Fig. 4A and B). However, CD4⁺ CD25⁺ CD103⁺ T cells, in particular, released significant amounts of IL-10, arguing for the effector/memory-like T_reg cell subset as an important source of IL-10 (Fig. 4C). Incubation of the CD4⁺ subsets with naïve DC resulted in marginal release of cytokines by T cells, clearly showing the antigen specificity of the cytokine response. Analysis of active TGF-β1 in culture supernatants revealed only marginal increases in cultures with antigen-loaded DC compared to those with naïve DC (data not shown). The generally lower cytokine levels in cultures of separated T_eff cells compared to complete MLNC cultures (Fig. 2) is probably due to fewer CD4⁺ T cells present in DC cocultures than in preparations of whole MLNC and to other cytokine-producing cells present in MLNC cultures, like basophils, mast cells, and eosinophils (16).

FIG. 4.

Cytokine production by CD4⁺ T-cell subsets. CD4⁺ T cells were isolated from pooled MLNC and splenocytes of eight mice in the chronic phase of infection (28 days p.i.) according to the indicated surface marker expression. T-cell subsets were incubated for 72 h with naïve bone marrow-derived DC (nDC) or DC pretreated with H. polygyrus adult worm antigen (HpDC). (A to C) Release of IL-4 (A), IL-13 (B), and IL-10 (C) was detected by ELISA. Means plus standard deviations (SD) of triplicate determinations are shown, and the data are representative of three independent experiments. (D to F) Release of IL-4 (D), IL-13 (E), and IL-10 (F) after addition of recombinant mouse IL-2 and αCD28 to cultures. Mean values plus SD of triplicate determinations of one of two independent experiments with similar results are shown.

While only CD4⁺ CD25⁺ CD103⁺ T cells produced IL-10 in an antigen-specific manner, we found that both CD25-expressing T-cell subsets produced large amounts of IL-10 when optimized conditions were provided by addition of exogenous IL-2 and enhanced costimulation through αCD28 antibodies (Fig. 4F). Similarly, IL-4 and IL-13 were produced in much larger amounts almost exclusively by CD4⁺ CD25⁻ CD103⁻ T_eff cells after enhanced costimulation was provided (Fig. 4D and E). Levels of active TGF-β1 were only marginally affected (data not shown). None of the CD4⁺ T-cell subsets from H. polygyrus-infected mice was found to produce relevant amounts of the Th1 cytokine IFN-γ (data not shown). Hence, our data indicate a dominant and parasite-specific Th2 response by CD4⁺ effector cells from the chronic phase of infection, while effector/memory-like CD4⁺ CD25⁺ CD103⁺ T_reg cells probably produce IL-10 in response to H. polygyrus antigen.

Distribution of antigen-specific CD4⁺ T cells and their cytokine production.

To assess the distribution of H. polygyrus-specific CD4⁺ T cells within the T_eff and T_reg cell populations, we examined the expression of CD154 as a marker of antigen-specific CD4⁺ T-cell activation (15, 26). As the detection of CD154 directly ex vivo is not possible, we used an optimized in vitro protocol (26). Comparison of CD4⁺ T cells from spleen and MLN showed that in both sites, about 0.2% (MLN, 0.233 ± 0.036 naïve versus infected [P < 0.001]; spleen, 0.246 ± 0.040 naïve versus infected [P < 0.0003]) of CD4⁺ CD103⁻ T_eff cells expressed CD154 when restimulated with adult worm antigen in vitro (Fig. 5A and B). By additional detection of CD103 and Foxp3, we were able to distinguish between CD154 expression on T_eff and T_reg cells.

FIG. 5.

Distribution of antigen-specific CD4⁺ T cells (CD154⁺) and their cytokine responses. (A) Shown are examples of fluorescence-activated cell sorter plots of MLNC from naïve mice (upper row) and H. polygyrus-infected mice at day 28 p.i. (lower row). The cells were restimulated with H. polygyrus antigen in vitro and stained for CD4, CD154, CD103, and Foxp3. CD4⁺ cells were plotted for expression of CD103 and Foxp3 (left). The middle plots show CD154 expression within the CD4⁺ CD103⁺ Foxp3⁻ effector population. The right-hand plots show CD154 expression by CD4⁺ CD103⁻ Foxp3⁻ effectors. (B and C) Frequencies of CD154-expressing cells within the CD4⁺ CD103⁻ Foxp3⁻ (B) and CD4⁺ CD103⁺ Foxp3⁻ (C) effector populations of MLNs (black bars) and spleens (open bars) after restimulation in vitro. The data were obtained from eight infected and seven naïve mice. Means plus standard errors of the mean (SEM) are shown. The data are representative of three independent experiments. (D) Intracellular staining of cytokines within restimulated CD4⁺ T cells from spleens (28 days p.i.). The cells were gated for expression of CD4 and CD154 (left), and IL-4 and IL-13 (upper row) or IFN-γ and IL-10 (lower row) were determined in CD4⁺ CD154⁻ cells (center) and CD4⁺ CD154⁺ cells (right). The cytometry plots shown are representative of a group of eight infected mice. (E) IL-4/IL-13 and IFN-γ/IL-10 (F) responses in CD4⁺ CD154⁺ cells after restimulation with H. polygyrus antigen. Means plus SEM of eight animals are shown. The asterisks show statistical significance comparing cells from naïve and infected animals as determined by the Mann-Whitney test: **, P < 0.01; ***, P < 0.001.

Interestingly, when analyzing the level of CD154 on CD103⁺ cells not coexpressing Foxp3 and therefore not of regulatory phenotype, we found clearly higher percentages of antigen-specific cells within this T_eff subpopulation in MLNs (1.701% ± 0.306% CD4⁺ CD103⁻ versus CD4⁺ CD103⁺; P < 0.0006) and spleens (0.586% ± 0.144% CD4⁺ CD103⁻ versus CD4⁺ CD103⁺; P < 0.01) (Fig. 5A and C) of infected animals in comparison to noninfected mice. In contrast, only low levels of CD154 expression were detected on Foxp3⁺ T_reg cells from MLN (<0.03%) and spleen (<0.05%) (data not shown). These data indicate that mice that received T cells sorted for coexpression of CD4 and CD103 (irrespective of CD25 expression) in transfer studies (Fig. 3B) received a highly enriched T_reg population that still contained a small number of antigen-specific effector cells not expressing Foxp3. The percentage of antigen-specific effectors within the transferred CD4⁺ CD103⁺ cell compartment was as low as 0.2%, according to CD154 expression, comparable to the values for antigen-specific cells within the CD4⁺ CD103⁻ effector compartment. These CD103⁺ effectors may have led to protection, even though a large number of T_reg cells was cotransferred, arguing for an insufficient capacity of the T_reg cells to control the antigen-specific effectors.

We next determined the cytokine production of the antigen-specific CD154⁺ CD4⁺ T cells. Investigation of IL-4, IL-13, IL-10, and IFN-γ production after in vitro restimulation (Fig. 5D) revealed a dominant Th2 response in CD154⁺ CD4⁺ T cells of chronically infected mice, characterized by high expression levels of IL-4 (24.55% ± 1.32%), IL-13 (10.48% ± 1.99%), or both (10.26% ± 1.02%) (Fig. 5D and E). Only low levels of IFN-γ-producing cells were detected in the antigen-specific T-cell compartment (4.47% ± 0.82%), and the frequencies of IL-10-producing cells were hardly distinguishable from nonspecific background (1.60% ± 0.37%) (Fig. 5D and F). These data again indicate that the lower worm burden in recipients receiving CD103-expressing CD4⁺ T cells might be due to cotransfer of antigen-specific effector cells that were able to produce IL-4 and IL-13, thereby mediating worm expulsion.

Suppressive effect of T_reg cells from H. polygyrus-infected mice in vitro.

To further investigate T_reg cells from worm-infected animals, we analyzed the suppressive activities of sorted T_reg subsets in vitro. CD4⁺ CD25⁺ CD103⁺ and CD4⁺ CD25⁺ CD103⁻ T_reg cells were isolated from naïve and worm-infected mice in the acute (12 days p.i.) and chronic (28 days p.i.) phases of infection. The cells were added to T_reg-depleted CD4⁺ T cells from naïve mice, which were stimulated polyclonally by αCD3 antibodies. CD4⁺ CD25⁺ CD103⁺ T_reg cells from infected mice suppressed the proliferation of naïve responder CD4⁺ T cells more vigorously than their counterparts from naïve controls (Fig. 6A and Table 2). For the lowest ratio of CD4⁺ CD25⁺ CD103⁺ T_reg cells to responder cells (1:20), reflecting the in vivo situation, we found that T_reg cells from the chronic phase of infection exhibited the highest suppressive efficiency (P < 0.001 compared to naïve cells and day 12 p.i.). The CD4⁺ CD25⁺ CD103⁺ T_reg cells derived from infected animals were more efficient in mediating suppression than CD4⁺ CD25⁺ CD103⁻ T_reg cells (P < 0.03 for all tested ratios) (Fig. 6B). As expected, both T_reg subsets showed an anergic phenotype after αCD3 stimulation (data not shown). These data clearly indicate the high in vitro suppressive capacity of CD4⁺ CD25⁺ CD103⁺ T_reg cells derived from the chronic phase of infection with regard to activation and proliferation of CD4⁺ T cells.

FIG. 6.

Suppressive capacities of T_reg cells in vitro. CD4⁺ CD25⁺ CD103⁻ and CD4⁺ CD25⁺ CD103⁺ T_reg cells were purified from splenocytes and MLNC of H. polygyrus-infected (12 and 28 days p.i.) and naïve mice. CD4⁺ CD25⁻ CD103⁻ responder cells were isolated from naïve animals. (A and B) The suppressive capacities of CD4⁺ CD25⁺ CD103⁺ T_reg cells (A) and of CD4⁺ CD25⁺ CD103⁻ T_reg cells (B) from naïve mice (open bars) and infected animals at the acute phase (12 days p.i.) (gray bars) or chronic phase (28 days p.i.) (black bars) of infection were analyzed. The ratios of T_reg cells and responder CD4⁺ T cells are indicated. Proliferation of CD4⁺ T cells after stimulation with αCD3 antibodies was detected by [³H]thymidine uptake. Means plus standard errors of the mean of quintuple determinations are shown. The data shown are representative of two independent experiments.

TABLE 2.

Statistical analysis of suppressive efficiencies of T_reg cell subsets from naive and H. polygyrus-infected mice

Treg/CD4Respa ratio	CD4+ CD25+ CD103−b			CD4+ CD25+ CD103+b
	Naïve vs. 12 days p.i.	Naïve vs. 28 days p.i.	12 vs. 28 days p.i.	Naïve vs. 12 days p.i.	Naïve vs. 28 days p.i.	12 vs. 28 days p.i.
1:20	NS	P < 0.001	P < 0.01	P < 0.001	P < 0.001	P < 0.001
1:10	P < 0.001	P < 0.001	NS	P < 0.001	P < 0.001	NS
1:5	NS	P < 0.01	NS	P < 0.001	P < 0.001	NS
1:2	P < 0.001	NS	P < 0.05	NS	NS	NS
1:1	NS	NS	NS	NS	NS	NS

^aRatio of T_reg cells to naïve CD4⁺ responders.

^bT_reg cell subset analyzed for suppressive efficiency. Statistical significance as determined by two-way ANOVA, followed by a Bonferroni posttest. NS, not significant.

DISCUSSION

In this study, we investigated the roles of different CD4⁺ T-cell subsets in a chronic intestinal worm infection. Adoptive transfer of CD4⁺ cells from the chronic phase of an H. polygyrus infection containing small numbers of antigen-specific T_eff cells led to significant protection in recipients. In contrast, adoptive transfer of T_reg cells did not alter the worm burden, although T_reg cells significantly suppressed CD4⁺ T-cell proliferation in vitro. Hence, our data clearly point to the importance of CD4⁺ T_eff cells in mediating host protection and worm expulsion. Our data provide evidence for the following traits of T_eff/T_reg interaction. First, T_eff cells that arise during a primary H. polygyrus infection exhibit the ability to mediate protection. Second, T_eff cells persist in chronically infected mice, despite strong cellular suppression (10, 49, 52). Third, about 0.24% of CD4⁺ T_eff cells are antigen-specific cells, based on expression of CD154, containing highly enriched frequencies of IL-4- and IL-13-producing cells responding to parasite antigens. Fourth, T_reg cells are probably an important source of IL-10 during chronic infection. Fifth, elevated numbers of effector/memory-like T_reg cells persist in lymphatic organs until the chronic phase, and T_reg cells derived from infected animals show an augmented suppressive capacity in vitro.

Our study revealed a significant and permanent increase in effector/memory-like (CD103⁺) T_reg cell numbers in lymphatic organs as early as day 6 p.i. However, only a transient increase in T_reg cells was detected at the site of inflammation, which might represent a host reaction to control inflammatory responses induced by innate immune and T_eff cells. The peak of T_reg cells in the small intestine at day 6 p.i. is in accordance with the time point of intense inflammation around tissue-invading larvae, characterized by accumulation of mainly granulocytes and, to a lesser extent, macrophages and CD4⁺ T cells (4, 43). Interestingly, it has been shown that T_reg cells not only control inflammatory responses directly or indirectly driven by T_eff cells, but also suppress innate immune responses in the absence of T_eff cells (36). Therefore, it is conceivable that T_reg cells induced by the nematodes are recruited to modulate innate and adaptive immune responses. The resulting immunosuppression could control excessive pathology and favor prolonged parasite survival. Helminth-induced T_reg cells have been shown to be involved in immunomodulation in various helminth infections, such as human infections with Onchocerca volvulus (51) or murine infections with Litomosoides sigmodontis (59), H. polygyrus (14), and Schistosoma mansoni (6, 29, 37). The general notion is that during helminth infections, T_reg cells might have the function of facilitating parasite survival and diminishing immunopathology (34, 35), but their concrete role is not yet fully elucidated.

We provide evidence that during the early phase of infection, activated T_eff and T_reg cells are both important sources of IL-10. However, during the chronic phase of infection, T_reg cells probably represent the main T-cell source of IL-10. Hence, our data indicate that not only T_reg cells, but also effector cells provide IL-10 that could have a role in dampening exaggerated inflammation, as shown previously for Th1 and Th2 effector cells (3, 8, 25, 61). Whether T_reg cells in nematode infections mediate suppression via IL-10, TGF-β, or both is still under investigation (8, 27, 39, 59). Increases in TGF-β1 expression by CD4⁺ and CD4⁻ cells and surface-bound TGF-β1 on CD4⁺ cells, as well as increased plasma levels of active TGF-β1 during infection with H. polygyrus, have been shown by others (14, 56, 68). In our system, T_reg cells from the chronic stage of infection seem to exert their effects via IL-10, supported by the fact that we detected only low levels of active T_reg-derived TGF-β1. Of note, our analysis of the TGF-β1 production by CD4⁺ T-cell subsets is restricted to the secreted active form, and currently we cannot exclude differences with respect to the secreted inactive form of the molecule. By analyzing tissue samples from the small intestine, we found elevated TGF-β1 levels at the site of infection in the early phase (6 days p.i.) (data not shown), perhaps indicating tissue repair in this phase of active inflammation.

In addition to T cells, B cells can also serve as major sources of IL-10 (45, 46), and B cells in protozoan and helminth infections have been shown to produce IL-10 in response to pathogen-derived antigens (2, 17, 46). To date, few data on B cells as a source of IL-10 in intestinal nematode infections are available. We detected large increases in B-cell numbers in MLNs of infected animals, and further analysis of these cells as possible source of IL-10 during H. polygyrus infection is needed.

This study showed that especially CD4⁺ CD25⁺ CD103⁺ T cells from mice chronically infected with H. polygyrus produce IL-10 in a parasite-specific manner. The high frequency of Foxp3-expressing cells within this T-cell population strongly suggests effector/memory-like T_reg cells as a parasite-specific source for IL-10, although we cannot completely rule out the possibility that Foxp3⁻ effector cells or adaptive T_reg cells may contribute to the detected IL-10 production. In mice infected with S. mansoni, CD4⁺ CD25⁺ T cells, whether Foxp3⁺ or Foxp3⁻, have been identified as sources of IL-10 (6, 58).

We revealed a disproportionate increase in CD25⁺ CD103⁺ T_reg cells within the CD4⁺ T-cell compartments of infected animals. Similar to what has been described for infection with S. mansoni (6), we detected only minor changes in the total T_reg cell numbers as determined from Foxp3 expression in lymphatic organs. However, we found a significant increase in cells expressing CD103 within the regulatory T-cell compartment, arguing for a specific role of these effector/memory-like T_reg cells in modulating the immune response to H. polygyrus. To test this hypothesis, we performed suppression assays with responder CD4⁺ T cells from naïve mice. These in vitro assays demonstrated that especially CD4⁺ CD25⁺ CD103⁺ T_reg cells isolated from the chronic phase of infection strongly suppressed the proliferation of CD4⁺ T cells when added in a physiological ratio of T_reg to responder cells (1:20). A particularly potent suppressive capability of CD4⁺ CD25⁺ CD103⁺ T_reg cells in comparison to CD4⁺ CD25⁺ CD103⁻ T_reg cells in vitro and in vivo has also been shown by others (30, 57). We expected that adoptive transfer of T_reg cells might interfere with the primary Th2 response to H. polygyrus, entailing increased worm burdens, as studies on protective immunity against H. polygyrus have shown that severe combined immunodeficient mice lacking T and B cells and mice depleted of CD4⁺ T cells harbor higher worm burdens than wild-type or untreated controls (63, 64). However, in spite of the high suppressive capacity of CD103⁺ T_reg cells in vitro, our adoptive-transfer model did not reveal an influence of T_reg cells on the worm burden in vivo. Transfer of CD4⁺ CD25⁺ CD103⁺ T_reg cells, as well as CD4⁺ CD25⁺ CD103⁻ T_reg cells, did not alter worm numbers, whereas transfer of a mixture of T_reg and T_eff cells (CD4⁺ CD103⁺ T cells) significantly lowered worm burdens, arguing against a major control of effector responses by T_reg cells in vivo. The discrepancy between the suppressive activity of CD103⁺ T_reg cells in vitro and the failure to interfere with worm expulsion in vivo might be due to multiple factors. First, T_reg cells might be involved in suppressing pathology, but not in facilitating worm survival. Second, the transferred CD4⁺ CD103⁺ T-cell population represented a heterogeneous pool of T_reg and T_eff cells containing the lowest percentage of Foxp3⁺ cells (<70%) in comparison to the CD4⁺ CD25⁺ CD103⁺ and CD4⁺ CD25⁺ CD103⁻ subsets (both ≥90%). Third, reduction of the adult worm burden in recipients receiving CD4⁺ CD103⁺ T cells coincided with a significant proportion of parasite-specific effector cells within the CD103⁺ Foxp3⁻ cell compartment, as shown by expression of CD154 after antigen stimulation. Finally, the conditions in vitro versus in vivo were clearly different. In vitro, T_reg cells drastically inhibited the proliferation of naïve CD4⁺ responder T cells after polyclonal stimulation, whereas in vivo, T_reg cells had to combat antigen-specifically activated T_eff cells.

We have provided data on the distribution and cytokine profile of parasite-specific CD4⁺ cells using the marker CD154, recently shown to exhibit exquisite specificity for antigen-activated CD4⁺ cells in the human and mouse systems (15, 26). Our approach revealed low percentages (<1%) of antigen-specific cells in the chronic phase, producing predominantly IL-4 and IL-13. Although cytokine-producing CD4⁺ T cells were also detected at low percentages among CD154⁻ cells, the responders to worm antigens were highly enriched within the CD154⁺ population. In contrast, much higher frequencies of IL-4-producing cells are described for infections with H. polygyrus in GFP-IL-4 reporter mice (41). Several factors may have led to the marked discrepancies. First, Mohrs et al. reported that not all GFP⁺ cells secrete IL-4 in response to stimulation with H. polygyrus antigen extracts and provided evidence that CD4⁺ T cells recently producing IL-4 in vivo are impaired in their cytokine response in vitro (41). This finding might, in part, explain the lower frequencies of antigen-specific IL-4 producers detected in our in vitro assays. Second, different conditions, such as the infection time point analyzed, the source of antigen-presenting cells, in vitro cell numbers during restimulation, or antigen dose, might have led to an underestimation of specific CD4 T-cell numbers based on CD154 expression. However, despite the marked difference between the frequencies of antigen-specific CD4⁺ T cells presented here and for the IL-4 reporter system, our data show that the CD154 technique facilitates the analysis of a broader spectrum of pathogen-specific cytokine responses in a nonmanipulated immune system.

Adoptive transfer of CD4⁻ cells did not show an effect on the worm burden (data not shown). This observation further supports the concept that resistance to gastrointestinal helminths is dependent on CD4⁺ Th2-type immune responses, as shown in animal models with H. polygyrus or Nippostrongylus brasiliensis (11, 64, 65), as well as in humans infected with Ascaris lumbricoides and Trichuris trichiura (24, 62).

Hence, our study is in line with the recent finding that memory CD4⁺ T cells develop after infection with the gastrointestinal nematode Trichuris muris and mediate protection against rechallenge (69). In a recent publication introducing alternatively activated macrophages as an effector population essential for protective immunity to challenge infections with H. polygyrus, it was shown that memory Th2 cells derived from H. polygyrus-cured mice transferred protection against a primary infection (4). The protective effect of Th2 memory cells was more pronounced than what we found in our transfer model using T_eff cells from an ongoing infection. It would be interesting to investigate whether differences in, e.g., the homing receptor repertoires or susceptibility to induced cell death between memory and effector T cells might be responsible for this phenomenon.

In conclusion, our data indicate that in our model system, tipping the balance of effector T cells during infection strongly influences the survival of parasitic nematodes, while T_reg cells may have functions that are not directly related to worm persistence.