Hookworm Excretory/Secretory Products Induce Interleukin-4 (IL-4)⁺ IL-10⁺ CD4⁺ T Cell Responses and Suppress Pathology in a Mouse Model of Colitis

A. caninum adult worms were cultured in serum free medium containing 100 U of penicillin/μl and 100 μg of streptomycin/ml (pen/strep) for 24 h. The supernatant (AcES) was collected, filter sterilized through a 0.22-μm-pore-size filter (Pall), and concentrated and buffer exchanged to phosphate-buffered saline (PBS) using a 10-kDa spin column (Pall). Removal of lipopolysaccharide from AcES and ovalbumin (OVA; Sigma) was then performed using one of two methods, Endotrap Blue (Hyglos) according to the manufacturer's instructions or Triton X-114 (Sigma) as previously described (20) with some minor changes. Briefly, AcES was incubated with 5% Triton X-114 at 4°C on a rotating wheel for 30 min, followed by heating to 37°C for 10 min and centrifugation at 1,600 × g for 15 min at room temperature. The upper endotoxin-depleted phase was collected, and the process was repeated twice to ensure thorough removal of endotoxin. A Limulus amebocyte lysate (Lonza) assay was used to ascertain the adequate removal of endotoxin, and the protein concentration was calculated by using a micro-BCA protein assay kit (Pierce). Some experiments used boiled and trypsinized AcES (bES) as a control. Briefly, AcES was digested with 1 μg of trypsin (Sigma)/ml at 37°C for 24 h, followed by boiling at 95°C for 15 min to denature both trypsin and the AcES protein constituents.

Female 6- to 10-week-old C57BL/6 mice were purchased from the Animal Resources Centre (Perth, Australia) and were housed according to Australian animal rights and regulations standards. Mice received food and water ad libitum. All injections were administered via the intraperitoneal (i.p.) route without adjuvant. In some experiments, mice received i.p. injections of PBS or 1 to 25 μg of AcES, bES, or OVA at various time points as indicated in the text. All procedures involving mice were approved by the James Cook University Animal Ethics Committee.

A 3.5% (wt/vol) solution of dextran sodium sulfate (DSS; 36,000 to 50,000 molecular weight; MP Biomedicals) was administered to mice as a substitute for normal drinking water. The mice were weighed and scored daily to assess disease progression based on a modified scoring system (21). Mice were scored on weight (percent change; 0 to 4), the level of fecal consistency (0 to 4), rectal bleeding (0 to 2), and general appearance (0 to 3) for a daily score out of a total of 13.

Upon termination of the experiment the mouse colons were given a macroscopic score for severity of adhesion (0 to 3), ulceration (0 to 3), wall thickening (0 to 3), and edema (0 to 3) for a total possible score of 12 as previously described (22). A small piece of the proximal colon was fixed in 4% formaldehyde for histological processing. Cross-sections of the colons were stained with hematoxylin and eosin (H&E) for microscopic visualization of inflammation. Histological scoring of the cross-sections was performed in a blinded fashion using a modified scoring system (23). Colon cross-sections were assessed for number of ulcers (no ulcers = 0, 1 ulcer = 1, 2 ulcers = 2, 3 ulcers = 3, and >3 ulcers = 4). Each ulcer was ∼200 μm in length; where ulceration was bigger than this, scoring was performed in 200-μm intervals. Epithelium integrity was scored follows: 0 = normal morphology, 1 = loss of goblet cells in 1 area, 2 = loss of goblet cells in more than one area, 3 = loss of crypts in 1 area, and 4 = loss of crypts in more than one area. Cellular infiltrate was scored as follows: 0 = no infiltrate, 1 = infiltrate around crypt bases, 2 = infiltrate reaching to muscularis mucosae, 3 = extensive infiltration reaching the muscularis, and 4 = infiltration of the submucosa with edema. Finally, lymphoid follicles were scored as none = 0, 1 = 1, 2 = 2, 3 = 3, >3 = 4. Together, these criteria could achieve a total possible score of 16.

Peritoneal cells were collected by washing the peritoneal cavity with 10 ml of ice-cold complete medium (RPMI 1640 plus 10% heat-inactivated fetal calf serum, 100 U of penicillin/ml, 100 μg of streptomycin/ml, and 2 mM l-glutamine; Invitrogen). Splenocyte restimulations and cytokine assays were performed as previously described (24). Briefly, spleens were macerated through 70-μm-pore-size nylon filters (BD Biosciences), and red blood cells were lysed using red blood cell lysis buffer (Sigma). Splenocytes were cultured in triplicate in flat-bottom 96-well plates (10⁶ cells/well) either in medium alone or in medium supplemented with AcES (10 μg/ml), OVA (10 μg/ml), or anti-CD3 (1 μg/ml) for 72 h at 37°C and 5% CO₂. Colon lysates were produced by flushing colons with PBS and placing a small piece of known weight into 1 ml of PBS and lysing on a TissueLyser (Qiagen) with the use of a metal bead. Cell-free supernatants were removed and concentrations of IL-4, IL-5, IL-10, gamma interferon (IFN-γ), IL-17A, and tumor necrosis factor alpha (TNF-α) were measured by using a sandwich enzyme-linked immunosorbent assay (ELISA; OptEIA kits; BD Biosciences).

Peritoneal cells were stained for CD11c-FITC (clone HL3), SIGLEC-F-PE (clone E50-2440) (BD Biosciences), and F4/80-APC (clone BM8; Caltag/Invitrogen), acquired on a FACSCanto flow cytometer (BD Biosciences), and analyzed using FlowJo software (TreeStar). Intracellular cytokine stains were performed on spleen and lymph node cells. Prior to staining, the cells were cultured for 4 h at 37°C and 5% CO₂ in complete medium containing phorbol myristate acetate (500 ng/ml), ionomycin (1 μg/ml), and brefeldin A (10 μg/ml). The cells were stained for CD4-FITC (clone RM4-5; BD Biosciences) and then permeabilized with Fix/Perm buffer (BD Biosciences) and stained for IL-4-PE (clone 11B11; BD Biosciences), IL-10-APC (clone JES5-16E3; eBioscience), and IFN-γ-eF450 (clone XMG1.2; eBioscience).

For peritoneal cells, 10⁶ cells were pelleted by centrifugation and resuspended in 1 ml of TRIzol (Invitrogen). Similarly, a small 0.5-cm piece of colon was washed in PBS, placed in 1 ml of TRIzol, and macerated on a TissueLyser (Qiagen) for 10 min with the use of a metal bead. Total RNA extraction was performed by phenol-chloroform separation according to the manufacturer's instructions. After treatment of RNA with RQ1 DNase (Promega), first-strand cDNA was produced with random hexamers (Invitrogen) from 0.5 to 1 μg of total RNA by using SuperScript III reverse transcriptase (Invitrogen) according to the manufacturer's instructions. The levels of transcription were measured by comparing cross-threshold values to a standard curve made of a pool of all samples. Samples were tested in dilutions of up to 1:600 using SYBR green (Applied Biosystems/Qiagen). A Rotor Gene 6000 (Qiagen) was used for real-time thermal cycling. Melting-curve analysis was used to confirm that a single product had been amplified. All genes were normalized for levels of transcription relative to the housekeeping gene β-actin. All primers were purchased from Sigma-Aldrich and were diluted to a 10 μM final concentration. The primers used were as follows: β-actin, sense (TGGAATCCTGTGGCATCCATGAAAC) and antisense (TAAAACGCAGCTCAGTAACAGTCCG); Fizz-1, sense (GTCCTGGAACCTTTCCTGAG) and antisense (AGCTGGATTGGCAAGAAGTT); Ym1, sense (CTGAGAAGCTCATTGTGGGA) and antisense (CTCAGTGGCTCCTTCATTCA); Arg-1, sense (CAGAAGAATGGAAGAGTCAG) and antisense (CAGATATGCAGGGAGTCACC); NOS-2, sense (ACCTTGTTCAGCTACGCCTT) and antisense (CATTCCCAAATGTGCTTGTC); IL-6, sense (CCGGAGAGGAGACTTCACAG) and antisense (TCCACGATTTCCCAGAGAAC); IL-17A, sense (CCTCCAGAATGTGAAGGTCA) and antisense (CTATCAGGGTCTTCATTGCG); and IFN-γ, sense (AGCTCTTCCTCATGGCTGTT) and antisense (TTTGCCAGTTCCTCCAGATA).

All data were analyzed with GraphPad (version 5; Prism). When three or more groups were compared, one-way analysis of variance (ANOVA) was used with a Bonferroni post-test with a 95% confidence interval to compare all columns. When the effect of a treatment over time was compared for different treatment groups, two-way ANOVA was used with a Bonferroni post-test to compare replicate means over time. P values of <0.05 were considered significant. When only two groups were compared, a Mann-Whitney test was used. All results stated in the text are means ± the standard errors of the mean (SEM). None of the figures presented here are pooled from multiple runs, and all data are representative of at least three repeat experiments.

Ingestion of DSS via the drinking water by mice injected i.p. with vehicle (PBS) or a control protein (OVA) caused rapid weight loss beginning at day 5 posttreatment compared to mice receiving normal drinking water (Fig. 1A). In contrast, i.p. administration of AcES protected against DSS-mediated weight loss (Fig. 1A). Upon euthanasia, the colons were scored for pathology on a macroscopic level (Fig. 1B), and H&E-stained tissue sections scored at a microscopic level (Fig. 1C), showing that AcES-treated mice had significantly lower colon pathology than control mice. Representative transverse sections of the colons demonstrate that mice receiving DSS and either PBS or OVA control treatments had increased cellular infiltrate and edema in the submucosa, whereas mice that had been treated with AcES had markedly less infiltrate and swelling (Fig. 1D).

We next addressed the impact of AcES administration on the expression levels of proinflammatory cytokines associated with DSS-induced pathology. Mice received 3.5% DSS in their drinking water for 8 days and received either a vehicle injection of PBS or 1, 10, or 25 μg of AcES daily i.p. Mesenteric lymph node (MLN) and spleen cells were polyclonally stimulated with anti-CD3 in vitro, and cytokine production was measured by ELISA. The results showed a dose-dependent suppression of IFN-γ and IL-17A expression by MLN cells (Fig. 2A) and splenocytes (data not shown).

We next compared the levels of gene expression at the site of inflammation, the colon, in mice treated with either vehicle or 25 μg of AcES. Critically, expression of the proinflammatory mediators iNOS, IL-6, and IL-17A was significantly reduced by AcES treatment (Fig. 2B), and the levels of IFN-γ tended to be lower but did not reach statistical significance (P = 0.0585). In contrast, IL-4 and IL-10 protein levels in the colon were increased in AcES-treated mice compared to control-treated mice (Fig. 2C). Together, these data indicate that AcES causes downregulation of proinflammatory type 1/type 17 responses potentially by inducing a bias toward a type 2 or regulatory cytokine.

To further investigate the induction of a type 2 cytokine response in mice receiving AcES, mice were injected with either PBS alone or 10 μg of AcES every 2 days for a total of 2 weeks in the absence of any other stimulus. A 10-μg dose was chosen as pilot studies had indicated that 10 μg produced a similar Th2 response as a 25-μg dose in the absence of DSS. Analysis of intracellular cytokine staining in peritoneal lavage cells demonstrated that injection of AcES into mice caused significantly reduced frequencies of CD4⁺ T cells that expressed IFN-γ (Fig. 3A) and significantly increased frequencies of CD4⁺ T cells expressing IL-4 and IL-10, including a prominent population expressing both of these cytokines (Fig. 3B). AcES injection also induced populations of IL-4/IL-10 double-positive CD4 T cells in the spleen and MLN (data not shown). Significant increases in AcES-specific IL-4, IL-5, and IL-10 production by restimulated splenocytes were observed in AcES-treated mice compared to PBS- or OVA-treated control mice (Fig. 3C). Splenocytes from OVA-injected animals restimulated with OVA did not produce significantly elevated levels of IL-4, IL-5, or IL-10 (IL-4, 25.34 ± 4.49 pg/ml; IL-5, 0.0; IL-10, 302.1 ± 151.8 pg/ml). Finally, no significant differences were noted in the production of either TNF or IFN-γ.

Given the ability of AcES to induce a type 2 cytokine bias, we explored whether AcES elicits a downstream innate effector eosinophil and M2 (alternatively activated) macrophage response at the site of injection. Mice injected with AcES had significantly more cells at site of injection than control-treated animals [(20.33 ± 3.69) × 10⁶ versus (4.40 ± 0.74) × 10⁶, P = 0.0055]. Flow cytometric analysis revealed that AcES injection resulted in significantly higher frequencies and total numbers of F4/80⁺ macrophages (P < 0.001) and Siglec-F⁺ eosinophils (P < 0.001) in the peritoneal cavity compared to mice injected with OVA or PBS (Fig. 3D-3F).

M2 macrophages are associated with suppression of T cell responses, and anti-parasite responses (25). Therefore, we analyzed the expression of various M2 macrophage markers in peritoneal cells by reverse transcription-PCR (RT-PCR). Consistent with the macrophages recruited to the peritoneal cavity after AcES injection being of an M2 macrophage phenotype, we detected a significant increase (P < 0.001) in YM1, FIZZ-1, and Arg-1 expression (Fig. 3G) in cells from mice that were injected with AcES compared to control groups. Together, these data indicate that the injection of AcES alone is able to potently modulate the immune status of mice toward a type 2 cytokine response, thereby limiting proinflammatory Th1 and Th17 cytokine responses.

To determine whether the factor(s) within AcES responsible for inducing type 2 cytokine responses are of a protein nature, we digested AcES with trypsin, followed by heat denaturation, a preparation we termed “boiled ES” (bES). Although the injection of AcES in DSS-treated mice resulted in characteristic reductions in IFN-γ expression and increased IL-4 and IL-10 coexpression by CD4⁺ T cells, bES administration had a significantly diminished effect (Fig. 4A). Consistent with an impaired ability to induce a Th2 cytokine response, bES administration resulted in a less-pronounced eosinophil response (Fig. 4B) but still induced significant expansion of macrophages in the peritoneum (Fig. 4C). However, the transcription of M2 macrophage activation markers was significantly reduced in the bES mice (Fig. 4D), suggesting a reduction in alternative activation of macrophages in these mice. Similar results to these were seen when bES was administered to mice in the absence of DSS (data not shown). Hence, the ability of AcES to induce optimal type 2 cytokine responses in mice is predominantly due to heat-labile protein factors.

To assess whether the impaired ability of bES to induce a type 2 cytokine bias results in a reduced capacity to limit disease severity during colitis, we assessed weight loss and intestinal pathology in DSS-treated mice coadministered either AcES or bES. Although AcES treatment resulted in less pronounced weight loss than when mice were treated with PBS, mice injected with bES lost significantly more weight by day 7 than did mice treated with AcES (Fig. 5A). Histological analysis showed that bES-treated mice had significant edema and cellular infiltrate in the submucosa of the colon, whereas the colons of AcES-treated mice appeared relatively healthy (Fig. 5B). These data demonstrate that the component(s) of AcES that mediate protection against DSS-induced colitis is likely a protein.